diff options
Diffstat (limited to 'fs')
622 files changed, 27405 insertions, 16341 deletions
diff --git a/fs/9p/fid.c b/fs/9p/fid.c index b364da70ff2..dfebdbe7440 100644 --- a/fs/9p/fid.c +++ b/fs/9p/fid.c @@ -175,7 +175,7 @@ struct p9_fid *v9fs_fid_lookup(struct dentry *dentry) if (!wnames) return ERR_PTR(-ENOMEM); - for (d = dentry, i = n; i >= 0; i--, d = d->d_parent) + for (d = dentry, i = (n-1); i >= 0; i--, d = d->d_parent) wnames[i] = (char *) d->d_name.name; clone = 1; @@ -183,7 +183,7 @@ struct p9_fid *v9fs_fid_lookup(struct dentry *dentry) while (i < n) { l = min(n - i, P9_MAXWELEM); fid = p9_client_walk(fid, l, &wnames[i], clone); - if (!fid) { + if (IS_ERR(fid)) { kfree(wnames); return fid; } diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c index fbb12dadba8..9b0f0222e8b 100644 --- a/fs/9p/v9fs.c +++ b/fs/9p/v9fs.c @@ -3,7 +3,7 @@ * * This file contains functions assisting in mapping VFS to 9P2000 * - * Copyright (C) 2004 by Eric Van Hensbergen <ericvh@gmail.com> + * Copyright (C) 2004-2008 by Eric Van Hensbergen <ericvh@gmail.com> * Copyright (C) 2002 by Ron Minnich <rminnich@lanl.gov> * * This program is free software; you can redistribute it and/or modify @@ -31,7 +31,6 @@ #include <linux/idr.h> #include <net/9p/9p.h> #include <net/9p/transport.h> -#include <net/9p/conn.h> #include <net/9p/client.h> #include "v9fs.h" #include "v9fs_vfs.h" @@ -43,11 +42,11 @@ enum { /* Options that take integer arguments */ - Opt_debug, Opt_msize, Opt_dfltuid, Opt_dfltgid, Opt_afid, + Opt_debug, Opt_dfltuid, Opt_dfltgid, Opt_afid, /* String options */ Opt_uname, Opt_remotename, Opt_trans, /* Options that take no arguments */ - Opt_legacy, Opt_nodevmap, + Opt_nodevmap, /* Cache options */ Opt_cache_loose, /* Access options */ @@ -58,14 +57,11 @@ enum { static match_table_t tokens = { {Opt_debug, "debug=%x"}, - {Opt_msize, "msize=%u"}, {Opt_dfltuid, "dfltuid=%u"}, {Opt_dfltgid, "dfltgid=%u"}, {Opt_afid, "afid=%u"}, {Opt_uname, "uname=%s"}, {Opt_remotename, "aname=%s"}, - {Opt_trans, "trans=%s"}, - {Opt_legacy, "noextend"}, {Opt_nodevmap, "nodevmap"}, {Opt_cache_loose, "cache=loose"}, {Opt_cache_loose, "loose"}, @@ -85,16 +81,14 @@ static void v9fs_parse_options(struct v9fs_session_info *v9ses) char *options; substring_t args[MAX_OPT_ARGS]; char *p; - int option; - int ret; + int option = 0; char *s, *e; + int ret; /* setup defaults */ - v9ses->maxdata = 8192; v9ses->afid = ~0; v9ses->debug = 0; v9ses->cache = 0; - v9ses->trans = v9fs_default_trans(); if (!v9ses->options) return; @@ -106,7 +100,8 @@ static void v9fs_parse_options(struct v9fs_session_info *v9ses) continue; token = match_token(p, tokens, args); if (token < Opt_uname) { - if ((ret = match_int(&args[0], &option)) < 0) { + ret = match_int(&args[0], &option); + if (ret < 0) { P9_DPRINTK(P9_DEBUG_ERROR, "integer field, but no integer?\n"); continue; @@ -119,9 +114,7 @@ static void v9fs_parse_options(struct v9fs_session_info *v9ses) p9_debug_level = option; #endif break; - case Opt_msize: - v9ses->maxdata = option; - break; + case Opt_dfltuid: v9ses->dfltuid = option; break; @@ -131,18 +124,12 @@ static void v9fs_parse_options(struct v9fs_session_info *v9ses) case Opt_afid: v9ses->afid = option; break; - case Opt_trans: - v9ses->trans = v9fs_match_trans(&args[0]); - break; case Opt_uname: match_strcpy(v9ses->uname, &args[0]); break; case Opt_remotename: match_strcpy(v9ses->aname, &args[0]); break; - case Opt_legacy: - v9ses->flags &= ~V9FS_EXTENDED; - break; case Opt_nodevmap: v9ses->nodev = 1; break; @@ -185,7 +172,6 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses, const char *dev_name, char *data) { int retval = -EINVAL; - struct p9_trans *trans = NULL; struct p9_fid *fid; v9ses->uname = __getname(); @@ -207,24 +193,7 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses, v9ses->options = kstrdup(data, GFP_KERNEL); v9fs_parse_options(v9ses); - if (v9ses->trans == NULL) { - retval = -EPROTONOSUPPORT; - P9_DPRINTK(P9_DEBUG_ERROR, - "No transport defined or default transport\n"); - goto error; - } - - trans = v9ses->trans->create(dev_name, v9ses->options); - if (IS_ERR(trans)) { - retval = PTR_ERR(trans); - trans = NULL; - goto error; - } - if ((v9ses->maxdata+P9_IOHDRSZ) > v9ses->trans->maxsize) - v9ses->maxdata = v9ses->trans->maxsize-P9_IOHDRSZ; - - v9ses->clnt = p9_client_create(trans, v9ses->maxdata+P9_IOHDRSZ, - v9fs_extended(v9ses)); + v9ses->clnt = p9_client_create(dev_name, v9ses->options); if (IS_ERR(v9ses->clnt)) { retval = PTR_ERR(v9ses->clnt); @@ -236,6 +205,8 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses, if (!v9ses->clnt->dotu) v9ses->flags &= ~V9FS_EXTENDED; + v9ses->maxdata = v9ses->clnt->msize; + /* for legacy mode, fall back to V9FS_ACCESS_ANY */ if (!v9fs_extended(v9ses) && ((v9ses->flags&V9FS_ACCESS_MASK) == V9FS_ACCESS_USER)) { diff --git a/fs/9p/v9fs.h b/fs/9p/v9fs.h index db4b4193f2e..7d3a1018db5 100644 --- a/fs/9p/v9fs.h +++ b/fs/9p/v9fs.h @@ -1,7 +1,7 @@ /* * V9FS definitions. * - * Copyright (C) 2004 by Eric Van Hensbergen <ericvh@gmail.com> + * Copyright (C) 2004-2008 by Eric Van Hensbergen <ericvh@gmail.com> * Copyright (C) 2002 by Ron Minnich <rminnich@lanl.gov> * * This program is free software; you can redistribute it and/or modify @@ -28,7 +28,6 @@ struct v9fs_session_info { /* options */ - unsigned int maxdata; unsigned char flags; /* session flags */ unsigned char nodev; /* set to 1 if no disable device mapping */ unsigned short debug; /* debug level */ @@ -38,10 +37,10 @@ struct v9fs_session_info { char *options; /* copy of mount options */ char *uname; /* user name to mount as */ char *aname; /* name of remote hierarchy being mounted */ + unsigned int maxdata; /* max data for client interface */ unsigned int dfltuid; /* default uid/muid for legacy support */ unsigned int dfltgid; /* default gid for legacy support */ u32 uid; /* if ACCESS_SINGLE, the uid that has access */ - struct p9_trans_module *trans; /* 9p transport */ struct p9_client *clnt; /* 9p client */ struct dentry *debugfs_dir; }; diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c index ba4b1caa9c4..a616fff8906 100644 --- a/fs/9p/vfs_file.c +++ b/fs/9p/vfs_file.c @@ -184,7 +184,7 @@ static const struct file_operations v9fs_cached_file_operations = { .open = v9fs_file_open, .release = v9fs_dir_release, .lock = v9fs_file_lock, - .mmap = generic_file_mmap, + .mmap = generic_file_readonly_mmap, }; const struct file_operations v9fs_file_operations = { @@ -194,5 +194,5 @@ const struct file_operations v9fs_file_operations = { .open = v9fs_file_open, .release = v9fs_dir_release, .lock = v9fs_file_lock, - .mmap = generic_file_mmap, + .mmap = generic_file_readonly_mmap, }; diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 23581bcb599..6a28842052e 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -77,6 +77,8 @@ static int unixmode2p9mode(struct v9fs_session_info *v9ses, int mode) res |= P9_DMSETUID; if ((mode & S_ISGID) == S_ISGID) res |= P9_DMSETGID; + if ((mode & S_ISVTX) == S_ISVTX) + res |= P9_DMSETVTX; if ((mode & P9_DMLINK)) res |= P9_DMLINK; } @@ -119,6 +121,9 @@ static int p9mode2unixmode(struct v9fs_session_info *v9ses, int mode) if ((mode & P9_DMSETGID) == P9_DMSETGID) res |= S_ISGID; + + if ((mode & P9_DMSETVTX) == P9_DMSETVTX) + res |= S_ISVTX; } return res; @@ -568,7 +573,7 @@ static struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, v9ses = v9fs_inode2v9ses(dir); dfid = v9fs_fid_lookup(dentry->d_parent); if (IS_ERR(dfid)) - return ERR_PTR(PTR_ERR(dfid)); + return ERR_CAST(dfid); name = (char *) dentry->d_name.name; fid = p9_client_walk(dfid, 1, &name, 1); diff --git a/fs/Kconfig b/fs/Kconfig index 781b47d2f9f..c509123bea4 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -236,6 +236,7 @@ config JBD_DEBUG config JBD2 tristate + select CRC32 help This is a generic journaling layer for block devices that support both 32-bit and 64-bit block numbers. It is currently used by @@ -440,14 +441,8 @@ config OCFS2_FS Tools web page: http://oss.oracle.com/projects/ocfs2-tools OCFS2 mailing lists: http://oss.oracle.com/projects/ocfs2/mailman/ - Note: Features which OCFS2 does not support yet: - - extended attributes - - quotas - - cluster aware flock - - Directory change notification (F_NOTIFY) - - Distributed Caching (F_SETLEASE/F_GETLEASE/break_lease) - - POSIX ACLs - - readpages / writepages (not user visible) + For more information on OCFS2, see the file + <file:Documentation/filesystems/ocfs2.txt>. config OCFS2_DEBUG_MASKLOG bool "OCFS2 logging support" @@ -468,40 +463,18 @@ config OCFS2_DEBUG_FS this option for debugging only as it is likely to decrease performance of the filesystem. -config MINIX_FS - tristate "Minix fs support" - help - Minix is a simple operating system used in many classes about OS's. - The minix file system (method to organize files on a hard disk - partition or a floppy disk) was the original file system for Linux, - but has been superseded by the second extended file system ext2fs. - You don't want to use the minix file system on your hard disk - because of certain built-in restrictions, but it is sometimes found - on older Linux floppy disks. This option will enlarge your kernel - by about 28 KB. If unsure, say N. - - To compile this file system support as a module, choose M here: the - module will be called minix. Note that the file system of your root - partition (the one containing the directory /) cannot be compiled as - a module. - -config ROMFS_FS - tristate "ROM file system support" - ---help--- - This is a very small read-only file system mainly intended for - initial ram disks of installation disks, but it could be used for - other read-only media as well. Read - <file:Documentation/filesystems/romfs.txt> for details. - - To compile this file system support as a module, choose M here: the - module will be called romfs. Note that the file system of your - root partition (the one containing the directory /) cannot be a - module. +endif # BLOCK - If you don't know whether you need it, then you don't need it: - answer N. +config DNOTIFY + bool "Dnotify support" + default y + help + Dnotify is a directory-based per-fd file change notification system + that uses signals to communicate events to user-space. There exist + superior alternatives, but some applications may still rely on + dnotify. -endif + If unsure, say Y. config INOTIFY bool "Inotify file change notification support" @@ -582,17 +555,6 @@ config QUOTACTL depends on XFS_QUOTA || QUOTA default y -config DNOTIFY - bool "Dnotify support" - default y - help - Dnotify is a directory-based per-fd file change notification system - that uses signals to communicate events to user-space. There exist - superior alternatives, but some applications may still rely on - dnotify. - - If unsure, say Y. - config AUTOFS_FS tristate "Kernel automounter support" help @@ -718,7 +680,7 @@ config UDF_NLS depends on (UDF_FS=m && NLS) || (UDF_FS=y && NLS=y) endmenu -endif +endif # BLOCK if BLOCK menu "DOS/FAT/NT Filesystems" @@ -901,7 +863,7 @@ config NTFS_RW It is perfectly safe to say N here. endmenu -endif +endif # BLOCK menu "Pseudo filesystems" @@ -1028,8 +990,8 @@ config HUGETLB_PAGE def_bool HUGETLBFS config CONFIGFS_FS - tristate "Userspace-driven configuration filesystem (EXPERIMENTAL)" - depends on SYSFS && EXPERIMENTAL + tristate "Userspace-driven configuration filesystem" + depends on SYSFS help configfs is a ram-based filesystem that provides the converse of sysfs's functionality. Where sysfs is a filesystem-based @@ -1157,7 +1119,7 @@ config BEFS_DEBUG depends on BEFS_FS help If you say Y here, you can use the 'debug' mount option to enable - debugging output from the driver. + debugging output from the driver. config BFS_FS tristate "BFS file system support (EXPERIMENTAL)" @@ -1268,7 +1230,7 @@ config JFFS2_FS_XATTR Extended attributes are name:value pairs associated with inodes by the kernel or by users (see the attr(5) manual page, or visit <http://acl.bestbits.at/> for details). - + If unsure, say N. config JFFS2_FS_POSIX_ACL @@ -1279,10 +1241,10 @@ config JFFS2_FS_POSIX_ACL help Posix Access Control Lists (ACLs) support permissions for users and groups beyond the owner/group/world scheme. - + To learn more about Access Control Lists, visit the Posix ACLs for Linux website <http://acl.bestbits.at/>. - + If you don't know what Access Control Lists are, say N config JFFS2_FS_SECURITY @@ -1294,7 +1256,7 @@ config JFFS2_FS_SECURITY implemented by security modules like SELinux. This option enables an extended attribute handler for file security labels in the jffs2 filesystem. - + If you are not using a security module that requires using extended attributes for file security labels, say N. @@ -1422,6 +1384,24 @@ config VXFS_FS To compile this as a module, choose M here: the module will be called freevxfs. If unsure, say N. +config MINIX_FS + tristate "Minix file system support" + depends on BLOCK + help + Minix is a simple operating system used in many classes about OS's. + The minix file system (method to organize files on a hard disk + partition or a floppy disk) was the original file system for Linux, + but has been superseded by the second extended file system ext2fs. + You don't want to use the minix file system on your hard disk + because of certain built-in restrictions, but it is sometimes found + on older Linux floppy disks. This option will enlarge your kernel + by about 28 KB. If unsure, say N. + + To compile this file system support as a module, choose M here: the + module will be called minix. Note that the file system of your root + partition (the one containing the directory /) cannot be compiled as + a module. + config HPFS_FS tristate "OS/2 HPFS file system support" @@ -1439,7 +1419,6 @@ config HPFS_FS module will be called hpfs. If unsure, say N. - config QNX4FS_FS tristate "QNX4 file system support (read only)" depends on BLOCK @@ -1466,6 +1445,22 @@ config QNX4FS_RW It's currently broken, so for now: answer N. +config ROMFS_FS + tristate "ROM file system support" + depends on BLOCK + ---help--- + This is a very small read-only file system mainly intended for + initial ram disks of installation disks, but it could be used for + other read-only media as well. Read + <file:Documentation/filesystems/romfs.txt> for details. + + To compile this file system support as a module, choose M here: the + module will be called romfs. Note that the file system of your + root partition (the one containing the directory /) cannot be a + module. + + If you don't know whether you need it, then you don't need it: + answer N. config SYSV_FS @@ -1506,7 +1501,6 @@ config SYSV_FS If you haven't heard about all of this before, it's safe to say N. - config UFS_FS tristate "UFS file system support (read only)" depends on BLOCK @@ -1679,6 +1673,8 @@ config NFSD select CRYPTO_MD5 if NFSD_V4 select CRYPTO if NFSD_V4 select FS_POSIX_ACL if NFSD_V4 + select PROC_FS if NFSD_V4 + select PROC_FS if SUNRPC_GSS help If you want your Linux box to act as an NFS *server*, so that other computers on your local network which support NFS can access certain @@ -1748,10 +1744,10 @@ config ROOT_NFS If you want your Linux box to mount its whole root file system (the one containing the directory /) from some other computer over the net via NFS (presumably because your box doesn't have a hard disk), - say Y. Read <file:Documentation/nfsroot.txt> for details. It is - likely that in this case, you also want to say Y to "Kernel level IP - autoconfiguration" so that your box can discover its network address - at boot time. + say Y. Read <file:Documentation/filesystems/nfsroot.txt> for + details. It is likely that in this case, you also want to say Y to + "Kernel level IP autoconfiguration" so that your box can discover + its network address at boot time. Most people say N here. @@ -1782,12 +1778,9 @@ config SUNRPC_GSS tristate config SUNRPC_XPRT_RDMA - tristate "RDMA transport for sunrpc (EXPERIMENTAL)" + tristate depends on SUNRPC && INFINIBAND && EXPERIMENTAL - default m - help - Adds a client RPC transport for supporting kernel NFS over RDMA - mounts, including Infiniband and iWARP. Experimental. + default SUNRPC && INFINIBAND config SUNRPC_BIND34 bool "Support for rpcbind versions 3 & 4 (EXPERIMENTAL)" @@ -1838,7 +1831,7 @@ config RPCSEC_GSS_SPKM3 If unsure, say N. config SMB_FS - tristate "SMB file system support (to mount Windows shares etc.)" + tristate "SMB file system support (OBSOLETE, please use CIFS)" depends on INET select NLS help @@ -1861,8 +1854,8 @@ config SMB_FS General information about how to connect Linux, Windows machines and Macs is on the WWW at <http://www.eats.com/linux_mac_win.html>. - To compile the SMB support as a module, choose M here: the module will - be called smbfs. Most people say N, however. + To compile the SMB support as a module, choose M here: + the module will be called smbfs. Most people say N, however. config SMB_NLS_DEFAULT bool "Use a default NLS" @@ -1894,7 +1887,7 @@ config SMB_NLS_REMOTE smbmount from samba 2.2.0 or later supports this. config CIFS - tristate "CIFS support (advanced network filesystem for Samba, Window and other CIFS compliant servers)" + tristate "CIFS support (advanced network filesystem, SMBFS successor)" depends on INET select NLS help @@ -1905,13 +1898,15 @@ config CIFS file servers such as Windows 2000 (including Windows 2003, NT 4 and Windows XP) as well by Samba (which provides excellent CIFS server support for Linux and many other operating systems). Limited - support for OS/2 and Windows ME and similar servers is provided as well. - - The intent of the cifs module is to provide an advanced - network file system client for mounting to CIFS compliant servers, - including support for dfs (hierarchical name space), secure per-user - session establishment, safe distributed caching (oplock), optional - packet signing, Unicode and other internationalization improvements. + support for OS/2 and Windows ME and similar servers is provided as + well. + + The cifs module provides an advanced network file system + client for mounting to CIFS compliant servers. It includes + support for DFS (hierarchical name space), secure per-user + session establishment via Kerberos or NTLM or NTLMv2, + safe distributed caching (oplock), optional packet + signing, Unicode and other internationalization improvements. If you need to mount to Samba or Windows from this machine, say Y. config CIFS_STATS @@ -1943,22 +1938,23 @@ config CIFS_WEAK_PW_HASH (since 1997) support stronger NTLM (and even NTLMv2 and Kerberos) security mechanisms. These hash the password more securely than the mechanisms used in the older LANMAN version of the - SMB protocol needed to establish sessions with old SMB servers. + SMB protocol but LANMAN based authentication is needed to + establish sessions with some old SMB servers. Enabling this option allows the cifs module to mount to older LANMAN based servers such as OS/2 and Windows 95, but such mounts may be less secure than mounts using NTLM or more recent security mechanisms if you are on a public network. Unless you - have a need to access old SMB servers (and are on a private + have a need to access old SMB servers (and are on a private network) you probably want to say N. Even if this support - is enabled in the kernel build, they will not be used - automatically. At runtime LANMAN mounts are disabled but + is enabled in the kernel build, LANMAN authentication will not be + used automatically. At runtime LANMAN mounts are disabled but can be set to required (or optional) either in /proc/fs/cifs (see fs/cifs/README for more detail) or via an - option on the mount command. This support is disabled by + option on the mount command. This support is disabled by default in order to reduce the possibility of a downgrade attack. - + If unsure, say N. config CIFS_XATTR @@ -1999,7 +1995,7 @@ config CIFS_DEBUG2 messages in some error paths, slowing performance. This option can be turned off unless you are debugging cifs problems. If unsure, say N. - + config CIFS_EXPERIMENTAL bool "CIFS Experimental Features (EXPERIMENTAL)" depends on CIFS && EXPERIMENTAL @@ -2018,12 +2014,22 @@ config CIFS_UPCALL depends on CIFS_EXPERIMENTAL depends on KEYS help - Enables an upcall mechanism for CIFS which will be used to contact - userspace helper utilities to provide SPNEGO packaged Kerberos - tickets which are needed to mount to certain secure servers + Enables an upcall mechanism for CIFS which accesses + userspace helper utilities to provide SPNEGO packaged (RFC 4178) + Kerberos tickets which are needed to mount to certain secure servers (for which more secure Kerberos authentication is required). If unsure, say N. +config CIFS_DFS_UPCALL + bool "DFS feature support (EXPERIMENTAL)" + depends on CIFS_EXPERIMENTAL + depends on KEYS + help + Enables an upcall mechanism for CIFS which contacts userspace + helper utilities to provide server name resolution (host names to + IP addresses) which is needed for implicit mounts of DFS junction + points. If unsure, say N. + config NCP_FS tristate "NCP file system support (to mount NetWare volumes)" depends on IPX!=n || INET @@ -2080,7 +2086,7 @@ config CODA_FS_OLD_API However this new API is not backward compatible with older clients. If you really need to run the old Coda userspace cache manager then say Y. - + For most cases you probably want to say N. config AFS_FS @@ -2130,4 +2136,3 @@ source "fs/nls/Kconfig" source "fs/dlm/Kconfig" endmenu - diff --git a/fs/Kconfig.binfmt b/fs/Kconfig.binfmt index d4fc6095466..b5c3b6114ad 100644 --- a/fs/Kconfig.binfmt +++ b/fs/Kconfig.binfmt @@ -23,6 +23,10 @@ config BINFMT_ELF ld.so (check the file <file:Documentation/Changes> for location and latest version). +config COMPAT_BINFMT_ELF + bool + depends on COMPAT && MMU + config BINFMT_ELF_FDPIC bool "Kernel support for FDPIC ELF binaries" default y @@ -57,7 +61,8 @@ config BINFMT_SHARED_FLAT config BINFMT_AOUT tristate "Kernel support for a.out and ECOFF binaries" - depends on X86_32 || ALPHA || ARM || M68K || SPARC32 + depends on ARCH_SUPPORTS_AOUT && \ + (X86_32 || ALPHA || ARM || M68K || SPARC32) ---help--- A.out (Assembler.OUTput) is a set of formats for libraries and executables used in the earliest versions of UNIX. Linux used diff --git a/fs/Makefile b/fs/Makefile index 500cf15cdb4..1e7a11bd4da 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -39,6 +39,7 @@ obj-$(CONFIG_BINFMT_MISC) += binfmt_misc.o obj-y += binfmt_script.o obj-$(CONFIG_BINFMT_ELF) += binfmt_elf.o +obj-$(CONFIG_COMPAT_BINFMT_ELF) += compat_binfmt_elf.o obj-$(CONFIG_BINFMT_ELF_FDPIC) += binfmt_elf_fdpic.o obj-$(CONFIG_BINFMT_SOM) += binfmt_som.o obj-$(CONFIG_BINFMT_FLAT) += binfmt_flat.o diff --git a/fs/adfs/super.c b/fs/adfs/super.c index b36695ae5c2..9e421eeb672 100644 --- a/fs/adfs/super.c +++ b/fs/adfs/super.c @@ -20,6 +20,8 @@ #include <linux/vfs.h> #include <linux/parser.h> #include <linux/bitops.h> +#include <linux/mount.h> +#include <linux/seq_file.h> #include <asm/uaccess.h> #include <asm/system.h> @@ -30,6 +32,9 @@ #include "dir_f.h" #include "dir_fplus.h" +#define ADFS_DEFAULT_OWNER_MASK S_IRWXU +#define ADFS_DEFAULT_OTHER_MASK (S_IRWXG | S_IRWXO) + void __adfs_error(struct super_block *sb, const char *function, const char *fmt, ...) { char error_buf[128]; @@ -134,6 +139,22 @@ static void adfs_put_super(struct super_block *sb) sb->s_fs_info = NULL; } +static int adfs_show_options(struct seq_file *seq, struct vfsmount *mnt) +{ + struct adfs_sb_info *asb = ADFS_SB(mnt->mnt_sb); + + if (asb->s_uid != 0) + seq_printf(seq, ",uid=%u", asb->s_uid); + if (asb->s_gid != 0) + seq_printf(seq, ",gid=%u", asb->s_gid); + if (asb->s_owner_mask != ADFS_DEFAULT_OWNER_MASK) + seq_printf(seq, ",ownmask=%o", asb->s_owner_mask); + if (asb->s_other_mask != ADFS_DEFAULT_OTHER_MASK) + seq_printf(seq, ",othmask=%o", asb->s_other_mask); + + return 0; +} + enum {Opt_uid, Opt_gid, Opt_ownmask, Opt_othmask, Opt_err}; static match_table_t tokens = { @@ -259,6 +280,7 @@ static const struct super_operations adfs_sops = { .put_super = adfs_put_super, .statfs = adfs_statfs, .remount_fs = adfs_remount, + .show_options = adfs_show_options, }; static struct adfs_discmap *adfs_read_map(struct super_block *sb, struct adfs_discrecord *dr) @@ -344,8 +366,8 @@ static int adfs_fill_super(struct super_block *sb, void *data, int silent) /* set default options */ asb->s_uid = 0; asb->s_gid = 0; - asb->s_owner_mask = S_IRWXU; - asb->s_other_mask = S_IRWXG | S_IRWXO; + asb->s_owner_mask = ADFS_DEFAULT_OWNER_MASK; + asb->s_other_mask = ADFS_DEFAULT_OTHER_MASK; if (parse_options(sb, data)) goto error; diff --git a/fs/affs/affs.h b/fs/affs/affs.h index 232c6949368..d5bd497ab9c 100644 --- a/fs/affs/affs.h +++ b/fs/affs/affs.h @@ -174,7 +174,8 @@ extern void affs_put_inode(struct inode *inode); extern void affs_drop_inode(struct inode *inode); extern void affs_delete_inode(struct inode *inode); extern void affs_clear_inode(struct inode *inode); -extern void affs_read_inode(struct inode *inode); +extern struct inode *affs_iget(struct super_block *sb, + unsigned long ino); extern int affs_write_inode(struct inode *inode, int); extern int affs_add_entry(struct inode *dir, struct inode *inode, struct dentry *dentry, s32 type); diff --git a/fs/affs/amigaffs.c b/fs/affs/amigaffs.c index f4de4b98004..805573005de 100644 --- a/fs/affs/amigaffs.c +++ b/fs/affs/amigaffs.c @@ -170,9 +170,11 @@ affs_remove_link(struct dentry *dentry) if (!link_bh) goto done; - dir = iget(sb, be32_to_cpu(AFFS_TAIL(sb, link_bh)->parent)); - if (!dir) + dir = affs_iget(sb, be32_to_cpu(AFFS_TAIL(sb, link_bh)->parent)); + if (IS_ERR(dir)) { + retval = PTR_ERR(dir); goto done; + } affs_lock_dir(dir); affs_fix_dcache(dentry, link_ino); diff --git a/fs/affs/inode.c b/fs/affs/inode.c index 4609a6c13fe..27fe6cbe43a 100644 --- a/fs/affs/inode.c +++ b/fs/affs/inode.c @@ -15,20 +15,25 @@ extern const struct inode_operations affs_symlink_inode_operations; extern struct timezone sys_tz; -void -affs_read_inode(struct inode *inode) +struct inode *affs_iget(struct super_block *sb, unsigned long ino) { - struct super_block *sb = inode->i_sb; struct affs_sb_info *sbi = AFFS_SB(sb); struct buffer_head *bh; struct affs_head *head; struct affs_tail *tail; + struct inode *inode; u32 block; u32 size; u32 prot; u16 id; - pr_debug("AFFS: read_inode(%lu)\n",inode->i_ino); + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; + + pr_debug("AFFS: affs_iget(%lu)\n", inode->i_ino); block = inode->i_ino; bh = affs_bread(sb, block); @@ -154,12 +159,13 @@ affs_read_inode(struct inode *inode) sys_tz.tz_minuteswest * 60; inode->i_mtime.tv_nsec = inode->i_ctime.tv_nsec = inode->i_atime.tv_nsec = 0; affs_brelse(bh); - return; + unlock_new_inode(inode); + return inode; bad_inode: - make_bad_inode(inode); affs_brelse(bh); - return; + iget_failed(inode); + return ERR_PTR(-EIO); } int diff --git a/fs/affs/namei.c b/fs/affs/namei.c index a42143ca016..2218f1ee71c 100644 --- a/fs/affs/namei.c +++ b/fs/affs/namei.c @@ -208,9 +208,8 @@ affs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) affs_lock_dir(dir); bh = affs_find_entry(dir, dentry); affs_unlock_dir(dir); - if (IS_ERR(bh)) { - return ERR_PTR(PTR_ERR(bh)); - } + if (IS_ERR(bh)) + return ERR_CAST(bh); if (bh) { u32 ino = bh->b_blocknr; @@ -223,10 +222,9 @@ affs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) ino = be32_to_cpu(AFFS_TAIL(sb, bh)->original); } affs_brelse(bh); - inode = iget(sb, ino); - if (!inode) { - return ERR_PTR(-EACCES); - } + inode = affs_iget(sb, ino); + if (IS_ERR(inode)) + return ERR_PTR(PTR_ERR(inode)); } dentry->d_op = AFFS_SB(sb)->s_flags & SF_INTL ? &affs_intl_dentry_operations : &affs_dentry_operations; d_add(dentry, inode); diff --git a/fs/affs/super.c b/fs/affs/super.c index b53e5d0ec65..d2dc047cb47 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -113,7 +113,6 @@ static void destroy_inodecache(void) static const struct super_operations affs_sops = { .alloc_inode = affs_alloc_inode, .destroy_inode = affs_destroy_inode, - .read_inode = affs_read_inode, .write_inode = affs_write_inode, .put_inode = affs_put_inode, .drop_inode = affs_drop_inode, @@ -123,6 +122,7 @@ static const struct super_operations affs_sops = { .write_super = affs_write_super, .statfs = affs_statfs, .remount_fs = affs_remount, + .show_options = generic_show_options, }; enum { @@ -271,6 +271,9 @@ static int affs_fill_super(struct super_block *sb, void *data, int silent) unsigned long mount_flags; int tmp_flags; /* fix remount prototype... */ u8 sig[4]; + int ret = -EINVAL; + + save_mount_options(sb, data); pr_debug("AFFS: read_super(%s)\n",data ? (const char *)data : "no options"); @@ -444,7 +447,12 @@ got_root: /* set up enough so that it can read an inode */ - root_inode = iget(sb, root_block); + root_inode = affs_iget(sb, root_block); + if (IS_ERR(root_inode)) { + ret = PTR_ERR(root_inode); + goto out_error_noinode; + } + sb->s_root = d_alloc_root(root_inode); if (!sb->s_root) { printk(KERN_ERR "AFFS: Get root inode failed\n"); @@ -461,12 +469,13 @@ got_root: out_error: if (root_inode) iput(root_inode); +out_error_noinode: kfree(sbi->s_bitmap); affs_brelse(root_bh); kfree(sbi->s_prefix); kfree(sbi); sb->s_fs_info = NULL; - return -EINVAL; + return ret; } static int @@ -481,14 +490,21 @@ affs_remount(struct super_block *sb, int *flags, char *data) int root_block; unsigned long mount_flags; int res = 0; + char *new_opts = kstrdup(data, GFP_KERNEL); pr_debug("AFFS: remount(flags=0x%x,opts=\"%s\")\n",*flags,data); *flags |= MS_NODIRATIME; - if (!parse_options(data,&uid,&gid,&mode,&reserved,&root_block, - &blocksize,&sbi->s_prefix,sbi->s_volume,&mount_flags)) + if (!parse_options(data, &uid, &gid, &mode, &reserved, &root_block, + &blocksize, &sbi->s_prefix, sbi->s_volume, + &mount_flags)) { + kfree(new_opts); return -EINVAL; + } + kfree(sb->s_options); + sb->s_options = new_opts; + sbi->s_flags = mount_flags; sbi->s_mode = mode; sbi->s_uid = uid; diff --git a/fs/afs/cell.c b/fs/afs/cell.c index 970d38f3056..584bb0f9c36 100644 --- a/fs/afs/cell.c +++ b/fs/afs/cell.c @@ -127,14 +127,21 @@ struct afs_cell *afs_cell_create(const char *name, char *vllist) _enter("%s,%s", name, vllist); + down_write(&afs_cells_sem); + read_lock(&afs_cells_lock); + list_for_each_entry(cell, &afs_cells, link) { + if (strcasecmp(cell->name, name) == 0) + goto duplicate_name; + } + read_unlock(&afs_cells_lock); + cell = afs_cell_alloc(name, vllist); if (IS_ERR(cell)) { _leave(" = %ld", PTR_ERR(cell)); + up_write(&afs_cells_sem); return cell; } - down_write(&afs_cells_sem); - /* add a proc directory for this cell */ ret = afs_proc_cell_setup(cell); if (ret < 0) @@ -167,6 +174,11 @@ error: kfree(cell); _leave(" = %d", ret); return ERR_PTR(ret); + +duplicate_name: + read_unlock(&afs_cells_lock); + up_write(&afs_cells_sem); + return ERR_PTR(-EEXIST); } /* diff --git a/fs/afs/dir.c b/fs/afs/dir.c index 33fe39ad4e0..b58af8f18bc 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -512,7 +512,7 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, key = afs_request_key(vnode->volume->cell); if (IS_ERR(key)) { _leave(" = %ld [key]", PTR_ERR(key)); - return ERR_PTR(PTR_ERR(key)); + return ERR_CAST(key); } ret = afs_validate(vnode, key); @@ -540,17 +540,17 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, key_put(key); if (IS_ERR(inode)) { _leave(" = %ld", PTR_ERR(inode)); - return ERR_PTR(PTR_ERR(inode)); + return ERR_CAST(inode); } dentry->d_op = &afs_fs_dentry_operations; d_add(dentry, inode); - _leave(" = 0 { vn=%u u=%u } -> { ino=%lu v=%lu }", + _leave(" = 0 { vn=%u u=%u } -> { ino=%lu v=%llu }", fid.vnode, fid.unique, dentry->d_inode->i_ino, - dentry->d_inode->i_version); + (unsigned long long)dentry->d_inode->i_version); return NULL; } @@ -630,9 +630,10 @@ static int afs_d_revalidate(struct dentry *dentry, struct nameidata *nd) * been deleted and replaced, and the original vnode ID has * been reused */ if (fid.unique != vnode->fid.unique) { - _debug("%s: file deleted (uq %u -> %u I:%lu)", + _debug("%s: file deleted (uq %u -> %u I:%llu)", dentry->d_name.name, fid.unique, - vnode->fid.unique, dentry->d_inode->i_version); + vnode->fid.unique, + (unsigned long long)dentry->d_inode->i_version); spin_lock(&vnode->lock); set_bit(AFS_VNODE_DELETED, &vnode->flags); spin_unlock(&vnode->lock); diff --git a/fs/afs/inode.c b/fs/afs/inode.c index d196840127c..08db82e1343 100644 --- a/fs/afs/inode.c +++ b/fs/afs/inode.c @@ -196,10 +196,7 @@ struct inode *afs_iget(struct super_block *sb, struct key *key, /* failure */ bad_inode: - make_bad_inode(inode); - unlock_new_inode(inode); - iput(inode); - + iget_failed(inode); _leave(" = %d [bad]", ret); return ERR_PTR(ret); } @@ -301,7 +298,8 @@ int afs_getattr(struct vfsmount *mnt, struct dentry *dentry, inode = dentry->d_inode; - _enter("{ ino=%lu v=%lu }", inode->i_ino, inode->i_version); + _enter("{ ino=%lu v=%llu }", inode->i_ino, + (unsigned long long)inode->i_version); generic_fillattr(inode, stat); return 0; diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 5ca3625cd39..eec41c76de7 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -573,7 +573,6 @@ extern const struct file_operations afs_mntpt_file_operations; extern int afs_mntpt_check_symlink(struct afs_vnode *, struct key *); extern void afs_mntpt_kill_timer(void); -extern void afs_umount_begin(struct vfsmount *, int); /* * proc.c @@ -750,7 +749,7 @@ extern int afs_fsync(struct file *, struct dentry *, int); extern unsigned afs_debug; #define dbgprintk(FMT,...) \ - printk("[%x%-6.6s] "FMT"\n", smp_processor_id(), current->comm ,##__VA_ARGS__) + printk("[%-6.6s] "FMT"\n", current->comm ,##__VA_ARGS__) /* make sure we maintain the format strings, even when debugging is disabled */ static inline __attribute__((format(printf,1,2))) diff --git a/fs/afs/main.c b/fs/afs/main.c index 0f60f6b3576..2d3e5d4fb9f 100644 --- a/fs/afs/main.c +++ b/fs/afs/main.c @@ -22,7 +22,7 @@ MODULE_LICENSE("GPL"); unsigned afs_debug; module_param_named(debug, afs_debug, uint, S_IWUSR | S_IRUGO); -MODULE_PARM_DESC(afs_debug, "AFS debugging mask"); +MODULE_PARM_DESC(debug, "AFS debugging mask"); static char *rootcell; diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c index 5ce43b63c60..2f5503902c3 100644 --- a/fs/afs/mntpt.c +++ b/fs/afs/mntpt.c @@ -218,16 +218,16 @@ static void *afs_mntpt_follow_link(struct dentry *dentry, struct nameidata *nd) _enter("%p{%s},{%s:%p{%s},}", dentry, dentry->d_name.name, - nd->mnt->mnt_devname, + nd->path.mnt->mnt_devname, dentry, - nd->dentry->d_name.name); + nd->path.dentry->d_name.name); - dput(nd->dentry); - nd->dentry = dget(dentry); + dput(nd->path.dentry); + nd->path.dentry = dget(dentry); - newmnt = afs_mntpt_do_automount(nd->dentry); + newmnt = afs_mntpt_do_automount(nd->path.dentry); if (IS_ERR(newmnt)) { - path_release(nd); + path_put(&nd->path); return (void *)newmnt; } @@ -235,17 +235,16 @@ static void *afs_mntpt_follow_link(struct dentry *dentry, struct nameidata *nd) err = do_add_mount(newmnt, nd, MNT_SHRINKABLE, &afs_vfsmounts); switch (err) { case 0: - dput(nd->dentry); - mntput(nd->mnt); - nd->mnt = newmnt; - nd->dentry = dget(newmnt->mnt_root); + path_put(&nd->path); + nd->path.mnt = newmnt; + nd->path.dentry = dget(newmnt->mnt_root); schedule_delayed_work(&afs_mntpt_expiry_timer, afs_mntpt_expiry_timeout * HZ); break; case -EBUSY: /* someone else made a mount here whilst we were busy */ - while (d_mountpoint(nd->dentry) && - follow_down(&nd->mnt, &nd->dentry)) + while (d_mountpoint(nd->path.dentry) && + follow_down(&nd->path.mnt, &nd->path.dentry)) ; err = 0; default: @@ -284,11 +283,3 @@ void afs_mntpt_kill_timer(void) cancel_delayed_work(&afs_mntpt_expiry_timer); flush_scheduled_work(); } - -/* - * begin unmount by attempting to remove all automounted mountpoints we added - */ -void afs_umount_begin(struct vfsmount *vfsmnt, int flags) -{ - shrink_submounts(vfsmnt, &afs_vfsmounts); -} diff --git a/fs/afs/security.c b/fs/afs/security.c index 566fe712c68..3bcbeceba1b 100644 --- a/fs/afs/security.c +++ b/fs/afs/security.c @@ -95,7 +95,7 @@ static struct afs_vnode *afs_get_auth_inode(struct afs_vnode *vnode, auth_inode = afs_iget(vnode->vfs_inode.i_sb, key, &vnode->status.parent, NULL, NULL); if (IS_ERR(auth_inode)) - return ERR_PTR(PTR_ERR(auth_inode)); + return ERR_CAST(auth_inode); } auth_vnode = AFS_FS_I(auth_inode); @@ -287,7 +287,7 @@ static int afs_check_permit(struct afs_vnode *vnode, struct key *key, int afs_permission(struct inode *inode, int mask, struct nameidata *nd) { struct afs_vnode *vnode = AFS_FS_I(inode); - afs_access_t access; + afs_access_t uninitialized_var(access); struct key *key; int ret; diff --git a/fs/afs/super.c b/fs/afs/super.c index 4b2558c4221..4b572b801d8 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -50,8 +50,8 @@ static const struct super_operations afs_super_ops = { .write_inode = afs_write_inode, .destroy_inode = afs_destroy_inode, .clear_inode = afs_clear_inode, - .umount_begin = afs_umount_begin, .put_super = afs_put_super, + .show_options = generic_show_options, }; static struct kmem_cache *afs_inode_cachep; @@ -357,6 +357,7 @@ static int afs_get_sb(struct file_system_type *fs_type, struct super_block *sb; struct afs_volume *vol; struct key *key; + char *new_opts = kstrdup(options, GFP_KERNEL); int ret; _enter(",,%s,%p", dev_name, options); @@ -408,9 +409,11 @@ static int afs_get_sb(struct file_system_type *fs_type, deactivate_super(sb); goto error; } + sb->s_options = new_opts; sb->s_flags |= MS_ACTIVE; } else { _debug("reuse"); + kfree(new_opts); ASSERTCMP(sb->s_flags, &, MS_ACTIVE); } @@ -424,6 +427,7 @@ error: afs_put_volume(params.volume); afs_put_cell(params.cell); key_put(params.key); + kfree(new_opts); _leave(" = %d", ret); return ret; } @@ -317,7 +317,7 @@ out: /* wait_on_sync_kiocb: * Waits on the given sync kiocb to complete. */ -ssize_t fastcall wait_on_sync_kiocb(struct kiocb *iocb) +ssize_t wait_on_sync_kiocb(struct kiocb *iocb) { while (iocb->ki_users) { set_current_state(TASK_UNINTERRUPTIBLE); @@ -336,7 +336,7 @@ ssize_t fastcall wait_on_sync_kiocb(struct kiocb *iocb) * go away, they will call put_ioctx and release any pinned memory * associated with the request (held via struct page * references). */ -void fastcall exit_aio(struct mm_struct *mm) +void exit_aio(struct mm_struct *mm) { struct kioctx *ctx = mm->ioctx_list; mm->ioctx_list = NULL; @@ -365,7 +365,7 @@ void fastcall exit_aio(struct mm_struct *mm) * Called when the last user of an aio context has gone away, * and the struct needs to be freed. */ -void fastcall __put_ioctx(struct kioctx *ctx) +void __put_ioctx(struct kioctx *ctx) { unsigned nr_events = ctx->max_reqs; @@ -397,8 +397,7 @@ void fastcall __put_ioctx(struct kioctx *ctx) * This prevents races between the aio code path referencing the * req (after submitting it) and aio_complete() freeing the req. */ -static struct kiocb *FASTCALL(__aio_get_req(struct kioctx *ctx)); -static struct kiocb fastcall *__aio_get_req(struct kioctx *ctx) +static struct kiocb *__aio_get_req(struct kioctx *ctx) { struct kiocb *req = NULL; struct aio_ring *ring; @@ -533,7 +532,7 @@ static int __aio_put_req(struct kioctx *ctx, struct kiocb *req) * Returns true if this put was the last user of the kiocb, * false if the request is still in use. */ -int fastcall aio_put_req(struct kiocb *req) +int aio_put_req(struct kiocb *req) { struct kioctx *ctx = req->ki_ctx; int ret; @@ -893,7 +892,7 @@ static void try_queue_kicked_iocb(struct kiocb *iocb) * The retry is usually executed by aio workqueue * threads (See aio_kick_handler). */ -void fastcall kick_iocb(struct kiocb *iocb) +void kick_iocb(struct kiocb *iocb) { /* sync iocbs are easy: they can only ever be executing from a * single context. */ @@ -912,7 +911,7 @@ EXPORT_SYMBOL(kick_iocb); * Returns true if this is the last user of the request. The * only other user of the request can be the cancellation code. */ -int fastcall aio_complete(struct kiocb *iocb, long res, long res2) +int aio_complete(struct kiocb *iocb, long res, long res2) { struct kioctx *ctx = iocb->ki_ctx; struct aio_ring_info *info; @@ -937,14 +936,6 @@ int fastcall aio_complete(struct kiocb *iocb, long res, long res2) return 1; } - /* - * Check if the user asked us to deliver the result through an - * eventfd. The eventfd_signal() function is safe to be called - * from IRQ context. - */ - if (!IS_ERR(iocb->ki_eventfd)) - eventfd_signal(iocb->ki_eventfd, 1); - info = &ctx->ring_info; /* add a completion event to the ring buffer. @@ -993,10 +984,27 @@ int fastcall aio_complete(struct kiocb *iocb, long res, long res2) kunmap_atomic(ring, KM_IRQ1); pr_debug("added to ring %p at [%lu]\n", iocb, tail); + + /* + * Check if the user asked us to deliver the result through an + * eventfd. The eventfd_signal() function is safe to be called + * from IRQ context. + */ + if (!IS_ERR(iocb->ki_eventfd)) + eventfd_signal(iocb->ki_eventfd, 1); + put_rq: /* everything turned out well, dispose of the aiocb. */ ret = __aio_put_req(ctx, iocb); + /* + * We have to order our ring_info tail store above and test + * of the wait list below outside the wait lock. This is + * like in wake_up_bit() where clearing a bit has to be + * ordered with the unlocked test. + */ + smp_mb(); + if (waitqueue_active(&ctx->wait)) wake_up(&ctx->wait); @@ -1330,6 +1338,10 @@ static ssize_t aio_rw_vect_retry(struct kiocb *iocb) opcode = IOCB_CMD_PWRITEV; } + /* This matches the pread()/pwrite() logic */ + if (iocb->ki_pos < 0) + return -EINVAL; + do { ret = rw_op(iocb, &iocb->ki_iovec[iocb->ki_cur_seg], iocb->ki_nr_segs - iocb->ki_cur_seg, @@ -1348,6 +1360,13 @@ static ssize_t aio_rw_vect_retry(struct kiocb *iocb) if ((ret == 0) || (iocb->ki_left == 0)) ret = iocb->ki_nbytes - iocb->ki_left; + /* If we managed to write some out we return that, rather than + * the eventual error. */ + if (opcode == IOCB_CMD_PWRITEV + && ret < 0 && ret != -EIOCBQUEUED && ret != -EIOCBRETRY + && iocb->ki_nbytes - iocb->ki_left) + ret = iocb->ki_nbytes - iocb->ki_left; + return ret; } @@ -1523,7 +1542,7 @@ static int aio_wake_function(wait_queue_t *wait, unsigned mode, return 1; } -int fastcall io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, +int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, struct iocb *iocb) { struct kiocb *req; @@ -1772,6 +1791,7 @@ asmlinkage long sys_io_getevents(aio_context_t ctx_id, put_ioctx(ioctx); } + asmlinkage_protect(5, ret, ctx_id, min_nr, nr, events, timeout); return ret; } diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c index 23321889d9b..f42be069e08 100644 --- a/fs/anon_inodes.c +++ b/fs/anon_inodes.c @@ -81,13 +81,10 @@ int anon_inode_getfd(int *pfd, struct inode **pinode, struct file **pfile, if (IS_ERR(anon_inode_inode)) return -ENODEV; - file = get_empty_filp(); - if (!file) - return -ENFILE; error = get_unused_fd(); if (error < 0) - goto err_put_filp; + return error; fd = error; /* @@ -114,14 +111,15 @@ int anon_inode_getfd(int *pfd, struct inode **pinode, struct file **pfile, dentry->d_flags &= ~DCACHE_UNHASHED; d_instantiate(dentry, anon_inode_inode); - file->f_path.mnt = mntget(anon_inode_mnt); - file->f_path.dentry = dentry; + error = -ENFILE; + file = alloc_file(anon_inode_mnt, dentry, + FMODE_READ | FMODE_WRITE, fops); + if (!file) + goto err_dput; file->f_mapping = anon_inode_inode->i_mapping; file->f_pos = 0; file->f_flags = O_RDWR; - file->f_op = fops; - file->f_mode = FMODE_READ | FMODE_WRITE; file->f_version = 0; file->private_data = priv; @@ -132,10 +130,10 @@ int anon_inode_getfd(int *pfd, struct inode **pinode, struct file **pfile, *pfile = file; return 0; +err_dput: + dput(dentry); err_put_unused_fd: put_unused_fd(fd); -err_put_filp: - put_filp(file); return error; } EXPORT_SYMBOL_GPL(anon_inode_getfd); diff --git a/fs/autofs/autofs_i.h b/fs/autofs/autofs_i.h index 8b4cca3c470..901a3e67ec4 100644 --- a/fs/autofs/autofs_i.h +++ b/fs/autofs/autofs_i.h @@ -150,6 +150,7 @@ extern const struct file_operations autofs_root_operations; int autofs_fill_super(struct super_block *, void *, int); void autofs_kill_sb(struct super_block *sb); +struct inode *autofs_iget(struct super_block *, unsigned long); /* Queue management functions */ diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c index 45f5992a095..dda510d31f8 100644 --- a/fs/autofs/inode.c +++ b/fs/autofs/inode.c @@ -52,11 +52,9 @@ out_kill_sb: kill_anon_super(sb); } -static void autofs_read_inode(struct inode *inode); - static const struct super_operations autofs_sops = { - .read_inode = autofs_read_inode, .statfs = simple_statfs, + .show_options = generic_show_options, }; enum {Opt_err, Opt_fd, Opt_uid, Opt_gid, Opt_pgrp, Opt_minproto, Opt_maxproto}; @@ -143,6 +141,8 @@ int autofs_fill_super(struct super_block *s, void *data, int silent) int minproto, maxproto; pid_t pgid; + save_mount_options(s, data); + sbi = kzalloc(sizeof(*sbi), GFP_KERNEL); if (!sbi) goto fail_unlock; @@ -164,7 +164,9 @@ int autofs_fill_super(struct super_block *s, void *data, int silent) s->s_time_gran = 1; sbi->sb = s; - root_inode = iget(s, AUTOFS_ROOT_INO); + root_inode = autofs_iget(s, AUTOFS_ROOT_INO); + if (IS_ERR(root_inode)) + goto fail_free; root = d_alloc_root(root_inode); pipe = NULL; @@ -230,11 +232,17 @@ fail_unlock: return -EINVAL; } -static void autofs_read_inode(struct inode *inode) +struct inode *autofs_iget(struct super_block *sb, unsigned long ino) { - ino_t ino = inode->i_ino; unsigned int n; - struct autofs_sb_info *sbi = autofs_sbi(inode->i_sb); + struct autofs_sb_info *sbi = autofs_sbi(sb); + struct inode *inode; + + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; /* Initialize to the default case (stub directory) */ @@ -250,7 +258,7 @@ static void autofs_read_inode(struct inode *inode) inode->i_op = &autofs_root_inode_operations; inode->i_fop = &autofs_root_operations; inode->i_uid = inode->i_gid = 0; /* Changed in read_super */ - return; + goto done; } inode->i_uid = inode->i_sb->s_root->d_inode->i_uid; @@ -263,7 +271,7 @@ static void autofs_read_inode(struct inode *inode) n = ino - AUTOFS_FIRST_SYMLINK; if (n >= AUTOFS_MAX_SYMLINKS || !test_bit(n,sbi->symlink_bitmap)) { printk("autofs: Looking for bad symlink inode %u\n", (unsigned int) ino); - return; + goto done; } inode->i_op = &autofs_symlink_inode_operations; @@ -275,4 +283,8 @@ static void autofs_read_inode(struct inode *inode) inode->i_size = sl->len; inode->i_nlink = 1; } + +done: + unlock_new_inode(inode); + return inode; } diff --git a/fs/autofs/root.c b/fs/autofs/root.c index 5efff3c0d88..8aacade5695 100644 --- a/fs/autofs/root.c +++ b/fs/autofs/root.c @@ -114,8 +114,8 @@ static int try_to_fill_dentry(struct dentry *dentry, struct super_block *sb, str dentry->d_time = (unsigned long) ent; if (!dentry->d_inode) { - inode = iget(sb, ent->ino); - if (!inode) { + inode = autofs_iget(sb, ent->ino); + if (IS_ERR(inode)) { /* Failed, but leave pending for next time */ return 1; } @@ -274,6 +274,7 @@ static int autofs_root_symlink(struct inode *dir, struct dentry *dentry, const c unsigned int n; int slsize; struct autofs_symlink *sl; + struct inode *inode; DPRINTK(("autofs_root_symlink: %s <- ", symname)); autofs_say(dentry->d_name.name,dentry->d_name.len); @@ -331,7 +332,12 @@ static int autofs_root_symlink(struct inode *dir, struct dentry *dentry, const c ent->dentry = NULL; /* We don't keep the dentry for symlinks */ autofs_hash_insert(dh,ent); - d_instantiate(dentry, iget(dir->i_sb,ent->ino)); + + inode = autofs_iget(dir->i_sb, ent->ino); + if (IS_ERR(inode)) + return PTR_ERR(inode); + + d_instantiate(dentry, inode); unlock_kernel(); return 0; } @@ -428,6 +434,7 @@ static int autofs_root_mkdir(struct inode *dir, struct dentry *dentry, int mode) struct autofs_sb_info *sbi = autofs_sbi(dir->i_sb); struct autofs_dirhash *dh = &sbi->dirhash; struct autofs_dir_ent *ent; + struct inode *inode; ino_t ino; lock_kernel(); @@ -469,7 +476,14 @@ static int autofs_root_mkdir(struct inode *dir, struct dentry *dentry, int mode) autofs_hash_insert(dh,ent); inc_nlink(dir); - d_instantiate(dentry, iget(dir->i_sb,ino)); + + inode = autofs_iget(dir->i_sb, ino); + if (IS_ERR(inode)) { + drop_nlink(dir); + return PTR_ERR(inode); + } + + d_instantiate(dentry, inode); unlock_kernel(); return 0; diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index 7f05d6ccdb1..2fdcf5e1d23 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -176,11 +176,16 @@ out_kill_sb: static int autofs4_show_options(struct seq_file *m, struct vfsmount *mnt) { struct autofs_sb_info *sbi = autofs4_sbi(mnt->mnt_sb); + struct inode *root_inode = mnt->mnt_sb->s_root->d_inode; if (!sbi) return 0; seq_printf(m, ",fd=%d", sbi->pipefd); + if (root_inode->i_uid != 0) + seq_printf(m, ",uid=%u", root_inode->i_uid); + if (root_inode->i_gid != 0) + seq_printf(m, ",gid=%u", root_inode->i_gid); seq_printf(m, ",pgrp=%d", sbi->oz_pgrp); seq_printf(m, ",timeout=%lu", sbi->exp_timeout/HZ); seq_printf(m, ",minproto=%d", sbi->min_proto); diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 2bbcc8151dc..a54a946a50a 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -368,7 +368,8 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) * so we don't need to follow the mount. */ if (d_mountpoint(dentry)) { - if (!autofs4_follow_mount(&nd->mnt, &nd->dentry)) { + if (!autofs4_follow_mount(&nd->path.mnt, + &nd->path.dentry)) { status = -ENOENT; goto out_error; } @@ -382,7 +383,7 @@ done: return NULL; out_error: - path_release(nd); + path_put(&nd->path); return ERR_PTR(status); } diff --git a/fs/bad_inode.c b/fs/bad_inode.c index 521ff7caadb..f1c2ea8342f 100644 --- a/fs/bad_inode.c +++ b/fs/bad_inode.c @@ -359,3 +359,17 @@ int is_bad_inode(struct inode *inode) } EXPORT_SYMBOL(is_bad_inode); + +/** + * iget_failed - Mark an under-construction inode as dead and release it + * @inode: The inode to discard + * + * Mark an under-construction inode as dead and release it. + */ +void iget_failed(struct inode *inode) +{ + make_bad_inode(inode); + unlock_new_inode(inode); + iput(inode); +} +EXPORT_SYMBOL(iget_failed); diff --git a/fs/befs/btree.c b/fs/befs/btree.c index af5bb93276f..4202db7496c 100644 --- a/fs/befs/btree.c +++ b/fs/befs/btree.c @@ -232,7 +232,7 @@ befs_bt_read_node(struct super_block *sb, befs_data_stream * ds, * @key: Key string to lookup in btree * @value: Value stored with @key * - * On sucess, returns BEFS_OK and sets *@value to the value stored + * On success, returns BEFS_OK and sets *@value to the value stored * with @key (usually the disk block number of an inode). * * On failure, returns BEFS_ERR or BEFS_BT_NOT_FOUND. diff --git a/fs/befs/datastream.c b/fs/befs/datastream.c index aacb4da6298..e3287d0d1a5 100644 --- a/fs/befs/datastream.c +++ b/fs/befs/datastream.c @@ -236,7 +236,7 @@ befs_count_blocks(struct super_block * sb, befs_data_stream * ds) as in the indirect region code). When/if blockno is found, if blockno is inside of a block - run as stored on disk, we offset the start and lenght members + run as stored on disk, we offset the start and length members of the block run, so that blockno is the start and len is still valid (the run ends in the same place). diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index b28a20e61b8..82123ff3e1d 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -35,7 +35,7 @@ static int befs_get_block(struct inode *, sector_t, struct buffer_head *, int); static int befs_readpage(struct file *file, struct page *page); static sector_t befs_bmap(struct address_space *mapping, sector_t block); static struct dentry *befs_lookup(struct inode *, struct dentry *, struct nameidata *); -static void befs_read_inode(struct inode *ino); +static struct inode *befs_iget(struct super_block *, unsigned long); static struct inode *befs_alloc_inode(struct super_block *sb); static void befs_destroy_inode(struct inode *inode); static int befs_init_inodecache(void); @@ -52,12 +52,12 @@ static int befs_statfs(struct dentry *, struct kstatfs *); static int parse_options(char *, befs_mount_options *); static const struct super_operations befs_sops = { - .read_inode = befs_read_inode, /* initialize & read inode */ .alloc_inode = befs_alloc_inode, /* allocate a new inode */ .destroy_inode = befs_destroy_inode, /* deallocate an inode */ .put_super = befs_put_super, /* uninit super */ .statfs = befs_statfs, /* statfs */ .remount_fs = befs_remount, + .show_options = generic_show_options, }; /* slab cache for befs_inode_info objects */ @@ -198,9 +198,9 @@ befs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) return ERR_PTR(-ENODATA); } - inode = iget(dir->i_sb, (ino_t) offset); - if (!inode) - return ERR_PTR(-EACCES); + inode = befs_iget(dir->i_sb, (ino_t) offset); + if (IS_ERR(inode)) + return ERR_CAST(inode); d_add(dentry, inode); @@ -296,17 +296,23 @@ static void init_once(struct kmem_cache *cachep, void *foo) inode_init_once(&bi->vfs_inode); } -static void -befs_read_inode(struct inode *inode) +static struct inode *befs_iget(struct super_block *sb, unsigned long ino) { struct buffer_head *bh = NULL; befs_inode *raw_inode = NULL; - struct super_block *sb = inode->i_sb; befs_sb_info *befs_sb = BEFS_SB(sb); befs_inode_info *befs_ino = NULL; + struct inode *inode; + long ret = -EIO; + + befs_debug(sb, "---> befs_read_inode() " "inode = %lu", ino); - befs_debug(sb, "---> befs_read_inode() " "inode = %lu", inode->i_ino); + inode = iget_locked(sb, ino); + if (IS_ERR(inode)) + return inode; + if (!(inode->i_state & I_NEW)) + return inode; befs_ino = BEFS_I(inode); @@ -402,15 +408,16 @@ befs_read_inode(struct inode *inode) brelse(bh); befs_debug(sb, "<--- befs_read_inode()"); - return; + unlock_new_inode(inode); + return inode; unacquire_bh: brelse(bh); unacquire_none: - make_bad_inode(inode); + iget_failed(inode); befs_debug(sb, "<--- befs_read_inode() - Bad inode"); - return; + return ERR_PTR(ret); } /* Initialize the inode cache. Called at fs setup. @@ -752,10 +759,12 @@ befs_fill_super(struct super_block *sb, void *data, int silent) befs_sb_info *befs_sb; befs_super_block *disk_sb; struct inode *root; - + long ret = -EINVAL; const unsigned long sb_block = 0; const off_t x86_sb_off = 512; + save_mount_options(sb, data); + sb->s_fs_info = kmalloc(sizeof (*befs_sb), GFP_KERNEL); if (sb->s_fs_info == NULL) { printk(KERN_ERR @@ -833,7 +842,11 @@ befs_fill_super(struct super_block *sb, void *data, int silent) /* Set real blocksize of fs */ sb_set_blocksize(sb, (ulong) befs_sb->block_size); sb->s_op = (struct super_operations *) &befs_sops; - root = iget(sb, iaddr2blockno(sb, &(befs_sb->root_dir))); + root = befs_iget(sb, iaddr2blockno(sb, &(befs_sb->root_dir))); + if (IS_ERR(root)) { + ret = PTR_ERR(root); + goto unacquire_priv_sbp; + } sb->s_root = d_alloc_root(root); if (!sb->s_root) { iput(root); @@ -868,7 +881,7 @@ befs_fill_super(struct super_block *sb, void *data, int silent) unacquire_none: sb->s_fs_info = NULL; - return -EINVAL; + return ret; } static int diff --git a/fs/bfs/bfs.h b/fs/bfs/bfs.h index ac7a8b1d6c3..71faf4d2390 100644 --- a/fs/bfs/bfs.h +++ b/fs/bfs/bfs.h @@ -44,6 +44,8 @@ static inline struct bfs_inode_info *BFS_I(struct inode *inode) #define printf(format, args...) \ printk(KERN_ERR "BFS-fs: %s(): " format, __FUNCTION__, ## args) +/* inode.c */ +extern struct inode *bfs_iget(struct super_block *sb, unsigned long ino); /* file.c */ extern const struct inode_operations bfs_file_inops; diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c index 1fd056d0fc3..034950cb3cb 100644 --- a/fs/bfs/dir.c +++ b/fs/bfs/dir.c @@ -148,10 +148,10 @@ static struct dentry *bfs_lookup(struct inode *dir, struct dentry *dentry, if (bh) { unsigned long ino = (unsigned long)le16_to_cpu(de->ino); brelse(bh); - inode = iget(dir->i_sb, ino); - if (!inode) { + inode = bfs_iget(dir->i_sb, ino); + if (IS_ERR(inode)) { unlock_kernel(); - return ERR_PTR(-EACCES); + return ERR_CAST(inode); } } unlock_kernel(); diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index a64a71d444f..8db623838b5 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -32,17 +32,22 @@ MODULE_LICENSE("GPL"); void dump_imap(const char *prefix, struct super_block *s); -static void bfs_read_inode(struct inode *inode) +struct inode *bfs_iget(struct super_block *sb, unsigned long ino) { - unsigned long ino = inode->i_ino; struct bfs_inode *di; + struct inode *inode; struct buffer_head *bh; int block, off; + inode = iget_locked(sb, ino); + if (IS_ERR(inode)) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; + if ((ino < BFS_ROOT_INO) || (ino > BFS_SB(inode->i_sb)->si_lasti)) { printf("Bad inode number %s:%08lx\n", inode->i_sb->s_id, ino); - make_bad_inode(inode); - return; + goto error; } block = (ino - BFS_ROOT_INO) / BFS_INODES_PER_BLOCK + 1; @@ -50,8 +55,7 @@ static void bfs_read_inode(struct inode *inode) if (!bh) { printf("Unable to read inode %s:%08lx\n", inode->i_sb->s_id, ino); - make_bad_inode(inode); - return; + goto error; } off = (ino - BFS_ROOT_INO) % BFS_INODES_PER_BLOCK; @@ -85,6 +89,12 @@ static void bfs_read_inode(struct inode *inode) inode->i_ctime.tv_nsec = 0; brelse(bh); + unlock_new_inode(inode); + return inode; + +error: + iget_failed(inode); + return ERR_PTR(-EIO); } static int bfs_write_inode(struct inode *inode, int unused) @@ -276,7 +286,6 @@ static void destroy_inodecache(void) static const struct super_operations bfs_sops = { .alloc_inode = bfs_alloc_inode, .destroy_inode = bfs_destroy_inode, - .read_inode = bfs_read_inode, .write_inode = bfs_write_inode, .delete_inode = bfs_delete_inode, .put_super = bfs_put_super, @@ -312,6 +321,7 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent) struct inode *inode; unsigned i, imap_len; struct bfs_sb_info *info; + long ret = -EINVAL; info = kzalloc(sizeof(*info), GFP_KERNEL); if (!info) @@ -346,14 +356,16 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent) set_bit(i, info->si_imap); s->s_op = &bfs_sops; - inode = iget(s, BFS_ROOT_INO); - if (!inode) { + inode = bfs_iget(s, BFS_ROOT_INO); + if (IS_ERR(inode)) { + ret = PTR_ERR(inode); kfree(info->si_imap); goto out; } s->s_root = d_alloc_root(inode); if (!s->s_root) { iput(inode); + ret = -ENOMEM; kfree(info->si_imap); goto out; } @@ -404,7 +416,7 @@ out: brelse(bh); kfree(info); s->s_fs_info = NULL; - return -EINVAL; + return ret; } static int bfs_get_sb(struct file_system_type *fs_type, diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c index 7596e1e94cd..a1bb2244cac 100644 --- a/fs/binfmt_aout.c +++ b/fs/binfmt_aout.c @@ -28,6 +28,7 @@ #include <asm/system.h> #include <asm/uaccess.h> #include <asm/cacheflush.h> +#include <asm/a.out-core.h> static int load_aout_binary(struct linux_binprm *, struct pt_regs * regs); static int load_aout_library(struct file*); @@ -115,10 +116,10 @@ static int aout_core_dump(long signr, struct pt_regs *regs, struct file *file, u current->flags |= PF_DUMPCORE; strncpy(dump.u_comm, current->comm, sizeof(dump.u_comm)); #ifndef __sparc__ - dump.u_ar0 = (void *)(((unsigned long)(&dump.regs)) - ((unsigned long)(&dump))); + dump.u_ar0 = offsetof(struct user, regs); #endif dump.signal = signr; - dump_thread(regs, &dump); + aout_dump_thread(regs, &dump); /* If the size of the dump file exceeds the rlimit, then see what would happen if we wrote the stack, but not the data area. */ diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index f0b3171842f..5e1a4fb5cac 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -45,7 +45,8 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs); static int load_elf_library(struct file *); -static unsigned long elf_map (struct file *, unsigned long, struct elf_phdr *, int, int); +static unsigned long elf_map(struct file *, unsigned long, struct elf_phdr *, + int, int, unsigned long); /* * If we don't support core dumping, then supply a NULL so we @@ -116,7 +117,7 @@ static int padzero(unsigned long elf_bss) return 0; } -/* Let's use some macros to make this stack manipulation a litle clearer */ +/* Let's use some macros to make this stack manipulation a little clearer */ #ifdef CONFIG_STACK_GROWSUP #define STACK_ADD(sp, items) ((elf_addr_t __user *)(sp) + (items)) #define STACK_ROUND(sp, items) \ @@ -133,8 +134,7 @@ static int padzero(unsigned long elf_bss) static int create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec, - int interp_aout, unsigned long load_addr, - unsigned long interp_load_addr) + unsigned long load_addr, unsigned long interp_load_addr) { unsigned long p = bprm->p; int argc = bprm->argc; @@ -222,12 +222,7 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec, sp = STACK_ADD(p, ei_index); - items = (argc + 1) + (envc + 1); - if (interp_aout) { - items += 3; /* a.out interpreters require argv & envp too */ - } else { - items += 1; /* ELF interpreters only put argc on the stack */ - } + items = (argc + 1) + (envc + 1) + 1; bprm->p = STACK_ROUND(sp, items); /* Point sp at the lowest address on the stack */ @@ -250,16 +245,8 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec, /* Now, let's put argc (and argv, envp if appropriate) on the stack */ if (__put_user(argc, sp++)) return -EFAULT; - if (interp_aout) { - argv = sp + 2; - envp = argv + argc + 1; - if (__put_user((elf_addr_t)(unsigned long)argv, sp++) || - __put_user((elf_addr_t)(unsigned long)envp, sp++)) - return -EFAULT; - } else { - argv = sp; - envp = argv + argc + 1; - } + argv = sp; + envp = argv + argc + 1; /* Populate argv and envp */ p = current->mm->arg_end = current->mm->arg_start; @@ -298,33 +285,70 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec, #ifndef elf_map static unsigned long elf_map(struct file *filep, unsigned long addr, - struct elf_phdr *eppnt, int prot, int type) + struct elf_phdr *eppnt, int prot, int type, + unsigned long total_size) { unsigned long map_addr; - unsigned long pageoffset = ELF_PAGEOFFSET(eppnt->p_vaddr); + unsigned long size = eppnt->p_filesz + ELF_PAGEOFFSET(eppnt->p_vaddr); + unsigned long off = eppnt->p_offset - ELF_PAGEOFFSET(eppnt->p_vaddr); + addr = ELF_PAGESTART(addr); + size = ELF_PAGEALIGN(size); - down_write(¤t->mm->mmap_sem); /* mmap() will return -EINVAL if given a zero size, but a * segment with zero filesize is perfectly valid */ - if (eppnt->p_filesz + pageoffset) - map_addr = do_mmap(filep, ELF_PAGESTART(addr), - eppnt->p_filesz + pageoffset, prot, type, - eppnt->p_offset - pageoffset); - else - map_addr = ELF_PAGESTART(addr); + if (!size) + return addr; + + down_write(¤t->mm->mmap_sem); + /* + * total_size is the size of the ELF (interpreter) image. + * The _first_ mmap needs to know the full size, otherwise + * randomization might put this image into an overlapping + * position with the ELF binary image. (since size < total_size) + * So we first map the 'big' image - and unmap the remainder at + * the end. (which unmap is needed for ELF images with holes.) + */ + if (total_size) { + total_size = ELF_PAGEALIGN(total_size); + map_addr = do_mmap(filep, addr, total_size, prot, type, off); + if (!BAD_ADDR(map_addr)) + do_munmap(current->mm, map_addr+size, total_size-size); + } else + map_addr = do_mmap(filep, addr, size, prot, type, off); + up_write(¤t->mm->mmap_sem); return(map_addr); } #endif /* !elf_map */ +static unsigned long total_mapping_size(struct elf_phdr *cmds, int nr) +{ + int i, first_idx = -1, last_idx = -1; + + for (i = 0; i < nr; i++) { + if (cmds[i].p_type == PT_LOAD) { + last_idx = i; + if (first_idx == -1) + first_idx = i; + } + } + if (first_idx == -1) + return 0; + + return cmds[last_idx].p_vaddr + cmds[last_idx].p_memsz - + ELF_PAGESTART(cmds[first_idx].p_vaddr); +} + + /* This is much more generalized than the library routine read function, so we keep this separate. Technically the library read function is only provided so that we can read a.out libraries that have an ELF header */ static unsigned long load_elf_interp(struct elfhdr *interp_elf_ex, - struct file *interpreter, unsigned long *interp_load_addr) + struct file *interpreter, unsigned long *interp_map_addr, + unsigned long no_base) { struct elf_phdr *elf_phdata; struct elf_phdr *eppnt; @@ -332,6 +356,7 @@ static unsigned long load_elf_interp(struct elfhdr *interp_elf_ex, int load_addr_set = 0; unsigned long last_bss = 0, elf_bss = 0; unsigned long error = ~0UL; + unsigned long total_size; int retval, i, size; /* First of all, some simple consistency checks */ @@ -370,6 +395,12 @@ static unsigned long load_elf_interp(struct elfhdr *interp_elf_ex, goto out_close; } + total_size = total_mapping_size(elf_phdata, interp_elf_ex->e_phnum); + if (!total_size) { + error = -EINVAL; + goto out_close; + } + eppnt = elf_phdata; for (i = 0; i < interp_elf_ex->e_phnum; i++, eppnt++) { if (eppnt->p_type == PT_LOAD) { @@ -387,9 +418,14 @@ static unsigned long load_elf_interp(struct elfhdr *interp_elf_ex, vaddr = eppnt->p_vaddr; if (interp_elf_ex->e_type == ET_EXEC || load_addr_set) elf_type |= MAP_FIXED; + else if (no_base && interp_elf_ex->e_type == ET_DYN) + load_addr = -vaddr; map_addr = elf_map(interpreter, load_addr + vaddr, - eppnt, elf_prot, elf_type); + eppnt, elf_prot, elf_type, total_size); + total_size = 0; + if (!*interp_map_addr) + *interp_map_addr = map_addr; error = map_addr; if (BAD_ADDR(map_addr)) goto out_close; @@ -455,8 +491,7 @@ static unsigned long load_elf_interp(struct elfhdr *interp_elf_ex, goto out_close; } - *interp_load_addr = load_addr; - error = ((unsigned long)interp_elf_ex->e_entry) + load_addr; + error = load_addr; out_close: kfree(elf_phdata); @@ -464,59 +499,12 @@ out: return error; } -static unsigned long load_aout_interp(struct exec *interp_ex, - struct file *interpreter) -{ - unsigned long text_data, elf_entry = ~0UL; - char __user * addr; - loff_t offset; - - current->mm->end_code = interp_ex->a_text; - text_data = interp_ex->a_text + interp_ex->a_data; - current->mm->end_data = text_data; - current->mm->brk = interp_ex->a_bss + text_data; - - switch (N_MAGIC(*interp_ex)) { - case OMAGIC: - offset = 32; - addr = (char __user *)0; - break; - case ZMAGIC: - case QMAGIC: - offset = N_TXTOFF(*interp_ex); - addr = (char __user *)N_TXTADDR(*interp_ex); - break; - default: - goto out; - } - - down_write(¤t->mm->mmap_sem); - do_brk(0, text_data); - up_write(¤t->mm->mmap_sem); - if (!interpreter->f_op || !interpreter->f_op->read) - goto out; - if (interpreter->f_op->read(interpreter, addr, text_data, &offset) < 0) - goto out; - flush_icache_range((unsigned long)addr, - (unsigned long)addr + text_data); - - down_write(¤t->mm->mmap_sem); - do_brk(ELF_PAGESTART(text_data + ELF_MIN_ALIGN - 1), - interp_ex->a_bss); - up_write(¤t->mm->mmap_sem); - elf_entry = interp_ex->a_entry; - -out: - return elf_entry; -} - /* * These are the functions used to load ELF style executables and shared * libraries. There is no binary dependent code anywhere else. */ #define INTERPRETER_NONE 0 -#define INTERPRETER_AOUT 1 #define INTERPRETER_ELF 2 #ifndef STACK_RND_MASK @@ -545,18 +533,16 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) unsigned long load_addr = 0, load_bias = 0; int load_addr_set = 0; char * elf_interpreter = NULL; - unsigned int interpreter_type = INTERPRETER_NONE; - unsigned char ibcs2_interpreter = 0; unsigned long error; struct elf_phdr *elf_ppnt, *elf_phdata; unsigned long elf_bss, elf_brk; int elf_exec_fileno; int retval, i; unsigned int size; - unsigned long elf_entry, interp_load_addr = 0; + unsigned long elf_entry; + unsigned long interp_load_addr = 0; unsigned long start_code, end_code, start_data, end_data; unsigned long reloc_func_desc = 0; - char passed_fileno[6]; struct files_struct *files; int executable_stack = EXSTACK_DEFAULT; unsigned long def_flags = 0; @@ -663,14 +649,6 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) if (elf_interpreter[elf_ppnt->p_filesz - 1] != '\0') goto out_free_interp; - /* If the program interpreter is one of these two, - * then assume an iBCS2 image. Otherwise assume - * a native linux image. - */ - if (strcmp(elf_interpreter,"/usr/lib/libc.so.1") == 0 || - strcmp(elf_interpreter,"/usr/lib/ld.so.1") == 0) - ibcs2_interpreter = 1; - /* * The early SET_PERSONALITY here is so that the lookup * for the interpreter happens in the namespace of the @@ -690,7 +668,7 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) * switch really is going to happen - do this in * flush_thread(). - akpm */ - SET_PERSONALITY(loc->elf_ex, ibcs2_interpreter); + SET_PERSONALITY(loc->elf_ex, 0); interpreter = open_exec(elf_interpreter); retval = PTR_ERR(interpreter); @@ -733,57 +711,16 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) /* Some simple consistency checks for the interpreter */ if (elf_interpreter) { - static int warn; - interpreter_type = INTERPRETER_ELF | INTERPRETER_AOUT; - - /* Now figure out which format our binary is */ - if ((N_MAGIC(loc->interp_ex) != OMAGIC) && - (N_MAGIC(loc->interp_ex) != ZMAGIC) && - (N_MAGIC(loc->interp_ex) != QMAGIC)) - interpreter_type = INTERPRETER_ELF; - - if (memcmp(loc->interp_elf_ex.e_ident, ELFMAG, SELFMAG) != 0) - interpreter_type &= ~INTERPRETER_ELF; - - if (interpreter_type == INTERPRETER_AOUT && warn < 10) { - printk(KERN_WARNING "a.out ELF interpreter %s is " - "deprecated and will not be supported " - "after Linux 2.6.25\n", elf_interpreter); - warn++; - } - retval = -ELIBBAD; - if (!interpreter_type) + /* Not an ELF interpreter */ + if (memcmp(loc->interp_elf_ex.e_ident, ELFMAG, SELFMAG) != 0) goto out_free_dentry; - - /* Make sure only one type was selected */ - if ((interpreter_type & INTERPRETER_ELF) && - interpreter_type != INTERPRETER_ELF) { - // FIXME - ratelimit this before re-enabling - // printk(KERN_WARNING "ELF: Ambiguous type, using ELF\n"); - interpreter_type = INTERPRETER_ELF; - } /* Verify the interpreter has a valid arch */ - if ((interpreter_type == INTERPRETER_ELF) && - !elf_check_arch(&loc->interp_elf_ex)) + if (!elf_check_arch(&loc->interp_elf_ex)) goto out_free_dentry; } else { /* Executables without an interpreter also need a personality */ - SET_PERSONALITY(loc->elf_ex, ibcs2_interpreter); - } - - /* OK, we are done with that, now set up the arg stuff, - and then start this sucker up */ - if ((!bprm->sh_bang) && (interpreter_type == INTERPRETER_AOUT)) { - char *passed_p = passed_fileno; - sprintf(passed_fileno, "%d", elf_exec_fileno); - - if (elf_interpreter) { - retval = copy_strings_kernel(1, &passed_p, bprm); - if (retval) - goto out_free_dentry; - bprm->argc++; - } + SET_PERSONALITY(loc->elf_ex, 0); } /* Flush all traces of the currently running executable */ @@ -803,7 +740,7 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) /* Do this immediately, since STACK_TOP as used in setup_arg_pages may depend on the personality. */ - SET_PERSONALITY(loc->elf_ex, ibcs2_interpreter); + SET_PERSONALITY(loc->elf_ex, 0); if (elf_read_implies_exec(loc->elf_ex, executable_stack)) current->personality |= READ_IMPLIES_EXEC; @@ -825,9 +762,7 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) current->mm->start_stack = bprm->p; /* Now we do a little grungy work by mmaping the ELF image into - the correct location in memory. At this point, we assume that - the image should be loaded at fixed address, not at a variable - address. */ + the correct location in memory. */ for(i = 0, elf_ppnt = elf_phdata; i < loc->elf_ex.e_phnum; i++, elf_ppnt++) { int elf_prot = 0, elf_flags; @@ -881,11 +816,15 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) * default mmap base, as well as whatever program they * might try to exec. This is because the brk will * follow the loader, and is not movable. */ +#ifdef CONFIG_X86 + load_bias = 0; +#else load_bias = ELF_PAGESTART(ELF_ET_DYN_BASE - vaddr); +#endif } error = elf_map(bprm->file, load_bias + vaddr, elf_ppnt, - elf_prot, elf_flags); + elf_prot, elf_flags, 0); if (BAD_ADDR(error)) { send_sig(SIGKILL, current, 0); retval = IS_ERR((void *)error) ? @@ -961,13 +900,20 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) } if (elf_interpreter) { - if (interpreter_type == INTERPRETER_AOUT) - elf_entry = load_aout_interp(&loc->interp_ex, - interpreter); - else - elf_entry = load_elf_interp(&loc->interp_elf_ex, - interpreter, - &interp_load_addr); + unsigned long uninitialized_var(interp_map_addr); + + elf_entry = load_elf_interp(&loc->interp_elf_ex, + interpreter, + &interp_map_addr, + load_bias); + if (!IS_ERR((void *)elf_entry)) { + /* + * load_elf_interp() returns relocation + * adjustment + */ + interp_load_addr = elf_entry; + elf_entry += loc->interp_elf_ex.e_entry; + } if (BAD_ADDR(elf_entry)) { force_sig(SIGSEGV, current); retval = IS_ERR((void *)elf_entry) ? @@ -990,8 +936,7 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) kfree(elf_phdata); - if (interpreter_type != INTERPRETER_AOUT) - sys_close(elf_exec_fileno); + sys_close(elf_exec_fileno); set_binfmt(&elf_format); @@ -1006,21 +951,24 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) compute_creds(bprm); current->flags &= ~PF_FORKNOEXEC; retval = create_elf_tables(bprm, &loc->elf_ex, - (interpreter_type == INTERPRETER_AOUT), load_addr, interp_load_addr); if (retval < 0) { send_sig(SIGKILL, current, 0); goto out; } /* N.B. passed_fileno might not be initialized? */ - if (interpreter_type == INTERPRETER_AOUT) - current->mm->arg_start += strlen(passed_fileno) + 1; current->mm->end_code = end_code; current->mm->start_code = start_code; current->mm->start_data = start_data; current->mm->end_data = end_data; current->mm->start_stack = bprm->p; +#ifdef arch_randomize_brk + if ((current->flags & PF_RANDOMIZE) && (randomize_va_space > 1)) + current->mm->brk = current->mm->start_brk = + arch_randomize_brk(current->mm); +#endif + if (current->personality & MMAP_PAGE_ZERO) { /* Why this, you ask??? Well SVr4 maps page 0 as read-only, and some applications "depend" upon this behavior. @@ -1325,7 +1273,8 @@ static int writenote(struct memelfnote *men, struct file *file, if (!dump_seek(file, (off))) \ goto end_coredump; -static void fill_elf_header(struct elfhdr *elf, int segs) +static void fill_elf_header(struct elfhdr *elf, int segs, + u16 machine, u32 flags, u8 osabi) { memcpy(elf->e_ident, ELFMAG, SELFMAG); elf->e_ident[EI_CLASS] = ELF_CLASS; @@ -1335,12 +1284,12 @@ static void fill_elf_header(struct elfhdr *elf, int segs) memset(elf->e_ident+EI_PAD, 0, EI_NIDENT-EI_PAD); elf->e_type = ET_CORE; - elf->e_machine = ELF_ARCH; + elf->e_machine = machine; elf->e_version = EV_CURRENT; elf->e_entry = 0; elf->e_phoff = sizeof(struct elfhdr); elf->e_shoff = 0; - elf->e_flags = ELF_CORE_EFLAGS; + elf->e_flags = flags; elf->e_ehsize = sizeof(struct elfhdr); elf->e_phentsize = sizeof(struct elf_phdr); elf->e_phnum = segs; @@ -1447,6 +1396,253 @@ static int fill_psinfo(struct elf_prpsinfo *psinfo, struct task_struct *p, return 0; } +static void fill_auxv_note(struct memelfnote *note, struct mm_struct *mm) +{ + elf_addr_t *auxv = (elf_addr_t *) mm->saved_auxv; + int i = 0; + do + i += 2; + while (auxv[i - 2] != AT_NULL); + fill_note(note, "CORE", NT_AUXV, i * sizeof(elf_addr_t), auxv); +} + +#ifdef CORE_DUMP_USE_REGSET +#include <linux/regset.h> + +struct elf_thread_core_info { + struct elf_thread_core_info *next; + struct task_struct *task; + struct elf_prstatus prstatus; + struct memelfnote notes[0]; +}; + +struct elf_note_info { + struct elf_thread_core_info *thread; + struct memelfnote psinfo; + struct memelfnote auxv; + size_t size; + int thread_notes; +}; + +/* + * When a regset has a writeback hook, we call it on each thread before + * dumping user memory. On register window machines, this makes sure the + * user memory backing the register data is up to date before we read it. + */ +static void do_thread_regset_writeback(struct task_struct *task, + const struct user_regset *regset) +{ + if (regset->writeback) + regset->writeback(task, regset, 1); +} + +static int fill_thread_core_info(struct elf_thread_core_info *t, + const struct user_regset_view *view, + long signr, size_t *total) +{ + unsigned int i; + + /* + * NT_PRSTATUS is the one special case, because the regset data + * goes into the pr_reg field inside the note contents, rather + * than being the whole note contents. We fill the reset in here. + * We assume that regset 0 is NT_PRSTATUS. + */ + fill_prstatus(&t->prstatus, t->task, signr); + (void) view->regsets[0].get(t->task, &view->regsets[0], + 0, sizeof(t->prstatus.pr_reg), + &t->prstatus.pr_reg, NULL); + + fill_note(&t->notes[0], "CORE", NT_PRSTATUS, + sizeof(t->prstatus), &t->prstatus); + *total += notesize(&t->notes[0]); + + do_thread_regset_writeback(t->task, &view->regsets[0]); + + /* + * Each other regset might generate a note too. For each regset + * that has no core_note_type or is inactive, we leave t->notes[i] + * all zero and we'll know to skip writing it later. + */ + for (i = 1; i < view->n; ++i) { + const struct user_regset *regset = &view->regsets[i]; + do_thread_regset_writeback(t->task, regset); + if (regset->core_note_type && + (!regset->active || regset->active(t->task, regset))) { + int ret; + size_t size = regset->n * regset->size; + void *data = kmalloc(size, GFP_KERNEL); + if (unlikely(!data)) + return 0; + ret = regset->get(t->task, regset, + 0, size, data, NULL); + if (unlikely(ret)) + kfree(data); + else { + if (regset->core_note_type != NT_PRFPREG) + fill_note(&t->notes[i], "LINUX", + regset->core_note_type, + size, data); + else { + t->prstatus.pr_fpvalid = 1; + fill_note(&t->notes[i], "CORE", + NT_PRFPREG, size, data); + } + *total += notesize(&t->notes[i]); + } + } + } + + return 1; +} + +static int fill_note_info(struct elfhdr *elf, int phdrs, + struct elf_note_info *info, + long signr, struct pt_regs *regs) +{ + struct task_struct *dump_task = current; + const struct user_regset_view *view = task_user_regset_view(dump_task); + struct elf_thread_core_info *t; + struct elf_prpsinfo *psinfo; + struct task_struct *g, *p; + unsigned int i; + + info->size = 0; + info->thread = NULL; + + psinfo = kmalloc(sizeof(*psinfo), GFP_KERNEL); + fill_note(&info->psinfo, "CORE", NT_PRPSINFO, sizeof(*psinfo), psinfo); + + if (psinfo == NULL) + return 0; + + /* + * Figure out how many notes we're going to need for each thread. + */ + info->thread_notes = 0; + for (i = 0; i < view->n; ++i) + if (view->regsets[i].core_note_type != 0) + ++info->thread_notes; + + /* + * Sanity check. We rely on regset 0 being in NT_PRSTATUS, + * since it is our one special case. + */ + if (unlikely(info->thread_notes == 0) || + unlikely(view->regsets[0].core_note_type != NT_PRSTATUS)) { + WARN_ON(1); + return 0; + } + + /* + * Initialize the ELF file header. + */ + fill_elf_header(elf, phdrs, + view->e_machine, view->e_flags, view->ei_osabi); + + /* + * Allocate a structure for each thread. + */ + rcu_read_lock(); + do_each_thread(g, p) + if (p->mm == dump_task->mm) { + t = kzalloc(offsetof(struct elf_thread_core_info, + notes[info->thread_notes]), + GFP_ATOMIC); + if (unlikely(!t)) { + rcu_read_unlock(); + return 0; + } + t->task = p; + if (p == dump_task || !info->thread) { + t->next = info->thread; + info->thread = t; + } else { + /* + * Make sure to keep the original task at + * the head of the list. + */ + t->next = info->thread->next; + info->thread->next = t; + } + } + while_each_thread(g, p); + rcu_read_unlock(); + + /* + * Now fill in each thread's information. + */ + for (t = info->thread; t != NULL; t = t->next) + if (!fill_thread_core_info(t, view, signr, &info->size)) + return 0; + + /* + * Fill in the two process-wide notes. + */ + fill_psinfo(psinfo, dump_task->group_leader, dump_task->mm); + info->size += notesize(&info->psinfo); + + fill_auxv_note(&info->auxv, current->mm); + info->size += notesize(&info->auxv); + + return 1; +} + +static size_t get_note_info_size(struct elf_note_info *info) +{ + return info->size; +} + +/* + * Write all the notes for each thread. When writing the first thread, the + * process-wide notes are interleaved after the first thread-specific note. + */ +static int write_note_info(struct elf_note_info *info, + struct file *file, loff_t *foffset) +{ + bool first = 1; + struct elf_thread_core_info *t = info->thread; + + do { + int i; + + if (!writenote(&t->notes[0], file, foffset)) + return 0; + + if (first && !writenote(&info->psinfo, file, foffset)) + return 0; + if (first && !writenote(&info->auxv, file, foffset)) + return 0; + + for (i = 1; i < info->thread_notes; ++i) + if (t->notes[i].data && + !writenote(&t->notes[i], file, foffset)) + return 0; + + first = 0; + t = t->next; + } while (t); + + return 1; +} + +static void free_note_info(struct elf_note_info *info) +{ + struct elf_thread_core_info *threads = info->thread; + while (threads) { + unsigned int i; + struct elf_thread_core_info *t = threads; + threads = t->next; + WARN_ON(t->notes[0].data && t->notes[0].data != &t->prstatus); + for (i = 1; i < info->thread_notes; ++i) + kfree(t->notes[i].data); + kfree(t); + } + kfree(info->psinfo.data); +} + +#else + /* Here is the structure in which status of each thread is captured. */ struct elf_thread_status { @@ -1499,6 +1695,176 @@ static int elf_dump_thread_status(long signr, struct elf_thread_status *t) return sz; } +struct elf_note_info { + struct memelfnote *notes; + struct elf_prstatus *prstatus; /* NT_PRSTATUS */ + struct elf_prpsinfo *psinfo; /* NT_PRPSINFO */ + struct list_head thread_list; + elf_fpregset_t *fpu; +#ifdef ELF_CORE_COPY_XFPREGS + elf_fpxregset_t *xfpu; +#endif + int thread_status_size; + int numnote; +}; + +static int fill_note_info(struct elfhdr *elf, int phdrs, + struct elf_note_info *info, + long signr, struct pt_regs *regs) +{ +#define NUM_NOTES 6 + struct list_head *t; + struct task_struct *g, *p; + + info->notes = NULL; + info->prstatus = NULL; + info->psinfo = NULL; + info->fpu = NULL; +#ifdef ELF_CORE_COPY_XFPREGS + info->xfpu = NULL; +#endif + INIT_LIST_HEAD(&info->thread_list); + + info->notes = kmalloc(NUM_NOTES * sizeof(struct memelfnote), + GFP_KERNEL); + if (!info->notes) + return 0; + info->psinfo = kmalloc(sizeof(*info->psinfo), GFP_KERNEL); + if (!info->psinfo) + return 0; + info->prstatus = kmalloc(sizeof(*info->prstatus), GFP_KERNEL); + if (!info->prstatus) + return 0; + info->fpu = kmalloc(sizeof(*info->fpu), GFP_KERNEL); + if (!info->fpu) + return 0; +#ifdef ELF_CORE_COPY_XFPREGS + info->xfpu = kmalloc(sizeof(*info->xfpu), GFP_KERNEL); + if (!info->xfpu) + return 0; +#endif + + info->thread_status_size = 0; + if (signr) { + struct elf_thread_status *tmp; + rcu_read_lock(); + do_each_thread(g, p) + if (current->mm == p->mm && current != p) { + tmp = kzalloc(sizeof(*tmp), GFP_ATOMIC); + if (!tmp) { + rcu_read_unlock(); + return 0; + } + tmp->thread = p; + list_add(&tmp->list, &info->thread_list); + } + while_each_thread(g, p); + rcu_read_unlock(); + list_for_each(t, &info->thread_list) { + struct elf_thread_status *tmp; + int sz; + + tmp = list_entry(t, struct elf_thread_status, list); + sz = elf_dump_thread_status(signr, tmp); + info->thread_status_size += sz; + } + } + /* now collect the dump for the current */ + memset(info->prstatus, 0, sizeof(*info->prstatus)); + fill_prstatus(info->prstatus, current, signr); + elf_core_copy_regs(&info->prstatus->pr_reg, regs); + + /* Set up header */ + fill_elf_header(elf, phdrs, ELF_ARCH, ELF_CORE_EFLAGS, ELF_OSABI); + + /* + * Set up the notes in similar form to SVR4 core dumps made + * with info from their /proc. + */ + + fill_note(info->notes + 0, "CORE", NT_PRSTATUS, + sizeof(*info->prstatus), info->prstatus); + fill_psinfo(info->psinfo, current->group_leader, current->mm); + fill_note(info->notes + 1, "CORE", NT_PRPSINFO, + sizeof(*info->psinfo), info->psinfo); + + info->numnote = 2; + + fill_auxv_note(&info->notes[info->numnote++], current->mm); + + /* Try to dump the FPU. */ + info->prstatus->pr_fpvalid = elf_core_copy_task_fpregs(current, regs, + info->fpu); + if (info->prstatus->pr_fpvalid) + fill_note(info->notes + info->numnote++, + "CORE", NT_PRFPREG, sizeof(*info->fpu), info->fpu); +#ifdef ELF_CORE_COPY_XFPREGS + if (elf_core_copy_task_xfpregs(current, info->xfpu)) + fill_note(info->notes + info->numnote++, + "LINUX", ELF_CORE_XFPREG_TYPE, + sizeof(*info->xfpu), info->xfpu); +#endif + + return 1; + +#undef NUM_NOTES +} + +static size_t get_note_info_size(struct elf_note_info *info) +{ + int sz = 0; + int i; + + for (i = 0; i < info->numnote; i++) + sz += notesize(info->notes + i); + + sz += info->thread_status_size; + + return sz; +} + +static int write_note_info(struct elf_note_info *info, + struct file *file, loff_t *foffset) +{ + int i; + struct list_head *t; + + for (i = 0; i < info->numnote; i++) + if (!writenote(info->notes + i, file, foffset)) + return 0; + + /* write out the thread status notes section */ + list_for_each(t, &info->thread_list) { + struct elf_thread_status *tmp = + list_entry(t, struct elf_thread_status, list); + + for (i = 0; i < tmp->num_notes; i++) + if (!writenote(&tmp->notes[i], file, foffset)) + return 0; + } + + return 1; +} + +static void free_note_info(struct elf_note_info *info) +{ + while (!list_empty(&info->thread_list)) { + struct list_head *tmp = info->thread_list.next; + list_del(tmp); + kfree(list_entry(tmp, struct elf_thread_status, list)); + } + + kfree(info->prstatus); + kfree(info->psinfo); + kfree(info->notes); + kfree(info->fpu); +#ifdef ELF_CORE_COPY_XFPREGS + kfree(info->xfpu); +#endif +} + +#endif + static struct vm_area_struct *first_vma(struct task_struct *tsk, struct vm_area_struct *gate_vma) { @@ -1534,29 +1900,15 @@ static struct vm_area_struct *next_vma(struct vm_area_struct *this_vma, */ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file, unsigned long limit) { -#define NUM_NOTES 6 int has_dumped = 0; mm_segment_t fs; int segs; size_t size = 0; - int i; struct vm_area_struct *vma, *gate_vma; struct elfhdr *elf = NULL; loff_t offset = 0, dataoff, foffset; - int numnote; - struct memelfnote *notes = NULL; - struct elf_prstatus *prstatus = NULL; /* NT_PRSTATUS */ - struct elf_prpsinfo *psinfo = NULL; /* NT_PRPSINFO */ - struct task_struct *g, *p; - LIST_HEAD(thread_list); - struct list_head *t; - elf_fpregset_t *fpu = NULL; -#ifdef ELF_CORE_COPY_XFPREGS - elf_fpxregset_t *xfpu = NULL; -#endif - int thread_status_size = 0; - elf_addr_t *auxv; unsigned long mm_flags; + struct elf_note_info info; /* * We no longer stop all VM operations. @@ -1574,52 +1926,6 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file, un elf = kmalloc(sizeof(*elf), GFP_KERNEL); if (!elf) goto cleanup; - prstatus = kmalloc(sizeof(*prstatus), GFP_KERNEL); - if (!prstatus) - goto cleanup; - psinfo = kmalloc(sizeof(*psinfo), GFP_KERNEL); - if (!psinfo) - goto cleanup; - notes = kmalloc(NUM_NOTES * sizeof(struct memelfnote), GFP_KERNEL); - if (!notes) - goto cleanup; - fpu = kmalloc(sizeof(*fpu), GFP_KERNEL); - if (!fpu) - goto cleanup; -#ifdef ELF_CORE_COPY_XFPREGS - xfpu = kmalloc(sizeof(*xfpu), GFP_KERNEL); - if (!xfpu) - goto cleanup; -#endif - - if (signr) { - struct elf_thread_status *tmp; - rcu_read_lock(); - do_each_thread(g,p) - if (current->mm == p->mm && current != p) { - tmp = kzalloc(sizeof(*tmp), GFP_ATOMIC); - if (!tmp) { - rcu_read_unlock(); - goto cleanup; - } - tmp->thread = p; - list_add(&tmp->list, &thread_list); - } - while_each_thread(g,p); - rcu_read_unlock(); - list_for_each(t, &thread_list) { - struct elf_thread_status *tmp; - int sz; - - tmp = list_entry(t, struct elf_thread_status, list); - sz = elf_dump_thread_status(signr, tmp); - thread_status_size += sz; - } - } - /* now collect the dump for the current */ - memset(prstatus, 0, sizeof(*prstatus)); - fill_prstatus(prstatus, current, signr); - elf_core_copy_regs(&prstatus->pr_reg, regs); segs = current->mm->map_count; #ifdef ELF_CORE_EXTRA_PHDRS @@ -1630,42 +1936,16 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file, un if (gate_vma != NULL) segs++; - /* Set up header */ - fill_elf_header(elf, segs + 1); /* including notes section */ - - has_dumped = 1; - current->flags |= PF_DUMPCORE; - /* - * Set up the notes in similar form to SVR4 core dumps made - * with info from their /proc. + * Collect all the non-memory information about the process for the + * notes. This also sets up the file header. */ + if (!fill_note_info(elf, segs + 1, /* including notes section */ + &info, signr, regs)) + goto cleanup; - fill_note(notes + 0, "CORE", NT_PRSTATUS, sizeof(*prstatus), prstatus); - fill_psinfo(psinfo, current->group_leader, current->mm); - fill_note(notes + 1, "CORE", NT_PRPSINFO, sizeof(*psinfo), psinfo); - - numnote = 2; - - auxv = (elf_addr_t *)current->mm->saved_auxv; - - i = 0; - do - i += 2; - while (auxv[i - 2] != AT_NULL); - fill_note(¬es[numnote++], "CORE", NT_AUXV, - i * sizeof(elf_addr_t), auxv); - - /* Try to dump the FPU. */ - if ((prstatus->pr_fpvalid = - elf_core_copy_task_fpregs(current, regs, fpu))) - fill_note(notes + numnote++, - "CORE", NT_PRFPREG, sizeof(*fpu), fpu); -#ifdef ELF_CORE_COPY_XFPREGS - if (elf_core_copy_task_xfpregs(current, xfpu)) - fill_note(notes + numnote++, - "LINUX", ELF_CORE_XFPREG_TYPE, sizeof(*xfpu), xfpu); -#endif + has_dumped = 1; + current->flags |= PF_DUMPCORE; fs = get_fs(); set_fs(KERNEL_DS); @@ -1678,12 +1958,7 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file, un /* Write notes phdr entry */ { struct elf_phdr phdr; - int sz = 0; - - for (i = 0; i < numnote; i++) - sz += notesize(notes + i); - - sz += thread_status_size; + size_t sz = get_note_info_size(&info); sz += elf_coredump_extra_notes_size(); @@ -1728,23 +2003,12 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file, un #endif /* write out the notes section */ - for (i = 0; i < numnote; i++) - if (!writenote(notes + i, file, &foffset)) - goto end_coredump; + if (!write_note_info(&info, file, &foffset)) + goto end_coredump; if (elf_coredump_extra_notes_write(file, &foffset)) goto end_coredump; - /* write out the thread status notes section */ - list_for_each(t, &thread_list) { - struct elf_thread_status *tmp = - list_entry(t, struct elf_thread_status, list); - - for (i = 0; i < tmp->num_notes; i++) - if (!writenote(&tmp->notes[i], file, &foffset)) - goto end_coredump; - } - /* Align to page */ DUMP_SEEK(dataoff - foffset); @@ -1795,22 +2059,9 @@ end_coredump: set_fs(fs); cleanup: - while (!list_empty(&thread_list)) { - struct list_head *tmp = thread_list.next; - list_del(tmp); - kfree(list_entry(tmp, struct elf_thread_status, list)); - } - kfree(elf); - kfree(prstatus); - kfree(psinfo); - kfree(notes); - kfree(fpu); -#ifdef ELF_CORE_COPY_XFPREGS - kfree(xfpu); -#endif + free_note_info(&info); return has_dumped; -#undef NUM_NOTES } #endif /* USE_ELF_CORE_DUMP */ diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c index 33764fd6db6..0498b181dd5 100644 --- a/fs/binfmt_flat.c +++ b/fs/binfmt_flat.c @@ -20,7 +20,6 @@ #include <linux/sched.h> #include <linux/mm.h> #include <linux/mman.h> -#include <linux/a.out.h> #include <linux/errno.h> #include <linux/signal.h> #include <linux/string.h> @@ -444,12 +443,12 @@ static int load_flat_file(struct linux_binprm * bprm, if (strncmp(hdr->magic, "bFLT", 4)) { /* + * Previously, here was a printk to tell people + * "BINFMT_FLAT: bad header magic". + * But for the kernel which also use ELF FD-PIC format, this + * error message is confusing. * because a lot of people do not manage to produce good - * flat binaries, we leave this printk to help them realise - * the problem. We only print the error if its not a script file */ - if (strncmp(hdr->magic, "#!", 2)) - printk("BINFMT_FLAT: bad header magic\n"); ret = -ENOEXEC; goto err; } diff --git a/fs/binfmt_som.c b/fs/binfmt_som.c index 9208c41209f..14c63527c76 100644 --- a/fs/binfmt_som.c +++ b/fs/binfmt_som.c @@ -29,7 +29,6 @@ #include <linux/personality.h> #include <linux/init.h> -#include <asm/a.out.h> #include <asm/uaccess.h> #include <asm/pgtable.h> @@ -248,11 +248,13 @@ inline int bio_hw_segments(struct request_queue *q, struct bio *bio) */ void __bio_clone(struct bio *bio, struct bio *bio_src) { - struct request_queue *q = bdev_get_queue(bio_src->bi_bdev); - memcpy(bio->bi_io_vec, bio_src->bi_io_vec, bio_src->bi_max_vecs * sizeof(struct bio_vec)); + /* + * most users will be overriding ->bi_bdev with a new target, + * so we don't set nor calculate new physical/hw segment counts here + */ bio->bi_sector = bio_src->bi_sector; bio->bi_bdev = bio_src->bi_bdev; bio->bi_flags |= 1 << BIO_CLONED; @@ -260,8 +262,6 @@ void __bio_clone(struct bio *bio, struct bio *bio_src) bio->bi_vcnt = bio_src->bi_vcnt; bio->bi_size = bio_src->bi_size; bio->bi_idx = bio_src->bi_idx; - bio_phys_segments(q, bio); - bio_hw_segments(q, bio); } /** @@ -903,7 +903,7 @@ void bio_set_pages_dirty(struct bio *bio) } } -void bio_release_pages(struct bio *bio) +static void bio_release_pages(struct bio *bio) { struct bio_vec *bvec = bio->bi_io_vec; int i; @@ -1194,6 +1194,8 @@ EXPORT_SYMBOL(bio_hw_segments); EXPORT_SYMBOL(bio_add_page); EXPORT_SYMBOL(bio_add_pc_page); EXPORT_SYMBOL(bio_get_nr_vecs); +EXPORT_SYMBOL(bio_map_user); +EXPORT_SYMBOL(bio_unmap_user); EXPORT_SYMBOL(bio_map_kern); EXPORT_SYMBOL(bio_pair_release); EXPORT_SYMBOL(bio_split); diff --git a/fs/block_dev.c b/fs/block_dev.c index 993f78c5522..7d822fae776 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -31,6 +31,8 @@ struct bdev_inode { struct inode vfs_inode; }; +static const struct address_space_operations def_blk_aops; + static inline struct bdev_inode *BDEV_I(struct inode *inode) { return container_of(inode, struct bdev_inode, vfs_inode); @@ -171,203 +173,6 @@ blkdev_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, iov, offset, nr_segs, blkdev_get_blocks, NULL); } -#if 0 -static void blk_end_aio(struct bio *bio, int error) -{ - struct kiocb *iocb = bio->bi_private; - atomic_t *bio_count = &iocb->ki_bio_count; - - if (bio_data_dir(bio) == READ) - bio_check_pages_dirty(bio); - else { - bio_release_pages(bio); - bio_put(bio); - } - - /* iocb->ki_nbytes stores error code from LLDD */ - if (error) - iocb->ki_nbytes = -EIO; - - if (atomic_dec_and_test(bio_count)) { - if ((long)iocb->ki_nbytes < 0) - aio_complete(iocb, iocb->ki_nbytes, 0); - else - aio_complete(iocb, iocb->ki_left, 0); - } - - return 0; -} - -#define VEC_SIZE 16 -struct pvec { - unsigned short nr; - unsigned short idx; - struct page *page[VEC_SIZE]; -}; - -#define PAGES_SPANNED(addr, len) \ - (DIV_ROUND_UP((addr) + (len), PAGE_SIZE) - (addr) / PAGE_SIZE); - -/* - * get page pointer for user addr, we internally cache struct page array for - * (addr, count) range in pvec to avoid frequent call to get_user_pages. If - * internal page list is exhausted, a batch count of up to VEC_SIZE is used - * to get next set of page struct. - */ -static struct page *blk_get_page(unsigned long addr, size_t count, int rw, - struct pvec *pvec) -{ - int ret, nr_pages; - if (pvec->idx == pvec->nr) { - nr_pages = PAGES_SPANNED(addr, count); - nr_pages = min(nr_pages, VEC_SIZE); - down_read(¤t->mm->mmap_sem); - ret = get_user_pages(current, current->mm, addr, nr_pages, - rw == READ, 0, pvec->page, NULL); - up_read(¤t->mm->mmap_sem); - if (ret < 0) - return ERR_PTR(ret); - pvec->nr = ret; - pvec->idx = 0; - } - return pvec->page[pvec->idx++]; -} - -/* return a page back to pvec array */ -static void blk_unget_page(struct page *page, struct pvec *pvec) -{ - pvec->page[--pvec->idx] = page; -} - -static ssize_t -blkdev_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, - loff_t pos, unsigned long nr_segs) -{ - struct inode *inode = iocb->ki_filp->f_mapping->host; - unsigned blkbits = blksize_bits(bdev_hardsect_size(I_BDEV(inode))); - unsigned blocksize_mask = (1 << blkbits) - 1; - unsigned long seg = 0; /* iov segment iterator */ - unsigned long nvec; /* number of bio vec needed */ - unsigned long cur_off; /* offset into current page */ - unsigned long cur_len; /* I/O len of current page, up to PAGE_SIZE */ - - unsigned long addr; /* user iovec address */ - size_t count; /* user iovec len */ - size_t nbytes = iocb->ki_nbytes = iocb->ki_left; /* total xfer size */ - loff_t size; /* size of block device */ - struct bio *bio; - atomic_t *bio_count = &iocb->ki_bio_count; - struct page *page; - struct pvec pvec; - - pvec.nr = 0; - pvec.idx = 0; - - if (pos & blocksize_mask) - return -EINVAL; - - size = i_size_read(inode); - if (pos + nbytes > size) { - nbytes = size - pos; - iocb->ki_left = nbytes; - } - - /* - * check first non-zero iov alignment, the remaining - * iov alignment is checked inside bio loop below. - */ - do { - addr = (unsigned long) iov[seg].iov_base; - count = min(iov[seg].iov_len, nbytes); - if (addr & blocksize_mask || count & blocksize_mask) - return -EINVAL; - } while (!count && ++seg < nr_segs); - atomic_set(bio_count, 1); - - while (nbytes) { - /* roughly estimate number of bio vec needed */ - nvec = (nbytes + PAGE_SIZE - 1) / PAGE_SIZE; - nvec = max(nvec, nr_segs - seg); - nvec = min(nvec, (unsigned long) BIO_MAX_PAGES); - - /* bio_alloc should not fail with GFP_KERNEL flag */ - bio = bio_alloc(GFP_KERNEL, nvec); - bio->bi_bdev = I_BDEV(inode); - bio->bi_end_io = blk_end_aio; - bio->bi_private = iocb; - bio->bi_sector = pos >> blkbits; -same_bio: - cur_off = addr & ~PAGE_MASK; - cur_len = PAGE_SIZE - cur_off; - if (count < cur_len) - cur_len = count; - - page = blk_get_page(addr, count, rw, &pvec); - if (unlikely(IS_ERR(page))) - goto backout; - - if (bio_add_page(bio, page, cur_len, cur_off)) { - pos += cur_len; - addr += cur_len; - count -= cur_len; - nbytes -= cur_len; - - if (count) - goto same_bio; - while (++seg < nr_segs) { - addr = (unsigned long) iov[seg].iov_base; - count = iov[seg].iov_len; - if (!count) - continue; - if (unlikely(addr & blocksize_mask || - count & blocksize_mask)) { - page = ERR_PTR(-EINVAL); - goto backout; - } - count = min(count, nbytes); - goto same_bio; - } - } else { - blk_unget_page(page, &pvec); - } - - /* bio is ready, submit it */ - if (rw == READ) - bio_set_pages_dirty(bio); - atomic_inc(bio_count); - submit_bio(rw, bio); - } - -completion: - iocb->ki_left -= nbytes; - nbytes = iocb->ki_left; - iocb->ki_pos += nbytes; - - blk_run_address_space(inode->i_mapping); - if (atomic_dec_and_test(bio_count)) - aio_complete(iocb, nbytes, 0); - - return -EIOCBQUEUED; - -backout: - /* - * back out nbytes count constructed so far for this bio, - * we will throw away current bio. - */ - nbytes += bio->bi_size; - bio_release_pages(bio); - bio_put(bio); - - /* - * if no bio was submmitted, return the error code. - * otherwise, proceed with pending I/O completion. - */ - if (atomic_read(bio_count) == 1) - return PTR_ERR(page); - goto completion; -} -#endif - static int blkdev_writepage(struct page *page, struct writeback_control *wbc) { return block_write_full_page(page, blkdev_get_block, wbc); @@ -534,7 +339,6 @@ void __init bdev_cache_init(void) if (err) panic("Cannot register bdev pseudo-fs"); bd_mnt = kern_mount(&bd_type); - err = PTR_ERR(bd_mnt); if (IS_ERR(bd_mnt)) panic("Cannot create bdev pseudo-fs"); blockdev_superblock = bd_mnt->mnt_sb; /* For writeback */ @@ -738,9 +542,9 @@ EXPORT_SYMBOL(bd_release); static struct kobject *bdev_get_kobj(struct block_device *bdev) { if (bdev->bd_contains != bdev) - return kobject_get(&bdev->bd_part->kobj); + return kobject_get(&bdev->bd_part->dev.kobj); else - return kobject_get(&bdev->bd_disk->kobj); + return kobject_get(&bdev->bd_disk->dev.kobj); } static struct kobject *bdev_get_holder(struct block_device *bdev) @@ -1176,7 +980,7 @@ static int do_open(struct block_device *bdev, struct file *file, int for_part) ret = -ENXIO; goto out_first; } - kobject_get(&p->kobj); + kobject_get(&p->dev.kobj); bdev->bd_part = p; bd_set_size(bdev, (loff_t) p->nr_sects << 9); } @@ -1299,7 +1103,7 @@ static int __blkdev_put(struct block_device *bdev, int for_part) module_put(owner); if (bdev->bd_contains != bdev) { - kobject_put(&bdev->bd_part->kobj); + kobject_put(&bdev->bd_part->dev.kobj); bdev->bd_part = NULL; } bdev->bd_disk = NULL; @@ -1335,7 +1139,7 @@ static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg) return blkdev_ioctl(file->f_mapping->host, file, cmd, arg); } -const struct address_space_operations def_blk_aops = { +static const struct address_space_operations def_blk_aops = { .readpage = blkdev_readpage, .writepage = blkdev_writepage, .sync_page = block_sync_page, @@ -1398,19 +1202,19 @@ struct block_device *lookup_bdev(const char *path) if (error) return ERR_PTR(error); - inode = nd.dentry->d_inode; + inode = nd.path.dentry->d_inode; error = -ENOTBLK; if (!S_ISBLK(inode->i_mode)) goto fail; error = -EACCES; - if (nd.mnt->mnt_flags & MNT_NODEV) + if (nd.path.mnt->mnt_flags & MNT_NODEV) goto fail; error = -ENOMEM; bdev = bd_acquire(inode); if (!bdev) goto fail; out: - path_release(&nd); + path_put(&nd.path); return bdev; fail: bdev = ERR_PTR(error); diff --git a/fs/buffer.c b/fs/buffer.c index 7249e014819..39ff14403d1 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -67,14 +67,14 @@ static int sync_buffer(void *word) return 0; } -void fastcall __lock_buffer(struct buffer_head *bh) +void __lock_buffer(struct buffer_head *bh) { wait_on_bit_lock(&bh->b_state, BH_Lock, sync_buffer, TASK_UNINTERRUPTIBLE); } EXPORT_SYMBOL(__lock_buffer); -void fastcall unlock_buffer(struct buffer_head *bh) +void unlock_buffer(struct buffer_head *bh) { smp_mb__before_clear_bit(); clear_buffer_locked(bh); @@ -627,8 +627,7 @@ repeat: } /** - * sync_mapping_buffers - write out and wait upon a mapping's "associated" - * buffers + * sync_mapping_buffers - write out & wait upon a mapping's "associated" buffers * @mapping: the mapping which wants those buffers written * * Starts I/O against the buffers at mapping->private_list, and waits upon @@ -678,7 +677,7 @@ void mark_buffer_dirty_inode(struct buffer_head *bh, struct inode *inode) } else { BUG_ON(mapping->assoc_mapping != buffer_mapping); } - if (list_empty(&bh->b_assoc_buffers)) { + if (!bh->b_assoc_map) { spin_lock(&buffer_mapping->private_lock); list_move_tail(&bh->b_assoc_buffers, &mapping->private_list); @@ -794,6 +793,7 @@ static int fsync_buffers_list(spinlock_t *lock, struct list_head *list) { struct buffer_head *bh; struct list_head tmp; + struct address_space *mapping; int err = 0, err2; INIT_LIST_HEAD(&tmp); @@ -801,9 +801,14 @@ static int fsync_buffers_list(spinlock_t *lock, struct list_head *list) spin_lock(lock); while (!list_empty(list)) { bh = BH_ENTRY(list->next); + mapping = bh->b_assoc_map; __remove_assoc_queue(bh); + /* Avoid race with mark_buffer_dirty_inode() which does + * a lockless check and we rely on seeing the dirty bit */ + smp_mb(); if (buffer_dirty(bh) || buffer_locked(bh)) { list_add(&bh->b_assoc_buffers, &tmp); + bh->b_assoc_map = mapping; if (buffer_dirty(bh)) { get_bh(bh); spin_unlock(lock); @@ -822,8 +827,17 @@ static int fsync_buffers_list(spinlock_t *lock, struct list_head *list) while (!list_empty(&tmp)) { bh = BH_ENTRY(tmp.prev); - list_del_init(&bh->b_assoc_buffers); get_bh(bh); + mapping = bh->b_assoc_map; + __remove_assoc_queue(bh); + /* Avoid race with mark_buffer_dirty_inode() which does + * a lockless check and we rely on seeing the dirty bit */ + smp_mb(); + if (buffer_dirty(bh)) { + list_add(&bh->b_assoc_buffers, + &mapping->private_list); + bh->b_assoc_map = mapping; + } spin_unlock(lock); wait_on_buffer(bh); if (!buffer_uptodate(bh)) @@ -1164,10 +1178,23 @@ __getblk_slow(struct block_device *bdev, sector_t block, int size) * mark_buffer_dirty() is atomic. It takes bh->b_page->mapping->private_lock, * mapping->tree_lock and the global inode_lock. */ -void fastcall mark_buffer_dirty(struct buffer_head *bh) +void mark_buffer_dirty(struct buffer_head *bh) { WARN_ON_ONCE(!buffer_uptodate(bh)); - if (!buffer_dirty(bh) && !test_set_buffer_dirty(bh)) + + /* + * Very *carefully* optimize the it-is-already-dirty case. + * + * Don't let the final "is it dirty" escape to before we + * perhaps modified the buffer. + */ + if (buffer_dirty(bh)) { + smp_mb(); + if (buffer_dirty(bh)) + return; + } + + if (!test_set_buffer_dirty(bh)) __set_page_dirty(bh->b_page, page_mapping(bh->b_page), 0); } @@ -1195,7 +1222,7 @@ void __brelse(struct buffer_head * buf) void __bforget(struct buffer_head *bh) { clear_buffer_dirty(bh); - if (!list_empty(&bh->b_assoc_buffers)) { + if (bh->b_assoc_map) { struct address_space *buffer_mapping = bh->b_page->mapping; spin_lock(&buffer_mapping->private_lock); @@ -1436,6 +1463,7 @@ void invalidate_bh_lrus(void) { on_each_cpu(invalidate_bh_lru, NULL, 1, 1); } +EXPORT_SYMBOL_GPL(invalidate_bh_lrus); void set_bh_page(struct buffer_head *bh, struct page *page, unsigned long offset) @@ -1798,7 +1826,7 @@ void page_zero_new_buffers(struct page *page, unsigned from, unsigned to) start = max(from, block_start); size = min(to, block_end) - start; - zero_user_page(page, start, size, KM_USER0); + zero_user(page, start, size); set_buffer_uptodate(bh); } @@ -1861,19 +1889,10 @@ static int __block_prepare_write(struct inode *inode, struct page *page, mark_buffer_dirty(bh); continue; } - if (block_end > to || block_start < from) { - void *kaddr; - - kaddr = kmap_atomic(page, KM_USER0); - if (block_end > to) - memset(kaddr+to, 0, - block_end-to); - if (block_start < from) - memset(kaddr+block_start, - 0, from-block_start); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); - } + if (block_end > to || block_start < from) + zero_user_segments(page, + to, block_end, + block_start, from); continue; } } @@ -2104,8 +2123,7 @@ int block_read_full_page(struct page *page, get_block_t *get_block) SetPageError(page); } if (!buffer_mapped(bh)) { - zero_user_page(page, i * blocksize, blocksize, - KM_USER0); + zero_user(page, i * blocksize, blocksize); if (!err) set_buffer_uptodate(bh); continue; @@ -2218,7 +2236,7 @@ int cont_expand_zero(struct file *file, struct address_space *mapping, &page, &fsdata); if (err) goto out; - zero_user_page(page, zerofrom, len, KM_USER0); + zero_user(page, zerofrom, len); err = pagecache_write_end(file, mapping, curpos, len, len, page, fsdata); if (err < 0) @@ -2245,7 +2263,7 @@ int cont_expand_zero(struct file *file, struct address_space *mapping, &page, &fsdata); if (err) goto out; - zero_user_page(page, zerofrom, len, KM_USER0); + zero_user(page, zerofrom, len); err = pagecache_write_end(file, mapping, curpos, len, len, page, fsdata); if (err < 0) @@ -2422,7 +2440,6 @@ int nobh_write_begin(struct file *file, struct address_space *mapping, unsigned block_in_page; unsigned block_start, block_end; sector_t block_in_file; - char *kaddr; int nr_reads = 0; int ret = 0; int is_mapped_to_disk = 1; @@ -2493,13 +2510,8 @@ int nobh_write_begin(struct file *file, struct address_space *mapping, continue; } if (buffer_new(bh) || !buffer_mapped(bh)) { - kaddr = kmap_atomic(page, KM_USER0); - if (block_start < from) - memset(kaddr+block_start, 0, from-block_start); - if (block_end > to) - memset(kaddr + to, 0, block_end - to); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); + zero_user_segments(page, block_start, from, + to, block_end); continue; } if (buffer_uptodate(bh)) @@ -2565,14 +2577,13 @@ int nobh_write_end(struct file *file, struct address_space *mapping, struct inode *inode = page->mapping->host; struct buffer_head *head = fsdata; struct buffer_head *bh; + BUG_ON(fsdata != NULL && page_has_buffers(page)); - if (!PageMappedToDisk(page)) { - if (unlikely(copied < len) && !page_has_buffers(page)) - attach_nobh_buffers(page, head); - if (page_has_buffers(page)) - return generic_write_end(file, mapping, pos, len, - copied, page, fsdata); - } + if (unlikely(copied < len) && !page_has_buffers(page)) + attach_nobh_buffers(page, head); + if (page_has_buffers(page)) + return generic_write_end(file, mapping, pos, len, + copied, page, fsdata); SetPageUptodate(page); set_page_dirty(page); @@ -2636,7 +2647,7 @@ int nobh_writepage(struct page *page, get_block_t *get_block, * the page size, the remaining memory is zeroed when mapped, and * writes to that region are not written out to the file." */ - zero_user_page(page, offset, PAGE_CACHE_SIZE - offset, KM_USER0); + zero_user_segment(page, offset, PAGE_CACHE_SIZE); out: ret = mpage_writepage(page, get_block, wbc); if (ret == -EAGAIN) @@ -2709,7 +2720,7 @@ has_buffers: if (page_has_buffers(page)) goto has_buffers; } - zero_user_page(page, offset, length, KM_USER0); + zero_user(page, offset, length); set_page_dirty(page); err = 0; @@ -2785,7 +2796,7 @@ int block_truncate_page(struct address_space *mapping, goto unlock; } - zero_user_page(page, offset, length, KM_USER0); + zero_user(page, offset, length); mark_buffer_dirty(bh); err = 0; @@ -2831,7 +2842,7 @@ int block_write_full_page(struct page *page, get_block_t *get_block, * the page size, the remaining memory is zeroed when mapped, and * writes to that region are not written out to the file." */ - zero_user_page(page, offset, PAGE_CACHE_SIZE - offset, KM_USER0); + zero_user_segment(page, offset, PAGE_CACHE_SIZE); return __block_write_full_page(inode, page, get_block, wbc); } @@ -3037,7 +3048,7 @@ drop_buffers(struct page *page, struct buffer_head **buffers_to_free) do { struct buffer_head *next = bh->b_this_page; - if (!list_empty(&bh->b_assoc_buffers)) + if (bh->b_assoc_map) __remove_assoc_queue(bh); bh = next; } while (bh != head); @@ -3169,7 +3180,7 @@ static void recalc_bh_state(void) struct buffer_head *alloc_buffer_head(gfp_t gfp_flags) { - struct buffer_head *ret = kmem_cache_zalloc(bh_cachep, + struct buffer_head *ret = kmem_cache_alloc(bh_cachep, set_migrateflags(gfp_flags, __GFP_RECLAIMABLE)); if (ret) { INIT_LIST_HEAD(&ret->b_assoc_buffers); @@ -3213,12 +3224,68 @@ static int buffer_cpu_notify(struct notifier_block *self, return NOTIFY_OK; } +/** + * bh_uptodate_or_lock - Test whether the buffer is uptodate + * @bh: struct buffer_head + * + * Return true if the buffer is up-to-date and false, + * with the buffer locked, if not. + */ +int bh_uptodate_or_lock(struct buffer_head *bh) +{ + if (!buffer_uptodate(bh)) { + lock_buffer(bh); + if (!buffer_uptodate(bh)) + return 0; + unlock_buffer(bh); + } + return 1; +} +EXPORT_SYMBOL(bh_uptodate_or_lock); + +/** + * bh_submit_read - Submit a locked buffer for reading + * @bh: struct buffer_head + * + * Returns zero on success and -EIO on error. + */ +int bh_submit_read(struct buffer_head *bh) +{ + BUG_ON(!buffer_locked(bh)); + + if (buffer_uptodate(bh)) { + unlock_buffer(bh); + return 0; + } + + get_bh(bh); + bh->b_end_io = end_buffer_read_sync; + submit_bh(READ, bh); + wait_on_buffer(bh); + if (buffer_uptodate(bh)) + return 0; + return -EIO; +} +EXPORT_SYMBOL(bh_submit_read); + +static void +init_buffer_head(struct kmem_cache *cachep, void *data) +{ + struct buffer_head *bh = data; + + memset(bh, 0, sizeof(*bh)); + INIT_LIST_HEAD(&bh->b_assoc_buffers); +} + void __init buffer_init(void) { int nrpages; - bh_cachep = KMEM_CACHE(buffer_head, - SLAB_RECLAIM_ACCOUNT|SLAB_PANIC|SLAB_MEM_SPREAD); + bh_cachep = kmem_cache_create("buffer_head", + sizeof(struct buffer_head), 0, + (SLAB_RECLAIM_ACCOUNT|SLAB_PANIC| + SLAB_MEM_SPREAD), + init_buffer_head); /* * Limit the bh occupancy to 10% of ZONE_NORMAL diff --git a/fs/char_dev.c b/fs/char_dev.c index c3bfa76765c..038674aa88a 100644 --- a/fs/char_dev.c +++ b/fs/char_dev.c @@ -357,7 +357,7 @@ void cdev_put(struct cdev *p) /* * Called every time a character special file is opened */ -int chrdev_open(struct inode * inode, struct file * filp) +static int chrdev_open(struct inode *inode, struct file *filp) { struct cdev *p; struct cdev *new = NULL; @@ -510,9 +510,8 @@ struct cdev *cdev_alloc(void) { struct cdev *p = kzalloc(sizeof(struct cdev), GFP_KERNEL); if (p) { - p->kobj.ktype = &ktype_cdev_dynamic; INIT_LIST_HEAD(&p->list); - kobject_init(&p->kobj); + kobject_init(&p->kobj, &ktype_cdev_dynamic); } return p; } @@ -529,8 +528,7 @@ void cdev_init(struct cdev *cdev, const struct file_operations *fops) { memset(cdev, 0, sizeof *cdev); INIT_LIST_HEAD(&cdev->list); - cdev->kobj.ktype = &ktype_cdev_default; - kobject_init(&cdev->kobj); + kobject_init(&cdev->kobj, &ktype_cdev_default); cdev->ops = fops; } diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index a609599287a..dbd91461853 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -3,7 +3,12 @@ Version 1.52 Fix oops on second mount to server when null auth is used. Enable experimental Kerberos support. Return writebehind errors on flush and sync so that events like out of disk space get reported properly on -cached files. +cached files. Fix setxattr failure to certain Samba versions. Fix mount +of second share to disconnected server session (autoreconnect on this). +Add ability to modify cifs acls for handling chmod (when mounted with +cifsacl flag). Fix prefixpath path separator so we can handle mounts +with prefixpaths longer than one directory (one path component) when +mounted to Windows servers. Version 1.51 ------------ diff --git a/fs/cifs/Makefile b/fs/cifs/Makefile index 45e42fb97c1..6ba43fb346f 100644 --- a/fs/cifs/Makefile +++ b/fs/cifs/Makefile @@ -9,3 +9,5 @@ cifs-y := cifsfs.o cifssmb.o cifs_debug.o connect.o dir.o file.o inode.o \ readdir.o ioctl.o sess.o export.o cifsacl.o cifs-$(CONFIG_CIFS_UPCALL) += cifs_spnego.o + +cifs-$(CONFIG_CIFS_DFS_UPCALL) += dns_resolve.o cifs_dfs_ref.o diff --git a/fs/cifs/README b/fs/cifs/README index bf11329ac78..50306229b0f 100644 --- a/fs/cifs/README +++ b/fs/cifs/README @@ -56,7 +56,8 @@ the CIFS VFS web site) copy it to the same directory in which mount.smbfs and similar files reside (usually /sbin). Although the helper software is not required, mount.cifs is recommended. Eventually the Samba 3.0 utility program "net" may also be helpful since it may someday provide easier mount syntax for -users who are used to Windows e.g. net use <mount point> <UNC name or cifs URL> +users who are used to Windows e.g. + net use <mount point> <UNC name or cifs URL> Note that running the Winbind pam/nss module (logon service) on all of your Linux clients is useful in mapping Uids and Gids consistently across the domain to the proper network user. The mount.cifs mount helper can be @@ -248,7 +249,7 @@ A partial list of the supported mount options follows: the CIFS session. password The user password. If the mount helper is installed, the user will be prompted for password - if it is not supplied. + if not supplied. ip The ip address of the target server unc The target server Universal Network Name (export) to mount. @@ -283,7 +284,7 @@ A partial list of the supported mount options follows: can be enabled by specifying file_mode and dir_mode on the client. Note that the mount.cifs helper must be at version 1.10 or higher to support specifying the uid - (or gid) in non-numberic form. + (or gid) in non-numeric form. gid Set the default gid for inodes (similar to above). file_mode If CIFS Unix extensions are not supported by the server this overrides the default mode for file inodes. @@ -417,9 +418,10 @@ A partial list of the supported mount options follows: acl Allow setfacl and getfacl to manage posix ACLs if server supports them. (default) noacl Do not allow setfacl and getfacl calls on this mount - user_xattr Allow getting and setting user xattrs as OS/2 EAs (extended - attributes) to the server (default) e.g. via setfattr - and getfattr utilities. + user_xattr Allow getting and setting user xattrs (those attributes whose + name begins with "user." or "os2.") as OS/2 EAs (extended + attributes) to the server. This allows support of the + setfattr and getfattr utilities. (default) nouser_xattr Do not allow getfattr/setfattr to get/set/list xattrs mapchars Translate six of the seven reserved characters (not backslash) *?<>|: @@ -434,6 +436,7 @@ A partial list of the supported mount options follows: nomapchars Do not translate any of these seven characters (default). nocase Request case insensitive path name matching (case sensitive is the default if the server suports it). + (mount option "ignorecase" is identical to "nocase") posixpaths If CIFS Unix extensions are supported, attempt to negotiate posix path name support which allows certain characters forbidden in typical CIFS filenames, without @@ -458,7 +461,7 @@ A partial list of the supported mount options follows: cifsacl Report mode bits (e.g. on stat) based on the Windows ACL for the file. (EXPERIMENTAL) servern Specify the server 's netbios name (RFC1001 name) to use - when attempting to setup a session to the server. This is + when attempting to setup a session to the server. This is needed for mounting to some older servers (such as OS/2 or Windows 98 and Windows ME) since they do not support a default server name. A server name can be up @@ -485,6 +488,9 @@ A partial list of the supported mount options follows: ntlmv2i Use NTLMv2 password hashing with packet signing lanman (if configured in kernel config) use older lanman hash +hard Retry file operations if server is not responding +soft Limit retries to unresponsive servers (usually only + one retry) before returning an error. (default) The mount.cifs mount helper also accepts a few mount options before -o including: @@ -535,8 +541,8 @@ SecurityFlags Flags which control security negotiation and must use NTLM 0x02002 may use NTLMv2 0x00004 must use NTLMv2 0x04004 - may use Kerberos security (not implemented yet) 0x00008 - must use Kerberos (not implemented yet) 0x08008 + may use Kerberos security 0x00008 + must use Kerberos 0x08008 may use lanman (weak) password hash 0x00010 must use lanman password hash 0x10010 may use plaintext passwords 0x00020 @@ -626,6 +632,6 @@ returned success. Also note that "cat /proc/fs/cifs/DebugData" will display information about the active sessions and the shares that are mounted. -Enabling Kerberos (extended security) works when CONFIG_CIFS_EXPERIMENTAL is enabled -but requires a user space helper (from the Samba project). NTLM and NTLMv2 and -LANMAN support do not require this helpr. +Enabling Kerberos (extended security) works when CONFIG_CIFS_EXPERIMENTAL is +on but requires a user space helper (from the Samba project). NTLM and NTLMv2 and +LANMAN support do not require this helper. diff --git a/fs/cifs/TODO b/fs/cifs/TODO index a8852c20072..92c9feac440 100644 --- a/fs/cifs/TODO +++ b/fs/cifs/TODO @@ -1,4 +1,4 @@ -Version 1.49 April 26, 2007 +Version 1.52 January 3, 2008 A Partial List of Missing Features ================================== @@ -16,16 +16,14 @@ SecurityDescriptors c) Better pam/winbind integration (e.g. to handle uid mapping better) -d) Verify that Kerberos signing works - -e) Cleanup now unneeded SessSetup code in +d) Cleanup now unneeded SessSetup code in fs/cifs/connect.c and add back in NTLMSSP code if any servers need it -f) MD5-HMAC signing SMB PDUs when SPNEGO style SessionSetup -used (Kerberos or NTLMSSP). Signing alreadyimplemented for NTLM -and raw NTLMSSP already. This is important when enabling -extended security and mounting to Windows 2003 Servers +e) ms-dfs and ms-dfs host name resolution cleanup + +f) fix NTLMv2 signing when two mounts with different users to same +server. g) Directory entry caching relies on a 1 second timer, rather than using FindNotify or equivalent. - (started) diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index 73c4c419663..0228ed06069 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -98,8 +98,7 @@ void cifs_dump_mids(struct TCP_Server_Info *server) if (mid_entry->resp_buf) { cifs_dump_detail(mid_entry->resp_buf); cifs_dump_mem("existing buf: ", - mid_entry->resp_buf, - 62 /* fixme */); + mid_entry->resp_buf, 62); } } } @@ -439,7 +438,7 @@ cifs_stats_read(char *buf, char **beginBuffer, off_t offset, return length; } -#endif +#endif /* STATS */ static struct proc_dir_entry *proc_fs_cifs; read_proc_t cifs_txanchor_read; @@ -482,7 +481,7 @@ cifs_proc_init(void) cifs_stats_read, NULL); if (pde) pde->write_proc = cifs_stats_write; -#endif +#endif /* STATS */ pde = create_proc_read_entry("cifsFYI", 0, proc_fs_cifs, cifsFYI_read, NULL); if (pde) @@ -918,4 +917,12 @@ security_flags_write(struct file *file, const char __user *buffer, /* BB should we turn on MAY flags for other MUST options? */ return count; } -#endif +#else +inline void cifs_proc_init(void) +{ +} + +inline void cifs_proc_clean(void) +{ +} +#endif /* PROC_FS */ diff --git a/fs/cifs/cifs_debug.h b/fs/cifs/cifs_debug.h index c26cd0d2c6d..5eb3b83bbfa 100644 --- a/fs/cifs/cifs_debug.h +++ b/fs/cifs/cifs_debug.h @@ -25,8 +25,11 @@ void cifs_dump_mem(char *label, void *data, int length); #ifdef CONFIG_CIFS_DEBUG2 +#define DBG2 2 void cifs_dump_detail(struct smb_hdr *); void cifs_dump_mids(struct TCP_Server_Info *); +#else +#define DBG2 0 #endif extern int traceSMB; /* flag which enables the function below */ void dump_smb(struct smb_hdr *, int); @@ -64,10 +67,10 @@ extern int cifsERROR; * --------- */ #else /* _CIFS_DEBUG */ -#define cERROR(button,prspec) -#define cEVENT(format,arg...) +#define cERROR(button, prspec) +#define cEVENT(format, arg...) #define cFYI(button, prspec) -#define cifserror(format,arg...) +#define cifserror(format, arg...) #endif /* _CIFS_DEBUG */ #endif /* _H_CIFS_DEBUG */ diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c new file mode 100644 index 00000000000..56c924033b7 --- /dev/null +++ b/fs/cifs/cifs_dfs_ref.c @@ -0,0 +1,377 @@ +/* + * Contains the CIFS DFS referral mounting routines used for handling + * traversal via DFS junction point + * + * Copyright (c) 2007 Igor Mammedov + * Copyright (C) International Business Machines Corp., 2008 + * Author(s): Igor Mammedov (niallain@gmail.com) + * Steve French (sfrench@us.ibm.com) + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include <linux/dcache.h> +#include <linux/mount.h> +#include <linux/namei.h> +#include <linux/vfs.h> +#include <linux/fs.h> +#include "cifsglob.h" +#include "cifsproto.h" +#include "cifsfs.h" +#include "dns_resolve.h" +#include "cifs_debug.h" + +LIST_HEAD(cifs_dfs_automount_list); + +/* + * DFS functions +*/ + +void dfs_shrink_umount_helper(struct vfsmount *vfsmnt) +{ + mark_mounts_for_expiry(&cifs_dfs_automount_list); + mark_mounts_for_expiry(&cifs_dfs_automount_list); +} + +/** + * cifs_get_share_name - extracts share name from UNC + * @node_name: pointer to UNC string + * + * Extracts sharename form full UNC. + * i.e. strips from UNC trailing path that is not part of share + * name and fixup missing '\' in the begining of DFS node refferal + * if neccessary. + * Returns pointer to share name on success or NULL on error. + * Caller is responsible for freeing returned string. + */ +static char *cifs_get_share_name(const char *node_name) +{ + int len; + char *UNC; + char *pSep; + + len = strlen(node_name); + UNC = kmalloc(len+2 /*for term null and additional \ if it's missed */, + GFP_KERNEL); + if (!UNC) + return NULL; + + /* get share name and server name */ + if (node_name[1] != '\\') { + UNC[0] = '\\'; + strncpy(UNC+1, node_name, len); + len++; + UNC[len] = 0; + } else { + strncpy(UNC, node_name, len); + UNC[len] = 0; + } + + /* find server name end */ + pSep = memchr(UNC+2, '\\', len-2); + if (!pSep) { + cERROR(1, ("%s: no server name end in node name: %s", + __func__, node_name)); + kfree(UNC); + return NULL; + } + + /* find sharename end */ + pSep++; + pSep = memchr(UNC+(pSep-UNC), '\\', len-(pSep-UNC)); + if (!pSep) { + cERROR(1, ("%s:2 cant find share name in node name: %s", + __func__, node_name)); + kfree(UNC); + return NULL; + } + /* trim path up to sharename end + * * now we have share name in UNC */ + *pSep = 0; + + return UNC; +} + + +/** + * compose_mount_options - creates mount options for refferral + * @sb_mountdata: parent/root DFS mount options (template) + * @ref_unc: refferral server UNC + * @devname: pointer for saving device name + * + * creates mount options for submount based on template options sb_mountdata + * and replacing unc,ip,prefixpath options with ones we've got form ref_unc. + * + * Returns: pointer to new mount options or ERR_PTR. + * Caller is responcible for freeing retunrned value if it is not error. + */ +static char *compose_mount_options(const char *sb_mountdata, + const char *ref_unc, + char **devname) +{ + int rc; + char *mountdata; + int md_len; + char *tkn_e; + char *srvIP = NULL; + char sep = ','; + int off, noff; + + if (sb_mountdata == NULL) + return ERR_PTR(-EINVAL); + + *devname = cifs_get_share_name(ref_unc); + rc = dns_resolve_server_name_to_ip(*devname, &srvIP); + if (rc != 0) { + cERROR(1, ("%s: Failed to resolve server part of %s to IP", + __func__, *devname)); + mountdata = ERR_PTR(rc); + goto compose_mount_options_out; + } + md_len = strlen(sb_mountdata) + strlen(srvIP) + strlen(ref_unc) + 3; + mountdata = kzalloc(md_len+1, GFP_KERNEL); + if (mountdata == NULL) { + mountdata = ERR_PTR(-ENOMEM); + goto compose_mount_options_out; + } + + /* copy all options except of unc,ip,prefixpath */ + off = 0; + if (strncmp(sb_mountdata, "sep=", 4) == 0) { + sep = sb_mountdata[4]; + strncpy(mountdata, sb_mountdata, 5); + off += 5; + } + while ((tkn_e = strchr(sb_mountdata+off, sep))) { + noff = (tkn_e - (sb_mountdata+off)) + 1; + if (strnicmp(sb_mountdata+off, "unc=", 4) == 0) { + off += noff; + continue; + } + if (strnicmp(sb_mountdata+off, "ip=", 3) == 0) { + off += noff; + continue; + } + if (strnicmp(sb_mountdata+off, "prefixpath=", 3) == 0) { + off += noff; + continue; + } + strncat(mountdata, sb_mountdata+off, noff); + off += noff; + } + strcat(mountdata, sb_mountdata+off); + mountdata[md_len] = '\0'; + + /* copy new IP and ref share name */ + strcat(mountdata, ",ip="); + strcat(mountdata, srvIP); + strcat(mountdata, ",unc="); + strcat(mountdata, *devname); + + /* find & copy prefixpath */ + tkn_e = strchr(ref_unc+2, '\\'); + if (tkn_e) { + tkn_e = strchr(tkn_e+1, '\\'); + if (tkn_e) { + strcat(mountdata, ",prefixpath="); + strcat(mountdata, tkn_e); + } + } + + /*cFYI(1,("%s: parent mountdata: %s", __func__,sb_mountdata));*/ + /*cFYI(1, ("%s: submount mountdata: %s", __func__, mountdata ));*/ + +compose_mount_options_out: + kfree(srvIP); + return mountdata; +} + + +static struct vfsmount *cifs_dfs_do_refmount(const struct vfsmount *mnt_parent, + struct dentry *dentry, char *ref_unc) +{ + struct cifs_sb_info *cifs_sb; + struct vfsmount *mnt; + char *mountdata; + char *devname = NULL; + + cifs_sb = CIFS_SB(dentry->d_inode->i_sb); + mountdata = compose_mount_options(cifs_sb->mountdata, + ref_unc, &devname); + + if (IS_ERR(mountdata)) + return (struct vfsmount *)mountdata; + + mnt = vfs_kern_mount(&cifs_fs_type, 0, devname, mountdata); + kfree(mountdata); + kfree(devname); + return mnt; + +} + +static char *build_full_dfs_path_from_dentry(struct dentry *dentry) +{ + char *full_path = NULL; + char *search_path; + char *tmp_path; + size_t l_max_len; + struct cifs_sb_info *cifs_sb; + + if (dentry->d_inode == NULL) + return NULL; + + cifs_sb = CIFS_SB(dentry->d_inode->i_sb); + + if (cifs_sb->tcon == NULL) + return NULL; + + search_path = build_path_from_dentry(dentry); + if (search_path == NULL) + return NULL; + + if (cifs_sb->tcon->Flags & SMB_SHARE_IS_IN_DFS) { + /* we should use full path name to correct working with DFS */ + l_max_len = strnlen(cifs_sb->tcon->treeName, MAX_TREE_SIZE+1) + + strnlen(search_path, MAX_PATHCONF) + 1; + tmp_path = kmalloc(l_max_len, GFP_KERNEL); + if (tmp_path == NULL) { + kfree(search_path); + return NULL; + } + strncpy(tmp_path, cifs_sb->tcon->treeName, l_max_len); + strcat(tmp_path, search_path); + tmp_path[l_max_len-1] = 0; + full_path = tmp_path; + kfree(search_path); + } else { + full_path = search_path; + } + return full_path; +} + +static int add_mount_helper(struct vfsmount *newmnt, struct nameidata *nd, + struct list_head *mntlist) +{ + /* stolen from afs code */ + int err; + + mntget(newmnt); + err = do_add_mount(newmnt, nd, nd->path.mnt->mnt_flags, mntlist); + switch (err) { + case 0: + dput(nd->path.dentry); + mntput(nd->path.mnt); + nd->path.mnt = newmnt; + nd->path.dentry = dget(newmnt->mnt_root); + break; + case -EBUSY: + /* someone else made a mount here whilst we were busy */ + while (d_mountpoint(nd->path.dentry) && + follow_down(&nd->path.mnt, &nd->path.dentry)) + ; + err = 0; + default: + mntput(newmnt); + break; + } + return err; +} + +static void dump_referral(const struct dfs_info3_param *ref) +{ + cFYI(1, ("DFS: ref path: %s", ref->path_name)); + cFYI(1, ("DFS: node path: %s", ref->node_name)); + cFYI(1, ("DFS: fl: %hd, srv_type: %hd", ref->flags, ref->server_type)); + cFYI(1, ("DFS: ref_flags: %hd, path_consumed: %hd", ref->ref_flag, + ref->path_consumed)); +} + + +static void* +cifs_dfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd) +{ + struct dfs_info3_param *referrals = NULL; + unsigned int num_referrals = 0; + struct cifs_sb_info *cifs_sb; + struct cifsSesInfo *ses; + char *full_path = NULL; + int xid, i; + int rc = 0; + struct vfsmount *mnt = ERR_PTR(-ENOENT); + + cFYI(1, ("in %s", __func__)); + BUG_ON(IS_ROOT(dentry)); + + xid = GetXid(); + + dput(nd->path.dentry); + nd->path.dentry = dget(dentry); + + cifs_sb = CIFS_SB(dentry->d_inode->i_sb); + ses = cifs_sb->tcon->ses; + + if (!ses) { + rc = -EINVAL; + goto out_err; + } + + full_path = build_full_dfs_path_from_dentry(dentry); + if (full_path == NULL) { + rc = -ENOMEM; + goto out_err; + } + + rc = get_dfs_path(xid, ses , full_path, cifs_sb->local_nls, + &num_referrals, &referrals, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + + for (i = 0; i < num_referrals; i++) { + dump_referral(referrals+i); + /* connect to a storage node */ + if (referrals[i].flags & DFSREF_STORAGE_SERVER) { + int len; + len = strlen(referrals[i].node_name); + if (len < 2) { + cERROR(1, ("%s: Net Address path too short: %s", + __func__, referrals[i].node_name)); + rc = -EINVAL; + goto out_err; + } + mnt = cifs_dfs_do_refmount(nd->path.mnt, + nd->path.dentry, + referrals[i].node_name); + cFYI(1, ("%s: cifs_dfs_do_refmount:%s , mnt:%p", + __func__, + referrals[i].node_name, mnt)); + + /* complete mount procedure if we accured submount */ + if (!IS_ERR(mnt)) + break; + } + } + + /* we need it cause for() above could exit without valid submount */ + rc = PTR_ERR(mnt); + if (IS_ERR(mnt)) + goto out_err; + + nd->path.mnt->mnt_flags |= MNT_SHRINKABLE; + rc = add_mount_helper(mnt, nd, &cifs_dfs_automount_list); + +out: + FreeXid(xid); + free_dfs_info_array(referrals, num_referrals); + kfree(full_path); + cFYI(1, ("leaving %s" , __func__)); + return ERR_PTR(rc); +out_err: + path_put(&nd->path); + goto out; +} + +struct inode_operations cifs_dfs_referral_inode_operations = { + .follow_link = cifs_dfs_follow_mountpoint, +}; + diff --git a/fs/cifs/cifs_fs_sb.h b/fs/cifs/cifs_fs_sb.h index 34af556cdd8..8ad2330ba06 100644 --- a/fs/cifs/cifs_fs_sb.h +++ b/fs/cifs/cifs_fs_sb.h @@ -43,6 +43,9 @@ struct cifs_sb_info { mode_t mnt_dir_mode; int mnt_cifs_flags; int prepathlen; - char *prepath; + char *prepath; /* relative path under the share to mount to */ +#ifdef CONFIG_CIFS_DFS_UPCALL + char *mountdata; /* mount options received at mount time */ +#endif }; #endif /* _CIFS_FS_SB_H */ diff --git a/fs/cifs/cifs_spnego.c b/fs/cifs/cifs_spnego.c index 1529d2b12e9..6653e29637a 100644 --- a/fs/cifs/cifs_spnego.c +++ b/fs/cifs/cifs_spnego.c @@ -122,11 +122,13 @@ cifs_get_spnego_key(struct cifsSesInfo *sesInfo) cFYI(1, ("key description = %s", description)); spnego_key = request_key(&cifs_spnego_key_type, description, ""); +#ifdef CONFIG_CIFS_DEBUG2 if (cifsFYI && !IS_ERR(spnego_key)) { struct cifs_spnego_msg *msg = spnego_key->payload.data; - cifs_dump_mem("SPNEGO reply blob:", msg->data, - msg->secblob_len + msg->sesskey_len); + cifs_dump_mem("SPNEGO reply blob:", msg->data, min(1024U, + msg->secblob_len + msg->sesskey_len)); } +#endif /* CONFIG_CIFS_DEBUG2 */ out: kfree(description); diff --git a/fs/cifs/cifs_unicode.c b/fs/cifs/cifs_unicode.c index b5903b89250..7d75272a6b3 100644 --- a/fs/cifs/cifs_unicode.c +++ b/fs/cifs/cifs_unicode.c @@ -32,7 +32,7 @@ * */ int -cifs_strfromUCS_le(char *to, const __le16 * from, +cifs_strfromUCS_le(char *to, const __le16 *from, int len, const struct nls_table *codepage) { int i; @@ -61,7 +61,7 @@ cifs_strfromUCS_le(char *to, const __le16 * from, * */ int -cifs_strtoUCS(__le16 * to, const char *from, int len, +cifs_strtoUCS(__le16 *to, const char *from, int len, const struct nls_table *codepage) { int charlen; diff --git a/fs/cifs/cifs_unicode.h b/fs/cifs/cifs_unicode.h index 614c11fcdcb..14eb9a2395d 100644 --- a/fs/cifs/cifs_unicode.h +++ b/fs/cifs/cifs_unicode.h @@ -254,7 +254,8 @@ UniStrstr(const wchar_t *ucs1, const wchar_t *ucs2) const wchar_t *anchor2 = ucs2; while (*ucs1) { - if (*ucs1 == *ucs2) { /* Partial match found */ + if (*ucs1 == *ucs2) { + /* Partial match found */ ucs1++; ucs2++; } else { @@ -279,7 +280,8 @@ UniToupper(register wchar_t uc) { register const struct UniCaseRange *rp; - if (uc < sizeof (CifsUniUpperTable)) { /* Latin characters */ + if (uc < sizeof(CifsUniUpperTable)) { + /* Latin characters */ return uc + CifsUniUpperTable[uc]; /* Use base tables */ } else { rp = CifsUniUpperRange; /* Use range tables */ @@ -320,7 +322,8 @@ UniTolower(wchar_t uc) { register struct UniCaseRange *rp; - if (uc < sizeof (UniLowerTable)) { /* Latin characters */ + if (uc < sizeof(UniLowerTable)) { + /* Latin characters */ return uc + UniLowerTable[uc]; /* Use base tables */ } else { rp = UniLowerRange; /* Use range tables */ diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c index c312adcba4f..1cb5b0a9f2a 100644 --- a/fs/cifs/cifsacl.c +++ b/fs/cifs/cifsacl.c @@ -1,7 +1,7 @@ /* * fs/cifs/cifsacl.c * - * Copyright (C) International Business Machines Corp., 2007 + * Copyright (C) International Business Machines Corp., 2007,2008 * Author(s): Steve French (sfrench@us.ibm.com) * * Contains the routines for mapping CIFS/NTFS ACLs @@ -46,8 +46,7 @@ static struct cifs_wksid wksidarr[NUM_WK_SIDS] = { static const struct cifs_sid sid_everyone = { 1, 1, {0, 0, 0, 0, 0, 1}, {0} }; /* group users */ -static const struct cifs_sid sid_user = - {1, 2 , {0, 0, 0, 0, 0, 5}, {} }; +static const struct cifs_sid sid_user = {1, 2 , {0, 0, 0, 0, 0, 5}, {} }; int match_sid(struct cifs_sid *ctsid) @@ -129,6 +128,54 @@ int compare_sids(const struct cifs_sid *ctsid, const struct cifs_sid *cwsid) return (1); /* sids compare/match */ } + +/* copy ntsd, owner sid, and group sid from a security descriptor to another */ +static void copy_sec_desc(const struct cifs_ntsd *pntsd, + struct cifs_ntsd *pnntsd, __u32 sidsoffset) +{ + int i; + + struct cifs_sid *owner_sid_ptr, *group_sid_ptr; + struct cifs_sid *nowner_sid_ptr, *ngroup_sid_ptr; + + /* copy security descriptor control portion */ + pnntsd->revision = pntsd->revision; + pnntsd->type = pntsd->type; + pnntsd->dacloffset = cpu_to_le32(sizeof(struct cifs_ntsd)); + pnntsd->sacloffset = 0; + pnntsd->osidoffset = cpu_to_le32(sidsoffset); + pnntsd->gsidoffset = cpu_to_le32(sidsoffset + sizeof(struct cifs_sid)); + + /* copy owner sid */ + owner_sid_ptr = (struct cifs_sid *)((char *)pntsd + + le32_to_cpu(pntsd->osidoffset)); + nowner_sid_ptr = (struct cifs_sid *)((char *)pnntsd + sidsoffset); + + nowner_sid_ptr->revision = owner_sid_ptr->revision; + nowner_sid_ptr->num_subauth = owner_sid_ptr->num_subauth; + for (i = 0; i < 6; i++) + nowner_sid_ptr->authority[i] = owner_sid_ptr->authority[i]; + for (i = 0; i < 5; i++) + nowner_sid_ptr->sub_auth[i] = owner_sid_ptr->sub_auth[i]; + + /* copy group sid */ + group_sid_ptr = (struct cifs_sid *)((char *)pntsd + + le32_to_cpu(pntsd->gsidoffset)); + ngroup_sid_ptr = (struct cifs_sid *)((char *)pnntsd + sidsoffset + + sizeof(struct cifs_sid)); + + ngroup_sid_ptr->revision = group_sid_ptr->revision; + ngroup_sid_ptr->num_subauth = group_sid_ptr->num_subauth; + for (i = 0; i < 6; i++) + ngroup_sid_ptr->authority[i] = group_sid_ptr->authority[i]; + for (i = 0; i < 5; i++) + ngroup_sid_ptr->sub_auth[i] = + cpu_to_le32(group_sid_ptr->sub_auth[i]); + + return; +} + + /* change posix mode to reflect permissions pmode is the existing mode (we only want to overwrite part of this @@ -147,9 +194,9 @@ static void access_flags_to_mode(__le32 ace_flags, int type, umode_t *pmode, /* For deny ACEs we change the mask so that subsequent allow access control entries do not turn on the bits we are denying */ if (type == ACCESS_DENIED) { - if (flags & GENERIC_ALL) { + if (flags & GENERIC_ALL) *pbits_to_set &= ~S_IRWXUGO; - } + if ((flags & GENERIC_WRITE) || ((flags & FILE_WRITE_RIGHTS) == FILE_WRITE_RIGHTS)) *pbits_to_set &= ~S_IWUGO; @@ -168,9 +215,7 @@ static void access_flags_to_mode(__le32 ace_flags, int type, umode_t *pmode, if (flags & GENERIC_ALL) { *pmode |= (S_IRWXUGO & (*pbits_to_set)); -#ifdef CONFIG_CIFS_DEBUG2 - cFYI(1, ("all perms")); -#endif + cFYI(DBG2, ("all perms")); return; } if ((flags & GENERIC_WRITE) || @@ -183,9 +228,7 @@ static void access_flags_to_mode(__le32 ace_flags, int type, umode_t *pmode, ((flags & FILE_EXEC_RIGHTS) == FILE_EXEC_RIGHTS)) *pmode |= (S_IXUGO & (*pbits_to_set)); -#ifdef CONFIG_CIFS_DEBUG2 - cFYI(1, ("access flags 0x%x mode now 0x%x", flags, *pmode)); -#endif + cFYI(DBG2, ("access flags 0x%x mode now 0x%x", flags, *pmode)); return; } @@ -214,12 +257,37 @@ static void mode_to_access_flags(umode_t mode, umode_t bits_to_use, if (mode & S_IXUGO) *pace_flags |= SET_FILE_EXEC_RIGHTS; -#ifdef CONFIG_CIFS_DEBUG2 - cFYI(1, ("mode: 0x%x, access flags now 0x%x", mode, *pace_flags)); -#endif + cFYI(DBG2, ("mode: 0x%x, access flags now 0x%x", mode, *pace_flags)); return; } +static __u16 fill_ace_for_sid(struct cifs_ace *pntace, + const struct cifs_sid *psid, __u64 nmode, umode_t bits) +{ + int i; + __u16 size = 0; + __u32 access_req = 0; + + pntace->type = ACCESS_ALLOWED; + pntace->flags = 0x0; + mode_to_access_flags(nmode, bits, &access_req); + if (!access_req) + access_req = SET_MINIMUM_RIGHTS; + pntace->access_req = cpu_to_le32(access_req); + + pntace->sid.revision = psid->revision; + pntace->sid.num_subauth = psid->num_subauth; + for (i = 0; i < 6; i++) + pntace->sid.authority[i] = psid->authority[i]; + for (i = 0; i < psid->num_subauth; i++) + pntace->sid.sub_auth[i] = psid->sub_auth[i]; + + size = 1 + 1 + 2 + 4 + 1 + 1 + 6 + (psid->num_subauth * 4); + pntace->size = cpu_to_le16(size); + + return (size); +} + #ifdef CONFIG_CIFS_DEBUG2 static void dump_ace(struct cifs_ace *pace, char *end_of_acl) @@ -243,7 +311,7 @@ static void dump_ace(struct cifs_ace *pace, char *end_of_acl) int i; cFYI(1, ("ACE revision %d num_auth %d type %d flags %d size %d", pace->sid.revision, pace->sid.num_subauth, pace->type, - pace->flags, pace->size)); + pace->flags, le16_to_cpu(pace->size))); for (i = 0; i < num_subauth; ++i) { cFYI(1, ("ACE sub_auth[%d]: 0x%x", i, le32_to_cpu(pace->sid.sub_auth[i]))); @@ -283,11 +351,9 @@ static void parse_dacl(struct cifs_acl *pdacl, char *end_of_acl, return; } -#ifdef CONFIG_CIFS_DEBUG2 - cFYI(1, ("DACL revision %d size %d num aces %d", + cFYI(DBG2, ("DACL revision %d size %d num aces %d", le16_to_cpu(pdacl->revision), le16_to_cpu(pdacl->size), le32_to_cpu(pdacl->num_aces))); -#endif /* reset rwx permissions for user/group/other. Also, if num_aces is 0 i.e. DACL has no ACEs, @@ -306,10 +372,6 @@ static void parse_dacl(struct cifs_acl *pdacl, char *end_of_acl, ppace = kmalloc(num_aces * sizeof(struct cifs_ace *), GFP_KERNEL); -/* cifscred->cecount = pdacl->num_aces; - cifscred->aces = kmalloc(num_aces * - sizeof(struct cifs_ace *), GFP_KERNEL);*/ - for (i = 0; i < num_aces; ++i) { ppace[i] = (struct cifs_ace *) (acl_base + acl_size); #ifdef CONFIG_CIFS_DEBUG2 @@ -346,6 +408,28 @@ static void parse_dacl(struct cifs_acl *pdacl, char *end_of_acl, } +static int set_chmod_dacl(struct cifs_acl *pndacl, struct cifs_sid *pownersid, + struct cifs_sid *pgrpsid, __u64 nmode) +{ + u16 size = 0; + struct cifs_acl *pnndacl; + + pnndacl = (struct cifs_acl *)((char *)pndacl + sizeof(struct cifs_acl)); + + size += fill_ace_for_sid((struct cifs_ace *) ((char *)pnndacl + size), + pownersid, nmode, S_IRWXU); + size += fill_ace_for_sid((struct cifs_ace *)((char *)pnndacl + size), + pgrpsid, nmode, S_IRWXG); + size += fill_ace_for_sid((struct cifs_ace *)((char *)pnndacl + size), + &sid_everyone, nmode, S_IRWXO); + + pndacl->size = cpu_to_le16(size + sizeof(struct cifs_acl)); + pndacl->num_aces = cpu_to_le32(3); + + return (0); +} + + static int parse_sid(struct cifs_sid *psid, char *end_of_acl) { /* BB need to add parm so we can store the SID BB */ @@ -398,13 +482,11 @@ static int parse_sec_desc(struct cifs_ntsd *pntsd, int acl_len, le32_to_cpu(pntsd->gsidoffset)); dacloffset = le32_to_cpu(pntsd->dacloffset); dacl_ptr = (struct cifs_acl *)((char *)pntsd + dacloffset); -#ifdef CONFIG_CIFS_DEBUG2 - cFYI(1, ("revision %d type 0x%x ooffset 0x%x goffset 0x%x " + cFYI(DBG2, ("revision %d type 0x%x ooffset 0x%x goffset 0x%x " "sacloffset 0x%x dacloffset 0x%x", pntsd->revision, pntsd->type, le32_to_cpu(pntsd->osidoffset), le32_to_cpu(pntsd->gsidoffset), le32_to_cpu(pntsd->sacloffset), dacloffset)); -#endif /* cifs_dump_mem("owner_sid: ", owner_sid_ptr, 64); */ rc = parse_sid(owner_sid_ptr, end_of_acl); if (rc) @@ -432,11 +514,51 @@ static int parse_sec_desc(struct cifs_ntsd *pntsd, int acl_len, } +/* Convert permission bits from mode to equivalent CIFS ACL */ +static int build_sec_desc(struct cifs_ntsd *pntsd, struct cifs_ntsd *pnntsd, + int acl_len, struct inode *inode, __u64 nmode) +{ + int rc = 0; + __u32 dacloffset; + __u32 ndacloffset; + __u32 sidsoffset; + struct cifs_sid *owner_sid_ptr, *group_sid_ptr; + struct cifs_acl *dacl_ptr = NULL; /* no need for SACL ptr */ + struct cifs_acl *ndacl_ptr = NULL; /* no need for SACL ptr */ + + if ((inode == NULL) || (pntsd == NULL) || (pnntsd == NULL)) + return (-EIO); + + owner_sid_ptr = (struct cifs_sid *)((char *)pntsd + + le32_to_cpu(pntsd->osidoffset)); + group_sid_ptr = (struct cifs_sid *)((char *)pntsd + + le32_to_cpu(pntsd->gsidoffset)); + + dacloffset = le32_to_cpu(pntsd->dacloffset); + dacl_ptr = (struct cifs_acl *)((char *)pntsd + dacloffset); + + ndacloffset = sizeof(struct cifs_ntsd); + ndacl_ptr = (struct cifs_acl *)((char *)pnntsd + ndacloffset); + ndacl_ptr->revision = dacl_ptr->revision; + ndacl_ptr->size = 0; + ndacl_ptr->num_aces = 0; + + rc = set_chmod_dacl(ndacl_ptr, owner_sid_ptr, group_sid_ptr, nmode); + + sidsoffset = ndacloffset + le16_to_cpu(ndacl_ptr->size); + + /* copy security descriptor control portion and owner and group sid */ + copy_sec_desc(pntsd, pnntsd, sidsoffset); + + return (rc); +} + + /* Retrieve an ACL from the server */ static struct cifs_ntsd *get_cifs_acl(u32 *pacllen, struct inode *inode, - const char *path) + const char *path, const __u16 *pfid) { - struct cifsFileInfo *open_file; + struct cifsFileInfo *open_file = NULL; int unlock_file = FALSE; int xid; int rc = -EIO; @@ -451,7 +573,11 @@ static struct cifs_ntsd *get_cifs_acl(u32 *pacllen, struct inode *inode, return NULL; xid = GetXid(); - open_file = find_readable_file(CIFS_I(inode)); + if (pfid == NULL) + open_file = find_readable_file(CIFS_I(inode)); + else + fid = *pfid; + sb = inode->i_sb; if (sb == NULL) { FreeXid(xid); @@ -462,7 +588,7 @@ static struct cifs_ntsd *get_cifs_acl(u32 *pacllen, struct inode *inode, if (open_file) { unlock_file = TRUE; fid = open_file->netfid; - } else { + } else if (pfid == NULL) { int oplock = FALSE; /* open file */ rc = CIFSSMBOpen(xid, cifs_sb->tcon, path, FILE_OPEN, @@ -478,26 +604,79 @@ static struct cifs_ntsd *get_cifs_acl(u32 *pacllen, struct inode *inode, rc = CIFSSMBGetCIFSACL(xid, cifs_sb->tcon, fid, &pntsd, pacllen); cFYI(1, ("GetCIFSACL rc = %d ACL len %d", rc, *pacllen)); + if (unlock_file == TRUE) /* find_readable_file increments ref count */ + atomic_dec(&open_file->wrtPending); + else if (pfid == NULL) /* if opened above we have to close the handle */ + CIFSSMBClose(xid, cifs_sb->tcon, fid); + /* else handle was passed in by caller */ + + FreeXid(xid); + return pntsd; +} + +/* Set an ACL on the server */ +static int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen, + struct inode *inode, const char *path) +{ + struct cifsFileInfo *open_file; + int unlock_file = FALSE; + int xid; + int rc = -EIO; + __u16 fid; + struct super_block *sb; + struct cifs_sb_info *cifs_sb; + + cFYI(DBG2, ("set ACL for %s from mode 0x%x", path, inode->i_mode)); + + if (!inode) + return (rc); + + sb = inode->i_sb; + if (sb == NULL) + return (rc); + + cifs_sb = CIFS_SB(sb); + xid = GetXid(); + + open_file = find_readable_file(CIFS_I(inode)); + if (open_file) { + unlock_file = TRUE; + fid = open_file->netfid; + } else { + int oplock = FALSE; + /* open file */ + rc = CIFSSMBOpen(xid, cifs_sb->tcon, path, FILE_OPEN, + WRITE_DAC, 0, &fid, &oplock, NULL, + cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + if (rc != 0) { + cERROR(1, ("Unable to open file to set ACL")); + FreeXid(xid); + return (rc); + } + } + + rc = CIFSSMBSetCIFSACL(xid, cifs_sb->tcon, fid, pnntsd, acllen); + cFYI(DBG2, ("SetCIFSACL rc = %d", rc)); if (unlock_file == TRUE) atomic_dec(&open_file->wrtPending); else CIFSSMBClose(xid, cifs_sb->tcon, fid); FreeXid(xid); - return pntsd; + + return (rc); } /* Translate the CIFS ACL (simlar to NTFS ACL) for a file into mode bits */ -void acl_to_uid_mode(struct inode *inode, const char *path) +void acl_to_uid_mode(struct inode *inode, const char *path, const __u16 *pfid) { struct cifs_ntsd *pntsd = NULL; u32 acllen = 0; int rc = 0; -#ifdef CONFIG_CIFS_DEBUG2 - cFYI(1, ("converting ACL to mode for %s", path)); -#endif - pntsd = get_cifs_acl(&acllen, inode, path); + cFYI(DBG2, ("converting ACL to mode for %s", path)); + pntsd = get_cifs_acl(&acllen, inode, path, pfid); /* if we can retrieve the ACL, now parse Access Control Entries, ACEs */ if (pntsd) @@ -510,24 +689,47 @@ void acl_to_uid_mode(struct inode *inode, const char *path) } /* Convert mode bits to an ACL so we can update the ACL on the server */ -int mode_to_acl(struct inode *inode, const char *path) +int mode_to_acl(struct inode *inode, const char *path, __u64 nmode) { int rc = 0; __u32 acllen = 0; - struct cifs_ntsd *pntsd = NULL; + struct cifs_ntsd *pntsd = NULL; /* acl obtained from server */ + struct cifs_ntsd *pnntsd = NULL; /* modified acl to be sent to server */ - cFYI(1, ("set ACL from mode for %s", path)); + cFYI(DBG2, ("set ACL from mode for %s", path)); /* Get the security descriptor */ - pntsd = get_cifs_acl(&acllen, inode, path); + pntsd = get_cifs_acl(&acllen, inode, path, NULL); - /* Add/Modify the three ACEs for owner, group, everyone - while retaining the other ACEs */ + /* Add three ACEs for owner, group, everyone getting rid of + other ACEs as chmod disables ACEs and set the security descriptor */ - /* Set the security descriptor */ + if (pntsd) { + /* allocate memory for the smb header, + set security descriptor request security descriptor + parameters, and secuirty descriptor itself */ + pnntsd = kmalloc(acllen, GFP_KERNEL); + if (!pnntsd) { + cERROR(1, ("Unable to allocate security descriptor")); + kfree(pntsd); + return (-ENOMEM); + } - kfree(pntsd); - return rc; + rc = build_sec_desc(pntsd, pnntsd, acllen, inode, nmode); + + cFYI(DBG2, ("build_sec_desc rc: %d", rc)); + + if (!rc) { + /* Set the security descriptor */ + rc = set_cifs_acl(pnntsd, acllen, inode, path); + cFYI(DBG2, ("set_cifs_acl rc: %d", rc)); + } + + kfree(pnntsd); + kfree(pntsd); + } + + return (rc); } #endif /* CONFIG_CIFS_EXPERIMENTAL */ diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 093beaa3900..a04b17e5a9d 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -44,6 +44,7 @@ #include "cifs_fs_sb.h" #include <linux/mm.h> #include <linux/key-type.h> +#include "dns_resolve.h" #include "cifs_spnego.h" #define CIFS_MAGIC_NUMBER 0xFF534D42 /* the first four bytes of SMB PDUs */ @@ -96,6 +97,9 @@ cifs_read_super(struct super_block *sb, void *data, { struct inode *inode; struct cifs_sb_info *cifs_sb; +#ifdef CONFIG_CIFS_DFS_UPCALL + int len; +#endif int rc = 0; /* BB should we make this contingent on mount parm? */ @@ -105,6 +109,25 @@ cifs_read_super(struct super_block *sb, void *data, if (cifs_sb == NULL) return -ENOMEM; +#ifdef CONFIG_CIFS_DFS_UPCALL + /* copy mount params to sb for use in submounts */ + /* BB: should we move this after the mount so we + * do not have to do the copy on failed mounts? + * BB: May be it is better to do simple copy before + * complex operation (mount), and in case of fail + * just exit instead of doing mount and attempting + * undo it if this copy fails?*/ + len = strlen(data); + cifs_sb->mountdata = kzalloc(len + 1, GFP_KERNEL); + if (cifs_sb->mountdata == NULL) { + kfree(sb->s_fs_info); + sb->s_fs_info = NULL; + return -ENOMEM; + } + strncpy(cifs_sb->mountdata, data, len + 1); + cifs_sb->mountdata[len] = '\0'; +#endif + rc = cifs_mount(sb, cifs_sb, data, devname); if (rc) { @@ -124,10 +147,11 @@ cifs_read_super(struct super_block *sb, void *data, #endif sb->s_blocksize = CIFS_MAX_MSGSIZE; sb->s_blocksize_bits = 14; /* default 2**14 = CIFS_MAX_MSGSIZE */ - inode = iget(sb, ROOT_I); + inode = cifs_iget(sb, ROOT_I); - if (!inode) { - rc = -ENOMEM; + if (IS_ERR(inode)) { + rc = PTR_ERR(inode); + inode = NULL; goto out_no_root; } @@ -154,6 +178,12 @@ out_no_root: out_mount_failed: if (cifs_sb) { +#ifdef CONFIG_CIFS_DFS_UPCALL + if (cifs_sb->mountdata) { + kfree(cifs_sb->mountdata); + cifs_sb->mountdata = NULL; + } +#endif if (cifs_sb->local_nls) unload_nls(cifs_sb->local_nls); kfree(cifs_sb); @@ -174,9 +204,15 @@ cifs_put_super(struct super_block *sb) return; } rc = cifs_umount(sb, cifs_sb); - if (rc) { + if (rc) cERROR(1, ("cifs_umount failed with return code %d", rc)); +#ifdef CONFIG_CIFS_DFS_UPCALL + if (cifs_sb->mountdata) { + kfree(cifs_sb->mountdata); + cifs_sb->mountdata = NULL; } +#endif + unload_nls(cifs_sb->local_nls); kfree(cifs_sb); return; @@ -424,7 +460,7 @@ int cifs_xstate_get(struct super_block *sb, struct fs_quota_stat *qstats) static struct quotactl_ops cifs_quotactl_ops = { .set_xquota = cifs_xquota_set, - .get_xquota = cifs_xquota_set, + .get_xquota = cifs_xquota_get, .set_xstate = cifs_xstate_set, .get_xstate = cifs_xstate_get, }; @@ -435,6 +471,8 @@ static void cifs_umount_begin(struct vfsmount *vfsmnt, int flags) struct cifs_sb_info *cifs_sb; struct cifsTconInfo *tcon; + dfs_shrink_umount_helper(vfsmnt); + if (!(flags & MNT_FORCE)) return; cifs_sb = CIFS_SB(vfsmnt->mnt_sb); @@ -480,7 +518,6 @@ static int cifs_remount(struct super_block *sb, int *flags, char *data) } static const struct super_operations cifs_super_ops = { - .read_inode = cifs_read_inode, .put_super = cifs_put_super, .statfs = cifs_statfs, .alloc_inode = cifs_alloc_inode, @@ -552,7 +589,7 @@ static loff_t cifs_llseek(struct file *file, loff_t offset, int origin) return remote_llseek(file, offset, origin); } -static struct file_system_type cifs_fs_type = { +struct file_system_type cifs_fs_type = { .owner = THIS_MODULE, .name = "cifs", .get_sb = cifs_get_sb, @@ -952,9 +989,7 @@ static int __init init_cifs(void) { int rc = 0; -#ifdef CONFIG_PROC_FS cifs_proc_init(); -#endif /* INIT_LIST_HEAD(&GlobalServerList);*/ /* BB not implemented yet */ INIT_LIST_HEAD(&GlobalSMBSessionList); INIT_LIST_HEAD(&GlobalTreeConnectionList); @@ -1015,11 +1050,16 @@ init_cifs(void) if (rc) goto out_unregister_filesystem; #endif +#ifdef CONFIG_CIFS_DFS_UPCALL + rc = register_key_type(&key_type_dns_resolver); + if (rc) + goto out_unregister_key_type; +#endif oplockThread = kthread_run(cifs_oplock_thread, NULL, "cifsoplockd"); if (IS_ERR(oplockThread)) { rc = PTR_ERR(oplockThread); cERROR(1, ("error %d create oplock thread", rc)); - goto out_unregister_key_type; + goto out_unregister_dfs_key_type; } dnotifyThread = kthread_run(cifs_dnotify_thread, NULL, "cifsdnotifyd"); @@ -1033,7 +1073,11 @@ init_cifs(void) out_stop_oplock_thread: kthread_stop(oplockThread); + out_unregister_dfs_key_type: +#ifdef CONFIG_CIFS_DFS_UPCALL + unregister_key_type(&key_type_dns_resolver); out_unregister_key_type: +#endif #ifdef CONFIG_CIFS_UPCALL unregister_key_type(&cifs_spnego_key_type); out_unregister_filesystem: @@ -1046,18 +1090,17 @@ init_cifs(void) out_destroy_inodecache: cifs_destroy_inodecache(); out_clean_proc: -#ifdef CONFIG_PROC_FS cifs_proc_clean(); -#endif return rc; } static void __exit exit_cifs(void) { - cFYI(0, ("exit_cifs")); -#ifdef CONFIG_PROC_FS + cFYI(DBG2, ("exit_cifs")); cifs_proc_clean(); +#ifdef CONFIG_CIFS_DFS_UPCALL + unregister_key_type(&key_type_dns_resolver); #endif #ifdef CONFIG_CIFS_UPCALL unregister_key_type(&cifs_spnego_key_type); diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 2a21dc66f0d..68978306c3c 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -32,6 +32,7 @@ #define TRUE 1 #endif +extern struct file_system_type cifs_fs_type; extern const struct address_space_operations cifs_addr_ops; extern const struct address_space_operations cifs_addr_ops_smallbuf; @@ -43,6 +44,7 @@ extern void cifs_read_inode(struct inode *); /* Functions related to inodes */ extern const struct inode_operations cifs_dir_inode_ops; +extern struct inode *cifs_iget(struct super_block *, unsigned long); extern int cifs_create(struct inode *, struct dentry *, int, struct nameidata *); extern struct dentry *cifs_lookup(struct inode *, struct dentry *, @@ -60,6 +62,10 @@ extern int cifs_setattr(struct dentry *, struct iattr *); extern const struct inode_operations cifs_file_inode_ops; extern const struct inode_operations cifs_symlink_inode_ops; +extern struct list_head cifs_dfs_automount_list; +extern struct inode_operations cifs_dfs_referral_inode_operations; + + /* Functions related to files and directories */ extern const struct file_operations cifs_file_ops; diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 1fde2197ad7..69a2e194254 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -1,7 +1,7 @@ /* * fs/cifs/cifsglob.h * - * Copyright (C) International Business Machines Corp., 2002,2007 + * Copyright (C) International Business Machines Corp., 2002,2008 * Author(s): Steve French (sfrench@us.ibm.com) * Jeremy Allison (jra@samba.org) * @@ -70,14 +70,6 @@ #endif /* - * This information is kept on every Server we know about. - * - * Some things to note: - * - */ -#define SERVER_NAME_LEN_WITH_NULL (SERVER_NAME_LENGTH + 1) - -/* * CIFS vfs client Status information (based on what we know.) */ @@ -460,6 +452,37 @@ struct dir_notify_req { struct file *pfile; }; +struct dfs_info3_param { + int flags; /* DFSREF_REFERRAL_SERVER, DFSREF_STORAGE_SERVER*/ + int path_consumed; + int server_type; + int ref_flag; + char *path_name; + char *node_name; +}; + +static inline void free_dfs_info_param(struct dfs_info3_param *param) +{ + if (param) { + kfree(param->path_name); + kfree(param->node_name); + kfree(param); + } +} + +static inline void free_dfs_info_array(struct dfs_info3_param *param, + int number_of_items) +{ + int i; + if ((number_of_items == 0) || (param == NULL)) + return; + for (i = 0; i < number_of_items; i++) { + kfree(param[i].path_name); + kfree(param[i].node_name); + } + kfree(param); +} + #define MID_FREE 0 #define MID_REQUEST_ALLOCATED 1 #define MID_REQUEST_SUBMITTED 2 diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index dbe6b846f37..47f79504f57 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -237,6 +237,9 @@ | DELETE | READ_CONTROL | WRITE_DAC \ | WRITE_OWNER | SYNCHRONIZE) +#define SET_MINIMUM_RIGHTS (FILE_READ_EA | FILE_READ_ATTRIBUTES \ + | READ_CONTROL | SYNCHRONIZE) + /* * Invalid readdir handle diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 8350eec4966..7e5e0e78cd7 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -1,7 +1,7 @@ /* * fs/cifs/cifsproto.h * - * Copyright (c) International Business Machines Corp., 2002,2007 + * Copyright (c) International Business Machines Corp., 2002,2008 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify @@ -39,8 +39,8 @@ extern int smb_send(struct socket *, struct smb_hdr *, unsigned int /* length */ , struct sockaddr *); extern unsigned int _GetXid(void); extern void _FreeXid(unsigned int); -#define GetXid() (int)_GetXid(); cFYI(1,("CIFS VFS: in %s as Xid: %d with uid: %d",__FUNCTION__, xid,current->fsuid)); -#define FreeXid(curr_xid) {_FreeXid(curr_xid); cFYI(1,("CIFS VFS: leaving %s (xid = %d) rc = %d",__FUNCTION__,curr_xid,(int)rc));} +#define GetXid() (int)_GetXid(); cFYI(1,("CIFS VFS: in %s as Xid: %d with uid: %d",__func__, xid,current->fsuid)); +#define FreeXid(curr_xid) {_FreeXid(curr_xid); cFYI(1,("CIFS VFS: leaving %s (xid = %d) rc = %d",__func__,curr_xid,(int)rc));} extern char *build_path_from_dentry(struct dentry *); extern char *build_wildcard_path_from_dentry(struct dentry *direntry); /* extern void renew_parental_timestamps(struct dentry *direntry);*/ @@ -53,11 +53,11 @@ extern int SendReceiveNoRsp(const unsigned int xid, struct cifsSesInfo *ses, extern int SendReceive2(const unsigned int /* xid */ , struct cifsSesInfo *, struct kvec *, int /* nvec to send */, int * /* type of buf returned */ , const int flags); -extern int SendReceiveBlockingLock(const unsigned int /* xid */ , - struct cifsTconInfo *, - struct smb_hdr * /* input */ , - struct smb_hdr * /* out */ , - int * /* bytes returned */); +extern int SendReceiveBlockingLock(const unsigned int xid, + struct cifsTconInfo *ptcon, + struct smb_hdr *in_buf , + struct smb_hdr *out_buf, + int *bytes_returned); extern int checkSMB(struct smb_hdr *smb, __u16 mid, unsigned int length); extern int is_valid_oplock_break(struct smb_hdr *smb, struct TCP_Server_Info *); extern int is_size_safe_to_change(struct cifsInodeInfo *, __u64 eof); @@ -84,7 +84,7 @@ extern __u16 GetNextMid(struct TCP_Server_Info *server); extern struct oplock_q_entry *AllocOplockQEntry(struct inode *, u16, struct cifsTconInfo *); extern void DeleteOplockQEntry(struct oplock_q_entry *); -extern struct timespec cifs_NTtimeToUnix(u64 /* utc nanoseconds since 1601 */ ); +extern struct timespec cifs_NTtimeToUnix(u64 utc_nanoseconds_since_1601); extern u64 cifs_UnixTimeToNT(struct timespec); extern __le64 cnvrtDosCifsTm(__u16 date, __u16 time); extern struct timespec cnvrtDosUnixTm(__u16 date, __u16 time); @@ -92,16 +92,24 @@ extern struct timespec cnvrtDosUnixTm(__u16 date, __u16 time); extern int cifs_get_inode_info(struct inode **pinode, const unsigned char *search_path, FILE_ALL_INFO * pfile_info, - struct super_block *sb, int xid); + struct super_block *sb, int xid, const __u16 *pfid); extern int cifs_get_inode_info_unix(struct inode **pinode, const unsigned char *search_path, struct super_block *sb, int xid); -extern void acl_to_uid_mode(struct inode *inode, const char *search_path); -extern int mode_to_acl(struct inode *inode, const char *path); +extern void acl_to_uid_mode(struct inode *inode, const char *path, + const __u16 *pfid); +extern int mode_to_acl(struct inode *inode, const char *path, __u64); extern int cifs_mount(struct super_block *, struct cifs_sb_info *, char *, const char *); extern int cifs_umount(struct super_block *, struct cifs_sb_info *); +#ifdef CONFIG_CIFS_DFS_UPCALL +extern void dfs_shrink_umount_helper(struct vfsmount *vfsmnt); +#else +static inline void dfs_shrink_umount_helper(struct vfsmount *vfsmnt) +{ +} +#endif /* DFS_UPCALL */ void cifs_proc_init(void); void cifs_proc_clean(void); @@ -153,7 +161,7 @@ extern int get_dfs_path(int xid, struct cifsSesInfo *pSesInfo, const char *old_path, const struct nls_table *nls_codepage, unsigned int *pnum_referrals, - unsigned char **preferrals, + struct dfs_info3_param **preferrals, int remap); extern void reset_cifs_unix_caps(int xid, struct cifsTconInfo *tcon, struct super_block *sb, struct smb_vol *vol); @@ -172,11 +180,11 @@ extern int CIFSSMBQFSPosixInfo(const int xid, struct cifsTconInfo *tcon, struct kstatfs *FSData); extern int CIFSSMBSetTimes(const int xid, struct cifsTconInfo *tcon, - const char *fileName, const FILE_BASIC_INFO * data, + const char *fileName, const FILE_BASIC_INFO *data, const struct nls_table *nls_codepage, int remap_special_chars); extern int CIFSSMBSetFileTimes(const int xid, struct cifsTconInfo *tcon, - const FILE_BASIC_INFO * data, __u16 fid); + const FILE_BASIC_INFO *data, __u16 fid); #if 0 extern int CIFSSMBSetAttrLegacy(int xid, struct cifsTconInfo *tcon, char *fileName, __u16 dos_attributes, @@ -342,6 +350,8 @@ extern int CIFSSMBSetEA(const int xid, struct cifsTconInfo *tcon, const struct nls_table *nls_codepage, int remap_special_chars); extern int CIFSSMBGetCIFSACL(const int xid, struct cifsTconInfo *tcon, __u16 fid, struct cifs_ntsd **acl_inf, __u32 *buflen); +extern int CIFSSMBSetCIFSACL(const int, struct cifsTconInfo *, __u16, + struct cifs_ntsd *, __u32); extern int CIFSSMBGetPosixACL(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, char *acl_inf, const int buflen, const int acl_type, diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 9e8a6bef029..30bbe448e26 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -1,7 +1,7 @@ /* * fs/cifs/cifssmb.c * - * Copyright (C) International Business Machines Corp., 2002,2007 + * Copyright (C) International Business Machines Corp., 2002,2008 * Author(s): Steve French (sfrench@us.ibm.com) * * Contains the routines for constructing the SMB PDUs themselves @@ -102,10 +102,12 @@ static void mark_open_files_invalid(struct cifsTconInfo *pTcon) to this tcon */ } -/* If the return code is zero, this function must fill in request_buf pointer */ +/* Allocate and return pointer to an SMB request buffer, and set basic + SMB information in the SMB header. If the return code is zero, this + function must have filled in request_buf pointer */ static int small_smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, - void **request_buf /* returned */) + void **request_buf) { int rc = 0; @@ -363,7 +365,7 @@ smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, *response_buf = *request_buf; header_assemble((struct smb_hdr *) *request_buf, smb_command, tcon, - wct /*wct */ ); + wct); if (tcon != NULL) cifs_stats_inc(&tcon->num_smbs_sent); @@ -523,7 +525,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) if (remain >= (MIN_TZ_ADJ / 2)) result += MIN_TZ_ADJ; if (val < 0) - result = - result; + result = -result; server->timeAdj = result; } else { server->timeAdj = (int)tmp; @@ -600,7 +602,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) server->maxBuf = min(le32_to_cpu(pSMBr->MaxBufferSize), (__u32) CIFSMaxBufSize + MAX_CIFS_HDR_SIZE); server->maxRw = le32_to_cpu(pSMBr->MaxRawSize); - cFYI(0, ("Max buf = %d", ses->server->maxBuf)); + cFYI(DBG2, ("Max buf = %d", ses->server->maxBuf)); GETU32(ses->server->sessid) = le32_to_cpu(pSMBr->SessionKey); server->capabilities = le32_to_cpu(pSMBr->Capabilities); server->timeAdj = (int)(__s16)le16_to_cpu(pSMBr->ServerTimeZone); @@ -868,9 +870,8 @@ PsxDelete: pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); - if (rc) { + if (rc) cFYI(1, ("Posix delete returned %d", rc)); - } cifs_buf_release(pSMB); cifs_stats_inc(&tcon->num_deletes); @@ -916,9 +917,8 @@ DelFileRetry: rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); cifs_stats_inc(&tcon->num_deletes); - if (rc) { + if (rc) cFYI(1, ("Error in RMFile = %d", rc)); - } cifs_buf_release(pSMB); if (rc == -EAGAIN) @@ -961,9 +961,8 @@ RmDirRetry: rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); cifs_stats_inc(&tcon->num_rmdirs); - if (rc) { + if (rc) cFYI(1, ("Error in RMDir = %d", rc)); - } cifs_buf_release(pSMB); if (rc == -EAGAIN) @@ -1005,9 +1004,8 @@ MkDirRetry: rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); cifs_stats_inc(&tcon->num_mkdirs); - if (rc) { + if (rc) cFYI(1, ("Error in Mkdir = %d", rc)); - } cifs_buf_release(pSMB); if (rc == -EAGAIN) @@ -1017,7 +1015,7 @@ MkDirRetry: int CIFSPOSIXCreate(const int xid, struct cifsTconInfo *tcon, __u32 posix_flags, - __u64 mode, __u16 * netfid, FILE_UNIX_BASIC_INFO *pRetData, + __u64 mode, __u16 *netfid, FILE_UNIX_BASIC_INFO *pRetData, __u32 *pOplock, const char *name, const struct nls_table *nls_codepage, int remap) { @@ -1027,8 +1025,8 @@ CIFSPOSIXCreate(const int xid, struct cifsTconInfo *tcon, __u32 posix_flags, int rc = 0; int bytes_returned = 0; __u16 params, param_offset, offset, byte_count, count; - OPEN_PSX_REQ * pdata; - OPEN_PSX_RSP * psx_rsp; + OPEN_PSX_REQ *pdata; + OPEN_PSX_RSP *psx_rsp; cFYI(1, ("In POSIX Create")); PsxCreat: @@ -1110,9 +1108,7 @@ PsxCreat: /* check to make sure response data is there */ if (psx_rsp->ReturnedLevel != cpu_to_le16(SMB_QUERY_FILE_UNIX_BASIC)) { pRetData->Type = cpu_to_le32(-1); /* unknown */ -#ifdef CONFIG_CIFS_DEBUG2 - cFYI(1, ("unknown type")); -#endif + cFYI(DBG2, ("unknown type")); } else { if (pSMBr->ByteCount < sizeof(OPEN_PSX_RSP) + sizeof(FILE_UNIX_BASIC_INFO)) { @@ -1169,8 +1165,8 @@ static __u16 convert_disposition(int disposition) int SMBLegacyOpen(const int xid, struct cifsTconInfo *tcon, const char *fileName, const int openDisposition, - const int access_flags, const int create_options, __u16 * netfid, - int *pOplock, FILE_ALL_INFO * pfile_info, + const int access_flags, const int create_options, __u16 *netfid, + int *pOplock, FILE_ALL_INFO *pfile_info, const struct nls_table *nls_codepage, int remap) { int rc = -EACCES; @@ -1221,8 +1217,8 @@ OldOpenRetry: if (create_options & CREATE_OPTION_SPECIAL) pSMB->FileAttributes = cpu_to_le16(ATTR_SYSTEM); - else - pSMB->FileAttributes = cpu_to_le16(0/*ATTR_NORMAL*/); /* BB FIXME */ + else /* BB FIXME BB */ + pSMB->FileAttributes = cpu_to_le16(0/*ATTR_NORMAL*/); /* if ((omode & S_IWUGO) == 0) pSMB->FileAttributes |= cpu_to_le32(ATTR_READONLY);*/ @@ -1284,8 +1280,8 @@ OldOpenRetry: int CIFSSMBOpen(const int xid, struct cifsTconInfo *tcon, const char *fileName, const int openDisposition, - const int access_flags, const int create_options, __u16 * netfid, - int *pOplock, FILE_ALL_INFO * pfile_info, + const int access_flags, const int create_options, __u16 *netfid, + int *pOplock, FILE_ALL_INFO *pfile_info, const struct nls_table *nls_codepage, int remap) { int rc = -EACCES; @@ -1556,9 +1552,9 @@ CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, } /* else setting file size with write of zero bytes */ if (wct == 14) byte_count = bytes_sent + 1; /* pad */ - else /* wct == 12 */ { + else /* wct == 12 */ byte_count = bytes_sent + 5; /* bigger pad, smaller smb hdr */ - } + pSMB->DataLengthLow = cpu_to_le16(bytes_sent & 0xFFFF); pSMB->DataLengthHigh = cpu_to_le16(bytes_sent >> 16); pSMB->hdr.smb_buf_length += byte_count; @@ -1663,7 +1659,7 @@ CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, rc = -EIO; *nbytes = 0; } else { - WRITE_RSP * pSMBr = (WRITE_RSP *)iov[0].iov_base; + WRITE_RSP *pSMBr = (WRITE_RSP *)iov[0].iov_base; *nbytes = le16_to_cpu(pSMBr->CountHigh); *nbytes = (*nbytes) << 16; *nbytes += le16_to_cpu(pSMBr->Count); @@ -1744,9 +1740,8 @@ CIFSSMBLock(const int xid, struct cifsTconInfo *tcon, /* SMB buffer freed by function above */ } cifs_stats_inc(&tcon->num_locks); - if (rc) { + if (rc) cFYI(1, ("Send error in Lock = %d", rc)); - } /* Note: On -EAGAIN error only caller can retry on handle based calls since file handle passed in no longer valid */ @@ -1791,7 +1786,7 @@ CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon, count = sizeof(struct cifs_posix_lock); pSMB->MaxParameterCount = cpu_to_le16(2); - pSMB->MaxDataCount = cpu_to_le16(1000); /* BB find max SMB PDU from sess */ + pSMB->MaxDataCount = cpu_to_le16(1000); /* BB find max SMB from sess */ pSMB->SetupCount = 1; pSMB->Reserved3 = 0; if (get_flag) @@ -1972,9 +1967,8 @@ renameRetry: rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); cifs_stats_inc(&tcon->num_renames); - if (rc) { + if (rc) cFYI(1, ("Send error in rename = %d", rc)); - } cifs_buf_release(pSMB); @@ -2016,7 +2010,7 @@ int CIFSSMBRenameOpenFile(const int xid, struct cifsTconInfo *pTcon, data_offset = (char *) (&pSMB->hdr.Protocol) + offset; rename_info = (struct set_file_rename *) data_offset; pSMB->MaxParameterCount = cpu_to_le16(2); - pSMB->MaxDataCount = cpu_to_le16(1000); /* BB find max SMB PDU from sess */ + pSMB->MaxDataCount = cpu_to_le16(1000); /* BB find max SMB from sess */ pSMB->SetupCount = 1; pSMB->Reserved3 = 0; pSMB->SubCommand = cpu_to_le16(TRANS2_SET_FILE_INFORMATION); @@ -2052,9 +2046,8 @@ int CIFSSMBRenameOpenFile(const int xid, struct cifsTconInfo *pTcon, rc = SendReceive(xid, pTcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); cifs_stats_inc(&pTcon->num_t2renames); - if (rc) { + if (rc) cFYI(1, ("Send error in Rename (by file handle) = %d", rc)); - } cifs_buf_release(pSMB); @@ -2211,9 +2204,8 @@ createSymLinkRetry: rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); cifs_stats_inc(&tcon->num_symlinks); - if (rc) { + if (rc) cFYI(1, ("Send error in SetPathInfo create symlink = %d", rc)); - } if (pSMB) cifs_buf_release(pSMB); @@ -2299,9 +2291,8 @@ createHardLinkRetry: rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); cifs_stats_inc(&tcon->num_hardlinks); - if (rc) { + if (rc) cFYI(1, ("Send error in SetPathInfo (hard link) = %d", rc)); - } cifs_buf_release(pSMB); if (rc == -EAGAIN) @@ -2370,9 +2361,9 @@ winCreateHardLinkRetry: rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); cifs_stats_inc(&tcon->num_hardlinks); - if (rc) { + if (rc) cFYI(1, ("Send error in hard link (NT rename) = %d", rc)); - } + cifs_buf_release(pSMB); if (rc == -EAGAIN) goto winCreateHardLinkRetry; @@ -2968,9 +2959,8 @@ setAclRetry: pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); - if (rc) { + if (rc) cFYI(1, ("Set POSIX ACL returned %d", rc)); - } setACLerrorExit: cifs_buf_release(pSMB); @@ -2982,7 +2972,7 @@ setACLerrorExit: /* BB fix tabs in this function FIXME BB */ int CIFSGetExtAttr(const int xid, struct cifsTconInfo *tcon, - const int netfid, __u64 * pExtAttrBits, __u64 *pMask) + const int netfid, __u64 *pExtAttrBits, __u64 *pMask) { int rc = 0; struct smb_t2_qfi_req *pSMB = NULL; @@ -3000,7 +2990,7 @@ GetExtAttrRetry: if (rc) return rc; - params = 2 /* level */ +2 /* fid */; + params = 2 /* level */ + 2 /* fid */; pSMB->t2.TotalDataCount = 0; pSMB->t2.MaxParameterCount = cpu_to_le16(4); /* BB find exact max data count below from sess structure BB */ @@ -3071,7 +3061,7 @@ CIFSSMBGetCIFSACL(const int xid, struct cifsTconInfo *tcon, __u16 fid, { int rc = 0; int buf_type = 0; - QUERY_SEC_DESC_REQ * pSMB; + QUERY_SEC_DESC_REQ *pSMB; struct kvec iov[1]; cFYI(1, ("GetCifsACL")); @@ -3101,7 +3091,7 @@ CIFSSMBGetCIFSACL(const int xid, struct cifsTconInfo *tcon, __u16 fid, if (rc) { cFYI(1, ("Send error in QuerySecDesc = %d", rc)); } else { /* decode response */ - __le32 * parm; + __le32 *parm; __u32 parm_len; __u32 acl_len; struct smb_com_ntransact_rsp *pSMBr; @@ -3156,6 +3146,71 @@ qsec_out: /* cifs_small_buf_release(pSMB); */ /* Freed earlier now in SendReceive2 */ return rc; } + +int +CIFSSMBSetCIFSACL(const int xid, struct cifsTconInfo *tcon, __u16 fid, + struct cifs_ntsd *pntsd, __u32 acllen) +{ + __u16 byte_count, param_count, data_count, param_offset, data_offset; + int rc = 0; + int bytes_returned = 0; + SET_SEC_DESC_REQ *pSMB = NULL; + NTRANSACT_RSP *pSMBr = NULL; + +setCifsAclRetry: + rc = smb_init(SMB_COM_NT_TRANSACT, 19, tcon, (void **) &pSMB, + (void **) &pSMBr); + if (rc) + return (rc); + + pSMB->MaxSetupCount = 0; + pSMB->Reserved = 0; + + param_count = 8; + param_offset = offsetof(struct smb_com_transaction_ssec_req, Fid) - 4; + data_count = acllen; + data_offset = param_offset + param_count; + byte_count = 3 /* pad */ + param_count; + + pSMB->DataCount = cpu_to_le32(data_count); + pSMB->TotalDataCount = pSMB->DataCount; + pSMB->MaxParameterCount = cpu_to_le32(4); + pSMB->MaxDataCount = cpu_to_le32(16384); + pSMB->ParameterCount = cpu_to_le32(param_count); + pSMB->ParameterOffset = cpu_to_le32(param_offset); + pSMB->TotalParameterCount = pSMB->ParameterCount; + pSMB->DataOffset = cpu_to_le32(data_offset); + pSMB->SetupCount = 0; + pSMB->SubCommand = cpu_to_le16(NT_TRANSACT_SET_SECURITY_DESC); + pSMB->ByteCount = cpu_to_le16(byte_count+data_count); + + pSMB->Fid = fid; /* file handle always le */ + pSMB->Reserved2 = 0; + pSMB->AclFlags = cpu_to_le32(CIFS_ACL_DACL); + + if (pntsd && acllen) { + memcpy((char *) &pSMBr->hdr.Protocol + data_offset, + (char *) pntsd, + acllen); + pSMB->hdr.smb_buf_length += (byte_count + data_count); + + } else + pSMB->hdr.smb_buf_length += byte_count; + + rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, + (struct smb_hdr *) pSMBr, &bytes_returned, 0); + + cFYI(1, ("SetCIFSACL bytes_returned: %d, rc: %d", bytes_returned, rc)); + if (rc) + cFYI(1, ("Set CIFS ACL returned %d", rc)); + cifs_buf_release(pSMB); + + if (rc == -EAGAIN) + goto setCifsAclRetry; + + return (rc); +} + #endif /* CONFIG_CIFS_EXPERIMENTAL */ /* Legacy Query Path Information call for lookup to old servers such @@ -3165,8 +3220,8 @@ int SMBQueryInformation(const int xid, struct cifsTconInfo *tcon, FILE_ALL_INFO *pFinfo, const struct nls_table *nls_codepage, int remap) { - QUERY_INFORMATION_REQ * pSMB; - QUERY_INFORMATION_RSP * pSMBr; + QUERY_INFORMATION_REQ *pSMB; + QUERY_INFORMATION_RSP *pSMBr; int rc = 0; int bytes_returned; int name_len; @@ -3198,9 +3253,11 @@ QInfRetry: (struct smb_hdr *) pSMBr, &bytes_returned, 0); if (rc) { cFYI(1, ("Send error in QueryInfo = %d", rc)); - } else if (pFinfo) { /* decode response */ + } else if (pFinfo) { struct timespec ts; __u32 time = le32_to_cpu(pSMBr->last_write_time); + + /* decode response */ /* BB FIXME - add time zone adjustment BB */ memset(pFinfo, 0, sizeof(FILE_ALL_INFO)); ts.tv_nsec = 0; @@ -3231,7 +3288,7 @@ QInfRetry: int CIFSSMBQPathInfo(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, - FILE_ALL_INFO * pFindData, + FILE_ALL_INFO *pFindData, int legacy /* old style infolevel */, const struct nls_table *nls_codepage, int remap) { @@ -3306,10 +3363,12 @@ QPathInfoRetry: else if (pFindData) { int size; __u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset); - if (legacy) /* we do not read the last field, EAsize, - fortunately since it varies by subdialect - and on Set vs. Get, is two bytes or 4 - bytes depending but we don't care here */ + + /* On legacy responses we do not read the last field, + EAsize, fortunately since it varies by subdialect and + also note it differs on Set vs. Get, ie two bytes or 4 + bytes depending but we don't care here */ + if (legacy) size = sizeof(FILE_INFO_STANDARD); else size = sizeof(FILE_ALL_INFO); @@ -3411,85 +3470,6 @@ UnixQPathInfoRetry: return rc; } -#if 0 /* function unused at present */ -int CIFSFindSingle(const int xid, struct cifsTconInfo *tcon, - const char *searchName, FILE_ALL_INFO * findData, - const struct nls_table *nls_codepage) -{ -/* level 257 SMB_ */ - TRANSACTION2_FFIRST_REQ *pSMB = NULL; - TRANSACTION2_FFIRST_RSP *pSMBr = NULL; - int rc = 0; - int bytes_returned; - int name_len; - __u16 params, byte_count; - - cFYI(1, ("In FindUnique")); -findUniqueRetry: - rc = smb_init(SMB_COM_TRANSACTION2, 15, tcon, (void **) &pSMB, - (void **) &pSMBr); - if (rc) - return rc; - - if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { - name_len = - cifsConvertToUCS((__le16 *) pSMB->FileName, searchName, - PATH_MAX, nls_codepage); - name_len++; /* trailing null */ - name_len *= 2; - } else { /* BB improve the check for buffer overruns BB */ - name_len = strnlen(searchName, PATH_MAX); - name_len++; /* trailing null */ - strncpy(pSMB->FileName, searchName, name_len); - } - - params = 12 + name_len /* includes null */ ; - pSMB->TotalDataCount = 0; /* no EAs */ - pSMB->MaxParameterCount = cpu_to_le16(2); - pSMB->MaxDataCount = cpu_to_le16(4000); /* BB find exact max SMB PDU from sess structure BB */ - pSMB->MaxSetupCount = 0; - pSMB->Reserved = 0; - pSMB->Flags = 0; - pSMB->Timeout = 0; - pSMB->Reserved2 = 0; - pSMB->ParameterOffset = cpu_to_le16( - offsetof(struct smb_com_transaction2_ffirst_req, InformationLevel)-4); - pSMB->DataCount = 0; - pSMB->DataOffset = 0; - pSMB->SetupCount = 1; /* one byte, no need to le convert */ - pSMB->Reserved3 = 0; - pSMB->SubCommand = cpu_to_le16(TRANS2_FIND_FIRST); - byte_count = params + 1 /* pad */ ; - pSMB->TotalParameterCount = cpu_to_le16(params); - pSMB->ParameterCount = pSMB->TotalParameterCount; - pSMB->SearchAttributes = - cpu_to_le16(ATTR_READONLY | ATTR_HIDDEN | ATTR_SYSTEM | - ATTR_DIRECTORY); - pSMB->SearchCount = cpu_to_le16(16); /* BB increase */ - pSMB->SearchFlags = cpu_to_le16(1); - pSMB->InformationLevel = cpu_to_le16(SMB_FIND_FILE_DIRECTORY_INFO); - pSMB->SearchStorageType = 0; /* BB what should we set this to? BB */ - pSMB->hdr.smb_buf_length += byte_count; - pSMB->ByteCount = cpu_to_le16(byte_count); - - rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, - (struct smb_hdr *) pSMBr, &bytes_returned, 0); - - if (rc) { - cFYI(1, ("Send error in FindFileDirInfo = %d", rc)); - } else { /* decode response */ - cifs_stats_inc(&tcon->num_ffirst); - /* BB fill in */ - } - - cifs_buf_release(pSMB); - if (rc == -EAGAIN) - goto findUniqueRetry; - - return rc; -} -#endif /* end unused (temporarily) function */ - /* xid, tcon, searchName and codepage are input parms, rest are returned */ int CIFSFindFirst(const int xid, struct cifsTconInfo *tcon, @@ -3501,7 +3481,7 @@ CIFSFindFirst(const int xid, struct cifsTconInfo *tcon, /* level 257 SMB_ */ TRANSACTION2_FFIRST_REQ *pSMB = NULL; TRANSACTION2_FFIRST_RSP *pSMBr = NULL; - T2_FFIRST_RSP_PARMS * parms; + T2_FFIRST_RSP_PARMS *parms; int rc = 0; int bytes_returned = 0; int name_len; @@ -3632,7 +3612,7 @@ int CIFSFindNext(const int xid, struct cifsTconInfo *tcon, { TRANSACTION2_FNEXT_REQ *pSMB = NULL; TRANSACTION2_FNEXT_RSP *pSMBr = NULL; - T2_FNEXT_RSP_PARMS * parms; + T2_FNEXT_RSP_PARMS *parms; char *response_data; int rc = 0; int bytes_returned, name_len; @@ -3771,9 +3751,9 @@ CIFSFindClose(const int xid, struct cifsTconInfo *tcon, pSMB->FileID = searchHandle; pSMB->ByteCount = 0; rc = SendReceiveNoRsp(xid, tcon->ses, (struct smb_hdr *) pSMB, 0); - if (rc) { + if (rc) cERROR(1, ("Send error in FindClose = %d", rc)); - } + cifs_stats_inc(&tcon->num_fclose); /* Since session is dead, search handle closed on server already */ @@ -3786,7 +3766,7 @@ CIFSFindClose(const int xid, struct cifsTconInfo *tcon, int CIFSGetSrvInodeNumber(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, - __u64 * inode_number, + __u64 *inode_number, const struct nls_table *nls_codepage, int remap) { int rc = 0; @@ -4495,9 +4475,8 @@ SETFSUnixRetry: cERROR(1, ("Send error in SETFSUnixInfo = %d", rc)); } else { /* decode response */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc) { + if (rc) rc = -EIO; /* bad smb */ - } } cifs_buf_release(pSMB); @@ -4679,9 +4658,8 @@ SetEOFRetry: pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); - if (rc) { + if (rc) cFYI(1, ("SetPathInfo (file size) returned %d", rc)); - } cifs_buf_release(pSMB); @@ -4832,9 +4810,8 @@ CIFSSMBSetFileTimes(const int xid, struct cifsTconInfo *tcon, pSMB->ByteCount = cpu_to_le16(byte_count); memcpy(data_offset, data, sizeof(FILE_BASIC_INFO)); rc = SendReceiveNoRsp(xid, tcon->ses, (struct smb_hdr *) pSMB, 0); - if (rc) { + if (rc) cFYI(1, ("Send error in Set Time (SetFileInfo) = %d", rc)); - } /* Note: On -EAGAIN error only caller can retry on handle based calls since file handle passed in no longer valid */ @@ -4910,9 +4887,8 @@ SetTimesRetry: pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); - if (rc) { + if (rc) cFYI(1, ("SetPathInfo (times) returned %d", rc)); - } cifs_buf_release(pSMB); @@ -4962,9 +4938,8 @@ SetAttrLgcyRetry: pSMB->ByteCount = cpu_to_le16(name_len + 1); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); - if (rc) { + if (rc) cFYI(1, ("Error in LegacySetAttr = %d", rc)); - } cifs_buf_release(pSMB); @@ -5073,9 +5048,8 @@ setPermsRetry: pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); - if (rc) { + if (rc) cFYI(1, ("SetPathInfo (perms) returned %d", rc)); - } if (pSMB) cifs_buf_release(pSMB); @@ -5499,7 +5473,7 @@ SetEARetry: else name_len = strnlen(ea_name, 255); - count = sizeof(*parm_data) + ea_value_len + name_len + 1; + count = sizeof(*parm_data) + ea_value_len + name_len; pSMB->MaxParameterCount = cpu_to_le16(2); pSMB->MaxDataCount = cpu_to_le16(1000); /* BB find max SMB size from sess */ pSMB->MaxSetupCount = 0; @@ -5550,9 +5524,8 @@ SetEARetry: pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); - if (rc) { + if (rc) cFYI(1, ("SetPathInfo (EA) returned %d", rc)); - } cifs_buf_release(pSMB); diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index fd9147cdb5a..8dbfa97cd18 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -1,7 +1,7 @@ /* * fs/cifs/connect.c * - * Copyright (C) International Business Machines Corp., 2002,2007 + * Copyright (C) International Business Machines Corp., 2002,2008 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify @@ -1410,7 +1410,7 @@ connect_to_dfs_path(int xid, struct cifsSesInfo *pSesInfo, const char *old_path, const struct nls_table *nls_codepage, int remap) { - unsigned char *referrals = NULL; + struct dfs_info3_param *referrals = NULL; unsigned int num_referrals; int rc = 0; @@ -1429,12 +1429,14 @@ connect_to_dfs_path(int xid, struct cifsSesInfo *pSesInfo, int get_dfs_path(int xid, struct cifsSesInfo *pSesInfo, const char *old_path, const struct nls_table *nls_codepage, unsigned int *pnum_referrals, - unsigned char **preferrals, int remap) + struct dfs_info3_param **preferrals, int remap) { char *temp_unc; int rc = 0; + unsigned char *targetUNCs; *pnum_referrals = 0; + *preferrals = NULL; if (pSesInfo->ipc_tid == 0) { temp_unc = kmalloc(2 /* for slashes */ + @@ -1454,8 +1456,10 @@ get_dfs_path(int xid, struct cifsSesInfo *pSesInfo, const char *old_path, kfree(temp_unc); } if (rc == 0) - rc = CIFSGetDFSRefer(xid, pSesInfo, old_path, preferrals, + rc = CIFSGetDFSRefer(xid, pSesInfo, old_path, &targetUNCs, pnum_referrals, nls_codepage, remap); + /* BB map targetUNCs to dfs_info3 structures, here or + in CIFSGetDFSRefer BB */ return rc; } @@ -1718,8 +1722,15 @@ void reset_cifs_unix_caps(int xid, struct cifsTconInfo *tcon, originally at mount time */ if ((saved_cap & CIFS_UNIX_POSIX_ACL_CAP) == 0) cap &= ~CIFS_UNIX_POSIX_ACL_CAP; - if ((saved_cap & CIFS_UNIX_POSIX_PATHNAMES_CAP) == 0) + if ((saved_cap & CIFS_UNIX_POSIX_PATHNAMES_CAP) == 0) { + if (cap & CIFS_UNIX_POSIX_PATHNAMES_CAP) + cERROR(1, ("POSIXPATH support change")); cap &= ~CIFS_UNIX_POSIX_PATHNAMES_CAP; + } else if ((cap & CIFS_UNIX_POSIX_PATHNAMES_CAP) == 0) { + cERROR(1, ("possible reconnect error")); + cERROR(1, + ("server disabled POSIX path support")); + } } cap &= CIFS_UNIX_CAP_MASK; @@ -1749,9 +1760,8 @@ void reset_cifs_unix_caps(int xid, struct cifsTconInfo *tcon, if (sb && (CIFS_SB(sb)->rsize > 127 * 1024)) { if ((cap & CIFS_UNIX_LARGE_READ_CAP) == 0) { CIFS_SB(sb)->rsize = 127 * 1024; -#ifdef CONFIG_CIFS_DEBUG2 - cFYI(1, ("larger reads not supported by srv")); -#endif + cFYI(DBG2, + ("larger reads not supported by srv")); } } @@ -1788,6 +1798,26 @@ void reset_cifs_unix_caps(int xid, struct cifsTconInfo *tcon, } } +static void +convert_delimiter(char *path, char delim) +{ + int i; + char old_delim; + + if (path == NULL) + return; + + if (delim == '/') + old_delim = '\\'; + else + old_delim = '/'; + + for (i = 0; path[i] != '\0'; i++) { + if (path[i] == old_delim) + path[i] = delim; + } +} + int cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, char *mount_data, const char *devname) @@ -1964,7 +1994,15 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, if (existingCifsSes) { pSesInfo = existingCifsSes; - cFYI(1, ("Existing smb sess found")); + cFYI(1, ("Existing smb sess found (status=%d)", + pSesInfo->status)); + down(&pSesInfo->sesSem); + if (pSesInfo->status == CifsNeedReconnect) { + cFYI(1, ("Session needs reconnect")); + rc = cifs_setup_session(xid, pSesInfo, + cifs_sb->local_nls); + } + up(&pSesInfo->sesSem); } else if (!rc) { cFYI(1, ("Existing smb sess not found")); pSesInfo = sesInfoAlloc(); @@ -2045,7 +2083,11 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, cifs_sb->prepath = volume_info.prepath; if (cifs_sb->prepath) { cifs_sb->prepathlen = strlen(cifs_sb->prepath); - cifs_sb->prepath[0] = CIFS_DIR_SEP(cifs_sb); + /* we can not convert the / to \ in the path + separators in the prefixpath yet because we do not + know (until reset_cifs_unix_caps is called later) + whether POSIX PATH CAP is available. We normalize + the / to \ after reset_cifs_unix_caps is called */ volume_info.prepath = NULL; } else cifs_sb->prepathlen = 0; @@ -2213,11 +2255,15 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, else tcon->unix_ext = 0; /* server does not support them */ + /* convert forward to back slashes in prepath here if needed */ + if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) == 0) + convert_delimiter(cifs_sb->prepath, + CIFS_DIR_SEP(cifs_sb)); + if ((tcon->unix_ext == 0) && (cifs_sb->rsize > (1024 * 127))) { cifs_sb->rsize = 1024 * 127; -#ifdef CONFIG_CIFS_DEBUG2 - cFYI(1, ("no very large read support, rsize now 127K")); -#endif + cFYI(DBG2, + ("no very large read support, rsize now 127K")); } if (!(tcon->ses->capabilities & CAP_LARGE_WRITE_X)) cifs_sb->wsize = min(cifs_sb->wsize, @@ -3514,7 +3560,7 @@ cifs_umount(struct super_block *sb, struct cifs_sb_info *cifs_sb) sesInfoFree(ses); FreeXid(xid); - return rc; /* BB check if we should always return zero here */ + return rc; } int cifs_setup_session(unsigned int xid, struct cifsSesInfo *pSesInfo, diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index 37dc97af148..0f5c62ba403 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -3,7 +3,7 @@ * * vfs operations that deal with dentries * - * Copyright (C) International Business Machines Corp., 2002,2007 + * Copyright (C) International Business Machines Corp., 2002,2008 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify @@ -111,16 +111,6 @@ cifs_bp_rename_retry: return full_path; } -/* char * build_wildcard_path_from_dentry(struct dentry *direntry) -{ - if(full_path == NULL) - return full_path; - - full_path[namelen] = '\\'; - full_path[namelen+1] = '*'; - full_path[namelen+2] = 0; -BB remove above eight lines BB */ - /* Inode operations in similar order to how they appear in Linux file fs.h */ int @@ -171,9 +161,8 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, disposition = FILE_OVERWRITE_IF; else if ((oflags & O_CREAT) == O_CREAT) disposition = FILE_OPEN_IF; - else { + else cFYI(1, ("Create flag not set in create function")); - } } /* BB add processing to set equivalent of mode - e.g. via CreateX with @@ -240,7 +229,8 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, inode->i_sb, xid); else { rc = cifs_get_inode_info(&newinode, full_path, - buf, inode->i_sb, xid); + buf, inode->i_sb, xid, + &fileHandle); if (newinode) { newinode->i_mode = mode; if ((oplock & CIFS_CREATE_ACTION) && @@ -367,7 +357,7 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode, if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) { int oplock = 0; u16 fileHandle; - FILE_ALL_INFO * buf; + FILE_ALL_INFO *buf; cFYI(1, ("sfu compat create special file")); @@ -494,7 +484,7 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, parent_dir_inode->i_sb, xid); else rc = cifs_get_inode_info(&newInode, full_path, NULL, - parent_dir_inode->i_sb, xid); + parent_dir_inode->i_sb, xid, NULL); if ((rc == 0) && (newInode != NULL)) { if (pTcon->nocase) @@ -517,12 +507,10 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, d_add(direntry, NULL); /* if it was once a directory (but how can we tell?) we could do shrink_dcache_parent(direntry); */ - } else { - cERROR(1, ("Error 0x%x on cifs_get_inode_info in lookup of %s", - rc, full_path)); - /* BB special case check for Access Denied - watch security - exposure of returning dir info implicitly via different rc - if file exists or not but no access BB */ + } else if (rc != -EACCES) { + cERROR(1, ("Unexpected lookup error %d", rc)); + /* We special case check for Access Denied - since that + is a common return code */ } kfree(full_path); @@ -536,9 +524,8 @@ cifs_d_revalidate(struct dentry *direntry, struct nameidata *nd) int isValid = 1; if (direntry->d_inode) { - if (cifs_revalidate(direntry)) { + if (cifs_revalidate(direntry)) return 0; - } } else { cFYI(1, ("neg dentry 0x%p name = %s", direntry, direntry->d_name.name)); diff --git a/fs/cifs/dns_resolve.c b/fs/cifs/dns_resolve.c new file mode 100644 index 00000000000..7cc86c41818 --- /dev/null +++ b/fs/cifs/dns_resolve.c @@ -0,0 +1,124 @@ +/* + * fs/cifs/dns_resolve.c + * + * Copyright (c) 2007 Igor Mammedov + * Author(s): Igor Mammedov (niallain@gmail.com) + * Steve French (sfrench@us.ibm.com) + * + * Contains the CIFS DFS upcall routines used for hostname to + * IP address translation. + * + * This library is free software; you can redistribute it and/or modify + * it under the terms of the GNU Lesser General Public License as published + * by the Free Software Foundation; either version 2.1 of the License, or + * (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See + * the GNU Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public License + * along with this library; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#include <keys/user-type.h> +#include "dns_resolve.h" +#include "cifsglob.h" +#include "cifsproto.h" +#include "cifs_debug.h" + +static int dns_resolver_instantiate(struct key *key, const void *data, + size_t datalen) +{ + int rc = 0; + char *ip; + + ip = kmalloc(datalen+1, GFP_KERNEL); + if (!ip) + return -ENOMEM; + + memcpy(ip, data, datalen); + ip[datalen] = '\0'; + + rcu_assign_pointer(key->payload.data, ip); + + return rc; +} + +struct key_type key_type_dns_resolver = { + .name = "dns_resolver", + .def_datalen = sizeof(struct in_addr), + .describe = user_describe, + .instantiate = dns_resolver_instantiate, + .match = user_match, +}; + + +/* Resolves server name to ip address. + * input: + * unc - server UNC + * output: + * *ip_addr - pointer to server ip, caller responcible for freeing it. + * return 0 on success + */ +int +dns_resolve_server_name_to_ip(const char *unc, char **ip_addr) +{ + int rc = -EAGAIN; + struct key *rkey; + char *name; + int len; + + if (!ip_addr || !unc) + return -EINVAL; + + /* search for server name delimiter */ + len = strlen(unc); + if (len < 3) { + cFYI(1, ("%s: unc is too short: %s", __func__, unc)); + return -EINVAL; + } + len -= 2; + name = memchr(unc+2, '\\', len); + if (!name) { + cFYI(1, ("%s: probably server name is whole unc: %s", + __func__, unc)); + } else { + len = (name - unc) - 2/* leading // */; + } + + name = kmalloc(len+1, GFP_KERNEL); + if (!name) { + rc = -ENOMEM; + return rc; + } + memcpy(name, unc+2, len); + name[len] = 0; + + rkey = request_key(&key_type_dns_resolver, name, ""); + if (!IS_ERR(rkey)) { + len = strlen(rkey->payload.data); + *ip_addr = kmalloc(len+1, GFP_KERNEL); + if (*ip_addr) { + memcpy(*ip_addr, rkey->payload.data, len); + (*ip_addr)[len] = '\0'; + cFYI(1, ("%s: resolved: %s to %s", __func__, + rkey->description, + *ip_addr + )); + rc = 0; + } else { + rc = -ENOMEM; + } + key_put(rkey); + } else { + cERROR(1, ("%s: unable to resolve: %s", __func__, name)); + } + + kfree(name); + return rc; +} + + diff --git a/fs/cifs/dns_resolve.h b/fs/cifs/dns_resolve.h new file mode 100644 index 00000000000..966e9288930 --- /dev/null +++ b/fs/cifs/dns_resolve.h @@ -0,0 +1,32 @@ +/* + * fs/cifs/dns_resolve.h -- DNS Resolver upcall management for CIFS DFS + * Handles host name to IP address resolution + * + * Copyright (c) International Business Machines Corp., 2008 + * Author(s): Steve French (sfrench@us.ibm.com) + * + * This library is free software; you can redistribute it and/or modify + * it under the terms of the GNU Lesser General Public License as published + * by the Free Software Foundation; either version 2.1 of the License, or + * (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See + * the GNU Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public License + * along with this library; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#ifndef _DNS_RESOLVE_H +#define _DNS_RESOLVE_H + +#ifdef __KERNEL__ +#include <linux/key-type.h> +extern struct key_type key_type_dns_resolver; +extern int dns_resolve_server_name_to_ip(const char *unc, char **ip_addr); +#endif /* KERNEL */ + +#endif /* _DNS_RESOLVE_H */ diff --git a/fs/cifs/fcntl.c b/fs/cifs/fcntl.c index 995474c9088..7d1d5aa4c43 100644 --- a/fs/cifs/fcntl.c +++ b/fs/cifs/fcntl.c @@ -35,9 +35,8 @@ static __u32 convert_to_cifs_notify_flags(unsigned long fcntl_notify_flags) /* No way on Linux VFS to ask to monitor xattr changes (and no stream support either */ - if (fcntl_notify_flags & DN_ACCESS) { + if (fcntl_notify_flags & DN_ACCESS) cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_LAST_ACCESS; - } if (fcntl_notify_flags & DN_MODIFY) { /* What does this mean on directories? */ cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_LAST_WRITE | @@ -47,9 +46,8 @@ static __u32 convert_to_cifs_notify_flags(unsigned long fcntl_notify_flags) cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_CREATION | FILE_NOTIFY_CHANGE_LAST_WRITE; } - if (fcntl_notify_flags & DN_DELETE) { + if (fcntl_notify_flags & DN_DELETE) cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_LAST_WRITE; - } if (fcntl_notify_flags & DN_RENAME) { /* BB review this - checking various server behaviors */ cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_DIR_NAME | diff --git a/fs/cifs/file.c b/fs/cifs/file.c index dd26e2759b1..40b690073fc 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -145,7 +145,7 @@ client_can_cache: full_path, inode->i_sb, xid); else rc = cifs_get_inode_info(&file->f_path.dentry->d_inode, - full_path, buf, inode->i_sb, xid); + full_path, buf, inode->i_sb, xid, NULL); if ((*oplock & 0xF) == OPLOCK_EXCLUSIVE) { pCifsInode->clientCanCacheAll = TRUE; @@ -353,9 +353,9 @@ static int cifs_reopen_file(struct file *file, int can_flush) int disposition = FILE_OPEN; __u16 netfid; - if (file->private_data) { + if (file->private_data) pCifsFile = (struct cifsFileInfo *)file->private_data; - } else + else return -EBADF; xid = GetXid(); @@ -440,7 +440,7 @@ reopen_error_exit: else rc = cifs_get_inode_info(&inode, full_path, NULL, inode->i_sb, - xid); + xid, NULL); } /* else we are writing out data to server already and could deadlock if we tried to flush data, and since we do not know if we have data that would @@ -499,9 +499,8 @@ int cifs_close(struct inode *inode, struct file *file) the struct would be in each open file, but this should give enough time to clear the socket */ -#ifdef CONFIG_CIFS_DEBUG2 - cFYI(1, ("close delay, write pending")); -#endif /* DEBUG2 */ + cFYI(DBG2, + ("close delay, write pending")); msleep(timeout); timeout *= 4; } @@ -1179,12 +1178,10 @@ static int cifs_partialpagewrite(struct page *page, unsigned from, unsigned to) atomic_dec(&open_file->wrtPending); /* Does mm or vfs already set times? */ inode->i_atime = inode->i_mtime = current_fs_time(inode->i_sb); - if ((bytes_written > 0) && (offset)) { + if ((bytes_written > 0) && (offset)) rc = 0; - } else if (bytes_written < 0) { - if (rc != -EBADF) - rc = bytes_written; - } + else if (bytes_written < 0) + rc = bytes_written; } else { cFYI(1, ("No writeable filehandles for inode")); rc = -EIO; @@ -1425,9 +1422,8 @@ static int cifs_writepage(struct page *page, struct writeback_control *wbc) xid = GetXid(); /* BB add check for wbc flags */ page_cache_get(page); - if (!PageUptodate(page)) { + if (!PageUptodate(page)) cFYI(1, ("ppw - page not up to date")); - } /* * Set the "writeback" flag, and clear "dirty" in the radix tree. @@ -1462,9 +1458,9 @@ static int cifs_commit_write(struct file *file, struct page *page, cFYI(1, ("commit write for page %p up to position %lld for %d", page, position, to)); spin_lock(&inode->i_lock); - if (position > inode->i_size) { + if (position > inode->i_size) i_size_write(inode, position); - } + spin_unlock(&inode->i_lock); if (!PageUptodate(page)) { position = ((loff_t)page->index << PAGE_CACHE_SHIFT) + offset; @@ -1598,9 +1594,9 @@ ssize_t cifs_user_read(struct file *file, char __user *read_data, } open_file = (struct cifsFileInfo *)file->private_data; - if ((file->f_flags & O_ACCMODE) == O_WRONLY) { + if ((file->f_flags & O_ACCMODE) == O_WRONLY) cFYI(1, ("attempting read on write only file instance")); - } + for (total_read = 0, current_offset = read_data; read_size > total_read; total_read += bytes_read, current_offset += bytes_read) { @@ -1627,9 +1623,8 @@ ssize_t cifs_user_read(struct file *file, char __user *read_data, smb_read_data + 4 /* RFC1001 length field */ + le16_to_cpu(pSMBr->DataOffset), - bytes_read)) { + bytes_read)) rc = -EFAULT; - } if (buf_type == CIFS_SMALL_BUFFER) cifs_small_buf_release(smb_read_data); @@ -1816,9 +1811,7 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, pTcon = cifs_sb->tcon; pagevec_init(&lru_pvec, 0); -#ifdef CONFIG_CIFS_DEBUG2 - cFYI(1, ("rpages: num pages %d", num_pages)); -#endif + cFYI(DBG2, ("rpages: num pages %d", num_pages)); for (i = 0; i < num_pages; ) { unsigned contig_pages; struct page *tmp_page; @@ -1851,10 +1844,8 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, /* Read size needs to be in multiples of one page */ read_size = min_t(const unsigned int, read_size, cifs_sb->rsize & PAGE_CACHE_MASK); -#ifdef CONFIG_CIFS_DEBUG2 - cFYI(1, ("rpages: read size 0x%x contiguous pages %d", + cFYI(DBG2, ("rpages: read size 0x%x contiguous pages %d", read_size, contig_pages)); -#endif rc = -EAGAIN; while (rc == -EAGAIN) { if ((open_file->invalidHandle) && @@ -2028,7 +2019,7 @@ int is_size_safe_to_change(struct cifsInodeInfo *cifsInode, __u64 end_of_file) struct cifs_sb_info *cifs_sb; cifs_sb = CIFS_SB(cifsInode->vfs_inode.i_sb); - if ( cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO ) { + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) { /* since no page cache to corrupt on directio we can change size safely */ return 1; diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index e915eb1d2e6..bc673c8c1e6 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -29,6 +29,162 @@ #include "cifs_debug.h" #include "cifs_fs_sb.h" + +static void cifs_set_ops(struct inode *inode, const bool is_dfs_referral) +{ + struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + + switch (inode->i_mode & S_IFMT) { + case S_IFREG: + inode->i_op = &cifs_file_inode_ops; + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) { + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) + inode->i_fop = &cifs_file_direct_nobrl_ops; + else + inode->i_fop = &cifs_file_direct_ops; + } else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) + inode->i_fop = &cifs_file_nobrl_ops; + else { /* not direct, send byte range locks */ + inode->i_fop = &cifs_file_ops; + } + + + /* check if server can support readpages */ + if (cifs_sb->tcon->ses->server->maxBuf < + PAGE_CACHE_SIZE + MAX_CIFS_HDR_SIZE) + inode->i_data.a_ops = &cifs_addr_ops_smallbuf; + else + inode->i_data.a_ops = &cifs_addr_ops; + break; + case S_IFDIR: +#ifdef CONFIG_CIFS_DFS_UPCALL + if (is_dfs_referral) { + inode->i_op = &cifs_dfs_referral_inode_operations; + } else { +#else /* NO DFS support, treat as a directory */ + { +#endif + inode->i_op = &cifs_dir_inode_ops; + inode->i_fop = &cifs_dir_ops; + } + break; + case S_IFLNK: + inode->i_op = &cifs_symlink_inode_ops; + break; + default: + init_special_inode(inode, inode->i_mode, inode->i_rdev); + break; + } +} + +static void cifs_unix_info_to_inode(struct inode *inode, + FILE_UNIX_BASIC_INFO *info, int force_uid_gid) +{ + struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifsInodeInfo *cifsInfo = CIFS_I(inode); + __u64 num_of_bytes = le64_to_cpu(info->NumOfBytes); + __u64 end_of_file = le64_to_cpu(info->EndOfFile); + + inode->i_atime = cifs_NTtimeToUnix(le64_to_cpu(info->LastAccessTime)); + inode->i_mtime = + cifs_NTtimeToUnix(le64_to_cpu(info->LastModificationTime)); + inode->i_ctime = cifs_NTtimeToUnix(le64_to_cpu(info->LastStatusChange)); + inode->i_mode = le64_to_cpu(info->Permissions); + + /* + * Since we set the inode type below we need to mask off + * to avoid strange results if bits set above. + */ + inode->i_mode &= ~S_IFMT; + switch (le32_to_cpu(info->Type)) { + case UNIX_FILE: + inode->i_mode |= S_IFREG; + break; + case UNIX_SYMLINK: + inode->i_mode |= S_IFLNK; + break; + case UNIX_DIR: + inode->i_mode |= S_IFDIR; + break; + case UNIX_CHARDEV: + inode->i_mode |= S_IFCHR; + inode->i_rdev = MKDEV(le64_to_cpu(info->DevMajor), + le64_to_cpu(info->DevMinor) & MINORMASK); + break; + case UNIX_BLOCKDEV: + inode->i_mode |= S_IFBLK; + inode->i_rdev = MKDEV(le64_to_cpu(info->DevMajor), + le64_to_cpu(info->DevMinor) & MINORMASK); + break; + case UNIX_FIFO: + inode->i_mode |= S_IFIFO; + break; + case UNIX_SOCKET: + inode->i_mode |= S_IFSOCK; + break; + default: + /* safest to call it a file if we do not know */ + inode->i_mode |= S_IFREG; + cFYI(1, ("unknown type %d", le32_to_cpu(info->Type))); + break; + } + + if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_UID) && + !force_uid_gid) + inode->i_uid = cifs_sb->mnt_uid; + else + inode->i_uid = le64_to_cpu(info->Uid); + + if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_GID) && + !force_uid_gid) + inode->i_gid = cifs_sb->mnt_gid; + else + inode->i_gid = le64_to_cpu(info->Gid); + + inode->i_nlink = le64_to_cpu(info->Nlinks); + + spin_lock(&inode->i_lock); + if (is_size_safe_to_change(cifsInfo, end_of_file)) { + /* + * We can not safely change the file size here if the client + * is writing to it due to potential races. + */ + i_size_write(inode, end_of_file); + + /* + * i_blocks is not related to (i_size / i_blksize), + * but instead 512 byte (2**9) size is required for + * calculating num blocks. + */ + inode->i_blocks = (512 - 1 + num_of_bytes) >> 9; + } + spin_unlock(&inode->i_lock); +} + +static const unsigned char *cifs_get_search_path(struct cifsTconInfo *pTcon, + const char *search_path) +{ + int tree_len; + int path_len; + char *tmp_path; + + if (!(pTcon->Flags & SMB_SHARE_IS_IN_DFS)) + return search_path; + + /* use full path name for working with DFS */ + tree_len = strnlen(pTcon->treeName, MAX_TREE_SIZE + 1); + path_len = strnlen(search_path, MAX_PATHCONF); + + tmp_path = kmalloc(tree_len+path_len+1, GFP_KERNEL); + if (tmp_path == NULL) + return search_path; + + strncpy(tmp_path, pTcon->treeName, tree_len); + strncpy(tmp_path+tree_len, search_path, path_len); + tmp_path[tree_len+path_len] = 0; + return tmp_path; +} + int cifs_get_inode_info_unix(struct inode **pinode, const unsigned char *search_path, struct super_block *sb, int xid) { @@ -37,52 +193,43 @@ int cifs_get_inode_info_unix(struct inode **pinode, struct cifsTconInfo *pTcon; struct inode *inode; struct cifs_sb_info *cifs_sb = CIFS_SB(sb); - char *tmp_path; + const unsigned char *full_path; + bool is_dfs_referral = false; pTcon = cifs_sb->tcon; cFYI(1, ("Getting info on %s", search_path)); + + full_path = cifs_get_search_path(pTcon, search_path); + +try_again_CIFSSMBUnixQPathInfo: /* could have done a find first instead but this returns more info */ - rc = CIFSSMBUnixQPathInfo(xid, pTcon, search_path, &findData, + rc = CIFSSMBUnixQPathInfo(xid, pTcon, full_path, &findData, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); /* dump_mem("\nUnixQPathInfo return data", &findData, sizeof(findData)); */ if (rc) { - if (rc == -EREMOTE) { - tmp_path = - kmalloc(strnlen(pTcon->treeName, - MAX_TREE_SIZE + 1) + - strnlen(search_path, MAX_PATHCONF) + 1, - GFP_KERNEL); - if (tmp_path == NULL) { - return -ENOMEM; + if (rc == -EREMOTE && !is_dfs_referral) { + is_dfs_referral = true; + if (full_path != search_path) { + kfree(full_path); + full_path = search_path; } - /* have to skip first of the double backslash of - UNC name */ - strncpy(tmp_path, pTcon->treeName, MAX_TREE_SIZE); - strncat(tmp_path, search_path, MAX_PATHCONF); - rc = connect_to_dfs_path(xid, pTcon->ses, - /* treename + */ tmp_path, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - kfree(tmp_path); - - /* BB fix up inode etc. */ - } else if (rc) { - return rc; + goto try_again_CIFSSMBUnixQPathInfo; } + goto cgiiu_exit; } else { struct cifsInodeInfo *cifsInfo; - __u32 type = le32_to_cpu(findData.Type); __u64 num_of_bytes = le64_to_cpu(findData.NumOfBytes); __u64 end_of_file = le64_to_cpu(findData.EndOfFile); /* get new inode */ if (*pinode == NULL) { *pinode = new_inode(sb); - if (*pinode == NULL) - return -ENOMEM; + if (*pinode == NULL) { + rc = -ENOMEM; + goto cgiiu_exit; + } /* Is an i_ino of zero legal? */ /* Are there sanity checks we can use to ensure that the server is really filling in that field? */ @@ -105,113 +252,20 @@ int cifs_get_inode_info_unix(struct inode **pinode, /* this is ok to set on every inode revalidate */ atomic_set(&cifsInfo->inUse, 1); - inode->i_atime = - cifs_NTtimeToUnix(le64_to_cpu(findData.LastAccessTime)); - inode->i_mtime = - cifs_NTtimeToUnix(le64_to_cpu - (findData.LastModificationTime)); - inode->i_ctime = - cifs_NTtimeToUnix(le64_to_cpu(findData.LastStatusChange)); - inode->i_mode = le64_to_cpu(findData.Permissions); - /* since we set the inode type below we need to mask off - to avoid strange results if bits set above */ - inode->i_mode &= ~S_IFMT; - if (type == UNIX_FILE) { - inode->i_mode |= S_IFREG; - } else if (type == UNIX_SYMLINK) { - inode->i_mode |= S_IFLNK; - } else if (type == UNIX_DIR) { - inode->i_mode |= S_IFDIR; - } else if (type == UNIX_CHARDEV) { - inode->i_mode |= S_IFCHR; - inode->i_rdev = MKDEV(le64_to_cpu(findData.DevMajor), - le64_to_cpu(findData.DevMinor) & MINORMASK); - } else if (type == UNIX_BLOCKDEV) { - inode->i_mode |= S_IFBLK; - inode->i_rdev = MKDEV(le64_to_cpu(findData.DevMajor), - le64_to_cpu(findData.DevMinor) & MINORMASK); - } else if (type == UNIX_FIFO) { - inode->i_mode |= S_IFIFO; - } else if (type == UNIX_SOCKET) { - inode->i_mode |= S_IFSOCK; - } else { - /* safest to call it a file if we do not know */ - inode->i_mode |= S_IFREG; - cFYI(1, ("unknown type %d", type)); - } + cifs_unix_info_to_inode(inode, &findData, 0); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_UID) - inode->i_uid = cifs_sb->mnt_uid; - else - inode->i_uid = le64_to_cpu(findData.Uid); - - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_GID) - inode->i_gid = cifs_sb->mnt_gid; - else - inode->i_gid = le64_to_cpu(findData.Gid); - - inode->i_nlink = le64_to_cpu(findData.Nlinks); - - spin_lock(&inode->i_lock); - if (is_size_safe_to_change(cifsInfo, end_of_file)) { - /* can not safely change the file size here if the - client is writing to it due to potential races */ - i_size_write(inode, end_of_file); - - /* blksize needs to be multiple of two. So safer to default to - blksize and blkbits set in superblock so 2**blkbits and blksize - will match rather than setting to: - (pTcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE) & 0xFFFFFE00;*/ - - /* This seems incredibly stupid but it turns out that i_blocks - is not related to (i_size / i_blksize), instead 512 byte size - is required for calculating num blocks */ - - /* 512 bytes (2**9) is the fake blocksize that must be used */ - /* for this calculation */ - inode->i_blocks = (512 - 1 + num_of_bytes) >> 9; - } - spin_unlock(&inode->i_lock); if (num_of_bytes < end_of_file) cFYI(1, ("allocation size less than end of file")); cFYI(1, ("Size %ld and blocks %llu", (unsigned long) inode->i_size, (unsigned long long)inode->i_blocks)); - if (S_ISREG(inode->i_mode)) { - cFYI(1, ("File inode")); - inode->i_op = &cifs_file_inode_ops; - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) { - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) - inode->i_fop = - &cifs_file_direct_nobrl_ops; - else - inode->i_fop = &cifs_file_direct_ops; - } else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) - inode->i_fop = &cifs_file_nobrl_ops; - else /* not direct, send byte range locks */ - inode->i_fop = &cifs_file_ops; - - /* check if server can support readpages */ - if (pTcon->ses->server->maxBuf < - PAGE_CACHE_SIZE + MAX_CIFS_HDR_SIZE) - inode->i_data.a_ops = &cifs_addr_ops_smallbuf; - else - inode->i_data.a_ops = &cifs_addr_ops; - } else if (S_ISDIR(inode->i_mode)) { - cFYI(1, ("Directory inode")); - inode->i_op = &cifs_dir_inode_ops; - inode->i_fop = &cifs_dir_ops; - } else if (S_ISLNK(inode->i_mode)) { - cFYI(1, ("Symbolic Link inode")); - inode->i_op = &cifs_symlink_inode_ops; - /* tmp_inode->i_fop = */ /* do not need to set to anything */ - } else { - cFYI(1, ("Init special inode")); - init_special_inode(inode, inode->i_mode, - inode->i_rdev); - } + + cifs_set_ops(inode, is_dfs_referral); } +cgiiu_exit: + if (full_path != search_path) + kfree(full_path); return rc; } @@ -320,15 +374,16 @@ static int get_sfu_mode(struct inode *inode, int cifs_get_inode_info(struct inode **pinode, const unsigned char *search_path, FILE_ALL_INFO *pfindData, - struct super_block *sb, int xid) + struct super_block *sb, int xid, const __u16 *pfid) { int rc = 0; struct cifsTconInfo *pTcon; struct inode *inode; struct cifs_sb_info *cifs_sb = CIFS_SB(sb); - char *tmp_path; + const unsigned char *full_path = NULL; char *buf = NULL; int adjustTZ = FALSE; + bool is_dfs_referral = false; pTcon = cifs_sb->tcon; cFYI(1, ("Getting info on %s", search_path)); @@ -346,8 +401,12 @@ int cifs_get_inode_info(struct inode **pinode, if (buf == NULL) return -ENOMEM; pfindData = (FILE_ALL_INFO *)buf; + + full_path = cifs_get_search_path(pTcon, search_path); + +try_again_CIFSSMBQPathInfo: /* could do find first instead but this returns more info */ - rc = CIFSSMBQPathInfo(xid, pTcon, search_path, pfindData, + rc = CIFSSMBQPathInfo(xid, pTcon, full_path, pfindData, 0 /* not legacy */, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); @@ -355,7 +414,7 @@ int cifs_get_inode_info(struct inode **pinode, when server claims no NT SMB support and the above call failed at least once - set flag in tcon or mount */ if ((rc == -EOPNOTSUPP) || (rc == -EINVAL)) { - rc = SMBQueryInformation(xid, pTcon, search_path, + rc = SMBQueryInformation(xid, pTcon, full_path, pfindData, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); @@ -364,31 +423,15 @@ int cifs_get_inode_info(struct inode **pinode, } /* dump_mem("\nQPathInfo return data",&findData, sizeof(findData)); */ if (rc) { - if (rc == -EREMOTE) { - tmp_path = - kmalloc(strnlen - (pTcon->treeName, - MAX_TREE_SIZE + 1) + - strnlen(search_path, MAX_PATHCONF) + 1, - GFP_KERNEL); - if (tmp_path == NULL) { - kfree(buf); - return -ENOMEM; + if (rc == -EREMOTE && !is_dfs_referral) { + is_dfs_referral = true; + if (full_path != search_path) { + kfree(full_path); + full_path = search_path; } - - strncpy(tmp_path, pTcon->treeName, MAX_TREE_SIZE); - strncat(tmp_path, search_path, MAX_PATHCONF); - rc = connect_to_dfs_path(xid, pTcon->ses, - /* treename + */ tmp_path, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - kfree(tmp_path); - /* BB fix up inode etc. */ - } else if (rc) { - kfree(buf); - return rc; + goto try_again_CIFSSMBQPathInfo; } + goto cgii_exit; } else { struct cifsInodeInfo *cifsInfo; __u32 attr = le32_to_cpu(pfindData->Attributes); @@ -397,8 +440,8 @@ int cifs_get_inode_info(struct inode **pinode, if (*pinode == NULL) { *pinode = new_inode(sb); if (*pinode == NULL) { - kfree(buf); - return -ENOMEM; + rc = -ENOMEM; + goto cgii_exit; } /* Is an i_ino of zero legal? Can we use that to check if the server supports returning inode numbers? Are @@ -490,9 +533,9 @@ int cifs_get_inode_info(struct inode **pinode, if (decode_sfu_inode(inode, le64_to_cpu(pfindData->EndOfFile), search_path, - cifs_sb, xid)) { + cifs_sb, xid)) cFYI(1, ("Unrecognized sfu inode type")); - } + cFYI(1, ("sfu mode 0%o", inode->i_mode)); } else { inode->i_mode |= S_IFREG; @@ -511,7 +554,8 @@ int cifs_get_inode_info(struct inode **pinode, } spin_lock(&inode->i_lock); - if (is_size_safe_to_change(cifsInfo, le64_to_cpu(pfindData->EndOfFile))) { + if (is_size_safe_to_change(cifsInfo, + le64_to_cpu(pfindData->EndOfFile))) { /* can not safely shrink the file size here if the client is writing to it due to potential races */ i_size_write(inode, le64_to_cpu(pfindData->EndOfFile)); @@ -531,7 +575,7 @@ int cifs_get_inode_info(struct inode **pinode, /* fill in 0777 bits from ACL */ if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) { cFYI(1, ("Getting mode bits from ACL")); - acl_to_uid_mode(inode, search_path); + acl_to_uid_mode(inode, search_path, pfid); } #endif if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) { @@ -545,37 +589,11 @@ int cifs_get_inode_info(struct inode **pinode, atomic_set(&cifsInfo->inUse, 1); } - if (S_ISREG(inode->i_mode)) { - cFYI(1, ("File inode")); - inode->i_op = &cifs_file_inode_ops; - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) { - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) - inode->i_fop = - &cifs_file_direct_nobrl_ops; - else - inode->i_fop = &cifs_file_direct_ops; - } else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) - inode->i_fop = &cifs_file_nobrl_ops; - else /* not direct, send byte range locks */ - inode->i_fop = &cifs_file_ops; - - if (pTcon->ses->server->maxBuf < - PAGE_CACHE_SIZE + MAX_CIFS_HDR_SIZE) - inode->i_data.a_ops = &cifs_addr_ops_smallbuf; - else - inode->i_data.a_ops = &cifs_addr_ops; - } else if (S_ISDIR(inode->i_mode)) { - cFYI(1, ("Directory inode")); - inode->i_op = &cifs_dir_inode_ops; - inode->i_fop = &cifs_dir_ops; - } else if (S_ISLNK(inode->i_mode)) { - cFYI(1, ("Symbolic Link inode")); - inode->i_op = &cifs_symlink_inode_ops; - } else { - init_special_inode(inode, inode->i_mode, - inode->i_rdev); - } + cifs_set_ops(inode, is_dfs_referral); } +cgii_exit: + if (full_path != search_path) + kfree(full_path); kfree(buf); return rc; } @@ -585,10 +603,18 @@ static const struct inode_operations cifs_ipc_inode_ops = { }; /* gets root inode */ -void cifs_read_inode(struct inode *inode) +struct inode *cifs_iget(struct super_block *sb, unsigned long ino) { - int xid, rc; + int xid; struct cifs_sb_info *cifs_sb; + struct inode *inode; + long rc; + + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; cifs_sb = CIFS_SB(inode->i_sb); xid = GetXid(); @@ -596,7 +622,8 @@ void cifs_read_inode(struct inode *inode) if (cifs_sb->tcon->unix_ext) rc = cifs_get_inode_info_unix(&inode, "", inode->i_sb, xid); else - rc = cifs_get_inode_info(&inode, "", NULL, inode->i_sb, xid); + rc = cifs_get_inode_info(&inode, "", NULL, inode->i_sb, xid, + NULL); if (rc && cifs_sb->tcon->ipc) { cFYI(1, ("ipc connection - fake read inode")); inode->i_mode |= S_IFDIR; @@ -605,10 +632,18 @@ void cifs_read_inode(struct inode *inode) inode->i_fop = &simple_dir_operations; inode->i_uid = cifs_sb->mnt_uid; inode->i_gid = cifs_sb->mnt_gid; + _FreeXid(xid); + iget_failed(inode); + return ERR_PTR(rc); } - /* can not call macro FreeXid here since in a void func */ + unlock_new_inode(inode); + + /* can not call macro FreeXid here since in a void func + * TODO: This is no longer true + */ _FreeXid(xid); + return inode; } int cifs_unlink(struct inode *inode, struct dentry *direntry) @@ -775,17 +810,12 @@ psx_del_no_retry: } static void posix_fill_in_inode(struct inode *tmp_inode, - FILE_UNIX_BASIC_INFO *pData, int *pobject_type, int isNewInode) + FILE_UNIX_BASIC_INFO *pData, int isNewInode) { + struct cifsInodeInfo *cifsInfo = CIFS_I(tmp_inode); loff_t local_size; struct timespec local_mtime; - struct cifsInodeInfo *cifsInfo = CIFS_I(tmp_inode); - struct cifs_sb_info *cifs_sb = CIFS_SB(tmp_inode->i_sb); - - __u32 type = le32_to_cpu(pData->Type); - __u64 num_of_bytes = le64_to_cpu(pData->NumOfBytes); - __u64 end_of_file = le64_to_cpu(pData->EndOfFile); cifsInfo->time = jiffies; atomic_inc(&cifsInfo->inUse); @@ -793,115 +823,27 @@ static void posix_fill_in_inode(struct inode *tmp_inode, local_mtime = tmp_inode->i_mtime; local_size = tmp_inode->i_size; - tmp_inode->i_atime = - cifs_NTtimeToUnix(le64_to_cpu(pData->LastAccessTime)); - tmp_inode->i_mtime = - cifs_NTtimeToUnix(le64_to_cpu(pData->LastModificationTime)); - tmp_inode->i_ctime = - cifs_NTtimeToUnix(le64_to_cpu(pData->LastStatusChange)); - - tmp_inode->i_mode = le64_to_cpu(pData->Permissions); - /* since we set the inode type below we need to mask off type - to avoid strange results if bits above were corrupt */ - tmp_inode->i_mode &= ~S_IFMT; - if (type == UNIX_FILE) { - *pobject_type = DT_REG; - tmp_inode->i_mode |= S_IFREG; - } else if (type == UNIX_SYMLINK) { - *pobject_type = DT_LNK; - tmp_inode->i_mode |= S_IFLNK; - } else if (type == UNIX_DIR) { - *pobject_type = DT_DIR; - tmp_inode->i_mode |= S_IFDIR; - } else if (type == UNIX_CHARDEV) { - *pobject_type = DT_CHR; - tmp_inode->i_mode |= S_IFCHR; - tmp_inode->i_rdev = MKDEV(le64_to_cpu(pData->DevMajor), - le64_to_cpu(pData->DevMinor) & MINORMASK); - } else if (type == UNIX_BLOCKDEV) { - *pobject_type = DT_BLK; - tmp_inode->i_mode |= S_IFBLK; - tmp_inode->i_rdev = MKDEV(le64_to_cpu(pData->DevMajor), - le64_to_cpu(pData->DevMinor) & MINORMASK); - } else if (type == UNIX_FIFO) { - *pobject_type = DT_FIFO; - tmp_inode->i_mode |= S_IFIFO; - } else if (type == UNIX_SOCKET) { - *pobject_type = DT_SOCK; - tmp_inode->i_mode |= S_IFSOCK; - } else { - /* safest to just call it a file */ - *pobject_type = DT_REG; - tmp_inode->i_mode |= S_IFREG; - cFYI(1, ("unknown inode type %d", type)); - } - -#ifdef CONFIG_CIFS_DEBUG2 - cFYI(1, ("object type: %d", type)); -#endif - tmp_inode->i_uid = le64_to_cpu(pData->Uid); - tmp_inode->i_gid = le64_to_cpu(pData->Gid); - tmp_inode->i_nlink = le64_to_cpu(pData->Nlinks); - - spin_lock(&tmp_inode->i_lock); - if (is_size_safe_to_change(cifsInfo, end_of_file)) { - /* can not safely change the file size here if the - client is writing to it due to potential races */ - i_size_write(tmp_inode, end_of_file); - - /* 512 bytes (2**9) is the fake blocksize that must be used */ - /* for this calculation, not the real blocksize */ - tmp_inode->i_blocks = (512 - 1 + num_of_bytes) >> 9; - } - spin_unlock(&tmp_inode->i_lock); - - if (S_ISREG(tmp_inode->i_mode)) { - cFYI(1, ("File inode")); - tmp_inode->i_op = &cifs_file_inode_ops; - - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) { - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) - tmp_inode->i_fop = &cifs_file_direct_nobrl_ops; - else - tmp_inode->i_fop = &cifs_file_direct_ops; - - } else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) - tmp_inode->i_fop = &cifs_file_nobrl_ops; - else - tmp_inode->i_fop = &cifs_file_ops; + cifs_unix_info_to_inode(tmp_inode, pData, 1); + cifs_set_ops(tmp_inode, false); - if ((cifs_sb->tcon) && (cifs_sb->tcon->ses) && - (cifs_sb->tcon->ses->server->maxBuf < - PAGE_CACHE_SIZE + MAX_CIFS_HDR_SIZE)) - tmp_inode->i_data.a_ops = &cifs_addr_ops_smallbuf; - else - tmp_inode->i_data.a_ops = &cifs_addr_ops; + if (!S_ISREG(tmp_inode->i_mode)) + return; - if (isNewInode) - return; /* No sense invalidating pages for new inode - since we we have not started caching - readahead file data yet */ + /* + * No sense invalidating pages for new inode + * since we we have not started caching + * readahead file data yet. + */ + if (isNewInode) + return; - if (timespec_equal(&tmp_inode->i_mtime, &local_mtime) && - (local_size == tmp_inode->i_size)) { - cFYI(1, ("inode exists but unchanged")); - } else { - /* file may have changed on server */ - cFYI(1, ("invalidate inode, readdir detected change")); - invalidate_remote_inode(tmp_inode); - } - } else if (S_ISDIR(tmp_inode->i_mode)) { - cFYI(1, ("Directory inode")); - tmp_inode->i_op = &cifs_dir_inode_ops; - tmp_inode->i_fop = &cifs_dir_ops; - } else if (S_ISLNK(tmp_inode->i_mode)) { - cFYI(1, ("Symbolic Link inode")); - tmp_inode->i_op = &cifs_symlink_inode_ops; -/* tmp_inode->i_fop = *//* do not need to set to anything */ + if (timespec_equal(&tmp_inode->i_mtime, &local_mtime) && + (local_size == tmp_inode->i_size)) { + cFYI(1, ("inode exists but unchanged")); } else { - cFYI(1, ("Special inode")); - init_special_inode(tmp_inode, tmp_inode->i_mode, - tmp_inode->i_rdev); + /* file may have changed on server */ + cFYI(1, ("invalidate inode, readdir detected change")); + invalidate_remote_inode(tmp_inode); } } @@ -931,7 +873,7 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) (CIFS_UNIX_POSIX_PATH_OPS_CAP & le64_to_cpu(pTcon->fsUnixInfo.Capability))) { u32 oplock = 0; - FILE_UNIX_BASIC_INFO * pInfo = + FILE_UNIX_BASIC_INFO *pInfo = kzalloc(sizeof(FILE_UNIX_BASIC_INFO), GFP_KERNEL); if (pInfo == NULL) { rc = -ENOMEM; @@ -951,7 +893,6 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) cFYI(1, ("posix mkdir returned 0x%x", rc)); d_drop(direntry); } else { - int obj_type; if (pInfo->Type == cpu_to_le32(-1)) { /* no return info, go query for it */ kfree(pInfo); @@ -987,7 +928,7 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) /* we already checked in POSIXCreate whether frame was long enough */ posix_fill_in_inode(direntry->d_inode, - pInfo, &obj_type, 1 /* NewInode */); + pInfo, 1 /* NewInode */); #ifdef CONFIG_CIFS_DEBUG2 cFYI(1, ("instantiated dentry %p %s to inode %p", direntry, direntry->d_name.name, newinode)); @@ -1015,7 +956,7 @@ mkdir_get_info: inode->i_sb, xid); else rc = cifs_get_inode_info(&newinode, full_path, NULL, - inode->i_sb, xid); + inode->i_sb, xid, NULL); if (pTcon->nocase) direntry->d_op = &cifs_ci_dentry_ops; @@ -1197,9 +1138,8 @@ int cifs_rename(struct inode *source_inode, struct dentry *source_direntry, } /* if we can not get memory just leave rc as EEXIST */ } - if (rc) { + if (rc) cFYI(1, ("rename rc %d", rc)); - } if ((rc == -EIO) || (rc == -EEXIST)) { int oplock = FALSE; @@ -1298,7 +1238,7 @@ int cifs_revalidate(struct dentry *direntry) } } else { rc = cifs_get_inode_info(&direntry->d_inode, full_path, NULL, - direntry->d_sb, xid); + direntry->d_sb, xid, NULL); if (rc) { cFYI(1, ("error on getting revalidate info %d", rc)); /* if (rc != -ENOENT) @@ -1385,7 +1325,7 @@ static int cifs_truncate_page(struct address_space *mapping, loff_t from) if (!page) return -ENOMEM; - zero_user_page(page, offset, PAGE_CACHE_SIZE - offset, KM_USER0); + zero_user_segment(page, offset, PAGE_CACHE_SIZE); unlock_page(page); page_cache_release(page); return rc; @@ -1487,11 +1427,10 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) } cifsInode = CIFS_I(direntry->d_inode); - /* BB check if we need to refresh inode from server now ? BB */ - - if (attrs->ia_valid & ATTR_SIZE) { + if ((attrs->ia_valid & ATTR_MTIME) || (attrs->ia_valid & ATTR_SIZE)) { /* - Flush data before changing file size on server. If the + Flush data before changing file size or changing the last + write time of the file on the server. If the flush returns error, store it to report later and continue. BB: This should be smarter. Why bother flushing pages that will be truncated anyway? Also, should we error out here if @@ -1502,7 +1441,9 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) CIFS_I(direntry->d_inode)->write_behind_rc = rc; rc = 0; } + } + if (attrs->ia_valid & ATTR_SIZE) { /* To avoid spurious oplock breaks from server, in the case of inodes that we already have open, avoid doing path based setting of file size if we can do it by handle. @@ -1607,7 +1548,14 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) CIFS_MOUNT_MAP_SPECIAL_CHR); else if (attrs->ia_valid & ATTR_MODE) { rc = 0; - if ((mode & S_IWUGO) == 0) /* not writeable */ { +#ifdef CONFIG_CIFS_EXPERIMENTAL + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) + rc = mode_to_acl(direntry->d_inode, full_path, mode); + else if ((mode & S_IWUGO) == 0) { +#else + if ((mode & S_IWUGO) == 0) { +#endif + /* not writeable */ if ((cifsInode->cifsAttrs & ATTR_READONLY) == 0) { set_dosattr = TRUE; time_buf.Attributes = @@ -1626,10 +1574,10 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) if (time_buf.Attributes == 0) time_buf.Attributes |= cpu_to_le32(ATTR_NORMAL); } - /* BB to be implemented - - via Windows security descriptors or streams */ - /* CIFSSMBWinSetPerms(xid, pTcon, full_path, mode, uid, gid, - cifs_sb->local_nls); */ +#ifdef CONFIG_CIFS_EXPERIMENTAL + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) + mode_to_acl(direntry->d_inode, full_path, mode); +#endif } if (attrs->ia_valid & ATTR_ATIME) { diff --git a/fs/cifs/ioctl.c b/fs/cifs/ioctl.c index d24fe6880a0..5c792df13d6 100644 --- a/fs/cifs/ioctl.c +++ b/fs/cifs/ioctl.c @@ -30,7 +30,7 @@ #define CIFS_IOC_CHECKUMOUNT _IO(0xCF, 2) -int cifs_ioctl (struct inode *inode, struct file *filep, +int cifs_ioctl(struct inode *inode, struct file *filep, unsigned int command, unsigned long arg) { int rc = -ENOTTY; /* strange error - but the precedent */ diff --git a/fs/cifs/link.c b/fs/cifs/link.c index 11f265726db..d4e7ec93285 100644 --- a/fs/cifs/link.c +++ b/fs/cifs/link.c @@ -1,7 +1,7 @@ /* * fs/cifs/link.c * - * Copyright (C) International Business Machines Corp., 2002,2003 + * Copyright (C) International Business Machines Corp., 2002,2008 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify @@ -205,7 +205,7 @@ cifs_symlink(struct inode *inode, struct dentry *direntry, const char *symname) inode->i_sb, xid); else rc = cifs_get_inode_info(&newinode, full_path, NULL, - inode->i_sb, xid); + inode->i_sb, xid, NULL); if (rc != 0) { cFYI(1, ("Create symlink ok, getinodeinfo fail rc = %d", @@ -236,8 +236,6 @@ cifs_readlink(struct dentry *direntry, char __user *pBuffer, int buflen) char *full_path = NULL; char *tmp_path = NULL; char *tmpbuffer; - unsigned char *referrals = NULL; - unsigned int num_referrals = 0; int len; __u16 fid; @@ -297,8 +295,11 @@ cifs_readlink(struct dentry *direntry, char __user *pBuffer, int buflen) cFYI(1, ("Error closing junction point " "(open for ioctl)")); } + /* BB unwind this long, nested function, or remove BB */ if (rc == -EIO) { /* Query if DFS Junction */ + unsigned int num_referrals = 0; + struct dfs_info3_param *refs = NULL; tmp_path = kmalloc(MAX_TREE_SIZE + MAX_PATHCONF + 1, GFP_KERNEL); @@ -310,7 +311,7 @@ cifs_readlink(struct dentry *direntry, char __user *pBuffer, int buflen) rc = get_dfs_path(xid, pTcon->ses, tmp_path, cifs_sb->local_nls, - &num_referrals, &referrals, + &num_referrals, &refs, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); cFYI(1, ("Get DFS for %s rc = %d ", @@ -320,14 +321,13 @@ cifs_readlink(struct dentry *direntry, char __user *pBuffer, int buflen) else { cFYI(1, ("num referral: %d", num_referrals)); - if (referrals) { - cFYI(1,("referral string: %s", referrals)); + if (refs && refs->path_name) { strncpy(tmpbuffer, - referrals, + refs->path_name, len-1); } } - kfree(referrals); + kfree(refs); kfree(tmp_path); } /* BB add code like else decode referrals diff --git a/fs/cifs/md4.c b/fs/cifs/md4.c index a2415c1a14d..a725c2609d6 100644 --- a/fs/cifs/md4.c +++ b/fs/cifs/md4.c @@ -56,7 +56,7 @@ lshift(__u32 x, int s) /* this applies md4 to 64 byte chunks */ static void -mdfour64(__u32 * M, __u32 * A, __u32 *B, __u32 * C, __u32 *D) +mdfour64(__u32 *M, __u32 *A, __u32 *B, __u32 *C, __u32 *D) { int j; __u32 AA, BB, CC, DD; @@ -137,7 +137,7 @@ mdfour64(__u32 * M, __u32 * A, __u32 *B, __u32 * C, __u32 *D) } static void -copy64(__u32 * M, unsigned char *in) +copy64(__u32 *M, unsigned char *in) { int i; diff --git a/fs/cifs/md5.c b/fs/cifs/md5.c index f13f96d42fc..462bbfefd4b 100644 --- a/fs/cifs/md5.c +++ b/fs/cifs/md5.c @@ -161,7 +161,7 @@ MD5Final(unsigned char digest[16], struct MD5Context *ctx) /* This is the central step in the MD5 algorithm. */ #define MD5STEP(f, w, x, y, z, data, s) \ - ( w += f(x, y, z) + data, w = w<<s | w>>(32-s), w += x ) + (w += f(x, y, z) + data, w = w<<s | w>>(32-s), w += x) /* * The core of the MD5 algorithm, this alters an existing MD5 hash to @@ -302,9 +302,8 @@ hmac_md5_init_limK_to_64(const unsigned char *key, int key_len, int i; /* if key is longer than 64 bytes truncate it */ - if (key_len > 64) { + if (key_len > 64) key_len = 64; - } /* start out by storing key in pads */ memset(ctx->k_ipad, 0, sizeof(ctx->k_ipad)); @@ -359,9 +358,9 @@ hmac_md5(unsigned char key[16], unsigned char *data, int data_len, { struct HMACMD5Context ctx; hmac_md5_init_limK_to_64(key, 16, &ctx); - if (data_len != 0) { + if (data_len != 0) hmac_md5_update(data, data_len, &ctx); - } + hmac_md5_final(digest, &ctx); } #endif diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 15546c2354c..2a42d9fedbb 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -1,7 +1,7 @@ /* * fs/cifs/misc.c * - * Copyright (C) International Business Machines Corp., 2002,2007 + * Copyright (C) International Business Machines Corp., 2002,2008 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify @@ -320,9 +320,9 @@ header_assemble(struct smb_hdr *buffer, char smb_command /* command */ , if (treeCon->ses) { if (treeCon->ses->capabilities & CAP_UNICODE) buffer->Flags2 |= SMBFLG2_UNICODE; - if (treeCon->ses->capabilities & CAP_STATUS32) { + if (treeCon->ses->capabilities & CAP_STATUS32) buffer->Flags2 |= SMBFLG2_ERR_STATUS; - } + /* Uid is not converted */ buffer->Uid = treeCon->ses->Suid; buffer->Mid = GetNextMid(treeCon->ses->server); @@ -610,7 +610,8 @@ dump_smb(struct smb_hdr *smb_buf, int smb_buf_length) buffer = (unsigned char *) smb_buf; for (i = 0, j = 0; i < smb_buf_length; i++, j++) { - if (i % 8 == 0) { /* have reached the beginning of line */ + if (i % 8 == 0) { + /* have reached the beginning of line */ printk(KERN_DEBUG "| "); j = 0; } @@ -621,7 +622,8 @@ dump_smb(struct smb_hdr *smb_buf, int smb_buf_length) else debug_line[1 + (2 * j)] = '_'; - if (i % 8 == 7) { /* reached end of line, time to print ascii */ + if (i % 8 == 7) { + /* reached end of line, time to print ascii */ debug_line[16] = 0; printk(" | %s\n", debug_line); } @@ -631,7 +633,7 @@ dump_smb(struct smb_hdr *smb_buf, int smb_buf_length) debug_line[2 * j] = ' '; debug_line[1 + (2 * j)] = ' '; } - printk( " | %s\n", debug_line); + printk(" | %s\n", debug_line); return; } diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c index 646e1f06941..3b5a5ce882b 100644 --- a/fs/cifs/netmisc.c +++ b/fs/cifs/netmisc.c @@ -1,7 +1,7 @@ /* * fs/cifs/netmisc.c * - * Copyright (c) International Business Machines Corp., 2002 + * Copyright (c) International Business Machines Corp., 2002,2008 * Author(s): Steve French (sfrench@us.ibm.com) * * Error mapping routines from Samba libsmb/errormap.c @@ -150,9 +150,7 @@ static int canonicalize_unc(char *cp) if (cp[i] == '\\') break; if (cp[i] == '/') { -#ifdef CONFIG_CIFS_DEBUG2 - cFYI(1, ("change slash to backslash in malformed UNC")); -#endif + cFYI(DBG2, ("change slash to \\ in malformed UNC")); cp[i] = '\\'; return 1; } @@ -178,9 +176,7 @@ cifs_inet_pton(int address_family, char *cp, void *dst) } else if (address_family == AF_INET6) { ret = in6_pton(cp, -1 /* len */, dst , '\\', NULL); } -#ifdef CONFIG_CIFS_DEBUG2 - cFYI(1, ("address conversion returned %d for %s", ret, cp)); -#endif + cFYI(DBG2, ("address conversion returned %d for %s", ret, cp)); if (ret > 0) ret = 1; return ret; @@ -253,7 +249,8 @@ static const struct { ERRDOS, 87, NT_STATUS_INVALID_PARAMETER_MIX}, { ERRHRD, ERRgeneral, NT_STATUS_INVALID_QUOTA_LOWER}, { ERRHRD, ERRgeneral, NT_STATUS_DISK_CORRUPT_ERROR}, { - ERRDOS, ERRbadfile, NT_STATUS_OBJECT_NAME_INVALID}, { /* mapping changed since shell does lookup on * and expects file not found */ + /* mapping changed since shell does lookup on * expects FileNotFound */ + ERRDOS, ERRbadfile, NT_STATUS_OBJECT_NAME_INVALID}, { ERRDOS, ERRbadfile, NT_STATUS_OBJECT_NAME_NOT_FOUND}, { ERRDOS, ERRalreadyexists, NT_STATUS_OBJECT_NAME_COLLISION}, { ERRHRD, ERRgeneral, NT_STATUS_HANDLE_NOT_WAITABLE}, { @@ -820,7 +817,8 @@ map_smb_to_linux_error(struct smb_hdr *smb, int logErr) /* old style errors */ /* DOS class smb error codes - map DOS */ - if (smberrclass == ERRDOS) { /* 1 byte field no need to byte reverse */ + if (smberrclass == ERRDOS) { + /* 1 byte field no need to byte reverse */ for (i = 0; i < sizeof(mapping_table_ERRDOS) / @@ -834,7 +832,8 @@ map_smb_to_linux_error(struct smb_hdr *smb, int logErr) } /* else try next error mapping one to see if match */ } - } else if (smberrclass == ERRSRV) { /* server class of error codes */ + } else if (smberrclass == ERRSRV) { + /* server class of error codes */ for (i = 0; i < sizeof(mapping_table_ERRSRV) / @@ -922,8 +921,8 @@ struct timespec cnvrtDosUnixTm(__u16 date, __u16 time) { struct timespec ts; int sec, min, days, month, year; - SMB_TIME * st = (SMB_TIME *)&time; - SMB_DATE * sd = (SMB_DATE *)&date; + SMB_TIME *st = (SMB_TIME *)&time; + SMB_DATE *sd = (SMB_DATE *)&date; cFYI(1, ("date %d time %d", date, time)); diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 0f22def4bdf..32b445edc88 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -3,7 +3,7 @@ * * Directory search handling * - * Copyright (C) International Business Machines Corp., 2004, 2007 + * Copyright (C) International Business Machines Corp., 2004, 2008 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify @@ -42,17 +42,18 @@ static void dump_cifs_file_struct(struct file *file, char *label) cFYI(1, ("empty cifs private file data")); return; } - if (cf->invalidHandle) { + if (cf->invalidHandle) cFYI(1, ("invalid handle")); - } - if (cf->srch_inf.endOfSearch) { + if (cf->srch_inf.endOfSearch) cFYI(1, ("end of search")); - } - if (cf->srch_inf.emptyDir) { + if (cf->srch_inf.emptyDir) cFYI(1, ("empty dir")); - } } } +#else +static inline void dump_cifs_file_struct(struct file *file, char *label) +{ +} #endif /* DEBUG2 */ /* Returns one if new inode created (which therefore needs to be hashed) */ @@ -150,7 +151,7 @@ static void fill_in_inode(struct inode *tmp_inode, int new_buf_type, cifs_NTtimeToUnix(le64_to_cpu(pfindData->ChangeTime)); } else { /* legacy, OS2 and DOS style */ /* struct timespec ts;*/ - FIND_FILE_STANDARD_INFO * pfindData = + FIND_FILE_STANDARD_INFO *pfindData = (FIND_FILE_STANDARD_INFO *)buf; tmp_inode->i_mtime = cnvrtDosUnixTm( @@ -198,9 +199,8 @@ static void fill_in_inode(struct inode *tmp_inode, int new_buf_type, if (attr & ATTR_DIRECTORY) { *pobject_type = DT_DIR; /* override default perms since we do not lock dirs */ - if (atomic_read(&cifsInfo->inUse) == 0) { + if (atomic_read(&cifsInfo->inUse) == 0) tmp_inode->i_mode = cifs_sb->mnt_dir_mode; - } tmp_inode->i_mode |= S_IFDIR; } else if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) && (attr & ATTR_SYSTEM)) { @@ -231,9 +231,8 @@ static void fill_in_inode(struct inode *tmp_inode, int new_buf_type, } /* could add code here - to validate if device or weird share type? */ /* can not fill in nlink here as in qpathinfo version and Unx search */ - if (atomic_read(&cifsInfo->inUse) == 0) { + if (atomic_read(&cifsInfo->inUse) == 0) atomic_set(&cifsInfo->inUse, 1); - } spin_lock(&tmp_inode->i_lock); if (is_size_safe_to_change(cifsInfo, end_of_file)) { @@ -461,9 +460,8 @@ static int initiate_cifs_search(const int xid, struct file *file) full_path = build_path_from_dentry(file->f_path.dentry); - if (full_path == NULL) { + if (full_path == NULL) return -ENOMEM; - } cFYI(1, ("Full path: %s start at: %lld", full_path, file->f_pos)); @@ -471,9 +469,9 @@ ffirst_retry: /* test for Unix extensions */ /* but now check for them on the share/mount not on the SMB session */ /* if (pTcon->ses->capabilities & CAP_UNIX) { */ - if (pTcon->unix_ext) { + if (pTcon->unix_ext) cifsFile->srch_inf.info_level = SMB_FIND_FILE_UNIX; - } else if ((pTcon->ses->capabilities & + else if ((pTcon->ses->capabilities & (CAP_NT_SMBS | CAP_NT_FIND)) == 0) { cifsFile->srch_inf.info_level = SMB_FIND_FILE_INFO_STANDARD; } else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) { @@ -514,10 +512,10 @@ static int cifs_unicode_bytelen(char *str) static char *nxt_dir_entry(char *old_entry, char *end_of_smb, int level) { char *new_entry; - FILE_DIRECTORY_INFO * pDirInfo = (FILE_DIRECTORY_INFO *)old_entry; + FILE_DIRECTORY_INFO *pDirInfo = (FILE_DIRECTORY_INFO *)old_entry; if (level == SMB_FIND_FILE_INFO_STANDARD) { - FIND_FILE_STANDARD_INFO * pfData; + FIND_FILE_STANDARD_INFO *pfData; pfData = (FIND_FILE_STANDARD_INFO *)pDirInfo; new_entry = old_entry + sizeof(FIND_FILE_STANDARD_INFO) + @@ -553,7 +551,7 @@ static int cifs_entry_is_dot(char *current_entry, struct cifsFileInfo *cfile) int len = 0; if (cfile->srch_inf.info_level == SMB_FIND_FILE_UNIX) { - FILE_UNIX_INFO * pFindData = (FILE_UNIX_INFO *)current_entry; + FILE_UNIX_INFO *pFindData = (FILE_UNIX_INFO *)current_entry; filename = &pFindData->FileName[0]; if (cfile->srch_inf.unicode) { len = cifs_unicode_bytelen(filename); @@ -562,30 +560,30 @@ static int cifs_entry_is_dot(char *current_entry, struct cifsFileInfo *cfile) len = strnlen(filename, 5); } } else if (cfile->srch_inf.info_level == SMB_FIND_FILE_DIRECTORY_INFO) { - FILE_DIRECTORY_INFO * pFindData = + FILE_DIRECTORY_INFO *pFindData = (FILE_DIRECTORY_INFO *)current_entry; filename = &pFindData->FileName[0]; len = le32_to_cpu(pFindData->FileNameLength); } else if (cfile->srch_inf.info_level == SMB_FIND_FILE_FULL_DIRECTORY_INFO) { - FILE_FULL_DIRECTORY_INFO * pFindData = + FILE_FULL_DIRECTORY_INFO *pFindData = (FILE_FULL_DIRECTORY_INFO *)current_entry; filename = &pFindData->FileName[0]; len = le32_to_cpu(pFindData->FileNameLength); } else if (cfile->srch_inf.info_level == SMB_FIND_FILE_ID_FULL_DIR_INFO) { - SEARCH_ID_FULL_DIR_INFO * pFindData = + SEARCH_ID_FULL_DIR_INFO *pFindData = (SEARCH_ID_FULL_DIR_INFO *)current_entry; filename = &pFindData->FileName[0]; len = le32_to_cpu(pFindData->FileNameLength); } else if (cfile->srch_inf.info_level == SMB_FIND_FILE_BOTH_DIRECTORY_INFO) { - FILE_BOTH_DIRECTORY_INFO * pFindData = + FILE_BOTH_DIRECTORY_INFO *pFindData = (FILE_BOTH_DIRECTORY_INFO *)current_entry; filename = &pFindData->FileName[0]; len = le32_to_cpu(pFindData->FileNameLength); } else if (cfile->srch_inf.info_level == SMB_FIND_FILE_INFO_STANDARD) { - FIND_FILE_STANDARD_INFO * pFindData = + FIND_FILE_STANDARD_INFO *pFindData = (FIND_FILE_STANDARD_INFO *)current_entry; filename = &pFindData->FileName[0]; len = pFindData->FileNameLength; @@ -666,9 +664,7 @@ static int find_cifs_entry(const int xid, struct cifsTconInfo *pTcon, . and .. for the root of a drive and for those we need to start two entries earlier */ -#ifdef CONFIG_CIFS_DEBUG2 dump_cifs_file_struct(file, "In fce "); -#endif if (((index_to_find < cifsFile->srch_inf.index_of_last_entry) && is_dir_changed(file)) || (index_to_find < first_entry_in_buffer)) { @@ -718,7 +714,7 @@ static int find_cifs_entry(const int xid, struct cifsTconInfo *pTcon, pos_in_buf = index_to_find - first_entry_in_buffer; cFYI(1, ("found entry - pos_in_buf %d", pos_in_buf)); - for (i=0; (i < (pos_in_buf)) && (current_entry != NULL); i++) { + for (i = 0; (i < (pos_in_buf)) && (current_entry != NULL); i++) { /* go entry by entry figuring out which is first */ current_entry = nxt_dir_entry(current_entry, end_of_smb, cifsFile->srch_inf.info_level); @@ -793,7 +789,7 @@ static int cifs_get_name_from_search_buf(struct qstr *pqst, filename = &pFindData->FileName[0]; len = le32_to_cpu(pFindData->FileNameLength); } else if (level == SMB_FIND_FILE_INFO_STANDARD) { - FIND_FILE_STANDARD_INFO * pFindData = + FIND_FILE_STANDARD_INFO *pFindData = (FIND_FILE_STANDARD_INFO *)current_entry; filename = &pFindData->FileName[0]; /* one byte length, no name conversion */ @@ -928,7 +924,7 @@ static int cifs_save_resume_key(const char *current_entry, level = cifsFile->srch_inf.info_level; if (level == SMB_FIND_FILE_UNIX) { - FILE_UNIX_INFO * pFindData = (FILE_UNIX_INFO *)current_entry; + FILE_UNIX_INFO *pFindData = (FILE_UNIX_INFO *)current_entry; filename = &pFindData->FileName[0]; if (cifsFile->srch_inf.unicode) { diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index d0cb469daab..ed150efbe27 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -417,10 +417,6 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, calc_lanman_hash(ses, lnm_session_key); ses->flags |= CIFS_SES_LANMAN; -/* #ifdef CONFIG_CIFS_DEBUG2 - cifs_dump_mem("cryptkey: ",ses->server->cryptKey, - CIFS_SESS_KEY_SIZE); -#endif */ memcpy(bcc_ptr, (char *)lnm_session_key, CIFS_SESS_KEY_SIZE); bcc_ptr += CIFS_SESS_KEY_SIZE; @@ -528,9 +524,11 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, rc = -EOVERFLOW; goto ssetup_exit; } - ses->server->mac_signing_key.len = msg->sesskey_len; - memcpy(ses->server->mac_signing_key.data.krb5, msg->data, - msg->sesskey_len); + if (first_time) { + ses->server->mac_signing_key.len = msg->sesskey_len; + memcpy(ses->server->mac_signing_key.data.krb5, + msg->data, msg->sesskey_len); + } pSMB->req.hdr.Flags2 |= SMBFLG2_EXT_SEC; capabilities |= CAP_EXTENDED_SECURITY; pSMB->req.Capabilities = cpu_to_le32(capabilities); @@ -540,7 +538,7 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, if (ses->capabilities & CAP_UNICODE) { /* unicode strings must be word aligned */ - if (iov[0].iov_len % 2) { + if ((iov[0].iov_len + iov[1].iov_len) % 2) { *bcc_ptr = 0; bcc_ptr++; } diff --git a/fs/cifs/smbdes.c b/fs/cifs/smbdes.c index cfa6d21fb4e..04943c976f9 100644 --- a/fs/cifs/smbdes.c +++ b/fs/cifs/smbdes.c @@ -114,42 +114,42 @@ static uchar sbox[8][4][16] = { {{14, 4, 13, 1, 2, 15, 11, 8, 3, 10, 6, 12, 5, 9, 0, 7}, {0, 15, 7, 4, 14, 2, 13, 1, 10, 6, 12, 11, 9, 5, 3, 8}, {4, 1, 14, 8, 13, 6, 2, 11, 15, 12, 9, 7, 3, 10, 5, 0}, - {15, 12, 8, 2, 4, 9, 1, 7, 5, 11, 3, 14, 10, 0, 6, 13}}, + {15, 12, 8, 2, 4, 9, 1, 7, 5, 11, 3, 14, 10, 0, 6, 13} }, {{15, 1, 8, 14, 6, 11, 3, 4, 9, 7, 2, 13, 12, 0, 5, 10}, {3, 13, 4, 7, 15, 2, 8, 14, 12, 0, 1, 10, 6, 9, 11, 5}, {0, 14, 7, 11, 10, 4, 13, 1, 5, 8, 12, 6, 9, 3, 2, 15}, - {13, 8, 10, 1, 3, 15, 4, 2, 11, 6, 7, 12, 0, 5, 14, 9}}, + {13, 8, 10, 1, 3, 15, 4, 2, 11, 6, 7, 12, 0, 5, 14, 9} }, {{10, 0, 9, 14, 6, 3, 15, 5, 1, 13, 12, 7, 11, 4, 2, 8}, {13, 7, 0, 9, 3, 4, 6, 10, 2, 8, 5, 14, 12, 11, 15, 1}, {13, 6, 4, 9, 8, 15, 3, 0, 11, 1, 2, 12, 5, 10, 14, 7}, - {1, 10, 13, 0, 6, 9, 8, 7, 4, 15, 14, 3, 11, 5, 2, 12}}, + {1, 10, 13, 0, 6, 9, 8, 7, 4, 15, 14, 3, 11, 5, 2, 12} }, {{7, 13, 14, 3, 0, 6, 9, 10, 1, 2, 8, 5, 11, 12, 4, 15}, {13, 8, 11, 5, 6, 15, 0, 3, 4, 7, 2, 12, 1, 10, 14, 9}, {10, 6, 9, 0, 12, 11, 7, 13, 15, 1, 3, 14, 5, 2, 8, 4}, - {3, 15, 0, 6, 10, 1, 13, 8, 9, 4, 5, 11, 12, 7, 2, 14}}, + {3, 15, 0, 6, 10, 1, 13, 8, 9, 4, 5, 11, 12, 7, 2, 14} }, {{2, 12, 4, 1, 7, 10, 11, 6, 8, 5, 3, 15, 13, 0, 14, 9}, {14, 11, 2, 12, 4, 7, 13, 1, 5, 0, 15, 10, 3, 9, 8, 6}, {4, 2, 1, 11, 10, 13, 7, 8, 15, 9, 12, 5, 6, 3, 0, 14}, - {11, 8, 12, 7, 1, 14, 2, 13, 6, 15, 0, 9, 10, 4, 5, 3}}, + {11, 8, 12, 7, 1, 14, 2, 13, 6, 15, 0, 9, 10, 4, 5, 3} }, {{12, 1, 10, 15, 9, 2, 6, 8, 0, 13, 3, 4, 14, 7, 5, 11}, {10, 15, 4, 2, 7, 12, 9, 5, 6, 1, 13, 14, 0, 11, 3, 8}, {9, 14, 15, 5, 2, 8, 12, 3, 7, 0, 4, 10, 1, 13, 11, 6}, - {4, 3, 2, 12, 9, 5, 15, 10, 11, 14, 1, 7, 6, 0, 8, 13}}, + {4, 3, 2, 12, 9, 5, 15, 10, 11, 14, 1, 7, 6, 0, 8, 13} }, {{4, 11, 2, 14, 15, 0, 8, 13, 3, 12, 9, 7, 5, 10, 6, 1}, {13, 0, 11, 7, 4, 9, 1, 10, 14, 3, 5, 12, 2, 15, 8, 6}, {1, 4, 11, 13, 12, 3, 7, 14, 10, 15, 6, 8, 0, 5, 9, 2}, - {6, 11, 13, 8, 1, 4, 10, 7, 9, 5, 0, 15, 14, 2, 3, 12}}, + {6, 11, 13, 8, 1, 4, 10, 7, 9, 5, 0, 15, 14, 2, 3, 12} }, {{13, 2, 8, 4, 6, 15, 11, 1, 10, 9, 3, 14, 5, 0, 12, 7}, {1, 15, 13, 8, 10, 3, 7, 4, 12, 5, 6, 11, 0, 14, 9, 2}, {7, 11, 4, 1, 9, 12, 14, 2, 0, 6, 10, 13, 15, 3, 5, 8}, - {2, 1, 14, 7, 4, 10, 8, 13, 15, 12, 9, 0, 3, 5, 6, 11}} + {2, 1, 14, 7, 4, 10, 8, 13, 15, 12, 9, 0, 3, 5, 6, 11} } }; static void @@ -313,9 +313,8 @@ str_to_key(unsigned char *str, unsigned char *key) key[5] = ((str[4] & 0x1F) << 2) | (str[5] >> 6); key[6] = ((str[5] & 0x3F) << 1) | (str[6] >> 7); key[7] = str[6] & 0x7F; - for (i = 0; i < 8; i++) { + for (i = 0; i < 8; i++) key[i] = (key[i] << 1); - } } static void @@ -344,9 +343,8 @@ smbhash(unsigned char *out, unsigned char *in, unsigned char *key, int forw) dohash(outb, inb, keyb, forw); - for (i = 0; i < 8; i++) { + for (i = 0; i < 8; i++) out[i] = 0; - } for (i = 0; i < 64; i++) { if (outb[i]) diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 50b623ad932..3612d6c0a0b 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -1,7 +1,7 @@ /* * fs/cifs/transport.c * - * Copyright (C) International Business Machines Corp., 2002,2007 + * Copyright (C) International Business Machines Corp., 2002,2008 * Author(s): Steve French (sfrench@us.ibm.com) * Jeremy Allison (jra@samba.org) 2006. * @@ -358,9 +358,9 @@ static int allocate_mid(struct cifsSesInfo *ses, struct smb_hdr *in_buf, } else if (ses->status != CifsGood) { /* check if SMB session is bad because we are setting it up */ if ((in_buf->Command != SMB_COM_SESSION_SETUP_ANDX) && - (in_buf->Command != SMB_COM_NEGOTIATE)) { + (in_buf->Command != SMB_COM_NEGOTIATE)) return -EAGAIN; - } /* else ok - we are setting up session */ + /* else ok - we are setting up session */ } *ppmidQ = AllocMidQEntry(in_buf, ses); if (*ppmidQ == NULL) @@ -437,9 +437,8 @@ SendReceiveNoRsp(const unsigned int xid, struct cifsSesInfo *ses, iov[0].iov_len = in_buf->smb_buf_length + 4; flags |= CIFS_NO_RESP; rc = SendReceive2(xid, ses, iov, 1, &resp_buf_type, flags); -#ifdef CONFIG_CIFS_DEBUG2 - cFYI(1, ("SendRcvNoR flags %d rc %d", flags, rc)); -#endif + cFYI(DBG2, ("SendRcvNoRsp flags %d rc %d", flags, rc)); + return rc; } diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c index 54e8ef96cb7..8cd6a445b01 100644 --- a/fs/cifs/xattr.c +++ b/fs/cifs/xattr.c @@ -139,9 +139,9 @@ int cifs_setxattr(struct dentry *direntry, const char *ea_name, } else if (strncmp(ea_name, CIFS_XATTR_USER_PREFIX, 5) == 0) { if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_XATTR) goto set_ea_exit; - if (strncmp(ea_name, CIFS_XATTR_DOS_ATTRIB, 14) == 0) { + if (strncmp(ea_name, CIFS_XATTR_DOS_ATTRIB, 14) == 0) cFYI(1, ("attempt to set cifs inode metadata")); - } + ea_name += 5; /* skip past user. prefix */ rc = CIFSSMBSetEA(xid, pTcon, full_path, ea_name, ea_value, (__u16)value_size, cifs_sb->local_nls, @@ -262,7 +262,7 @@ ssize_t cifs_getxattr(struct dentry *direntry, const char *ea_name, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); #ifdef CONFIG_CIFS_EXPERIMENTAL - else if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) { + else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) { __u16 fid; int oplock = FALSE; struct cifs_ntsd *pacl = NULL; @@ -303,11 +303,10 @@ ssize_t cifs_getxattr(struct dentry *direntry, const char *ea_name, } else if (strncmp(ea_name, CIFS_XATTR_SECURITY_PREFIX, XATTR_SECURITY_PREFIX_LEN) == 0) { cFYI(1, ("Security xattr namespace not supported yet")); - } else { + } else cFYI(1, ("illegal xattr request %s (only user namespace supported)", ea_name)); - } /* We could add an additional check for streams ie if proc/fs/cifs/streamstoxattr is set then diff --git a/fs/coda/pioctl.c b/fs/coda/pioctl.c index 2bf3026adc8..c21a1f552a6 100644 --- a/fs/coda/pioctl.c +++ b/fs/coda/pioctl.c @@ -75,12 +75,12 @@ static int coda_pioctl(struct inode * inode, struct file * filp, if ( error ) { return error; } else { - target_inode = nd.dentry->d_inode; + target_inode = nd.path.dentry->d_inode; } /* return if it is not a Coda inode */ if ( target_inode->i_sb != inode->i_sb ) { - path_release(&nd); + path_put(&nd.path); return -EINVAL; } @@ -89,7 +89,7 @@ static int coda_pioctl(struct inode * inode, struct file * filp, error = venus_pioctl(inode->i_sb, &(cnp->c_fid), cmd, &data); - path_release(&nd); + path_put(&nd.path); return error; } diff --git a/fs/coda/psdev.c b/fs/coda/psdev.c index dcc6aead70f..e3eb3556622 100644 --- a/fs/coda/psdev.c +++ b/fs/coda/psdev.c @@ -362,8 +362,8 @@ static int init_coda_psdev(void) goto out_chrdev; } for (i = 0; i < MAX_CODADEVS; i++) - class_device_create(coda_psdev_class, NULL, - MKDEV(CODA_PSDEV_MAJOR,i), NULL, "cfs%d", i); + device_create(coda_psdev_class, NULL, + MKDEV(CODA_PSDEV_MAJOR,i), "cfs%d", i); coda_sysctl_init(); goto out; @@ -405,7 +405,7 @@ static int __init init_coda(void) return 0; out: for (i = 0; i < MAX_CODADEVS; i++) - class_device_destroy(coda_psdev_class, MKDEV(CODA_PSDEV_MAJOR, i)); + device_destroy(coda_psdev_class, MKDEV(CODA_PSDEV_MAJOR, i)); class_destroy(coda_psdev_class); unregister_chrdev(CODA_PSDEV_MAJOR, "coda"); coda_sysctl_clean(); @@ -424,7 +424,7 @@ static void __exit exit_coda(void) printk("coda: failed to unregister filesystem\n"); } for (i = 0; i < MAX_CODADEVS; i++) - class_device_destroy(coda_psdev_class, MKDEV(CODA_PSDEV_MAJOR, i)); + device_destroy(coda_psdev_class, MKDEV(CODA_PSDEV_MAJOR, i)); class_destroy(coda_psdev_class); unregister_chrdev(CODA_PSDEV_MAJOR, "coda"); coda_sysctl_clean(); diff --git a/fs/compat.c b/fs/compat.c index 15078ce4c04..2ce4456aad3 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -241,10 +241,10 @@ asmlinkage long compat_sys_statfs(const char __user *path, struct compat_statfs error = user_path_walk(path, &nd); if (!error) { struct kstatfs tmp; - error = vfs_statfs(nd.dentry, &tmp); + error = vfs_statfs(nd.path.dentry, &tmp); if (!error) error = put_compat_statfs(buf, &tmp); - path_release(&nd); + path_put(&nd.path); } return error; } @@ -309,10 +309,10 @@ asmlinkage long compat_sys_statfs64(const char __user *path, compat_size_t sz, s error = user_path_walk(path, &nd); if (!error) { struct kstatfs tmp; - error = vfs_statfs(nd.dentry, &tmp); + error = vfs_statfs(nd.path.dentry, &tmp); if (!error) error = put_compat_statfs64(buf, &tmp); - path_release(&nd); + path_put(&nd.path); } return error; } @@ -702,9 +702,6 @@ static int do_nfs4_super_data_conv(void *raw_data) real->flags = raw->flags; real->version = raw->version; } - else { - return -EINVAL; - } return 0; } @@ -1104,10 +1101,6 @@ static ssize_t compat_do_readv_writev(int type, struct file *file, if (ret < 0) goto out; - ret = security_file_permission(file, type == READ ? MAY_READ:MAY_WRITE); - if (ret) - goto out; - fnv = NULL; if (type == READ) { fn = file->f_op->read; @@ -2087,51 +2080,6 @@ long asmlinkage compat_sys_nfsservctl(int cmd, void *notused, void *notused2) #ifdef CONFIG_EPOLL -#ifdef CONFIG_HAS_COMPAT_EPOLL_EVENT -asmlinkage long compat_sys_epoll_ctl(int epfd, int op, int fd, - struct compat_epoll_event __user *event) -{ - long err = 0; - struct compat_epoll_event user; - struct epoll_event __user *kernel = NULL; - - if (event) { - if (copy_from_user(&user, event, sizeof(user))) - return -EFAULT; - kernel = compat_alloc_user_space(sizeof(struct epoll_event)); - err |= __put_user(user.events, &kernel->events); - err |= __put_user(user.data, &kernel->data); - } - - return err ? err : sys_epoll_ctl(epfd, op, fd, kernel); -} - - -asmlinkage long compat_sys_epoll_wait(int epfd, - struct compat_epoll_event __user *events, - int maxevents, int timeout) -{ - long i, ret, err = 0; - struct epoll_event __user *kbuf; - struct epoll_event ev; - - if ((maxevents <= 0) || - (maxevents > (INT_MAX / sizeof(struct epoll_event)))) - return -EINVAL; - kbuf = compat_alloc_user_space(sizeof(struct epoll_event) * maxevents); - ret = sys_epoll_wait(epfd, kbuf, maxevents, timeout); - for (i = 0; i < ret; i++) { - err |= __get_user(ev.events, &kbuf[i].events); - err |= __get_user(ev.data, &kbuf[i].data); - err |= __put_user(ev.events, &events->events); - err |= __put_user_unaligned(ev.data, &events->data); - events++; - } - - return err ? -EFAULT: ret; -} -#endif /* CONFIG_HAS_COMPAT_EPOLL_EVENT */ - #ifdef TIF_RESTORE_SIGMASK asmlinkage long compat_sys_epoll_pwait(int epfd, struct compat_epoll_event __user *events, @@ -2157,11 +2105,7 @@ asmlinkage long compat_sys_epoll_pwait(int epfd, sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); } -#ifdef CONFIG_HAS_COMPAT_EPOLL_EVENT - err = compat_sys_epoll_wait(epfd, events, maxevents, timeout); -#else err = sys_epoll_wait(epfd, events, maxevents, timeout); -#endif /* * If we changed the signal mask, we need to restore the original one. @@ -2210,19 +2154,41 @@ asmlinkage long compat_sys_signalfd(int ufd, #ifdef CONFIG_TIMERFD -asmlinkage long compat_sys_timerfd(int ufd, int clockid, int flags, - const struct compat_itimerspec __user *utmr) +asmlinkage long compat_sys_timerfd_settime(int ufd, int flags, + const struct compat_itimerspec __user *utmr, + struct compat_itimerspec __user *otmr) { + int error; struct itimerspec t; struct itimerspec __user *ut; if (get_compat_itimerspec(&t, utmr)) return -EFAULT; - ut = compat_alloc_user_space(sizeof(*ut)); - if (copy_to_user(ut, &t, sizeof(t))) + ut = compat_alloc_user_space(2 * sizeof(struct itimerspec)); + if (copy_to_user(&ut[0], &t, sizeof(t))) return -EFAULT; + error = sys_timerfd_settime(ufd, flags, &ut[0], &ut[1]); + if (!error && otmr) + error = (copy_from_user(&t, &ut[1], sizeof(struct itimerspec)) || + put_compat_itimerspec(otmr, &t)) ? -EFAULT: 0; + + return error; +} - return sys_timerfd(ufd, clockid, flags, ut); +asmlinkage long compat_sys_timerfd_gettime(int ufd, + struct compat_itimerspec __user *otmr) +{ + int error; + struct itimerspec t; + struct itimerspec __user *ut; + + ut = compat_alloc_user_space(sizeof(struct itimerspec)); + error = sys_timerfd_gettime(ufd, ut); + if (!error) + error = (copy_from_user(&t, ut, sizeof(struct itimerspec)) || + put_compat_itimerspec(otmr, &t)) ? -EFAULT: 0; + + return error; } #endif /* CONFIG_TIMERFD */ diff --git a/fs/compat_binfmt_elf.c b/fs/compat_binfmt_elf.c new file mode 100644 index 00000000000..0adced2f296 --- /dev/null +++ b/fs/compat_binfmt_elf.c @@ -0,0 +1,131 @@ +/* + * 32-bit compatibility support for ELF format executables and core dumps. + * + * Copyright (C) 2007 Red Hat, Inc. All rights reserved. + * + * This copyrighted material is made available to anyone wishing to use, + * modify, copy, or redistribute it subject to the terms and conditions + * of the GNU General Public License v.2. + * + * Red Hat Author: Roland McGrath. + * + * This file is used in a 64-bit kernel that wants to support 32-bit ELF. + * asm/elf.h is responsible for defining the compat_* and COMPAT_* macros + * used below, with definitions appropriate for 32-bit ABI compatibility. + * + * We use macros to rename the ABI types and machine-dependent + * functions used in binfmt_elf.c to compat versions. + */ + +#include <linux/elfcore-compat.h> +#include <linux/time.h> + +/* + * Rename the basic ELF layout types to refer to the 32-bit class of files. + */ +#undef ELF_CLASS +#define ELF_CLASS ELFCLASS32 + +#undef elfhdr +#undef elf_phdr +#undef elf_note +#undef elf_addr_t +#define elfhdr elf32_hdr +#define elf_phdr elf32_phdr +#define elf_note elf32_note +#define elf_addr_t Elf32_Addr + +/* + * The machine-dependent core note format types are defined in elfcore-compat.h, + * which requires asm/elf.h to define compat_elf_gregset_t et al. + */ +#define elf_prstatus compat_elf_prstatus +#define elf_prpsinfo compat_elf_prpsinfo + +/* + * Compat version of cputime_to_compat_timeval, perhaps this + * should be an inline in <linux/compat.h>. + */ +static void cputime_to_compat_timeval(const cputime_t cputime, + struct compat_timeval *value) +{ + struct timeval tv; + cputime_to_timeval(cputime, &tv); + value->tv_sec = tv.tv_sec; + value->tv_usec = tv.tv_usec; +} + +#undef cputime_to_timeval +#define cputime_to_timeval cputime_to_compat_timeval + + +/* + * To use this file, asm/elf.h must define compat_elf_check_arch. + * The other following macros can be defined if the compat versions + * differ from the native ones, or omitted when they match. + */ + +#undef ELF_ARCH +#undef elf_check_arch +#define elf_check_arch compat_elf_check_arch + +#ifdef COMPAT_ELF_PLATFORM +#undef ELF_PLATFORM +#define ELF_PLATFORM COMPAT_ELF_PLATFORM +#endif + +#ifdef COMPAT_ELF_HWCAP +#undef ELF_HWCAP +#define ELF_HWCAP COMPAT_ELF_HWCAP +#endif + +#ifdef COMPAT_ARCH_DLINFO +#undef ARCH_DLINFO +#define ARCH_DLINFO COMPAT_ARCH_DLINFO +#endif + +#ifdef COMPAT_ELF_ET_DYN_BASE +#undef ELF_ET_DYN_BASE +#define ELF_ET_DYN_BASE COMPAT_ELF_ET_DYN_BASE +#endif + +#ifdef COMPAT_ELF_EXEC_PAGESIZE +#undef ELF_EXEC_PAGESIZE +#define ELF_EXEC_PAGESIZE COMPAT_ELF_EXEC_PAGESIZE +#endif + +#ifdef COMPAT_ELF_PLAT_INIT +#undef ELF_PLAT_INIT +#define ELF_PLAT_INIT COMPAT_ELF_PLAT_INIT +#endif + +#ifdef COMPAT_SET_PERSONALITY +#undef SET_PERSONALITY +#define SET_PERSONALITY COMPAT_SET_PERSONALITY +#endif + +#ifdef compat_start_thread +#undef start_thread +#define start_thread compat_start_thread +#endif + +#ifdef compat_arch_setup_additional_pages +#undef ARCH_HAS_SETUP_ADDITIONAL_PAGES +#define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1 +#undef arch_setup_additional_pages +#define arch_setup_additional_pages compat_arch_setup_additional_pages +#endif + +/* + * Rename a few of the symbols that binfmt_elf.c will define. + * These are all local so the names don't really matter, but it + * might make some debugging less confusing not to duplicate them. + */ +#define elf_format compat_elf_format +#define init_elf_binfmt init_compat_elf_binfmt +#define exit_elf_binfmt exit_compat_elf_binfmt + +/* + * We share all the actual code with the native (64-bit) version. + */ +#include "binfmt_elf.c" diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index da8cb3b3592..c6e72aebd16 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -78,7 +78,6 @@ #include <linux/mii.h> #include <linux/if_bonding.h> #include <linux/watchdog.h> -#include <linux/dm-ioctl.h> #include <linux/soundcard.h> #include <linux/lp.h> @@ -1376,7 +1375,7 @@ static int do_atm_ioctl(unsigned int fd, unsigned int cmd32, unsigned long arg) return -EINVAL; } -static __attribute_used__ int +static __used int ret_einval(unsigned int fd, unsigned int cmd, unsigned long arg) { return -EINVAL; @@ -1993,39 +1992,6 @@ COMPATIBLE_IOCTL(STOP_ARRAY_RO) COMPATIBLE_IOCTL(RESTART_ARRAY_RW) COMPATIBLE_IOCTL(GET_BITMAP_FILE) ULONG_IOCTL(SET_BITMAP_FILE) -/* DM */ -COMPATIBLE_IOCTL(DM_VERSION_32) -COMPATIBLE_IOCTL(DM_REMOVE_ALL_32) -COMPATIBLE_IOCTL(DM_LIST_DEVICES_32) -COMPATIBLE_IOCTL(DM_DEV_CREATE_32) -COMPATIBLE_IOCTL(DM_DEV_REMOVE_32) -COMPATIBLE_IOCTL(DM_DEV_RENAME_32) -COMPATIBLE_IOCTL(DM_DEV_SUSPEND_32) -COMPATIBLE_IOCTL(DM_DEV_STATUS_32) -COMPATIBLE_IOCTL(DM_DEV_WAIT_32) -COMPATIBLE_IOCTL(DM_TABLE_LOAD_32) -COMPATIBLE_IOCTL(DM_TABLE_CLEAR_32) -COMPATIBLE_IOCTL(DM_TABLE_DEPS_32) -COMPATIBLE_IOCTL(DM_TABLE_STATUS_32) -COMPATIBLE_IOCTL(DM_LIST_VERSIONS_32) -COMPATIBLE_IOCTL(DM_TARGET_MSG_32) -COMPATIBLE_IOCTL(DM_DEV_SET_GEOMETRY_32) -COMPATIBLE_IOCTL(DM_VERSION) -COMPATIBLE_IOCTL(DM_REMOVE_ALL) -COMPATIBLE_IOCTL(DM_LIST_DEVICES) -COMPATIBLE_IOCTL(DM_DEV_CREATE) -COMPATIBLE_IOCTL(DM_DEV_REMOVE) -COMPATIBLE_IOCTL(DM_DEV_RENAME) -COMPATIBLE_IOCTL(DM_DEV_SUSPEND) -COMPATIBLE_IOCTL(DM_DEV_STATUS) -COMPATIBLE_IOCTL(DM_DEV_WAIT) -COMPATIBLE_IOCTL(DM_TABLE_LOAD) -COMPATIBLE_IOCTL(DM_TABLE_CLEAR) -COMPATIBLE_IOCTL(DM_TABLE_DEPS) -COMPATIBLE_IOCTL(DM_TABLE_STATUS) -COMPATIBLE_IOCTL(DM_LIST_VERSIONS) -COMPATIBLE_IOCTL(DM_TARGET_MSG) -COMPATIBLE_IOCTL(DM_DEV_SET_GEOMETRY) /* Big K */ COMPATIBLE_IOCTL(PIO_FONT) COMPATIBLE_IOCTL(GIO_FONT) @@ -2887,7 +2853,7 @@ static void compat_ioctl_error(struct file *filp, unsigned int fd, /* find the name of the device. */ path = (char *)__get_free_page(GFP_KERNEL); if (path) { - fn = d_path(filp->f_path.dentry, filp->f_path.mnt, path, PAGE_SIZE); + fn = d_path(&filp->f_path, path, PAGE_SIZE); if (IS_ERR(fn)) fn = "?"; } @@ -2986,7 +2952,7 @@ asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd, } do_ioctl: - error = vfs_ioctl(filp, fd, cmd, arg); + error = do_vfs_ioctl(filp, fd, cmd, arg); out_fput: fput_light(filp, fput_needed); out: diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c index 50ed691098b..a48dc7dd876 100644 --- a/fs/configfs/dir.c +++ b/fs/configfs/dir.c @@ -546,7 +546,7 @@ static int populate_groups(struct config_group *group) * That said, taking our i_mutex is closer to mkdir * emulation, and shouldn't hurt. */ - mutex_lock(&dentry->d_inode->i_mutex); + mutex_lock_nested(&dentry->d_inode->i_mutex, I_MUTEX_CHILD); for (i = 0; group->default_groups[i]; i++) { new_group = group->default_groups[i]; @@ -1405,7 +1405,8 @@ int configfs_register_subsystem(struct configfs_subsystem *subsys) sd = configfs_sb->s_root->d_fsdata; link_group(to_config_group(sd->s_element), group); - mutex_lock(&configfs_sb->s_root->d_inode->i_mutex); + mutex_lock_nested(&configfs_sb->s_root->d_inode->i_mutex, + I_MUTEX_PARENT); name.name = group->cg_item.ci_name; name.len = strlen(name.name); diff --git a/fs/configfs/file.c b/fs/configfs/file.c index a3658f9a082..397cb503a18 100644 --- a/fs/configfs/file.c +++ b/fs/configfs/file.c @@ -320,7 +320,7 @@ int configfs_add_file(struct dentry * dir, const struct configfs_attribute * att umode_t mode = (attr->ca_mode & S_IALLUGO) | S_IFREG; int error = 0; - mutex_lock(&dir->d_inode->i_mutex); + mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_NORMAL); error = configfs_make_dirent(parent_sd, NULL, (void *) attr, mode, type); mutex_unlock(&dir->d_inode->i_mutex); diff --git a/fs/configfs/mount.c b/fs/configfs/mount.c index 3bf0278ea84..de3b31d0a37 100644 --- a/fs/configfs/mount.c +++ b/fs/configfs/mount.c @@ -128,7 +128,7 @@ void configfs_release_fs(void) } -static decl_subsys(config, NULL, NULL); +static struct kobject *config_kobj; static int __init configfs_init(void) { @@ -140,9 +140,8 @@ static int __init configfs_init(void) if (!configfs_dir_cachep) goto out; - kobj_set_kset_s(&config_subsys, kernel_subsys); - err = subsystem_register(&config_subsys); - if (err) { + config_kobj = kobject_create_and_add("config", kernel_kobj); + if (!config_kobj) { kmem_cache_destroy(configfs_dir_cachep); configfs_dir_cachep = NULL; goto out; @@ -151,7 +150,7 @@ static int __init configfs_init(void) err = register_filesystem(&configfs_fs_type); if (err) { printk(KERN_ERR "configfs: Unable to register filesystem!\n"); - subsystem_unregister(&config_subsys); + kobject_put(config_kobj); kmem_cache_destroy(configfs_dir_cachep); configfs_dir_cachep = NULL; goto out; @@ -160,7 +159,7 @@ static int __init configfs_init(void) err = configfs_inode_init(); if (err) { unregister_filesystem(&configfs_fs_type); - subsystem_unregister(&config_subsys); + kobject_put(config_kobj); kmem_cache_destroy(configfs_dir_cachep); configfs_dir_cachep = NULL; } @@ -171,7 +170,7 @@ out: static void __exit configfs_exit(void) { unregister_filesystem(&configfs_fs_type); - subsystem_unregister(&config_subsys); + kobject_put(config_kobj); kmem_cache_destroy(configfs_dir_cachep); configfs_dir_cachep = NULL; configfs_inode_exit(); diff --git a/fs/configfs/symlink.c b/fs/configfs/symlink.c index 22700d2857d..78929ea84ff 100644 --- a/fs/configfs/symlink.c +++ b/fs/configfs/symlink.c @@ -99,11 +99,11 @@ static int get_target(const char *symname, struct nameidata *nd, ret = path_lookup(symname, LOOKUP_FOLLOW|LOOKUP_DIRECTORY, nd); if (!ret) { - if (nd->dentry->d_sb == configfs_sb) { - *target = configfs_get_config_item(nd->dentry); + if (nd->path.dentry->d_sb == configfs_sb) { + *target = configfs_get_config_item(nd->path.dentry); if (!*target) { ret = -ENOENT; - path_release(nd); + path_put(&nd->path); } } else ret = -EPERM; @@ -141,7 +141,7 @@ int configfs_symlink(struct inode *dir, struct dentry *dentry, const char *symna ret = create_link(parent_item, target_item, dentry); config_item_put(target_item); - path_release(&nd); + path_put(&nd.path); out_put: config_item_put(parent_item); diff --git a/fs/dcache.c b/fs/dcache.c index d9ca1e5ceb9..43455776711 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -89,12 +89,20 @@ static void d_free(struct dentry *dentry) if (dentry->d_op && dentry->d_op->d_release) dentry->d_op->d_release(dentry); /* if dentry was never inserted into hash, immediate free is OK */ - if (dentry->d_hash.pprev == NULL) + if (hlist_unhashed(&dentry->d_hash)) __d_free(dentry); else call_rcu(&dentry->d_u.d_rcu, d_callback); } +static void dentry_lru_remove(struct dentry *dentry) +{ + if (!list_empty(&dentry->d_lru)) { + list_del_init(&dentry->d_lru); + dentry_stat.nr_unused--; + } +} + /* * Release the dentry's inode, using the filesystem * d_iput() operation if defined. @@ -211,13 +219,7 @@ repeat: unhash_it: __d_drop(dentry); kill_it: - /* If dentry was on d_lru list - * delete it from there - */ - if (!list_empty(&dentry->d_lru)) { - list_del(&dentry->d_lru); - dentry_stat.nr_unused--; - } + dentry_lru_remove(dentry); dentry = d_kill(dentry); if (dentry) goto repeat; @@ -285,10 +287,7 @@ int d_invalidate(struct dentry * dentry) static inline struct dentry * __dget_locked(struct dentry *dentry) { atomic_inc(&dentry->d_count); - if (!list_empty(&dentry->d_lru)) { - dentry_stat.nr_unused--; - list_del_init(&dentry->d_lru); - } + dentry_lru_remove(dentry); return dentry; } @@ -404,10 +403,7 @@ static void prune_one_dentry(struct dentry * dentry) if (dentry->d_op && dentry->d_op->d_delete) dentry->d_op->d_delete(dentry); - if (!list_empty(&dentry->d_lru)) { - list_del(&dentry->d_lru); - dentry_stat.nr_unused--; - } + dentry_lru_remove(dentry); __d_drop(dentry); dentry = d_kill(dentry); spin_lock(&dcache_lock); @@ -596,10 +592,7 @@ static void shrink_dcache_for_umount_subtree(struct dentry *dentry) /* detach this root from the system */ spin_lock(&dcache_lock); - if (!list_empty(&dentry->d_lru)) { - dentry_stat.nr_unused--; - list_del_init(&dentry->d_lru); - } + dentry_lru_remove(dentry); __d_drop(dentry); spin_unlock(&dcache_lock); @@ -613,11 +606,7 @@ static void shrink_dcache_for_umount_subtree(struct dentry *dentry) spin_lock(&dcache_lock); list_for_each_entry(loop, &dentry->d_subdirs, d_u.d_child) { - if (!list_empty(&loop->d_lru)) { - dentry_stat.nr_unused--; - list_del_init(&loop->d_lru); - } - + dentry_lru_remove(loop); __d_drop(loop); cond_resched_lock(&dcache_lock); } @@ -799,10 +788,7 @@ resume: struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child); next = tmp->next; - if (!list_empty(&dentry->d_lru)) { - dentry_stat.nr_unused--; - list_del_init(&dentry->d_lru); - } + dentry_lru_remove(dentry); /* * move only zero ref count dentries to the end * of the unused list for prune_dcache @@ -1408,9 +1394,6 @@ void d_delete(struct dentry * dentry) if (atomic_read(&dentry->d_count) == 1) { dentry_iput(dentry); fsnotify_nameremove(dentry, isdir); - - /* remove this and other inotify debug checks after 2.6.18 */ - dentry->d_flags &= ~DCACHE_INOTIFY_PARENT_WATCHED; return; } @@ -1779,9 +1762,8 @@ shouldnt_be_hashed: * * "buflen" should be positive. Caller holds the dcache_lock. */ -static char * __d_path( struct dentry *dentry, struct vfsmount *vfsmnt, - struct dentry *root, struct vfsmount *rootmnt, - char *buffer, int buflen) +static char *__d_path(struct dentry *dentry, struct vfsmount *vfsmnt, + struct path *root, char *buffer, int buflen) { char * end = buffer+buflen; char * retval; @@ -1806,7 +1788,7 @@ static char * __d_path( struct dentry *dentry, struct vfsmount *vfsmnt, for (;;) { struct dentry * parent; - if (dentry == root && vfsmnt == rootmnt) + if (dentry == root->dentry && vfsmnt == root->mnt) break; if (dentry == vfsmnt->mnt_root || IS_ROOT(dentry)) { /* Global root? */ @@ -1847,13 +1829,23 @@ Elong: return ERR_PTR(-ENAMETOOLONG); } -/* write full pathname into buffer and return start of pathname */ -char * d_path(struct dentry *dentry, struct vfsmount *vfsmnt, - char *buf, int buflen) +/** + * d_path - return the path of a dentry + * @path: path to report + * @buf: buffer to return value in + * @buflen: buffer length + * + * Convert a dentry into an ASCII path name. If the entry has been deleted + * the string " (deleted)" is appended. Note that this is ambiguous. + * + * Returns the buffer or an error code if the path was too long. + * + * "buflen" should be positive. Caller holds the dcache_lock. + */ +char *d_path(struct path *path, char *buf, int buflen) { char *res; - struct vfsmount *rootmnt; - struct dentry *root; + struct path root; /* * We have various synthetic filesystems that never get mounted. On @@ -1862,18 +1854,17 @@ char * d_path(struct dentry *dentry, struct vfsmount *vfsmnt, * user wants to identify the object in /proc/pid/fd/. The little hack * below allows us to generate a name for these objects on demand: */ - if (dentry->d_op && dentry->d_op->d_dname) - return dentry->d_op->d_dname(dentry, buf, buflen); + if (path->dentry->d_op && path->dentry->d_op->d_dname) + return path->dentry->d_op->d_dname(path->dentry, buf, buflen); read_lock(¤t->fs->lock); - rootmnt = mntget(current->fs->rootmnt); - root = dget(current->fs->root); + root = current->fs->root; + path_get(¤t->fs->root); read_unlock(¤t->fs->lock); spin_lock(&dcache_lock); - res = __d_path(dentry, vfsmnt, root, rootmnt, buf, buflen); + res = __d_path(path->dentry, path->mnt, &root, buf, buflen); spin_unlock(&dcache_lock); - dput(root); - mntput(rootmnt); + path_put(&root); return res; } @@ -1919,28 +1910,27 @@ char *dynamic_dname(struct dentry *dentry, char *buffer, int buflen, asmlinkage long sys_getcwd(char __user *buf, unsigned long size) { int error; - struct vfsmount *pwdmnt, *rootmnt; - struct dentry *pwd, *root; + struct path pwd, root; char *page = (char *) __get_free_page(GFP_USER); if (!page) return -ENOMEM; read_lock(¤t->fs->lock); - pwdmnt = mntget(current->fs->pwdmnt); - pwd = dget(current->fs->pwd); - rootmnt = mntget(current->fs->rootmnt); - root = dget(current->fs->root); + pwd = current->fs->pwd; + path_get(¤t->fs->pwd); + root = current->fs->root; + path_get(¤t->fs->root); read_unlock(¤t->fs->lock); error = -ENOENT; /* Has the current directory has been unlinked? */ spin_lock(&dcache_lock); - if (pwd->d_parent == pwd || !d_unhashed(pwd)) { + if (pwd.dentry->d_parent == pwd.dentry || !d_unhashed(pwd.dentry)) { unsigned long len; char * cwd; - cwd = __d_path(pwd, pwdmnt, root, rootmnt, page, PAGE_SIZE); + cwd = __d_path(pwd.dentry, pwd.mnt, &root, page, PAGE_SIZE); spin_unlock(&dcache_lock); error = PTR_ERR(cwd); @@ -1958,10 +1948,8 @@ asmlinkage long sys_getcwd(char __user *buf, unsigned long size) spin_unlock(&dcache_lock); out: - dput(pwd); - mntput(pwdmnt); - dput(root); - mntput(rootmnt); + path_put(&pwd); + path_put(&root); free_page((unsigned long) page); return error; } diff --git a/fs/dcookies.c b/fs/dcookies.c index 792cbf55fa9..855d4b1d619 100644 --- a/fs/dcookies.c +++ b/fs/dcookies.c @@ -24,6 +24,7 @@ #include <linux/errno.h> #include <linux/dcookies.h> #include <linux/mutex.h> +#include <linux/path.h> #include <asm/uaccess.h> /* The dcookies are allocated from a kmem_cache and @@ -31,8 +32,7 @@ * code here is particularly performance critical */ struct dcookie_struct { - struct dentry * dentry; - struct vfsmount * vfsmnt; + struct path path; struct list_head hash_list; }; @@ -51,7 +51,7 @@ static inline int is_live(void) /* The dentry is locked, its address will do for the cookie */ static inline unsigned long dcookie_value(struct dcookie_struct * dcs) { - return (unsigned long)dcs->dentry; + return (unsigned long)dcs->path.dentry; } @@ -89,19 +89,17 @@ static void hash_dcookie(struct dcookie_struct * dcs) } -static struct dcookie_struct * alloc_dcookie(struct dentry * dentry, - struct vfsmount * vfsmnt) +static struct dcookie_struct *alloc_dcookie(struct path *path) { - struct dcookie_struct * dcs = kmem_cache_alloc(dcookie_cache, GFP_KERNEL); + struct dcookie_struct *dcs = kmem_cache_alloc(dcookie_cache, + GFP_KERNEL); if (!dcs) return NULL; - dentry->d_cookie = dcs; - - dcs->dentry = dget(dentry); - dcs->vfsmnt = mntget(vfsmnt); + path->dentry->d_cookie = dcs; + dcs->path = *path; + path_get(path); hash_dcookie(dcs); - return dcs; } @@ -109,8 +107,7 @@ static struct dcookie_struct * alloc_dcookie(struct dentry * dentry, /* This is the main kernel-side routine that retrieves the cookie * value for a dentry/vfsmnt pair. */ -int get_dcookie(struct dentry * dentry, struct vfsmount * vfsmnt, - unsigned long * cookie) +int get_dcookie(struct path *path, unsigned long *cookie) { int err = 0; struct dcookie_struct * dcs; @@ -122,10 +119,10 @@ int get_dcookie(struct dentry * dentry, struct vfsmount * vfsmnt, goto out; } - dcs = dentry->d_cookie; + dcs = path->dentry->d_cookie; if (!dcs) - dcs = alloc_dcookie(dentry, vfsmnt); + dcs = alloc_dcookie(path); if (!dcs) { err = -ENOMEM; @@ -174,7 +171,7 @@ asmlinkage long sys_lookup_dcookie(u64 cookie64, char __user * buf, size_t len) goto out; /* FIXME: (deleted) ? */ - path = d_path(dcs->dentry, dcs->vfsmnt, kbuf, PAGE_SIZE); + path = d_path(&dcs->path, kbuf, PAGE_SIZE); if (IS_ERR(path)) { err = PTR_ERR(path); @@ -254,9 +251,8 @@ out_kmem: static void free_dcookie(struct dcookie_struct * dcs) { - dcs->dentry->d_cookie = NULL; - dput(dcs->dentry); - mntput(dcs->vfsmnt); + dcs->path.dentry->d_cookie = NULL; + path_put(&dcs->path); kmem_cache_free(dcookie_cache, dcs); } diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c index fa6b7f7ff91..fddffe4851f 100644 --- a/fs/debugfs/file.c +++ b/fs/debugfs/file.c @@ -56,13 +56,15 @@ const struct inode_operations debugfs_link_operations = { .follow_link = debugfs_follow_link, }; -static void debugfs_u8_set(void *data, u64 val) +static int debugfs_u8_set(void *data, u64 val) { *(u8 *)data = val; + return 0; } -static u64 debugfs_u8_get(void *data) +static int debugfs_u8_get(void *data, u64 *val) { - return *(u8 *)data; + *val = *(u8 *)data; + return 0; } DEFINE_SIMPLE_ATTRIBUTE(fops_u8, debugfs_u8_get, debugfs_u8_set, "%llu\n"); @@ -97,13 +99,15 @@ struct dentry *debugfs_create_u8(const char *name, mode_t mode, } EXPORT_SYMBOL_GPL(debugfs_create_u8); -static void debugfs_u16_set(void *data, u64 val) +static int debugfs_u16_set(void *data, u64 val) { *(u16 *)data = val; + return 0; } -static u64 debugfs_u16_get(void *data) +static int debugfs_u16_get(void *data, u64 *val) { - return *(u16 *)data; + *val = *(u16 *)data; + return 0; } DEFINE_SIMPLE_ATTRIBUTE(fops_u16, debugfs_u16_get, debugfs_u16_set, "%llu\n"); @@ -138,13 +142,15 @@ struct dentry *debugfs_create_u16(const char *name, mode_t mode, } EXPORT_SYMBOL_GPL(debugfs_create_u16); -static void debugfs_u32_set(void *data, u64 val) +static int debugfs_u32_set(void *data, u64 val) { *(u32 *)data = val; + return 0; } -static u64 debugfs_u32_get(void *data) +static int debugfs_u32_get(void *data, u64 *val) { - return *(u32 *)data; + *val = *(u32 *)data; + return 0; } DEFINE_SIMPLE_ATTRIBUTE(fops_u32, debugfs_u32_get, debugfs_u32_set, "%llu\n"); @@ -179,14 +185,16 @@ struct dentry *debugfs_create_u32(const char *name, mode_t mode, } EXPORT_SYMBOL_GPL(debugfs_create_u32); -static void debugfs_u64_set(void *data, u64 val) +static int debugfs_u64_set(void *data, u64 val) { *(u64 *)data = val; + return 0; } -static u64 debugfs_u64_get(void *data) +static int debugfs_u64_get(void *data, u64 *val) { - return *(u64 *)data; + *val = *(u64 *)data; + return 0; } DEFINE_SIMPLE_ATTRIBUTE(fops_u64, debugfs_u64_get, debugfs_u64_set, "%llu\n"); diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index 6a713b33992..e9602d85c11 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -29,10 +29,6 @@ #define DEBUGFS_MAGIC 0x64626720 -/* declared over in file.c */ -extern struct file_operations debugfs_file_operations; -extern struct inode_operations debugfs_link_operations; - static struct vfsmount *debugfs_mount; static int debugfs_mount_count; @@ -426,20 +422,19 @@ exit: } EXPORT_SYMBOL_GPL(debugfs_rename); -static decl_subsys(debug, NULL, NULL); +static struct kobject *debug_kobj; static int __init debugfs_init(void) { int retval; - kobj_set_kset_s(&debug_subsys, kernel_subsys); - retval = subsystem_register(&debug_subsys); - if (retval) - return retval; + debug_kobj = kobject_create_and_add("debug", kernel_kobj); + if (!debug_kobj) + return -EINVAL; retval = register_filesystem(&debug_fs_type); if (retval) - subsystem_unregister(&debug_subsys); + kobject_put(debug_kobj); return retval; } @@ -447,7 +442,7 @@ static void __exit debugfs_exit(void) { simple_release_fs(&debugfs_mount, &debugfs_mount_count); unregister_filesystem(&debug_fs_type); - subsystem_unregister(&debug_subsys); + kobject_put(debug_kobj); } core_initcall(debugfs_init); diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c index 06ef9a255c7..f120e120787 100644 --- a/fs/devpts/inode.c +++ b/fs/devpts/inode.c @@ -20,9 +20,12 @@ #include <linux/devpts_fs.h> #include <linux/parser.h> #include <linux/fsnotify.h> +#include <linux/seq_file.h> #define DEVPTS_SUPER_MAGIC 0x1cd1 +#define DEVPTS_DEFAULT_MODE 0600 + static struct vfsmount *devpts_mnt; static struct dentry *devpts_root; @@ -32,7 +35,7 @@ static struct { uid_t uid; gid_t gid; umode_t mode; -} config = {.mode = 0600}; +} config = {.mode = DEVPTS_DEFAULT_MODE}; enum { Opt_uid, Opt_gid, Opt_mode, @@ -54,7 +57,7 @@ static int devpts_remount(struct super_block *sb, int *flags, char *data) config.setgid = 0; config.uid = 0; config.gid = 0; - config.mode = 0600; + config.mode = DEVPTS_DEFAULT_MODE; while ((p = strsep(&data, ",")) != NULL) { substring_t args[MAX_OPT_ARGS]; @@ -81,7 +84,7 @@ static int devpts_remount(struct super_block *sb, int *flags, char *data) case Opt_mode: if (match_octal(&args[0], &option)) return -EINVAL; - config.mode = option & ~S_IFMT; + config.mode = option & S_IALLUGO; break; default: printk(KERN_ERR "devpts: called with bogus options\n"); @@ -92,9 +95,21 @@ static int devpts_remount(struct super_block *sb, int *flags, char *data) return 0; } +static int devpts_show_options(struct seq_file *seq, struct vfsmount *vfs) +{ + if (config.setuid) + seq_printf(seq, ",uid=%u", config.uid); + if (config.setgid) + seq_printf(seq, ",gid=%u", config.gid); + seq_printf(seq, ",mode=%03o", config.mode); + + return 0; +} + static const struct super_operations devpts_sops = { .statfs = simple_statfs, .remount_fs = devpts_remount, + .show_options = devpts_show_options, }; static int diff --git a/fs/direct-io.c b/fs/direct-io.c index acf0da1bd25..9e81addbd6e 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -878,8 +878,8 @@ do_holes: page_cache_release(page); goto out; } - zero_user_page(page, block_in_page << blkbits, - 1 << blkbits, KM_USER0); + zero_user(page, block_in_page << blkbits, + 1 << blkbits); dio->block_in_file++; block_in_page++; goto next_block; diff --git a/fs/dlm/ast.c b/fs/dlm/ast.c index 6308122890c..8bf31e3fbf0 100644 --- a/fs/dlm/ast.c +++ b/fs/dlm/ast.c @@ -39,7 +39,6 @@ void dlm_add_ast(struct dlm_lkb *lkb, int type) dlm_user_add_ast(lkb, type); return; } - DLM_ASSERT(lkb->lkb_astaddr != DLM_FAKE_USER_AST, dlm_print_lkb(lkb);); spin_lock(&ast_queue_lock); if (!(lkb->lkb_ast_type & (AST_COMP | AST_BAST))) { @@ -58,8 +57,8 @@ static void process_asts(void) struct dlm_ls *ls = NULL; struct dlm_rsb *r = NULL; struct dlm_lkb *lkb; - void (*cast) (long param); - void (*bast) (long param, int mode); + void (*cast) (void *astparam); + void (*bast) (void *astparam, int mode); int type = 0, found, bmode; for (;;) { @@ -83,8 +82,8 @@ static void process_asts(void) if (!found) break; - cast = lkb->lkb_astaddr; - bast = lkb->lkb_bastaddr; + cast = lkb->lkb_astfn; + bast = lkb->lkb_bastfn; bmode = lkb->lkb_bastmode; if ((type & AST_COMP) && cast) diff --git a/fs/dlm/config.c b/fs/dlm/config.c index 2f8e3c81bc1..c3ad1dff3b2 100644 --- a/fs/dlm/config.c +++ b/fs/dlm/config.c @@ -604,7 +604,7 @@ static struct clusters clusters_root = { }, }; -int dlm_config_init(void) +int __init dlm_config_init(void) { config_group_init(&clusters_root.subsys.su_group); mutex_init(&clusters_root.subsys.su_mutex); diff --git a/fs/dlm/debug_fs.c b/fs/dlm/debug_fs.c index 12c3bfd5e66..8fc24f4507a 100644 --- a/fs/dlm/debug_fs.c +++ b/fs/dlm/debug_fs.c @@ -162,14 +162,12 @@ static int print_resource(struct dlm_rsb *res, struct seq_file *s) static void print_lock(struct seq_file *s, struct dlm_lkb *lkb, struct dlm_rsb *r) { - struct dlm_user_args *ua; unsigned int waiting = 0; uint64_t xid = 0; if (lkb->lkb_flags & DLM_IFL_USER) { - ua = (struct dlm_user_args *) lkb->lkb_astparam; - if (ua) - xid = ua->xid; + if (lkb->lkb_ua) + xid = lkb->lkb_ua->xid; } if (lkb->lkb_timestamp) @@ -543,7 +541,7 @@ void dlm_delete_debug_file(struct dlm_ls *ls) debugfs_remove(ls->ls_debug_locks_dentry); } -int dlm_register_debugfs(void) +int __init dlm_register_debugfs(void) { mutex_init(&debug_buf_lock); dlm_root = debugfs_create_dir("dlm", NULL); diff --git a/fs/dlm/dir.c b/fs/dlm/dir.c index 46754553fdc..85defeb64df 100644 --- a/fs/dlm/dir.c +++ b/fs/dlm/dir.c @@ -49,7 +49,7 @@ static struct dlm_direntry *get_free_de(struct dlm_ls *ls, int len) spin_unlock(&ls->ls_recover_list_lock); if (!found) - de = allocate_direntry(ls, len); + de = kzalloc(sizeof(struct dlm_direntry) + len, GFP_KERNEL); return de; } @@ -62,7 +62,7 @@ void dlm_clear_free_entries(struct dlm_ls *ls) de = list_entry(ls->ls_recover_list.next, struct dlm_direntry, list); list_del(&de->list); - free_direntry(de); + kfree(de); } spin_unlock(&ls->ls_recover_list_lock); } @@ -171,7 +171,7 @@ void dlm_dir_remove_entry(struct dlm_ls *ls, int nodeid, char *name, int namelen } list_del(&de->list); - free_direntry(de); + kfree(de); out: write_unlock(&ls->ls_dirtbl[bucket].lock); } @@ -220,6 +220,7 @@ int dlm_recover_directory(struct dlm_ls *ls) last_len = 0; for (;;) { + int left; error = dlm_recovery_stopped(ls); if (error) goto out_free; @@ -235,12 +236,21 @@ int dlm_recover_directory(struct dlm_ls *ls) * pick namelen/name pairs out of received buffer */ - b = ls->ls_recover_buf + sizeof(struct dlm_rcom); + b = ls->ls_recover_buf->rc_buf; + left = ls->ls_recover_buf->rc_header.h_length; + left -= sizeof(struct dlm_rcom); for (;;) { - memcpy(&namelen, b, sizeof(uint16_t)); - namelen = be16_to_cpu(namelen); - b += sizeof(uint16_t); + __be16 v; + + error = -EINVAL; + if (left < sizeof(__be16)) + goto out_free; + + memcpy(&v, b, sizeof(__be16)); + namelen = be16_to_cpu(v); + b += sizeof(__be16); + left -= sizeof(__be16); /* namelen of 0xFFFFF marks end of names for this node; namelen of 0 marks end of the @@ -251,6 +261,12 @@ int dlm_recover_directory(struct dlm_ls *ls) if (!namelen) break; + if (namelen > left) + goto out_free; + + if (namelen > DLM_RESNAME_MAXLEN) + goto out_free; + error = -ENOMEM; de = get_free_de(ls, namelen); if (!de) @@ -262,6 +278,7 @@ int dlm_recover_directory(struct dlm_ls *ls) memcpy(de->name, b, namelen); memcpy(last_name, b, namelen); b += namelen; + left -= namelen; add_entry_to_hash(ls, de); count++; @@ -302,7 +319,10 @@ static int get_entry(struct dlm_ls *ls, int nodeid, char *name, write_unlock(&ls->ls_dirtbl[bucket].lock); - de = allocate_direntry(ls, namelen); + if (namelen > DLM_RESNAME_MAXLEN) + return -EINVAL; + + de = kzalloc(sizeof(struct dlm_direntry) + namelen, GFP_KERNEL); if (!de) return -ENOMEM; @@ -313,7 +333,7 @@ static int get_entry(struct dlm_ls *ls, int nodeid, char *name, write_lock(&ls->ls_dirtbl[bucket].lock); tmp = search_bucket(ls, name, namelen, bucket); if (tmp) { - free_direntry(de); + kfree(de); de = tmp; } else { list_add_tail(&de->list, &ls->ls_dirtbl[bucket].list); @@ -329,49 +349,47 @@ int dlm_dir_lookup(struct dlm_ls *ls, int nodeid, char *name, int namelen, return get_entry(ls, nodeid, name, namelen, r_nodeid); } -/* Copy the names of master rsb's into the buffer provided. - Only select names whose dir node is the given nodeid. */ +static struct dlm_rsb *find_rsb_root(struct dlm_ls *ls, char *name, int len) +{ + struct dlm_rsb *r; + + down_read(&ls->ls_root_sem); + list_for_each_entry(r, &ls->ls_root_list, res_root_list) { + if (len == r->res_length && !memcmp(name, r->res_name, len)) { + up_read(&ls->ls_root_sem); + return r; + } + } + up_read(&ls->ls_root_sem); + return NULL; +} + +/* Find the rsb where we left off (or start again), then send rsb names + for rsb's we're master of and whose directory node matches the requesting + node. inbuf is the rsb name last sent, inlen is the name's length */ void dlm_copy_master_names(struct dlm_ls *ls, char *inbuf, int inlen, char *outbuf, int outlen, int nodeid) { struct list_head *list; - struct dlm_rsb *start_r = NULL, *r = NULL; - int offset = 0, start_namelen, error, dir_nodeid; - char *start_name; + struct dlm_rsb *r; + int offset = 0, dir_nodeid; uint16_t be_namelen; - /* - * Find the rsb where we left off (or start again) - */ - - start_namelen = inlen; - start_name = inbuf; - - if (start_namelen > 1) { - /* - * We could also use a find_rsb_root() function here that - * searched the ls_root_list. - */ - error = dlm_find_rsb(ls, start_name, start_namelen, R_MASTER, - &start_r); - DLM_ASSERT(!error && start_r, - printk("error %d\n", error);); - DLM_ASSERT(!list_empty(&start_r->res_root_list), - dlm_print_rsb(start_r);); - dlm_put_rsb(start_r); - } - - /* - * Send rsb names for rsb's we're master of and whose directory node - * matches the requesting node. - */ - down_read(&ls->ls_root_sem); - if (start_r) - list = start_r->res_root_list.next; - else + + if (inlen > 1) { + r = find_rsb_root(ls, inbuf, inlen); + if (!r) { + inbuf[inlen - 1] = '\0'; + log_error(ls, "copy_master_names from %d start %d %s", + nodeid, inlen, inbuf); + goto out; + } + list = r->res_root_list.next; + } else { list = ls->ls_root_list.next; + } for (offset = 0; list != &ls->ls_root_list; list = list->next) { r = list_entry(list, struct dlm_rsb, res_root_list); diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index d2fc2384c3b..d30ea8b433a 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -92,8 +92,6 @@ do { \ } \ } -#define DLM_FAKE_USER_AST ERR_PTR(-EINVAL) - struct dlm_direntry { struct list_head list; @@ -146,9 +144,9 @@ struct dlm_recover { struct dlm_args { uint32_t flags; - void *astaddr; - long astparam; - void *bastaddr; + void (*astfn) (void *astparam); + void *astparam; + void (*bastfn) (void *astparam, int mode); int mode; struct dlm_lksb *lksb; unsigned long timeout; @@ -253,9 +251,12 @@ struct dlm_lkb { char *lkb_lvbptr; struct dlm_lksb *lkb_lksb; /* caller's status block */ - void *lkb_astaddr; /* caller's ast function */ - void *lkb_bastaddr; /* caller's bast function */ - long lkb_astparam; /* caller's ast arg */ + void (*lkb_astfn) (void *astparam); + void (*lkb_bastfn) (void *astparam, int mode); + union { + void *lkb_astparam; /* caller's ast arg */ + struct dlm_user_args *lkb_ua; + }; }; @@ -403,28 +404,34 @@ struct dlm_rcom { char rc_buf[0]; }; +union dlm_packet { + struct dlm_header header; /* common to other two */ + struct dlm_message message; + struct dlm_rcom rcom; +}; + struct rcom_config { - uint32_t rf_lvblen; - uint32_t rf_lsflags; - uint64_t rf_unused; + __le32 rf_lvblen; + __le32 rf_lsflags; + __le64 rf_unused; }; struct rcom_lock { - uint32_t rl_ownpid; - uint32_t rl_lkid; - uint32_t rl_remid; - uint32_t rl_parent_lkid; - uint32_t rl_parent_remid; - uint32_t rl_exflags; - uint32_t rl_flags; - uint32_t rl_lvbseq; - int rl_result; + __le32 rl_ownpid; + __le32 rl_lkid; + __le32 rl_remid; + __le32 rl_parent_lkid; + __le32 rl_parent_remid; + __le32 rl_exflags; + __le32 rl_flags; + __le32 rl_lvbseq; + __le32 rl_result; int8_t rl_rqmode; int8_t rl_grmode; int8_t rl_status; int8_t rl_asts; - uint16_t rl_wait_type; - uint16_t rl_namelen; + __le16 rl_wait_type; + __le16 rl_namelen; char rl_name[DLM_RESNAME_MAXLEN]; char rl_lvb[0]; }; @@ -494,7 +501,7 @@ struct dlm_ls { struct rw_semaphore ls_recv_active; /* block dlm_recv */ struct list_head ls_requestqueue;/* queue remote requests */ struct mutex ls_requestqueue_mutex; - char *ls_recover_buf; + struct dlm_rcom *ls_recover_buf; int ls_recover_nodeid; /* for debugging */ uint64_t ls_rcom_seq; spinlock_t ls_rcom_spin; @@ -570,5 +577,21 @@ static inline int dlm_no_directory(struct dlm_ls *ls) return (ls->ls_exflags & DLM_LSFL_NODIR) ? 1 : 0; } +int dlm_netlink_init(void); +void dlm_netlink_exit(void); +void dlm_timeout_warn(struct dlm_lkb *lkb); + +#ifdef CONFIG_DLM_DEBUG +int dlm_register_debugfs(void); +void dlm_unregister_debugfs(void); +int dlm_create_debug_file(struct dlm_ls *ls); +void dlm_delete_debug_file(struct dlm_ls *ls); +#else +static inline int dlm_register_debugfs(void) { return 0; } +static inline void dlm_unregister_debugfs(void) { } +static inline int dlm_create_debug_file(struct dlm_ls *ls) { return 0; } +static inline void dlm_delete_debug_file(struct dlm_ls *ls) { } +#endif + #endif /* __DLM_INTERNAL_DOT_H__ */ diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 3915b8e1414..8f250ac8b92 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -1,7 +1,7 @@ /****************************************************************************** ******************************************************************************* ** -** Copyright (C) 2005-2007 Red Hat, Inc. All rights reserved. +** Copyright (C) 2005-2008 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -88,7 +88,6 @@ static void __receive_convert_reply(struct dlm_rsb *r, struct dlm_lkb *lkb, static int receive_extralen(struct dlm_message *ms); static void do_purge(struct dlm_ls *ls, int nodeid, int pid); static void del_timeout(struct dlm_lkb *lkb); -void dlm_timeout_warn(struct dlm_lkb *lkb); /* * Lock compatibilty matrix - thanks Steve @@ -335,7 +334,7 @@ static struct dlm_rsb *create_rsb(struct dlm_ls *ls, char *name, int len) { struct dlm_rsb *r; - r = allocate_rsb(ls, len); + r = dlm_allocate_rsb(ls, len); if (!r) return NULL; @@ -437,11 +436,15 @@ static int find_rsb(struct dlm_ls *ls, char *name, int namelen, { struct dlm_rsb *r, *tmp; uint32_t hash, bucket; - int error = 0; + int error = -EINVAL; + + if (namelen > DLM_RESNAME_MAXLEN) + goto out; if (dlm_no_directory(ls)) flags |= R_CREATE; + error = 0; hash = jhash(name, namelen, 0); bucket = hash & (ls->ls_rsbtbl_size - 1); @@ -478,7 +481,7 @@ static int find_rsb(struct dlm_ls *ls, char *name, int namelen, error = _search_rsb(ls, name, namelen, bucket, 0, &tmp); if (!error) { write_unlock(&ls->ls_rsbtbl[bucket].lock); - free_rsb(r); + dlm_free_rsb(r); r = tmp; goto out; } @@ -490,12 +493,6 @@ static int find_rsb(struct dlm_ls *ls, char *name, int namelen, return error; } -int dlm_find_rsb(struct dlm_ls *ls, char *name, int namelen, - unsigned int flags, struct dlm_rsb **r_ret) -{ - return find_rsb(ls, name, namelen, flags, r_ret); -} - /* This is only called to add a reference when the code already holds a valid reference to the rsb, so there's no need for locking. */ @@ -519,7 +516,7 @@ static void toss_rsb(struct kref *kref) list_move(&r->res_hashchain, &ls->ls_rsbtbl[r->res_bucket].toss); r->res_toss_time = jiffies; if (r->res_lvbptr) { - free_lvb(r->res_lvbptr); + dlm_free_lvb(r->res_lvbptr); r->res_lvbptr = NULL; } } @@ -589,7 +586,7 @@ static int create_lkb(struct dlm_ls *ls, struct dlm_lkb **lkb_ret) uint32_t lkid = 0; uint16_t bucket; - lkb = allocate_lkb(ls); + lkb = dlm_allocate_lkb(ls); if (!lkb) return -ENOMEM; @@ -683,8 +680,8 @@ static int __put_lkb(struct dlm_ls *ls, struct dlm_lkb *lkb) /* for local/process lkbs, lvbptr points to caller's lksb */ if (lkb->lkb_lvbptr && is_master_copy(lkb)) - free_lvb(lkb->lkb_lvbptr); - free_lkb(lkb); + dlm_free_lvb(lkb->lkb_lvbptr); + dlm_free_lkb(lkb); return 1; } else { write_unlock(&ls->ls_lkbtbl[bucket].lock); @@ -988,7 +985,7 @@ static int shrink_bucket(struct dlm_ls *ls, int b) if (is_master(r)) dir_remove(r); - free_rsb(r); + dlm_free_rsb(r); count++; } else { write_unlock(&ls->ls_rsbtbl[b].lock); @@ -1171,7 +1168,7 @@ static void set_lvb_lock(struct dlm_rsb *r, struct dlm_lkb *lkb) return; if (!r->res_lvbptr) - r->res_lvbptr = allocate_lvb(r->res_ls); + r->res_lvbptr = dlm_allocate_lvb(r->res_ls); if (!r->res_lvbptr) return; @@ -1203,7 +1200,7 @@ static void set_lvb_unlock(struct dlm_rsb *r, struct dlm_lkb *lkb) return; if (!r->res_lvbptr) - r->res_lvbptr = allocate_lvb(r->res_ls); + r->res_lvbptr = dlm_allocate_lvb(r->res_ls); if (!r->res_lvbptr) return; @@ -1229,6 +1226,8 @@ static void set_lvb_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb, b = dlm_lvb_operations[lkb->lkb_grmode + 1][lkb->lkb_rqmode + 1]; if (b == 1) { int len = receive_extralen(ms); + if (len > DLM_RESNAME_MAXLEN) + len = DLM_RESNAME_MAXLEN; memcpy(lkb->lkb_lvbptr, ms->m_extra, len); lkb->lkb_lvbseq = ms->m_lvbseq; } @@ -1782,7 +1781,7 @@ static void grant_pending_locks(struct dlm_rsb *r) */ list_for_each_entry_safe(lkb, s, &r->res_grantqueue, lkb_statequeue) { - if (lkb->lkb_bastaddr && lock_requires_bast(lkb, high, cw)) { + if (lkb->lkb_bastfn && lock_requires_bast(lkb, high, cw)) { if (cw && high == DLM_LOCK_PR) queue_bast(r, lkb, DLM_LOCK_CW); else @@ -1812,7 +1811,7 @@ static void send_bast_queue(struct dlm_rsb *r, struct list_head *head, struct dlm_lkb *gr; list_for_each_entry(gr, head, lkb_statequeue) { - if (gr->lkb_bastaddr && modes_require_bast(gr, lkb)) { + if (gr->lkb_bastfn && modes_require_bast(gr, lkb)) { queue_bast(r, gr, lkb->lkb_rqmode); gr->lkb_highbast = lkb->lkb_rqmode; } @@ -1852,7 +1851,7 @@ static void send_blocking_asts_all(struct dlm_rsb *r, struct dlm_lkb *lkb) static int set_master(struct dlm_rsb *r, struct dlm_lkb *lkb) { struct dlm_ls *ls = r->res_ls; - int error, dir_nodeid, ret_nodeid, our_nodeid = dlm_our_nodeid(); + int i, error, dir_nodeid, ret_nodeid, our_nodeid = dlm_our_nodeid(); if (rsb_flag(r, RSB_MASTER_UNCERTAIN)) { rsb_clear_flag(r, RSB_MASTER_UNCERTAIN); @@ -1886,7 +1885,7 @@ static int set_master(struct dlm_rsb *r, struct dlm_lkb *lkb) return 1; } - for (;;) { + for (i = 0; i < 2; i++) { /* It's possible for dlm_scand to remove an old rsb for this same resource from the toss list, us to create a new one, look up the master locally, and find it @@ -1900,6 +1899,8 @@ static int set_master(struct dlm_rsb *r, struct dlm_lkb *lkb) log_debug(ls, "dir_lookup error %d %s", error, r->res_name); schedule(); } + if (error && error != -EEXIST) + return error; if (ret_nodeid == our_nodeid) { r->res_first_lkid = 0; @@ -1941,8 +1942,11 @@ static void confirm_master(struct dlm_rsb *r, int error) break; case -EAGAIN: - /* the remote master didn't queue our NOQUEUE request; - make a waiting lkb the first_lkid */ + case -EBADR: + case -ENOTBLK: + /* the remote request failed and won't be retried (it was + a NOQUEUE, or has been canceled/unlocked); make a waiting + lkb the first_lkid */ r->res_first_lkid = 0; @@ -1962,8 +1966,11 @@ static void confirm_master(struct dlm_rsb *r, int error) } static int set_lock_args(int mode, struct dlm_lksb *lksb, uint32_t flags, - int namelen, unsigned long timeout_cs, void *ast, - void *astarg, void *bast, struct dlm_args *args) + int namelen, unsigned long timeout_cs, + void (*ast) (void *astparam), + void *astparam, + void (*bast) (void *astparam, int mode), + struct dlm_args *args) { int rv = -EINVAL; @@ -2013,9 +2020,9 @@ static int set_lock_args(int mode, struct dlm_lksb *lksb, uint32_t flags, an active lkb cannot be modified before locking the rsb */ args->flags = flags; - args->astaddr = ast; - args->astparam = (long) astarg; - args->bastaddr = bast; + args->astfn = ast; + args->astparam = astparam; + args->bastfn = bast; args->timeout = timeout_cs; args->mode = mode; args->lksb = lksb; @@ -2034,7 +2041,7 @@ static int set_unlock_args(uint32_t flags, void *astarg, struct dlm_args *args) return -EINVAL; args->flags = flags; - args->astparam = (long) astarg; + args->astparam = astarg; return 0; } @@ -2064,9 +2071,9 @@ static int validate_lock_args(struct dlm_ls *ls, struct dlm_lkb *lkb, lkb->lkb_exflags = args->flags; lkb->lkb_sbflags = 0; - lkb->lkb_astaddr = args->astaddr; + lkb->lkb_astfn = args->astfn; lkb->lkb_astparam = args->astparam; - lkb->lkb_bastaddr = args->bastaddr; + lkb->lkb_bastfn = args->bastfn; lkb->lkb_rqmode = args->mode; lkb->lkb_lksb = args->lksb; lkb->lkb_lvbptr = args->lksb->sb_lvbptr; @@ -2108,17 +2115,18 @@ static int validate_unlock_args(struct dlm_lkb *lkb, struct dlm_args *args) /* an lkb may be waiting for an rsb lookup to complete where the lookup was initiated by another lock */ - if (args->flags & (DLM_LKF_CANCEL | DLM_LKF_FORCEUNLOCK)) { - if (!list_empty(&lkb->lkb_rsb_lookup)) { + if (!list_empty(&lkb->lkb_rsb_lookup)) { + if (args->flags & (DLM_LKF_CANCEL | DLM_LKF_FORCEUNLOCK)) { log_debug(ls, "unlock on rsb_lookup %x", lkb->lkb_id); list_del_init(&lkb->lkb_rsb_lookup); queue_cast(lkb->lkb_resource, lkb, args->flags & DLM_LKF_CANCEL ? -DLM_ECANCEL : -DLM_EUNLOCK); unhold_lkb(lkb); /* undoes create_lkb() */ - rv = -EBUSY; - goto out; } + /* caller changes -EBUSY to 0 for CANCEL and FORCEUNLOCK */ + rv = -EBUSY; + goto out; } /* cancel not allowed with another cancel/unlock in progress */ @@ -2712,9 +2720,9 @@ static void send_args(struct dlm_rsb *r, struct dlm_lkb *lkb, /* m_result and m_bastmode are set from function args, not from lkb fields */ - if (lkb->lkb_bastaddr) + if (lkb->lkb_bastfn) ms->m_asts |= AST_BAST; - if (lkb->lkb_astaddr) + if (lkb->lkb_astfn) ms->m_asts |= AST_COMP; /* compare with switch in create_message; send_remove() doesn't @@ -2986,15 +2994,27 @@ static int receive_lvb(struct dlm_ls *ls, struct dlm_lkb *lkb, if (lkb->lkb_exflags & DLM_LKF_VALBLK) { if (!lkb->lkb_lvbptr) - lkb->lkb_lvbptr = allocate_lvb(ls); + lkb->lkb_lvbptr = dlm_allocate_lvb(ls); if (!lkb->lkb_lvbptr) return -ENOMEM; len = receive_extralen(ms); + if (len > DLM_RESNAME_MAXLEN) + len = DLM_RESNAME_MAXLEN; memcpy(lkb->lkb_lvbptr, ms->m_extra, len); } return 0; } +static void fake_bastfn(void *astparam, int mode) +{ + log_print("fake_bastfn should not be called"); +} + +static void fake_astfn(void *astparam) +{ + log_print("fake_astfn should not be called"); +} + static int receive_request_args(struct dlm_ls *ls, struct dlm_lkb *lkb, struct dlm_message *ms) { @@ -3003,14 +3023,13 @@ static int receive_request_args(struct dlm_ls *ls, struct dlm_lkb *lkb, lkb->lkb_remid = ms->m_lkid; lkb->lkb_grmode = DLM_LOCK_IV; lkb->lkb_rqmode = ms->m_rqmode; - lkb->lkb_bastaddr = (void *) (long) (ms->m_asts & AST_BAST); - lkb->lkb_astaddr = (void *) (long) (ms->m_asts & AST_COMP); - DLM_ASSERT(is_master_copy(lkb), dlm_print_lkb(lkb);); + lkb->lkb_bastfn = (ms->m_asts & AST_BAST) ? &fake_bastfn : NULL; + lkb->lkb_astfn = (ms->m_asts & AST_COMP) ? &fake_astfn : NULL; if (lkb->lkb_exflags & DLM_LKF_VALBLK) { /* lkb was just created so there won't be an lvb yet */ - lkb->lkb_lvbptr = allocate_lvb(ls); + lkb->lkb_lvbptr = dlm_allocate_lvb(ls); if (!lkb->lkb_lvbptr) return -ENOMEM; } @@ -3021,16 +3040,6 @@ static int receive_request_args(struct dlm_ls *ls, struct dlm_lkb *lkb, static int receive_convert_args(struct dlm_ls *ls, struct dlm_lkb *lkb, struct dlm_message *ms) { - if (lkb->lkb_nodeid != ms->m_header.h_nodeid) { - log_error(ls, "convert_args nodeid %d %d lkid %x %x", - lkb->lkb_nodeid, ms->m_header.h_nodeid, - lkb->lkb_id, lkb->lkb_remid); - return -EINVAL; - } - - if (!is_master_copy(lkb)) - return -EINVAL; - if (lkb->lkb_status != DLM_LKSTS_GRANTED) return -EBUSY; @@ -3046,8 +3055,6 @@ static int receive_convert_args(struct dlm_ls *ls, struct dlm_lkb *lkb, static int receive_unlock_args(struct dlm_ls *ls, struct dlm_lkb *lkb, struct dlm_message *ms) { - if (!is_master_copy(lkb)) - return -EINVAL; if (receive_lvb(ls, lkb, ms)) return -ENOMEM; return 0; @@ -3063,6 +3070,50 @@ static void setup_stub_lkb(struct dlm_ls *ls, struct dlm_message *ms) lkb->lkb_remid = ms->m_lkid; } +/* This is called after the rsb is locked so that we can safely inspect + fields in the lkb. */ + +static int validate_message(struct dlm_lkb *lkb, struct dlm_message *ms) +{ + int from = ms->m_header.h_nodeid; + int error = 0; + + switch (ms->m_type) { + case DLM_MSG_CONVERT: + case DLM_MSG_UNLOCK: + case DLM_MSG_CANCEL: + if (!is_master_copy(lkb) || lkb->lkb_nodeid != from) + error = -EINVAL; + break; + + case DLM_MSG_CONVERT_REPLY: + case DLM_MSG_UNLOCK_REPLY: + case DLM_MSG_CANCEL_REPLY: + case DLM_MSG_GRANT: + case DLM_MSG_BAST: + if (!is_process_copy(lkb) || lkb->lkb_nodeid != from) + error = -EINVAL; + break; + + case DLM_MSG_REQUEST_REPLY: + if (!is_process_copy(lkb)) + error = -EINVAL; + else if (lkb->lkb_nodeid != -1 && lkb->lkb_nodeid != from) + error = -EINVAL; + break; + + default: + error = -EINVAL; + } + + if (error) + log_error(lkb->lkb_resource->res_ls, + "ignore invalid message %d from %d %x %x %x %d", + ms->m_type, from, lkb->lkb_id, lkb->lkb_remid, + lkb->lkb_flags, lkb->lkb_nodeid); + return error; +} + static void receive_request(struct dlm_ls *ls, struct dlm_message *ms) { struct dlm_lkb *lkb; @@ -3124,17 +3175,21 @@ static void receive_convert(struct dlm_ls *ls, struct dlm_message *ms) hold_rsb(r); lock_rsb(r); + error = validate_message(lkb, ms); + if (error) + goto out; + receive_flags(lkb, ms); error = receive_convert_args(ls, lkb, ms); if (error) - goto out; + goto out_reply; reply = !down_conversion(lkb); error = do_convert(r, lkb); - out: + out_reply: if (reply) send_convert_reply(r, lkb, error); - + out: unlock_rsb(r); put_rsb(r); dlm_put_lkb(lkb); @@ -3160,15 +3215,19 @@ static void receive_unlock(struct dlm_ls *ls, struct dlm_message *ms) hold_rsb(r); lock_rsb(r); + error = validate_message(lkb, ms); + if (error) + goto out; + receive_flags(lkb, ms); error = receive_unlock_args(ls, lkb, ms); if (error) - goto out; + goto out_reply; error = do_unlock(r, lkb); - out: + out_reply: send_unlock_reply(r, lkb, error); - + out: unlock_rsb(r); put_rsb(r); dlm_put_lkb(lkb); @@ -3196,9 +3255,13 @@ static void receive_cancel(struct dlm_ls *ls, struct dlm_message *ms) hold_rsb(r); lock_rsb(r); + error = validate_message(lkb, ms); + if (error) + goto out; + error = do_cancel(r, lkb); send_cancel_reply(r, lkb, error); - + out: unlock_rsb(r); put_rsb(r); dlm_put_lkb(lkb); @@ -3217,22 +3280,26 @@ static void receive_grant(struct dlm_ls *ls, struct dlm_message *ms) error = find_lkb(ls, ms->m_remid, &lkb); if (error) { - log_error(ls, "receive_grant no lkb"); + log_debug(ls, "receive_grant from %d no lkb %x", + ms->m_header.h_nodeid, ms->m_remid); return; } - DLM_ASSERT(is_process_copy(lkb), dlm_print_lkb(lkb);); r = lkb->lkb_resource; hold_rsb(r); lock_rsb(r); + error = validate_message(lkb, ms); + if (error) + goto out; + receive_flags_reply(lkb, ms); if (is_altmode(lkb)) munge_altmode(lkb, ms); grant_lock_pc(r, lkb, ms); queue_cast(r, lkb, 0); - + out: unlock_rsb(r); put_rsb(r); dlm_put_lkb(lkb); @@ -3246,18 +3313,22 @@ static void receive_bast(struct dlm_ls *ls, struct dlm_message *ms) error = find_lkb(ls, ms->m_remid, &lkb); if (error) { - log_error(ls, "receive_bast no lkb"); + log_debug(ls, "receive_bast from %d no lkb %x", + ms->m_header.h_nodeid, ms->m_remid); return; } - DLM_ASSERT(is_process_copy(lkb), dlm_print_lkb(lkb);); r = lkb->lkb_resource; hold_rsb(r); lock_rsb(r); - queue_bast(r, lkb, ms->m_bastmode); + error = validate_message(lkb, ms); + if (error) + goto out; + queue_bast(r, lkb, ms->m_bastmode); + out: unlock_rsb(r); put_rsb(r); dlm_put_lkb(lkb); @@ -3323,15 +3394,19 @@ static void receive_request_reply(struct dlm_ls *ls, struct dlm_message *ms) error = find_lkb(ls, ms->m_remid, &lkb); if (error) { - log_error(ls, "receive_request_reply no lkb"); + log_debug(ls, "receive_request_reply from %d no lkb %x", + ms->m_header.h_nodeid, ms->m_remid); return; } - DLM_ASSERT(is_process_copy(lkb), dlm_print_lkb(lkb);); r = lkb->lkb_resource; hold_rsb(r); lock_rsb(r); + error = validate_message(lkb, ms); + if (error) + goto out; + mstype = lkb->lkb_wait_type; error = remove_from_waiters(lkb, DLM_MSG_REQUEST_REPLY); if (error) @@ -3383,6 +3458,7 @@ static void receive_request_reply(struct dlm_ls *ls, struct dlm_message *ms) if (is_overlap(lkb)) { /* we'll ignore error in cancel/unlock reply */ queue_cast_overlap(r, lkb); + confirm_master(r, result); unhold_lkb(lkb); /* undoes create_lkb() */ } else _request_lock(r, lkb); @@ -3463,6 +3539,10 @@ static void _receive_convert_reply(struct dlm_lkb *lkb, struct dlm_message *ms) hold_rsb(r); lock_rsb(r); + error = validate_message(lkb, ms); + if (error) + goto out; + /* stub reply can happen with waiters_mutex held */ error = remove_from_waiters_ms(lkb, ms); if (error) @@ -3481,10 +3561,10 @@ static void receive_convert_reply(struct dlm_ls *ls, struct dlm_message *ms) error = find_lkb(ls, ms->m_remid, &lkb); if (error) { - log_error(ls, "receive_convert_reply no lkb"); + log_debug(ls, "receive_convert_reply from %d no lkb %x", + ms->m_header.h_nodeid, ms->m_remid); return; } - DLM_ASSERT(is_process_copy(lkb), dlm_print_lkb(lkb);); _receive_convert_reply(lkb, ms); dlm_put_lkb(lkb); @@ -3498,6 +3578,10 @@ static void _receive_unlock_reply(struct dlm_lkb *lkb, struct dlm_message *ms) hold_rsb(r); lock_rsb(r); + error = validate_message(lkb, ms); + if (error) + goto out; + /* stub reply can happen with waiters_mutex held */ error = remove_from_waiters_ms(lkb, ms); if (error) @@ -3529,10 +3613,10 @@ static void receive_unlock_reply(struct dlm_ls *ls, struct dlm_message *ms) error = find_lkb(ls, ms->m_remid, &lkb); if (error) { - log_error(ls, "receive_unlock_reply no lkb"); + log_debug(ls, "receive_unlock_reply from %d no lkb %x", + ms->m_header.h_nodeid, ms->m_remid); return; } - DLM_ASSERT(is_process_copy(lkb), dlm_print_lkb(lkb);); _receive_unlock_reply(lkb, ms); dlm_put_lkb(lkb); @@ -3546,6 +3630,10 @@ static void _receive_cancel_reply(struct dlm_lkb *lkb, struct dlm_message *ms) hold_rsb(r); lock_rsb(r); + error = validate_message(lkb, ms); + if (error) + goto out; + /* stub reply can happen with waiters_mutex held */ error = remove_from_waiters_ms(lkb, ms); if (error) @@ -3577,10 +3665,10 @@ static void receive_cancel_reply(struct dlm_ls *ls, struct dlm_message *ms) error = find_lkb(ls, ms->m_remid, &lkb); if (error) { - log_error(ls, "receive_cancel_reply no lkb"); + log_debug(ls, "receive_cancel_reply from %d no lkb %x", + ms->m_header.h_nodeid, ms->m_remid); return; } - DLM_ASSERT(is_process_copy(lkb), dlm_print_lkb(lkb);); _receive_cancel_reply(lkb, ms); dlm_put_lkb(lkb); @@ -3640,6 +3728,13 @@ static void receive_lookup_reply(struct dlm_ls *ls, struct dlm_message *ms) static void _receive_message(struct dlm_ls *ls, struct dlm_message *ms) { + if (!dlm_is_member(ls, ms->m_header.h_nodeid)) { + log_debug(ls, "ignore non-member message %d from %d %x %x %d", + ms->m_type, ms->m_header.h_nodeid, ms->m_lkid, + ms->m_remid, ms->m_result); + return; + } + switch (ms->m_type) { /* messages sent to a master node */ @@ -3729,7 +3824,7 @@ static void dlm_receive_message(struct dlm_ls *ls, struct dlm_message *ms, int nodeid) { if (dlm_locking_stopped(ls)) { - dlm_add_requestqueue(ls, nodeid, (struct dlm_header *) ms); + dlm_add_requestqueue(ls, nodeid, ms); } else { dlm_wait_requestqueue(ls); _receive_message(ls, ms); @@ -3749,21 +3844,20 @@ void dlm_receive_message_saved(struct dlm_ls *ls, struct dlm_message *ms) standard locking activity) or an RCOM (recovery message sent as part of lockspace recovery). */ -void dlm_receive_buffer(struct dlm_header *hd, int nodeid) +void dlm_receive_buffer(union dlm_packet *p, int nodeid) { - struct dlm_message *ms = (struct dlm_message *) hd; - struct dlm_rcom *rc = (struct dlm_rcom *) hd; + struct dlm_header *hd = &p->header; struct dlm_ls *ls; int type = 0; switch (hd->h_cmd) { case DLM_MSG: - dlm_message_in(ms); - type = ms->m_type; + dlm_message_in(&p->message); + type = p->message.m_type; break; case DLM_RCOM: - dlm_rcom_in(rc); - type = rc->rc_type; + dlm_rcom_in(&p->rcom); + type = p->rcom.rc_type; break; default: log_print("invalid h_cmd %d from %u", hd->h_cmd, nodeid); @@ -3778,11 +3872,12 @@ void dlm_receive_buffer(struct dlm_header *hd, int nodeid) ls = dlm_find_lockspace_global(hd->h_lockspace); if (!ls) { - log_print("invalid h_lockspace %x from %d cmd %d type %d", - hd->h_lockspace, nodeid, hd->h_cmd, type); + if (dlm_config.ci_log_debug) + log_print("invalid lockspace %x from %d cmd %d type %d", + hd->h_lockspace, nodeid, hd->h_cmd, type); if (hd->h_cmd == DLM_RCOM && type == DLM_RCOM_STATUS) - dlm_send_ls_not_ready(nodeid, rc); + dlm_send_ls_not_ready(nodeid, &p->rcom); return; } @@ -3791,9 +3886,9 @@ void dlm_receive_buffer(struct dlm_header *hd, int nodeid) down_read(&ls->ls_recv_active); if (hd->h_cmd == DLM_MSG) - dlm_receive_message(ls, ms, nodeid); + dlm_receive_message(ls, &p->message, nodeid); else - dlm_receive_rcom(ls, rc, nodeid); + dlm_receive_rcom(ls, &p->rcom, nodeid); up_read(&ls->ls_recv_active); dlm_put_lockspace(ls); @@ -3806,6 +3901,7 @@ static void recover_convert_waiter(struct dlm_ls *ls, struct dlm_lkb *lkb) ls->ls_stub_ms.m_type = DLM_MSG_CONVERT_REPLY; ls->ls_stub_ms.m_result = -EINPROGRESS; ls->ls_stub_ms.m_flags = lkb->lkb_flags; + ls->ls_stub_ms.m_header.h_nodeid = lkb->lkb_nodeid; _receive_convert_reply(lkb, &ls->ls_stub_ms); /* Same special case as in receive_rcom_lock_args() */ @@ -3847,6 +3943,7 @@ static int waiter_needs_recovery(struct dlm_ls *ls, struct dlm_lkb *lkb) void dlm_recover_waiters_pre(struct dlm_ls *ls) { struct dlm_lkb *lkb, *safe; + int wait_type, stub_unlock_result, stub_cancel_result; mutex_lock(&ls->ls_waiters_mutex); @@ -3865,7 +3962,33 @@ void dlm_recover_waiters_pre(struct dlm_ls *ls) if (!waiter_needs_recovery(ls, lkb)) continue; - switch (lkb->lkb_wait_type) { + wait_type = lkb->lkb_wait_type; + stub_unlock_result = -DLM_EUNLOCK; + stub_cancel_result = -DLM_ECANCEL; + + /* Main reply may have been received leaving a zero wait_type, + but a reply for the overlapping op may not have been + received. In that case we need to fake the appropriate + reply for the overlap op. */ + + if (!wait_type) { + if (is_overlap_cancel(lkb)) { + wait_type = DLM_MSG_CANCEL; + if (lkb->lkb_grmode == DLM_LOCK_IV) + stub_cancel_result = 0; + } + if (is_overlap_unlock(lkb)) { + wait_type = DLM_MSG_UNLOCK; + if (lkb->lkb_grmode == DLM_LOCK_IV) + stub_unlock_result = -ENOENT; + } + + log_debug(ls, "rwpre overlap %x %x %d %d %d", + lkb->lkb_id, lkb->lkb_flags, wait_type, + stub_cancel_result, stub_unlock_result); + } + + switch (wait_type) { case DLM_MSG_REQUEST: lkb->lkb_flags |= DLM_IFL_RESEND; @@ -3878,8 +4001,9 @@ void dlm_recover_waiters_pre(struct dlm_ls *ls) case DLM_MSG_UNLOCK: hold_lkb(lkb); ls->ls_stub_ms.m_type = DLM_MSG_UNLOCK_REPLY; - ls->ls_stub_ms.m_result = -DLM_EUNLOCK; + ls->ls_stub_ms.m_result = stub_unlock_result; ls->ls_stub_ms.m_flags = lkb->lkb_flags; + ls->ls_stub_ms.m_header.h_nodeid = lkb->lkb_nodeid; _receive_unlock_reply(lkb, &ls->ls_stub_ms); dlm_put_lkb(lkb); break; @@ -3887,15 +4011,16 @@ void dlm_recover_waiters_pre(struct dlm_ls *ls) case DLM_MSG_CANCEL: hold_lkb(lkb); ls->ls_stub_ms.m_type = DLM_MSG_CANCEL_REPLY; - ls->ls_stub_ms.m_result = -DLM_ECANCEL; + ls->ls_stub_ms.m_result = stub_cancel_result; ls->ls_stub_ms.m_flags = lkb->lkb_flags; + ls->ls_stub_ms.m_header.h_nodeid = lkb->lkb_nodeid; _receive_cancel_reply(lkb, &ls->ls_stub_ms); dlm_put_lkb(lkb); break; default: - log_error(ls, "invalid lkb wait_type %d", - lkb->lkb_wait_type); + log_error(ls, "invalid lkb wait_type %d %d", + lkb->lkb_wait_type, wait_type); } schedule(); } @@ -4163,32 +4288,34 @@ static struct dlm_lkb *search_remid(struct dlm_rsb *r, int nodeid, return NULL; } +/* needs at least dlm_rcom + rcom_lock */ static int receive_rcom_lock_args(struct dlm_ls *ls, struct dlm_lkb *lkb, struct dlm_rsb *r, struct dlm_rcom *rc) { struct rcom_lock *rl = (struct rcom_lock *) rc->rc_buf; - int lvblen; lkb->lkb_nodeid = rc->rc_header.h_nodeid; - lkb->lkb_ownpid = rl->rl_ownpid; - lkb->lkb_remid = rl->rl_lkid; - lkb->lkb_exflags = rl->rl_exflags; - lkb->lkb_flags = rl->rl_flags & 0x0000FFFF; + lkb->lkb_ownpid = le32_to_cpu(rl->rl_ownpid); + lkb->lkb_remid = le32_to_cpu(rl->rl_lkid); + lkb->lkb_exflags = le32_to_cpu(rl->rl_exflags); + lkb->lkb_flags = le32_to_cpu(rl->rl_flags) & 0x0000FFFF; lkb->lkb_flags |= DLM_IFL_MSTCPY; - lkb->lkb_lvbseq = rl->rl_lvbseq; + lkb->lkb_lvbseq = le32_to_cpu(rl->rl_lvbseq); lkb->lkb_rqmode = rl->rl_rqmode; lkb->lkb_grmode = rl->rl_grmode; /* don't set lkb_status because add_lkb wants to itself */ - lkb->lkb_bastaddr = (void *) (long) (rl->rl_asts & AST_BAST); - lkb->lkb_astaddr = (void *) (long) (rl->rl_asts & AST_COMP); + lkb->lkb_bastfn = (rl->rl_asts & AST_BAST) ? &fake_bastfn : NULL; + lkb->lkb_astfn = (rl->rl_asts & AST_COMP) ? &fake_astfn : NULL; if (lkb->lkb_exflags & DLM_LKF_VALBLK) { - lkb->lkb_lvbptr = allocate_lvb(ls); + int lvblen = rc->rc_header.h_length - sizeof(struct dlm_rcom) - + sizeof(struct rcom_lock); + if (lvblen > ls->ls_lvblen) + return -EINVAL; + lkb->lkb_lvbptr = dlm_allocate_lvb(ls); if (!lkb->lkb_lvbptr) return -ENOMEM; - lvblen = rc->rc_header.h_length - sizeof(struct dlm_rcom) - - sizeof(struct rcom_lock); memcpy(lkb->lkb_lvbptr, rl->rl_lvb, lvblen); } @@ -4196,7 +4323,8 @@ static int receive_rcom_lock_args(struct dlm_ls *ls, struct dlm_lkb *lkb, The real granted mode of these converting locks cannot be determined until all locks have been rebuilt on the rsb (recover_conversion) */ - if (rl->rl_wait_type == DLM_MSG_CONVERT && middle_conversion(lkb)) { + if (rl->rl_wait_type == cpu_to_le16(DLM_MSG_CONVERT) && + middle_conversion(lkb)) { rl->rl_status = DLM_LKSTS_CONVERT; lkb->lkb_grmode = DLM_LOCK_IV; rsb_set_flag(r, RSB_RECOVER_CONVERT); @@ -4211,6 +4339,7 @@ static int receive_rcom_lock_args(struct dlm_ls *ls, struct dlm_lkb *lkb, the given values and send back our lkid. We send back our lkid by sending back the rcom_lock struct we got but with the remid field filled in. */ +/* needs at least dlm_rcom + rcom_lock */ int dlm_recover_master_copy(struct dlm_ls *ls, struct dlm_rcom *rc) { struct rcom_lock *rl = (struct rcom_lock *) rc->rc_buf; @@ -4223,13 +4352,14 @@ int dlm_recover_master_copy(struct dlm_ls *ls, struct dlm_rcom *rc) goto out; } - error = find_rsb(ls, rl->rl_name, rl->rl_namelen, R_MASTER, &r); + error = find_rsb(ls, rl->rl_name, le16_to_cpu(rl->rl_namelen), + R_MASTER, &r); if (error) goto out; lock_rsb(r); - lkb = search_remid(r, rc->rc_header.h_nodeid, rl->rl_lkid); + lkb = search_remid(r, rc->rc_header.h_nodeid, le32_to_cpu(rl->rl_lkid)); if (lkb) { error = -EEXIST; goto out_remid; @@ -4252,18 +4382,20 @@ int dlm_recover_master_copy(struct dlm_ls *ls, struct dlm_rcom *rc) out_remid: /* this is the new value returned to the lock holder for saving in its process-copy lkb */ - rl->rl_remid = lkb->lkb_id; + rl->rl_remid = cpu_to_le32(lkb->lkb_id); out_unlock: unlock_rsb(r); put_rsb(r); out: if (error) - log_print("recover_master_copy %d %x", error, rl->rl_lkid); - rl->rl_result = error; + log_debug(ls, "recover_master_copy %d %x", error, + le32_to_cpu(rl->rl_lkid)); + rl->rl_result = cpu_to_le32(error); return error; } +/* needs at least dlm_rcom + rcom_lock */ int dlm_recover_process_copy(struct dlm_ls *ls, struct dlm_rcom *rc) { struct rcom_lock *rl = (struct rcom_lock *) rc->rc_buf; @@ -4271,15 +4403,16 @@ int dlm_recover_process_copy(struct dlm_ls *ls, struct dlm_rcom *rc) struct dlm_lkb *lkb; int error; - error = find_lkb(ls, rl->rl_lkid, &lkb); + error = find_lkb(ls, le32_to_cpu(rl->rl_lkid), &lkb); if (error) { - log_error(ls, "recover_process_copy no lkid %x", rl->rl_lkid); + log_error(ls, "recover_process_copy no lkid %x", + le32_to_cpu(rl->rl_lkid)); return error; } DLM_ASSERT(is_process_copy(lkb), dlm_print_lkb(lkb);); - error = rl->rl_result; + error = le32_to_cpu(rl->rl_result); r = lkb->lkb_resource; hold_rsb(r); @@ -4298,7 +4431,7 @@ int dlm_recover_process_copy(struct dlm_ls *ls, struct dlm_rcom *rc) log_debug(ls, "master copy exists %x", lkb->lkb_id); /* fall through */ case 0: - lkb->lkb_remid = rl->rl_remid; + lkb->lkb_remid = le32_to_cpu(rl->rl_remid); break; default: log_error(ls, "dlm_recover_process_copy unknown error %d %x", @@ -4342,12 +4475,12 @@ int dlm_user_request(struct dlm_ls *ls, struct dlm_user_args *ua, } } - /* After ua is attached to lkb it will be freed by free_lkb(). + /* After ua is attached to lkb it will be freed by dlm_free_lkb(). When DLM_IFL_USER is set, the dlm knows that this is a userspace lock and that lkb_astparam is the dlm_user_args structure. */ error = set_lock_args(mode, &ua->lksb, flags, namelen, timeout_cs, - DLM_FAKE_USER_AST, ua, DLM_FAKE_USER_AST, &args); + fake_astfn, ua, fake_bastfn, &args); lkb->lkb_flags |= DLM_IFL_USER; ua->old_mode = DLM_LOCK_IV; @@ -4400,7 +4533,7 @@ int dlm_user_convert(struct dlm_ls *ls, struct dlm_user_args *ua_tmp, /* user can change the params on its lock when it converts it, or add an lvb that didn't exist before */ - ua = (struct dlm_user_args *)lkb->lkb_astparam; + ua = lkb->lkb_ua; if (flags & DLM_LKF_VALBLK && !ua->lksb.sb_lvbptr) { ua->lksb.sb_lvbptr = kzalloc(DLM_USER_LVB_LEN, GFP_KERNEL); @@ -4421,7 +4554,7 @@ int dlm_user_convert(struct dlm_ls *ls, struct dlm_user_args *ua_tmp, ua->old_mode = lkb->lkb_grmode; error = set_lock_args(mode, &ua->lksb, flags, 0, timeout_cs, - DLM_FAKE_USER_AST, ua, DLM_FAKE_USER_AST, &args); + fake_astfn, ua, fake_bastfn, &args); if (error) goto out_put; @@ -4451,7 +4584,7 @@ int dlm_user_unlock(struct dlm_ls *ls, struct dlm_user_args *ua_tmp, if (error) goto out; - ua = (struct dlm_user_args *)lkb->lkb_astparam; + ua = lkb->lkb_ua; if (lvb_in && ua->lksb.sb_lvbptr) memcpy(ua->lksb.sb_lvbptr, lvb_in, DLM_USER_LVB_LEN); @@ -4500,7 +4633,7 @@ int dlm_user_cancel(struct dlm_ls *ls, struct dlm_user_args *ua_tmp, if (error) goto out; - ua = (struct dlm_user_args *)lkb->lkb_astparam; + ua = lkb->lkb_ua; if (ua_tmp->castparam) ua->castparam = ua_tmp->castparam; ua->user_lksb = ua_tmp->user_lksb; @@ -4538,7 +4671,7 @@ int dlm_user_deadlock(struct dlm_ls *ls, uint32_t flags, uint32_t lkid) if (error) goto out; - ua = (struct dlm_user_args *)lkb->lkb_astparam; + ua = lkb->lkb_ua; error = set_unlock_args(flags, ua, &args); if (error) @@ -4577,7 +4710,6 @@ int dlm_user_deadlock(struct dlm_ls *ls, uint32_t flags, uint32_t lkid) static int orphan_proc_lock(struct dlm_ls *ls, struct dlm_lkb *lkb) { - struct dlm_user_args *ua = (struct dlm_user_args *)lkb->lkb_astparam; struct dlm_args args; int error; @@ -4586,7 +4718,7 @@ static int orphan_proc_lock(struct dlm_ls *ls, struct dlm_lkb *lkb) list_add_tail(&lkb->lkb_ownqueue, &ls->ls_orphans); mutex_unlock(&ls->ls_orphans_mutex); - set_unlock_args(0, ua, &args); + set_unlock_args(0, lkb->lkb_ua, &args); error = cancel_lock(ls, lkb, &args); if (error == -DLM_ECANCEL) @@ -4599,11 +4731,10 @@ static int orphan_proc_lock(struct dlm_ls *ls, struct dlm_lkb *lkb) static int unlock_proc_lock(struct dlm_ls *ls, struct dlm_lkb *lkb) { - struct dlm_user_args *ua = (struct dlm_user_args *)lkb->lkb_astparam; struct dlm_args args; int error; - set_unlock_args(DLM_LKF_FORCEUNLOCK, ua, &args); + set_unlock_args(DLM_LKF_FORCEUNLOCK, lkb->lkb_ua, &args); error = unlock_lock(ls, lkb, &args); if (error == -DLM_EUNLOCK) @@ -4679,6 +4810,7 @@ void dlm_clear_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc) } list_for_each_entry_safe(lkb, safe, &proc->asts, lkb_astqueue) { + lkb->lkb_ast_type = 0; list_del(&lkb->lkb_astqueue); dlm_put_lkb(lkb); } diff --git a/fs/dlm/lock.h b/fs/dlm/lock.h index ada04680a1e..05d9c82e646 100644 --- a/fs/dlm/lock.h +++ b/fs/dlm/lock.h @@ -17,10 +17,8 @@ void dlm_print_rsb(struct dlm_rsb *r); void dlm_dump_rsb(struct dlm_rsb *r); void dlm_print_lkb(struct dlm_lkb *lkb); void dlm_receive_message_saved(struct dlm_ls *ls, struct dlm_message *ms); -void dlm_receive_buffer(struct dlm_header *hd, int nodeid); +void dlm_receive_buffer(union dlm_packet *p, int nodeid); int dlm_modes_compat(int mode1, int mode2); -int dlm_find_rsb(struct dlm_ls *ls, char *name, int namelen, - unsigned int flags, struct dlm_rsb **r_ret); void dlm_put_rsb(struct dlm_rsb *r); void dlm_hold_rsb(struct dlm_rsb *r); int dlm_put_lkb(struct dlm_lkb *lkb); diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index 6353a838452..b64e55e0515 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -24,14 +24,6 @@ #include "recover.h" #include "requestqueue.h" -#ifdef CONFIG_DLM_DEBUG -int dlm_create_debug_file(struct dlm_ls *ls); -void dlm_delete_debug_file(struct dlm_ls *ls); -#else -static inline int dlm_create_debug_file(struct dlm_ls *ls) { return 0; } -static inline void dlm_delete_debug_file(struct dlm_ls *ls) { } -#endif - static int ls_count; static struct mutex ls_lock; static struct list_head lslist; @@ -166,26 +158,7 @@ static struct kobj_type dlm_ktype = { .release = lockspace_kobj_release, }; -static struct kset dlm_kset = { - .ktype = &dlm_ktype, -}; - -static int kobject_setup(struct dlm_ls *ls) -{ - char lsname[DLM_LOCKSPACE_LEN]; - int error; - - memset(lsname, 0, DLM_LOCKSPACE_LEN); - snprintf(lsname, DLM_LOCKSPACE_LEN, "%s", ls->ls_name); - - error = kobject_set_name(&ls->ls_kobj, "%s", lsname); - if (error) - return error; - - ls->ls_kobj.kset = &dlm_kset; - ls->ls_kobj.ktype = &dlm_ktype; - return 0; -} +static struct kset *dlm_kset; static int do_uevent(struct dlm_ls *ls, int in) { @@ -218,26 +191,24 @@ static int do_uevent(struct dlm_ls *ls, int in) } -int dlm_lockspace_init(void) +int __init dlm_lockspace_init(void) { - int error; - ls_count = 0; mutex_init(&ls_lock); INIT_LIST_HEAD(&lslist); spin_lock_init(&lslist_lock); - kobject_set_name(&dlm_kset.kobj, "dlm"); - kobj_set_kset_s(&dlm_kset, kernel_subsys); - error = kset_register(&dlm_kset); - if (error) - printk("dlm_lockspace_init: cannot register kset %d\n", error); - return error; + dlm_kset = kset_create_and_add("dlm", NULL, kernel_kobj); + if (!dlm_kset) { + printk(KERN_WARNING "%s: can not create kset\n", __FUNCTION__); + return -ENOMEM; + } + return 0; } void dlm_lockspace_exit(void) { - kset_unregister(&dlm_kset); + kset_unregister(dlm_kset); } static int dlm_scand(void *data) @@ -549,13 +520,12 @@ static int new_lockspace(char *name, int namelen, void **lockspace, goto out_delist; } - error = kobject_setup(ls); - if (error) - goto out_stop; - - error = kobject_register(&ls->ls_kobj); + ls->ls_kobj.kset = dlm_kset; + error = kobject_init_and_add(&ls->ls_kobj, &dlm_ktype, NULL, + "%s", ls->ls_name); if (error) goto out_stop; + kobject_uevent(&ls->ls_kobj, KOBJ_ADD); /* let kobject handle freeing of ls if there's an error */ do_unreg = 1; @@ -601,7 +571,7 @@ static int new_lockspace(char *name, int namelen, void **lockspace, kfree(ls->ls_rsbtbl); out_lsfree: if (do_unreg) - kobject_unregister(&ls->ls_kobj); + kobject_put(&ls->ls_kobj); else kfree(ls); out: @@ -706,9 +676,9 @@ static int release_lockspace(struct dlm_ls *ls, int force) dlm_del_ast(lkb); if (lkb->lkb_lvbptr && lkb->lkb_flags & DLM_IFL_MSTCPY) - free_lvb(lkb->lkb_lvbptr); + dlm_free_lvb(lkb->lkb_lvbptr); - free_lkb(lkb); + dlm_free_lkb(lkb); } } dlm_astd_resume(); @@ -726,7 +696,7 @@ static int release_lockspace(struct dlm_ls *ls, int force) res_hashchain); list_del(&rsb->res_hashchain); - free_rsb(rsb); + dlm_free_rsb(rsb); } head = &ls->ls_rsbtbl[i].toss; @@ -734,7 +704,7 @@ static int release_lockspace(struct dlm_ls *ls, int force) rsb = list_entry(head->next, struct dlm_rsb, res_hashchain); list_del(&rsb->res_hashchain); - free_rsb(rsb); + dlm_free_rsb(rsb); } } @@ -750,7 +720,7 @@ static int release_lockspace(struct dlm_ls *ls, int force) dlm_clear_members(ls); dlm_clear_members_gone(ls); kfree(ls->ls_node_array); - kobject_unregister(&ls->ls_kobj); + kobject_put(&ls->ls_kobj); /* The ls structure will be freed when the kobject is done with */ mutex_lock(&ls_lock); diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index e9923ca9c2d..7c1e5e5cccd 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -864,7 +864,7 @@ static void sctp_init_assoc(struct connection *con) static void tcp_connect_to_sock(struct connection *con) { int result = -EHOSTUNREACH; - struct sockaddr_storage saddr; + struct sockaddr_storage saddr, src_addr; int addr_len; struct socket *sock; @@ -898,6 +898,17 @@ static void tcp_connect_to_sock(struct connection *con) con->connect_action = tcp_connect_to_sock; add_sock(sock, con); + /* Bind to our cluster-known address connecting to avoid + routing problems */ + memcpy(&src_addr, dlm_local_addr[0], sizeof(src_addr)); + make_sockaddr(&src_addr, 0, &addr_len); + result = sock->ops->bind(sock, (struct sockaddr *) &src_addr, + addr_len); + if (result < 0) { + log_print("could not bind for connect: %d", result); + /* This *may* not indicate a critical error */ + } + make_sockaddr(&saddr, dlm_config.ci_tcp_port, &addr_len); log_print("connecting to %d", con->nodeid); @@ -1426,6 +1437,8 @@ void dlm_lowcomms_stop(void) con = __nodeid2con(i, 0); if (con) { close_connection(con, true); + if (con->othercon) + kmem_cache_free(con_cache, con->othercon); kmem_cache_free(con_cache, con); } } diff --git a/fs/dlm/main.c b/fs/dlm/main.c index eca2907f238..58487fb95a4 100644 --- a/fs/dlm/main.c +++ b/fs/dlm/main.c @@ -18,16 +18,6 @@ #include "memory.h" #include "config.h" -#ifdef CONFIG_DLM_DEBUG -int dlm_register_debugfs(void); -void dlm_unregister_debugfs(void); -#else -static inline int dlm_register_debugfs(void) { return 0; } -static inline void dlm_unregister_debugfs(void) { } -#endif -int dlm_netlink_init(void); -void dlm_netlink_exit(void); - static int __init init_dlm(void) { int error; diff --git a/fs/dlm/member.c b/fs/dlm/member.c index e9cdcab306e..fa17f5a2788 100644 --- a/fs/dlm/member.c +++ b/fs/dlm/member.c @@ -1,7 +1,7 @@ /****************************************************************************** ******************************************************************************* ** -** Copyright (C) 2005-2007 Red Hat, Inc. All rights reserved. +** Copyright (C) 2005-2008 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -70,7 +70,7 @@ static void dlm_remove_member(struct dlm_ls *ls, struct dlm_member *memb) ls->ls_num_nodes--; } -static int dlm_is_member(struct dlm_ls *ls, int nodeid) +int dlm_is_member(struct dlm_ls *ls, int nodeid) { struct dlm_member *memb; diff --git a/fs/dlm/member.h b/fs/dlm/member.h index 927c08c1921..7a26fca1e0b 100644 --- a/fs/dlm/member.h +++ b/fs/dlm/member.h @@ -1,7 +1,7 @@ /****************************************************************************** ******************************************************************************* ** -** Copyright (C) 2005 Red Hat, Inc. All rights reserved. +** Copyright (C) 2005-2008 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -19,6 +19,7 @@ void dlm_clear_members(struct dlm_ls *ls); void dlm_clear_members_gone(struct dlm_ls *ls); int dlm_recover_members(struct dlm_ls *ls, struct dlm_recover *rv,int *neg_out); int dlm_is_removed(struct dlm_ls *ls, int nodeid); +int dlm_is_member(struct dlm_ls *ls, int nodeid); #endif /* __MEMBER_DOT_H__ */ diff --git a/fs/dlm/memory.c b/fs/dlm/memory.c index ecf0e5cb203..54c14c6d06c 100644 --- a/fs/dlm/memory.c +++ b/fs/dlm/memory.c @@ -2,7 +2,7 @@ ******************************************************************************* ** ** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. -** Copyright (C) 2004-2005 Red Hat, Inc. All rights reserved. +** Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -18,7 +18,7 @@ static struct kmem_cache *lkb_cache; -int dlm_memory_init(void) +int __init dlm_memory_init(void) { int ret = 0; @@ -35,7 +35,7 @@ void dlm_memory_exit(void) kmem_cache_destroy(lkb_cache); } -char *allocate_lvb(struct dlm_ls *ls) +char *dlm_allocate_lvb(struct dlm_ls *ls) { char *p; @@ -43,7 +43,7 @@ char *allocate_lvb(struct dlm_ls *ls) return p; } -void free_lvb(char *p) +void dlm_free_lvb(char *p) { kfree(p); } @@ -51,7 +51,7 @@ void free_lvb(char *p) /* FIXME: have some minimal space built-in to rsb for the name and kmalloc a separate name if needed, like dentries are done */ -struct dlm_rsb *allocate_rsb(struct dlm_ls *ls, int namelen) +struct dlm_rsb *dlm_allocate_rsb(struct dlm_ls *ls, int namelen) { struct dlm_rsb *r; @@ -61,14 +61,14 @@ struct dlm_rsb *allocate_rsb(struct dlm_ls *ls, int namelen) return r; } -void free_rsb(struct dlm_rsb *r) +void dlm_free_rsb(struct dlm_rsb *r) { if (r->res_lvbptr) - free_lvb(r->res_lvbptr); + dlm_free_lvb(r->res_lvbptr); kfree(r); } -struct dlm_lkb *allocate_lkb(struct dlm_ls *ls) +struct dlm_lkb *dlm_allocate_lkb(struct dlm_ls *ls) { struct dlm_lkb *lkb; @@ -76,11 +76,11 @@ struct dlm_lkb *allocate_lkb(struct dlm_ls *ls) return lkb; } -void free_lkb(struct dlm_lkb *lkb) +void dlm_free_lkb(struct dlm_lkb *lkb) { if (lkb->lkb_flags & DLM_IFL_USER) { struct dlm_user_args *ua; - ua = (struct dlm_user_args *)lkb->lkb_astparam; + ua = lkb->lkb_ua; if (ua) { if (ua->lksb.sb_lvbptr) kfree(ua->lksb.sb_lvbptr); @@ -90,19 +90,3 @@ void free_lkb(struct dlm_lkb *lkb) kmem_cache_free(lkb_cache, lkb); } -struct dlm_direntry *allocate_direntry(struct dlm_ls *ls, int namelen) -{ - struct dlm_direntry *de; - - DLM_ASSERT(namelen <= DLM_RESNAME_MAXLEN, - printk("namelen = %d\n", namelen);); - - de = kzalloc(sizeof(*de) + namelen, GFP_KERNEL); - return de; -} - -void free_direntry(struct dlm_direntry *de) -{ - kfree(de); -} - diff --git a/fs/dlm/memory.h b/fs/dlm/memory.h index 6ead158ccc5..485fb29143b 100644 --- a/fs/dlm/memory.h +++ b/fs/dlm/memory.h @@ -2,7 +2,7 @@ ******************************************************************************* ** ** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. -** Copyright (C) 2004-2005 Red Hat, Inc. All rights reserved. +** Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -16,14 +16,12 @@ int dlm_memory_init(void); void dlm_memory_exit(void); -struct dlm_rsb *allocate_rsb(struct dlm_ls *ls, int namelen); -void free_rsb(struct dlm_rsb *r); -struct dlm_lkb *allocate_lkb(struct dlm_ls *ls); -void free_lkb(struct dlm_lkb *l); -struct dlm_direntry *allocate_direntry(struct dlm_ls *ls, int namelen); -void free_direntry(struct dlm_direntry *de); -char *allocate_lvb(struct dlm_ls *ls); -void free_lvb(char *l); +struct dlm_rsb *dlm_allocate_rsb(struct dlm_ls *ls, int namelen); +void dlm_free_rsb(struct dlm_rsb *r); +struct dlm_lkb *dlm_allocate_lkb(struct dlm_ls *ls); +void dlm_free_lkb(struct dlm_lkb *l); +char *dlm_allocate_lvb(struct dlm_ls *ls); +void dlm_free_lvb(char *l); #endif /* __MEMORY_DOT_H__ */ diff --git a/fs/dlm/midcomms.c b/fs/dlm/midcomms.c index f8c69dda16a..07ac709f3ed 100644 --- a/fs/dlm/midcomms.c +++ b/fs/dlm/midcomms.c @@ -2,7 +2,7 @@ ******************************************************************************* ** ** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. -** Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved. +** Copyright (C) 2004-2008 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -58,8 +58,12 @@ static void copy_from_cb(void *dst, const void *base, unsigned offset, int dlm_process_incoming_buffer(int nodeid, const void *base, unsigned offset, unsigned len, unsigned limit) { - unsigned char __tmp[DLM_INBUF_LEN]; - struct dlm_header *msg = (struct dlm_header *) __tmp; + union { + unsigned char __buf[DLM_INBUF_LEN]; + /* this is to force proper alignment on some arches */ + union dlm_packet p; + } __tmp; + union dlm_packet *p = &__tmp.p; int ret = 0; int err = 0; uint16_t msglen; @@ -71,15 +75,22 @@ int dlm_process_incoming_buffer(int nodeid, const void *base, message may wrap around the end of the buffer back to the start, so we need to use a temp buffer and copy_from_cb. */ - copy_from_cb(msg, base, offset, sizeof(struct dlm_header), + copy_from_cb(p, base, offset, sizeof(struct dlm_header), limit); - msglen = le16_to_cpu(msg->h_length); - lockspace = msg->h_lockspace; + msglen = le16_to_cpu(p->header.h_length); + lockspace = p->header.h_lockspace; err = -EINVAL; if (msglen < sizeof(struct dlm_header)) break; + if (p->header.h_cmd == DLM_MSG) { + if (msglen < sizeof(struct dlm_message)) + break; + } else { + if (msglen < sizeof(struct dlm_rcom)) + break; + } err = -E2BIG; if (msglen > dlm_config.ci_buffer_size) { log_print("message size %d from %d too big, buf len %d", @@ -100,27 +111,26 @@ int dlm_process_incoming_buffer(int nodeid, const void *base, in the buffer on the stack (which should work for most ordinary messages). */ - if (msglen > sizeof(__tmp) && - msg == (struct dlm_header *) __tmp) { - msg = kmalloc(dlm_config.ci_buffer_size, GFP_KERNEL); - if (msg == NULL) + if (msglen > sizeof(__tmp) && p == &__tmp.p) { + p = kmalloc(dlm_config.ci_buffer_size, GFP_KERNEL); + if (p == NULL) return ret; } - copy_from_cb(msg, base, offset, msglen, limit); + copy_from_cb(p, base, offset, msglen, limit); - BUG_ON(lockspace != msg->h_lockspace); + BUG_ON(lockspace != p->header.h_lockspace); ret += msglen; offset += msglen; offset &= (limit - 1); len -= msglen; - dlm_receive_buffer(msg, nodeid); + dlm_receive_buffer(p, nodeid); } - if (msg != (struct dlm_header *) __tmp) - kfree(msg); + if (p != &__tmp.p) + kfree(p); return err ? err : ret; } diff --git a/fs/dlm/netlink.c b/fs/dlm/netlink.c index 863b87d0dc7..714593621f4 100644 --- a/fs/dlm/netlink.c +++ b/fs/dlm/netlink.c @@ -78,7 +78,7 @@ static struct genl_ops dlm_nl_ops = { .doit = user_cmd, }; -int dlm_netlink_init(void) +int __init dlm_netlink_init(void) { int rv; @@ -95,7 +95,7 @@ int dlm_netlink_init(void) return rv; } -void dlm_netlink_exit(void) +void __exit dlm_netlink_exit(void) { genl_unregister_ops(&family, &dlm_nl_ops); genl_unregister_family(&family); @@ -104,7 +104,6 @@ void dlm_netlink_exit(void) static void fill_data(struct dlm_lock_data *data, struct dlm_lkb *lkb) { struct dlm_rsb *r = lkb->lkb_resource; - struct dlm_user_args *ua = (struct dlm_user_args *) lkb->lkb_astparam; memset(data, 0, sizeof(struct dlm_lock_data)); @@ -117,8 +116,8 @@ static void fill_data(struct dlm_lock_data *data, struct dlm_lkb *lkb) data->grmode = lkb->lkb_grmode; data->rqmode = lkb->lkb_rqmode; data->timestamp = lkb->lkb_timestamp; - if (ua) - data->xid = ua->xid; + if (lkb->lkb_ua) + data->xid = lkb->lkb_ua->xid; if (r) { data->lockspace_id = r->res_ls->ls_global_id; data->resource_namelen = r->res_length; diff --git a/fs/dlm/rcom.c b/fs/dlm/rcom.c index ae2fd97fa4a..67522c268c1 100644 --- a/fs/dlm/rcom.c +++ b/fs/dlm/rcom.c @@ -2,7 +2,7 @@ ******************************************************************************* ** ** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. -** Copyright (C) 2005-2007 Red Hat, Inc. All rights reserved. +** Copyright (C) 2005-2008 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -78,13 +78,14 @@ static void send_rcom(struct dlm_ls *ls, struct dlm_mhandle *mh, static void make_config(struct dlm_ls *ls, struct rcom_config *rf) { - rf->rf_lvblen = ls->ls_lvblen; - rf->rf_lsflags = ls->ls_exflags; + rf->rf_lvblen = cpu_to_le32(ls->ls_lvblen); + rf->rf_lsflags = cpu_to_le32(ls->ls_exflags); } static int check_config(struct dlm_ls *ls, struct dlm_rcom *rc, int nodeid) { struct rcom_config *rf = (struct rcom_config *) rc->rc_buf; + size_t conf_size = sizeof(struct dlm_rcom) + sizeof(struct rcom_config); if ((rc->rc_header.h_version & 0xFFFF0000) != DLM_HEADER_MAJOR) { log_error(ls, "version mismatch: %x nodeid %d: %x", @@ -93,11 +94,18 @@ static int check_config(struct dlm_ls *ls, struct dlm_rcom *rc, int nodeid) return -EPROTO; } - if (rf->rf_lvblen != ls->ls_lvblen || - rf->rf_lsflags != ls->ls_exflags) { + if (rc->rc_header.h_length < conf_size) { + log_error(ls, "config too short: %d nodeid %d", + rc->rc_header.h_length, nodeid); + return -EPROTO; + } + + if (le32_to_cpu(rf->rf_lvblen) != ls->ls_lvblen || + le32_to_cpu(rf->rf_lsflags) != ls->ls_exflags) { log_error(ls, "config mismatch: %d,%x nodeid %d: %d,%x", - ls->ls_lvblen, ls->ls_exflags, - nodeid, rf->rf_lvblen, rf->rf_lsflags); + ls->ls_lvblen, ls->ls_exflags, nodeid, + le32_to_cpu(rf->rf_lvblen), + le32_to_cpu(rf->rf_lsflags)); return -EPROTO; } return 0; @@ -128,7 +136,7 @@ int dlm_rcom_status(struct dlm_ls *ls, int nodeid) ls->ls_recover_nodeid = nodeid; if (nodeid == dlm_our_nodeid()) { - rc = (struct dlm_rcom *) ls->ls_recover_buf; + rc = ls->ls_recover_buf; rc->rc_result = dlm_recover_status(ls); goto out; } @@ -147,7 +155,7 @@ int dlm_rcom_status(struct dlm_ls *ls, int nodeid) if (error) goto out; - rc = (struct dlm_rcom *) ls->ls_recover_buf; + rc = ls->ls_recover_buf; if (rc->rc_result == -ESRCH) { /* we pretend the remote lockspace exists with 0 status */ @@ -197,23 +205,21 @@ static void receive_sync_reply(struct dlm_ls *ls, struct dlm_rcom *rc_in) spin_unlock(&ls->ls_rcom_spin); } -static void receive_rcom_status_reply(struct dlm_ls *ls, struct dlm_rcom *rc_in) -{ - receive_sync_reply(ls, rc_in); -} - int dlm_rcom_names(struct dlm_ls *ls, int nodeid, char *last_name, int last_len) { struct dlm_rcom *rc; struct dlm_mhandle *mh; - int error = 0, len = sizeof(struct dlm_rcom); + int error = 0; + int max_size = dlm_config.ci_buffer_size - sizeof(struct dlm_rcom); ls->ls_recover_nodeid = nodeid; if (nodeid == dlm_our_nodeid()) { + ls->ls_recover_buf->rc_header.h_length = + dlm_config.ci_buffer_size; dlm_copy_master_names(ls, last_name, last_len, - ls->ls_recover_buf + len, - dlm_config.ci_buffer_size - len, nodeid); + ls->ls_recover_buf->rc_buf, + max_size, nodeid); goto out; } @@ -254,11 +260,6 @@ static void receive_rcom_names(struct dlm_ls *ls, struct dlm_rcom *rc_in) send_rcom(ls, mh, rc); } -static void receive_rcom_names_reply(struct dlm_ls *ls, struct dlm_rcom *rc_in) -{ - receive_sync_reply(ls, rc_in); -} - int dlm_send_rcom_lookup(struct dlm_rsb *r, int dir_nodeid) { struct dlm_rcom *rc; @@ -309,22 +310,22 @@ static void pack_rcom_lock(struct dlm_rsb *r, struct dlm_lkb *lkb, { memset(rl, 0, sizeof(*rl)); - rl->rl_ownpid = lkb->lkb_ownpid; - rl->rl_lkid = lkb->lkb_id; - rl->rl_exflags = lkb->lkb_exflags; - rl->rl_flags = lkb->lkb_flags; - rl->rl_lvbseq = lkb->lkb_lvbseq; + rl->rl_ownpid = cpu_to_le32(lkb->lkb_ownpid); + rl->rl_lkid = cpu_to_le32(lkb->lkb_id); + rl->rl_exflags = cpu_to_le32(lkb->lkb_exflags); + rl->rl_flags = cpu_to_le32(lkb->lkb_flags); + rl->rl_lvbseq = cpu_to_le32(lkb->lkb_lvbseq); rl->rl_rqmode = lkb->lkb_rqmode; rl->rl_grmode = lkb->lkb_grmode; rl->rl_status = lkb->lkb_status; - rl->rl_wait_type = lkb->lkb_wait_type; + rl->rl_wait_type = cpu_to_le16(lkb->lkb_wait_type); - if (lkb->lkb_bastaddr) + if (lkb->lkb_bastfn) rl->rl_asts |= AST_BAST; - if (lkb->lkb_astaddr) + if (lkb->lkb_astfn) rl->rl_asts |= AST_COMP; - rl->rl_namelen = r->res_length; + rl->rl_namelen = cpu_to_le16(r->res_length); memcpy(rl->rl_name, r->res_name, r->res_length); /* FIXME: might we have an lvb without DLM_LKF_VALBLK set ? @@ -358,6 +359,7 @@ int dlm_send_rcom_lock(struct dlm_rsb *r, struct dlm_lkb *lkb) return error; } +/* needs at least dlm_rcom + rcom_lock */ static void receive_rcom_lock(struct dlm_ls *ls, struct dlm_rcom *rc_in) { struct dlm_rcom *rc; @@ -381,11 +383,6 @@ static void receive_rcom_lock(struct dlm_ls *ls, struct dlm_rcom *rc_in) send_rcom(ls, mh, rc); } -static void receive_rcom_lock_reply(struct dlm_ls *ls, struct dlm_rcom *rc_in) -{ - dlm_recover_process_copy(ls, rc_in); -} - /* If the lockspace doesn't exist then still send a status message back; it's possible that it just doesn't have its global_id yet. */ @@ -416,7 +413,7 @@ int dlm_send_ls_not_ready(int nodeid, struct dlm_rcom *rc_in) rc->rc_result = -ESRCH; rf = (struct rcom_config *) rc->rc_buf; - rf->rf_lvblen = -1; + rf->rf_lvblen = cpu_to_le32(~0U); dlm_rcom_out(rc); dlm_lowcomms_commit_buffer(mh); @@ -454,6 +451,8 @@ static int is_old_reply(struct dlm_ls *ls, struct dlm_rcom *rc) void dlm_receive_rcom(struct dlm_ls *ls, struct dlm_rcom *rc, int nodeid) { + int lock_size = sizeof(struct dlm_rcom) + sizeof(struct rcom_lock); + if (dlm_recovery_stopped(ls) && (rc->rc_type != DLM_RCOM_STATUS)) { log_debug(ls, "ignoring recovery message %x from %d", rc->rc_type, nodeid); @@ -477,15 +476,17 @@ void dlm_receive_rcom(struct dlm_ls *ls, struct dlm_rcom *rc, int nodeid) break; case DLM_RCOM_LOCK: + if (rc->rc_header.h_length < lock_size) + goto Eshort; receive_rcom_lock(ls, rc); break; case DLM_RCOM_STATUS_REPLY: - receive_rcom_status_reply(ls, rc); + receive_sync_reply(ls, rc); break; case DLM_RCOM_NAMES_REPLY: - receive_rcom_names_reply(ls, rc); + receive_sync_reply(ls, rc); break; case DLM_RCOM_LOOKUP_REPLY: @@ -493,13 +494,18 @@ void dlm_receive_rcom(struct dlm_ls *ls, struct dlm_rcom *rc, int nodeid) break; case DLM_RCOM_LOCK_REPLY: - receive_rcom_lock_reply(ls, rc); + if (rc->rc_header.h_length < lock_size) + goto Eshort; + dlm_recover_process_copy(ls, rc); break; default: - DLM_ASSERT(0, printk("rc_type=%x\n", rc->rc_type);); + log_error(ls, "receive_rcom bad type %d", rc->rc_type); } - out: +out: return; +Eshort: + log_error(ls, "recovery message %x from %d is too short", + rc->rc_type, nodeid); } diff --git a/fs/dlm/recover.c b/fs/dlm/recover.c index c2cc7694cd1..80aba5bdd4a 100644 --- a/fs/dlm/recover.c +++ b/fs/dlm/recover.c @@ -94,7 +94,7 @@ void dlm_set_recover_status(struct dlm_ls *ls, uint32_t status) static int wait_status_all(struct dlm_ls *ls, uint32_t wait_status) { - struct dlm_rcom *rc = (struct dlm_rcom *) ls->ls_recover_buf; + struct dlm_rcom *rc = ls->ls_recover_buf; struct dlm_member *memb; int error = 0, delay; @@ -123,7 +123,7 @@ static int wait_status_all(struct dlm_ls *ls, uint32_t wait_status) static int wait_status_low(struct dlm_ls *ls, uint32_t wait_status) { - struct dlm_rcom *rc = (struct dlm_rcom *) ls->ls_recover_buf; + struct dlm_rcom *rc = ls->ls_recover_buf; int error = 0, delay = 0, nodeid = ls->ls_low_nodeid; for (;;) { @@ -629,7 +629,7 @@ static void recover_lvb(struct dlm_rsb *r) goto out; if (!r->res_lvbptr) { - r->res_lvbptr = allocate_lvb(r->res_ls); + r->res_lvbptr = dlm_allocate_lvb(r->res_ls); if (!r->res_lvbptr) goto out; } @@ -731,6 +731,20 @@ int dlm_create_root_list(struct dlm_ls *ls) list_add(&r->res_root_list, &ls->ls_root_list); dlm_hold_rsb(r); } + + /* If we're using a directory, add tossed rsbs to the root + list; they'll have entries created in the new directory, + but no other recovery steps should do anything with them. */ + + if (dlm_no_directory(ls)) { + read_unlock(&ls->ls_rsbtbl[i].lock); + continue; + } + + list_for_each_entry(r, &ls->ls_rsbtbl[i].toss, res_hashchain) { + list_add(&r->res_root_list, &ls->ls_root_list); + dlm_hold_rsb(r); + } read_unlock(&ls->ls_rsbtbl[i].lock); } out: @@ -750,6 +764,11 @@ void dlm_release_root_list(struct dlm_ls *ls) up_write(&ls->ls_root_sem); } +/* If not using a directory, clear the entire toss list, there's no benefit to + caching the master value since it's fixed. If we are using a dir, keep the + rsb's we're the master of. Recovery will add them to the root list and from + there they'll be entered in the rebuilt directory. */ + void dlm_clear_toss_list(struct dlm_ls *ls) { struct dlm_rsb *r, *safe; @@ -759,8 +778,10 @@ void dlm_clear_toss_list(struct dlm_ls *ls) write_lock(&ls->ls_rsbtbl[i].lock); list_for_each_entry_safe(r, safe, &ls->ls_rsbtbl[i].toss, res_hashchain) { - list_del(&r->res_hashchain); - free_rsb(r); + if (dlm_no_directory(ls) || !is_master(r)) { + list_del(&r->res_hashchain); + dlm_free_rsb(r); + } } write_unlock(&ls->ls_rsbtbl[i].lock); } diff --git a/fs/dlm/recoverd.c b/fs/dlm/recoverd.c index 4b89e20eebe..997f9531d59 100644 --- a/fs/dlm/recoverd.c +++ b/fs/dlm/recoverd.c @@ -67,17 +67,18 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) dlm_astd_resume(); /* - * This list of root rsb's will be the basis of most of the recovery - * routines. + * Free non-master tossed rsb's. Master rsb's are kept on toss + * list and put on root list to be included in resdir recovery. */ - dlm_create_root_list(ls); + dlm_clear_toss_list(ls); /* - * Free all the tossed rsb's so we don't have to recover them. + * This list of root rsb's will be the basis of most of the recovery + * routines. */ - dlm_clear_toss_list(ls); + dlm_create_root_list(ls); /* * Add or remove nodes from the lockspace's ls_nodes list. diff --git a/fs/dlm/requestqueue.c b/fs/dlm/requestqueue.c index 0de04f17cce..daa4183fbb8 100644 --- a/fs/dlm/requestqueue.c +++ b/fs/dlm/requestqueue.c @@ -20,7 +20,7 @@ struct rq_entry { struct list_head list; int nodeid; - char request[0]; + struct dlm_message request; }; /* @@ -30,10 +30,10 @@ struct rq_entry { * lockspace is enabled on some while still suspended on others. */ -void dlm_add_requestqueue(struct dlm_ls *ls, int nodeid, struct dlm_header *hd) +void dlm_add_requestqueue(struct dlm_ls *ls, int nodeid, struct dlm_message *ms) { struct rq_entry *e; - int length = hd->h_length; + int length = ms->m_header.h_length - sizeof(struct dlm_message); e = kmalloc(sizeof(struct rq_entry) + length, GFP_KERNEL); if (!e) { @@ -42,7 +42,7 @@ void dlm_add_requestqueue(struct dlm_ls *ls, int nodeid, struct dlm_header *hd) } e->nodeid = nodeid; - memcpy(e->request, hd, length); + memcpy(&e->request, ms, ms->m_header.h_length); mutex_lock(&ls->ls_requestqueue_mutex); list_add_tail(&e->list, &ls->ls_requestqueue); @@ -76,7 +76,7 @@ int dlm_process_requestqueue(struct dlm_ls *ls) e = list_entry(ls->ls_requestqueue.next, struct rq_entry, list); mutex_unlock(&ls->ls_requestqueue_mutex); - dlm_receive_message_saved(ls, (struct dlm_message *)e->request); + dlm_receive_message_saved(ls, &e->request); mutex_lock(&ls->ls_requestqueue_mutex); list_del(&e->list); @@ -176,7 +176,7 @@ void dlm_purge_requestqueue(struct dlm_ls *ls) mutex_lock(&ls->ls_requestqueue_mutex); list_for_each_entry_safe(e, safe, &ls->ls_requestqueue, list) { - ms = (struct dlm_message *) e->request; + ms = &e->request; if (purge_request(ls, ms, e->nodeid)) { list_del(&e->list); diff --git a/fs/dlm/requestqueue.h b/fs/dlm/requestqueue.h index aba34fc05ee..10ce449b77d 100644 --- a/fs/dlm/requestqueue.h +++ b/fs/dlm/requestqueue.h @@ -13,7 +13,7 @@ #ifndef __REQUESTQUEUE_DOT_H__ #define __REQUESTQUEUE_DOT_H__ -void dlm_add_requestqueue(struct dlm_ls *ls, int nodeid, struct dlm_header *hd); +void dlm_add_requestqueue(struct dlm_ls *ls, int nodeid, struct dlm_message *ms); int dlm_process_requestqueue(struct dlm_ls *ls); void dlm_wait_requestqueue(struct dlm_ls *ls); void dlm_purge_requestqueue(struct dlm_ls *ls); diff --git a/fs/dlm/user.c b/fs/dlm/user.c index 4f741546f4b..ebbcf38fd33 100644 --- a/fs/dlm/user.c +++ b/fs/dlm/user.c @@ -24,8 +24,7 @@ #include "lvb_table.h" #include "user.h" -static const char *name_prefix="dlm"; -static struct miscdevice ctl_device; +static const char name_prefix[] = "dlm"; static const struct file_operations device_fops; #ifdef CONFIG_COMPAT @@ -82,7 +81,8 @@ struct dlm_lock_result32 { }; static void compat_input(struct dlm_write_request *kb, - struct dlm_write_request32 *kb32) + struct dlm_write_request32 *kb32, + size_t count) { kb->version[0] = kb32->version[0]; kb->version[1] = kb32->version[1]; @@ -94,7 +94,8 @@ static void compat_input(struct dlm_write_request *kb, kb->cmd == DLM_USER_REMOVE_LOCKSPACE) { kb->i.lspace.flags = kb32->i.lspace.flags; kb->i.lspace.minor = kb32->i.lspace.minor; - strcpy(kb->i.lspace.name, kb32->i.lspace.name); + memcpy(kb->i.lspace.name, kb32->i.lspace.name, count - + offsetof(struct dlm_write_request32, i.lspace.name)); } else if (kb->cmd == DLM_USER_PURGE) { kb->i.purge.nodeid = kb32->i.purge.nodeid; kb->i.purge.pid = kb32->i.purge.pid; @@ -112,7 +113,8 @@ static void compat_input(struct dlm_write_request *kb, kb->i.lock.bastaddr = (void *)(long)kb32->i.lock.bastaddr; kb->i.lock.lksb = (void *)(long)kb32->i.lock.lksb; memcpy(kb->i.lock.lvb, kb32->i.lock.lvb, DLM_USER_LVB_LEN); - memcpy(kb->i.lock.name, kb32->i.lock.name, kb->i.lock.namelen); + memcpy(kb->i.lock.name, kb32->i.lock.name, count - + offsetof(struct dlm_write_request32, i.lock.name)); } } @@ -193,8 +195,8 @@ void dlm_user_add_ast(struct dlm_lkb *lkb, int type) if (lkb->lkb_flags & (DLM_IFL_ORPHAN | DLM_IFL_DEAD)) goto out; - DLM_ASSERT(lkb->lkb_astparam, dlm_print_lkb(lkb);); - ua = (struct dlm_user_args *)lkb->lkb_astparam; + DLM_ASSERT(lkb->lkb_ua, dlm_print_lkb(lkb);); + ua = lkb->lkb_ua; proc = ua->proc; if (type == AST_BAST && ua->bastaddr == NULL) @@ -236,12 +238,12 @@ void dlm_user_add_ast(struct dlm_lkb *lkb, int type) spin_unlock(&proc->asts_spin); if (eol) { - spin_lock(&ua->proc->locks_spin); + spin_lock(&proc->locks_spin); if (!list_empty(&lkb->lkb_ownqueue)) { list_del_init(&lkb->lkb_ownqueue); dlm_put_lkb(lkb); } - spin_unlock(&ua->proc->locks_spin); + spin_unlock(&proc->locks_spin); } out: mutex_unlock(&ls->ls_clear_proc_locks); @@ -504,7 +506,7 @@ static ssize_t device_write(struct file *file, const char __user *buf, #endif return -EINVAL; - kbuf = kmalloc(count, GFP_KERNEL); + kbuf = kzalloc(count + 1, GFP_KERNEL); if (!kbuf) return -ENOMEM; @@ -522,14 +524,14 @@ static ssize_t device_write(struct file *file, const char __user *buf, if (!kbuf->is64bit) { struct dlm_write_request32 *k32buf; k32buf = (struct dlm_write_request32 *)kbuf; - kbuf = kmalloc(count + (sizeof(struct dlm_write_request) - + kbuf = kmalloc(count + 1 + (sizeof(struct dlm_write_request) - sizeof(struct dlm_write_request32)), GFP_KERNEL); if (!kbuf) return -ENOMEM; if (proc) set_bit(DLM_PROC_FLAGS_COMPAT, &proc->flags); - compat_input(kbuf, k32buf); + compat_input(kbuf, k32buf, count + 1); kfree(k32buf); } #endif @@ -769,7 +771,6 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count, { struct dlm_user_proc *proc = file->private_data; struct dlm_lkb *lkb; - struct dlm_user_args *ua; DECLARE_WAITQUEUE(wait, current); int error, type=0, bmode=0, removed = 0; @@ -840,8 +841,7 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count, } spin_unlock(&proc->asts_spin); - ua = (struct dlm_user_args *)lkb->lkb_astparam; - error = copy_result_to_user(ua, + error = copy_result_to_user(lkb->lkb_ua, test_bit(DLM_PROC_FLAGS_COMPAT, &proc->flags), type, bmode, buf, count); @@ -896,14 +896,16 @@ static const struct file_operations ctl_device_fops = { .owner = THIS_MODULE, }; -int dlm_user_init(void) +static struct miscdevice ctl_device = { + .name = "dlm-control", + .fops = &ctl_device_fops, + .minor = MISC_DYNAMIC_MINOR, +}; + +int __init dlm_user_init(void) { int error; - ctl_device.name = "dlm-control"; - ctl_device.fops = &ctl_device_fops; - ctl_device.minor = MISC_DYNAMIC_MINOR; - error = misc_register(&ctl_device); if (error) log_print("misc_register failed for control device"); diff --git a/fs/dlm/util.c b/fs/dlm/util.c index 963889cf674..e36520af7cc 100644 --- a/fs/dlm/util.c +++ b/fs/dlm/util.c @@ -1,7 +1,7 @@ /****************************************************************************** ******************************************************************************* ** -** Copyright (C) 2005 Red Hat, Inc. All rights reserved. +** Copyright (C) 2005-2008 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -14,6 +14,14 @@ #include "rcom.h" #include "util.h" +#define DLM_ERRNO_EDEADLK 35 +#define DLM_ERRNO_EBADR 53 +#define DLM_ERRNO_EBADSLT 57 +#define DLM_ERRNO_EPROTO 71 +#define DLM_ERRNO_EOPNOTSUPP 95 +#define DLM_ERRNO_ETIMEDOUT 110 +#define DLM_ERRNO_EINPROGRESS 115 + static void header_out(struct dlm_header *hd) { hd->h_version = cpu_to_le32(hd->h_version); @@ -30,11 +38,54 @@ static void header_in(struct dlm_header *hd) hd->h_length = le16_to_cpu(hd->h_length); } -void dlm_message_out(struct dlm_message *ms) +/* higher errno values are inconsistent across architectures, so select + one set of values for on the wire */ + +static int to_dlm_errno(int err) { - struct dlm_header *hd = (struct dlm_header *) ms; + switch (err) { + case -EDEADLK: + return -DLM_ERRNO_EDEADLK; + case -EBADR: + return -DLM_ERRNO_EBADR; + case -EBADSLT: + return -DLM_ERRNO_EBADSLT; + case -EPROTO: + return -DLM_ERRNO_EPROTO; + case -EOPNOTSUPP: + return -DLM_ERRNO_EOPNOTSUPP; + case -ETIMEDOUT: + return -DLM_ERRNO_ETIMEDOUT; + case -EINPROGRESS: + return -DLM_ERRNO_EINPROGRESS; + } + return err; +} - header_out(hd); +static int from_dlm_errno(int err) +{ + switch (err) { + case -DLM_ERRNO_EDEADLK: + return -EDEADLK; + case -DLM_ERRNO_EBADR: + return -EBADR; + case -DLM_ERRNO_EBADSLT: + return -EBADSLT; + case -DLM_ERRNO_EPROTO: + return -EPROTO; + case -DLM_ERRNO_EOPNOTSUPP: + return -EOPNOTSUPP; + case -DLM_ERRNO_ETIMEDOUT: + return -ETIMEDOUT; + case -DLM_ERRNO_EINPROGRESS: + return -EINPROGRESS; + } + return err; +} + +void dlm_message_out(struct dlm_message *ms) +{ + header_out(&ms->m_header); ms->m_type = cpu_to_le32(ms->m_type); ms->m_nodeid = cpu_to_le32(ms->m_nodeid); @@ -53,14 +104,12 @@ void dlm_message_out(struct dlm_message *ms) ms->m_rqmode = cpu_to_le32(ms->m_rqmode); ms->m_bastmode = cpu_to_le32(ms->m_bastmode); ms->m_asts = cpu_to_le32(ms->m_asts); - ms->m_result = cpu_to_le32(ms->m_result); + ms->m_result = cpu_to_le32(to_dlm_errno(ms->m_result)); } void dlm_message_in(struct dlm_message *ms) { - struct dlm_header *hd = (struct dlm_header *) ms; - - header_in(hd); + header_in(&ms->m_header); ms->m_type = le32_to_cpu(ms->m_type); ms->m_nodeid = le32_to_cpu(ms->m_nodeid); @@ -79,87 +128,27 @@ void dlm_message_in(struct dlm_message *ms) ms->m_rqmode = le32_to_cpu(ms->m_rqmode); ms->m_bastmode = le32_to_cpu(ms->m_bastmode); ms->m_asts = le32_to_cpu(ms->m_asts); - ms->m_result = le32_to_cpu(ms->m_result); -} - -static void rcom_lock_out(struct rcom_lock *rl) -{ - rl->rl_ownpid = cpu_to_le32(rl->rl_ownpid); - rl->rl_lkid = cpu_to_le32(rl->rl_lkid); - rl->rl_remid = cpu_to_le32(rl->rl_remid); - rl->rl_parent_lkid = cpu_to_le32(rl->rl_parent_lkid); - rl->rl_parent_remid = cpu_to_le32(rl->rl_parent_remid); - rl->rl_exflags = cpu_to_le32(rl->rl_exflags); - rl->rl_flags = cpu_to_le32(rl->rl_flags); - rl->rl_lvbseq = cpu_to_le32(rl->rl_lvbseq); - rl->rl_result = cpu_to_le32(rl->rl_result); - rl->rl_wait_type = cpu_to_le16(rl->rl_wait_type); - rl->rl_namelen = cpu_to_le16(rl->rl_namelen); -} - -static void rcom_lock_in(struct rcom_lock *rl) -{ - rl->rl_ownpid = le32_to_cpu(rl->rl_ownpid); - rl->rl_lkid = le32_to_cpu(rl->rl_lkid); - rl->rl_remid = le32_to_cpu(rl->rl_remid); - rl->rl_parent_lkid = le32_to_cpu(rl->rl_parent_lkid); - rl->rl_parent_remid = le32_to_cpu(rl->rl_parent_remid); - rl->rl_exflags = le32_to_cpu(rl->rl_exflags); - rl->rl_flags = le32_to_cpu(rl->rl_flags); - rl->rl_lvbseq = le32_to_cpu(rl->rl_lvbseq); - rl->rl_result = le32_to_cpu(rl->rl_result); - rl->rl_wait_type = le16_to_cpu(rl->rl_wait_type); - rl->rl_namelen = le16_to_cpu(rl->rl_namelen); -} - -static void rcom_config_out(struct rcom_config *rf) -{ - rf->rf_lvblen = cpu_to_le32(rf->rf_lvblen); - rf->rf_lsflags = cpu_to_le32(rf->rf_lsflags); -} - -static void rcom_config_in(struct rcom_config *rf) -{ - rf->rf_lvblen = le32_to_cpu(rf->rf_lvblen); - rf->rf_lsflags = le32_to_cpu(rf->rf_lsflags); + ms->m_result = from_dlm_errno(le32_to_cpu(ms->m_result)); } void dlm_rcom_out(struct dlm_rcom *rc) { - struct dlm_header *hd = (struct dlm_header *) rc; - int type = rc->rc_type; - - header_out(hd); + header_out(&rc->rc_header); rc->rc_type = cpu_to_le32(rc->rc_type); rc->rc_result = cpu_to_le32(rc->rc_result); rc->rc_id = cpu_to_le64(rc->rc_id); rc->rc_seq = cpu_to_le64(rc->rc_seq); rc->rc_seq_reply = cpu_to_le64(rc->rc_seq_reply); - - if (type == DLM_RCOM_LOCK) - rcom_lock_out((struct rcom_lock *) rc->rc_buf); - - else if (type == DLM_RCOM_STATUS_REPLY) - rcom_config_out((struct rcom_config *) rc->rc_buf); } void dlm_rcom_in(struct dlm_rcom *rc) { - struct dlm_header *hd = (struct dlm_header *) rc; - - header_in(hd); + header_in(&rc->rc_header); rc->rc_type = le32_to_cpu(rc->rc_type); rc->rc_result = le32_to_cpu(rc->rc_result); rc->rc_id = le64_to_cpu(rc->rc_id); rc->rc_seq = le64_to_cpu(rc->rc_seq); rc->rc_seq_reply = le64_to_cpu(rc->rc_seq_reply); - - if (rc->rc_type == DLM_RCOM_LOCK) - rcom_lock_in((struct rcom_lock *) rc->rc_buf); - - else if (rc->rc_type == DLM_RCOM_STATUS_REPLY) - rcom_config_in((struct rcom_config *) rc->rc_buf); } - diff --git a/fs/dquot.c b/fs/dquot.c index cee7c6f428f..41b9dbd68b0 100644 --- a/fs/dquot.c +++ b/fs/dquot.c @@ -696,9 +696,8 @@ static int dqinit_needed(struct inode *inode, int type) /* This routine is guarded by dqonoff_mutex mutex */ static void add_dquot_ref(struct super_block *sb, int type) { - struct inode *inode; + struct inode *inode, *old_inode = NULL; -restart: spin_lock(&inode_lock); list_for_each_entry(inode, &sb->s_inodes, i_sb_list) { if (!atomic_read(&inode->i_writecount)) @@ -711,12 +710,18 @@ restart: __iget(inode); spin_unlock(&inode_lock); + iput(old_inode); sb->dq_op->initialize(inode, type); - iput(inode); - /* As we may have blocked we had better restart... */ - goto restart; + /* We hold a reference to 'inode' so it couldn't have been + * removed from s_inodes list while we dropped the inode_lock. + * We cannot iput the inode now as we can be holding the last + * reference and we cannot iput it under inode_lock. So we + * keep the reference and iput it later. */ + old_inode = inode; + spin_lock(&inode_lock); } spin_unlock(&inode_lock); + iput(old_inode); } /* Return 0 if dqput() won't block (note that 1 doesn't necessarily mean blocking) */ @@ -1517,8 +1522,8 @@ int vfs_quota_off(struct super_block *sb, int type) truncate_inode_pages(&toputinode[cnt]->i_data, 0); mutex_unlock(&toputinode[cnt]->i_mutex); mark_inode_dirty(toputinode[cnt]); - iput(toputinode[cnt]); } + iput(toputinode[cnt]); mutex_unlock(&dqopt->dqonoff_mutex); } if (sb->s_bdev) @@ -1628,16 +1633,17 @@ int vfs_quota_on(struct super_block *sb, int type, int format_id, char *path) error = path_lookup(path, LOOKUP_FOLLOW, &nd); if (error < 0) return error; - error = security_quota_on(nd.dentry); + error = security_quota_on(nd.path.dentry); if (error) goto out_path; /* Quota file not on the same filesystem? */ - if (nd.mnt->mnt_sb != sb) + if (nd.path.mnt->mnt_sb != sb) error = -EXDEV; else - error = vfs_quota_on_inode(nd.dentry->d_inode, type, format_id); + error = vfs_quota_on_inode(nd.path.dentry->d_inode, type, + format_id); out_path: - path_release(&nd); + path_put(&nd.path); return error; } diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index f8ef0af919e..a066e109ad9 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -355,8 +355,11 @@ static int encrypt_scatterlist(struct ecryptfs_crypt_stat *crypt_stat, } /* Consider doing this once, when the file is opened */ mutex_lock(&crypt_stat->cs_tfm_mutex); - rc = crypto_blkcipher_setkey(crypt_stat->tfm, crypt_stat->key, - crypt_stat->key_size); + if (!(crypt_stat->flags & ECRYPTFS_KEY_SET)) { + rc = crypto_blkcipher_setkey(crypt_stat->tfm, crypt_stat->key, + crypt_stat->key_size); + crypt_stat->flags |= ECRYPTFS_KEY_SET; + } if (rc) { ecryptfs_printk(KERN_ERR, "Error setting key; rc = [%d]\n", rc); @@ -376,11 +379,10 @@ out: * * Convert an eCryptfs page index into a lower byte offset */ -void ecryptfs_lower_offset_for_extent(loff_t *offset, loff_t extent_num, - struct ecryptfs_crypt_stat *crypt_stat) +static void ecryptfs_lower_offset_for_extent(loff_t *offset, loff_t extent_num, + struct ecryptfs_crypt_stat *crypt_stat) { - (*offset) = ((crypt_stat->extent_size - * crypt_stat->num_header_extents_at_front) + (*offset) = (crypt_stat->num_header_bytes_at_front + (crypt_stat->extent_size * extent_num)); } @@ -842,15 +844,13 @@ void ecryptfs_set_default_sizes(struct ecryptfs_crypt_stat *crypt_stat) set_extent_mask_and_shift(crypt_stat); crypt_stat->iv_bytes = ECRYPTFS_DEFAULT_IV_BYTES; if (crypt_stat->flags & ECRYPTFS_METADATA_IN_XATTR) - crypt_stat->num_header_extents_at_front = 0; + crypt_stat->num_header_bytes_at_front = 0; else { if (PAGE_CACHE_SIZE <= ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE) - crypt_stat->num_header_extents_at_front = - (ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE - / crypt_stat->extent_size); + crypt_stat->num_header_bytes_at_front = + ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE; else - crypt_stat->num_header_extents_at_front = - (PAGE_CACHE_SIZE / crypt_stat->extent_size); + crypt_stat->num_header_bytes_at_front = PAGE_CACHE_SIZE; } } @@ -1128,7 +1128,7 @@ write_ecryptfs_flags(char *page_virt, struct ecryptfs_crypt_stat *crypt_stat, struct ecryptfs_cipher_code_str_map_elem { char cipher_str[16]; - u16 cipher_code; + u8 cipher_code; }; /* Add support for additional ciphers by adding elements here. The @@ -1152,10 +1152,10 @@ ecryptfs_cipher_code_str_map[] = { * * Returns zero on no match, or the cipher code on match */ -u16 ecryptfs_code_for_cipher_string(struct ecryptfs_crypt_stat *crypt_stat) +u8 ecryptfs_code_for_cipher_string(struct ecryptfs_crypt_stat *crypt_stat) { int i; - u16 code = 0; + u8 code = 0; struct ecryptfs_cipher_code_str_map_elem *map = ecryptfs_cipher_code_str_map; @@ -1187,7 +1187,7 @@ u16 ecryptfs_code_for_cipher_string(struct ecryptfs_crypt_stat *crypt_stat) * * Returns zero on success */ -int ecryptfs_cipher_code_to_string(char *str, u16 cipher_code) +int ecryptfs_cipher_code_to_string(char *str, u8 cipher_code) { int rc = 0; int i; @@ -1236,7 +1236,8 @@ ecryptfs_write_header_metadata(char *virt, header_extent_size = (u32)crypt_stat->extent_size; num_header_extents_at_front = - (u16)crypt_stat->num_header_extents_at_front; + (u16)(crypt_stat->num_header_bytes_at_front + / crypt_stat->extent_size); header_extent_size = cpu_to_be32(header_extent_size); memcpy(virt, &header_extent_size, 4); virt += 4; @@ -1311,40 +1312,16 @@ static int ecryptfs_write_headers_virt(char *page_virt, size_t *size, static int ecryptfs_write_metadata_to_contents(struct ecryptfs_crypt_stat *crypt_stat, struct dentry *ecryptfs_dentry, - char *page_virt) + char *virt) { - int current_header_page; - int header_pages; int rc; - rc = ecryptfs_write_lower(ecryptfs_dentry->d_inode, page_virt, - 0, PAGE_CACHE_SIZE); - if (rc) { + rc = ecryptfs_write_lower(ecryptfs_dentry->d_inode, virt, + 0, crypt_stat->num_header_bytes_at_front); + if (rc) printk(KERN_ERR "%s: Error attempting to write header " "information to lower file; rc = [%d]\n", __FUNCTION__, rc); - goto out; - } - header_pages = ((crypt_stat->extent_size - * crypt_stat->num_header_extents_at_front) - / PAGE_CACHE_SIZE); - memset(page_virt, 0, PAGE_CACHE_SIZE); - current_header_page = 1; - while (current_header_page < header_pages) { - loff_t offset; - - offset = (((loff_t)current_header_page) << PAGE_CACHE_SHIFT); - if ((rc = ecryptfs_write_lower(ecryptfs_dentry->d_inode, - page_virt, offset, - PAGE_CACHE_SIZE))) { - printk(KERN_ERR "%s: Error attempting to write header " - "information to lower file; rc = [%d]\n", - __FUNCTION__, rc); - goto out; - } - current_header_page++; - } -out: return rc; } @@ -1370,15 +1347,13 @@ ecryptfs_write_metadata_to_xattr(struct dentry *ecryptfs_dentry, * retrieved via a prompt. Exactly what happens at this point should * be policy-dependent. * - * TODO: Support header information spanning multiple pages - * * Returns zero on success; non-zero on error */ int ecryptfs_write_metadata(struct dentry *ecryptfs_dentry) { struct ecryptfs_crypt_stat *crypt_stat = &ecryptfs_inode_to_private(ecryptfs_dentry->d_inode)->crypt_stat; - char *page_virt; + char *virt; size_t size = 0; int rc = 0; @@ -1389,40 +1364,39 @@ int ecryptfs_write_metadata(struct dentry *ecryptfs_dentry) goto out; } } else { + printk(KERN_WARNING "%s: Encrypted flag not set\n", + __FUNCTION__); rc = -EINVAL; - ecryptfs_printk(KERN_WARNING, - "Called with crypt_stat->encrypted == 0\n"); goto out; } /* Released in this function */ - page_virt = kmem_cache_zalloc(ecryptfs_header_cache_0, GFP_USER); - if (!page_virt) { - ecryptfs_printk(KERN_ERR, "Out of memory\n"); + virt = kzalloc(crypt_stat->num_header_bytes_at_front, GFP_KERNEL); + if (!virt) { + printk(KERN_ERR "%s: Out of memory\n", __FUNCTION__); rc = -ENOMEM; goto out; } - rc = ecryptfs_write_headers_virt(page_virt, &size, crypt_stat, - ecryptfs_dentry); + rc = ecryptfs_write_headers_virt(virt, &size, crypt_stat, + ecryptfs_dentry); if (unlikely(rc)) { - ecryptfs_printk(KERN_ERR, "Error whilst writing headers\n"); - memset(page_virt, 0, PAGE_CACHE_SIZE); + printk(KERN_ERR "%s: Error whilst writing headers; rc = [%d]\n", + __FUNCTION__, rc); goto out_free; } if (crypt_stat->flags & ECRYPTFS_METADATA_IN_XATTR) rc = ecryptfs_write_metadata_to_xattr(ecryptfs_dentry, - crypt_stat, page_virt, - size); + crypt_stat, virt, size); else rc = ecryptfs_write_metadata_to_contents(crypt_stat, - ecryptfs_dentry, - page_virt); + ecryptfs_dentry, virt); if (rc) { - printk(KERN_ERR "Error writing metadata out to lower file; " - "rc = [%d]\n", rc); + printk(KERN_ERR "%s: Error writing metadata out to lower file; " + "rc = [%d]\n", __FUNCTION__, rc); goto out_free; } out_free: - kmem_cache_free(ecryptfs_header_cache_0, page_virt); + memset(virt, 0, crypt_stat->num_header_bytes_at_front); + kfree(virt); out: return rc; } @@ -1442,16 +1416,16 @@ static int parse_header_metadata(struct ecryptfs_crypt_stat *crypt_stat, virt += sizeof(u32); memcpy(&num_header_extents_at_front, virt, sizeof(u16)); num_header_extents_at_front = be16_to_cpu(num_header_extents_at_front); - crypt_stat->num_header_extents_at_front = - (int)num_header_extents_at_front; + crypt_stat->num_header_bytes_at_front = + (((size_t)num_header_extents_at_front + * (size_t)header_extent_size)); (*bytes_read) = (sizeof(u32) + sizeof(u16)); if ((validate_header_size == ECRYPTFS_VALIDATE_HEADER_SIZE) - && ((crypt_stat->extent_size - * crypt_stat->num_header_extents_at_front) + && (crypt_stat->num_header_bytes_at_front < ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE)) { rc = -EINVAL; - printk(KERN_WARNING "Invalid number of header extents: [%zd]\n", - crypt_stat->num_header_extents_at_front); + printk(KERN_WARNING "Invalid header size: [%zd]\n", + crypt_stat->num_header_bytes_at_front); } return rc; } @@ -1466,7 +1440,8 @@ static int parse_header_metadata(struct ecryptfs_crypt_stat *crypt_stat, */ static void set_default_header_data(struct ecryptfs_crypt_stat *crypt_stat) { - crypt_stat->num_header_extents_at_front = 2; + crypt_stat->num_header_bytes_at_front = + ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE; } /** @@ -1552,9 +1527,10 @@ int ecryptfs_read_xattr_region(char *page_virt, struct inode *ecryptfs_inode) size = ecryptfs_getxattr_lower(lower_dentry, ECRYPTFS_XATTR_NAME, page_virt, ECRYPTFS_DEFAULT_EXTENT_SIZE); if (size < 0) { - printk(KERN_ERR "Error attempting to read the [%s] " - "xattr from the lower file; return value = [%zd]\n", - ECRYPTFS_XATTR_NAME, size); + if (unlikely(ecryptfs_verbosity > 0)) + printk(KERN_INFO "Error attempting to read the [%s] " + "xattr from the lower file; return value = " + "[%zd]\n", ECRYPTFS_XATTR_NAME, size); rc = -EINVAL; goto out; } @@ -1802,7 +1778,7 @@ out: } struct kmem_cache *ecryptfs_key_tfm_cache; -struct list_head key_tfm_list; +static struct list_head key_tfm_list; struct mutex key_tfm_list_mutex; int ecryptfs_init_crypto(void) @@ -1812,6 +1788,11 @@ int ecryptfs_init_crypto(void) return 0; } +/** + * ecryptfs_destroy_crypto - free all cached key_tfms on key_tfm_list + * + * Called only at module unload time + */ int ecryptfs_destroy_crypto(void) { struct ecryptfs_key_tfm *key_tfm, *key_tfm_tmp; @@ -1835,6 +1816,8 @@ ecryptfs_add_new_key_tfm(struct ecryptfs_key_tfm **key_tfm, char *cipher_name, struct ecryptfs_key_tfm *tmp_tfm; int rc = 0; + BUG_ON(!mutex_is_locked(&key_tfm_list_mutex)); + tmp_tfm = kmem_cache_alloc(ecryptfs_key_tfm_cache, GFP_KERNEL); if (key_tfm != NULL) (*key_tfm) = tmp_tfm; @@ -1861,13 +1844,50 @@ ecryptfs_add_new_key_tfm(struct ecryptfs_key_tfm **key_tfm, char *cipher_name, (*key_tfm) = NULL; goto out; } - mutex_lock(&key_tfm_list_mutex); list_add(&tmp_tfm->key_tfm_list, &key_tfm_list); - mutex_unlock(&key_tfm_list_mutex); out: return rc; } +/** + * ecryptfs_tfm_exists - Search for existing tfm for cipher_name. + * @cipher_name: the name of the cipher to search for + * @key_tfm: set to corresponding tfm if found + * + * Searches for cached key_tfm matching @cipher_name + * Must be called with &key_tfm_list_mutex held + * Returns 1 if found, with @key_tfm set + * Returns 0 if not found, with @key_tfm set to NULL + */ +int ecryptfs_tfm_exists(char *cipher_name, struct ecryptfs_key_tfm **key_tfm) +{ + struct ecryptfs_key_tfm *tmp_key_tfm; + + BUG_ON(!mutex_is_locked(&key_tfm_list_mutex)); + + list_for_each_entry(tmp_key_tfm, &key_tfm_list, key_tfm_list) { + if (strcmp(tmp_key_tfm->cipher_name, cipher_name) == 0) { + if (key_tfm) + (*key_tfm) = tmp_key_tfm; + return 1; + } + } + if (key_tfm) + (*key_tfm) = NULL; + return 0; +} + +/** + * ecryptfs_get_tfm_and_mutex_for_cipher_name + * + * @tfm: set to cached tfm found, or new tfm created + * @tfm_mutex: set to mutex for cached tfm found, or new tfm created + * @cipher_name: the name of the cipher to search for and/or add + * + * Sets pointers to @tfm & @tfm_mutex matching @cipher_name. + * Searches for cached item first, and creates new if not found. + * Returns 0 on success, non-zero if adding new cipher failed + */ int ecryptfs_get_tfm_and_mutex_for_cipher_name(struct crypto_blkcipher **tfm, struct mutex **tfm_mutex, char *cipher_name) @@ -1877,22 +1897,17 @@ int ecryptfs_get_tfm_and_mutex_for_cipher_name(struct crypto_blkcipher **tfm, (*tfm) = NULL; (*tfm_mutex) = NULL; + mutex_lock(&key_tfm_list_mutex); - list_for_each_entry(key_tfm, &key_tfm_list, key_tfm_list) { - if (strcmp(key_tfm->cipher_name, cipher_name) == 0) { - (*tfm) = key_tfm->key_tfm; - (*tfm_mutex) = &key_tfm->key_tfm_mutex; - mutex_unlock(&key_tfm_list_mutex); + if (!ecryptfs_tfm_exists(cipher_name, &key_tfm)) { + rc = ecryptfs_add_new_key_tfm(&key_tfm, cipher_name, 0); + if (rc) { + printk(KERN_ERR "Error adding new key_tfm to list; " + "rc = [%d]\n", rc); goto out; } } mutex_unlock(&key_tfm_list_mutex); - rc = ecryptfs_add_new_key_tfm(&key_tfm, cipher_name, 0); - if (rc) { - printk(KERN_ERR "Error adding new key_tfm to list; rc = [%d]\n", - rc); - goto out; - } (*tfm) = key_tfm->key_tfm; (*tfm_mutex) = &key_tfm->key_tfm_mutex; out: diff --git a/fs/ecryptfs/dentry.c b/fs/ecryptfs/dentry.c index cb20b964419..5e596583946 100644 --- a/fs/ecryptfs/dentry.c +++ b/fs/ecryptfs/dentry.c @@ -51,13 +51,13 @@ static int ecryptfs_d_revalidate(struct dentry *dentry, struct nameidata *nd) if (!lower_dentry->d_op || !lower_dentry->d_op->d_revalidate) goto out; - dentry_save = nd->dentry; - vfsmount_save = nd->mnt; - nd->dentry = lower_dentry; - nd->mnt = lower_mnt; + dentry_save = nd->path.dentry; + vfsmount_save = nd->path.mnt; + nd->path.dentry = lower_dentry; + nd->path.mnt = lower_mnt; rc = lower_dentry->d_op->d_revalidate(lower_dentry, nd); - nd->dentry = dentry_save; - nd->mnt = vfsmount_save; + nd->path.dentry = dentry_save; + nd->path.mnt = vfsmount_save; if (dentry->d_inode) { struct inode *lower_inode = ecryptfs_inode_to_lower(dentry->d_inode); @@ -80,8 +80,8 @@ static void ecryptfs_d_release(struct dentry *dentry) { if (ecryptfs_dentry_to_private(dentry)) { if (ecryptfs_dentry_to_lower(dentry)) { - mntput(ecryptfs_dentry_to_lower_mnt(dentry)); dput(ecryptfs_dentry_to_lower(dentry)); + mntput(ecryptfs_dentry_to_lower_mnt(dentry)); } kmem_cache_free(ecryptfs_dentry_info_cache, ecryptfs_dentry_to_private(dentry)); diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index ce7a5d4aec3..5007f788da0 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -234,10 +234,11 @@ struct ecryptfs_crypt_stat { #define ECRYPTFS_KEY_VALID 0x00000080 #define ECRYPTFS_METADATA_IN_XATTR 0x00000100 #define ECRYPTFS_VIEW_AS_ENCRYPTED 0x00000200 +#define ECRYPTFS_KEY_SET 0x00000400 u32 flags; unsigned int file_version; size_t iv_bytes; - size_t num_header_extents_at_front; + size_t num_header_bytes_at_front; size_t extent_size; /* Data extent size; default is 4096 */ size_t key_size; size_t extent_shift; @@ -322,7 +323,6 @@ struct ecryptfs_key_tfm { unsigned char cipher_name[ECRYPTFS_MAX_CIPHER_NAME_SIZE + 1]; }; -extern struct list_head key_tfm_list; extern struct mutex key_tfm_list_mutex; /** @@ -521,11 +521,9 @@ extern struct kmem_cache *ecryptfs_file_info_cache; extern struct kmem_cache *ecryptfs_dentry_info_cache; extern struct kmem_cache *ecryptfs_inode_info_cache; extern struct kmem_cache *ecryptfs_sb_info_cache; -extern struct kmem_cache *ecryptfs_header_cache_0; extern struct kmem_cache *ecryptfs_header_cache_1; extern struct kmem_cache *ecryptfs_header_cache_2; extern struct kmem_cache *ecryptfs_xattr_cache; -extern struct kmem_cache *ecryptfs_lower_page_cache; extern struct kmem_cache *ecryptfs_key_record_cache; extern struct kmem_cache *ecryptfs_key_sig_cache; extern struct kmem_cache *ecryptfs_global_auth_tok_cache; @@ -562,8 +560,8 @@ int ecryptfs_read_and_validate_header_region(char *data, struct inode *ecryptfs_inode); int ecryptfs_read_and_validate_xattr_region(char *page_virt, struct dentry *ecryptfs_dentry); -u16 ecryptfs_code_for_cipher_string(struct ecryptfs_crypt_stat *crypt_stat); -int ecryptfs_cipher_code_to_string(char *str, u16 cipher_code); +u8 ecryptfs_code_for_cipher_string(struct ecryptfs_crypt_stat *crypt_stat); +int ecryptfs_cipher_code_to_string(char *str, u8 cipher_code); void ecryptfs_set_default_sizes(struct ecryptfs_crypt_stat *crypt_stat); int ecryptfs_generate_key_packet_set(char *dest_base, struct ecryptfs_crypt_stat *crypt_stat, @@ -576,8 +574,6 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length); int ecryptfs_inode_test(struct inode *inode, void *candidate_lower_inode); int ecryptfs_inode_set(struct inode *inode, void *lower_inode); void ecryptfs_init_inode(struct inode *inode, struct inode *lower_inode); -ssize_t ecryptfs_getxattr(struct dentry *dentry, const char *name, void *value, - size_t size); ssize_t ecryptfs_getxattr_lower(struct dentry *lower_dentry, const char *name, void *value, size_t size); @@ -623,6 +619,7 @@ ecryptfs_add_new_key_tfm(struct ecryptfs_key_tfm **key_tfm, char *cipher_name, size_t key_size); int ecryptfs_init_crypto(void); int ecryptfs_destroy_crypto(void); +int ecryptfs_tfm_exists(char *cipher_name, struct ecryptfs_key_tfm **key_tfm); int ecryptfs_get_tfm_and_mutex_for_cipher_name(struct crypto_blkcipher **tfm, struct mutex **tfm_mutex, char *cipher_name); @@ -631,8 +628,6 @@ int ecryptfs_keyring_auth_tok_for_sig(struct key **auth_tok_key, char *sig); int ecryptfs_write_zeros(struct file *file, pgoff_t index, int start, int num_zeros); -void ecryptfs_lower_offset_for_extent(loff_t *offset, loff_t extent_num, - struct ecryptfs_crypt_stat *crypt_stat); int ecryptfs_write_lower(struct inode *ecryptfs_inode, char *data, loff_t offset, size_t size); int ecryptfs_write_lower_page_segment(struct inode *ecryptfs_inode, @@ -646,8 +641,6 @@ int ecryptfs_read_lower_page_segment(struct page *page_for_ecryptfs, pgoff_t page_index, size_t offset_in_page, size_t size, struct inode *ecryptfs_inode); -int ecryptfs_read(char *data, loff_t offset, size_t size, - struct file *ecryptfs_file); struct page *ecryptfs_get_locked_page(struct file *file, loff_t index); #endif /* #ifndef ECRYPTFS_KERNEL_H */ diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index c98c4690a77..2b8f5ed4ade 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -209,9 +209,10 @@ static int ecryptfs_open(struct inode *inode, struct file *file) if (!(mount_crypt_stat->flags & ECRYPTFS_PLAINTEXT_PASSTHROUGH_ENABLED)) { rc = -EIO; - printk(KERN_WARNING "Attempt to read file that " + printk(KERN_WARNING "Either the lower file " "is not in a valid eCryptfs format, " - "and plaintext passthrough mode is not " + "or the key could not be retrieved. " + "Plaintext passthrough mode is not " "enabled; returning -EIO\n"); mutex_unlock(&crypt_stat->cs_mutex); goto out_free; diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 5a719180983..e2386115210 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -77,13 +77,13 @@ ecryptfs_create_underlying_file(struct inode *lower_dir_inode, struct vfsmount *vfsmount_save; int rc; - dentry_save = nd->dentry; - vfsmount_save = nd->mnt; - nd->dentry = lower_dentry; - nd->mnt = lower_mnt; + dentry_save = nd->path.dentry; + vfsmount_save = nd->path.mnt; + nd->path.dentry = lower_dentry; + nd->path.mnt = lower_mnt; rc = vfs_create(lower_dir_inode, lower_dentry, mode, nd); - nd->dentry = dentry_save; - nd->mnt = vfsmount_save; + nd->path.dentry = dentry_save; + nd->path.mnt = vfsmount_save; return rc; } @@ -365,8 +365,7 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, dentry->d_sb)->mount_crypt_stat; if (mount_crypt_stat->flags & ECRYPTFS_ENCRYPTED_VIEW_ENABLED) { if (crypt_stat->flags & ECRYPTFS_METADATA_IN_XATTR) - file_size = ((crypt_stat->extent_size - * crypt_stat->num_header_extents_at_front) + file_size = (crypt_stat->num_header_bytes_at_front + i_size_read(lower_dentry->d_inode)); else file_size = i_size_read(lower_dentry->d_inode); @@ -685,7 +684,7 @@ ecryptfs_put_link(struct dentry *dentry, struct nameidata *nd, void *ptr) * @crypt_stat: Crypt_stat associated with file * @upper_size: Size of the upper file * - * Calculate the requried size of the lower file based on the + * Calculate the required size of the lower file based on the * specified size of the upper file. This calculation is based on the * number of headers in the underlying file and the extent size. * @@ -697,8 +696,7 @@ upper_size_to_lower_size(struct ecryptfs_crypt_stat *crypt_stat, { loff_t lower_size; - lower_size = (crypt_stat->extent_size - * crypt_stat->num_header_extents_at_front); + lower_size = crypt_stat->num_header_bytes_at_front; if (upper_size != 0) { loff_t num_extents; @@ -821,14 +819,14 @@ ecryptfs_permission(struct inode *inode, int mask, struct nameidata *nd) int rc; if (nd) { - struct vfsmount *vfsmnt_save = nd->mnt; - struct dentry *dentry_save = nd->dentry; + struct vfsmount *vfsmnt_save = nd->path.mnt; + struct dentry *dentry_save = nd->path.dentry; - nd->mnt = ecryptfs_dentry_to_lower_mnt(nd->dentry); - nd->dentry = ecryptfs_dentry_to_lower(nd->dentry); + nd->path.mnt = ecryptfs_dentry_to_lower_mnt(nd->path.dentry); + nd->path.dentry = ecryptfs_dentry_to_lower(nd->path.dentry); rc = permission(ecryptfs_inode_to_lower(inode), mask, nd); - nd->mnt = vfsmnt_save; - nd->dentry = dentry_save; + nd->path.mnt = vfsmnt_save; + nd->path.dentry = dentry_save; } else rc = permission(ecryptfs_inode_to_lower(inode), mask, NULL); return rc; @@ -875,11 +873,11 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia) if (!(mount_crypt_stat->flags & ECRYPTFS_PLAINTEXT_PASSTHROUGH_ENABLED)) { rc = -EIO; - printk(KERN_WARNING "Attempt to read file that " + printk(KERN_WARNING "Either the lower file " "is not in a valid eCryptfs format, " - "and plaintext passthrough mode is not " + "or the key could not be retrieved. " + "Plaintext passthrough mode is not " "enabled; returning -EIO\n"); - mutex_unlock(&crypt_stat->cs_mutex); goto out; } @@ -954,7 +952,7 @@ out: return rc; } -ssize_t +static ssize_t ecryptfs_getxattr(struct dentry *dentry, const char *name, void *value, size_t size) { diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index f458c1f3556..682b1b2482c 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -189,7 +189,7 @@ out: } static int -parse_tag_65_packet(struct ecryptfs_session_key *session_key, u16 *cipher_code, +parse_tag_65_packet(struct ecryptfs_session_key *session_key, u8 *cipher_code, struct ecryptfs_message *msg) { size_t i = 0; @@ -275,7 +275,7 @@ out: static int -write_tag_66_packet(char *signature, size_t cipher_code, +write_tag_66_packet(char *signature, u8 cipher_code, struct ecryptfs_crypt_stat *crypt_stat, char **packet, size_t *packet_len) { @@ -428,7 +428,7 @@ static int decrypt_pki_encrypted_session_key(struct ecryptfs_auth_tok *auth_tok, struct ecryptfs_crypt_stat *crypt_stat) { - u16 cipher_code = 0; + u8 cipher_code = 0; struct ecryptfs_msg_ctx *msg_ctx; struct ecryptfs_message *msg = NULL; char *auth_tok_sig; @@ -1537,7 +1537,7 @@ write_tag_3_packet(char *dest, size_t *remaining_bytes, struct scatterlist dst_sg; struct scatterlist src_sg; struct mutex *tfm_mutex = NULL; - size_t cipher_code; + u8 cipher_code; size_t packet_size_length; size_t max_packet_size; struct ecryptfs_mount_crypt_stat *mount_crypt_stat = diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index e5580bcb923..d25ac9500a9 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -117,7 +117,7 @@ void __ecryptfs_printk(const char *fmt, ...) * * Returns zero on success; non-zero otherwise */ -int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry) +static int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry) { struct ecryptfs_inode_info *inode_info = ecryptfs_inode_to_private(ecryptfs_dentry->d_inode); @@ -226,17 +226,15 @@ out: return rc; } -enum { ecryptfs_opt_sig, ecryptfs_opt_ecryptfs_sig, ecryptfs_opt_debug, - ecryptfs_opt_ecryptfs_debug, ecryptfs_opt_cipher, - ecryptfs_opt_ecryptfs_cipher, ecryptfs_opt_ecryptfs_key_bytes, +enum { ecryptfs_opt_sig, ecryptfs_opt_ecryptfs_sig, + ecryptfs_opt_cipher, ecryptfs_opt_ecryptfs_cipher, + ecryptfs_opt_ecryptfs_key_bytes, ecryptfs_opt_passthrough, ecryptfs_opt_xattr_metadata, ecryptfs_opt_encrypted_view, ecryptfs_opt_err }; static match_table_t tokens = { {ecryptfs_opt_sig, "sig=%s"}, {ecryptfs_opt_ecryptfs_sig, "ecryptfs_sig=%s"}, - {ecryptfs_opt_debug, "debug=%u"}, - {ecryptfs_opt_ecryptfs_debug, "ecryptfs_debug=%u"}, {ecryptfs_opt_cipher, "cipher=%s"}, {ecryptfs_opt_ecryptfs_cipher, "ecryptfs_cipher=%s"}, {ecryptfs_opt_ecryptfs_key_bytes, "ecryptfs_key_bytes=%u"}, @@ -313,7 +311,6 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) substring_t args[MAX_OPT_ARGS]; int token; char *sig_src; - char *debug_src; char *cipher_name_dst; char *cipher_name_src; char *cipher_key_bytes_src; @@ -341,16 +338,6 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) } sig_set = 1; break; - case ecryptfs_opt_debug: - case ecryptfs_opt_ecryptfs_debug: - debug_src = args[0].from; - ecryptfs_verbosity = - (int)simple_strtol(debug_src, &debug_src, - 0); - ecryptfs_printk(KERN_DEBUG, - "Verbosity set to [%d]" "\n", - ecryptfs_verbosity); - break; case ecryptfs_opt_cipher: case ecryptfs_opt_ecryptfs_cipher: cipher_name_src = args[0].from; @@ -423,9 +410,13 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) if (!cipher_key_bytes_set) { mount_crypt_stat->global_default_cipher_key_size = 0; } - rc = ecryptfs_add_new_key_tfm( - NULL, mount_crypt_stat->global_default_cipher_name, - mount_crypt_stat->global_default_cipher_key_size); + mutex_lock(&key_tfm_list_mutex); + if (!ecryptfs_tfm_exists(mount_crypt_stat->global_default_cipher_name, + NULL)) + rc = ecryptfs_add_new_key_tfm( + NULL, mount_crypt_stat->global_default_cipher_name, + mount_crypt_stat->global_default_cipher_key_size); + mutex_unlock(&key_tfm_list_mutex); if (rc) { printk(KERN_ERR "Error attempting to initialize cipher with " "name = [%s] and key size = [%td]; rc = [%d]\n", @@ -522,8 +513,8 @@ static int ecryptfs_read_super(struct super_block *sb, const char *dev_name) ecryptfs_printk(KERN_WARNING, "path_lookup() failed\n"); goto out; } - lower_root = nd.dentry; - lower_mnt = nd.mnt; + lower_root = nd.path.dentry; + lower_mnt = nd.path.mnt; ecryptfs_set_superblock_lower(sb, lower_root->d_sb); sb->s_maxbytes = lower_root->d_sb->s_maxbytes; sb->s_blocksize = lower_root->d_sb->s_blocksize; @@ -535,7 +526,7 @@ static int ecryptfs_read_super(struct super_block *sb, const char *dev_name) rc = 0; goto out; out_free: - path_release(&nd); + path_put(&nd.path); out: return rc; } @@ -654,11 +645,6 @@ static struct ecryptfs_cache_info { .size = sizeof(struct ecryptfs_sb_info), }, { - .cache = &ecryptfs_header_cache_0, - .name = "ecryptfs_headers_0", - .size = PAGE_CACHE_SIZE, - }, - { .cache = &ecryptfs_header_cache_1, .name = "ecryptfs_headers_1", .size = PAGE_CACHE_SIZE, @@ -734,127 +720,40 @@ static int ecryptfs_init_kmem_caches(void) return 0; } -struct ecryptfs_obj { - char *name; - struct list_head slot_list; - struct kobject kobj; -}; - -struct ecryptfs_attribute { - struct attribute attr; - ssize_t(*show) (struct ecryptfs_obj *, char *); - ssize_t(*store) (struct ecryptfs_obj *, const char *, size_t); -}; - -static ssize_t -ecryptfs_attr_store(struct kobject *kobj, - struct attribute *attr, const char *buf, size_t len) -{ - struct ecryptfs_obj *obj = container_of(kobj, struct ecryptfs_obj, - kobj); - struct ecryptfs_attribute *attribute = - container_of(attr, struct ecryptfs_attribute, attr); - - return (attribute->store ? attribute->store(obj, buf, len) : 0); -} +static struct kobject *ecryptfs_kobj; -static ssize_t -ecryptfs_attr_show(struct kobject *kobj, struct attribute *attr, char *buf) +static ssize_t version_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buff) { - struct ecryptfs_obj *obj = container_of(kobj, struct ecryptfs_obj, - kobj); - struct ecryptfs_attribute *attribute = - container_of(attr, struct ecryptfs_attribute, attr); - - return (attribute->show ? attribute->show(obj, buf) : 0); + return snprintf(buff, PAGE_SIZE, "%d\n", ECRYPTFS_VERSIONING_MASK); } -static struct sysfs_ops ecryptfs_sysfs_ops = { - .show = ecryptfs_attr_show, - .store = ecryptfs_attr_store -}; +static struct kobj_attribute version_attr = __ATTR_RO(version); -static struct kobj_type ecryptfs_ktype = { - .sysfs_ops = &ecryptfs_sysfs_ops +static struct attribute *attributes[] = { + &version_attr.attr, + NULL, }; -static decl_subsys(ecryptfs, &ecryptfs_ktype, NULL); - -static ssize_t version_show(struct ecryptfs_obj *obj, char *buff) -{ - return snprintf(buff, PAGE_SIZE, "%d\n", ECRYPTFS_VERSIONING_MASK); -} - -static struct ecryptfs_attribute sysfs_attr_version = __ATTR_RO(version); - -static struct ecryptfs_version_str_map_elem { - u32 flag; - char *str; -} ecryptfs_version_str_map[] = { - {ECRYPTFS_VERSIONING_PASSPHRASE, "passphrase"}, - {ECRYPTFS_VERSIONING_PUBKEY, "pubkey"}, - {ECRYPTFS_VERSIONING_PLAINTEXT_PASSTHROUGH, "plaintext passthrough"}, - {ECRYPTFS_VERSIONING_POLICY, "policy"}, - {ECRYPTFS_VERSIONING_XATTR, "metadata in extended attribute"}, - {ECRYPTFS_VERSIONING_MULTKEY, "multiple keys per file"} +static struct attribute_group attr_group = { + .attrs = attributes, }; -static ssize_t version_str_show(struct ecryptfs_obj *obj, char *buff) -{ - int i; - int remaining = PAGE_SIZE; - int total_written = 0; - - buff[0] = '\0'; - for (i = 0; i < ARRAY_SIZE(ecryptfs_version_str_map); i++) { - int entry_size; - - if (!(ECRYPTFS_VERSIONING_MASK - & ecryptfs_version_str_map[i].flag)) - continue; - entry_size = strlen(ecryptfs_version_str_map[i].str); - if ((entry_size + 2) > remaining) - goto out; - memcpy(buff, ecryptfs_version_str_map[i].str, entry_size); - buff[entry_size++] = '\n'; - buff[entry_size] = '\0'; - buff += entry_size; - total_written += entry_size; - remaining -= entry_size; - } -out: - return total_written; -} - -static struct ecryptfs_attribute sysfs_attr_version_str = __ATTR_RO(version_str); - static int do_sysfs_registration(void) { int rc; - rc = subsystem_register(&ecryptfs_subsys); - if (rc) { - printk(KERN_ERR - "Unable to register ecryptfs sysfs subsystem\n"); - goto out; - } - rc = sysfs_create_file(&ecryptfs_subsys.kobj, - &sysfs_attr_version.attr); - if (rc) { - printk(KERN_ERR - "Unable to create ecryptfs version attribute\n"); - subsystem_unregister(&ecryptfs_subsys); + ecryptfs_kobj = kobject_create_and_add("ecryptfs", fs_kobj); + if (!ecryptfs_kobj) { + printk(KERN_ERR "Unable to create ecryptfs kset\n"); + rc = -ENOMEM; goto out; } - rc = sysfs_create_file(&ecryptfs_subsys.kobj, - &sysfs_attr_version_str.attr); + rc = sysfs_create_group(ecryptfs_kobj, &attr_group); if (rc) { printk(KERN_ERR - "Unable to create ecryptfs version_str attribute\n"); - sysfs_remove_file(&ecryptfs_subsys.kobj, - &sysfs_attr_version.attr); - subsystem_unregister(&ecryptfs_subsys); - goto out; + "Unable to create ecryptfs version attributes\n"); + kobject_put(ecryptfs_kobj); } out: return rc; @@ -862,11 +761,8 @@ out: static void do_sysfs_unregistration(void) { - sysfs_remove_file(&ecryptfs_subsys.kobj, - &sysfs_attr_version.attr); - sysfs_remove_file(&ecryptfs_subsys.kobj, - &sysfs_attr_version_str.attr); - subsystem_unregister(&ecryptfs_subsys); + sysfs_remove_group(ecryptfs_kobj, &attr_group); + kobject_put(ecryptfs_kobj); } static int __init ecryptfs_init(void) @@ -894,7 +790,6 @@ static int __init ecryptfs_init(void) printk(KERN_ERR "Failed to register filesystem\n"); goto out_free_kmem_caches; } - kobj_set_kset_s(&ecryptfs_subsys, fs_subsys); rc = do_sysfs_registration(); if (rc) { printk(KERN_ERR "sysfs registration failed\n"); @@ -912,6 +807,10 @@ static int __init ecryptfs_init(void) "rc = [%d]\n", rc); goto out_release_messaging; } + if (ecryptfs_verbosity > 0) + printk(KERN_CRIT "eCryptfs verbosity set to %d. Secret values " + "will be written to the syslog!\n", ecryptfs_verbosity); + goto out; out_release_messaging: ecryptfs_release_messaging(ecryptfs_transport); diff --git a/fs/ecryptfs/mmap.c b/fs/ecryptfs/mmap.c index 32c5711d79a..6df1debdccc 100644 --- a/fs/ecryptfs/mmap.c +++ b/fs/ecryptfs/mmap.c @@ -34,8 +34,6 @@ #include <linux/scatterlist.h> #include "ecryptfs_kernel.h" -struct kmem_cache *ecryptfs_lower_page_cache; - /** * ecryptfs_get_locked_page * @@ -102,13 +100,14 @@ static void set_header_info(char *page_virt, struct ecryptfs_crypt_stat *crypt_stat) { size_t written; - int save_num_header_extents_at_front = - crypt_stat->num_header_extents_at_front; + size_t save_num_header_bytes_at_front = + crypt_stat->num_header_bytes_at_front; - crypt_stat->num_header_extents_at_front = 1; + crypt_stat->num_header_bytes_at_front = + ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE; ecryptfs_write_header_metadata(page_virt + 20, crypt_stat, &written); - crypt_stat->num_header_extents_at_front = - save_num_header_extents_at_front; + crypt_stat->num_header_bytes_at_front = + save_num_header_bytes_at_front; } /** @@ -134,8 +133,11 @@ ecryptfs_copy_up_encrypted_with_header(struct page *page, loff_t view_extent_num = ((((loff_t)page->index) * num_extents_per_page) + extent_num_in_page); + size_t num_header_extents_at_front = + (crypt_stat->num_header_bytes_at_front + / crypt_stat->extent_size); - if (view_extent_num < crypt_stat->num_header_extents_at_front) { + if (view_extent_num < num_header_extents_at_front) { /* This is a header extent */ char *page_virt; @@ -157,9 +159,8 @@ ecryptfs_copy_up_encrypted_with_header(struct page *page, } else { /* This is an encrypted data extent */ loff_t lower_offset = - ((view_extent_num - - crypt_stat->num_header_extents_at_front) - * crypt_stat->extent_size); + ((view_extent_num * crypt_stat->extent_size) + - crypt_stat->num_header_bytes_at_front); rc = ecryptfs_read_lower_page_segment( page, (lower_offset >> PAGE_CACHE_SHIFT), @@ -257,58 +258,107 @@ static int fill_zeros_to_end_of_page(struct page *page, unsigned int to) end_byte_in_page = i_size_read(inode) % PAGE_CACHE_SIZE; if (to > end_byte_in_page) end_byte_in_page = to; - zero_user_page(page, end_byte_in_page, - PAGE_CACHE_SIZE - end_byte_in_page, KM_USER0); + zero_user_segment(page, end_byte_in_page, PAGE_CACHE_SIZE); out: return 0; } -/* This function must zero any hole we create */ +/** + * ecryptfs_prepare_write + * @file: The eCryptfs file + * @page: The eCryptfs page + * @from: The start byte from which we will write + * @to: The end byte to which we will write + * + * This function must zero any hole we create + * + * Returns zero on success; non-zero otherwise + */ static int ecryptfs_prepare_write(struct file *file, struct page *page, unsigned from, unsigned to) { - int rc = 0; loff_t prev_page_end_size; + int rc = 0; if (!PageUptodate(page)) { - rc = ecryptfs_read_lower_page_segment(page, page->index, 0, - PAGE_CACHE_SIZE, - page->mapping->host); - if (rc) { - printk(KERN_ERR "%s: Error attemping to read lower " - "page segment; rc = [%d]\n", __FUNCTION__, rc); - ClearPageUptodate(page); - goto out; - } else + struct ecryptfs_crypt_stat *crypt_stat = + &ecryptfs_inode_to_private( + file->f_path.dentry->d_inode)->crypt_stat; + + if (!(crypt_stat->flags & ECRYPTFS_ENCRYPTED) + || (crypt_stat->flags & ECRYPTFS_NEW_FILE)) { + rc = ecryptfs_read_lower_page_segment( + page, page->index, 0, PAGE_CACHE_SIZE, + page->mapping->host); + if (rc) { + printk(KERN_ERR "%s: Error attemping to read " + "lower page segment; rc = [%d]\n", + __FUNCTION__, rc); + ClearPageUptodate(page); + goto out; + } else + SetPageUptodate(page); + } else if (crypt_stat->flags & ECRYPTFS_VIEW_AS_ENCRYPTED) { + if (crypt_stat->flags & ECRYPTFS_METADATA_IN_XATTR) { + rc = ecryptfs_copy_up_encrypted_with_header( + page, crypt_stat); + if (rc) { + printk(KERN_ERR "%s: Error attempting " + "to copy the encrypted content " + "from the lower file whilst " + "inserting the metadata from " + "the xattr into the header; rc " + "= [%d]\n", __FUNCTION__, rc); + ClearPageUptodate(page); + goto out; + } + SetPageUptodate(page); + } else { + rc = ecryptfs_read_lower_page_segment( + page, page->index, 0, PAGE_CACHE_SIZE, + page->mapping->host); + if (rc) { + printk(KERN_ERR "%s: Error reading " + "page; rc = [%d]\n", + __FUNCTION__, rc); + ClearPageUptodate(page); + goto out; + } + SetPageUptodate(page); + } + } else { + rc = ecryptfs_decrypt_page(page); + if (rc) { + printk(KERN_ERR "%s: Error decrypting page " + "at index [%ld]; rc = [%d]\n", + __FUNCTION__, page->index, rc); + ClearPageUptodate(page); + goto out; + } SetPageUptodate(page); + } } - prev_page_end_size = ((loff_t)page->index << PAGE_CACHE_SHIFT); - - /* - * If creating a page or more of holes, zero them out via truncate. - * Note, this will increase i_size. - */ + /* If creating a page or more of holes, zero them out via truncate. + * Note, this will increase i_size. */ if (page->index != 0) { if (prev_page_end_size > i_size_read(page->mapping->host)) { rc = ecryptfs_truncate(file->f_path.dentry, prev_page_end_size); if (rc) { - printk(KERN_ERR "Error on attempt to " + printk(KERN_ERR "%s: Error on attempt to " "truncate to (higher) offset [%lld];" - " rc = [%d]\n", prev_page_end_size, rc); + " rc = [%d]\n", __FUNCTION__, + prev_page_end_size, rc); goto out; } } } - /* - * Writing to a new page, and creating a small hole from start of page? - * Zero it out. - */ - if ((i_size_read(page->mapping->host) == prev_page_end_size) && - (from != 0)) { - zero_user_page(page, 0, PAGE_CACHE_SIZE, KM_USER0); - } + /* Writing to a new page, and creating a small hole from start + * of page? Zero it out. */ + if ((i_size_read(page->mapping->host) == prev_page_end_size) + && (from != 0)) + zero_user(page, 0, PAGE_CACHE_SIZE); out: return rc; } diff --git a/fs/ecryptfs/netlink.c b/fs/ecryptfs/netlink.c index 9aa345121e0..f638a698dc5 100644 --- a/fs/ecryptfs/netlink.c +++ b/fs/ecryptfs/netlink.c @@ -237,7 +237,6 @@ out: */ void ecryptfs_release_netlink(void) { - if (ecryptfs_nl_sock && ecryptfs_nl_sock->sk_socket) - sock_release(ecryptfs_nl_sock->sk_socket); + netlink_kernel_release(ecryptfs_nl_sock); ecryptfs_nl_sock = NULL; } diff --git a/fs/ecryptfs/read_write.c b/fs/ecryptfs/read_write.c index 948f57624c0..0c4928623bb 100644 --- a/fs/ecryptfs/read_write.c +++ b/fs/ecryptfs/read_write.c @@ -293,6 +293,7 @@ int ecryptfs_read_lower_page_segment(struct page *page_for_ecryptfs, return rc; } +#if 0 /** * ecryptfs_read * @data: The virtual address into which to write the data read (and @@ -371,3 +372,4 @@ int ecryptfs_read(char *data, loff_t offset, size_t size, out: return rc; } +#endif /* 0 */ diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c index 4859c4eecd6..c27ac2b358a 100644 --- a/fs/ecryptfs/super.c +++ b/fs/ecryptfs/super.c @@ -156,32 +156,38 @@ static void ecryptfs_clear_inode(struct inode *inode) /** * ecryptfs_show_options * - * Prints the directory we are currently mounted over. - * Returns zero on success; non-zero otherwise + * Prints the mount options for a given superblock. + * Returns zero; does not fail. */ static int ecryptfs_show_options(struct seq_file *m, struct vfsmount *mnt) { struct super_block *sb = mnt->mnt_sb; - struct dentry *lower_root_dentry = ecryptfs_dentry_to_lower(sb->s_root); - struct vfsmount *lower_mnt = ecryptfs_dentry_to_lower_mnt(sb->s_root); - char *tmp_page; - char *path; - int rc = 0; - - tmp_page = (char *)__get_free_page(GFP_KERNEL); - if (!tmp_page) { - rc = -ENOMEM; - goto out; - } - path = d_path(lower_root_dentry, lower_mnt, tmp_page, PAGE_SIZE); - if (IS_ERR(path)) { - rc = PTR_ERR(path); - goto out; + struct ecryptfs_mount_crypt_stat *mount_crypt_stat = + &ecryptfs_superblock_to_private(sb)->mount_crypt_stat; + struct ecryptfs_global_auth_tok *walker; + + mutex_lock(&mount_crypt_stat->global_auth_tok_list_mutex); + list_for_each_entry(walker, + &mount_crypt_stat->global_auth_tok_list, + mount_crypt_stat_list) { + seq_printf(m, ",ecryptfs_sig=%s", walker->sig); } - seq_printf(m, ",dir=%s", path); - free_page((unsigned long)tmp_page); -out: - return rc; + mutex_unlock(&mount_crypt_stat->global_auth_tok_list_mutex); + + seq_printf(m, ",ecryptfs_cipher=%s", + mount_crypt_stat->global_default_cipher_name); + + if (mount_crypt_stat->global_default_cipher_key_size) + seq_printf(m, ",ecryptfs_key_bytes=%zd", + mount_crypt_stat->global_default_cipher_key_size); + if (mount_crypt_stat->flags & ECRYPTFS_PLAINTEXT_PASSTHROUGH_ENABLED) + seq_printf(m, ",ecryptfs_passthrough"); + if (mount_crypt_stat->flags & ECRYPTFS_XATTR_METADATA_ENABLED) + seq_printf(m, ",ecryptfs_xattr_metadata"); + if (mount_crypt_stat->flags & ECRYPTFS_ENCRYPTED_VIEW_ENABLED) + seq_printf(m, ",ecryptfs_encrypted_view"); + + return 0; } const struct super_operations ecryptfs_sops = { diff --git a/fs/efs/dir.c b/fs/efs/dir.c index dfb5cb40021..49308a29798 100644 --- a/fs/efs/dir.c +++ b/fs/efs/dir.c @@ -5,8 +5,8 @@ */ #include <linux/buffer_head.h> -#include <linux/efs_fs.h> #include <linux/smp_lock.h> +#include "efs.h" static int efs_readdir(struct file *, void *, filldir_t); diff --git a/fs/efs/efs.h b/fs/efs/efs.h new file mode 100644 index 00000000000..d8305b582ab --- /dev/null +++ b/fs/efs/efs.h @@ -0,0 +1,140 @@ +/* + * Copyright (c) 1999 Al Smith + * + * Portions derived from work (c) 1995,1996 Christian Vogelgsang. + * Portions derived from IRIX header files (c) 1988 Silicon Graphics + */ +#ifndef _EFS_EFS_H_ +#define _EFS_EFS_H_ + +#include <linux/fs.h> +#include <asm/uaccess.h> + +#define EFS_VERSION "1.0a" + +static const char cprt[] = "EFS: "EFS_VERSION" - (c) 1999 Al Smith <Al.Smith@aeschi.ch.eu.org>"; + + +/* 1 block is 512 bytes */ +#define EFS_BLOCKSIZE_BITS 9 +#define EFS_BLOCKSIZE (1 << EFS_BLOCKSIZE_BITS) + +typedef int32_t efs_block_t; +typedef uint32_t efs_ino_t; + +#define EFS_DIRECTEXTENTS 12 + +/* + * layout of an extent, in memory and on disk. 8 bytes exactly. + */ +typedef union extent_u { + unsigned char raw[8]; + struct extent_s { + unsigned int ex_magic:8; /* magic # (zero) */ + unsigned int ex_bn:24; /* basic block */ + unsigned int ex_length:8; /* numblocks in this extent */ + unsigned int ex_offset:24; /* logical offset into file */ + } cooked; +} efs_extent; + +typedef struct edevs { + __be16 odev; + __be32 ndev; +} efs_devs; + +/* + * extent based filesystem inode as it appears on disk. The efs inode + * is exactly 128 bytes long. + */ +struct efs_dinode { + __be16 di_mode; /* mode and type of file */ + __be16 di_nlink; /* number of links to file */ + __be16 di_uid; /* owner's user id */ + __be16 di_gid; /* owner's group id */ + __be32 di_size; /* number of bytes in file */ + __be32 di_atime; /* time last accessed */ + __be32 di_mtime; /* time last modified */ + __be32 di_ctime; /* time created */ + __be32 di_gen; /* generation number */ + __be16 di_numextents; /* # of extents */ + u_char di_version; /* version of inode */ + u_char di_spare; /* spare - used by AFS */ + union di_addr { + efs_extent di_extents[EFS_DIRECTEXTENTS]; + efs_devs di_dev; /* device for IFCHR/IFBLK */ + } di_u; +}; + +/* efs inode storage in memory */ +struct efs_inode_info { + int numextents; + int lastextent; + + efs_extent extents[EFS_DIRECTEXTENTS]; + struct inode vfs_inode; +}; + +#include <linux/efs_fs_sb.h> + +#define EFS_DIRBSIZE_BITS EFS_BLOCKSIZE_BITS +#define EFS_DIRBSIZE (1 << EFS_DIRBSIZE_BITS) + +struct efs_dentry { + __be32 inode; + unsigned char namelen; + char name[3]; +}; + +#define EFS_DENTSIZE (sizeof(struct efs_dentry) - 3 + 1) +#define EFS_MAXNAMELEN ((1 << (sizeof(char) * 8)) - 1) + +#define EFS_DIRBLK_HEADERSIZE 4 +#define EFS_DIRBLK_MAGIC 0xbeef /* moo */ + +struct efs_dir { + __be16 magic; + unsigned char firstused; + unsigned char slots; + + unsigned char space[EFS_DIRBSIZE - EFS_DIRBLK_HEADERSIZE]; +}; + +#define EFS_MAXENTS \ + ((EFS_DIRBSIZE - EFS_DIRBLK_HEADERSIZE) / \ + (EFS_DENTSIZE + sizeof(char))) + +#define EFS_SLOTAT(dir, slot) EFS_REALOFF((dir)->space[slot]) + +#define EFS_REALOFF(offset) ((offset << 1)) + + +static inline struct efs_inode_info *INODE_INFO(struct inode *inode) +{ + return container_of(inode, struct efs_inode_info, vfs_inode); +} + +static inline struct efs_sb_info *SUPER_INFO(struct super_block *sb) +{ + return sb->s_fs_info; +} + +struct statfs; +struct fid; + +extern const struct inode_operations efs_dir_inode_operations; +extern const struct file_operations efs_dir_operations; +extern const struct address_space_operations efs_symlink_aops; + +extern struct inode *efs_iget(struct super_block *, unsigned long); +extern efs_block_t efs_map_block(struct inode *, efs_block_t); +extern int efs_get_block(struct inode *, sector_t, struct buffer_head *, int); + +extern struct dentry *efs_lookup(struct inode *, struct dentry *, struct nameidata *); +extern struct dentry *efs_fh_to_dentry(struct super_block *sb, struct fid *fid, + int fh_len, int fh_type); +extern struct dentry *efs_fh_to_parent(struct super_block *sb, struct fid *fid, + int fh_len, int fh_type); +extern struct dentry *efs_get_parent(struct dentry *); +extern int efs_bmap(struct inode *, int); + +#endif /* _EFS_EFS_H_ */ diff --git a/fs/efs/file.c b/fs/efs/file.c index 5db20129681..1ccb364ffa6 100644 --- a/fs/efs/file.c +++ b/fs/efs/file.c @@ -7,7 +7,7 @@ */ #include <linux/buffer_head.h> -#include <linux/efs_fs.h> +#include "efs.h" int efs_get_block(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) diff --git a/fs/efs/inode.c b/fs/efs/inode.c index 174696f9bf1..a8e7797b947 100644 --- a/fs/efs/inode.c +++ b/fs/efs/inode.c @@ -7,11 +7,11 @@ * and from work (c) 1998 Mike Shaver. */ -#include <linux/efs_fs.h> -#include <linux/efs_fs_sb.h> #include <linux/buffer_head.h> #include <linux/module.h> #include <linux/fs.h> +#include "efs.h" +#include <linux/efs_fs_sb.h> static int efs_readpage(struct file *file, struct page *page) { @@ -45,17 +45,26 @@ static inline void extent_copy(efs_extent *src, efs_extent *dst) { return; } -void efs_read_inode(struct inode *inode) +struct inode *efs_iget(struct super_block *super, unsigned long ino) { int i, inode_index; dev_t device; u32 rdev; struct buffer_head *bh; - struct efs_sb_info *sb = SUPER_INFO(inode->i_sb); - struct efs_inode_info *in = INODE_INFO(inode); + struct efs_sb_info *sb = SUPER_INFO(super); + struct efs_inode_info *in; efs_block_t block, offset; struct efs_dinode *efs_inode; - + struct inode *inode; + + inode = iget_locked(super, ino); + if (IS_ERR(inode)) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; + + in = INODE_INFO(inode); + /* ** EFS layout: ** @@ -131,7 +140,7 @@ void efs_read_inode(struct inode *inode) brelse(bh); #ifdef DEBUG - printk(KERN_DEBUG "EFS: read_inode(): inode %lu, extents %d, mode %o\n", + printk(KERN_DEBUG "EFS: efs_iget(): inode %lu, extents %d, mode %o\n", inode->i_ino, in->numextents, inode->i_mode); #endif @@ -159,13 +168,13 @@ void efs_read_inode(struct inode *inode) break; } - return; + unlock_new_inode(inode); + return inode; read_inode_error: printk(KERN_WARNING "EFS: failed to read inode %lu\n", inode->i_ino); - make_bad_inode(inode); - - return; + iget_failed(inode); + return ERR_PTR(-EIO); } static inline efs_block_t diff --git a/fs/efs/namei.c b/fs/efs/namei.c index f7f407075be..3a404e7fad5 100644 --- a/fs/efs/namei.c +++ b/fs/efs/namei.c @@ -8,9 +8,9 @@ #include <linux/buffer_head.h> #include <linux/string.h> -#include <linux/efs_fs.h> #include <linux/smp_lock.h> #include <linux/exportfs.h> +#include "efs.h" static efs_ino_t efs_find_entry(struct inode *inode, const char *name, int len) { @@ -66,9 +66,10 @@ struct dentry *efs_lookup(struct inode *dir, struct dentry *dentry, struct namei lock_kernel(); inodenum = efs_find_entry(dir, dentry->d_name.name, dentry->d_name.len); if (inodenum) { - if (!(inode = iget(dir->i_sb, inodenum))) { + inode = efs_iget(dir->i_sb, inodenum); + if (IS_ERR(inode)) { unlock_kernel(); - return ERR_PTR(-EACCES); + return ERR_CAST(inode); } } unlock_kernel(); @@ -84,12 +85,11 @@ static struct inode *efs_nfs_get_inode(struct super_block *sb, u64 ino, if (ino == 0) return ERR_PTR(-ESTALE); - inode = iget(sb, ino); - if (inode == NULL) - return ERR_PTR(-ENOMEM); + inode = efs_iget(sb, ino); + if (IS_ERR(inode)) + return ERR_CAST(inode); - if (is_bad_inode(inode) || - (generation && inode->i_generation != generation)) { + if (generation && inode->i_generation != generation) { iput(inode); return ERR_PTR(-ESTALE); } @@ -116,7 +116,7 @@ struct dentry *efs_get_parent(struct dentry *child) struct dentry *parent; struct inode *inode; efs_ino_t ino; - int error; + long error; lock_kernel(); @@ -125,10 +125,11 @@ struct dentry *efs_get_parent(struct dentry *child) if (!ino) goto fail; - error = -EACCES; - inode = iget(child->d_inode->i_sb, ino); - if (!inode) + inode = efs_iget(child->d_inode->i_sb, ino); + if (IS_ERR(inode)) { + error = PTR_ERR(inode); goto fail; + } error = -ENOMEM; parent = d_alloc_anon(inode); diff --git a/fs/efs/super.c b/fs/efs/super.c index c79bc627f10..d733531b55e 100644 --- a/fs/efs/super.c +++ b/fs/efs/super.c @@ -8,14 +8,15 @@ #include <linux/init.h> #include <linux/module.h> -#include <linux/efs_fs.h> -#include <linux/efs_vh.h> -#include <linux/efs_fs_sb.h> #include <linux/exportfs.h> #include <linux/slab.h> #include <linux/buffer_head.h> #include <linux/vfs.h> +#include "efs.h" +#include <linux/efs_vh.h> +#include <linux/efs_fs_sb.h> + static int efs_statfs(struct dentry *dentry, struct kstatfs *buf); static int efs_fill_super(struct super_block *s, void *d, int silent); @@ -107,7 +108,6 @@ static int efs_remount(struct super_block *sb, int *flags, char *data) static const struct super_operations efs_superblock_operations = { .alloc_inode = efs_alloc_inode, .destroy_inode = efs_destroy_inode, - .read_inode = efs_read_inode, .put_super = efs_put_super, .statfs = efs_statfs, .remount_fs = efs_remount, @@ -247,6 +247,7 @@ static int efs_fill_super(struct super_block *s, void *d, int silent) struct efs_sb_info *sb; struct buffer_head *bh; struct inode *root; + int ret = -EINVAL; sb = kzalloc(sizeof(struct efs_sb_info), GFP_KERNEL); if (!sb) @@ -303,12 +304,18 @@ static int efs_fill_super(struct super_block *s, void *d, int silent) } s->s_op = &efs_superblock_operations; s->s_export_op = &efs_export_ops; - root = iget(s, EFS_ROOTINODE); + root = efs_iget(s, EFS_ROOTINODE); + if (IS_ERR(root)) { + printk(KERN_ERR "EFS: get root inode failed\n"); + ret = PTR_ERR(root); + goto out_no_fs; + } + s->s_root = d_alloc_root(root); - if (!(s->s_root)) { - printk(KERN_ERR "EFS: get root inode failed\n"); + printk(KERN_ERR "EFS: get root dentry failed\n"); iput(root); + ret = -ENOMEM; goto out_no_fs; } @@ -318,7 +325,7 @@ out_no_fs_ul: out_no_fs: s->s_fs_info = NULL; kfree(sb); - return -EINVAL; + return ret; } static int efs_statfs(struct dentry *dentry, struct kstatfs *buf) { diff --git a/fs/efs/symlink.c b/fs/efs/symlink.c index 1d30d2ff440..41911ec83aa 100644 --- a/fs/efs/symlink.c +++ b/fs/efs/symlink.c @@ -7,10 +7,10 @@ */ #include <linux/string.h> -#include <linux/efs_fs.h> #include <linux/pagemap.h> #include <linux/buffer_head.h> #include <linux/smp_lock.h> +#include "efs.h" static int efs_symlink_readpage(struct file *file, struct page *page) { diff --git a/fs/eventfd.c b/fs/eventfd.c index 2ce19c000d2..a9f130cd50a 100644 --- a/fs/eventfd.c +++ b/fs/eventfd.c @@ -15,6 +15,7 @@ #include <linux/spinlock.h> #include <linux/anon_inodes.h> #include <linux/eventfd.h> +#include <linux/syscalls.h> struct eventfd_ctx { wait_queue_head_t wqh; diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 34f68f3a069..a415f42d32c 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -353,7 +353,7 @@ static void ep_poll_safewake(struct poll_safewake *psw, wait_queue_head_t *wq) spin_unlock_irqrestore(&psw->lock, flags); /* Do really wake up now */ - wake_up(wq); + wake_up_nested(wq, 1 + wake_nests); /* Remove the current task from the list */ spin_lock_irqsave(&psw->lock, flags); @@ -656,8 +656,7 @@ is_linked: * wait list. */ if (waitqueue_active(&ep->wq)) - __wake_up_locked(&ep->wq, TASK_UNINTERRUPTIBLE | - TASK_INTERRUPTIBLE); + wake_up_locked(&ep->wq); if (waitqueue_active(&ep->poll_wait)) pwake++; @@ -780,7 +779,7 @@ static int ep_insert(struct eventpoll *ep, struct epoll_event *event, /* Notify waiting tasks that events are available */ if (waitqueue_active(&ep->wq)) - __wake_up_locked(&ep->wq, TASK_UNINTERRUPTIBLE | TASK_INTERRUPTIBLE); + wake_up_locked(&ep->wq); if (waitqueue_active(&ep->poll_wait)) pwake++; } @@ -854,8 +853,7 @@ static int ep_modify(struct eventpoll *ep, struct epitem *epi, struct epoll_even /* Notify waiting tasks that events are available */ if (waitqueue_active(&ep->wq)) - __wake_up_locked(&ep->wq, TASK_UNINTERRUPTIBLE | - TASK_INTERRUPTIBLE); + wake_up_locked(&ep->wq); if (waitqueue_active(&ep->poll_wait)) pwake++; } @@ -978,8 +976,7 @@ errxit: * wait list (delayed after we release the lock). */ if (waitqueue_active(&ep->wq)) - __wake_up_locked(&ep->wq, TASK_UNINTERRUPTIBLE | - TASK_INTERRUPTIBLE); + wake_up_locked(&ep->wq); if (waitqueue_active(&ep->poll_wait)) pwake++; } diff --git a/fs/exec.c b/fs/exec.c index 282240afe99..54a0a557b67 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -112,14 +112,14 @@ asmlinkage long sys_uselib(const char __user * library) goto out; error = -EINVAL; - if (!S_ISREG(nd.dentry->d_inode->i_mode)) + if (!S_ISREG(nd.path.dentry->d_inode->i_mode)) goto exit; error = vfs_permission(&nd, MAY_READ | MAY_EXEC); if (error) goto exit; - file = nameidata_to_filp(&nd, O_RDONLY); + file = nameidata_to_filp(&nd, O_RDONLY|O_LARGEFILE); error = PTR_ERR(file); if (IS_ERR(file)) goto out; @@ -148,7 +148,7 @@ out: return error; exit: release_open_intent(&nd); - path_release(&nd); + path_put(&nd.path); goto out; } @@ -173,8 +173,15 @@ static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, return NULL; if (write) { - struct rlimit *rlim = current->signal->rlim; unsigned long size = bprm->vma->vm_end - bprm->vma->vm_start; + struct rlimit *rlim; + + /* + * We've historically supported up to 32 pages (ARG_MAX) + * of argument strings even with small stacks + */ + if (size <= ARG_MAX) + return page; /* * Limit to 1/4-th the stack size for the argv+env strings. @@ -183,6 +190,7 @@ static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, * - the program will have a reasonable amount of stack left * to work from. */ + rlim = current->signal->rlim; if (size > rlim[RLIMIT_STACK].rlim_cur / 4) { put_page(page); return NULL; @@ -652,13 +660,14 @@ struct file *open_exec(const char *name) file = ERR_PTR(err); if (!err) { - struct inode *inode = nd.dentry->d_inode; + struct inode *inode = nd.path.dentry->d_inode; file = ERR_PTR(-EACCES); if (S_ISREG(inode->i_mode)) { int err = vfs_permission(&nd, MAY_EXEC); file = ERR_PTR(err); if (!err) { - file = nameidata_to_filp(&nd, O_RDONLY); + file = nameidata_to_filp(&nd, + O_RDONLY|O_LARGEFILE); if (!IS_ERR(file)) { err = deny_write_access(file); if (err) { @@ -671,7 +680,7 @@ out: } } release_open_intent(&nd); - path_release(&nd); + path_put(&nd.path); } goto out; } @@ -760,7 +769,7 @@ static int de_thread(struct task_struct *tsk) */ read_lock(&tasklist_lock); spin_lock_irq(lock); - if (sig->flags & SIGNAL_GROUP_EXIT) { + if (signal_group_exit(sig)) { /* * Another group action in progress, just * return so that the signal is processed. @@ -778,31 +787,13 @@ static int de_thread(struct task_struct *tsk) if (unlikely(tsk->group_leader == task_child_reaper(tsk))) task_active_pid_ns(tsk)->child_reaper = tsk; + sig->group_exit_task = tsk; zap_other_threads(tsk); read_unlock(&tasklist_lock); - /* - * Account for the thread group leader hanging around: - */ - count = 1; - if (!thread_group_leader(tsk)) { - count = 2; - /* - * The SIGALRM timer survives the exec, but needs to point - * at us as the new group leader now. We have a race with - * a timer firing now getting the old leader, so we need to - * synchronize with any firing (by calling del_timer_sync) - * before we can safely let the old group leader die. - */ - sig->tsk = tsk; - spin_unlock_irq(lock); - if (hrtimer_cancel(&sig->real_timer)) - hrtimer_restart(&sig->real_timer); - spin_lock_irq(lock); - } - + /* Account for the thread group leader hanging around: */ + count = thread_group_leader(tsk) ? 1 : 2; sig->notify_count = count; - sig->group_exit_task = tsk; while (atomic_read(&sig->count) > count) { __set_current_state(TASK_UNINTERRUPTIBLE); spin_unlock_irq(lock); @@ -871,15 +862,10 @@ static int de_thread(struct task_struct *tsk) leader->exit_state = EXIT_DEAD; write_unlock_irq(&tasklist_lock); - } + } sig->group_exit_task = NULL; sig->notify_count = 0; - /* - * There may be one thread left which is just exiting, - * but it's safe to stop telling the group to kill themselves. - */ - sig->flags = 0; no_thread_group: exit_itimers(sig); @@ -947,12 +933,13 @@ static void flush_old_files(struct files_struct * files) spin_unlock(&files->file_lock); } -void get_task_comm(char *buf, struct task_struct *tsk) +char *get_task_comm(char *buf, struct task_struct *tsk) { /* buf must be at least sizeof(tsk->comm) in size */ task_lock(tsk); strncpy(buf, tsk->comm, sizeof(tsk->comm)); task_unlock(tsk); + return buf; } void set_task_comm(struct task_struct *tsk, char *buf) @@ -1188,7 +1175,7 @@ int search_binary_handler(struct linux_binprm *bprm,struct pt_regs *regs) { int try,retval; struct linux_binfmt *fmt; -#ifdef __alpha__ +#if defined(__alpha__) && defined(CONFIG_ARCH_SUPPORTS_AOUT) /* handle /sbin/loader.. */ { struct exec * eh = (struct exec *) bprm->buf; @@ -1548,7 +1535,7 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm, int err = -EAGAIN; spin_lock_irq(&tsk->sighand->siglock); - if (!(tsk->signal->flags & SIGNAL_GROUP_EXIT)) { + if (!signal_group_exit(tsk->signal)) { tsk->signal->group_exit_code = exit_code; zap_process(tsk); err = 0; diff --git a/fs/ext2/balloc.c b/fs/ext2/balloc.c index 377ad172d74..e7b2bafa1dd 100644 --- a/fs/ext2/balloc.c +++ b/fs/ext2/balloc.c @@ -69,9 +69,53 @@ struct ext2_group_desc * ext2_get_group_desc(struct super_block * sb, return desc + offset; } +static int ext2_valid_block_bitmap(struct super_block *sb, + struct ext2_group_desc *desc, + unsigned int block_group, + struct buffer_head *bh) +{ + ext2_grpblk_t offset; + ext2_grpblk_t next_zero_bit; + ext2_fsblk_t bitmap_blk; + ext2_fsblk_t group_first_block; + + group_first_block = ext2_group_first_block_no(sb, block_group); + + /* check whether block bitmap block number is set */ + bitmap_blk = le32_to_cpu(desc->bg_block_bitmap); + offset = bitmap_blk - group_first_block; + if (!ext2_test_bit(offset, bh->b_data)) + /* bad block bitmap */ + goto err_out; + + /* check whether the inode bitmap block number is set */ + bitmap_blk = le32_to_cpu(desc->bg_inode_bitmap); + offset = bitmap_blk - group_first_block; + if (!ext2_test_bit(offset, bh->b_data)) + /* bad block bitmap */ + goto err_out; + + /* check whether the inode table block number is set */ + bitmap_blk = le32_to_cpu(desc->bg_inode_table); + offset = bitmap_blk - group_first_block; + next_zero_bit = ext2_find_next_zero_bit(bh->b_data, + offset + EXT2_SB(sb)->s_itb_per_group, + offset); + if (next_zero_bit >= offset + EXT2_SB(sb)->s_itb_per_group) + /* good bitmap for inode tables */ + return 1; + +err_out: + ext2_error(sb, __FUNCTION__, + "Invalid block bitmap - " + "block_group = %d, block = %lu", + block_group, bitmap_blk); + return 0; +} + /* - * Read the bitmap for a given block_group, reading into the specified - * slot in the superblock's bitmap cache. + * Read the bitmap for a given block_group,and validate the + * bits for block/inode/inode tables are set in the bitmaps * * Return buffer_head on success or NULL in case of failure. */ @@ -80,17 +124,36 @@ read_block_bitmap(struct super_block *sb, unsigned int block_group) { struct ext2_group_desc * desc; struct buffer_head * bh = NULL; - - desc = ext2_get_group_desc (sb, block_group, NULL); + ext2_fsblk_t bitmap_blk; + + desc = ext2_get_group_desc(sb, block_group, NULL); if (!desc) - goto error_out; - bh = sb_bread(sb, le32_to_cpu(desc->bg_block_bitmap)); - if (!bh) - ext2_error (sb, "read_block_bitmap", + return NULL; + bitmap_blk = le32_to_cpu(desc->bg_block_bitmap); + bh = sb_getblk(sb, bitmap_blk); + if (unlikely(!bh)) { + ext2_error(sb, __FUNCTION__, + "Cannot read block bitmap - " + "block_group = %d, block_bitmap = %u", + block_group, le32_to_cpu(desc->bg_block_bitmap)); + return NULL; + } + if (likely(bh_uptodate_or_lock(bh))) + return bh; + + if (bh_submit_read(bh) < 0) { + brelse(bh); + ext2_error(sb, __FUNCTION__, "Cannot read block bitmap - " "block_group = %d, block_bitmap = %u", block_group, le32_to_cpu(desc->bg_block_bitmap)); -error_out: + return NULL; + } + if (!ext2_valid_block_bitmap(sb, desc, block_group, bh)) { + brelse(bh); + return NULL; + } + return bh; } @@ -474,11 +537,13 @@ do_more: in_range (block, le32_to_cpu(desc->bg_inode_table), sbi->s_itb_per_group) || in_range (block + count - 1, le32_to_cpu(desc->bg_inode_table), - sbi->s_itb_per_group)) + sbi->s_itb_per_group)) { ext2_error (sb, "ext2_free_blocks", "Freeing blocks in system zones - " "Block = %lu, count = %lu", block, count); + goto error_return; + } for (i = 0, group_freed = 0; i < count; i++) { if (!ext2_clear_bit_atomic(sb_bgl_lock(sbi, block_group), @@ -1250,8 +1315,8 @@ retry_alloc: smp_rmb(); /* - * Now search the rest of the groups. We assume that - * i and gdp correctly point to the last group visited. + * Now search the rest of the groups. We assume that + * group_no and gdp correctly point to the last group visited. */ for (bgi = 0; bgi < ngroups; bgi++) { group_no++; @@ -1311,11 +1376,13 @@ allocated: in_range(ret_block, le32_to_cpu(gdp->bg_inode_table), EXT2_SB(sb)->s_itb_per_group) || in_range(ret_block + num - 1, le32_to_cpu(gdp->bg_inode_table), - EXT2_SB(sb)->s_itb_per_group)) + EXT2_SB(sb)->s_itb_per_group)) { ext2_error(sb, "ext2_new_blocks", "Allocating block in system zone - " "blocks from "E2FSBLK", length %lu", ret_block, num); + goto out; + } performed_allocation = 1; @@ -1466,9 +1533,6 @@ int ext2_bg_has_super(struct super_block *sb, int group) */ unsigned long ext2_bg_num_gdb(struct super_block *sb, int group) { - if (EXT2_HAS_RO_COMPAT_FEATURE(sb,EXT2_FEATURE_RO_COMPAT_SPARSE_SUPER)&& - !ext2_group_sparse(group)) - return 0; - return EXT2_SB(sb)->s_gdb_count; + return ext2_bg_has_super(sb, group) ? EXT2_SB(sb)->s_gdb_count : 0; } diff --git a/fs/ext2/dir.c b/fs/ext2/dir.c index d868e26c15e..8dededd80fe 100644 --- a/fs/ext2/dir.c +++ b/fs/ext2/dir.c @@ -703,7 +703,7 @@ const struct file_operations ext2_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, .readdir = ext2_readdir, - .ioctl = ext2_ioctl, + .unlocked_ioctl = ext2_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = ext2_compat_ioctl, #endif diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h index c87ae29c19c..47d88da2d33 100644 --- a/fs/ext2/ext2.h +++ b/fs/ext2/ext2.h @@ -124,9 +124,8 @@ extern void ext2_check_inodes_bitmap (struct super_block *); extern unsigned long ext2_count_free (struct buffer_head *, unsigned); /* inode.c */ -extern void ext2_read_inode (struct inode *); +extern struct inode *ext2_iget (struct super_block *, unsigned long); extern int ext2_write_inode (struct inode *, int); -extern void ext2_put_inode (struct inode *); extern void ext2_delete_inode (struct inode *); extern int ext2_sync_inode (struct inode *); extern int ext2_get_block(struct inode *, sector_t, struct buffer_head *, int); @@ -139,8 +138,7 @@ int __ext2_write_begin(struct file *file, struct address_space *mapping, struct page **pagep, void **fsdata); /* ioctl.c */ -extern int ext2_ioctl (struct inode *, struct file *, unsigned int, - unsigned long); +extern long ext2_ioctl(struct file *, unsigned int, unsigned long); extern long ext2_compat_ioctl(struct file *, unsigned int, unsigned long); /* namei.c */ diff --git a/fs/ext2/file.c b/fs/ext2/file.c index c051798459a..5f2fa9c3629 100644 --- a/fs/ext2/file.c +++ b/fs/ext2/file.c @@ -48,7 +48,7 @@ const struct file_operations ext2_file_operations = { .write = do_sync_write, .aio_read = generic_file_aio_read, .aio_write = generic_file_aio_write, - .ioctl = ext2_ioctl, + .unlocked_ioctl = ext2_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = ext2_compat_ioctl, #endif @@ -65,7 +65,7 @@ const struct file_operations ext2_xip_file_operations = { .llseek = generic_file_llseek, .read = xip_file_read, .write = xip_file_write, - .ioctl = ext2_ioctl, + .unlocked_ioctl = ext2_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = ext2_compat_ioctl, #endif diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index b1ab32ab5a7..c6200680542 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -286,15 +286,12 @@ static unsigned long ext2_find_near(struct inode *inode, Indirect *ind) * ext2_find_goal - find a prefered place for allocation. * @inode: owner * @block: block we want - * @chain: chain of indirect blocks * @partial: pointer to the last triple within a chain * * Returns preferred place for a block (the goal). */ -static inline int ext2_find_goal(struct inode *inode, - long block, - Indirect chain[4], +static inline int ext2_find_goal(struct inode *inode, long block, Indirect *partial) { struct ext2_block_alloc_info *block_i; @@ -569,7 +566,6 @@ static void ext2_splice_branch(struct inode *inode, * * `handle' can be NULL if create == 0. * - * The BKL may not be held on entry here. Be sure to take it early. * return > 0, # of blocks mapped or allocated. * return = 0, if plain lookup failed. * return < 0, error case. @@ -639,7 +635,7 @@ reread: if (S_ISREG(inode->i_mode) && (!ei->i_block_alloc_info)) ext2_init_block_alloc_info(inode); - goal = ext2_find_goal(inode, iblock, chain, partial); + goal = ext2_find_goal(inode, iblock, partial); /* the number of blocks need to allocate for [d,t]indirect blocks */ indirect_blks = (chain + depth) - partial - 1; @@ -1185,22 +1181,33 @@ void ext2_get_inode_flags(struct ext2_inode_info *ei) ei->i_flags |= EXT2_DIRSYNC_FL; } -void ext2_read_inode (struct inode * inode) +struct inode *ext2_iget (struct super_block *sb, unsigned long ino) { - struct ext2_inode_info *ei = EXT2_I(inode); - ino_t ino = inode->i_ino; + struct ext2_inode_info *ei; struct buffer_head * bh; - struct ext2_inode * raw_inode = ext2_get_inode(inode->i_sb, ino, &bh); + struct ext2_inode *raw_inode; + struct inode *inode; + long ret = -EIO; int n; + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; + + ei = EXT2_I(inode); #ifdef CONFIG_EXT2_FS_POSIX_ACL ei->i_acl = EXT2_ACL_NOT_CACHED; ei->i_default_acl = EXT2_ACL_NOT_CACHED; #endif ei->i_block_alloc_info = NULL; - if (IS_ERR(raw_inode)) + raw_inode = ext2_get_inode(inode->i_sb, ino, &bh); + if (IS_ERR(raw_inode)) { + ret = PTR_ERR(raw_inode); goto bad_inode; + } inode->i_mode = le16_to_cpu(raw_inode->i_mode); inode->i_uid = (uid_t)le16_to_cpu(raw_inode->i_uid_low); @@ -1224,6 +1231,7 @@ void ext2_read_inode (struct inode * inode) if (inode->i_nlink == 0 && (inode->i_mode == 0 || ei->i_dtime)) { /* this inode is deleted */ brelse (bh); + ret = -ESTALE; goto bad_inode; } inode->i_blocks = le32_to_cpu(raw_inode->i_blocks); @@ -1290,11 +1298,12 @@ void ext2_read_inode (struct inode * inode) } brelse (bh); ext2_set_inode_flags(inode); - return; + unlock_new_inode(inode); + return inode; bad_inode: - make_bad_inode(inode); - return; + iget_failed(inode); + return ERR_PTR(ret); } static int ext2_update_inode(struct inode * inode, int do_sync) diff --git a/fs/ext2/ioctl.c b/fs/ext2/ioctl.c index 320b2cb3d4d..b8ea11fee5c 100644 --- a/fs/ext2/ioctl.c +++ b/fs/ext2/ioctl.c @@ -17,9 +17,9 @@ #include <asm/uaccess.h> -int ext2_ioctl (struct inode * inode, struct file * filp, unsigned int cmd, - unsigned long arg) +long ext2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { + struct inode *inode = filp->f_dentry->d_inode; struct ext2_inode_info *ei = EXT2_I(inode); unsigned int flags; unsigned short rsv_window_size; @@ -141,9 +141,6 @@ int ext2_ioctl (struct inode * inode, struct file * filp, unsigned int cmd, #ifdef CONFIG_COMPAT long ext2_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { - struct inode *inode = file->f_path.dentry->d_inode; - int ret; - /* These are just misnamed, they actually get/put from/to user an int */ switch (cmd) { case EXT2_IOC32_GETFLAGS: @@ -161,9 +158,6 @@ long ext2_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) default: return -ENOIOCTLCMD; } - lock_kernel(); - ret = ext2_ioctl(inode, file, cmd, (unsigned long) compat_ptr(arg)); - unlock_kernel(); - return ret; + return ext2_ioctl(file, cmd, (unsigned long) compat_ptr(arg)); } #endif diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c index e69beed839a..80c97fd8c57 100644 --- a/fs/ext2/namei.c +++ b/fs/ext2/namei.c @@ -63,9 +63,9 @@ static struct dentry *ext2_lookup(struct inode * dir, struct dentry *dentry, str ino = ext2_inode_by_name(dir, dentry); inode = NULL; if (ino) { - inode = iget(dir->i_sb, ino); - if (!inode) - return ERR_PTR(-EACCES); + inode = ext2_iget(dir->i_sb, ino); + if (IS_ERR(inode)) + return ERR_CAST(inode); } return d_splice_alias(inode, dentry); } @@ -83,10 +83,10 @@ struct dentry *ext2_get_parent(struct dentry *child) ino = ext2_inode_by_name(child->d_inode, &dotdot); if (!ino) return ERR_PTR(-ENOENT); - inode = iget(child->d_inode->i_sb, ino); + inode = ext2_iget(child->d_inode->i_sb, ino); - if (!inode) - return ERR_PTR(-EACCES); + if (IS_ERR(inode)) + return ERR_CAST(inode); parent = d_alloc_anon(inode); if (!parent) { iput(inode); diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 154e25f13d7..088b011bb97 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -234,16 +234,16 @@ static int ext2_show_options(struct seq_file *seq, struct vfsmount *vfs) le16_to_cpu(es->s_def_resgid) != EXT2_DEF_RESGID) { seq_printf(seq, ",resgid=%u", sbi->s_resgid); } - if (test_opt(sb, ERRORS_CONT)) { + if (test_opt(sb, ERRORS_RO)) { int def_errors = le16_to_cpu(es->s_errors); if (def_errors == EXT2_ERRORS_PANIC || - def_errors == EXT2_ERRORS_RO) { - seq_puts(seq, ",errors=continue"); + def_errors == EXT2_ERRORS_CONTINUE) { + seq_puts(seq, ",errors=remount-ro"); } } - if (test_opt(sb, ERRORS_RO)) - seq_puts(seq, ",errors=remount-ro"); + if (test_opt(sb, ERRORS_CONT)) + seq_puts(seq, ",errors=continue"); if (test_opt(sb, ERRORS_PANIC)) seq_puts(seq, ",errors=panic"); if (test_opt(sb, NO_UID32)) @@ -285,6 +285,9 @@ static int ext2_show_options(struct seq_file *seq, struct vfsmount *vfs) seq_puts(seq, ",xip"); #endif + if (!test_opt(sb, RESERVATION)) + seq_puts(seq, ",noreservation"); + return 0; } @@ -296,7 +299,6 @@ static ssize_t ext2_quota_write(struct super_block *sb, int type, const char *da static const struct super_operations ext2_sops = { .alloc_inode = ext2_alloc_inode, .destroy_inode = ext2_destroy_inode, - .read_inode = ext2_read_inode, .write_inode = ext2_write_inode, .delete_inode = ext2_delete_inode, .put_super = ext2_put_super, @@ -326,11 +328,10 @@ static struct inode *ext2_nfs_get_inode(struct super_block *sb, * it might be "neater" to call ext2_get_inode first and check * if the inode is valid..... */ - inode = iget(sb, ino); - if (inode == NULL) - return ERR_PTR(-ENOMEM); - if (is_bad_inode(inode) || - (generation && inode->i_generation != generation)) { + inode = ext2_iget(sb, ino); + if (IS_ERR(inode)) + return ERR_CAST(inode); + if (generation && inode->i_generation != generation) { /* we didn't find the right inode.. */ iput(inode); return ERR_PTR(-ESTALE); @@ -617,27 +618,24 @@ static int ext2_setup_super (struct super_block * sb, return res; } -static int ext2_check_descriptors (struct super_block * sb) +static int ext2_check_descriptors(struct super_block *sb) { int i; - int desc_block = 0; struct ext2_sb_info *sbi = EXT2_SB(sb); unsigned long first_block = le32_to_cpu(sbi->s_es->s_first_data_block); unsigned long last_block; - struct ext2_group_desc * gdp = NULL; ext2_debug ("Checking group descriptors"); - for (i = 0; i < sbi->s_groups_count; i++) - { + for (i = 0; i < sbi->s_groups_count; i++) { + struct ext2_group_desc *gdp = ext2_get_group_desc(sb, i, NULL); + if (i == sbi->s_groups_count - 1) last_block = le32_to_cpu(sbi->s_es->s_blocks_count) - 1; else last_block = first_block + (EXT2_BLOCKS_PER_GROUP(sb) - 1); - if ((i % EXT2_DESC_PER_BLOCK(sb)) == 0) - gdp = (struct ext2_group_desc *) sbi->s_group_desc[desc_block++]->b_data; if (le32_to_cpu(gdp->bg_block_bitmap) < first_block || le32_to_cpu(gdp->bg_block_bitmap) > last_block) { @@ -667,7 +665,6 @@ static int ext2_check_descriptors (struct super_block * sb) return 0; } first_block += EXT2_BLOCKS_PER_GROUP(sb); - gdp++; } return 1; } @@ -680,11 +677,31 @@ static int ext2_check_descriptors (struct super_block * sb) static loff_t ext2_max_size(int bits) { loff_t res = EXT2_NDIR_BLOCKS; - /* This constant is calculated to be the largest file size for a - * dense, 4k-blocksize file such that the total number of + int meta_blocks; + loff_t upper_limit; + + /* This is calculated to be the largest file size for a + * dense, file such that the total number of * sectors in the file, including data and all indirect blocks, - * does not exceed 2^32. */ - const loff_t upper_limit = 0x1ff7fffd000LL; + * does not exceed 2^32 -1 + * __u32 i_blocks representing the total number of + * 512 bytes blocks of the file + */ + upper_limit = (1LL << 32) - 1; + + /* total blocks in file system block size */ + upper_limit >>= (bits - 9); + + + /* indirect blocks */ + meta_blocks = 1; + /* double indirect blocks */ + meta_blocks += 1 + (1LL << (bits-2)); + /* tripple indirect blocks */ + meta_blocks += 1 + (1LL << (bits-2)) + (1LL << (2*(bits-2))); + + upper_limit -= meta_blocks; + upper_limit <<= bits; res += 1LL << (bits-2); res += 1LL << (2*(bits-2)); @@ -692,6 +709,10 @@ static loff_t ext2_max_size(int bits) res <<= bits; if (res > upper_limit) res = upper_limit; + + if (res > MAX_LFS_FILESIZE) + res = MAX_LFS_FILESIZE; + return res; } @@ -726,6 +747,7 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) unsigned long logic_sb_block; unsigned long offset = 0; unsigned long def_mount_opts; + long ret = -EINVAL; int blocksize = BLOCK_SIZE; int db_count; int i, j; @@ -796,10 +818,10 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) if (le16_to_cpu(sbi->s_es->s_errors) == EXT2_ERRORS_PANIC) set_opt(sbi->s_mount_opt, ERRORS_PANIC); - else if (le16_to_cpu(sbi->s_es->s_errors) == EXT2_ERRORS_RO) - set_opt(sbi->s_mount_opt, ERRORS_RO); - else + else if (le16_to_cpu(sbi->s_es->s_errors) == EXT2_ERRORS_CONTINUE) set_opt(sbi->s_mount_opt, ERRORS_CONT); + else + set_opt(sbi->s_mount_opt, ERRORS_RO); sbi->s_resuid = le16_to_cpu(es->s_def_resuid); sbi->s_resgid = le16_to_cpu(es->s_def_resgid); @@ -844,8 +866,7 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) blocksize = BLOCK_SIZE << le32_to_cpu(sbi->s_es->s_log_block_size); - if ((ext2_use_xip(sb)) && ((blocksize != PAGE_SIZE) || - (sb->s_blocksize != blocksize))) { + if (ext2_use_xip(sb) && blocksize != PAGE_SIZE) { if (!silent) printk("XIP: Unsupported blocksize\n"); goto failed_mount; @@ -1022,19 +1043,24 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) sb->s_op = &ext2_sops; sb->s_export_op = &ext2_export_ops; sb->s_xattr = ext2_xattr_handlers; - root = iget(sb, EXT2_ROOT_INO); - sb->s_root = d_alloc_root(root); - if (!sb->s_root) { - iput(root); - printk(KERN_ERR "EXT2-fs: get root inode failed\n"); + root = ext2_iget(sb, EXT2_ROOT_INO); + if (IS_ERR(root)) { + ret = PTR_ERR(root); goto failed_mount3; } if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) { - dput(sb->s_root); - sb->s_root = NULL; + iput(root); printk(KERN_ERR "EXT2-fs: corrupt root inode, run e2fsck\n"); goto failed_mount3; } + + sb->s_root = d_alloc_root(root); + if (!sb->s_root) { + iput(root); + printk(KERN_ERR "EXT2-fs: get root inode failed\n"); + ret = -ENOMEM; + goto failed_mount3; + } if (EXT2_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_HAS_JOURNAL)) ext2_warning(sb, __FUNCTION__, "mounting ext3 filesystem as ext2"); @@ -1061,7 +1087,7 @@ failed_mount: failed_sbi: sb->s_fs_info = NULL; kfree(sbi); - return -EINVAL; + return ret; } static void ext2_commit_super (struct super_block * sb, diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c index 3e8683dbb13..a99d46f3b26 100644 --- a/fs/ext2/xattr.c +++ b/fs/ext2/xattr.c @@ -835,7 +835,7 @@ ext2_xattr_cache_insert(struct buffer_head *bh) struct mb_cache_entry *ce; int error; - ce = mb_cache_entry_alloc(ext2_xattr_cache); + ce = mb_cache_entry_alloc(ext2_xattr_cache, GFP_NOFS); if (!ce) return -ENOMEM; error = mb_cache_entry_insert(ce, bh->b_bdev, bh->b_blocknr, &hash); diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c index d34e9967430..a754d184817 100644 --- a/fs/ext3/acl.c +++ b/fs/ext3/acl.c @@ -37,7 +37,7 @@ ext3_acl_from_disk(const void *value, size_t size) return ERR_PTR(-EINVAL); if (count == 0) return NULL; - acl = posix_acl_alloc(count, GFP_KERNEL); + acl = posix_acl_alloc(count, GFP_NOFS); if (!acl) return ERR_PTR(-ENOMEM); for (n=0; n < count; n++) { @@ -91,7 +91,7 @@ ext3_acl_to_disk(const struct posix_acl *acl, size_t *size) *size = ext3_acl_size(acl->a_count); ext_acl = kmalloc(sizeof(ext3_acl_header) + acl->a_count * - sizeof(ext3_acl_entry), GFP_KERNEL); + sizeof(ext3_acl_entry), GFP_NOFS); if (!ext_acl) return ERR_PTR(-ENOMEM); ext_acl->a_version = cpu_to_le32(EXT3_ACL_VERSION); @@ -187,7 +187,7 @@ ext3_get_acl(struct inode *inode, int type) } retval = ext3_xattr_get(inode, name_index, "", NULL, 0); if (retval > 0) { - value = kmalloc(retval, GFP_KERNEL); + value = kmalloc(retval, GFP_NOFS); if (!value) return ERR_PTR(-ENOMEM); retval = ext3_xattr_get(inode, name_index, "", value, retval); @@ -335,7 +335,7 @@ ext3_init_acl(handle_t *handle, struct inode *inode, struct inode *dir) if (error) goto cleanup; } - clone = posix_acl_clone(acl, GFP_KERNEL); + clone = posix_acl_clone(acl, GFP_NOFS); error = -ENOMEM; if (!clone) goto cleanup; diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c index a8ba7e83127..da0cb2c0e43 100644 --- a/fs/ext3/balloc.c +++ b/fs/ext3/balloc.c @@ -80,13 +80,57 @@ struct ext3_group_desc * ext3_get_group_desc(struct super_block * sb, return desc + offset; } +static int ext3_valid_block_bitmap(struct super_block *sb, + struct ext3_group_desc *desc, + unsigned int block_group, + struct buffer_head *bh) +{ + ext3_grpblk_t offset; + ext3_grpblk_t next_zero_bit; + ext3_fsblk_t bitmap_blk; + ext3_fsblk_t group_first_block; + + group_first_block = ext3_group_first_block_no(sb, block_group); + + /* check whether block bitmap block number is set */ + bitmap_blk = le32_to_cpu(desc->bg_block_bitmap); + offset = bitmap_blk - group_first_block; + if (!ext3_test_bit(offset, bh->b_data)) + /* bad block bitmap */ + goto err_out; + + /* check whether the inode bitmap block number is set */ + bitmap_blk = le32_to_cpu(desc->bg_inode_bitmap); + offset = bitmap_blk - group_first_block; + if (!ext3_test_bit(offset, bh->b_data)) + /* bad block bitmap */ + goto err_out; + + /* check whether the inode table block number is set */ + bitmap_blk = le32_to_cpu(desc->bg_inode_table); + offset = bitmap_blk - group_first_block; + next_zero_bit = ext3_find_next_zero_bit(bh->b_data, + offset + EXT3_SB(sb)->s_itb_per_group, + offset); + if (next_zero_bit >= offset + EXT3_SB(sb)->s_itb_per_group) + /* good bitmap for inode tables */ + return 1; + +err_out: + ext3_error(sb, __FUNCTION__, + "Invalid block bitmap - " + "block_group = %d, block = %lu", + block_group, bitmap_blk); + return 0; +} + /** * read_block_bitmap() * @sb: super block * @block_group: given block group * - * Read the bitmap for a given block_group, reading into the specified - * slot in the superblock's bitmap cache. + * Read the bitmap for a given block_group,and validate the + * bits for block/inode/inode tables are set in the bitmaps * * Return buffer_head on success or NULL in case of failure. */ @@ -95,17 +139,35 @@ read_block_bitmap(struct super_block *sb, unsigned int block_group) { struct ext3_group_desc * desc; struct buffer_head * bh = NULL; + ext3_fsblk_t bitmap_blk; - desc = ext3_get_group_desc (sb, block_group, NULL); + desc = ext3_get_group_desc(sb, block_group, NULL); if (!desc) - goto error_out; - bh = sb_bread(sb, le32_to_cpu(desc->bg_block_bitmap)); - if (!bh) - ext3_error (sb, "read_block_bitmap", + return NULL; + bitmap_blk = le32_to_cpu(desc->bg_block_bitmap); + bh = sb_getblk(sb, bitmap_blk); + if (unlikely(!bh)) { + ext3_error(sb, __FUNCTION__, "Cannot read block bitmap - " "block_group = %d, block_bitmap = %u", block_group, le32_to_cpu(desc->bg_block_bitmap)); -error_out: + return NULL; + } + if (likely(bh_uptodate_or_lock(bh))) + return bh; + + if (bh_submit_read(bh) < 0) { + brelse(bh); + ext3_error(sb, __FUNCTION__, + "Cannot read block bitmap - " + "block_group = %d, block_bitmap = %u", + block_group, le32_to_cpu(desc->bg_block_bitmap)); + return NULL; + } + if (!ext3_valid_block_bitmap(sb, desc, block_group, bh)) { + brelse(bh); + return NULL; + } return bh; } /* @@ -468,11 +530,13 @@ do_more: in_range (block, le32_to_cpu(desc->bg_inode_table), sbi->s_itb_per_group) || in_range (block + count - 1, le32_to_cpu(desc->bg_inode_table), - sbi->s_itb_per_group)) + sbi->s_itb_per_group)) { ext3_error (sb, "ext3_free_blocks", "Freeing blocks in system zones - " "Block = "E3FSBLK", count = %lu", block, count); + goto error_return; + } /* * We are about to start releasing blocks in the bitmap, @@ -566,9 +630,7 @@ do_more: jbd_unlock_bh_state(bitmap_bh); spin_lock(sb_bgl_lock(sbi, block_group)); - desc->bg_free_blocks_count = - cpu_to_le16(le16_to_cpu(desc->bg_free_blocks_count) + - group_freed); + le16_add_cpu(&desc->bg_free_blocks_count, group_freed); spin_unlock(sb_bgl_lock(sbi, block_group)); percpu_counter_add(&sbi->s_freeblocks_counter, count); @@ -1508,7 +1570,7 @@ retry_alloc: /* * Now search the rest of the groups. We assume that - * i and gdp correctly point to the last group visited. + * group_no and gdp correctly point to the last group visited. */ for (bgi = 0; bgi < ngroups; bgi++) { group_no++; @@ -1575,11 +1637,13 @@ allocated: in_range(ret_block, le32_to_cpu(gdp->bg_inode_table), EXT3_SB(sb)->s_itb_per_group) || in_range(ret_block + num - 1, le32_to_cpu(gdp->bg_inode_table), - EXT3_SB(sb)->s_itb_per_group)) + EXT3_SB(sb)->s_itb_per_group)) { ext3_error(sb, "ext3_new_block", "Allocating block in system zone - " "blocks from "E3FSBLK", length %lu", ret_block, num); + goto out; + } performed_allocation = 1; @@ -1630,8 +1694,7 @@ allocated: ret_block, goal_hits, goal_attempts); spin_lock(sb_bgl_lock(sbi, group_no)); - gdp->bg_free_blocks_count = - cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count)-num); + le16_add_cpu(&gdp->bg_free_blocks_count, -num); spin_unlock(sb_bgl_lock(sbi, group_no)); percpu_counter_sub(&sbi->s_freeblocks_counter, num); @@ -1782,11 +1845,7 @@ static unsigned long ext3_bg_num_gdb_meta(struct super_block *sb, int group) static unsigned long ext3_bg_num_gdb_nometa(struct super_block *sb, int group) { - if (EXT3_HAS_RO_COMPAT_FEATURE(sb, - EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER) && - !ext3_group_sparse(group)) - return 0; - return EXT3_SB(sb)->s_gdb_count; + return ext3_bg_has_super(sb, group) ? EXT3_SB(sb)->s_gdb_count : 0; } /** diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c index 1bc8cd89c51..4f4020c5468 100644 --- a/fs/ext3/ialloc.c +++ b/fs/ext3/ialloc.c @@ -164,11 +164,9 @@ void ext3_free_inode (handle_t *handle, struct inode * inode) if (gdp) { spin_lock(sb_bgl_lock(sbi, block_group)); - gdp->bg_free_inodes_count = cpu_to_le16( - le16_to_cpu(gdp->bg_free_inodes_count) + 1); + le16_add_cpu(&gdp->bg_free_inodes_count, 1); if (is_directory) - gdp->bg_used_dirs_count = cpu_to_le16( - le16_to_cpu(gdp->bg_used_dirs_count) - 1); + le16_add_cpu(&gdp->bg_used_dirs_count, -1); spin_unlock(sb_bgl_lock(sbi, block_group)); percpu_counter_inc(&sbi->s_freeinodes_counter); if (is_directory) @@ -527,11 +525,9 @@ got: err = ext3_journal_get_write_access(handle, bh2); if (err) goto fail; spin_lock(sb_bgl_lock(sbi, group)); - gdp->bg_free_inodes_count = - cpu_to_le16(le16_to_cpu(gdp->bg_free_inodes_count) - 1); + le16_add_cpu(&gdp->bg_free_inodes_count, -1); if (S_ISDIR(mode)) { - gdp->bg_used_dirs_count = - cpu_to_le16(le16_to_cpu(gdp->bg_used_dirs_count) + 1); + le16_add_cpu(&gdp->bg_used_dirs_count, 1); } spin_unlock(sb_bgl_lock(sbi, group)); BUFFER_TRACE(bh2, "call ext3_journal_dirty_metadata"); @@ -642,14 +638,15 @@ struct inode *ext3_orphan_get(struct super_block *sb, unsigned long ino) unsigned long max_ino = le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count); unsigned long block_group; int bit; - struct buffer_head *bitmap_bh = NULL; + struct buffer_head *bitmap_bh; struct inode *inode = NULL; + long err = -EIO; /* Error cases - e2fsck has already cleaned up for us */ if (ino > max_ino) { ext3_warning(sb, __FUNCTION__, "bad orphan ino %lu! e2fsck was run?", ino); - goto out; + goto error; } block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb); @@ -658,38 +655,49 @@ struct inode *ext3_orphan_get(struct super_block *sb, unsigned long ino) if (!bitmap_bh) { ext3_warning(sb, __FUNCTION__, "inode bitmap error for orphan %lu", ino); - goto out; + goto error; } /* Having the inode bit set should be a 100% indicator that this * is a valid orphan (no e2fsck run on fs). Orphans also include * inodes that were being truncated, so we can't check i_nlink==0. */ - if (!ext3_test_bit(bit, bitmap_bh->b_data) || - !(inode = iget(sb, ino)) || is_bad_inode(inode) || - NEXT_ORPHAN(inode) > max_ino) { - ext3_warning(sb, __FUNCTION__, - "bad orphan inode %lu! e2fsck was run?", ino); - printk(KERN_NOTICE "ext3_test_bit(bit=%d, block=%llu) = %d\n", - bit, (unsigned long long)bitmap_bh->b_blocknr, - ext3_test_bit(bit, bitmap_bh->b_data)); - printk(KERN_NOTICE "inode=%p\n", inode); - if (inode) { - printk(KERN_NOTICE "is_bad_inode(inode)=%d\n", - is_bad_inode(inode)); - printk(KERN_NOTICE "NEXT_ORPHAN(inode)=%u\n", - NEXT_ORPHAN(inode)); - printk(KERN_NOTICE "max_ino=%lu\n", max_ino); - } + if (!ext3_test_bit(bit, bitmap_bh->b_data)) + goto bad_orphan; + + inode = ext3_iget(sb, ino); + if (IS_ERR(inode)) + goto iget_failed; + + if (NEXT_ORPHAN(inode) > max_ino) + goto bad_orphan; + brelse(bitmap_bh); + return inode; + +iget_failed: + err = PTR_ERR(inode); + inode = NULL; +bad_orphan: + ext3_warning(sb, __FUNCTION__, + "bad orphan inode %lu! e2fsck was run?", ino); + printk(KERN_NOTICE "ext3_test_bit(bit=%d, block=%llu) = %d\n", + bit, (unsigned long long)bitmap_bh->b_blocknr, + ext3_test_bit(bit, bitmap_bh->b_data)); + printk(KERN_NOTICE "inode=%p\n", inode); + if (inode) { + printk(KERN_NOTICE "is_bad_inode(inode)=%d\n", + is_bad_inode(inode)); + printk(KERN_NOTICE "NEXT_ORPHAN(inode)=%u\n", + NEXT_ORPHAN(inode)); + printk(KERN_NOTICE "max_ino=%lu\n", max_ino); /* Avoid freeing blocks if we got a bad deleted inode */ - if (inode && inode->i_nlink == 0) + if (inode->i_nlink == 0) inode->i_blocks = 0; iput(inode); - inode = NULL; } -out: brelse(bitmap_bh); - return inode; +error: + return ERR_PTR(err); } unsigned long ext3_count_free_inodes (struct super_block * sb) diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 9b162cd6c16..eb95670a27e 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -439,16 +439,14 @@ static ext3_fsblk_t ext3_find_near(struct inode *inode, Indirect *ind) * ext3_find_goal - find a prefered place for allocation. * @inode: owner * @block: block we want - * @chain: chain of indirect blocks * @partial: pointer to the last triple within a chain - * @goal: place to store the result. * * Normally this function find the prefered place for block allocation, - * stores it in *@goal and returns zero. + * returns it. */ static ext3_fsblk_t ext3_find_goal(struct inode *inode, long block, - Indirect chain[4], Indirect *partial) + Indirect *partial) { struct ext3_block_alloc_info *block_i; @@ -884,7 +882,7 @@ int ext3_get_blocks_handle(handle_t *handle, struct inode *inode, if (S_ISREG(inode->i_mode) && (!ei->i_block_alloc_info)) ext3_init_block_alloc_info(inode); - goal = ext3_find_goal(inode, iblock, chain, partial); + goal = ext3_find_goal(inode, iblock, partial); /* the number of blocks need to allocate for [d,t]indirect blocks */ indirect_blks = (chain + depth) - partial - 1; @@ -941,55 +939,45 @@ out: return err; } -#define DIO_CREDITS (EXT3_RESERVE_TRANS_BLOCKS + 32) +/* Maximum number of blocks we map for direct IO at once. */ +#define DIO_MAX_BLOCKS 4096 +/* + * Number of credits we need for writing DIO_MAX_BLOCKS: + * We need sb + group descriptor + bitmap + inode -> 4 + * For B blocks with A block pointers per block we need: + * 1 (triple ind.) + (B/A/A + 2) (doubly ind.) + (B/A + 2) (indirect). + * If we plug in 4096 for B and 256 for A (for 1KB block size), we get 25. + */ +#define DIO_CREDITS 25 static int ext3_get_block(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) { handle_t *handle = ext3_journal_current_handle(); - int ret = 0; + int ret = 0, started = 0; unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; - if (!create) - goto get_block; /* A read */ - - if (max_blocks == 1) - goto get_block; /* A single block get */ - - if (handle->h_transaction->t_state == T_LOCKED) { - /* - * Huge direct-io writes can hold off commits for long - * periods of time. Let this commit run. - */ - ext3_journal_stop(handle); - handle = ext3_journal_start(inode, DIO_CREDITS); - if (IS_ERR(handle)) + if (create && !handle) { /* Direct IO write... */ + if (max_blocks > DIO_MAX_BLOCKS) + max_blocks = DIO_MAX_BLOCKS; + handle = ext3_journal_start(inode, DIO_CREDITS + + 2 * EXT3_QUOTA_TRANS_BLOCKS(inode->i_sb)); + if (IS_ERR(handle)) { ret = PTR_ERR(handle); - goto get_block; - } - - if (handle->h_buffer_credits <= EXT3_RESERVE_TRANS_BLOCKS) { - /* - * Getting low on buffer credits... - */ - ret = ext3_journal_extend(handle, DIO_CREDITS); - if (ret > 0) { - /* - * Couldn't extend the transaction. Start a new one. - */ - ret = ext3_journal_restart(handle, DIO_CREDITS); + goto out; } + started = 1; } -get_block: - if (ret == 0) { - ret = ext3_get_blocks_handle(handle, inode, iblock, + ret = ext3_get_blocks_handle(handle, inode, iblock, max_blocks, bh_result, create, 0); - if (ret > 0) { - bh_result->b_size = (ret << inode->i_blkbits); - ret = 0; - } + if (ret > 0) { + bh_result->b_size = (ret << inode->i_blkbits); + ret = 0; } + if (started) + ext3_journal_stop(handle); +out: return ret; } @@ -1680,7 +1668,8 @@ static int ext3_releasepage(struct page *page, gfp_t wait) * if the machine crashes during the write. * * If the O_DIRECT write is intantiating holes inside i_size and the machine - * crashes then stale disk data _may_ be exposed inside the file. + * crashes then stale disk data _may_ be exposed inside the file. But current + * VFS code falls back into buffered path in that case so we are safe. */ static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, loff_t offset, @@ -1689,7 +1678,7 @@ static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb, struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; struct ext3_inode_info *ei = EXT3_I(inode); - handle_t *handle = NULL; + handle_t *handle; ssize_t ret; int orphan = 0; size_t count = iov_length(iov, nr_segs); @@ -1697,17 +1686,21 @@ static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb, if (rw == WRITE) { loff_t final_size = offset + count; - handle = ext3_journal_start(inode, DIO_CREDITS); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - goto out; - } if (final_size > inode->i_size) { + /* Credits for sb + inode write */ + handle = ext3_journal_start(inode, 2); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; + } ret = ext3_orphan_add(handle, inode); - if (ret) - goto out_stop; + if (ret) { + ext3_journal_stop(handle); + goto out; + } orphan = 1; ei->i_disksize = inode->i_size; + ext3_journal_stop(handle); } } @@ -1715,18 +1708,21 @@ static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb, offset, nr_segs, ext3_get_block, NULL); - /* - * Reacquire the handle: ext3_get_block() can restart the transaction - */ - handle = ext3_journal_current_handle(); - -out_stop: - if (handle) { + if (orphan) { int err; - if (orphan && inode->i_nlink) + /* Credits for sb + inode write */ + handle = ext3_journal_start(inode, 2); + if (IS_ERR(handle)) { + /* This is really bad luck. We've written the data + * but cannot extend i_size. Bail out and pretend + * the write failed... */ + ret = PTR_ERR(handle); + goto out; + } + if (inode->i_nlink) ext3_orphan_del(handle, inode); - if (orphan && ret > 0) { + if (ret > 0) { loff_t end = offset + ret; if (end > inode->i_size) { ei->i_disksize = end; @@ -1845,7 +1841,7 @@ static int ext3_block_truncate_page(handle_t *handle, struct page *page, */ if (!page_has_buffers(page) && test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode) && PageUptodate(page)) { - zero_user_page(page, offset, length, KM_USER0); + zero_user(page, offset, length); set_page_dirty(page); goto unlock; } @@ -1898,7 +1894,7 @@ static int ext3_block_truncate_page(handle_t *handle, struct page *page, goto unlock; } - zero_user_page(page, offset, length, KM_USER0); + zero_user(page, offset, length); BUFFER_TRACE(bh, "zeroed end of block"); err = 0; @@ -2658,21 +2654,31 @@ void ext3_get_inode_flags(struct ext3_inode_info *ei) ei->i_flags |= EXT3_DIRSYNC_FL; } -void ext3_read_inode(struct inode * inode) +struct inode *ext3_iget(struct super_block *sb, unsigned long ino) { struct ext3_iloc iloc; struct ext3_inode *raw_inode; - struct ext3_inode_info *ei = EXT3_I(inode); + struct ext3_inode_info *ei; struct buffer_head *bh; + struct inode *inode; + long ret; int block; + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; + + ei = EXT3_I(inode); #ifdef CONFIG_EXT3_FS_POSIX_ACL ei->i_acl = EXT3_ACL_NOT_CACHED; ei->i_default_acl = EXT3_ACL_NOT_CACHED; #endif ei->i_block_alloc_info = NULL; - if (__ext3_get_inode_loc(inode, &iloc, 0)) + ret = __ext3_get_inode_loc(inode, &iloc, 0); + if (ret < 0) goto bad_inode; bh = iloc.bh; raw_inode = ext3_raw_inode(&iloc); @@ -2703,6 +2709,7 @@ void ext3_read_inode(struct inode * inode) !(EXT3_SB(inode->i_sb)->s_mount_state & EXT3_ORPHAN_FS)) { /* this inode is deleted */ brelse (bh); + ret = -ESTALE; goto bad_inode; } /* The only unlinked inodes we let through here have @@ -2746,6 +2753,7 @@ void ext3_read_inode(struct inode * inode) if (EXT3_GOOD_OLD_INODE_SIZE + ei->i_extra_isize > EXT3_INODE_SIZE(inode->i_sb)) { brelse (bh); + ret = -EIO; goto bad_inode; } if (ei->i_extra_isize == 0) { @@ -2787,11 +2795,12 @@ void ext3_read_inode(struct inode * inode) } brelse (iloc.bh); ext3_set_inode_flags(inode); - return; + unlock_new_inode(inode); + return inode; bad_inode: - make_bad_inode(inode); - return; + iget_failed(inode); + return ERR_PTR(ret); } /* diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index 4ab6f76e63d..dec3e0d88ab 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -860,14 +860,10 @@ static struct buffer_head * ext3_find_entry (struct dentry *dentry, int nblocks, i, err; struct inode *dir = dentry->d_parent->d_inode; int namelen; - const u8 *name; - unsigned blocksize; *res_dir = NULL; sb = dir->i_sb; - blocksize = sb->s_blocksize; namelen = dentry->d_name.len; - name = dentry->d_name.name; if (namelen > EXT3_NAME_LEN) return NULL; if (is_dx(dir)) { @@ -1041,17 +1037,11 @@ static struct dentry *ext3_lookup(struct inode * dir, struct dentry *dentry, str if (!ext3_valid_inum(dir->i_sb, ino)) { ext3_error(dir->i_sb, "ext3_lookup", "bad inode number: %lu", ino); - inode = NULL; - } else - inode = iget(dir->i_sb, ino); - - if (!inode) - return ERR_PTR(-EACCES); - - if (is_bad_inode(inode)) { - iput(inode); - return ERR_PTR(-ENOENT); + return ERR_PTR(-EIO); } + inode = ext3_iget(dir->i_sb, ino); + if (IS_ERR(inode)) + return ERR_CAST(inode); } return d_splice_alias(inode, dentry); } @@ -1080,18 +1070,13 @@ struct dentry *ext3_get_parent(struct dentry *child) if (!ext3_valid_inum(child->d_inode->i_sb, ino)) { ext3_error(child->d_inode->i_sb, "ext3_get_parent", "bad inode number: %lu", ino); - inode = NULL; - } else - inode = iget(child->d_inode->i_sb, ino); - - if (!inode) - return ERR_PTR(-EACCES); - - if (is_bad_inode(inode)) { - iput(inode); - return ERR_PTR(-ENOENT); + return ERR_PTR(-EIO); } + inode = ext3_iget(child->d_inode->i_sb, ino); + if (IS_ERR(inode)) + return ERR_CAST(inode); + parent = d_alloc_anon(inode); if (!parent) { iput(inode); diff --git a/fs/ext3/resize.c b/fs/ext3/resize.c index 44de1453c30..0e97b6e07cb 100644 --- a/fs/ext3/resize.c +++ b/fs/ext3/resize.c @@ -485,7 +485,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, goto exit_dindj; n_group_desc = kmalloc((gdb_num + 1) * sizeof(struct buffer_head *), - GFP_KERNEL); + GFP_NOFS); if (!n_group_desc) { err = -ENOMEM; ext3_warning (sb, __FUNCTION__, @@ -518,8 +518,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, EXT3_SB(sb)->s_gdb_count++; kfree(o_group_desc); - es->s_reserved_gdt_blocks = - cpu_to_le16(le16_to_cpu(es->s_reserved_gdt_blocks) - 1); + le16_add_cpu(&es->s_reserved_gdt_blocks, -1); ext3_journal_dirty_metadata(handle, EXT3_SB(sb)->s_sbh); return 0; @@ -569,7 +568,7 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode, int res, i; int err; - primary = kmalloc(reserved_gdb * sizeof(*primary), GFP_KERNEL); + primary = kmalloc(reserved_gdb * sizeof(*primary), GFP_NOFS); if (!primary) return -ENOMEM; @@ -795,12 +794,11 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) "No reserved GDT blocks, can't resize"); return -EPERM; } - inode = iget(sb, EXT3_RESIZE_INO); - if (!inode || is_bad_inode(inode)) { + inode = ext3_iget(sb, EXT3_RESIZE_INO); + if (IS_ERR(inode)) { ext3_warning(sb, __FUNCTION__, "Error opening resize inode"); - iput(inode); - return -ENOENT; + return PTR_ERR(inode); } } @@ -891,10 +889,8 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) * blocks/inodes before the group is live won't actually let us * allocate the new space yet. */ - es->s_blocks_count = cpu_to_le32(le32_to_cpu(es->s_blocks_count) + - input->blocks_count); - es->s_inodes_count = cpu_to_le32(le32_to_cpu(es->s_inodes_count) + - EXT3_INODES_PER_GROUP(sb)); + le32_add_cpu(&es->s_blocks_count, input->blocks_count); + le32_add_cpu(&es->s_inodes_count, EXT3_INODES_PER_GROUP(sb)); /* * We need to protect s_groups_count against other CPUs seeing @@ -927,8 +923,7 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) /* Update the reserved block counts only once the new group is * active. */ - es->s_r_blocks_count = cpu_to_le32(le32_to_cpu(es->s_r_blocks_count) + - input->reserved_blocks); + le32_add_cpu(&es->s_r_blocks_count, input->reserved_blocks); /* Update the free space counts */ percpu_counter_add(&sbi->s_freeblocks_counter, diff --git a/fs/ext3/super.c b/fs/ext3/super.c index cb14de1502c..ad536066408 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -575,16 +575,16 @@ static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs) le16_to_cpu(es->s_def_resgid) != EXT3_DEF_RESGID) { seq_printf(seq, ",resgid=%u", sbi->s_resgid); } - if (test_opt(sb, ERRORS_CONT)) { + if (test_opt(sb, ERRORS_RO)) { int def_errors = le16_to_cpu(es->s_errors); if (def_errors == EXT3_ERRORS_PANIC || - def_errors == EXT3_ERRORS_RO) { - seq_puts(seq, ",errors=continue"); + def_errors == EXT3_ERRORS_CONTINUE) { + seq_puts(seq, ",errors=remount-ro"); } } - if (test_opt(sb, ERRORS_RO)) - seq_puts(seq, ",errors=remount-ro"); + if (test_opt(sb, ERRORS_CONT)) + seq_puts(seq, ",errors=continue"); if (test_opt(sb, ERRORS_PANIC)) seq_puts(seq, ",errors=panic"); if (test_opt(sb, NO_UID32)) @@ -649,11 +649,10 @@ static struct inode *ext3_nfs_get_inode(struct super_block *sb, * Currently we don't know the generation for parent directory, so * a generation of 0 means "accept any" */ - inode = iget(sb, ino); - if (inode == NULL) - return ERR_PTR(-ENOMEM); - if (is_bad_inode(inode) || - (generation && inode->i_generation != generation)) { + inode = ext3_iget(sb, ino); + if (IS_ERR(inode)) + return ERR_CAST(inode); + if (generation && inode->i_generation != generation) { iput(inode); return ERR_PTR(-ESTALE); } @@ -722,7 +721,6 @@ static struct quotactl_ops ext3_qctl_operations = { static const struct super_operations ext3_sops = { .alloc_inode = ext3_alloc_inode, .destroy_inode = ext3_destroy_inode, - .read_inode = ext3_read_inode, .write_inode = ext3_write_inode, .dirty_inode = ext3_dirty_inode, .delete_inode = ext3_delete_inode, @@ -808,8 +806,8 @@ static match_table_t tokens = { {Opt_quota, "quota"}, {Opt_usrquota, "usrquota"}, {Opt_barrier, "barrier=%u"}, - {Opt_err, NULL}, {Opt_resize, "resize"}, + {Opt_err, NULL}, }; static ext3_fsblk_t get_sb_block(void **data) @@ -1224,7 +1222,7 @@ static int ext3_setup_super(struct super_block *sb, struct ext3_super_block *es, #endif if (!(__s16) le16_to_cpu(es->s_max_mnt_count)) es->s_max_mnt_count = cpu_to_le16(EXT3_DFL_MAX_MNT_COUNT); - es->s_mnt_count=cpu_to_le16(le16_to_cpu(es->s_mnt_count) + 1); + le16_add_cpu(&es->s_mnt_count, 1); es->s_mtime = cpu_to_le32(get_seconds()); ext3_update_dynamic_rev(sb); EXT3_SET_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER); @@ -1252,28 +1250,24 @@ static int ext3_setup_super(struct super_block *sb, struct ext3_super_block *es, } /* Called at mount-time, super-block is locked */ -static int ext3_check_descriptors (struct super_block * sb) +static int ext3_check_descriptors(struct super_block *sb) { struct ext3_sb_info *sbi = EXT3_SB(sb); ext3_fsblk_t first_block = le32_to_cpu(sbi->s_es->s_first_data_block); ext3_fsblk_t last_block; - struct ext3_group_desc * gdp = NULL; - int desc_block = 0; int i; ext3_debug ("Checking group descriptors"); - for (i = 0; i < sbi->s_groups_count; i++) - { + for (i = 0; i < sbi->s_groups_count; i++) { + struct ext3_group_desc *gdp = ext3_get_group_desc(sb, i, NULL); + if (i == sbi->s_groups_count - 1) last_block = le32_to_cpu(sbi->s_es->s_blocks_count) - 1; else last_block = first_block + (EXT3_BLOCKS_PER_GROUP(sb) - 1); - if ((i % EXT3_DESC_PER_BLOCK(sb)) == 0) - gdp = (struct ext3_group_desc *) - sbi->s_group_desc[desc_block++]->b_data; if (le32_to_cpu(gdp->bg_block_bitmap) < first_block || le32_to_cpu(gdp->bg_block_bitmap) > last_block) { @@ -1306,7 +1300,6 @@ static int ext3_check_descriptors (struct super_block * sb) return 0; } first_block += EXT3_BLOCKS_PER_GROUP(sb); - gdp++; } sbi->s_es->s_free_blocks_count=cpu_to_le32(ext3_count_free_blocks(sb)); @@ -1383,8 +1376,8 @@ static void ext3_orphan_cleanup (struct super_block * sb, while (es->s_last_orphan) { struct inode *inode; - if (!(inode = - ext3_orphan_get(sb, le32_to_cpu(es->s_last_orphan)))) { + inode = ext3_orphan_get(sb, le32_to_cpu(es->s_last_orphan)); + if (IS_ERR(inode)) { es->s_last_orphan = 0; break; } @@ -1436,11 +1429,31 @@ static void ext3_orphan_cleanup (struct super_block * sb, static loff_t ext3_max_size(int bits) { loff_t res = EXT3_NDIR_BLOCKS; - /* This constant is calculated to be the largest file size for a - * dense, 4k-blocksize file such that the total number of + int meta_blocks; + loff_t upper_limit; + + /* This is calculated to be the largest file size for a + * dense, file such that the total number of * sectors in the file, including data and all indirect blocks, - * does not exceed 2^32. */ - const loff_t upper_limit = 0x1ff7fffd000LL; + * does not exceed 2^32 -1 + * __u32 i_blocks representing the total number of + * 512 bytes blocks of the file + */ + upper_limit = (1LL << 32) - 1; + + /* total blocks in file system block size */ + upper_limit >>= (bits - 9); + + + /* indirect blocks */ + meta_blocks = 1; + /* double indirect blocks */ + meta_blocks += 1 + (1LL << (bits-2)); + /* tripple indirect blocks */ + meta_blocks += 1 + (1LL << (bits-2)) + (1LL << (2*(bits-2))); + + upper_limit -= meta_blocks; + upper_limit <<= bits; res += 1LL << (bits-2); res += 1LL << (2*(bits-2)); @@ -1448,6 +1461,10 @@ static loff_t ext3_max_size(int bits) res <<= bits; if (res > upper_limit) res = upper_limit; + + if (res > MAX_LFS_FILESIZE) + res = MAX_LFS_FILESIZE; + return res; } @@ -1489,6 +1506,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) int db_count; int i; int needs_recovery; + int ret = -EINVAL; __le32 features; int err; @@ -1559,10 +1577,10 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) if (le16_to_cpu(sbi->s_es->s_errors) == EXT3_ERRORS_PANIC) set_opt(sbi->s_mount_opt, ERRORS_PANIC); - else if (le16_to_cpu(sbi->s_es->s_errors) == EXT3_ERRORS_RO) - set_opt(sbi->s_mount_opt, ERRORS_RO); - else + else if (le16_to_cpu(sbi->s_es->s_errors) == EXT3_ERRORS_CONTINUE) set_opt(sbi->s_mount_opt, ERRORS_CONT); + else + set_opt(sbi->s_mount_opt, ERRORS_RO); sbi->s_resuid = le16_to_cpu(es->s_def_resuid); sbi->s_resgid = le16_to_cpu(es->s_def_resgid); @@ -1858,19 +1876,24 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) * so we can safely mount the rest of the filesystem now. */ - root = iget(sb, EXT3_ROOT_INO); - sb->s_root = d_alloc_root(root); - if (!sb->s_root) { + root = ext3_iget(sb, EXT3_ROOT_INO); + if (IS_ERR(root)) { printk(KERN_ERR "EXT3-fs: get root inode failed\n"); - iput(root); + ret = PTR_ERR(root); goto failed_mount4; } if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) { - dput(sb->s_root); - sb->s_root = NULL; + iput(root); printk(KERN_ERR "EXT3-fs: corrupt root inode, run e2fsck\n"); goto failed_mount4; } + sb->s_root = d_alloc_root(root); + if (!sb->s_root) { + printk(KERN_ERR "EXT3-fs: get root dentry failed\n"); + iput(root); + ret = -ENOMEM; + goto failed_mount4; + } ext3_setup_super (sb, es, sb->s_flags & MS_RDONLY); /* @@ -1922,7 +1945,7 @@ out_fail: sb->s_fs_info = NULL; kfree(sbi); lock_kernel(); - return -EINVAL; + return ret; } /* @@ -1958,8 +1981,8 @@ static journal_t *ext3_get_journal(struct super_block *sb, * things happen if we iget() an unused inode, as the subsequent * iput() will try to delete it. */ - journal_inode = iget(sb, journal_inum); - if (!journal_inode) { + journal_inode = ext3_iget(sb, journal_inum); + if (IS_ERR(journal_inode)) { printk(KERN_ERR "EXT3-fs: no journal found.\n"); return NULL; } @@ -1972,7 +1995,7 @@ static journal_t *ext3_get_journal(struct super_block *sb, jbd_debug(2, "Journal inode found at %p: %Ld bytes\n", journal_inode, journal_inode->i_size); - if (is_bad_inode(journal_inode) || !S_ISREG(journal_inode->i_mode)) { + if (!S_ISREG(journal_inode->i_mode)) { printk(KERN_ERR "EXT3-fs: invalid journal inode.\n"); iput(journal_inode); return NULL; @@ -2735,16 +2758,16 @@ static int ext3_quota_on(struct super_block *sb, int type, int format_id, if (err) return err; /* Quotafile not on the same filesystem? */ - if (nd.mnt->mnt_sb != sb) { - path_release(&nd); + if (nd.path.mnt->mnt_sb != sb) { + path_put(&nd.path); return -EXDEV; } /* Quotafile not of fs root? */ - if (nd.dentry->d_parent->d_inode != sb->s_root->d_inode) + if (nd.path.dentry->d_parent->d_inode != sb->s_root->d_inode) printk(KERN_WARNING "EXT3-fs: Quota file not on filesystem root. " "Journalled quota will not work.\n"); - path_release(&nd); + path_put(&nd.path); return vfs_quota_on(sb, type, format_id, path); } diff --git a/fs/ext3/xattr.c b/fs/ext3/xattr.c index 408373819e3..42856541e9a 100644 --- a/fs/ext3/xattr.c +++ b/fs/ext3/xattr.c @@ -492,8 +492,7 @@ ext3_xattr_release_block(handle_t *handle, struct inode *inode, get_bh(bh); ext3_forget(handle, 1, inode, bh, bh->b_blocknr); } else { - BHDR(bh)->h_refcount = cpu_to_le32( - le32_to_cpu(BHDR(bh)->h_refcount) - 1); + le32_add_cpu(&BHDR(bh)->h_refcount, -1); error = ext3_journal_dirty_metadata(handle, bh); if (IS_SYNC(inode)) handle->h_sync = 1; @@ -729,7 +728,7 @@ ext3_xattr_block_set(handle_t *handle, struct inode *inode, ce = NULL; } ea_bdebug(bs->bh, "cloning"); - s->base = kmalloc(bs->bh->b_size, GFP_KERNEL); + s->base = kmalloc(bs->bh->b_size, GFP_NOFS); error = -ENOMEM; if (s->base == NULL) goto cleanup; @@ -741,7 +740,7 @@ ext3_xattr_block_set(handle_t *handle, struct inode *inode, } } else { /* Allocate a buffer where we construct the new block. */ - s->base = kzalloc(sb->s_blocksize, GFP_KERNEL); + s->base = kzalloc(sb->s_blocksize, GFP_NOFS); /* assert(header == s->base) */ error = -ENOMEM; if (s->base == NULL) @@ -780,8 +779,7 @@ inserted: if (error) goto cleanup_dquot; lock_buffer(new_bh); - BHDR(new_bh)->h_refcount = cpu_to_le32(1 + - le32_to_cpu(BHDR(new_bh)->h_refcount)); + le32_add_cpu(&BHDR(new_bh)->h_refcount, 1); ea_bdebug(new_bh, "reusing; refcount now=%d", le32_to_cpu(BHDR(new_bh)->h_refcount)); unlock_buffer(new_bh); @@ -1128,7 +1126,7 @@ ext3_xattr_cache_insert(struct buffer_head *bh) struct mb_cache_entry *ce; int error; - ce = mb_cache_entry_alloc(ext3_xattr_cache); + ce = mb_cache_entry_alloc(ext3_xattr_cache, GFP_NOFS); if (!ce) { ea_bdebug(bh, "out of memory"); return; diff --git a/fs/ext4/Makefile b/fs/ext4/Makefile index ae6e7e502ac..ac6fa8ca0a2 100644 --- a/fs/ext4/Makefile +++ b/fs/ext4/Makefile @@ -6,7 +6,7 @@ obj-$(CONFIG_EXT4DEV_FS) += ext4dev.o ext4dev-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ ioctl.o namei.o super.o symlink.o hash.o resize.o extents.o \ - ext4_jbd2.o + ext4_jbd2.o migrate.o mballoc.o ext4dev-$(CONFIG_EXT4DEV_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o ext4dev-$(CONFIG_EXT4DEV_FS_POSIX_ACL) += acl.o diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 71ee95e534f..0737e05ba3d 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -29,7 +29,7 @@ * Calculate the block group number and offset, given a block number */ void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr, - unsigned long *blockgrpp, ext4_grpblk_t *offsetp) + ext4_group_t *blockgrpp, ext4_grpblk_t *offsetp) { struct ext4_super_block *es = EXT4_SB(sb)->s_es; ext4_grpblk_t offset; @@ -46,7 +46,7 @@ void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr, /* Initializes an uninitialized block bitmap if given, and returns the * number of blocks free in the group. */ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, - int block_group, struct ext4_group_desc *gdp) + ext4_group_t block_group, struct ext4_group_desc *gdp) { unsigned long start; int bit, bit_max; @@ -60,7 +60,7 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, * essentially implementing a per-group read-only flag. */ if (!ext4_group_desc_csum_verify(sbi, block_group, gdp)) { ext4_error(sb, __FUNCTION__, - "Checksum bad for group %u\n", block_group); + "Checksum bad for group %lu\n", block_group); gdp->bg_free_blocks_count = 0; gdp->bg_free_inodes_count = 0; gdp->bg_itable_unused = 0; @@ -153,7 +153,7 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, * group descriptor */ struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb, - unsigned int block_group, + ext4_group_t block_group, struct buffer_head ** bh) { unsigned long group_desc; @@ -164,7 +164,7 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb, if (block_group >= sbi->s_groups_count) { ext4_error (sb, "ext4_get_group_desc", "block_group >= groups_count - " - "block_group = %d, groups_count = %lu", + "block_group = %lu, groups_count = %lu", block_group, sbi->s_groups_count); return NULL; @@ -176,7 +176,7 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb, if (!sbi->s_group_desc[group_desc]) { ext4_error (sb, "ext4_get_group_desc", "Group descriptor not loaded - " - "block_group = %d, group_desc = %lu, desc = %lu", + "block_group = %lu, group_desc = %lu, desc = %lu", block_group, group_desc, offset); return NULL; } @@ -189,18 +189,70 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb, return desc; } +static int ext4_valid_block_bitmap(struct super_block *sb, + struct ext4_group_desc *desc, + unsigned int block_group, + struct buffer_head *bh) +{ + ext4_grpblk_t offset; + ext4_grpblk_t next_zero_bit; + ext4_fsblk_t bitmap_blk; + ext4_fsblk_t group_first_block; + + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG)) { + /* with FLEX_BG, the inode/block bitmaps and itable + * blocks may not be in the group at all + * so the bitmap validation will be skipped for those groups + * or it has to also read the block group where the bitmaps + * are located to verify they are set. + */ + return 1; + } + group_first_block = ext4_group_first_block_no(sb, block_group); + + /* check whether block bitmap block number is set */ + bitmap_blk = ext4_block_bitmap(sb, desc); + offset = bitmap_blk - group_first_block; + if (!ext4_test_bit(offset, bh->b_data)) + /* bad block bitmap */ + goto err_out; + + /* check whether the inode bitmap block number is set */ + bitmap_blk = ext4_inode_bitmap(sb, desc); + offset = bitmap_blk - group_first_block; + if (!ext4_test_bit(offset, bh->b_data)) + /* bad block bitmap */ + goto err_out; + + /* check whether the inode table block number is set */ + bitmap_blk = ext4_inode_table(sb, desc); + offset = bitmap_blk - group_first_block; + next_zero_bit = ext4_find_next_zero_bit(bh->b_data, + offset + EXT4_SB(sb)->s_itb_per_group, + offset); + if (next_zero_bit >= offset + EXT4_SB(sb)->s_itb_per_group) + /* good bitmap for inode tables */ + return 1; + +err_out: + ext4_error(sb, __FUNCTION__, + "Invalid block bitmap - " + "block_group = %d, block = %llu", + block_group, bitmap_blk); + return 0; +} /** * read_block_bitmap() * @sb: super block * @block_group: given block group * - * Read the bitmap for a given block_group, reading into the specified - * slot in the superblock's bitmap cache. + * Read the bitmap for a given block_group,and validate the + * bits for block/inode/inode tables are set in the bitmaps * * Return buffer_head on success or NULL in case of failure. */ struct buffer_head * -read_block_bitmap(struct super_block *sb, unsigned int block_group) +read_block_bitmap(struct super_block *sb, ext4_group_t block_group) { struct ext4_group_desc * desc; struct buffer_head * bh = NULL; @@ -210,25 +262,36 @@ read_block_bitmap(struct super_block *sb, unsigned int block_group) if (!desc) return NULL; bitmap_blk = ext4_block_bitmap(sb, desc); + bh = sb_getblk(sb, bitmap_blk); + if (unlikely(!bh)) { + ext4_error(sb, __FUNCTION__, + "Cannot read block bitmap - " + "block_group = %d, block_bitmap = %llu", + (int)block_group, (unsigned long long)bitmap_blk); + return NULL; + } + if (bh_uptodate_or_lock(bh)) + return bh; + if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) { - bh = sb_getblk(sb, bitmap_blk); - if (!buffer_uptodate(bh)) { - lock_buffer(bh); - if (!buffer_uptodate(bh)) { - ext4_init_block_bitmap(sb, bh, block_group, - desc); - set_buffer_uptodate(bh); - } - unlock_buffer(bh); - } - } else { - bh = sb_bread(sb, bitmap_blk); + ext4_init_block_bitmap(sb, bh, block_group, desc); + set_buffer_uptodate(bh); + unlock_buffer(bh); + return bh; } - if (!bh) - ext4_error (sb, __FUNCTION__, + if (bh_submit_read(bh) < 0) { + put_bh(bh); + ext4_error(sb, __FUNCTION__, "Cannot read block bitmap - " "block_group = %d, block_bitmap = %llu", - block_group, bitmap_blk); + (int)block_group, (unsigned long long)bitmap_blk); + return NULL; + } + if (!ext4_valid_block_bitmap(sb, desc, block_group, bh)) { + put_bh(bh); + return NULL; + } + return bh; } /* @@ -320,7 +383,7 @@ restart: */ static int goal_in_my_reservation(struct ext4_reserve_window *rsv, ext4_grpblk_t grp_goal, - unsigned int group, struct super_block * sb) + ext4_group_t group, struct super_block *sb) { ext4_fsblk_t group_first_block, group_last_block; @@ -463,7 +526,7 @@ static inline int rsv_is_empty(struct ext4_reserve_window *rsv) * when setting the reservation window size through ioctl before the file * is open for write (needs block allocation). * - * Needs truncate_mutex protection prior to call this function. + * Needs down_write(i_data_sem) protection prior to call this function. */ void ext4_init_block_alloc_info(struct inode *inode) { @@ -514,6 +577,8 @@ void ext4_discard_reservation(struct inode *inode) struct ext4_reserve_window_node *rsv; spinlock_t *rsv_lock = &EXT4_SB(inode->i_sb)->s_rsv_window_lock; + ext4_mb_discard_inode_preallocations(inode); + if (!block_i) return; @@ -540,7 +605,7 @@ void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb, { struct buffer_head *bitmap_bh = NULL; struct buffer_head *gd_bh; - unsigned long block_group; + ext4_group_t block_group; ext4_grpblk_t bit; unsigned long i; unsigned long overflow; @@ -587,11 +652,13 @@ do_more: in_range(ext4_inode_bitmap(sb, desc), block, count) || in_range(block, ext4_inode_table(sb, desc), sbi->s_itb_per_group) || in_range(block + count - 1, ext4_inode_table(sb, desc), - sbi->s_itb_per_group)) + sbi->s_itb_per_group)) { ext4_error (sb, "ext4_free_blocks", "Freeing blocks in system zones - " "Block = %llu, count = %lu", block, count); + goto error_return; + } /* * We are about to start releasing blocks in the bitmap, @@ -720,19 +787,29 @@ error_return: * @inode: inode * @block: start physical block to free * @count: number of blocks to count + * @metadata: Are these metadata blocks */ void ext4_free_blocks(handle_t *handle, struct inode *inode, - ext4_fsblk_t block, unsigned long count) + ext4_fsblk_t block, unsigned long count, + int metadata) { struct super_block * sb; unsigned long dquot_freed_blocks; + /* this isn't the right place to decide whether block is metadata + * inode.c/extents.c knows better, but for safety ... */ + if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode) || + ext4_should_journal_data(inode)) + metadata = 1; + sb = inode->i_sb; - if (!sb) { - printk ("ext4_free_blocks: nonexistent device"); - return; - } - ext4_free_blocks_sb(handle, sb, block, count, &dquot_freed_blocks); + + if (!test_opt(sb, MBALLOC) || !EXT4_SB(sb)->s_group_info) + ext4_free_blocks_sb(handle, sb, block, count, + &dquot_freed_blocks); + else + ext4_mb_free_blocks(handle, inode, block, count, + metadata, &dquot_freed_blocks); if (dquot_freed_blocks) DQUOT_FREE_BLOCK(inode, dquot_freed_blocks); return; @@ -920,9 +997,10 @@ claim_block(spinlock_t *lock, ext4_grpblk_t block, struct buffer_head *bh) * ext4_journal_release_buffer(), else we'll run out of credits. */ static ext4_grpblk_t -ext4_try_to_allocate(struct super_block *sb, handle_t *handle, int group, - struct buffer_head *bitmap_bh, ext4_grpblk_t grp_goal, - unsigned long *count, struct ext4_reserve_window *my_rsv) +ext4_try_to_allocate(struct super_block *sb, handle_t *handle, + ext4_group_t group, struct buffer_head *bitmap_bh, + ext4_grpblk_t grp_goal, unsigned long *count, + struct ext4_reserve_window *my_rsv) { ext4_fsblk_t group_first_block; ext4_grpblk_t start, end; @@ -1156,7 +1234,7 @@ static int find_next_reservable_window( */ static int alloc_new_reservation(struct ext4_reserve_window_node *my_rsv, ext4_grpblk_t grp_goal, struct super_block *sb, - unsigned int group, struct buffer_head *bitmap_bh) + ext4_group_t group, struct buffer_head *bitmap_bh) { struct ext4_reserve_window_node *search_head; ext4_fsblk_t group_first_block, group_end_block, start_block; @@ -1354,7 +1432,7 @@ static void try_to_extend_reservation(struct ext4_reserve_window_node *my_rsv, */ static ext4_grpblk_t ext4_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle, - unsigned int group, struct buffer_head *bitmap_bh, + ext4_group_t group, struct buffer_head *bitmap_bh, ext4_grpblk_t grp_goal, struct ext4_reserve_window_node * my_rsv, unsigned long *count, int *errp) @@ -1510,7 +1588,7 @@ int ext4_should_retry_alloc(struct super_block *sb, int *retries) } /** - * ext4_new_blocks() -- core block(s) allocation function + * ext4_new_blocks_old() -- core block(s) allocation function * @handle: handle to this transaction * @inode: file inode * @goal: given target block(filesystem wide) @@ -1523,17 +1601,17 @@ int ext4_should_retry_alloc(struct super_block *sb, int *retries) * any specific goal block. * */ -ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode, +ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode, ext4_fsblk_t goal, unsigned long *count, int *errp) { struct buffer_head *bitmap_bh = NULL; struct buffer_head *gdp_bh; - unsigned long group_no; - int goal_group; + ext4_group_t group_no; + ext4_group_t goal_group; ext4_grpblk_t grp_target_blk; /* blockgroup relative goal block */ ext4_grpblk_t grp_alloc_blk; /* blockgroup-relative allocated block*/ ext4_fsblk_t ret_block; /* filesyetem-wide allocated block */ - int bgi; /* blockgroup iteration index */ + ext4_group_t bgi; /* blockgroup iteration index */ int fatal = 0, err; int performed_allocation = 0; ext4_grpblk_t free_blocks; /* number of free blocks in a group */ @@ -1544,10 +1622,7 @@ ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode, struct ext4_reserve_window_node *my_rsv = NULL; struct ext4_block_alloc_info *block_i; unsigned short windowsz = 0; -#ifdef EXT4FS_DEBUG - static int goal_hits, goal_attempts; -#endif - unsigned long ngroups; + ext4_group_t ngroups; unsigned long num = *count; *errp = -ENOSPC; @@ -1567,7 +1642,7 @@ ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode, sbi = EXT4_SB(sb); es = EXT4_SB(sb)->s_es; - ext4_debug("goal=%lu.\n", goal); + ext4_debug("goal=%llu.\n", goal); /* * Allocate a block from reservation only when * filesystem is mounted with reservation(default,-o reservation), and @@ -1625,7 +1700,7 @@ retry_alloc: /* * Now search the rest of the groups. We assume that - * i and gdp correctly point to the last group visited. + * group_no and gdp correctly point to the last group visited. */ for (bgi = 0; bgi < ngroups; bgi++) { group_no++; @@ -1677,7 +1752,7 @@ retry_alloc: allocated: - ext4_debug("using block group %d(%d)\n", + ext4_debug("using block group %lu(%d)\n", group_no, gdp->bg_free_blocks_count); BUFFER_TRACE(gdp_bh, "get_write_access"); @@ -1692,11 +1767,13 @@ allocated: in_range(ret_block, ext4_inode_table(sb, gdp), EXT4_SB(sb)->s_itb_per_group) || in_range(ret_block + num - 1, ext4_inode_table(sb, gdp), - EXT4_SB(sb)->s_itb_per_group)) + EXT4_SB(sb)->s_itb_per_group)) { ext4_error(sb, "ext4_new_block", "Allocating block in system zone - " "blocks from %llu, length %lu", ret_block, num); + goto out; + } performed_allocation = 1; @@ -1743,9 +1820,6 @@ allocated: * list of some description. We don't know in advance whether * the caller wants to use it as metadata or data. */ - ext4_debug("allocating block %lu. Goal hits %d of %d.\n", - ret_block, goal_hits, goal_attempts); - spin_lock(sb_bgl_lock(sbi, group_no)); if (gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT); @@ -1787,13 +1861,46 @@ out: } ext4_fsblk_t ext4_new_block(handle_t *handle, struct inode *inode, - ext4_fsblk_t goal, int *errp) + ext4_fsblk_t goal, int *errp) +{ + struct ext4_allocation_request ar; + ext4_fsblk_t ret; + + if (!test_opt(inode->i_sb, MBALLOC)) { + unsigned long count = 1; + ret = ext4_new_blocks_old(handle, inode, goal, &count, errp); + return ret; + } + + memset(&ar, 0, sizeof(ar)); + ar.inode = inode; + ar.goal = goal; + ar.len = 1; + ret = ext4_mb_new_blocks(handle, &ar, errp); + return ret; +} + +ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode, + ext4_fsblk_t goal, unsigned long *count, int *errp) { - unsigned long count = 1; + struct ext4_allocation_request ar; + ext4_fsblk_t ret; - return ext4_new_blocks(handle, inode, goal, &count, errp); + if (!test_opt(inode->i_sb, MBALLOC)) { + ret = ext4_new_blocks_old(handle, inode, goal, count, errp); + return ret; + } + + memset(&ar, 0, sizeof(ar)); + ar.inode = inode; + ar.goal = goal; + ar.len = *count; + ret = ext4_mb_new_blocks(handle, &ar, errp); + *count = ar.len; + return ret; } + /** * ext4_count_free_blocks() -- count filesystem free blocks * @sb: superblock @@ -1804,8 +1911,8 @@ ext4_fsblk_t ext4_count_free_blocks(struct super_block *sb) { ext4_fsblk_t desc_count; struct ext4_group_desc *gdp; - int i; - unsigned long ngroups = EXT4_SB(sb)->s_groups_count; + ext4_group_t i; + ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count; #ifdef EXT4FS_DEBUG struct ext4_super_block *es; ext4_fsblk_t bitmap_count; @@ -1829,14 +1936,14 @@ ext4_fsblk_t ext4_count_free_blocks(struct super_block *sb) continue; x = ext4_count_free(bitmap_bh, sb->s_blocksize); - printk("group %d: stored = %d, counted = %lu\n", + printk(KERN_DEBUG "group %lu: stored = %d, counted = %lu\n", i, le16_to_cpu(gdp->bg_free_blocks_count), x); bitmap_count += x; } brelse(bitmap_bh); printk("ext4_count_free_blocks: stored = %llu" ", computed = %llu, %llu\n", - EXT4_FREE_BLOCKS_COUNT(es), + ext4_free_blocks_count(es), desc_count, bitmap_count); return bitmap_count; #else @@ -1853,7 +1960,7 @@ ext4_fsblk_t ext4_count_free_blocks(struct super_block *sb) #endif } -static inline int test_root(int a, int b) +static inline int test_root(ext4_group_t a, int b) { int num = b; @@ -1862,7 +1969,7 @@ static inline int test_root(int a, int b) return num == a; } -static int ext4_group_sparse(int group) +static int ext4_group_sparse(ext4_group_t group) { if (group <= 1) return 1; @@ -1880,7 +1987,7 @@ static int ext4_group_sparse(int group) * Return the number of blocks used by the superblock (primary or backup) * in this group. Currently this will be only 0 or 1. */ -int ext4_bg_has_super(struct super_block *sb, int group) +int ext4_bg_has_super(struct super_block *sb, ext4_group_t group) { if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER) && @@ -1889,24 +1996,22 @@ int ext4_bg_has_super(struct super_block *sb, int group) return 1; } -static unsigned long ext4_bg_num_gdb_meta(struct super_block *sb, int group) +static unsigned long ext4_bg_num_gdb_meta(struct super_block *sb, + ext4_group_t group) { unsigned long metagroup = group / EXT4_DESC_PER_BLOCK(sb); - unsigned long first = metagroup * EXT4_DESC_PER_BLOCK(sb); - unsigned long last = first + EXT4_DESC_PER_BLOCK(sb) - 1; + ext4_group_t first = metagroup * EXT4_DESC_PER_BLOCK(sb); + ext4_group_t last = first + EXT4_DESC_PER_BLOCK(sb) - 1; if (group == first || group == first + 1 || group == last) return 1; return 0; } -static unsigned long ext4_bg_num_gdb_nometa(struct super_block *sb, int group) +static unsigned long ext4_bg_num_gdb_nometa(struct super_block *sb, + ext4_group_t group) { - if (EXT4_HAS_RO_COMPAT_FEATURE(sb, - EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER) && - !ext4_group_sparse(group)) - return 0; - return EXT4_SB(sb)->s_gdb_count; + return ext4_bg_has_super(sb, group) ? EXT4_SB(sb)->s_gdb_count : 0; } /** @@ -1918,7 +2023,7 @@ static unsigned long ext4_bg_num_gdb_nometa(struct super_block *sb, int group) * (primary or backup) in this group. In the future there may be a * different number of descriptor blocks in each group. */ -unsigned long ext4_bg_num_gdb(struct super_block *sb, int group) +unsigned long ext4_bg_num_gdb(struct super_block *sb, ext4_group_t group) { unsigned long first_meta_bg = le32_to_cpu(EXT4_SB(sb)->s_es->s_first_meta_bg); diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index f612bef9831..2c23bade9aa 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -46,7 +46,7 @@ const struct file_operations ext4_dir_operations = { #ifdef CONFIG_COMPAT .compat_ioctl = ext4_compat_ioctl, #endif - .fsync = ext4_sync_file, /* BKL held */ + .fsync = ext4_sync_file, .release = ext4_release_dir, }; @@ -67,7 +67,7 @@ int ext4_check_dir_entry (const char * function, struct inode * dir, unsigned long offset) { const char * error_msg = NULL; - const int rlen = le16_to_cpu(de->rec_len); + const int rlen = ext4_rec_len_from_disk(de->rec_len); if (rlen < EXT4_DIR_REC_LEN(1)) error_msg = "rec_len is smaller than minimal"; @@ -124,7 +124,7 @@ static int ext4_readdir(struct file * filp, offset = filp->f_pos & (sb->s_blocksize - 1); while (!error && !stored && filp->f_pos < inode->i_size) { - unsigned long blk = filp->f_pos >> EXT4_BLOCK_SIZE_BITS(sb); + ext4_lblk_t blk = filp->f_pos >> EXT4_BLOCK_SIZE_BITS(sb); struct buffer_head map_bh; struct buffer_head *bh = NULL; @@ -172,10 +172,10 @@ revalidate: * least that it is non-zero. A * failure will be detected in the * dirent test below. */ - if (le16_to_cpu(de->rec_len) < - EXT4_DIR_REC_LEN(1)) + if (ext4_rec_len_from_disk(de->rec_len) + < EXT4_DIR_REC_LEN(1)) break; - i += le16_to_cpu(de->rec_len); + i += ext4_rec_len_from_disk(de->rec_len); } offset = i; filp->f_pos = (filp->f_pos & ~(sb->s_blocksize - 1)) @@ -197,7 +197,7 @@ revalidate: ret = stored; goto out; } - offset += le16_to_cpu(de->rec_len); + offset += ext4_rec_len_from_disk(de->rec_len); if (le32_to_cpu(de->inode)) { /* We might block in the next section * if the data destination is @@ -219,7 +219,7 @@ revalidate: goto revalidate; stored ++; } - filp->f_pos += le16_to_cpu(de->rec_len); + filp->f_pos += ext4_rec_len_from_disk(de->rec_len); } offset = 0; brelse (bh); diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 85287742f2a..9ae6e67090c 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -61,7 +61,7 @@ static ext4_fsblk_t ext_pblock(struct ext4_extent *ex) * idx_pblock: * combine low and high parts of a leaf physical block number into ext4_fsblk_t */ -static ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix) +ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix) { ext4_fsblk_t block; @@ -75,7 +75,7 @@ static ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix) * stores a large physical block number into an extent struct, * breaking it into parts */ -static void ext4_ext_store_pblock(struct ext4_extent *ex, ext4_fsblk_t pb) +void ext4_ext_store_pblock(struct ext4_extent *ex, ext4_fsblk_t pb) { ex->ee_start_lo = cpu_to_le32((unsigned long) (pb & 0xffffffff)); ex->ee_start_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff); @@ -144,10 +144,11 @@ static int ext4_ext_dirty(handle_t *handle, struct inode *inode, static ext4_fsblk_t ext4_ext_find_goal(struct inode *inode, struct ext4_ext_path *path, - ext4_fsblk_t block) + ext4_lblk_t block) { struct ext4_inode_info *ei = EXT4_I(inode); ext4_fsblk_t bg_start; + ext4_fsblk_t last_block; ext4_grpblk_t colour; int depth; @@ -169,8 +170,13 @@ static ext4_fsblk_t ext4_ext_find_goal(struct inode *inode, /* OK. use inode's group */ bg_start = (ei->i_block_group * EXT4_BLOCKS_PER_GROUP(inode->i_sb)) + le32_to_cpu(EXT4_SB(inode->i_sb)->s_es->s_first_data_block); - colour = (current->pid % 16) * + last_block = ext4_blocks_count(EXT4_SB(inode->i_sb)->s_es) - 1; + + if (bg_start + EXT4_BLOCKS_PER_GROUP(inode->i_sb) <= last_block) + colour = (current->pid % 16) * (EXT4_BLOCKS_PER_GROUP(inode->i_sb) / 16); + else + colour = (current->pid % 16) * ((last_block - bg_start) / 16); return bg_start + colour + block; } @@ -349,7 +355,7 @@ static void ext4_ext_show_leaf(struct inode *inode, struct ext4_ext_path *path) #define ext4_ext_show_leaf(inode,path) #endif -static void ext4_ext_drop_refs(struct ext4_ext_path *path) +void ext4_ext_drop_refs(struct ext4_ext_path *path) { int depth = path->p_depth; int i; @@ -367,13 +373,14 @@ static void ext4_ext_drop_refs(struct ext4_ext_path *path) * the header must be checked before calling this */ static void -ext4_ext_binsearch_idx(struct inode *inode, struct ext4_ext_path *path, int block) +ext4_ext_binsearch_idx(struct inode *inode, + struct ext4_ext_path *path, ext4_lblk_t block) { struct ext4_extent_header *eh = path->p_hdr; struct ext4_extent_idx *r, *l, *m; - ext_debug("binsearch for %d(idx): ", block); + ext_debug("binsearch for %u(idx): ", block); l = EXT_FIRST_INDEX(eh) + 1; r = EXT_LAST_INDEX(eh); @@ -425,7 +432,8 @@ ext4_ext_binsearch_idx(struct inode *inode, struct ext4_ext_path *path, int bloc * the header must be checked before calling this */ static void -ext4_ext_binsearch(struct inode *inode, struct ext4_ext_path *path, int block) +ext4_ext_binsearch(struct inode *inode, + struct ext4_ext_path *path, ext4_lblk_t block) { struct ext4_extent_header *eh = path->p_hdr; struct ext4_extent *r, *l, *m; @@ -438,7 +446,7 @@ ext4_ext_binsearch(struct inode *inode, struct ext4_ext_path *path, int block) return; } - ext_debug("binsearch for %d: ", block); + ext_debug("binsearch for %u: ", block); l = EXT_FIRST_EXTENT(eh) + 1; r = EXT_LAST_EXTENT(eh); @@ -494,7 +502,8 @@ int ext4_ext_tree_init(handle_t *handle, struct inode *inode) } struct ext4_ext_path * -ext4_ext_find_extent(struct inode *inode, int block, struct ext4_ext_path *path) +ext4_ext_find_extent(struct inode *inode, ext4_lblk_t block, + struct ext4_ext_path *path) { struct ext4_extent_header *eh; struct buffer_head *bh; @@ -763,7 +772,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, while (k--) { oldblock = newblock; newblock = ablocks[--a]; - bh = sb_getblk(inode->i_sb, (ext4_fsblk_t)newblock); + bh = sb_getblk(inode->i_sb, newblock); if (!bh) { err = -EIO; goto cleanup; @@ -783,9 +792,8 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, fidx->ei_block = border; ext4_idx_store_pblock(fidx, oldblock); - ext_debug("int.index at %d (block %llu): %lu -> %llu\n", i, - newblock, (unsigned long) le32_to_cpu(border), - oldblock); + ext_debug("int.index at %d (block %llu): %u -> %llu\n", + i, newblock, le32_to_cpu(border), oldblock); /* copy indexes */ m = 0; path[i].p_idx++; @@ -851,7 +859,7 @@ cleanup: for (i = 0; i < depth; i++) { if (!ablocks[i]) continue; - ext4_free_blocks(handle, inode, ablocks[i], 1); + ext4_free_blocks(handle, inode, ablocks[i], 1, 1); } } kfree(ablocks); @@ -979,8 +987,8 @@ repeat: /* refill path */ ext4_ext_drop_refs(path); path = ext4_ext_find_extent(inode, - le32_to_cpu(newext->ee_block), - path); + (ext4_lblk_t)le32_to_cpu(newext->ee_block), + path); if (IS_ERR(path)) err = PTR_ERR(path); } else { @@ -992,8 +1000,8 @@ repeat: /* refill path */ ext4_ext_drop_refs(path); path = ext4_ext_find_extent(inode, - le32_to_cpu(newext->ee_block), - path); + (ext4_lblk_t)le32_to_cpu(newext->ee_block), + path); if (IS_ERR(path)) { err = PTR_ERR(path); goto out; @@ -1015,13 +1023,157 @@ out: } /* + * search the closest allocated block to the left for *logical + * and returns it at @logical + it's physical address at @phys + * if *logical is the smallest allocated block, the function + * returns 0 at @phys + * return value contains 0 (success) or error code + */ +int +ext4_ext_search_left(struct inode *inode, struct ext4_ext_path *path, + ext4_lblk_t *logical, ext4_fsblk_t *phys) +{ + struct ext4_extent_idx *ix; + struct ext4_extent *ex; + int depth, ee_len; + + BUG_ON(path == NULL); + depth = path->p_depth; + *phys = 0; + + if (depth == 0 && path->p_ext == NULL) + return 0; + + /* usually extent in the path covers blocks smaller + * then *logical, but it can be that extent is the + * first one in the file */ + + ex = path[depth].p_ext; + ee_len = ext4_ext_get_actual_len(ex); + if (*logical < le32_to_cpu(ex->ee_block)) { + BUG_ON(EXT_FIRST_EXTENT(path[depth].p_hdr) != ex); + while (--depth >= 0) { + ix = path[depth].p_idx; + BUG_ON(ix != EXT_FIRST_INDEX(path[depth].p_hdr)); + } + return 0; + } + + BUG_ON(*logical < (le32_to_cpu(ex->ee_block) + ee_len)); + + *logical = le32_to_cpu(ex->ee_block) + ee_len - 1; + *phys = ext_pblock(ex) + ee_len - 1; + return 0; +} + +/* + * search the closest allocated block to the right for *logical + * and returns it at @logical + it's physical address at @phys + * if *logical is the smallest allocated block, the function + * returns 0 at @phys + * return value contains 0 (success) or error code + */ +int +ext4_ext_search_right(struct inode *inode, struct ext4_ext_path *path, + ext4_lblk_t *logical, ext4_fsblk_t *phys) +{ + struct buffer_head *bh = NULL; + struct ext4_extent_header *eh; + struct ext4_extent_idx *ix; + struct ext4_extent *ex; + ext4_fsblk_t block; + int depth, ee_len; + + BUG_ON(path == NULL); + depth = path->p_depth; + *phys = 0; + + if (depth == 0 && path->p_ext == NULL) + return 0; + + /* usually extent in the path covers blocks smaller + * then *logical, but it can be that extent is the + * first one in the file */ + + ex = path[depth].p_ext; + ee_len = ext4_ext_get_actual_len(ex); + if (*logical < le32_to_cpu(ex->ee_block)) { + BUG_ON(EXT_FIRST_EXTENT(path[depth].p_hdr) != ex); + while (--depth >= 0) { + ix = path[depth].p_idx; + BUG_ON(ix != EXT_FIRST_INDEX(path[depth].p_hdr)); + } + *logical = le32_to_cpu(ex->ee_block); + *phys = ext_pblock(ex); + return 0; + } + + BUG_ON(*logical < (le32_to_cpu(ex->ee_block) + ee_len)); + + if (ex != EXT_LAST_EXTENT(path[depth].p_hdr)) { + /* next allocated block in this leaf */ + ex++; + *logical = le32_to_cpu(ex->ee_block); + *phys = ext_pblock(ex); + return 0; + } + + /* go up and search for index to the right */ + while (--depth >= 0) { + ix = path[depth].p_idx; + if (ix != EXT_LAST_INDEX(path[depth].p_hdr)) + break; + } + + if (depth < 0) { + /* we've gone up to the root and + * found no index to the right */ + return 0; + } + + /* we've found index to the right, let's + * follow it and find the closest allocated + * block to the right */ + ix++; + block = idx_pblock(ix); + while (++depth < path->p_depth) { + bh = sb_bread(inode->i_sb, block); + if (bh == NULL) + return -EIO; + eh = ext_block_hdr(bh); + if (ext4_ext_check_header(inode, eh, depth)) { + put_bh(bh); + return -EIO; + } + ix = EXT_FIRST_INDEX(eh); + block = idx_pblock(ix); + put_bh(bh); + } + + bh = sb_bread(inode->i_sb, block); + if (bh == NULL) + return -EIO; + eh = ext_block_hdr(bh); + if (ext4_ext_check_header(inode, eh, path->p_depth - depth)) { + put_bh(bh); + return -EIO; + } + ex = EXT_FIRST_EXTENT(eh); + *logical = le32_to_cpu(ex->ee_block); + *phys = ext_pblock(ex); + put_bh(bh); + return 0; + +} + +/* * ext4_ext_next_allocated_block: * returns allocated block in subsequent extent or EXT_MAX_BLOCK. * NOTE: it considers block number from index entry as * allocated block. Thus, index entries have to be consistent * with leaves. */ -static unsigned long +static ext4_lblk_t ext4_ext_next_allocated_block(struct ext4_ext_path *path) { int depth; @@ -1054,7 +1206,7 @@ ext4_ext_next_allocated_block(struct ext4_ext_path *path) * ext4_ext_next_leaf_block: * returns first allocated block from next leaf or EXT_MAX_BLOCK */ -static unsigned ext4_ext_next_leaf_block(struct inode *inode, +static ext4_lblk_t ext4_ext_next_leaf_block(struct inode *inode, struct ext4_ext_path *path) { int depth; @@ -1072,7 +1224,8 @@ static unsigned ext4_ext_next_leaf_block(struct inode *inode, while (depth >= 0) { if (path[depth].p_idx != EXT_LAST_INDEX(path[depth].p_hdr)) - return le32_to_cpu(path[depth].p_idx[1].ei_block); + return (ext4_lblk_t) + le32_to_cpu(path[depth].p_idx[1].ei_block); depth--; } @@ -1085,7 +1238,7 @@ static unsigned ext4_ext_next_leaf_block(struct inode *inode, * then we have to correct all indexes above. * TODO: do we need to correct tree in all cases? */ -int ext4_ext_correct_indexes(handle_t *handle, struct inode *inode, +static int ext4_ext_correct_indexes(handle_t *handle, struct inode *inode, struct ext4_ext_path *path) { struct ext4_extent_header *eh; @@ -1171,7 +1324,7 @@ ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1, if (ext1_ee_len + ext2_ee_len > max_len) return 0; #ifdef AGGRESSIVE_TEST - if (le16_to_cpu(ex1->ee_len) >= 4) + if (ext1_ee_len >= 4) return 0; #endif @@ -1239,7 +1392,7 @@ unsigned int ext4_ext_check_overlap(struct inode *inode, struct ext4_extent *newext, struct ext4_ext_path *path) { - unsigned long b1, b2; + ext4_lblk_t b1, b2; unsigned int depth, len1; unsigned int ret = 0; @@ -1260,7 +1413,7 @@ unsigned int ext4_ext_check_overlap(struct inode *inode, goto out; } - /* check for wrap through zero */ + /* check for wrap through zero on extent logical start block*/ if (b1 + len1 < b1) { len1 = EXT_MAX_BLOCK - b1; newext->ee_len = cpu_to_le16(len1); @@ -1290,7 +1443,8 @@ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode, struct ext4_extent *ex, *fex; struct ext4_extent *nearex; /* nearest extent */ struct ext4_ext_path *npath = NULL; - int depth, len, err, next; + int depth, len, err; + ext4_lblk_t next; unsigned uninitialized = 0; BUG_ON(ext4_ext_get_actual_len(newext) == 0); @@ -1435,114 +1589,8 @@ cleanup: return err; } -int ext4_ext_walk_space(struct inode *inode, unsigned long block, - unsigned long num, ext_prepare_callback func, - void *cbdata) -{ - struct ext4_ext_path *path = NULL; - struct ext4_ext_cache cbex; - struct ext4_extent *ex; - unsigned long next, start = 0, end = 0; - unsigned long last = block + num; - int depth, exists, err = 0; - - BUG_ON(func == NULL); - BUG_ON(inode == NULL); - - while (block < last && block != EXT_MAX_BLOCK) { - num = last - block; - /* find extent for this block */ - path = ext4_ext_find_extent(inode, block, path); - if (IS_ERR(path)) { - err = PTR_ERR(path); - path = NULL; - break; - } - - depth = ext_depth(inode); - BUG_ON(path[depth].p_hdr == NULL); - ex = path[depth].p_ext; - next = ext4_ext_next_allocated_block(path); - - exists = 0; - if (!ex) { - /* there is no extent yet, so try to allocate - * all requested space */ - start = block; - end = block + num; - } else if (le32_to_cpu(ex->ee_block) > block) { - /* need to allocate space before found extent */ - start = block; - end = le32_to_cpu(ex->ee_block); - if (block + num < end) - end = block + num; - } else if (block >= le32_to_cpu(ex->ee_block) - + ext4_ext_get_actual_len(ex)) { - /* need to allocate space after found extent */ - start = block; - end = block + num; - if (end >= next) - end = next; - } else if (block >= le32_to_cpu(ex->ee_block)) { - /* - * some part of requested space is covered - * by found extent - */ - start = block; - end = le32_to_cpu(ex->ee_block) - + ext4_ext_get_actual_len(ex); - if (block + num < end) - end = block + num; - exists = 1; - } else { - BUG(); - } - BUG_ON(end <= start); - - if (!exists) { - cbex.ec_block = start; - cbex.ec_len = end - start; - cbex.ec_start = 0; - cbex.ec_type = EXT4_EXT_CACHE_GAP; - } else { - cbex.ec_block = le32_to_cpu(ex->ee_block); - cbex.ec_len = ext4_ext_get_actual_len(ex); - cbex.ec_start = ext_pblock(ex); - cbex.ec_type = EXT4_EXT_CACHE_EXTENT; - } - - BUG_ON(cbex.ec_len == 0); - err = func(inode, path, &cbex, cbdata); - ext4_ext_drop_refs(path); - - if (err < 0) - break; - if (err == EXT_REPEAT) - continue; - else if (err == EXT_BREAK) { - err = 0; - break; - } - - if (ext_depth(inode) != depth) { - /* depth was changed. we have to realloc path */ - kfree(path); - path = NULL; - } - - block = cbex.ec_block + cbex.ec_len; - } - - if (path) { - ext4_ext_drop_refs(path); - kfree(path); - } - - return err; -} - static void -ext4_ext_put_in_cache(struct inode *inode, __u32 block, +ext4_ext_put_in_cache(struct inode *inode, ext4_lblk_t block, __u32 len, ext4_fsblk_t start, int type) { struct ext4_ext_cache *cex; @@ -1561,10 +1609,11 @@ ext4_ext_put_in_cache(struct inode *inode, __u32 block, */ static void ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path, - unsigned long block) + ext4_lblk_t block) { int depth = ext_depth(inode); - unsigned long lblock, len; + unsigned long len; + ext4_lblk_t lblock; struct ext4_extent *ex; ex = path[depth].p_ext; @@ -1576,32 +1625,34 @@ ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path, } else if (block < le32_to_cpu(ex->ee_block)) { lblock = block; len = le32_to_cpu(ex->ee_block) - block; - ext_debug("cache gap(before): %lu [%lu:%lu]", - (unsigned long) block, - (unsigned long) le32_to_cpu(ex->ee_block), - (unsigned long) ext4_ext_get_actual_len(ex)); + ext_debug("cache gap(before): %u [%u:%u]", + block, + le32_to_cpu(ex->ee_block), + ext4_ext_get_actual_len(ex)); } else if (block >= le32_to_cpu(ex->ee_block) + ext4_ext_get_actual_len(ex)) { + ext4_lblk_t next; lblock = le32_to_cpu(ex->ee_block) + ext4_ext_get_actual_len(ex); - len = ext4_ext_next_allocated_block(path); - ext_debug("cache gap(after): [%lu:%lu] %lu", - (unsigned long) le32_to_cpu(ex->ee_block), - (unsigned long) ext4_ext_get_actual_len(ex), - (unsigned long) block); - BUG_ON(len == lblock); - len = len - lblock; + + next = ext4_ext_next_allocated_block(path); + ext_debug("cache gap(after): [%u:%u] %u", + le32_to_cpu(ex->ee_block), + ext4_ext_get_actual_len(ex), + block); + BUG_ON(next == lblock); + len = next - lblock; } else { lblock = len = 0; BUG(); } - ext_debug(" -> %lu:%lu\n", (unsigned long) lblock, len); + ext_debug(" -> %u:%lu\n", lblock, len); ext4_ext_put_in_cache(inode, lblock, len, 0, EXT4_EXT_CACHE_GAP); } static int -ext4_ext_in_cache(struct inode *inode, unsigned long block, +ext4_ext_in_cache(struct inode *inode, ext4_lblk_t block, struct ext4_extent *ex) { struct ext4_ext_cache *cex; @@ -1618,11 +1669,9 @@ ext4_ext_in_cache(struct inode *inode, unsigned long block, ex->ee_block = cpu_to_le32(cex->ec_block); ext4_ext_store_pblock(ex, cex->ec_start); ex->ee_len = cpu_to_le16(cex->ec_len); - ext_debug("%lu cached by %lu:%lu:%llu\n", - (unsigned long) block, - (unsigned long) cex->ec_block, - (unsigned long) cex->ec_len, - cex->ec_start); + ext_debug("%u cached by %u:%u:%llu\n", + block, + cex->ec_block, cex->ec_len, cex->ec_start); return cex->ec_type; } @@ -1636,7 +1685,7 @@ ext4_ext_in_cache(struct inode *inode, unsigned long block, * It's used in truncate case only, thus all requests are for * last index in the block only. */ -int ext4_ext_rm_idx(handle_t *handle, struct inode *inode, +static int ext4_ext_rm_idx(handle_t *handle, struct inode *inode, struct ext4_ext_path *path) { struct buffer_head *bh; @@ -1657,7 +1706,7 @@ int ext4_ext_rm_idx(handle_t *handle, struct inode *inode, ext_debug("index is empty, remove it, free block %llu\n", leaf); bh = sb_find_get_block(inode->i_sb, leaf); ext4_forget(handle, 1, inode, bh, leaf); - ext4_free_blocks(handle, inode, leaf, 1); + ext4_free_blocks(handle, inode, leaf, 1, 1); return err; } @@ -1666,7 +1715,7 @@ int ext4_ext_rm_idx(handle_t *handle, struct inode *inode, * This routine returns max. credits that the extent tree can consume. * It should be OK for low-performance paths like ->writepage() * To allow many writing processes to fit into a single transaction, - * the caller should calculate credits under truncate_mutex and + * the caller should calculate credits under i_data_sem and * pass the actual path. */ int ext4_ext_calc_credits_for_insert(struct inode *inode, @@ -1714,12 +1763,14 @@ int ext4_ext_calc_credits_for_insert(struct inode *inode, static int ext4_remove_blocks(handle_t *handle, struct inode *inode, struct ext4_extent *ex, - unsigned long from, unsigned long to) + ext4_lblk_t from, ext4_lblk_t to) { struct buffer_head *bh; unsigned short ee_len = ext4_ext_get_actual_len(ex); - int i; + int i, metadata = 0; + if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) + metadata = 1; #ifdef EXTENTS_STATS { struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); @@ -1738,42 +1789,45 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode, if (from >= le32_to_cpu(ex->ee_block) && to == le32_to_cpu(ex->ee_block) + ee_len - 1) { /* tail removal */ - unsigned long num; + ext4_lblk_t num; ext4_fsblk_t start; + num = le32_to_cpu(ex->ee_block) + ee_len - from; start = ext_pblock(ex) + ee_len - num; - ext_debug("free last %lu blocks starting %llu\n", num, start); + ext_debug("free last %u blocks starting %llu\n", num, start); for (i = 0; i < num; i++) { bh = sb_find_get_block(inode->i_sb, start + i); ext4_forget(handle, 0, inode, bh, start + i); } - ext4_free_blocks(handle, inode, start, num); + ext4_free_blocks(handle, inode, start, num, metadata); } else if (from == le32_to_cpu(ex->ee_block) && to <= le32_to_cpu(ex->ee_block) + ee_len - 1) { - printk("strange request: removal %lu-%lu from %u:%u\n", + printk(KERN_INFO "strange request: removal %u-%u from %u:%u\n", from, to, le32_to_cpu(ex->ee_block), ee_len); } else { - printk("strange request: removal(2) %lu-%lu from %u:%u\n", - from, to, le32_to_cpu(ex->ee_block), ee_len); + printk(KERN_INFO "strange request: removal(2) " + "%u-%u from %u:%u\n", + from, to, le32_to_cpu(ex->ee_block), ee_len); } return 0; } static int ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, - struct ext4_ext_path *path, unsigned long start) + struct ext4_ext_path *path, ext4_lblk_t start) { int err = 0, correct_index = 0; int depth = ext_depth(inode), credits; struct ext4_extent_header *eh; - unsigned a, b, block, num; - unsigned long ex_ee_block; + ext4_lblk_t a, b, block; + unsigned num; + ext4_lblk_t ex_ee_block; unsigned short ex_ee_len; unsigned uninitialized = 0; struct ext4_extent *ex; /* the header must be checked already in ext4_ext_remove_space() */ - ext_debug("truncate since %lu in leaf\n", start); + ext_debug("truncate since %u in leaf\n", start); if (!path[depth].p_hdr) path[depth].p_hdr = ext_block_hdr(path[depth].p_bh); eh = path[depth].p_hdr; @@ -1904,7 +1958,7 @@ ext4_ext_more_to_rm(struct ext4_ext_path *path) return 1; } -int ext4_ext_remove_space(struct inode *inode, unsigned long start) +static int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start) { struct super_block *sb = inode->i_sb; int depth = ext_depth(inode); @@ -1912,7 +1966,7 @@ int ext4_ext_remove_space(struct inode *inode, unsigned long start) handle_t *handle; int i = 0, err = 0; - ext_debug("truncate since %lu\n", start); + ext_debug("truncate since %u\n", start); /* probably first extent we're gonna free will be last in block */ handle = ext4_journal_start(inode, depth + 1); @@ -2094,17 +2148,19 @@ void ext4_ext_release(struct super_block *sb) * b> Splits in two extents: Write is happening at either end of the extent * c> Splits in three extents: Somone is writing in middle of the extent */ -int ext4_ext_convert_to_initialized(handle_t *handle, struct inode *inode, - struct ext4_ext_path *path, - ext4_fsblk_t iblock, - unsigned long max_blocks) +static int ext4_ext_convert_to_initialized(handle_t *handle, + struct inode *inode, + struct ext4_ext_path *path, + ext4_lblk_t iblock, + unsigned long max_blocks) { struct ext4_extent *ex, newex; struct ext4_extent *ex1 = NULL; struct ext4_extent *ex2 = NULL; struct ext4_extent *ex3 = NULL; struct ext4_extent_header *eh; - unsigned int allocated, ee_block, ee_len, depth; + ext4_lblk_t ee_block; + unsigned int allocated, ee_len, depth; ext4_fsblk_t newblock; int err = 0; int ret = 0; @@ -2118,6 +2174,10 @@ int ext4_ext_convert_to_initialized(handle_t *handle, struct inode *inode, newblock = iblock - ee_block + ext_pblock(ex); ex2 = ex; + err = ext4_ext_get_access(handle, inode, path + depth); + if (err) + goto out; + /* ex1: ee_block to iblock - 1 : uninitialized */ if (iblock > ee_block) { ex1 = ex; @@ -2150,16 +2210,20 @@ int ext4_ext_convert_to_initialized(handle_t *handle, struct inode *inode, newdepth = ext_depth(inode); if (newdepth != depth) { depth = newdepth; - path = ext4_ext_find_extent(inode, iblock, NULL); + ext4_ext_drop_refs(path); + path = ext4_ext_find_extent(inode, iblock, path); if (IS_ERR(path)) { err = PTR_ERR(path); - path = NULL; goto out; } eh = path[depth].p_hdr; ex = path[depth].p_ext; if (ex2 != &newex) ex2 = ex; + + err = ext4_ext_get_access(handle, inode, path + depth); + if (err) + goto out; } allocated = max_blocks; } @@ -2180,9 +2244,6 @@ int ext4_ext_convert_to_initialized(handle_t *handle, struct inode *inode, ex2->ee_len = cpu_to_le16(allocated); if (ex2 != ex) goto insert; - err = ext4_ext_get_access(handle, inode, path + depth); - if (err) - goto out; /* * New (initialized) extent starts from the first block * in the current extent. i.e., ex2 == ex @@ -2225,8 +2286,26 @@ out: return err ? err : allocated; } +/* + * Block allocation/map/preallocation routine for extents based files + * + * + * Need to be called with + * down_read(&EXT4_I(inode)->i_data_sem) if not allocating file system block + * (ie, create is zero). Otherwise down_write(&EXT4_I(inode)->i_data_sem) + * + * return > 0, number of of blocks already mapped/allocated + * if create == 0 and these are pre-allocated blocks + * buffer head is unmapped + * otherwise blocks are mapped + * + * return = 0, if plain look up failed (blocks have not been allocated) + * buffer head is unmapped + * + * return < 0, error case. + */ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, - ext4_fsblk_t iblock, + ext4_lblk_t iblock, unsigned long max_blocks, struct buffer_head *bh_result, int create, int extend_disksize) { @@ -2236,11 +2315,11 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, ext4_fsblk_t goal, newblock; int err = 0, depth, ret; unsigned long allocated = 0; + struct ext4_allocation_request ar; __clear_bit(BH_New, &bh_result->b_state); - ext_debug("blocks %d/%lu requested for inode %u\n", (int) iblock, - max_blocks, (unsigned) inode->i_ino); - mutex_lock(&EXT4_I(inode)->truncate_mutex); + ext_debug("blocks %u/%lu requested for inode %u\n", + iblock, max_blocks, inode->i_ino); /* check in cache */ goal = ext4_ext_in_cache(inode, iblock, &newex); @@ -2260,7 +2339,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, - le32_to_cpu(newex.ee_block) + ext_pblock(&newex); /* number of remaining blocks in the extent */ - allocated = le16_to_cpu(newex.ee_len) - + allocated = ext4_ext_get_actual_len(&newex) - (iblock - le32_to_cpu(newex.ee_block)); goto out; } else { @@ -2288,7 +2367,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, ex = path[depth].p_ext; if (ex) { - unsigned long ee_block = le32_to_cpu(ex->ee_block); + ext4_lblk_t ee_block = le32_to_cpu(ex->ee_block); ext4_fsblk_t ee_start = ext_pblock(ex); unsigned short ee_len; @@ -2302,7 +2381,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, newblock = iblock - ee_block + ee_start; /* number of remaining blocks in the extent */ allocated = ee_len - (iblock - ee_block); - ext_debug("%d fit into %lu:%d -> %llu\n", (int) iblock, + ext_debug("%u fit into %lu:%d -> %llu\n", iblock, ee_block, ee_len, newblock); /* Do not put uninitialized extent in the cache */ @@ -2320,9 +2399,10 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, ret = ext4_ext_convert_to_initialized(handle, inode, path, iblock, max_blocks); - if (ret <= 0) + if (ret <= 0) { + err = ret; goto out2; - else + } else allocated = ret; goto outnew; } @@ -2347,8 +2427,15 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, if (S_ISREG(inode->i_mode) && (!EXT4_I(inode)->i_block_alloc_info)) ext4_init_block_alloc_info(inode); - /* allocate new block */ - goal = ext4_ext_find_goal(inode, path, iblock); + /* find neighbour allocated blocks */ + ar.lleft = iblock; + err = ext4_ext_search_left(inode, path, &ar.lleft, &ar.pleft); + if (err) + goto out2; + ar.lright = iblock; + err = ext4_ext_search_right(inode, path, &ar.lright, &ar.pright); + if (err) + goto out2; /* * See if request is beyond maximum number of blocks we can have in @@ -2368,10 +2455,21 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, newex.ee_len = cpu_to_le16(max_blocks); err = ext4_ext_check_overlap(inode, &newex, path); if (err) - allocated = le16_to_cpu(newex.ee_len); + allocated = ext4_ext_get_actual_len(&newex); else allocated = max_blocks; - newblock = ext4_new_blocks(handle, inode, goal, &allocated, &err); + + /* allocate new block */ + ar.inode = inode; + ar.goal = ext4_ext_find_goal(inode, path, iblock); + ar.logical = iblock; + ar.len = allocated; + if (S_ISREG(inode->i_mode)) + ar.flags = EXT4_MB_HINT_DATA; + else + /* disable in-core preallocation for non-regular files */ + ar.flags = 0; + newblock = ext4_mb_new_blocks(handle, &ar, &err); if (!newblock) goto out2; ext_debug("allocate new block: goal %llu, found %llu/%lu\n", @@ -2379,14 +2477,17 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, /* try to insert new extent into found leaf and return */ ext4_ext_store_pblock(&newex, newblock); - newex.ee_len = cpu_to_le16(allocated); + newex.ee_len = cpu_to_le16(ar.len); if (create == EXT4_CREATE_UNINITIALIZED_EXT) /* Mark uninitialized */ ext4_ext_mark_uninitialized(&newex); err = ext4_ext_insert_extent(handle, inode, path, &newex); if (err) { /* free data blocks we just allocated */ + /* not a good idea to call discard here directly, + * but otherwise we'd need to call it every free() */ + ext4_mb_discard_inode_preallocations(inode); ext4_free_blocks(handle, inode, ext_pblock(&newex), - le16_to_cpu(newex.ee_len)); + ext4_ext_get_actual_len(&newex), 0); goto out2; } @@ -2395,6 +2496,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, /* previous routine could use block we allocated */ newblock = ext_pblock(&newex); + allocated = ext4_ext_get_actual_len(&newex); outnew: __set_bit(BH_New, &bh_result->b_state); @@ -2414,8 +2516,6 @@ out2: ext4_ext_drop_refs(path); kfree(path); } - mutex_unlock(&EXT4_I(inode)->truncate_mutex); - return err ? err : allocated; } @@ -2423,7 +2523,7 @@ void ext4_ext_truncate(struct inode * inode, struct page *page) { struct address_space *mapping = inode->i_mapping; struct super_block *sb = inode->i_sb; - unsigned long last_block; + ext4_lblk_t last_block; handle_t *handle; int err = 0; @@ -2445,9 +2545,11 @@ void ext4_ext_truncate(struct inode * inode, struct page *page) if (page) ext4_block_truncate_page(handle, page, mapping, inode->i_size); - mutex_lock(&EXT4_I(inode)->truncate_mutex); + down_write(&EXT4_I(inode)->i_data_sem); ext4_ext_invalidate_cache(inode); + ext4_mb_discard_inode_preallocations(inode); + /* * TODO: optimization is possible here. * Probably we need not scan at all, @@ -2481,7 +2583,7 @@ out_stop: if (inode->i_nlink) ext4_orphan_del(handle, inode); - mutex_unlock(&EXT4_I(inode)->truncate_mutex); + up_write(&EXT4_I(inode)->i_data_sem); ext4_journal_stop(handle); } @@ -2516,7 +2618,8 @@ int ext4_ext_writepage_trans_blocks(struct inode *inode, int num) long ext4_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len) { handle_t *handle; - ext4_fsblk_t block, max_blocks; + ext4_lblk_t block; + unsigned long max_blocks; ext4_fsblk_t nblocks = 0; int ret = 0; int ret2 = 0; @@ -2544,6 +2647,7 @@ long ext4_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len) * modify 1 super block, 1 block bitmap and 1 group descriptor. */ credits = EXT4_DATA_TRANS_BLOCKS(inode->i_sb) + 3; + mutex_lock(&inode->i_mutex); retry: while (ret >= 0 && ret < max_blocks) { block = block + ret; @@ -2554,16 +2658,17 @@ retry: break; } - ret = ext4_ext_get_blocks(handle, inode, block, + ret = ext4_get_blocks_wrap(handle, inode, block, max_blocks, &map_bh, EXT4_CREATE_UNINITIALIZED_EXT, 0); - WARN_ON(!ret); - if (!ret) { - ext4_error(inode->i_sb, "ext4_fallocate", - "ext4_ext_get_blocks returned 0! inode#%lu" - ", block=%llu, max_blocks=%llu", - inode->i_ino, block, max_blocks); - ret = -EIO; + if (ret <= 0) { +#ifdef EXT4FS_DEBUG + WARN_ON(ret <= 0); + printk(KERN_ERR "%s: ext4_ext_get_blocks " + "returned error inode#%lu, block=%u, " + "max_blocks=%lu", __func__, + inode->i_ino, block, max_blocks); +#endif ext4_mark_inode_dirty(handle, inode); ret2 = ext4_journal_stop(handle); break; @@ -2611,21 +2716,18 @@ retry: * if no error, we assume preallocation succeeded * completely */ - mutex_lock(&inode->i_mutex); i_size_write(inode, offset + len); EXT4_I(inode)->i_disksize = i_size_read(inode); - mutex_unlock(&inode->i_mutex); } else if (ret < 0 && nblocks) { /* Handle partial allocation scenario */ loff_t newsize; - mutex_lock(&inode->i_mutex); newsize = (nblocks << blkbits) + i_size_read(inode); i_size_write(inode, EXT4_BLOCK_ALIGN(newsize, blkbits)); EXT4_I(inode)->i_disksize = i_size_read(inode); - mutex_unlock(&inode->i_mutex); } } + mutex_unlock(&inode->i_mutex); return ret > 0 ? ret2 : ret; } diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 1a81cd66d63..ac35ec58db5 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -37,9 +37,9 @@ static int ext4_release_file (struct inode * inode, struct file * filp) if ((filp->f_mode & FMODE_WRITE) && (atomic_read(&inode->i_writecount) == 1)) { - mutex_lock(&EXT4_I(inode)->truncate_mutex); + down_write(&EXT4_I(inode)->i_data_sem); ext4_discard_reservation(inode); - mutex_unlock(&EXT4_I(inode)->truncate_mutex); + up_write(&EXT4_I(inode)->i_data_sem); } if (is_dx(inode) && filp->private_data) ext4_htree_free_dir_info(filp->private_data); @@ -56,8 +56,25 @@ ext4_file_write(struct kiocb *iocb, const struct iovec *iov, ssize_t ret; int err; - ret = generic_file_aio_write(iocb, iov, nr_segs, pos); + /* + * If we have encountered a bitmap-format file, the size limit + * is smaller than s_maxbytes, which is for extent-mapped files. + */ + + if (!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)) { + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + size_t length = iov_length(iov, nr_segs); + if (pos > sbi->s_bitmap_maxbytes) + return -EFBIG; + + if (pos + length > sbi->s_bitmap_maxbytes) { + nr_segs = iov_shorten((struct iovec *)iov, nr_segs, + sbi->s_bitmap_maxbytes - pos); + } + } + + ret = generic_file_aio_write(iocb, iov, nr_segs, pos); /* * Skip flushing if there was an error, or if nothing was written. */ diff --git a/fs/ext4/group.h b/fs/ext4/group.h index 1577910bb58..7eb0604e7ee 100644 --- a/fs/ext4/group.h +++ b/fs/ext4/group.h @@ -14,14 +14,16 @@ extern __le16 ext4_group_desc_csum(struct ext4_sb_info *sbi, __u32 group, extern int ext4_group_desc_csum_verify(struct ext4_sb_info *sbi, __u32 group, struct ext4_group_desc *gdp); struct buffer_head *read_block_bitmap(struct super_block *sb, - unsigned int block_group); + ext4_group_t block_group); extern unsigned ext4_init_block_bitmap(struct super_block *sb, - struct buffer_head *bh, int group, + struct buffer_head *bh, + ext4_group_t group, struct ext4_group_desc *desc); #define ext4_free_blocks_after_init(sb, group, desc) \ ext4_init_block_bitmap(sb, NULL, group, desc) extern unsigned ext4_init_inode_bitmap(struct super_block *sb, - struct buffer_head *bh, int group, + struct buffer_head *bh, + ext4_group_t group, struct ext4_group_desc *desc); extern void mark_bitmap_end(int start_bit, int end_bit, char *bitmap); #endif /* _LINUX_EXT4_GROUP_H */ diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index c61f37fd3f0..8036b9b5376 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -64,8 +64,8 @@ void mark_bitmap_end(int start_bit, int end_bit, char *bitmap) } /* Initializes an uninitialized inode bitmap */ -unsigned ext4_init_inode_bitmap(struct super_block *sb, - struct buffer_head *bh, int block_group, +unsigned ext4_init_inode_bitmap(struct super_block *sb, struct buffer_head *bh, + ext4_group_t block_group, struct ext4_group_desc *gdp) { struct ext4_sb_info *sbi = EXT4_SB(sb); @@ -75,7 +75,7 @@ unsigned ext4_init_inode_bitmap(struct super_block *sb, /* If checksum is bad mark all blocks and inodes use to prevent * allocation, essentially implementing a per-group read-only flag. */ if (!ext4_group_desc_csum_verify(sbi, block_group, gdp)) { - ext4_error(sb, __FUNCTION__, "Checksum bad for group %u\n", + ext4_error(sb, __FUNCTION__, "Checksum bad for group %lu\n", block_group); gdp->bg_free_blocks_count = 0; gdp->bg_free_inodes_count = 0; @@ -98,7 +98,7 @@ unsigned ext4_init_inode_bitmap(struct super_block *sb, * Return buffer_head of bitmap on success or NULL. */ static struct buffer_head * -read_inode_bitmap(struct super_block * sb, unsigned long block_group) +read_inode_bitmap(struct super_block *sb, ext4_group_t block_group) { struct ext4_group_desc *desc; struct buffer_head *bh = NULL; @@ -152,7 +152,7 @@ void ext4_free_inode (handle_t *handle, struct inode * inode) unsigned long ino; struct buffer_head *bitmap_bh = NULL; struct buffer_head *bh2; - unsigned long block_group; + ext4_group_t block_group; unsigned long bit; struct ext4_group_desc * gdp; struct ext4_super_block * es; @@ -260,12 +260,14 @@ error_return: * For other inodes, search forward from the parent directory\'s block * group to find a free inode. */ -static int find_group_dir(struct super_block *sb, struct inode *parent) +static int find_group_dir(struct super_block *sb, struct inode *parent, + ext4_group_t *best_group) { - int ngroups = EXT4_SB(sb)->s_groups_count; + ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count; unsigned int freei, avefreei; struct ext4_group_desc *desc, *best_desc = NULL; - int group, best_group = -1; + ext4_group_t group; + int ret = -1; freei = percpu_counter_read_positive(&EXT4_SB(sb)->s_freeinodes_counter); avefreei = freei / ngroups; @@ -279,11 +281,12 @@ static int find_group_dir(struct super_block *sb, struct inode *parent) if (!best_desc || (le16_to_cpu(desc->bg_free_blocks_count) > le16_to_cpu(best_desc->bg_free_blocks_count))) { - best_group = group; + *best_group = group; best_desc = desc; + ret = 0; } } - return best_group; + return ret; } /* @@ -314,12 +317,13 @@ static int find_group_dir(struct super_block *sb, struct inode *parent) #define INODE_COST 64 #define BLOCK_COST 256 -static int find_group_orlov(struct super_block *sb, struct inode *parent) +static int find_group_orlov(struct super_block *sb, struct inode *parent, + ext4_group_t *group) { - int parent_group = EXT4_I(parent)->i_block_group; + ext4_group_t parent_group = EXT4_I(parent)->i_block_group; struct ext4_sb_info *sbi = EXT4_SB(sb); struct ext4_super_block *es = sbi->s_es; - int ngroups = sbi->s_groups_count; + ext4_group_t ngroups = sbi->s_groups_count; int inodes_per_group = EXT4_INODES_PER_GROUP(sb); unsigned int freei, avefreei; ext4_fsblk_t freeb, avefreeb; @@ -327,7 +331,7 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent) unsigned int ndirs; int max_debt, max_dirs, min_inodes; ext4_grpblk_t min_blocks; - int group = -1, i; + ext4_group_t i; struct ext4_group_desc *desc; freei = percpu_counter_read_positive(&sbi->s_freeinodes_counter); @@ -340,13 +344,14 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent) if ((parent == sb->s_root->d_inode) || (EXT4_I(parent)->i_flags & EXT4_TOPDIR_FL)) { int best_ndir = inodes_per_group; - int best_group = -1; + ext4_group_t grp; + int ret = -1; - get_random_bytes(&group, sizeof(group)); - parent_group = (unsigned)group % ngroups; + get_random_bytes(&grp, sizeof(grp)); + parent_group = (unsigned)grp % ngroups; for (i = 0; i < ngroups; i++) { - group = (parent_group + i) % ngroups; - desc = ext4_get_group_desc (sb, group, NULL); + grp = (parent_group + i) % ngroups; + desc = ext4_get_group_desc(sb, grp, NULL); if (!desc || !desc->bg_free_inodes_count) continue; if (le16_to_cpu(desc->bg_used_dirs_count) >= best_ndir) @@ -355,11 +360,12 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent) continue; if (le16_to_cpu(desc->bg_free_blocks_count) < avefreeb) continue; - best_group = group; + *group = grp; + ret = 0; best_ndir = le16_to_cpu(desc->bg_used_dirs_count); } - if (best_group >= 0) - return best_group; + if (ret == 0) + return ret; goto fallback; } @@ -380,8 +386,8 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent) max_debt = 1; for (i = 0; i < ngroups; i++) { - group = (parent_group + i) % ngroups; - desc = ext4_get_group_desc (sb, group, NULL); + *group = (parent_group + i) % ngroups; + desc = ext4_get_group_desc(sb, *group, NULL); if (!desc || !desc->bg_free_inodes_count) continue; if (le16_to_cpu(desc->bg_used_dirs_count) >= max_dirs) @@ -390,17 +396,16 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent) continue; if (le16_to_cpu(desc->bg_free_blocks_count) < min_blocks) continue; - return group; + return 0; } fallback: for (i = 0; i < ngroups; i++) { - group = (parent_group + i) % ngroups; - desc = ext4_get_group_desc (sb, group, NULL); - if (!desc || !desc->bg_free_inodes_count) - continue; - if (le16_to_cpu(desc->bg_free_inodes_count) >= avefreei) - return group; + *group = (parent_group + i) % ngroups; + desc = ext4_get_group_desc(sb, *group, NULL); + if (desc && desc->bg_free_inodes_count && + le16_to_cpu(desc->bg_free_inodes_count) >= avefreei) + return 0; } if (avefreei) { @@ -415,21 +420,22 @@ fallback: return -1; } -static int find_group_other(struct super_block *sb, struct inode *parent) +static int find_group_other(struct super_block *sb, struct inode *parent, + ext4_group_t *group) { - int parent_group = EXT4_I(parent)->i_block_group; - int ngroups = EXT4_SB(sb)->s_groups_count; + ext4_group_t parent_group = EXT4_I(parent)->i_block_group; + ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count; struct ext4_group_desc *desc; - int group, i; + ext4_group_t i; /* * Try to place the inode in its parent directory */ - group = parent_group; - desc = ext4_get_group_desc (sb, group, NULL); + *group = parent_group; + desc = ext4_get_group_desc(sb, *group, NULL); if (desc && le16_to_cpu(desc->bg_free_inodes_count) && le16_to_cpu(desc->bg_free_blocks_count)) - return group; + return 0; /* * We're going to place this inode in a different blockgroup from its @@ -440,33 +446,33 @@ static int find_group_other(struct super_block *sb, struct inode *parent) * * So add our directory's i_ino into the starting point for the hash. */ - group = (group + parent->i_ino) % ngroups; + *group = (*group + parent->i_ino) % ngroups; /* * Use a quadratic hash to find a group with a free inode and some free * blocks. */ for (i = 1; i < ngroups; i <<= 1) { - group += i; - if (group >= ngroups) - group -= ngroups; - desc = ext4_get_group_desc (sb, group, NULL); + *group += i; + if (*group >= ngroups) + *group -= ngroups; + desc = ext4_get_group_desc(sb, *group, NULL); if (desc && le16_to_cpu(desc->bg_free_inodes_count) && le16_to_cpu(desc->bg_free_blocks_count)) - return group; + return 0; } /* * That failed: try linear search for a free inode, even if that group * has no free blocks. */ - group = parent_group; + *group = parent_group; for (i = 0; i < ngroups; i++) { - if (++group >= ngroups) - group = 0; - desc = ext4_get_group_desc (sb, group, NULL); + if (++*group >= ngroups) + *group = 0; + desc = ext4_get_group_desc(sb, *group, NULL); if (desc && le16_to_cpu(desc->bg_free_inodes_count)) - return group; + return 0; } return -1; @@ -487,16 +493,17 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode) struct super_block *sb; struct buffer_head *bitmap_bh = NULL; struct buffer_head *bh2; - int group; + ext4_group_t group = 0; unsigned long ino = 0; struct inode * inode; struct ext4_group_desc * gdp = NULL; struct ext4_super_block * es; struct ext4_inode_info *ei; struct ext4_sb_info *sbi; - int err = 0; + int ret2, err = 0; struct inode *ret; - int i, free = 0; + ext4_group_t i; + int free = 0; /* Cannot create files in a deleted directory */ if (!dir || !dir->i_nlink) @@ -512,14 +519,14 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode) es = sbi->s_es; if (S_ISDIR(mode)) { if (test_opt (sb, OLDALLOC)) - group = find_group_dir(sb, dir); + ret2 = find_group_dir(sb, dir, &group); else - group = find_group_orlov(sb, dir); + ret2 = find_group_orlov(sb, dir, &group); } else - group = find_group_other(sb, dir); + ret2 = find_group_other(sb, dir, &group); err = -ENOSPC; - if (group == -1) + if (ret2 == -1) goto out; for (i = 0; i < sbi->s_groups_count; i++) { @@ -583,7 +590,7 @@ got: ino > EXT4_INODES_PER_GROUP(sb)) { ext4_error(sb, __FUNCTION__, "reserved inode or inode > inodes count - " - "block_group = %d, inode=%lu", group, + "block_group = %lu, inode=%lu", group, ino + group * EXT4_INODES_PER_GROUP(sb)); err = -EIO; goto fail; @@ -695,14 +702,18 @@ got: ei->i_dir_start_lookup = 0; ei->i_disksize = 0; - ei->i_flags = EXT4_I(dir)->i_flags & ~EXT4_INDEX_FL; + /* + * Don't inherit extent flag from directory. We set extent flag on + * newly created directory and file only if -o extent mount option is + * specified + */ + ei->i_flags = EXT4_I(dir)->i_flags & ~(EXT4_INDEX_FL|EXT4_EXTENTS_FL); if (S_ISLNK(mode)) ei->i_flags &= ~(EXT4_IMMUTABLE_FL|EXT4_APPEND_FL); /* dirsync only applies to directories */ if (!S_ISDIR(mode)) ei->i_flags &= ~EXT4_DIRSYNC_FL; ei->i_file_acl = 0; - ei->i_dir_acl = 0; ei->i_dtime = 0; ei->i_block_alloc_info = NULL; ei->i_block_group = group; @@ -739,14 +750,14 @@ got: goto fail_free_drop; } if (test_opt(sb, EXTENTS)) { - EXT4_I(inode)->i_flags |= EXT4_EXTENTS_FL; - ext4_ext_tree_init(handle, inode); - if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS)) { - err = ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh); - if (err) goto fail; - EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS); - BUFFER_TRACE(EXT4_SB(sb)->s_sbh, "call ext4_journal_dirty_metadata"); - err = ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh); + /* set extent flag only for directory and file */ + if (S_ISDIR(mode) || S_ISREG(mode)) { + EXT4_I(inode)->i_flags |= EXT4_EXTENTS_FL; + ext4_ext_tree_init(handle, inode); + err = ext4_update_incompat_feature(handle, sb, + EXT4_FEATURE_INCOMPAT_EXTENTS); + if (err) + goto fail; } } @@ -777,16 +788,17 @@ fail_drop: struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino) { unsigned long max_ino = le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count); - unsigned long block_group; + ext4_group_t block_group; int bit; - struct buffer_head *bitmap_bh = NULL; + struct buffer_head *bitmap_bh; struct inode *inode = NULL; + long err = -EIO; /* Error cases - e2fsck has already cleaned up for us */ if (ino > max_ino) { ext4_warning(sb, __FUNCTION__, "bad orphan ino %lu! e2fsck was run?", ino); - goto out; + goto error; } block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb); @@ -795,45 +807,56 @@ struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino) if (!bitmap_bh) { ext4_warning(sb, __FUNCTION__, "inode bitmap error for orphan %lu", ino); - goto out; + goto error; } /* Having the inode bit set should be a 100% indicator that this * is a valid orphan (no e2fsck run on fs). Orphans also include * inodes that were being truncated, so we can't check i_nlink==0. */ - if (!ext4_test_bit(bit, bitmap_bh->b_data) || - !(inode = iget(sb, ino)) || is_bad_inode(inode) || - NEXT_ORPHAN(inode) > max_ino) { - ext4_warning(sb, __FUNCTION__, - "bad orphan inode %lu! e2fsck was run?", ino); - printk(KERN_NOTICE "ext4_test_bit(bit=%d, block=%llu) = %d\n", - bit, (unsigned long long)bitmap_bh->b_blocknr, - ext4_test_bit(bit, bitmap_bh->b_data)); - printk(KERN_NOTICE "inode=%p\n", inode); - if (inode) { - printk(KERN_NOTICE "is_bad_inode(inode)=%d\n", - is_bad_inode(inode)); - printk(KERN_NOTICE "NEXT_ORPHAN(inode)=%u\n", - NEXT_ORPHAN(inode)); - printk(KERN_NOTICE "max_ino=%lu\n", max_ino); - } + if (!ext4_test_bit(bit, bitmap_bh->b_data)) + goto bad_orphan; + + inode = ext4_iget(sb, ino); + if (IS_ERR(inode)) + goto iget_failed; + + if (NEXT_ORPHAN(inode) > max_ino) + goto bad_orphan; + brelse(bitmap_bh); + return inode; + +iget_failed: + err = PTR_ERR(inode); + inode = NULL; +bad_orphan: + ext4_warning(sb, __FUNCTION__, + "bad orphan inode %lu! e2fsck was run?", ino); + printk(KERN_NOTICE "ext4_test_bit(bit=%d, block=%llu) = %d\n", + bit, (unsigned long long)bitmap_bh->b_blocknr, + ext4_test_bit(bit, bitmap_bh->b_data)); + printk(KERN_NOTICE "inode=%p\n", inode); + if (inode) { + printk(KERN_NOTICE "is_bad_inode(inode)=%d\n", + is_bad_inode(inode)); + printk(KERN_NOTICE "NEXT_ORPHAN(inode)=%u\n", + NEXT_ORPHAN(inode)); + printk(KERN_NOTICE "max_ino=%lu\n", max_ino); /* Avoid freeing blocks if we got a bad deleted inode */ - if (inode && inode->i_nlink == 0) + if (inode->i_nlink == 0) inode->i_blocks = 0; iput(inode); - inode = NULL; } -out: brelse(bitmap_bh); - return inode; +error: + return ERR_PTR(err); } unsigned long ext4_count_free_inodes (struct super_block * sb) { unsigned long desc_count; struct ext4_group_desc *gdp; - int i; + ext4_group_t i; #ifdef EXT4FS_DEBUG struct ext4_super_block *es; unsigned long bitmap_count, x; @@ -854,7 +877,7 @@ unsigned long ext4_count_free_inodes (struct super_block * sb) continue; x = ext4_count_free(bitmap_bh, EXT4_INODES_PER_GROUP(sb) / 8); - printk("group %d: stored = %d, counted = %lu\n", + printk(KERN_DEBUG "group %lu: stored = %d, counted = %lu\n", i, le16_to_cpu(gdp->bg_free_inodes_count), x); bitmap_count += x; } @@ -879,7 +902,7 @@ unsigned long ext4_count_free_inodes (struct super_block * sb) unsigned long ext4_count_dirs (struct super_block * sb) { unsigned long count = 0; - int i; + ext4_group_t i; for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) { struct ext4_group_desc *gdp = ext4_get_group_desc (sb, i, NULL); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 5489703d957..945cbf6cb1f 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -105,7 +105,7 @@ int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode, */ static unsigned long blocks_for_truncate(struct inode *inode) { - unsigned long needed; + ext4_lblk_t needed; needed = inode->i_blocks >> (inode->i_sb->s_blocksize_bits - 9); @@ -243,13 +243,6 @@ static inline void add_chain(Indirect *p, struct buffer_head *bh, __le32 *v) p->bh = bh; } -static int verify_chain(Indirect *from, Indirect *to) -{ - while (from <= to && from->key == *from->p) - from++; - return (from > to); -} - /** * ext4_block_to_path - parse the block number into array of offsets * @inode: inode in question (we are only interested in its superblock) @@ -282,7 +275,8 @@ static int verify_chain(Indirect *from, Indirect *to) */ static int ext4_block_to_path(struct inode *inode, - long i_block, int offsets[4], int *boundary) + ext4_lblk_t i_block, + ext4_lblk_t offsets[4], int *boundary) { int ptrs = EXT4_ADDR_PER_BLOCK(inode->i_sb); int ptrs_bits = EXT4_ADDR_PER_BLOCK_BITS(inode->i_sb); @@ -313,7 +307,10 @@ static int ext4_block_to_path(struct inode *inode, offsets[n++] = i_block & (ptrs - 1); final = ptrs; } else { - ext4_warning(inode->i_sb, "ext4_block_to_path", "block > big"); + ext4_warning(inode->i_sb, "ext4_block_to_path", + "block %lu > max", + i_block + direct_blocks + + indirect_blocks + double_blocks); } if (boundary) *boundary = final - 1 - (i_block & (ptrs - 1)); @@ -344,12 +341,14 @@ static int ext4_block_to_path(struct inode *inode, * (pointer to last triple returned, *@err == 0) * or when it gets an IO error reading an indirect block * (ditto, *@err == -EIO) - * or when it notices that chain had been changed while it was reading - * (ditto, *@err == -EAGAIN) * or when it reads all @depth-1 indirect blocks successfully and finds * the whole chain, all way to the data (returns %NULL, *err == 0). + * + * Need to be called with + * down_read(&EXT4_I(inode)->i_data_sem) */ -static Indirect *ext4_get_branch(struct inode *inode, int depth, int *offsets, +static Indirect *ext4_get_branch(struct inode *inode, int depth, + ext4_lblk_t *offsets, Indirect chain[4], int *err) { struct super_block *sb = inode->i_sb; @@ -365,9 +364,6 @@ static Indirect *ext4_get_branch(struct inode *inode, int depth, int *offsets, bh = sb_bread(sb, le32_to_cpu(p->key)); if (!bh) goto failure; - /* Reader: pointers */ - if (!verify_chain(chain, p)) - goto changed; add_chain(++p, bh, (__le32*)bh->b_data + *++offsets); /* Reader: end */ if (!p->key) @@ -375,10 +371,6 @@ static Indirect *ext4_get_branch(struct inode *inode, int depth, int *offsets, } return NULL; -changed: - brelse(bh); - *err = -EAGAIN; - goto no_block; failure: *err = -EIO; no_block: @@ -411,6 +403,7 @@ static ext4_fsblk_t ext4_find_near(struct inode *inode, Indirect *ind) __le32 *start = ind->bh ? (__le32*) ind->bh->b_data : ei->i_data; __le32 *p; ext4_fsblk_t bg_start; + ext4_fsblk_t last_block; ext4_grpblk_t colour; /* Try to find previous block */ @@ -428,8 +421,13 @@ static ext4_fsblk_t ext4_find_near(struct inode *inode, Indirect *ind) * into the same cylinder group then. */ bg_start = ext4_group_first_block_no(inode->i_sb, ei->i_block_group); - colour = (current->pid % 16) * + last_block = ext4_blocks_count(EXT4_SB(inode->i_sb)->s_es) - 1; + + if (bg_start + EXT4_BLOCKS_PER_GROUP(inode->i_sb) <= last_block) + colour = (current->pid % 16) * (EXT4_BLOCKS_PER_GROUP(inode->i_sb) / 16); + else + colour = (current->pid % 16) * ((last_block - bg_start) / 16); return bg_start + colour; } @@ -437,16 +435,13 @@ static ext4_fsblk_t ext4_find_near(struct inode *inode, Indirect *ind) * ext4_find_goal - find a prefered place for allocation. * @inode: owner * @block: block we want - * @chain: chain of indirect blocks * @partial: pointer to the last triple within a chain - * @goal: place to store the result. * * Normally this function find the prefered place for block allocation, - * stores it in *@goal and returns zero. + * returns it. */ - -static ext4_fsblk_t ext4_find_goal(struct inode *inode, long block, - Indirect chain[4], Indirect *partial) +static ext4_fsblk_t ext4_find_goal(struct inode *inode, ext4_lblk_t block, + Indirect *partial) { struct ext4_block_alloc_info *block_i; @@ -559,7 +554,7 @@ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, return ret; failed_out: for (i = 0; i <index; i++) - ext4_free_blocks(handle, inode, new_blocks[i], 1); + ext4_free_blocks(handle, inode, new_blocks[i], 1, 0); return ret; } @@ -590,7 +585,7 @@ failed_out: */ static int ext4_alloc_branch(handle_t *handle, struct inode *inode, int indirect_blks, int *blks, ext4_fsblk_t goal, - int *offsets, Indirect *branch) + ext4_lblk_t *offsets, Indirect *branch) { int blocksize = inode->i_sb->s_blocksize; int i, n = 0; @@ -658,9 +653,9 @@ failed: ext4_journal_forget(handle, branch[i].bh); } for (i = 0; i <indirect_blks; i++) - ext4_free_blocks(handle, inode, new_blocks[i], 1); + ext4_free_blocks(handle, inode, new_blocks[i], 1, 0); - ext4_free_blocks(handle, inode, new_blocks[i], num); + ext4_free_blocks(handle, inode, new_blocks[i], num, 0); return err; } @@ -680,7 +675,7 @@ failed: * chain to new block and return 0. */ static int ext4_splice_branch(handle_t *handle, struct inode *inode, - long block, Indirect *where, int num, int blks) + ext4_lblk_t block, Indirect *where, int num, int blks) { int i; int err = 0; @@ -757,9 +752,10 @@ err_out: for (i = 1; i <= num; i++) { BUFFER_TRACE(where[i].bh, "call jbd2_journal_forget"); ext4_journal_forget(handle, where[i].bh); - ext4_free_blocks(handle,inode,le32_to_cpu(where[i-1].key),1); + ext4_free_blocks(handle, inode, + le32_to_cpu(where[i-1].key), 1, 0); } - ext4_free_blocks(handle, inode, le32_to_cpu(where[num].key), blks); + ext4_free_blocks(handle, inode, le32_to_cpu(where[num].key), blks, 0); return err; } @@ -778,18 +774,22 @@ err_out: * * `handle' can be NULL if create == 0. * - * The BKL may not be held on entry here. Be sure to take it early. * return > 0, # of blocks mapped or allocated. * return = 0, if plain lookup failed. * return < 0, error case. + * + * + * Need to be called with + * down_read(&EXT4_I(inode)->i_data_sem) if not allocating file system block + * (ie, create is zero). Otherwise down_write(&EXT4_I(inode)->i_data_sem) */ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, - sector_t iblock, unsigned long maxblocks, + ext4_lblk_t iblock, unsigned long maxblocks, struct buffer_head *bh_result, int create, int extend_disksize) { int err = -EIO; - int offsets[4]; + ext4_lblk_t offsets[4]; Indirect chain[4]; Indirect *partial; ext4_fsblk_t goal; @@ -803,7 +803,8 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, J_ASSERT(!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)); J_ASSERT(handle != NULL || create == 0); - depth = ext4_block_to_path(inode,iblock,offsets,&blocks_to_boundary); + depth = ext4_block_to_path(inode, iblock, offsets, + &blocks_to_boundary); if (depth == 0) goto out; @@ -819,18 +820,6 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, while (count < maxblocks && count <= blocks_to_boundary) { ext4_fsblk_t blk; - if (!verify_chain(chain, partial)) { - /* - * Indirect block might be removed by - * truncate while we were reading it. - * Handling of that case: forget what we've - * got now. Flag the err as EAGAIN, so it - * will reread. - */ - err = -EAGAIN; - count = 0; - break; - } blk = le32_to_cpu(*(chain[depth-1].p + count)); if (blk == first_block + count) @@ -838,44 +827,13 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, else break; } - if (err != -EAGAIN) - goto got_it; + goto got_it; } /* Next simple case - plain lookup or failed read of indirect block */ if (!create || err == -EIO) goto cleanup; - mutex_lock(&ei->truncate_mutex); - - /* - * If the indirect block is missing while we are reading - * the chain(ext4_get_branch() returns -EAGAIN err), or - * if the chain has been changed after we grab the semaphore, - * (either because another process truncated this branch, or - * another get_block allocated this branch) re-grab the chain to see if - * the request block has been allocated or not. - * - * Since we already block the truncate/other get_block - * at this point, we will have the current copy of the chain when we - * splice the branch into the tree. - */ - if (err == -EAGAIN || !verify_chain(chain, partial)) { - while (partial > chain) { - brelse(partial->bh); - partial--; - } - partial = ext4_get_branch(inode, depth, offsets, chain, &err); - if (!partial) { - count++; - mutex_unlock(&ei->truncate_mutex); - if (err) - goto cleanup; - clear_buffer_new(bh_result); - goto got_it; - } - } - /* * Okay, we need to do block allocation. Lazily initialize the block * allocation info here if necessary @@ -883,7 +841,7 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, if (S_ISREG(inode->i_mode) && (!ei->i_block_alloc_info)) ext4_init_block_alloc_info(inode); - goal = ext4_find_goal(inode, iblock, chain, partial); + goal = ext4_find_goal(inode, iblock, partial); /* the number of blocks need to allocate for [d,t]indirect blocks */ indirect_blks = (chain + depth) - partial - 1; @@ -911,13 +869,12 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, err = ext4_splice_branch(handle, inode, iblock, partial, indirect_blks, count); /* - * i_disksize growing is protected by truncate_mutex. Don't forget to + * i_disksize growing is protected by i_data_sem. Don't forget to * protect it if you're about to implement concurrent * ext4_get_block() -bzzz */ if (!err && extend_disksize && inode->i_size > ei->i_disksize) ei->i_disksize = inode->i_size; - mutex_unlock(&ei->truncate_mutex); if (err) goto cleanup; @@ -940,55 +897,128 @@ out: return err; } -#define DIO_CREDITS (EXT4_RESERVE_TRANS_BLOCKS + 32) +/* Maximum number of blocks we map for direct IO at once. */ +#define DIO_MAX_BLOCKS 4096 +/* + * Number of credits we need for writing DIO_MAX_BLOCKS: + * We need sb + group descriptor + bitmap + inode -> 4 + * For B blocks with A block pointers per block we need: + * 1 (triple ind.) + (B/A/A + 2) (doubly ind.) + (B/A + 2) (indirect). + * If we plug in 4096 for B and 256 for A (for 1KB block size), we get 25. + */ +#define DIO_CREDITS 25 -static int ext4_get_block(struct inode *inode, sector_t iblock, - struct buffer_head *bh_result, int create) + +/* + * + * + * ext4_ext4 get_block() wrapper function + * It will do a look up first, and returns if the blocks already mapped. + * Otherwise it takes the write lock of the i_data_sem and allocate blocks + * and store the allocated blocks in the result buffer head and mark it + * mapped. + * + * If file type is extents based, it will call ext4_ext_get_blocks(), + * Otherwise, call with ext4_get_blocks_handle() to handle indirect mapping + * based files + * + * On success, it returns the number of blocks being mapped or allocate. + * if create==0 and the blocks are pre-allocated and uninitialized block, + * the result buffer head is unmapped. If the create ==1, it will make sure + * the buffer head is mapped. + * + * It returns 0 if plain look up failed (blocks have not been allocated), in + * that casem, buffer head is unmapped + * + * It returns the error in case of allocation failure. + */ +int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, + unsigned long max_blocks, struct buffer_head *bh, + int create, int extend_disksize) { - handle_t *handle = ext4_journal_current_handle(); - int ret = 0; - unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; + int retval; + + clear_buffer_mapped(bh); + + /* + * Try to see if we can get the block without requesting + * for new file system block. + */ + down_read((&EXT4_I(inode)->i_data_sem)); + if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) { + retval = ext4_ext_get_blocks(handle, inode, block, max_blocks, + bh, 0, 0); + } else { + retval = ext4_get_blocks_handle(handle, + inode, block, max_blocks, bh, 0, 0); + } + up_read((&EXT4_I(inode)->i_data_sem)); + /* If it is only a block(s) look up */ if (!create) - goto get_block; /* A read */ + return retval; - if (max_blocks == 1) - goto get_block; /* A single block get */ + /* + * Returns if the blocks have already allocated + * + * Note that if blocks have been preallocated + * ext4_ext_get_block() returns th create = 0 + * with buffer head unmapped. + */ + if (retval > 0 && buffer_mapped(bh)) + return retval; - if (handle->h_transaction->t_state == T_LOCKED) { - /* - * Huge direct-io writes can hold off commits for long - * periods of time. Let this commit run. - */ - ext4_journal_stop(handle); - handle = ext4_journal_start(inode, DIO_CREDITS); - if (IS_ERR(handle)) - ret = PTR_ERR(handle); - goto get_block; + /* + * New blocks allocate and/or writing to uninitialized extent + * will possibly result in updating i_data, so we take + * the write lock of i_data_sem, and call get_blocks() + * with create == 1 flag. + */ + down_write((&EXT4_I(inode)->i_data_sem)); + /* + * We need to check for EXT4 here because migrate + * could have changed the inode type in between + */ + if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) { + retval = ext4_ext_get_blocks(handle, inode, block, max_blocks, + bh, create, extend_disksize); + } else { + retval = ext4_get_blocks_handle(handle, inode, block, + max_blocks, bh, create, extend_disksize); } + up_write((&EXT4_I(inode)->i_data_sem)); + return retval; +} - if (handle->h_buffer_credits <= EXT4_RESERVE_TRANS_BLOCKS) { - /* - * Getting low on buffer credits... - */ - ret = ext4_journal_extend(handle, DIO_CREDITS); - if (ret > 0) { - /* - * Couldn't extend the transaction. Start a new one. - */ - ret = ext4_journal_restart(handle, DIO_CREDITS); +static int ext4_get_block(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int create) +{ + handle_t *handle = ext4_journal_current_handle(); + int ret = 0, started = 0; + unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; + + if (create && !handle) { + /* Direct IO write... */ + if (max_blocks > DIO_MAX_BLOCKS) + max_blocks = DIO_MAX_BLOCKS; + handle = ext4_journal_start(inode, DIO_CREDITS + + 2 * EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb)); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; } + started = 1; } -get_block: - if (ret == 0) { - ret = ext4_get_blocks_wrap(handle, inode, iblock, + ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, bh_result, create, 0); - if (ret > 0) { - bh_result->b_size = (ret << inode->i_blkbits); - ret = 0; - } + if (ret > 0) { + bh_result->b_size = (ret << inode->i_blkbits); + ret = 0; } + if (started) + ext4_journal_stop(handle); +out: return ret; } @@ -996,7 +1026,7 @@ get_block: * `handle' can be NULL if create is zero */ struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode, - long block, int create, int *errp) + ext4_lblk_t block, int create, int *errp) { struct buffer_head dummy; int fatal = 0, err; @@ -1063,7 +1093,7 @@ err: } struct buffer_head *ext4_bread(handle_t *handle, struct inode *inode, - int block, int create, int *err) + ext4_lblk_t block, int create, int *err) { struct buffer_head * bh; @@ -1446,7 +1476,7 @@ static int jbd2_journal_dirty_data_fn(handle_t *handle, struct buffer_head *bh) * ext4_file_write() -> generic_file_write() -> __alloc_pages() -> ... * * Same applies to ext4_get_block(). We will deadlock on various things like - * lock_journal and i_truncate_mutex. + * lock_journal and i_data_sem * * Setting PF_MEMALLOC here doesn't work - too many internal memory * allocations fail. @@ -1678,7 +1708,8 @@ static int ext4_releasepage(struct page *page, gfp_t wait) * if the machine crashes during the write. * * If the O_DIRECT write is intantiating holes inside i_size and the machine - * crashes then stale disk data _may_ be exposed inside the file. + * crashes then stale disk data _may_ be exposed inside the file. But current + * VFS code falls back into buffered path in that case so we are safe. */ static ssize_t ext4_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, loff_t offset, @@ -1687,7 +1718,7 @@ static ssize_t ext4_direct_IO(int rw, struct kiocb *iocb, struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; struct ext4_inode_info *ei = EXT4_I(inode); - handle_t *handle = NULL; + handle_t *handle; ssize_t ret; int orphan = 0; size_t count = iov_length(iov, nr_segs); @@ -1695,17 +1726,21 @@ static ssize_t ext4_direct_IO(int rw, struct kiocb *iocb, if (rw == WRITE) { loff_t final_size = offset + count; - handle = ext4_journal_start(inode, DIO_CREDITS); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - goto out; - } if (final_size > inode->i_size) { + /* Credits for sb + inode write */ + handle = ext4_journal_start(inode, 2); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; + } ret = ext4_orphan_add(handle, inode); - if (ret) - goto out_stop; + if (ret) { + ext4_journal_stop(handle); + goto out; + } orphan = 1; ei->i_disksize = inode->i_size; + ext4_journal_stop(handle); } } @@ -1713,18 +1748,21 @@ static ssize_t ext4_direct_IO(int rw, struct kiocb *iocb, offset, nr_segs, ext4_get_block, NULL); - /* - * Reacquire the handle: ext4_get_block() can restart the transaction - */ - handle = ext4_journal_current_handle(); - -out_stop: - if (handle) { + if (orphan) { int err; - if (orphan && inode->i_nlink) + /* Credits for sb + inode write */ + handle = ext4_journal_start(inode, 2); + if (IS_ERR(handle)) { + /* This is really bad luck. We've written the data + * but cannot extend i_size. Bail out and pretend + * the write failed... */ + ret = PTR_ERR(handle); + goto out; + } + if (inode->i_nlink) ext4_orphan_del(handle, inode); - if (orphan && ret > 0) { + if (ret > 0) { loff_t end = offset + ret; if (end > inode->i_size) { ei->i_disksize = end; @@ -1828,7 +1866,8 @@ int ext4_block_truncate_page(handle_t *handle, struct page *page, { ext4_fsblk_t index = from >> PAGE_CACHE_SHIFT; unsigned offset = from & (PAGE_CACHE_SIZE-1); - unsigned blocksize, iblock, length, pos; + unsigned blocksize, length, pos; + ext4_lblk_t iblock; struct inode *inode = mapping->host; struct buffer_head *bh; int err = 0; @@ -1843,7 +1882,7 @@ int ext4_block_truncate_page(handle_t *handle, struct page *page, */ if (!page_has_buffers(page) && test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode) && PageUptodate(page)) { - zero_user_page(page, offset, length, KM_USER0); + zero_user(page, offset, length); set_page_dirty(page); goto unlock; } @@ -1896,7 +1935,7 @@ int ext4_block_truncate_page(handle_t *handle, struct page *page, goto unlock; } - zero_user_page(page, offset, length, KM_USER0); + zero_user(page, offset, length); BUFFER_TRACE(bh, "zeroed end of block"); @@ -1964,7 +2003,7 @@ static inline int all_zeroes(__le32 *p, __le32 *q) * (no partially truncated stuff there). */ static Indirect *ext4_find_shared(struct inode *inode, int depth, - int offsets[4], Indirect chain[4], __le32 *top) + ext4_lblk_t offsets[4], Indirect chain[4], __le32 *top) { Indirect *partial, *p; int k, err; @@ -2048,15 +2087,15 @@ static void ext4_clear_blocks(handle_t *handle, struct inode *inode, for (p = first; p < last; p++) { u32 nr = le32_to_cpu(*p); if (nr) { - struct buffer_head *bh; + struct buffer_head *tbh; *p = 0; - bh = sb_find_get_block(inode->i_sb, nr); - ext4_forget(handle, 0, inode, bh, nr); + tbh = sb_find_get_block(inode->i_sb, nr); + ext4_forget(handle, 0, inode, tbh, nr); } } - ext4_free_blocks(handle, inode, block_to_free, count); + ext4_free_blocks(handle, inode, block_to_free, count, 0); } /** @@ -2229,7 +2268,7 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode, ext4_journal_test_restart(handle, inode); } - ext4_free_blocks(handle, inode, nr, 1); + ext4_free_blocks(handle, inode, nr, 1, 1); if (parent_bh) { /* @@ -2289,12 +2328,12 @@ void ext4_truncate(struct inode *inode) __le32 *i_data = ei->i_data; int addr_per_block = EXT4_ADDR_PER_BLOCK(inode->i_sb); struct address_space *mapping = inode->i_mapping; - int offsets[4]; + ext4_lblk_t offsets[4]; Indirect chain[4]; Indirect *partial; __le32 nr = 0; int n; - long last_block; + ext4_lblk_t last_block; unsigned blocksize = inode->i_sb->s_blocksize; struct page *page; @@ -2320,8 +2359,10 @@ void ext4_truncate(struct inode *inode) return; } - if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) - return ext4_ext_truncate(inode, page); + if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) { + ext4_ext_truncate(inode, page); + return; + } handle = start_transaction(inode); if (IS_ERR(handle)) { @@ -2369,7 +2410,7 @@ void ext4_truncate(struct inode *inode) * From here we block out all ext4_get_block() callers who want to * modify the block allocation tree. */ - mutex_lock(&ei->truncate_mutex); + down_write(&ei->i_data_sem); if (n == 1) { /* direct blocks */ ext4_free_data(handle, inode, NULL, i_data+offsets[0], @@ -2433,7 +2474,7 @@ do_indirects: ext4_discard_reservation(inode); - mutex_unlock(&ei->truncate_mutex); + up_write(&ei->i_data_sem); inode->i_mtime = inode->i_ctime = ext4_current_time(inode); ext4_mark_inode_dirty(handle, inode); @@ -2460,7 +2501,8 @@ out_stop: static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb, unsigned long ino, struct ext4_iloc *iloc) { - unsigned long desc, group_desc, block_group; + unsigned long desc, group_desc; + ext4_group_t block_group; unsigned long offset; ext4_fsblk_t block; struct buffer_head *bh; @@ -2547,7 +2589,7 @@ static int __ext4_get_inode_loc(struct inode *inode, struct ext4_group_desc *desc; int inodes_per_buffer; int inode_offset, i; - int block_group; + ext4_group_t block_group; int start; block_group = (inode->i_ino - 1) / @@ -2660,22 +2702,54 @@ void ext4_get_inode_flags(struct ext4_inode_info *ei) if (flags & S_DIRSYNC) ei->i_flags |= EXT4_DIRSYNC_FL; } +static blkcnt_t ext4_inode_blocks(struct ext4_inode *raw_inode, + struct ext4_inode_info *ei) +{ + blkcnt_t i_blocks ; + struct inode *inode = &(ei->vfs_inode); + struct super_block *sb = inode->i_sb; + + if (EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_HUGE_FILE)) { + /* we are using combined 48 bit field */ + i_blocks = ((u64)le16_to_cpu(raw_inode->i_blocks_high)) << 32 | + le32_to_cpu(raw_inode->i_blocks_lo); + if (ei->i_flags & EXT4_HUGE_FILE_FL) { + /* i_blocks represent file system block size */ + return i_blocks << (inode->i_blkbits - 9); + } else { + return i_blocks; + } + } else { + return le32_to_cpu(raw_inode->i_blocks_lo); + } +} -void ext4_read_inode(struct inode * inode) +struct inode *ext4_iget(struct super_block *sb, unsigned long ino) { struct ext4_iloc iloc; struct ext4_inode *raw_inode; - struct ext4_inode_info *ei = EXT4_I(inode); + struct ext4_inode_info *ei; struct buffer_head *bh; + struct inode *inode; + long ret; int block; + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; + + ei = EXT4_I(inode); #ifdef CONFIG_EXT4DEV_FS_POSIX_ACL ei->i_acl = EXT4_ACL_NOT_CACHED; ei->i_default_acl = EXT4_ACL_NOT_CACHED; #endif ei->i_block_alloc_info = NULL; - if (__ext4_get_inode_loc(inode, &iloc, 0)) + ret = __ext4_get_inode_loc(inode, &iloc, 0); + if (ret < 0) goto bad_inode; bh = iloc.bh; raw_inode = ext4_raw_inode(&iloc); @@ -2687,7 +2761,6 @@ void ext4_read_inode(struct inode * inode) inode->i_gid |= le16_to_cpu(raw_inode->i_gid_high) << 16; } inode->i_nlink = le16_to_cpu(raw_inode->i_links_count); - inode->i_size = le32_to_cpu(raw_inode->i_size); ei->i_state = 0; ei->i_dir_start_lookup = 0; @@ -2702,6 +2775,7 @@ void ext4_read_inode(struct inode * inode) !(EXT4_SB(inode->i_sb)->s_mount_state & EXT4_ORPHAN_FS)) { /* this inode is deleted */ brelse (bh); + ret = -ESTALE; goto bad_inode; } /* The only unlinked inodes we let through here have @@ -2709,19 +2783,15 @@ void ext4_read_inode(struct inode * inode) * recovery code: that's fine, we're about to complete * the process of deleting those. */ } - inode->i_blocks = le32_to_cpu(raw_inode->i_blocks); ei->i_flags = le32_to_cpu(raw_inode->i_flags); - ei->i_file_acl = le32_to_cpu(raw_inode->i_file_acl); + inode->i_blocks = ext4_inode_blocks(raw_inode, ei); + ei->i_file_acl = le32_to_cpu(raw_inode->i_file_acl_lo); if (EXT4_SB(inode->i_sb)->s_es->s_creator_os != - cpu_to_le32(EXT4_OS_HURD)) + cpu_to_le32(EXT4_OS_HURD)) { ei->i_file_acl |= ((__u64)le16_to_cpu(raw_inode->i_file_acl_high)) << 32; - if (!S_ISREG(inode->i_mode)) { - ei->i_dir_acl = le32_to_cpu(raw_inode->i_dir_acl); - } else { - inode->i_size |= - ((__u64)le32_to_cpu(raw_inode->i_size_high)) << 32; } + inode->i_size = ext4_isize(raw_inode); ei->i_disksize = inode->i_size; inode->i_generation = le32_to_cpu(raw_inode->i_generation); ei->i_block_group = iloc.block_group; @@ -2733,17 +2803,12 @@ void ext4_read_inode(struct inode * inode) ei->i_data[block] = raw_inode->i_block[block]; INIT_LIST_HEAD(&ei->i_orphan); - if (inode->i_ino >= EXT4_FIRST_INO(inode->i_sb) + 1 && - EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE) { - /* - * When mke2fs creates big inodes it does not zero out - * the unused bytes above EXT4_GOOD_OLD_INODE_SIZE, - * so ignore those first few inodes. - */ + if (EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE) { ei->i_extra_isize = le16_to_cpu(raw_inode->i_extra_isize); if (EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize > EXT4_INODE_SIZE(inode->i_sb)) { brelse (bh); + ret = -EIO; goto bad_inode; } if (ei->i_extra_isize == 0) { @@ -2765,6 +2830,13 @@ void ext4_read_inode(struct inode * inode) EXT4_INODE_GET_XTIME(i_atime, inode, raw_inode); EXT4_EINODE_GET_XTIME(i_crtime, ei, raw_inode); + inode->i_version = le32_to_cpu(raw_inode->i_disk_version); + if (EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE) { + if (EXT4_FITS_IN_INODE(raw_inode, ei, i_version_hi)) + inode->i_version |= + (__u64)(le32_to_cpu(raw_inode->i_version_hi)) << 32; + } + if (S_ISREG(inode->i_mode)) { inode->i_op = &ext4_file_inode_operations; inode->i_fop = &ext4_file_operations; @@ -2790,11 +2862,61 @@ void ext4_read_inode(struct inode * inode) } brelse (iloc.bh); ext4_set_inode_flags(inode); - return; + unlock_new_inode(inode); + return inode; bad_inode: - make_bad_inode(inode); - return; + iget_failed(inode); + return ERR_PTR(ret); +} + +static int ext4_inode_blocks_set(handle_t *handle, + struct ext4_inode *raw_inode, + struct ext4_inode_info *ei) +{ + struct inode *inode = &(ei->vfs_inode); + u64 i_blocks = inode->i_blocks; + struct super_block *sb = inode->i_sb; + int err = 0; + + if (i_blocks <= ~0U) { + /* + * i_blocks can be represnted in a 32 bit variable + * as multiple of 512 bytes + */ + raw_inode->i_blocks_lo = cpu_to_le32(i_blocks); + raw_inode->i_blocks_high = 0; + ei->i_flags &= ~EXT4_HUGE_FILE_FL; + } else if (i_blocks <= 0xffffffffffffULL) { + /* + * i_blocks can be represented in a 48 bit variable + * as multiple of 512 bytes + */ + err = ext4_update_rocompat_feature(handle, sb, + EXT4_FEATURE_RO_COMPAT_HUGE_FILE); + if (err) + goto err_out; + /* i_block is stored in the split 48 bit fields */ + raw_inode->i_blocks_lo = cpu_to_le32(i_blocks); + raw_inode->i_blocks_high = cpu_to_le16(i_blocks >> 32); + ei->i_flags &= ~EXT4_HUGE_FILE_FL; + } else { + /* + * i_blocks should be represented in a 48 bit variable + * as multiple of file system block size + */ + err = ext4_update_rocompat_feature(handle, sb, + EXT4_FEATURE_RO_COMPAT_HUGE_FILE); + if (err) + goto err_out; + ei->i_flags |= EXT4_HUGE_FILE_FL; + /* i_block is stored in file system block size */ + i_blocks = i_blocks >> (inode->i_blkbits - 9); + raw_inode->i_blocks_lo = cpu_to_le32(i_blocks); + raw_inode->i_blocks_high = cpu_to_le16(i_blocks >> 32); + } +err_out: + return err; } /* @@ -2845,47 +2967,42 @@ static int ext4_do_update_inode(handle_t *handle, raw_inode->i_gid_high = 0; } raw_inode->i_links_count = cpu_to_le16(inode->i_nlink); - raw_inode->i_size = cpu_to_le32(ei->i_disksize); EXT4_INODE_SET_XTIME(i_ctime, inode, raw_inode); EXT4_INODE_SET_XTIME(i_mtime, inode, raw_inode); EXT4_INODE_SET_XTIME(i_atime, inode, raw_inode); EXT4_EINODE_SET_XTIME(i_crtime, ei, raw_inode); - raw_inode->i_blocks = cpu_to_le32(inode->i_blocks); + if (ext4_inode_blocks_set(handle, raw_inode, ei)) + goto out_brelse; raw_inode->i_dtime = cpu_to_le32(ei->i_dtime); raw_inode->i_flags = cpu_to_le32(ei->i_flags); if (EXT4_SB(inode->i_sb)->s_es->s_creator_os != cpu_to_le32(EXT4_OS_HURD)) raw_inode->i_file_acl_high = cpu_to_le16(ei->i_file_acl >> 32); - raw_inode->i_file_acl = cpu_to_le32(ei->i_file_acl); - if (!S_ISREG(inode->i_mode)) { - raw_inode->i_dir_acl = cpu_to_le32(ei->i_dir_acl); - } else { - raw_inode->i_size_high = - cpu_to_le32(ei->i_disksize >> 32); - if (ei->i_disksize > 0x7fffffffULL) { - struct super_block *sb = inode->i_sb; - if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, - EXT4_FEATURE_RO_COMPAT_LARGE_FILE) || - EXT4_SB(sb)->s_es->s_rev_level == - cpu_to_le32(EXT4_GOOD_OLD_REV)) { - /* If this is the first large file - * created, add a flag to the superblock. - */ - err = ext4_journal_get_write_access(handle, - EXT4_SB(sb)->s_sbh); - if (err) - goto out_brelse; - ext4_update_dynamic_rev(sb); - EXT4_SET_RO_COMPAT_FEATURE(sb, + raw_inode->i_file_acl_lo = cpu_to_le32(ei->i_file_acl); + ext4_isize_set(raw_inode, ei->i_disksize); + if (ei->i_disksize > 0x7fffffffULL) { + struct super_block *sb = inode->i_sb; + if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_LARGE_FILE) || + EXT4_SB(sb)->s_es->s_rev_level == + cpu_to_le32(EXT4_GOOD_OLD_REV)) { + /* If this is the first large file + * created, add a flag to the superblock. + */ + err = ext4_journal_get_write_access(handle, + EXT4_SB(sb)->s_sbh); + if (err) + goto out_brelse; + ext4_update_dynamic_rev(sb); + EXT4_SET_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_LARGE_FILE); - sb->s_dirt = 1; - handle->h_sync = 1; - err = ext4_journal_dirty_metadata(handle, - EXT4_SB(sb)->s_sbh); - } + sb->s_dirt = 1; + handle->h_sync = 1; + err = ext4_journal_dirty_metadata(handle, + EXT4_SB(sb)->s_sbh); } } raw_inode->i_generation = cpu_to_le32(inode->i_generation); @@ -2903,8 +3020,14 @@ static int ext4_do_update_inode(handle_t *handle, } else for (block = 0; block < EXT4_N_BLOCKS; block++) raw_inode->i_block[block] = ei->i_data[block]; - if (ei->i_extra_isize) + raw_inode->i_disk_version = cpu_to_le32(inode->i_version); + if (ei->i_extra_isize) { + if (EXT4_FITS_IN_INODE(raw_inode, ei, i_version_hi)) + raw_inode->i_version_hi = + cpu_to_le32(inode->i_version >> 32); raw_inode->i_extra_isize = cpu_to_le16(ei->i_extra_isize); + } + BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata"); rc = ext4_journal_dirty_metadata(handle, bh); @@ -3024,6 +3147,17 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) ext4_journal_stop(handle); } + if (attr->ia_valid & ATTR_SIZE) { + if (!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)) { + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + + if (attr->ia_size > sbi->s_bitmap_maxbytes) { + error = -EFBIG; + goto err_out; + } + } + } + if (S_ISREG(inode->i_mode) && attr->ia_valid & ATTR_SIZE && attr->ia_size < inode->i_size) { handle_t *handle; @@ -3120,6 +3254,9 @@ int ext4_mark_iloc_dirty(handle_t *handle, { int err = 0; + if (test_opt(inode->i_sb, I_VERSION)) + inode_inc_iversion(inode); + /* the do_update_inode consumes one bh->b_count */ get_bh(iloc->bh); @@ -3158,8 +3295,10 @@ ext4_reserve_inode_write(handle_t *handle, struct inode *inode, * Expand an inode by new_extra_isize bytes. * Returns 0 on success or negative error number on failure. */ -int ext4_expand_extra_isize(struct inode *inode, unsigned int new_extra_isize, - struct ext4_iloc iloc, handle_t *handle) +static int ext4_expand_extra_isize(struct inode *inode, + unsigned int new_extra_isize, + struct ext4_iloc iloc, + handle_t *handle) { struct ext4_inode *raw_inode; struct ext4_xattr_ibody_header *header; diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index e7f894bdb42..2ed7c37f897 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -199,7 +199,7 @@ flags_err: * need to allocate reservation structure for this inode * before set the window size */ - mutex_lock(&ei->truncate_mutex); + down_write(&ei->i_data_sem); if (!ei->i_block_alloc_info) ext4_init_block_alloc_info(inode); @@ -207,7 +207,7 @@ flags_err: struct ext4_reserve_window_node *rsv = &ei->i_block_alloc_info->rsv_window_node; rsv->rsv_goal_size = rsv_window_size; } - mutex_unlock(&ei->truncate_mutex); + up_write(&ei->i_data_sem); return 0; } case EXT4_IOC_GROUP_EXTEND: { @@ -254,6 +254,9 @@ flags_err: return err; } + case EXT4_IOC_MIGRATE: + return ext4_ext_migrate(inode, filp, cmd, arg); + default: return -ENOTTY; } diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c new file mode 100644 index 00000000000..ef97f19c2f9 --- /dev/null +++ b/fs/ext4/mballoc.c @@ -0,0 +1,4628 @@ +/* + * Copyright (c) 2003-2006, Cluster File Systems, Inc, info@clusterfs.com + * Written by Alex Tomas <alex@clusterfs.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public Licens + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111- + */ + + +/* + * mballoc.c contains the multiblocks allocation routines + */ + +#include <linux/time.h> +#include <linux/fs.h> +#include <linux/namei.h> +#include <linux/ext4_jbd2.h> +#include <linux/ext4_fs.h> +#include <linux/quotaops.h> +#include <linux/buffer_head.h> +#include <linux/module.h> +#include <linux/swap.h> +#include <linux/proc_fs.h> +#include <linux/pagemap.h> +#include <linux/seq_file.h> +#include <linux/version.h> +#include "group.h" + +/* + * MUSTDO: + * - test ext4_ext_search_left() and ext4_ext_search_right() + * - search for metadata in few groups + * + * TODO v4: + * - normalization should take into account whether file is still open + * - discard preallocations if no free space left (policy?) + * - don't normalize tails + * - quota + * - reservation for superuser + * + * TODO v3: + * - bitmap read-ahead (proposed by Oleg Drokin aka green) + * - track min/max extents in each group for better group selection + * - mb_mark_used() may allocate chunk right after splitting buddy + * - tree of groups sorted by number of free blocks + * - error handling + */ + +/* + * The allocation request involve request for multiple number of blocks + * near to the goal(block) value specified. + * + * During initialization phase of the allocator we decide to use the group + * preallocation or inode preallocation depending on the size file. The + * size of the file could be the resulting file size we would have after + * allocation or the current file size which ever is larger. If the size is + * less that sbi->s_mb_stream_request we select the group + * preallocation. The default value of s_mb_stream_request is 16 + * blocks. This can also be tuned via + * /proc/fs/ext4/<partition>/stream_req. The value is represented in terms + * of number of blocks. + * + * The main motivation for having small file use group preallocation is to + * ensure that we have small file closer in the disk. + * + * First stage the allocator looks at the inode prealloc list + * ext4_inode_info->i_prealloc_list contain list of prealloc spaces for + * this particular inode. The inode prealloc space is represented as: + * + * pa_lstart -> the logical start block for this prealloc space + * pa_pstart -> the physical start block for this prealloc space + * pa_len -> lenght for this prealloc space + * pa_free -> free space available in this prealloc space + * + * The inode preallocation space is used looking at the _logical_ start + * block. If only the logical file block falls within the range of prealloc + * space we will consume the particular prealloc space. This make sure that + * that the we have contiguous physical blocks representing the file blocks + * + * The important thing to be noted in case of inode prealloc space is that + * we don't modify the values associated to inode prealloc space except + * pa_free. + * + * If we are not able to find blocks in the inode prealloc space and if we + * have the group allocation flag set then we look at the locality group + * prealloc space. These are per CPU prealloc list repreasented as + * + * ext4_sb_info.s_locality_groups[smp_processor_id()] + * + * The reason for having a per cpu locality group is to reduce the contention + * between CPUs. It is possible to get scheduled at this point. + * + * The locality group prealloc space is used looking at whether we have + * enough free space (pa_free) withing the prealloc space. + * + * If we can't allocate blocks via inode prealloc or/and locality group + * prealloc then we look at the buddy cache. The buddy cache is represented + * by ext4_sb_info.s_buddy_cache (struct inode) whose file offset gets + * mapped to the buddy and bitmap information regarding different + * groups. The buddy information is attached to buddy cache inode so that + * we can access them through the page cache. The information regarding + * each group is loaded via ext4_mb_load_buddy. The information involve + * block bitmap and buddy information. The information are stored in the + * inode as: + * + * { page } + * [ group 0 buddy][ group 0 bitmap] [group 1][ group 1]... + * + * + * one block each for bitmap and buddy information. So for each group we + * take up 2 blocks. A page can contain blocks_per_page (PAGE_CACHE_SIZE / + * blocksize) blocks. So it can have information regarding groups_per_page + * which is blocks_per_page/2 + * + * The buddy cache inode is not stored on disk. The inode is thrown + * away when the filesystem is unmounted. + * + * We look for count number of blocks in the buddy cache. If we were able + * to locate that many free blocks we return with additional information + * regarding rest of the contiguous physical block available + * + * Before allocating blocks via buddy cache we normalize the request + * blocks. This ensure we ask for more blocks that we needed. The extra + * blocks that we get after allocation is added to the respective prealloc + * list. In case of inode preallocation we follow a list of heuristics + * based on file size. This can be found in ext4_mb_normalize_request. If + * we are doing a group prealloc we try to normalize the request to + * sbi->s_mb_group_prealloc. Default value of s_mb_group_prealloc is set to + * 512 blocks. This can be tuned via + * /proc/fs/ext4/<partition/group_prealloc. The value is represented in + * terms of number of blocks. If we have mounted the file system with -O + * stripe=<value> option the group prealloc request is normalized to the + * stripe value (sbi->s_stripe) + * + * The regular allocator(using the buddy cache) support few tunables. + * + * /proc/fs/ext4/<partition>/min_to_scan + * /proc/fs/ext4/<partition>/max_to_scan + * /proc/fs/ext4/<partition>/order2_req + * + * The regular allocator use buddy scan only if the request len is power of + * 2 blocks and the order of allocation is >= sbi->s_mb_order2_reqs. The + * value of s_mb_order2_reqs can be tuned via + * /proc/fs/ext4/<partition>/order2_req. If the request len is equal to + * stripe size (sbi->s_stripe), we try to search for contigous block in + * stripe size. This should result in better allocation on RAID setup. If + * not we search in the specific group using bitmap for best extents. The + * tunable min_to_scan and max_to_scan controll the behaviour here. + * min_to_scan indicate how long the mballoc __must__ look for a best + * extent and max_to_scanindicate how long the mballoc __can__ look for a + * best extent in the found extents. Searching for the blocks starts with + * the group specified as the goal value in allocation context via + * ac_g_ex. Each group is first checked based on the criteria whether it + * can used for allocation. ext4_mb_good_group explains how the groups are + * checked. + * + * Both the prealloc space are getting populated as above. So for the first + * request we will hit the buddy cache which will result in this prealloc + * space getting filled. The prealloc space is then later used for the + * subsequent request. + */ + +/* + * mballoc operates on the following data: + * - on-disk bitmap + * - in-core buddy (actually includes buddy and bitmap) + * - preallocation descriptors (PAs) + * + * there are two types of preallocations: + * - inode + * assiged to specific inode and can be used for this inode only. + * it describes part of inode's space preallocated to specific + * physical blocks. any block from that preallocated can be used + * independent. the descriptor just tracks number of blocks left + * unused. so, before taking some block from descriptor, one must + * make sure corresponded logical block isn't allocated yet. this + * also means that freeing any block within descriptor's range + * must discard all preallocated blocks. + * - locality group + * assigned to specific locality group which does not translate to + * permanent set of inodes: inode can join and leave group. space + * from this type of preallocation can be used for any inode. thus + * it's consumed from the beginning to the end. + * + * relation between them can be expressed as: + * in-core buddy = on-disk bitmap + preallocation descriptors + * + * this mean blocks mballoc considers used are: + * - allocated blocks (persistent) + * - preallocated blocks (non-persistent) + * + * consistency in mballoc world means that at any time a block is either + * free or used in ALL structures. notice: "any time" should not be read + * literally -- time is discrete and delimited by locks. + * + * to keep it simple, we don't use block numbers, instead we count number of + * blocks: how many blocks marked used/free in on-disk bitmap, buddy and PA. + * + * all operations can be expressed as: + * - init buddy: buddy = on-disk + PAs + * - new PA: buddy += N; PA = N + * - use inode PA: on-disk += N; PA -= N + * - discard inode PA buddy -= on-disk - PA; PA = 0 + * - use locality group PA on-disk += N; PA -= N + * - discard locality group PA buddy -= PA; PA = 0 + * note: 'buddy -= on-disk - PA' is used to show that on-disk bitmap + * is used in real operation because we can't know actual used + * bits from PA, only from on-disk bitmap + * + * if we follow this strict logic, then all operations above should be atomic. + * given some of them can block, we'd have to use something like semaphores + * killing performance on high-end SMP hardware. let's try to relax it using + * the following knowledge: + * 1) if buddy is referenced, it's already initialized + * 2) while block is used in buddy and the buddy is referenced, + * nobody can re-allocate that block + * 3) we work on bitmaps and '+' actually means 'set bits'. if on-disk has + * bit set and PA claims same block, it's OK. IOW, one can set bit in + * on-disk bitmap if buddy has same bit set or/and PA covers corresponded + * block + * + * so, now we're building a concurrency table: + * - init buddy vs. + * - new PA + * blocks for PA are allocated in the buddy, buddy must be referenced + * until PA is linked to allocation group to avoid concurrent buddy init + * - use inode PA + * we need to make sure that either on-disk bitmap or PA has uptodate data + * given (3) we care that PA-=N operation doesn't interfere with init + * - discard inode PA + * the simplest way would be to have buddy initialized by the discard + * - use locality group PA + * again PA-=N must be serialized with init + * - discard locality group PA + * the simplest way would be to have buddy initialized by the discard + * - new PA vs. + * - use inode PA + * i_data_sem serializes them + * - discard inode PA + * discard process must wait until PA isn't used by another process + * - use locality group PA + * some mutex should serialize them + * - discard locality group PA + * discard process must wait until PA isn't used by another process + * - use inode PA + * - use inode PA + * i_data_sem or another mutex should serializes them + * - discard inode PA + * discard process must wait until PA isn't used by another process + * - use locality group PA + * nothing wrong here -- they're different PAs covering different blocks + * - discard locality group PA + * discard process must wait until PA isn't used by another process + * + * now we're ready to make few consequences: + * - PA is referenced and while it is no discard is possible + * - PA is referenced until block isn't marked in on-disk bitmap + * - PA changes only after on-disk bitmap + * - discard must not compete with init. either init is done before + * any discard or they're serialized somehow + * - buddy init as sum of on-disk bitmap and PAs is done atomically + * + * a special case when we've used PA to emptiness. no need to modify buddy + * in this case, but we should care about concurrent init + * + */ + + /* + * Logic in few words: + * + * - allocation: + * load group + * find blocks + * mark bits in on-disk bitmap + * release group + * + * - use preallocation: + * find proper PA (per-inode or group) + * load group + * mark bits in on-disk bitmap + * release group + * release PA + * + * - free: + * load group + * mark bits in on-disk bitmap + * release group + * + * - discard preallocations in group: + * mark PAs deleted + * move them onto local list + * load on-disk bitmap + * load group + * remove PA from object (inode or locality group) + * mark free blocks in-core + * + * - discard inode's preallocations: + */ + +/* + * Locking rules + * + * Locks: + * - bitlock on a group (group) + * - object (inode/locality) (object) + * - per-pa lock (pa) + * + * Paths: + * - new pa + * object + * group + * + * - find and use pa: + * pa + * + * - release consumed pa: + * pa + * group + * object + * + * - generate in-core bitmap: + * group + * pa + * + * - discard all for given object (inode, locality group): + * object + * pa + * group + * + * - discard all for given group: + * group + * pa + * group + * object + * + */ + +/* + * with AGGRESSIVE_CHECK allocator runs consistency checks over + * structures. these checks slow things down a lot + */ +#define AGGRESSIVE_CHECK__ + +/* + * with DOUBLE_CHECK defined mballoc creates persistent in-core + * bitmaps, maintains and uses them to check for double allocations + */ +#define DOUBLE_CHECK__ + +/* + */ +#define MB_DEBUG__ +#ifdef MB_DEBUG +#define mb_debug(fmt, a...) printk(fmt, ##a) +#else +#define mb_debug(fmt, a...) +#endif + +/* + * with EXT4_MB_HISTORY mballoc stores last N allocations in memory + * and you can monitor it in /proc/fs/ext4/<dev>/mb_history + */ +#define EXT4_MB_HISTORY +#define EXT4_MB_HISTORY_ALLOC 1 /* allocation */ +#define EXT4_MB_HISTORY_PREALLOC 2 /* preallocated blocks used */ +#define EXT4_MB_HISTORY_DISCARD 4 /* preallocation discarded */ +#define EXT4_MB_HISTORY_FREE 8 /* free */ + +#define EXT4_MB_HISTORY_DEFAULT (EXT4_MB_HISTORY_ALLOC | \ + EXT4_MB_HISTORY_PREALLOC) + +/* + * How long mballoc can look for a best extent (in found extents) + */ +#define MB_DEFAULT_MAX_TO_SCAN 200 + +/* + * How long mballoc must look for a best extent + */ +#define MB_DEFAULT_MIN_TO_SCAN 10 + +/* + * How many groups mballoc will scan looking for the best chunk + */ +#define MB_DEFAULT_MAX_GROUPS_TO_SCAN 5 + +/* + * with 'ext4_mb_stats' allocator will collect stats that will be + * shown at umount. The collecting costs though! + */ +#define MB_DEFAULT_STATS 1 + +/* + * files smaller than MB_DEFAULT_STREAM_THRESHOLD are served + * by the stream allocator, which purpose is to pack requests + * as close each to other as possible to produce smooth I/O traffic + * We use locality group prealloc space for stream request. + * We can tune the same via /proc/fs/ext4/<parition>/stream_req + */ +#define MB_DEFAULT_STREAM_THRESHOLD 16 /* 64K */ + +/* + * for which requests use 2^N search using buddies + */ +#define MB_DEFAULT_ORDER2_REQS 2 + +/* + * default group prealloc size 512 blocks + */ +#define MB_DEFAULT_GROUP_PREALLOC 512 + +static struct kmem_cache *ext4_pspace_cachep; +static struct kmem_cache *ext4_ac_cachep; + +#ifdef EXT4_BB_MAX_BLOCKS +#undef EXT4_BB_MAX_BLOCKS +#endif +#define EXT4_BB_MAX_BLOCKS 30 + +struct ext4_free_metadata { + ext4_group_t group; + unsigned short num; + ext4_grpblk_t blocks[EXT4_BB_MAX_BLOCKS]; + struct list_head list; +}; + +struct ext4_group_info { + unsigned long bb_state; + unsigned long bb_tid; + struct ext4_free_metadata *bb_md_cur; + unsigned short bb_first_free; + unsigned short bb_free; + unsigned short bb_fragments; + struct list_head bb_prealloc_list; +#ifdef DOUBLE_CHECK + void *bb_bitmap; +#endif + unsigned short bb_counters[]; +}; + +#define EXT4_GROUP_INFO_NEED_INIT_BIT 0 +#define EXT4_GROUP_INFO_LOCKED_BIT 1 + +#define EXT4_MB_GRP_NEED_INIT(grp) \ + (test_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &((grp)->bb_state))) + + +struct ext4_prealloc_space { + struct list_head pa_inode_list; + struct list_head pa_group_list; + union { + struct list_head pa_tmp_list; + struct rcu_head pa_rcu; + } u; + spinlock_t pa_lock; + atomic_t pa_count; + unsigned pa_deleted; + ext4_fsblk_t pa_pstart; /* phys. block */ + ext4_lblk_t pa_lstart; /* log. block */ + unsigned short pa_len; /* len of preallocated chunk */ + unsigned short pa_free; /* how many blocks are free */ + unsigned short pa_linear; /* consumed in one direction + * strictly, for grp prealloc */ + spinlock_t *pa_obj_lock; + struct inode *pa_inode; /* hack, for history only */ +}; + + +struct ext4_free_extent { + ext4_lblk_t fe_logical; + ext4_grpblk_t fe_start; + ext4_group_t fe_group; + int fe_len; +}; + +/* + * Locality group: + * we try to group all related changes together + * so that writeback can flush/allocate them together as well + */ +struct ext4_locality_group { + /* for allocator */ + struct mutex lg_mutex; /* to serialize allocates */ + struct list_head lg_prealloc_list;/* list of preallocations */ + spinlock_t lg_prealloc_lock; +}; + +struct ext4_allocation_context { + struct inode *ac_inode; + struct super_block *ac_sb; + + /* original request */ + struct ext4_free_extent ac_o_ex; + + /* goal request (after normalization) */ + struct ext4_free_extent ac_g_ex; + + /* the best found extent */ + struct ext4_free_extent ac_b_ex; + + /* copy of the bext found extent taken before preallocation efforts */ + struct ext4_free_extent ac_f_ex; + + /* number of iterations done. we have to track to limit searching */ + unsigned long ac_ex_scanned; + __u16 ac_groups_scanned; + __u16 ac_found; + __u16 ac_tail; + __u16 ac_buddy; + __u16 ac_flags; /* allocation hints */ + __u8 ac_status; + __u8 ac_criteria; + __u8 ac_repeats; + __u8 ac_2order; /* if request is to allocate 2^N blocks and + * N > 0, the field stores N, otherwise 0 */ + __u8 ac_op; /* operation, for history only */ + struct page *ac_bitmap_page; + struct page *ac_buddy_page; + struct ext4_prealloc_space *ac_pa; + struct ext4_locality_group *ac_lg; +}; + +#define AC_STATUS_CONTINUE 1 +#define AC_STATUS_FOUND 2 +#define AC_STATUS_BREAK 3 + +struct ext4_mb_history { + struct ext4_free_extent orig; /* orig allocation */ + struct ext4_free_extent goal; /* goal allocation */ + struct ext4_free_extent result; /* result allocation */ + unsigned pid; + unsigned ino; + __u16 found; /* how many extents have been found */ + __u16 groups; /* how many groups have been scanned */ + __u16 tail; /* what tail broke some buddy */ + __u16 buddy; /* buddy the tail ^^^ broke */ + __u16 flags; + __u8 cr:3; /* which phase the result extent was found at */ + __u8 op:4; + __u8 merged:1; +}; + +struct ext4_buddy { + struct page *bd_buddy_page; + void *bd_buddy; + struct page *bd_bitmap_page; + void *bd_bitmap; + struct ext4_group_info *bd_info; + struct super_block *bd_sb; + __u16 bd_blkbits; + ext4_group_t bd_group; +}; +#define EXT4_MB_BITMAP(e4b) ((e4b)->bd_bitmap) +#define EXT4_MB_BUDDY(e4b) ((e4b)->bd_buddy) + +#ifndef EXT4_MB_HISTORY +static inline void ext4_mb_store_history(struct ext4_allocation_context *ac) +{ + return; +} +#else +static void ext4_mb_store_history(struct ext4_allocation_context *ac); +#endif + +#define in_range(b, first, len) ((b) >= (first) && (b) <= (first) + (len) - 1) + +static struct proc_dir_entry *proc_root_ext4; +struct buffer_head *read_block_bitmap(struct super_block *, ext4_group_t); +ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode, + ext4_fsblk_t goal, unsigned long *count, int *errp); + +static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap, + ext4_group_t group); +static void ext4_mb_poll_new_transaction(struct super_block *, handle_t *); +static void ext4_mb_free_committed_blocks(struct super_block *); +static void ext4_mb_return_to_preallocation(struct inode *inode, + struct ext4_buddy *e4b, sector_t block, + int count); +static void ext4_mb_put_pa(struct ext4_allocation_context *, + struct super_block *, struct ext4_prealloc_space *pa); +static int ext4_mb_init_per_dev_proc(struct super_block *sb); +static int ext4_mb_destroy_per_dev_proc(struct super_block *sb); + + +static inline void ext4_lock_group(struct super_block *sb, ext4_group_t group) +{ + struct ext4_group_info *grinfo = ext4_get_group_info(sb, group); + + bit_spin_lock(EXT4_GROUP_INFO_LOCKED_BIT, &(grinfo->bb_state)); +} + +static inline void ext4_unlock_group(struct super_block *sb, + ext4_group_t group) +{ + struct ext4_group_info *grinfo = ext4_get_group_info(sb, group); + + bit_spin_unlock(EXT4_GROUP_INFO_LOCKED_BIT, &(grinfo->bb_state)); +} + +static inline int ext4_is_group_locked(struct super_block *sb, + ext4_group_t group) +{ + struct ext4_group_info *grinfo = ext4_get_group_info(sb, group); + + return bit_spin_is_locked(EXT4_GROUP_INFO_LOCKED_BIT, + &(grinfo->bb_state)); +} + +static ext4_fsblk_t ext4_grp_offs_to_block(struct super_block *sb, + struct ext4_free_extent *fex) +{ + ext4_fsblk_t block; + + block = (ext4_fsblk_t) fex->fe_group * EXT4_BLOCKS_PER_GROUP(sb) + + fex->fe_start + + le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block); + return block; +} + +static inline void *mb_correct_addr_and_bit(int *bit, void *addr) +{ +#if BITS_PER_LONG == 64 + *bit += ((unsigned long) addr & 7UL) << 3; + addr = (void *) ((unsigned long) addr & ~7UL); +#elif BITS_PER_LONG == 32 + *bit += ((unsigned long) addr & 3UL) << 3; + addr = (void *) ((unsigned long) addr & ~3UL); +#else +#error "how many bits you are?!" +#endif + return addr; +} + +static inline int mb_test_bit(int bit, void *addr) +{ + /* + * ext4_test_bit on architecture like powerpc + * needs unsigned long aligned address + */ + addr = mb_correct_addr_and_bit(&bit, addr); + return ext4_test_bit(bit, addr); +} + +static inline void mb_set_bit(int bit, void *addr) +{ + addr = mb_correct_addr_and_bit(&bit, addr); + ext4_set_bit(bit, addr); +} + +static inline void mb_set_bit_atomic(spinlock_t *lock, int bit, void *addr) +{ + addr = mb_correct_addr_and_bit(&bit, addr); + ext4_set_bit_atomic(lock, bit, addr); +} + +static inline void mb_clear_bit(int bit, void *addr) +{ + addr = mb_correct_addr_and_bit(&bit, addr); + ext4_clear_bit(bit, addr); +} + +static inline void mb_clear_bit_atomic(spinlock_t *lock, int bit, void *addr) +{ + addr = mb_correct_addr_and_bit(&bit, addr); + ext4_clear_bit_atomic(lock, bit, addr); +} + +static inline int mb_find_next_zero_bit(void *addr, int max, int start) +{ + int fix = 0; + addr = mb_correct_addr_and_bit(&fix, addr); + max += fix; + start += fix; + + return ext4_find_next_zero_bit(addr, max, start) - fix; +} + +static inline int mb_find_next_bit(void *addr, int max, int start) +{ + int fix = 0; + addr = mb_correct_addr_and_bit(&fix, addr); + max += fix; + start += fix; + + return ext4_find_next_bit(addr, max, start) - fix; +} + +static void *mb_find_buddy(struct ext4_buddy *e4b, int order, int *max) +{ + char *bb; + + BUG_ON(EXT4_MB_BITMAP(e4b) == EXT4_MB_BUDDY(e4b)); + BUG_ON(max == NULL); + + if (order > e4b->bd_blkbits + 1) { + *max = 0; + return NULL; + } + + /* at order 0 we see each particular block */ + *max = 1 << (e4b->bd_blkbits + 3); + if (order == 0) + return EXT4_MB_BITMAP(e4b); + + bb = EXT4_MB_BUDDY(e4b) + EXT4_SB(e4b->bd_sb)->s_mb_offsets[order]; + *max = EXT4_SB(e4b->bd_sb)->s_mb_maxs[order]; + + return bb; +} + +#ifdef DOUBLE_CHECK +static void mb_free_blocks_double(struct inode *inode, struct ext4_buddy *e4b, + int first, int count) +{ + int i; + struct super_block *sb = e4b->bd_sb; + + if (unlikely(e4b->bd_info->bb_bitmap == NULL)) + return; + BUG_ON(!ext4_is_group_locked(sb, e4b->bd_group)); + for (i = 0; i < count; i++) { + if (!mb_test_bit(first + i, e4b->bd_info->bb_bitmap)) { + ext4_fsblk_t blocknr; + blocknr = e4b->bd_group * EXT4_BLOCKS_PER_GROUP(sb); + blocknr += first + i; + blocknr += + le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block); + + ext4_error(sb, __FUNCTION__, "double-free of inode" + " %lu's block %llu(bit %u in group %lu)\n", + inode ? inode->i_ino : 0, blocknr, + first + i, e4b->bd_group); + } + mb_clear_bit(first + i, e4b->bd_info->bb_bitmap); + } +} + +static void mb_mark_used_double(struct ext4_buddy *e4b, int first, int count) +{ + int i; + + if (unlikely(e4b->bd_info->bb_bitmap == NULL)) + return; + BUG_ON(!ext4_is_group_locked(e4b->bd_sb, e4b->bd_group)); + for (i = 0; i < count; i++) { + BUG_ON(mb_test_bit(first + i, e4b->bd_info->bb_bitmap)); + mb_set_bit(first + i, e4b->bd_info->bb_bitmap); + } +} + +static void mb_cmp_bitmaps(struct ext4_buddy *e4b, void *bitmap) +{ + if (memcmp(e4b->bd_info->bb_bitmap, bitmap, e4b->bd_sb->s_blocksize)) { + unsigned char *b1, *b2; + int i; + b1 = (unsigned char *) e4b->bd_info->bb_bitmap; + b2 = (unsigned char *) bitmap; + for (i = 0; i < e4b->bd_sb->s_blocksize; i++) { + if (b1[i] != b2[i]) { + printk("corruption in group %lu at byte %u(%u):" + " %x in copy != %x on disk/prealloc\n", + e4b->bd_group, i, i * 8, b1[i], b2[i]); + BUG(); + } + } + } +} + +#else +static inline void mb_free_blocks_double(struct inode *inode, + struct ext4_buddy *e4b, int first, int count) +{ + return; +} +static inline void mb_mark_used_double(struct ext4_buddy *e4b, + int first, int count) +{ + return; +} +static inline void mb_cmp_bitmaps(struct ext4_buddy *e4b, void *bitmap) +{ + return; +} +#endif + +#ifdef AGGRESSIVE_CHECK + +#define MB_CHECK_ASSERT(assert) \ +do { \ + if (!(assert)) { \ + printk(KERN_EMERG \ + "Assertion failure in %s() at %s:%d: \"%s\"\n", \ + function, file, line, # assert); \ + BUG(); \ + } \ +} while (0) + +static int __mb_check_buddy(struct ext4_buddy *e4b, char *file, + const char *function, int line) +{ + struct super_block *sb = e4b->bd_sb; + int order = e4b->bd_blkbits + 1; + int max; + int max2; + int i; + int j; + int k; + int count; + struct ext4_group_info *grp; + int fragments = 0; + int fstart; + struct list_head *cur; + void *buddy; + void *buddy2; + + if (!test_opt(sb, MBALLOC)) + return 0; + + { + static int mb_check_counter; + if (mb_check_counter++ % 100 != 0) + return 0; + } + + while (order > 1) { + buddy = mb_find_buddy(e4b, order, &max); + MB_CHECK_ASSERT(buddy); + buddy2 = mb_find_buddy(e4b, order - 1, &max2); + MB_CHECK_ASSERT(buddy2); + MB_CHECK_ASSERT(buddy != buddy2); + MB_CHECK_ASSERT(max * 2 == max2); + + count = 0; + for (i = 0; i < max; i++) { + + if (mb_test_bit(i, buddy)) { + /* only single bit in buddy2 may be 1 */ + if (!mb_test_bit(i << 1, buddy2)) { + MB_CHECK_ASSERT( + mb_test_bit((i<<1)+1, buddy2)); + } else if (!mb_test_bit((i << 1) + 1, buddy2)) { + MB_CHECK_ASSERT( + mb_test_bit(i << 1, buddy2)); + } + continue; + } + + /* both bits in buddy2 must be 0 */ + MB_CHECK_ASSERT(mb_test_bit(i << 1, buddy2)); + MB_CHECK_ASSERT(mb_test_bit((i << 1) + 1, buddy2)); + + for (j = 0; j < (1 << order); j++) { + k = (i * (1 << order)) + j; + MB_CHECK_ASSERT( + !mb_test_bit(k, EXT4_MB_BITMAP(e4b))); + } + count++; + } + MB_CHECK_ASSERT(e4b->bd_info->bb_counters[order] == count); + order--; + } + + fstart = -1; + buddy = mb_find_buddy(e4b, 0, &max); + for (i = 0; i < max; i++) { + if (!mb_test_bit(i, buddy)) { + MB_CHECK_ASSERT(i >= e4b->bd_info->bb_first_free); + if (fstart == -1) { + fragments++; + fstart = i; + } + continue; + } + fstart = -1; + /* check used bits only */ + for (j = 0; j < e4b->bd_blkbits + 1; j++) { + buddy2 = mb_find_buddy(e4b, j, &max2); + k = i >> j; + MB_CHECK_ASSERT(k < max2); + MB_CHECK_ASSERT(mb_test_bit(k, buddy2)); + } + } + MB_CHECK_ASSERT(!EXT4_MB_GRP_NEED_INIT(e4b->bd_info)); + MB_CHECK_ASSERT(e4b->bd_info->bb_fragments == fragments); + + grp = ext4_get_group_info(sb, e4b->bd_group); + buddy = mb_find_buddy(e4b, 0, &max); + list_for_each(cur, &grp->bb_prealloc_list) { + ext4_group_t groupnr; + struct ext4_prealloc_space *pa; + pa = list_entry(cur, struct ext4_prealloc_space, group_list); + ext4_get_group_no_and_offset(sb, pa->pstart, &groupnr, &k); + MB_CHECK_ASSERT(groupnr == e4b->bd_group); + for (i = 0; i < pa->len; i++) + MB_CHECK_ASSERT(mb_test_bit(k + i, buddy)); + } + return 0; +} +#undef MB_CHECK_ASSERT +#define mb_check_buddy(e4b) __mb_check_buddy(e4b, \ + __FILE__, __FUNCTION__, __LINE__) +#else +#define mb_check_buddy(e4b) +#endif + +/* FIXME!! need more doc */ +static void ext4_mb_mark_free_simple(struct super_block *sb, + void *buddy, unsigned first, int len, + struct ext4_group_info *grp) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + unsigned short min; + unsigned short max; + unsigned short chunk; + unsigned short border; + + BUG_ON(len > EXT4_BLOCKS_PER_GROUP(sb)); + + border = 2 << sb->s_blocksize_bits; + + while (len > 0) { + /* find how many blocks can be covered since this position */ + max = ffs(first | border) - 1; + + /* find how many blocks of power 2 we need to mark */ + min = fls(len) - 1; + + if (max < min) + min = max; + chunk = 1 << min; + + /* mark multiblock chunks only */ + grp->bb_counters[min]++; + if (min > 0) + mb_clear_bit(first >> min, + buddy + sbi->s_mb_offsets[min]); + + len -= chunk; + first += chunk; + } +} + +static void ext4_mb_generate_buddy(struct super_block *sb, + void *buddy, void *bitmap, ext4_group_t group) +{ + struct ext4_group_info *grp = ext4_get_group_info(sb, group); + unsigned short max = EXT4_BLOCKS_PER_GROUP(sb); + unsigned short i = 0; + unsigned short first; + unsigned short len; + unsigned free = 0; + unsigned fragments = 0; + unsigned long long period = get_cycles(); + + /* initialize buddy from bitmap which is aggregation + * of on-disk bitmap and preallocations */ + i = mb_find_next_zero_bit(bitmap, max, 0); + grp->bb_first_free = i; + while (i < max) { + fragments++; + first = i; + i = mb_find_next_bit(bitmap, max, i); + len = i - first; + free += len; + if (len > 1) + ext4_mb_mark_free_simple(sb, buddy, first, len, grp); + else + grp->bb_counters[0]++; + if (i < max) + i = mb_find_next_zero_bit(bitmap, max, i); + } + grp->bb_fragments = fragments; + + if (free != grp->bb_free) { + ext4_error(sb, __FUNCTION__, + "EXT4-fs: group %lu: %u blocks in bitmap, %u in gd\n", + group, free, grp->bb_free); + /* + * If we intent to continue, we consider group descritor + * corrupt and update bb_free using bitmap value + */ + grp->bb_free = free; + } + + clear_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &(grp->bb_state)); + + period = get_cycles() - period; + spin_lock(&EXT4_SB(sb)->s_bal_lock); + EXT4_SB(sb)->s_mb_buddies_generated++; + EXT4_SB(sb)->s_mb_generation_time += period; + spin_unlock(&EXT4_SB(sb)->s_bal_lock); +} + +/* The buddy information is attached the buddy cache inode + * for convenience. The information regarding each group + * is loaded via ext4_mb_load_buddy. The information involve + * block bitmap and buddy information. The information are + * stored in the inode as + * + * { page } + * [ group 0 buddy][ group 0 bitmap] [group 1][ group 1]... + * + * + * one block each for bitmap and buddy information. + * So for each group we take up 2 blocks. A page can + * contain blocks_per_page (PAGE_CACHE_SIZE / blocksize) blocks. + * So it can have information regarding groups_per_page which + * is blocks_per_page/2 + */ + +static int ext4_mb_init_cache(struct page *page, char *incore) +{ + int blocksize; + int blocks_per_page; + int groups_per_page; + int err = 0; + int i; + ext4_group_t first_group; + int first_block; + struct super_block *sb; + struct buffer_head *bhs; + struct buffer_head **bh; + struct inode *inode; + char *data; + char *bitmap; + + mb_debug("init page %lu\n", page->index); + + inode = page->mapping->host; + sb = inode->i_sb; + blocksize = 1 << inode->i_blkbits; + blocks_per_page = PAGE_CACHE_SIZE / blocksize; + + groups_per_page = blocks_per_page >> 1; + if (groups_per_page == 0) + groups_per_page = 1; + + /* allocate buffer_heads to read bitmaps */ + if (groups_per_page > 1) { + err = -ENOMEM; + i = sizeof(struct buffer_head *) * groups_per_page; + bh = kzalloc(i, GFP_NOFS); + if (bh == NULL) + goto out; + } else + bh = &bhs; + + first_group = page->index * blocks_per_page / 2; + + /* read all groups the page covers into the cache */ + for (i = 0; i < groups_per_page; i++) { + struct ext4_group_desc *desc; + + if (first_group + i >= EXT4_SB(sb)->s_groups_count) + break; + + err = -EIO; + desc = ext4_get_group_desc(sb, first_group + i, NULL); + if (desc == NULL) + goto out; + + err = -ENOMEM; + bh[i] = sb_getblk(sb, ext4_block_bitmap(sb, desc)); + if (bh[i] == NULL) + goto out; + + if (bh_uptodate_or_lock(bh[i])) + continue; + + if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) { + ext4_init_block_bitmap(sb, bh[i], + first_group + i, desc); + set_buffer_uptodate(bh[i]); + unlock_buffer(bh[i]); + continue; + } + get_bh(bh[i]); + bh[i]->b_end_io = end_buffer_read_sync; + submit_bh(READ, bh[i]); + mb_debug("read bitmap for group %lu\n", first_group + i); + } + + /* wait for I/O completion */ + for (i = 0; i < groups_per_page && bh[i]; i++) + wait_on_buffer(bh[i]); + + err = -EIO; + for (i = 0; i < groups_per_page && bh[i]; i++) + if (!buffer_uptodate(bh[i])) + goto out; + + first_block = page->index * blocks_per_page; + for (i = 0; i < blocks_per_page; i++) { + int group; + struct ext4_group_info *grinfo; + + group = (first_block + i) >> 1; + if (group >= EXT4_SB(sb)->s_groups_count) + break; + + /* + * data carry information regarding this + * particular group in the format specified + * above + * + */ + data = page_address(page) + (i * blocksize); + bitmap = bh[group - first_group]->b_data; + + /* + * We place the buddy block and bitmap block + * close together + */ + if ((first_block + i) & 1) { + /* this is block of buddy */ + BUG_ON(incore == NULL); + mb_debug("put buddy for group %u in page %lu/%x\n", + group, page->index, i * blocksize); + memset(data, 0xff, blocksize); + grinfo = ext4_get_group_info(sb, group); + grinfo->bb_fragments = 0; + memset(grinfo->bb_counters, 0, + sizeof(unsigned short)*(sb->s_blocksize_bits+2)); + /* + * incore got set to the group block bitmap below + */ + ext4_mb_generate_buddy(sb, data, incore, group); + incore = NULL; + } else { + /* this is block of bitmap */ + BUG_ON(incore != NULL); + mb_debug("put bitmap for group %u in page %lu/%x\n", + group, page->index, i * blocksize); + + /* see comments in ext4_mb_put_pa() */ + ext4_lock_group(sb, group); + memcpy(data, bitmap, blocksize); + + /* mark all preallocated blks used in in-core bitmap */ + ext4_mb_generate_from_pa(sb, data, group); + ext4_unlock_group(sb, group); + + /* set incore so that the buddy information can be + * generated using this + */ + incore = data; + } + } + SetPageUptodate(page); + +out: + if (bh) { + for (i = 0; i < groups_per_page && bh[i]; i++) + brelse(bh[i]); + if (bh != &bhs) + kfree(bh); + } + return err; +} + +static int ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group, + struct ext4_buddy *e4b) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct inode *inode = sbi->s_buddy_cache; + int blocks_per_page; + int block; + int pnum; + int poff; + struct page *page; + + mb_debug("load group %lu\n", group); + + blocks_per_page = PAGE_CACHE_SIZE / sb->s_blocksize; + + e4b->bd_blkbits = sb->s_blocksize_bits; + e4b->bd_info = ext4_get_group_info(sb, group); + e4b->bd_sb = sb; + e4b->bd_group = group; + e4b->bd_buddy_page = NULL; + e4b->bd_bitmap_page = NULL; + + /* + * the buddy cache inode stores the block bitmap + * and buddy information in consecutive blocks. + * So for each group we need two blocks. + */ + block = group * 2; + pnum = block / blocks_per_page; + poff = block % blocks_per_page; + + /* we could use find_or_create_page(), but it locks page + * what we'd like to avoid in fast path ... */ + page = find_get_page(inode->i_mapping, pnum); + if (page == NULL || !PageUptodate(page)) { + if (page) + page_cache_release(page); + page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS); + if (page) { + BUG_ON(page->mapping != inode->i_mapping); + if (!PageUptodate(page)) { + ext4_mb_init_cache(page, NULL); + mb_cmp_bitmaps(e4b, page_address(page) + + (poff * sb->s_blocksize)); + } + unlock_page(page); + } + } + if (page == NULL || !PageUptodate(page)) + goto err; + e4b->bd_bitmap_page = page; + e4b->bd_bitmap = page_address(page) + (poff * sb->s_blocksize); + mark_page_accessed(page); + + block++; + pnum = block / blocks_per_page; + poff = block % blocks_per_page; + + page = find_get_page(inode->i_mapping, pnum); + if (page == NULL || !PageUptodate(page)) { + if (page) + page_cache_release(page); + page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS); + if (page) { + BUG_ON(page->mapping != inode->i_mapping); + if (!PageUptodate(page)) + ext4_mb_init_cache(page, e4b->bd_bitmap); + + unlock_page(page); + } + } + if (page == NULL || !PageUptodate(page)) + goto err; + e4b->bd_buddy_page = page; + e4b->bd_buddy = page_address(page) + (poff * sb->s_blocksize); + mark_page_accessed(page); + + BUG_ON(e4b->bd_bitmap_page == NULL); + BUG_ON(e4b->bd_buddy_page == NULL); + + return 0; + +err: + if (e4b->bd_bitmap_page) + page_cache_release(e4b->bd_bitmap_page); + if (e4b->bd_buddy_page) + page_cache_release(e4b->bd_buddy_page); + e4b->bd_buddy = NULL; + e4b->bd_bitmap = NULL; + return -EIO; +} + +static void ext4_mb_release_desc(struct ext4_buddy *e4b) +{ + if (e4b->bd_bitmap_page) + page_cache_release(e4b->bd_bitmap_page); + if (e4b->bd_buddy_page) + page_cache_release(e4b->bd_buddy_page); +} + + +static int mb_find_order_for_block(struct ext4_buddy *e4b, int block) +{ + int order = 1; + void *bb; + + BUG_ON(EXT4_MB_BITMAP(e4b) == EXT4_MB_BUDDY(e4b)); + BUG_ON(block >= (1 << (e4b->bd_blkbits + 3))); + + bb = EXT4_MB_BUDDY(e4b); + while (order <= e4b->bd_blkbits + 1) { + block = block >> 1; + if (!mb_test_bit(block, bb)) { + /* this block is part of buddy of order 'order' */ + return order; + } + bb += 1 << (e4b->bd_blkbits - order); + order++; + } + return 0; +} + +static void mb_clear_bits(spinlock_t *lock, void *bm, int cur, int len) +{ + __u32 *addr; + + len = cur + len; + while (cur < len) { + if ((cur & 31) == 0 && (len - cur) >= 32) { + /* fast path: clear whole word at once */ + addr = bm + (cur >> 3); + *addr = 0; + cur += 32; + continue; + } + mb_clear_bit_atomic(lock, cur, bm); + cur++; + } +} + +static void mb_set_bits(spinlock_t *lock, void *bm, int cur, int len) +{ + __u32 *addr; + + len = cur + len; + while (cur < len) { + if ((cur & 31) == 0 && (len - cur) >= 32) { + /* fast path: set whole word at once */ + addr = bm + (cur >> 3); + *addr = 0xffffffff; + cur += 32; + continue; + } + mb_set_bit_atomic(lock, cur, bm); + cur++; + } +} + +static int mb_free_blocks(struct inode *inode, struct ext4_buddy *e4b, + int first, int count) +{ + int block = 0; + int max = 0; + int order; + void *buddy; + void *buddy2; + struct super_block *sb = e4b->bd_sb; + + BUG_ON(first + count > (sb->s_blocksize << 3)); + BUG_ON(!ext4_is_group_locked(sb, e4b->bd_group)); + mb_check_buddy(e4b); + mb_free_blocks_double(inode, e4b, first, count); + + e4b->bd_info->bb_free += count; + if (first < e4b->bd_info->bb_first_free) + e4b->bd_info->bb_first_free = first; + + /* let's maintain fragments counter */ + if (first != 0) + block = !mb_test_bit(first - 1, EXT4_MB_BITMAP(e4b)); + if (first + count < EXT4_SB(sb)->s_mb_maxs[0]) + max = !mb_test_bit(first + count, EXT4_MB_BITMAP(e4b)); + if (block && max) + e4b->bd_info->bb_fragments--; + else if (!block && !max) + e4b->bd_info->bb_fragments++; + + /* let's maintain buddy itself */ + while (count-- > 0) { + block = first++; + order = 0; + + if (!mb_test_bit(block, EXT4_MB_BITMAP(e4b))) { + ext4_fsblk_t blocknr; + blocknr = e4b->bd_group * EXT4_BLOCKS_PER_GROUP(sb); + blocknr += block; + blocknr += + le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block); + + ext4_error(sb, __FUNCTION__, "double-free of inode" + " %lu's block %llu(bit %u in group %lu)\n", + inode ? inode->i_ino : 0, blocknr, block, + e4b->bd_group); + } + mb_clear_bit(block, EXT4_MB_BITMAP(e4b)); + e4b->bd_info->bb_counters[order]++; + + /* start of the buddy */ + buddy = mb_find_buddy(e4b, order, &max); + + do { + block &= ~1UL; + if (mb_test_bit(block, buddy) || + mb_test_bit(block + 1, buddy)) + break; + + /* both the buddies are free, try to coalesce them */ + buddy2 = mb_find_buddy(e4b, order + 1, &max); + + if (!buddy2) + break; + + if (order > 0) { + /* for special purposes, we don't set + * free bits in bitmap */ + mb_set_bit(block, buddy); + mb_set_bit(block + 1, buddy); + } + e4b->bd_info->bb_counters[order]--; + e4b->bd_info->bb_counters[order]--; + + block = block >> 1; + order++; + e4b->bd_info->bb_counters[order]++; + + mb_clear_bit(block, buddy2); + buddy = buddy2; + } while (1); + } + mb_check_buddy(e4b); + + return 0; +} + +static int mb_find_extent(struct ext4_buddy *e4b, int order, int block, + int needed, struct ext4_free_extent *ex) +{ + int next = block; + int max; + int ord; + void *buddy; + + BUG_ON(!ext4_is_group_locked(e4b->bd_sb, e4b->bd_group)); + BUG_ON(ex == NULL); + + buddy = mb_find_buddy(e4b, order, &max); + BUG_ON(buddy == NULL); + BUG_ON(block >= max); + if (mb_test_bit(block, buddy)) { + ex->fe_len = 0; + ex->fe_start = 0; + ex->fe_group = 0; + return 0; + } + + /* FIXME dorp order completely ? */ + if (likely(order == 0)) { + /* find actual order */ + order = mb_find_order_for_block(e4b, block); + block = block >> order; + } + + ex->fe_len = 1 << order; + ex->fe_start = block << order; + ex->fe_group = e4b->bd_group; + + /* calc difference from given start */ + next = next - ex->fe_start; + ex->fe_len -= next; + ex->fe_start += next; + + while (needed > ex->fe_len && + (buddy = mb_find_buddy(e4b, order, &max))) { + + if (block + 1 >= max) + break; + + next = (block + 1) * (1 << order); + if (mb_test_bit(next, EXT4_MB_BITMAP(e4b))) + break; + + ord = mb_find_order_for_block(e4b, next); + + order = ord; + block = next >> order; + ex->fe_len += 1 << order; + } + + BUG_ON(ex->fe_start + ex->fe_len > (1 << (e4b->bd_blkbits + 3))); + return ex->fe_len; +} + +static int mb_mark_used(struct ext4_buddy *e4b, struct ext4_free_extent *ex) +{ + int ord; + int mlen = 0; + int max = 0; + int cur; + int start = ex->fe_start; + int len = ex->fe_len; + unsigned ret = 0; + int len0 = len; + void *buddy; + + BUG_ON(start + len > (e4b->bd_sb->s_blocksize << 3)); + BUG_ON(e4b->bd_group != ex->fe_group); + BUG_ON(!ext4_is_group_locked(e4b->bd_sb, e4b->bd_group)); + mb_check_buddy(e4b); + mb_mark_used_double(e4b, start, len); + + e4b->bd_info->bb_free -= len; + if (e4b->bd_info->bb_first_free == start) + e4b->bd_info->bb_first_free += len; + + /* let's maintain fragments counter */ + if (start != 0) + mlen = !mb_test_bit(start - 1, EXT4_MB_BITMAP(e4b)); + if (start + len < EXT4_SB(e4b->bd_sb)->s_mb_maxs[0]) + max = !mb_test_bit(start + len, EXT4_MB_BITMAP(e4b)); + if (mlen && max) + e4b->bd_info->bb_fragments++; + else if (!mlen && !max) + e4b->bd_info->bb_fragments--; + + /* let's maintain buddy itself */ + while (len) { + ord = mb_find_order_for_block(e4b, start); + + if (((start >> ord) << ord) == start && len >= (1 << ord)) { + /* the whole chunk may be allocated at once! */ + mlen = 1 << ord; + buddy = mb_find_buddy(e4b, ord, &max); + BUG_ON((start >> ord) >= max); + mb_set_bit(start >> ord, buddy); + e4b->bd_info->bb_counters[ord]--; + start += mlen; + len -= mlen; + BUG_ON(len < 0); + continue; + } + + /* store for history */ + if (ret == 0) + ret = len | (ord << 16); + + /* we have to split large buddy */ + BUG_ON(ord <= 0); + buddy = mb_find_buddy(e4b, ord, &max); + mb_set_bit(start >> ord, buddy); + e4b->bd_info->bb_counters[ord]--; + + ord--; + cur = (start >> ord) & ~1U; + buddy = mb_find_buddy(e4b, ord, &max); + mb_clear_bit(cur, buddy); + mb_clear_bit(cur + 1, buddy); + e4b->bd_info->bb_counters[ord]++; + e4b->bd_info->bb_counters[ord]++; + } + + mb_set_bits(sb_bgl_lock(EXT4_SB(e4b->bd_sb), ex->fe_group), + EXT4_MB_BITMAP(e4b), ex->fe_start, len0); + mb_check_buddy(e4b); + + return ret; +} + +/* + * Must be called under group lock! + */ +static void ext4_mb_use_best_found(struct ext4_allocation_context *ac, + struct ext4_buddy *e4b) +{ + struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb); + int ret; + + BUG_ON(ac->ac_b_ex.fe_group != e4b->bd_group); + BUG_ON(ac->ac_status == AC_STATUS_FOUND); + + ac->ac_b_ex.fe_len = min(ac->ac_b_ex.fe_len, ac->ac_g_ex.fe_len); + ac->ac_b_ex.fe_logical = ac->ac_g_ex.fe_logical; + ret = mb_mark_used(e4b, &ac->ac_b_ex); + + /* preallocation can change ac_b_ex, thus we store actually + * allocated blocks for history */ + ac->ac_f_ex = ac->ac_b_ex; + + ac->ac_status = AC_STATUS_FOUND; + ac->ac_tail = ret & 0xffff; + ac->ac_buddy = ret >> 16; + + /* XXXXXXX: SUCH A HORRIBLE **CK */ + /*FIXME!! Why ? */ + ac->ac_bitmap_page = e4b->bd_bitmap_page; + get_page(ac->ac_bitmap_page); + ac->ac_buddy_page = e4b->bd_buddy_page; + get_page(ac->ac_buddy_page); + + /* store last allocated for subsequent stream allocation */ + if ((ac->ac_flags & EXT4_MB_HINT_DATA)) { + spin_lock(&sbi->s_md_lock); + sbi->s_mb_last_group = ac->ac_f_ex.fe_group; + sbi->s_mb_last_start = ac->ac_f_ex.fe_start; + spin_unlock(&sbi->s_md_lock); + } +} + +/* + * regular allocator, for general purposes allocation + */ + +static void ext4_mb_check_limits(struct ext4_allocation_context *ac, + struct ext4_buddy *e4b, + int finish_group) +{ + struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb); + struct ext4_free_extent *bex = &ac->ac_b_ex; + struct ext4_free_extent *gex = &ac->ac_g_ex; + struct ext4_free_extent ex; + int max; + + /* + * We don't want to scan for a whole year + */ + if (ac->ac_found > sbi->s_mb_max_to_scan && + !(ac->ac_flags & EXT4_MB_HINT_FIRST)) { + ac->ac_status = AC_STATUS_BREAK; + return; + } + + /* + * Haven't found good chunk so far, let's continue + */ + if (bex->fe_len < gex->fe_len) + return; + + if ((finish_group || ac->ac_found > sbi->s_mb_min_to_scan) + && bex->fe_group == e4b->bd_group) { + /* recheck chunk's availability - we don't know + * when it was found (within this lock-unlock + * period or not) */ + max = mb_find_extent(e4b, 0, bex->fe_start, gex->fe_len, &ex); + if (max >= gex->fe_len) { + ext4_mb_use_best_found(ac, e4b); + return; + } + } +} + +/* + * The routine checks whether found extent is good enough. If it is, + * then the extent gets marked used and flag is set to the context + * to stop scanning. Otherwise, the extent is compared with the + * previous found extent and if new one is better, then it's stored + * in the context. Later, the best found extent will be used, if + * mballoc can't find good enough extent. + * + * FIXME: real allocation policy is to be designed yet! + */ +static void ext4_mb_measure_extent(struct ext4_allocation_context *ac, + struct ext4_free_extent *ex, + struct ext4_buddy *e4b) +{ + struct ext4_free_extent *bex = &ac->ac_b_ex; + struct ext4_free_extent *gex = &ac->ac_g_ex; + + BUG_ON(ex->fe_len <= 0); + BUG_ON(ex->fe_len >= EXT4_BLOCKS_PER_GROUP(ac->ac_sb)); + BUG_ON(ex->fe_start >= EXT4_BLOCKS_PER_GROUP(ac->ac_sb)); + BUG_ON(ac->ac_status != AC_STATUS_CONTINUE); + + ac->ac_found++; + + /* + * The special case - take what you catch first + */ + if (unlikely(ac->ac_flags & EXT4_MB_HINT_FIRST)) { + *bex = *ex; + ext4_mb_use_best_found(ac, e4b); + return; + } + + /* + * Let's check whether the chuck is good enough + */ + if (ex->fe_len == gex->fe_len) { + *bex = *ex; + ext4_mb_use_best_found(ac, e4b); + return; + } + + /* + * If this is first found extent, just store it in the context + */ + if (bex->fe_len == 0) { + *bex = *ex; + return; + } + + /* + * If new found extent is better, store it in the context + */ + if (bex->fe_len < gex->fe_len) { + /* if the request isn't satisfied, any found extent + * larger than previous best one is better */ + if (ex->fe_len > bex->fe_len) + *bex = *ex; + } else if (ex->fe_len > gex->fe_len) { + /* if the request is satisfied, then we try to find + * an extent that still satisfy the request, but is + * smaller than previous one */ + if (ex->fe_len < bex->fe_len) + *bex = *ex; + } + + ext4_mb_check_limits(ac, e4b, 0); +} + +static int ext4_mb_try_best_found(struct ext4_allocation_context *ac, + struct ext4_buddy *e4b) +{ + struct ext4_free_extent ex = ac->ac_b_ex; + ext4_group_t group = ex.fe_group; + int max; + int err; + + BUG_ON(ex.fe_len <= 0); + err = ext4_mb_load_buddy(ac->ac_sb, group, e4b); + if (err) + return err; + + ext4_lock_group(ac->ac_sb, group); + max = mb_find_extent(e4b, 0, ex.fe_start, ex.fe_len, &ex); + + if (max > 0) { + ac->ac_b_ex = ex; + ext4_mb_use_best_found(ac, e4b); + } + + ext4_unlock_group(ac->ac_sb, group); + ext4_mb_release_desc(e4b); + + return 0; +} + +static int ext4_mb_find_by_goal(struct ext4_allocation_context *ac, + struct ext4_buddy *e4b) +{ + ext4_group_t group = ac->ac_g_ex.fe_group; + int max; + int err; + struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb); + struct ext4_super_block *es = sbi->s_es; + struct ext4_free_extent ex; + + if (!(ac->ac_flags & EXT4_MB_HINT_TRY_GOAL)) + return 0; + + err = ext4_mb_load_buddy(ac->ac_sb, group, e4b); + if (err) + return err; + + ext4_lock_group(ac->ac_sb, group); + max = mb_find_extent(e4b, 0, ac->ac_g_ex.fe_start, + ac->ac_g_ex.fe_len, &ex); + + if (max >= ac->ac_g_ex.fe_len && ac->ac_g_ex.fe_len == sbi->s_stripe) { + ext4_fsblk_t start; + + start = (e4b->bd_group * EXT4_BLOCKS_PER_GROUP(ac->ac_sb)) + + ex.fe_start + le32_to_cpu(es->s_first_data_block); + /* use do_div to get remainder (would be 64-bit modulo) */ + if (do_div(start, sbi->s_stripe) == 0) { + ac->ac_found++; + ac->ac_b_ex = ex; + ext4_mb_use_best_found(ac, e4b); + } + } else if (max >= ac->ac_g_ex.fe_len) { + BUG_ON(ex.fe_len <= 0); + BUG_ON(ex.fe_group != ac->ac_g_ex.fe_group); + BUG_ON(ex.fe_start != ac->ac_g_ex.fe_start); + ac->ac_found++; + ac->ac_b_ex = ex; + ext4_mb_use_best_found(ac, e4b); + } else if (max > 0 && (ac->ac_flags & EXT4_MB_HINT_MERGE)) { + /* Sometimes, caller may want to merge even small + * number of blocks to an existing extent */ + BUG_ON(ex.fe_len <= 0); + BUG_ON(ex.fe_group != ac->ac_g_ex.fe_group); + BUG_ON(ex.fe_start != ac->ac_g_ex.fe_start); + ac->ac_found++; + ac->ac_b_ex = ex; + ext4_mb_use_best_found(ac, e4b); + } + ext4_unlock_group(ac->ac_sb, group); + ext4_mb_release_desc(e4b); + + return 0; +} + +/* + * The routine scans buddy structures (not bitmap!) from given order + * to max order and tries to find big enough chunk to satisfy the req + */ +static void ext4_mb_simple_scan_group(struct ext4_allocation_context *ac, + struct ext4_buddy *e4b) +{ + struct super_block *sb = ac->ac_sb; + struct ext4_group_info *grp = e4b->bd_info; + void *buddy; + int i; + int k; + int max; + + BUG_ON(ac->ac_2order <= 0); + for (i = ac->ac_2order; i <= sb->s_blocksize_bits + 1; i++) { + if (grp->bb_counters[i] == 0) + continue; + + buddy = mb_find_buddy(e4b, i, &max); + BUG_ON(buddy == NULL); + + k = mb_find_next_zero_bit(buddy, max, 0); + BUG_ON(k >= max); + + ac->ac_found++; + + ac->ac_b_ex.fe_len = 1 << i; + ac->ac_b_ex.fe_start = k << i; + ac->ac_b_ex.fe_group = e4b->bd_group; + + ext4_mb_use_best_found(ac, e4b); + + BUG_ON(ac->ac_b_ex.fe_len != ac->ac_g_ex.fe_len); + + if (EXT4_SB(sb)->s_mb_stats) + atomic_inc(&EXT4_SB(sb)->s_bal_2orders); + + break; + } +} + +/* + * The routine scans the group and measures all found extents. + * In order to optimize scanning, caller must pass number of + * free blocks in the group, so the routine can know upper limit. + */ +static void ext4_mb_complex_scan_group(struct ext4_allocation_context *ac, + struct ext4_buddy *e4b) +{ + struct super_block *sb = ac->ac_sb; + void *bitmap = EXT4_MB_BITMAP(e4b); + struct ext4_free_extent ex; + int i; + int free; + + free = e4b->bd_info->bb_free; + BUG_ON(free <= 0); + + i = e4b->bd_info->bb_first_free; + + while (free && ac->ac_status == AC_STATUS_CONTINUE) { + i = mb_find_next_zero_bit(bitmap, + EXT4_BLOCKS_PER_GROUP(sb), i); + if (i >= EXT4_BLOCKS_PER_GROUP(sb)) { + /* + * IF we have corrupt bitmap, we won't find any + * free blocks even though group info says we + * we have free blocks + */ + ext4_error(sb, __FUNCTION__, "%d free blocks as per " + "group info. But bitmap says 0\n", + free); + break; + } + + mb_find_extent(e4b, 0, i, ac->ac_g_ex.fe_len, &ex); + BUG_ON(ex.fe_len <= 0); + if (free < ex.fe_len) { + ext4_error(sb, __FUNCTION__, "%d free blocks as per " + "group info. But got %d blocks\n", + free, ex.fe_len); + /* + * The number of free blocks differs. This mostly + * indicate that the bitmap is corrupt. So exit + * without claiming the space. + */ + break; + } + + ext4_mb_measure_extent(ac, &ex, e4b); + + i += ex.fe_len; + free -= ex.fe_len; + } + + ext4_mb_check_limits(ac, e4b, 1); +} + +/* + * This is a special case for storages like raid5 + * we try to find stripe-aligned chunks for stripe-size requests + * XXX should do so at least for multiples of stripe size as well + */ +static void ext4_mb_scan_aligned(struct ext4_allocation_context *ac, + struct ext4_buddy *e4b) +{ + struct super_block *sb = ac->ac_sb; + struct ext4_sb_info *sbi = EXT4_SB(sb); + void *bitmap = EXT4_MB_BITMAP(e4b); + struct ext4_free_extent ex; + ext4_fsblk_t first_group_block; + ext4_fsblk_t a; + ext4_grpblk_t i; + int max; + + BUG_ON(sbi->s_stripe == 0); + + /* find first stripe-aligned block in group */ + first_group_block = e4b->bd_group * EXT4_BLOCKS_PER_GROUP(sb) + + le32_to_cpu(sbi->s_es->s_first_data_block); + a = first_group_block + sbi->s_stripe - 1; + do_div(a, sbi->s_stripe); + i = (a * sbi->s_stripe) - first_group_block; + + while (i < EXT4_BLOCKS_PER_GROUP(sb)) { + if (!mb_test_bit(i, bitmap)) { + max = mb_find_extent(e4b, 0, i, sbi->s_stripe, &ex); + if (max >= sbi->s_stripe) { + ac->ac_found++; + ac->ac_b_ex = ex; + ext4_mb_use_best_found(ac, e4b); + break; + } + } + i += sbi->s_stripe; + } +} + +static int ext4_mb_good_group(struct ext4_allocation_context *ac, + ext4_group_t group, int cr) +{ + unsigned free, fragments; + unsigned i, bits; + struct ext4_group_desc *desc; + struct ext4_group_info *grp = ext4_get_group_info(ac->ac_sb, group); + + BUG_ON(cr < 0 || cr >= 4); + BUG_ON(EXT4_MB_GRP_NEED_INIT(grp)); + + free = grp->bb_free; + fragments = grp->bb_fragments; + if (free == 0) + return 0; + if (fragments == 0) + return 0; + + switch (cr) { + case 0: + BUG_ON(ac->ac_2order == 0); + /* If this group is uninitialized, skip it initially */ + desc = ext4_get_group_desc(ac->ac_sb, group, NULL); + if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) + return 0; + + bits = ac->ac_sb->s_blocksize_bits + 1; + for (i = ac->ac_2order; i <= bits; i++) + if (grp->bb_counters[i] > 0) + return 1; + break; + case 1: + if ((free / fragments) >= ac->ac_g_ex.fe_len) + return 1; + break; + case 2: + if (free >= ac->ac_g_ex.fe_len) + return 1; + break; + case 3: + return 1; + default: + BUG(); + } + + return 0; +} + +static int ext4_mb_regular_allocator(struct ext4_allocation_context *ac) +{ + ext4_group_t group; + ext4_group_t i; + int cr; + int err = 0; + int bsbits; + struct ext4_sb_info *sbi; + struct super_block *sb; + struct ext4_buddy e4b; + loff_t size, isize; + + sb = ac->ac_sb; + sbi = EXT4_SB(sb); + BUG_ON(ac->ac_status == AC_STATUS_FOUND); + + /* first, try the goal */ + err = ext4_mb_find_by_goal(ac, &e4b); + if (err || ac->ac_status == AC_STATUS_FOUND) + goto out; + + if (unlikely(ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY)) + goto out; + + /* + * ac->ac2_order is set only if the fe_len is a power of 2 + * if ac2_order is set we also set criteria to 0 so that we + * try exact allocation using buddy. + */ + i = fls(ac->ac_g_ex.fe_len); + ac->ac_2order = 0; + /* + * We search using buddy data only if the order of the request + * is greater than equal to the sbi_s_mb_order2_reqs + * You can tune it via /proc/fs/ext4/<partition>/order2_req + */ + if (i >= sbi->s_mb_order2_reqs) { + /* + * This should tell if fe_len is exactly power of 2 + */ + if ((ac->ac_g_ex.fe_len & (~(1 << (i - 1)))) == 0) + ac->ac_2order = i - 1; + } + + bsbits = ac->ac_sb->s_blocksize_bits; + /* if stream allocation is enabled, use global goal */ + size = ac->ac_o_ex.fe_logical + ac->ac_o_ex.fe_len; + isize = i_size_read(ac->ac_inode) >> bsbits; + if (size < isize) + size = isize; + + if (size < sbi->s_mb_stream_request && + (ac->ac_flags & EXT4_MB_HINT_DATA)) { + /* TBD: may be hot point */ + spin_lock(&sbi->s_md_lock); + ac->ac_g_ex.fe_group = sbi->s_mb_last_group; + ac->ac_g_ex.fe_start = sbi->s_mb_last_start; + spin_unlock(&sbi->s_md_lock); + } + + /* searching for the right group start from the goal value specified */ + group = ac->ac_g_ex.fe_group; + + /* Let's just scan groups to find more-less suitable blocks */ + cr = ac->ac_2order ? 0 : 1; + /* + * cr == 0 try to get exact allocation, + * cr == 3 try to get anything + */ +repeat: + for (; cr < 4 && ac->ac_status == AC_STATUS_CONTINUE; cr++) { + ac->ac_criteria = cr; + for (i = 0; i < EXT4_SB(sb)->s_groups_count; group++, i++) { + struct ext4_group_info *grp; + struct ext4_group_desc *desc; + + if (group == EXT4_SB(sb)->s_groups_count) + group = 0; + + /* quick check to skip empty groups */ + grp = ext4_get_group_info(ac->ac_sb, group); + if (grp->bb_free == 0) + continue; + + /* + * if the group is already init we check whether it is + * a good group and if not we don't load the buddy + */ + if (EXT4_MB_GRP_NEED_INIT(grp)) { + /* + * we need full data about the group + * to make a good selection + */ + err = ext4_mb_load_buddy(sb, group, &e4b); + if (err) + goto out; + ext4_mb_release_desc(&e4b); + } + + /* + * If the particular group doesn't satisfy our + * criteria we continue with the next group + */ + if (!ext4_mb_good_group(ac, group, cr)) + continue; + + err = ext4_mb_load_buddy(sb, group, &e4b); + if (err) + goto out; + + ext4_lock_group(sb, group); + if (!ext4_mb_good_group(ac, group, cr)) { + /* someone did allocation from this group */ + ext4_unlock_group(sb, group); + ext4_mb_release_desc(&e4b); + continue; + } + + ac->ac_groups_scanned++; + desc = ext4_get_group_desc(sb, group, NULL); + if (cr == 0 || (desc->bg_flags & + cpu_to_le16(EXT4_BG_BLOCK_UNINIT) && + ac->ac_2order != 0)) + ext4_mb_simple_scan_group(ac, &e4b); + else if (cr == 1 && + ac->ac_g_ex.fe_len == sbi->s_stripe) + ext4_mb_scan_aligned(ac, &e4b); + else + ext4_mb_complex_scan_group(ac, &e4b); + + ext4_unlock_group(sb, group); + ext4_mb_release_desc(&e4b); + + if (ac->ac_status != AC_STATUS_CONTINUE) + break; + } + } + + if (ac->ac_b_ex.fe_len > 0 && ac->ac_status != AC_STATUS_FOUND && + !(ac->ac_flags & EXT4_MB_HINT_FIRST)) { + /* + * We've been searching too long. Let's try to allocate + * the best chunk we've found so far + */ + + ext4_mb_try_best_found(ac, &e4b); + if (ac->ac_status != AC_STATUS_FOUND) { + /* + * Someone more lucky has already allocated it. + * The only thing we can do is just take first + * found block(s) + printk(KERN_DEBUG "EXT4-fs: someone won our chunk\n"); + */ + ac->ac_b_ex.fe_group = 0; + ac->ac_b_ex.fe_start = 0; + ac->ac_b_ex.fe_len = 0; + ac->ac_status = AC_STATUS_CONTINUE; + ac->ac_flags |= EXT4_MB_HINT_FIRST; + cr = 3; + atomic_inc(&sbi->s_mb_lost_chunks); + goto repeat; + } + } +out: + return err; +} + +#ifdef EXT4_MB_HISTORY +struct ext4_mb_proc_session { + struct ext4_mb_history *history; + struct super_block *sb; + int start; + int max; +}; + +static void *ext4_mb_history_skip_empty(struct ext4_mb_proc_session *s, + struct ext4_mb_history *hs, + int first) +{ + if (hs == s->history + s->max) + hs = s->history; + if (!first && hs == s->history + s->start) + return NULL; + while (hs->orig.fe_len == 0) { + hs++; + if (hs == s->history + s->max) + hs = s->history; + if (hs == s->history + s->start) + return NULL; + } + return hs; +} + +static void *ext4_mb_seq_history_start(struct seq_file *seq, loff_t *pos) +{ + struct ext4_mb_proc_session *s = seq->private; + struct ext4_mb_history *hs; + int l = *pos; + + if (l == 0) + return SEQ_START_TOKEN; + hs = ext4_mb_history_skip_empty(s, s->history + s->start, 1); + if (!hs) + return NULL; + while (--l && (hs = ext4_mb_history_skip_empty(s, ++hs, 0)) != NULL); + return hs; +} + +static void *ext4_mb_seq_history_next(struct seq_file *seq, void *v, + loff_t *pos) +{ + struct ext4_mb_proc_session *s = seq->private; + struct ext4_mb_history *hs = v; + + ++*pos; + if (v == SEQ_START_TOKEN) + return ext4_mb_history_skip_empty(s, s->history + s->start, 1); + else + return ext4_mb_history_skip_empty(s, ++hs, 0); +} + +static int ext4_mb_seq_history_show(struct seq_file *seq, void *v) +{ + char buf[25], buf2[25], buf3[25], *fmt; + struct ext4_mb_history *hs = v; + + if (v == SEQ_START_TOKEN) { + seq_printf(seq, "%-5s %-8s %-23s %-23s %-23s %-5s " + "%-5s %-2s %-5s %-5s %-5s %-6s\n", + "pid", "inode", "original", "goal", "result", "found", + "grps", "cr", "flags", "merge", "tail", "broken"); + return 0; + } + + if (hs->op == EXT4_MB_HISTORY_ALLOC) { + fmt = "%-5u %-8u %-23s %-23s %-23s %-5u %-5u %-2u " + "%-5u %-5s %-5u %-6u\n"; + sprintf(buf2, "%lu/%d/%u@%u", hs->result.fe_group, + hs->result.fe_start, hs->result.fe_len, + hs->result.fe_logical); + sprintf(buf, "%lu/%d/%u@%u", hs->orig.fe_group, + hs->orig.fe_start, hs->orig.fe_len, + hs->orig.fe_logical); + sprintf(buf3, "%lu/%d/%u@%u", hs->goal.fe_group, + hs->goal.fe_start, hs->goal.fe_len, + hs->goal.fe_logical); + seq_printf(seq, fmt, hs->pid, hs->ino, buf, buf3, buf2, + hs->found, hs->groups, hs->cr, hs->flags, + hs->merged ? "M" : "", hs->tail, + hs->buddy ? 1 << hs->buddy : 0); + } else if (hs->op == EXT4_MB_HISTORY_PREALLOC) { + fmt = "%-5u %-8u %-23s %-23s %-23s\n"; + sprintf(buf2, "%lu/%d/%u@%u", hs->result.fe_group, + hs->result.fe_start, hs->result.fe_len, + hs->result.fe_logical); + sprintf(buf, "%lu/%d/%u@%u", hs->orig.fe_group, + hs->orig.fe_start, hs->orig.fe_len, + hs->orig.fe_logical); + seq_printf(seq, fmt, hs->pid, hs->ino, buf, "", buf2); + } else if (hs->op == EXT4_MB_HISTORY_DISCARD) { + sprintf(buf2, "%lu/%d/%u", hs->result.fe_group, + hs->result.fe_start, hs->result.fe_len); + seq_printf(seq, "%-5u %-8u %-23s discard\n", + hs->pid, hs->ino, buf2); + } else if (hs->op == EXT4_MB_HISTORY_FREE) { + sprintf(buf2, "%lu/%d/%u", hs->result.fe_group, + hs->result.fe_start, hs->result.fe_len); + seq_printf(seq, "%-5u %-8u %-23s free\n", + hs->pid, hs->ino, buf2); + } + return 0; +} + +static void ext4_mb_seq_history_stop(struct seq_file *seq, void *v) +{ +} + +static struct seq_operations ext4_mb_seq_history_ops = { + .start = ext4_mb_seq_history_start, + .next = ext4_mb_seq_history_next, + .stop = ext4_mb_seq_history_stop, + .show = ext4_mb_seq_history_show, +}; + +static int ext4_mb_seq_history_open(struct inode *inode, struct file *file) +{ + struct super_block *sb = PDE(inode)->data; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_mb_proc_session *s; + int rc; + int size; + + s = kmalloc(sizeof(*s), GFP_KERNEL); + if (s == NULL) + return -ENOMEM; + s->sb = sb; + size = sizeof(struct ext4_mb_history) * sbi->s_mb_history_max; + s->history = kmalloc(size, GFP_KERNEL); + if (s->history == NULL) { + kfree(s); + return -ENOMEM; + } + + spin_lock(&sbi->s_mb_history_lock); + memcpy(s->history, sbi->s_mb_history, size); + s->max = sbi->s_mb_history_max; + s->start = sbi->s_mb_history_cur % s->max; + spin_unlock(&sbi->s_mb_history_lock); + + rc = seq_open(file, &ext4_mb_seq_history_ops); + if (rc == 0) { + struct seq_file *m = (struct seq_file *)file->private_data; + m->private = s; + } else { + kfree(s->history); + kfree(s); + } + return rc; + +} + +static int ext4_mb_seq_history_release(struct inode *inode, struct file *file) +{ + struct seq_file *seq = (struct seq_file *)file->private_data; + struct ext4_mb_proc_session *s = seq->private; + kfree(s->history); + kfree(s); + return seq_release(inode, file); +} + +static ssize_t ext4_mb_seq_history_write(struct file *file, + const char __user *buffer, + size_t count, loff_t *ppos) +{ + struct seq_file *seq = (struct seq_file *)file->private_data; + struct ext4_mb_proc_session *s = seq->private; + struct super_block *sb = s->sb; + char str[32]; + int value; + + if (count >= sizeof(str)) { + printk(KERN_ERR "EXT4-fs: %s string too long, max %u bytes\n", + "mb_history", (int)sizeof(str)); + return -EOVERFLOW; + } + + if (copy_from_user(str, buffer, count)) + return -EFAULT; + + value = simple_strtol(str, NULL, 0); + if (value < 0) + return -ERANGE; + EXT4_SB(sb)->s_mb_history_filter = value; + + return count; +} + +static struct file_operations ext4_mb_seq_history_fops = { + .owner = THIS_MODULE, + .open = ext4_mb_seq_history_open, + .read = seq_read, + .write = ext4_mb_seq_history_write, + .llseek = seq_lseek, + .release = ext4_mb_seq_history_release, +}; + +static void *ext4_mb_seq_groups_start(struct seq_file *seq, loff_t *pos) +{ + struct super_block *sb = seq->private; + struct ext4_sb_info *sbi = EXT4_SB(sb); + ext4_group_t group; + + if (*pos < 0 || *pos >= sbi->s_groups_count) + return NULL; + + group = *pos + 1; + return (void *) group; +} + +static void *ext4_mb_seq_groups_next(struct seq_file *seq, void *v, loff_t *pos) +{ + struct super_block *sb = seq->private; + struct ext4_sb_info *sbi = EXT4_SB(sb); + ext4_group_t group; + + ++*pos; + if (*pos < 0 || *pos >= sbi->s_groups_count) + return NULL; + group = *pos + 1; + return (void *) group;; +} + +static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v) +{ + struct super_block *sb = seq->private; + long group = (long) v; + int i; + int err; + struct ext4_buddy e4b; + struct sg { + struct ext4_group_info info; + unsigned short counters[16]; + } sg; + + group--; + if (group == 0) + seq_printf(seq, "#%-5s: %-5s %-5s %-5s " + "[ %-5s %-5s %-5s %-5s %-5s %-5s %-5s " + "%-5s %-5s %-5s %-5s %-5s %-5s %-5s ]\n", + "group", "free", "frags", "first", + "2^0", "2^1", "2^2", "2^3", "2^4", "2^5", "2^6", + "2^7", "2^8", "2^9", "2^10", "2^11", "2^12", "2^13"); + + i = (sb->s_blocksize_bits + 2) * sizeof(sg.info.bb_counters[0]) + + sizeof(struct ext4_group_info); + err = ext4_mb_load_buddy(sb, group, &e4b); + if (err) { + seq_printf(seq, "#%-5lu: I/O error\n", group); + return 0; + } + ext4_lock_group(sb, group); + memcpy(&sg, ext4_get_group_info(sb, group), i); + ext4_unlock_group(sb, group); + ext4_mb_release_desc(&e4b); + + seq_printf(seq, "#%-5lu: %-5u %-5u %-5u [", group, sg.info.bb_free, + sg.info.bb_fragments, sg.info.bb_first_free); + for (i = 0; i <= 13; i++) + seq_printf(seq, " %-5u", i <= sb->s_blocksize_bits + 1 ? + sg.info.bb_counters[i] : 0); + seq_printf(seq, " ]\n"); + + return 0; +} + +static void ext4_mb_seq_groups_stop(struct seq_file *seq, void *v) +{ +} + +static struct seq_operations ext4_mb_seq_groups_ops = { + .start = ext4_mb_seq_groups_start, + .next = ext4_mb_seq_groups_next, + .stop = ext4_mb_seq_groups_stop, + .show = ext4_mb_seq_groups_show, +}; + +static int ext4_mb_seq_groups_open(struct inode *inode, struct file *file) +{ + struct super_block *sb = PDE(inode)->data; + int rc; + + rc = seq_open(file, &ext4_mb_seq_groups_ops); + if (rc == 0) { + struct seq_file *m = (struct seq_file *)file->private_data; + m->private = sb; + } + return rc; + +} + +static struct file_operations ext4_mb_seq_groups_fops = { + .owner = THIS_MODULE, + .open = ext4_mb_seq_groups_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +static void ext4_mb_history_release(struct super_block *sb) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + + remove_proc_entry("mb_groups", sbi->s_mb_proc); + remove_proc_entry("mb_history", sbi->s_mb_proc); + + kfree(sbi->s_mb_history); +} + +static void ext4_mb_history_init(struct super_block *sb) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + int i; + + if (sbi->s_mb_proc != NULL) { + struct proc_dir_entry *p; + p = create_proc_entry("mb_history", S_IRUGO, sbi->s_mb_proc); + if (p) { + p->proc_fops = &ext4_mb_seq_history_fops; + p->data = sb; + } + p = create_proc_entry("mb_groups", S_IRUGO, sbi->s_mb_proc); + if (p) { + p->proc_fops = &ext4_mb_seq_groups_fops; + p->data = sb; + } + } + + sbi->s_mb_history_max = 1000; + sbi->s_mb_history_cur = 0; + spin_lock_init(&sbi->s_mb_history_lock); + i = sbi->s_mb_history_max * sizeof(struct ext4_mb_history); + sbi->s_mb_history = kmalloc(i, GFP_KERNEL); + if (likely(sbi->s_mb_history != NULL)) + memset(sbi->s_mb_history, 0, i); + /* if we can't allocate history, then we simple won't use it */ +} + +static void ext4_mb_store_history(struct ext4_allocation_context *ac) +{ + struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb); + struct ext4_mb_history h; + + if (unlikely(sbi->s_mb_history == NULL)) + return; + + if (!(ac->ac_op & sbi->s_mb_history_filter)) + return; + + h.op = ac->ac_op; + h.pid = current->pid; + h.ino = ac->ac_inode ? ac->ac_inode->i_ino : 0; + h.orig = ac->ac_o_ex; + h.result = ac->ac_b_ex; + h.flags = ac->ac_flags; + h.found = ac->ac_found; + h.groups = ac->ac_groups_scanned; + h.cr = ac->ac_criteria; + h.tail = ac->ac_tail; + h.buddy = ac->ac_buddy; + h.merged = 0; + if (ac->ac_op == EXT4_MB_HISTORY_ALLOC) { + if (ac->ac_g_ex.fe_start == ac->ac_b_ex.fe_start && + ac->ac_g_ex.fe_group == ac->ac_b_ex.fe_group) + h.merged = 1; + h.goal = ac->ac_g_ex; + h.result = ac->ac_f_ex; + } + + spin_lock(&sbi->s_mb_history_lock); + memcpy(sbi->s_mb_history + sbi->s_mb_history_cur, &h, sizeof(h)); + if (++sbi->s_mb_history_cur >= sbi->s_mb_history_max) + sbi->s_mb_history_cur = 0; + spin_unlock(&sbi->s_mb_history_lock); +} + +#else +#define ext4_mb_history_release(sb) +#define ext4_mb_history_init(sb) +#endif + +static int ext4_mb_init_backend(struct super_block *sb) +{ + ext4_group_t i; + int j, len, metalen; + struct ext4_sb_info *sbi = EXT4_SB(sb); + int num_meta_group_infos = + (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) >> + EXT4_DESC_PER_BLOCK_BITS(sb); + struct ext4_group_info **meta_group_info; + + /* An 8TB filesystem with 64-bit pointers requires a 4096 byte + * kmalloc. A 128kb malloc should suffice for a 256TB filesystem. + * So a two level scheme suffices for now. */ + sbi->s_group_info = kmalloc(sizeof(*sbi->s_group_info) * + num_meta_group_infos, GFP_KERNEL); + if (sbi->s_group_info == NULL) { + printk(KERN_ERR "EXT4-fs: can't allocate buddy meta group\n"); + return -ENOMEM; + } + sbi->s_buddy_cache = new_inode(sb); + if (sbi->s_buddy_cache == NULL) { + printk(KERN_ERR "EXT4-fs: can't get new inode\n"); + goto err_freesgi; + } + EXT4_I(sbi->s_buddy_cache)->i_disksize = 0; + + metalen = sizeof(*meta_group_info) << EXT4_DESC_PER_BLOCK_BITS(sb); + for (i = 0; i < num_meta_group_infos; i++) { + if ((i + 1) == num_meta_group_infos) + metalen = sizeof(*meta_group_info) * + (sbi->s_groups_count - + (i << EXT4_DESC_PER_BLOCK_BITS(sb))); + meta_group_info = kmalloc(metalen, GFP_KERNEL); + if (meta_group_info == NULL) { + printk(KERN_ERR "EXT4-fs: can't allocate mem for a " + "buddy group\n"); + goto err_freemeta; + } + sbi->s_group_info[i] = meta_group_info; + } + + /* + * calculate needed size. if change bb_counters size, + * don't forget about ext4_mb_generate_buddy() + */ + len = sizeof(struct ext4_group_info); + len += sizeof(unsigned short) * (sb->s_blocksize_bits + 2); + for (i = 0; i < sbi->s_groups_count; i++) { + struct ext4_group_desc *desc; + + meta_group_info = + sbi->s_group_info[i >> EXT4_DESC_PER_BLOCK_BITS(sb)]; + j = i & (EXT4_DESC_PER_BLOCK(sb) - 1); + + meta_group_info[j] = kzalloc(len, GFP_KERNEL); + if (meta_group_info[j] == NULL) { + printk(KERN_ERR "EXT4-fs: can't allocate buddy mem\n"); + i--; + goto err_freebuddy; + } + desc = ext4_get_group_desc(sb, i, NULL); + if (desc == NULL) { + printk(KERN_ERR + "EXT4-fs: can't read descriptor %lu\n", i); + goto err_freebuddy; + } + memset(meta_group_info[j], 0, len); + set_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, + &(meta_group_info[j]->bb_state)); + + /* + * initialize bb_free to be able to skip + * empty groups without initialization + */ + if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) { + meta_group_info[j]->bb_free = + ext4_free_blocks_after_init(sb, i, desc); + } else { + meta_group_info[j]->bb_free = + le16_to_cpu(desc->bg_free_blocks_count); + } + + INIT_LIST_HEAD(&meta_group_info[j]->bb_prealloc_list); + +#ifdef DOUBLE_CHECK + { + struct buffer_head *bh; + meta_group_info[j]->bb_bitmap = + kmalloc(sb->s_blocksize, GFP_KERNEL); + BUG_ON(meta_group_info[j]->bb_bitmap == NULL); + bh = read_block_bitmap(sb, i); + BUG_ON(bh == NULL); + memcpy(meta_group_info[j]->bb_bitmap, bh->b_data, + sb->s_blocksize); + put_bh(bh); + } +#endif + + } + + return 0; + +err_freebuddy: + while (i >= 0) { + kfree(ext4_get_group_info(sb, i)); + i--; + } + i = num_meta_group_infos; +err_freemeta: + while (--i >= 0) + kfree(sbi->s_group_info[i]); + iput(sbi->s_buddy_cache); +err_freesgi: + kfree(sbi->s_group_info); + return -ENOMEM; +} + +int ext4_mb_init(struct super_block *sb, int needs_recovery) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + unsigned i; + unsigned offset; + unsigned max; + + if (!test_opt(sb, MBALLOC)) + return 0; + + i = (sb->s_blocksize_bits + 2) * sizeof(unsigned short); + + sbi->s_mb_offsets = kmalloc(i, GFP_KERNEL); + if (sbi->s_mb_offsets == NULL) { + clear_opt(sbi->s_mount_opt, MBALLOC); + return -ENOMEM; + } + sbi->s_mb_maxs = kmalloc(i, GFP_KERNEL); + if (sbi->s_mb_maxs == NULL) { + clear_opt(sbi->s_mount_opt, MBALLOC); + kfree(sbi->s_mb_maxs); + return -ENOMEM; + } + + /* order 0 is regular bitmap */ + sbi->s_mb_maxs[0] = sb->s_blocksize << 3; + sbi->s_mb_offsets[0] = 0; + + i = 1; + offset = 0; + max = sb->s_blocksize << 2; + do { + sbi->s_mb_offsets[i] = offset; + sbi->s_mb_maxs[i] = max; + offset += 1 << (sb->s_blocksize_bits - i); + max = max >> 1; + i++; + } while (i <= sb->s_blocksize_bits + 1); + + /* init file for buddy data */ + i = ext4_mb_init_backend(sb); + if (i) { + clear_opt(sbi->s_mount_opt, MBALLOC); + kfree(sbi->s_mb_offsets); + kfree(sbi->s_mb_maxs); + return i; + } + + spin_lock_init(&sbi->s_md_lock); + INIT_LIST_HEAD(&sbi->s_active_transaction); + INIT_LIST_HEAD(&sbi->s_closed_transaction); + INIT_LIST_HEAD(&sbi->s_committed_transaction); + spin_lock_init(&sbi->s_bal_lock); + + sbi->s_mb_max_to_scan = MB_DEFAULT_MAX_TO_SCAN; + sbi->s_mb_min_to_scan = MB_DEFAULT_MIN_TO_SCAN; + sbi->s_mb_stats = MB_DEFAULT_STATS; + sbi->s_mb_stream_request = MB_DEFAULT_STREAM_THRESHOLD; + sbi->s_mb_order2_reqs = MB_DEFAULT_ORDER2_REQS; + sbi->s_mb_history_filter = EXT4_MB_HISTORY_DEFAULT; + sbi->s_mb_group_prealloc = MB_DEFAULT_GROUP_PREALLOC; + + i = sizeof(struct ext4_locality_group) * NR_CPUS; + sbi->s_locality_groups = kmalloc(i, GFP_KERNEL); + if (sbi->s_locality_groups == NULL) { + clear_opt(sbi->s_mount_opt, MBALLOC); + kfree(sbi->s_mb_offsets); + kfree(sbi->s_mb_maxs); + return -ENOMEM; + } + for (i = 0; i < NR_CPUS; i++) { + struct ext4_locality_group *lg; + lg = &sbi->s_locality_groups[i]; + mutex_init(&lg->lg_mutex); + INIT_LIST_HEAD(&lg->lg_prealloc_list); + spin_lock_init(&lg->lg_prealloc_lock); + } + + ext4_mb_init_per_dev_proc(sb); + ext4_mb_history_init(sb); + + printk("EXT4-fs: mballoc enabled\n"); + return 0; +} + +/* need to called with ext4 group lock (ext4_lock_group) */ +static void ext4_mb_cleanup_pa(struct ext4_group_info *grp) +{ + struct ext4_prealloc_space *pa; + struct list_head *cur, *tmp; + int count = 0; + + list_for_each_safe(cur, tmp, &grp->bb_prealloc_list) { + pa = list_entry(cur, struct ext4_prealloc_space, pa_group_list); + list_del(&pa->pa_group_list); + count++; + kfree(pa); + } + if (count) + mb_debug("mballoc: %u PAs left\n", count); + +} + +int ext4_mb_release(struct super_block *sb) +{ + ext4_group_t i; + int num_meta_group_infos; + struct ext4_group_info *grinfo; + struct ext4_sb_info *sbi = EXT4_SB(sb); + + if (!test_opt(sb, MBALLOC)) + return 0; + + /* release freed, non-committed blocks */ + spin_lock(&sbi->s_md_lock); + list_splice_init(&sbi->s_closed_transaction, + &sbi->s_committed_transaction); + list_splice_init(&sbi->s_active_transaction, + &sbi->s_committed_transaction); + spin_unlock(&sbi->s_md_lock); + ext4_mb_free_committed_blocks(sb); + + if (sbi->s_group_info) { + for (i = 0; i < sbi->s_groups_count; i++) { + grinfo = ext4_get_group_info(sb, i); +#ifdef DOUBLE_CHECK + kfree(grinfo->bb_bitmap); +#endif + ext4_lock_group(sb, i); + ext4_mb_cleanup_pa(grinfo); + ext4_unlock_group(sb, i); + kfree(grinfo); + } + num_meta_group_infos = (sbi->s_groups_count + + EXT4_DESC_PER_BLOCK(sb) - 1) >> + EXT4_DESC_PER_BLOCK_BITS(sb); + for (i = 0; i < num_meta_group_infos; i++) + kfree(sbi->s_group_info[i]); + kfree(sbi->s_group_info); + } + kfree(sbi->s_mb_offsets); + kfree(sbi->s_mb_maxs); + if (sbi->s_buddy_cache) + iput(sbi->s_buddy_cache); + if (sbi->s_mb_stats) { + printk(KERN_INFO + "EXT4-fs: mballoc: %u blocks %u reqs (%u success)\n", + atomic_read(&sbi->s_bal_allocated), + atomic_read(&sbi->s_bal_reqs), + atomic_read(&sbi->s_bal_success)); + printk(KERN_INFO + "EXT4-fs: mballoc: %u extents scanned, %u goal hits, " + "%u 2^N hits, %u breaks, %u lost\n", + atomic_read(&sbi->s_bal_ex_scanned), + atomic_read(&sbi->s_bal_goals), + atomic_read(&sbi->s_bal_2orders), + atomic_read(&sbi->s_bal_breaks), + atomic_read(&sbi->s_mb_lost_chunks)); + printk(KERN_INFO + "EXT4-fs: mballoc: %lu generated and it took %Lu\n", + sbi->s_mb_buddies_generated++, + sbi->s_mb_generation_time); + printk(KERN_INFO + "EXT4-fs: mballoc: %u preallocated, %u discarded\n", + atomic_read(&sbi->s_mb_preallocated), + atomic_read(&sbi->s_mb_discarded)); + } + + kfree(sbi->s_locality_groups); + + ext4_mb_history_release(sb); + ext4_mb_destroy_per_dev_proc(sb); + + return 0; +} + +static void ext4_mb_free_committed_blocks(struct super_block *sb) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + int err; + int i; + int count = 0; + int count2 = 0; + struct ext4_free_metadata *md; + struct ext4_buddy e4b; + + if (list_empty(&sbi->s_committed_transaction)) + return; + + /* there is committed blocks to be freed yet */ + do { + /* get next array of blocks */ + md = NULL; + spin_lock(&sbi->s_md_lock); + if (!list_empty(&sbi->s_committed_transaction)) { + md = list_entry(sbi->s_committed_transaction.next, + struct ext4_free_metadata, list); + list_del(&md->list); + } + spin_unlock(&sbi->s_md_lock); + + if (md == NULL) + break; + + mb_debug("gonna free %u blocks in group %lu (0x%p):", + md->num, md->group, md); + + err = ext4_mb_load_buddy(sb, md->group, &e4b); + /* we expect to find existing buddy because it's pinned */ + BUG_ON(err != 0); + + /* there are blocks to put in buddy to make them really free */ + count += md->num; + count2++; + ext4_lock_group(sb, md->group); + for (i = 0; i < md->num; i++) { + mb_debug(" %u", md->blocks[i]); + err = mb_free_blocks(NULL, &e4b, md->blocks[i], 1); + BUG_ON(err != 0); + } + mb_debug("\n"); + ext4_unlock_group(sb, md->group); + + /* balance refcounts from ext4_mb_free_metadata() */ + page_cache_release(e4b.bd_buddy_page); + page_cache_release(e4b.bd_bitmap_page); + + kfree(md); + ext4_mb_release_desc(&e4b); + + } while (md); + + mb_debug("freed %u blocks in %u structures\n", count, count2); +} + +#define EXT4_ROOT "ext4" +#define EXT4_MB_STATS_NAME "stats" +#define EXT4_MB_MAX_TO_SCAN_NAME "max_to_scan" +#define EXT4_MB_MIN_TO_SCAN_NAME "min_to_scan" +#define EXT4_MB_ORDER2_REQ "order2_req" +#define EXT4_MB_STREAM_REQ "stream_req" +#define EXT4_MB_GROUP_PREALLOC "group_prealloc" + + + +#define MB_PROC_VALUE_READ(name) \ +static int ext4_mb_read_##name(char *page, char **start, \ + off_t off, int count, int *eof, void *data) \ +{ \ + struct ext4_sb_info *sbi = data; \ + int len; \ + *eof = 1; \ + if (off != 0) \ + return 0; \ + len = sprintf(page, "%ld\n", sbi->s_mb_##name); \ + *start = page; \ + return len; \ +} + +#define MB_PROC_VALUE_WRITE(name) \ +static int ext4_mb_write_##name(struct file *file, \ + const char __user *buf, unsigned long cnt, void *data) \ +{ \ + struct ext4_sb_info *sbi = data; \ + char str[32]; \ + long value; \ + if (cnt >= sizeof(str)) \ + return -EINVAL; \ + if (copy_from_user(str, buf, cnt)) \ + return -EFAULT; \ + value = simple_strtol(str, NULL, 0); \ + if (value <= 0) \ + return -ERANGE; \ + sbi->s_mb_##name = value; \ + return cnt; \ +} + +MB_PROC_VALUE_READ(stats); +MB_PROC_VALUE_WRITE(stats); +MB_PROC_VALUE_READ(max_to_scan); +MB_PROC_VALUE_WRITE(max_to_scan); +MB_PROC_VALUE_READ(min_to_scan); +MB_PROC_VALUE_WRITE(min_to_scan); +MB_PROC_VALUE_READ(order2_reqs); +MB_PROC_VALUE_WRITE(order2_reqs); +MB_PROC_VALUE_READ(stream_request); +MB_PROC_VALUE_WRITE(stream_request); +MB_PROC_VALUE_READ(group_prealloc); +MB_PROC_VALUE_WRITE(group_prealloc); + +#define MB_PROC_HANDLER(name, var) \ +do { \ + proc = create_proc_entry(name, mode, sbi->s_mb_proc); \ + if (proc == NULL) { \ + printk(KERN_ERR "EXT4-fs: can't to create %s\n", name); \ + goto err_out; \ + } \ + proc->data = sbi; \ + proc->read_proc = ext4_mb_read_##var ; \ + proc->write_proc = ext4_mb_write_##var; \ +} while (0) + +static int ext4_mb_init_per_dev_proc(struct super_block *sb) +{ + mode_t mode = S_IFREG | S_IRUGO | S_IWUSR; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct proc_dir_entry *proc; + char devname[64]; + + snprintf(devname, sizeof(devname) - 1, "%s", + bdevname(sb->s_bdev, devname)); + sbi->s_mb_proc = proc_mkdir(devname, proc_root_ext4); + + MB_PROC_HANDLER(EXT4_MB_STATS_NAME, stats); + MB_PROC_HANDLER(EXT4_MB_MAX_TO_SCAN_NAME, max_to_scan); + MB_PROC_HANDLER(EXT4_MB_MIN_TO_SCAN_NAME, min_to_scan); + MB_PROC_HANDLER(EXT4_MB_ORDER2_REQ, order2_reqs); + MB_PROC_HANDLER(EXT4_MB_STREAM_REQ, stream_request); + MB_PROC_HANDLER(EXT4_MB_GROUP_PREALLOC, group_prealloc); + + return 0; + +err_out: + printk(KERN_ERR "EXT4-fs: Unable to create %s\n", devname); + remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_mb_proc); + remove_proc_entry(devname, proc_root_ext4); + sbi->s_mb_proc = NULL; + + return -ENOMEM; +} + +static int ext4_mb_destroy_per_dev_proc(struct super_block *sb) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + char devname[64]; + + if (sbi->s_mb_proc == NULL) + return -EINVAL; + + snprintf(devname, sizeof(devname) - 1, "%s", + bdevname(sb->s_bdev, devname)); + remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_mb_proc); + remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_mb_proc); + remove_proc_entry(devname, proc_root_ext4); + + return 0; +} + +int __init init_ext4_mballoc(void) +{ + ext4_pspace_cachep = + kmem_cache_create("ext4_prealloc_space", + sizeof(struct ext4_prealloc_space), + 0, SLAB_RECLAIM_ACCOUNT, NULL); + if (ext4_pspace_cachep == NULL) + return -ENOMEM; + + ext4_ac_cachep = + kmem_cache_create("ext4_alloc_context", + sizeof(struct ext4_allocation_context), + 0, SLAB_RECLAIM_ACCOUNT, NULL); + if (ext4_ac_cachep == NULL) { + kmem_cache_destroy(ext4_pspace_cachep); + return -ENOMEM; + } +#ifdef CONFIG_PROC_FS + proc_root_ext4 = proc_mkdir(EXT4_ROOT, proc_root_fs); + if (proc_root_ext4 == NULL) + printk(KERN_ERR "EXT4-fs: Unable to create %s\n", EXT4_ROOT); +#endif + return 0; +} + +void exit_ext4_mballoc(void) +{ + /* XXX: synchronize_rcu(); */ + kmem_cache_destroy(ext4_pspace_cachep); + kmem_cache_destroy(ext4_ac_cachep); +#ifdef CONFIG_PROC_FS + remove_proc_entry(EXT4_ROOT, proc_root_fs); +#endif +} + + +/* + * Check quota and mark choosed space (ac->ac_b_ex) non-free in bitmaps + * Returns 0 if success or error code + */ +static int ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac, + handle_t *handle) +{ + struct buffer_head *bitmap_bh = NULL; + struct ext4_super_block *es; + struct ext4_group_desc *gdp; + struct buffer_head *gdp_bh; + struct ext4_sb_info *sbi; + struct super_block *sb; + ext4_fsblk_t block; + int err; + + BUG_ON(ac->ac_status != AC_STATUS_FOUND); + BUG_ON(ac->ac_b_ex.fe_len <= 0); + + sb = ac->ac_sb; + sbi = EXT4_SB(sb); + es = sbi->s_es; + + ext4_debug("using block group %lu(%d)\n", ac->ac_b_ex.fe_group, + gdp->bg_free_blocks_count); + + err = -EIO; + bitmap_bh = read_block_bitmap(sb, ac->ac_b_ex.fe_group); + if (!bitmap_bh) + goto out_err; + + err = ext4_journal_get_write_access(handle, bitmap_bh); + if (err) + goto out_err; + + err = -EIO; + gdp = ext4_get_group_desc(sb, ac->ac_b_ex.fe_group, &gdp_bh); + if (!gdp) + goto out_err; + + err = ext4_journal_get_write_access(handle, gdp_bh); + if (err) + goto out_err; + + block = ac->ac_b_ex.fe_group * EXT4_BLOCKS_PER_GROUP(sb) + + ac->ac_b_ex.fe_start + + le32_to_cpu(es->s_first_data_block); + + if (block == ext4_block_bitmap(sb, gdp) || + block == ext4_inode_bitmap(sb, gdp) || + in_range(block, ext4_inode_table(sb, gdp), + EXT4_SB(sb)->s_itb_per_group)) { + + ext4_error(sb, __FUNCTION__, + "Allocating block in system zone - block = %llu", + block); + } +#ifdef AGGRESSIVE_CHECK + { + int i; + for (i = 0; i < ac->ac_b_ex.fe_len; i++) { + BUG_ON(mb_test_bit(ac->ac_b_ex.fe_start + i, + bitmap_bh->b_data)); + } + } +#endif + mb_set_bits(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group), bitmap_bh->b_data, + ac->ac_b_ex.fe_start, ac->ac_b_ex.fe_len); + + spin_lock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group)); + if (gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) { + gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT); + gdp->bg_free_blocks_count = + cpu_to_le16(ext4_free_blocks_after_init(sb, + ac->ac_b_ex.fe_group, + gdp)); + } + gdp->bg_free_blocks_count = + cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count) + - ac->ac_b_ex.fe_len); + gdp->bg_checksum = ext4_group_desc_csum(sbi, ac->ac_b_ex.fe_group, gdp); + spin_unlock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group)); + percpu_counter_sub(&sbi->s_freeblocks_counter, ac->ac_b_ex.fe_len); + + err = ext4_journal_dirty_metadata(handle, bitmap_bh); + if (err) + goto out_err; + err = ext4_journal_dirty_metadata(handle, gdp_bh); + +out_err: + sb->s_dirt = 1; + brelse(bitmap_bh); + return err; +} + +/* + * here we normalize request for locality group + * Group request are normalized to s_strip size if we set the same via mount + * option. If not we set it to s_mb_group_prealloc which can be configured via + * /proc/fs/ext4/<partition>/group_prealloc + * + * XXX: should we try to preallocate more than the group has now? + */ +static void ext4_mb_normalize_group_request(struct ext4_allocation_context *ac) +{ + struct super_block *sb = ac->ac_sb; + struct ext4_locality_group *lg = ac->ac_lg; + + BUG_ON(lg == NULL); + if (EXT4_SB(sb)->s_stripe) + ac->ac_g_ex.fe_len = EXT4_SB(sb)->s_stripe; + else + ac->ac_g_ex.fe_len = EXT4_SB(sb)->s_mb_group_prealloc; + mb_debug("#%u: goal %lu blocks for locality group\n", + current->pid, ac->ac_g_ex.fe_len); +} + +/* + * Normalization means making request better in terms of + * size and alignment + */ +static void ext4_mb_normalize_request(struct ext4_allocation_context *ac, + struct ext4_allocation_request *ar) +{ + int bsbits, max; + ext4_lblk_t end; + struct list_head *cur; + loff_t size, orig_size, start_off; + ext4_lblk_t start, orig_start; + struct ext4_inode_info *ei = EXT4_I(ac->ac_inode); + + /* do normalize only data requests, metadata requests + do not need preallocation */ + if (!(ac->ac_flags & EXT4_MB_HINT_DATA)) + return; + + /* sometime caller may want exact blocks */ + if (unlikely(ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY)) + return; + + /* caller may indicate that preallocation isn't + * required (it's a tail, for example) */ + if (ac->ac_flags & EXT4_MB_HINT_NOPREALLOC) + return; + + if (ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC) { + ext4_mb_normalize_group_request(ac); + return ; + } + + bsbits = ac->ac_sb->s_blocksize_bits; + + /* first, let's learn actual file size + * given current request is allocated */ + size = ac->ac_o_ex.fe_logical + ac->ac_o_ex.fe_len; + size = size << bsbits; + if (size < i_size_read(ac->ac_inode)) + size = i_size_read(ac->ac_inode); + + /* max available blocks in a free group */ + max = EXT4_BLOCKS_PER_GROUP(ac->ac_sb) - 1 - 1 - + EXT4_SB(ac->ac_sb)->s_itb_per_group; + +#define NRL_CHECK_SIZE(req, size, max,bits) \ + (req <= (size) || max <= ((size) >> bits)) + + /* first, try to predict filesize */ + /* XXX: should this table be tunable? */ + start_off = 0; + if (size <= 16 * 1024) { + size = 16 * 1024; + } else if (size <= 32 * 1024) { + size = 32 * 1024; + } else if (size <= 64 * 1024) { + size = 64 * 1024; + } else if (size <= 128 * 1024) { + size = 128 * 1024; + } else if (size <= 256 * 1024) { + size = 256 * 1024; + } else if (size <= 512 * 1024) { + size = 512 * 1024; + } else if (size <= 1024 * 1024) { + size = 1024 * 1024; + } else if (NRL_CHECK_SIZE(size, 4 * 1024 * 1024, max, bsbits)) { + start_off = ((loff_t)ac->ac_o_ex.fe_logical >> + (20 - bsbits)) << 20; + size = 1024 * 1024; + } else if (NRL_CHECK_SIZE(size, 8 * 1024 * 1024, max, bsbits)) { + start_off = ((loff_t)ac->ac_o_ex.fe_logical >> + (22 - bsbits)) << 22; + size = 4 * 1024 * 1024; + } else if (NRL_CHECK_SIZE(ac->ac_o_ex.fe_len, + (8<<20)>>bsbits, max, bsbits)) { + start_off = ((loff_t)ac->ac_o_ex.fe_logical >> + (23 - bsbits)) << 23; + size = 8 * 1024 * 1024; + } else { + start_off = (loff_t)ac->ac_o_ex.fe_logical << bsbits; + size = ac->ac_o_ex.fe_len << bsbits; + } + orig_size = size = size >> bsbits; + orig_start = start = start_off >> bsbits; + + /* don't cover already allocated blocks in selected range */ + if (ar->pleft && start <= ar->lleft) { + size -= ar->lleft + 1 - start; + start = ar->lleft + 1; + } + if (ar->pright && start + size - 1 >= ar->lright) + size -= start + size - ar->lright; + + end = start + size; + + /* check we don't cross already preallocated blocks */ + rcu_read_lock(); + list_for_each_rcu(cur, &ei->i_prealloc_list) { + struct ext4_prealloc_space *pa; + unsigned long pa_end; + + pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list); + + if (pa->pa_deleted) + continue; + spin_lock(&pa->pa_lock); + if (pa->pa_deleted) { + spin_unlock(&pa->pa_lock); + continue; + } + + pa_end = pa->pa_lstart + pa->pa_len; + + /* PA must not overlap original request */ + BUG_ON(!(ac->ac_o_ex.fe_logical >= pa_end || + ac->ac_o_ex.fe_logical < pa->pa_lstart)); + + /* skip PA normalized request doesn't overlap with */ + if (pa->pa_lstart >= end) { + spin_unlock(&pa->pa_lock); + continue; + } + if (pa_end <= start) { + spin_unlock(&pa->pa_lock); + continue; + } + BUG_ON(pa->pa_lstart <= start && pa_end >= end); + + if (pa_end <= ac->ac_o_ex.fe_logical) { + BUG_ON(pa_end < start); + start = pa_end; + } + + if (pa->pa_lstart > ac->ac_o_ex.fe_logical) { + BUG_ON(pa->pa_lstart > end); + end = pa->pa_lstart; + } + spin_unlock(&pa->pa_lock); + } + rcu_read_unlock(); + size = end - start; + + /* XXX: extra loop to check we really don't overlap preallocations */ + rcu_read_lock(); + list_for_each_rcu(cur, &ei->i_prealloc_list) { + struct ext4_prealloc_space *pa; + unsigned long pa_end; + pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list); + spin_lock(&pa->pa_lock); + if (pa->pa_deleted == 0) { + pa_end = pa->pa_lstart + pa->pa_len; + BUG_ON(!(start >= pa_end || end <= pa->pa_lstart)); + } + spin_unlock(&pa->pa_lock); + } + rcu_read_unlock(); + + if (start + size <= ac->ac_o_ex.fe_logical && + start > ac->ac_o_ex.fe_logical) { + printk(KERN_ERR "start %lu, size %lu, fe_logical %lu\n", + (unsigned long) start, (unsigned long) size, + (unsigned long) ac->ac_o_ex.fe_logical); + } + BUG_ON(start + size <= ac->ac_o_ex.fe_logical && + start > ac->ac_o_ex.fe_logical); + BUG_ON(size <= 0 || size >= EXT4_BLOCKS_PER_GROUP(ac->ac_sb)); + + /* now prepare goal request */ + + /* XXX: is it better to align blocks WRT to logical + * placement or satisfy big request as is */ + ac->ac_g_ex.fe_logical = start; + ac->ac_g_ex.fe_len = size; + + /* define goal start in order to merge */ + if (ar->pright && (ar->lright == (start + size))) { + /* merge to the right */ + ext4_get_group_no_and_offset(ac->ac_sb, ar->pright - size, + &ac->ac_f_ex.fe_group, + &ac->ac_f_ex.fe_start); + ac->ac_flags |= EXT4_MB_HINT_TRY_GOAL; + } + if (ar->pleft && (ar->lleft + 1 == start)) { + /* merge to the left */ + ext4_get_group_no_and_offset(ac->ac_sb, ar->pleft + 1, + &ac->ac_f_ex.fe_group, + &ac->ac_f_ex.fe_start); + ac->ac_flags |= EXT4_MB_HINT_TRY_GOAL; + } + + mb_debug("goal: %u(was %u) blocks at %u\n", (unsigned) size, + (unsigned) orig_size, (unsigned) start); +} + +static void ext4_mb_collect_stats(struct ext4_allocation_context *ac) +{ + struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb); + + if (sbi->s_mb_stats && ac->ac_g_ex.fe_len > 1) { + atomic_inc(&sbi->s_bal_reqs); + atomic_add(ac->ac_b_ex.fe_len, &sbi->s_bal_allocated); + if (ac->ac_o_ex.fe_len >= ac->ac_g_ex.fe_len) + atomic_inc(&sbi->s_bal_success); + atomic_add(ac->ac_found, &sbi->s_bal_ex_scanned); + if (ac->ac_g_ex.fe_start == ac->ac_b_ex.fe_start && + ac->ac_g_ex.fe_group == ac->ac_b_ex.fe_group) + atomic_inc(&sbi->s_bal_goals); + if (ac->ac_found > sbi->s_mb_max_to_scan) + atomic_inc(&sbi->s_bal_breaks); + } + + ext4_mb_store_history(ac); +} + +/* + * use blocks preallocated to inode + */ +static void ext4_mb_use_inode_pa(struct ext4_allocation_context *ac, + struct ext4_prealloc_space *pa) +{ + ext4_fsblk_t start; + ext4_fsblk_t end; + int len; + + /* found preallocated blocks, use them */ + start = pa->pa_pstart + (ac->ac_o_ex.fe_logical - pa->pa_lstart); + end = min(pa->pa_pstart + pa->pa_len, start + ac->ac_o_ex.fe_len); + len = end - start; + ext4_get_group_no_and_offset(ac->ac_sb, start, &ac->ac_b_ex.fe_group, + &ac->ac_b_ex.fe_start); + ac->ac_b_ex.fe_len = len; + ac->ac_status = AC_STATUS_FOUND; + ac->ac_pa = pa; + + BUG_ON(start < pa->pa_pstart); + BUG_ON(start + len > pa->pa_pstart + pa->pa_len); + BUG_ON(pa->pa_free < len); + pa->pa_free -= len; + + mb_debug("use %llu/%lu from inode pa %p\n", start, len, pa); +} + +/* + * use blocks preallocated to locality group + */ +static void ext4_mb_use_group_pa(struct ext4_allocation_context *ac, + struct ext4_prealloc_space *pa) +{ + unsigned len = ac->ac_o_ex.fe_len; + + ext4_get_group_no_and_offset(ac->ac_sb, pa->pa_pstart, + &ac->ac_b_ex.fe_group, + &ac->ac_b_ex.fe_start); + ac->ac_b_ex.fe_len = len; + ac->ac_status = AC_STATUS_FOUND; + ac->ac_pa = pa; + + /* we don't correct pa_pstart or pa_plen here to avoid + * possible race when the group is being loaded concurrently + * instead we correct pa later, after blocks are marked + * in on-disk bitmap -- see ext4_mb_release_context() + * Other CPUs are prevented from allocating from this pa by lg_mutex + */ + mb_debug("use %u/%u from group pa %p\n", pa->pa_lstart-len, len, pa); +} + +/* + * search goal blocks in preallocated space + */ +static int ext4_mb_use_preallocated(struct ext4_allocation_context *ac) +{ + struct ext4_inode_info *ei = EXT4_I(ac->ac_inode); + struct ext4_locality_group *lg; + struct ext4_prealloc_space *pa; + struct list_head *cur; + + /* only data can be preallocated */ + if (!(ac->ac_flags & EXT4_MB_HINT_DATA)) + return 0; + + /* first, try per-file preallocation */ + rcu_read_lock(); + list_for_each_rcu(cur, &ei->i_prealloc_list) { + pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list); + + /* all fields in this condition don't change, + * so we can skip locking for them */ + if (ac->ac_o_ex.fe_logical < pa->pa_lstart || + ac->ac_o_ex.fe_logical >= pa->pa_lstart + pa->pa_len) + continue; + + /* found preallocated blocks, use them */ + spin_lock(&pa->pa_lock); + if (pa->pa_deleted == 0 && pa->pa_free) { + atomic_inc(&pa->pa_count); + ext4_mb_use_inode_pa(ac, pa); + spin_unlock(&pa->pa_lock); + ac->ac_criteria = 10; + rcu_read_unlock(); + return 1; + } + spin_unlock(&pa->pa_lock); + } + rcu_read_unlock(); + + /* can we use group allocation? */ + if (!(ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC)) + return 0; + + /* inode may have no locality group for some reason */ + lg = ac->ac_lg; + if (lg == NULL) + return 0; + + rcu_read_lock(); + list_for_each_rcu(cur, &lg->lg_prealloc_list) { + pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list); + spin_lock(&pa->pa_lock); + if (pa->pa_deleted == 0 && pa->pa_free >= ac->ac_o_ex.fe_len) { + atomic_inc(&pa->pa_count); + ext4_mb_use_group_pa(ac, pa); + spin_unlock(&pa->pa_lock); + ac->ac_criteria = 20; + rcu_read_unlock(); + return 1; + } + spin_unlock(&pa->pa_lock); + } + rcu_read_unlock(); + + return 0; +} + +/* + * the function goes through all preallocation in this group and marks them + * used in in-core bitmap. buddy must be generated from this bitmap + * Need to be called with ext4 group lock (ext4_lock_group) + */ +static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap, + ext4_group_t group) +{ + struct ext4_group_info *grp = ext4_get_group_info(sb, group); + struct ext4_prealloc_space *pa; + struct list_head *cur; + ext4_group_t groupnr; + ext4_grpblk_t start; + int preallocated = 0; + int count = 0; + int len; + + /* all form of preallocation discards first load group, + * so the only competing code is preallocation use. + * we don't need any locking here + * notice we do NOT ignore preallocations with pa_deleted + * otherwise we could leave used blocks available for + * allocation in buddy when concurrent ext4_mb_put_pa() + * is dropping preallocation + */ + list_for_each(cur, &grp->bb_prealloc_list) { + pa = list_entry(cur, struct ext4_prealloc_space, pa_group_list); + spin_lock(&pa->pa_lock); + ext4_get_group_no_and_offset(sb, pa->pa_pstart, + &groupnr, &start); + len = pa->pa_len; + spin_unlock(&pa->pa_lock); + if (unlikely(len == 0)) + continue; + BUG_ON(groupnr != group); + mb_set_bits(sb_bgl_lock(EXT4_SB(sb), group), + bitmap, start, len); + preallocated += len; + count++; + } + mb_debug("prellocated %u for group %lu\n", preallocated, group); +} + +static void ext4_mb_pa_callback(struct rcu_head *head) +{ + struct ext4_prealloc_space *pa; + pa = container_of(head, struct ext4_prealloc_space, u.pa_rcu); + kmem_cache_free(ext4_pspace_cachep, pa); +} + +/* + * drops a reference to preallocated space descriptor + * if this was the last reference and the space is consumed + */ +static void ext4_mb_put_pa(struct ext4_allocation_context *ac, + struct super_block *sb, struct ext4_prealloc_space *pa) +{ + unsigned long grp; + + if (!atomic_dec_and_test(&pa->pa_count) || pa->pa_free != 0) + return; + + /* in this short window concurrent discard can set pa_deleted */ + spin_lock(&pa->pa_lock); + if (pa->pa_deleted == 1) { + spin_unlock(&pa->pa_lock); + return; + } + + pa->pa_deleted = 1; + spin_unlock(&pa->pa_lock); + + /* -1 is to protect from crossing allocation group */ + ext4_get_group_no_and_offset(sb, pa->pa_pstart - 1, &grp, NULL); + + /* + * possible race: + * + * P1 (buddy init) P2 (regular allocation) + * find block B in PA + * copy on-disk bitmap to buddy + * mark B in on-disk bitmap + * drop PA from group + * mark all PAs in buddy + * + * thus, P1 initializes buddy with B available. to prevent this + * we make "copy" and "mark all PAs" atomic and serialize "drop PA" + * against that pair + */ + ext4_lock_group(sb, grp); + list_del(&pa->pa_group_list); + ext4_unlock_group(sb, grp); + + spin_lock(pa->pa_obj_lock); + list_del_rcu(&pa->pa_inode_list); + spin_unlock(pa->pa_obj_lock); + + call_rcu(&(pa)->u.pa_rcu, ext4_mb_pa_callback); +} + +/* + * creates new preallocated space for given inode + */ +static int ext4_mb_new_inode_pa(struct ext4_allocation_context *ac) +{ + struct super_block *sb = ac->ac_sb; + struct ext4_prealloc_space *pa; + struct ext4_group_info *grp; + struct ext4_inode_info *ei; + + /* preallocate only when found space is larger then requested */ + BUG_ON(ac->ac_o_ex.fe_len >= ac->ac_b_ex.fe_len); + BUG_ON(ac->ac_status != AC_STATUS_FOUND); + BUG_ON(!S_ISREG(ac->ac_inode->i_mode)); + + pa = kmem_cache_alloc(ext4_pspace_cachep, GFP_NOFS); + if (pa == NULL) + return -ENOMEM; + + if (ac->ac_b_ex.fe_len < ac->ac_g_ex.fe_len) { + int winl; + int wins; + int win; + int offs; + + /* we can't allocate as much as normalizer wants. + * so, found space must get proper lstart + * to cover original request */ + BUG_ON(ac->ac_g_ex.fe_logical > ac->ac_o_ex.fe_logical); + BUG_ON(ac->ac_g_ex.fe_len < ac->ac_o_ex.fe_len); + + /* we're limited by original request in that + * logical block must be covered any way + * winl is window we can move our chunk within */ + winl = ac->ac_o_ex.fe_logical - ac->ac_g_ex.fe_logical; + + /* also, we should cover whole original request */ + wins = ac->ac_b_ex.fe_len - ac->ac_o_ex.fe_len; + + /* the smallest one defines real window */ + win = min(winl, wins); + + offs = ac->ac_o_ex.fe_logical % ac->ac_b_ex.fe_len; + if (offs && offs < win) + win = offs; + + ac->ac_b_ex.fe_logical = ac->ac_o_ex.fe_logical - win; + BUG_ON(ac->ac_o_ex.fe_logical < ac->ac_b_ex.fe_logical); + BUG_ON(ac->ac_o_ex.fe_len > ac->ac_b_ex.fe_len); + } + + /* preallocation can change ac_b_ex, thus we store actually + * allocated blocks for history */ + ac->ac_f_ex = ac->ac_b_ex; + + pa->pa_lstart = ac->ac_b_ex.fe_logical; + pa->pa_pstart = ext4_grp_offs_to_block(sb, &ac->ac_b_ex); + pa->pa_len = ac->ac_b_ex.fe_len; + pa->pa_free = pa->pa_len; + atomic_set(&pa->pa_count, 1); + spin_lock_init(&pa->pa_lock); + pa->pa_deleted = 0; + pa->pa_linear = 0; + + mb_debug("new inode pa %p: %llu/%u for %u\n", pa, + pa->pa_pstart, pa->pa_len, pa->pa_lstart); + + ext4_mb_use_inode_pa(ac, pa); + atomic_add(pa->pa_free, &EXT4_SB(sb)->s_mb_preallocated); + + ei = EXT4_I(ac->ac_inode); + grp = ext4_get_group_info(sb, ac->ac_b_ex.fe_group); + + pa->pa_obj_lock = &ei->i_prealloc_lock; + pa->pa_inode = ac->ac_inode; + + ext4_lock_group(sb, ac->ac_b_ex.fe_group); + list_add(&pa->pa_group_list, &grp->bb_prealloc_list); + ext4_unlock_group(sb, ac->ac_b_ex.fe_group); + + spin_lock(pa->pa_obj_lock); + list_add_rcu(&pa->pa_inode_list, &ei->i_prealloc_list); + spin_unlock(pa->pa_obj_lock); + + return 0; +} + +/* + * creates new preallocated space for locality group inodes belongs to + */ +static int ext4_mb_new_group_pa(struct ext4_allocation_context *ac) +{ + struct super_block *sb = ac->ac_sb; + struct ext4_locality_group *lg; + struct ext4_prealloc_space *pa; + struct ext4_group_info *grp; + + /* preallocate only when found space is larger then requested */ + BUG_ON(ac->ac_o_ex.fe_len >= ac->ac_b_ex.fe_len); + BUG_ON(ac->ac_status != AC_STATUS_FOUND); + BUG_ON(!S_ISREG(ac->ac_inode->i_mode)); + + BUG_ON(ext4_pspace_cachep == NULL); + pa = kmem_cache_alloc(ext4_pspace_cachep, GFP_NOFS); + if (pa == NULL) + return -ENOMEM; + + /* preallocation can change ac_b_ex, thus we store actually + * allocated blocks for history */ + ac->ac_f_ex = ac->ac_b_ex; + + pa->pa_pstart = ext4_grp_offs_to_block(sb, &ac->ac_b_ex); + pa->pa_lstart = pa->pa_pstart; + pa->pa_len = ac->ac_b_ex.fe_len; + pa->pa_free = pa->pa_len; + atomic_set(&pa->pa_count, 1); + spin_lock_init(&pa->pa_lock); + pa->pa_deleted = 0; + pa->pa_linear = 1; + + mb_debug("new group pa %p: %llu/%u for %u\n", pa, + pa->pa_pstart, pa->pa_len, pa->pa_lstart); + + ext4_mb_use_group_pa(ac, pa); + atomic_add(pa->pa_free, &EXT4_SB(sb)->s_mb_preallocated); + + grp = ext4_get_group_info(sb, ac->ac_b_ex.fe_group); + lg = ac->ac_lg; + BUG_ON(lg == NULL); + + pa->pa_obj_lock = &lg->lg_prealloc_lock; + pa->pa_inode = NULL; + + ext4_lock_group(sb, ac->ac_b_ex.fe_group); + list_add(&pa->pa_group_list, &grp->bb_prealloc_list); + ext4_unlock_group(sb, ac->ac_b_ex.fe_group); + + spin_lock(pa->pa_obj_lock); + list_add_tail_rcu(&pa->pa_inode_list, &lg->lg_prealloc_list); + spin_unlock(pa->pa_obj_lock); + + return 0; +} + +static int ext4_mb_new_preallocation(struct ext4_allocation_context *ac) +{ + int err; + + if (ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC) + err = ext4_mb_new_group_pa(ac); + else + err = ext4_mb_new_inode_pa(ac); + return err; +} + +/* + * finds all unused blocks in on-disk bitmap, frees them in + * in-core bitmap and buddy. + * @pa must be unlinked from inode and group lists, so that + * nobody else can find/use it. + * the caller MUST hold group/inode locks. + * TODO: optimize the case when there are no in-core structures yet + */ +static int ext4_mb_release_inode_pa(struct ext4_buddy *e4b, + struct buffer_head *bitmap_bh, + struct ext4_prealloc_space *pa) +{ + struct ext4_allocation_context *ac; + struct super_block *sb = e4b->bd_sb; + struct ext4_sb_info *sbi = EXT4_SB(sb); + unsigned long end; + unsigned long next; + ext4_group_t group; + ext4_grpblk_t bit; + sector_t start; + int err = 0; + int free = 0; + + BUG_ON(pa->pa_deleted == 0); + ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, &bit); + BUG_ON(group != e4b->bd_group && pa->pa_len != 0); + end = bit + pa->pa_len; + + ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS); + + if (ac) { + ac->ac_sb = sb; + ac->ac_inode = pa->pa_inode; + ac->ac_op = EXT4_MB_HISTORY_DISCARD; + } + + while (bit < end) { + bit = mb_find_next_zero_bit(bitmap_bh->b_data, end, bit); + if (bit >= end) + break; + next = mb_find_next_bit(bitmap_bh->b_data, end, bit); + if (next > end) + next = end; + start = group * EXT4_BLOCKS_PER_GROUP(sb) + bit + + le32_to_cpu(sbi->s_es->s_first_data_block); + mb_debug(" free preallocated %u/%u in group %u\n", + (unsigned) start, (unsigned) next - bit, + (unsigned) group); + free += next - bit; + + if (ac) { + ac->ac_b_ex.fe_group = group; + ac->ac_b_ex.fe_start = bit; + ac->ac_b_ex.fe_len = next - bit; + ac->ac_b_ex.fe_logical = 0; + ext4_mb_store_history(ac); + } + + mb_free_blocks(pa->pa_inode, e4b, bit, next - bit); + bit = next + 1; + } + if (free != pa->pa_free) { + printk(KERN_CRIT "pa %p: logic %lu, phys. %lu, len %lu\n", + pa, (unsigned long) pa->pa_lstart, + (unsigned long) pa->pa_pstart, + (unsigned long) pa->pa_len); + ext4_error(sb, __FUNCTION__, "free %u, pa_free %u\n", + free, pa->pa_free); + /* + * pa is already deleted so we use the value obtained + * from the bitmap and continue. + */ + } + atomic_add(free, &sbi->s_mb_discarded); + if (ac) + kmem_cache_free(ext4_ac_cachep, ac); + + return err; +} + +static int ext4_mb_release_group_pa(struct ext4_buddy *e4b, + struct ext4_prealloc_space *pa) +{ + struct ext4_allocation_context *ac; + struct super_block *sb = e4b->bd_sb; + ext4_group_t group; + ext4_grpblk_t bit; + + ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS); + + if (ac) + ac->ac_op = EXT4_MB_HISTORY_DISCARD; + + BUG_ON(pa->pa_deleted == 0); + ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, &bit); + BUG_ON(group != e4b->bd_group && pa->pa_len != 0); + mb_free_blocks(pa->pa_inode, e4b, bit, pa->pa_len); + atomic_add(pa->pa_len, &EXT4_SB(sb)->s_mb_discarded); + + if (ac) { + ac->ac_sb = sb; + ac->ac_inode = NULL; + ac->ac_b_ex.fe_group = group; + ac->ac_b_ex.fe_start = bit; + ac->ac_b_ex.fe_len = pa->pa_len; + ac->ac_b_ex.fe_logical = 0; + ext4_mb_store_history(ac); + kmem_cache_free(ext4_ac_cachep, ac); + } + + return 0; +} + +/* + * releases all preallocations in given group + * + * first, we need to decide discard policy: + * - when do we discard + * 1) ENOSPC + * - how many do we discard + * 1) how many requested + */ +static int ext4_mb_discard_group_preallocations(struct super_block *sb, + ext4_group_t group, int needed) +{ + struct ext4_group_info *grp = ext4_get_group_info(sb, group); + struct buffer_head *bitmap_bh = NULL; + struct ext4_prealloc_space *pa, *tmp; + struct list_head list; + struct ext4_buddy e4b; + int err; + int busy = 0; + int free = 0; + + mb_debug("discard preallocation for group %lu\n", group); + + if (list_empty(&grp->bb_prealloc_list)) + return 0; + + bitmap_bh = read_block_bitmap(sb, group); + if (bitmap_bh == NULL) { + /* error handling here */ + ext4_mb_release_desc(&e4b); + BUG_ON(bitmap_bh == NULL); + } + + err = ext4_mb_load_buddy(sb, group, &e4b); + BUG_ON(err != 0); /* error handling here */ + + if (needed == 0) + needed = EXT4_BLOCKS_PER_GROUP(sb) + 1; + + grp = ext4_get_group_info(sb, group); + INIT_LIST_HEAD(&list); + +repeat: + ext4_lock_group(sb, group); + list_for_each_entry_safe(pa, tmp, + &grp->bb_prealloc_list, pa_group_list) { + spin_lock(&pa->pa_lock); + if (atomic_read(&pa->pa_count)) { + spin_unlock(&pa->pa_lock); + busy = 1; + continue; + } + if (pa->pa_deleted) { + spin_unlock(&pa->pa_lock); + continue; + } + + /* seems this one can be freed ... */ + pa->pa_deleted = 1; + + /* we can trust pa_free ... */ + free += pa->pa_free; + + spin_unlock(&pa->pa_lock); + + list_del(&pa->pa_group_list); + list_add(&pa->u.pa_tmp_list, &list); + } + + /* if we still need more blocks and some PAs were used, try again */ + if (free < needed && busy) { + busy = 0; + ext4_unlock_group(sb, group); + /* + * Yield the CPU here so that we don't get soft lockup + * in non preempt case. + */ + yield(); + goto repeat; + } + + /* found anything to free? */ + if (list_empty(&list)) { + BUG_ON(free != 0); + goto out; + } + + /* now free all selected PAs */ + list_for_each_entry_safe(pa, tmp, &list, u.pa_tmp_list) { + + /* remove from object (inode or locality group) */ + spin_lock(pa->pa_obj_lock); + list_del_rcu(&pa->pa_inode_list); + spin_unlock(pa->pa_obj_lock); + + if (pa->pa_linear) + ext4_mb_release_group_pa(&e4b, pa); + else + ext4_mb_release_inode_pa(&e4b, bitmap_bh, pa); + + list_del(&pa->u.pa_tmp_list); + call_rcu(&(pa)->u.pa_rcu, ext4_mb_pa_callback); + } + +out: + ext4_unlock_group(sb, group); + ext4_mb_release_desc(&e4b); + put_bh(bitmap_bh); + return free; +} + +/* + * releases all non-used preallocated blocks for given inode + * + * It's important to discard preallocations under i_data_sem + * We don't want another block to be served from the prealloc + * space when we are discarding the inode prealloc space. + * + * FIXME!! Make sure it is valid at all the call sites + */ +void ext4_mb_discard_inode_preallocations(struct inode *inode) +{ + struct ext4_inode_info *ei = EXT4_I(inode); + struct super_block *sb = inode->i_sb; + struct buffer_head *bitmap_bh = NULL; + struct ext4_prealloc_space *pa, *tmp; + ext4_group_t group = 0; + struct list_head list; + struct ext4_buddy e4b; + int err; + + if (!test_opt(sb, MBALLOC) || !S_ISREG(inode->i_mode)) { + /*BUG_ON(!list_empty(&ei->i_prealloc_list));*/ + return; + } + + mb_debug("discard preallocation for inode %lu\n", inode->i_ino); + + INIT_LIST_HEAD(&list); + +repeat: + /* first, collect all pa's in the inode */ + spin_lock(&ei->i_prealloc_lock); + while (!list_empty(&ei->i_prealloc_list)) { + pa = list_entry(ei->i_prealloc_list.next, + struct ext4_prealloc_space, pa_inode_list); + BUG_ON(pa->pa_obj_lock != &ei->i_prealloc_lock); + spin_lock(&pa->pa_lock); + if (atomic_read(&pa->pa_count)) { + /* this shouldn't happen often - nobody should + * use preallocation while we're discarding it */ + spin_unlock(&pa->pa_lock); + spin_unlock(&ei->i_prealloc_lock); + printk(KERN_ERR "uh-oh! used pa while discarding\n"); + WARN_ON(1); + schedule_timeout_uninterruptible(HZ); + goto repeat; + + } + if (pa->pa_deleted == 0) { + pa->pa_deleted = 1; + spin_unlock(&pa->pa_lock); + list_del_rcu(&pa->pa_inode_list); + list_add(&pa->u.pa_tmp_list, &list); + continue; + } + + /* someone is deleting pa right now */ + spin_unlock(&pa->pa_lock); + spin_unlock(&ei->i_prealloc_lock); + + /* we have to wait here because pa_deleted + * doesn't mean pa is already unlinked from + * the list. as we might be called from + * ->clear_inode() the inode will get freed + * and concurrent thread which is unlinking + * pa from inode's list may access already + * freed memory, bad-bad-bad */ + + /* XXX: if this happens too often, we can + * add a flag to force wait only in case + * of ->clear_inode(), but not in case of + * regular truncate */ + schedule_timeout_uninterruptible(HZ); + goto repeat; + } + spin_unlock(&ei->i_prealloc_lock); + + list_for_each_entry_safe(pa, tmp, &list, u.pa_tmp_list) { + BUG_ON(pa->pa_linear != 0); + ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, NULL); + + err = ext4_mb_load_buddy(sb, group, &e4b); + BUG_ON(err != 0); /* error handling here */ + + bitmap_bh = read_block_bitmap(sb, group); + if (bitmap_bh == NULL) { + /* error handling here */ + ext4_mb_release_desc(&e4b); + BUG_ON(bitmap_bh == NULL); + } + + ext4_lock_group(sb, group); + list_del(&pa->pa_group_list); + ext4_mb_release_inode_pa(&e4b, bitmap_bh, pa); + ext4_unlock_group(sb, group); + + ext4_mb_release_desc(&e4b); + put_bh(bitmap_bh); + + list_del(&pa->u.pa_tmp_list); + call_rcu(&(pa)->u.pa_rcu, ext4_mb_pa_callback); + } +} + +/* + * finds all preallocated spaces and return blocks being freed to them + * if preallocated space becomes full (no block is used from the space) + * then the function frees space in buddy + * XXX: at the moment, truncate (which is the only way to free blocks) + * discards all preallocations + */ +static void ext4_mb_return_to_preallocation(struct inode *inode, + struct ext4_buddy *e4b, + sector_t block, int count) +{ + BUG_ON(!list_empty(&EXT4_I(inode)->i_prealloc_list)); +} +#ifdef MB_DEBUG +static void ext4_mb_show_ac(struct ext4_allocation_context *ac) +{ + struct super_block *sb = ac->ac_sb; + ext4_group_t i; + + printk(KERN_ERR "EXT4-fs: Can't allocate:" + " Allocation context details:\n"); + printk(KERN_ERR "EXT4-fs: status %d flags %d\n", + ac->ac_status, ac->ac_flags); + printk(KERN_ERR "EXT4-fs: orig %lu/%lu/%lu@%lu, goal %lu/%lu/%lu@%lu, " + "best %lu/%lu/%lu@%lu cr %d\n", + (unsigned long)ac->ac_o_ex.fe_group, + (unsigned long)ac->ac_o_ex.fe_start, + (unsigned long)ac->ac_o_ex.fe_len, + (unsigned long)ac->ac_o_ex.fe_logical, + (unsigned long)ac->ac_g_ex.fe_group, + (unsigned long)ac->ac_g_ex.fe_start, + (unsigned long)ac->ac_g_ex.fe_len, + (unsigned long)ac->ac_g_ex.fe_logical, + (unsigned long)ac->ac_b_ex.fe_group, + (unsigned long)ac->ac_b_ex.fe_start, + (unsigned long)ac->ac_b_ex.fe_len, + (unsigned long)ac->ac_b_ex.fe_logical, + (int)ac->ac_criteria); + printk(KERN_ERR "EXT4-fs: %lu scanned, %d found\n", ac->ac_ex_scanned, + ac->ac_found); + printk(KERN_ERR "EXT4-fs: groups: \n"); + for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) { + struct ext4_group_info *grp = ext4_get_group_info(sb, i); + struct ext4_prealloc_space *pa; + ext4_grpblk_t start; + struct list_head *cur; + ext4_lock_group(sb, i); + list_for_each(cur, &grp->bb_prealloc_list) { + pa = list_entry(cur, struct ext4_prealloc_space, + pa_group_list); + spin_lock(&pa->pa_lock); + ext4_get_group_no_and_offset(sb, pa->pa_pstart, + NULL, &start); + spin_unlock(&pa->pa_lock); + printk(KERN_ERR "PA:%lu:%d:%u \n", i, + start, pa->pa_len); + } + ext4_lock_group(sb, i); + + if (grp->bb_free == 0) + continue; + printk(KERN_ERR "%lu: %d/%d \n", + i, grp->bb_free, grp->bb_fragments); + } + printk(KERN_ERR "\n"); +} +#else +static inline void ext4_mb_show_ac(struct ext4_allocation_context *ac) +{ + return; +} +#endif + +/* + * We use locality group preallocation for small size file. The size of the + * file is determined by the current size or the resulting size after + * allocation which ever is larger + * + * One can tune this size via /proc/fs/ext4/<partition>/stream_req + */ +static void ext4_mb_group_or_file(struct ext4_allocation_context *ac) +{ + struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb); + int bsbits = ac->ac_sb->s_blocksize_bits; + loff_t size, isize; + + if (!(ac->ac_flags & EXT4_MB_HINT_DATA)) + return; + + size = ac->ac_o_ex.fe_logical + ac->ac_o_ex.fe_len; + isize = i_size_read(ac->ac_inode) >> bsbits; + size = max(size, isize); + + /* don't use group allocation for large files */ + if (size >= sbi->s_mb_stream_request) + return; + + if (unlikely(ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY)) + return; + + BUG_ON(ac->ac_lg != NULL); + /* + * locality group prealloc space are per cpu. The reason for having + * per cpu locality group is to reduce the contention between block + * request from multiple CPUs. + */ + ac->ac_lg = &sbi->s_locality_groups[get_cpu()]; + put_cpu(); + + /* we're going to use group allocation */ + ac->ac_flags |= EXT4_MB_HINT_GROUP_ALLOC; + + /* serialize all allocations in the group */ + mutex_lock(&ac->ac_lg->lg_mutex); +} + +static int ext4_mb_initialize_context(struct ext4_allocation_context *ac, + struct ext4_allocation_request *ar) +{ + struct super_block *sb = ar->inode->i_sb; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; + ext4_group_t group; + unsigned long len; + unsigned long goal; + ext4_grpblk_t block; + + /* we can't allocate > group size */ + len = ar->len; + + /* just a dirty hack to filter too big requests */ + if (len >= EXT4_BLOCKS_PER_GROUP(sb) - 10) + len = EXT4_BLOCKS_PER_GROUP(sb) - 10; + + /* start searching from the goal */ + goal = ar->goal; + if (goal < le32_to_cpu(es->s_first_data_block) || + goal >= ext4_blocks_count(es)) + goal = le32_to_cpu(es->s_first_data_block); + ext4_get_group_no_and_offset(sb, goal, &group, &block); + + /* set up allocation goals */ + ac->ac_b_ex.fe_logical = ar->logical; + ac->ac_b_ex.fe_group = 0; + ac->ac_b_ex.fe_start = 0; + ac->ac_b_ex.fe_len = 0; + ac->ac_status = AC_STATUS_CONTINUE; + ac->ac_groups_scanned = 0; + ac->ac_ex_scanned = 0; + ac->ac_found = 0; + ac->ac_sb = sb; + ac->ac_inode = ar->inode; + ac->ac_o_ex.fe_logical = ar->logical; + ac->ac_o_ex.fe_group = group; + ac->ac_o_ex.fe_start = block; + ac->ac_o_ex.fe_len = len; + ac->ac_g_ex.fe_logical = ar->logical; + ac->ac_g_ex.fe_group = group; + ac->ac_g_ex.fe_start = block; + ac->ac_g_ex.fe_len = len; + ac->ac_f_ex.fe_len = 0; + ac->ac_flags = ar->flags; + ac->ac_2order = 0; + ac->ac_criteria = 0; + ac->ac_pa = NULL; + ac->ac_bitmap_page = NULL; + ac->ac_buddy_page = NULL; + ac->ac_lg = NULL; + + /* we have to define context: we'll we work with a file or + * locality group. this is a policy, actually */ + ext4_mb_group_or_file(ac); + + mb_debug("init ac: %u blocks @ %u, goal %u, flags %x, 2^%d, " + "left: %u/%u, right %u/%u to %swritable\n", + (unsigned) ar->len, (unsigned) ar->logical, + (unsigned) ar->goal, ac->ac_flags, ac->ac_2order, + (unsigned) ar->lleft, (unsigned) ar->pleft, + (unsigned) ar->lright, (unsigned) ar->pright, + atomic_read(&ar->inode->i_writecount) ? "" : "non-"); + return 0; + +} + +/* + * release all resource we used in allocation + */ +static int ext4_mb_release_context(struct ext4_allocation_context *ac) +{ + if (ac->ac_pa) { + if (ac->ac_pa->pa_linear) { + /* see comment in ext4_mb_use_group_pa() */ + spin_lock(&ac->ac_pa->pa_lock); + ac->ac_pa->pa_pstart += ac->ac_b_ex.fe_len; + ac->ac_pa->pa_lstart += ac->ac_b_ex.fe_len; + ac->ac_pa->pa_free -= ac->ac_b_ex.fe_len; + ac->ac_pa->pa_len -= ac->ac_b_ex.fe_len; + spin_unlock(&ac->ac_pa->pa_lock); + } + ext4_mb_put_pa(ac, ac->ac_sb, ac->ac_pa); + } + if (ac->ac_bitmap_page) + page_cache_release(ac->ac_bitmap_page); + if (ac->ac_buddy_page) + page_cache_release(ac->ac_buddy_page); + if (ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC) + mutex_unlock(&ac->ac_lg->lg_mutex); + ext4_mb_collect_stats(ac); + return 0; +} + +static int ext4_mb_discard_preallocations(struct super_block *sb, int needed) +{ + ext4_group_t i; + int ret; + int freed = 0; + + for (i = 0; i < EXT4_SB(sb)->s_groups_count && needed > 0; i++) { + ret = ext4_mb_discard_group_preallocations(sb, i, needed); + freed += ret; + needed -= ret; + } + + return freed; +} + +/* + * Main entry point into mballoc to allocate blocks + * it tries to use preallocation first, then falls back + * to usual allocation + */ +ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle, + struct ext4_allocation_request *ar, int *errp) +{ + struct ext4_allocation_context *ac = NULL; + struct ext4_sb_info *sbi; + struct super_block *sb; + ext4_fsblk_t block = 0; + int freed; + int inquota; + + sb = ar->inode->i_sb; + sbi = EXT4_SB(sb); + + if (!test_opt(sb, MBALLOC)) { + block = ext4_new_blocks_old(handle, ar->inode, ar->goal, + &(ar->len), errp); + return block; + } + + while (ar->len && DQUOT_ALLOC_BLOCK(ar->inode, ar->len)) { + ar->flags |= EXT4_MB_HINT_NOPREALLOC; + ar->len--; + } + if (ar->len == 0) { + *errp = -EDQUOT; + return 0; + } + inquota = ar->len; + + ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS); + if (!ac) { + *errp = -ENOMEM; + return 0; + } + + ext4_mb_poll_new_transaction(sb, handle); + + *errp = ext4_mb_initialize_context(ac, ar); + if (*errp) { + ar->len = 0; + goto out; + } + + ac->ac_op = EXT4_MB_HISTORY_PREALLOC; + if (!ext4_mb_use_preallocated(ac)) { + + ac->ac_op = EXT4_MB_HISTORY_ALLOC; + ext4_mb_normalize_request(ac, ar); + +repeat: + /* allocate space in core */ + ext4_mb_regular_allocator(ac); + + /* as we've just preallocated more space than + * user requested orinally, we store allocated + * space in a special descriptor */ + if (ac->ac_status == AC_STATUS_FOUND && + ac->ac_o_ex.fe_len < ac->ac_b_ex.fe_len) + ext4_mb_new_preallocation(ac); + } + + if (likely(ac->ac_status == AC_STATUS_FOUND)) { + ext4_mb_mark_diskspace_used(ac, handle); + *errp = 0; + block = ext4_grp_offs_to_block(sb, &ac->ac_b_ex); + ar->len = ac->ac_b_ex.fe_len; + } else { + freed = ext4_mb_discard_preallocations(sb, ac->ac_o_ex.fe_len); + if (freed) + goto repeat; + *errp = -ENOSPC; + ac->ac_b_ex.fe_len = 0; + ar->len = 0; + ext4_mb_show_ac(ac); + } + + ext4_mb_release_context(ac); + +out: + if (ar->len < inquota) + DQUOT_FREE_BLOCK(ar->inode, inquota - ar->len); + + kmem_cache_free(ext4_ac_cachep, ac); + return block; +} +static void ext4_mb_poll_new_transaction(struct super_block *sb, + handle_t *handle) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + + if (sbi->s_last_transaction == handle->h_transaction->t_tid) + return; + + /* new transaction! time to close last one and free blocks for + * committed transaction. we know that only transaction can be + * active, so previos transaction can be being logged and we + * know that transaction before previous is known to be already + * logged. this means that now we may free blocks freed in all + * transactions before previous one. hope I'm clear enough ... */ + + spin_lock(&sbi->s_md_lock); + if (sbi->s_last_transaction != handle->h_transaction->t_tid) { + mb_debug("new transaction %lu, old %lu\n", + (unsigned long) handle->h_transaction->t_tid, + (unsigned long) sbi->s_last_transaction); + list_splice_init(&sbi->s_closed_transaction, + &sbi->s_committed_transaction); + list_splice_init(&sbi->s_active_transaction, + &sbi->s_closed_transaction); + sbi->s_last_transaction = handle->h_transaction->t_tid; + } + spin_unlock(&sbi->s_md_lock); + + ext4_mb_free_committed_blocks(sb); +} + +static int ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b, + ext4_group_t group, ext4_grpblk_t block, int count) +{ + struct ext4_group_info *db = e4b->bd_info; + struct super_block *sb = e4b->bd_sb; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_free_metadata *md; + int i; + + BUG_ON(e4b->bd_bitmap_page == NULL); + BUG_ON(e4b->bd_buddy_page == NULL); + + ext4_lock_group(sb, group); + for (i = 0; i < count; i++) { + md = db->bb_md_cur; + if (md && db->bb_tid != handle->h_transaction->t_tid) { + db->bb_md_cur = NULL; + md = NULL; + } + + if (md == NULL) { + ext4_unlock_group(sb, group); + md = kmalloc(sizeof(*md), GFP_NOFS); + if (md == NULL) + return -ENOMEM; + md->num = 0; + md->group = group; + + ext4_lock_group(sb, group); + if (db->bb_md_cur == NULL) { + spin_lock(&sbi->s_md_lock); + list_add(&md->list, &sbi->s_active_transaction); + spin_unlock(&sbi->s_md_lock); + /* protect buddy cache from being freed, + * otherwise we'll refresh it from + * on-disk bitmap and lose not-yet-available + * blocks */ + page_cache_get(e4b->bd_buddy_page); + page_cache_get(e4b->bd_bitmap_page); + db->bb_md_cur = md; + db->bb_tid = handle->h_transaction->t_tid; + mb_debug("new md 0x%p for group %lu\n", + md, md->group); + } else { + kfree(md); + md = db->bb_md_cur; + } + } + + BUG_ON(md->num >= EXT4_BB_MAX_BLOCKS); + md->blocks[md->num] = block + i; + md->num++; + if (md->num == EXT4_BB_MAX_BLOCKS) { + /* no more space, put full container on a sb's list */ + db->bb_md_cur = NULL; + } + } + ext4_unlock_group(sb, group); + return 0; +} + +/* + * Main entry point into mballoc to free blocks + */ +void ext4_mb_free_blocks(handle_t *handle, struct inode *inode, + unsigned long block, unsigned long count, + int metadata, unsigned long *freed) +{ + struct buffer_head *bitmap_bh = NULL; + struct super_block *sb = inode->i_sb; + struct ext4_allocation_context *ac = NULL; + struct ext4_group_desc *gdp; + struct ext4_super_block *es; + unsigned long overflow; + ext4_grpblk_t bit; + struct buffer_head *gd_bh; + ext4_group_t block_group; + struct ext4_sb_info *sbi; + struct ext4_buddy e4b; + int err = 0; + int ret; + + *freed = 0; + + ext4_mb_poll_new_transaction(sb, handle); + + sbi = EXT4_SB(sb); + es = EXT4_SB(sb)->s_es; + if (block < le32_to_cpu(es->s_first_data_block) || + block + count < block || + block + count > ext4_blocks_count(es)) { + ext4_error(sb, __FUNCTION__, + "Freeing blocks not in datazone - " + "block = %lu, count = %lu", block, count); + goto error_return; + } + + ext4_debug("freeing block %lu\n", block); + + ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS); + if (ac) { + ac->ac_op = EXT4_MB_HISTORY_FREE; + ac->ac_inode = inode; + ac->ac_sb = sb; + } + +do_more: + overflow = 0; + ext4_get_group_no_and_offset(sb, block, &block_group, &bit); + + /* + * Check to see if we are freeing blocks across a group + * boundary. + */ + if (bit + count > EXT4_BLOCKS_PER_GROUP(sb)) { + overflow = bit + count - EXT4_BLOCKS_PER_GROUP(sb); + count -= overflow; + } + bitmap_bh = read_block_bitmap(sb, block_group); + if (!bitmap_bh) + goto error_return; + gdp = ext4_get_group_desc(sb, block_group, &gd_bh); + if (!gdp) + goto error_return; + + if (in_range(ext4_block_bitmap(sb, gdp), block, count) || + in_range(ext4_inode_bitmap(sb, gdp), block, count) || + in_range(block, ext4_inode_table(sb, gdp), + EXT4_SB(sb)->s_itb_per_group) || + in_range(block + count - 1, ext4_inode_table(sb, gdp), + EXT4_SB(sb)->s_itb_per_group)) { + + ext4_error(sb, __FUNCTION__, + "Freeing blocks in system zone - " + "Block = %lu, count = %lu", block, count); + } + + BUFFER_TRACE(bitmap_bh, "getting write access"); + err = ext4_journal_get_write_access(handle, bitmap_bh); + if (err) + goto error_return; + + /* + * We are about to modify some metadata. Call the journal APIs + * to unshare ->b_data if a currently-committing transaction is + * using it + */ + BUFFER_TRACE(gd_bh, "get_write_access"); + err = ext4_journal_get_write_access(handle, gd_bh); + if (err) + goto error_return; + + err = ext4_mb_load_buddy(sb, block_group, &e4b); + if (err) + goto error_return; + +#ifdef AGGRESSIVE_CHECK + { + int i; + for (i = 0; i < count; i++) + BUG_ON(!mb_test_bit(bit + i, bitmap_bh->b_data)); + } +#endif + mb_clear_bits(sb_bgl_lock(sbi, block_group), bitmap_bh->b_data, + bit, count); + + /* We dirtied the bitmap block */ + BUFFER_TRACE(bitmap_bh, "dirtied bitmap block"); + err = ext4_journal_dirty_metadata(handle, bitmap_bh); + + if (ac) { + ac->ac_b_ex.fe_group = block_group; + ac->ac_b_ex.fe_start = bit; + ac->ac_b_ex.fe_len = count; + ext4_mb_store_history(ac); + } + + if (metadata) { + /* blocks being freed are metadata. these blocks shouldn't + * be used until this transaction is committed */ + ext4_mb_free_metadata(handle, &e4b, block_group, bit, count); + } else { + ext4_lock_group(sb, block_group); + err = mb_free_blocks(inode, &e4b, bit, count); + ext4_mb_return_to_preallocation(inode, &e4b, block, count); + ext4_unlock_group(sb, block_group); + BUG_ON(err != 0); + } + + spin_lock(sb_bgl_lock(sbi, block_group)); + gdp->bg_free_blocks_count = + cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count) + count); + gdp->bg_checksum = ext4_group_desc_csum(sbi, block_group, gdp); + spin_unlock(sb_bgl_lock(sbi, block_group)); + percpu_counter_add(&sbi->s_freeblocks_counter, count); + + ext4_mb_release_desc(&e4b); + + *freed += count; + + /* And the group descriptor block */ + BUFFER_TRACE(gd_bh, "dirtied group descriptor block"); + ret = ext4_journal_dirty_metadata(handle, gd_bh); + if (!err) + err = ret; + + if (overflow && !err) { + block += count; + count = overflow; + put_bh(bitmap_bh); + goto do_more; + } + sb->s_dirt = 1; +error_return: + brelse(bitmap_bh); + ext4_std_error(sb, err); + if (ac) + kmem_cache_free(ext4_ac_cachep, ac); + return; +} diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c new file mode 100644 index 00000000000..5c1e27de775 --- /dev/null +++ b/fs/ext4/migrate.c @@ -0,0 +1,602 @@ +/* + * Copyright IBM Corporation, 2007 + * Author Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2.1 of the GNU Lesser General Public License + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + */ + +#include <linux/module.h> +#include <linux/ext4_jbd2.h> +#include <linux/ext4_fs_extents.h> + +/* + * The contiguous blocks details which can be + * represented by a single extent + */ +struct list_blocks_struct { + ext4_lblk_t first_block, last_block; + ext4_fsblk_t first_pblock, last_pblock; +}; + +static int finish_range(handle_t *handle, struct inode *inode, + struct list_blocks_struct *lb) + +{ + int retval = 0, needed; + struct ext4_extent newext; + struct ext4_ext_path *path; + if (lb->first_pblock == 0) + return 0; + + /* Add the extent to temp inode*/ + newext.ee_block = cpu_to_le32(lb->first_block); + newext.ee_len = cpu_to_le16(lb->last_block - lb->first_block + 1); + ext4_ext_store_pblock(&newext, lb->first_pblock); + path = ext4_ext_find_extent(inode, lb->first_block, NULL); + + if (IS_ERR(path)) { + retval = PTR_ERR(path); + path = NULL; + goto err_out; + } + + /* + * Calculate the credit needed to inserting this extent + * Since we are doing this in loop we may accumalate extra + * credit. But below we try to not accumalate too much + * of them by restarting the journal. + */ + needed = ext4_ext_calc_credits_for_insert(inode, path); + + /* + * Make sure the credit we accumalated is not really high + */ + if (needed && handle->h_buffer_credits >= EXT4_RESERVE_TRANS_BLOCKS) { + retval = ext4_journal_restart(handle, needed); + if (retval) + goto err_out; + } else if (needed) { + retval = ext4_journal_extend(handle, needed); + if (retval) { + /* + * IF not able to extend the journal restart the journal + */ + retval = ext4_journal_restart(handle, needed); + if (retval) + goto err_out; + } + } + retval = ext4_ext_insert_extent(handle, inode, path, &newext); +err_out: + if (path) { + ext4_ext_drop_refs(path); + kfree(path); + } + lb->first_pblock = 0; + return retval; +} + +static int update_extent_range(handle_t *handle, struct inode *inode, + ext4_fsblk_t pblock, ext4_lblk_t blk_num, + struct list_blocks_struct *lb) +{ + int retval; + /* + * See if we can add on to the existing range (if it exists) + */ + if (lb->first_pblock && + (lb->last_pblock+1 == pblock) && + (lb->last_block+1 == blk_num)) { + lb->last_pblock = pblock; + lb->last_block = blk_num; + return 0; + } + /* + * Start a new range. + */ + retval = finish_range(handle, inode, lb); + lb->first_pblock = lb->last_pblock = pblock; + lb->first_block = lb->last_block = blk_num; + + return retval; +} + +static int update_ind_extent_range(handle_t *handle, struct inode *inode, + ext4_fsblk_t pblock, ext4_lblk_t *blk_nump, + struct list_blocks_struct *lb) +{ + struct buffer_head *bh; + __le32 *i_data; + int i, retval = 0; + ext4_lblk_t blk_count = *blk_nump; + unsigned long max_entries = inode->i_sb->s_blocksize >> 2; + + if (!pblock) { + /* Only update the file block number */ + *blk_nump += max_entries; + return 0; + } + + bh = sb_bread(inode->i_sb, pblock); + if (!bh) + return -EIO; + + i_data = (__le32 *)bh->b_data; + for (i = 0; i < max_entries; i++, blk_count++) { + if (i_data[i]) { + retval = update_extent_range(handle, inode, + le32_to_cpu(i_data[i]), + blk_count, lb); + if (retval) + break; + } + } + + /* Update the file block number */ + *blk_nump = blk_count; + put_bh(bh); + return retval; + +} + +static int update_dind_extent_range(handle_t *handle, struct inode *inode, + ext4_fsblk_t pblock, ext4_lblk_t *blk_nump, + struct list_blocks_struct *lb) +{ + struct buffer_head *bh; + __le32 *i_data; + int i, retval = 0; + ext4_lblk_t blk_count = *blk_nump; + unsigned long max_entries = inode->i_sb->s_blocksize >> 2; + + if (!pblock) { + /* Only update the file block number */ + *blk_nump += max_entries * max_entries; + return 0; + } + bh = sb_bread(inode->i_sb, pblock); + if (!bh) + return -EIO; + + i_data = (__le32 *)bh->b_data; + for (i = 0; i < max_entries; i++) { + if (i_data[i]) { + retval = update_ind_extent_range(handle, inode, + le32_to_cpu(i_data[i]), + &blk_count, lb); + if (retval) + break; + } else { + /* Only update the file block number */ + blk_count += max_entries; + } + } + + /* Update the file block number */ + *blk_nump = blk_count; + put_bh(bh); + return retval; + +} + +static int update_tind_extent_range(handle_t *handle, struct inode *inode, + ext4_fsblk_t pblock, ext4_lblk_t *blk_nump, + struct list_blocks_struct *lb) +{ + struct buffer_head *bh; + __le32 *i_data; + int i, retval = 0; + ext4_lblk_t blk_count = *blk_nump; + unsigned long max_entries = inode->i_sb->s_blocksize >> 2; + + if (!pblock) { + /* Only update the file block number */ + *blk_nump += max_entries * max_entries * max_entries; + return 0; + } + bh = sb_bread(inode->i_sb, pblock); + if (!bh) + return -EIO; + + i_data = (__le32 *)bh->b_data; + for (i = 0; i < max_entries; i++) { + if (i_data[i]) { + retval = update_dind_extent_range(handle, inode, + le32_to_cpu(i_data[i]), + &blk_count, lb); + if (retval) + break; + } else + /* Only update the file block number */ + blk_count += max_entries * max_entries; + } + /* Update the file block number */ + *blk_nump = blk_count; + put_bh(bh); + return retval; + +} + +static int extend_credit_for_blkdel(handle_t *handle, struct inode *inode) +{ + int retval = 0, needed; + + if (handle->h_buffer_credits > EXT4_RESERVE_TRANS_BLOCKS) + return 0; + /* + * We are freeing a blocks. During this we touch + * superblock, group descriptor and block bitmap. + * So allocate a credit of 3. We may update + * quota (user and group). + */ + needed = 3 + 2*EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb); + + if (ext4_journal_extend(handle, needed) != 0) + retval = ext4_journal_restart(handle, needed); + + return retval; +} + +static int free_dind_blocks(handle_t *handle, + struct inode *inode, __le32 i_data) +{ + int i; + __le32 *tmp_idata; + struct buffer_head *bh; + unsigned long max_entries = inode->i_sb->s_blocksize >> 2; + + bh = sb_bread(inode->i_sb, le32_to_cpu(i_data)); + if (!bh) + return -EIO; + + tmp_idata = (__le32 *)bh->b_data; + for (i = 0; i < max_entries; i++) { + if (tmp_idata[i]) { + extend_credit_for_blkdel(handle, inode); + ext4_free_blocks(handle, inode, + le32_to_cpu(tmp_idata[i]), 1, 1); + } + } + put_bh(bh); + extend_credit_for_blkdel(handle, inode); + ext4_free_blocks(handle, inode, le32_to_cpu(i_data), 1, 1); + return 0; +} + +static int free_tind_blocks(handle_t *handle, + struct inode *inode, __le32 i_data) +{ + int i, retval = 0; + __le32 *tmp_idata; + struct buffer_head *bh; + unsigned long max_entries = inode->i_sb->s_blocksize >> 2; + + bh = sb_bread(inode->i_sb, le32_to_cpu(i_data)); + if (!bh) + return -EIO; + + tmp_idata = (__le32 *)bh->b_data; + for (i = 0; i < max_entries; i++) { + if (tmp_idata[i]) { + retval = free_dind_blocks(handle, + inode, tmp_idata[i]); + if (retval) { + put_bh(bh); + return retval; + } + } + } + put_bh(bh); + extend_credit_for_blkdel(handle, inode); + ext4_free_blocks(handle, inode, le32_to_cpu(i_data), 1, 1); + return 0; +} + +static int free_ind_block(handle_t *handle, struct inode *inode, __le32 *i_data) +{ + int retval; + + /* ei->i_data[EXT4_IND_BLOCK] */ + if (i_data[0]) { + extend_credit_for_blkdel(handle, inode); + ext4_free_blocks(handle, inode, + le32_to_cpu(i_data[0]), 1, 1); + } + + /* ei->i_data[EXT4_DIND_BLOCK] */ + if (i_data[1]) { + retval = free_dind_blocks(handle, inode, i_data[1]); + if (retval) + return retval; + } + + /* ei->i_data[EXT4_TIND_BLOCK] */ + if (i_data[2]) { + retval = free_tind_blocks(handle, inode, i_data[2]); + if (retval) + return retval; + } + return 0; +} + +static int ext4_ext_swap_inode_data(handle_t *handle, struct inode *inode, + struct inode *tmp_inode) +{ + int retval; + __le32 i_data[3]; + struct ext4_inode_info *ei = EXT4_I(inode); + struct ext4_inode_info *tmp_ei = EXT4_I(tmp_inode); + + /* + * One credit accounted for writing the + * i_data field of the original inode + */ + retval = ext4_journal_extend(handle, 1); + if (retval != 0) { + retval = ext4_journal_restart(handle, 1); + if (retval) + goto err_out; + } + + i_data[0] = ei->i_data[EXT4_IND_BLOCK]; + i_data[1] = ei->i_data[EXT4_DIND_BLOCK]; + i_data[2] = ei->i_data[EXT4_TIND_BLOCK]; + + down_write(&EXT4_I(inode)->i_data_sem); + /* + * We have the extent map build with the tmp inode. + * Now copy the i_data across + */ + ei->i_flags |= EXT4_EXTENTS_FL; + memcpy(ei->i_data, tmp_ei->i_data, sizeof(ei->i_data)); + + /* + * Update i_blocks with the new blocks that got + * allocated while adding extents for extent index + * blocks. + * + * While converting to extents we need not + * update the orignal inode i_blocks for extent blocks + * via quota APIs. The quota update happened via tmp_inode already. + */ + spin_lock(&inode->i_lock); + inode->i_blocks += tmp_inode->i_blocks; + spin_unlock(&inode->i_lock); + up_write(&EXT4_I(inode)->i_data_sem); + + /* + * We mark the inode dirty after, because we decrement the + * i_blocks when freeing the indirect meta-data blocks + */ + retval = free_ind_block(handle, inode, i_data); + ext4_mark_inode_dirty(handle, inode); + +err_out: + return retval; +} + +static int free_ext_idx(handle_t *handle, struct inode *inode, + struct ext4_extent_idx *ix) +{ + int i, retval = 0; + ext4_fsblk_t block; + struct buffer_head *bh; + struct ext4_extent_header *eh; + + block = idx_pblock(ix); + bh = sb_bread(inode->i_sb, block); + if (!bh) + return -EIO; + + eh = (struct ext4_extent_header *)bh->b_data; + if (eh->eh_depth != 0) { + ix = EXT_FIRST_INDEX(eh); + for (i = 0; i < le16_to_cpu(eh->eh_entries); i++, ix++) { + retval = free_ext_idx(handle, inode, ix); + if (retval) + break; + } + } + put_bh(bh); + extend_credit_for_blkdel(handle, inode); + ext4_free_blocks(handle, inode, block, 1, 1); + return retval; +} + +/* + * Free the extent meta data blocks only + */ +static int free_ext_block(handle_t *handle, struct inode *inode) +{ + int i, retval = 0; + struct ext4_inode_info *ei = EXT4_I(inode); + struct ext4_extent_header *eh = (struct ext4_extent_header *)ei->i_data; + struct ext4_extent_idx *ix; + if (eh->eh_depth == 0) + /* + * No extra blocks allocated for extent meta data + */ + return 0; + ix = EXT_FIRST_INDEX(eh); + for (i = 0; i < le16_to_cpu(eh->eh_entries); i++, ix++) { + retval = free_ext_idx(handle, inode, ix); + if (retval) + return retval; + } + return retval; + +} + +int ext4_ext_migrate(struct inode *inode, struct file *filp, + unsigned int cmd, unsigned long arg) +{ + handle_t *handle; + int retval = 0, i; + __le32 *i_data; + ext4_lblk_t blk_count = 0; + struct ext4_inode_info *ei; + struct inode *tmp_inode = NULL; + struct list_blocks_struct lb; + unsigned long max_entries; + + if (!test_opt(inode->i_sb, EXTENTS)) + /* + * if mounted with noextents we don't allow the migrate + */ + return -EINVAL; + + if ((EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)) + return -EINVAL; + + if (S_ISLNK(inode->i_mode) && inode->i_blocks == 0) + /* + * don't migrate fast symlink + */ + return retval; + + handle = ext4_journal_start(inode, + EXT4_DATA_TRANS_BLOCKS(inode->i_sb) + + EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3 + + 2 * EXT4_QUOTA_INIT_BLOCKS(inode->i_sb) + + 1); + if (IS_ERR(handle)) { + retval = PTR_ERR(handle); + goto err_out; + } + tmp_inode = ext4_new_inode(handle, + inode->i_sb->s_root->d_inode, + S_IFREG); + if (IS_ERR(tmp_inode)) { + retval = -ENOMEM; + ext4_journal_stop(handle); + tmp_inode = NULL; + goto err_out; + } + i_size_write(tmp_inode, i_size_read(inode)); + /* + * We don't want the inode to be reclaimed + * if we got interrupted in between. We have + * this tmp inode carrying reference to the + * data blocks of the original file. We set + * the i_nlink to zero at the last stage after + * switching the original file to extent format + */ + tmp_inode->i_nlink = 1; + + ext4_ext_tree_init(handle, tmp_inode); + ext4_orphan_add(handle, tmp_inode); + ext4_journal_stop(handle); + + /* + * start with one credit accounted for + * superblock modification. + * + * For the tmp_inode we already have commited the + * trascation that created the inode. Later as and + * when we add extents we extent the journal + */ + /* + * inode_mutex prevent write and truncate on the file. Read still goes + * through. We take i_data_sem in ext4_ext_swap_inode_data before we + * switch the inode format to prevent read. + */ + mutex_lock(&(inode->i_mutex)); + handle = ext4_journal_start(inode, 1); + + ei = EXT4_I(inode); + i_data = ei->i_data; + memset(&lb, 0, sizeof(lb)); + + /* 32 bit block address 4 bytes */ + max_entries = inode->i_sb->s_blocksize >> 2; + for (i = 0; i < EXT4_NDIR_BLOCKS; i++, blk_count++) { + if (i_data[i]) { + retval = update_extent_range(handle, tmp_inode, + le32_to_cpu(i_data[i]), + blk_count, &lb); + if (retval) + goto err_out; + } + } + if (i_data[EXT4_IND_BLOCK]) { + retval = update_ind_extent_range(handle, tmp_inode, + le32_to_cpu(i_data[EXT4_IND_BLOCK]), + &blk_count, &lb); + if (retval) + goto err_out; + } else + blk_count += max_entries; + if (i_data[EXT4_DIND_BLOCK]) { + retval = update_dind_extent_range(handle, tmp_inode, + le32_to_cpu(i_data[EXT4_DIND_BLOCK]), + &blk_count, &lb); + if (retval) + goto err_out; + } else + blk_count += max_entries * max_entries; + if (i_data[EXT4_TIND_BLOCK]) { + retval = update_tind_extent_range(handle, tmp_inode, + le32_to_cpu(i_data[EXT4_TIND_BLOCK]), + &blk_count, &lb); + if (retval) + goto err_out; + } + /* + * Build the last extent + */ + retval = finish_range(handle, tmp_inode, &lb); +err_out: + if (retval) + /* + * Failure case delete the extent information with the + * tmp_inode + */ + free_ext_block(handle, tmp_inode); + else + retval = ext4_ext_swap_inode_data(handle, inode, + tmp_inode); + + /* We mark the tmp_inode dirty via ext4_ext_tree_init. */ + if (ext4_journal_extend(handle, 1) != 0) + ext4_journal_restart(handle, 1); + + /* + * Mark the tmp_inode as of size zero + */ + i_size_write(tmp_inode, 0); + + /* + * set the i_blocks count to zero + * so that the ext4_delete_inode does the + * right job + * + * We don't need to take the i_lock because + * the inode is not visible to user space. + */ + tmp_inode->i_blocks = 0; + + /* Reset the extent details */ + ext4_ext_tree_init(handle, tmp_inode); + + /* + * Set the i_nlink to zero so that + * generic_drop_inode really deletes the + * inode + */ + tmp_inode->i_nlink = 0; + + ext4_journal_stop(handle); + mutex_unlock(&(inode->i_mutex)); + + if (tmp_inode) + iput(tmp_inode); + + return retval; +} diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 94ee6f315dc..28aa2ed4297 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -51,7 +51,7 @@ static struct buffer_head *ext4_append(handle_t *handle, struct inode *inode, - u32 *block, int *err) + ext4_lblk_t *block, int *err) { struct buffer_head *bh; @@ -144,8 +144,8 @@ struct dx_map_entry u16 size; }; -static inline unsigned dx_get_block (struct dx_entry *entry); -static void dx_set_block (struct dx_entry *entry, unsigned value); +static inline ext4_lblk_t dx_get_block(struct dx_entry *entry); +static void dx_set_block(struct dx_entry *entry, ext4_lblk_t value); static inline unsigned dx_get_hash (struct dx_entry *entry); static void dx_set_hash (struct dx_entry *entry, unsigned value); static unsigned dx_get_count (struct dx_entry *entries); @@ -166,7 +166,8 @@ static void dx_sort_map(struct dx_map_entry *map, unsigned count); static struct ext4_dir_entry_2 *dx_move_dirents (char *from, char *to, struct dx_map_entry *offsets, int count); static struct ext4_dir_entry_2* dx_pack_dirents (char *base, int size); -static void dx_insert_block (struct dx_frame *frame, u32 hash, u32 block); +static void dx_insert_block(struct dx_frame *frame, + u32 hash, ext4_lblk_t block); static int ext4_htree_next_block(struct inode *dir, __u32 hash, struct dx_frame *frame, struct dx_frame *frames, @@ -181,12 +182,12 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, * Mask them off for now. */ -static inline unsigned dx_get_block (struct dx_entry *entry) +static inline ext4_lblk_t dx_get_block(struct dx_entry *entry) { return le32_to_cpu(entry->block) & 0x00ffffff; } -static inline void dx_set_block (struct dx_entry *entry, unsigned value) +static inline void dx_set_block(struct dx_entry *entry, ext4_lblk_t value) { entry->block = cpu_to_le32(value); } @@ -243,8 +244,8 @@ static void dx_show_index (char * label, struct dx_entry *entries) int i, n = dx_get_count (entries); printk("%s index ", label); for (i = 0; i < n; i++) { - printk("%x->%u ", i? dx_get_hash(entries + i) : - 0, dx_get_block(entries + i)); + printk("%x->%lu ", i? dx_get_hash(entries + i) : + 0, (unsigned long)dx_get_block(entries + i)); } printk("\n"); } @@ -280,7 +281,7 @@ static struct stats dx_show_leaf(struct dx_hash_info *hinfo, struct ext4_dir_ent space += EXT4_DIR_REC_LEN(de->name_len); names++; } - de = (struct ext4_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len)); + de = ext4_next_entry(de); } printk("(%i)\n", names); return (struct stats) { names, space, 1 }; @@ -297,7 +298,8 @@ struct stats dx_show_entries(struct dx_hash_info *hinfo, struct inode *dir, printk("%i indexed blocks...\n", count); for (i = 0; i < count; i++, entries++) { - u32 block = dx_get_block(entries), hash = i? dx_get_hash(entries): 0; + ext4_lblk_t block = dx_get_block(entries); + ext4_lblk_t hash = i ? dx_get_hash(entries): 0; u32 range = i < count - 1? (dx_get_hash(entries + 1) - hash): ~hash; struct stats stats; printk("%s%3u:%03u hash %8x/%8x ",levels?"":" ", i, block, hash, range); @@ -551,7 +553,8 @@ static int ext4_htree_next_block(struct inode *dir, __u32 hash, */ static inline struct ext4_dir_entry_2 *ext4_next_entry(struct ext4_dir_entry_2 *p) { - return (struct ext4_dir_entry_2 *)((char*)p + le16_to_cpu(p->rec_len)); + return (struct ext4_dir_entry_2 *)((char *)p + + ext4_rec_len_from_disk(p->rec_len)); } /* @@ -560,7 +563,7 @@ static inline struct ext4_dir_entry_2 *ext4_next_entry(struct ext4_dir_entry_2 * * into the tree. If there is an error it is returned in err. */ static int htree_dirblock_to_tree(struct file *dir_file, - struct inode *dir, int block, + struct inode *dir, ext4_lblk_t block, struct dx_hash_info *hinfo, __u32 start_hash, __u32 start_minor_hash) { @@ -568,7 +571,8 @@ static int htree_dirblock_to_tree(struct file *dir_file, struct ext4_dir_entry_2 *de, *top; int err, count = 0; - dxtrace(printk("In htree dirblock_to_tree: block %d\n", block)); + dxtrace(printk(KERN_INFO "In htree dirblock_to_tree: block %lu\n", + (unsigned long)block)); if (!(bh = ext4_bread (NULL, dir, block, 0, &err))) return err; @@ -620,9 +624,9 @@ int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash, struct ext4_dir_entry_2 *de; struct dx_frame frames[2], *frame; struct inode *dir; - int block, err; + ext4_lblk_t block; int count = 0; - int ret; + int ret, err; __u32 hashval; dxtrace(printk("In htree_fill_tree, start hash: %x:%x\n", start_hash, @@ -720,7 +724,7 @@ static int dx_make_map (struct ext4_dir_entry_2 *de, int size, cond_resched(); } /* XXX: do we need to check rec_len == 0 case? -Chris */ - de = (struct ext4_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len)); + de = ext4_next_entry(de); } return count; } @@ -752,7 +756,7 @@ static void dx_sort_map (struct dx_map_entry *map, unsigned count) } while(more); } -static void dx_insert_block(struct dx_frame *frame, u32 hash, u32 block) +static void dx_insert_block(struct dx_frame *frame, u32 hash, ext4_lblk_t block) { struct dx_entry *entries = frame->entries; struct dx_entry *old = frame->at, *new = old + 1; @@ -820,7 +824,7 @@ static inline int search_dirblock(struct buffer_head * bh, return 1; } /* prevent looping on a bad block */ - de_len = le16_to_cpu(de->rec_len); + de_len = ext4_rec_len_from_disk(de->rec_len); if (de_len <= 0) return -1; offset += de_len; @@ -847,23 +851,20 @@ static struct buffer_head * ext4_find_entry (struct dentry *dentry, struct super_block * sb; struct buffer_head * bh_use[NAMEI_RA_SIZE]; struct buffer_head * bh, *ret = NULL; - unsigned long start, block, b; + ext4_lblk_t start, block, b; int ra_max = 0; /* Number of bh's in the readahead buffer, bh_use[] */ int ra_ptr = 0; /* Current index into readahead buffer */ int num = 0; - int nblocks, i, err; + ext4_lblk_t nblocks; + int i, err; struct inode *dir = dentry->d_parent->d_inode; int namelen; - const u8 *name; - unsigned blocksize; *res_dir = NULL; sb = dir->i_sb; - blocksize = sb->s_blocksize; namelen = dentry->d_name.len; - name = dentry->d_name.name; if (namelen > EXT4_NAME_LEN) return NULL; if (is_dx(dir)) { @@ -914,7 +915,8 @@ restart: if (!buffer_uptodate(bh)) { /* read error, skip block & hope for the best */ ext4_error(sb, __FUNCTION__, "reading directory #%lu " - "offset %lu", dir->i_ino, block); + "offset %lu", dir->i_ino, + (unsigned long)block); brelse(bh); goto next; } @@ -961,7 +963,7 @@ static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, struct dx_frame frames[2], *frame; struct ext4_dir_entry_2 *de, *top; struct buffer_head *bh; - unsigned long block; + ext4_lblk_t block; int retval; int namelen = dentry->d_name.len; const u8 *name = dentry->d_name.name; @@ -1037,17 +1039,11 @@ static struct dentry *ext4_lookup(struct inode * dir, struct dentry *dentry, str if (!ext4_valid_inum(dir->i_sb, ino)) { ext4_error(dir->i_sb, "ext4_lookup", "bad inode number: %lu", ino); - inode = NULL; - } else - inode = iget(dir->i_sb, ino); - - if (!inode) - return ERR_PTR(-EACCES); - - if (is_bad_inode(inode)) { - iput(inode); - return ERR_PTR(-ENOENT); + return ERR_PTR(-EIO); } + inode = ext4_iget(dir->i_sb, ino); + if (IS_ERR(inode)) + return ERR_CAST(inode); } return d_splice_alias(inode, dentry); } @@ -1076,18 +1072,13 @@ struct dentry *ext4_get_parent(struct dentry *child) if (!ext4_valid_inum(child->d_inode->i_sb, ino)) { ext4_error(child->d_inode->i_sb, "ext4_get_parent", "bad inode number: %lu", ino); - inode = NULL; - } else - inode = iget(child->d_inode->i_sb, ino); - - if (!inode) - return ERR_PTR(-EACCES); - - if (is_bad_inode(inode)) { - iput(inode); - return ERR_PTR(-ENOENT); + return ERR_PTR(-EIO); } + inode = ext4_iget(child->d_inode->i_sb, ino); + if (IS_ERR(inode)) + return ERR_CAST(inode); + parent = d_alloc_anon(inode); if (!parent) { iput(inode); @@ -1128,7 +1119,7 @@ dx_move_dirents(char *from, char *to, struct dx_map_entry *map, int count) rec_len = EXT4_DIR_REC_LEN(de->name_len); memcpy (to, de, rec_len); ((struct ext4_dir_entry_2 *) to)->rec_len = - cpu_to_le16(rec_len); + ext4_rec_len_to_disk(rec_len); de->inode = 0; map++; to += rec_len; @@ -1147,13 +1138,12 @@ static struct ext4_dir_entry_2* dx_pack_dirents(char *base, int size) prev = to = de; while ((char*)de < base + size) { - next = (struct ext4_dir_entry_2 *) ((char *) de + - le16_to_cpu(de->rec_len)); + next = ext4_next_entry(de); if (de->inode && de->name_len) { rec_len = EXT4_DIR_REC_LEN(de->name_len); if (de > to) memmove(to, de, rec_len); - to->rec_len = cpu_to_le16(rec_len); + to->rec_len = ext4_rec_len_to_disk(rec_len); prev = to; to = (struct ext4_dir_entry_2 *) (((char *) to) + rec_len); } @@ -1174,7 +1164,7 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, unsigned blocksize = dir->i_sb->s_blocksize; unsigned count, continued; struct buffer_head *bh2; - u32 newblock; + ext4_lblk_t newblock; u32 hash2; struct dx_map_entry *map; char *data1 = (*bh)->b_data, *data2; @@ -1221,14 +1211,15 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, split = count - move; hash2 = map[split].hash; continued = hash2 == map[split - 1].hash; - dxtrace(printk("Split block %i at %x, %i/%i\n", - dx_get_block(frame->at), hash2, split, count-split)); + dxtrace(printk(KERN_INFO "Split block %lu at %x, %i/%i\n", + (unsigned long)dx_get_block(frame->at), + hash2, split, count-split)); /* Fancy dance to stay within two buffers */ de2 = dx_move_dirents(data1, data2, map + split, count - split); de = dx_pack_dirents(data1,blocksize); - de->rec_len = cpu_to_le16(data1 + blocksize - (char *) de); - de2->rec_len = cpu_to_le16(data2 + blocksize - (char *) de2); + de->rec_len = ext4_rec_len_to_disk(data1 + blocksize - (char *) de); + de2->rec_len = ext4_rec_len_to_disk(data2 + blocksize - (char *) de2); dxtrace(dx_show_leaf (hinfo, (struct ext4_dir_entry_2 *) data1, blocksize, 1)); dxtrace(dx_show_leaf (hinfo, (struct ext4_dir_entry_2 *) data2, blocksize, 1)); @@ -1297,7 +1288,7 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry, return -EEXIST; } nlen = EXT4_DIR_REC_LEN(de->name_len); - rlen = le16_to_cpu(de->rec_len); + rlen = ext4_rec_len_from_disk(de->rec_len); if ((de->inode? rlen - nlen: rlen) >= reclen) break; de = (struct ext4_dir_entry_2 *)((char *)de + rlen); @@ -1316,11 +1307,11 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry, /* By now the buffer is marked for journaling */ nlen = EXT4_DIR_REC_LEN(de->name_len); - rlen = le16_to_cpu(de->rec_len); + rlen = ext4_rec_len_from_disk(de->rec_len); if (de->inode) { struct ext4_dir_entry_2 *de1 = (struct ext4_dir_entry_2 *)((char *)de + nlen); - de1->rec_len = cpu_to_le16(rlen - nlen); - de->rec_len = cpu_to_le16(nlen); + de1->rec_len = ext4_rec_len_to_disk(rlen - nlen); + de->rec_len = ext4_rec_len_to_disk(nlen); de = de1; } de->file_type = EXT4_FT_UNKNOWN; @@ -1374,7 +1365,7 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, int retval; unsigned blocksize; struct dx_hash_info hinfo; - u32 block; + ext4_lblk_t block; struct fake_dirent *fde; blocksize = dir->i_sb->s_blocksize; @@ -1397,17 +1388,18 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, /* The 0th block becomes the root, move the dirents out */ fde = &root->dotdot; - de = (struct ext4_dir_entry_2 *)((char *)fde + le16_to_cpu(fde->rec_len)); + de = (struct ext4_dir_entry_2 *)((char *)fde + + ext4_rec_len_from_disk(fde->rec_len)); len = ((char *) root) + blocksize - (char *) de; memcpy (data1, de, len); de = (struct ext4_dir_entry_2 *) data1; top = data1 + len; - while ((char *)(de2=(void*)de+le16_to_cpu(de->rec_len)) < top) + while ((char *)(de2 = ext4_next_entry(de)) < top) de = de2; - de->rec_len = cpu_to_le16(data1 + blocksize - (char *) de); + de->rec_len = ext4_rec_len_to_disk(data1 + blocksize - (char *) de); /* Initialize the root; the dot dirents already exist */ de = (struct ext4_dir_entry_2 *) (&root->dotdot); - de->rec_len = cpu_to_le16(blocksize - EXT4_DIR_REC_LEN(2)); + de->rec_len = ext4_rec_len_to_disk(blocksize - EXT4_DIR_REC_LEN(2)); memset (&root->info, 0, sizeof(root->info)); root->info.info_length = sizeof(root->info); root->info.hash_version = EXT4_SB(dir->i_sb)->s_def_hash_version; @@ -1454,7 +1446,7 @@ static int ext4_add_entry (handle_t *handle, struct dentry *dentry, int retval; int dx_fallback=0; unsigned blocksize; - u32 block, blocks; + ext4_lblk_t block, blocks; sb = dir->i_sb; blocksize = sb->s_blocksize; @@ -1487,7 +1479,7 @@ static int ext4_add_entry (handle_t *handle, struct dentry *dentry, return retval; de = (struct ext4_dir_entry_2 *) bh->b_data; de->inode = 0; - de->rec_len = cpu_to_le16(blocksize); + de->rec_len = ext4_rec_len_to_disk(blocksize); return add_dirent_to_buf(handle, dentry, inode, de, bh); } @@ -1531,7 +1523,7 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, dx_get_count(entries), dx_get_limit(entries))); /* Need to split index? */ if (dx_get_count(entries) == dx_get_limit(entries)) { - u32 newblock; + ext4_lblk_t newblock; unsigned icount = dx_get_count(entries); int levels = frame - frames; struct dx_entry *entries2; @@ -1550,7 +1542,7 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, goto cleanup; node2 = (struct dx_node *)(bh2->b_data); entries2 = node2->entries; - node2->fake.rec_len = cpu_to_le16(sb->s_blocksize); + node2->fake.rec_len = ext4_rec_len_to_disk(sb->s_blocksize); node2->fake.inode = 0; BUFFER_TRACE(frame->bh, "get_write_access"); err = ext4_journal_get_write_access(handle, frame->bh); @@ -1648,9 +1640,9 @@ static int ext4_delete_entry (handle_t *handle, BUFFER_TRACE(bh, "get_write_access"); ext4_journal_get_write_access(handle, bh); if (pde) - pde->rec_len = - cpu_to_le16(le16_to_cpu(pde->rec_len) + - le16_to_cpu(de->rec_len)); + pde->rec_len = ext4_rec_len_to_disk( + ext4_rec_len_from_disk(pde->rec_len) + + ext4_rec_len_from_disk(de->rec_len)); else de->inode = 0; dir->i_version++; @@ -1658,10 +1650,9 @@ static int ext4_delete_entry (handle_t *handle, ext4_journal_dirty_metadata(handle, bh); return 0; } - i += le16_to_cpu(de->rec_len); + i += ext4_rec_len_from_disk(de->rec_len); pde = de; - de = (struct ext4_dir_entry_2 *) - ((char *) de + le16_to_cpu(de->rec_len)); + de = ext4_next_entry(de); } return -ENOENT; } @@ -1813,24 +1804,20 @@ retry: inode->i_fop = &ext4_dir_operations; inode->i_size = EXT4_I(inode)->i_disksize = inode->i_sb->s_blocksize; dir_block = ext4_bread (handle, inode, 0, 1, &err); - if (!dir_block) { - ext4_dec_count(handle, inode); /* is this nlink == 0? */ - ext4_mark_inode_dirty(handle, inode); - iput (inode); - goto out_stop; - } + if (!dir_block) + goto out_clear_inode; BUFFER_TRACE(dir_block, "get_write_access"); ext4_journal_get_write_access(handle, dir_block); de = (struct ext4_dir_entry_2 *) dir_block->b_data; de->inode = cpu_to_le32(inode->i_ino); de->name_len = 1; - de->rec_len = cpu_to_le16(EXT4_DIR_REC_LEN(de->name_len)); + de->rec_len = ext4_rec_len_to_disk(EXT4_DIR_REC_LEN(de->name_len)); strcpy (de->name, "."); ext4_set_de_type(dir->i_sb, de, S_IFDIR); - de = (struct ext4_dir_entry_2 *) - ((char *) de + le16_to_cpu(de->rec_len)); + de = ext4_next_entry(de); de->inode = cpu_to_le32(dir->i_ino); - de->rec_len = cpu_to_le16(inode->i_sb->s_blocksize-EXT4_DIR_REC_LEN(1)); + de->rec_len = ext4_rec_len_to_disk(inode->i_sb->s_blocksize - + EXT4_DIR_REC_LEN(1)); de->name_len = 2; strcpy (de->name, ".."); ext4_set_de_type(dir->i_sb, de, S_IFDIR); @@ -1841,7 +1828,8 @@ retry: ext4_mark_inode_dirty(handle, inode); err = ext4_add_entry (handle, dentry, inode); if (err) { - inode->i_nlink = 0; +out_clear_inode: + clear_nlink(inode); ext4_mark_inode_dirty(handle, inode); iput (inode); goto out_stop; @@ -1882,8 +1870,7 @@ static int empty_dir (struct inode * inode) return 1; } de = (struct ext4_dir_entry_2 *) bh->b_data; - de1 = (struct ext4_dir_entry_2 *) - ((char *) de + le16_to_cpu(de->rec_len)); + de1 = ext4_next_entry(de); if (le32_to_cpu(de->inode) != inode->i_ino || !le32_to_cpu(de1->inode) || strcmp (".", de->name) || @@ -1894,9 +1881,9 @@ static int empty_dir (struct inode * inode) brelse (bh); return 1; } - offset = le16_to_cpu(de->rec_len) + le16_to_cpu(de1->rec_len); - de = (struct ext4_dir_entry_2 *) - ((char *) de1 + le16_to_cpu(de1->rec_len)); + offset = ext4_rec_len_from_disk(de->rec_len) + + ext4_rec_len_from_disk(de1->rec_len); + de = ext4_next_entry(de1); while (offset < inode->i_size ) { if (!bh || (void *) de >= (void *) (bh->b_data+sb->s_blocksize)) { @@ -1925,9 +1912,8 @@ static int empty_dir (struct inode * inode) brelse (bh); return 0; } - offset += le16_to_cpu(de->rec_len); - de = (struct ext4_dir_entry_2 *) - ((char *) de + le16_to_cpu(de->rec_len)); + offset += ext4_rec_len_from_disk(de->rec_len); + de = ext4_next_entry(de); } brelse (bh); return 1; @@ -2175,7 +2161,7 @@ static int ext4_unlink(struct inode * dir, struct dentry *dentry) dir->i_ctime = dir->i_mtime = ext4_current_time(dir); ext4_update_dx_flag(dir); ext4_mark_inode_dirty(handle, dir); - ext4_dec_count(handle, inode); + drop_nlink(inode); if (!inode->i_nlink) ext4_orphan_add(handle, inode); inode->i_ctime = ext4_current_time(inode); @@ -2225,7 +2211,7 @@ retry: err = __page_symlink(inode, symname, l, mapping_gfp_mask(inode->i_mapping) & ~__GFP_FS); if (err) { - ext4_dec_count(handle, inode); + clear_nlink(inode); ext4_mark_inode_dirty(handle, inode); iput (inode); goto out_stop; @@ -2282,8 +2268,7 @@ retry: } #define PARENT_INO(buffer) \ - ((struct ext4_dir_entry_2 *) ((char *) buffer + \ - le16_to_cpu(((struct ext4_dir_entry_2 *) buffer)->rec_len)))->inode + (ext4_next_entry((struct ext4_dir_entry_2 *)(buffer))->inode) /* * Anybody can rename anything with this: the permission checks are left to the @@ -2418,7 +2403,7 @@ static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry, ext4_dec_count(handle, old_dir); if (new_inode) { /* checked empty_dir above, can't have another parent, - * ext3_dec_count() won't work for many-linked dirs */ + * ext4_dec_count() won't work for many-linked dirs */ new_inode->i_nlink = 0; } else { ext4_inc_count(handle, new_dir); diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index bd8a52bb399..e29efa0f9d6 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -28,7 +28,7 @@ static int verify_group_input(struct super_block *sb, struct ext4_super_block *es = sbi->s_es; ext4_fsblk_t start = ext4_blocks_count(es); ext4_fsblk_t end = start + input->blocks_count; - unsigned group = input->group; + ext4_group_t group = input->group; ext4_fsblk_t itend = input->inode_table + sbi->s_itb_per_group; unsigned overhead = ext4_bg_has_super(sb, group) ? (1 + ext4_bg_num_gdb(sb, group) + @@ -206,7 +206,7 @@ static int setup_new_group_blocks(struct super_block *sb, } if (ext4_bg_has_super(sb, input->group)) { - ext4_debug("mark backup superblock %#04lx (+0)\n", start); + ext4_debug("mark backup superblock %#04llx (+0)\n", start); ext4_set_bit(0, bh->b_data); } @@ -215,7 +215,7 @@ static int setup_new_group_blocks(struct super_block *sb, i < gdblocks; i++, block++, bit++) { struct buffer_head *gdb; - ext4_debug("update backup group %#04lx (+%d)\n", block, bit); + ext4_debug("update backup group %#04llx (+%d)\n", block, bit); if ((err = extend_or_restart_transaction(handle, 1, bh))) goto exit_bh; @@ -243,7 +243,7 @@ static int setup_new_group_blocks(struct super_block *sb, i < reserved_gdb; i++, block++, bit++) { struct buffer_head *gdb; - ext4_debug("clear reserved block %#04lx (+%d)\n", block, bit); + ext4_debug("clear reserved block %#04llx (+%d)\n", block, bit); if ((err = extend_or_restart_transaction(handle, 1, bh))) goto exit_bh; @@ -256,10 +256,10 @@ static int setup_new_group_blocks(struct super_block *sb, ext4_set_bit(bit, bh->b_data); brelse(gdb); } - ext4_debug("mark block bitmap %#04x (+%ld)\n", input->block_bitmap, + ext4_debug("mark block bitmap %#04llx (+%llu)\n", input->block_bitmap, input->block_bitmap - start); ext4_set_bit(input->block_bitmap - start, bh->b_data); - ext4_debug("mark inode bitmap %#04x (+%ld)\n", input->inode_bitmap, + ext4_debug("mark inode bitmap %#04llx (+%llu)\n", input->inode_bitmap, input->inode_bitmap - start); ext4_set_bit(input->inode_bitmap - start, bh->b_data); @@ -268,7 +268,7 @@ static int setup_new_group_blocks(struct super_block *sb, i < sbi->s_itb_per_group; i++, bit++, block++) { struct buffer_head *it; - ext4_debug("clear inode block %#04lx (+%d)\n", block, bit); + ext4_debug("clear inode block %#04llx (+%d)\n", block, bit); if ((err = extend_or_restart_transaction(handle, 1, bh))) goto exit_bh; @@ -291,7 +291,7 @@ static int setup_new_group_blocks(struct super_block *sb, brelse(bh); /* Mark unused entries in inode bitmap used */ - ext4_debug("clear inode bitmap %#04x (+%ld)\n", + ext4_debug("clear inode bitmap %#04llx (+%llu)\n", input->inode_bitmap, input->inode_bitmap - start); if (IS_ERR(bh = bclean(handle, sb, input->inode_bitmap))) { err = PTR_ERR(bh); @@ -357,7 +357,7 @@ static int verify_reserved_gdb(struct super_block *sb, struct buffer_head *primary) { const ext4_fsblk_t blk = primary->b_blocknr; - const unsigned long end = EXT4_SB(sb)->s_groups_count; + const ext4_group_t end = EXT4_SB(sb)->s_groups_count; unsigned three = 1; unsigned five = 5; unsigned seven = 7; @@ -656,12 +656,12 @@ static void update_backups(struct super_block *sb, int blk_off, char *data, int size) { struct ext4_sb_info *sbi = EXT4_SB(sb); - const unsigned long last = sbi->s_groups_count; + const ext4_group_t last = sbi->s_groups_count; const int bpg = EXT4_BLOCKS_PER_GROUP(sb); unsigned three = 1; unsigned five = 5; unsigned seven = 7; - unsigned group; + ext4_group_t group; int rest = sb->s_blocksize - size; handle_t *handle; int err = 0, err2; @@ -716,7 +716,7 @@ static void update_backups(struct super_block *sb, exit_err: if (err) { ext4_warning(sb, __FUNCTION__, - "can't update backup for group %d (err %d), " + "can't update backup for group %lu (err %d), " "forcing fsck on next reboot", group, err); sbi->s_mount_state &= ~EXT4_VALID_FS; sbi->s_es->s_state &= cpu_to_le16(~EXT4_VALID_FS); @@ -779,12 +779,11 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) "No reserved GDT blocks, can't resize"); return -EPERM; } - inode = iget(sb, EXT4_RESIZE_INO); - if (!inode || is_bad_inode(inode)) { + inode = ext4_iget(sb, EXT4_RESIZE_INO); + if (IS_ERR(inode)) { ext4_warning(sb, __FUNCTION__, "Error opening resize inode"); - iput(inode); - return -ENOENT; + return PTR_ERR(inode); } } @@ -952,7 +951,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, ext4_fsblk_t n_blocks_count) { ext4_fsblk_t o_blocks_count; - unsigned long o_groups_count; + ext4_group_t o_groups_count; ext4_grpblk_t last; ext4_grpblk_t add; struct buffer_head * bh; @@ -1038,6 +1037,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, ext4_warning(sb, __FUNCTION__, "multiple resizers run on filesystem!"); unlock_super(sb); + ext4_journal_stop(handle); err = -EBUSY; goto exit_put; } @@ -1054,7 +1054,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh); sb->s_dirt = 1; unlock_super(sb); - ext4_debug("freeing blocks %lu through %llu\n", o_blocks_count, + ext4_debug("freeing blocks %llu through %llu\n", o_blocks_count, o_blocks_count + add); ext4_free_blocks_sb(handle, sb, o_blocks_count, add, &freed_blocks); ext4_debug("freed blocks %llu through %llu\n", o_blocks_count, diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 1ca0f546c46..13383ba18f1 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -373,6 +373,66 @@ void ext4_update_dynamic_rev(struct super_block *sb) */ } +int ext4_update_compat_feature(handle_t *handle, + struct super_block *sb, __u32 compat) +{ + int err = 0; + if (!EXT4_HAS_COMPAT_FEATURE(sb, compat)) { + err = ext4_journal_get_write_access(handle, + EXT4_SB(sb)->s_sbh); + if (err) + return err; + EXT4_SET_COMPAT_FEATURE(sb, compat); + sb->s_dirt = 1; + handle->h_sync = 1; + BUFFER_TRACE(EXT4_SB(sb)->s_sbh, + "call ext4_journal_dirty_met adata"); + err = ext4_journal_dirty_metadata(handle, + EXT4_SB(sb)->s_sbh); + } + return err; +} + +int ext4_update_rocompat_feature(handle_t *handle, + struct super_block *sb, __u32 rocompat) +{ + int err = 0; + if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, rocompat)) { + err = ext4_journal_get_write_access(handle, + EXT4_SB(sb)->s_sbh); + if (err) + return err; + EXT4_SET_RO_COMPAT_FEATURE(sb, rocompat); + sb->s_dirt = 1; + handle->h_sync = 1; + BUFFER_TRACE(EXT4_SB(sb)->s_sbh, + "call ext4_journal_dirty_met adata"); + err = ext4_journal_dirty_metadata(handle, + EXT4_SB(sb)->s_sbh); + } + return err; +} + +int ext4_update_incompat_feature(handle_t *handle, + struct super_block *sb, __u32 incompat) +{ + int err = 0; + if (!EXT4_HAS_INCOMPAT_FEATURE(sb, incompat)) { + err = ext4_journal_get_write_access(handle, + EXT4_SB(sb)->s_sbh); + if (err) + return err; + EXT4_SET_INCOMPAT_FEATURE(sb, incompat); + sb->s_dirt = 1; + handle->h_sync = 1; + BUFFER_TRACE(EXT4_SB(sb)->s_sbh, + "call ext4_journal_dirty_met adata"); + err = ext4_journal_dirty_metadata(handle, + EXT4_SB(sb)->s_sbh); + } + return err; +} + /* * Open the external journal device */ @@ -443,6 +503,7 @@ static void ext4_put_super (struct super_block * sb) struct ext4_super_block *es = sbi->s_es; int i; + ext4_mb_release(sb); ext4_ext_release(sb); ext4_xattr_put_super(sb); jbd2_journal_destroy(sbi->s_journal); @@ -509,6 +570,8 @@ static struct inode *ext4_alloc_inode(struct super_block *sb) ei->i_block_alloc_info = NULL; ei->vfs_inode.i_version = 1; memset(&ei->i_cached_extent, 0, sizeof(struct ext4_ext_cache)); + INIT_LIST_HEAD(&ei->i_prealloc_list); + spin_lock_init(&ei->i_prealloc_lock); return &ei->vfs_inode; } @@ -533,7 +596,7 @@ static void init_once(struct kmem_cache *cachep, void *foo) #ifdef CONFIG_EXT4DEV_FS_XATTR init_rwsem(&ei->xattr_sem); #endif - mutex_init(&ei->truncate_mutex); + init_rwsem(&ei->i_data_sem); inode_init_once(&ei->vfs_inode); } @@ -605,18 +668,20 @@ static inline void ext4_show_quota_options(struct seq_file *seq, struct super_bl */ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) { + int def_errors; + unsigned long def_mount_opts; struct super_block *sb = vfs->mnt_sb; struct ext4_sb_info *sbi = EXT4_SB(sb); struct ext4_super_block *es = sbi->s_es; - unsigned long def_mount_opts; def_mount_opts = le32_to_cpu(es->s_default_mount_opts); + def_errors = le16_to_cpu(es->s_errors); if (sbi->s_sb_block != 1) seq_printf(seq, ",sb=%llu", sbi->s_sb_block); if (test_opt(sb, MINIX_DF)) seq_puts(seq, ",minixdf"); - if (test_opt(sb, GRPID)) + if (test_opt(sb, GRPID) && !(def_mount_opts & EXT4_DEFM_BSDGROUPS)) seq_puts(seq, ",grpid"); if (!test_opt(sb, GRPID) && (def_mount_opts & EXT4_DEFM_BSDGROUPS)) seq_puts(seq, ",nogrpid"); @@ -628,34 +693,33 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) le16_to_cpu(es->s_def_resgid) != EXT4_DEF_RESGID) { seq_printf(seq, ",resgid=%u", sbi->s_resgid); } - if (test_opt(sb, ERRORS_CONT)) { - int def_errors = le16_to_cpu(es->s_errors); - + if (test_opt(sb, ERRORS_RO)) { if (def_errors == EXT4_ERRORS_PANIC || - def_errors == EXT4_ERRORS_RO) { - seq_puts(seq, ",errors=continue"); + def_errors == EXT4_ERRORS_CONTINUE) { + seq_puts(seq, ",errors=remount-ro"); } } - if (test_opt(sb, ERRORS_RO)) - seq_puts(seq, ",errors=remount-ro"); - if (test_opt(sb, ERRORS_PANIC)) + if (test_opt(sb, ERRORS_CONT) && def_errors != EXT4_ERRORS_CONTINUE) + seq_puts(seq, ",errors=continue"); + if (test_opt(sb, ERRORS_PANIC) && def_errors != EXT4_ERRORS_PANIC) seq_puts(seq, ",errors=panic"); - if (test_opt(sb, NO_UID32)) + if (test_opt(sb, NO_UID32) && !(def_mount_opts & EXT4_DEFM_UID16)) seq_puts(seq, ",nouid32"); - if (test_opt(sb, DEBUG)) + if (test_opt(sb, DEBUG) && !(def_mount_opts & EXT4_DEFM_DEBUG)) seq_puts(seq, ",debug"); if (test_opt(sb, OLDALLOC)) seq_puts(seq, ",oldalloc"); -#ifdef CONFIG_EXT4_FS_XATTR - if (test_opt(sb, XATTR_USER)) +#ifdef CONFIG_EXT4DEV_FS_XATTR + if (test_opt(sb, XATTR_USER) && + !(def_mount_opts & EXT4_DEFM_XATTR_USER)) seq_puts(seq, ",user_xattr"); if (!test_opt(sb, XATTR_USER) && (def_mount_opts & EXT4_DEFM_XATTR_USER)) { seq_puts(seq, ",nouser_xattr"); } #endif -#ifdef CONFIG_EXT4_FS_POSIX_ACL - if (test_opt(sb, POSIX_ACL)) +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL + if (test_opt(sb, POSIX_ACL) && !(def_mount_opts & EXT4_DEFM_ACL)) seq_puts(seq, ",acl"); if (!test_opt(sb, POSIX_ACL) && (def_mount_opts & EXT4_DEFM_ACL)) seq_puts(seq, ",noacl"); @@ -672,7 +736,17 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) seq_puts(seq, ",nobh"); if (!test_opt(sb, EXTENTS)) seq_puts(seq, ",noextents"); + if (!test_opt(sb, MBALLOC)) + seq_puts(seq, ",nomballoc"); + if (test_opt(sb, I_VERSION)) + seq_puts(seq, ",i_version"); + if (sbi->s_stripe) + seq_printf(seq, ",stripe=%lu", sbi->s_stripe); + /* + * journal mode get enabled in different ways + * So just print the value even if we didn't specify it + */ if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) seq_puts(seq, ",data=journal"); else if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA) @@ -681,7 +755,6 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) seq_puts(seq, ",data=writeback"); ext4_show_quota_options(seq, sb); - return 0; } @@ -704,11 +777,10 @@ static struct inode *ext4_nfs_get_inode(struct super_block *sb, * Currently we don't know the generation for parent directory, so * a generation of 0 means "accept any" */ - inode = iget(sb, ino); - if (inode == NULL) - return ERR_PTR(-ENOMEM); - if (is_bad_inode(inode) || - (generation && inode->i_generation != generation)) { + inode = ext4_iget(sb, ino); + if (IS_ERR(inode)) + return ERR_CAST(inode); + if (generation && inode->i_generation != generation) { iput(inode); return ERR_PTR(-ESTALE); } @@ -777,7 +849,6 @@ static struct quotactl_ops ext4_qctl_operations = { static const struct super_operations ext4_sops = { .alloc_inode = ext4_alloc_inode, .destroy_inode = ext4_destroy_inode, - .read_inode = ext4_read_inode, .write_inode = ext4_write_inode, .dirty_inode = ext4_dirty_inode, .delete_inode = ext4_delete_inode, @@ -809,11 +880,13 @@ enum { Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl, Opt_reservation, Opt_noreservation, Opt_noload, Opt_nobh, Opt_bh, Opt_commit, Opt_journal_update, Opt_journal_inum, Opt_journal_dev, + Opt_journal_checksum, Opt_journal_async_commit, Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback, Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota, Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota, Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota, - Opt_grpquota, Opt_extents, Opt_noextents, + Opt_grpquota, Opt_extents, Opt_noextents, Opt_i_version, + Opt_mballoc, Opt_nomballoc, Opt_stripe, }; static match_table_t tokens = { @@ -848,6 +921,8 @@ static match_table_t tokens = { {Opt_journal_update, "journal=update"}, {Opt_journal_inum, "journal=%u"}, {Opt_journal_dev, "journal_dev=%u"}, + {Opt_journal_checksum, "journal_checksum"}, + {Opt_journal_async_commit, "journal_async_commit"}, {Opt_abort, "abort"}, {Opt_data_journal, "data=journal"}, {Opt_data_ordered, "data=ordered"}, @@ -865,6 +940,10 @@ static match_table_t tokens = { {Opt_barrier, "barrier=%u"}, {Opt_extents, "extents"}, {Opt_noextents, "noextents"}, + {Opt_i_version, "i_version"}, + {Opt_mballoc, "mballoc"}, + {Opt_nomballoc, "nomballoc"}, + {Opt_stripe, "stripe=%u"}, {Opt_err, NULL}, {Opt_resize, "resize"}, }; @@ -1035,6 +1114,13 @@ static int parse_options (char *options, struct super_block *sb, return 0; *journal_devnum = option; break; + case Opt_journal_checksum: + set_opt(sbi->s_mount_opt, JOURNAL_CHECKSUM); + break; + case Opt_journal_async_commit: + set_opt(sbi->s_mount_opt, JOURNAL_ASYNC_COMMIT); + set_opt(sbi->s_mount_opt, JOURNAL_CHECKSUM); + break; case Opt_noload: set_opt (sbi->s_mount_opt, NOLOAD); break; @@ -1203,6 +1289,23 @@ clear_qf_name: case Opt_noextents: clear_opt (sbi->s_mount_opt, EXTENTS); break; + case Opt_i_version: + set_opt(sbi->s_mount_opt, I_VERSION); + sb->s_flags |= MS_I_VERSION; + break; + case Opt_mballoc: + set_opt(sbi->s_mount_opt, MBALLOC); + break; + case Opt_nomballoc: + clear_opt(sbi->s_mount_opt, MBALLOC); + break; + case Opt_stripe: + if (match_int(&args[0], &option)) + return 0; + if (option < 0) + return 0; + sbi->s_stripe = option; + break; default: printk (KERN_ERR "EXT4-fs: Unrecognized mount option \"%s\" " @@ -1353,7 +1456,7 @@ int ext4_group_desc_csum_verify(struct ext4_sb_info *sbi, __u32 block_group, } /* Called at mount-time, super-block is locked */ -static int ext4_check_descriptors (struct super_block * sb) +static int ext4_check_descriptors(struct super_block *sb) { struct ext4_sb_info *sbi = EXT4_SB(sb); ext4_fsblk_t first_block = le32_to_cpu(sbi->s_es->s_first_data_block); @@ -1361,32 +1464,28 @@ static int ext4_check_descriptors (struct super_block * sb) ext4_fsblk_t block_bitmap; ext4_fsblk_t inode_bitmap; ext4_fsblk_t inode_table; - struct ext4_group_desc * gdp = NULL; - int desc_block = 0; int flexbg_flag = 0; - int i; + ext4_group_t i; if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG)) flexbg_flag = 1; ext4_debug ("Checking group descriptors"); - for (i = 0; i < sbi->s_groups_count; i++) - { + for (i = 0; i < sbi->s_groups_count; i++) { + struct ext4_group_desc *gdp = ext4_get_group_desc(sb, i, NULL); + if (i == sbi->s_groups_count - 1 || flexbg_flag) last_block = ext4_blocks_count(sbi->s_es) - 1; else last_block = first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1); - if ((i % EXT4_DESC_PER_BLOCK(sb)) == 0) - gdp = (struct ext4_group_desc *) - sbi->s_group_desc[desc_block++]->b_data; block_bitmap = ext4_block_bitmap(sb, gdp); if (block_bitmap < first_block || block_bitmap > last_block) { ext4_error (sb, "ext4_check_descriptors", - "Block bitmap for group %d" + "Block bitmap for group %lu" " not in group (block %llu)!", i, block_bitmap); return 0; @@ -1395,7 +1494,7 @@ static int ext4_check_descriptors (struct super_block * sb) if (inode_bitmap < first_block || inode_bitmap > last_block) { ext4_error (sb, "ext4_check_descriptors", - "Inode bitmap for group %d" + "Inode bitmap for group %lu" " not in group (block %llu)!", i, inode_bitmap); return 0; @@ -1405,23 +1504,20 @@ static int ext4_check_descriptors (struct super_block * sb) inode_table + sbi->s_itb_per_group - 1 > last_block) { ext4_error (sb, "ext4_check_descriptors", - "Inode table for group %d" + "Inode table for group %lu" " not in group (block %llu)!", i, inode_table); return 0; } if (!ext4_group_desc_csum_verify(sbi, i, gdp)) { ext4_error(sb, __FUNCTION__, - "Checksum for group %d failed (%u!=%u)\n", i, - le16_to_cpu(ext4_group_desc_csum(sbi, i, - gdp)), - le16_to_cpu(gdp->bg_checksum)); + "Checksum for group %lu failed (%u!=%u)\n", + i, le16_to_cpu(ext4_group_desc_csum(sbi, i, + gdp)), le16_to_cpu(gdp->bg_checksum)); return 0; } if (!flexbg_flag) first_block += EXT4_BLOCKS_PER_GROUP(sb); - gdp = (struct ext4_group_desc *) - ((__u8 *)gdp + EXT4_DESC_SIZE(sb)); } ext4_free_blocks_count_set(sbi->s_es, ext4_count_free_blocks(sb)); @@ -1429,7 +1525,6 @@ static int ext4_check_descriptors (struct super_block * sb) return 1; } - /* ext4_orphan_cleanup() walks a singly-linked list of inodes (starting at * the superblock) which were deleted from all directories, but held open by * a process at the time of a crash. We walk the list and try to delete these @@ -1542,20 +1637,95 @@ static void ext4_orphan_cleanup (struct super_block * sb, #endif sb->s_flags = s_flags; /* Restore MS_RDONLY status */ } +/* + * Maximal extent format file size. + * Resulting logical blkno at s_maxbytes must fit in our on-disk + * extent format containers, within a sector_t, and within i_blocks + * in the vfs. ext4 inode has 48 bits of i_block in fsblock units, + * so that won't be a limiting factor. + * + * Note, this does *not* consider any metadata overhead for vfs i_blocks. + */ +static loff_t ext4_max_size(int blkbits) +{ + loff_t res; + loff_t upper_limit = MAX_LFS_FILESIZE; + + /* small i_blocks in vfs inode? */ + if (sizeof(blkcnt_t) < sizeof(u64)) { + /* + * CONFIG_LSF is not enabled implies the inode + * i_block represent total blocks in 512 bytes + * 32 == size of vfs inode i_blocks * 8 + */ + upper_limit = (1LL << 32) - 1; + + /* total blocks in file system block size */ + upper_limit >>= (blkbits - 9); + upper_limit <<= blkbits; + } + + /* 32-bit extent-start container, ee_block */ + res = 1LL << 32; + res <<= blkbits; + res -= 1; + + /* Sanity check against vm- & vfs- imposed limits */ + if (res > upper_limit) + res = upper_limit; + + return res; +} /* - * Maximal file size. There is a direct, and {,double-,triple-}indirect - * block limit, and also a limit of (2^32 - 1) 512-byte sectors in i_blocks. - * We need to be 1 filesystem block less than the 2^32 sector limit. + * Maximal bitmap file size. There is a direct, and {,double-,triple-}indirect + * block limit, and also a limit of (2^48 - 1) 512-byte sectors in i_blocks. + * We need to be 1 filesystem block less than the 2^48 sector limit. */ -static loff_t ext4_max_size(int bits) +static loff_t ext4_max_bitmap_size(int bits) { loff_t res = EXT4_NDIR_BLOCKS; - /* This constant is calculated to be the largest file size for a - * dense, 4k-blocksize file such that the total number of + int meta_blocks; + loff_t upper_limit; + /* This is calculated to be the largest file size for a + * dense, bitmapped file such that the total number of * sectors in the file, including data and all indirect blocks, - * does not exceed 2^32. */ - const loff_t upper_limit = 0x1ff7fffd000LL; + * does not exceed 2^48 -1 + * __u32 i_blocks_lo and _u16 i_blocks_high representing the + * total number of 512 bytes blocks of the file + */ + + if (sizeof(blkcnt_t) < sizeof(u64)) { + /* + * CONFIG_LSF is not enabled implies the inode + * i_block represent total blocks in 512 bytes + * 32 == size of vfs inode i_blocks * 8 + */ + upper_limit = (1LL << 32) - 1; + + /* total blocks in file system block size */ + upper_limit >>= (bits - 9); + + } else { + /* + * We use 48 bit ext4_inode i_blocks + * With EXT4_HUGE_FILE_FL set the i_blocks + * represent total number of blocks in + * file system block size + */ + upper_limit = (1LL << 48) - 1; + + } + + /* indirect blocks */ + meta_blocks = 1; + /* double indirect blocks */ + meta_blocks += 1 + (1LL << (bits-2)); + /* tripple indirect blocks */ + meta_blocks += 1 + (1LL << (bits-2)) + (1LL << (2*(bits-2))); + + upper_limit -= meta_blocks; + upper_limit <<= bits; res += 1LL << (bits-2); res += 1LL << (2*(bits-2)); @@ -1563,6 +1733,10 @@ static loff_t ext4_max_size(int bits) res <<= bits; if (res > upper_limit) res = upper_limit; + + if (res > MAX_LFS_FILESIZE) + res = MAX_LFS_FILESIZE; + return res; } @@ -1570,7 +1744,7 @@ static ext4_fsblk_t descriptor_loc(struct super_block *sb, ext4_fsblk_t logical_sb_block, int nr) { struct ext4_sb_info *sbi = EXT4_SB(sb); - unsigned long bg, first_meta_bg; + ext4_group_t bg, first_meta_bg; int has_super = 0; first_meta_bg = le32_to_cpu(sbi->s_es->s_first_meta_bg); @@ -1584,8 +1758,39 @@ static ext4_fsblk_t descriptor_loc(struct super_block *sb, return (has_super + ext4_group_first_block_no(sb, bg)); } +/** + * ext4_get_stripe_size: Get the stripe size. + * @sbi: In memory super block info + * + * If we have specified it via mount option, then + * use the mount option value. If the value specified at mount time is + * greater than the blocks per group use the super block value. + * If the super block value is greater than blocks per group return 0. + * Allocator needs it be less than blocks per group. + * + */ +static unsigned long ext4_get_stripe_size(struct ext4_sb_info *sbi) +{ + unsigned long stride = le16_to_cpu(sbi->s_es->s_raid_stride); + unsigned long stripe_width = + le32_to_cpu(sbi->s_es->s_raid_stripe_width); + + if (sbi->s_stripe && sbi->s_stripe <= sbi->s_blocks_per_group) + return sbi->s_stripe; + + if (stripe_width <= sbi->s_blocks_per_group) + return stripe_width; + + if (stride <= sbi->s_blocks_per_group) + return stride; + + return 0; +} static int ext4_fill_super (struct super_block *sb, void *data, int silent) + __releases(kernel_sem) + __acquires(kernel_sem) + { struct buffer_head * bh; struct ext4_super_block *es = NULL; @@ -1598,8 +1803,8 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) unsigned long journal_devnum = 0; unsigned long def_mount_opts; struct inode *root; + int ret = -EINVAL; int blocksize; - int hblock; int db_count; int i; int needs_recovery; @@ -1624,6 +1829,11 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) goto out_fail; } + if (!sb_set_blocksize(sb, blocksize)) { + printk(KERN_ERR "EXT4-fs: bad blocksize %d.\n", blocksize); + goto out_fail; + } + /* * The ext4 superblock will not be buffer aligned for other than 1kB * block sizes. We need to calculate the offset from buffer start. @@ -1674,10 +1884,10 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_PANIC) set_opt(sbi->s_mount_opt, ERRORS_PANIC); - else if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_RO) - set_opt(sbi->s_mount_opt, ERRORS_RO); - else + else if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_CONTINUE) set_opt(sbi->s_mount_opt, ERRORS_CONT); + else + set_opt(sbi->s_mount_opt, ERRORS_RO); sbi->s_resuid = le16_to_cpu(es->s_def_resuid); sbi->s_resgid = le16_to_cpu(es->s_def_resgid); @@ -1689,6 +1899,11 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) * User -o noextents to turn it off */ set_opt(sbi->s_mount_opt, EXTENTS); + /* + * turn on mballoc feature by default in ext4 filesystem + * User -o nomballoc to turn it off + */ + set_opt(sbi->s_mount_opt, MBALLOC); if (!parse_options ((char *) data, sb, &journal_inum, &journal_devnum, NULL, 0)) @@ -1704,6 +1919,17 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) printk(KERN_WARNING "EXT4-fs warning: feature flags set on rev 0 fs, " "running e2fsck is recommended\n"); + + /* + * Since ext4 is still considered development code, we require + * that the TEST_FILESYS flag in s->flags be set. + */ + if (!(le32_to_cpu(es->s_flags) & EXT2_FLAGS_TEST_FILESYS)) { + printk(KERN_WARNING "EXT4-fs: %s: not marked " + "OK to use with test code.\n", sb->s_id); + goto failed_mount; + } + /* * Check feature flags regardless of the revision level, since we * previously didn't change the revision level when setting the flags, @@ -1723,6 +1949,19 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) sb->s_id, le32_to_cpu(features)); goto failed_mount; } + if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_HUGE_FILE)) { + /* + * Large file size enabled file system can only be + * mount if kernel is build with CONFIG_LSF + */ + if (sizeof(root->i_blocks) < sizeof(u64) && + !(sb->s_flags & MS_RDONLY)) { + printk(KERN_ERR "EXT4-fs: %s: Filesystem with huge " + "files cannot be mounted read-write " + "without CONFIG_LSF.\n", sb->s_id); + goto failed_mount; + } + } blocksize = BLOCK_SIZE << le32_to_cpu(es->s_log_block_size); if (blocksize < EXT4_MIN_BLOCK_SIZE || @@ -1733,20 +1972,16 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) goto failed_mount; } - hblock = bdev_hardsect_size(sb->s_bdev); if (sb->s_blocksize != blocksize) { - /* - * Make sure the blocksize for the filesystem is larger - * than the hardware sectorsize for the machine. - */ - if (blocksize < hblock) { - printk(KERN_ERR "EXT4-fs: blocksize %d too small for " - "device blocksize %d.\n", blocksize, hblock); + + /* Validate the filesystem blocksize */ + if (!sb_set_blocksize(sb, blocksize)) { + printk(KERN_ERR "EXT4-fs: bad block size %d.\n", + blocksize); goto failed_mount; } brelse (bh); - sb_set_blocksize(sb, blocksize); logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE; offset = do_div(logical_sb_block, blocksize); bh = sb_bread(sb, logical_sb_block); @@ -1764,6 +1999,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) } } + sbi->s_bitmap_maxbytes = ext4_max_bitmap_size(sb->s_blocksize_bits); sb->s_maxbytes = ext4_max_size(sb->s_blocksize_bits); if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV) { @@ -1838,6 +2074,17 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) if (EXT4_BLOCKS_PER_GROUP(sb) == 0) goto cantfind_ext4; + + /* ensure blocks_count calculation below doesn't sign-extend */ + if (ext4_blocks_count(es) + EXT4_BLOCKS_PER_GROUP(sb) < + le32_to_cpu(es->s_first_data_block) + 1) { + printk(KERN_WARNING "EXT4-fs: bad geometry: block count %llu, " + "first data block %u, blocks per group %lu\n", + ext4_blocks_count(es), + le32_to_cpu(es->s_first_data_block), + EXT4_BLOCKS_PER_GROUP(sb)); + goto failed_mount; + } blocks_count = (ext4_blocks_count(es) - le32_to_cpu(es->s_first_data_block) + EXT4_BLOCKS_PER_GROUP(sb) - 1); @@ -1900,6 +2147,8 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) sbi->s_rsv_window_head.rsv_goal_size = 0; ext4_rsv_window_add(sb, &sbi->s_rsv_window_head); + sbi->s_stripe = ext4_get_stripe_size(sbi); + /* * set up enough so that it can read an inode */ @@ -1944,6 +2193,21 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) goto failed_mount4; } + if (test_opt(sb, JOURNAL_ASYNC_COMMIT)) { + jbd2_journal_set_features(sbi->s_journal, + JBD2_FEATURE_COMPAT_CHECKSUM, 0, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT); + } else if (test_opt(sb, JOURNAL_CHECKSUM)) { + jbd2_journal_set_features(sbi->s_journal, + JBD2_FEATURE_COMPAT_CHECKSUM, 0, 0); + jbd2_journal_clear_features(sbi->s_journal, 0, 0, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT); + } else { + jbd2_journal_clear_features(sbi->s_journal, + JBD2_FEATURE_COMPAT_CHECKSUM, 0, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT); + } + /* We have now updated the journal if required, so we can * validate the data journaling mode. */ switch (test_opt(sb, DATA_FLAGS)) { @@ -1983,19 +2247,24 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) * so we can safely mount the rest of the filesystem now. */ - root = iget(sb, EXT4_ROOT_INO); - sb->s_root = d_alloc_root(root); - if (!sb->s_root) { + root = ext4_iget(sb, EXT4_ROOT_INO); + if (IS_ERR(root)) { printk(KERN_ERR "EXT4-fs: get root inode failed\n"); - iput(root); + ret = PTR_ERR(root); goto failed_mount4; } if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) { - dput(sb->s_root); - sb->s_root = NULL; + iput(root); printk(KERN_ERR "EXT4-fs: corrupt root inode, run e2fsck\n"); goto failed_mount4; } + sb->s_root = d_alloc_root(root); + if (!sb->s_root) { + printk(KERN_ERR "EXT4-fs: get root dentry failed\n"); + iput(root); + ret = -ENOMEM; + goto failed_mount4; + } ext4_setup_super (sb, es, sb->s_flags & MS_RDONLY); @@ -2044,6 +2313,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) "writeback"); ext4_ext_init(sb); + ext4_mb_init(sb, needs_recovery); lock_kernel(); return 0; @@ -2075,7 +2345,7 @@ out_fail: sb->s_fs_info = NULL; kfree(sbi); lock_kernel(); - return -EINVAL; + return ret; } /* @@ -2111,8 +2381,8 @@ static journal_t *ext4_get_journal(struct super_block *sb, * things happen if we iget() an unused inode, as the subsequent * iput() will try to delete it. */ - journal_inode = iget(sb, journal_inum); - if (!journal_inode) { + journal_inode = ext4_iget(sb, journal_inum); + if (IS_ERR(journal_inode)) { printk(KERN_ERR "EXT4-fs: no journal found.\n"); return NULL; } @@ -2125,7 +2395,7 @@ static journal_t *ext4_get_journal(struct super_block *sb, jbd_debug(2, "Journal inode found at %p: %Ld bytes\n", journal_inode, journal_inode->i_size); - if (is_bad_inode(journal_inode) || !S_ISREG(journal_inode->i_mode)) { + if (!S_ISREG(journal_inode->i_mode)) { printk(KERN_ERR "EXT4-fs: invalid journal inode.\n"); iput(journal_inode); return NULL; @@ -2673,7 +2943,7 @@ static int ext4_statfs (struct dentry * dentry, struct kstatfs * buf) if (test_opt(sb, MINIX_DF)) { sbi->s_overhead_last = 0; } else if (sbi->s_blocks_last != ext4_blocks_count(es)) { - unsigned long ngroups = sbi->s_groups_count, i; + ext4_group_t ngroups = sbi->s_groups_count, i; ext4_fsblk_t overhead = 0; smp_rmb(); @@ -2888,16 +3158,16 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id, if (err) return err; /* Quotafile not on the same filesystem? */ - if (nd.mnt->mnt_sb != sb) { - path_release(&nd); + if (nd.path.mnt->mnt_sb != sb) { + path_put(&nd.path); return -EXDEV; } /* Quotafile not of fs root? */ - if (nd.dentry->d_parent->d_inode != sb->s_root->d_inode) + if (nd.path.dentry->d_parent->d_inode != sb->s_root->d_inode) printk(KERN_WARNING "EXT4-fs: Quota file not on filesystem root. " "Journalled quota will not work.\n"); - path_release(&nd); + path_put(&nd.path); return vfs_quota_on(sb, type, format_id, path); } @@ -2909,7 +3179,7 @@ static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data, size_t len, loff_t off) { struct inode *inode = sb_dqopt(sb)->files[type]; - sector_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb); + ext4_lblk_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb); int err = 0; int offset = off & (sb->s_blocksize - 1); int tocopy; @@ -2947,7 +3217,7 @@ static ssize_t ext4_quota_write(struct super_block *sb, int type, const char *data, size_t len, loff_t off) { struct inode *inode = sb_dqopt(sb)->files[type]; - sector_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb); + ext4_lblk_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb); int err = 0; int offset = off & (sb->s_blocksize - 1); int tocopy; @@ -3002,7 +3272,6 @@ out: i_size_write(inode, off+len-towrite); EXT4_I(inode)->i_disksize = inode->i_size; } - inode->i_version++; inode->i_mtime = inode->i_ctime = CURRENT_TIME; ext4_mark_inode_dirty(handle, inode); mutex_unlock(&inode->i_mutex); @@ -3027,9 +3296,15 @@ static struct file_system_type ext4dev_fs_type = { static int __init init_ext4_fs(void) { - int err = init_ext4_xattr(); + int err; + + err = init_ext4_mballoc(); if (err) return err; + + err = init_ext4_xattr(); + if (err) + goto out2; err = init_inodecache(); if (err) goto out1; @@ -3041,6 +3316,8 @@ out: destroy_inodecache(); out1: exit_ext4_xattr(); +out2: + exit_ext4_mballoc(); return err; } @@ -3049,6 +3326,7 @@ static void __exit exit_ext4_fs(void) unregister_filesystem(&ext4dev_fs_type); destroy_inodecache(); exit_ext4_xattr(); + exit_ext4_mballoc(); } MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others"); diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 86387302c2a..e9054c1c7d9 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -480,7 +480,7 @@ ext4_xattr_release_block(handle_t *handle, struct inode *inode, ea_bdebug(bh, "refcount now=0; freeing"); if (ce) mb_cache_entry_free(ce); - ext4_free_blocks(handle, inode, bh->b_blocknr, 1); + ext4_free_blocks(handle, inode, bh->b_blocknr, 1, 1); get_bh(bh); ext4_forget(handle, 1, inode, bh, bh->b_blocknr); } else { @@ -821,7 +821,7 @@ inserted: new_bh = sb_getblk(sb, block); if (!new_bh) { getblk_failed: - ext4_free_blocks(handle, inode, block, 1); + ext4_free_blocks(handle, inode, block, 1, 1); error = -EIO; goto cleanup; } @@ -1386,7 +1386,7 @@ ext4_xattr_cache_insert(struct buffer_head *bh) struct mb_cache_entry *ce; int error; - ce = mb_cache_entry_alloc(ext4_xattr_cache); + ce = mb_cache_entry_alloc(ext4_xattr_cache, GFP_NOFS); if (!ce) { ea_bdebug(bh, "out of memory"); return; diff --git a/fs/fat/file.c b/fs/fat/file.c index 69a83b59dce..c614175876e 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -155,6 +155,42 @@ out: return err; } +static int check_mode(const struct msdos_sb_info *sbi, mode_t mode) +{ + mode_t req = mode & ~S_IFMT; + + /* + * Of the r and x bits, all (subject to umask) must be present. Of the + * w bits, either all (subject to umask) or none must be present. + */ + + if (S_ISREG(mode)) { + req &= ~sbi->options.fs_fmask; + + if ((req & (S_IRUGO | S_IXUGO)) != + ((S_IRUGO | S_IXUGO) & ~sbi->options.fs_fmask)) + return -EPERM; + + if ((req & S_IWUGO) != 0 && + (req & S_IWUGO) != (S_IWUGO & ~sbi->options.fs_fmask)) + return -EPERM; + } else if (S_ISDIR(mode)) { + req &= ~sbi->options.fs_dmask; + + if ((req & (S_IRUGO | S_IXUGO)) != + ((S_IRUGO | S_IXUGO) & ~sbi->options.fs_dmask)) + return -EPERM; + + if ((req & S_IWUGO) != 0 && + (req & S_IWUGO) != (S_IWUGO & ~sbi->options.fs_dmask)) + return -EPERM; + } else { + return -EPERM; + } + + return 0; +} + int fat_notify_change(struct dentry *dentry, struct iattr *attr) { struct msdos_sb_info *sbi = MSDOS_SB(dentry->d_sb); @@ -186,9 +222,7 @@ int fat_notify_change(struct dentry *dentry, struct iattr *attr) if (((attr->ia_valid & ATTR_UID) && (attr->ia_uid != sbi->options.fs_uid)) || ((attr->ia_valid & ATTR_GID) && - (attr->ia_gid != sbi->options.fs_gid)) || - ((attr->ia_valid & ATTR_MODE) && - (attr->ia_mode & ~MSDOS_VALID_MODE))) + (attr->ia_gid != sbi->options.fs_gid))) error = -EPERM; if (error) { @@ -196,6 +230,13 @@ int fat_notify_change(struct dentry *dentry, struct iattr *attr) error = 0; goto out; } + + if (attr->ia_valid & ATTR_MODE) { + error = check_mode(sbi, attr->ia_mode); + if (error != 0 && !sbi->options.quiet) + goto out; + } + error = inode_setattr(inode, attr); if (error) goto out; diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 920a576e1c2..53f3cf62b7c 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -634,8 +634,6 @@ static const struct super_operations fat_sops = { .clear_inode = fat_clear_inode, .remount_fs = fat_remount, - .read_inode = make_bad_inode, - .show_options = fat_show_options, }; @@ -663,8 +661,8 @@ static struct dentry *fat_fh_to_dentry(struct super_block *sb, if (fh_len < 5 || fh_type != 3) return NULL; - inode = iget(sb, fh[0]); - if (!inode || is_bad_inode(inode) || inode->i_generation != fh[1]) { + inode = ilookup(sb, fh[0]); + if (!inode || inode->i_generation != fh[1]) { if (inode) iput(inode); inode = NULL; @@ -760,7 +758,7 @@ static struct dentry *fat_get_parent(struct dentry *child) inode = fat_build_inode(child->d_sb, de, i_pos); brelse(bh); if (IS_ERR(inode)) { - parent = ERR_PTR(PTR_ERR(inode)); + parent = ERR_CAST(inode); goto out; } parent = d_alloc_anon(inode); @@ -839,6 +837,8 @@ static int fat_show_options(struct seq_file *m, struct vfsmount *mnt) if (!opts->numtail) seq_puts(m, ",nonumtail"); } + if (sbi->options.flush) + seq_puts(m, ",flush"); return 0; } @@ -1295,10 +1295,8 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, fsinfo = (struct fat_boot_fsinfo *)fsinfo_bh->b_data; if (!IS_FSINFO(fsinfo)) { - printk(KERN_WARNING - "FAT: Did not find valid FSINFO signature.\n" - " Found signature1 0x%08x signature2 0x%08x" - " (sector = %lu)\n", + printk(KERN_WARNING "FAT: Invalid FSINFO signature: " + "0x%08x, 0x%08x (sector = %lu)\n", le32_to_cpu(fsinfo->signature1), le32_to_cpu(fsinfo->signature2), sbi->fsinfo_sector); diff --git a/fs/fat/misc.c b/fs/fat/misc.c index 308f2b6b502..61f23511eac 100644 --- a/fs/fat/misc.c +++ b/fs/fat/misc.c @@ -55,9 +55,8 @@ void fat_clusters_flush(struct super_block *sb) fsinfo = (struct fat_boot_fsinfo *)bh->b_data; /* Sanity check */ if (!IS_FSINFO(fsinfo)) { - printk(KERN_ERR "FAT: Did not find valid FSINFO signature.\n" - " Found signature1 0x%08x signature2 0x%08x" - " (sector = %lu)\n", + printk(KERN_ERR "FAT: Invalid FSINFO signature: " + "0x%08x, 0x%08x (sector = %lu)\n", le32_to_cpu(fsinfo->signature1), le32_to_cpu(fsinfo->signature2), sbi->fsinfo_sector); diff --git a/fs/fcntl.c b/fs/fcntl.c index 8685263ccc4..e632da761fc 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -24,7 +24,7 @@ #include <asm/siginfo.h> #include <asm/uaccess.h> -void fastcall set_close_on_exec(unsigned int fd, int flag) +void set_close_on_exec(unsigned int fd, int flag) { struct files_struct *files = current->files; struct fdtable *fdt; @@ -309,7 +309,7 @@ pid_t f_getown(struct file *filp) { pid_t pid; read_lock(&filp->f_owner.lock); - pid = pid_nr_ns(filp->f_owner.pid, current->nsproxy->pid_ns); + pid = pid_vnr(filp->f_owner.pid); if (filp->f_owner.pid_type == PIDTYPE_PGID) pid = -pid; read_unlock(&filp->f_owner.lock); diff --git a/fs/file.c b/fs/file.c index c5575de0111..5110acb1c9e 100644 --- a/fs/file.c +++ b/fs/file.c @@ -24,6 +24,8 @@ struct fdtable_defer { struct fdtable *next; }; +int sysctl_nr_open __read_mostly = 1024*1024; + /* * We use this list to defer free fdtables that have vmalloced * sets/arrays. By keeping a per-cpu list, we avoid having to embed @@ -147,8 +149,8 @@ static struct fdtable * alloc_fdtable(unsigned int nr) nr /= (1024 / sizeof(struct file *)); nr = roundup_pow_of_two(nr + 1); nr *= (1024 / sizeof(struct file *)); - if (nr > NR_OPEN) - nr = NR_OPEN; + if (nr > sysctl_nr_open) + nr = sysctl_nr_open; fdt = kmalloc(sizeof(struct fdtable), GFP_KERNEL); if (!fdt) @@ -233,7 +235,7 @@ int expand_files(struct files_struct *files, int nr) if (nr < fdt->max_fds) return 0; /* Can we expand? */ - if (nr >= NR_OPEN) + if (nr >= sysctl_nr_open) return -EMFILE; /* All good, so we try */ diff --git a/fs/file_table.c b/fs/file_table.c index 664e3f2309b..986ff4ed0a7 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -83,6 +83,12 @@ int proc_nr_files(ctl_table *table, int write, struct file *filp, /* Find an unused file structure and return a pointer to it. * Returns NULL, if there are no more free file structures or * we run out of memory. + * + * Be very careful using this. You are responsible for + * getting write access to any mount that you might assign + * to this filp, if it is opened for write. If this is not + * done, you will imbalance int the mount's writer count + * and a warning at __fput() time. */ struct file *get_empty_filp(void) { @@ -197,7 +203,7 @@ int init_file(struct file *file, struct vfsmount *mnt, struct dentry *dentry, } EXPORT_SYMBOL(init_file); -void fastcall fput(struct file *file) +void fput(struct file *file) { if (atomic_dec_and_test(&file->f_count)) __fput(file); @@ -208,7 +214,7 @@ EXPORT_SYMBOL(fput); /* __fput is called from task context when aio completion releases the last * last use of a struct file *. Do not use otherwise. */ -void fastcall __fput(struct file *file) +void __fput(struct file *file) { struct dentry *dentry = file->f_path.dentry; struct vfsmount *mnt = file->f_path.mnt; @@ -241,7 +247,7 @@ void fastcall __fput(struct file *file) mntput(mnt); } -struct file fastcall *fget(unsigned int fd) +struct file *fget(unsigned int fd) { struct file *file; struct files_struct *files = current->files; @@ -269,7 +275,7 @@ EXPORT_SYMBOL(fget); * and a flag is returned to be passed to the corresponding fput_light(). * There must not be a cloning between an fget_light/fput_light pair. */ -struct file fastcall *fget_light(unsigned int fd, int *fput_needed) +struct file *fget_light(unsigned int fd, int *fput_needed) { struct file *file; struct files_struct *files = current->files; diff --git a/fs/freevxfs/vxfs_dir.h b/fs/freevxfs/vxfs_dir.h index 3c96d6e6397..aaf1fb09863 100644 --- a/fs/freevxfs/vxfs_dir.h +++ b/fs/freevxfs/vxfs_dir.h @@ -41,7 +41,7 @@ * VxFS directory block header. * * This entry is the head of every filesystem block in a directory. - * It is used for free space managment and additionally includes + * It is used for free space management and additionally includes * a hash for speeding up directory search (lookup). * * The hash may be empty and in fact we do not use it all in the diff --git a/fs/freevxfs/vxfs_extern.h b/fs/freevxfs/vxfs_extern.h index 91ccee8723f..2b46064f66b 100644 --- a/fs/freevxfs/vxfs_extern.h +++ b/fs/freevxfs/vxfs_extern.h @@ -58,7 +58,7 @@ extern struct inode * vxfs_get_fake_inode(struct super_block *, extern void vxfs_put_fake_inode(struct inode *); extern struct vxfs_inode_info * vxfs_blkiget(struct super_block *, u_long, ino_t); extern struct vxfs_inode_info * vxfs_stiget(struct super_block *, ino_t); -extern void vxfs_read_inode(struct inode *); +extern struct inode * vxfs_iget(struct super_block *, ino_t); extern void vxfs_clear_inode(struct inode *); /* vxfs_lookup.c */ diff --git a/fs/freevxfs/vxfs_immed.c b/fs/freevxfs/vxfs_immed.c index 24b5a775ff9..8a5959a61ba 100644 --- a/fs/freevxfs/vxfs_immed.c +++ b/fs/freevxfs/vxfs_immed.c @@ -54,7 +54,7 @@ const struct inode_operations vxfs_immed_symlink_iops = { }; /* - * Adress space operations for immed files and directories. + * Address space operations for immed files and directories. */ const struct address_space_operations vxfs_immed_aops = { .readpage = vxfs_immed_readpage, diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c index d1f7c5b5b3c..ad88d2364bc 100644 --- a/fs/freevxfs/vxfs_inode.c +++ b/fs/freevxfs/vxfs_inode.c @@ -129,7 +129,7 @@ fail: * Description: * Search the for inode number @ino in the filesystem * described by @sbp. Use the specified inode table (@ilistp). - * Returns the matching VxFS inode on success, else a NULL pointer. + * Returns the matching VxFS inode on success, else an error code. */ static struct vxfs_inode_info * __vxfs_iget(ino_t ino, struct inode *ilistp) @@ -157,12 +157,12 @@ __vxfs_iget(ino_t ino, struct inode *ilistp) } printk(KERN_WARNING "vxfs: error on page %p\n", pp); - return NULL; + return ERR_CAST(pp); fail: printk(KERN_WARNING "vxfs: unable to read inode %ld\n", (unsigned long)ino); vxfs_put_page(pp); - return NULL; + return ERR_PTR(-ENOMEM); } /** @@ -178,7 +178,10 @@ fail: struct vxfs_inode_info * vxfs_stiget(struct super_block *sbp, ino_t ino) { - return __vxfs_iget(ino, VXFS_SBI(sbp)->vsi_stilist); + struct vxfs_inode_info *vip; + + vip = __vxfs_iget(ino, VXFS_SBI(sbp)->vsi_stilist); + return IS_ERR(vip) ? NULL : vip; } /** @@ -282,23 +285,32 @@ vxfs_put_fake_inode(struct inode *ip) } /** - * vxfs_read_inode - fill in inode information - * @ip: inode pointer to fill + * vxfs_iget - get an inode + * @sbp: the superblock to get the inode for + * @ino: the number of the inode to get * * Description: - * vxfs_read_inode reads the disk inode for @ip and fills - * in all relevant fields in @ip. + * vxfs_read_inode creates an inode, reads the disk inode for @ino and fills + * in all relevant fields in the new inode. */ -void -vxfs_read_inode(struct inode *ip) +struct inode * +vxfs_iget(struct super_block *sbp, ino_t ino) { - struct super_block *sbp = ip->i_sb; struct vxfs_inode_info *vip; const struct address_space_operations *aops; - ino_t ino = ip->i_ino; - - if (!(vip = __vxfs_iget(ino, VXFS_SBI(sbp)->vsi_ilist))) - return; + struct inode *ip; + + ip = iget_locked(sbp, ino); + if (!ip) + return ERR_PTR(-ENOMEM); + if (!(ip->i_state & I_NEW)) + return ip; + + vip = __vxfs_iget(ino, VXFS_SBI(sbp)->vsi_ilist); + if (IS_ERR(vip)) { + iget_failed(ip); + return ERR_CAST(vip); + } vxfs_iinit(ip, vip); @@ -323,7 +335,8 @@ vxfs_read_inode(struct inode *ip) } else init_special_inode(ip, ip->i_mode, old_decode_dev(vip->vii_rdev)); - return; + unlock_new_inode(ip); + return ip; } /** diff --git a/fs/freevxfs/vxfs_lookup.c b/fs/freevxfs/vxfs_lookup.c index bf86e5444ea..aee049cb9f8 100644 --- a/fs/freevxfs/vxfs_lookup.c +++ b/fs/freevxfs/vxfs_lookup.c @@ -213,10 +213,10 @@ vxfs_lookup(struct inode *dip, struct dentry *dp, struct nameidata *nd) lock_kernel(); ino = vxfs_inode_by_name(dip, dp); if (ino) { - ip = iget(dip->i_sb, ino); - if (!ip) { + ip = vxfs_iget(dip->i_sb, ino); + if (IS_ERR(ip)) { unlock_kernel(); - return ERR_PTR(-EACCES); + return ERR_CAST(ip); } } unlock_kernel(); diff --git a/fs/freevxfs/vxfs_super.c b/fs/freevxfs/vxfs_super.c index 4f95572d272..1dacda83157 100644 --- a/fs/freevxfs/vxfs_super.c +++ b/fs/freevxfs/vxfs_super.c @@ -60,7 +60,6 @@ static int vxfs_statfs(struct dentry *, struct kstatfs *); static int vxfs_remount(struct super_block *, int *, char *); static const struct super_operations vxfs_super_ops = { - .read_inode = vxfs_read_inode, .clear_inode = vxfs_clear_inode, .put_super = vxfs_put_super, .statfs = vxfs_statfs, @@ -153,6 +152,7 @@ static int vxfs_fill_super(struct super_block *sbp, void *dp, int silent) struct buffer_head *bp = NULL; u_long bsize; struct inode *root; + int ret = -EINVAL; sbp->s_flags |= MS_RDONLY; @@ -219,7 +219,11 @@ static int vxfs_fill_super(struct super_block *sbp, void *dp, int silent) } sbp->s_op = &vxfs_super_ops; - root = iget(sbp, VXFS_ROOT_INO); + root = vxfs_iget(sbp, VXFS_ROOT_INO); + if (IS_ERR(root)) { + ret = PTR_ERR(root); + goto out; + } sbp->s_root = d_alloc_root(root); if (!sbp->s_root) { iput(root); @@ -236,7 +240,7 @@ out_free_ilist: out: brelse(bp); kfree(infp); - return -EINVAL; + return ret; } /* diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 300324bd563..06557679ca4 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -284,7 +284,17 @@ __sync_single_inode(struct inode *inode, struct writeback_control *wbc) * soon as the queue becomes uncongested. */ inode->i_state |= I_DIRTY_PAGES; - requeue_io(inode); + if (wbc->nr_to_write <= 0) { + /* + * slice used up: queue for next turn + */ + requeue_io(inode); + } else { + /* + * somehow blocked: retry later + */ + redirty_tail(inode); + } } else { /* * Otherwise fully redirty the inode so that @@ -334,9 +344,6 @@ __writeback_single_inode(struct inode *inode, struct writeback_control *wbc) WARN_ON(inode->i_state & I_WILL_FREE); if ((wbc->sync_mode != WB_SYNC_ALL) && (inode->i_state & I_SYNC)) { - struct address_space *mapping = inode->i_mapping; - int ret; - /* * We're skipping this inode because it's locked, and we're not * doing writeback-for-data-integrity. Move it to s_more_io so @@ -345,15 +352,7 @@ __writeback_single_inode(struct inode *inode, struct writeback_control *wbc) * completed a full scan of s_io. */ requeue_io(inode); - - /* - * Even if we don't actually write the inode itself here, - * we can at least start some of the data writeout.. - */ - spin_unlock(&inode_lock); - ret = do_writepages(mapping, wbc); - spin_lock(&inode_lock); - return ret; + return 0; } /* @@ -479,8 +478,12 @@ sync_sb_inodes(struct super_block *sb, struct writeback_control *wbc) iput(inode); cond_resched(); spin_lock(&inode_lock); - if (wbc->nr_to_write <= 0) + if (wbc->nr_to_write <= 0) { + wbc->more_io = 1; break; + } + if (!list_empty(&sb->s_more_io)) + wbc->more_io = 1; } return; /* Leave any unwritten inodes on s_io */ } @@ -512,8 +515,7 @@ writeback_inodes(struct writeback_control *wbc) might_sleep(); spin_lock(&sb_lock); restart: - sb = sb_entry(super_blocks.prev); - for (; sb != sb_entry(&super_blocks); sb = sb_entry(sb->s_list.prev)) { + list_for_each_entry_reverse(sb, &super_blocks, s_list) { if (sb_has_dirty_inodes(sb)) { /* we're making our own get_super here */ sb->s_count++; @@ -578,10 +580,8 @@ static void set_sb_syncing(int val) { struct super_block *sb; spin_lock(&sb_lock); - sb = sb_entry(super_blocks.prev); - for (; sb != sb_entry(&super_blocks); sb = sb_entry(sb->s_list.prev)) { + list_for_each_entry_reverse(sb, &super_blocks, s_list) sb->s_syncing = val; - } spin_unlock(&sb_lock); } @@ -655,7 +655,7 @@ int write_inode_now(struct inode *inode, int sync) int ret; struct writeback_control wbc = { .nr_to_write = LONG_MAX, - .sync_mode = WB_SYNC_ALL, + .sync_mode = sync ? WB_SYNC_ALL : WB_SYNC_NONE, .range_start = 0, .range_end = LLONG_MAX, }; @@ -751,7 +751,7 @@ int generic_osync_inode(struct inode *inode, struct address_space *mapping, int EXPORT_SYMBOL(generic_osync_inode); /** - * writeback_acquire: attempt to get exclusive writeback access to a device + * writeback_acquire - attempt to get exclusive writeback access to a device * @bdi: the device's backing_dev_info structure * * It is a waste of resources to have more than one pdflush thread blocked on @@ -768,7 +768,7 @@ int writeback_acquire(struct backing_dev_info *bdi) } /** - * writeback_in_progress: determine whether there is writeback in progress + * writeback_in_progress - determine whether there is writeback in progress * @bdi: the device's backing_dev_info structure. * * Determine whether there is writeback in progress against a backing device. @@ -779,7 +779,7 @@ int writeback_in_progress(struct backing_dev_info *bdi) } /** - * writeback_release: relinquish exclusive writeback access against a device. + * writeback_release - relinquish exclusive writeback access against a device. * @bdi: the device's backing_dev_info structure */ void writeback_release(struct backing_dev_info *bdi) diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index db534bcde45..af639807524 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -201,6 +201,55 @@ void fuse_put_request(struct fuse_conn *fc, struct fuse_req *req) } } +static unsigned len_args(unsigned numargs, struct fuse_arg *args) +{ + unsigned nbytes = 0; + unsigned i; + + for (i = 0; i < numargs; i++) + nbytes += args[i].size; + + return nbytes; +} + +static u64 fuse_get_unique(struct fuse_conn *fc) +{ + fc->reqctr++; + /* zero is special */ + if (fc->reqctr == 0) + fc->reqctr = 1; + + return fc->reqctr; +} + +static void queue_request(struct fuse_conn *fc, struct fuse_req *req) +{ + req->in.h.unique = fuse_get_unique(fc); + req->in.h.len = sizeof(struct fuse_in_header) + + len_args(req->in.numargs, (struct fuse_arg *) req->in.args); + list_add_tail(&req->list, &fc->pending); + req->state = FUSE_REQ_PENDING; + if (!req->waiting) { + req->waiting = 1; + atomic_inc(&fc->num_waiting); + } + wake_up(&fc->waitq); + kill_fasync(&fc->fasync, SIGIO, POLL_IN); +} + +static void flush_bg_queue(struct fuse_conn *fc) +{ + while (fc->active_background < FUSE_MAX_BACKGROUND && + !list_empty(&fc->bg_queue)) { + struct fuse_req *req; + + req = list_entry(fc->bg_queue.next, struct fuse_req, list); + list_del(&req->list); + fc->active_background++; + queue_request(fc, req); + } +} + /* * This function is called when a request is finished. Either a reply * has arrived or it was aborted (and not yet sent) or some error @@ -229,6 +278,8 @@ static void request_end(struct fuse_conn *fc, struct fuse_req *req) clear_bdi_congested(&fc->bdi, WRITE); } fc->num_background--; + fc->active_background--; + flush_bg_queue(fc); } spin_unlock(&fc->lock); wake_up(&req->waitq); @@ -320,42 +371,6 @@ static void request_wait_answer(struct fuse_conn *fc, struct fuse_req *req) } } -static unsigned len_args(unsigned numargs, struct fuse_arg *args) -{ - unsigned nbytes = 0; - unsigned i; - - for (i = 0; i < numargs; i++) - nbytes += args[i].size; - - return nbytes; -} - -static u64 fuse_get_unique(struct fuse_conn *fc) - { - fc->reqctr++; - /* zero is special */ - if (fc->reqctr == 0) - fc->reqctr = 1; - - return fc->reqctr; -} - -static void queue_request(struct fuse_conn *fc, struct fuse_req *req) -{ - req->in.h.unique = fuse_get_unique(fc); - req->in.h.len = sizeof(struct fuse_in_header) + - len_args(req->in.numargs, (struct fuse_arg *) req->in.args); - list_add_tail(&req->list, &fc->pending); - req->state = FUSE_REQ_PENDING; - if (!req->waiting) { - req->waiting = 1; - atomic_inc(&fc->num_waiting); - } - wake_up(&fc->waitq); - kill_fasync(&fc->fasync, SIGIO, POLL_IN); -} - void request_send(struct fuse_conn *fc, struct fuse_req *req) { req->isreply = 1; @@ -375,20 +390,26 @@ void request_send(struct fuse_conn *fc, struct fuse_req *req) spin_unlock(&fc->lock); } +static void request_send_nowait_locked(struct fuse_conn *fc, + struct fuse_req *req) +{ + req->background = 1; + fc->num_background++; + if (fc->num_background == FUSE_MAX_BACKGROUND) + fc->blocked = 1; + if (fc->num_background == FUSE_CONGESTION_THRESHOLD) { + set_bdi_congested(&fc->bdi, READ); + set_bdi_congested(&fc->bdi, WRITE); + } + list_add_tail(&req->list, &fc->bg_queue); + flush_bg_queue(fc); +} + static void request_send_nowait(struct fuse_conn *fc, struct fuse_req *req) { spin_lock(&fc->lock); if (fc->connected) { - req->background = 1; - fc->num_background++; - if (fc->num_background == FUSE_MAX_BACKGROUND) - fc->blocked = 1; - if (fc->num_background == FUSE_CONGESTION_THRESHOLD) { - set_bdi_congested(&fc->bdi, READ); - set_bdi_congested(&fc->bdi, WRITE); - } - - queue_request(fc, req); + request_send_nowait_locked(fc, req); spin_unlock(&fc->lock); } else { req->out.h.error = -ENOTCONN; diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 80d2f5292cf..c4807b3fc8a 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -269,12 +269,12 @@ static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry, req = fuse_get_req(fc); if (IS_ERR(req)) - return ERR_PTR(PTR_ERR(req)); + return ERR_CAST(req); forget_req = fuse_get_req(fc); if (IS_ERR(forget_req)) { fuse_put_request(fc, req); - return ERR_PTR(PTR_ERR(forget_req)); + return ERR_CAST(forget_req); } attr_version = fuse_get_attr_version(fc); @@ -416,6 +416,7 @@ static int fuse_create_open(struct inode *dir, struct dentry *entry, int mode, fuse_put_request(fc, forget_req); d_instantiate(entry, inode); fuse_change_entry_timeout(entry, &outentry); + fuse_invalidate_attr(dir); file = lookup_instantiate_filp(nd, entry, generic_file_open); if (IS_ERR(file)) { ff->fh = outopen.fh; @@ -905,7 +906,7 @@ static int fuse_permission(struct inode *inode, int mask, struct nameidata *nd) } if (fc->flags & FUSE_DEFAULT_PERMISSIONS) { - int err = generic_permission(inode, mask, NULL); + err = generic_permission(inode, mask, NULL); /* If permission is denied, try to refresh file attributes. This is also needed, because the root @@ -1005,7 +1006,7 @@ static char *read_link(struct dentry *dentry) char *link; if (IS_ERR(req)) - return ERR_PTR(PTR_ERR(req)); + return ERR_CAST(req); link = (char *) __get_free_page(GFP_KERNEL); if (!link) { diff --git a/fs/fuse/file.c b/fs/fuse/file.c index bb05d227cf3..676b0bc8a86 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -77,8 +77,8 @@ static struct fuse_file *fuse_file_get(struct fuse_file *ff) static void fuse_release_end(struct fuse_conn *fc, struct fuse_req *req) { - dput(req->dentry); - mntput(req->vfsmount); + dput(req->misc.release.dentry); + mntput(req->misc.release.vfsmount); fuse_put_request(fc, req); } @@ -86,7 +86,8 @@ static void fuse_file_put(struct fuse_file *ff) { if (atomic_dec_and_test(&ff->count)) { struct fuse_req *req = ff->reserved_req; - struct fuse_conn *fc = get_fuse_conn(req->dentry->d_inode); + struct inode *inode = req->misc.release.dentry->d_inode; + struct fuse_conn *fc = get_fuse_conn(inode); req->end = fuse_release_end; request_send_background(fc, req); kfree(ff); @@ -137,7 +138,7 @@ int fuse_open_common(struct inode *inode, struct file *file, int isdir) void fuse_release_fill(struct fuse_file *ff, u64 nodeid, int flags, int opcode) { struct fuse_req *req = ff->reserved_req; - struct fuse_release_in *inarg = &req->misc.release_in; + struct fuse_release_in *inarg = &req->misc.release.in; inarg->fh = ff->fh; inarg->flags = flags; @@ -153,13 +154,14 @@ int fuse_release_common(struct inode *inode, struct file *file, int isdir) struct fuse_file *ff = file->private_data; if (ff) { struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_req *req = ff->reserved_req; fuse_release_fill(ff, get_node_id(inode), file->f_flags, isdir ? FUSE_RELEASEDIR : FUSE_RELEASE); /* Hold vfsmount and dentry until release is finished */ - ff->reserved_req->vfsmount = mntget(file->f_path.mnt); - ff->reserved_req->dentry = dget(file->f_path.dentry); + req->misc.release.vfsmount = mntget(file->f_path.mnt); + req->misc.release.dentry = dget(file->f_path.dentry); spin_lock(&fc->lock); list_del(&ff->write_entry); diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index 3ab8a3048e8..67aaf6ee38e 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -215,7 +215,11 @@ struct fuse_req { /** Data for asynchronous requests */ union { struct fuse_forget_in forget_in; - struct fuse_release_in release_in; + struct { + struct fuse_release_in in; + struct vfsmount *vfsmount; + struct dentry *dentry; + } release; struct fuse_init_in init_in; struct fuse_init_out init_out; struct fuse_read_in read_in; @@ -238,12 +242,6 @@ struct fuse_req { /** File used in the request (or NULL) */ struct fuse_file *ff; - /** vfsmount used in release */ - struct vfsmount *vfsmount; - - /** dentry used in release */ - struct dentry *dentry; - /** Request completion callback */ void (*end)(struct fuse_conn *, struct fuse_req *); @@ -298,6 +296,12 @@ struct fuse_conn { /** Number of requests currently in the background */ unsigned num_background; + /** Number of background requests currently queued for userspace */ + unsigned active_background; + + /** The list of background requests set aside for later queuing */ + struct list_head bg_queue; + /** Pending interrupts */ struct list_head interrupts; diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 84f9f7dfdf5..033f7bdd47e 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -29,6 +29,8 @@ DEFINE_MUTEX(fuse_mutex); #define FUSE_SUPER_MAGIC 0x65735546 +#define FUSE_DEFAULT_BLKSIZE 512 + struct fuse_mount_data { int fd; unsigned rootmode; @@ -76,11 +78,6 @@ static void fuse_destroy_inode(struct inode *inode) kmem_cache_free(fuse_inode_cachep, inode); } -static void fuse_read_inode(struct inode *inode) -{ - /* No op */ -} - void fuse_send_forget(struct fuse_conn *fc, struct fuse_req *req, unsigned long nodeid, u64 nlookup) { @@ -360,7 +357,7 @@ static int parse_fuse_opt(char *opt, struct fuse_mount_data *d, int is_bdev) char *p; memset(d, 0, sizeof(struct fuse_mount_data)); d->max_read = ~0; - d->blksize = 512; + d->blksize = FUSE_DEFAULT_BLKSIZE; while ((p = strsep(&opt, ",")) != NULL) { int token; @@ -445,6 +442,9 @@ static int fuse_show_options(struct seq_file *m, struct vfsmount *mnt) seq_puts(m, ",allow_other"); if (fc->max_read != ~0) seq_printf(m, ",max_read=%u", fc->max_read); + if (mnt->mnt_sb->s_bdev && + mnt->mnt_sb->s_blocksize != FUSE_DEFAULT_BLKSIZE) + seq_printf(m, ",blksize=%lu", mnt->mnt_sb->s_blocksize); return 0; } @@ -465,6 +465,7 @@ static struct fuse_conn *new_conn(void) INIT_LIST_HEAD(&fc->processing); INIT_LIST_HEAD(&fc->io); INIT_LIST_HEAD(&fc->interrupts); + INIT_LIST_HEAD(&fc->bg_queue); atomic_set(&fc->num_waiting, 0); fc->bdi.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE; fc->bdi.unplug_io_fn = default_unplug_io_fn; @@ -514,7 +515,6 @@ static struct inode *get_root_inode(struct super_block *sb, unsigned mode) static const struct super_operations fuse_super_operations = { .alloc_inode = fuse_alloc_inode, .destroy_inode = fuse_destroy_inode, - .read_inode = fuse_read_inode, .clear_inode = fuse_clear_inode, .drop_inode = generic_delete_inode, .remount_fs = fuse_remount_fs, @@ -744,9 +744,6 @@ static inline void unregister_fuseblk(void) } #endif -static decl_subsys(fuse, NULL, NULL); -static decl_subsys(connections, NULL, NULL); - static void fuse_inode_init_once(struct kmem_cache *cachep, void *foo) { struct inode * inode = foo; @@ -791,32 +788,37 @@ static void fuse_fs_cleanup(void) kmem_cache_destroy(fuse_inode_cachep); } +static struct kobject *fuse_kobj; +static struct kobject *connections_kobj; + static int fuse_sysfs_init(void) { int err; - kobj_set_kset_s(&fuse_subsys, fs_subsys); - err = subsystem_register(&fuse_subsys); - if (err) + fuse_kobj = kobject_create_and_add("fuse", fs_kobj); + if (!fuse_kobj) { + err = -ENOMEM; goto out_err; + } - kobj_set_kset_s(&connections_subsys, fuse_subsys); - err = subsystem_register(&connections_subsys); - if (err) + connections_kobj = kobject_create_and_add("connections", fuse_kobj); + if (!connections_kobj) { + err = -ENOMEM; goto out_fuse_unregister; + } return 0; out_fuse_unregister: - subsystem_unregister(&fuse_subsys); + kobject_put(fuse_kobj); out_err: return err; } static void fuse_sysfs_cleanup(void) { - subsystem_unregister(&connections_subsys); - subsystem_unregister(&fuse_subsys); + kobject_put(connections_kobj); + kobject_put(fuse_kobj); } static int __init fuse_init(void) diff --git a/fs/gfs2/Makefile b/fs/gfs2/Makefile index 04ad0caebed..8fff11058ce 100644 --- a/fs/gfs2/Makefile +++ b/fs/gfs2/Makefile @@ -2,7 +2,7 @@ obj-$(CONFIG_GFS2_FS) += gfs2.o gfs2-y := acl.o bmap.o daemon.o dir.o eaops.o eattr.o glock.o \ glops.o inode.o lm.o log.o lops.o locking.o main.o meta_io.o \ mount.o ops_address.o ops_dentry.o ops_export.o ops_file.o \ - ops_fstype.o ops_inode.o ops_super.o ops_vm.o quota.o \ + ops_fstype.o ops_inode.o ops_super.o quota.o \ recovery.o rgrp.o super.o sys.o trans.o util.o obj-$(CONFIG_GFS2_FS_LOCKING_NOLOCK) += locking/nolock/ diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 93fa427bb5f..e9456ebd3bb 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -59,7 +59,6 @@ struct strip_mine { static int gfs2_unstuffer_page(struct gfs2_inode *ip, struct buffer_head *dibh, u64 block, struct page *page) { - struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct inode *inode = &ip->i_inode; struct buffer_head *bh; int release = 0; @@ -95,7 +94,7 @@ static int gfs2_unstuffer_page(struct gfs2_inode *ip, struct buffer_head *dibh, set_buffer_uptodate(bh); if (!gfs2_is_jdata(ip)) mark_buffer_dirty(bh); - if (sdp->sd_args.ar_data == GFS2_DATA_ORDERED || gfs2_is_jdata(ip)) + if (!gfs2_is_writeback(ip)) gfs2_trans_add_bh(ip->i_gl, bh, 0); if (release) { @@ -453,8 +452,8 @@ static inline void bmap_unlock(struct inode *inode, int create) * Returns: errno */ -int gfs2_block_map(struct inode *inode, u64 lblock, int create, - struct buffer_head *bh_map) +int gfs2_block_map(struct inode *inode, sector_t lblock, + struct buffer_head *bh_map, int create) { struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); @@ -470,6 +469,7 @@ int gfs2_block_map(struct inode *inode, u64 lblock, int create, unsigned int maxlen = bh_map->b_size >> inode->i_blkbits; struct metapath mp; u64 size; + struct buffer_head *dibh = NULL; BUG_ON(maxlen == 0); @@ -500,6 +500,8 @@ int gfs2_block_map(struct inode *inode, u64 lblock, int create, error = gfs2_meta_inode_buffer(ip, &bh); if (error) goto out_fail; + dibh = bh; + get_bh(dibh); for (x = 0; x < end_of_metadata; x++) { lookup_block(ip, bh, x, &mp, create, &new, &dblock); @@ -518,13 +520,8 @@ int gfs2_block_map(struct inode *inode, u64 lblock, int create, if (boundary) set_buffer_boundary(bh_map); if (new) { - struct buffer_head *dibh; - error = gfs2_meta_inode_buffer(ip, &dibh); - if (!error) { - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(ip, dibh->b_data); - brelse(dibh); - } + gfs2_trans_add_bh(ip->i_gl, dibh, 1); + gfs2_dinode_out(ip, dibh->b_data); set_buffer_new(bh_map); goto out_brelse; } @@ -545,6 +542,8 @@ out_brelse: out_ok: error = 0; out_fail: + if (dibh) + brelse(dibh); bmap_unlock(inode, create); return error; } @@ -560,7 +559,7 @@ int gfs2_extent_map(struct inode *inode, u64 lblock, int *new, u64 *dblock, unsi BUG_ON(!new); bh.b_size = 1 << (inode->i_blkbits + 5); - ret = gfs2_block_map(inode, lblock, create, &bh); + ret = gfs2_block_map(inode, lblock, &bh, create); *extlen = bh.b_size >> inode->i_blkbits; *dblock = bh.b_blocknr; if (buffer_new(&bh)) @@ -684,7 +683,7 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, if (metadata) revokes = (height) ? sdp->sd_inptrs : sdp->sd_diptrs; - error = gfs2_rindex_hold(sdp, &ip->i_alloc.al_ri_gh); + error = gfs2_rindex_hold(sdp, &ip->i_alloc->al_ri_gh); if (error) return error; @@ -786,7 +785,7 @@ out_rg_gunlock: out_rlist: gfs2_rlist_free(&rlist); out: - gfs2_glock_dq_uninit(&ip->i_alloc.al_ri_gh); + gfs2_glock_dq_uninit(&ip->i_alloc->al_ri_gh); return error; } @@ -879,7 +878,6 @@ static int gfs2_block_truncate_page(struct address_space *mapping) { struct inode *inode = mapping->host; struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_sbd *sdp = GFS2_SB(inode); loff_t from = inode->i_size; unsigned long index = from >> PAGE_CACHE_SHIFT; unsigned offset = from & (PAGE_CACHE_SIZE-1); @@ -911,7 +909,7 @@ static int gfs2_block_truncate_page(struct address_space *mapping) err = 0; if (!buffer_mapped(bh)) { - gfs2_get_block(inode, iblock, bh, 0); + gfs2_block_map(inode, iblock, bh, 0); /* unmapped? It's a hole - nothing to do */ if (!buffer_mapped(bh)) goto unlock; @@ -931,10 +929,10 @@ static int gfs2_block_truncate_page(struct address_space *mapping) err = 0; } - if (sdp->sd_args.ar_data == GFS2_DATA_ORDERED || gfs2_is_jdata(ip)) + if (!gfs2_is_writeback(ip)) gfs2_trans_add_bh(ip->i_gl, bh, 0); - zero_user_page(page, offset, length, KM_USER0); + zero_user(page, offset, length); unlock: unlock_page(page); @@ -1224,8 +1222,13 @@ int gfs2_write_alloc_required(struct gfs2_inode *ip, u64 offset, do_div(lblock_stop, bsize); } else { unsigned int shift = sdp->sd_sb.sb_bsize_shift; + u64 end_of_file = (ip->i_di.di_size + sdp->sd_sb.sb_bsize - 1) >> shift; lblock = offset >> shift; lblock_stop = (offset + len + sdp->sd_sb.sb_bsize - 1) >> shift; + if (lblock_stop > end_of_file) { + *alloc_required = 1; + return 0; + } } for (; lblock < lblock_stop; lblock += extlen) { diff --git a/fs/gfs2/bmap.h b/fs/gfs2/bmap.h index ac2fd04370d..4e6cde2943b 100644 --- a/fs/gfs2/bmap.h +++ b/fs/gfs2/bmap.h @@ -15,7 +15,7 @@ struct gfs2_inode; struct page; int gfs2_unstuff_dinode(struct gfs2_inode *ip, struct page *page); -int gfs2_block_map(struct inode *inode, u64 lblock, int create, struct buffer_head *bh); +int gfs2_block_map(struct inode *inode, sector_t lblock, struct buffer_head *bh, int create); int gfs2_extent_map(struct inode *inode, u64 lblock, int *new, u64 *dblock, unsigned *extlen); int gfs2_truncatei(struct gfs2_inode *ip, u64 size); diff --git a/fs/gfs2/daemon.c b/fs/gfs2/daemon.c index 3731ab0771d..e51991947d2 100644 --- a/fs/gfs2/daemon.c +++ b/fs/gfs2/daemon.c @@ -83,56 +83,6 @@ int gfs2_recoverd(void *data) } /** - * gfs2_logd - Update log tail as Active Items get flushed to in-place blocks - * @sdp: Pointer to GFS2 superblock - * - * Also, periodically check to make sure that we're using the most recent - * journal index. - */ - -int gfs2_logd(void *data) -{ - struct gfs2_sbd *sdp = data; - struct gfs2_holder ji_gh; - unsigned long t; - int need_flush; - - while (!kthread_should_stop()) { - /* Advance the log tail */ - - t = sdp->sd_log_flush_time + - gfs2_tune_get(sdp, gt_log_flush_secs) * HZ; - - gfs2_ail1_empty(sdp, DIO_ALL); - gfs2_log_lock(sdp); - need_flush = sdp->sd_log_num_buf > gfs2_tune_get(sdp, gt_incore_log_blocks); - gfs2_log_unlock(sdp); - if (need_flush || time_after_eq(jiffies, t)) { - gfs2_log_flush(sdp, NULL); - sdp->sd_log_flush_time = jiffies; - } - - /* Check for latest journal index */ - - t = sdp->sd_jindex_refresh_time + - gfs2_tune_get(sdp, gt_jindex_refresh_secs) * HZ; - - if (time_after_eq(jiffies, t)) { - if (!gfs2_jindex_hold(sdp, &ji_gh)) - gfs2_glock_dq_uninit(&ji_gh); - sdp->sd_jindex_refresh_time = jiffies; - } - - t = gfs2_tune_get(sdp, gt_logd_secs) * HZ; - if (freezing(current)) - refrigerator(); - schedule_timeout_interruptible(t); - } - - return 0; -} - -/** * gfs2_quotad - Write cached quota changes into the quota file * @sdp: Pointer to GFS2 superblock * diff --git a/fs/gfs2/daemon.h b/fs/gfs2/daemon.h index 0de9b355795..4be084fb6a6 100644 --- a/fs/gfs2/daemon.h +++ b/fs/gfs2/daemon.h @@ -12,7 +12,6 @@ int gfs2_glockd(void *data); int gfs2_recoverd(void *data); -int gfs2_logd(void *data); int gfs2_quotad(void *data); #endif /* __DAEMON_DOT_H__ */ diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index 9949bb746a5..c34709512b1 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -1498,7 +1498,7 @@ struct inode *gfs2_dir_search(struct inode *dir, const struct qstr *name) dent = gfs2_dirent_search(dir, name, gfs2_dirent_find, &bh); if (dent) { if (IS_ERR(dent)) - return ERR_PTR(PTR_ERR(dent)); + return ERR_CAST(dent); inode = gfs2_inode_lookup(dir->i_sb, be16_to_cpu(dent->de_type), be64_to_cpu(dent->de_inum.no_addr), @@ -1876,7 +1876,7 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len, if (error) goto out; - error = gfs2_rindex_hold(sdp, &dip->i_alloc.al_ri_gh); + error = gfs2_rindex_hold(sdp, &dip->i_alloc->al_ri_gh); if (error) goto out_qs; @@ -1949,7 +1949,7 @@ out_rg_gunlock: gfs2_glock_dq_m(rlist.rl_rgrps, rlist.rl_ghs); out_rlist: gfs2_rlist_free(&rlist); - gfs2_glock_dq_uninit(&dip->i_alloc.al_ri_gh); + gfs2_glock_dq_uninit(&dip->i_alloc->al_ri_gh); out_qs: gfs2_quota_unhold(dip); out: diff --git a/fs/gfs2/eaops.c b/fs/gfs2/eaops.c index aa8dbf303f6..f114ba2b355 100644 --- a/fs/gfs2/eaops.c +++ b/fs/gfs2/eaops.c @@ -56,46 +56,6 @@ unsigned int gfs2_ea_name2type(const char *name, const char **truncated_name) return type; } -static int user_eo_get(struct gfs2_inode *ip, struct gfs2_ea_request *er) -{ - struct inode *inode = &ip->i_inode; - int error = permission(inode, MAY_READ, NULL); - if (error) - return error; - - return gfs2_ea_get_i(ip, er); -} - -static int user_eo_set(struct gfs2_inode *ip, struct gfs2_ea_request *er) -{ - struct inode *inode = &ip->i_inode; - - if (S_ISREG(inode->i_mode) || - (S_ISDIR(inode->i_mode) && !(inode->i_mode & S_ISVTX))) { - int error = permission(inode, MAY_WRITE, NULL); - if (error) - return error; - } else - return -EPERM; - - return gfs2_ea_set_i(ip, er); -} - -static int user_eo_remove(struct gfs2_inode *ip, struct gfs2_ea_request *er) -{ - struct inode *inode = &ip->i_inode; - - if (S_ISREG(inode->i_mode) || - (S_ISDIR(inode->i_mode) && !(inode->i_mode & S_ISVTX))) { - int error = permission(inode, MAY_WRITE, NULL); - if (error) - return error; - } else - return -EPERM; - - return gfs2_ea_remove_i(ip, er); -} - static int system_eo_get(struct gfs2_inode *ip, struct gfs2_ea_request *er) { if (!GFS2_ACL_IS_ACCESS(er->er_name, er->er_name_len) && @@ -108,8 +68,6 @@ static int system_eo_get(struct gfs2_inode *ip, struct gfs2_ea_request *er) GFS2_ACL_IS_DEFAULT(er->er_name, er->er_name_len))) return -EOPNOTSUPP; - - return gfs2_ea_get_i(ip, er); } @@ -170,40 +128,10 @@ static int system_eo_remove(struct gfs2_inode *ip, struct gfs2_ea_request *er) return gfs2_ea_remove_i(ip, er); } -static int security_eo_get(struct gfs2_inode *ip, struct gfs2_ea_request *er) -{ - struct inode *inode = &ip->i_inode; - int error = permission(inode, MAY_READ, NULL); - if (error) - return error; - - return gfs2_ea_get_i(ip, er); -} - -static int security_eo_set(struct gfs2_inode *ip, struct gfs2_ea_request *er) -{ - struct inode *inode = &ip->i_inode; - int error = permission(inode, MAY_WRITE, NULL); - if (error) - return error; - - return gfs2_ea_set_i(ip, er); -} - -static int security_eo_remove(struct gfs2_inode *ip, struct gfs2_ea_request *er) -{ - struct inode *inode = &ip->i_inode; - int error = permission(inode, MAY_WRITE, NULL); - if (error) - return error; - - return gfs2_ea_remove_i(ip, er); -} - static const struct gfs2_eattr_operations gfs2_user_eaops = { - .eo_get = user_eo_get, - .eo_set = user_eo_set, - .eo_remove = user_eo_remove, + .eo_get = gfs2_ea_get_i, + .eo_set = gfs2_ea_set_i, + .eo_remove = gfs2_ea_remove_i, .eo_name = "user", }; @@ -215,9 +143,9 @@ const struct gfs2_eattr_operations gfs2_system_eaops = { }; static const struct gfs2_eattr_operations gfs2_security_eaops = { - .eo_get = security_eo_get, - .eo_set = security_eo_set, - .eo_remove = security_eo_remove, + .eo_get = gfs2_ea_get_i, + .eo_set = gfs2_ea_set_i, + .eo_remove = gfs2_ea_remove_i, .eo_name = "security", }; diff --git a/fs/gfs2/eattr.c b/fs/gfs2/eattr.c index 2a7435b5c4d..bee99704ea1 100644 --- a/fs/gfs2/eattr.c +++ b/fs/gfs2/eattr.c @@ -1418,7 +1418,7 @@ out: static int ea_dealloc_block(struct gfs2_inode *ip) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al = &ip->i_alloc; + struct gfs2_alloc *al = ip->i_alloc; struct gfs2_rgrpd *rgd; struct buffer_head *dibh; int error; diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index a37efe4aae6..7175a4d0643 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -1,6 +1,6 @@ /* * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. - * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved. + * Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved. * * This copyrighted material is made available to anyone wishing to use, * modify, copy, or redistribute it subject to the terms and conditions @@ -217,7 +217,6 @@ int gfs2_glock_put(struct gfs2_glock *gl) if (atomic_dec_and_test(&gl->gl_ref)) { hlist_del(&gl->gl_list); write_unlock(gl_lock_addr(gl->gl_hash)); - BUG_ON(spin_is_locked(&gl->gl_spin)); gfs2_assert(sdp, gl->gl_state == LM_ST_UNLOCKED); gfs2_assert(sdp, list_empty(&gl->gl_reclaim)); gfs2_assert(sdp, list_empty(&gl->gl_holders)); @@ -335,7 +334,7 @@ int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number, gl->gl_state = LM_ST_UNLOCKED; gl->gl_demote_state = LM_ST_EXCLUSIVE; gl->gl_hash = hash; - gl->gl_owner_pid = 0; + gl->gl_owner_pid = NULL; gl->gl_ip = 0; gl->gl_ops = glops; gl->gl_req_gh = NULL; @@ -346,7 +345,6 @@ int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number, gl->gl_object = NULL; gl->gl_sbd = sdp; gl->gl_aspace = NULL; - lops_init_le(&gl->gl_le, &gfs2_glock_lops); INIT_DELAYED_WORK(&gl->gl_work, glock_work_func); /* If this glock protects actual on-disk data or metadata blocks, @@ -401,7 +399,7 @@ void gfs2_holder_init(struct gfs2_glock *gl, unsigned int state, unsigned flags, INIT_LIST_HEAD(&gh->gh_list); gh->gh_gl = gl; gh->gh_ip = (unsigned long)__builtin_return_address(0); - gh->gh_owner_pid = current->pid; + gh->gh_owner_pid = get_pid(task_pid(current)); gh->gh_state = state; gh->gh_flags = flags; gh->gh_error = 0; @@ -435,6 +433,7 @@ void gfs2_holder_reinit(unsigned int state, unsigned flags, struct gfs2_holder * void gfs2_holder_uninit(struct gfs2_holder *gh) { + put_pid(gh->gh_owner_pid); gfs2_glock_put(gh->gh_gl); gh->gh_gl = NULL; gh->gh_ip = 0; @@ -461,7 +460,6 @@ static void wait_on_holder(struct gfs2_holder *gh) static void gfs2_demote_wake(struct gfs2_glock *gl) { - BUG_ON(!spin_is_locked(&gl->gl_spin)); gl->gl_demote_state = LM_ST_EXCLUSIVE; clear_bit(GLF_DEMOTE, &gl->gl_flags); smp_mb__after_clear_bit(); @@ -507,21 +505,12 @@ static int rq_mutex(struct gfs2_holder *gh) static int rq_promote(struct gfs2_holder *gh) { struct gfs2_glock *gl = gh->gh_gl; - struct gfs2_sbd *sdp = gl->gl_sbd; if (!relaxed_state_ok(gl->gl_state, gh->gh_state, gh->gh_flags)) { if (list_empty(&gl->gl_holders)) { gl->gl_req_gh = gh; set_bit(GLF_LOCK, &gl->gl_flags); spin_unlock(&gl->gl_spin); - - if (atomic_read(&sdp->sd_reclaim_count) > - gfs2_tune_get(sdp, gt_reclaim_limit) && - !(gh->gh_flags & LM_FLAG_PRIORITY)) { - gfs2_reclaim_glock(sdp); - gfs2_reclaim_glock(sdp); - } - gfs2_glock_xmote_th(gh->gh_gl, gh); spin_lock(&gl->gl_spin); } @@ -567,7 +556,10 @@ static int rq_demote(struct gfs2_glock *gl) gfs2_demote_wake(gl); return 0; } + set_bit(GLF_LOCK, &gl->gl_flags); + set_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags); + if (gl->gl_demote_state == LM_ST_UNLOCKED || gl->gl_state != LM_ST_EXCLUSIVE) { spin_unlock(&gl->gl_spin); @@ -576,7 +568,9 @@ static int rq_demote(struct gfs2_glock *gl) spin_unlock(&gl->gl_spin); gfs2_glock_xmote_th(gl, NULL); } + spin_lock(&gl->gl_spin); + clear_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags); return 0; } @@ -598,23 +592,18 @@ static void run_queue(struct gfs2_glock *gl) if (!list_empty(&gl->gl_waiters1)) { gh = list_entry(gl->gl_waiters1.next, struct gfs2_holder, gh_list); - - if (test_bit(HIF_MUTEX, &gh->gh_iflags)) - blocked = rq_mutex(gh); - else - gfs2_assert_warn(gl->gl_sbd, 0); - + blocked = rq_mutex(gh); } else if (test_bit(GLF_DEMOTE, &gl->gl_flags)) { blocked = rq_demote(gl); + if (gl->gl_waiters2 && !blocked) { + set_bit(GLF_DEMOTE, &gl->gl_flags); + gl->gl_demote_state = LM_ST_UNLOCKED; + } + gl->gl_waiters2 = 0; } else if (!list_empty(&gl->gl_waiters3)) { gh = list_entry(gl->gl_waiters3.next, struct gfs2_holder, gh_list); - - if (test_bit(HIF_PROMOTE, &gh->gh_iflags)) - blocked = rq_promote(gh); - else - gfs2_assert_warn(gl->gl_sbd, 0); - + blocked = rq_promote(gh); } else break; @@ -632,27 +621,21 @@ static void run_queue(struct gfs2_glock *gl) static void gfs2_glmutex_lock(struct gfs2_glock *gl) { - struct gfs2_holder gh; - - gfs2_holder_init(gl, 0, 0, &gh); - set_bit(HIF_MUTEX, &gh.gh_iflags); - if (test_and_set_bit(HIF_WAIT, &gh.gh_iflags)) - BUG(); - spin_lock(&gl->gl_spin); if (test_and_set_bit(GLF_LOCK, &gl->gl_flags)) { + struct gfs2_holder gh; + + gfs2_holder_init(gl, 0, 0, &gh); + set_bit(HIF_WAIT, &gh.gh_iflags); list_add_tail(&gh.gh_list, &gl->gl_waiters1); + spin_unlock(&gl->gl_spin); + wait_on_holder(&gh); + gfs2_holder_uninit(&gh); } else { - gl->gl_owner_pid = current->pid; + gl->gl_owner_pid = get_pid(task_pid(current)); gl->gl_ip = (unsigned long)__builtin_return_address(0); - clear_bit(HIF_WAIT, &gh.gh_iflags); - smp_mb(); - wake_up_bit(&gh.gh_iflags, HIF_WAIT); + spin_unlock(&gl->gl_spin); } - spin_unlock(&gl->gl_spin); - - wait_on_holder(&gh); - gfs2_holder_uninit(&gh); } /** @@ -670,7 +653,7 @@ static int gfs2_glmutex_trylock(struct gfs2_glock *gl) if (test_and_set_bit(GLF_LOCK, &gl->gl_flags)) { acquired = 0; } else { - gl->gl_owner_pid = current->pid; + gl->gl_owner_pid = get_pid(task_pid(current)); gl->gl_ip = (unsigned long)__builtin_return_address(0); } spin_unlock(&gl->gl_spin); @@ -686,13 +669,17 @@ static int gfs2_glmutex_trylock(struct gfs2_glock *gl) static void gfs2_glmutex_unlock(struct gfs2_glock *gl) { + struct pid *pid; + spin_lock(&gl->gl_spin); clear_bit(GLF_LOCK, &gl->gl_flags); - gl->gl_owner_pid = 0; + pid = gl->gl_owner_pid; + gl->gl_owner_pid = NULL; gl->gl_ip = 0; run_queue(gl); - BUG_ON(!spin_is_locked(&gl->gl_spin)); spin_unlock(&gl->gl_spin); + + put_pid(pid); } /** @@ -722,7 +709,10 @@ static void handle_callback(struct gfs2_glock *gl, unsigned int state, } } else if (gl->gl_demote_state != LM_ST_UNLOCKED && gl->gl_demote_state != state) { - gl->gl_demote_state = LM_ST_UNLOCKED; + if (test_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags)) + gl->gl_waiters2 = 1; + else + gl->gl_demote_state = LM_ST_UNLOCKED; } spin_unlock(&gl->gl_spin); } @@ -943,8 +933,8 @@ static void gfs2_glock_drop_th(struct gfs2_glock *gl) const struct gfs2_glock_operations *glops = gl->gl_ops; unsigned int ret; - if (glops->go_drop_th) - glops->go_drop_th(gl); + if (glops->go_xmote_th) + glops->go_xmote_th(gl); gfs2_assert_warn(sdp, test_bit(GLF_LOCK, &gl->gl_flags)); gfs2_assert_warn(sdp, list_empty(&gl->gl_holders)); @@ -1061,7 +1051,7 @@ static int glock_wait_internal(struct gfs2_holder *gh) } static inline struct gfs2_holder * -find_holder_by_owner(struct list_head *head, pid_t pid) +find_holder_by_owner(struct list_head *head, struct pid *pid) { struct gfs2_holder *gh; @@ -1098,7 +1088,7 @@ static void add_to_queue(struct gfs2_holder *gh) struct gfs2_glock *gl = gh->gh_gl; struct gfs2_holder *existing; - BUG_ON(!gh->gh_owner_pid); + BUG_ON(gh->gh_owner_pid == NULL); if (test_and_set_bit(HIF_WAIT, &gh->gh_iflags)) BUG(); @@ -1108,12 +1098,14 @@ static void add_to_queue(struct gfs2_holder *gh) if (existing) { print_symbol(KERN_WARNING "original: %s\n", existing->gh_ip); - printk(KERN_INFO "pid : %d\n", existing->gh_owner_pid); + printk(KERN_INFO "pid : %d\n", + pid_nr(existing->gh_owner_pid)); printk(KERN_INFO "lock type : %d lock state : %d\n", existing->gh_gl->gl_name.ln_type, existing->gh_gl->gl_state); print_symbol(KERN_WARNING "new: %s\n", gh->gh_ip); - printk(KERN_INFO "pid : %d\n", gh->gh_owner_pid); + printk(KERN_INFO "pid : %d\n", + pid_nr(gh->gh_owner_pid)); printk(KERN_INFO "lock type : %d lock state : %d\n", gl->gl_name.ln_type, gl->gl_state); BUG(); @@ -1156,8 +1148,6 @@ restart: return -EIO; } - set_bit(HIF_PROMOTE, &gh->gh_iflags); - spin_lock(&gl->gl_spin); add_to_queue(gh); run_queue(gl); @@ -1248,12 +1238,11 @@ void gfs2_glock_dq(struct gfs2_holder *gh) list_del_init(&gh->gh_list); if (list_empty(&gl->gl_holders)) { - spin_unlock(&gl->gl_spin); - - if (glops->go_unlock) + if (glops->go_unlock) { + spin_unlock(&gl->gl_spin); glops->go_unlock(gh); - - spin_lock(&gl->gl_spin); + spin_lock(&gl->gl_spin); + } gl->gl_stamp = jiffies; } @@ -1817,8 +1806,9 @@ static int dump_holder(struct glock_iter *gi, char *str, print_dbg(gi, " %s\n", str); if (gh->gh_owner_pid) { - print_dbg(gi, " owner = %ld ", (long)gh->gh_owner_pid); - gh_owner = find_task_by_pid(gh->gh_owner_pid); + print_dbg(gi, " owner = %ld ", + (long)pid_nr(gh->gh_owner_pid)); + gh_owner = pid_task(gh->gh_owner_pid, PIDTYPE_PID); if (gh_owner) print_dbg(gi, "(%s)\n", gh_owner->comm); else @@ -1896,13 +1886,13 @@ static int dump_glock(struct glock_iter *gi, struct gfs2_glock *gl) print_dbg(gi, " gl_ref = %d\n", atomic_read(&gl->gl_ref)); print_dbg(gi, " gl_state = %u\n", gl->gl_state); if (gl->gl_owner_pid) { - gl_owner = find_task_by_pid(gl->gl_owner_pid); + gl_owner = pid_task(gl->gl_owner_pid, PIDTYPE_PID); if (gl_owner) print_dbg(gi, " gl_owner = pid %d (%s)\n", - gl->gl_owner_pid, gl_owner->comm); + pid_nr(gl->gl_owner_pid), gl_owner->comm); else print_dbg(gi, " gl_owner = %d (ended)\n", - gl->gl_owner_pid); + pid_nr(gl->gl_owner_pid)); } else print_dbg(gi, " gl_owner = -1\n"); print_dbg(gi, " gl_ip = %lu\n", gl->gl_ip); @@ -1910,8 +1900,6 @@ static int dump_glock(struct glock_iter *gi, struct gfs2_glock *gl) print_dbg(gi, " req_bh = %s\n", (gl->gl_req_bh) ? "yes" : "no"); print_dbg(gi, " lvb_count = %d\n", atomic_read(&gl->gl_lvb_count)); print_dbg(gi, " object = %s\n", (gl->gl_object) ? "yes" : "no"); - print_dbg(gi, " le = %s\n", - (list_empty(&gl->gl_le.le_list)) ? "no" : "yes"); print_dbg(gi, " reclaim = %s\n", (list_empty(&gl->gl_reclaim)) ? "no" : "yes"); if (gl->gl_aspace) diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h index b16f604eea9..2f9c6d136b3 100644 --- a/fs/gfs2/glock.h +++ b/fs/gfs2/glock.h @@ -36,11 +36,13 @@ static inline int gfs2_glock_is_locked_by_me(struct gfs2_glock *gl) { struct gfs2_holder *gh; int locked = 0; + struct pid *pid; /* Look in glock's list of holders for one with current task as owner */ spin_lock(&gl->gl_spin); + pid = task_pid(current); list_for_each_entry(gh, &gl->gl_holders, gh_list) { - if (gh->gh_owner_pid == current->pid) { + if (gh->gh_owner_pid == pid) { locked = 1; break; } diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c index 4670dcb2a87..c663b7a0f41 100644 --- a/fs/gfs2/glops.c +++ b/fs/gfs2/glops.c @@ -56,7 +56,7 @@ static void gfs2_ail_empty_gl(struct gfs2_glock *gl) bd = list_entry(head->next, struct gfs2_bufdata, bd_ail_gl_list); bh = bd->bd_bh; - gfs2_remove_from_ail(NULL, bd); + gfs2_remove_from_ail(bd); bd->bd_bh = NULL; bh->b_private = NULL; bd->bd_blkno = bh->b_blocknr; @@ -86,15 +86,10 @@ static void gfs2_pte_inval(struct gfs2_glock *gl) if (!ip || !S_ISREG(inode->i_mode)) return; - if (!test_bit(GIF_PAGED, &ip->i_flags)) - return; - unmap_shared_mapping_range(inode->i_mapping, 0, 0); - if (test_bit(GIF_SW_PAGED, &ip->i_flags)) set_bit(GLF_DIRTY, &gl->gl_flags); - clear_bit(GIF_SW_PAGED, &ip->i_flags); } /** @@ -143,44 +138,34 @@ static void meta_go_inval(struct gfs2_glock *gl, int flags) static void inode_go_sync(struct gfs2_glock *gl) { struct gfs2_inode *ip = gl->gl_object; + struct address_space *metamapping = gl->gl_aspace->i_mapping; + int error; + + if (gl->gl_state != LM_ST_UNLOCKED) + gfs2_pte_inval(gl); + if (gl->gl_state != LM_ST_EXCLUSIVE) + return; if (ip && !S_ISREG(ip->i_inode.i_mode)) ip = NULL; if (test_bit(GLF_DIRTY, &gl->gl_flags)) { - if (ip && !gfs2_is_jdata(ip)) - filemap_fdatawrite(ip->i_inode.i_mapping); gfs2_log_flush(gl->gl_sbd, gl); - if (ip && gfs2_is_jdata(ip)) - filemap_fdatawrite(ip->i_inode.i_mapping); - gfs2_meta_sync(gl); + filemap_fdatawrite(metamapping); if (ip) { struct address_space *mapping = ip->i_inode.i_mapping; - int error = filemap_fdatawait(mapping); + filemap_fdatawrite(mapping); + error = filemap_fdatawait(mapping); mapping_set_error(mapping, error); } + error = filemap_fdatawait(metamapping); + mapping_set_error(metamapping, error); clear_bit(GLF_DIRTY, &gl->gl_flags); gfs2_ail_empty_gl(gl); } } /** - * inode_go_xmote_th - promote/demote a glock - * @gl: the glock - * @state: the requested state - * @flags: - * - */ - -static void inode_go_xmote_th(struct gfs2_glock *gl) -{ - if (gl->gl_state != LM_ST_UNLOCKED) - gfs2_pte_inval(gl); - if (gl->gl_state == LM_ST_EXCLUSIVE) - inode_go_sync(gl); -} - -/** * inode_go_xmote_bh - After promoting/demoting a glock * @gl: the glock * @@ -201,22 +186,6 @@ static void inode_go_xmote_bh(struct gfs2_glock *gl) } /** - * inode_go_drop_th - unlock a glock - * @gl: the glock - * - * Invoked from rq_demote(). - * Another node needs the lock in EXCLUSIVE mode, or lock (unused for too long) - * is being purged from our node's glock cache; we're dropping lock. - */ - -static void inode_go_drop_th(struct gfs2_glock *gl) -{ - gfs2_pte_inval(gl); - if (gl->gl_state == LM_ST_EXCLUSIVE) - inode_go_sync(gl); -} - -/** * inode_go_inval - prepare a inode glock to be released * @gl: the glock * @flags: @@ -234,10 +203,8 @@ static void inode_go_inval(struct gfs2_glock *gl, int flags) set_bit(GIF_INVALID, &ip->i_flags); } - if (ip && S_ISREG(ip->i_inode.i_mode)) { + if (ip && S_ISREG(ip->i_inode.i_mode)) truncate_inode_pages(ip->i_inode.i_mapping, 0); - clear_bit(GIF_PAGED, &ip->i_flags); - } } /** @@ -294,23 +261,6 @@ static int inode_go_lock(struct gfs2_holder *gh) } /** - * inode_go_unlock - operation done before an inode lock is unlocked by a - * process - * @gl: the glock - * @flags: - * - */ - -static void inode_go_unlock(struct gfs2_holder *gh) -{ - struct gfs2_glock *gl = gh->gh_gl; - struct gfs2_inode *ip = gl->gl_object; - - if (ip) - gfs2_meta_cache_flush(ip); -} - -/** * rgrp_go_demote_ok - Check to see if it's ok to unlock a RG's glock * @gl: the glock * @@ -350,14 +300,14 @@ static void rgrp_go_unlock(struct gfs2_holder *gh) } /** - * trans_go_xmote_th - promote/demote the transaction glock + * trans_go_sync - promote/demote the transaction glock * @gl: the glock * @state: the requested state * @flags: * */ -static void trans_go_xmote_th(struct gfs2_glock *gl) +static void trans_go_sync(struct gfs2_glock *gl) { struct gfs2_sbd *sdp = gl->gl_sbd; @@ -384,7 +334,6 @@ static void trans_go_xmote_bh(struct gfs2_glock *gl) if (gl->gl_state != LM_ST_UNLOCKED && test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)) { - gfs2_meta_cache_flush(GFS2_I(sdp->sd_jdesc->jd_inode)); j_gl->gl_ops->go_inval(j_gl, DIO_METADATA); error = gfs2_find_jhead(sdp->sd_jdesc, &head); @@ -402,24 +351,6 @@ static void trans_go_xmote_bh(struct gfs2_glock *gl) } /** - * trans_go_drop_th - unlock the transaction glock - * @gl: the glock - * - * We want to sync the device even with localcaching. Remember - * that localcaching journal replay only marks buffers dirty. - */ - -static void trans_go_drop_th(struct gfs2_glock *gl) -{ - struct gfs2_sbd *sdp = gl->gl_sbd; - - if (test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)) { - gfs2_meta_syncfs(sdp); - gfs2_log_shutdown(sdp); - } -} - -/** * quota_go_demote_ok - Check to see if it's ok to unlock a quota glock * @gl: the glock * @@ -433,25 +364,21 @@ static int quota_go_demote_ok(struct gfs2_glock *gl) const struct gfs2_glock_operations gfs2_meta_glops = { .go_xmote_th = meta_go_sync, - .go_drop_th = meta_go_sync, .go_type = LM_TYPE_META, }; const struct gfs2_glock_operations gfs2_inode_glops = { - .go_xmote_th = inode_go_xmote_th, + .go_xmote_th = inode_go_sync, .go_xmote_bh = inode_go_xmote_bh, - .go_drop_th = inode_go_drop_th, .go_inval = inode_go_inval, .go_demote_ok = inode_go_demote_ok, .go_lock = inode_go_lock, - .go_unlock = inode_go_unlock, .go_type = LM_TYPE_INODE, .go_min_hold_time = HZ / 10, }; const struct gfs2_glock_operations gfs2_rgrp_glops = { .go_xmote_th = meta_go_sync, - .go_drop_th = meta_go_sync, .go_inval = meta_go_inval, .go_demote_ok = rgrp_go_demote_ok, .go_lock = rgrp_go_lock, @@ -461,9 +388,8 @@ const struct gfs2_glock_operations gfs2_rgrp_glops = { }; const struct gfs2_glock_operations gfs2_trans_glops = { - .go_xmote_th = trans_go_xmote_th, + .go_xmote_th = trans_go_sync, .go_xmote_bh = trans_go_xmote_bh, - .go_drop_th = trans_go_drop_th, .go_type = LM_TYPE_NONDISK, }; diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index eaddfb5a8e6..525dcae352d 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -1,6 +1,6 @@ /* * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. - * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved. + * Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved. * * This copyrighted material is made available to anyone wishing to use, * modify, copy, or redistribute it subject to the terms and conditions @@ -131,7 +131,6 @@ struct gfs2_bufdata { struct gfs2_glock_operations { void (*go_xmote_th) (struct gfs2_glock *gl); void (*go_xmote_bh) (struct gfs2_glock *gl); - void (*go_drop_th) (struct gfs2_glock *gl); void (*go_inval) (struct gfs2_glock *gl, int flags); int (*go_demote_ok) (struct gfs2_glock *gl); int (*go_lock) (struct gfs2_holder *gh); @@ -141,10 +140,6 @@ struct gfs2_glock_operations { }; enum { - /* Actions */ - HIF_MUTEX = 0, - HIF_PROMOTE = 1, - /* States */ HIF_HOLDER = 6, HIF_FIRST = 7, @@ -156,7 +151,7 @@ struct gfs2_holder { struct list_head gh_list; struct gfs2_glock *gh_gl; - pid_t gh_owner_pid; + struct pid *gh_owner_pid; unsigned int gh_state; unsigned gh_flags; @@ -171,6 +166,8 @@ enum { GLF_DEMOTE = 3, GLF_PENDING_DEMOTE = 4, GLF_DIRTY = 5, + GLF_DEMOTE_IN_PROGRESS = 6, + GLF_LFLUSH = 7, }; struct gfs2_glock { @@ -185,11 +182,12 @@ struct gfs2_glock { unsigned int gl_hash; unsigned int gl_demote_state; /* state requested by remote node */ unsigned long gl_demote_time; /* time of first demote request */ - pid_t gl_owner_pid; + struct pid *gl_owner_pid; unsigned long gl_ip; struct list_head gl_holders; struct list_head gl_waiters1; /* HIF_MUTEX */ struct list_head gl_waiters3; /* HIF_PROMOTE */ + int gl_waiters2; /* GIF_DEMOTE */ const struct gfs2_glock_operations *gl_ops; @@ -210,7 +208,6 @@ struct gfs2_glock { struct gfs2_sbd *gl_sbd; struct inode *gl_aspace; - struct gfs2_log_element gl_le; struct list_head gl_ail_list; atomic_t gl_ail_count; struct delayed_work gl_work; @@ -239,7 +236,6 @@ struct gfs2_alloc { enum { GIF_INVALID = 0, GIF_QD_LOCKED = 1, - GIF_PAGED = 2, GIF_SW_PAGED = 3, }; @@ -268,14 +264,10 @@ struct gfs2_inode { struct gfs2_glock *i_gl; /* Move into i_gh? */ struct gfs2_holder i_iopen_gh; struct gfs2_holder i_gh; /* for prepare/commit_write only */ - struct gfs2_alloc i_alloc; + struct gfs2_alloc *i_alloc; u64 i_last_rg_alloc; - spinlock_t i_spin; struct rw_semaphore i_rw_mutex; - unsigned long i_last_pfault; - - struct buffer_head *i_cache[GFS2_MAX_META_HEIGHT]; }; /* @@ -287,19 +279,12 @@ static inline struct gfs2_inode *GFS2_I(struct inode *inode) return container_of(inode, struct gfs2_inode, i_inode); } -/* To be removed? */ -static inline struct gfs2_sbd *GFS2_SB(struct inode *inode) +static inline struct gfs2_sbd *GFS2_SB(const struct inode *inode) { return inode->i_sb->s_fs_info; } -enum { - GFF_DID_DIRECT_ALLOC = 0, - GFF_EXLOCK = 1, -}; - struct gfs2_file { - unsigned long f_flags; /* GFF_... */ struct mutex f_fl_mutex; struct gfs2_holder f_fl_gh; }; @@ -373,8 +358,17 @@ struct gfs2_ail { u64 ai_sync_gen; }; +struct gfs2_journal_extent { + struct list_head extent_list; + + unsigned int lblock; /* First logical block */ + u64 dblock; /* First disk block */ + u64 blocks; +}; + struct gfs2_jdesc { struct list_head jd_list; + struct list_head extent_list; struct inode *jd_inode; unsigned int jd_jid; @@ -421,13 +415,9 @@ struct gfs2_args { struct gfs2_tune { spinlock_t gt_spin; - unsigned int gt_ilimit; - unsigned int gt_ilimit_tries; - unsigned int gt_ilimit_min; unsigned int gt_demote_secs; /* Cache retention for unheld glock */ unsigned int gt_incore_log_blocks; unsigned int gt_log_flush_secs; - unsigned int gt_jindex_refresh_secs; /* Check for new journal index */ unsigned int gt_recoverd_secs; unsigned int gt_logd_secs; @@ -443,10 +433,8 @@ struct gfs2_tune { unsigned int gt_new_files_jdata; unsigned int gt_new_files_directio; unsigned int gt_max_readahead; /* Max bytes to read-ahead from disk */ - unsigned int gt_lockdump_size; unsigned int gt_stall_secs; /* Detects trouble! */ unsigned int gt_complain_secs; - unsigned int gt_reclaim_limit; /* Max num of glocks in reclaim list */ unsigned int gt_statfs_quantum; unsigned int gt_statfs_slow; }; @@ -539,7 +527,6 @@ struct gfs2_sbd { /* StatFS stuff */ spinlock_t sd_statfs_spin; - struct mutex sd_statfs_mutex; struct gfs2_statfs_change_host sd_statfs_master; struct gfs2_statfs_change_host sd_statfs_local; unsigned long sd_statfs_sync_time; @@ -602,20 +589,18 @@ struct gfs2_sbd { unsigned int sd_log_commited_databuf; unsigned int sd_log_commited_revoke; - unsigned int sd_log_num_gl; unsigned int sd_log_num_buf; unsigned int sd_log_num_revoke; unsigned int sd_log_num_rg; unsigned int sd_log_num_databuf; - struct list_head sd_log_le_gl; struct list_head sd_log_le_buf; struct list_head sd_log_le_revoke; struct list_head sd_log_le_rg; struct list_head sd_log_le_databuf; struct list_head sd_log_le_ordered; - unsigned int sd_log_blks_free; + atomic_t sd_log_blks_free; struct mutex sd_log_reserve_mutex; u64 sd_log_sequence; diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 5f6dc32946c..37725ade3c5 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -31,7 +31,6 @@ #include "log.h" #include "meta_io.h" #include "ops_address.h" -#include "ops_file.h" #include "ops_inode.h" #include "quota.h" #include "rgrp.h" @@ -132,15 +131,21 @@ static struct inode *gfs2_iget_skip(struct super_block *sb, void gfs2_set_iop(struct inode *inode) { + struct gfs2_sbd *sdp = GFS2_SB(inode); umode_t mode = inode->i_mode; if (S_ISREG(mode)) { inode->i_op = &gfs2_file_iops; - inode->i_fop = &gfs2_file_fops; - inode->i_mapping->a_ops = &gfs2_file_aops; + if (sdp->sd_args.ar_localflocks) + inode->i_fop = &gfs2_file_fops_nolock; + else + inode->i_fop = &gfs2_file_fops; } else if (S_ISDIR(mode)) { inode->i_op = &gfs2_dir_iops; - inode->i_fop = &gfs2_dir_fops; + if (sdp->sd_args.ar_localflocks) + inode->i_fop = &gfs2_dir_fops_nolock; + else + inode->i_fop = &gfs2_dir_fops; } else if (S_ISLNK(mode)) { inode->i_op = &gfs2_symlink_iops; } else { @@ -235,7 +240,7 @@ fail_put: ip->i_gl->gl_object = NULL; gfs2_glock_put(ip->i_gl); fail: - iput(inode); + iget_failed(inode); return ERR_PTR(error); } @@ -291,12 +296,10 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) di->di_entries = be32_to_cpu(str->di_entries); di->di_eattr = be64_to_cpu(str->di_eattr); - return 0; -} + if (S_ISREG(ip->i_inode.i_mode)) + gfs2_set_aops(&ip->i_inode); -static void gfs2_inode_bh(struct gfs2_inode *ip, struct buffer_head *bh) -{ - ip->i_cache[0] = bh; + return 0; } /** @@ -366,7 +369,8 @@ int gfs2_dinode_dealloc(struct gfs2_inode *ip) if (error) goto out_rg_gunlock; - gfs2_trans_add_gl(ip->i_gl); + set_bit(GLF_DIRTY, &ip->i_gl->gl_flags); + set_bit(GLF_LFLUSH, &ip->i_gl->gl_flags); gfs2_free_di(rgd, ip); @@ -707,9 +711,10 @@ static int alloc_dinode(struct gfs2_inode *dip, u64 *no_addr, u64 *generation) struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); int error; - gfs2_alloc_get(dip); + if (gfs2_alloc_get(dip) == NULL) + return -ENOMEM; - dip->i_alloc.al_requested = RES_DINODE; + dip->i_alloc->al_requested = RES_DINODE; error = gfs2_inplace_reserve(dip); if (error) goto out; @@ -855,7 +860,7 @@ static int link_dinode(struct gfs2_inode *dip, const struct qstr *name, error = alloc_required = gfs2_diradd_alloc_required(&dip->i_inode, name); if (alloc_required < 0) - goto fail; + goto fail_quota_locks; if (alloc_required) { error = gfs2_quota_check(dip, dip->i_inode.i_uid, dip->i_inode.i_gid); if (error) @@ -896,7 +901,7 @@ fail_end_trans: gfs2_trans_end(sdp); fail_ipreserv: - if (dip->i_alloc.al_rgd) + if (dip->i_alloc->al_rgd) gfs2_inplace_release(dip); fail_quota_locks: @@ -966,7 +971,7 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, struct gfs2_inum_host inum = { .no_addr = 0, .no_formal_ino = 0 }; int error; u64 generation; - struct buffer_head *bh=NULL; + struct buffer_head *bh = NULL; if (!name->len || name->len > GFS2_FNAMESIZE) return ERR_PTR(-ENAMETOOLONG); @@ -1003,8 +1008,6 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, if (IS_ERR(inode)) goto fail_gunlock2; - gfs2_inode_bh(GFS2_I(inode), bh); - error = gfs2_inode_refresh(GFS2_I(inode)); if (error) goto fail_gunlock2; @@ -1021,6 +1024,8 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, if (error) goto fail_gunlock2; + if (bh) + brelse(bh); if (!inode) return ERR_PTR(-ENOMEM); return inode; @@ -1032,6 +1037,8 @@ fail_gunlock2: fail_gunlock: gfs2_glock_dq(ghs); fail: + if (bh) + brelse(bh); return ERR_PTR(error); } diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h index 351ac87ab38..d4465066261 100644 --- a/fs/gfs2/inode.h +++ b/fs/gfs2/inode.h @@ -20,6 +20,18 @@ static inline int gfs2_is_jdata(const struct gfs2_inode *ip) return ip->i_di.di_flags & GFS2_DIF_JDATA; } +static inline int gfs2_is_writeback(const struct gfs2_inode *ip) +{ + const struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); + return (sdp->sd_args.ar_data == GFS2_DATA_WRITEBACK) && !gfs2_is_jdata(ip); +} + +static inline int gfs2_is_ordered(const struct gfs2_inode *ip) +{ + const struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); + return (sdp->sd_args.ar_data == GFS2_DATA_ORDERED) && !gfs2_is_jdata(ip); +} + static inline int gfs2_is_dir(const struct gfs2_inode *ip) { return S_ISDIR(ip->i_inode.i_mode); diff --git a/fs/gfs2/locking/dlm/mount.c b/fs/gfs2/locking/dlm/mount.c index 41c5b04caab..f2efff42422 100644 --- a/fs/gfs2/locking/dlm/mount.c +++ b/fs/gfs2/locking/dlm/mount.c @@ -67,6 +67,11 @@ static int make_args(struct gdlm_ls *ls, char *data_arg, int *nodir) memset(data, 0, 256); strncpy(data, data_arg, 255); + if (!strlen(data)) { + log_error("no mount options, (u)mount helpers not installed"); + return -EINVAL; + } + for (options = data; (x = strsep(&options, ":")); ) { if (!*x) continue; diff --git a/fs/gfs2/locking/dlm/plock.c b/fs/gfs2/locking/dlm/plock.c index 1f7b038530b..2ebd374b314 100644 --- a/fs/gfs2/locking/dlm/plock.c +++ b/fs/gfs2/locking/dlm/plock.c @@ -89,15 +89,19 @@ int gdlm_plock(void *lockspace, struct lm_lockname *name, op->info.number = name->ln_number; op->info.start = fl->fl_start; op->info.end = fl->fl_end; - op->info.owner = (__u64)(long) fl->fl_owner; if (fl->fl_lmops && fl->fl_lmops->fl_grant) { + /* fl_owner is lockd which doesn't distinguish + processes on the nfs client */ + op->info.owner = (__u64) fl->fl_pid; xop->callback = fl->fl_lmops->fl_grant; locks_init_lock(&xop->flc); locks_copy_lock(&xop->flc, fl); xop->fl = fl; xop->file = file; - } else + } else { + op->info.owner = (__u64)(long) fl->fl_owner; xop->callback = NULL; + } send_op(op); @@ -203,7 +207,10 @@ int gdlm_punlock(void *lockspace, struct lm_lockname *name, op->info.number = name->ln_number; op->info.start = fl->fl_start; op->info.end = fl->fl_end; - op->info.owner = (__u64)(long) fl->fl_owner; + if (fl->fl_lmops && fl->fl_lmops->fl_grant) + op->info.owner = (__u64) fl->fl_pid; + else + op->info.owner = (__u64)(long) fl->fl_owner; send_op(op); wait_event(recv_wq, (op->done != 0)); @@ -242,7 +249,10 @@ int gdlm_plock_get(void *lockspace, struct lm_lockname *name, op->info.number = name->ln_number; op->info.start = fl->fl_start; op->info.end = fl->fl_end; - op->info.owner = (__u64)(long) fl->fl_owner; + if (fl->fl_lmops && fl->fl_lmops->fl_grant) + op->info.owner = (__u64) fl->fl_pid; + else + op->info.owner = (__u64)(long) fl->fl_owner; send_op(op); wait_event(recv_wq, (op->done != 0)); diff --git a/fs/gfs2/locking/dlm/sysfs.c b/fs/gfs2/locking/dlm/sysfs.c index ae9e6a25fe2..a87b0983976 100644 --- a/fs/gfs2/locking/dlm/sysfs.c +++ b/fs/gfs2/locking/dlm/sysfs.c @@ -189,51 +189,39 @@ static struct kobj_type gdlm_ktype = { .sysfs_ops = &gdlm_attr_ops, }; -static struct kset gdlm_kset = { - .ktype = &gdlm_ktype, -}; +static struct kset *gdlm_kset; int gdlm_kobject_setup(struct gdlm_ls *ls, struct kobject *fskobj) { int error; - error = kobject_set_name(&ls->kobj, "%s", "lock_module"); - if (error) { - log_error("can't set kobj name %d", error); - return error; - } - - ls->kobj.kset = &gdlm_kset; - ls->kobj.ktype = &gdlm_ktype; - ls->kobj.parent = fskobj; - - error = kobject_register(&ls->kobj); + ls->kobj.kset = gdlm_kset; + error = kobject_init_and_add(&ls->kobj, &gdlm_ktype, fskobj, + "lock_module"); if (error) log_error("can't register kobj %d", error); + kobject_uevent(&ls->kobj, KOBJ_ADD); return error; } void gdlm_kobject_release(struct gdlm_ls *ls) { - kobject_unregister(&ls->kobj); + kobject_put(&ls->kobj); } int gdlm_sysfs_init(void) { - int error; - - kobject_set_name(&gdlm_kset.kobj, "lock_dlm"); - kobj_set_kset_s(&gdlm_kset, kernel_subsys); - error = kset_register(&gdlm_kset); - if (error) - printk("lock_dlm: cannot register kset %d\n", error); - - return error; + gdlm_kset = kset_create_and_add("lock_dlm", NULL, kernel_kobj); + if (!gdlm_kset) { + printk(KERN_WARNING "%s: can not create kset\n", __FUNCTION__); + return -ENOMEM; + } + return 0; } void gdlm_sysfs_exit(void) { - kset_unregister(&gdlm_kset); + kset_unregister(gdlm_kset); } diff --git a/fs/gfs2/locking/dlm/thread.c b/fs/gfs2/locking/dlm/thread.c index bd938f06481..521694fc19d 100644 --- a/fs/gfs2/locking/dlm/thread.c +++ b/fs/gfs2/locking/dlm/thread.c @@ -273,18 +273,13 @@ static int gdlm_thread(void *data, int blist) struct gdlm_ls *ls = (struct gdlm_ls *) data; struct gdlm_lock *lp = NULL; uint8_t complete, blocking, submit, drop; - DECLARE_WAITQUEUE(wait, current); /* Only thread1 is allowed to do blocking callbacks since gfs may wait for a completion callback within a blocking cb. */ while (!kthread_should_stop()) { - set_current_state(TASK_INTERRUPTIBLE); - add_wait_queue(&ls->thread_wait, &wait); - if (no_work(ls, blist)) - schedule(); - remove_wait_queue(&ls->thread_wait, &wait); - set_current_state(TASK_RUNNING); + wait_event_interruptible(ls->thread_wait, + !no_work(ls, blist) || kthread_should_stop()); complete = blocking = submit = drop = 0; diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 7df70247325..161ab6f2058 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -1,6 +1,6 @@ /* * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. - * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved. + * Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved. * * This copyrighted material is made available to anyone wishing to use, * modify, copy, or redistribute it subject to the terms and conditions @@ -16,6 +16,8 @@ #include <linux/crc32.h> #include <linux/lm_interface.h> #include <linux/delay.h> +#include <linux/kthread.h> +#include <linux/freezer.h> #include "gfs2.h" #include "incore.h" @@ -68,14 +70,12 @@ unsigned int gfs2_struct2blk(struct gfs2_sbd *sdp, unsigned int nstruct, * */ -void gfs2_remove_from_ail(struct address_space *mapping, struct gfs2_bufdata *bd) +void gfs2_remove_from_ail(struct gfs2_bufdata *bd) { bd->bd_ail = NULL; list_del_init(&bd->bd_ail_st_list); list_del_init(&bd->bd_ail_gl_list); atomic_dec(&bd->bd_gl->gl_ail_count); - if (mapping) - gfs2_meta_cache_flush(GFS2_I(mapping->host)); brelse(bd->bd_bh); } @@ -92,8 +92,6 @@ static void gfs2_ail1_start_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai) struct buffer_head *bh; int retry; - BUG_ON(!spin_is_locked(&sdp->sd_log_lock)); - do { retry = 0; @@ -210,7 +208,7 @@ static void gfs2_ail1_start(struct gfs2_sbd *sdp, int flags) gfs2_log_unlock(sdp); } -int gfs2_ail1_empty(struct gfs2_sbd *sdp, int flags) +static int gfs2_ail1_empty(struct gfs2_sbd *sdp, int flags) { struct gfs2_ail *ai, *s; int ret; @@ -248,7 +246,7 @@ static void gfs2_ail2_empty_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai) bd = list_entry(head->prev, struct gfs2_bufdata, bd_ail_st_list); gfs2_assert(sdp, bd->bd_ail == ai); - gfs2_remove_from_ail(bd->bd_bh->b_page->mapping, bd); + gfs2_remove_from_ail(bd); } } @@ -303,7 +301,7 @@ int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks) mutex_lock(&sdp->sd_log_reserve_mutex); gfs2_log_lock(sdp); - while(sdp->sd_log_blks_free <= (blks + reserved_blks)) { + while(atomic_read(&sdp->sd_log_blks_free) <= (blks + reserved_blks)) { gfs2_log_unlock(sdp); gfs2_ail1_empty(sdp, 0); gfs2_log_flush(sdp, NULL); @@ -312,7 +310,7 @@ int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks) gfs2_ail1_start(sdp, 0); gfs2_log_lock(sdp); } - sdp->sd_log_blks_free -= blks; + atomic_sub(blks, &sdp->sd_log_blks_free); gfs2_log_unlock(sdp); mutex_unlock(&sdp->sd_log_reserve_mutex); @@ -332,27 +330,23 @@ void gfs2_log_release(struct gfs2_sbd *sdp, unsigned int blks) { gfs2_log_lock(sdp); - sdp->sd_log_blks_free += blks; + atomic_add(blks, &sdp->sd_log_blks_free); gfs2_assert_withdraw(sdp, - sdp->sd_log_blks_free <= sdp->sd_jdesc->jd_blocks); + atomic_read(&sdp->sd_log_blks_free) <= sdp->sd_jdesc->jd_blocks); gfs2_log_unlock(sdp); up_read(&sdp->sd_log_flush_lock); } static u64 log_bmap(struct gfs2_sbd *sdp, unsigned int lbn) { - struct inode *inode = sdp->sd_jdesc->jd_inode; - int error; - struct buffer_head bh_map = { .b_state = 0, .b_blocknr = 0 }; - - bh_map.b_size = 1 << inode->i_blkbits; - error = gfs2_block_map(inode, lbn, 0, &bh_map); - if (error || !bh_map.b_blocknr) - printk(KERN_INFO "error=%d, dbn=%llu lbn=%u", error, - (unsigned long long)bh_map.b_blocknr, lbn); - gfs2_assert_withdraw(sdp, !error && bh_map.b_blocknr); - - return bh_map.b_blocknr; + struct gfs2_journal_extent *je; + + list_for_each_entry(je, &sdp->sd_jdesc->extent_list, extent_list) { + if (lbn >= je->lblock && lbn < je->lblock + je->blocks) + return je->dblock + lbn - je->lblock; + } + + return -1; } /** @@ -561,8 +555,8 @@ static void log_pull_tail(struct gfs2_sbd *sdp, unsigned int new_tail) ail2_empty(sdp, new_tail); gfs2_log_lock(sdp); - sdp->sd_log_blks_free += dist; - gfs2_assert_withdraw(sdp, sdp->sd_log_blks_free <= sdp->sd_jdesc->jd_blocks); + atomic_add(dist, &sdp->sd_log_blks_free); + gfs2_assert_withdraw(sdp, atomic_read(&sdp->sd_log_blks_free) <= sdp->sd_jdesc->jd_blocks); gfs2_log_unlock(sdp); sdp->sd_log_tail = new_tail; @@ -652,7 +646,7 @@ static void gfs2_ordered_write(struct gfs2_sbd *sdp) get_bh(bh); gfs2_log_unlock(sdp); lock_buffer(bh); - if (test_clear_buffer_dirty(bh)) { + if (buffer_mapped(bh) && test_clear_buffer_dirty(bh)) { bh->b_end_io = end_buffer_write_sync; submit_bh(WRITE, bh); } else { @@ -694,20 +688,16 @@ static void gfs2_ordered_wait(struct gfs2_sbd *sdp) * */ -void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl) +void __gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl) { struct gfs2_ail *ai; down_write(&sdp->sd_log_flush_lock); - if (gl) { - gfs2_log_lock(sdp); - if (list_empty(&gl->gl_le.le_list)) { - gfs2_log_unlock(sdp); - up_write(&sdp->sd_log_flush_lock); - return; - } - gfs2_log_unlock(sdp); + /* Log might have been flushed while we waited for the flush lock */ + if (gl && !test_bit(GLF_LFLUSH, &gl->gl_flags)) { + up_write(&sdp->sd_log_flush_lock); + return; } ai = kzalloc(sizeof(struct gfs2_ail), GFP_NOFS | __GFP_NOFAIL); @@ -739,7 +729,7 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl) log_flush_commit(sdp); else if (sdp->sd_log_tail != current_tail(sdp) && !sdp->sd_log_idle){ gfs2_log_lock(sdp); - sdp->sd_log_blks_free--; /* Adjust for unreserved buffer */ + atomic_dec(&sdp->sd_log_blks_free); /* Adjust for unreserved buffer */ gfs2_log_unlock(sdp); log_write_header(sdp, 0, PULL); } @@ -767,7 +757,7 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl) static void log_refund(struct gfs2_sbd *sdp, struct gfs2_trans *tr) { unsigned int reserved; - unsigned int old; + unsigned int unused; gfs2_log_lock(sdp); @@ -779,14 +769,11 @@ static void log_refund(struct gfs2_sbd *sdp, struct gfs2_trans *tr) sdp->sd_log_commited_revoke += tr->tr_num_revoke - tr->tr_num_revoke_rm; gfs2_assert_withdraw(sdp, ((int)sdp->sd_log_commited_revoke) >= 0); reserved = calc_reserved(sdp); - old = sdp->sd_log_blks_free; - sdp->sd_log_blks_free += tr->tr_reserved - - (reserved - sdp->sd_log_blks_reserved); - - gfs2_assert_withdraw(sdp, sdp->sd_log_blks_free >= old); - gfs2_assert_withdraw(sdp, sdp->sd_log_blks_free <= + unused = sdp->sd_log_blks_reserved - reserved + tr->tr_reserved; + gfs2_assert_withdraw(sdp, unused >= 0); + atomic_add(unused, &sdp->sd_log_blks_free); + gfs2_assert_withdraw(sdp, atomic_read(&sdp->sd_log_blks_free) <= sdp->sd_jdesc->jd_blocks); - sdp->sd_log_blks_reserved = reserved; gfs2_log_unlock(sdp); @@ -825,7 +812,6 @@ void gfs2_log_shutdown(struct gfs2_sbd *sdp) down_write(&sdp->sd_log_flush_lock); gfs2_assert_withdraw(sdp, !sdp->sd_log_blks_reserved); - gfs2_assert_withdraw(sdp, !sdp->sd_log_num_gl); gfs2_assert_withdraw(sdp, !sdp->sd_log_num_buf); gfs2_assert_withdraw(sdp, !sdp->sd_log_num_revoke); gfs2_assert_withdraw(sdp, !sdp->sd_log_num_rg); @@ -838,7 +824,7 @@ void gfs2_log_shutdown(struct gfs2_sbd *sdp) log_write_header(sdp, GFS2_LOG_HEAD_UNMOUNT, (sdp->sd_log_tail == current_tail(sdp)) ? 0 : PULL); - gfs2_assert_warn(sdp, sdp->sd_log_blks_free == sdp->sd_jdesc->jd_blocks); + gfs2_assert_warn(sdp, atomic_read(&sdp->sd_log_blks_free) == sdp->sd_jdesc->jd_blocks); gfs2_assert_warn(sdp, sdp->sd_log_head == sdp->sd_log_tail); gfs2_assert_warn(sdp, list_empty(&sdp->sd_ail2_list)); @@ -866,3 +852,42 @@ void gfs2_meta_syncfs(struct gfs2_sbd *sdp) } } + +/** + * gfs2_logd - Update log tail as Active Items get flushed to in-place blocks + * @sdp: Pointer to GFS2 superblock + * + * Also, periodically check to make sure that we're using the most recent + * journal index. + */ + +int gfs2_logd(void *data) +{ + struct gfs2_sbd *sdp = data; + unsigned long t; + int need_flush; + + while (!kthread_should_stop()) { + /* Advance the log tail */ + + t = sdp->sd_log_flush_time + + gfs2_tune_get(sdp, gt_log_flush_secs) * HZ; + + gfs2_ail1_empty(sdp, DIO_ALL); + gfs2_log_lock(sdp); + need_flush = sdp->sd_log_num_buf > gfs2_tune_get(sdp, gt_incore_log_blocks); + gfs2_log_unlock(sdp); + if (need_flush || time_after_eq(jiffies, t)) { + gfs2_log_flush(sdp, NULL); + sdp->sd_log_flush_time = jiffies; + } + + t = gfs2_tune_get(sdp, gt_logd_secs) * HZ; + if (freezing(current)) + refrigerator(); + schedule_timeout_interruptible(t); + } + + return 0; +} + diff --git a/fs/gfs2/log.h b/fs/gfs2/log.h index dae28240062..77115281650 100644 --- a/fs/gfs2/log.h +++ b/fs/gfs2/log.h @@ -48,8 +48,6 @@ static inline void gfs2_log_pointers_init(struct gfs2_sbd *sdp, unsigned int gfs2_struct2blk(struct gfs2_sbd *sdp, unsigned int nstruct, unsigned int ssize); -int gfs2_ail1_empty(struct gfs2_sbd *sdp, int flags); - int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks); void gfs2_log_release(struct gfs2_sbd *sdp, unsigned int blks); void gfs2_log_incr_head(struct gfs2_sbd *sdp); @@ -57,11 +55,19 @@ void gfs2_log_incr_head(struct gfs2_sbd *sdp); struct buffer_head *gfs2_log_get_buf(struct gfs2_sbd *sdp); struct buffer_head *gfs2_log_fake_buf(struct gfs2_sbd *sdp, struct buffer_head *real); -void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl); +void __gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl); + +static inline void gfs2_log_flush(struct gfs2_sbd *sbd, struct gfs2_glock *gl) +{ + if (!gl || test_bit(GLF_LFLUSH, &gl->gl_flags)) + __gfs2_log_flush(sbd, gl); +} + void gfs2_log_commit(struct gfs2_sbd *sdp, struct gfs2_trans *trans); -void gfs2_remove_from_ail(struct address_space *mapping, struct gfs2_bufdata *bd); +void gfs2_remove_from_ail(struct gfs2_bufdata *bd); void gfs2_log_shutdown(struct gfs2_sbd *sdp); void gfs2_meta_syncfs(struct gfs2_sbd *sdp); +int gfs2_logd(void *data); #endif /* __LOG_DOT_H__ */ diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c index 6c27cea761c..fae59d69d01 100644 --- a/fs/gfs2/lops.c +++ b/fs/gfs2/lops.c @@ -87,6 +87,7 @@ static void gfs2_unpin(struct gfs2_sbd *sdp, struct buffer_head *bh, } bd->bd_ail = ai; list_add(&bd->bd_ail_st_list, &ai->ai_ail1_list); + clear_bit(GLF_LFLUSH, &bd->bd_gl->gl_flags); gfs2_log_unlock(sdp); unlock_buffer(bh); } @@ -124,49 +125,6 @@ static struct buffer_head *gfs2_get_log_desc(struct gfs2_sbd *sdp, u32 ld_type) return bh; } -static void __glock_lo_add(struct gfs2_sbd *sdp, struct gfs2_log_element *le) -{ - struct gfs2_glock *gl; - struct gfs2_trans *tr = current->journal_info; - - tr->tr_touched = 1; - - gl = container_of(le, struct gfs2_glock, gl_le); - if (gfs2_assert_withdraw(sdp, gfs2_glock_is_held_excl(gl))) - return; - - if (!list_empty(&le->le_list)) - return; - - gfs2_glock_hold(gl); - set_bit(GLF_DIRTY, &gl->gl_flags); - sdp->sd_log_num_gl++; - list_add(&le->le_list, &sdp->sd_log_le_gl); -} - -static void glock_lo_add(struct gfs2_sbd *sdp, struct gfs2_log_element *le) -{ - gfs2_log_lock(sdp); - __glock_lo_add(sdp, le); - gfs2_log_unlock(sdp); -} - -static void glock_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_ail *ai) -{ - struct list_head *head = &sdp->sd_log_le_gl; - struct gfs2_glock *gl; - - while (!list_empty(head)) { - gl = list_entry(head->next, struct gfs2_glock, gl_le.le_list); - list_del_init(&gl->gl_le.le_list); - sdp->sd_log_num_gl--; - - gfs2_assert_withdraw(sdp, gfs2_glock_is_held_excl(gl)); - gfs2_glock_put(gl); - } - gfs2_assert_warn(sdp, !sdp->sd_log_num_gl); -} - static void buf_lo_add(struct gfs2_sbd *sdp, struct gfs2_log_element *le) { struct gfs2_bufdata *bd = container_of(le, struct gfs2_bufdata, bd_le); @@ -182,7 +140,8 @@ static void buf_lo_add(struct gfs2_sbd *sdp, struct gfs2_log_element *le) list_add(&bd->bd_list_tr, &tr->tr_list_buf); if (!list_empty(&le->le_list)) goto out; - __glock_lo_add(sdp, &bd->bd_gl->gl_le); + set_bit(GLF_LFLUSH, &bd->bd_gl->gl_flags); + set_bit(GLF_DIRTY, &bd->bd_gl->gl_flags); gfs2_meta_check(sdp, bd->bd_bh); gfs2_pin(sdp, bd->bd_bh); sdp->sd_log_num_buf++; @@ -556,17 +515,20 @@ static void databuf_lo_add(struct gfs2_sbd *sdp, struct gfs2_log_element *le) lock_buffer(bd->bd_bh); gfs2_log_lock(sdp); - if (!list_empty(&bd->bd_list_tr)) - goto out; - tr->tr_touched = 1; - if (gfs2_is_jdata(ip)) { - tr->tr_num_buf++; - list_add(&bd->bd_list_tr, &tr->tr_list_buf); + if (tr) { + if (!list_empty(&bd->bd_list_tr)) + goto out; + tr->tr_touched = 1; + if (gfs2_is_jdata(ip)) { + tr->tr_num_buf++; + list_add(&bd->bd_list_tr, &tr->tr_list_buf); + } } if (!list_empty(&le->le_list)) goto out; - __glock_lo_add(sdp, &bd->bd_gl->gl_le); + set_bit(GLF_LFLUSH, &bd->bd_gl->gl_flags); + set_bit(GLF_DIRTY, &bd->bd_gl->gl_flags); if (gfs2_is_jdata(ip)) { gfs2_pin(sdp, bd->bd_bh); tr->tr_num_databuf_new++; @@ -773,12 +735,6 @@ static void databuf_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_ail *ai) } -const struct gfs2_log_operations gfs2_glock_lops = { - .lo_add = glock_lo_add, - .lo_after_commit = glock_lo_after_commit, - .lo_name = "glock", -}; - const struct gfs2_log_operations gfs2_buf_lops = { .lo_add = buf_lo_add, .lo_incore_commit = buf_lo_incore_commit, @@ -816,7 +772,6 @@ const struct gfs2_log_operations gfs2_databuf_lops = { }; const struct gfs2_log_operations *gfs2_log_ops[] = { - &gfs2_glock_lops, &gfs2_databuf_lops, &gfs2_buf_lops, &gfs2_rg_lops, diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c index 7ecfe0d3a49..9c7765c12d6 100644 --- a/fs/gfs2/main.c +++ b/fs/gfs2/main.c @@ -29,9 +29,8 @@ static void gfs2_init_inode_once(struct kmem_cache *cachep, void *foo) struct gfs2_inode *ip = foo; inode_init_once(&ip->i_inode); - spin_lock_init(&ip->i_spin); init_rwsem(&ip->i_rw_mutex); - memset(ip->i_cache, 0, sizeof(ip->i_cache)); + ip->i_alloc = NULL; } static void gfs2_init_glock_once(struct kmem_cache *cachep, void *foo) diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c index 4da423985e4..85aea27b4a8 100644 --- a/fs/gfs2/meta_io.c +++ b/fs/gfs2/meta_io.c @@ -50,6 +50,7 @@ static int gfs2_aspace_writepage(struct page *page, static const struct address_space_operations aspace_aops = { .writepage = gfs2_aspace_writepage, .releasepage = gfs2_releasepage, + .sync_page = block_sync_page, }; /** @@ -221,13 +222,14 @@ int gfs2_meta_read(struct gfs2_glock *gl, u64 blkno, int flags, struct buffer_head **bhp) { *bhp = getbuf(gl, blkno, CREATE); - if (!buffer_uptodate(*bhp)) + if (!buffer_uptodate(*bhp)) { ll_rw_block(READ_META, 1, bhp); - if (flags & DIO_WAIT) { - int error = gfs2_meta_wait(gl->gl_sbd, *bhp); - if (error) { - brelse(*bhp); - return error; + if (flags & DIO_WAIT) { + int error = gfs2_meta_wait(gl->gl_sbd, *bhp); + if (error) { + brelse(*bhp); + return error; + } } } @@ -282,7 +284,7 @@ void gfs2_attach_bufdata(struct gfs2_glock *gl, struct buffer_head *bh, return; } - bd = kmem_cache_zalloc(gfs2_bufdata_cachep, GFP_NOFS | __GFP_NOFAIL), + bd = kmem_cache_zalloc(gfs2_bufdata_cachep, GFP_NOFS | __GFP_NOFAIL); bd->bd_bh = bh; bd->bd_gl = gl; @@ -317,7 +319,7 @@ void gfs2_remove_from_journal(struct buffer_head *bh, struct gfs2_trans *tr, int } if (bd) { if (bd->bd_ail) { - gfs2_remove_from_ail(NULL, bd); + gfs2_remove_from_ail(bd); bh->b_private = NULL; bd->bd_bh = NULL; bd->bd_blkno = bh->b_blocknr; @@ -358,32 +360,6 @@ void gfs2_meta_wipe(struct gfs2_inode *ip, u64 bstart, u32 blen) } /** - * gfs2_meta_cache_flush - get rid of any references on buffers for this inode - * @ip: The GFS2 inode - * - * This releases buffers that are in the most-recently-used array of - * blocks used for indirect block addressing for this inode. - */ - -void gfs2_meta_cache_flush(struct gfs2_inode *ip) -{ - struct buffer_head **bh_slot; - unsigned int x; - - spin_lock(&ip->i_spin); - - for (x = 0; x < GFS2_MAX_META_HEIGHT; x++) { - bh_slot = &ip->i_cache[x]; - if (*bh_slot) { - brelse(*bh_slot); - *bh_slot = NULL; - } - } - - spin_unlock(&ip->i_spin); -} - -/** * gfs2_meta_indirect_buffer - Get a metadata buffer * @ip: The GFS2 inode * @height: The level of this buf in the metadata (indir addr) tree (if any) @@ -391,8 +367,6 @@ void gfs2_meta_cache_flush(struct gfs2_inode *ip) * @new: Non-zero if we may create a new buffer * @bhp: the buffer is returned here * - * Try to use the gfs2_inode's MRU metadata tree cache. - * * Returns: errno */ @@ -401,58 +375,25 @@ int gfs2_meta_indirect_buffer(struct gfs2_inode *ip, int height, u64 num, { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct gfs2_glock *gl = ip->i_gl; - struct buffer_head *bh = NULL, **bh_slot = ip->i_cache + height; - int in_cache = 0; - - BUG_ON(!gl); - BUG_ON(!sdp); - - spin_lock(&ip->i_spin); - if (*bh_slot && (*bh_slot)->b_blocknr == num) { - bh = *bh_slot; - get_bh(bh); - in_cache = 1; - } - spin_unlock(&ip->i_spin); - - if (!bh) - bh = getbuf(gl, num, CREATE); - - if (!bh) - return -ENOBUFS; + struct buffer_head *bh; + int ret = 0; if (new) { - if (gfs2_assert_warn(sdp, height)) - goto err; - meta_prep_new(bh); + BUG_ON(height == 0); + bh = gfs2_meta_new(gl, num); gfs2_trans_add_bh(ip->i_gl, bh, 1); gfs2_metatype_set(bh, GFS2_METATYPE_IN, GFS2_FORMAT_IN); gfs2_buffer_clear_tail(bh, sizeof(struct gfs2_meta_header)); } else { u32 mtype = height ? GFS2_METATYPE_IN : GFS2_METATYPE_DI; - if (!buffer_uptodate(bh)) { - ll_rw_block(READ_META, 1, &bh); - if (gfs2_meta_wait(sdp, bh)) - goto err; + ret = gfs2_meta_read(gl, num, DIO_WAIT, &bh); + if (ret == 0 && gfs2_metatype_check(sdp, bh, mtype)) { + brelse(bh); + ret = -EIO; } - if (gfs2_metatype_check(sdp, bh, mtype)) - goto err; - } - - if (!in_cache) { - spin_lock(&ip->i_spin); - if (*bh_slot) - brelse(*bh_slot); - *bh_slot = bh; - get_bh(bh); - spin_unlock(&ip->i_spin); } - *bhp = bh; - return 0; -err: - brelse(bh); - return -EIO; + return ret; } /** diff --git a/fs/gfs2/meta_io.h b/fs/gfs2/meta_io.h index b7048222ebb..73e3b1c76fe 100644 --- a/fs/gfs2/meta_io.h +++ b/fs/gfs2/meta_io.h @@ -56,7 +56,6 @@ void gfs2_remove_from_journal(struct buffer_head *bh, struct gfs2_trans *tr, void gfs2_meta_wipe(struct gfs2_inode *ip, u64 bstart, u32 blen); -void gfs2_meta_cache_flush(struct gfs2_inode *ip); int gfs2_meta_indirect_buffer(struct gfs2_inode *ip, int height, u64 num, int new, struct buffer_head **bhp); diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index 9679f8b9870..ac772b6d9db 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -20,6 +20,8 @@ #include <linux/swap.h> #include <linux/gfs2_ondisk.h> #include <linux/lm_interface.h> +#include <linux/backing-dev.h> +#include <linux/pagevec.h> #include "gfs2.h" #include "incore.h" @@ -32,7 +34,6 @@ #include "quota.h" #include "trans.h" #include "rgrp.h" -#include "ops_file.h" #include "super.h" #include "util.h" #include "glops.h" @@ -58,22 +59,6 @@ static void gfs2_page_add_databufs(struct gfs2_inode *ip, struct page *page, } /** - * gfs2_get_block - Fills in a buffer head with details about a block - * @inode: The inode - * @lblock: The block number to look up - * @bh_result: The buffer head to return the result in - * @create: Non-zero if we may add block to the file - * - * Returns: errno - */ - -int gfs2_get_block(struct inode *inode, sector_t lblock, - struct buffer_head *bh_result, int create) -{ - return gfs2_block_map(inode, lblock, create, bh_result); -} - -/** * gfs2_get_block_noalloc - Fills in a buffer head with details about a block * @inode: The inode * @lblock: The block number to look up @@ -88,7 +73,7 @@ static int gfs2_get_block_noalloc(struct inode *inode, sector_t lblock, { int error; - error = gfs2_block_map(inode, lblock, 0, bh_result); + error = gfs2_block_map(inode, lblock, bh_result, 0); if (error) return error; if (!buffer_mapped(bh_result)) @@ -99,20 +84,19 @@ static int gfs2_get_block_noalloc(struct inode *inode, sector_t lblock, static int gfs2_get_block_direct(struct inode *inode, sector_t lblock, struct buffer_head *bh_result, int create) { - return gfs2_block_map(inode, lblock, 0, bh_result); + return gfs2_block_map(inode, lblock, bh_result, 0); } /** - * gfs2_writepage - Write complete page - * @page: Page to write + * gfs2_writepage_common - Common bits of writepage + * @page: The page to be written + * @wbc: The writeback control * - * Returns: errno - * - * Some of this is copied from block_write_full_page() although we still - * call it to do most of the work. + * Returns: 1 if writepage is ok, otherwise an error code or zero if no error. */ -static int gfs2_writepage(struct page *page, struct writeback_control *wbc) +static int gfs2_writepage_common(struct page *page, + struct writeback_control *wbc) { struct inode *inode = page->mapping->host; struct gfs2_inode *ip = GFS2_I(inode); @@ -120,41 +104,133 @@ static int gfs2_writepage(struct page *page, struct writeback_control *wbc) loff_t i_size = i_size_read(inode); pgoff_t end_index = i_size >> PAGE_CACHE_SHIFT; unsigned offset; - int error; - int done_trans = 0; + int ret = -EIO; - if (gfs2_assert_withdraw(sdp, gfs2_glock_is_held_excl(ip->i_gl))) { - unlock_page(page); - return -EIO; - } + if (gfs2_assert_withdraw(sdp, gfs2_glock_is_held_excl(ip->i_gl))) + goto out; + ret = 0; if (current->journal_info) - goto out_ignore; - + goto redirty; /* Is the page fully outside i_size? (truncate in progress) */ - offset = i_size & (PAGE_CACHE_SIZE-1); + offset = i_size & (PAGE_CACHE_SIZE-1); if (page->index > end_index || (page->index == end_index && !offset)) { page->mapping->a_ops->invalidatepage(page, 0); - unlock_page(page); - return 0; /* don't care */ + goto out; + } + return 1; +redirty: + redirty_page_for_writepage(wbc, page); +out: + unlock_page(page); + return 0; +} + +/** + * gfs2_writeback_writepage - Write page for writeback mappings + * @page: The page + * @wbc: The writeback control + * + */ + +static int gfs2_writeback_writepage(struct page *page, + struct writeback_control *wbc) +{ + int ret; + + ret = gfs2_writepage_common(page, wbc); + if (ret <= 0) + return ret; + + ret = mpage_writepage(page, gfs2_get_block_noalloc, wbc); + if (ret == -EAGAIN) + ret = block_write_full_page(page, gfs2_get_block_noalloc, wbc); + return ret; +} + +/** + * gfs2_ordered_writepage - Write page for ordered data files + * @page: The page to write + * @wbc: The writeback control + * + */ + +static int gfs2_ordered_writepage(struct page *page, + struct writeback_control *wbc) +{ + struct inode *inode = page->mapping->host; + struct gfs2_inode *ip = GFS2_I(inode); + int ret; + + ret = gfs2_writepage_common(page, wbc); + if (ret <= 0) + return ret; + + if (!page_has_buffers(page)) { + create_empty_buffers(page, inode->i_sb->s_blocksize, + (1 << BH_Dirty)|(1 << BH_Uptodate)); } + gfs2_page_add_databufs(ip, page, 0, inode->i_sb->s_blocksize-1); + return block_write_full_page(page, gfs2_get_block_noalloc, wbc); +} - if ((sdp->sd_args.ar_data == GFS2_DATA_ORDERED || gfs2_is_jdata(ip)) && - PageChecked(page)) { +/** + * __gfs2_jdata_writepage - The core of jdata writepage + * @page: The page to write + * @wbc: The writeback control + * + * This is shared between writepage and writepages and implements the + * core of the writepage operation. If a transaction is required then + * PageChecked will have been set and the transaction will have + * already been started before this is called. + */ + +static int __gfs2_jdata_writepage(struct page *page, struct writeback_control *wbc) +{ + struct inode *inode = page->mapping->host; + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_sbd *sdp = GFS2_SB(inode); + + if (PageChecked(page)) { ClearPageChecked(page); - error = gfs2_trans_begin(sdp, RES_DINODE + 1, 0); - if (error) - goto out_ignore; if (!page_has_buffers(page)) { create_empty_buffers(page, inode->i_sb->s_blocksize, (1 << BH_Dirty)|(1 << BH_Uptodate)); } gfs2_page_add_databufs(ip, page, 0, sdp->sd_vfs->s_blocksize-1); + } + return block_write_full_page(page, gfs2_get_block_noalloc, wbc); +} + +/** + * gfs2_jdata_writepage - Write complete page + * @page: Page to write + * + * Returns: errno + * + */ + +static int gfs2_jdata_writepage(struct page *page, struct writeback_control *wbc) +{ + struct inode *inode = page->mapping->host; + struct gfs2_sbd *sdp = GFS2_SB(inode); + int error; + int done_trans = 0; + + error = gfs2_writepage_common(page, wbc); + if (error <= 0) + return error; + + if (PageChecked(page)) { + if (wbc->sync_mode != WB_SYNC_ALL) + goto out_ignore; + error = gfs2_trans_begin(sdp, RES_DINODE + 1, 0); + if (error) + goto out_ignore; done_trans = 1; } - error = block_write_full_page(page, gfs2_get_block_noalloc, wbc); + error = __gfs2_jdata_writepage(page, wbc); if (done_trans) gfs2_trans_end(sdp); - gfs2_meta_cache_flush(ip); return error; out_ignore: @@ -164,29 +240,190 @@ out_ignore: } /** - * gfs2_writepages - Write a bunch of dirty pages back to disk + * gfs2_writeback_writepages - Write a bunch of dirty pages back to disk * @mapping: The mapping to write * @wbc: Write-back control * - * For journaled files and/or ordered writes this just falls back to the - * kernel's default writepages path for now. We will probably want to change - * that eventually (i.e. when we look at allocate on flush). - * - * For the data=writeback case though we can already ignore buffer heads + * For the data=writeback case we can already ignore buffer heads * and write whole extents at once. This is a big reduction in the * number of I/O requests we send and the bmap calls we make in this case. */ -static int gfs2_writepages(struct address_space *mapping, - struct writeback_control *wbc) +static int gfs2_writeback_writepages(struct address_space *mapping, + struct writeback_control *wbc) +{ + return mpage_writepages(mapping, wbc, gfs2_get_block_noalloc); +} + +/** + * gfs2_write_jdata_pagevec - Write back a pagevec's worth of pages + * @mapping: The mapping + * @wbc: The writeback control + * @writepage: The writepage function to call for each page + * @pvec: The vector of pages + * @nr_pages: The number of pages to write + * + * Returns: non-zero if loop should terminate, zero otherwise + */ + +static int gfs2_write_jdata_pagevec(struct address_space *mapping, + struct writeback_control *wbc, + struct pagevec *pvec, + int nr_pages, pgoff_t end) { struct inode *inode = mapping->host; - struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); + loff_t i_size = i_size_read(inode); + pgoff_t end_index = i_size >> PAGE_CACHE_SHIFT; + unsigned offset = i_size & (PAGE_CACHE_SIZE-1); + unsigned nrblocks = nr_pages * (PAGE_CACHE_SIZE/inode->i_sb->s_blocksize); + struct backing_dev_info *bdi = mapping->backing_dev_info; + int i; + int ret; + + ret = gfs2_trans_begin(sdp, nrblocks, 0); + if (ret < 0) + return ret; + + for(i = 0; i < nr_pages; i++) { + struct page *page = pvec->pages[i]; + + lock_page(page); + + if (unlikely(page->mapping != mapping)) { + unlock_page(page); + continue; + } + + if (!wbc->range_cyclic && page->index > end) { + ret = 1; + unlock_page(page); + continue; + } + + if (wbc->sync_mode != WB_SYNC_NONE) + wait_on_page_writeback(page); + + if (PageWriteback(page) || + !clear_page_dirty_for_io(page)) { + unlock_page(page); + continue; + } + + /* Is the page fully outside i_size? (truncate in progress) */ + if (page->index > end_index || (page->index == end_index && !offset)) { + page->mapping->a_ops->invalidatepage(page, 0); + unlock_page(page); + continue; + } + + ret = __gfs2_jdata_writepage(page, wbc); + + if (ret || (--(wbc->nr_to_write) <= 0)) + ret = 1; + if (wbc->nonblocking && bdi_write_congested(bdi)) { + wbc->encountered_congestion = 1; + ret = 1; + } + + } + gfs2_trans_end(sdp); + return ret; +} + +/** + * gfs2_write_cache_jdata - Like write_cache_pages but different + * @mapping: The mapping to write + * @wbc: The writeback control + * @writepage: The writepage function to call + * @data: The data to pass to writepage + * + * The reason that we use our own function here is that we need to + * start transactions before we grab page locks. This allows us + * to get the ordering right. + */ + +static int gfs2_write_cache_jdata(struct address_space *mapping, + struct writeback_control *wbc) +{ + struct backing_dev_info *bdi = mapping->backing_dev_info; + int ret = 0; + int done = 0; + struct pagevec pvec; + int nr_pages; + pgoff_t index; + pgoff_t end; + int scanned = 0; + int range_whole = 0; + + if (wbc->nonblocking && bdi_write_congested(bdi)) { + wbc->encountered_congestion = 1; + return 0; + } + + pagevec_init(&pvec, 0); + if (wbc->range_cyclic) { + index = mapping->writeback_index; /* Start from prev offset */ + end = -1; + } else { + index = wbc->range_start >> PAGE_CACHE_SHIFT; + end = wbc->range_end >> PAGE_CACHE_SHIFT; + if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX) + range_whole = 1; + scanned = 1; + } - if (sdp->sd_args.ar_data == GFS2_DATA_WRITEBACK && !gfs2_is_jdata(ip)) - return mpage_writepages(mapping, wbc, gfs2_get_block_noalloc); +retry: + while (!done && (index <= end) && + (nr_pages = pagevec_lookup_tag(&pvec, mapping, &index, + PAGECACHE_TAG_DIRTY, + min(end - index, (pgoff_t)PAGEVEC_SIZE-1) + 1))) { + scanned = 1; + ret = gfs2_write_jdata_pagevec(mapping, wbc, &pvec, nr_pages, end); + if (ret) + done = 1; + if (ret > 0) + ret = 0; + + pagevec_release(&pvec); + cond_resched(); + } + + if (!scanned && !done) { + /* + * We hit the last page and there is more work to be done: wrap + * back to the start of the file + */ + scanned = 1; + index = 0; + goto retry; + } + + if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0)) + mapping->writeback_index = index; + return ret; +} + + +/** + * gfs2_jdata_writepages - Write a bunch of dirty pages back to disk + * @mapping: The mapping to write + * @wbc: The writeback control + * + */ - return generic_writepages(mapping, wbc); +static int gfs2_jdata_writepages(struct address_space *mapping, + struct writeback_control *wbc) +{ + struct gfs2_inode *ip = GFS2_I(mapping->host); + struct gfs2_sbd *sdp = GFS2_SB(mapping->host); + int ret; + + ret = gfs2_write_cache_jdata(mapping, wbc); + if (ret == 0 && wbc->sync_mode == WB_SYNC_ALL) { + gfs2_log_flush(sdp, ip->i_gl); + ret = gfs2_write_cache_jdata(mapping, wbc); + } + return ret; } /** @@ -209,7 +446,7 @@ static int stuffed_readpage(struct gfs2_inode *ip, struct page *page) * so we need to supply one here. It doesn't happen often. */ if (unlikely(page->index)) { - zero_user_page(page, 0, PAGE_CACHE_SIZE, KM_USER0); + zero_user(page, 0, PAGE_CACHE_SIZE); return 0; } @@ -231,62 +468,107 @@ static int stuffed_readpage(struct gfs2_inode *ip, struct page *page) /** - * gfs2_readpage - readpage with locking - * @file: The file to read a page for. N.B. This may be NULL if we are - * reading an internal file. + * __gfs2_readpage - readpage + * @file: The file to read a page for * @page: The page to read * - * Returns: errno + * This is the core of gfs2's readpage. Its used by the internal file + * reading code as in that case we already hold the glock. Also its + * called by gfs2_readpage() once the required lock has been granted. + * */ -static int gfs2_readpage(struct file *file, struct page *page) +static int __gfs2_readpage(void *file, struct page *page) { struct gfs2_inode *ip = GFS2_I(page->mapping->host); struct gfs2_sbd *sdp = GFS2_SB(page->mapping->host); - struct gfs2_file *gf = NULL; - struct gfs2_holder gh; int error; - int do_unlock = 0; - - if (likely(file != &gfs2_internal_file_sentinel)) { - if (file) { - gf = file->private_data; - if (test_bit(GFF_EXLOCK, &gf->f_flags)) - /* gfs2_sharewrite_fault has grabbed the ip->i_gl already */ - goto skip_lock; - } - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME|LM_FLAG_TRY_1CB, &gh); - do_unlock = 1; - error = gfs2_glock_nq_atime(&gh); - if (unlikely(error)) - goto out_unlock; - } -skip_lock: if (gfs2_is_stuffed(ip)) { error = stuffed_readpage(ip, page); unlock_page(page); - } else - error = mpage_readpage(page, gfs2_get_block); + } else { + error = mpage_readpage(page, gfs2_block_map); + } if (unlikely(test_bit(SDF_SHUTDOWN, &sdp->sd_flags))) - error = -EIO; + return -EIO; + + return error; +} + +/** + * gfs2_readpage - read a page of a file + * @file: The file to read + * @page: The page of the file + * + * This deals with the locking required. We use a trylock in order to + * avoid the page lock / glock ordering problems returning AOP_TRUNCATED_PAGE + * in the event that we are unable to get the lock. + */ + +static int gfs2_readpage(struct file *file, struct page *page) +{ + struct gfs2_inode *ip = GFS2_I(page->mapping->host); + struct gfs2_holder gh; + int error; - if (do_unlock) { - gfs2_glock_dq_m(1, &gh); - gfs2_holder_uninit(&gh); + gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME|LM_FLAG_TRY_1CB, &gh); + error = gfs2_glock_nq_atime(&gh); + if (unlikely(error)) { + unlock_page(page); + goto out; } + error = __gfs2_readpage(file, page); + gfs2_glock_dq(&gh); out: - return error; -out_unlock: - unlock_page(page); + gfs2_holder_uninit(&gh); if (error == GLR_TRYFAILED) { - error = AOP_TRUNCATED_PAGE; yield(); + return AOP_TRUNCATED_PAGE; } - if (do_unlock) - gfs2_holder_uninit(&gh); - goto out; + return error; +} + +/** + * gfs2_internal_read - read an internal file + * @ip: The gfs2 inode + * @ra_state: The readahead state (or NULL for no readahead) + * @buf: The buffer to fill + * @pos: The file position + * @size: The amount to read + * + */ + +int gfs2_internal_read(struct gfs2_inode *ip, struct file_ra_state *ra_state, + char *buf, loff_t *pos, unsigned size) +{ + struct address_space *mapping = ip->i_inode.i_mapping; + unsigned long index = *pos / PAGE_CACHE_SIZE; + unsigned offset = *pos & (PAGE_CACHE_SIZE - 1); + unsigned copied = 0; + unsigned amt; + struct page *page; + void *p; + + do { + amt = size - copied; + if (offset + size > PAGE_CACHE_SIZE) + amt = PAGE_CACHE_SIZE - offset; + page = read_cache_page(mapping, index, __gfs2_readpage, NULL); + if (IS_ERR(page)) + return PTR_ERR(page); + p = kmap_atomic(page, KM_USER0); + memcpy(buf + copied, p + offset, amt); + kunmap_atomic(p, KM_USER0); + mark_page_accessed(page); + page_cache_release(page); + copied += amt; + index++; + offset = 0; + } while(copied < size); + (*pos) += size; + return size; } /** @@ -300,10 +582,9 @@ out_unlock: * Any I/O we ignore at this time will be done via readpage later. * 2. We don't handle stuffed files here we let readpage do the honours. * 3. mpage_readpages() does most of the heavy lifting in the common case. - * 4. gfs2_get_block() is relied upon to set BH_Boundary in the right places. - * 5. We use LM_FLAG_TRY_1CB here, effectively we then have lock-ahead as - * well as read-ahead. + * 4. gfs2_block_map() is relied upon to set BH_Boundary in the right places. */ + static int gfs2_readpages(struct file *file, struct address_space *mapping, struct list_head *pages, unsigned nr_pages) { @@ -311,42 +592,20 @@ static int gfs2_readpages(struct file *file, struct address_space *mapping, struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); struct gfs2_holder gh; - int ret = 0; - int do_unlock = 0; + int ret; - if (likely(file != &gfs2_internal_file_sentinel)) { - if (file) { - struct gfs2_file *gf = file->private_data; - if (test_bit(GFF_EXLOCK, &gf->f_flags)) - goto skip_lock; - } - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, - LM_FLAG_TRY_1CB|GL_ATIME, &gh); - do_unlock = 1; - ret = gfs2_glock_nq_atime(&gh); - if (ret == GLR_TRYFAILED) - goto out_noerror; - if (unlikely(ret)) - goto out_unlock; - } -skip_lock: + gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME, &gh); + ret = gfs2_glock_nq_atime(&gh); + if (unlikely(ret)) + goto out_uninit; if (!gfs2_is_stuffed(ip)) - ret = mpage_readpages(mapping, pages, nr_pages, gfs2_get_block); - - if (do_unlock) { - gfs2_glock_dq_m(1, &gh); - gfs2_holder_uninit(&gh); - } -out: + ret = mpage_readpages(mapping, pages, nr_pages, gfs2_block_map); + gfs2_glock_dq(&gh); +out_uninit: + gfs2_holder_uninit(&gh); if (unlikely(test_bit(SDF_SHUTDOWN, &sdp->sd_flags))) ret = -EIO; return ret; -out_noerror: - ret = 0; -out_unlock: - if (do_unlock) - gfs2_holder_uninit(&gh); - goto out; } /** @@ -382,20 +641,11 @@ static int gfs2_write_begin(struct file *file, struct address_space *mapping, if (unlikely(error)) goto out_uninit; - error = -ENOMEM; - page = __grab_cache_page(mapping, index); - *pagep = page; - if (!page) - goto out_unlock; - gfs2_write_calc_reserv(ip, len, &data_blocks, &ind_blocks); - error = gfs2_write_alloc_required(ip, pos, len, &alloc_required); if (error) - goto out_putpage; - + goto out_unlock; - ip->i_alloc.al_requested = 0; if (alloc_required) { al = gfs2_alloc_get(ip); @@ -424,40 +674,47 @@ static int gfs2_write_begin(struct file *file, struct address_space *mapping, if (error) goto out_trans_fail; + error = -ENOMEM; + page = __grab_cache_page(mapping, index); + *pagep = page; + if (unlikely(!page)) + goto out_endtrans; + if (gfs2_is_stuffed(ip)) { + error = 0; if (pos + len > sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode)) { error = gfs2_unstuff_dinode(ip, page); if (error == 0) goto prepare_write; - } else if (!PageUptodate(page)) + } else if (!PageUptodate(page)) { error = stuffed_readpage(ip, page); + } goto out; } prepare_write: - error = block_prepare_write(page, from, to, gfs2_get_block); - + error = block_prepare_write(page, from, to, gfs2_block_map); out: - if (error) { - gfs2_trans_end(sdp); + if (error == 0) + return 0; + + page_cache_release(page); + if (pos + len > ip->i_inode.i_size) + vmtruncate(&ip->i_inode, ip->i_inode.i_size); +out_endtrans: + gfs2_trans_end(sdp); out_trans_fail: - if (alloc_required) { - gfs2_inplace_release(ip); + if (alloc_required) { + gfs2_inplace_release(ip); out_qunlock: - gfs2_quota_unlock(ip); + gfs2_quota_unlock(ip); out_alloc_put: - gfs2_alloc_put(ip); - } -out_putpage: - page_cache_release(page); - if (pos + len > ip->i_inode.i_size) - vmtruncate(&ip->i_inode, ip->i_inode.i_size); + gfs2_alloc_put(ip); + } out_unlock: - gfs2_glock_dq_m(1, &ip->i_gh); + gfs2_glock_dq(&ip->i_gh); out_uninit: - gfs2_holder_uninit(&ip->i_gh); - } - + gfs2_holder_uninit(&ip->i_gh); return error; } @@ -565,7 +822,7 @@ static int gfs2_write_end(struct file *file, struct address_space *mapping, struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); struct buffer_head *dibh; - struct gfs2_alloc *al = &ip->i_alloc; + struct gfs2_alloc *al = ip->i_alloc; struct gfs2_dinode *di; unsigned int from = pos & (PAGE_CACHE_SIZE - 1); unsigned int to = from + len; @@ -585,19 +842,16 @@ static int gfs2_write_end(struct file *file, struct address_space *mapping, if (gfs2_is_stuffed(ip)) return gfs2_stuffed_write_end(inode, dibh, pos, len, copied, page); - if (sdp->sd_args.ar_data == GFS2_DATA_ORDERED || gfs2_is_jdata(ip)) + if (!gfs2_is_writeback(ip)) gfs2_page_add_databufs(ip, page, from, to); ret = generic_write_end(file, mapping, pos, len, copied, page, fsdata); - if (likely(ret >= 0)) { - copied = ret; - if ((pos + copied) > inode->i_size) { - di = (struct gfs2_dinode *)dibh->b_data; - ip->i_di.di_size = inode->i_size; - di->di_size = cpu_to_be64(inode->i_size); - mark_inode_dirty(inode); - } + if (likely(ret >= 0) && (inode->i_size > ip->i_di.di_size)) { + di = (struct gfs2_dinode *)dibh->b_data; + ip->i_di.di_size = inode->i_size; + di->di_size = cpu_to_be64(inode->i_size); + mark_inode_dirty(inode); } if (inode == sdp->sd_rindex) @@ -606,7 +860,7 @@ static int gfs2_write_end(struct file *file, struct address_space *mapping, brelse(dibh); gfs2_trans_end(sdp); failed: - if (al->al_requested) { + if (al) { gfs2_inplace_release(ip); gfs2_quota_unlock(ip); gfs2_alloc_put(ip); @@ -625,11 +879,7 @@ failed: static int gfs2_set_page_dirty(struct page *page) { - struct gfs2_inode *ip = GFS2_I(page->mapping->host); - struct gfs2_sbd *sdp = GFS2_SB(page->mapping->host); - - if (sdp->sd_args.ar_data == GFS2_DATA_ORDERED || gfs2_is_jdata(ip)) - SetPageChecked(page); + SetPageChecked(page); return __set_page_dirty_buffers(page); } @@ -653,7 +903,7 @@ static sector_t gfs2_bmap(struct address_space *mapping, sector_t lblock) return 0; if (!gfs2_is_stuffed(ip)) - dblock = generic_block_bmap(mapping, lblock, gfs2_get_block); + dblock = generic_block_bmap(mapping, lblock, gfs2_block_map); gfs2_glock_dq_uninit(&i_gh); @@ -719,13 +969,9 @@ static int gfs2_ok_for_dio(struct gfs2_inode *ip, int rw, loff_t offset) { /* * Should we return an error here? I can't see that O_DIRECT for - * a journaled file makes any sense. For now we'll silently fall - * back to buffered I/O, likewise we do the same for stuffed - * files since they are (a) small and (b) unaligned. + * a stuffed file makes any sense. For now we'll silently fall + * back to buffered I/O */ - if (gfs2_is_jdata(ip)) - return 0; - if (gfs2_is_stuffed(ip)) return 0; @@ -836,9 +1082,23 @@ cannot_release: return 0; } -const struct address_space_operations gfs2_file_aops = { - .writepage = gfs2_writepage, - .writepages = gfs2_writepages, +static const struct address_space_operations gfs2_writeback_aops = { + .writepage = gfs2_writeback_writepage, + .writepages = gfs2_writeback_writepages, + .readpage = gfs2_readpage, + .readpages = gfs2_readpages, + .sync_page = block_sync_page, + .write_begin = gfs2_write_begin, + .write_end = gfs2_write_end, + .bmap = gfs2_bmap, + .invalidatepage = gfs2_invalidatepage, + .releasepage = gfs2_releasepage, + .direct_IO = gfs2_direct_IO, + .migratepage = buffer_migrate_page, +}; + +static const struct address_space_operations gfs2_ordered_aops = { + .writepage = gfs2_ordered_writepage, .readpage = gfs2_readpage, .readpages = gfs2_readpages, .sync_page = block_sync_page, @@ -849,5 +1109,34 @@ const struct address_space_operations gfs2_file_aops = { .invalidatepage = gfs2_invalidatepage, .releasepage = gfs2_releasepage, .direct_IO = gfs2_direct_IO, + .migratepage = buffer_migrate_page, }; +static const struct address_space_operations gfs2_jdata_aops = { + .writepage = gfs2_jdata_writepage, + .writepages = gfs2_jdata_writepages, + .readpage = gfs2_readpage, + .readpages = gfs2_readpages, + .sync_page = block_sync_page, + .write_begin = gfs2_write_begin, + .write_end = gfs2_write_end, + .set_page_dirty = gfs2_set_page_dirty, + .bmap = gfs2_bmap, + .invalidatepage = gfs2_invalidatepage, + .releasepage = gfs2_releasepage, +}; + +void gfs2_set_aops(struct inode *inode) +{ + struct gfs2_inode *ip = GFS2_I(inode); + + if (gfs2_is_writeback(ip)) + inode->i_mapping->a_ops = &gfs2_writeback_aops; + else if (gfs2_is_ordered(ip)) + inode->i_mapping->a_ops = &gfs2_ordered_aops; + else if (gfs2_is_jdata(ip)) + inode->i_mapping->a_ops = &gfs2_jdata_aops; + else + BUG(); +} + diff --git a/fs/gfs2/ops_address.h b/fs/gfs2/ops_address.h index fa1b5b3d28b..5da21285bba 100644 --- a/fs/gfs2/ops_address.h +++ b/fs/gfs2/ops_address.h @@ -14,9 +14,10 @@ #include <linux/buffer_head.h> #include <linux/mm.h> -extern const struct address_space_operations gfs2_file_aops; -extern int gfs2_get_block(struct inode *inode, sector_t lblock, - struct buffer_head *bh_result, int create); extern int gfs2_releasepage(struct page *page, gfp_t gfp_mask); +extern int gfs2_internal_read(struct gfs2_inode *ip, + struct file_ra_state *ra_state, + char *buf, loff_t *pos, unsigned size); +extern void gfs2_set_aops(struct inode *inode); #endif /* __OPS_ADDRESS_DOT_H__ */ diff --git a/fs/gfs2/ops_export.c b/fs/gfs2/ops_export.c index b9da62348a8..334c7f85351 100644 --- a/fs/gfs2/ops_export.c +++ b/fs/gfs2/ops_export.c @@ -143,7 +143,7 @@ static struct dentry *gfs2_get_parent(struct dentry *child) * have to return that as a(n invalid) pointer to dentry. */ if (IS_ERR(inode)) - return ERR_PTR(PTR_ERR(inode)); + return ERR_CAST(inode); dentry = d_alloc_anon(inode); if (!dentry) { diff --git a/fs/gfs2/ops_file.c b/fs/gfs2/ops_file.c index bb11fd6752d..f4842f2548c 100644 --- a/fs/gfs2/ops_file.c +++ b/fs/gfs2/ops_file.c @@ -33,57 +33,12 @@ #include "lm.h" #include "log.h" #include "meta_io.h" -#include "ops_file.h" -#include "ops_vm.h" #include "quota.h" #include "rgrp.h" #include "trans.h" #include "util.h" #include "eaops.h" - -/* - * Most fields left uninitialised to catch anybody who tries to - * use them. f_flags set to prevent file_accessed() from touching - * any other part of this. Its use is purely as a flag so that we - * know (in readpage()) whether or not do to locking. - */ -struct file gfs2_internal_file_sentinel = { - .f_flags = O_NOATIME|O_RDONLY, -}; - -static int gfs2_read_actor(read_descriptor_t *desc, struct page *page, - unsigned long offset, unsigned long size) -{ - char *kaddr; - unsigned long count = desc->count; - - if (size > count) - size = count; - - kaddr = kmap(page); - memcpy(desc->arg.data, kaddr + offset, size); - kunmap(page); - - desc->count = count - size; - desc->written += size; - desc->arg.buf += size; - return size; -} - -int gfs2_internal_read(struct gfs2_inode *ip, struct file_ra_state *ra_state, - char *buf, loff_t *pos, unsigned size) -{ - struct inode *inode = &ip->i_inode; - read_descriptor_t desc; - desc.written = 0; - desc.arg.data = buf; - desc.count = size; - desc.error = 0; - do_generic_mapping_read(inode->i_mapping, ra_state, - &gfs2_internal_file_sentinel, pos, &desc, - gfs2_read_actor); - return desc.written ? desc.written : desc.error; -} +#include "ops_address.h" /** * gfs2_llseek - seek to a location in a file @@ -214,7 +169,7 @@ static int gfs2_get_flags(struct file *filp, u32 __user *ptr) if (put_user(fsflags, ptr)) error = -EFAULT; - gfs2_glock_dq_m(1, &gh); + gfs2_glock_dq(&gh); gfs2_holder_uninit(&gh); return error; } @@ -291,7 +246,16 @@ static int do_gfs2_set_flags(struct file *filp, u32 reqflags, u32 mask) if (error) goto out; } - + if ((flags ^ new_flags) & GFS2_DIF_JDATA) { + if (flags & GFS2_DIF_JDATA) + gfs2_log_flush(sdp, ip->i_gl); + error = filemap_fdatawrite(inode->i_mapping); + if (error) + goto out; + error = filemap_fdatawait(inode->i_mapping); + if (error) + goto out; + } error = gfs2_trans_begin(sdp, RES_DINODE, 0); if (error) goto out; @@ -303,6 +267,7 @@ static int do_gfs2_set_flags(struct file *filp, u32 reqflags, u32 mask) gfs2_dinode_out(ip, bh->b_data); brelse(bh); gfs2_set_inode_flags(inode); + gfs2_set_aops(inode); out_trans_end: gfs2_trans_end(sdp); out: @@ -338,6 +303,128 @@ static long gfs2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) return -ENOTTY; } +/** + * gfs2_allocate_page_backing - Use bmap to allocate blocks + * @page: The (locked) page to allocate backing for + * + * We try to allocate all the blocks required for the page in + * one go. This might fail for various reasons, so we keep + * trying until all the blocks to back this page are allocated. + * If some of the blocks are already allocated, thats ok too. + */ + +static int gfs2_allocate_page_backing(struct page *page) +{ + struct inode *inode = page->mapping->host; + struct buffer_head bh; + unsigned long size = PAGE_CACHE_SIZE; + u64 lblock = page->index << (PAGE_CACHE_SHIFT - inode->i_blkbits); + + do { + bh.b_state = 0; + bh.b_size = size; + gfs2_block_map(inode, lblock, &bh, 1); + if (!buffer_mapped(&bh)) + return -EIO; + size -= bh.b_size; + lblock += (bh.b_size >> inode->i_blkbits); + } while(size > 0); + return 0; +} + +/** + * gfs2_page_mkwrite - Make a shared, mmap()ed, page writable + * @vma: The virtual memory area + * @page: The page which is about to become writable + * + * When the page becomes writable, we need to ensure that we have + * blocks allocated on disk to back that page. + */ + +static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct page *page) +{ + struct inode *inode = vma->vm_file->f_path.dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_sbd *sdp = GFS2_SB(inode); + unsigned long last_index; + u64 pos = page->index << (PAGE_CACHE_SIZE - inode->i_blkbits); + unsigned int data_blocks, ind_blocks, rblocks; + int alloc_required = 0; + struct gfs2_holder gh; + struct gfs2_alloc *al; + int ret; + + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_ATIME, &gh); + ret = gfs2_glock_nq_atime(&gh); + if (ret) + goto out; + + set_bit(GIF_SW_PAGED, &ip->i_flags); + gfs2_write_calc_reserv(ip, PAGE_CACHE_SIZE, &data_blocks, &ind_blocks); + ret = gfs2_write_alloc_required(ip, pos, PAGE_CACHE_SIZE, &alloc_required); + if (ret || !alloc_required) + goto out_unlock; + ret = -ENOMEM; + al = gfs2_alloc_get(ip); + if (al == NULL) + goto out_unlock; + + ret = gfs2_quota_lock(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE); + if (ret) + goto out_alloc_put; + ret = gfs2_quota_check(ip, ip->i_inode.i_uid, ip->i_inode.i_gid); + if (ret) + goto out_quota_unlock; + al->al_requested = data_blocks + ind_blocks; + ret = gfs2_inplace_reserve(ip); + if (ret) + goto out_quota_unlock; + + rblocks = RES_DINODE + ind_blocks; + if (gfs2_is_jdata(ip)) + rblocks += data_blocks ? data_blocks : 1; + if (ind_blocks || data_blocks) + rblocks += RES_STATFS + RES_QUOTA; + ret = gfs2_trans_begin(sdp, rblocks, 0); + if (ret) + goto out_trans_fail; + + lock_page(page); + ret = -EINVAL; + last_index = ip->i_inode.i_size >> PAGE_CACHE_SHIFT; + if (page->index > last_index) + goto out_unlock_page; + ret = 0; + if (!PageUptodate(page) || page->mapping != ip->i_inode.i_mapping) + goto out_unlock_page; + if (gfs2_is_stuffed(ip)) { + ret = gfs2_unstuff_dinode(ip, page); + if (ret) + goto out_unlock_page; + } + ret = gfs2_allocate_page_backing(page); + +out_unlock_page: + unlock_page(page); + gfs2_trans_end(sdp); +out_trans_fail: + gfs2_inplace_release(ip); +out_quota_unlock: + gfs2_quota_unlock(ip); +out_alloc_put: + gfs2_alloc_put(ip); +out_unlock: + gfs2_glock_dq(&gh); +out: + gfs2_holder_uninit(&gh); + return ret; +} + +static struct vm_operations_struct gfs2_vm_ops = { + .fault = filemap_fault, + .page_mkwrite = gfs2_page_mkwrite, +}; + /** * gfs2_mmap - @@ -360,14 +447,7 @@ static int gfs2_mmap(struct file *file, struct vm_area_struct *vma) return error; } - /* This is VM_MAYWRITE instead of VM_WRITE because a call - to mprotect() can turn on VM_WRITE later. */ - - if ((vma->vm_flags & (VM_MAYSHARE | VM_MAYWRITE)) == - (VM_MAYSHARE | VM_MAYWRITE)) - vma->vm_ops = &gfs2_vm_ops_sharewrite; - else - vma->vm_ops = &gfs2_vm_ops_private; + vma->vm_ops = &gfs2_vm_ops; gfs2_glock_dq_uninit(&i_gh); @@ -538,15 +618,6 @@ static int gfs2_lock(struct file *file, int cmd, struct file_lock *fl) if (__mandatory_lock(&ip->i_inode)) return -ENOLCK; - if (sdp->sd_args.ar_localflocks) { - if (IS_GETLK(cmd)) { - posix_test_lock(file, fl); - return 0; - } else { - return posix_lock_file_wait(file, fl); - } - } - if (cmd == F_CANCELLK) { /* Hack: */ cmd = F_SETLK; @@ -632,16 +703,12 @@ static void do_unflock(struct file *file, struct file_lock *fl) static int gfs2_flock(struct file *file, int cmd, struct file_lock *fl) { struct gfs2_inode *ip = GFS2_I(file->f_mapping->host); - struct gfs2_sbd *sdp = GFS2_SB(file->f_mapping->host); if (!(fl->fl_flags & FL_FLOCK)) return -ENOLCK; if (__mandatory_lock(&ip->i_inode)) return -ENOLCK; - if (sdp->sd_args.ar_localflocks) - return flock_lock_file_wait(file, fl); - if (fl->fl_type == F_UNLCK) { do_unflock(file, fl); return 0; @@ -678,3 +745,27 @@ const struct file_operations gfs2_dir_fops = { .flock = gfs2_flock, }; +const struct file_operations gfs2_file_fops_nolock = { + .llseek = gfs2_llseek, + .read = do_sync_read, + .aio_read = generic_file_aio_read, + .write = do_sync_write, + .aio_write = generic_file_aio_write, + .unlocked_ioctl = gfs2_ioctl, + .mmap = gfs2_mmap, + .open = gfs2_open, + .release = gfs2_close, + .fsync = gfs2_fsync, + .splice_read = generic_file_splice_read, + .splice_write = generic_file_splice_write, + .setlease = gfs2_setlease, +}; + +const struct file_operations gfs2_dir_fops_nolock = { + .readdir = gfs2_readdir, + .unlocked_ioctl = gfs2_ioctl, + .open = gfs2_open, + .release = gfs2_close, + .fsync = gfs2_fsync, +}; + diff --git a/fs/gfs2/ops_file.h b/fs/gfs2/ops_file.h deleted file mode 100644 index 7e5d8ec9c84..00000000000 --- a/fs/gfs2/ops_file.h +++ /dev/null @@ -1,24 +0,0 @@ -/* - * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. - * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved. - * - * This copyrighted material is made available to anyone wishing to use, - * modify, copy, or redistribute it subject to the terms and conditions - * of the GNU General Public License version 2. - */ - -#ifndef __OPS_FILE_DOT_H__ -#define __OPS_FILE_DOT_H__ - -#include <linux/fs.h> -struct gfs2_inode; - -extern struct file gfs2_internal_file_sentinel; -extern int gfs2_internal_read(struct gfs2_inode *ip, - struct file_ra_state *ra_state, - char *buf, loff_t *pos, unsigned size); -extern void gfs2_set_inode_flags(struct inode *inode); -extern const struct file_operations gfs2_file_fops; -extern const struct file_operations gfs2_dir_fops; - -#endif /* __OPS_FILE_DOT_H__ */ diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 17de58e83d9..4bee6aa845e 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -1,6 +1,6 @@ /* * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. - * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved. + * Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved. * * This copyrighted material is made available to anyone wishing to use, * modify, copy, or redistribute it subject to the terms and conditions @@ -21,6 +21,7 @@ #include "gfs2.h" #include "incore.h" +#include "bmap.h" #include "daemon.h" #include "glock.h" #include "glops.h" @@ -59,7 +60,6 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb) mutex_init(&sdp->sd_inum_mutex); spin_lock_init(&sdp->sd_statfs_spin); - mutex_init(&sdp->sd_statfs_mutex); spin_lock_init(&sdp->sd_rindex_spin); mutex_init(&sdp->sd_rindex_mutex); @@ -77,7 +77,6 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb) spin_lock_init(&sdp->sd_log_lock); - INIT_LIST_HEAD(&sdp->sd_log_le_gl); INIT_LIST_HEAD(&sdp->sd_log_le_buf); INIT_LIST_HEAD(&sdp->sd_log_le_revoke); INIT_LIST_HEAD(&sdp->sd_log_le_rg); @@ -303,6 +302,67 @@ out: return error; } +/** + * map_journal_extents - create a reusable "extent" mapping from all logical + * blocks to all physical blocks for the given journal. This will save + * us time when writing journal blocks. Most journals will have only one + * extent that maps all their logical blocks. That's because gfs2.mkfs + * arranges the journal blocks sequentially to maximize performance. + * So the extent would map the first block for the entire file length. + * However, gfs2_jadd can happen while file activity is happening, so + * those journals may not be sequential. Less likely is the case where + * the users created their own journals by mounting the metafs and + * laying it out. But it's still possible. These journals might have + * several extents. + * + * TODO: This should be done in bigger chunks rather than one block at a time, + * but since it's only done at mount time, I'm not worried about the + * time it takes. + */ +static int map_journal_extents(struct gfs2_sbd *sdp) +{ + struct gfs2_jdesc *jd = sdp->sd_jdesc; + unsigned int lb; + u64 db, prev_db; /* logical block, disk block, prev disk block */ + struct gfs2_inode *ip = GFS2_I(jd->jd_inode); + struct gfs2_journal_extent *jext = NULL; + struct buffer_head bh; + int rc = 0; + + prev_db = 0; + + for (lb = 0; lb < ip->i_di.di_size >> sdp->sd_sb.sb_bsize_shift; lb++) { + bh.b_state = 0; + bh.b_blocknr = 0; + bh.b_size = 1 << ip->i_inode.i_blkbits; + rc = gfs2_block_map(jd->jd_inode, lb, &bh, 0); + db = bh.b_blocknr; + if (rc || !db) { + printk(KERN_INFO "GFS2 journal mapping error %d: lb=" + "%u db=%llu\n", rc, lb, (unsigned long long)db); + break; + } + if (!prev_db || db != prev_db + 1) { + jext = kzalloc(sizeof(struct gfs2_journal_extent), + GFP_KERNEL); + if (!jext) { + printk(KERN_INFO "GFS2 error: out of memory " + "mapping journal extents.\n"); + rc = -ENOMEM; + break; + } + jext->dblock = db; + jext->lblock = lb; + jext->blocks = 1; + list_add_tail(&jext->extent_list, &jd->extent_list); + } else { + jext->blocks++; + } + prev_db = db; + } + return rc; +} + static int init_journal(struct gfs2_sbd *sdp, int undo) { struct gfs2_holder ji_gh; @@ -340,7 +400,7 @@ static int init_journal(struct gfs2_sbd *sdp, int undo) if (sdp->sd_args.ar_spectator) { sdp->sd_jdesc = gfs2_jdesc_find(sdp, 0); - sdp->sd_log_blks_free = sdp->sd_jdesc->jd_blocks; + atomic_set(&sdp->sd_log_blks_free, sdp->sd_jdesc->jd_blocks); } else { if (sdp->sd_lockstruct.ls_jid >= gfs2_jindex_size(sdp)) { fs_err(sdp, "can't mount journal #%u\n", @@ -377,7 +437,10 @@ static int init_journal(struct gfs2_sbd *sdp, int undo) sdp->sd_jdesc->jd_jid, error); goto fail_jinode_gh; } - sdp->sd_log_blks_free = sdp->sd_jdesc->jd_blocks; + atomic_set(&sdp->sd_log_blks_free, sdp->sd_jdesc->jd_blocks); + + /* Map the extents for this journal's blocks */ + map_journal_extents(sdp); } if (sdp->sd_lockstruct.ls_first) { @@ -821,12 +884,13 @@ static struct super_block* get_gfs2_sb(const char *dev_name) dev_name); goto out; } - error = vfs_getattr(nd.mnt, nd.dentry, &stat); + error = vfs_getattr(nd.path.mnt, nd.path.dentry, &stat); fstype = get_fs_type("gfs2"); list_for_each_entry(s, &fstype->fs_supers, s_instances) { if ((S_ISBLK(stat.mode) && s->s_dev == stat.rdev) || - (S_ISDIR(stat.mode) && s == nd.dentry->d_inode->i_sb)) { + (S_ISDIR(stat.mode) && + s == nd.path.dentry->d_inode->i_sb)) { sb = s; goto free_nd; } @@ -836,7 +900,7 @@ static struct super_block* get_gfs2_sb(const char *dev_name) "mount point %s\n", dev_name); free_nd: - path_release(&nd); + path_put(&nd.path); out: return sb; } diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index 291f0c7eaa3..e87412902be 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -61,7 +61,7 @@ static int gfs2_create(struct inode *dir, struct dentry *dentry, inode = gfs2_createi(ghs, &dentry->d_name, S_IFREG | mode, 0); if (!IS_ERR(inode)) { gfs2_trans_end(sdp); - if (dip->i_alloc.al_rgd) + if (dip->i_alloc->al_rgd) gfs2_inplace_release(dip); gfs2_quota_unlock(dip); gfs2_alloc_put(dip); @@ -111,10 +111,20 @@ static struct dentry *gfs2_lookup(struct inode *dir, struct dentry *dentry, inode = gfs2_lookupi(dir, &dentry->d_name, 0, nd); if (inode && IS_ERR(inode)) - return ERR_PTR(PTR_ERR(inode)); - - if (inode) + return ERR_CAST(inode); + + if (inode) { + struct gfs2_glock *gl = GFS2_I(inode)->i_gl; + struct gfs2_holder gh; + int error; + error = gfs2_glock_nq_init(gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); + if (error) { + iput(inode); + return ERR_PTR(error); + } + gfs2_glock_dq_uninit(&gh); return d_splice_alias(inode, dentry); + } d_add(dentry, inode); return NULL; @@ -366,7 +376,7 @@ static int gfs2_symlink(struct inode *dir, struct dentry *dentry, } gfs2_trans_end(sdp); - if (dip->i_alloc.al_rgd) + if (dip->i_alloc->al_rgd) gfs2_inplace_release(dip); gfs2_quota_unlock(dip); gfs2_alloc_put(dip); @@ -442,7 +452,7 @@ static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode) gfs2_assert_withdraw(sdp, !error); /* dip already pinned */ gfs2_trans_end(sdp); - if (dip->i_alloc.al_rgd) + if (dip->i_alloc->al_rgd) gfs2_inplace_release(dip); gfs2_quota_unlock(dip); gfs2_alloc_put(dip); @@ -548,7 +558,7 @@ static int gfs2_mknod(struct inode *dir, struct dentry *dentry, int mode, } gfs2_trans_end(sdp); - if (dip->i_alloc.al_rgd) + if (dip->i_alloc->al_rgd) gfs2_inplace_release(dip); gfs2_quota_unlock(dip); gfs2_alloc_put(dip); diff --git a/fs/gfs2/ops_inode.h b/fs/gfs2/ops_inode.h index 34f0caac1a0..fd8cee231e1 100644 --- a/fs/gfs2/ops_inode.h +++ b/fs/gfs2/ops_inode.h @@ -16,5 +16,11 @@ extern const struct inode_operations gfs2_file_iops; extern const struct inode_operations gfs2_dir_iops; extern const struct inode_operations gfs2_symlink_iops; extern const struct inode_operations gfs2_dev_iops; +extern const struct file_operations gfs2_file_fops; +extern const struct file_operations gfs2_dir_fops; +extern const struct file_operations gfs2_file_fops_nolock; +extern const struct file_operations gfs2_dir_fops_nolock; + +extern void gfs2_set_inode_flags(struct inode *inode); #endif /* __OPS_INODE_DOT_H__ */ diff --git a/fs/gfs2/ops_super.c b/fs/gfs2/ops_super.c index 950f31460e8..5e524217944 100644 --- a/fs/gfs2/ops_super.c +++ b/fs/gfs2/ops_super.c @@ -487,7 +487,6 @@ static struct inode *gfs2_alloc_inode(struct super_block *sb) if (ip) { ip->i_flags = 0; ip->i_gl = NULL; - ip->i_last_pfault = jiffies; } return &ip->i_inode; } diff --git a/fs/gfs2/ops_vm.c b/fs/gfs2/ops_vm.c deleted file mode 100644 index 927d739d468..00000000000 --- a/fs/gfs2/ops_vm.c +++ /dev/null @@ -1,169 +0,0 @@ -/* - * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. - * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved. - * - * This copyrighted material is made available to anyone wishing to use, - * modify, copy, or redistribute it subject to the terms and conditions - * of the GNU General Public License version 2. - */ - -#include <linux/slab.h> -#include <linux/spinlock.h> -#include <linux/completion.h> -#include <linux/buffer_head.h> -#include <linux/mm.h> -#include <linux/pagemap.h> -#include <linux/gfs2_ondisk.h> -#include <linux/lm_interface.h> - -#include "gfs2.h" -#include "incore.h" -#include "bmap.h" -#include "glock.h" -#include "inode.h" -#include "ops_vm.h" -#include "quota.h" -#include "rgrp.h" -#include "trans.h" -#include "util.h" - -static int gfs2_private_fault(struct vm_area_struct *vma, struct vm_fault *vmf) -{ - struct gfs2_inode *ip = GFS2_I(vma->vm_file->f_mapping->host); - - set_bit(GIF_PAGED, &ip->i_flags); - return filemap_fault(vma, vmf); -} - -static int alloc_page_backing(struct gfs2_inode *ip, struct page *page) -{ - struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - unsigned long index = page->index; - u64 lblock = index << (PAGE_CACHE_SHIFT - - sdp->sd_sb.sb_bsize_shift); - unsigned int blocks = PAGE_CACHE_SIZE >> sdp->sd_sb.sb_bsize_shift; - struct gfs2_alloc *al; - unsigned int data_blocks, ind_blocks; - unsigned int x; - int error; - - al = gfs2_alloc_get(ip); - - error = gfs2_quota_lock(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE); - if (error) - goto out; - - error = gfs2_quota_check(ip, ip->i_inode.i_uid, ip->i_inode.i_gid); - if (error) - goto out_gunlock_q; - - gfs2_write_calc_reserv(ip, PAGE_CACHE_SIZE, &data_blocks, &ind_blocks); - - al->al_requested = data_blocks + ind_blocks; - - error = gfs2_inplace_reserve(ip); - if (error) - goto out_gunlock_q; - - error = gfs2_trans_begin(sdp, al->al_rgd->rd_length + - ind_blocks + RES_DINODE + - RES_STATFS + RES_QUOTA, 0); - if (error) - goto out_ipres; - - if (gfs2_is_stuffed(ip)) { - error = gfs2_unstuff_dinode(ip, NULL); - if (error) - goto out_trans; - } - - for (x = 0; x < blocks; ) { - u64 dblock; - unsigned int extlen; - int new = 1; - - error = gfs2_extent_map(&ip->i_inode, lblock, &new, &dblock, &extlen); - if (error) - goto out_trans; - - lblock += extlen; - x += extlen; - } - - gfs2_assert_warn(sdp, al->al_alloced); - -out_trans: - gfs2_trans_end(sdp); -out_ipres: - gfs2_inplace_release(ip); -out_gunlock_q: - gfs2_quota_unlock(ip); -out: - gfs2_alloc_put(ip); - return error; -} - -static int gfs2_sharewrite_fault(struct vm_area_struct *vma, - struct vm_fault *vmf) -{ - struct file *file = vma->vm_file; - struct gfs2_file *gf = file->private_data; - struct gfs2_inode *ip = GFS2_I(file->f_mapping->host); - struct gfs2_holder i_gh; - int alloc_required; - int error; - int ret = 0; - - error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &i_gh); - if (error) - goto out; - - set_bit(GIF_PAGED, &ip->i_flags); - set_bit(GIF_SW_PAGED, &ip->i_flags); - - error = gfs2_write_alloc_required(ip, - (u64)vmf->pgoff << PAGE_CACHE_SHIFT, - PAGE_CACHE_SIZE, &alloc_required); - if (error) { - ret = VM_FAULT_OOM; /* XXX: are these right? */ - goto out_unlock; - } - - set_bit(GFF_EXLOCK, &gf->f_flags); - ret = filemap_fault(vma, vmf); - clear_bit(GFF_EXLOCK, &gf->f_flags); - if (ret & VM_FAULT_ERROR) - goto out_unlock; - - if (alloc_required) { - /* XXX: do we need to drop page lock around alloc_page_backing?*/ - error = alloc_page_backing(ip, vmf->page); - if (error) { - /* - * VM_FAULT_LOCKED should always be the case for - * filemap_fault, but it may not be in a future - * implementation. - */ - if (ret & VM_FAULT_LOCKED) - unlock_page(vmf->page); - page_cache_release(vmf->page); - ret = VM_FAULT_OOM; - goto out_unlock; - } - set_page_dirty(vmf->page); - } - -out_unlock: - gfs2_glock_dq_uninit(&i_gh); -out: - return ret; -} - -struct vm_operations_struct gfs2_vm_ops_private = { - .fault = gfs2_private_fault, -}; - -struct vm_operations_struct gfs2_vm_ops_sharewrite = { - .fault = gfs2_sharewrite_fault, -}; - diff --git a/fs/gfs2/ops_vm.h b/fs/gfs2/ops_vm.h deleted file mode 100644 index 4ae8f43ed5e..00000000000 --- a/fs/gfs2/ops_vm.h +++ /dev/null @@ -1,18 +0,0 @@ -/* - * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. - * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved. - * - * This copyrighted material is made available to anyone wishing to use, - * modify, copy, or redistribute it subject to the terms and conditions - * of the GNU General Public License version 2. - */ - -#ifndef __OPS_VM_DOT_H__ -#define __OPS_VM_DOT_H__ - -#include <linux/mm.h> - -extern struct vm_operations_struct gfs2_vm_ops_private; -extern struct vm_operations_struct gfs2_vm_ops_sharewrite; - -#endif /* __OPS_VM_DOT_H__ */ diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index addb51e0f13..a08dabd6ce9 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -1,6 +1,6 @@ /* * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. - * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved. + * Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved. * * This copyrighted material is made available to anyone wishing to use, * modify, copy, or redistribute it subject to the terms and conditions @@ -59,7 +59,6 @@ #include "super.h" #include "trans.h" #include "inode.h" -#include "ops_file.h" #include "ops_address.h" #include "util.h" @@ -274,10 +273,10 @@ static int bh_get(struct gfs2_quota_data *qd) } block = qd->qd_slot / sdp->sd_qc_per_block; - offset = qd->qd_slot % sdp->sd_qc_per_block;; + offset = qd->qd_slot % sdp->sd_qc_per_block; bh_map.b_size = 1 << ip->i_inode.i_blkbits; - error = gfs2_block_map(&ip->i_inode, block, 0, &bh_map); + error = gfs2_block_map(&ip->i_inode, block, &bh_map, 0); if (error) goto fail; error = gfs2_meta_read(ip->i_gl, bh_map.b_blocknr, DIO_WAIT, &bh); @@ -454,7 +453,7 @@ static void qdsb_put(struct gfs2_quota_data *qd) int gfs2_quota_hold(struct gfs2_inode *ip, u32 uid, u32 gid) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al = &ip->i_alloc; + struct gfs2_alloc *al = ip->i_alloc; struct gfs2_quota_data **qd = al->al_qd; int error; @@ -502,7 +501,7 @@ out: void gfs2_quota_unhold(struct gfs2_inode *ip) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al = &ip->i_alloc; + struct gfs2_alloc *al = ip->i_alloc; unsigned int x; gfs2_assert_warn(sdp, !test_bit(GIF_QD_LOCKED, &ip->i_flags)); @@ -646,7 +645,7 @@ static int gfs2_adjust_quota(struct gfs2_inode *ip, loff_t loc, } if (!buffer_mapped(bh)) { - gfs2_get_block(inode, iblock, bh, 1); + gfs2_block_map(inode, iblock, bh, 1); if (!buffer_mapped(bh)) goto unlock; } @@ -793,11 +792,9 @@ static int do_glock(struct gfs2_quota_data *qd, int force_refresh, struct gfs2_holder i_gh; struct gfs2_quota_host q; char buf[sizeof(struct gfs2_quota)]; - struct file_ra_state ra_state; int error; struct gfs2_quota_lvb *qlvb; - file_ra_state_init(&ra_state, sdp->sd_quota_inode->i_mapping); restart: error = gfs2_glock_nq_init(qd->qd_gl, LM_ST_SHARED, 0, q_gh); if (error) @@ -820,8 +817,8 @@ restart: memset(buf, 0, sizeof(struct gfs2_quota)); pos = qd2offset(qd); - error = gfs2_internal_read(ip, &ra_state, buf, - &pos, sizeof(struct gfs2_quota)); + error = gfs2_internal_read(ip, NULL, buf, &pos, + sizeof(struct gfs2_quota)); if (error < 0) goto fail_gunlock; @@ -856,7 +853,7 @@ fail: int gfs2_quota_lock(struct gfs2_inode *ip, u32 uid, u32 gid) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al = &ip->i_alloc; + struct gfs2_alloc *al = ip->i_alloc; unsigned int x; int error = 0; @@ -924,7 +921,7 @@ static int need_sync(struct gfs2_quota_data *qd) void gfs2_quota_unlock(struct gfs2_inode *ip) { - struct gfs2_alloc *al = &ip->i_alloc; + struct gfs2_alloc *al = ip->i_alloc; struct gfs2_quota_data *qda[4]; unsigned int count = 0; unsigned int x; @@ -972,7 +969,7 @@ static int print_message(struct gfs2_quota_data *qd, char *type) int gfs2_quota_check(struct gfs2_inode *ip, u32 uid, u32 gid) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al = &ip->i_alloc; + struct gfs2_alloc *al = ip->i_alloc; struct gfs2_quota_data *qd; s64 value; unsigned int x; @@ -1016,10 +1013,9 @@ int gfs2_quota_check(struct gfs2_inode *ip, u32 uid, u32 gid) void gfs2_quota_change(struct gfs2_inode *ip, s64 change, u32 uid, u32 gid) { - struct gfs2_alloc *al = &ip->i_alloc; + struct gfs2_alloc *al = ip->i_alloc; struct gfs2_quota_data *qd; unsigned int x; - unsigned int found = 0; if (gfs2_assert_warn(GFS2_SB(&ip->i_inode), change)) return; @@ -1032,7 +1028,6 @@ void gfs2_quota_change(struct gfs2_inode *ip, s64 change, if ((qd->qd_id == uid && test_bit(QDF_USER, &qd->qd_flags)) || (qd->qd_id == gid && !test_bit(QDF_USER, &qd->qd_flags))) { do_qc(qd, change); - found++; } } } diff --git a/fs/gfs2/recovery.c b/fs/gfs2/recovery.c index beb6c7ac008..6fb07d67ca8 100644 --- a/fs/gfs2/recovery.c +++ b/fs/gfs2/recovery.c @@ -391,7 +391,7 @@ static int clean_journal(struct gfs2_jdesc *jd, struct gfs2_log_header_host *hea lblock = head->lh_blkno; gfs2_replay_incr_blk(sdp, &lblock); bh_map.b_size = 1 << ip->i_inode.i_blkbits; - error = gfs2_block_map(&ip->i_inode, lblock, 0, &bh_map); + error = gfs2_block_map(&ip->i_inode, lblock, &bh_map, 0); if (error) return error; if (!bh_map.b_blocknr) { @@ -450,7 +450,7 @@ int gfs2_recover_journal(struct gfs2_jdesc *jd) fs_info(sdp, "jid=%u: Trying to acquire journal lock...\n", jd->jd_jid); - /* Aquire the journal lock so we can do recovery */ + /* Acquire the journal lock so we can do recovery */ error = gfs2_glock_nq_num(sdp, jd->jd_jid, &gfs2_journal_glops, LM_ST_EXCLUSIVE, @@ -504,13 +504,21 @@ int gfs2_recover_journal(struct gfs2_jdesc *jd) if (!test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)) ro = 1; } else { - if (sdp->sd_vfs->s_flags & MS_RDONLY) - ro = 1; + if (sdp->sd_vfs->s_flags & MS_RDONLY) { + /* check if device itself is read-only */ + ro = bdev_read_only(sdp->sd_vfs->s_bdev); + if (!ro) { + fs_info(sdp, "recovery required on " + "read-only filesystem.\n"); + fs_info(sdp, "write access will be " + "enabled during recovery.\n"); + } + } } if (ro) { - fs_warn(sdp, "jid=%u: Can't replay: read-only FS\n", - jd->jd_jid); + fs_warn(sdp, "jid=%u: Can't replay: read-only block " + "device\n", jd->jd_jid); error = -EROFS; goto fail_gunlock_tr; } diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index 708c287e1d0..3552110b2e5 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -25,10 +25,10 @@ #include "rgrp.h" #include "super.h" #include "trans.h" -#include "ops_file.h" #include "util.h" #include "log.h" #include "inode.h" +#include "ops_address.h" #define BFITNOENT ((u32)~0) #define NO_BLOCK ((u64)~0) @@ -126,41 +126,43 @@ static unsigned char gfs2_testbit(struct gfs2_rgrpd *rgd, unsigned char *buffer, * Return: the block number (bitmap buffer scope) that was found */ -static u32 gfs2_bitfit(struct gfs2_rgrpd *rgd, unsigned char *buffer, - unsigned int buflen, u32 goal, - unsigned char old_state) +static u32 gfs2_bitfit(unsigned char *buffer, unsigned int buflen, u32 goal, + unsigned char old_state) { - unsigned char *byte, *end, alloc; + unsigned char *byte; u32 blk = goal; - unsigned int bit; + unsigned int bit, bitlong; + unsigned long *plong, plong55; byte = buffer + (goal / GFS2_NBBY); + plong = (unsigned long *)(buffer + (goal / GFS2_NBBY)); bit = (goal % GFS2_NBBY) * GFS2_BIT_SIZE; - end = buffer + buflen; - alloc = (old_state == GFS2_BLKST_FREE) ? 0x55 : 0; - - while (byte < end) { - /* If we're looking for a free block we can eliminate all - bitmap settings with 0x55, which represents four data - blocks in a row. If we're looking for a data block, we can - eliminate 0x00 which corresponds to four free blocks. */ - if ((*byte & 0x55) == alloc) { - blk += (8 - bit) >> 1; - - bit = 0; - byte++; - + bitlong = bit; +#if BITS_PER_LONG == 32 + plong55 = 0x55555555; +#else + plong55 = 0x5555555555555555; +#endif + while (byte < buffer + buflen) { + + if (bitlong == 0 && old_state == 0 && *plong == plong55) { + plong++; + byte += sizeof(unsigned long); + blk += sizeof(unsigned long) * GFS2_NBBY; continue; } - if (((*byte >> bit) & GFS2_BIT_MASK) == old_state) return blk; - bit += GFS2_BIT_SIZE; if (bit >= 8) { bit = 0; byte++; } + bitlong += GFS2_BIT_SIZE; + if (bitlong >= sizeof(unsigned long) * 8) { + bitlong = 0; + plong++; + } blk++; } @@ -817,11 +819,9 @@ void gfs2_rgrp_repolish_clones(struct gfs2_rgrpd *rgd) struct gfs2_alloc *gfs2_alloc_get(struct gfs2_inode *ip) { - struct gfs2_alloc *al = &ip->i_alloc; - - /* FIXME: Should assert that the correct locks are held here... */ - memset(al, 0, sizeof(*al)); - return al; + BUG_ON(ip->i_alloc != NULL); + ip->i_alloc = kzalloc(sizeof(struct gfs2_alloc), GFP_KERNEL); + return ip->i_alloc; } /** @@ -1059,26 +1059,34 @@ static struct inode *get_local_rgrp(struct gfs2_inode *ip, u64 *last_unlinked) struct inode *inode = NULL; struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct gfs2_rgrpd *rgd, *begin = NULL; - struct gfs2_alloc *al = &ip->i_alloc; + struct gfs2_alloc *al = ip->i_alloc; int flags = LM_FLAG_TRY; int skipped = 0; int loops = 0; - int error; + int error, rg_locked; /* Try recently successful rgrps */ rgd = recent_rgrp_first(sdp, ip->i_last_rg_alloc); while (rgd) { - error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, - LM_FLAG_TRY, &al->al_rgd_gh); + rg_locked = 0; + + if (gfs2_glock_is_locked_by_me(rgd->rd_gl)) { + rg_locked = 1; + error = 0; + } else { + error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, + LM_FLAG_TRY, &al->al_rgd_gh); + } switch (error) { case 0: if (try_rgrp_fit(rgd, al)) goto out; if (rgd->rd_flags & GFS2_RDF_CHECK) inode = try_rgrp_unlink(rgd, last_unlinked); - gfs2_glock_dq_uninit(&al->al_rgd_gh); + if (!rg_locked) + gfs2_glock_dq_uninit(&al->al_rgd_gh); if (inode) return inode; rgd = recent_rgrp_next(rgd, 1); @@ -1098,15 +1106,23 @@ static struct inode *get_local_rgrp(struct gfs2_inode *ip, u64 *last_unlinked) begin = rgd = forward_rgrp_get(sdp); for (;;) { - error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, flags, - &al->al_rgd_gh); + rg_locked = 0; + + if (gfs2_glock_is_locked_by_me(rgd->rd_gl)) { + rg_locked = 1; + error = 0; + } else { + error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, flags, + &al->al_rgd_gh); + } switch (error) { case 0: if (try_rgrp_fit(rgd, al)) goto out; if (rgd->rd_flags & GFS2_RDF_CHECK) inode = try_rgrp_unlink(rgd, last_unlinked); - gfs2_glock_dq_uninit(&al->al_rgd_gh); + if (!rg_locked) + gfs2_glock_dq_uninit(&al->al_rgd_gh); if (inode) return inode; break; @@ -1158,7 +1174,7 @@ out: int gfs2_inplace_reserve_i(struct gfs2_inode *ip, char *file, unsigned int line) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al = &ip->i_alloc; + struct gfs2_alloc *al = ip->i_alloc; struct inode *inode; int error = 0; u64 last_unlinked = NO_BLOCK; @@ -1204,7 +1220,7 @@ try_again: void gfs2_inplace_release(struct gfs2_inode *ip) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al = &ip->i_alloc; + struct gfs2_alloc *al = ip->i_alloc; if (gfs2_assert_warn(sdp, al->al_alloced <= al->al_requested) == -1) fs_warn(sdp, "al_alloced = %u, al_requested = %u " @@ -1213,7 +1229,8 @@ void gfs2_inplace_release(struct gfs2_inode *ip) al->al_line); al->al_rgd = NULL; - gfs2_glock_dq_uninit(&al->al_rgd_gh); + if (al->al_rgd_gh.gh_gl) + gfs2_glock_dq_uninit(&al->al_rgd_gh); if (ip != GFS2_I(sdp->sd_rindex)) gfs2_glock_dq_uninit(&al->al_ri_gh); } @@ -1301,11 +1318,10 @@ static u32 rgblk_search(struct gfs2_rgrpd *rgd, u32 goal, /* The GFS2_BLKST_UNLINKED state doesn't apply to the clone bitmaps, so we must search the originals for that. */ if (old_state != GFS2_BLKST_UNLINKED && bi->bi_clone) - blk = gfs2_bitfit(rgd, bi->bi_clone + bi->bi_offset, + blk = gfs2_bitfit(bi->bi_clone + bi->bi_offset, bi->bi_len, goal, old_state); else - blk = gfs2_bitfit(rgd, - bi->bi_bh->b_data + bi->bi_offset, + blk = gfs2_bitfit(bi->bi_bh->b_data + bi->bi_offset, bi->bi_len, goal, old_state); if (blk != BFITNOENT) break; @@ -1394,7 +1410,7 @@ static struct gfs2_rgrpd *rgblk_free(struct gfs2_sbd *sdp, u64 bstart, u64 gfs2_alloc_data(struct gfs2_inode *ip) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al = &ip->i_alloc; + struct gfs2_alloc *al = ip->i_alloc; struct gfs2_rgrpd *rgd = al->al_rgd; u32 goal, blk; u64 block; @@ -1439,7 +1455,7 @@ u64 gfs2_alloc_data(struct gfs2_inode *ip) u64 gfs2_alloc_meta(struct gfs2_inode *ip) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al = &ip->i_alloc; + struct gfs2_alloc *al = ip->i_alloc; struct gfs2_rgrpd *rgd = al->al_rgd; u32 goal, blk; u64 block; @@ -1485,7 +1501,7 @@ u64 gfs2_alloc_meta(struct gfs2_inode *ip) u64 gfs2_alloc_di(struct gfs2_inode *dip, u64 *generation) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); - struct gfs2_alloc *al = &dip->i_alloc; + struct gfs2_alloc *al = dip->i_alloc; struct gfs2_rgrpd *rgd = al->al_rgd; u32 blk; u64 block; diff --git a/fs/gfs2/rgrp.h b/fs/gfs2/rgrp.h index b4c6adfc6f2..149bb161f4b 100644 --- a/fs/gfs2/rgrp.h +++ b/fs/gfs2/rgrp.h @@ -32,7 +32,9 @@ void gfs2_rgrp_repolish_clones(struct gfs2_rgrpd *rgd); struct gfs2_alloc *gfs2_alloc_get(struct gfs2_inode *ip); static inline void gfs2_alloc_put(struct gfs2_inode *ip) { - return; /* So we can see where ip->i_alloc is used */ + BUG_ON(ip->i_alloc == NULL); + kfree(ip->i_alloc); + ip->i_alloc = NULL; } int gfs2_inplace_reserve_i(struct gfs2_inode *ip, diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index dd3e737f528..ef0562c3bc7 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -1,6 +1,6 @@ /* * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. - * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved. + * Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved. * * This copyrighted material is made available to anyone wishing to use, * modify, copy, or redistribute it subject to the terms and conditions @@ -51,13 +51,9 @@ void gfs2_tune_init(struct gfs2_tune *gt) { spin_lock_init(>->gt_spin); - gt->gt_ilimit = 100; - gt->gt_ilimit_tries = 3; - gt->gt_ilimit_min = 1; gt->gt_demote_secs = 300; gt->gt_incore_log_blocks = 1024; gt->gt_log_flush_secs = 60; - gt->gt_jindex_refresh_secs = 60; gt->gt_recoverd_secs = 60; gt->gt_logd_secs = 1; gt->gt_quotad_secs = 5; @@ -71,10 +67,8 @@ void gfs2_tune_init(struct gfs2_tune *gt) gt->gt_new_files_jdata = 0; gt->gt_new_files_directio = 0; gt->gt_max_readahead = 1 << 18; - gt->gt_lockdump_size = 131072; gt->gt_stall_secs = 600; gt->gt_complain_secs = 10; - gt->gt_reclaim_limit = 5000; gt->gt_statfs_quantum = 30; gt->gt_statfs_slow = 0; } @@ -393,6 +387,7 @@ int gfs2_jindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ji_gh) if (!jd) break; + INIT_LIST_HEAD(&jd->extent_list); jd->jd_inode = gfs2_lookupi(sdp->sd_jindex, &name, 1, NULL); if (!jd->jd_inode || IS_ERR(jd->jd_inode)) { if (!jd->jd_inode) @@ -422,8 +417,9 @@ int gfs2_jindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ji_gh) void gfs2_jindex_free(struct gfs2_sbd *sdp) { - struct list_head list; + struct list_head list, *head; struct gfs2_jdesc *jd; + struct gfs2_journal_extent *jext; spin_lock(&sdp->sd_jindex_spin); list_add(&list, &sdp->sd_jindex_list); @@ -433,6 +429,14 @@ void gfs2_jindex_free(struct gfs2_sbd *sdp) while (!list_empty(&list)) { jd = list_entry(list.next, struct gfs2_jdesc, jd_list); + head = &jd->extent_list; + while (!list_empty(head)) { + jext = list_entry(head->next, + struct gfs2_journal_extent, + extent_list); + list_del(&jext->extent_list); + kfree(jext); + } list_del(&jd->jd_list); iput(jd->jd_inode); kfree(jd); @@ -543,7 +547,6 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp) if (error) return error; - gfs2_meta_cache_flush(ip); j_gl->gl_ops->go_inval(j_gl, DIO_METADATA); error = gfs2_find_jhead(sdp->sd_jdesc, &head); @@ -686,9 +689,7 @@ void gfs2_statfs_change(struct gfs2_sbd *sdp, s64 total, s64 free, if (error) return; - mutex_lock(&sdp->sd_statfs_mutex); gfs2_trans_add_bh(l_ip->i_gl, l_bh, 1); - mutex_unlock(&sdp->sd_statfs_mutex); spin_lock(&sdp->sd_statfs_spin); l_sc->sc_total += total; @@ -736,9 +737,7 @@ int gfs2_statfs_sync(struct gfs2_sbd *sdp) if (error) goto out_bh2; - mutex_lock(&sdp->sd_statfs_mutex); gfs2_trans_add_bh(l_ip->i_gl, l_bh, 1); - mutex_unlock(&sdp->sd_statfs_mutex); spin_lock(&sdp->sd_statfs_spin); m_sc->sc_total += l_sc->sc_total; diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c index 06e0b7768d9..eaa3b7b2f99 100644 --- a/fs/gfs2/sys.c +++ b/fs/gfs2/sys.c @@ -32,7 +32,8 @@ spinlock_t gfs2_sys_margs_lock; static ssize_t id_show(struct gfs2_sbd *sdp, char *buf) { - return snprintf(buf, PAGE_SIZE, "%s\n", sdp->sd_vfs->s_id); + return snprintf(buf, PAGE_SIZE, "%u:%u\n", + MAJOR(sdp->sd_vfs->s_dev), MINOR(sdp->sd_vfs->s_dev)); } static ssize_t fsname_show(struct gfs2_sbd *sdp, char *buf) @@ -221,9 +222,7 @@ static struct kobj_type gfs2_ktype = { .sysfs_ops = &gfs2_attr_ops, }; -static struct kset gfs2_kset = { - .ktype = &gfs2_ktype, -}; +static struct kset *gfs2_kset; /* * display struct lm_lockstruct fields @@ -427,13 +426,11 @@ TUNE_ATTR_2(name, name##_store) TUNE_ATTR(demote_secs, 0); TUNE_ATTR(incore_log_blocks, 0); TUNE_ATTR(log_flush_secs, 0); -TUNE_ATTR(jindex_refresh_secs, 0); TUNE_ATTR(quota_warn_period, 0); TUNE_ATTR(quota_quantum, 0); TUNE_ATTR(atime_quantum, 0); TUNE_ATTR(max_readahead, 0); TUNE_ATTR(complain_secs, 0); -TUNE_ATTR(reclaim_limit, 0); TUNE_ATTR(statfs_slow, 0); TUNE_ATTR(new_files_jdata, 0); TUNE_ATTR(new_files_directio, 0); @@ -450,13 +447,11 @@ static struct attribute *tune_attrs[] = { &tune_attr_demote_secs.attr, &tune_attr_incore_log_blocks.attr, &tune_attr_log_flush_secs.attr, - &tune_attr_jindex_refresh_secs.attr, &tune_attr_quota_warn_period.attr, &tune_attr_quota_quantum.attr, &tune_attr_atime_quantum.attr, &tune_attr_max_readahead.attr, &tune_attr_complain_secs.attr, - &tune_attr_reclaim_limit.attr, &tune_attr_statfs_slow.attr, &tune_attr_quota_simul_sync.attr, &tune_attr_quota_cache_secs.attr, @@ -495,14 +490,9 @@ int gfs2_sys_fs_add(struct gfs2_sbd *sdp) { int error; - sdp->sd_kobj.kset = &gfs2_kset; - sdp->sd_kobj.ktype = &gfs2_ktype; - - error = kobject_set_name(&sdp->sd_kobj, "%s", sdp->sd_table_name); - if (error) - goto fail; - - error = kobject_register(&sdp->sd_kobj); + sdp->sd_kobj.kset = gfs2_kset; + error = kobject_init_and_add(&sdp->sd_kobj, &gfs2_ktype, NULL, + "%s", sdp->sd_table_name); if (error) goto fail; @@ -522,6 +512,7 @@ int gfs2_sys_fs_add(struct gfs2_sbd *sdp) if (error) goto fail_args; + kobject_uevent(&sdp->sd_kobj, KOBJ_ADD); return 0; fail_args: @@ -531,7 +522,7 @@ fail_counters: fail_lockstruct: sysfs_remove_group(&sdp->sd_kobj, &lockstruct_group); fail_reg: - kobject_unregister(&sdp->sd_kobj); + kobject_put(&sdp->sd_kobj); fail: fs_err(sdp, "error %d adding sysfs files", error); return error; @@ -543,21 +534,22 @@ void gfs2_sys_fs_del(struct gfs2_sbd *sdp) sysfs_remove_group(&sdp->sd_kobj, &args_group); sysfs_remove_group(&sdp->sd_kobj, &counters_group); sysfs_remove_group(&sdp->sd_kobj, &lockstruct_group); - kobject_unregister(&sdp->sd_kobj); + kobject_put(&sdp->sd_kobj); } int gfs2_sys_init(void) { gfs2_sys_margs = NULL; spin_lock_init(&gfs2_sys_margs_lock); - kobject_set_name(&gfs2_kset.kobj, "gfs2"); - kobj_set_kset_s(&gfs2_kset, fs_subsys); - return kset_register(&gfs2_kset); + gfs2_kset = kset_create_and_add("gfs2", NULL, fs_kobj); + if (!gfs2_kset) + return -ENOMEM; + return 0; } void gfs2_sys_uninit(void) { kfree(gfs2_sys_margs); - kset_unregister(&gfs2_kset); + kset_unregister(gfs2_kset); } diff --git a/fs/gfs2/trans.c b/fs/gfs2/trans.c index 717983e2c2a..73e5d92a657 100644 --- a/fs/gfs2/trans.c +++ b/fs/gfs2/trans.c @@ -114,11 +114,6 @@ void gfs2_trans_end(struct gfs2_sbd *sdp) gfs2_log_flush(sdp, NULL); } -void gfs2_trans_add_gl(struct gfs2_glock *gl) -{ - lops_add(gl->gl_sbd, &gl->gl_le); -} - /** * gfs2_trans_add_bh - Add a to-be-modified buffer to the current transaction * @gl: the glock the buffer belongs to diff --git a/fs/gfs2/trans.h b/fs/gfs2/trans.h index 043d5f4b9c4..e826f0dab80 100644 --- a/fs/gfs2/trans.h +++ b/fs/gfs2/trans.h @@ -30,7 +30,6 @@ int gfs2_trans_begin(struct gfs2_sbd *sdp, unsigned int blocks, void gfs2_trans_end(struct gfs2_sbd *sdp); -void gfs2_trans_add_gl(struct gfs2_glock *gl); void gfs2_trans_add_bh(struct gfs2_glock *gl, struct buffer_head *bh, int meta); void gfs2_trans_add_revoke(struct gfs2_sbd *sdp, struct gfs2_bufdata *bd); void gfs2_trans_add_unrevoke(struct gfs2_sbd *sdp, u64 blkno); diff --git a/fs/hfs/bfind.c b/fs/hfs/bfind.c index f8452a0eab5..4129cdb3f0d 100644 --- a/fs/hfs/bfind.c +++ b/fs/hfs/bfind.c @@ -52,9 +52,9 @@ int __hfs_brec_find(struct hfs_bnode *bnode, struct hfs_find_data *fd) rec = (e + b) / 2; len = hfs_brec_lenoff(bnode, rec, &off); keylen = hfs_brec_keylen(bnode, rec); - if (keylen == HFS_BAD_KEYLEN) { + if (keylen == 0) { res = -EINVAL; - goto done; + goto fail; } hfs_bnode_read(bnode, fd->key, off, keylen); cmpval = bnode->tree->keycmp(fd->key, fd->search_key); @@ -71,9 +71,9 @@ int __hfs_brec_find(struct hfs_bnode *bnode, struct hfs_find_data *fd) if (rec != e && e >= 0) { len = hfs_brec_lenoff(bnode, e, &off); keylen = hfs_brec_keylen(bnode, e); - if (keylen == HFS_BAD_KEYLEN) { + if (keylen == 0) { res = -EINVAL; - goto done; + goto fail; } hfs_bnode_read(bnode, fd->key, off, keylen); } @@ -83,6 +83,7 @@ done: fd->keylength = keylen; fd->entryoffset = off + keylen; fd->entrylength = len - keylen; +fail: return res; } @@ -206,7 +207,7 @@ int hfs_brec_goto(struct hfs_find_data *fd, int cnt) len = hfs_brec_lenoff(bnode, fd->record, &off); keylen = hfs_brec_keylen(bnode, fd->record); - if (keylen == HFS_BAD_KEYLEN) { + if (keylen == 0) { res = -EINVAL; goto out; } diff --git a/fs/hfs/brec.c b/fs/hfs/brec.c index 8626ee375ea..92fb358ce82 100644 --- a/fs/hfs/brec.c +++ b/fs/hfs/brec.c @@ -49,14 +49,14 @@ u16 hfs_brec_keylen(struct hfs_bnode *node, u16 rec) if (retval > node->tree->max_key_len + 2) { printk(KERN_ERR "hfs: keylen %d too large\n", retval); - retval = HFS_BAD_KEYLEN; + retval = 0; } } else { retval = (hfs_bnode_read_u8(node, recoff) | 1) + 1; if (retval > node->tree->max_key_len + 1) { printk(KERN_ERR "hfs: keylen %d too large\n", retval); - retval = HFS_BAD_KEYLEN; + retval = 0; } } } @@ -229,7 +229,7 @@ skip: static struct hfs_bnode *hfs_bnode_split(struct hfs_find_data *fd) { struct hfs_btree *tree; - struct hfs_bnode *node, *new_node; + struct hfs_bnode *node, *new_node, *next_node; struct hfs_bnode_desc node_desc; int num_recs, new_rec_off, new_off, old_rec_off; int data_start, data_end, size; @@ -248,6 +248,17 @@ static struct hfs_bnode *hfs_bnode_split(struct hfs_find_data *fd) new_node->type = node->type; new_node->height = node->height; + if (node->next) + next_node = hfs_bnode_find(tree, node->next); + else + next_node = NULL; + + if (IS_ERR(next_node)) { + hfs_bnode_put(node); + hfs_bnode_put(new_node); + return next_node; + } + size = tree->node_size / 2 - node->num_recs * 2 - 14; old_rec_off = tree->node_size - 4; num_recs = 1; @@ -261,6 +272,8 @@ static struct hfs_bnode *hfs_bnode_split(struct hfs_find_data *fd) /* panic? */ hfs_bnode_put(node); hfs_bnode_put(new_node); + if (next_node) + hfs_bnode_put(next_node); return ERR_PTR(-ENOSPC); } @@ -315,8 +328,7 @@ static struct hfs_bnode *hfs_bnode_split(struct hfs_find_data *fd) hfs_bnode_write(node, &node_desc, 0, sizeof(node_desc)); /* update next bnode header */ - if (new_node->next) { - struct hfs_bnode *next_node = hfs_bnode_find(tree, new_node->next); + if (next_node) { next_node->prev = new_node->this; hfs_bnode_read(next_node, &node_desc, 0, sizeof(node_desc)); node_desc.prev = cpu_to_be32(next_node->prev); diff --git a/fs/hfs/btree.c b/fs/hfs/btree.c index 110dd3515dc..24cf6fc4302 100644 --- a/fs/hfs/btree.c +++ b/fs/hfs/btree.c @@ -81,15 +81,23 @@ struct hfs_btree *hfs_btree_open(struct super_block *sb, u32 id, btree_keycmp ke goto fail_page; if (!tree->node_count) goto fail_page; - if ((id == HFS_EXT_CNID) && (tree->max_key_len != HFS_MAX_EXT_KEYLEN)) { - printk(KERN_ERR "hfs: invalid extent max_key_len %d\n", - tree->max_key_len); - goto fail_page; - } - if ((id == HFS_CAT_CNID) && (tree->max_key_len != HFS_MAX_CAT_KEYLEN)) { - printk(KERN_ERR "hfs: invalid catalog max_key_len %d\n", - tree->max_key_len); - goto fail_page; + switch (id) { + case HFS_EXT_CNID: + if (tree->max_key_len != HFS_MAX_EXT_KEYLEN) { + printk(KERN_ERR "hfs: invalid extent max_key_len %d\n", + tree->max_key_len); + goto fail_page; + } + break; + case HFS_CAT_CNID: + if (tree->max_key_len != HFS_MAX_CAT_KEYLEN) { + printk(KERN_ERR "hfs: invalid catalog max_key_len %d\n", + tree->max_key_len); + goto fail_page; + } + break; + default: + BUG(); } tree->node_size_shift = ffs(size) - 1; diff --git a/fs/hfs/hfs.h b/fs/hfs/hfs.h index c6aae61adfe..6f194d0768b 100644 --- a/fs/hfs/hfs.h +++ b/fs/hfs/hfs.h @@ -28,8 +28,6 @@ #define HFS_MAX_NAMELEN 128 #define HFS_MAX_VALENCE 32767U -#define HFS_BAD_KEYLEN 0xFF - /* Meanings of the drAtrb field of the MDB, * Reference: _Inside Macintosh: Files_ p. 2-61 */ diff --git a/fs/hfs/super.c b/fs/hfs/super.c index 16cbd902f8b..32de44ed002 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -6,7 +6,7 @@ * This file may be distributed under the terms of the GNU General Public License. * * This file contains hfs_read_super(), some of the super_ops and - * init_module() and cleanup_module(). The remaining super_ops are in + * init_hfs_fs() and exit_hfs_fs(). The remaining super_ops are in * inode.c since they deal with inodes. * * Based on the minix file system code, (C) 1991, 1992 by Linus Torvalds diff --git a/fs/hfsplus/btree.c b/fs/hfsplus/btree.c index 050d29c0a5b..bb5433608a4 100644 --- a/fs/hfsplus/btree.c +++ b/fs/hfsplus/btree.c @@ -22,6 +22,7 @@ struct hfs_btree *hfs_btree_open(struct super_block *sb, u32 id) struct hfs_btree *tree; struct hfs_btree_header_rec *head; struct address_space *mapping; + struct inode *inode; struct page *page; unsigned int size; @@ -33,9 +34,10 @@ struct hfs_btree *hfs_btree_open(struct super_block *sb, u32 id) spin_lock_init(&tree->hash_lock); tree->sb = sb; tree->cnid = id; - tree->inode = iget(sb, id); - if (!tree->inode) + inode = hfsplus_iget(sb, id); + if (IS_ERR(inode)) goto free_tree; + tree->inode = inode; mapping = tree->inode->i_mapping; page = read_mapping_page(mapping, 0, NULL); diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c index 1955ee61251..5f402367825 100644 --- a/fs/hfsplus/dir.c +++ b/fs/hfsplus/dir.c @@ -97,9 +97,9 @@ again: goto fail; } hfs_find_exit(&fd); - inode = iget(dir->i_sb, cnid); - if (!inode) - return ERR_PTR(-EACCES); + inode = hfsplus_iget(dir->i_sb, cnid); + if (IS_ERR(inode)) + return ERR_CAST(inode); if (S_ISREG(inode->i_mode)) HFSPLUS_I(inode).dev = linkid; out: @@ -340,16 +340,23 @@ static int hfsplus_unlink(struct inode *dir, struct dentry *dentry) if (inode->i_nlink > 0) drop_nlink(inode); - hfsplus_delete_inode(inode); - if (inode->i_ino != cnid && !inode->i_nlink) { - if (!atomic_read(&HFSPLUS_I(inode).opencnt)) { - res = hfsplus_delete_cat(inode->i_ino, HFSPLUS_SB(sb).hidden_dir, NULL); - if (!res) - hfsplus_delete_inode(inode); + if (inode->i_ino == cnid) + clear_nlink(inode); + if (!inode->i_nlink) { + if (inode->i_ino != cnid) { + HFSPLUS_SB(sb).file_count--; + if (!atomic_read(&HFSPLUS_I(inode).opencnt)) { + res = hfsplus_delete_cat(inode->i_ino, + HFSPLUS_SB(sb).hidden_dir, + NULL); + if (!res) + hfsplus_delete_inode(inode); + } else + inode->i_flags |= S_DEAD; } else - inode->i_flags |= S_DEAD; + hfsplus_delete_inode(inode); } else - clear_nlink(inode); + HFSPLUS_SB(sb).file_count--; inode->i_ctime = CURRENT_TIME_SEC; mark_inode_dirty(inode); diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h index d9f5eda6d03..d72d0a8b25a 100644 --- a/fs/hfsplus/hfsplus_fs.h +++ b/fs/hfsplus/hfsplus_fs.h @@ -345,6 +345,9 @@ int hfsplus_parse_options(char *, struct hfsplus_sb_info *); void hfsplus_fill_defaults(struct hfsplus_sb_info *); int hfsplus_show_options(struct seq_file *, struct vfsmount *); +/* super.c */ +struct inode *hfsplus_iget(struct super_block *, unsigned long); + /* tables.c */ extern u16 hfsplus_case_fold_table[]; extern u16 hfsplus_decompose_table[]; diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index ecf70dafb64..b0f9ad362d1 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -20,11 +20,18 @@ static void hfsplus_destroy_inode(struct inode *inode); #include "hfsplus_fs.h" -static void hfsplus_read_inode(struct inode *inode) +struct inode *hfsplus_iget(struct super_block *sb, unsigned long ino) { struct hfs_find_data fd; struct hfsplus_vh *vhdr; - int err; + struct inode *inode; + long err = -EIO; + + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; INIT_LIST_HEAD(&HFSPLUS_I(inode).open_dir_list); init_MUTEX(&HFSPLUS_I(inode).extents_lock); @@ -41,7 +48,7 @@ static void hfsplus_read_inode(struct inode *inode) hfs_find_exit(&fd); if (err) goto bad_inode; - return; + goto done; } vhdr = HFSPLUS_SB(inode->i_sb).s_vhdr; switch(inode->i_ino) { @@ -70,10 +77,13 @@ static void hfsplus_read_inode(struct inode *inode) goto bad_inode; } - return; +done: + unlock_new_inode(inode); + return inode; - bad_inode: - make_bad_inode(inode); +bad_inode: + iget_failed(inode); + return ERR_PTR(err); } static int hfsplus_write_inode(struct inode *inode, int unused) @@ -262,7 +272,6 @@ static int hfsplus_remount(struct super_block *sb, int *flags, char *data) static const struct super_operations hfsplus_sops = { .alloc_inode = hfsplus_alloc_inode, .destroy_inode = hfsplus_destroy_inode, - .read_inode = hfsplus_read_inode, .write_inode = hfsplus_write_inode, .clear_inode = hfsplus_clear_inode, .put_super = hfsplus_put_super, @@ -278,7 +287,7 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) struct hfsplus_sb_info *sbi; hfsplus_cat_entry entry; struct hfs_find_data fd; - struct inode *root; + struct inode *root, *inode; struct qstr str; struct nls_table *nls = NULL; int err = -EINVAL; @@ -366,18 +375,25 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) goto cleanup; } - HFSPLUS_SB(sb).alloc_file = iget(sb, HFSPLUS_ALLOC_CNID); - if (!HFSPLUS_SB(sb).alloc_file) { + inode = hfsplus_iget(sb, HFSPLUS_ALLOC_CNID); + if (IS_ERR(inode)) { printk(KERN_ERR "hfs: failed to load allocation file\n"); + err = PTR_ERR(inode); goto cleanup; } + HFSPLUS_SB(sb).alloc_file = inode; /* Load the root directory */ - root = iget(sb, HFSPLUS_ROOT_CNID); + root = hfsplus_iget(sb, HFSPLUS_ROOT_CNID); + if (IS_ERR(root)) { + printk(KERN_ERR "hfs: failed to load root directory\n"); + err = PTR_ERR(root); + goto cleanup; + } sb->s_root = d_alloc_root(root); if (!sb->s_root) { - printk(KERN_ERR "hfs: failed to load root directory\n"); iput(root); + err = -ENOMEM; goto cleanup; } sb->s_root->d_op = &hfsplus_dentry_operations; @@ -390,9 +406,12 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) hfs_find_exit(&fd); if (entry.type != cpu_to_be16(HFSPLUS_FOLDER)) goto cleanup; - HFSPLUS_SB(sb).hidden_dir = iget(sb, be32_to_cpu(entry.folder.id)); - if (!HFSPLUS_SB(sb).hidden_dir) + inode = hfsplus_iget(sb, be32_to_cpu(entry.folder.id)); + if (IS_ERR(inode)) { + err = PTR_ERR(inode); goto cleanup; + } + HFSPLUS_SB(sb).hidden_dir = inode; } else hfs_find_exit(&fd); diff --git a/fs/hfsplus/unicode.c b/fs/hfsplus/unicode.c index 9e10f9444b6..628ccf6fa40 100644 --- a/fs/hfsplus/unicode.c +++ b/fs/hfsplus/unicode.c @@ -325,7 +325,7 @@ int hfsplus_hash_dentry(struct dentry *dentry, struct qstr *str) struct super_block *sb = dentry->d_sb; const char *astr; const u16 *dstr; - int casefold, decompose, size, dsize, len; + int casefold, decompose, size, len; unsigned long hash; wchar_t c; u16 c2; @@ -336,6 +336,7 @@ int hfsplus_hash_dentry(struct dentry *dentry, struct qstr *str) astr = str->name; len = str->len; while (len > 0) { + int uninitialized_var(dsize); size = asc2unichar(sb, astr, len, &c); astr += size; len -= size; diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index 8966b050196..5222345ddcc 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -11,6 +11,8 @@ #include <linux/mm.h> #include <linux/pagemap.h> #include <linux/statfs.h> +#include <linux/seq_file.h> +#include <linux/mount.h> #include "hostfs.h" #include "init.h" #include "kern.h" @@ -202,7 +204,7 @@ static char *follow_link(char *link) return ERR_PTR(n); } -static int read_inode(struct inode *ino) +static int hostfs_read_inode(struct inode *ino) { char *name; int err = 0; @@ -233,6 +235,25 @@ static int read_inode(struct inode *ino) return err; } +static struct inode *hostfs_iget(struct super_block *sb) +{ + struct inode *inode; + long ret; + + inode = iget_locked(sb, 0); + if (!inode) + return ERR_PTR(-ENOMEM); + if (inode->i_state & I_NEW) { + ret = hostfs_read_inode(inode); + if (ret < 0) { + iget_failed(inode); + return ERR_PTR(ret); + } + unlock_new_inode(inode); + } + return inode; +} + int hostfs_statfs(struct dentry *dentry, struct kstatfs *sf) { /* @@ -303,9 +324,16 @@ static void hostfs_destroy_inode(struct inode *inode) kfree(HOSTFS_I(inode)); } -static void hostfs_read_inode(struct inode *inode) +static int hostfs_show_options(struct seq_file *seq, struct vfsmount *vfs) { - read_inode(inode); + struct inode *root = vfs->mnt_sb->s_root->d_inode; + const char *root_path = HOSTFS_I(root)->host_filename; + size_t offset = strlen(root_ino) + 1; + + if (strlen(root_path) > offset) + seq_printf(seq, ",%s", root_path + offset); + + return 0; } static const struct super_operations hostfs_sbops = { @@ -313,8 +341,8 @@ static const struct super_operations hostfs_sbops = { .drop_inode = generic_delete_inode, .delete_inode = hostfs_delete_inode, .destroy_inode = hostfs_destroy_inode, - .read_inode = hostfs_read_inode, .statfs = hostfs_statfs, + .show_options = hostfs_show_options, }; int hostfs_readdir(struct file *file, void *ent, filldir_t filldir) @@ -571,10 +599,11 @@ int hostfs_create(struct inode *dir, struct dentry *dentry, int mode, char *name; int error, fd; - error = -ENOMEM; - inode = iget(dir->i_sb, 0); - if (inode == NULL) + inode = hostfs_iget(dir->i_sb); + if (IS_ERR(inode)) { + error = PTR_ERR(inode); goto out; + } error = init_inode(inode, dentry); if (error) @@ -615,10 +644,11 @@ struct dentry *hostfs_lookup(struct inode *ino, struct dentry *dentry, char *name; int err; - err = -ENOMEM; - inode = iget(ino->i_sb, 0); - if (inode == NULL) + inode = hostfs_iget(ino->i_sb); + if (IS_ERR(inode)) { + err = PTR_ERR(inode); goto out; + } err = init_inode(inode, dentry); if (err) @@ -736,11 +766,13 @@ int hostfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) { struct inode *inode; char *name; - int err = -ENOMEM; + int err; - inode = iget(dir->i_sb, 0); - if (inode == NULL) + inode = hostfs_iget(dir->i_sb); + if (IS_ERR(inode)) { + err = PTR_ERR(inode); goto out; + } err = init_inode(inode, dentry); if (err) @@ -952,9 +984,11 @@ static int hostfs_fill_sb_common(struct super_block *sb, void *d, int silent) sprintf(host_root_path, "%s/%s", root_ino, req_root); - root_inode = iget(sb, 0); - if (root_inode == NULL) + root_inode = hostfs_iget(sb); + if (IS_ERR(root_inode)) { + err = PTR_ERR(root_inode); goto out_free; + } err = init_inode(root_inode, NULL); if (err) @@ -972,7 +1006,7 @@ static int hostfs_fill_sb_common(struct super_block *sb, void *d, int silent) if (sb->s_root == NULL) goto out_put; - err = read_inode(root_inode); + err = hostfs_read_inode(root_inode); if (err) { /* No iput in this case because the dput does that for us */ dput(sb->s_root); diff --git a/fs/hostfs/hostfs_user.c b/fs/hostfs/hostfs_user.c index 35c1a9f33f4..53fd0a67c11 100644 --- a/fs/hostfs/hostfs_user.c +++ b/fs/hostfs/hostfs_user.c @@ -285,17 +285,17 @@ int set_attr(const char *file, struct hostfs_iattr *attrs, int fd) return err; times[0].tv_sec = atime_ts.tv_sec; - times[0].tv_usec = atime_ts.tv_nsec * 1000; + times[0].tv_usec = atime_ts.tv_nsec / 1000; times[1].tv_sec = mtime_ts.tv_sec; - times[1].tv_usec = mtime_ts.tv_nsec * 1000; + times[1].tv_usec = mtime_ts.tv_nsec / 1000; if (attrs->ia_valid & HOSTFS_ATTR_ATIME_SET) { times[0].tv_sec = attrs->ia_atime.tv_sec; - times[0].tv_usec = attrs->ia_atime.tv_nsec * 1000; + times[0].tv_usec = attrs->ia_atime.tv_nsec / 1000; } if (attrs->ia_valid & HOSTFS_ATTR_MTIME_SET) { times[1].tv_sec = attrs->ia_mtime.tv_sec; - times[1].tv_usec = attrs->ia_mtime.tv_nsec * 1000; + times[1].tv_usec = attrs->ia_mtime.tv_nsec / 1000; } if (fd >= 0) { diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c index 00971d99996..f63a699ec65 100644 --- a/fs/hpfs/super.c +++ b/fs/hpfs/super.c @@ -386,6 +386,7 @@ static int hpfs_remount_fs(struct super_block *s, int *flags, char *data) int lowercase, conv, eas, chk, errs, chkdsk, timeshift; int o; struct hpfs_sb_info *sbi = hpfs_sb(s); + char *new_opts = kstrdup(data, GFP_KERNEL); *flags |= MS_NOATIME; @@ -398,15 +399,15 @@ static int hpfs_remount_fs(struct super_block *s, int *flags, char *data) if (!(o = parse_opts(data, &uid, &gid, &umask, &lowercase, &conv, &eas, &chk, &errs, &chkdsk, ×hift))) { printk("HPFS: bad mount options.\n"); - return 1; + goto out_err; } if (o == 2) { hpfs_help(); - return 1; + goto out_err; } if (timeshift != sbi->sb_timeshift) { printk("HPFS: timeshift can't be changed using remount.\n"); - return 1; + goto out_err; } unmark_dirty(s); @@ -419,7 +420,14 @@ static int hpfs_remount_fs(struct super_block *s, int *flags, char *data) if (!(*flags & MS_RDONLY)) mark_dirty(s); + kfree(s->s_options); + s->s_options = new_opts; + return 0; + +out_err: + kfree(new_opts); + return -EINVAL; } /* Super operations */ @@ -432,6 +440,7 @@ static const struct super_operations hpfs_sops = .put_super = hpfs_put_super, .statfs = hpfs_statfs, .remount_fs = hpfs_remount_fs, + .show_options = generic_show_options, }; static int hpfs_fill_super(struct super_block *s, void *options, int silent) @@ -454,6 +463,8 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent) int o; + save_mount_options(s, options); + sbi = kzalloc(sizeof(*sbi), GFP_KERNEL); if (!sbi) return -ENOMEM; diff --git a/fs/hppfs/hppfs_kern.c b/fs/hppfs/hppfs_kern.c index affb7412125..8601d8ef3b5 100644 --- a/fs/hppfs/hppfs_kern.c +++ b/fs/hppfs/hppfs_kern.c @@ -1,23 +1,24 @@ /* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -#include <linux/fs.h> +#include <linux/ctype.h> +#include <linux/dcache.h> #include <linux/file.h> -#include <linux/module.h> +#include <linux/fs.h> #include <linux/init.h> -#include <linux/slab.h> -#include <linux/list.h> #include <linux/kernel.h> -#include <linux/ctype.h> -#include <linux/dcache.h> +#include <linux/list.h> +#include <linux/module.h> +#include <linux/mount.h> +#include <linux/slab.h> #include <linux/statfs.h> +#include <linux/types.h> #include <asm/uaccess.h> -#include <asm/fcntl.h> #include "os.h" -static int init_inode(struct inode *inode, struct dentry *dentry); +static struct inode *get_inode(struct super_block *, struct dentry *); struct hppfs_data { struct list_head list; @@ -51,14 +52,14 @@ static int is_pid(struct dentry *dentry) int i; sb = dentry->d_sb; - if((sb->s_op != &hppfs_sbops) || (dentry->d_parent != sb->s_root)) - return(0); + if ((sb->s_op != &hppfs_sbops) || (dentry->d_parent != sb->s_root)) + return 0; - for(i = 0; i < dentry->d_name.len; i++){ - if(!isdigit(dentry->d_name.name[i])) - return(0); + for (i = 0; i < dentry->d_name.len; i++) { + if (!isdigit(dentry->d_name.name[i])) + return 0; } - return(1); + return 1; } static char *dentry_name(struct dentry *dentry, int extra) @@ -70,8 +71,8 @@ static char *dentry_name(struct dentry *dentry, int extra) len = 0; parent = dentry; - while(parent->d_parent != parent){ - if(is_pid(parent)) + while (parent->d_parent != parent) { + if (is_pid(parent)) len += strlen("pid") + 1; else len += parent->d_name.len + 1; parent = parent->d_parent; @@ -80,12 +81,13 @@ static char *dentry_name(struct dentry *dentry, int extra) root = "proc"; len += strlen(root); name = kmalloc(len + extra + 1, GFP_KERNEL); - if(name == NULL) return(NULL); + if (name == NULL) + return NULL; name[len] = '\0'; parent = dentry; - while(parent->d_parent != parent){ - if(is_pid(parent)){ + while (parent->d_parent != parent) { + if (is_pid(parent)) { seg_name = "pid"; seg_len = strlen("pid"); } @@ -100,27 +102,25 @@ static char *dentry_name(struct dentry *dentry, int extra) parent = parent->d_parent; } strncpy(name, root, strlen(root)); - return(name); + return name; } -struct dentry_operations hppfs_dentry_ops = { -}; - static int file_removed(struct dentry *dentry, const char *file) { char *host_file; int extra, fd; extra = 0; - if(file != NULL) extra += strlen(file) + 1; + if (file != NULL) + extra += strlen(file) + 1; host_file = dentry_name(dentry, extra + strlen("/remove")); - if(host_file == NULL){ - printk("file_removed : allocation failed\n"); - return(-ENOMEM); + if (host_file == NULL) { + printk(KERN_ERR "file_removed : allocation failed\n"); + return -ENOMEM; } - if(file != NULL){ + if (file != NULL) { strcat(host_file, "/"); strcat(host_file, file); } @@ -128,31 +128,11 @@ static int file_removed(struct dentry *dentry, const char *file) fd = os_open_file(host_file, of_read(OPENFLAGS()), 0); kfree(host_file); - if(fd > 0){ + if (fd > 0) { os_close_file(fd); - return(1); + return 1; } - return(0); -} - -static void hppfs_read_inode(struct inode *ino) -{ - struct inode *proc_ino; - - if(HPPFS_I(ino)->proc_dentry == NULL) - return; - - proc_ino = HPPFS_I(ino)->proc_dentry->d_inode; - ino->i_uid = proc_ino->i_uid; - ino->i_gid = proc_ino->i_gid; - ino->i_atime = proc_ino->i_atime; - ino->i_mtime = proc_ino->i_mtime; - ino->i_ctime = proc_ino->i_ctime; - ino->i_ino = proc_ino->i_ino; - ino->i_mode = proc_ino->i_mode; - ino->i_nlink = proc_ino->i_nlink; - ino->i_size = proc_ino->i_size; - ino->i_blocks = proc_ino->i_blocks; + return 0; } static struct dentry *hppfs_lookup(struct inode *ino, struct dentry *dentry, @@ -163,53 +143,45 @@ static struct dentry *hppfs_lookup(struct inode *ino, struct dentry *dentry, int err, deleted; deleted = file_removed(dentry, NULL); - if(deleted < 0) - return(ERR_PTR(deleted)); - else if(deleted) - return(ERR_PTR(-ENOENT)); + if (deleted < 0) + return ERR_PTR(deleted); + else if (deleted) + return ERR_PTR(-ENOENT); err = -ENOMEM; parent = HPPFS_I(ino)->proc_dentry; mutex_lock(&parent->d_inode->i_mutex); proc_dentry = d_lookup(parent, &dentry->d_name); - if(proc_dentry == NULL){ + if (proc_dentry == NULL) { proc_dentry = d_alloc(parent, &dentry->d_name); - if(proc_dentry == NULL){ + if (proc_dentry == NULL) { mutex_unlock(&parent->d_inode->i_mutex); goto out; } new = (*parent->d_inode->i_op->lookup)(parent->d_inode, proc_dentry, NULL); - if(new){ + if (new) { dput(proc_dentry); proc_dentry = new; } } mutex_unlock(&parent->d_inode->i_mutex); - if(IS_ERR(proc_dentry)) - return(proc_dentry); + if (IS_ERR(proc_dentry)) + return proc_dentry; - inode = iget(ino->i_sb, 0); - if(inode == NULL) + err = -ENOMEM; + inode = get_inode(ino->i_sb, proc_dentry); + if (!inode) goto out_dput; - err = init_inode(inode, proc_dentry); - if(err) - goto out_put; - - hppfs_read_inode(inode); - d_add(dentry, inode); - dentry->d_op = &hppfs_dentry_ops; - return(NULL); + return NULL; - out_put: - iput(inode); out_dput: dput(proc_dentry); out: - return(ERR_PTR(err)); + return ERR_PTR(err); } static const struct inode_operations hppfs_file_iops = { @@ -223,15 +195,16 @@ static ssize_t read_proc(struct file *file, char __user *buf, ssize_t count, read = file->f_path.dentry->d_inode->i_fop->read; - if(!is_user) + if (!is_user) set_fs(KERNEL_DS); n = (*read)(file, buf, count, &file->f_pos); - if(!is_user) + if (!is_user) set_fs(USER_DS); - if(ppos) *ppos = file->f_pos; + if (ppos) + *ppos = file->f_pos; return n; } @@ -243,24 +216,23 @@ static ssize_t hppfs_read_file(int fd, char __user *buf, ssize_t count) n = -ENOMEM; new_buf = kmalloc(PAGE_SIZE, GFP_KERNEL); - if(new_buf == NULL){ - printk("hppfs_read_file : kmalloc failed\n"); + if (new_buf == NULL) { + printk(KERN_ERR "hppfs_read_file : kmalloc failed\n"); goto out; } n = 0; - while(count > 0){ + while (count > 0) { cur = min_t(ssize_t, count, PAGE_SIZE); err = os_read_file(fd, new_buf, cur); - if(err < 0){ - printk("hppfs_read : read failed, errno = %d\n", - err); + if (err < 0) { + printk(KERN_ERR "hppfs_read : read failed, " + "errno = %d\n", err); n = err; goto out_free; - } - else if(err == 0) + } else if (err == 0) break; - if(copy_to_user(buf, new_buf, err)){ + if (copy_to_user(buf, new_buf, err)) { n = -EFAULT; goto out_free; } @@ -281,35 +253,36 @@ static ssize_t hppfs_read(struct file *file, char __user *buf, size_t count, loff_t off; int err; - if(hppfs->contents != NULL){ - if(*ppos >= hppfs->len) return(0); + if (hppfs->contents != NULL) { + if (*ppos >= hppfs->len) + return 0; data = hppfs->contents; off = *ppos; - while(off >= sizeof(data->contents)){ + while (off >= sizeof(data->contents)) { data = list_entry(data->list.next, struct hppfs_data, list); off -= sizeof(data->contents); } - if(off + count > hppfs->len) + if (off + count > hppfs->len) count = hppfs->len - off; copy_to_user(buf, &data->contents[off], count); *ppos += count; - } - else if(hppfs->host_fd != -1){ + } else if (hppfs->host_fd != -1) { err = os_seek_file(hppfs->host_fd, *ppos); - if(err){ - printk("hppfs_read : seek failed, errno = %d\n", err); - return(err); + if (err) { + printk(KERN_ERR "hppfs_read : seek failed, " + "errno = %d\n", err); + return err; } count = hppfs_read_file(hppfs->host_fd, buf, count); - if(count > 0) + if (count > 0) *ppos += count; } else count = read_proc(hppfs->proc_file, buf, count, ppos, 1); - return(count); + return count; } static ssize_t hppfs_write(struct file *file, const char __user *buf, size_t len, @@ -326,7 +299,7 @@ static ssize_t hppfs_write(struct file *file, const char __user *buf, size_t len err = (*write)(proc_file, buf, len, &proc_file->f_pos); file->f_pos = proc_file->f_pos; - return(err); + return err; } static int open_host_sock(char *host_file, int *filter_out) @@ -338,13 +311,13 @@ static int open_host_sock(char *host_file, int *filter_out) strcpy(end, "/rw"); *filter_out = 1; fd = os_connect_socket(host_file); - if(fd > 0) - return(fd); + if (fd > 0) + return fd; strcpy(end, "/r"); *filter_out = 0; fd = os_connect_socket(host_file); - return(fd); + return fd; } static void free_contents(struct hppfs_data *head) @@ -352,9 +325,10 @@ static void free_contents(struct hppfs_data *head) struct hppfs_data *data; struct list_head *ele, *next; - if(head == NULL) return; + if (head == NULL) + return; - list_for_each_safe(ele, next, &head->list){ + list_for_each_safe(ele, next, &head->list) { data = list_entry(ele, struct hppfs_data, list); kfree(data); } @@ -371,8 +345,8 @@ static struct hppfs_data *hppfs_get_data(int fd, int filter, err = -ENOMEM; data = kmalloc(sizeof(*data), GFP_KERNEL); - if(data == NULL){ - printk("hppfs_get_data : head allocation failed\n"); + if (data == NULL) { + printk(KERN_ERR "hppfs_get_data : head allocation failed\n"); goto failed; } @@ -381,36 +355,36 @@ static struct hppfs_data *hppfs_get_data(int fd, int filter, head = data; *size_out = 0; - if(filter){ - while((n = read_proc(proc_file, data->contents, + if (filter) { + while ((n = read_proc(proc_file, data->contents, sizeof(data->contents), NULL, 0)) > 0) os_write_file(fd, data->contents, n); err = os_shutdown_socket(fd, 0, 1); - if(err){ - printk("hppfs_get_data : failed to shut down " + if (err) { + printk(KERN_ERR "hppfs_get_data : failed to shut down " "socket\n"); goto failed_free; } } - while(1){ + while (1) { n = os_read_file(fd, data->contents, sizeof(data->contents)); - if(n < 0){ + if (n < 0) { err = n; - printk("hppfs_get_data : read failed, errno = %d\n", - err); + printk(KERN_ERR "hppfs_get_data : read failed, " + "errno = %d\n", err); goto failed_free; - } - else if(n == 0) + } else if (n == 0) break; *size_out += n; - if(n < sizeof(data->contents)) + if (n < sizeof(data->contents)) break; new = kmalloc(sizeof(*data), GFP_KERNEL); - if(new == 0){ - printk("hppfs_get_data : data allocation failed\n"); + if (new == 0) { + printk(KERN_ERR "hppfs_get_data : data allocation " + "failed\n"); err = -ENOMEM; goto failed_free; } @@ -419,12 +393,12 @@ static struct hppfs_data *hppfs_get_data(int fd, int filter, list_add(&new->list, &data->list); data = new; } - return(head); + return head; failed_free: free_contents(head); failed: - return(ERR_PTR(err)); + return ERR_PTR(err); } static struct hppfs_private *hppfs_data(void) @@ -432,77 +406,79 @@ static struct hppfs_private *hppfs_data(void) struct hppfs_private *data; data = kmalloc(sizeof(*data), GFP_KERNEL); - if(data == NULL) - return(data); + if (data == NULL) + return data; *data = ((struct hppfs_private ) { .host_fd = -1, .len = -1, .contents = NULL } ); - return(data); + return data; } static int file_mode(int fmode) { - if(fmode == (FMODE_READ | FMODE_WRITE)) - return(O_RDWR); - if(fmode == FMODE_READ) - return(O_RDONLY); - if(fmode == FMODE_WRITE) - return(O_WRONLY); - return(0); + if (fmode == (FMODE_READ | FMODE_WRITE)) + return O_RDWR; + if (fmode == FMODE_READ) + return O_RDONLY; + if (fmode == FMODE_WRITE) + return O_WRONLY; + return 0; } static int hppfs_open(struct inode *inode, struct file *file) { struct hppfs_private *data; struct dentry *proc_dentry; + struct vfsmount *proc_mnt; char *host_file; int err, fd, type, filter; err = -ENOMEM; data = hppfs_data(); - if(data == NULL) + if (data == NULL) goto out; host_file = dentry_name(file->f_path.dentry, strlen("/rw")); - if(host_file == NULL) + if (host_file == NULL) goto out_free2; proc_dentry = HPPFS_I(inode)->proc_dentry; + proc_mnt = inode->i_sb->s_fs_info; /* XXX This isn't closed anywhere */ - data->proc_file = dentry_open(dget(proc_dentry), NULL, + data->proc_file = dentry_open(dget(proc_dentry), mntget(proc_mnt), file_mode(file->f_mode)); err = PTR_ERR(data->proc_file); - if(IS_ERR(data->proc_file)) + if (IS_ERR(data->proc_file)) goto out_free1; type = os_file_type(host_file); - if(type == OS_TYPE_FILE){ + if (type == OS_TYPE_FILE) { fd = os_open_file(host_file, of_read(OPENFLAGS()), 0); - if(fd >= 0) + if (fd >= 0) data->host_fd = fd; - else printk("hppfs_open : failed to open '%s', errno = %d\n", - host_file, -fd); + else + printk(KERN_ERR "hppfs_open : failed to open '%s', " + "errno = %d\n", host_file, -fd); data->contents = NULL; - } - else if(type == OS_TYPE_DIR){ + } else if (type == OS_TYPE_DIR) { fd = open_host_sock(host_file, &filter); - if(fd > 0){ + if (fd > 0) { data->contents = hppfs_get_data(fd, filter, data->proc_file, file, &data->len); - if(!IS_ERR(data->contents)) + if (!IS_ERR(data->contents)) data->host_fd = fd; - } - else printk("hppfs_open : failed to open a socket in " - "'%s', errno = %d\n", host_file, -fd); + } else + printk(KERN_ERR "hppfs_open : failed to open a socket " + "in '%s', errno = %d\n", host_file, -fd); } kfree(host_file); file->private_data = data; - return(0); + return 0; out_free1: kfree(host_file); @@ -510,34 +486,36 @@ static int hppfs_open(struct inode *inode, struct file *file) free_contents(data->contents); kfree(data); out: - return(err); + return err; } static int hppfs_dir_open(struct inode *inode, struct file *file) { struct hppfs_private *data; struct dentry *proc_dentry; + struct vfsmount *proc_mnt; int err; err = -ENOMEM; data = hppfs_data(); - if(data == NULL) + if (data == NULL) goto out; proc_dentry = HPPFS_I(inode)->proc_dentry; - data->proc_file = dentry_open(dget(proc_dentry), NULL, + proc_mnt = inode->i_sb->s_fs_info; + data->proc_file = dentry_open(dget(proc_dentry), mntget(proc_mnt), file_mode(file->f_mode)); err = PTR_ERR(data->proc_file); - if(IS_ERR(data->proc_file)) + if (IS_ERR(data->proc_file)) goto out_free; file->private_data = data; - return(0); + return 0; out_free: kfree(data); out: - return(err); + return err; } static loff_t hppfs_llseek(struct file *file, loff_t off, int where) @@ -548,13 +526,13 @@ static loff_t hppfs_llseek(struct file *file, loff_t off, int where) loff_t ret; llseek = proc_file->f_path.dentry->d_inode->i_fop->llseek; - if(llseek != NULL){ + if (llseek != NULL) { ret = (*llseek)(proc_file, off, where); - if(ret < 0) - return(ret); + if (ret < 0) + return ret; } - return(default_llseek(file, off, where)); + return default_llseek(file, off, where); } static const struct file_operations hppfs_file_fops = { @@ -576,11 +554,11 @@ static int hppfs_filldir(void *d, const char *name, int size, { struct hppfs_dirent *dirent = d; - if(file_removed(dirent->dentry, name)) - return(0); + if (file_removed(dirent->dentry, name)) + return 0; - return((*dirent->filldir)(dirent->vfs_dirent, name, size, offset, - inode, type)); + return (*dirent->filldir)(dirent->vfs_dirent, name, size, offset, + inode, type); } static int hppfs_readdir(struct file *file, void *ent, filldir_t filldir) @@ -591,7 +569,8 @@ static int hppfs_readdir(struct file *file, void *ent, filldir_t filldir) struct hppfs_dirent dirent = ((struct hppfs_dirent) { .vfs_dirent = ent, .filldir = filldir, - .dentry = file->f_path.dentry } ); + .dentry = file->f_path.dentry + }); int err; readdir = proc_file->f_path.dentry->d_inode->i_fop->readdir; @@ -600,12 +579,12 @@ static int hppfs_readdir(struct file *file, void *ent, filldir_t filldir) err = (*readdir)(proc_file, &dirent, hppfs_filldir); file->f_pos = proc_file->f_pos; - return(err); + return err; } static int hppfs_fsync(struct file *file, struct dentry *dentry, int datasync) { - return(0); + return 0; } static const struct file_operations hppfs_dir_fops = { @@ -623,7 +602,7 @@ static int hppfs_statfs(struct dentry *dentry, struct kstatfs *sf) sf->f_files = 0; sf->f_ffree = 0; sf->f_type = HPPFS_SUPER_MAGIC; - return(0); + return 0; } static struct inode *hppfs_alloc_inode(struct super_block *sb) @@ -631,12 +610,12 @@ static struct inode *hppfs_alloc_inode(struct super_block *sb) struct hppfs_inode_info *hi; hi = kmalloc(sizeof(*hi), GFP_KERNEL); - if(hi == NULL) - return(NULL); + if (!hi) + return NULL; - *hi = ((struct hppfs_inode_info) { .proc_dentry = NULL }); + hi->proc_dentry = NULL; inode_init_once(&hi->vfs_inode); - return(&hi->vfs_inode); + return &hi->vfs_inode; } void hppfs_delete_inode(struct inode *ino) @@ -649,22 +628,31 @@ static void hppfs_destroy_inode(struct inode *inode) kfree(HPPFS_I(inode)); } +static void hppfs_put_super(struct super_block *sb) +{ + mntput(sb->s_fs_info); +} + static const struct super_operations hppfs_sbops = { .alloc_inode = hppfs_alloc_inode, .destroy_inode = hppfs_destroy_inode, - .read_inode = hppfs_read_inode, .delete_inode = hppfs_delete_inode, .statfs = hppfs_statfs, + .put_super = hppfs_put_super, }; -static int hppfs_readlink(struct dentry *dentry, char __user *buffer, int buflen) +static int hppfs_readlink(struct dentry *dentry, char __user *buffer, + int buflen) { struct file *proc_file; struct dentry *proc_dentry; + struct vfsmount *proc_mnt; int ret; proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry; - proc_file = dentry_open(dget(proc_dentry), NULL, O_RDONLY); + proc_mnt = dentry->d_sb->s_fs_info; + + proc_file = dentry_open(dget(proc_dentry), mntget(proc_mnt), O_RDONLY); if (IS_ERR(proc_file)) return PTR_ERR(proc_file); @@ -679,10 +667,13 @@ static void* hppfs_follow_link(struct dentry *dentry, struct nameidata *nd) { struct file *proc_file; struct dentry *proc_dentry; + struct vfsmount *proc_mnt; void *ret; proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry; - proc_file = dentry_open(dget(proc_dentry), NULL, O_RDONLY); + proc_mnt = dentry->d_sb->s_fs_info; + + proc_file = dentry_open(dget(proc_dentry), mntget(proc_mnt), O_RDONLY); if (IS_ERR(proc_file)) return proc_file; @@ -702,68 +693,72 @@ static const struct inode_operations hppfs_link_iops = { .follow_link = hppfs_follow_link, }; -static int init_inode(struct inode *inode, struct dentry *dentry) +static struct inode *get_inode(struct super_block *sb, struct dentry *dentry) { - if(S_ISDIR(dentry->d_inode->i_mode)){ + struct inode *proc_ino = dentry->d_inode; + struct inode *inode = new_inode(sb); + + if (!inode) + return ERR_PTR(-ENOMEM); + + if (S_ISDIR(dentry->d_inode->i_mode)) { inode->i_op = &hppfs_dir_iops; inode->i_fop = &hppfs_dir_fops; - } - else if(S_ISLNK(dentry->d_inode->i_mode)){ + } else if (S_ISLNK(dentry->d_inode->i_mode)) { inode->i_op = &hppfs_link_iops; inode->i_fop = &hppfs_file_fops; - } - else { + } else { inode->i_op = &hppfs_file_iops; inode->i_fop = &hppfs_file_fops; } HPPFS_I(inode)->proc_dentry = dentry; - return(0); + inode->i_uid = proc_ino->i_uid; + inode->i_gid = proc_ino->i_gid; + inode->i_atime = proc_ino->i_atime; + inode->i_mtime = proc_ino->i_mtime; + inode->i_ctime = proc_ino->i_ctime; + inode->i_ino = proc_ino->i_ino; + inode->i_mode = proc_ino->i_mode; + inode->i_nlink = proc_ino->i_nlink; + inode->i_size = proc_ino->i_size; + inode->i_blocks = proc_ino->i_blocks; + + return 0; } static int hppfs_fill_super(struct super_block *sb, void *d, int silent) { struct inode *root_inode; - struct file_system_type *procfs; - struct super_block *proc_sb; - int err; - - err = -ENOENT; - procfs = get_fs_type("proc"); - if(procfs == NULL) - goto out; + struct vfsmount *proc_mnt; + int err = -ENOENT; - if(list_empty(&procfs->fs_supers)) + proc_mnt = do_kern_mount("proc", 0, "proc", NULL); + if (IS_ERR(proc_mnt)) goto out; - proc_sb = list_entry(procfs->fs_supers.next, struct super_block, - s_instances); - sb->s_blocksize = 1024; sb->s_blocksize_bits = 10; sb->s_magic = HPPFS_SUPER_MAGIC; sb->s_op = &hppfs_sbops; - - root_inode = iget(sb, 0); - if(root_inode == NULL) - goto out; - - err = init_inode(root_inode, proc_sb->s_root); - if(err) - goto out_put; + sb->s_fs_info = proc_mnt; err = -ENOMEM; - sb->s_root = d_alloc_root(root_inode); - if(sb->s_root == NULL) - goto out_put; + root_inode = get_inode(sb, proc_mnt->mnt_sb->s_root); + if (!root_inode) + goto out_mntput; - hppfs_read_inode(root_inode); + sb->s_root = d_alloc_root(root_inode); + if (!sb->s_root) + goto out_iput; - return(0); + return 0; - out_put: + out_iput: iput(root_inode); + out_mntput: + mntput(proc_mnt); out: return(err); } @@ -785,7 +780,7 @@ static struct file_system_type hppfs_type = { static int __init init_hppfs(void) { - return(register_filesystem(&hppfs_type)); + return register_filesystem(&hppfs_type); } static void __exit exit_hppfs(void) @@ -796,14 +791,3 @@ static void __exit exit_hppfs(void) module_init(init_hppfs) module_exit(exit_hppfs) MODULE_LICENSE("GPL"); - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 09ee07f0266..6846785fe90 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -734,6 +734,7 @@ static const struct super_operations hugetlbfs_ops = { .delete_inode = hugetlbfs_delete_inode, .drop_inode = hugetlbfs_drop_inode, .put_super = hugetlbfs_put_super, + .show_options = generic_show_options, }; static int @@ -768,7 +769,7 @@ hugetlbfs_parse_options(char *options, struct hugetlbfs_config *pconfig) case Opt_mode: if (match_octal(&args[0], &option)) goto bad_val; - pconfig->mode = option & 0777U; + pconfig->mode = option & 01777U; break; case Opt_size: { @@ -817,6 +818,8 @@ hugetlbfs_fill_super(struct super_block *sb, void *data, int silent) struct hugetlbfs_config config; struct hugetlbfs_sb_info *sbinfo; + save_mount_options(sb, data); + config.nr_blocks = -1; /* No limit on size by default */ config.nr_inodes = -1; /* No limit on number of inodes by default */ config.uid = current->fsuid; @@ -951,7 +954,7 @@ struct file *hugetlb_file_setup(const char *name, size_t size) FMODE_WRITE | FMODE_READ, &hugetlbfs_file_operations); if (!file) - goto out_inode; + goto out_dentry; /* inode is already attached */ return file; diff --git a/fs/inode.c b/fs/inode.c index ed35383d0b6..53245ffcf93 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -928,8 +928,6 @@ EXPORT_SYMBOL(ilookup); * @set: callback used to initialize a new struct inode * @data: opaque data pointer to pass to @test and @set * - * This is iget() without the read_inode() portion of get_new_inode(). - * * iget5_locked() uses ifind() to search for the inode specified by @hashval * and @data in the inode cache and if present it is returned with an increased * reference count. This is a generalized version of iget_locked() for file @@ -966,8 +964,6 @@ EXPORT_SYMBOL(iget5_locked); * @sb: super block of file system * @ino: inode number to get * - * This is iget() without the read_inode() portion of get_new_inode_fast(). - * * iget_locked() uses ifind_fast() to search for the inode specified by @ino in * the inode cache and if present it is returned with an increased reference * count. This is for file systems where the inode number is sufficient for @@ -1276,6 +1272,11 @@ void file_update_time(struct file *file) sync_it = 1; } + if (IS_I_VERSION(inode)) { + inode_inc_iversion(inode); + sync_it = 1; + } + if (sync_it) mark_inode_dirty_sync(inode); } diff --git a/fs/inotify.c b/fs/inotify.c index 2c5b9215287..690e72595e6 100644 --- a/fs/inotify.c +++ b/fs/inotify.c @@ -168,20 +168,14 @@ static void set_dentry_child_flags(struct inode *inode, int watched) struct dentry *child; list_for_each_entry(child, &alias->d_subdirs, d_u.d_child) { - if (!child->d_inode) { - WARN_ON(child->d_flags & DCACHE_INOTIFY_PARENT_WATCHED); + if (!child->d_inode) continue; - } + spin_lock(&child->d_lock); - if (watched) { - WARN_ON(child->d_flags & - DCACHE_INOTIFY_PARENT_WATCHED); + if (watched) child->d_flags |= DCACHE_INOTIFY_PARENT_WATCHED; - } else { - WARN_ON(!(child->d_flags & - DCACHE_INOTIFY_PARENT_WATCHED)); - child->d_flags&=~DCACHE_INOTIFY_PARENT_WATCHED; - } + else + child->d_flags &=~DCACHE_INOTIFY_PARENT_WATCHED; spin_unlock(&child->d_lock); } } @@ -253,7 +247,6 @@ void inotify_d_instantiate(struct dentry *entry, struct inode *inode) if (!inode) return; - WARN_ON(entry->d_flags & DCACHE_INOTIFY_PARENT_WATCHED); spin_lock(&entry->d_lock); parent = entry->d_parent; if (parent->d_inode && inotify_inode_watched(parent->d_inode)) @@ -627,6 +620,7 @@ s32 inotify_add_watch(struct inotify_handle *ih, struct inotify_watch *watch, struct inode *inode, u32 mask) { int ret = 0; + int newly_watched; /* don't allow invalid bits: we don't want flags set */ mask &= IN_ALL_EVENTS | IN_ONESHOT; @@ -653,12 +647,18 @@ s32 inotify_add_watch(struct inotify_handle *ih, struct inotify_watch *watch, */ watch->inode = igrab(inode); - if (!inotify_inode_watched(inode)) - set_dentry_child_flags(inode, 1); - /* Add the watch to the handle's and the inode's list */ + newly_watched = !inotify_inode_watched(inode); list_add(&watch->h_list, &ih->watches); list_add(&watch->i_list, &inode->inotify_watches); + /* + * Set child flags _after_ adding the watch, so there is no race + * windows where newly instantiated children could miss their parent's + * watched flag. + */ + if (newly_watched) + set_dentry_child_flags(inode, 1); + out: mutex_unlock(&ih->mutex); mutex_unlock(&inode->inotify_mutex); diff --git a/fs/inotify_user.c b/fs/inotify_user.c index 5e009331c01..7b94a1e3c01 100644 --- a/fs/inotify_user.c +++ b/fs/inotify_user.c @@ -41,9 +41,9 @@ static struct kmem_cache *event_cachep __read_mostly; static struct vfsmount *inotify_mnt __read_mostly; /* these are configurable via /proc/sys/fs/inotify/ */ -int inotify_max_user_instances __read_mostly; -int inotify_max_user_watches __read_mostly; -int inotify_max_queued_events __read_mostly; +static int inotify_max_user_instances __read_mostly; +static int inotify_max_user_watches __read_mostly; +static int inotify_max_queued_events __read_mostly; /* * Lock ordering: @@ -79,6 +79,7 @@ struct inotify_device { atomic_t count; /* reference count */ struct user_struct *user; /* user who opened this dev */ struct inotify_handle *ih; /* inotify handle */ + struct fasync_struct *fa; /* async notification */ unsigned int queue_size; /* size of the queue (bytes) */ unsigned int event_count; /* number of pending events */ unsigned int max_events; /* maximum number of events */ @@ -248,6 +249,19 @@ inotify_dev_get_event(struct inotify_device *dev) } /* + * inotify_dev_get_last_event - return the last event in the given dev's queue + * + * Caller must hold dev->ev_mutex. + */ +static inline struct inotify_kernel_event * +inotify_dev_get_last_event(struct inotify_device *dev) +{ + if (list_empty(&dev->events)) + return NULL; + return list_entry(dev->events.prev, struct inotify_kernel_event, list); +} + +/* * inotify_dev_queue_event - event handler registered with core inotify, adds * a new event to the given device * @@ -269,11 +283,11 @@ static void inotify_dev_queue_event(struct inotify_watch *w, u32 wd, u32 mask, /* we can safely put the watch as we don't reference it while * generating the event */ - if (mask & IN_IGNORED || mask & IN_ONESHOT) + if (mask & IN_IGNORED || w->mask & IN_ONESHOT) put_inotify_watch(w); /* final put */ /* coalescing: drop this event if it is a dupe of the previous */ - last = inotify_dev_get_event(dev); + last = inotify_dev_get_last_event(dev); if (last && last->event.mask == mask && last->event.wd == wd && last->event.cookie == cookie) { const char *lastname = last->name; @@ -302,6 +316,7 @@ static void inotify_dev_queue_event(struct inotify_watch *w, u32 wd, u32 mask, dev->queue_size += sizeof(struct inotify_event) + kevent->event.len; list_add_tail(&kevent->list, &dev->events); wake_up_interruptible(&dev->wq); + kill_fasync(&dev->fa, SIGIO, POLL_IN); out: mutex_unlock(&dev->ev_mutex); @@ -352,7 +367,7 @@ static int find_inode(const char __user *dirname, struct nameidata *nd, /* you can only watch an inode if you have read permissions on it */ error = vfs_permission(nd, MAY_READ); if (error) - path_release(nd); + path_put(&nd->path); return error; } @@ -490,6 +505,13 @@ static ssize_t inotify_read(struct file *file, char __user *buf, return ret; } +static int inotify_fasync(int fd, struct file *file, int on) +{ + struct inotify_device *dev = file->private_data; + + return fasync_helper(fd, file, on, &dev->fa) >= 0 ? 0 : -EIO; +} + static int inotify_release(struct inode *ignored, struct file *file) { struct inotify_device *dev = file->private_data; @@ -502,6 +524,9 @@ static int inotify_release(struct inode *ignored, struct file *file) inotify_dev_event_dequeue(dev); mutex_unlock(&dev->ev_mutex); + if (file->f_flags & FASYNC) + inotify_fasync(-1, file, 0); + /* free this device: the put matching the get in inotify_init() */ put_inotify_dev(dev); @@ -530,6 +555,7 @@ static long inotify_ioctl(struct file *file, unsigned int cmd, static const struct file_operations inotify_fops = { .poll = inotify_poll, .read = inotify_read, + .fasync = inotify_fasync, .release = inotify_release, .unlocked_ioctl = inotify_ioctl, .compat_ioctl = inotify_ioctl, @@ -577,6 +603,7 @@ asmlinkage long sys_inotify_init(void) goto out_free_dev; } dev->ih = ih; + dev->fa = NULL; filp->f_op = &inotify_fops; filp->f_path.mnt = mntget(inotify_mnt); @@ -640,7 +667,7 @@ asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask) goto fput_and_out; /* inode held in place by reference to nd; dev by fget on fd */ - inode = nd.dentry->d_inode; + inode = nd.path.dentry->d_inode; dev = filp->private_data; mutex_lock(&dev->up_mutex); @@ -649,7 +676,7 @@ asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask) ret = create_watch(dev, inode, mask); mutex_unlock(&dev->up_mutex); - path_release(&nd); + path_put(&nd.path); fput_and_out: fput_light(filp, fput_needed); return ret; diff --git a/fs/ioctl.c b/fs/ioctl.c index c2a773e8620..f32fbde2175 100644 --- a/fs/ioctl.c +++ b/fs/ioctl.c @@ -12,12 +12,24 @@ #include <linux/fs.h> #include <linux/security.h> #include <linux/module.h> +#include <linux/uaccess.h> -#include <asm/uaccess.h> #include <asm/ioctls.h> -static long do_ioctl(struct file *filp, unsigned int cmd, - unsigned long arg) +/** + * vfs_ioctl - call filesystem specific ioctl methods + * @filp: open file to invoke ioctl method on + * @cmd: ioctl command to execute + * @arg: command-specific argument for ioctl + * + * Invokes filesystem specific ->unlocked_ioctl, if one exists; otherwise + * invokes filesystem specific ->ioctl method. If neither method exists, + * returns -ENOTTY. + * + * Returns 0 on success, -errno on error. + */ +long vfs_ioctl(struct file *filp, unsigned int cmd, + unsigned long arg) { int error = -ENOTTY; @@ -40,123 +52,148 @@ static long do_ioctl(struct file *filp, unsigned int cmd, return error; } +static int ioctl_fibmap(struct file *filp, int __user *p) +{ + struct address_space *mapping = filp->f_mapping; + int res, block; + + /* do we support this mess? */ + if (!mapping->a_ops->bmap) + return -EINVAL; + if (!capable(CAP_SYS_RAWIO)) + return -EPERM; + res = get_user(block, p); + if (res) + return res; + lock_kernel(); + res = mapping->a_ops->bmap(mapping, block); + unlock_kernel(); + return put_user(res, p); +} + static int file_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { - int error; - int block; - struct inode * inode = filp->f_path.dentry->d_inode; + struct inode *inode = filp->f_path.dentry->d_inode; int __user *p = (int __user *)arg; switch (cmd) { - case FIBMAP: - { - struct address_space *mapping = filp->f_mapping; - int res; - /* do we support this mess? */ - if (!mapping->a_ops->bmap) - return -EINVAL; - if (!capable(CAP_SYS_RAWIO)) - return -EPERM; - if ((error = get_user(block, p)) != 0) - return error; + case FIBMAP: + return ioctl_fibmap(filp, p); + case FIGETBSZ: + return put_user(inode->i_sb->s_blocksize, p); + case FIONREAD: + return put_user(i_size_read(inode) - filp->f_pos, p); + } + return vfs_ioctl(filp, cmd, arg); +} + +static int ioctl_fionbio(struct file *filp, int __user *argp) +{ + unsigned int flag; + int on, error; + + error = get_user(on, argp); + if (error) + return error; + flag = O_NONBLOCK; +#ifdef __sparc__ + /* SunOS compatibility item. */ + if (O_NONBLOCK != O_NDELAY) + flag |= O_NDELAY; +#endif + if (on) + filp->f_flags |= flag; + else + filp->f_flags &= ~flag; + return error; +} + +static int ioctl_fioasync(unsigned int fd, struct file *filp, + int __user *argp) +{ + unsigned int flag; + int on, error; + + error = get_user(on, argp); + if (error) + return error; + flag = on ? FASYNC : 0; + + /* Did FASYNC state change ? */ + if ((flag ^ filp->f_flags) & FASYNC) { + if (filp->f_op && filp->f_op->fasync) { lock_kernel(); - res = mapping->a_ops->bmap(mapping, block); + error = filp->f_op->fasync(fd, filp, on); unlock_kernel(); - return put_user(res, p); - } - case FIGETBSZ: - return put_user(inode->i_sb->s_blocksize, p); - case FIONREAD: - return put_user(i_size_read(inode) - filp->f_pos, p); + } else + error = -ENOTTY; } + if (error) + return error; - return do_ioctl(filp, cmd, arg); + if (on) + filp->f_flags |= FASYNC; + else + filp->f_flags &= ~FASYNC; + return error; } /* * When you add any new common ioctls to the switches above and below * please update compat_sys_ioctl() too. * - * vfs_ioctl() is not for drivers and not intended to be EXPORT_SYMBOL()'d. + * do_vfs_ioctl() is not for drivers and not intended to be EXPORT_SYMBOL()'d. * It's just a simple helper for sys_ioctl and compat_sys_ioctl. */ -int vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd, unsigned long arg) +int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd, + unsigned long arg) { - unsigned int flag; - int on, error = 0; + int error = 0; + int __user *argp = (int __user *)arg; switch (cmd) { - case FIOCLEX: - set_close_on_exec(fd, 1); - break; - - case FIONCLEX: - set_close_on_exec(fd, 0); - break; - - case FIONBIO: - if ((error = get_user(on, (int __user *)arg)) != 0) - break; - flag = O_NONBLOCK; -#ifdef __sparc__ - /* SunOS compatibility item. */ - if(O_NONBLOCK != O_NDELAY) - flag |= O_NDELAY; -#endif - if (on) - filp->f_flags |= flag; - else - filp->f_flags &= ~flag; - break; - - case FIOASYNC: - if ((error = get_user(on, (int __user *)arg)) != 0) - break; - flag = on ? FASYNC : 0; - - /* Did FASYNC state change ? */ - if ((flag ^ filp->f_flags) & FASYNC) { - if (filp->f_op && filp->f_op->fasync) { - lock_kernel(); - error = filp->f_op->fasync(fd, filp, on); - unlock_kernel(); - } - else error = -ENOTTY; - } - if (error != 0) - break; - - if (on) - filp->f_flags |= FASYNC; - else - filp->f_flags &= ~FASYNC; - break; - - case FIOQSIZE: - if (S_ISDIR(filp->f_path.dentry->d_inode->i_mode) || - S_ISREG(filp->f_path.dentry->d_inode->i_mode) || - S_ISLNK(filp->f_path.dentry->d_inode->i_mode)) { - loff_t res = inode_get_bytes(filp->f_path.dentry->d_inode); - error = copy_to_user((loff_t __user *)arg, &res, sizeof(res)) ? -EFAULT : 0; - } - else - error = -ENOTTY; - break; - default: - if (S_ISREG(filp->f_path.dentry->d_inode->i_mode)) - error = file_ioctl(filp, cmd, arg); - else - error = do_ioctl(filp, cmd, arg); - break; + case FIOCLEX: + set_close_on_exec(fd, 1); + break; + + case FIONCLEX: + set_close_on_exec(fd, 0); + break; + + case FIONBIO: + error = ioctl_fionbio(filp, argp); + break; + + case FIOASYNC: + error = ioctl_fioasync(fd, filp, argp); + break; + + case FIOQSIZE: + if (S_ISDIR(filp->f_path.dentry->d_inode->i_mode) || + S_ISREG(filp->f_path.dentry->d_inode->i_mode) || + S_ISLNK(filp->f_path.dentry->d_inode->i_mode)) { + loff_t res = + inode_get_bytes(filp->f_path.dentry->d_inode); + error = copy_to_user((loff_t __user *)arg, &res, + sizeof(res)) ? -EFAULT : 0; + } else + error = -ENOTTY; + break; + default: + if (S_ISREG(filp->f_path.dentry->d_inode->i_mode)) + error = file_ioctl(filp, cmd, arg); + else + error = vfs_ioctl(filp, cmd, arg); + break; } return error; } asmlinkage long sys_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg) { - struct file * filp; + struct file *filp; int error = -EBADF; int fput_needed; @@ -168,7 +205,7 @@ asmlinkage long sys_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg) if (error) goto out_fput; - error = vfs_ioctl(filp, fd, cmd, arg); + error = do_vfs_ioctl(filp, fd, cmd, arg); out_fput: fput_light(filp, fput_needed); out: diff --git a/fs/ioprio.c b/fs/ioprio.c index e4e01bc7f33..c4a1c3c65aa 100644 --- a/fs/ioprio.c +++ b/fs/ioprio.c @@ -41,18 +41,28 @@ static int set_task_ioprio(struct task_struct *task, int ioprio) return err; task_lock(task); + do { + ioc = task->io_context; + /* see wmb() in current_io_context() */ + smp_read_barrier_depends(); + if (ioc) + break; - task->ioprio = ioprio; - - ioc = task->io_context; - /* see wmb() in current_io_context() */ - smp_read_barrier_depends(); + ioc = alloc_io_context(GFP_ATOMIC, -1); + if (!ioc) { + err = -ENOMEM; + break; + } + task->io_context = ioc; + } while (1); - if (ioc) + if (!err) { + ioc->ioprio = ioprio; ioc->ioprio_changed = 1; + } task_unlock(task); - return 0; + return err; } asmlinkage long sys_ioprio_set(int which, int who, int ioprio) @@ -75,8 +85,6 @@ asmlinkage long sys_ioprio_set(int which, int who, int ioprio) break; case IOPRIO_CLASS_IDLE: - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; break; case IOPRIO_CLASS_NONE: if (data) @@ -148,7 +156,9 @@ static int get_task_ioprio(struct task_struct *p) ret = security_task_getioprio(p); if (ret) goto out; - ret = p->ioprio; + ret = IOPRIO_PRIO_VALUE(IOPRIO_CLASS_NONE, IOPRIO_NORM); + if (p->io_context) + ret = p->io_context->ioprio; out: return ret; } diff --git a/fs/isofs/compress.c b/fs/isofs/compress.c index 37dbd640478..defb932eee9 100644 --- a/fs/isofs/compress.c +++ b/fs/isofs/compress.c @@ -72,6 +72,17 @@ static int zisofs_readpage(struct file *file, struct page *page) offset = index & ~zisofs_block_page_mask; blockindex = offset >> zisofs_block_page_shift; maxpage = (inode->i_size + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + + /* + * If this page is wholly outside i_size we just return zero; + * do_generic_file_read() will handle this for us + */ + if (page->index >= maxpage) { + SetPageUptodate(page); + unlock_page(page); + return 0; + } + maxpage = min(zisofs_block_pages, maxpage-offset); for ( i = 0 ; i < maxpage ; i++, offset++ ) { diff --git a/fs/isofs/export.c b/fs/isofs/export.c index 29f9753ae5e..bb219138331 100644 --- a/fs/isofs/export.c +++ b/fs/isofs/export.c @@ -26,11 +26,9 @@ isofs_export_iget(struct super_block *sb, if (block == 0) return ERR_PTR(-ESTALE); inode = isofs_iget(sb, block, offset); - if (inode == NULL) - return ERR_PTR(-ENOMEM); - if (is_bad_inode(inode) - || (generation && inode->i_generation != generation)) - { + if (IS_ERR(inode)) + return ERR_CAST(inode); + if (generation && inode->i_generation != generation) { iput(inode); return ERR_PTR(-ESTALE); } @@ -110,8 +108,10 @@ static struct dentry *isofs_export_get_parent(struct dentry *child) parent_inode = isofs_iget(child_inode->i_sb, parent_block, parent_offset); - if (parent_inode == NULL) { - rv = ERR_PTR(-EACCES); + if (IS_ERR(parent_inode)) { + rv = ERR_CAST(parent_inode); + if (rv != ERR_PTR(-ENOMEM)) + rv = ERR_PTR(-EACCES); goto out; } diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index 09e3d306e96..044a254d526 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -54,7 +54,7 @@ static void isofs_put_super(struct super_block *sb) return; } -static void isofs_read_inode(struct inode *); +static int isofs_read_inode(struct inode *); static int isofs_statfs (struct dentry *, struct kstatfs *); static struct kmem_cache *isofs_inode_cachep; @@ -107,10 +107,10 @@ static int isofs_remount(struct super_block *sb, int *flags, char *data) static const struct super_operations isofs_sops = { .alloc_inode = isofs_alloc_inode, .destroy_inode = isofs_destroy_inode, - .read_inode = isofs_read_inode, .put_super = isofs_put_super, .statfs = isofs_statfs, .remount_fs = isofs_remount, + .show_options = generic_show_options, }; @@ -145,7 +145,8 @@ struct iso9660_options{ char nocompress; unsigned char check; unsigned int blocksize; - mode_t mode; + mode_t fmode; + mode_t dmode; gid_t gid; uid_t uid; char *iocharset; @@ -306,7 +307,7 @@ enum { Opt_block, Opt_check_r, Opt_check_s, Opt_cruft, Opt_gid, Opt_ignore, Opt_iocharset, Opt_map_a, Opt_map_n, Opt_map_o, Opt_mode, Opt_nojoliet, Opt_norock, Opt_sb, Opt_session, Opt_uid, Opt_unhide, Opt_utf8, Opt_err, - Opt_nocompress, Opt_hide, Opt_showassoc, + Opt_nocompress, Opt_hide, Opt_showassoc, Opt_dmode, }; static match_table_t tokens = { @@ -333,6 +334,7 @@ static match_table_t tokens = { {Opt_uid, "uid=%u"}, {Opt_gid, "gid=%u"}, {Opt_mode, "mode=%u"}, + {Opt_dmode, "dmode=%u"}, {Opt_block, "block=%u"}, {Opt_ignore, "conv=binary"}, {Opt_ignore, "conv=b"}, @@ -360,7 +362,7 @@ static int parse_options(char *options, struct iso9660_options *popt) popt->check = 'u'; /* unset */ popt->nocompress = 0; popt->blocksize = 1024; - popt->mode = S_IRUGO | S_IXUGO; /* + popt->fmode = popt->dmode = S_IRUGO | S_IXUGO; /* * r-x for all. The disc could * be shared with DOS machines so * virtually anything could be @@ -452,7 +454,12 @@ static int parse_options(char *options, struct iso9660_options *popt) case Opt_mode: if (match_int(&args[0], &option)) return 0; - popt->mode = option; + popt->fmode = option; + break; + case Opt_dmode: + if (match_int(&args[0], &option)) + return 0; + popt->dmode = option; break; case Opt_block: if (match_int(&args[0], &option)) @@ -552,9 +559,11 @@ static int isofs_fill_super(struct super_block *s, void *data, int silent) int joliet_level = 0; int iso_blknum, block; int orig_zonesize; - int table; + int table, error = -EINVAL; unsigned int vol_desc_start; + save_mount_options(s, data); + sbi = kzalloc(sizeof(*sbi), GFP_KERNEL); if (!sbi) return -ENOMEM; @@ -802,7 +811,8 @@ root_found: * on the disk as suid, so we merely allow them to set the default * permissions. */ - sbi->s_mode = opt.mode & 0777; + sbi->s_fmode = opt.fmode & 0777; + sbi->s_dmode = opt.dmode & 0777; /* * Read the root inode, which _may_ result in changing @@ -810,6 +820,8 @@ root_found: * we then decide whether to use the Joliet descriptor. */ inode = isofs_iget(s, sbi->s_firstdatazone, 0); + if (IS_ERR(inode)) + goto out_no_root; /* * If this disk has both Rock Ridge and Joliet on it, then we @@ -829,6 +841,8 @@ root_found: "ISOFS: changing to secondary root\n"); iput(inode); inode = isofs_iget(s, sbi->s_firstdatazone, 0); + if (IS_ERR(inode)) + goto out_no_root; } } @@ -842,8 +856,6 @@ root_found: sbi->s_joliet_level = joliet_level; /* check the root inode */ - if (!inode) - goto out_no_root; if (!inode->i_op) goto out_bad_root; @@ -876,11 +888,14 @@ root_found: */ out_bad_root: printk(KERN_WARNING "%s: root inode not initialized\n", __func__); - goto out_iput; -out_no_root: - printk(KERN_WARNING "%s: get root inode failed\n", __func__); out_iput: iput(inode); + goto out_no_inode; +out_no_root: + error = PTR_ERR(inode); + if (error != -ENOMEM) + printk(KERN_WARNING "%s: get root inode failed\n", __func__); +out_no_inode: #ifdef CONFIG_JOLIET if (sbi->s_nls_iocharset) unload_nls(sbi->s_nls_iocharset); @@ -908,7 +923,7 @@ out_freesbi: kfree(opt.iocharset); kfree(sbi); s->s_fs_info = NULL; - return -EINVAL; + return error; } static int isofs_statfs (struct dentry *dentry, struct kstatfs *buf) @@ -930,7 +945,7 @@ static int isofs_statfs (struct dentry *dentry, struct kstatfs *buf) /* * Get a set of blocks; filling in buffer_heads if already allocated * or getblk() if they are not. Returns the number of blocks inserted - * (0 == error.) + * (-ve == error.) */ int isofs_get_blocks(struct inode *inode, sector_t iblock_s, struct buffer_head **bh, unsigned long nblocks) @@ -940,11 +955,12 @@ int isofs_get_blocks(struct inode *inode, sector_t iblock_s, unsigned int firstext; unsigned long nextblk, nextoff; long iblock = (long)iblock_s; - int section, rv; + int section, rv, error; struct iso_inode_info *ei = ISOFS_I(inode); lock_kernel(); + error = -EIO; rv = 0; if (iblock < 0 || iblock != iblock_s) { printk(KERN_DEBUG "%s: block number too large\n", __func__); @@ -983,8 +999,10 @@ int isofs_get_blocks(struct inode *inode, sector_t iblock_s, offset += sect_size; ninode = isofs_iget(inode->i_sb, nextblk, nextoff); - if (!ninode) + if (IS_ERR(ninode)) { + error = PTR_ERR(ninode); goto abort; + } firstext = ISOFS_I(ninode)->i_first_extent; sect_size = ISOFS_I(ninode)->i_section_size >> ISOFS_BUFFER_BITS(ninode); nextblk = ISOFS_I(ninode)->i_next_section_block; @@ -1015,9 +1033,10 @@ int isofs_get_blocks(struct inode *inode, sector_t iblock_s, rv++; } + error = 0; abort: unlock_kernel(); - return rv; + return rv != 0 ? rv : error; } /* @@ -1026,12 +1045,15 @@ abort: static int isofs_get_block(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) { + int ret; + if (create) { printk(KERN_DEBUG "%s: Kernel tries to allocate a block\n", __func__); return -EROFS; } - return isofs_get_blocks(inode, iblock, &bh_result, 1) ? 0 : -EIO; + ret = isofs_get_blocks(inode, iblock, &bh_result, 1); + return ret < 0 ? ret : 0; } static int isofs_bmap(struct inode *inode, sector_t block) @@ -1186,7 +1208,7 @@ out_toomany: goto out; } -static void isofs_read_inode(struct inode *inode) +static int isofs_read_inode(struct inode *inode) { struct super_block *sb = inode->i_sb; struct isofs_sb_info *sbi = ISOFS_SB(sb); @@ -1199,6 +1221,7 @@ static void isofs_read_inode(struct inode *inode) unsigned int de_len; unsigned long offset; struct iso_inode_info *ei = ISOFS_I(inode); + int ret = -EIO; block = ei->i_iget5_block; bh = sb_bread(inode->i_sb, block); @@ -1216,6 +1239,7 @@ static void isofs_read_inode(struct inode *inode) tmpde = kmalloc(de_len, GFP_KERNEL); if (tmpde == NULL) { printk(KERN_INFO "%s: out of memory\n", __func__); + ret = -ENOMEM; goto fail; } memcpy(tmpde, bh->b_data + offset, frag1); @@ -1235,7 +1259,7 @@ static void isofs_read_inode(struct inode *inode) ei->i_file_format = isofs_file_normal; if (de->flags[-high_sierra] & 2) { - inode->i_mode = S_IRUGO | S_IXUGO | S_IFDIR; + inode->i_mode = sbi->s_dmode | S_IFDIR; inode->i_nlink = 1; /* * Set to 1. We know there are 2, but * the find utility tries to optimize @@ -1245,9 +1269,8 @@ static void isofs_read_inode(struct inode *inode) */ } else { /* Everybody gets to read the file. */ - inode->i_mode = sbi->s_mode; + inode->i_mode = sbi->s_fmode | S_IFREG; inode->i_nlink = 1; - inode->i_mode |= S_IFREG; } inode->i_uid = sbi->s_uid; inode->i_gid = sbi->s_gid; @@ -1259,8 +1282,10 @@ static void isofs_read_inode(struct inode *inode) ei->i_section_size = isonum_733(de->size); if (de->flags[-high_sierra] & 0x80) { - if(isofs_read_level3_size(inode)) + ret = isofs_read_level3_size(inode); + if (ret < 0) goto fail; + ret = -EIO; } else { ei->i_next_section_block = 0; ei->i_next_section_offset = 0; @@ -1346,16 +1371,16 @@ static void isofs_read_inode(struct inode *inode) /* XXX - parse_rock_ridge_inode() had already set i_rdev. */ init_special_inode(inode, inode->i_mode, inode->i_rdev); + ret = 0; out: kfree(tmpde); if (bh) brelse(bh); - return; + return ret; out_badread: printk(KERN_WARNING "ISOFS: unable to read i-node block\n"); fail: - make_bad_inode(inode); goto out; } @@ -1394,9 +1419,10 @@ struct inode *isofs_iget(struct super_block *sb, unsigned long hashval; struct inode *inode; struct isofs_iget5_callback_data data; + long ret; if (offset >= 1ul << sb->s_blocksize_bits) - return NULL; + return ERR_PTR(-EINVAL); data.block = block; data.offset = offset; @@ -1406,9 +1432,17 @@ struct inode *isofs_iget(struct super_block *sb, inode = iget5_locked(sb, hashval, &isofs_iget5_test, &isofs_iget5_set, &data); - if (inode && (inode->i_state & I_NEW)) { - sb->s_op->read_inode(inode); - unlock_new_inode(inode); + if (!inode) + return ERR_PTR(-ENOMEM); + + if (inode->i_state & I_NEW) { + ret = isofs_read_inode(inode); + if (ret < 0) { + iget_failed(inode); + inode = ERR_PTR(ret); + } else { + unlock_new_inode(inode); + } } return inode; diff --git a/fs/isofs/isofs.h b/fs/isofs/isofs.h index f3213f9f89a..d1bdf8adb35 100644 --- a/fs/isofs/isofs.h +++ b/fs/isofs/isofs.h @@ -51,7 +51,8 @@ struct isofs_sb_info { unsigned char s_hide; unsigned char s_showassoc; - mode_t s_mode; + mode_t s_fmode; + mode_t s_dmode; gid_t s_gid; uid_t s_uid; struct nls_table *s_nls_iocharset; /* Native language support table */ diff --git a/fs/isofs/namei.c b/fs/isofs/namei.c index e2b4dad39ca..344b247bc29 100644 --- a/fs/isofs/namei.c +++ b/fs/isofs/namei.c @@ -179,9 +179,9 @@ struct dentry *isofs_lookup(struct inode *dir, struct dentry *dentry, struct nam inode = NULL; if (found) { inode = isofs_iget(dir->i_sb, block, offset); - if (!inode) { + if (IS_ERR(inode)) { unlock_kernel(); - return ERR_PTR(-EACCES); + return ERR_CAST(inode); } } unlock_kernel(); diff --git a/fs/isofs/rock.c b/fs/isofs/rock.c index f3a1db3098d..6bd48f0a704 100644 --- a/fs/isofs/rock.c +++ b/fs/isofs/rock.c @@ -474,8 +474,10 @@ repeat: isofs_iget(inode->i_sb, ISOFS_I(inode)->i_first_extent, 0); - if (!reloc) + if (IS_ERR(reloc)) { + ret = PTR_ERR(reloc); goto out; + } inode->i_mode = reloc->i_mode; inode->i_nlink = reloc->i_nlink; inode->i_uid = reloc->i_uid; diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c index 0f69c416eeb..a5432bbbfb8 100644 --- a/fs/jbd/checkpoint.c +++ b/fs/jbd/checkpoint.c @@ -347,7 +347,8 @@ restart: break; } retry = __process_buffer(journal, jh, bhs,&batch_count); - if (!retry && lock_need_resched(&journal->j_list_lock)){ + if (!retry && (need_resched() || + spin_needbreak(&journal->j_list_lock))) { spin_unlock(&journal->j_list_lock); retry = 1; break; diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c index 610264b99a8..a38c7186c57 100644 --- a/fs/jbd/commit.c +++ b/fs/jbd/commit.c @@ -104,7 +104,8 @@ static int journal_write_commit_record(journal_t *journal, { struct journal_head *descriptor; struct buffer_head *bh; - int i, ret; + journal_header_t *header; + int ret; int barrier_done = 0; if (is_journal_aborted(journal)) @@ -116,13 +117,10 @@ static int journal_write_commit_record(journal_t *journal, bh = jh2bh(descriptor); - /* AKPM: buglet - add `i' to tmp! */ - for (i = 0; i < bh->b_size; i += 512) { - journal_header_t *tmp = (journal_header_t*)bh->b_data; - tmp->h_magic = cpu_to_be32(JFS_MAGIC_NUMBER); - tmp->h_blocktype = cpu_to_be32(JFS_COMMIT_BLOCK); - tmp->h_sequence = cpu_to_be32(commit_transaction->t_tid); - } + header = (journal_header_t *)(bh->b_data); + header->h_magic = cpu_to_be32(JFS_MAGIC_NUMBER); + header->h_blocktype = cpu_to_be32(JFS_COMMIT_BLOCK); + header->h_sequence = cpu_to_be32(commit_transaction->t_tid); JBUFFER_TRACE(descriptor, "write commit block"); set_buffer_dirty(bh); @@ -131,6 +129,8 @@ static int journal_write_commit_record(journal_t *journal, barrier_done = 1; } ret = sync_dirty_buffer(bh); + if (barrier_done) + clear_buffer_ordered(bh); /* is it possible for another commit to fail at roughly * the same time as this one? If so, we don't want to * trust the barrier flag in the super, but instead want @@ -148,7 +148,6 @@ static int journal_write_commit_record(journal_t *journal, spin_unlock(&journal->j_state_lock); /* And try again, without the barrier */ - clear_buffer_ordered(bh); set_buffer_uptodate(bh); set_buffer_dirty(bh); ret = sync_dirty_buffer(bh); @@ -265,7 +264,7 @@ write_out_data: put_bh(bh); } - if (lock_need_resched(&journal->j_list_lock)) { + if (need_resched() || spin_needbreak(&journal->j_list_lock)) { spin_unlock(&journal->j_list_lock); goto write_out_data; } diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index 5d14243499d..0e081d5f32e 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -697,13 +697,14 @@ fail: */ /** - * journal_t * journal_init_dev() - creates an initialises a journal structure + * journal_t * journal_init_dev() - creates and initialises a journal structure * @bdev: Block device on which to create the journal * @fs_dev: Device which hold journalled filesystem for this journal. * @start: Block nr Start of journal. * @len: Length of the journal in blocks. * @blocksize: blocksize of journalling device - * @returns: a newly created journal_t * + * + * Returns: a newly created journal_t * * * journal_init_dev creates a journal which maps a fixed contiguous * range of blocks on an arbitrary block device. @@ -1457,7 +1458,7 @@ static const char *journal_dev_name(journal_t *journal, char *buffer) * Aborts hard --- we mark the abort as occurred, but do _nothing_ else, * and don't attempt to make any other journal updates. */ -void __journal_abort_hard(journal_t *journal) +static void __journal_abort_hard(journal_t *journal) { transaction_t *transaction; char b[BDEVNAME_SIZE]; @@ -1619,14 +1620,14 @@ static int journal_init_journal_head_cache(void) { int retval; - J_ASSERT(journal_head_cache == 0); + J_ASSERT(journal_head_cache == NULL); journal_head_cache = kmem_cache_create("journal_head", sizeof(struct journal_head), 0, /* offset */ SLAB_TEMPORARY, /* flags */ NULL); /* ctor */ retval = 0; - if (journal_head_cache == 0) { + if (!journal_head_cache) { retval = -ENOMEM; printk(KERN_EMERG "JBD: no memory for journal_head cache\n"); } diff --git a/fs/jbd/recovery.c b/fs/jbd/recovery.c index c5d9694b6a2..43bc5e5ed06 100644 --- a/fs/jbd/recovery.c +++ b/fs/jbd/recovery.c @@ -354,7 +354,7 @@ static int do_one_pass(journal_t *journal, struct buffer_head * obh; struct buffer_head * nbh; - cond_resched(); /* We're under lock_kernel() */ + cond_resched(); /* If we already know where to stop the log traversal, * check right now that we haven't gone past the end of @@ -478,7 +478,7 @@ static int do_one_pass(journal_t *journal, memcpy(nbh->b_data, obh->b_data, journal->j_blocksize); if (flags & JFS_FLAG_ESCAPE) { - *((__be32 *)bh->b_data) = + *((__be32 *)nbh->b_data) = cpu_to_be32(JFS_MAGIC_NUMBER); } diff --git a/fs/jbd/revoke.c b/fs/jbd/revoke.c index ad2eacf570c..d5f8eee7c88 100644 --- a/fs/jbd/revoke.c +++ b/fs/jbd/revoke.c @@ -173,13 +173,13 @@ int __init journal_init_revoke_caches(void) 0, SLAB_HWCACHE_ALIGN|SLAB_TEMPORARY, NULL); - if (revoke_record_cache == 0) + if (!revoke_record_cache) return -ENOMEM; revoke_table_cache = kmem_cache_create("revoke_table", sizeof(struct jbd_revoke_table_s), 0, SLAB_TEMPORARY, NULL); - if (revoke_table_cache == 0) { + if (!revoke_table_cache) { kmem_cache_destroy(revoke_record_cache); revoke_record_cache = NULL; return -ENOMEM; diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index 038ed743619..2c9e8f5d13a 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -369,7 +369,7 @@ out: /** - * int journal_restart() - restart a handle . + * int journal_restart() - restart a handle. * @handle: handle to restart * @nblocks: nr credits requested * @@ -844,8 +844,7 @@ out: } /** - * int journal_get_undo_access() - Notify intent to modify metadata with - * non-rewindable consequences + * int journal_get_undo_access() - Notify intent to modify metadata with non-rewindable consequences * @handle: transaction * @bh: buffer to undo * @credits: store the number of taken credits here (if not NULL) @@ -921,12 +920,14 @@ out: } /** - * int journal_dirty_data() - mark a buffer as containing dirty data which - * needs to be flushed before we can commit the - * current transaction. + * int journal_dirty_data() - mark a buffer as containing dirty data to be flushed * @handle: transaction * @bh: bufferhead to mark * + * Description: + * Mark a buffer as containing dirty data which needs to be flushed before + * we can commit the current transaction. + * * The buffer is placed on the transaction's data list and is marked as * belonging to the transaction. * @@ -1098,11 +1099,11 @@ no_journal: } /** - * int journal_dirty_metadata() - mark a buffer as containing dirty metadata + * int journal_dirty_metadata() - mark a buffer as containing dirty metadata * @handle: transaction to add buffer to. * @bh: buffer to mark * - * mark dirty metadata which needs to be journaled as part of the current + * Mark dirty metadata which needs to be journaled as part of the current * transaction. * * The buffer is placed on the transaction's metadata list and is marked @@ -1425,7 +1426,8 @@ int journal_stop(handle_t *handle) return err; } -/**int journal_force_commit() - force any uncommitted transactions +/** + * int journal_force_commit() - force any uncommitted transactions * @journal: journal to force * * For synchronous operations: force any uncommitted transactions @@ -1902,13 +1904,12 @@ zap_buffer_unlocked: } /** - * void journal_invalidatepage() - * @journal: journal to use for flush... + * void journal_invalidatepage() - invalidate a journal page + * @journal: journal to use for flush * @page: page to flush * @offset: length of page to invalidate. * * Reap page buffers containing data after offset in page. - * */ void journal_invalidatepage(journal_t *journal, struct page *page, diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index 3fccde7ba00..6914598022c 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -232,7 +232,8 @@ __flush_batch(journal_t *journal, struct buffer_head **bhs, int *batch_count) * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it */ static int __process_buffer(journal_t *journal, struct journal_head *jh, - struct buffer_head **bhs, int *batch_count) + struct buffer_head **bhs, int *batch_count, + transaction_t *transaction) { struct buffer_head *bh = jh2bh(jh); int ret = 0; @@ -250,6 +251,7 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, transaction_t *t = jh->b_transaction; tid_t tid = t->t_tid; + transaction->t_chp_stats.cs_forced_to_close++; spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); jbd2_log_start_commit(journal, tid); @@ -279,6 +281,7 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, bhs[*batch_count] = bh; __buffer_relink_io(jh); jbd_unlock_bh_state(bh); + transaction->t_chp_stats.cs_written++; (*batch_count)++; if (*batch_count == NR_BATCH) { spin_unlock(&journal->j_list_lock); @@ -322,6 +325,8 @@ int jbd2_log_do_checkpoint(journal_t *journal) if (!journal->j_checkpoint_transactions) goto out; transaction = journal->j_checkpoint_transactions; + if (transaction->t_chp_stats.cs_chp_time == 0) + transaction->t_chp_stats.cs_chp_time = jiffies; this_tid = transaction->t_tid; restart: /* @@ -346,8 +351,10 @@ restart: retry = 1; break; } - retry = __process_buffer(journal, jh, bhs,&batch_count); - if (!retry && lock_need_resched(&journal->j_list_lock)){ + retry = __process_buffer(journal, jh, bhs, &batch_count, + transaction); + if (!retry && (need_resched() || + spin_needbreak(&journal->j_list_lock))) { spin_unlock(&journal->j_list_lock); retry = 1; break; @@ -602,15 +609,15 @@ int __jbd2_journal_remove_checkpoint(struct journal_head *jh) /* * There is one special case to worry about: if we have just pulled the - * buffer off a committing transaction's forget list, then even if the - * checkpoint list is empty, the transaction obviously cannot be - * dropped! + * buffer off a running or committing transaction's checkpoing list, + * then even if the checkpoint list is empty, the transaction obviously + * cannot be dropped! * - * The locking here around j_committing_transaction is a bit sleazy. + * The locking here around t_state is a bit sleazy. * See the comment at the end of jbd2_journal_commit_transaction(). */ - if (transaction == journal->j_committing_transaction) { - JBUFFER_TRACE(jh, "belongs to committing transaction"); + if (transaction->t_state != T_FINISHED) { + JBUFFER_TRACE(jh, "belongs to running/committing transaction"); goto out; } diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 6986f334c64..a8173081f83 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -20,6 +20,8 @@ #include <linux/slab.h> #include <linux/mm.h> #include <linux/pagemap.h> +#include <linux/jiffies.h> +#include <linux/crc32.h> /* * Default IO end handler for temporary BJ_IO buffer_heads. @@ -92,19 +94,23 @@ static int inverted_lock(journal_t *journal, struct buffer_head *bh) return 1; } -/* Done it all: now write the commit record. We should have +/* + * Done it all: now submit the commit record. We should have * cleaned up our previous buffers by now, so if we are in abort * mode we can now just skip the rest of the journal write * entirely. * * Returns 1 if the journal needs to be aborted or 0 on success */ -static int journal_write_commit_record(journal_t *journal, - transaction_t *commit_transaction) +static int journal_submit_commit_record(journal_t *journal, + transaction_t *commit_transaction, + struct buffer_head **cbh, + __u32 crc32_sum) { struct journal_head *descriptor; + struct commit_header *tmp; struct buffer_head *bh; - int i, ret; + int ret; int barrier_done = 0; if (is_journal_aborted(journal)) @@ -116,21 +122,35 @@ static int journal_write_commit_record(journal_t *journal, bh = jh2bh(descriptor); - /* AKPM: buglet - add `i' to tmp! */ - for (i = 0; i < bh->b_size; i += 512) { - journal_header_t *tmp = (journal_header_t*)bh->b_data; - tmp->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); - tmp->h_blocktype = cpu_to_be32(JBD2_COMMIT_BLOCK); - tmp->h_sequence = cpu_to_be32(commit_transaction->t_tid); + tmp = (struct commit_header *)bh->b_data; + tmp->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); + tmp->h_blocktype = cpu_to_be32(JBD2_COMMIT_BLOCK); + tmp->h_sequence = cpu_to_be32(commit_transaction->t_tid); + + if (JBD2_HAS_COMPAT_FEATURE(journal, + JBD2_FEATURE_COMPAT_CHECKSUM)) { + tmp->h_chksum_type = JBD2_CRC32_CHKSUM; + tmp->h_chksum_size = JBD2_CRC32_CHKSUM_SIZE; + tmp->h_chksum[0] = cpu_to_be32(crc32_sum); } - JBUFFER_TRACE(descriptor, "write commit block"); + JBUFFER_TRACE(descriptor, "submit commit block"); + lock_buffer(bh); + get_bh(bh); set_buffer_dirty(bh); - if (journal->j_flags & JBD2_BARRIER) { + set_buffer_uptodate(bh); + bh->b_end_io = journal_end_buffer_io_sync; + + if (journal->j_flags & JBD2_BARRIER && + !JBD2_HAS_INCOMPAT_FEATURE(journal, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) { set_buffer_ordered(bh); barrier_done = 1; } - ret = sync_dirty_buffer(bh); + ret = submit_bh(WRITE, bh); + if (barrier_done) + clear_buffer_ordered(bh); + /* is it possible for another commit to fail at roughly * the same time as this one? If so, we don't want to * trust the barrier flag in the super, but instead want @@ -148,17 +168,74 @@ static int journal_write_commit_record(journal_t *journal, spin_unlock(&journal->j_state_lock); /* And try again, without the barrier */ - clear_buffer_ordered(bh); set_buffer_uptodate(bh); set_buffer_dirty(bh); - ret = sync_dirty_buffer(bh); + ret = submit_bh(WRITE, bh); } - put_bh(bh); /* One for getblk() */ - jbd2_journal_put_journal_head(descriptor); + *cbh = bh; + return ret; +} + +/* + * This function along with journal_submit_commit_record + * allows to write the commit record asynchronously. + */ +static int journal_wait_on_commit_record(struct buffer_head *bh) +{ + int ret = 0; + + clear_buffer_dirty(bh); + wait_on_buffer(bh); + + if (unlikely(!buffer_uptodate(bh))) + ret = -EIO; + put_bh(bh); /* One for getblk() */ + jbd2_journal_put_journal_head(bh2jh(bh)); - return (ret == -EIO); + return ret; } +/* + * Wait for all submitted IO to complete. + */ +static int journal_wait_on_locked_list(journal_t *journal, + transaction_t *commit_transaction) +{ + int ret = 0; + struct journal_head *jh; + + while (commit_transaction->t_locked_list) { + struct buffer_head *bh; + + jh = commit_transaction->t_locked_list->b_tprev; + bh = jh2bh(jh); + get_bh(bh); + if (buffer_locked(bh)) { + spin_unlock(&journal->j_list_lock); + wait_on_buffer(bh); + if (unlikely(!buffer_uptodate(bh))) + ret = -EIO; + spin_lock(&journal->j_list_lock); + } + if (!inverted_lock(journal, bh)) { + put_bh(bh); + spin_lock(&journal->j_list_lock); + continue; + } + if (buffer_jbd(bh) && jh->b_jlist == BJ_Locked) { + __jbd2_journal_unfile_buffer(jh); + jbd_unlock_bh_state(bh); + jbd2_journal_remove_journal_head(bh); + put_bh(bh); + } else { + jbd_unlock_bh_state(bh); + } + put_bh(bh); + cond_resched_lock(&journal->j_list_lock); + } + return ret; + } + static void journal_do_submit_data(struct buffer_head **wbuf, int bufs) { int i; @@ -265,7 +342,7 @@ write_out_data: put_bh(bh); } - if (lock_need_resched(&journal->j_list_lock)) { + if (need_resched() || spin_needbreak(&journal->j_list_lock)) { spin_unlock(&journal->j_list_lock); goto write_out_data; } @@ -274,7 +351,21 @@ write_out_data: journal_do_submit_data(wbuf, bufs); } -static inline void write_tag_block(int tag_bytes, journal_block_tag_t *tag, +static __u32 jbd2_checksum_data(__u32 crc32_sum, struct buffer_head *bh) +{ + struct page *page = bh->b_page; + char *addr; + __u32 checksum; + + addr = kmap_atomic(page, KM_USER0); + checksum = crc32_be(crc32_sum, + (void *)(addr + offset_in_page(bh->b_data)), bh->b_size); + kunmap_atomic(addr, KM_USER0); + + return checksum; +} + +static void write_tag_block(int tag_bytes, journal_block_tag_t *tag, unsigned long long block) { tag->t_blocknr = cpu_to_be32(block & (u32)~0); @@ -290,6 +381,7 @@ static inline void write_tag_block(int tag_bytes, journal_block_tag_t *tag, */ void jbd2_journal_commit_transaction(journal_t *journal) { + struct transaction_stats_s stats; transaction_t *commit_transaction; struct journal_head *jh, *new_jh, *descriptor; struct buffer_head **wbuf = journal->j_wbuf; @@ -305,6 +397,8 @@ void jbd2_journal_commit_transaction(journal_t *journal) int tag_flag; int i; int tag_bytes = journal_tag_bytes(journal); + struct buffer_head *cbh = NULL; /* For transactional checksums */ + __u32 crc32_sum = ~0; /* * First job: lock down the current transaction and wait for @@ -337,6 +431,11 @@ void jbd2_journal_commit_transaction(journal_t *journal) spin_lock(&journal->j_state_lock); commit_transaction->t_state = T_LOCKED; + stats.u.run.rs_wait = commit_transaction->t_max_wait; + stats.u.run.rs_locked = jiffies; + stats.u.run.rs_running = jbd2_time_diff(commit_transaction->t_start, + stats.u.run.rs_locked); + spin_lock(&commit_transaction->t_handle_lock); while (commit_transaction->t_updates) { DEFINE_WAIT(wait); @@ -407,6 +506,10 @@ void jbd2_journal_commit_transaction(journal_t *journal) */ jbd2_journal_switch_revoke_table(journal); + stats.u.run.rs_flushing = jiffies; + stats.u.run.rs_locked = jbd2_time_diff(stats.u.run.rs_locked, + stats.u.run.rs_flushing); + commit_transaction->t_state = T_FLUSH; journal->j_committing_transaction = commit_transaction; journal->j_running_transaction = NULL; @@ -440,38 +543,15 @@ void jbd2_journal_commit_transaction(journal_t *journal) journal_submit_data_buffers(journal, commit_transaction); /* - * Wait for all previously submitted IO to complete. + * Wait for all previously submitted IO to complete if commit + * record is to be written synchronously. */ spin_lock(&journal->j_list_lock); - while (commit_transaction->t_locked_list) { - struct buffer_head *bh; + if (!JBD2_HAS_INCOMPAT_FEATURE(journal, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) + err = journal_wait_on_locked_list(journal, + commit_transaction); - jh = commit_transaction->t_locked_list->b_tprev; - bh = jh2bh(jh); - get_bh(bh); - if (buffer_locked(bh)) { - spin_unlock(&journal->j_list_lock); - wait_on_buffer(bh); - if (unlikely(!buffer_uptodate(bh))) - err = -EIO; - spin_lock(&journal->j_list_lock); - } - if (!inverted_lock(journal, bh)) { - put_bh(bh); - spin_lock(&journal->j_list_lock); - continue; - } - if (buffer_jbd(bh) && jh->b_jlist == BJ_Locked) { - __jbd2_journal_unfile_buffer(jh); - jbd_unlock_bh_state(bh); - jbd2_journal_remove_journal_head(bh); - put_bh(bh); - } else { - jbd_unlock_bh_state(bh); - } - put_bh(bh); - cond_resched_lock(&journal->j_list_lock); - } spin_unlock(&journal->j_list_lock); if (err) @@ -498,6 +578,12 @@ void jbd2_journal_commit_transaction(journal_t *journal) */ commit_transaction->t_state = T_COMMIT; + stats.u.run.rs_logging = jiffies; + stats.u.run.rs_flushing = jbd2_time_diff(stats.u.run.rs_flushing, + stats.u.run.rs_logging); + stats.u.run.rs_blocks = commit_transaction->t_outstanding_credits; + stats.u.run.rs_blocks_logged = 0; + descriptor = NULL; bufs = 0; while (commit_transaction->t_buffers) { @@ -639,6 +725,15 @@ void jbd2_journal_commit_transaction(journal_t *journal) start_journal_io: for (i = 0; i < bufs; i++) { struct buffer_head *bh = wbuf[i]; + /* + * Compute checksum. + */ + if (JBD2_HAS_COMPAT_FEATURE(journal, + JBD2_FEATURE_COMPAT_CHECKSUM)) { + crc32_sum = + jbd2_checksum_data(crc32_sum, bh); + } + lock_buffer(bh); clear_buffer_dirty(bh); set_buffer_uptodate(bh); @@ -646,6 +741,7 @@ start_journal_io: submit_bh(WRITE, bh); } cond_resched(); + stats.u.run.rs_blocks_logged += bufs; /* Force a new descriptor to be generated next time round the loop. */ @@ -654,6 +750,23 @@ start_journal_io: } } + /* Done it all: now write the commit record asynchronously. */ + + if (JBD2_HAS_INCOMPAT_FEATURE(journal, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) { + err = journal_submit_commit_record(journal, commit_transaction, + &cbh, crc32_sum); + if (err) + __jbd2_journal_abort_hard(journal); + + spin_lock(&journal->j_list_lock); + err = journal_wait_on_locked_list(journal, + commit_transaction); + spin_unlock(&journal->j_list_lock); + if (err) + __jbd2_journal_abort_hard(journal); + } + /* Lo and behold: we have just managed to send a transaction to the log. Before we can commit it, wait for the IO so far to complete. Control buffers being written are on the @@ -753,8 +866,15 @@ wait_for_iobuf: jbd_debug(3, "JBD: commit phase 6\n"); - if (journal_write_commit_record(journal, commit_transaction)) - err = -EIO; + if (!JBD2_HAS_INCOMPAT_FEATURE(journal, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) { + err = journal_submit_commit_record(journal, commit_transaction, + &cbh, crc32_sum); + if (err) + __jbd2_journal_abort_hard(journal); + } + if (!err && !is_journal_aborted(journal)) + err = journal_wait_on_commit_record(cbh); if (err) jbd2_journal_abort(journal, err); @@ -816,6 +936,7 @@ restart_loop: cp_transaction = jh->b_cp_transaction; if (cp_transaction) { JBUFFER_TRACE(jh, "remove from old cp transaction"); + cp_transaction->t_chp_stats.cs_dropped++; __jbd2_journal_remove_checkpoint(jh); } @@ -867,10 +988,10 @@ restart_loop: } spin_unlock(&journal->j_list_lock); /* - * This is a bit sleazy. We borrow j_list_lock to protect - * journal->j_committing_transaction in __jbd2_journal_remove_checkpoint. - * Really, __jbd2_journal_remove_checkpoint should be using j_state_lock but - * it's a bit hassle to hold that across __jbd2_journal_remove_checkpoint + * This is a bit sleazy. We use j_list_lock to protect transition + * of a transaction into T_FINISHED state and calling + * __jbd2_journal_drop_transaction(). Otherwise we could race with + * other checkpointing code processing the transaction... */ spin_lock(&journal->j_state_lock); spin_lock(&journal->j_list_lock); @@ -890,6 +1011,36 @@ restart_loop: J_ASSERT(commit_transaction->t_state == T_COMMIT); + commit_transaction->t_start = jiffies; + stats.u.run.rs_logging = jbd2_time_diff(stats.u.run.rs_logging, + commit_transaction->t_start); + + /* + * File the transaction for history + */ + stats.ts_type = JBD2_STATS_RUN; + stats.ts_tid = commit_transaction->t_tid; + stats.u.run.rs_handle_count = commit_transaction->t_handle_count; + spin_lock(&journal->j_history_lock); + memcpy(journal->j_history + journal->j_history_cur, &stats, + sizeof(stats)); + if (++journal->j_history_cur == journal->j_history_max) + journal->j_history_cur = 0; + + /* + * Calculate overall stats + */ + journal->j_stats.ts_tid++; + journal->j_stats.u.run.rs_wait += stats.u.run.rs_wait; + journal->j_stats.u.run.rs_running += stats.u.run.rs_running; + journal->j_stats.u.run.rs_locked += stats.u.run.rs_locked; + journal->j_stats.u.run.rs_flushing += stats.u.run.rs_flushing; + journal->j_stats.u.run.rs_logging += stats.u.run.rs_logging; + journal->j_stats.u.run.rs_handle_count += stats.u.run.rs_handle_count; + journal->j_stats.u.run.rs_blocks += stats.u.run.rs_blocks; + journal->j_stats.u.run.rs_blocks_logged += stats.u.run.rs_blocks_logged; + spin_unlock(&journal->j_history_lock); + commit_transaction->t_state = T_FINISHED; J_ASSERT(commit_transaction == journal->j_committing_transaction); journal->j_commit_sequence = commit_transaction->t_tid; diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 6ddc5531587..954cff001df 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -36,6 +36,7 @@ #include <linux/poison.h> #include <linux/proc_fs.h> #include <linux/debugfs.h> +#include <linux/seq_file.h> #include <asm/uaccess.h> #include <asm/page.h> @@ -218,7 +219,7 @@ static int jbd2_journal_start_thread(journal_t *journal) if (IS_ERR(t)) return PTR_ERR(t); - wait_event(journal->j_wait_done_commit, journal->j_task != 0); + wait_event(journal->j_wait_done_commit, journal->j_task != NULL); return 0; } @@ -230,7 +231,7 @@ static void journal_kill_thread(journal_t *journal) while (journal->j_task) { wake_up(&journal->j_wait_commit); spin_unlock(&journal->j_state_lock); - wait_event(journal->j_wait_done_commit, journal->j_task == 0); + wait_event(journal->j_wait_done_commit, journal->j_task == NULL); spin_lock(&journal->j_state_lock); } spin_unlock(&journal->j_state_lock); @@ -640,6 +641,312 @@ struct journal_head *jbd2_journal_get_descriptor_buffer(journal_t *journal) return jbd2_journal_add_journal_head(bh); } +struct jbd2_stats_proc_session { + journal_t *journal; + struct transaction_stats_s *stats; + int start; + int max; +}; + +static void *jbd2_history_skip_empty(struct jbd2_stats_proc_session *s, + struct transaction_stats_s *ts, + int first) +{ + if (ts == s->stats + s->max) + ts = s->stats; + if (!first && ts == s->stats + s->start) + return NULL; + while (ts->ts_type == 0) { + ts++; + if (ts == s->stats + s->max) + ts = s->stats; + if (ts == s->stats + s->start) + return NULL; + } + return ts; + +} + +static void *jbd2_seq_history_start(struct seq_file *seq, loff_t *pos) +{ + struct jbd2_stats_proc_session *s = seq->private; + struct transaction_stats_s *ts; + int l = *pos; + + if (l == 0) + return SEQ_START_TOKEN; + ts = jbd2_history_skip_empty(s, s->stats + s->start, 1); + if (!ts) + return NULL; + l--; + while (l) { + ts = jbd2_history_skip_empty(s, ++ts, 0); + if (!ts) + break; + l--; + } + return ts; +} + +static void *jbd2_seq_history_next(struct seq_file *seq, void *v, loff_t *pos) +{ + struct jbd2_stats_proc_session *s = seq->private; + struct transaction_stats_s *ts = v; + + ++*pos; + if (v == SEQ_START_TOKEN) + return jbd2_history_skip_empty(s, s->stats + s->start, 1); + else + return jbd2_history_skip_empty(s, ++ts, 0); +} + +static int jbd2_seq_history_show(struct seq_file *seq, void *v) +{ + struct transaction_stats_s *ts = v; + if (v == SEQ_START_TOKEN) { + seq_printf(seq, "%-4s %-5s %-5s %-5s %-5s %-5s %-5s %-6s %-5s " + "%-5s %-5s %-5s %-5s %-5s\n", "R/C", "tid", + "wait", "run", "lock", "flush", "log", "hndls", + "block", "inlog", "ctime", "write", "drop", + "close"); + return 0; + } + if (ts->ts_type == JBD2_STATS_RUN) + seq_printf(seq, "%-4s %-5lu %-5u %-5u %-5u %-5u %-5u " + "%-6lu %-5lu %-5lu\n", "R", ts->ts_tid, + jiffies_to_msecs(ts->u.run.rs_wait), + jiffies_to_msecs(ts->u.run.rs_running), + jiffies_to_msecs(ts->u.run.rs_locked), + jiffies_to_msecs(ts->u.run.rs_flushing), + jiffies_to_msecs(ts->u.run.rs_logging), + ts->u.run.rs_handle_count, + ts->u.run.rs_blocks, + ts->u.run.rs_blocks_logged); + else if (ts->ts_type == JBD2_STATS_CHECKPOINT) + seq_printf(seq, "%-4s %-5lu %48s %-5u %-5lu %-5lu %-5lu\n", + "C", ts->ts_tid, " ", + jiffies_to_msecs(ts->u.chp.cs_chp_time), + ts->u.chp.cs_written, ts->u.chp.cs_dropped, + ts->u.chp.cs_forced_to_close); + else + J_ASSERT(0); + return 0; +} + +static void jbd2_seq_history_stop(struct seq_file *seq, void *v) +{ +} + +static struct seq_operations jbd2_seq_history_ops = { + .start = jbd2_seq_history_start, + .next = jbd2_seq_history_next, + .stop = jbd2_seq_history_stop, + .show = jbd2_seq_history_show, +}; + +static int jbd2_seq_history_open(struct inode *inode, struct file *file) +{ + journal_t *journal = PDE(inode)->data; + struct jbd2_stats_proc_session *s; + int rc, size; + + s = kmalloc(sizeof(*s), GFP_KERNEL); + if (s == NULL) + return -ENOMEM; + size = sizeof(struct transaction_stats_s) * journal->j_history_max; + s->stats = kmalloc(size, GFP_KERNEL); + if (s->stats == NULL) { + kfree(s); + return -ENOMEM; + } + spin_lock(&journal->j_history_lock); + memcpy(s->stats, journal->j_history, size); + s->max = journal->j_history_max; + s->start = journal->j_history_cur % s->max; + spin_unlock(&journal->j_history_lock); + + rc = seq_open(file, &jbd2_seq_history_ops); + if (rc == 0) { + struct seq_file *m = file->private_data; + m->private = s; + } else { + kfree(s->stats); + kfree(s); + } + return rc; + +} + +static int jbd2_seq_history_release(struct inode *inode, struct file *file) +{ + struct seq_file *seq = file->private_data; + struct jbd2_stats_proc_session *s = seq->private; + + kfree(s->stats); + kfree(s); + return seq_release(inode, file); +} + +static struct file_operations jbd2_seq_history_fops = { + .owner = THIS_MODULE, + .open = jbd2_seq_history_open, + .read = seq_read, + .llseek = seq_lseek, + .release = jbd2_seq_history_release, +}; + +static void *jbd2_seq_info_start(struct seq_file *seq, loff_t *pos) +{ + return *pos ? NULL : SEQ_START_TOKEN; +} + +static void *jbd2_seq_info_next(struct seq_file *seq, void *v, loff_t *pos) +{ + return NULL; +} + +static int jbd2_seq_info_show(struct seq_file *seq, void *v) +{ + struct jbd2_stats_proc_session *s = seq->private; + + if (v != SEQ_START_TOKEN) + return 0; + seq_printf(seq, "%lu transaction, each upto %u blocks\n", + s->stats->ts_tid, + s->journal->j_max_transaction_buffers); + if (s->stats->ts_tid == 0) + return 0; + seq_printf(seq, "average: \n %ums waiting for transaction\n", + jiffies_to_msecs(s->stats->u.run.rs_wait / s->stats->ts_tid)); + seq_printf(seq, " %ums running transaction\n", + jiffies_to_msecs(s->stats->u.run.rs_running / s->stats->ts_tid)); + seq_printf(seq, " %ums transaction was being locked\n", + jiffies_to_msecs(s->stats->u.run.rs_locked / s->stats->ts_tid)); + seq_printf(seq, " %ums flushing data (in ordered mode)\n", + jiffies_to_msecs(s->stats->u.run.rs_flushing / s->stats->ts_tid)); + seq_printf(seq, " %ums logging transaction\n", + jiffies_to_msecs(s->stats->u.run.rs_logging / s->stats->ts_tid)); + seq_printf(seq, " %lu handles per transaction\n", + s->stats->u.run.rs_handle_count / s->stats->ts_tid); + seq_printf(seq, " %lu blocks per transaction\n", + s->stats->u.run.rs_blocks / s->stats->ts_tid); + seq_printf(seq, " %lu logged blocks per transaction\n", + s->stats->u.run.rs_blocks_logged / s->stats->ts_tid); + return 0; +} + +static void jbd2_seq_info_stop(struct seq_file *seq, void *v) +{ +} + +static struct seq_operations jbd2_seq_info_ops = { + .start = jbd2_seq_info_start, + .next = jbd2_seq_info_next, + .stop = jbd2_seq_info_stop, + .show = jbd2_seq_info_show, +}; + +static int jbd2_seq_info_open(struct inode *inode, struct file *file) +{ + journal_t *journal = PDE(inode)->data; + struct jbd2_stats_proc_session *s; + int rc, size; + + s = kmalloc(sizeof(*s), GFP_KERNEL); + if (s == NULL) + return -ENOMEM; + size = sizeof(struct transaction_stats_s); + s->stats = kmalloc(size, GFP_KERNEL); + if (s->stats == NULL) { + kfree(s); + return -ENOMEM; + } + spin_lock(&journal->j_history_lock); + memcpy(s->stats, &journal->j_stats, size); + s->journal = journal; + spin_unlock(&journal->j_history_lock); + + rc = seq_open(file, &jbd2_seq_info_ops); + if (rc == 0) { + struct seq_file *m = file->private_data; + m->private = s; + } else { + kfree(s->stats); + kfree(s); + } + return rc; + +} + +static int jbd2_seq_info_release(struct inode *inode, struct file *file) +{ + struct seq_file *seq = file->private_data; + struct jbd2_stats_proc_session *s = seq->private; + kfree(s->stats); + kfree(s); + return seq_release(inode, file); +} + +static struct file_operations jbd2_seq_info_fops = { + .owner = THIS_MODULE, + .open = jbd2_seq_info_open, + .read = seq_read, + .llseek = seq_lseek, + .release = jbd2_seq_info_release, +}; + +static struct proc_dir_entry *proc_jbd2_stats; + +static void jbd2_stats_proc_init(journal_t *journal) +{ + char name[BDEVNAME_SIZE]; + + snprintf(name, sizeof(name) - 1, "%s", bdevname(journal->j_dev, name)); + journal->j_proc_entry = proc_mkdir(name, proc_jbd2_stats); + if (journal->j_proc_entry) { + struct proc_dir_entry *p; + p = create_proc_entry("history", S_IRUGO, + journal->j_proc_entry); + if (p) { + p->proc_fops = &jbd2_seq_history_fops; + p->data = journal; + p = create_proc_entry("info", S_IRUGO, + journal->j_proc_entry); + if (p) { + p->proc_fops = &jbd2_seq_info_fops; + p->data = journal; + } + } + } +} + +static void jbd2_stats_proc_exit(journal_t *journal) +{ + char name[BDEVNAME_SIZE]; + + snprintf(name, sizeof(name) - 1, "%s", bdevname(journal->j_dev, name)); + remove_proc_entry("info", journal->j_proc_entry); + remove_proc_entry("history", journal->j_proc_entry); + remove_proc_entry(name, proc_jbd2_stats); +} + +static void journal_init_stats(journal_t *journal) +{ + int size; + + if (!proc_jbd2_stats) + return; + + journal->j_history_max = 100; + size = sizeof(struct transaction_stats_s) * journal->j_history_max; + journal->j_history = kzalloc(size, GFP_KERNEL); + if (!journal->j_history) { + journal->j_history_max = 0; + return; + } + spin_lock_init(&journal->j_history_lock); +} + /* * Management for journal control blocks: functions to create and * destroy journal_t structures, and to initialise and read existing @@ -681,6 +988,9 @@ static journal_t * journal_init_common (void) kfree(journal); goto fail; } + + journal_init_stats(journal); + return journal; fail: return NULL; @@ -735,6 +1045,7 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev, journal->j_fs_dev = fs_dev; journal->j_blk_offset = start; journal->j_maxlen = len; + jbd2_stats_proc_init(journal); bh = __getblk(journal->j_dev, start, journal->j_blocksize); J_ASSERT(bh != NULL); @@ -773,6 +1084,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode) journal->j_maxlen = inode->i_size >> inode->i_sb->s_blocksize_bits; journal->j_blocksize = inode->i_sb->s_blocksize; + jbd2_stats_proc_init(journal); /* journal descriptor can store up to n blocks -bzzz */ n = journal->j_blocksize / sizeof(journal_block_tag_t); @@ -1153,6 +1465,8 @@ void jbd2_journal_destroy(journal_t *journal) brelse(journal->j_sb_buffer); } + if (journal->j_proc_entry) + jbd2_stats_proc_exit(journal); if (journal->j_inode) iput(journal->j_inode); if (journal->j_revoke) @@ -1264,6 +1578,32 @@ int jbd2_journal_set_features (journal_t *journal, unsigned long compat, return 1; } +/* + * jbd2_journal_clear_features () - Clear a given journal feature in the + * superblock + * @journal: Journal to act on. + * @compat: bitmask of compatible features + * @ro: bitmask of features that force read-only mount + * @incompat: bitmask of incompatible features + * + * Clear a given journal feature as present on the + * superblock. + */ +void jbd2_journal_clear_features(journal_t *journal, unsigned long compat, + unsigned long ro, unsigned long incompat) +{ + journal_superblock_t *sb; + + jbd_debug(1, "Clear features 0x%lx/0x%lx/0x%lx\n", + compat, ro, incompat); + + sb = journal->j_superblock; + + sb->s_feature_compat &= ~cpu_to_be32(compat); + sb->s_feature_ro_compat &= ~cpu_to_be32(ro); + sb->s_feature_incompat &= ~cpu_to_be32(incompat); +} +EXPORT_SYMBOL(jbd2_journal_clear_features); /** * int jbd2_journal_update_format () - Update on-disk journal structure. @@ -1629,14 +1969,14 @@ static int journal_init_jbd2_journal_head_cache(void) { int retval; - J_ASSERT(jbd2_journal_head_cache == 0); + J_ASSERT(jbd2_journal_head_cache == NULL); jbd2_journal_head_cache = kmem_cache_create("jbd2_journal_head", sizeof(struct journal_head), 0, /* offset */ - 0, /* flags */ + SLAB_TEMPORARY, /* flags */ NULL); /* ctor */ retval = 0; - if (jbd2_journal_head_cache == 0) { + if (!jbd2_journal_head_cache) { retval = -ENOMEM; printk(KERN_EMERG "JBD: no memory for journal_head cache\n"); } @@ -1662,14 +2002,14 @@ static struct journal_head *journal_alloc_journal_head(void) atomic_inc(&nr_journal_heads); #endif ret = kmem_cache_alloc(jbd2_journal_head_cache, GFP_NOFS); - if (ret == 0) { + if (!ret) { jbd_debug(1, "out of memory for journal_head\n"); if (time_after(jiffies, last_warning + 5*HZ)) { printk(KERN_NOTICE "ENOMEM in %s, retrying.\n", __FUNCTION__); last_warning = jiffies; } - while (ret == 0) { + while (!ret) { yield(); ret = kmem_cache_alloc(jbd2_journal_head_cache, GFP_NOFS); } @@ -1900,6 +2240,28 @@ static void __exit jbd2_remove_debugfs_entry(void) #endif +#ifdef CONFIG_PROC_FS + +#define JBD2_STATS_PROC_NAME "fs/jbd2" + +static void __init jbd2_create_jbd_stats_proc_entry(void) +{ + proc_jbd2_stats = proc_mkdir(JBD2_STATS_PROC_NAME, NULL); +} + +static void __exit jbd2_remove_jbd_stats_proc_entry(void) +{ + if (proc_jbd2_stats) + remove_proc_entry(JBD2_STATS_PROC_NAME, NULL); +} + +#else + +#define jbd2_create_jbd_stats_proc_entry() do {} while (0) +#define jbd2_remove_jbd_stats_proc_entry() do {} while (0) + +#endif + struct kmem_cache *jbd2_handle_cache; static int __init journal_init_handle_cache(void) @@ -1907,7 +2269,7 @@ static int __init journal_init_handle_cache(void) jbd2_handle_cache = kmem_cache_create("jbd2_journal_handle", sizeof(handle_t), 0, /* offset */ - 0, /* flags */ + SLAB_TEMPORARY, /* flags */ NULL); /* ctor */ if (jbd2_handle_cache == NULL) { printk(KERN_EMERG "JBD: failed to create handle cache\n"); @@ -1955,6 +2317,7 @@ static int __init journal_init(void) if (ret != 0) jbd2_journal_destroy_caches(); jbd2_create_debugfs_entry(); + jbd2_create_jbd_stats_proc_entry(); return ret; } @@ -1966,6 +2329,7 @@ static void __exit journal_exit(void) printk(KERN_EMERG "JBD: leaked %d journal_heads!\n", n); #endif jbd2_remove_debugfs_entry(); + jbd2_remove_jbd_stats_proc_entry(); jbd2_journal_destroy_caches(); } diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index d0ce627539e..5d0405a9e7c 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -21,6 +21,7 @@ #include <linux/jbd2.h> #include <linux/errno.h> #include <linux/slab.h> +#include <linux/crc32.h> #endif /* @@ -316,6 +317,37 @@ static inline unsigned long long read_tag_block(int tag_bytes, journal_block_tag return block; } +/* + * calc_chksums calculates the checksums for the blocks described in the + * descriptor block. + */ +static int calc_chksums(journal_t *journal, struct buffer_head *bh, + unsigned long *next_log_block, __u32 *crc32_sum) +{ + int i, num_blks, err; + unsigned long io_block; + struct buffer_head *obh; + + num_blks = count_tags(journal, bh); + /* Calculate checksum of the descriptor block. */ + *crc32_sum = crc32_be(*crc32_sum, (void *)bh->b_data, bh->b_size); + + for (i = 0; i < num_blks; i++) { + io_block = (*next_log_block)++; + wrap(journal, *next_log_block); + err = jread(&obh, journal, io_block); + if (err) { + printk(KERN_ERR "JBD: IO error %d recovering block " + "%lu in log\n", err, io_block); + return 1; + } else { + *crc32_sum = crc32_be(*crc32_sum, (void *)obh->b_data, + obh->b_size); + } + } + return 0; +} + static int do_one_pass(journal_t *journal, struct recovery_info *info, enum passtype pass) { @@ -328,6 +360,7 @@ static int do_one_pass(journal_t *journal, unsigned int sequence; int blocktype; int tag_bytes = journal_tag_bytes(journal); + __u32 crc32_sum = ~0; /* Transactional Checksums */ /* Precompute the maximum metadata descriptors in a descriptor block */ int MAX_BLOCKS_PER_DESC; @@ -364,7 +397,7 @@ static int do_one_pass(journal_t *journal, struct buffer_head * obh; struct buffer_head * nbh; - cond_resched(); /* We're under lock_kernel() */ + cond_resched(); /* If we already know where to stop the log traversal, * check right now that we haven't gone past the end of @@ -419,12 +452,26 @@ static int do_one_pass(journal_t *journal, switch(blocktype) { case JBD2_DESCRIPTOR_BLOCK: /* If it is a valid descriptor block, replay it - * in pass REPLAY; otherwise, just skip over the - * blocks it describes. */ + * in pass REPLAY; if journal_checksums enabled, then + * calculate checksums in PASS_SCAN, otherwise, + * just skip over the blocks it describes. */ if (pass != PASS_REPLAY) { + if (pass == PASS_SCAN && + JBD2_HAS_COMPAT_FEATURE(journal, + JBD2_FEATURE_COMPAT_CHECKSUM) && + !info->end_transaction) { + if (calc_chksums(journal, bh, + &next_log_block, + &crc32_sum)) { + put_bh(bh); + break; + } + put_bh(bh); + continue; + } next_log_block += count_tags(journal, bh); wrap(journal, next_log_block); - brelse(bh); + put_bh(bh); continue; } @@ -488,7 +535,7 @@ static int do_one_pass(journal_t *journal, memcpy(nbh->b_data, obh->b_data, journal->j_blocksize); if (flags & JBD2_FLAG_ESCAPE) { - *((__be32 *)bh->b_data) = + *((__be32 *)nbh->b_data) = cpu_to_be32(JBD2_MAGIC_NUMBER); } @@ -516,9 +563,96 @@ static int do_one_pass(journal_t *journal, continue; case JBD2_COMMIT_BLOCK: - /* Found an expected commit block: not much to - * do other than move on to the next sequence + /* How to differentiate between interrupted commit + * and journal corruption ? + * + * {nth transaction} + * Checksum Verification Failed + * | + * ____________________ + * | | + * async_commit sync_commit + * | | + * | GO TO NEXT "Journal Corruption" + * | TRANSACTION + * | + * {(n+1)th transanction} + * | + * _______|______________ + * | | + * Commit block found Commit block not found + * | | + * "Journal Corruption" | + * _____________|_________ + * | | + * nth trans corrupt OR nth trans + * and (n+1)th interrupted interrupted + * before commit block + * could reach the disk. + * (Cannot find the difference in above + * mentioned conditions. Hence assume + * "Interrupted Commit".) + */ + + /* Found an expected commit block: if checksums + * are present verify them in PASS_SCAN; else not + * much to do other than move on to the next sequence * number. */ + if (pass == PASS_SCAN && + JBD2_HAS_COMPAT_FEATURE(journal, + JBD2_FEATURE_COMPAT_CHECKSUM)) { + int chksum_err, chksum_seen; + struct commit_header *cbh = + (struct commit_header *)bh->b_data; + unsigned found_chksum = + be32_to_cpu(cbh->h_chksum[0]); + + chksum_err = chksum_seen = 0; + + if (info->end_transaction) { + printk(KERN_ERR "JBD: Transaction %u " + "found to be corrupt.\n", + next_commit_ID - 1); + brelse(bh); + break; + } + + if (crc32_sum == found_chksum && + cbh->h_chksum_type == JBD2_CRC32_CHKSUM && + cbh->h_chksum_size == + JBD2_CRC32_CHKSUM_SIZE) + chksum_seen = 1; + else if (!(cbh->h_chksum_type == 0 && + cbh->h_chksum_size == 0 && + found_chksum == 0 && + !chksum_seen)) + /* + * If fs is mounted using an old kernel and then + * kernel with journal_chksum is used then we + * get a situation where the journal flag has + * checksum flag set but checksums are not + * present i.e chksum = 0, in the individual + * commit blocks. + * Hence to avoid checksum failures, in this + * situation, this extra check is added. + */ + chksum_err = 1; + + if (chksum_err) { + info->end_transaction = next_commit_ID; + + if (!JBD2_HAS_INCOMPAT_FEATURE(journal, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)){ + printk(KERN_ERR + "JBD: Transaction %u " + "found to be corrupt.\n", + next_commit_ID); + brelse(bh); + break; + } + } + crc32_sum = ~0; + } brelse(bh); next_commit_ID++; continue; @@ -554,9 +688,10 @@ static int do_one_pass(journal_t *journal, * transaction marks the end of the valid log. */ - if (pass == PASS_SCAN) - info->end_transaction = next_commit_ID; - else { + if (pass == PASS_SCAN) { + if (!info->end_transaction) + info->end_transaction = next_commit_ID; + } else { /* It's really bad news if different passes end up at * different places (but possible due to IO errors). */ if (info->end_transaction != next_commit_ID) { diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 3595fd432d5..2e1453a5e99 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -171,14 +171,16 @@ int __init jbd2_journal_init_revoke_caches(void) { jbd2_revoke_record_cache = kmem_cache_create("jbd2_revoke_record", sizeof(struct jbd2_revoke_record_s), - 0, SLAB_HWCACHE_ALIGN, NULL); - if (jbd2_revoke_record_cache == 0) + 0, + SLAB_HWCACHE_ALIGN|SLAB_TEMPORARY, + NULL); + if (!jbd2_revoke_record_cache) return -ENOMEM; jbd2_revoke_table_cache = kmem_cache_create("jbd2_revoke_table", sizeof(struct jbd2_revoke_table_s), - 0, 0, NULL); - if (jbd2_revoke_table_cache == 0) { + 0, SLAB_TEMPORARY, NULL); + if (!jbd2_revoke_table_cache) { kmem_cache_destroy(jbd2_revoke_record_cache); jbd2_revoke_record_cache = NULL; return -ENOMEM; diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index b1fcf2b3dca..b9b0b6f899b 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -54,11 +54,13 @@ jbd2_get_transaction(journal_t *journal, transaction_t *transaction) spin_lock_init(&transaction->t_handle_lock); /* Set up the commit timer for the new transaction. */ - journal->j_commit_timer.expires = transaction->t_expires; + journal->j_commit_timer.expires = round_jiffies(transaction->t_expires); add_timer(&journal->j_commit_timer); J_ASSERT(journal->j_running_transaction == NULL); journal->j_running_transaction = transaction; + transaction->t_max_wait = 0; + transaction->t_start = jiffies; return transaction; } @@ -85,6 +87,7 @@ static int start_this_handle(journal_t *journal, handle_t *handle) int nblocks = handle->h_buffer_credits; transaction_t *new_transaction = NULL; int ret = 0; + unsigned long ts = jiffies; if (nblocks > journal->j_max_transaction_buffers) { printk(KERN_ERR "JBD: %s wants too many credits (%d > %d)\n", @@ -217,6 +220,12 @@ repeat_locked: /* OK, account for the buffers that this operation expects to * use and add the handle to the running transaction. */ + if (time_after(transaction->t_start, ts)) { + ts = jbd2_time_diff(ts, transaction->t_start); + if (ts > transaction->t_max_wait) + transaction->t_max_wait = ts; + } + handle->h_transaction = transaction; transaction->t_outstanding_credits += nblocks; transaction->t_updates++; @@ -232,6 +241,8 @@ out: return ret; } +static struct lock_class_key jbd2_handle_key; + /* Allocate a new handle. This should probably be in a slab... */ static handle_t *new_handle(int nblocks) { @@ -242,6 +253,9 @@ static handle_t *new_handle(int nblocks) handle->h_buffer_credits = nblocks; handle->h_ref = 1; + lockdep_init_map(&handle->h_lockdep_map, "jbd2_handle", + &jbd2_handle_key, 0); + return handle; } @@ -284,7 +298,11 @@ handle_t *jbd2_journal_start(journal_t *journal, int nblocks) jbd2_free_handle(handle); current->journal_info = NULL; handle = ERR_PTR(err); + goto out; } + + lock_acquire(&handle->h_lockdep_map, 0, 0, 0, 2, _THIS_IP_); +out: return handle; } @@ -1164,7 +1182,7 @@ int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh) } /* That test should have eliminated the following case: */ - J_ASSERT_JH(jh, jh->b_frozen_data == 0); + J_ASSERT_JH(jh, jh->b_frozen_data == NULL); JBUFFER_TRACE(jh, "file as BJ_Metadata"); spin_lock(&journal->j_list_lock); @@ -1410,6 +1428,8 @@ int jbd2_journal_stop(handle_t *handle) spin_unlock(&journal->j_state_lock); } + lock_release(&handle->h_lockdep_map, 1, _THIS_IP_); + jbd2_free_handle(handle); return err; } @@ -1512,7 +1532,7 @@ void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh) J_ASSERT_JH(jh, jh->b_jlist < BJ_Types); if (jh->b_jlist != BJ_None) - J_ASSERT_JH(jh, transaction != 0); + J_ASSERT_JH(jh, transaction != NULL); switch (jh->b_jlist) { case BJ_None: @@ -1581,11 +1601,11 @@ __journal_try_to_free_buffer(journal_t *journal, struct buffer_head *bh) if (buffer_locked(bh) || buffer_dirty(bh)) goto out; - if (jh->b_next_transaction != 0) + if (jh->b_next_transaction != NULL) goto out; spin_lock(&journal->j_list_lock); - if (jh->b_transaction != 0 && jh->b_cp_transaction == 0) { + if (jh->b_transaction != NULL && jh->b_cp_transaction == NULL) { if (jh->b_jlist == BJ_SyncData || jh->b_jlist == BJ_Locked) { /* A written-back ordered data buffer */ JBUFFER_TRACE(jh, "release data"); @@ -1593,7 +1613,7 @@ __journal_try_to_free_buffer(journal_t *journal, struct buffer_head *bh) jbd2_journal_remove_journal_head(bh); __brelse(bh); } - } else if (jh->b_cp_transaction != 0 && jh->b_transaction == 0) { + } else if (jh->b_cp_transaction != NULL && jh->b_transaction == NULL) { /* written-back checkpointed metadata buffer */ if (jh->b_jlist == BJ_None) { JBUFFER_TRACE(jh, "remove from checkpoint list"); @@ -1953,7 +1973,7 @@ void __jbd2_journal_file_buffer(struct journal_head *jh, J_ASSERT_JH(jh, jh->b_jlist < BJ_Types); J_ASSERT_JH(jh, jh->b_transaction == transaction || - jh->b_transaction == 0); + jh->b_transaction == NULL); if (jh->b_transaction && jh->b_jlist == jlist) return; diff --git a/fs/jffs2/acl.c b/fs/jffs2/acl.c index 77fc5838609..4c80404a9ab 100644 --- a/fs/jffs2/acl.c +++ b/fs/jffs2/acl.c @@ -176,7 +176,7 @@ static void jffs2_iset_acl(struct inode *inode, struct posix_acl **i_acl, struct spin_unlock(&inode->i_lock); } -struct posix_acl *jffs2_get_acl(struct inode *inode, int type) +static struct posix_acl *jffs2_get_acl(struct inode *inode, int type) { struct jffs2_inode_info *f = JFFS2_INODE_INFO(inode); struct posix_acl *acl; @@ -345,8 +345,10 @@ int jffs2_init_acl_pre(struct inode *dir_i, struct inode *inode, int *i_mode) if (!clone) return -ENOMEM; rc = posix_acl_create_masq(clone, (mode_t *)i_mode); - if (rc < 0) + if (rc < 0) { + posix_acl_release(clone); return rc; + } if (rc > 0) jffs2_iset_acl(inode, &f->i_acl_access, clone); diff --git a/fs/jffs2/acl.h b/fs/jffs2/acl.h index 76c6ebd1acd..0bb7f003fd8 100644 --- a/fs/jffs2/acl.h +++ b/fs/jffs2/acl.h @@ -28,7 +28,6 @@ struct jffs2_acl_header { #define JFFS2_ACL_NOT_CACHED ((void *)-1) -extern struct posix_acl *jffs2_get_acl(struct inode *inode, int type); extern int jffs2_permission(struct inode *, int, struct nameidata *); extern int jffs2_acl_chmod(struct inode *); extern int jffs2_init_acl_pre(struct inode *, struct inode *, int *); @@ -40,7 +39,6 @@ extern struct xattr_handler jffs2_acl_default_xattr_handler; #else -#define jffs2_get_acl(inode, type) (NULL) #define jffs2_permission (NULL) #define jffs2_acl_chmod(inode) (0) #define jffs2_init_acl_pre(dir_i,inode,mode) (0) diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c index 787e392ffd4..f948f7e6ec8 100644 --- a/fs/jffs2/dir.c +++ b/fs/jffs2/dir.c @@ -101,10 +101,10 @@ static struct dentry *jffs2_lookup(struct inode *dir_i, struct dentry *target, ino = fd->ino; up(&dir_f->sem); if (ino) { - inode = iget(dir_i->i_sb, ino); - if (!inode) { + inode = jffs2_iget(dir_i->i_sb, ino); + if (IS_ERR(inode)) { printk(KERN_WARNING "iget() failed for ino #%u\n", ino); - return (ERR_PTR(-EIO)); + return ERR_CAST(inode); } } diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c index f9c5dd6f4b6..dcc2734e0b5 100644 --- a/fs/jffs2/file.c +++ b/fs/jffs2/file.c @@ -129,7 +129,7 @@ static int jffs2_write_begin(struct file *filp, struct address_space *mapping, struct inode *inode = mapping->host; struct jffs2_inode_info *f = JFFS2_INODE_INFO(inode); pgoff_t index = pos >> PAGE_CACHE_SHIFT; - uint32_t pageofs = pos & (PAGE_CACHE_SIZE - 1); + uint32_t pageofs = index << PAGE_CACHE_SHIFT; int ret = 0; pg = __grab_cache_page(mapping, index); diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c index d2e06f7ea96..e26ea78c789 100644 --- a/fs/jffs2/fs.c +++ b/fs/jffs2/fs.c @@ -97,11 +97,7 @@ int jffs2_do_setattr (struct inode *inode, struct iattr *iattr) ri->gid = cpu_to_je16((ivalid & ATTR_GID)?iattr->ia_gid:inode->i_gid); if (ivalid & ATTR_MODE) - if (iattr->ia_mode & S_ISGID && - !in_group_p(je16_to_cpu(ri->gid)) && !capable(CAP_FSETID)) - ri->mode = cpu_to_jemode(iattr->ia_mode & ~S_ISGID); - else - ri->mode = cpu_to_jemode(iattr->ia_mode); + ri->mode = cpu_to_jemode(iattr->ia_mode); else ri->mode = cpu_to_jemode(inode->i_mode); @@ -230,16 +226,23 @@ void jffs2_clear_inode (struct inode *inode) jffs2_do_clear_inode(c, f); } -void jffs2_read_inode (struct inode *inode) +struct inode *jffs2_iget(struct super_block *sb, unsigned long ino) { struct jffs2_inode_info *f; struct jffs2_sb_info *c; struct jffs2_raw_inode latest_node; union jffs2_device_node jdev; + struct inode *inode; dev_t rdev = 0; int ret; - D1(printk(KERN_DEBUG "jffs2_read_inode(): inode->i_ino == %lu\n", inode->i_ino)); + D1(printk(KERN_DEBUG "jffs2_iget(): ino == %lu\n", ino)); + + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; f = JFFS2_INODE_INFO(inode); c = JFFS2_SB_INFO(inode->i_sb); @@ -250,9 +253,9 @@ void jffs2_read_inode (struct inode *inode) ret = jffs2_do_read_inode(c, f, inode->i_ino, &latest_node); if (ret) { - make_bad_inode(inode); up(&f->sem); - return; + iget_failed(inode); + return ERR_PTR(ret); } inode->i_mode = jemode_to_cpu(latest_node.mode); inode->i_uid = je16_to_cpu(latest_node.uid); @@ -303,19 +306,14 @@ void jffs2_read_inode (struct inode *inode) if (f->metadata->size != sizeof(jdev.old) && f->metadata->size != sizeof(jdev.new)) { printk(KERN_NOTICE "Device node has strange size %d\n", f->metadata->size); - up(&f->sem); - jffs2_do_clear_inode(c, f); - make_bad_inode(inode); - return; + goto error_io; } D1(printk(KERN_DEBUG "Reading device numbers from flash\n")); - if (jffs2_read_dnode(c, f, f->metadata, (char *)&jdev, 0, f->metadata->size) < 0) { + ret = jffs2_read_dnode(c, f, f->metadata, (char *)&jdev, 0, f->metadata->size); + if (ret < 0) { /* Eep */ printk(KERN_NOTICE "Read device numbers for inode %lu failed\n", (unsigned long)inode->i_ino); - up(&f->sem); - jffs2_do_clear_inode(c, f); - make_bad_inode(inode); - return; + goto error; } if (f->metadata->size == sizeof(jdev.old)) rdev = old_decode_dev(je16_to_cpu(jdev.old)); @@ -335,6 +333,16 @@ void jffs2_read_inode (struct inode *inode) up(&f->sem); D1(printk(KERN_DEBUG "jffs2_read_inode() returning\n")); + unlock_new_inode(inode); + return inode; + +error_io: + ret = -EIO; +error: + up(&f->sem); + jffs2_do_clear_inode(c, f); + iget_failed(inode); + return ERR_PTR(ret); } void jffs2_dirty_inode(struct inode *inode) @@ -522,15 +530,16 @@ int jffs2_do_fill_super(struct super_block *sb, void *data, int silent) if ((ret = jffs2_do_mount_fs(c))) goto out_inohash; - ret = -EINVAL; - D1(printk(KERN_DEBUG "jffs2_do_fill_super(): Getting root inode\n")); - root_i = iget(sb, 1); - if (is_bad_inode(root_i)) { + root_i = jffs2_iget(sb, 1); + if (IS_ERR(root_i)) { D1(printk(KERN_WARNING "get root inode failed\n")); - goto out_root_i; + ret = PTR_ERR(root_i); + goto out_root; } + ret = -ENOMEM; + D1(printk(KERN_DEBUG "jffs2_do_fill_super(): d_alloc_root()\n")); sb->s_root = d_alloc_root(root_i); if (!sb->s_root) @@ -546,6 +555,7 @@ int jffs2_do_fill_super(struct super_block *sb, void *data, int silent) out_root_i: iput(root_i); +out_root: jffs2_free_ino_caches(c); jffs2_free_raw_node_refs(c); if (jffs2_blocks_use_vmalloc(c)) @@ -615,9 +625,9 @@ struct jffs2_inode_info *jffs2_gc_fetch_inode(struct jffs2_sb_info *c, jffs2_do_unlink() would need the alloc_sem and we have it. Just iget() it, and if read_inode() is necessary that's OK. */ - inode = iget(OFNI_BS_2SFFJ(c), inum); - if (!inode) - return ERR_PTR(-ENOMEM); + inode = jffs2_iget(OFNI_BS_2SFFJ(c), inum); + if (IS_ERR(inode)) + return ERR_CAST(inode); } if (is_bad_inode(inode)) { printk(KERN_NOTICE "Eep. read_inode() failed for ino #%u. nlink %d\n", diff --git a/fs/jffs2/nodelist.c b/fs/jffs2/nodelist.c index 4bf86088b3a..87c6f555e1a 100644 --- a/fs/jffs2/nodelist.c +++ b/fs/jffs2/nodelist.c @@ -32,15 +32,18 @@ void jffs2_add_fd_to_list(struct jffs2_sb_info *c, struct jffs2_full_dirent *new if ((*prev)->nhash == new->nhash && !strcmp((*prev)->name, new->name)) { /* Duplicate. Free one */ if (new->version < (*prev)->version) { - dbg_dentlist("Eep! Marking new dirent node is obsolete, old is \"%s\", ino #%u\n", + dbg_dentlist("Eep! Marking new dirent node obsolete, old is \"%s\", ino #%u\n", (*prev)->name, (*prev)->ino); jffs2_mark_node_obsolete(c, new->raw); jffs2_free_full_dirent(new); } else { - dbg_dentlist("marking old dirent \"%s\", ino #%u bsolete\n", + dbg_dentlist("marking old dirent \"%s\", ino #%u obsolete\n", (*prev)->name, (*prev)->ino); new->next = (*prev)->next; - jffs2_mark_node_obsolete(c, ((*prev)->raw)); + /* It may have been a 'placeholder' deletion dirent, + if jffs2_can_mark_obsolete() (see jffs2_do_unlink()) */ + if ((*prev)->raw) + jffs2_mark_node_obsolete(c, ((*prev)->raw)); jffs2_free_full_dirent(*prev); *prev = new; } diff --git a/fs/jffs2/os-linux.h b/fs/jffs2/os-linux.h index bf64686cf09..1b10d259409 100644 --- a/fs/jffs2/os-linux.h +++ b/fs/jffs2/os-linux.h @@ -175,7 +175,7 @@ extern const struct inode_operations jffs2_symlink_inode_operations; /* fs.c */ int jffs2_setattr (struct dentry *, struct iattr *); int jffs2_do_setattr (struct inode *, struct iattr *); -void jffs2_read_inode (struct inode *); +struct inode *jffs2_iget(struct super_block *, unsigned long); void jffs2_clear_inode (struct inode *); void jffs2_dirty_inode(struct inode *inode); struct inode *jffs2_new_inode (struct inode *dir_i, int mode, diff --git a/fs/jffs2/readinode.c b/fs/jffs2/readinode.c index 2eae5d2dbeb..e512a93d624 100644 --- a/fs/jffs2/readinode.c +++ b/fs/jffs2/readinode.c @@ -37,23 +37,24 @@ static int check_node_data(struct jffs2_sb_info *c, struct jffs2_tmp_dnode_info BUG_ON(tn->csize == 0); - if (!jffs2_is_writebuffered(c)) - goto adj_acc; - /* Calculate how many bytes were already checked */ ofs = ref_offset(ref) + sizeof(struct jffs2_raw_inode); - len = ofs % c->wbuf_pagesize; - if (likely(len)) - len = c->wbuf_pagesize - len; - - if (len >= tn->csize) { - dbg_readinode("no need to check node at %#08x, data length %u, data starts at %#08x - it has already been checked.\n", - ref_offset(ref), tn->csize, ofs); - goto adj_acc; - } + len = tn->csize; + + if (jffs2_is_writebuffered(c)) { + int adj = ofs % c->wbuf_pagesize; + if (likely(adj)) + adj = c->wbuf_pagesize - adj; + + if (adj >= tn->csize) { + dbg_readinode("no need to check node at %#08x, data length %u, data starts at %#08x - it has already been checked.\n", + ref_offset(ref), tn->csize, ofs); + goto adj_acc; + } - ofs += len; - len = tn->csize - len; + ofs += adj; + len -= adj; + } dbg_readinode("check node at %#08x, data length %u, partial CRC %#08x, correct CRC %#08x, data starts at %#08x, start checking from %#08x - %u bytes.\n", ref_offset(ref), tn->csize, tn->partial_crc, tn->data_crc, ofs - len, ofs, len); @@ -63,7 +64,7 @@ static int check_node_data(struct jffs2_sb_info *c, struct jffs2_tmp_dnode_info * adding and jffs2_flash_read_end() interface. */ if (c->mtd->point) { err = c->mtd->point(c->mtd, ofs, len, &retlen, &buffer); - if (!err && retlen < tn->csize) { + if (!err && retlen < len) { JFFS2_WARNING("MTD point returned len too short: %zu instead of %u.\n", retlen, tn->csize); c->mtd->unpoint(c->mtd, buffer, ofs, retlen); } else if (err) @@ -741,7 +742,7 @@ static inline int read_dnode(struct jffs2_sb_info *c, struct jffs2_raw_node_ref * are not obsolete. * * Of course, this optimization only makes sense in case - * of NAND flashes (or other flashes whith + * of NAND flashes (or other flashes with * !jffs2_can_mark_obsolete()), since on NOR flashes * nodes are marked obsolete physically. * diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index ffa447511e6..4677355996c 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -65,7 +65,6 @@ static const struct super_operations jffs2_super_operations = { .alloc_inode = jffs2_alloc_inode, .destroy_inode =jffs2_destroy_inode, - .read_inode = jffs2_read_inode, .put_super = jffs2_put_super, .write_super = jffs2_write_super, .statfs = jffs2_statfs, diff --git a/fs/jffs2/write.c b/fs/jffs2/write.c index 147e2cbee9e..776f13cbf2b 100644 --- a/fs/jffs2/write.c +++ b/fs/jffs2/write.c @@ -177,7 +177,7 @@ struct jffs2_full_dnode *jffs2_write_dnode(struct jffs2_sb_info *c, struct jffs2 void *hold_err = fn->raw; /* Release the full_dnode which is now useless, and return */ jffs2_free_full_dnode(fn); - return ERR_PTR(PTR_ERR(hold_err)); + return ERR_CAST(hold_err); } fn->ofs = je32_to_cpu(ri->offset); fn->size = je32_to_cpu(ri->dsize); @@ -313,7 +313,7 @@ struct jffs2_full_dirent *jffs2_write_dirent(struct jffs2_sb_info *c, struct jff void *hold_err = fd->raw; /* Release the full_dirent which is now useless, and return */ jffs2_free_full_dirent(fd); - return ERR_PTR(PTR_ERR(hold_err)); + return ERR_CAST(hold_err); } if (retried) { @@ -582,7 +582,7 @@ int jffs2_do_unlink(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, jffs2_add_fd_to_list(c, fd, &dir_f->dents); up(&dir_f->sem); } else { - struct jffs2_full_dirent **prev = &dir_f->dents; + struct jffs2_full_dirent *fd = dir_f->dents; uint32_t nhash = full_name_hash(name, namelen); /* We don't actually want to reserve any space, but we do @@ -590,21 +590,22 @@ int jffs2_do_unlink(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, down(&c->alloc_sem); down(&dir_f->sem); - while ((*prev) && (*prev)->nhash <= nhash) { - if ((*prev)->nhash == nhash && - !memcmp((*prev)->name, name, namelen) && - !(*prev)->name[namelen]) { - struct jffs2_full_dirent *this = *prev; + for (fd = dir_f->dents; fd; fd = fd->next) { + if (fd->nhash == nhash && + !memcmp(fd->name, name, namelen) && + !fd->name[namelen]) { D1(printk(KERN_DEBUG "Marking old dirent node (ino #%u) @%08x obsolete\n", - this->ino, ref_offset(this->raw))); - - *prev = this->next; - jffs2_mark_node_obsolete(c, (this->raw)); - jffs2_free_full_dirent(this); + fd->ino, ref_offset(fd->raw))); + jffs2_mark_node_obsolete(c, fd->raw); + /* We don't want to remove it from the list immediately, + because that screws up getdents()/seek() semantics even + more than they're screwed already. Turn it into a + node-less deletion dirent instead -- a placeholder */ + fd->raw = NULL; + fd->ino = 0; break; } - prev = &((*prev)->next); } up(&dir_f->sem); } @@ -630,7 +631,8 @@ int jffs2_do_unlink(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, D1(printk(KERN_DEBUG "Removing deletion dirent for \"%s\" from dir ino #%u\n", fd->name, dead_f->inocache->ino)); } - jffs2_mark_node_obsolete(c, fd->raw); + if (fd->raw) + jffs2_mark_node_obsolete(c, fd->raw); jffs2_free_full_dirent(fd); } } diff --git a/fs/jfs/file.c b/fs/jfs/file.c index 87eb93694af..7f6063acaa3 100644 --- a/fs/jfs/file.c +++ b/fs/jfs/file.c @@ -112,5 +112,8 @@ const struct file_operations jfs_file_operations = { .splice_write = generic_file_splice_write, .fsync = jfs_fsync, .release = jfs_release, - .ioctl = jfs_ioctl, + .unlocked_ioctl = jfs_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = jfs_compat_ioctl, +#endif }; diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c index 4672013802e..210339784b5 100644 --- a/fs/jfs/inode.c +++ b/fs/jfs/inode.c @@ -31,11 +31,21 @@ #include "jfs_debug.h" -void jfs_read_inode(struct inode *inode) +struct inode *jfs_iget(struct super_block *sb, unsigned long ino) { - if (diRead(inode)) { - make_bad_inode(inode); - return; + struct inode *inode; + int ret; + + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; + + ret = diRead(inode); + if (ret < 0) { + iget_failed(inode); + return ERR_PTR(ret); } if (S_ISREG(inode->i_mode)) { @@ -55,6 +65,8 @@ void jfs_read_inode(struct inode *inode) inode->i_op = &jfs_file_inode_operations; init_special_inode(inode, inode->i_mode, inode->i_rdev); } + unlock_new_inode(inode); + return inode; } /* diff --git a/fs/jfs/ioctl.c b/fs/jfs/ioctl.c index dfda12a073e..a1f8e375ad2 100644 --- a/fs/jfs/ioctl.c +++ b/fs/jfs/ioctl.c @@ -51,9 +51,9 @@ static long jfs_map_ext2(unsigned long flags, int from) } -int jfs_ioctl(struct inode * inode, struct file * filp, unsigned int cmd, - unsigned long arg) +long jfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { + struct inode *inode = filp->f_dentry->d_inode; struct jfs_inode_info *jfs_inode = JFS_IP(inode); unsigned int flags; @@ -82,6 +82,10 @@ int jfs_ioctl(struct inode * inode, struct file * filp, unsigned int cmd, /* Is it quota file? Do not allow user to mess with it */ if (IS_NOQUOTA(inode)) return -EPERM; + + /* Lock against other parallel changes of flags */ + mutex_lock(&inode->i_mutex); + jfs_get_inode_flags(jfs_inode); oldflags = jfs_inode->mode2; @@ -92,8 +96,10 @@ int jfs_ioctl(struct inode * inode, struct file * filp, unsigned int cmd, if ((oldflags & JFS_IMMUTABLE_FL) || ((flags ^ oldflags) & (JFS_APPEND_FL | JFS_IMMUTABLE_FL))) { - if (!capable(CAP_LINUX_IMMUTABLE)) + if (!capable(CAP_LINUX_IMMUTABLE)) { + mutex_unlock(&inode->i_mutex); return -EPERM; + } } flags = flags & JFS_FL_USER_MODIFIABLE; @@ -101,6 +107,7 @@ int jfs_ioctl(struct inode * inode, struct file * filp, unsigned int cmd, jfs_inode->mode2 = flags; jfs_set_inode_flags(inode); + mutex_unlock(&inode->i_mutex); inode->i_ctime = CURRENT_TIME_SEC; mark_inode_dirty(inode); return 0; @@ -110,3 +117,21 @@ int jfs_ioctl(struct inode * inode, struct file * filp, unsigned int cmd, } } +#ifdef CONFIG_COMPAT +long jfs_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) +{ + /* While these ioctl numbers defined with 'long' and have different + * numbers than the 64bit ABI, + * the actual implementation only deals with ints and is compatible. + */ + switch (cmd) { + case JFS_IOC_GETFLAGS32: + cmd = JFS_IOC_GETFLAGS; + break; + case JFS_IOC_SETFLAGS32: + cmd = JFS_IOC_SETFLAGS; + break; + } + return jfs_ioctl(filp, cmd, arg); +} +#endif diff --git a/fs/jfs/jfs_dinode.h b/fs/jfs/jfs_dinode.h index c387540d342..395c4c0d0f0 100644 --- a/fs/jfs/jfs_dinode.h +++ b/fs/jfs/jfs_dinode.h @@ -170,5 +170,7 @@ struct dinode { #define JFS_IOC_GETFLAGS _IOR('f', 1, long) #define JFS_IOC_SETFLAGS _IOW('f', 2, long) +#define JFS_IOC_GETFLAGS32 _IOR('f', 1, int) +#define JFS_IOC_SETFLAGS32 _IOW('f', 2, int) #endif /*_H_JFS_DINODE */ diff --git a/fs/jfs/jfs_dtree.c b/fs/jfs/jfs_dtree.c index df25ecc418a..4dcc0581999 100644 --- a/fs/jfs/jfs_dtree.c +++ b/fs/jfs/jfs_dtree.c @@ -284,11 +284,11 @@ static struct dir_table_slot *find_index(struct inode *ip, u32 index, release_metapage(*mp); *mp = NULL; } - if (*mp == 0) { + if (!(*mp)) { *lblock = blkno; *mp = read_index_page(ip, blkno); } - if (*mp == 0) { + if (!(*mp)) { jfs_err("free_index: error reading directory table"); return NULL; } @@ -413,7 +413,8 @@ static u32 add_index(tid_t tid, struct inode *ip, s64 bn, int slot) } ip->i_size = PSIZE; - if ((mp = get_index_page(ip, 0)) == 0) { + mp = get_index_page(ip, 0); + if (!mp) { jfs_err("add_index: get_metapage failed!"); xtTruncate(tid, ip, 0, COMMIT_PWMAP); memcpy(&jfs_ip->i_dirtable, temp_table, @@ -461,7 +462,7 @@ static u32 add_index(tid_t tid, struct inode *ip, s64 bn, int slot) } else mp = read_index_page(ip, blkno); - if (mp == 0) { + if (!mp) { jfs_err("add_index: get/read_metapage failed!"); goto clean_up; } @@ -499,7 +500,7 @@ static void free_index(tid_t tid, struct inode *ip, u32 index, u32 next) dirtab_slot = find_index(ip, index, &mp, &lblock); - if (dirtab_slot == 0) + if (!dirtab_slot) return; dirtab_slot->flag = DIR_INDEX_FREE; @@ -526,7 +527,7 @@ static void modify_index(tid_t tid, struct inode *ip, u32 index, s64 bn, dirtab_slot = find_index(ip, index, mp, lblock); - if (dirtab_slot == 0) + if (!dirtab_slot) return; DTSaddress(dirtab_slot, bn); @@ -552,7 +553,7 @@ static int read_index(struct inode *ip, u32 index, struct dir_table_slot *slot; slot = find_index(ip, index, &mp, &lblock); - if (slot == 0) { + if (!slot) { return -EIO; } @@ -592,10 +593,8 @@ int dtSearch(struct inode *ip, struct component_name * key, ino_t * data, struct component_name ciKey; struct super_block *sb = ip->i_sb; - ciKey.name = - (wchar_t *) kmalloc((JFS_NAME_MAX + 1) * sizeof(wchar_t), - GFP_NOFS); - if (ciKey.name == 0) { + ciKey.name = kmalloc((JFS_NAME_MAX + 1) * sizeof(wchar_t), GFP_NOFS); + if (!ciKey.name) { rc = -ENOMEM; goto dtSearch_Exit2; } @@ -957,10 +956,8 @@ static int dtSplitUp(tid_t tid, smp = split->mp; sp = DT_PAGE(ip, smp); - key.name = - (wchar_t *) kmalloc((JFS_NAME_MAX + 2) * sizeof(wchar_t), - GFP_NOFS); - if (key.name == 0) { + key.name = kmalloc((JFS_NAME_MAX + 2) * sizeof(wchar_t), GFP_NOFS); + if (!key.name) { DT_PUTPAGE(smp); rc = -ENOMEM; goto dtSplitUp_Exit; diff --git a/fs/jfs/jfs_dtree.h b/fs/jfs/jfs_dtree.h index 8561c6ecece..cdac2d5bafe 100644 --- a/fs/jfs/jfs_dtree.h +++ b/fs/jfs/jfs_dtree.h @@ -74,7 +74,7 @@ struct idtentry { #define DTIHDRDATALEN 11 /* compute number of slots for entry */ -#define NDTINTERNAL(klen) ( ((4 + (klen)) + (15 - 1)) / 15 ) +#define NDTINTERNAL(klen) (DIV_ROUND_UP((4 + (klen)), 15)) /* @@ -133,7 +133,7 @@ struct dir_table_slot { ( ((s64)((dts)->addr1)) << 32 | __le32_to_cpu((dts)->addr2) ) /* compute number of slots for entry */ -#define NDTLEAF_LEGACY(klen) ( ((2 + (klen)) + (15 - 1)) / 15 ) +#define NDTLEAF_LEGACY(klen) (DIV_ROUND_UP((2 + (klen)), 15)) #define NDTLEAF NDTINTERNAL diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c index 3870ba8b908..9bf29f77173 100644 --- a/fs/jfs/jfs_imap.c +++ b/fs/jfs/jfs_imap.c @@ -381,7 +381,7 @@ int diRead(struct inode *ip) /* read the page of disk inode */ mp = read_metapage(ipimap, pageno << sbi->l2nbperpage, PSIZE, 1); - if (mp == 0) { + if (!mp) { jfs_err("diRead: read_metapage failed"); return -EIO; } @@ -654,7 +654,7 @@ int diWrite(tid_t tid, struct inode *ip) /* read the page of disk inode */ retry: mp = read_metapage(ipimap, pageno << sbi->l2nbperpage, PSIZE, 1); - if (mp == 0) + if (!mp) return -EIO; /* get the pointer to the disk inode */ diff --git a/fs/jfs/jfs_inode.h b/fs/jfs/jfs_inode.h index 8e2cf2cde18..adb2fafcc54 100644 --- a/fs/jfs/jfs_inode.h +++ b/fs/jfs/jfs_inode.h @@ -22,9 +22,9 @@ struct fid; extern struct inode *ialloc(struct inode *, umode_t); extern int jfs_fsync(struct file *, struct dentry *, int); -extern int jfs_ioctl(struct inode *, struct file *, - unsigned int, unsigned long); -extern void jfs_read_inode(struct inode *); +extern long jfs_ioctl(struct file *, unsigned int, unsigned long); +extern long jfs_compat_ioctl(struct file *, unsigned int, unsigned long); +extern struct inode *jfs_iget(struct super_block *, unsigned long); extern int jfs_commit_inode(struct inode *, int); extern int jfs_write_inode(struct inode*, int); extern void jfs_delete_inode(struct inode *); diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c index 15a3974cdee..325a9679b95 100644 --- a/fs/jfs/jfs_logmgr.c +++ b/fs/jfs/jfs_logmgr.c @@ -208,6 +208,17 @@ static struct lmStat { } lmStat; #endif +static void write_special_inodes(struct jfs_log *log, + int (*writer)(struct address_space *)) +{ + struct jfs_sb_info *sbi; + + list_for_each_entry(sbi, &log->sb_list, log_list) { + writer(sbi->ipbmap->i_mapping); + writer(sbi->ipimap->i_mapping); + writer(sbi->direct_inode->i_mapping); + } +} /* * NAME: lmLog() @@ -935,22 +946,13 @@ static int lmLogSync(struct jfs_log * log, int hard_sync) struct lrd lrd; int lsn; struct logsyncblk *lp; - struct jfs_sb_info *sbi; unsigned long flags; /* push dirty metapages out to disk */ if (hard_sync) - list_for_each_entry(sbi, &log->sb_list, log_list) { - filemap_fdatawrite(sbi->ipbmap->i_mapping); - filemap_fdatawrite(sbi->ipimap->i_mapping); - filemap_fdatawrite(sbi->direct_inode->i_mapping); - } + write_special_inodes(log, filemap_fdatawrite); else - list_for_each_entry(sbi, &log->sb_list, log_list) { - filemap_flush(sbi->ipbmap->i_mapping); - filemap_flush(sbi->ipimap->i_mapping); - filemap_flush(sbi->direct_inode->i_mapping); - } + write_special_inodes(log, filemap_flush); /* * forward syncpt @@ -1536,7 +1538,6 @@ void jfs_flush_journal(struct jfs_log *log, int wait) { int i; struct tblock *target = NULL; - struct jfs_sb_info *sbi; /* jfs_write_inode may call us during read-only mount */ if (!log) @@ -1598,11 +1599,7 @@ void jfs_flush_journal(struct jfs_log *log, int wait) if (wait < 2) return; - list_for_each_entry(sbi, &log->sb_list, log_list) { - filemap_fdatawrite(sbi->ipbmap->i_mapping); - filemap_fdatawrite(sbi->ipimap->i_mapping); - filemap_fdatawrite(sbi->direct_inode->i_mapping); - } + write_special_inodes(log, filemap_fdatawrite); /* * If there was recent activity, we may need to wait @@ -1611,6 +1608,7 @@ void jfs_flush_journal(struct jfs_log *log, int wait) if ((!list_empty(&log->cqueue)) || !list_empty(&log->synclist)) { for (i = 0; i < 200; i++) { /* Too much? */ msleep(250); + write_special_inodes(log, filemap_fdatawrite); if (list_empty(&log->cqueue) && list_empty(&log->synclist)) break; @@ -2347,7 +2345,7 @@ int jfsIOWait(void *arg) do { spin_lock_irq(&log_redrive_lock); - while ((bp = log_redrive_list) != 0) { + while ((bp = log_redrive_list)) { log_redrive_list = bp->l_redrive_next; bp->l_redrive_next = NULL; spin_unlock_irq(&log_redrive_lock); diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c index f5cd8d38af7..d1e64f2f2fc 100644 --- a/fs/jfs/jfs_metapage.c +++ b/fs/jfs/jfs_metapage.c @@ -39,11 +39,11 @@ static struct { #endif #define metapage_locked(mp) test_bit(META_locked, &(mp)->flag) -#define trylock_metapage(mp) test_and_set_bit(META_locked, &(mp)->flag) +#define trylock_metapage(mp) test_and_set_bit_lock(META_locked, &(mp)->flag) static inline void unlock_metapage(struct metapage *mp) { - clear_bit(META_locked, &mp->flag); + clear_bit_unlock(META_locked, &mp->flag); wake_up(&mp->wait); } @@ -88,7 +88,7 @@ struct meta_anchor { }; #define mp_anchor(page) ((struct meta_anchor *)page_private(page)) -static inline struct metapage *page_to_mp(struct page *page, uint offset) +static inline struct metapage *page_to_mp(struct page *page, int offset) { if (!PagePrivate(page)) return NULL; @@ -153,7 +153,7 @@ static inline void dec_io(struct page *page, void (*handler) (struct page *)) } #else -static inline struct metapage *page_to_mp(struct page *page, uint offset) +static inline struct metapage *page_to_mp(struct page *page, int offset) { return PagePrivate(page) ? (struct metapage *)page_private(page) : NULL; } @@ -249,7 +249,7 @@ static inline void drop_metapage(struct page *page, struct metapage *mp) */ static sector_t metapage_get_blocks(struct inode *inode, sector_t lblock, - unsigned int *len) + int *len) { int rc = 0; int xflag; @@ -352,25 +352,27 @@ static void metapage_write_end_io(struct bio *bio, int err) static int metapage_writepage(struct page *page, struct writeback_control *wbc) { struct bio *bio = NULL; - unsigned int block_offset; /* block offset of mp within page */ + int block_offset; /* block offset of mp within page */ struct inode *inode = page->mapping->host; - unsigned int blocks_per_mp = JFS_SBI(inode->i_sb)->nbperpage; - unsigned int len; - unsigned int xlen; + int blocks_per_mp = JFS_SBI(inode->i_sb)->nbperpage; + int len; + int xlen; struct metapage *mp; int redirty = 0; sector_t lblock; + int nr_underway = 0; sector_t pblock; sector_t next_block = 0; sector_t page_start; unsigned long bio_bytes = 0; unsigned long bio_offset = 0; - unsigned int offset; + int offset; page_start = (sector_t)page->index << (PAGE_CACHE_SHIFT - inode->i_blkbits); BUG_ON(!PageLocked(page)); BUG_ON(PageWriteback(page)); + set_page_writeback(page); for (offset = 0; offset < PAGE_CACHE_SIZE; offset += PSIZE) { mp = page_to_mp(page, offset); @@ -413,11 +415,10 @@ static int metapage_writepage(struct page *page, struct writeback_control *wbc) if (!bio->bi_size) goto dump_bio; submit_bio(WRITE, bio); + nr_underway++; bio = NULL; - } else { - set_page_writeback(page); + } else inc_io(page); - } xlen = (PAGE_CACHE_SIZE - offset) >> inode->i_blkbits; pblock = metapage_get_blocks(inode, lblock, &xlen); if (!pblock) { @@ -427,7 +428,7 @@ static int metapage_writepage(struct page *page, struct writeback_control *wbc) continue; } set_bit(META_io, &mp->flag); - len = min(xlen, (uint) JFS_SBI(inode->i_sb)->nbperpage); + len = min(xlen, (int)JFS_SBI(inode->i_sb)->nbperpage); bio = bio_alloc(GFP_NOFS, 1); bio->bi_bdev = inode->i_sb->s_bdev; @@ -449,12 +450,16 @@ static int metapage_writepage(struct page *page, struct writeback_control *wbc) goto dump_bio; submit_bio(WRITE, bio); + nr_underway++; } if (redirty) redirty_page_for_writepage(wbc, page); unlock_page(page); + if (nr_underway == 0) + end_page_writeback(page); + return 0; add_failed: /* We should never reach here, since we're only adding one vec */ @@ -475,13 +480,13 @@ static int metapage_readpage(struct file *fp, struct page *page) { struct inode *inode = page->mapping->host; struct bio *bio = NULL; - unsigned int block_offset; - unsigned int blocks_per_page = PAGE_CACHE_SIZE >> inode->i_blkbits; + int block_offset; + int blocks_per_page = PAGE_CACHE_SIZE >> inode->i_blkbits; sector_t page_start; /* address of page in fs blocks */ sector_t pblock; - unsigned int xlen; + int xlen; unsigned int len; - unsigned int offset; + int offset; BUG_ON(!PageLocked(page)); page_start = (sector_t)page->index << @@ -530,7 +535,7 @@ static int metapage_releasepage(struct page *page, gfp_t gfp_mask) { struct metapage *mp; int ret = 1; - unsigned int offset; + int offset; for (offset = 0; offset < PAGE_CACHE_SIZE; offset += PSIZE) { mp = page_to_mp(page, offset); diff --git a/fs/jfs/jfs_mount.c b/fs/jfs/jfs_mount.c index 644429acb8c..7b698f2ec45 100644 --- a/fs/jfs/jfs_mount.c +++ b/fs/jfs/jfs_mount.c @@ -147,7 +147,7 @@ int jfs_mount(struct super_block *sb) */ if ((sbi->mntflag & JFS_BAD_SAIT) == 0) { ipaimap2 = diReadSpecial(sb, AGGREGATE_I, 1); - if (ipaimap2 == 0) { + if (!ipaimap2) { jfs_err("jfs_mount: Faild to read AGGREGATE_I"); rc = -EIO; goto errout35; diff --git a/fs/jfs/jfs_umount.c b/fs/jfs/jfs_umount.c index 7971f37534a..adcf92d3b60 100644 --- a/fs/jfs/jfs_umount.c +++ b/fs/jfs/jfs_umount.c @@ -68,7 +68,7 @@ int jfs_umount(struct super_block *sb) /* * Wait for outstanding transactions to be written to log: */ - jfs_flush_journal(log, 2); + jfs_flush_journal(log, 1); /* * close fileset inode allocation map (aka fileset inode) @@ -146,7 +146,7 @@ int jfs_umount_rw(struct super_block *sb) * * remove file system from log active file system list. */ - jfs_flush_journal(log, 2); + jfs_flush_journal(log, 1); /* * Make sure all metadata makes it to disk diff --git a/fs/jfs/jfs_xtree.c b/fs/jfs/jfs_xtree.c index 1543906a2e0..a000aaa7513 100644 --- a/fs/jfs/jfs_xtree.c +++ b/fs/jfs/jfs_xtree.c @@ -3965,7 +3965,7 @@ s64 xtTruncate(tid_t tid, struct inode *ip, s64 newsize, int flag) * xtTruncate_pmap() * * function: - * Perform truncate to zero lenghth for deleted file, leaving the + * Perform truncate to zero length for deleted file, leaving the * the xtree and working map untouched. This allows the file to * be accessed via open file handles, while the delete of the file * is committed to disk. diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c index 4e0a8493cef..0ba6778edaa 100644 --- a/fs/jfs/namei.c +++ b/fs/jfs/namei.c @@ -1103,8 +1103,8 @@ static int jfs_rename(struct inode *old_dir, struct dentry *old_dentry, * Make sure dest inode number (if any) is what we think it is */ rc = dtSearch(new_dir, &new_dname, &ino, &btstack, JFS_LOOKUP); - if (rc == 0) { - if ((new_ip == 0) || (ino != new_ip->i_ino)) { + if (!rc) { + if ((!new_ip) || (ino != new_ip->i_ino)) { rc = -ESTALE; goto out3; } @@ -1462,12 +1462,10 @@ static struct dentry *jfs_lookup(struct inode *dip, struct dentry *dentry, struc } } - ip = iget(dip->i_sb, inum); - if (ip == NULL || is_bad_inode(ip)) { + ip = jfs_iget(dip->i_sb, inum); + if (IS_ERR(ip)) { jfs_err("jfs_lookup: iget failed on inum %d", (uint) inum); - if (ip) - iput(ip); - return ERR_PTR(-EACCES); + return ERR_CAST(ip); } dentry = d_splice_alias(ip, dentry); @@ -1485,12 +1483,11 @@ static struct inode *jfs_nfs_get_inode(struct super_block *sb, if (ino == 0) return ERR_PTR(-ESTALE); - inode = iget(sb, ino); - if (inode == NULL) - return ERR_PTR(-ENOMEM); + inode = jfs_iget(sb, ino); + if (IS_ERR(inode)) + return ERR_CAST(inode); - if (is_bad_inode(inode) || - (generation && inode->i_generation != generation)) { + if (generation && inode->i_generation != generation) { iput(inode); return ERR_PTR(-ESTALE); } @@ -1521,17 +1518,14 @@ struct dentry *jfs_get_parent(struct dentry *dentry) parent_ino = le32_to_cpu(JFS_IP(dentry->d_inode)->i_dtroot.header.idotdot); - inode = iget(sb, parent_ino); - if (inode) { - if (is_bad_inode(inode)) { + inode = jfs_iget(sb, parent_ino); + if (IS_ERR(inode)) { + parent = ERR_CAST(inode); + } else { + parent = d_alloc_anon(inode); + if (!parent) { + parent = ERR_PTR(-ENOMEM); iput(inode); - parent = ERR_PTR(-EACCES); - } else { - parent = d_alloc_anon(inode); - if (!parent) { - parent = ERR_PTR(-ENOMEM); - iput(inode); - } } } @@ -1562,7 +1556,10 @@ const struct file_operations jfs_dir_operations = { .read = generic_read_dir, .readdir = jfs_readdir, .fsync = jfs_fsync, - .ioctl = jfs_ioctl, + .unlocked_ioctl = jfs_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = jfs_compat_ioctl, +#endif }; static int jfs_ci_hash(struct dentry *dir, struct qstr *this) diff --git a/fs/jfs/resize.c b/fs/jfs/resize.c index 71984ee9534..7f24a0bb08c 100644 --- a/fs/jfs/resize.c +++ b/fs/jfs/resize.c @@ -172,7 +172,7 @@ int jfs_extendfs(struct super_block *sb, s64 newLVSize, int newLogSize) */ t64 = ((newLVSize - newLogSize + BPERDMAP - 1) >> L2BPERDMAP) << L2BPERDMAP; - t32 = ((t64 + (BITSPERPAGE - 1)) / BITSPERPAGE) + 1 + 50; + t32 = DIV_ROUND_UP(t64, BITSPERPAGE) + 1 + 50; newFSCKSize = t32 << sbi->l2nbperpage; newFSCKAddress = newLogAddress - newFSCKSize; diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 314bb4ff1ba..50ea6545173 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -414,7 +414,7 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent) struct inode *inode; int rc; s64 newLVSize = 0; - int flag; + int flag, ret = -EINVAL; jfs_info("In jfs_read_super: s_flags=0x%lx", sb->s_flags); @@ -461,8 +461,10 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent) * Initialize direct-mapping inode/address-space */ inode = new_inode(sb); - if (inode == NULL) + if (inode == NULL) { + ret = -ENOMEM; goto out_kfree; + } inode->i_ino = 0; inode->i_nlink = 1; inode->i_size = sb->s_bdev->bd_inode->i_size; @@ -494,9 +496,11 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent) sb->s_magic = JFS_SUPER_MAGIC; - inode = iget(sb, ROOT_I); - if (!inode || is_bad_inode(inode)) + inode = jfs_iget(sb, ROOT_I); + if (IS_ERR(inode)) { + ret = PTR_ERR(inode); goto out_no_root; + } sb->s_root = d_alloc_root(inode); if (!sb->s_root) goto out_no_root; @@ -536,7 +540,7 @@ out_kfree: if (sbi->nls_tab) unload_nls(sbi->nls_tab); kfree(sbi); - return -EINVAL; + return ret; } static void jfs_write_super_lockfs(struct super_block *sb) @@ -598,6 +602,12 @@ static int jfs_show_options(struct seq_file *seq, struct vfsmount *vfs) seq_printf(seq, ",umask=%03o", sbi->umask); if (sbi->flag & JFS_NOINTEGRITY) seq_puts(seq, ",nointegrity"); + if (sbi->nls_tab) + seq_printf(seq, ",iocharset=%s", sbi->nls_tab->charset); + if (sbi->flag & JFS_ERR_CONTINUE) + seq_printf(seq, ",errors=continue"); + if (sbi->flag & JFS_ERR_PANIC) + seq_printf(seq, ",errors=panic"); #ifdef CONFIG_QUOTA if (sbi->flag & JFS_USRQUOTA) @@ -720,7 +730,6 @@ out: static const struct super_operations jfs_super_operations = { .alloc_inode = jfs_alloc_inode, .destroy_inode = jfs_destroy_inode, - .read_inode = jfs_read_inode, .dirty_inode = jfs_dirty_inode, .write_inode = jfs_write_inode, .delete_inode = jfs_delete_inode, diff --git a/fs/libfs.c b/fs/libfs.c index 6e68b700958..b004dfadd89 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -341,13 +341,10 @@ int simple_prepare_write(struct file *file, struct page *page, unsigned from, unsigned to) { if (!PageUptodate(page)) { - if (to - from != PAGE_CACHE_SIZE) { - void *kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr, 0, from); - memset(kaddr + to, 0, PAGE_CACHE_SIZE - to); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); - } + if (to - from != PAGE_CACHE_SIZE) + zero_user_segments(page, + 0, from, + to, PAGE_CACHE_SIZE); } return 0; } @@ -586,8 +583,8 @@ int simple_transaction_release(struct inode *inode, struct file *file) /* Simple attribute files */ struct simple_attr { - u64 (*get)(void *); - void (*set)(void *, u64); + int (*get)(void *, u64 *); + int (*set)(void *, u64); char get_buf[24]; /* enough to store a u64 and "\n\0" */ char set_buf[24]; void *data; @@ -598,7 +595,7 @@ struct simple_attr { /* simple_attr_open is called by an actual attribute open file operation * to set the attribute specific access operations. */ int simple_attr_open(struct inode *inode, struct file *file, - u64 (*get)(void *), void (*set)(void *, u64), + int (*get)(void *, u64 *), int (*set)(void *, u64), const char *fmt) { struct simple_attr *attr; @@ -618,7 +615,7 @@ int simple_attr_open(struct inode *inode, struct file *file, return nonseekable_open(inode, file); } -int simple_attr_close(struct inode *inode, struct file *file) +int simple_attr_release(struct inode *inode, struct file *file) { kfree(file->private_data); return 0; @@ -637,15 +634,24 @@ ssize_t simple_attr_read(struct file *file, char __user *buf, if (!attr->get) return -EACCES; - mutex_lock(&attr->mutex); - if (*ppos) /* continued read */ + ret = mutex_lock_interruptible(&attr->mutex); + if (ret) + return ret; + + if (*ppos) { /* continued read */ size = strlen(attr->get_buf); - else /* first read */ + } else { /* first read */ + u64 val; + ret = attr->get(attr->data, &val); + if (ret) + goto out; + size = scnprintf(attr->get_buf, sizeof(attr->get_buf), - attr->fmt, - (unsigned long long)attr->get(attr->data)); + attr->fmt, (unsigned long long)val); + } ret = simple_read_from_buffer(buf, len, ppos, attr->get_buf, size); +out: mutex_unlock(&attr->mutex); return ret; } @@ -660,11 +666,13 @@ ssize_t simple_attr_write(struct file *file, const char __user *buf, ssize_t ret; attr = file->private_data; - if (!attr->set) return -EACCES; - mutex_lock(&attr->mutex); + ret = mutex_lock_interruptible(&attr->mutex); + if (ret) + return ret; + ret = -EFAULT; size = min(sizeof(attr->set_buf) - 1, len); if (copy_from_user(attr->set_buf, buf, size)) @@ -796,6 +804,6 @@ EXPORT_SYMBOL(simple_transaction_get); EXPORT_SYMBOL(simple_transaction_read); EXPORT_SYMBOL(simple_transaction_release); EXPORT_SYMBOL_GPL(simple_attr_open); -EXPORT_SYMBOL_GPL(simple_attr_close); +EXPORT_SYMBOL_GPL(simple_attr_release); EXPORT_SYMBOL_GPL(simple_attr_read); EXPORT_SYMBOL_GPL(simple_attr_write); diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c index d070b18e539..0b45fd3a4bf 100644 --- a/fs/lockd/clntlock.c +++ b/fs/lockd/clntlock.c @@ -41,6 +41,48 @@ struct nlm_wait { static LIST_HEAD(nlm_blocked); +/** + * nlmclnt_init - Set up per-NFS mount point lockd data structures + * @nlm_init: pointer to arguments structure + * + * Returns pointer to an appropriate nlm_host struct, + * or an ERR_PTR value. + */ +struct nlm_host *nlmclnt_init(const struct nlmclnt_initdata *nlm_init) +{ + struct nlm_host *host; + u32 nlm_version = (nlm_init->nfs_version == 2) ? 1 : 4; + int status; + + status = lockd_up(nlm_init->protocol); + if (status < 0) + return ERR_PTR(status); + + host = nlmclnt_lookup_host((struct sockaddr_in *)nlm_init->address, + nlm_init->protocol, nlm_version, + nlm_init->hostname, + strlen(nlm_init->hostname)); + if (host == NULL) { + lockd_down(); + return ERR_PTR(-ENOLCK); + } + + return host; +} +EXPORT_SYMBOL_GPL(nlmclnt_init); + +/** + * nlmclnt_done - Release resources allocated by nlmclnt_init() + * @host: nlm_host structure reserved by nlmclnt_init() + * + */ +void nlmclnt_done(struct nlm_host *host) +{ + nlm_release_host(host); + lockd_down(); +} +EXPORT_SYMBOL_GPL(nlmclnt_done); + /* * Queue up a lock for blocking so that the GRANTED request can see it */ diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c index a10343bed16..b6b74a60e1e 100644 --- a/fs/lockd/clntproc.c +++ b/fs/lockd/clntproc.c @@ -145,34 +145,21 @@ static void nlmclnt_release_lockargs(struct nlm_rqst *req) BUG_ON(req->a_args.lock.fl.fl_ops != NULL); } -/* - * This is the main entry point for the NLM client. +/** + * nlmclnt_proc - Perform a single client-side lock request + * @host: address of a valid nlm_host context representing the NLM server + * @cmd: fcntl-style file lock operation to perform + * @fl: address of arguments for the lock operation + * */ -int -nlmclnt_proc(struct inode *inode, int cmd, struct file_lock *fl) +int nlmclnt_proc(struct nlm_host *host, int cmd, struct file_lock *fl) { - struct rpc_clnt *client = NFS_CLIENT(inode); - struct sockaddr_in addr; - struct nfs_server *nfssrv = NFS_SERVER(inode); - struct nlm_host *host; struct nlm_rqst *call; sigset_t oldset; unsigned long flags; - int status, vers; - - vers = (NFS_PROTO(inode)->version == 3) ? 4 : 1; - if (NFS_PROTO(inode)->version > 3) { - printk(KERN_NOTICE "NFSv4 file locking not implemented!\n"); - return -ENOLCK; - } - - rpc_peeraddr(client, (struct sockaddr *) &addr, sizeof(addr)); - host = nlmclnt_lookup_host(&addr, client->cl_xprt->prot, vers, - nfssrv->nfs_client->cl_hostname, - strlen(nfssrv->nfs_client->cl_hostname)); - if (host == NULL) - return -ENOLCK; + int status; + nlm_get_host(host); call = nlm_alloc_call(host); if (call == NULL) return -ENOMEM; @@ -219,7 +206,7 @@ nlmclnt_proc(struct inode *inode, int cmd, struct file_lock *fl) dprintk("lockd: clnt proc returns %d\n", status); return status; } -EXPORT_SYMBOL(nlmclnt_proc); +EXPORT_SYMBOL_GPL(nlmclnt_proc); /* * Allocate an NLM RPC call struct @@ -257,7 +244,7 @@ void nlm_release_call(struct nlm_rqst *call) static void nlmclnt_rpc_release(void *data) { - return nlm_release_call(data); + nlm_release_call(data); } static int nlm_wait_on_grace(wait_queue_head_t *queue) diff --git a/fs/lockd/host.c b/fs/lockd/host.c index 572601e98dc..f1ef49fff11 100644 --- a/fs/lockd/host.c +++ b/fs/lockd/host.c @@ -34,10 +34,10 @@ static DEFINE_MUTEX(nlm_host_mutex); static void nlm_gc_hosts(void); static struct nsm_handle * __nsm_find(const struct sockaddr_in *, - const char *, int, int); + const char *, unsigned int, int); static struct nsm_handle * nsm_find(const struct sockaddr_in *sin, const char *hostname, - int hostname_len); + unsigned int hostname_len); /* * Common host lookup routine for server & client @@ -45,7 +45,8 @@ static struct nsm_handle * nsm_find(const struct sockaddr_in *sin, static struct nlm_host * nlm_lookup_host(int server, const struct sockaddr_in *sin, int proto, int version, const char *hostname, - int hostname_len, const struct sockaddr_in *ssin) + unsigned int hostname_len, + const struct sockaddr_in *ssin) { struct hlist_head *chain; struct hlist_node *pos; @@ -176,7 +177,7 @@ nlm_destroy_host(struct nlm_host *host) */ struct nlm_host * nlmclnt_lookup_host(const struct sockaddr_in *sin, int proto, int version, - const char *hostname, int hostname_len) + const char *hostname, unsigned int hostname_len) { struct sockaddr_in ssin = {0}; @@ -189,7 +190,7 @@ nlmclnt_lookup_host(const struct sockaddr_in *sin, int proto, int version, */ struct nlm_host * nlmsvc_lookup_host(struct svc_rqst *rqstp, - const char *hostname, int hostname_len) + const char *hostname, unsigned int hostname_len) { struct sockaddr_in ssin = {0}; @@ -242,10 +243,18 @@ nlm_bind_host(struct nlm_host *host) .program = &nlm_program, .version = host->h_version, .authflavor = RPC_AUTH_UNIX, - .flags = (RPC_CLNT_CREATE_HARDRTRY | + .flags = (RPC_CLNT_CREATE_NOPING | RPC_CLNT_CREATE_AUTOBIND), }; + /* + * lockd retries server side blocks automatically so we want + * those to be soft RPC calls. Client side calls need to be + * hard RPC tasks. + */ + if (!host->h_server) + args.flags |= RPC_CLNT_CREATE_HARDRTRY; + clnt = rpc_create(&args); if (!IS_ERR(clnt)) host->h_rpcclnt = clnt; @@ -307,7 +316,8 @@ void nlm_release_host(struct nlm_host *host) * Release all resources held by that peer. */ void nlm_host_rebooted(const struct sockaddr_in *sin, - const char *hostname, int hostname_len, + const char *hostname, + unsigned int hostname_len, u32 new_state) { struct hlist_head *chain; @@ -377,8 +387,13 @@ nlm_shutdown_hosts(void) /* First, make all hosts eligible for gc */ dprintk("lockd: nuking all hosts...\n"); for (chain = nlm_hosts; chain < nlm_hosts + NLM_HOST_NRHASH; ++chain) { - hlist_for_each_entry(host, pos, chain, h_hash) + hlist_for_each_entry(host, pos, chain, h_hash) { host->h_expires = jiffies - 1; + if (host->h_rpcclnt) { + rpc_shutdown_client(host->h_rpcclnt); + host->h_rpcclnt = NULL; + } + } } /* Then, perform a garbage collection pass */ @@ -449,7 +464,7 @@ static DEFINE_MUTEX(nsm_mutex); static struct nsm_handle * __nsm_find(const struct sockaddr_in *sin, - const char *hostname, int hostname_len, + const char *hostname, unsigned int hostname_len, int create) { struct nsm_handle *nsm = NULL; @@ -503,7 +518,8 @@ out: } static struct nsm_handle * -nsm_find(const struct sockaddr_in *sin, const char *hostname, int hostname_len) +nsm_find(const struct sockaddr_in *sin, const char *hostname, + unsigned int hostname_len) { return __nsm_find(sin, hostname, hostname_len, 1); } diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c index 82e2192a0d5..1ed8bd4de94 100644 --- a/fs/lockd/svc.c +++ b/fs/lockd/svc.c @@ -153,7 +153,7 @@ lockd(struct svc_rqst *rqstp) */ while ((nlmsvc_users || !signalled()) && nlmsvc_pid == current->pid) { long timeout = MAX_SCHEDULE_TIMEOUT; - char buf[RPC_MAX_ADDRBUFLEN]; + RPC_IFDEBUG(char buf[RPC_MAX_ADDRBUFLEN]); if (signalled()) { flush_signals(current); @@ -219,19 +219,6 @@ lockd(struct svc_rqst *rqstp) module_put_and_exit(0); } - -static int find_socket(struct svc_serv *serv, int proto) -{ - struct svc_sock *svsk; - int found = 0; - list_for_each_entry(svsk, &serv->sv_permsocks, sk_list) - if (svsk->sk_sk->sk_protocol == proto) { - found = 1; - break; - } - return found; -} - /* * Make any sockets that are needed but not present. * If nlm_udpport or nlm_tcpport were set as module @@ -240,17 +227,25 @@ static int find_socket(struct svc_serv *serv, int proto) static int make_socks(struct svc_serv *serv, int proto) { static int warned; + struct svc_xprt *xprt; int err = 0; - if (proto == IPPROTO_UDP || nlm_udpport) - if (!find_socket(serv, IPPROTO_UDP)) - err = svc_makesock(serv, IPPROTO_UDP, nlm_udpport, - SVC_SOCK_DEFAULTS); - if (err >= 0 && (proto == IPPROTO_TCP || nlm_tcpport)) - if (!find_socket(serv, IPPROTO_TCP)) - err = svc_makesock(serv, IPPROTO_TCP, nlm_tcpport, - SVC_SOCK_DEFAULTS); - + if (proto == IPPROTO_UDP || nlm_udpport) { + xprt = svc_find_xprt(serv, "udp", 0, 0); + if (!xprt) + err = svc_create_xprt(serv, "udp", nlm_udpport, + SVC_SOCK_DEFAULTS); + else + svc_xprt_put(xprt); + } + if (err >= 0 && (proto == IPPROTO_TCP || nlm_tcpport)) { + xprt = svc_find_xprt(serv, "tcp", 0, 0); + if (!xprt) + err = svc_create_xprt(serv, "tcp", nlm_tcpport, + SVC_SOCK_DEFAULTS); + else + svc_xprt_put(xprt); + } if (err >= 0) { warned = 0; err = 0; diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c index bf27b6c6cb6..385437e3387 100644 --- a/fs/lockd/svc4proc.c +++ b/fs/lockd/svc4proc.c @@ -84,6 +84,7 @@ nlm4svc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, { struct nlm_host *host; struct nlm_file *file; + int rc = rpc_success; dprintk("lockd: TEST4 called\n"); resp->cookie = argp->cookie; @@ -91,7 +92,7 @@ nlm4svc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, /* Don't accept test requests during grace period */ if (nlmsvc_grace_period) { resp->status = nlm_lck_denied_grace_period; - return rpc_success; + return rc; } /* Obtain client and file */ @@ -101,12 +102,13 @@ nlm4svc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, /* Now check for conflicting locks */ resp->status = nlmsvc_testlock(rqstp, file, &argp->lock, &resp->lock, &resp->cookie); if (resp->status == nlm_drop_reply) - return rpc_drop_reply; + rc = rpc_drop_reply; + else + dprintk("lockd: TEST4 status %d\n", ntohl(resp->status)); - dprintk("lockd: TEST4 status %d\n", ntohl(resp->status)); nlm_release_host(host); nlm_release_file(file); - return rpc_success; + return rc; } static __be32 @@ -115,6 +117,7 @@ nlm4svc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, { struct nlm_host *host; struct nlm_file *file; + int rc = rpc_success; dprintk("lockd: LOCK called\n"); @@ -123,7 +126,7 @@ nlm4svc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, /* Don't accept new lock requests during grace period */ if (nlmsvc_grace_period && !argp->reclaim) { resp->status = nlm_lck_denied_grace_period; - return rpc_success; + return rc; } /* Obtain client and file */ @@ -146,12 +149,13 @@ nlm4svc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, resp->status = nlmsvc_lock(rqstp, file, &argp->lock, argp->block, &argp->cookie); if (resp->status == nlm_drop_reply) - return rpc_drop_reply; + rc = rpc_drop_reply; + else + dprintk("lockd: LOCK status %d\n", ntohl(resp->status)); - dprintk("lockd: LOCK status %d\n", ntohl(resp->status)); nlm_release_host(host); nlm_release_file(file); - return rpc_success; + return rc; } static __be32 diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c index d120ec39bcb..fe9bdb4a220 100644 --- a/fs/lockd/svclock.c +++ b/fs/lockd/svclock.c @@ -501,25 +501,29 @@ nlmsvc_testlock(struct svc_rqst *rqstp, struct nlm_file *file, block, block->b_flags, block->b_fl); if (block->b_flags & B_TIMED_OUT) { nlmsvc_unlink_block(block); - return nlm_lck_denied; + ret = nlm_lck_denied; + goto out; } if (block->b_flags & B_GOT_CALLBACK) { + nlmsvc_unlink_block(block); if (block->b_fl != NULL && block->b_fl->fl_type != F_UNLCK) { lock->fl = *block->b_fl; goto conf_lock; - } - else { - nlmsvc_unlink_block(block); - return nlm_granted; + } else { + ret = nlm_granted; + goto out; } } - return nlm_drop_reply; + ret = nlm_drop_reply; + goto out; } error = vfs_test_lock(file->f_file, &lock->fl); - if (error == -EINPROGRESS) - return nlmsvc_defer_lock_rqst(rqstp, block); + if (error == -EINPROGRESS) { + ret = nlmsvc_defer_lock_rqst(rqstp, block); + goto out; + } if (error) { ret = nlm_lck_denied_nolocks; goto out; @@ -759,11 +763,20 @@ callback: dprintk("lockd: GRANTing blocked lock.\n"); block->b_granted = 1; - /* Schedule next grant callback in 30 seconds */ - nlmsvc_insert_block(block, 30 * HZ); + /* keep block on the list, but don't reattempt until the RPC + * completes or the submission fails + */ + nlmsvc_insert_block(block, NLM_NEVER); + + /* Call the client -- use a soft RPC task since nlmsvc_retry_blocked + * will queue up a new one if this one times out + */ + error = nlm_async_call(block->b_call, NLMPROC_GRANTED_MSG, + &nlmsvc_grant_ops); - /* Call the client */ - nlm_async_call(block->b_call, NLMPROC_GRANTED_MSG, &nlmsvc_grant_ops); + /* RPC submission failed, wait a bit and retry */ + if (error < 0) + nlmsvc_insert_block(block, 10 * HZ); } /* @@ -782,6 +795,17 @@ static void nlmsvc_grant_callback(struct rpc_task *task, void *data) dprintk("lockd: GRANT_MSG RPC callback\n"); + /* if the block is not on a list at this point then it has + * been invalidated. Don't try to requeue it. + * + * FIXME: it's possible that the block is removed from the list + * after this check but before the nlmsvc_insert_block. In that + * case it will be added back. Perhaps we need better locking + * for nlm_blocked? + */ + if (list_empty(&block->b_list)) + return; + /* Technically, we should down the file semaphore here. Since we * move the block towards the head of the queue only, no harm * can be done, though. */ diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c index 9cd5c8b3759..88379cc6e0b 100644 --- a/fs/lockd/svcproc.c +++ b/fs/lockd/svcproc.c @@ -113,6 +113,7 @@ nlmsvc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, { struct nlm_host *host; struct nlm_file *file; + int rc = rpc_success; dprintk("lockd: TEST called\n"); resp->cookie = argp->cookie; @@ -120,7 +121,7 @@ nlmsvc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, /* Don't accept test requests during grace period */ if (nlmsvc_grace_period) { resp->status = nlm_lck_denied_grace_period; - return rpc_success; + return rc; } /* Obtain client and file */ @@ -130,13 +131,14 @@ nlmsvc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, /* Now check for conflicting locks */ resp->status = cast_status(nlmsvc_testlock(rqstp, file, &argp->lock, &resp->lock, &resp->cookie)); if (resp->status == nlm_drop_reply) - return rpc_drop_reply; + rc = rpc_drop_reply; + else + dprintk("lockd: TEST status %d vers %d\n", + ntohl(resp->status), rqstp->rq_vers); - dprintk("lockd: TEST status %d vers %d\n", - ntohl(resp->status), rqstp->rq_vers); nlm_release_host(host); nlm_release_file(file); - return rpc_success; + return rc; } static __be32 @@ -145,6 +147,7 @@ nlmsvc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, { struct nlm_host *host; struct nlm_file *file; + int rc = rpc_success; dprintk("lockd: LOCK called\n"); @@ -153,7 +156,7 @@ nlmsvc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, /* Don't accept new lock requests during grace period */ if (nlmsvc_grace_period && !argp->reclaim) { resp->status = nlm_lck_denied_grace_period; - return rpc_success; + return rc; } /* Obtain client and file */ @@ -176,12 +179,13 @@ nlmsvc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, resp->status = cast_status(nlmsvc_lock(rqstp, file, &argp->lock, argp->block, &argp->cookie)); if (resp->status == nlm_drop_reply) - return rpc_drop_reply; + rc = rpc_drop_reply; + else + dprintk("lockd: LOCK status %d\n", ntohl(resp->status)); - dprintk("lockd: LOCK status %d\n", ntohl(resp->status)); nlm_release_host(host); nlm_release_file(file); - return rpc_success; + return rc; } static __be32 diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c index 84ebba33b98..dbbefbcd671 100644 --- a/fs/lockd/svcsubs.c +++ b/fs/lockd/svcsubs.c @@ -87,7 +87,7 @@ nlm_lookup_file(struct svc_rqst *rqstp, struct nlm_file **result, unsigned int hash; __be32 nfserr; - nlm_debug_print_fh("nlm_file_lookup", f); + nlm_debug_print_fh("nlm_lookup_file", f); hash = file_hash(f); diff --git a/fs/lockd/xdr.c b/fs/lockd/xdr.c index 633653bff94..3e459e18cc3 100644 --- a/fs/lockd/xdr.c +++ b/fs/lockd/xdr.c @@ -612,8 +612,7 @@ const char *nlmdbg_cookie2a(const struct nlm_cookie *cookie) * called with BKL held. */ static char buf[2*NLM_MAXCOOKIELEN+1]; - int i; - int len = sizeof(buf); + unsigned int i, len = sizeof(buf); char *p = buf; len--; /* allow for trailing \0 */ diff --git a/fs/locks.c b/fs/locks.c index 8b8388eca05..43c0af21a0c 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -125,6 +125,7 @@ #include <linux/syscalls.h> #include <linux/time.h> #include <linux/rcupdate.h> +#include <linux/pid_namespace.h> #include <asm/semaphore.h> #include <asm/uaccess.h> @@ -185,6 +186,7 @@ void locks_init_lock(struct file_lock *fl) fl->fl_fasync = NULL; fl->fl_owner = NULL; fl->fl_pid = 0; + fl->fl_nspid = NULL; fl->fl_file = NULL; fl->fl_flags = 0; fl->fl_type = 0; @@ -553,6 +555,8 @@ static void locks_insert_lock(struct file_lock **pos, struct file_lock *fl) { list_add(&fl->fl_link, &file_lock_list); + fl->fl_nspid = get_pid(task_tgid(current)); + /* insert into file's list */ fl->fl_next = *pos; *pos = fl; @@ -584,6 +588,11 @@ static void locks_delete_lock(struct file_lock **thisfl_p) if (fl->fl_ops && fl->fl_ops->fl_remove) fl->fl_ops->fl_remove(fl); + if (fl->fl_nspid) { + put_pid(fl->fl_nspid); + fl->fl_nspid = NULL; + } + locks_wake_up_blocks(fl); locks_free_lock(fl); } @@ -634,33 +643,6 @@ static int flock_locks_conflict(struct file_lock *caller_fl, struct file_lock *s return (locks_conflict(caller_fl, sys_fl)); } -static int interruptible_sleep_on_locked(wait_queue_head_t *fl_wait, int timeout) -{ - int result = 0; - DECLARE_WAITQUEUE(wait, current); - - __set_current_state(TASK_INTERRUPTIBLE); - add_wait_queue(fl_wait, &wait); - if (timeout == 0) - schedule(); - else - result = schedule_timeout(timeout); - if (signal_pending(current)) - result = -ERESTARTSYS; - remove_wait_queue(fl_wait, &wait); - __set_current_state(TASK_RUNNING); - return result; -} - -static int locks_block_on_timeout(struct file_lock *blocker, struct file_lock *waiter, int time) -{ - int result; - locks_insert_block(blocker, waiter); - result = interruptible_sleep_on_locked(&waiter->fl_wait, time); - __locks_delete_block(waiter); - return result; -} - void posix_test_lock(struct file *filp, struct file_lock *fl) { @@ -673,55 +655,66 @@ posix_test_lock(struct file *filp, struct file_lock *fl) if (posix_locks_conflict(fl, cfl)) break; } - if (cfl) + if (cfl) { __locks_copy_lock(fl, cfl); - else + if (cfl->fl_nspid) + fl->fl_pid = pid_vnr(cfl->fl_nspid); + } else fl->fl_type = F_UNLCK; unlock_kernel(); return; } - EXPORT_SYMBOL(posix_test_lock); -/* This function tests for deadlock condition before putting a process to - * sleep. The detection scheme is no longer recursive. Recursive was neat, - * but dangerous - we risked stack corruption if the lock data was bad, or - * if the recursion was too deep for any other reason. +/* + * Deadlock detection: + * + * We attempt to detect deadlocks that are due purely to posix file + * locks. + * + * We assume that a task can be waiting for at most one lock at a time. + * So for any acquired lock, the process holding that lock may be + * waiting on at most one other lock. That lock in turns may be held by + * someone waiting for at most one other lock. Given a requested lock + * caller_fl which is about to wait for a conflicting lock block_fl, we + * follow this chain of waiters to ensure we are not about to create a + * cycle. * - * We rely on the fact that a task can only be on one lock's wait queue - * at a time. When we find blocked_task on a wait queue we can re-search - * with blocked_task equal to that queue's owner, until either blocked_task - * isn't found, or blocked_task is found on a queue owned by my_task. + * Since we do this before we ever put a process to sleep on a lock, we + * are ensured that there is never a cycle; that is what guarantees that + * the while() loop in posix_locks_deadlock() eventually completes. * - * Note: the above assumption may not be true when handling lock requests - * from a broken NFS client. But broken NFS clients have a lot more to - * worry about than proper deadlock detection anyway... --okir + * Note: the above assumption may not be true when handling lock + * requests from a broken NFS client. It may also fail in the presence + * of tasks (such as posix threads) sharing the same open file table. * - * However, the failure of this assumption (also possible in the case of - * multiple tasks sharing the same open file table) also means there's no - * guarantee that the loop below will terminate. As a hack, we give up - * after a few iterations. + * To handle those cases, we just bail out after a few iterations. */ #define MAX_DEADLK_ITERATIONS 10 +/* Find a lock that the owner of the given block_fl is blocking on. */ +static struct file_lock *what_owner_is_waiting_for(struct file_lock *block_fl) +{ + struct file_lock *fl; + + list_for_each_entry(fl, &blocked_list, fl_link) { + if (posix_same_owner(fl, block_fl)) + return fl->fl_next; + } + return NULL; +} + static int posix_locks_deadlock(struct file_lock *caller_fl, struct file_lock *block_fl) { - struct file_lock *fl; int i = 0; -next_task: - if (posix_same_owner(caller_fl, block_fl)) - return 1; - list_for_each_entry(fl, &blocked_list, fl_link) { - if (posix_same_owner(fl, block_fl)) { - if (i++ > MAX_DEADLK_ITERATIONS) - return 0; - fl = fl->fl_next; - block_fl = fl; - goto next_task; - } + while ((block_fl = what_owner_is_waiting_for(block_fl))) { + if (i++ > MAX_DEADLK_ITERATIONS) + return 0; + if (posix_same_owner(caller_fl, block_fl)) + return 1; } return 0; } @@ -1256,7 +1249,10 @@ restart: if (break_time == 0) break_time++; } - error = locks_block_on_timeout(flock, new_fl, break_time); + locks_insert_block(flock, new_fl); + error = wait_event_interruptible_timeout(new_fl->fl_wait, + !new_fl->fl_next, break_time); + __locks_delete_block(new_fl); if (error >= 0) { if (error == 0) time_out_leases(inode); @@ -1279,13 +1275,13 @@ out: EXPORT_SYMBOL(__break_lease); /** - * lease_get_mtime + * lease_get_mtime - get the last modified time of an inode * @inode: the inode * @time: pointer to a timespec which will contain the last modified time * * This is to force NFS clients to flush their caches for files with * exclusive leases. The justification is that if someone has an - * exclusive lease, then they could be modifiying it. + * exclusive lease, then they could be modifying it. */ void lease_get_mtime(struct inode *inode, struct timespec *time) { @@ -1805,17 +1801,21 @@ again: if (error) goto out; - for (;;) { - error = vfs_lock_file(filp, cmd, file_lock, NULL); - if (error != -EAGAIN || cmd == F_SETLK) - break; - error = wait_event_interruptible(file_lock->fl_wait, - !file_lock->fl_next); - if (!error) - continue; + if (filp->f_op && filp->f_op->lock != NULL) + error = filp->f_op->lock(filp, cmd, file_lock); + else { + for (;;) { + error = posix_lock_file(filp, file_lock, NULL); + if (error != -EAGAIN || cmd == F_SETLK) + break; + error = wait_event_interruptible(file_lock->fl_wait, + !file_lock->fl_next); + if (!error) + continue; - locks_delete_block(file_lock); - break; + locks_delete_block(file_lock); + break; + } } /* @@ -1929,17 +1929,21 @@ again: if (error) goto out; - for (;;) { - error = vfs_lock_file(filp, cmd, file_lock, NULL); - if (error != -EAGAIN || cmd == F_SETLK64) - break; - error = wait_event_interruptible(file_lock->fl_wait, - !file_lock->fl_next); - if (!error) - continue; + if (filp->f_op && filp->f_op->lock != NULL) + error = filp->f_op->lock(filp, cmd, file_lock); + else { + for (;;) { + error = posix_lock_file(filp, file_lock, NULL); + if (error != -EAGAIN || cmd == F_SETLK64) + break; + error = wait_event_interruptible(file_lock->fl_wait, + !file_lock->fl_next); + if (!error) + continue; - locks_delete_block(file_lock); - break; + locks_delete_block(file_lock); + break; + } } /* @@ -2084,6 +2088,12 @@ static void lock_get_status(struct seq_file *f, struct file_lock *fl, int id, char *pfx) { struct inode *inode = NULL; + unsigned int fl_pid; + + if (fl->fl_nspid) + fl_pid = pid_vnr(fl->fl_nspid); + else + fl_pid = fl->fl_pid; if (fl->fl_file != NULL) inode = fl->fl_file->f_path.dentry->d_inode; @@ -2124,16 +2134,16 @@ static void lock_get_status(struct seq_file *f, struct file_lock *fl, } if (inode) { #ifdef WE_CAN_BREAK_LSLK_NOW - seq_printf(f, "%d %s:%ld ", fl->fl_pid, + seq_printf(f, "%d %s:%ld ", fl_pid, inode->i_sb->s_id, inode->i_ino); #else /* userspace relies on this representation of dev_t ;-( */ - seq_printf(f, "%d %02x:%02x:%ld ", fl->fl_pid, + seq_printf(f, "%d %02x:%02x:%ld ", fl_pid, MAJOR(inode->i_sb->s_dev), MINOR(inode->i_sb->s_dev), inode->i_ino); #endif } else { - seq_printf(f, "%d <none>:0 ", fl->fl_pid); + seq_printf(f, "%d <none>:0 ", fl_pid); } if (IS_POSIX(fl)) { if (fl->fl_end == OFFSET_MAX) diff --git a/fs/mbcache.c b/fs/mbcache.c index eb31b73e7d6..ec88ff3d04a 100644 --- a/fs/mbcache.c +++ b/fs/mbcache.c @@ -399,11 +399,11 @@ mb_cache_destroy(struct mb_cache *cache) * if no more memory was available. */ struct mb_cache_entry * -mb_cache_entry_alloc(struct mb_cache *cache) +mb_cache_entry_alloc(struct mb_cache *cache, gfp_t gfp_flags) { struct mb_cache_entry *ce; - ce = kmem_cache_alloc(cache->c_entry_cache, GFP_KERNEL); + ce = kmem_cache_alloc(cache->c_entry_cache, gfp_flags); if (ce) { atomic_inc(&cache->c_entry_count); INIT_LIST_HEAD(&ce->e_lru_list); diff --git a/fs/minix/inode.c b/fs/minix/inode.c index bf4cd316af8..84f6242ba6f 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -18,7 +18,6 @@ #include <linux/highuid.h> #include <linux/vfs.h> -static void minix_read_inode(struct inode * inode); static int minix_write_inode(struct inode * inode, int wait); static int minix_statfs(struct dentry *dentry, struct kstatfs *buf); static int minix_remount (struct super_block * sb, int * flags, char * data); @@ -96,7 +95,6 @@ static void destroy_inodecache(void) static const struct super_operations minix_sops = { .alloc_inode = minix_alloc_inode, .destroy_inode = minix_destroy_inode, - .read_inode = minix_read_inode, .write_inode = minix_write_inode, .delete_inode = minix_delete_inode, .put_super = minix_put_super, @@ -149,6 +147,7 @@ static int minix_fill_super(struct super_block *s, void *data, int silent) unsigned long i, block; struct inode *root_inode; struct minix_sb_info *sbi; + int ret = -EINVAL; sbi = kzalloc(sizeof(struct minix_sb_info), GFP_KERNEL); if (!sbi) @@ -246,10 +245,13 @@ static int minix_fill_super(struct super_block *s, void *data, int silent) /* set up enough so that it can read an inode */ s->s_op = &minix_sops; - root_inode = iget(s, MINIX_ROOT_INO); - if (!root_inode || is_bad_inode(root_inode)) + root_inode = minix_iget(s, MINIX_ROOT_INO); + if (IS_ERR(root_inode)) { + ret = PTR_ERR(root_inode); goto out_no_root; + } + ret = -ENOMEM; s->s_root = d_alloc_root(root_inode); if (!s->s_root) goto out_iput; @@ -290,6 +292,7 @@ out_freemap: goto out_release; out_no_map: + ret = -ENOMEM; if (!silent) printk("MINIX-fs: can't allocate map\n"); goto out_release; @@ -316,7 +319,7 @@ out_bad_sb: out: s->s_fs_info = NULL; kfree(sbi); - return -EINVAL; + return ret; } static int minix_statfs(struct dentry *dentry, struct kstatfs *buf) @@ -409,7 +412,7 @@ void minix_set_inode(struct inode *inode, dev_t rdev) /* * The minix V1 function to read an inode. */ -static void V1_minix_read_inode(struct inode * inode) +static struct inode *V1_minix_iget(struct inode *inode) { struct buffer_head * bh; struct minix_inode * raw_inode; @@ -418,8 +421,8 @@ static void V1_minix_read_inode(struct inode * inode) raw_inode = minix_V1_raw_inode(inode->i_sb, inode->i_ino, &bh); if (!raw_inode) { - make_bad_inode(inode); - return; + iget_failed(inode); + return ERR_PTR(-EIO); } inode->i_mode = raw_inode->i_mode; inode->i_uid = (uid_t)raw_inode->i_uid; @@ -435,12 +438,14 @@ static void V1_minix_read_inode(struct inode * inode) minix_inode->u.i1_data[i] = raw_inode->i_zone[i]; minix_set_inode(inode, old_decode_dev(raw_inode->i_zone[0])); brelse(bh); + unlock_new_inode(inode); + return inode; } /* * The minix V2 function to read an inode. */ -static void V2_minix_read_inode(struct inode * inode) +static struct inode *V2_minix_iget(struct inode *inode) { struct buffer_head * bh; struct minix2_inode * raw_inode; @@ -449,8 +454,8 @@ static void V2_minix_read_inode(struct inode * inode) raw_inode = minix_V2_raw_inode(inode->i_sb, inode->i_ino, &bh); if (!raw_inode) { - make_bad_inode(inode); - return; + iget_failed(inode); + return ERR_PTR(-EIO); } inode->i_mode = raw_inode->i_mode; inode->i_uid = (uid_t)raw_inode->i_uid; @@ -468,17 +473,27 @@ static void V2_minix_read_inode(struct inode * inode) minix_inode->u.i2_data[i] = raw_inode->i_zone[i]; minix_set_inode(inode, old_decode_dev(raw_inode->i_zone[0])); brelse(bh); + unlock_new_inode(inode); + return inode; } /* * The global function to read an inode. */ -static void minix_read_inode(struct inode * inode) +struct inode *minix_iget(struct super_block *sb, unsigned long ino) { + struct inode *inode; + + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; + if (INODE_VERSION(inode) == MINIX_V1) - V1_minix_read_inode(inode); + return V1_minix_iget(inode); else - V2_minix_read_inode(inode); + return V2_minix_iget(inode); } /* diff --git a/fs/minix/minix.h b/fs/minix/minix.h index ac5d3a75cb0..326edfe9610 100644 --- a/fs/minix/minix.h +++ b/fs/minix/minix.h @@ -45,6 +45,7 @@ struct minix_sb_info { unsigned short s_version; }; +extern struct inode *minix_iget(struct super_block *, unsigned long); extern struct minix_inode * minix_V1_raw_inode(struct super_block *, ino_t, struct buffer_head **); extern struct minix2_inode * minix_V2_raw_inode(struct super_block *, ino_t, struct buffer_head **); extern struct inode * minix_new_inode(const struct inode * dir, int * error); diff --git a/fs/minix/namei.c b/fs/minix/namei.c index f4aa7a93904..102241bc9c7 100644 --- a/fs/minix/namei.c +++ b/fs/minix/namei.c @@ -54,10 +54,9 @@ static struct dentry *minix_lookup(struct inode * dir, struct dentry *dentry, st ino = minix_inode_by_name(dentry); if (ino) { - inode = iget(dir->i_sb, ino); - - if (!inode) - return ERR_PTR(-EACCES); + inode = minix_iget(dir->i_sb, ino); + if (IS_ERR(inode)) + return ERR_CAST(inode); } d_add(dentry, inode); return NULL; diff --git a/fs/mpage.c b/fs/mpage.c index d54f8f89722..235e4d3873a 100644 --- a/fs/mpage.c +++ b/fs/mpage.c @@ -276,9 +276,7 @@ do_mpage_readpage(struct bio *bio, struct page *page, unsigned nr_pages, } if (first_hole != blocks_per_page) { - zero_user_page(page, first_hole << blkbits, - PAGE_CACHE_SIZE - (first_hole << blkbits), - KM_USER0); + zero_user_segment(page, first_hole << blkbits, PAGE_CACHE_SIZE); if (first_hole == 0) { SetPageUptodate(page); unlock_page(page); @@ -327,16 +325,12 @@ confused: } /** - * mpage_readpages - populate an address space with some pages, and - * start reads against them. - * + * mpage_readpages - populate an address space with some pages & start reads against them * @mapping: the address_space * @pages: The address of a list_head which contains the target pages. These * pages have their ->index populated and are otherwise uninitialised. - * * The page at @pages->prev has the lowest file offset, and reads should be * issued in @pages->prev to @pages->next order. - * * @nr_pages: The number of pages at *@pages * @get_block: The filesystem's block mapper function. * @@ -362,6 +356,7 @@ confused: * So an mpage read of the first 16 blocks of an ext2 file will cause I/O to be * submitted in the following order: * 12 0 1 2 3 4 5 6 7 8 9 10 11 13 14 15 16 + * * because the indirect block has to be read to get the mappings of blocks * 13,14,15,16. Obviously, this impacts performance. * @@ -571,8 +566,7 @@ page_is_mapped: if (page->index > end_index || !offset) goto confused; - zero_user_page(page, offset, PAGE_CACHE_SIZE - offset, - KM_USER0); + zero_user_segment(page, offset, PAGE_CACHE_SIZE); } /* @@ -659,9 +653,7 @@ out: } /** - * mpage_writepages - walk the list of dirty pages of the given - * address space and writepage() all of them. - * + * mpage_writepages - walk the list of dirty pages of the given address space & writepage() all of them * @mapping: address space structure to write * @wbc: subtract the number of written pages from *@wbc->nr_to_write * @get_block: the filesystem's block mapper function. diff --git a/fs/namei.c b/fs/namei.c index 73e2e665817..8cf9bb9c2fc 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -106,7 +106,7 @@ * any extra contention... */ -static int fastcall link_path_walk(const char *name, struct nameidata *nd); +static int __link_path_walk(const char *name, struct nameidata *nd); /* In order to reduce some races, while at the same time doing additional * checking and hopefully speeding things up, we copy filenames to the @@ -231,7 +231,7 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) struct vfsmount *mnt = NULL; if (nd) - mnt = nd->mnt; + mnt = nd->path.mnt; if (mask & MAY_WRITE) { umode_t mode = inode->i_mode; @@ -296,7 +296,7 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) */ int vfs_permission(struct nameidata *nd, int mask) { - return permission(nd->dentry->d_inode, mask, nd); + return permission(nd->path.dentry->d_inode, mask, nd); } /** @@ -362,21 +362,31 @@ int deny_write_access(struct file * file) return 0; } -void path_release(struct nameidata *nd) +/** + * path_get - get a reference to a path + * @path: path to get the reference to + * + * Given a path increment the reference count to the dentry and the vfsmount. + */ +void path_get(struct path *path) { - dput(nd->dentry); - mntput(nd->mnt); + mntget(path->mnt); + dget(path->dentry); } +EXPORT_SYMBOL(path_get); -/* - * umount() mustn't call path_release()/mntput() as that would clear - * mnt_expiry_mark +/** + * path_put - put a reference to a path + * @path: path to put the reference to + * + * Given a path decrement the reference count to the dentry and the vfsmount. */ -void path_release_on_umount(struct nameidata *nd) +void path_put(struct path *path) { - dput(nd->dentry); - mntput_no_expire(nd->mnt); + dput(path->dentry); + mntput(path->mnt); } +EXPORT_SYMBOL(path_put); /** * release_open_intent - free up open intent resources @@ -539,20 +549,51 @@ walk_init_root(const char *name, struct nameidata *nd) struct fs_struct *fs = current->fs; read_lock(&fs->lock); - if (fs->altroot && !(nd->flags & LOOKUP_NOALT)) { - nd->mnt = mntget(fs->altrootmnt); - nd->dentry = dget(fs->altroot); + if (fs->altroot.dentry && !(nd->flags & LOOKUP_NOALT)) { + nd->path = fs->altroot; + path_get(&fs->altroot); read_unlock(&fs->lock); if (__emul_lookup_dentry(name,nd)) return 0; read_lock(&fs->lock); } - nd->mnt = mntget(fs->rootmnt); - nd->dentry = dget(fs->root); + nd->path = fs->root; + path_get(&fs->root); read_unlock(&fs->lock); return 1; } +/* + * Wrapper to retry pathname resolution whenever the underlying + * file system returns an ESTALE. + * + * Retry the whole path once, forcing real lookup requests + * instead of relying on the dcache. + */ +static __always_inline int link_path_walk(const char *name, struct nameidata *nd) +{ + struct path save = nd->path; + int result; + + /* make sure the stuff we saved doesn't go away */ + dget(save.dentry); + mntget(save.mnt); + + result = __link_path_walk(name, nd); + if (result == -ESTALE) { + /* nd->path had been dropped */ + nd->path = save; + dget(nd->path.dentry); + mntget(nd->path.mnt); + nd->flags |= LOOKUP_REVAL; + result = __link_path_walk(name, nd); + } + + path_put(&save); + + return result; +} + static __always_inline int __vfs_follow_link(struct nameidata *nd, const char *link) { int res = 0; @@ -561,7 +602,7 @@ static __always_inline int __vfs_follow_link(struct nameidata *nd, const char *l goto fail; if (*link == '/') { - path_release(nd); + path_put(&nd->path); if (!walk_init_root(link, nd)) /* weird __emul_prefix() stuff did it */ goto out; @@ -577,31 +618,31 @@ out: */ name = __getname(); if (unlikely(!name)) { - path_release(nd); + path_put(&nd->path); return -ENOMEM; } strcpy(name, nd->last.name); nd->last.name = name; return 0; fail: - path_release(nd); + path_put(&nd->path); return PTR_ERR(link); } -static inline void dput_path(struct path *path, struct nameidata *nd) +static void path_put_conditional(struct path *path, struct nameidata *nd) { dput(path->dentry); - if (path->mnt != nd->mnt) + if (path->mnt != nd->path.mnt) mntput(path->mnt); } static inline void path_to_nameidata(struct path *path, struct nameidata *nd) { - dput(nd->dentry); - if (nd->mnt != path->mnt) - mntput(nd->mnt); - nd->mnt = path->mnt; - nd->dentry = path->dentry; + dput(nd->path.dentry); + if (nd->path.mnt != path->mnt) + mntput(nd->path.mnt); + nd->path.mnt = path->mnt; + nd->path.dentry = path->dentry; } static __always_inline int __do_follow_link(struct path *path, struct nameidata *nd) @@ -613,7 +654,7 @@ static __always_inline int __do_follow_link(struct path *path, struct nameidata touch_atime(path->mnt, dentry); nd_set_link(nd, NULL); - if (path->mnt != nd->mnt) { + if (path->mnt != nd->path.mnt) { path_to_nameidata(path, nd); dget(dentry); } @@ -628,8 +669,7 @@ static __always_inline int __do_follow_link(struct path *path, struct nameidata if (dentry->d_inode->i_op->put_link) dentry->d_inode->i_op->put_link(dentry, nd, cookie); } - dput(dentry); - mntput(path->mnt); + path_put(path); return error; } @@ -661,8 +701,8 @@ static inline int do_follow_link(struct path *path, struct nameidata *nd) nd->depth--; return err; loop: - dput_path(path, nd); - path_release(nd); + path_put_conditional(path, nd); + path_put(&nd->path); return err; } @@ -743,37 +783,37 @@ static __always_inline void follow_dotdot(struct nameidata *nd) while(1) { struct vfsmount *parent; - struct dentry *old = nd->dentry; + struct dentry *old = nd->path.dentry; read_lock(&fs->lock); - if (nd->dentry == fs->root && - nd->mnt == fs->rootmnt) { + if (nd->path.dentry == fs->root.dentry && + nd->path.mnt == fs->root.mnt) { read_unlock(&fs->lock); break; } read_unlock(&fs->lock); spin_lock(&dcache_lock); - if (nd->dentry != nd->mnt->mnt_root) { - nd->dentry = dget(nd->dentry->d_parent); + if (nd->path.dentry != nd->path.mnt->mnt_root) { + nd->path.dentry = dget(nd->path.dentry->d_parent); spin_unlock(&dcache_lock); dput(old); break; } spin_unlock(&dcache_lock); spin_lock(&vfsmount_lock); - parent = nd->mnt->mnt_parent; - if (parent == nd->mnt) { + parent = nd->path.mnt->mnt_parent; + if (parent == nd->path.mnt) { spin_unlock(&vfsmount_lock); break; } mntget(parent); - nd->dentry = dget(nd->mnt->mnt_mountpoint); + nd->path.dentry = dget(nd->path.mnt->mnt_mountpoint); spin_unlock(&vfsmount_lock); dput(old); - mntput(nd->mnt); - nd->mnt = parent; + mntput(nd->path.mnt); + nd->path.mnt = parent; } - follow_mount(&nd->mnt, &nd->dentry); + follow_mount(&nd->path.mnt, &nd->path.dentry); } /* @@ -784,8 +824,8 @@ static __always_inline void follow_dotdot(struct nameidata *nd) static int do_lookup(struct nameidata *nd, struct qstr *name, struct path *path) { - struct vfsmount *mnt = nd->mnt; - struct dentry *dentry = __d_lookup(nd->dentry, name); + struct vfsmount *mnt = nd->path.mnt; + struct dentry *dentry = __d_lookup(nd->path.dentry, name); if (!dentry) goto need_lookup; @@ -798,7 +838,7 @@ done: return 0; need_lookup: - dentry = real_lookup(nd->dentry, name, nd); + dentry = real_lookup(nd->path.dentry, name, nd); if (IS_ERR(dentry)) goto fail; goto done; @@ -823,7 +863,7 @@ fail: * Returns 0 and nd will have valid dentry and mnt on success. * Returns error and drops reference to input namei data on failure. */ -static fastcall int __link_path_walk(const char * name, struct nameidata *nd) +static int __link_path_walk(const char *name, struct nameidata *nd) { struct path next; struct inode *inode; @@ -835,7 +875,7 @@ static fastcall int __link_path_walk(const char * name, struct nameidata *nd) if (!*name) goto return_reval; - inode = nd->dentry->d_inode; + inode = nd->path.dentry->d_inode; if (nd->depth) lookup_flags = LOOKUP_FOLLOW | (nd->flags & LOOKUP_CONTINUE); @@ -883,7 +923,7 @@ static fastcall int __link_path_walk(const char * name, struct nameidata *nd) if (this.name[1] != '.') break; follow_dotdot(nd); - inode = nd->dentry->d_inode; + inode = nd->path.dentry->d_inode; /* fallthrough */ case 1: continue; @@ -892,8 +932,9 @@ static fastcall int __link_path_walk(const char * name, struct nameidata *nd) * See if the low-level filesystem might want * to use its own hash.. */ - if (nd->dentry->d_op && nd->dentry->d_op->d_hash) { - err = nd->dentry->d_op->d_hash(nd->dentry, &this); + if (nd->path.dentry->d_op && nd->path.dentry->d_op->d_hash) { + err = nd->path.dentry->d_op->d_hash(nd->path.dentry, + &this); if (err < 0) break; } @@ -915,7 +956,7 @@ static fastcall int __link_path_walk(const char * name, struct nameidata *nd) if (err) goto return_err; err = -ENOENT; - inode = nd->dentry->d_inode; + inode = nd->path.dentry->d_inode; if (!inode) break; err = -ENOTDIR; @@ -943,13 +984,14 @@ last_component: if (this.name[1] != '.') break; follow_dotdot(nd); - inode = nd->dentry->d_inode; + inode = nd->path.dentry->d_inode; /* fallthrough */ case 1: goto return_reval; } - if (nd->dentry->d_op && nd->dentry->d_op->d_hash) { - err = nd->dentry->d_op->d_hash(nd->dentry, &this); + if (nd->path.dentry->d_op && nd->path.dentry->d_op->d_hash) { + err = nd->path.dentry->d_op->d_hash(nd->path.dentry, + &this); if (err < 0) break; } @@ -962,7 +1004,7 @@ last_component: err = do_follow_link(&next, nd); if (err) goto return_err; - inode = nd->dentry->d_inode; + inode = nd->path.dentry->d_inode; } else path_to_nameidata(&next, nd); err = -ENOENT; @@ -990,56 +1032,26 @@ return_reval: * We bypassed the ordinary revalidation routines. * We may need to check the cached dentry for staleness. */ - if (nd->dentry && nd->dentry->d_sb && - (nd->dentry->d_sb->s_type->fs_flags & FS_REVAL_DOT)) { + if (nd->path.dentry && nd->path.dentry->d_sb && + (nd->path.dentry->d_sb->s_type->fs_flags & FS_REVAL_DOT)) { err = -ESTALE; /* Note: we do not d_invalidate() */ - if (!nd->dentry->d_op->d_revalidate(nd->dentry, nd)) + if (!nd->path.dentry->d_op->d_revalidate( + nd->path.dentry, nd)) break; } return_base: return 0; out_dput: - dput_path(&next, nd); + path_put_conditional(&next, nd); break; } - path_release(nd); + path_put(&nd->path); return_err: return err; } -/* - * Wrapper to retry pathname resolution whenever the underlying - * file system returns an ESTALE. - * - * Retry the whole path once, forcing real lookup requests - * instead of relying on the dcache. - */ -static int fastcall link_path_walk(const char *name, struct nameidata *nd) -{ - struct nameidata save = *nd; - int result; - - /* make sure the stuff we saved doesn't go away */ - dget(save.dentry); - mntget(save.mnt); - - result = __link_path_walk(name, nd); - if (result == -ESTALE) { - *nd = save; - dget(nd->dentry); - mntget(nd->mnt); - nd->flags |= LOOKUP_REVAL; - result = __link_path_walk(name, nd); - } - - dput(save.dentry); - mntput(save.mnt); - - return result; -} - -static int fastcall path_walk(const char * name, struct nameidata *nd) +static int path_walk(const char *name, struct nameidata *nd) { current->total_link_count = 0; return link_path_walk(name, nd); @@ -1054,9 +1066,9 @@ static int __emul_lookup_dentry(const char *name, struct nameidata *nd) if (path_walk(name, nd)) return 0; /* something went wrong... */ - if (!nd->dentry->d_inode || S_ISDIR(nd->dentry->d_inode->i_mode)) { - struct dentry *old_dentry = nd->dentry; - struct vfsmount *old_mnt = nd->mnt; + if (!nd->path.dentry->d_inode || + S_ISDIR(nd->path.dentry->d_inode->i_mode)) { + struct path old_path = nd->path; struct qstr last = nd->last; int last_type = nd->last_type; struct fs_struct *fs = current->fs; @@ -1067,19 +1079,17 @@ static int __emul_lookup_dentry(const char *name, struct nameidata *nd) */ nd->last_type = LAST_ROOT; read_lock(&fs->lock); - nd->mnt = mntget(fs->rootmnt); - nd->dentry = dget(fs->root); + nd->path = fs->root; + path_get(&fs->root); read_unlock(&fs->lock); if (path_walk(name, nd) == 0) { - if (nd->dentry->d_inode) { - dput(old_dentry); - mntput(old_mnt); + if (nd->path.dentry->d_inode) { + path_put(&old_path); return 1; } - path_release(nd); + path_put(&nd->path); } - nd->dentry = old_dentry; - nd->mnt = old_mnt; + nd->path = old_path; nd->last = last; nd->last_type = last_type; } @@ -1090,33 +1100,26 @@ void set_fs_altroot(void) { char *emul = __emul_prefix(); struct nameidata nd; - struct vfsmount *mnt = NULL, *oldmnt; - struct dentry *dentry = NULL, *olddentry; + struct path path = {}, old_path; int err; struct fs_struct *fs = current->fs; if (!emul) goto set_it; err = path_lookup(emul, LOOKUP_FOLLOW|LOOKUP_DIRECTORY|LOOKUP_NOALT, &nd); - if (!err) { - mnt = nd.mnt; - dentry = nd.dentry; - } + if (!err) + path = nd.path; set_it: write_lock(&fs->lock); - oldmnt = fs->altrootmnt; - olddentry = fs->altroot; - fs->altrootmnt = mnt; - fs->altroot = dentry; + old_path = fs->altroot; + fs->altroot = path; write_unlock(&fs->lock); - if (olddentry) { - dput(olddentry); - mntput(oldmnt); - } + if (old_path.dentry) + path_put(&old_path); } /* Returns 0 and nd will be valid on success; Retuns error, otherwise. */ -static int fastcall do_path_lookup(int dfd, const char *name, +static int do_path_lookup(int dfd, const char *name, unsigned int flags, struct nameidata *nd) { int retval = 0; @@ -1130,21 +1133,21 @@ static int fastcall do_path_lookup(int dfd, const char *name, if (*name=='/') { read_lock(&fs->lock); - if (fs->altroot && !(nd->flags & LOOKUP_NOALT)) { - nd->mnt = mntget(fs->altrootmnt); - nd->dentry = dget(fs->altroot); + if (fs->altroot.dentry && !(nd->flags & LOOKUP_NOALT)) { + nd->path = fs->altroot; + path_get(&fs->altroot); read_unlock(&fs->lock); if (__emul_lookup_dentry(name,nd)) goto out; /* found in altroot */ read_lock(&fs->lock); } - nd->mnt = mntget(fs->rootmnt); - nd->dentry = dget(fs->root); + nd->path = fs->root; + path_get(&fs->root); read_unlock(&fs->lock); } else if (dfd == AT_FDCWD) { read_lock(&fs->lock); - nd->mnt = mntget(fs->pwdmnt); - nd->dentry = dget(fs->pwd); + nd->path = fs->pwd; + path_get(&fs->pwd); read_unlock(&fs->lock); } else { struct dentry *dentry; @@ -1164,17 +1167,17 @@ static int fastcall do_path_lookup(int dfd, const char *name, if (retval) goto fput_fail; - nd->mnt = mntget(file->f_path.mnt); - nd->dentry = dget(dentry); + nd->path = file->f_path; + path_get(&file->f_path); fput_light(file, fput_needed); } retval = path_walk(name, nd); out: - if (unlikely(!retval && !audit_dummy_context() && nd->dentry && - nd->dentry->d_inode)) - audit_inode(name, nd->dentry); + if (unlikely(!retval && !audit_dummy_context() && nd->path.dentry && + nd->path.dentry->d_inode)) + audit_inode(name, nd->path.dentry); out_fail: return retval; @@ -1183,7 +1186,7 @@ fput_fail: goto out_fail; } -int fastcall path_lookup(const char *name, unsigned int flags, +int path_lookup(const char *name, unsigned int flags, struct nameidata *nd) { return do_path_lookup(AT_FDCWD, name, flags, nd); @@ -1208,13 +1211,13 @@ int vfs_path_lookup(struct dentry *dentry, struct vfsmount *mnt, nd->flags = flags; nd->depth = 0; - nd->mnt = mntget(mnt); - nd->dentry = dget(dentry); + nd->path.mnt = mntget(mnt); + nd->path.dentry = dget(dentry); retval = path_walk(name, nd); - if (unlikely(!retval && !audit_dummy_context() && nd->dentry && - nd->dentry->d_inode)) - audit_inode(name, nd->dentry); + if (unlikely(!retval && !audit_dummy_context() && nd->path.dentry && + nd->path.dentry->d_inode)) + audit_inode(name, nd->path.dentry); return retval; @@ -1236,7 +1239,7 @@ static int __path_lookup_intent_open(int dfd, const char *name, if (IS_ERR(nd->intent.open.file)) { if (err == 0) { err = PTR_ERR(nd->intent.open.file); - path_release(nd); + path_put(&nd->path); } } else if (err != 0) release_open_intent(nd); @@ -1333,10 +1336,10 @@ static struct dentry *lookup_hash(struct nameidata *nd) { int err; - err = permission(nd->dentry->d_inode, MAY_EXEC, nd); + err = permission(nd->path.dentry->d_inode, MAY_EXEC, nd); if (err) return ERR_PTR(err); - return __lookup_hash(&nd->last, nd->dentry, nd); + return __lookup_hash(&nd->last, nd->path.dentry, nd); } static int __lookup_one_len(const char *name, struct qstr *this, @@ -1362,13 +1365,13 @@ static int __lookup_one_len(const char *name, struct qstr *this, } /** - * lookup_one_len: filesystem helper to lookup single pathname component + * lookup_one_len - filesystem helper to lookup single pathname component * @name: pathname component to lookup * @base: base directory to lookup from * @len: maximum length @len should be interpreted to * - * Note that this routine is purely a helper for filesystem useage and should - * not be called by generic code. Also note that by using this function to + * Note that this routine is purely a helper for filesystem usage and should + * not be called by generic code. Also note that by using this function the * nameidata argument is passed to the filesystem methods and a filesystem * using this helper needs to be prepared for that. */ @@ -1409,7 +1412,7 @@ struct dentry *lookup_one_noperm(const char *name, struct dentry *base) return __lookup_hash(&this, base, NULL); } -int fastcall __user_walk_fd(int dfd, const char __user *name, unsigned flags, +int __user_walk_fd(int dfd, const char __user *name, unsigned flags, struct nameidata *nd) { char *tmp = getname(name); @@ -1422,7 +1425,7 @@ int fastcall __user_walk_fd(int dfd, const char __user *name, unsigned flags, return err; } -int fastcall __user_walk(const char __user *name, unsigned flags, struct nameidata *nd) +int __user_walk(const char __user *name, unsigned flags, struct nameidata *nd) { return __user_walk_fd(AT_FDCWD, name, flags, nd); } @@ -1595,7 +1598,7 @@ int vfs_create(struct inode *dir, struct dentry *dentry, int mode, int may_open(struct nameidata *nd, int acc_mode, int flag) { - struct dentry *dentry = nd->dentry; + struct dentry *dentry = nd->path.dentry; struct inode *inode = dentry->d_inode; int error; @@ -1616,7 +1619,7 @@ int may_open(struct nameidata *nd, int acc_mode, int flag) if (S_ISFIFO(inode->i_mode) || S_ISSOCK(inode->i_mode)) { flag &= ~O_TRUNC; } else if (S_ISBLK(inode->i_mode) || S_ISCHR(inode->i_mode)) { - if (nd->mnt->mnt_flags & MNT_NODEV) + if (nd->path.mnt->mnt_flags & MNT_NODEV) return -EACCES; flag &= ~O_TRUNC; @@ -1678,14 +1681,14 @@ static int open_namei_create(struct nameidata *nd, struct path *path, int flag, int mode) { int error; - struct dentry *dir = nd->dentry; + struct dentry *dir = nd->path.dentry; if (!IS_POSIXACL(dir->d_inode)) mode &= ~current->fs->umask; error = vfs_create(dir->d_inode, path->dentry, mode, nd); mutex_unlock(&dir->d_inode->i_mutex); - dput(nd->dentry); - nd->dentry = path->dentry; + dput(nd->path.dentry); + nd->path.dentry = path->dentry; if (error) return error; /* Don't check for write permission, don't truncate */ @@ -1752,11 +1755,11 @@ int open_namei(int dfd, const char *pathname, int flag, if (nd->last_type != LAST_NORM || nd->last.name[nd->last.len]) goto exit; - dir = nd->dentry; + dir = nd->path.dentry; nd->flags &= ~LOOKUP_PARENT; mutex_lock(&dir->d_inode->i_mutex); path.dentry = lookup_hash(nd); - path.mnt = nd->mnt; + path.mnt = nd->path.mnt; do_last: error = PTR_ERR(path.dentry); @@ -1812,11 +1815,11 @@ ok: return 0; exit_dput: - dput_path(&path, nd); + path_put_conditional(&path, nd); exit: if (!IS_ERR(nd->intent.open.file)) release_open_intent(nd); - path_release(nd); + path_put(&nd->path); return error; do_link: @@ -1861,10 +1864,10 @@ do_link: __putname(nd->last.name); goto exit; } - dir = nd->dentry; + dir = nd->path.dentry; mutex_lock(&dir->d_inode->i_mutex); path.dentry = lookup_hash(nd); - path.mnt = nd->mnt; + path.mnt = nd->path.mnt; __putname(nd->last.name); goto do_last; } @@ -1877,13 +1880,13 @@ do_link: * Simple function to lookup and return a dentry and create it * if it doesn't exist. Is SMP-safe. * - * Returns with nd->dentry->d_inode->i_mutex locked. + * Returns with nd->path.dentry->d_inode->i_mutex locked. */ struct dentry *lookup_create(struct nameidata *nd, int is_dir) { struct dentry *dentry = ERR_PTR(-EEXIST); - mutex_lock_nested(&nd->dentry->d_inode->i_mutex, I_MUTEX_PARENT); + mutex_lock_nested(&nd->path.dentry->d_inode->i_mutex, I_MUTEX_PARENT); /* * Yucky last component or no last component at all? * (foo/., foo/.., /////) @@ -1962,19 +1965,19 @@ asmlinkage long sys_mknodat(int dfd, const char __user *filename, int mode, dentry = lookup_create(&nd, 0); error = PTR_ERR(dentry); - if (!IS_POSIXACL(nd.dentry->d_inode)) + if (!IS_POSIXACL(nd.path.dentry->d_inode)) mode &= ~current->fs->umask; if (!IS_ERR(dentry)) { switch (mode & S_IFMT) { case 0: case S_IFREG: - error = vfs_create(nd.dentry->d_inode,dentry,mode,&nd); + error = vfs_create(nd.path.dentry->d_inode,dentry,mode,&nd); break; case S_IFCHR: case S_IFBLK: - error = vfs_mknod(nd.dentry->d_inode,dentry,mode, + error = vfs_mknod(nd.path.dentry->d_inode,dentry,mode, new_decode_dev(dev)); break; case S_IFIFO: case S_IFSOCK: - error = vfs_mknod(nd.dentry->d_inode,dentry,mode,0); + error = vfs_mknod(nd.path.dentry->d_inode,dentry,mode,0); break; case S_IFDIR: error = -EPERM; @@ -1984,8 +1987,8 @@ asmlinkage long sys_mknodat(int dfd, const char __user *filename, int mode, } dput(dentry); } - mutex_unlock(&nd.dentry->d_inode->i_mutex); - path_release(&nd); + mutex_unlock(&nd.path.dentry->d_inode->i_mutex); + path_put(&nd.path); out: putname(tmp); @@ -2039,13 +2042,13 @@ asmlinkage long sys_mkdirat(int dfd, const char __user *pathname, int mode) if (IS_ERR(dentry)) goto out_unlock; - if (!IS_POSIXACL(nd.dentry->d_inode)) + if (!IS_POSIXACL(nd.path.dentry->d_inode)) mode &= ~current->fs->umask; - error = vfs_mkdir(nd.dentry->d_inode, dentry, mode); + error = vfs_mkdir(nd.path.dentry->d_inode, dentry, mode); dput(dentry); out_unlock: - mutex_unlock(&nd.dentry->d_inode->i_mutex); - path_release(&nd); + mutex_unlock(&nd.path.dentry->d_inode->i_mutex); + path_put(&nd.path); out: putname(tmp); out_err: @@ -2143,17 +2146,17 @@ static long do_rmdir(int dfd, const char __user *pathname) error = -EBUSY; goto exit1; } - mutex_lock_nested(&nd.dentry->d_inode->i_mutex, I_MUTEX_PARENT); + mutex_lock_nested(&nd.path.dentry->d_inode->i_mutex, I_MUTEX_PARENT); dentry = lookup_hash(&nd); error = PTR_ERR(dentry); if (IS_ERR(dentry)) goto exit2; - error = vfs_rmdir(nd.dentry->d_inode, dentry); + error = vfs_rmdir(nd.path.dentry->d_inode, dentry); dput(dentry); exit2: - mutex_unlock(&nd.dentry->d_inode->i_mutex); + mutex_unlock(&nd.path.dentry->d_inode->i_mutex); exit1: - path_release(&nd); + path_put(&nd.path); exit: putname(name); return error; @@ -2188,6 +2191,7 @@ int vfs_unlink(struct inode *dir, struct dentry *dentry) /* We don't d_delete() NFS sillyrenamed files--they still exist. */ if (!error && !(dentry->d_flags & DCACHE_NFSFS_RENAMED)) { + fsnotify_link_count(dentry->d_inode); d_delete(dentry); } @@ -2218,7 +2222,7 @@ static long do_unlinkat(int dfd, const char __user *pathname) error = -EISDIR; if (nd.last_type != LAST_NORM) goto exit1; - mutex_lock_nested(&nd.dentry->d_inode->i_mutex, I_MUTEX_PARENT); + mutex_lock_nested(&nd.path.dentry->d_inode->i_mutex, I_MUTEX_PARENT); dentry = lookup_hash(&nd); error = PTR_ERR(dentry); if (!IS_ERR(dentry)) { @@ -2228,15 +2232,15 @@ static long do_unlinkat(int dfd, const char __user *pathname) inode = dentry->d_inode; if (inode) atomic_inc(&inode->i_count); - error = vfs_unlink(nd.dentry->d_inode, dentry); + error = vfs_unlink(nd.path.dentry->d_inode, dentry); exit2: dput(dentry); } - mutex_unlock(&nd.dentry->d_inode->i_mutex); + mutex_unlock(&nd.path.dentry->d_inode->i_mutex); if (inode) iput(inode); /* truncate the inode here */ exit1: - path_release(&nd); + path_put(&nd.path); exit: putname(name); return error; @@ -2309,11 +2313,11 @@ asmlinkage long sys_symlinkat(const char __user *oldname, if (IS_ERR(dentry)) goto out_unlock; - error = vfs_symlink(nd.dentry->d_inode, dentry, from, S_IALLUGO); + error = vfs_symlink(nd.path.dentry->d_inode, dentry, from, S_IALLUGO); dput(dentry); out_unlock: - mutex_unlock(&nd.dentry->d_inode->i_mutex); - path_release(&nd); + mutex_unlock(&nd.path.dentry->d_inode->i_mutex); + path_put(&nd.path); out: putname(to); out_putname: @@ -2360,7 +2364,7 @@ int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_de error = dir->i_op->link(old_dentry, dir, new_dentry); mutex_unlock(&old_dentry->d_inode->i_mutex); if (!error) - fsnotify_create(dir, new_dentry); + fsnotify_link(dir, old_dentry->d_inode, new_dentry); return error; } @@ -2398,20 +2402,20 @@ asmlinkage long sys_linkat(int olddfd, const char __user *oldname, if (error) goto out; error = -EXDEV; - if (old_nd.mnt != nd.mnt) + if (old_nd.path.mnt != nd.path.mnt) goto out_release; new_dentry = lookup_create(&nd, 0); error = PTR_ERR(new_dentry); if (IS_ERR(new_dentry)) goto out_unlock; - error = vfs_link(old_nd.dentry, nd.dentry->d_inode, new_dentry); + error = vfs_link(old_nd.path.dentry, nd.path.dentry->d_inode, new_dentry); dput(new_dentry); out_unlock: - mutex_unlock(&nd.dentry->d_inode->i_mutex); + mutex_unlock(&nd.path.dentry->d_inode->i_mutex); out_release: - path_release(&nd); + path_put(&nd.path); out: - path_release(&old_nd); + path_put(&old_nd.path); exit: putname(to); @@ -2587,15 +2591,15 @@ static int do_rename(int olddfd, const char *oldname, goto exit1; error = -EXDEV; - if (oldnd.mnt != newnd.mnt) + if (oldnd.path.mnt != newnd.path.mnt) goto exit2; - old_dir = oldnd.dentry; + old_dir = oldnd.path.dentry; error = -EBUSY; if (oldnd.last_type != LAST_NORM) goto exit2; - new_dir = newnd.dentry; + new_dir = newnd.path.dentry; if (newnd.last_type != LAST_NORM) goto exit2; @@ -2639,9 +2643,9 @@ exit4: exit3: unlock_rename(new_dir, old_dir); exit2: - path_release(&newnd); + path_put(&newnd.path); exit1: - path_release(&oldnd); + path_put(&oldnd.path); exit: return error; } @@ -2815,7 +2819,6 @@ EXPORT_SYMBOL(page_symlink); EXPORT_SYMBOL(page_symlink_inode_operations); EXPORT_SYMBOL(path_lookup); EXPORT_SYMBOL(vfs_path_lookup); -EXPORT_SYMBOL(path_release); EXPORT_SYMBOL(permission); EXPORT_SYMBOL(vfs_permission); EXPORT_SYMBOL(file_permission); diff --git a/fs/namespace.c b/fs/namespace.c index 06083885b21..94f026ec990 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -25,31 +25,34 @@ #include <linux/security.h> #include <linux/mount.h> #include <linux/ramfs.h> +#include <linux/log2.h> #include <asm/uaccess.h> #include <asm/unistd.h> #include "pnode.h" #include "internal.h" +#define HASH_SHIFT ilog2(PAGE_SIZE / sizeof(struct list_head)) +#define HASH_SIZE (1UL << HASH_SHIFT) + /* spinlock for vfsmount related operations, inplace of dcache_lock */ __cacheline_aligned_in_smp DEFINE_SPINLOCK(vfsmount_lock); static int event; static struct list_head *mount_hashtable __read_mostly; -static int hash_mask __read_mostly, hash_bits __read_mostly; static struct kmem_cache *mnt_cache __read_mostly; static struct rw_semaphore namespace_sem; /* /sys/fs */ -decl_subsys(fs, NULL, NULL); -EXPORT_SYMBOL_GPL(fs_subsys); +struct kobject *fs_kobj; +EXPORT_SYMBOL_GPL(fs_kobj); static inline unsigned long hash(struct vfsmount *mnt, struct dentry *dentry) { unsigned long tmp = ((unsigned long)mnt / L1_CACHE_BYTES); tmp += ((unsigned long)dentry / L1_CACHE_BYTES); - tmp = tmp + (tmp >> hash_bits); - return tmp & hash_mask; + tmp = tmp + (tmp >> HASH_SHIFT); + return tmp & (HASH_SIZE - 1); } struct vfsmount *alloc_vfsmnt(const char *name) @@ -152,15 +155,15 @@ static void __touch_mnt_namespace(struct mnt_namespace *ns) } } -static void detach_mnt(struct vfsmount *mnt, struct nameidata *old_nd) +static void detach_mnt(struct vfsmount *mnt, struct path *old_path) { - old_nd->dentry = mnt->mnt_mountpoint; - old_nd->mnt = mnt->mnt_parent; + old_path->dentry = mnt->mnt_mountpoint; + old_path->mnt = mnt->mnt_parent; mnt->mnt_parent = mnt; mnt->mnt_mountpoint = mnt->mnt_root; list_del_init(&mnt->mnt_child); list_del_init(&mnt->mnt_hash); - old_nd->dentry->d_mounted--; + old_path->dentry->d_mounted--; } void mnt_set_mountpoint(struct vfsmount *mnt, struct dentry *dentry, @@ -171,12 +174,12 @@ void mnt_set_mountpoint(struct vfsmount *mnt, struct dentry *dentry, dentry->d_mounted++; } -static void attach_mnt(struct vfsmount *mnt, struct nameidata *nd) +static void attach_mnt(struct vfsmount *mnt, struct path *path) { - mnt_set_mountpoint(nd->mnt, nd->dentry, mnt); + mnt_set_mountpoint(path->mnt, path->dentry, mnt); list_add_tail(&mnt->mnt_hash, mount_hashtable + - hash(nd->mnt, nd->dentry)); - list_add_tail(&mnt->mnt_child, &nd->mnt->mnt_mounts); + hash(path->mnt, path->dentry)); + list_add_tail(&mnt->mnt_child, &path->mnt->mnt_mounts); } /* @@ -259,10 +262,8 @@ static struct vfsmount *clone_mnt(struct vfsmount *old, struct dentry *root, /* stick the duplicate mount on the same expiry list * as the original if that was on one */ if (flag & CL_EXPIRE) { - spin_lock(&vfsmount_lock); if (!list_empty(&old->mnt_expire)) list_add(&mnt->mnt_expire, &old->mnt_expire); - spin_unlock(&vfsmount_lock); } } return mnt; @@ -317,6 +318,50 @@ void mnt_unpin(struct vfsmount *mnt) EXPORT_SYMBOL(mnt_unpin); +static inline void mangle(struct seq_file *m, const char *s) +{ + seq_escape(m, s, " \t\n\\"); +} + +/* + * Simple .show_options callback for filesystems which don't want to + * implement more complex mount option showing. + * + * See also save_mount_options(). + */ +int generic_show_options(struct seq_file *m, struct vfsmount *mnt) +{ + const char *options = mnt->mnt_sb->s_options; + + if (options != NULL && options[0]) { + seq_putc(m, ','); + mangle(m, options); + } + + return 0; +} +EXPORT_SYMBOL(generic_show_options); + +/* + * If filesystem uses generic_show_options(), this function should be + * called from the fill_super() callback. + * + * The .remount_fs callback usually needs to be handled in a special + * way, to make sure, that previous options are not overwritten if the + * remount fails. + * + * Also note, that if the filesystem's .remount_fs function doesn't + * reset all options to their default value, but changes only newly + * given options, then the displayed options will not reflect reality + * any more. + */ +void save_mount_options(struct super_block *sb, char *options) +{ + kfree(sb->s_options); + sb->s_options = kstrdup(options, GFP_KERNEL); +} +EXPORT_SYMBOL(save_mount_options); + /* iterator */ static void *m_start(struct seq_file *m, loff_t *pos) { @@ -338,11 +383,6 @@ static void m_stop(struct seq_file *m, void *v) up_read(&namespace_sem); } -static inline void mangle(struct seq_file *m, const char *s) -{ - seq_escape(m, s, " \t\n\\"); -} - static int show_vfsmnt(struct seq_file *m, void *v) { struct vfsmount *mnt = list_entry(v, struct vfsmount, mnt_list); @@ -366,10 +406,11 @@ static int show_vfsmnt(struct seq_file *m, void *v) { 0, NULL } }; struct proc_fs_info *fs_infop; + struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt }; mangle(m, mnt->mnt_devname ? mnt->mnt_devname : "none"); seq_putc(m, ' '); - seq_path(m, mnt, mnt->mnt_root, " \t\n\\"); + seq_path(m, &mnt_path, " \t\n\\"); seq_putc(m, ' '); mangle(m, mnt->mnt_sb->s_type->name); if (mnt->mnt_sb->s_subtype && mnt->mnt_sb->s_subtype[0]) { @@ -401,6 +442,7 @@ struct seq_operations mounts_op = { static int show_vfsstat(struct seq_file *m, void *v) { struct vfsmount *mnt = list_entry(v, struct vfsmount, mnt_list); + struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt }; int err = 0; /* device */ @@ -412,7 +454,7 @@ static int show_vfsstat(struct seq_file *m, void *v) /* mount point */ seq_puts(m, " mounted on "); - seq_path(m, mnt, mnt->mnt_root, " \t\n\\"); + seq_path(m, &mnt_path, " \t\n\\"); seq_putc(m, ' '); /* file system type */ @@ -504,6 +546,7 @@ void release_mounts(struct list_head *head) m = mnt->mnt_parent; mnt->mnt_mountpoint = mnt->mnt_root; mnt->mnt_parent = mnt; + m->mnt_ghosts--; spin_unlock(&vfsmount_lock); dput(dentry); mntput(m); @@ -528,12 +571,16 @@ void umount_tree(struct vfsmount *mnt, int propagate, struct list_head *kill) __touch_mnt_namespace(p->mnt_ns); p->mnt_ns = NULL; list_del_init(&p->mnt_child); - if (p->mnt_parent != p) + if (p->mnt_parent != p) { + p->mnt_parent->mnt_ghosts++; p->mnt_mountpoint->d_mounted--; + } change_mnt_propagation(p, MS_PRIVATE); } } +static void shrink_submounts(struct vfsmount *mnt, struct list_head *umounts); + static int do_umount(struct vfsmount *mnt, int flags) { struct super_block *sb = mnt->mnt_sb; @@ -551,7 +598,7 @@ static int do_umount(struct vfsmount *mnt, int flags) * (2) the usage count == 1 [parent vfsmount] + 1 [sys_umount] */ if (flags & MNT_EXPIRE) { - if (mnt == current->fs->rootmnt || + if (mnt == current->fs->root.mnt || flags & (MNT_FORCE | MNT_DETACH)) return -EINVAL; @@ -586,7 +633,7 @@ static int do_umount(struct vfsmount *mnt, int flags) * /reboot - static binary that would close all descriptors and * call reboot(9). Then init(8) could umount root and exec /reboot. */ - if (mnt == current->fs->rootmnt && !(flags & MNT_DETACH)) { + if (mnt == current->fs->root.mnt && !(flags & MNT_DETACH)) { /* * Special case for "unmounting" root ... * we just try to remount it readonly. @@ -606,6 +653,9 @@ static int do_umount(struct vfsmount *mnt, int flags) spin_lock(&vfsmount_lock); event++; + if (!(flags & MNT_DETACH)) + shrink_submounts(mnt, &umount_list); + retval = -EBUSY; if (flags & MNT_DETACH || !propagate_mount_busy(mnt, 2)) { if (!list_empty(&mnt->mnt_list)) @@ -637,18 +687,20 @@ asmlinkage long sys_umount(char __user * name, int flags) if (retval) goto out; retval = -EINVAL; - if (nd.dentry != nd.mnt->mnt_root) + if (nd.path.dentry != nd.path.mnt->mnt_root) goto dput_and_out; - if (!check_mnt(nd.mnt)) + if (!check_mnt(nd.path.mnt)) goto dput_and_out; retval = -EPERM; if (!capable(CAP_SYS_ADMIN)) goto dput_and_out; - retval = do_umount(nd.mnt, flags); + retval = do_umount(nd.path.mnt, flags); dput_and_out: - path_release_on_umount(&nd); + /* we mustn't call path_put() as that would clear mnt_expiry_mark */ + dput(nd.path.dentry); + mntput_no_expire(nd.path.mnt); out: return retval; } @@ -671,10 +723,10 @@ static int mount_is_safe(struct nameidata *nd) return 0; return -EPERM; #ifdef notyet - if (S_ISLNK(nd->dentry->d_inode->i_mode)) + if (S_ISLNK(nd->path.dentry->d_inode->i_mode)) return -EPERM; - if (nd->dentry->d_inode->i_mode & S_ISVTX) { - if (current->uid != nd->dentry->d_inode->i_uid) + if (nd->path.dentry->d_inode->i_mode & S_ISVTX) { + if (current->uid != nd->path.dentry->d_inode->i_uid) return -EPERM; } if (vfs_permission(nd, MAY_WRITE)) @@ -698,7 +750,7 @@ struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry, int flag) { struct vfsmount *res, *p, *q, *r, *s; - struct nameidata nd; + struct path path; if (!(flag & CL_COPY_ALL) && IS_MNT_UNBINDABLE(mnt)) return NULL; @@ -723,14 +775,14 @@ struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry, q = q->mnt_parent; } p = s; - nd.mnt = q; - nd.dentry = p->mnt_mountpoint; + path.mnt = q; + path.dentry = p->mnt_mountpoint; q = clone_mnt(p, p->mnt_root, flag); if (!q) goto Enomem; spin_lock(&vfsmount_lock); list_add_tail(&q->mnt_list, &res->mnt_list); - attach_mnt(q, &nd); + attach_mnt(q, &path); spin_unlock(&vfsmount_lock); } } @@ -830,11 +882,11 @@ void drop_collected_mounts(struct vfsmount *mnt) * in allocations. */ static int attach_recursive_mnt(struct vfsmount *source_mnt, - struct nameidata *nd, struct nameidata *parent_nd) + struct path *path, struct path *parent_path) { LIST_HEAD(tree_list); - struct vfsmount *dest_mnt = nd->mnt; - struct dentry *dest_dentry = nd->dentry; + struct vfsmount *dest_mnt = path->mnt; + struct dentry *dest_dentry = path->dentry; struct vfsmount *child, *p; if (propagate_mnt(dest_mnt, dest_dentry, source_mnt, &tree_list)) @@ -846,9 +898,9 @@ static int attach_recursive_mnt(struct vfsmount *source_mnt, } spin_lock(&vfsmount_lock); - if (parent_nd) { - detach_mnt(source_mnt, parent_nd); - attach_mnt(source_mnt, nd); + if (parent_path) { + detach_mnt(source_mnt, parent_path); + attach_mnt(source_mnt, path); touch_mnt_namespace(current->nsproxy->mnt_ns); } else { mnt_set_mountpoint(dest_mnt, dest_dentry, source_mnt); @@ -869,13 +921,13 @@ static int graft_tree(struct vfsmount *mnt, struct nameidata *nd) if (mnt->mnt_sb->s_flags & MS_NOUSER) return -EINVAL; - if (S_ISDIR(nd->dentry->d_inode->i_mode) != + if (S_ISDIR(nd->path.dentry->d_inode->i_mode) != S_ISDIR(mnt->mnt_root->d_inode->i_mode)) return -ENOTDIR; err = -ENOENT; - mutex_lock(&nd->dentry->d_inode->i_mutex); - if (IS_DEADDIR(nd->dentry->d_inode)) + mutex_lock(&nd->path.dentry->d_inode->i_mutex); + if (IS_DEADDIR(nd->path.dentry->d_inode)) goto out_unlock; err = security_sb_check_sb(mnt, nd); @@ -883,10 +935,10 @@ static int graft_tree(struct vfsmount *mnt, struct nameidata *nd) goto out_unlock; err = -ENOENT; - if (IS_ROOT(nd->dentry) || !d_unhashed(nd->dentry)) - err = attach_recursive_mnt(mnt, nd, NULL); + if (IS_ROOT(nd->path.dentry) || !d_unhashed(nd->path.dentry)) + err = attach_recursive_mnt(mnt, &nd->path, NULL); out_unlock: - mutex_unlock(&nd->dentry->d_inode->i_mutex); + mutex_unlock(&nd->path.dentry->d_inode->i_mutex); if (!err) security_sb_post_addmount(mnt, nd); return err; @@ -894,17 +946,18 @@ out_unlock: /* * recursively change the type of the mountpoint. + * noinline this do_mount helper to save do_mount stack space. */ -static int do_change_type(struct nameidata *nd, int flag) +static noinline int do_change_type(struct nameidata *nd, int flag) { - struct vfsmount *m, *mnt = nd->mnt; + struct vfsmount *m, *mnt = nd->path.mnt; int recurse = flag & MS_REC; int type = flag & ~MS_REC; if (!capable(CAP_SYS_ADMIN)) return -EPERM; - if (nd->dentry != nd->mnt->mnt_root) + if (nd->path.dentry != nd->path.mnt->mnt_root) return -EINVAL; down_write(&namespace_sem); @@ -918,8 +971,10 @@ static int do_change_type(struct nameidata *nd, int flag) /* * do loopback mount. + * noinline this do_mount helper to save do_mount stack space. */ -static int do_loopback(struct nameidata *nd, char *old_name, int recurse) +static noinline int do_loopback(struct nameidata *nd, char *old_name, + int recurse) { struct nameidata old_nd; struct vfsmount *mnt = NULL; @@ -934,17 +989,17 @@ static int do_loopback(struct nameidata *nd, char *old_name, int recurse) down_write(&namespace_sem); err = -EINVAL; - if (IS_MNT_UNBINDABLE(old_nd.mnt)) - goto out; + if (IS_MNT_UNBINDABLE(old_nd.path.mnt)) + goto out; - if (!check_mnt(nd->mnt) || !check_mnt(old_nd.mnt)) + if (!check_mnt(nd->path.mnt) || !check_mnt(old_nd.path.mnt)) goto out; err = -ENOMEM; if (recurse) - mnt = copy_tree(old_nd.mnt, old_nd.dentry, 0); + mnt = copy_tree(old_nd.path.mnt, old_nd.path.dentry, 0); else - mnt = clone_mnt(old_nd.mnt, old_nd.dentry, 0); + mnt = clone_mnt(old_nd.path.mnt, old_nd.path.dentry, 0); if (!mnt) goto out; @@ -960,7 +1015,7 @@ static int do_loopback(struct nameidata *nd, char *old_name, int recurse) out: up_write(&namespace_sem); - path_release(&old_nd); + path_put(&old_nd.path); return err; } @@ -968,29 +1023,30 @@ out: * change filesystem flags. dir should be a physical root of filesystem. * If you've mounted a non-root directory somewhere and want to do remount * on it - tough luck. + * noinline this do_mount helper to save do_mount stack space. */ -static int do_remount(struct nameidata *nd, int flags, int mnt_flags, +static noinline int do_remount(struct nameidata *nd, int flags, int mnt_flags, void *data) { int err; - struct super_block *sb = nd->mnt->mnt_sb; + struct super_block *sb = nd->path.mnt->mnt_sb; if (!capable(CAP_SYS_ADMIN)) return -EPERM; - if (!check_mnt(nd->mnt)) + if (!check_mnt(nd->path.mnt)) return -EINVAL; - if (nd->dentry != nd->mnt->mnt_root) + if (nd->path.dentry != nd->path.mnt->mnt_root) return -EINVAL; down_write(&sb->s_umount); err = do_remount_sb(sb, flags, data, 0); if (!err) - nd->mnt->mnt_flags = mnt_flags; + nd->path.mnt->mnt_flags = mnt_flags; up_write(&sb->s_umount); if (!err) - security_sb_post_remount(nd->mnt, flags, data); + security_sb_post_remount(nd->path.mnt, flags, data); return err; } @@ -1004,9 +1060,13 @@ static inline int tree_contains_unbindable(struct vfsmount *mnt) return 0; } -static int do_move_mount(struct nameidata *nd, char *old_name) +/* + * noinline this do_mount helper to save do_mount stack space. + */ +static noinline int do_move_mount(struct nameidata *nd, char *old_name) { - struct nameidata old_nd, parent_nd; + struct nameidata old_nd; + struct path parent_path; struct vfsmount *p; int err = 0; if (!capable(CAP_SYS_ADMIN)) @@ -1018,69 +1078,72 @@ static int do_move_mount(struct nameidata *nd, char *old_name) return err; down_write(&namespace_sem); - while (d_mountpoint(nd->dentry) && follow_down(&nd->mnt, &nd->dentry)) + while (d_mountpoint(nd->path.dentry) && + follow_down(&nd->path.mnt, &nd->path.dentry)) ; err = -EINVAL; - if (!check_mnt(nd->mnt) || !check_mnt(old_nd.mnt)) + if (!check_mnt(nd->path.mnt) || !check_mnt(old_nd.path.mnt)) goto out; err = -ENOENT; - mutex_lock(&nd->dentry->d_inode->i_mutex); - if (IS_DEADDIR(nd->dentry->d_inode)) + mutex_lock(&nd->path.dentry->d_inode->i_mutex); + if (IS_DEADDIR(nd->path.dentry->d_inode)) goto out1; - if (!IS_ROOT(nd->dentry) && d_unhashed(nd->dentry)) + if (!IS_ROOT(nd->path.dentry) && d_unhashed(nd->path.dentry)) goto out1; err = -EINVAL; - if (old_nd.dentry != old_nd.mnt->mnt_root) + if (old_nd.path.dentry != old_nd.path.mnt->mnt_root) goto out1; - if (old_nd.mnt == old_nd.mnt->mnt_parent) + if (old_nd.path.mnt == old_nd.path.mnt->mnt_parent) goto out1; - if (S_ISDIR(nd->dentry->d_inode->i_mode) != - S_ISDIR(old_nd.dentry->d_inode->i_mode)) + if (S_ISDIR(nd->path.dentry->d_inode->i_mode) != + S_ISDIR(old_nd.path.dentry->d_inode->i_mode)) goto out1; /* * Don't move a mount residing in a shared parent. */ - if (old_nd.mnt->mnt_parent && IS_MNT_SHARED(old_nd.mnt->mnt_parent)) + if (old_nd.path.mnt->mnt_parent && + IS_MNT_SHARED(old_nd.path.mnt->mnt_parent)) goto out1; /* * Don't move a mount tree containing unbindable mounts to a destination * mount which is shared. */ - if (IS_MNT_SHARED(nd->mnt) && tree_contains_unbindable(old_nd.mnt)) + if (IS_MNT_SHARED(nd->path.mnt) && + tree_contains_unbindable(old_nd.path.mnt)) goto out1; err = -ELOOP; - for (p = nd->mnt; p->mnt_parent != p; p = p->mnt_parent) - if (p == old_nd.mnt) + for (p = nd->path.mnt; p->mnt_parent != p; p = p->mnt_parent) + if (p == old_nd.path.mnt) goto out1; - if ((err = attach_recursive_mnt(old_nd.mnt, nd, &parent_nd))) + err = attach_recursive_mnt(old_nd.path.mnt, &nd->path, &parent_path); + if (err) goto out1; - spin_lock(&vfsmount_lock); /* if the mount is moved, it should no longer be expire * automatically */ - list_del_init(&old_nd.mnt->mnt_expire); - spin_unlock(&vfsmount_lock); + list_del_init(&old_nd.path.mnt->mnt_expire); out1: - mutex_unlock(&nd->dentry->d_inode->i_mutex); + mutex_unlock(&nd->path.dentry->d_inode->i_mutex); out: up_write(&namespace_sem); if (!err) - path_release(&parent_nd); - path_release(&old_nd); + path_put(&parent_path); + path_put(&old_nd.path); return err; } /* * create a new mount for userspace and request it to be added into the * namespace's tree + * noinline this do_mount helper to save do_mount stack space. */ -static int do_new_mount(struct nameidata *nd, char *type, int flags, +static noinline int do_new_mount(struct nameidata *nd, char *type, int flags, int mnt_flags, char *name, void *data) { struct vfsmount *mnt; @@ -1110,16 +1173,17 @@ int do_add_mount(struct vfsmount *newmnt, struct nameidata *nd, down_write(&namespace_sem); /* Something was mounted here while we slept */ - while (d_mountpoint(nd->dentry) && follow_down(&nd->mnt, &nd->dentry)) + while (d_mountpoint(nd->path.dentry) && + follow_down(&nd->path.mnt, &nd->path.dentry)) ; err = -EINVAL; - if (!check_mnt(nd->mnt)) + if (!check_mnt(nd->path.mnt)) goto unlock; /* Refuse the same filesystem on the same mount point */ err = -EBUSY; - if (nd->mnt->mnt_sb == newmnt->mnt_sb && - nd->mnt->mnt_root == nd->dentry) + if (nd->path.mnt->mnt_sb == newmnt->mnt_sb && + nd->path.mnt->mnt_root == nd->path.dentry) goto unlock; err = -EINVAL; @@ -1130,12 +1194,9 @@ int do_add_mount(struct vfsmount *newmnt, struct nameidata *nd, if ((err = graft_tree(newmnt, nd))) goto unlock; - if (fslist) { - /* add to the specified expiration list */ - spin_lock(&vfsmount_lock); + if (fslist) /* add to the specified expiration list */ list_add_tail(&newmnt->mnt_expire, fslist); - spin_unlock(&vfsmount_lock); - } + up_write(&namespace_sem); return 0; @@ -1147,75 +1208,6 @@ unlock: EXPORT_SYMBOL_GPL(do_add_mount); -static void expire_mount(struct vfsmount *mnt, struct list_head *mounts, - struct list_head *umounts) -{ - spin_lock(&vfsmount_lock); - - /* - * Check if mount is still attached, if not, let whoever holds it deal - * with the sucker - */ - if (mnt->mnt_parent == mnt) { - spin_unlock(&vfsmount_lock); - return; - } - - /* - * Check that it is still dead: the count should now be 2 - as - * contributed by the vfsmount parent and the mntget above - */ - if (!propagate_mount_busy(mnt, 2)) { - /* delete from the namespace */ - touch_mnt_namespace(mnt->mnt_ns); - list_del_init(&mnt->mnt_list); - mnt->mnt_ns = NULL; - umount_tree(mnt, 1, umounts); - spin_unlock(&vfsmount_lock); - } else { - /* - * Someone brought it back to life whilst we didn't have any - * locks held so return it to the expiration list - */ - list_add_tail(&mnt->mnt_expire, mounts); - spin_unlock(&vfsmount_lock); - } -} - -/* - * go through the vfsmounts we've just consigned to the graveyard to - * - check that they're still dead - * - delete the vfsmount from the appropriate namespace under lock - * - dispose of the corpse - */ -static void expire_mount_list(struct list_head *graveyard, struct list_head *mounts) -{ - struct mnt_namespace *ns; - struct vfsmount *mnt; - - while (!list_empty(graveyard)) { - LIST_HEAD(umounts); - mnt = list_first_entry(graveyard, struct vfsmount, mnt_expire); - list_del_init(&mnt->mnt_expire); - - /* don't do anything if the namespace is dead - all the - * vfsmounts from it are going away anyway */ - ns = mnt->mnt_ns; - if (!ns || !ns->root) - continue; - get_mnt_ns(ns); - - spin_unlock(&vfsmount_lock); - down_write(&namespace_sem); - expire_mount(mnt, mounts, &umounts); - up_write(&namespace_sem); - release_mounts(&umounts); - mntput(mnt); - put_mnt_ns(ns); - spin_lock(&vfsmount_lock); - } -} - /* * process a list of expirable mountpoints with the intent of discarding any * mountpoints that aren't in use and haven't been touched since last we came @@ -1225,10 +1217,12 @@ void mark_mounts_for_expiry(struct list_head *mounts) { struct vfsmount *mnt, *next; LIST_HEAD(graveyard); + LIST_HEAD(umounts); if (list_empty(mounts)) return; + down_write(&namespace_sem); spin_lock(&vfsmount_lock); /* extract from the expiration list every vfsmount that matches the @@ -1239,16 +1233,19 @@ void mark_mounts_for_expiry(struct list_head *mounts) */ list_for_each_entry_safe(mnt, next, mounts, mnt_expire) { if (!xchg(&mnt->mnt_expiry_mark, 1) || - atomic_read(&mnt->mnt_count) != 1) + propagate_mount_busy(mnt, 1)) continue; - - mntget(mnt); list_move(&mnt->mnt_expire, &graveyard); } - - expire_mount_list(&graveyard, mounts); - + while (!list_empty(&graveyard)) { + mnt = list_first_entry(&graveyard, struct vfsmount, mnt_expire); + touch_mnt_namespace(mnt->mnt_ns); + umount_tree(mnt, 1, &umounts); + } spin_unlock(&vfsmount_lock); + up_write(&namespace_sem); + + release_mounts(&umounts); } EXPORT_SYMBOL_GPL(mark_mounts_for_expiry); @@ -1284,7 +1281,6 @@ resume: } if (!propagate_mount_busy(mnt, 1)) { - mntget(mnt); list_move_tail(&mnt->mnt_expire, graveyard); found++; } @@ -1304,22 +1300,22 @@ resume: * process a list of expirable mountpoints with the intent of discarding any * submounts of a specific parent mountpoint */ -void shrink_submounts(struct vfsmount *mountpoint, struct list_head *mounts) +static void shrink_submounts(struct vfsmount *mnt, struct list_head *umounts) { LIST_HEAD(graveyard); - int found; - - spin_lock(&vfsmount_lock); + struct vfsmount *m; /* extract submounts of 'mountpoint' from the expiration list */ - while ((found = select_submounts(mountpoint, &graveyard)) != 0) - expire_mount_list(&graveyard, mounts); - - spin_unlock(&vfsmount_lock); + while (select_submounts(mnt, &graveyard)) { + while (!list_empty(&graveyard)) { + m = list_first_entry(&graveyard, struct vfsmount, + mnt_expire); + touch_mnt_namespace(mnt->mnt_ns); + umount_tree(mnt, 1, umounts); + } + } } -EXPORT_SYMBOL_GPL(shrink_submounts); - /* * Some copy_from_user() implementations do not return the exact number of * bytes remaining to copy on a fault. But copy_mount_options() requires that. @@ -1455,7 +1451,7 @@ long do_mount(char *dev_name, char *dir_name, char *type_page, retval = do_new_mount(&nd, type_page, flags, mnt_flags, dev_name, data_page); dput_out: - path_release(&nd); + path_put(&nd.path); return retval; } @@ -1502,17 +1498,17 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, while (p) { q->mnt_ns = new_ns; if (fs) { - if (p == fs->rootmnt) { + if (p == fs->root.mnt) { rootmnt = p; - fs->rootmnt = mntget(q); + fs->root.mnt = mntget(q); } - if (p == fs->pwdmnt) { + if (p == fs->pwd.mnt) { pwdmnt = p; - fs->pwdmnt = mntget(q); + fs->pwd.mnt = mntget(q); } - if (p == fs->altrootmnt) { + if (p == fs->altroot.mnt) { altrootmnt = p; - fs->altrootmnt = mntget(q); + fs->altroot.mnt = mntget(q); } } p = next_mnt(p, mnt_ns->root); @@ -1593,47 +1589,38 @@ out1: * Replace the fs->{rootmnt,root} with {mnt,dentry}. Put the old values. * It can block. Requires the big lock held. */ -void set_fs_root(struct fs_struct *fs, struct vfsmount *mnt, - struct dentry *dentry) +void set_fs_root(struct fs_struct *fs, struct path *path) { - struct dentry *old_root; - struct vfsmount *old_rootmnt; + struct path old_root; + write_lock(&fs->lock); old_root = fs->root; - old_rootmnt = fs->rootmnt; - fs->rootmnt = mntget(mnt); - fs->root = dget(dentry); + fs->root = *path; + path_get(path); write_unlock(&fs->lock); - if (old_root) { - dput(old_root); - mntput(old_rootmnt); - } + if (old_root.dentry) + path_put(&old_root); } /* * Replace the fs->{pwdmnt,pwd} with {mnt,dentry}. Put the old values. * It can block. Requires the big lock held. */ -void set_fs_pwd(struct fs_struct *fs, struct vfsmount *mnt, - struct dentry *dentry) +void set_fs_pwd(struct fs_struct *fs, struct path *path) { - struct dentry *old_pwd; - struct vfsmount *old_pwdmnt; + struct path old_pwd; write_lock(&fs->lock); old_pwd = fs->pwd; - old_pwdmnt = fs->pwdmnt; - fs->pwdmnt = mntget(mnt); - fs->pwd = dget(dentry); + fs->pwd = *path; + path_get(path); write_unlock(&fs->lock); - if (old_pwd) { - dput(old_pwd); - mntput(old_pwdmnt); - } + if (old_pwd.dentry) + path_put(&old_pwd); } -static void chroot_fs_refs(struct nameidata *old_nd, struct nameidata *new_nd) +static void chroot_fs_refs(struct path *old_root, struct path *new_root) { struct task_struct *g, *p; struct fs_struct *fs; @@ -1645,12 +1632,12 @@ static void chroot_fs_refs(struct nameidata *old_nd, struct nameidata *new_nd) if (fs) { atomic_inc(&fs->count); task_unlock(p); - if (fs->root == old_nd->dentry - && fs->rootmnt == old_nd->mnt) - set_fs_root(fs, new_nd->mnt, new_nd->dentry); - if (fs->pwd == old_nd->dentry - && fs->pwdmnt == old_nd->mnt) - set_fs_pwd(fs, new_nd->mnt, new_nd->dentry); + if (fs->root.dentry == old_root->dentry + && fs->root.mnt == old_root->mnt) + set_fs_root(fs, new_root); + if (fs->pwd.dentry == old_root->dentry + && fs->pwd.mnt == old_root->mnt) + set_fs_pwd(fs, new_root); put_fs_struct(fs); } else task_unlock(p); @@ -1687,7 +1674,8 @@ asmlinkage long sys_pivot_root(const char __user * new_root, const char __user * put_old) { struct vfsmount *tmp; - struct nameidata new_nd, old_nd, parent_nd, root_parent, user_nd; + struct nameidata new_nd, old_nd, user_nd; + struct path parent_path, root_parent; int error; if (!capable(CAP_SYS_ADMIN)) @@ -1700,7 +1688,7 @@ asmlinkage long sys_pivot_root(const char __user * new_root, if (error) goto out0; error = -EINVAL; - if (!check_mnt(new_nd.mnt)) + if (!check_mnt(new_nd.path.mnt)) goto out1; error = __user_walk(put_old, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &old_nd); @@ -1709,74 +1697,78 @@ asmlinkage long sys_pivot_root(const char __user * new_root, error = security_sb_pivotroot(&old_nd, &new_nd); if (error) { - path_release(&old_nd); + path_put(&old_nd.path); goto out1; } read_lock(¤t->fs->lock); - user_nd.mnt = mntget(current->fs->rootmnt); - user_nd.dentry = dget(current->fs->root); + user_nd.path = current->fs->root; + path_get(¤t->fs->root); read_unlock(¤t->fs->lock); down_write(&namespace_sem); - mutex_lock(&old_nd.dentry->d_inode->i_mutex); + mutex_lock(&old_nd.path.dentry->d_inode->i_mutex); error = -EINVAL; - if (IS_MNT_SHARED(old_nd.mnt) || - IS_MNT_SHARED(new_nd.mnt->mnt_parent) || - IS_MNT_SHARED(user_nd.mnt->mnt_parent)) + if (IS_MNT_SHARED(old_nd.path.mnt) || + IS_MNT_SHARED(new_nd.path.mnt->mnt_parent) || + IS_MNT_SHARED(user_nd.path.mnt->mnt_parent)) goto out2; - if (!check_mnt(user_nd.mnt)) + if (!check_mnt(user_nd.path.mnt)) goto out2; error = -ENOENT; - if (IS_DEADDIR(new_nd.dentry->d_inode)) + if (IS_DEADDIR(new_nd.path.dentry->d_inode)) goto out2; - if (d_unhashed(new_nd.dentry) && !IS_ROOT(new_nd.dentry)) + if (d_unhashed(new_nd.path.dentry) && !IS_ROOT(new_nd.path.dentry)) goto out2; - if (d_unhashed(old_nd.dentry) && !IS_ROOT(old_nd.dentry)) + if (d_unhashed(old_nd.path.dentry) && !IS_ROOT(old_nd.path.dentry)) goto out2; error = -EBUSY; - if (new_nd.mnt == user_nd.mnt || old_nd.mnt == user_nd.mnt) + if (new_nd.path.mnt == user_nd.path.mnt || + old_nd.path.mnt == user_nd.path.mnt) goto out2; /* loop, on the same file system */ error = -EINVAL; - if (user_nd.mnt->mnt_root != user_nd.dentry) + if (user_nd.path.mnt->mnt_root != user_nd.path.dentry) goto out2; /* not a mountpoint */ - if (user_nd.mnt->mnt_parent == user_nd.mnt) + if (user_nd.path.mnt->mnt_parent == user_nd.path.mnt) goto out2; /* not attached */ - if (new_nd.mnt->mnt_root != new_nd.dentry) + if (new_nd.path.mnt->mnt_root != new_nd.path.dentry) goto out2; /* not a mountpoint */ - if (new_nd.mnt->mnt_parent == new_nd.mnt) + if (new_nd.path.mnt->mnt_parent == new_nd.path.mnt) goto out2; /* not attached */ - tmp = old_nd.mnt; /* make sure we can reach put_old from new_root */ + /* make sure we can reach put_old from new_root */ + tmp = old_nd.path.mnt; spin_lock(&vfsmount_lock); - if (tmp != new_nd.mnt) { + if (tmp != new_nd.path.mnt) { for (;;) { if (tmp->mnt_parent == tmp) goto out3; /* already mounted on put_old */ - if (tmp->mnt_parent == new_nd.mnt) + if (tmp->mnt_parent == new_nd.path.mnt) break; tmp = tmp->mnt_parent; } - if (!is_subdir(tmp->mnt_mountpoint, new_nd.dentry)) + if (!is_subdir(tmp->mnt_mountpoint, new_nd.path.dentry)) goto out3; - } else if (!is_subdir(old_nd.dentry, new_nd.dentry)) + } else if (!is_subdir(old_nd.path.dentry, new_nd.path.dentry)) goto out3; - detach_mnt(new_nd.mnt, &parent_nd); - detach_mnt(user_nd.mnt, &root_parent); - attach_mnt(user_nd.mnt, &old_nd); /* mount old root on put_old */ - attach_mnt(new_nd.mnt, &root_parent); /* mount new_root on / */ + detach_mnt(new_nd.path.mnt, &parent_path); + detach_mnt(user_nd.path.mnt, &root_parent); + /* mount old root on put_old */ + attach_mnt(user_nd.path.mnt, &old_nd.path); + /* mount new_root on / */ + attach_mnt(new_nd.path.mnt, &root_parent); touch_mnt_namespace(current->nsproxy->mnt_ns); spin_unlock(&vfsmount_lock); - chroot_fs_refs(&user_nd, &new_nd); + chroot_fs_refs(&user_nd.path, &new_nd.path); security_sb_post_pivotroot(&user_nd, &new_nd); error = 0; - path_release(&root_parent); - path_release(&parent_nd); + path_put(&root_parent); + path_put(&parent_path); out2: - mutex_unlock(&old_nd.dentry->d_inode->i_mutex); + mutex_unlock(&old_nd.path.dentry->d_inode->i_mutex); up_write(&namespace_sem); - path_release(&user_nd); - path_release(&old_nd); + path_put(&user_nd.path); + path_put(&old_nd.path); out1: - path_release(&new_nd); + path_put(&new_nd.path); out0: unlock_kernel(); return error; @@ -1789,6 +1781,7 @@ static void __init init_mount_tree(void) { struct vfsmount *mnt; struct mnt_namespace *ns; + struct path root; mnt = do_kern_mount("rootfs", 0, "rootfs", NULL); if (IS_ERR(mnt)) @@ -1807,15 +1800,16 @@ static void __init init_mount_tree(void) init_task.nsproxy->mnt_ns = ns; get_mnt_ns(ns); - set_fs_pwd(current->fs, ns->root, ns->root->mnt_root); - set_fs_root(current->fs, ns->root, ns->root->mnt_root); + root.mnt = ns->root; + root.dentry = ns->root->mnt_root; + + set_fs_pwd(current->fs, &root); + set_fs_root(current->fs, &root); } void __init mnt_init(void) { - struct list_head *d; - unsigned int nr_hash; - int i; + unsigned u; int err; init_rwsem(&namespace_sem); @@ -1828,43 +1822,18 @@ void __init mnt_init(void) if (!mount_hashtable) panic("Failed to allocate mount hash table\n"); - /* - * Find the power-of-two list-heads that can fit into the allocation.. - * We don't guarantee that "sizeof(struct list_head)" is necessarily - * a power-of-two. - */ - nr_hash = PAGE_SIZE / sizeof(struct list_head); - hash_bits = 0; - do { - hash_bits++; - } while ((nr_hash >> hash_bits) != 0); - hash_bits--; + printk("Mount-cache hash table entries: %lu\n", HASH_SIZE); + + for (u = 0; u < HASH_SIZE; u++) + INIT_LIST_HEAD(&mount_hashtable[u]); - /* - * Re-calculate the actual number of entries and the mask - * from the number of bits we can fit. - */ - nr_hash = 1UL << hash_bits; - hash_mask = nr_hash - 1; - - printk("Mount-cache hash table entries: %d\n", nr_hash); - - /* And initialize the newly allocated array */ - d = mount_hashtable; - i = nr_hash; - do { - INIT_LIST_HEAD(d); - d++; - i--; - } while (i); err = sysfs_init(); if (err) printk(KERN_WARNING "%s: sysfs_init error: %d\n", __FUNCTION__, err); - err = subsystem_register(&fs_subsys); - if (err) - printk(KERN_WARNING "%s: subsystem_register error: %d\n", - __FUNCTION__, err); + fs_kobj = kobject_create_and_add("fs", NULL); + if (!fs_kobj) + printk(KERN_WARNING "%s: kobj create error\n", __FUNCTION__); init_rootfs(); init_mount_tree(); } diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index e1cb70c643f..fbbb9f7afa1 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -28,6 +28,8 @@ #include <linux/init.h> #include <linux/smp_lock.h> #include <linux/vfs.h> +#include <linux/mount.h> +#include <linux/seq_file.h> #include <linux/ncp_fs.h> @@ -36,9 +38,15 @@ #include "ncplib_kernel.h" #include "getopt.h" +#define NCP_DEFAULT_FILE_MODE 0600 +#define NCP_DEFAULT_DIR_MODE 0700 +#define NCP_DEFAULT_TIME_OUT 10 +#define NCP_DEFAULT_RETRY_COUNT 20 + static void ncp_delete_inode(struct inode *); static void ncp_put_super(struct super_block *); static int ncp_statfs(struct dentry *, struct kstatfs *); +static int ncp_show_options(struct seq_file *, struct vfsmount *); static struct kmem_cache * ncp_inode_cachep; @@ -96,6 +104,7 @@ static const struct super_operations ncp_sops = .put_super = ncp_put_super, .statfs = ncp_statfs, .remount_fs = ncp_remount, + .show_options = ncp_show_options, }; extern struct dentry_operations ncp_root_dentry_operations; @@ -304,6 +313,37 @@ static void ncp_stop_tasks(struct ncp_server *server) { flush_scheduled_work(); } +static int ncp_show_options(struct seq_file *seq, struct vfsmount *mnt) +{ + struct ncp_server *server = NCP_SBP(mnt->mnt_sb); + unsigned int tmp; + + if (server->m.uid != 0) + seq_printf(seq, ",uid=%u", server->m.uid); + if (server->m.gid != 0) + seq_printf(seq, ",gid=%u", server->m.gid); + if (server->m.mounted_uid != 0) + seq_printf(seq, ",owner=%u", server->m.mounted_uid); + tmp = server->m.file_mode & S_IALLUGO; + if (tmp != NCP_DEFAULT_FILE_MODE) + seq_printf(seq, ",mode=0%o", tmp); + tmp = server->m.dir_mode & S_IALLUGO; + if (tmp != NCP_DEFAULT_DIR_MODE) + seq_printf(seq, ",dirmode=0%o", tmp); + if (server->m.time_out != NCP_DEFAULT_TIME_OUT * HZ / 100) { + tmp = server->m.time_out * 100 / HZ; + seq_printf(seq, ",timeout=%u", tmp); + } + if (server->m.retry_count != NCP_DEFAULT_RETRY_COUNT) + seq_printf(seq, ",retry=%u", server->m.retry_count); + if (server->m.flags != 0) + seq_printf(seq, ",flags=%lu", server->m.flags); + if (server->m.wdog_pid != NULL) + seq_printf(seq, ",wdogpid=%u", pid_vnr(server->m.wdog_pid)); + + return 0; +} + static const struct ncp_option ncp_opts[] = { { "uid", OPT_INT, 'u' }, { "gid", OPT_INT, 'g' }, @@ -331,12 +371,12 @@ static int ncp_parse_options(struct ncp_mount_data_kernel *data, char *options) data->mounted_uid = 0; data->wdog_pid = NULL; data->ncp_fd = ~0; - data->time_out = 10; - data->retry_count = 20; + data->time_out = NCP_DEFAULT_TIME_OUT; + data->retry_count = NCP_DEFAULT_RETRY_COUNT; data->uid = 0; data->gid = 0; - data->file_mode = 0600; - data->dir_mode = 0700; + data->file_mode = NCP_DEFAULT_FILE_MODE; + data->dir_mode = NCP_DEFAULT_DIR_MODE; data->info_fd = -1; data->mounted_vol[0] = 0; @@ -982,12 +1022,13 @@ static struct file_system_type ncp_fs_type = { .name = "ncpfs", .get_sb = ncp_get_sb, .kill_sb = kill_anon_super, + .fs_flags = FS_BINARY_MOUNTDATA, }; static int __init init_ncp_fs(void) { int err; - DPRINTK("ncpfs: init_module called\n"); + DPRINTK("ncpfs: init_ncp_fs called\n"); err = init_inodecache(); if (err) @@ -1004,7 +1045,7 @@ out1: static void __exit exit_ncp_fs(void) { - DPRINTK("ncpfs: cleanup_module called\n"); + DPRINTK("ncpfs: exit_ncp_fs called\n"); unregister_filesystem(&ncp_fs_type); destroy_inodecache(); } diff --git a/fs/ncpfs/mmap.c b/fs/ncpfs/mmap.c index a94473d3072..5d8dcb9ee32 100644 --- a/fs/ncpfs/mmap.c +++ b/fs/ncpfs/mmap.c @@ -50,10 +50,6 @@ static int ncp_file_mmap_fault(struct vm_area_struct *area, pos = vmf->pgoff << PAGE_SHIFT; count = PAGE_SIZE; - if ((unsigned long)vmf->virtual_address + PAGE_SIZE > area->vm_end) { - WARN_ON(1); /* shouldn't happen? */ - count = area->vm_end - (unsigned long)vmf->virtual_address; - } /* what we can read in one go */ bufsize = NCP_SERVER(inode)->buffer_size; diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c index a796be5051b..66648dd92d9 100644 --- a/fs/nfs/callback.c +++ b/fs/nfs/callback.c @@ -73,8 +73,6 @@ static void nfs_callback_svc(struct svc_rqst *rqstp) complete(&nfs_callback_info.started); for(;;) { - char buf[RPC_MAX_ADDRBUFLEN]; - if (signalled()) { if (nfs_callback_info.users == 0) break; @@ -92,11 +90,10 @@ static void nfs_callback_svc(struct svc_rqst *rqstp) __FUNCTION__, -err); break; } - dprintk("%s: request from %s\n", __FUNCTION__, - svc_print_addr(rqstp, buf, sizeof(buf))); svc_process(rqstp); } + flush_signals(current); svc_exit_thread(rqstp); nfs_callback_info.pid = 0; complete(&nfs_callback_info.stopped); @@ -109,7 +106,7 @@ static void nfs_callback_svc(struct svc_rqst *rqstp) */ int nfs_callback_up(void) { - struct svc_serv *serv; + struct svc_serv *serv = NULL; int ret = 0; lock_kernel(); @@ -123,27 +120,33 @@ int nfs_callback_up(void) if (!serv) goto out_err; - ret = svc_makesock(serv, IPPROTO_TCP, nfs_callback_set_tcpport, - SVC_SOCK_ANONYMOUS); + ret = svc_create_xprt(serv, "tcp", nfs_callback_set_tcpport, + SVC_SOCK_ANONYMOUS); if (ret <= 0) - goto out_destroy; + goto out_err; nfs_callback_tcpport = ret; dprintk("Callback port = 0x%x\n", nfs_callback_tcpport); ret = svc_create_thread(nfs_callback_svc, serv); if (ret < 0) - goto out_destroy; + goto out_err; nfs_callback_info.serv = serv; wait_for_completion(&nfs_callback_info.started); out: + /* + * svc_create creates the svc_serv with sv_nrthreads == 1, and then + * svc_create_thread increments that. So we need to call svc_destroy + * on both success and failure so that the refcount is 1 when the + * thread exits. + */ + if (serv) + svc_destroy(serv); mutex_unlock(&nfs_callback_mutex); unlock_kernel(); return ret; -out_destroy: +out_err: dprintk("Couldn't create callback socket or server thread; err = %d\n", ret); - svc_destroy(serv); -out_err: nfs_callback_info.users--; goto out; } @@ -168,12 +171,11 @@ void nfs_callback_down(void) static int nfs_callback_authenticate(struct svc_rqst *rqstp) { - struct sockaddr_in *addr = svc_addr_in(rqstp); struct nfs_client *clp; - char buf[RPC_MAX_ADDRBUFLEN]; + RPC_IFDEBUG(char buf[RPC_MAX_ADDRBUFLEN]); /* Don't talk to strangers */ - clp = nfs_find_client(addr, 4); + clp = nfs_find_client(svc_addr(rqstp), 4); if (clp == NULL) return SVC_DROP; diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h index c2bb14e053e..bb25d2135ff 100644 --- a/fs/nfs/callback.h +++ b/fs/nfs/callback.h @@ -38,7 +38,7 @@ struct cb_compound_hdr_res { }; struct cb_getattrargs { - struct sockaddr_in *addr; + struct sockaddr *addr; struct nfs_fh fh; uint32_t bitmap[2]; }; @@ -53,7 +53,7 @@ struct cb_getattrres { }; struct cb_recallargs { - struct sockaddr_in *addr; + struct sockaddr *addr; struct nfs_fh fh; nfs4_stateid stateid; uint32_t truncate; diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index 72e55d83756..15f7785048d 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -12,7 +12,9 @@ #include "delegation.h" #include "internal.h" +#ifdef NFS_DEBUG #define NFSDBG_FACILITY NFSDBG_CALLBACK +#endif __be32 nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res) { @@ -20,12 +22,16 @@ __be32 nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres * struct nfs_delegation *delegation; struct nfs_inode *nfsi; struct inode *inode; - + res->bitmap[0] = res->bitmap[1] = 0; res->status = htonl(NFS4ERR_BADHANDLE); clp = nfs_find_client(args->addr, 4); if (clp == NULL) goto out; + + dprintk("NFS: GETATTR callback request from %s\n", + rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_ADDR)); + inode = nfs_delegation_find_inode(clp, &args->fh); if (inode == NULL) goto out_putclient; @@ -65,23 +71,32 @@ __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy) clp = nfs_find_client(args->addr, 4); if (clp == NULL) goto out; - inode = nfs_delegation_find_inode(clp, &args->fh); - if (inode == NULL) - goto out_putclient; - /* Set up a helper thread to actually return the delegation */ - switch(nfs_async_inode_return_delegation(inode, &args->stateid)) { - case 0: - res = 0; - break; - case -ENOENT: - res = htonl(NFS4ERR_BAD_STATEID); - break; - default: - res = htonl(NFS4ERR_RESOURCE); - } - iput(inode); -out_putclient: - nfs_put_client(clp); + + dprintk("NFS: RECALL callback request from %s\n", + rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_ADDR)); + + do { + struct nfs_client *prev = clp; + + inode = nfs_delegation_find_inode(clp, &args->fh); + if (inode != NULL) { + /* Set up a helper thread to actually return the delegation */ + switch(nfs_async_inode_return_delegation(inode, &args->stateid)) { + case 0: + res = 0; + break; + case -ENOENT: + if (res != 0) + res = htonl(NFS4ERR_BAD_STATEID); + break; + default: + res = htonl(NFS4ERR_RESOURCE); + } + iput(inode); + } + clp = nfs_find_client_next(prev); + nfs_put_client(prev); + } while (clp != NULL); out: dprintk("%s: exit with status = %d\n", __FUNCTION__, ntohl(res)); return res; diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index 058ade7efe7..13619d24f02 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -139,7 +139,7 @@ static __be32 decode_compound_hdr_arg(struct xdr_stream *xdr, struct cb_compound if (unlikely(status != 0)) return status; /* We do not like overly long tags! */ - if (hdr->taglen > CB_OP_TAGLEN_MAXSZ-12 || hdr->taglen < 0) { + if (hdr->taglen > CB_OP_TAGLEN_MAXSZ - 12) { printk("NFSv4 CALLBACK %s: client sent tag of length %u\n", __FUNCTION__, hdr->taglen); return htonl(NFS4ERR_RESOURCE); @@ -176,7 +176,7 @@ static __be32 decode_getattr_args(struct svc_rqst *rqstp, struct xdr_stream *xdr status = decode_fh(xdr, &args->fh); if (unlikely(status != 0)) goto out; - args->addr = svc_addr_in(rqstp); + args->addr = svc_addr(rqstp); status = decode_bitmap(xdr, args->bitmap); out: dprintk("%s: exit with status = %d\n", __FUNCTION__, ntohl(status)); @@ -188,7 +188,7 @@ static __be32 decode_recall_args(struct svc_rqst *rqstp, struct xdr_stream *xdr, __be32 *p; __be32 status; - args->addr = svc_addr_in(rqstp); + args->addr = svc_addr(rqstp); status = decode_stateid(xdr, &args->stateid); if (unlikely(status != 0)) goto out; @@ -254,7 +254,7 @@ static __be32 encode_attr_change(struct xdr_stream *xdr, const uint32_t *bitmap, if (!(bitmap[0] & FATTR4_WORD0_CHANGE)) return 0; p = xdr_reserve_space(xdr, 8); - if (unlikely(p == 0)) + if (unlikely(!p)) return htonl(NFS4ERR_RESOURCE); p = xdr_encode_hyper(p, change); return 0; @@ -267,7 +267,7 @@ static __be32 encode_attr_size(struct xdr_stream *xdr, const uint32_t *bitmap, u if (!(bitmap[0] & FATTR4_WORD0_SIZE)) return 0; p = xdr_reserve_space(xdr, 8); - if (unlikely(p == 0)) + if (unlikely(!p)) return htonl(NFS4ERR_RESOURCE); p = xdr_encode_hyper(p, size); return 0; @@ -278,7 +278,7 @@ static __be32 encode_attr_time(struct xdr_stream *xdr, const struct timespec *ti __be32 *p; p = xdr_reserve_space(xdr, 12); - if (unlikely(p == 0)) + if (unlikely(!p)) return htonl(NFS4ERR_RESOURCE); p = xdr_encode_hyper(p, time->tv_sec); *p = htonl(time->tv_nsec); diff --git a/fs/nfs/client.c b/fs/nfs/client.c index a6f62549761..c5c0175898f 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -34,6 +34,8 @@ #include <linux/nfs_idmap.h> #include <linux/vfs.h> #include <linux/inet.h> +#include <linux/in6.h> +#include <net/ipv6.h> #include <linux/nfs_xdr.h> #include <asm/system.h> @@ -93,22 +95,30 @@ struct rpc_program nfsacl_program = { }; #endif /* CONFIG_NFS_V3_ACL */ +struct nfs_client_initdata { + const char *hostname; + const struct sockaddr *addr; + size_t addrlen; + const struct nfs_rpc_ops *rpc_ops; + int proto; +}; + /* * Allocate a shared client record * * Since these are allocated/deallocated very rarely, we don't * bother putting them in a slab cache... */ -static struct nfs_client *nfs_alloc_client(const char *hostname, - const struct sockaddr_in *addr, - int nfsversion) +static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_init) { struct nfs_client *clp; if ((clp = kzalloc(sizeof(*clp), GFP_KERNEL)) == NULL) goto error_0; - if (nfsversion == 4) { + clp->rpc_ops = cl_init->rpc_ops; + + if (cl_init->rpc_ops->version == 4) { if (nfs_callback_up() < 0) goto error_2; __set_bit(NFS_CS_CALLBACK, &clp->cl_res_state); @@ -117,11 +127,11 @@ static struct nfs_client *nfs_alloc_client(const char *hostname, atomic_set(&clp->cl_count, 1); clp->cl_cons_state = NFS_CS_INITING; - clp->cl_nfsversion = nfsversion; - memcpy(&clp->cl_addr, addr, sizeof(clp->cl_addr)); + memcpy(&clp->cl_addr, cl_init->addr, cl_init->addrlen); + clp->cl_addrlen = cl_init->addrlen; - if (hostname) { - clp->cl_hostname = kstrdup(hostname, GFP_KERNEL); + if (cl_init->hostname) { + clp->cl_hostname = kstrdup(cl_init->hostname, GFP_KERNEL); if (!clp->cl_hostname) goto error_3; } @@ -129,6 +139,8 @@ static struct nfs_client *nfs_alloc_client(const char *hostname, INIT_LIST_HEAD(&clp->cl_superblocks); clp->cl_rpcclient = ERR_PTR(-EINVAL); + clp->cl_proto = cl_init->proto; + #ifdef CONFIG_NFS_V4 init_rwsem(&clp->cl_sem); INIT_LIST_HEAD(&clp->cl_delegations); @@ -166,7 +178,7 @@ static void nfs4_shutdown_client(struct nfs_client *clp) */ static void nfs_free_client(struct nfs_client *clp) { - dprintk("--> nfs_free_client(%d)\n", clp->cl_nfsversion); + dprintk("--> nfs_free_client(%u)\n", clp->rpc_ops->version); nfs4_shutdown_client(clp); @@ -203,76 +215,148 @@ void nfs_put_client(struct nfs_client *clp) } } +static int nfs_sockaddr_match_ipaddr4(const struct sockaddr_in *sa1, + const struct sockaddr_in *sa2) +{ + return sa1->sin_addr.s_addr == sa2->sin_addr.s_addr; +} + +static int nfs_sockaddr_match_ipaddr6(const struct sockaddr_in6 *sa1, + const struct sockaddr_in6 *sa2) +{ + return ipv6_addr_equal(&sa1->sin6_addr, &sa2->sin6_addr); +} + +static int nfs_sockaddr_match_ipaddr(const struct sockaddr *sa1, + const struct sockaddr *sa2) +{ + switch (sa1->sa_family) { + case AF_INET: + return nfs_sockaddr_match_ipaddr4((const struct sockaddr_in *)sa1, + (const struct sockaddr_in *)sa2); + case AF_INET6: + return nfs_sockaddr_match_ipaddr6((const struct sockaddr_in6 *)sa1, + (const struct sockaddr_in6 *)sa2); + } + BUG(); +} + /* - * Find a client by address - * - caller must hold nfs_client_lock + * Find a client by IP address and protocol version + * - returns NULL if no such client */ -static struct nfs_client *__nfs_find_client(const struct sockaddr_in *addr, int nfsversion, int match_port) +struct nfs_client *nfs_find_client(const struct sockaddr *addr, u32 nfsversion) { struct nfs_client *clp; + spin_lock(&nfs_client_lock); list_for_each_entry(clp, &nfs_client_list, cl_share_link) { + struct sockaddr *clap = (struct sockaddr *)&clp->cl_addr; + /* Don't match clients that failed to initialise properly */ - if (clp->cl_cons_state < 0) + if (clp->cl_cons_state != NFS_CS_READY) continue; /* Different NFS versions cannot share the same nfs_client */ - if (clp->cl_nfsversion != nfsversion) + if (clp->rpc_ops->version != nfsversion) continue; - if (memcmp(&clp->cl_addr.sin_addr, &addr->sin_addr, - sizeof(clp->cl_addr.sin_addr)) != 0) + if (addr->sa_family != clap->sa_family) + continue; + /* Match only the IP address, not the port number */ + if (!nfs_sockaddr_match_ipaddr(addr, clap)) continue; - if (!match_port || clp->cl_addr.sin_port == addr->sin_port) - goto found; + atomic_inc(&clp->cl_count); + spin_unlock(&nfs_client_lock); + return clp; } - + spin_unlock(&nfs_client_lock); return NULL; - -found: - atomic_inc(&clp->cl_count); - return clp; } /* * Find a client by IP address and protocol version * - returns NULL if no such client */ -struct nfs_client *nfs_find_client(const struct sockaddr_in *addr, int nfsversion) +struct nfs_client *nfs_find_client_next(struct nfs_client *clp) { - struct nfs_client *clp; + struct sockaddr *sap = (struct sockaddr *)&clp->cl_addr; + u32 nfsvers = clp->rpc_ops->version; spin_lock(&nfs_client_lock); - clp = __nfs_find_client(addr, nfsversion, 0); + list_for_each_entry_continue(clp, &nfs_client_list, cl_share_link) { + struct sockaddr *clap = (struct sockaddr *)&clp->cl_addr; + + /* Don't match clients that failed to initialise properly */ + if (clp->cl_cons_state != NFS_CS_READY) + continue; + + /* Different NFS versions cannot share the same nfs_client */ + if (clp->rpc_ops->version != nfsvers) + continue; + + if (sap->sa_family != clap->sa_family) + continue; + /* Match only the IP address, not the port number */ + if (!nfs_sockaddr_match_ipaddr(sap, clap)) + continue; + + atomic_inc(&clp->cl_count); + spin_unlock(&nfs_client_lock); + return clp; + } spin_unlock(&nfs_client_lock); - if (clp != NULL && clp->cl_cons_state != NFS_CS_READY) { - nfs_put_client(clp); - clp = NULL; + return NULL; +} + +/* + * Find an nfs_client on the list that matches the initialisation data + * that is supplied. + */ +static struct nfs_client *nfs_match_client(const struct nfs_client_initdata *data) +{ + struct nfs_client *clp; + + list_for_each_entry(clp, &nfs_client_list, cl_share_link) { + /* Don't match clients that failed to initialise properly */ + if (clp->cl_cons_state < 0) + continue; + + /* Different NFS versions cannot share the same nfs_client */ + if (clp->rpc_ops != data->rpc_ops) + continue; + + if (clp->cl_proto != data->proto) + continue; + + /* Match the full socket address */ + if (memcmp(&clp->cl_addr, data->addr, sizeof(clp->cl_addr)) != 0) + continue; + + atomic_inc(&clp->cl_count); + return clp; } - return clp; + return NULL; } /* * Look up a client by IP address and protocol version * - creates a new record if one doesn't yet exist */ -static struct nfs_client *nfs_get_client(const char *hostname, - const struct sockaddr_in *addr, - int nfsversion) +static struct nfs_client *nfs_get_client(const struct nfs_client_initdata *cl_init) { struct nfs_client *clp, *new = NULL; int error; - dprintk("--> nfs_get_client(%s,"NIPQUAD_FMT":%d,%d)\n", - hostname ?: "", NIPQUAD(addr->sin_addr), - addr->sin_port, nfsversion); + dprintk("--> nfs_get_client(%s,v%u)\n", + cl_init->hostname ?: "", cl_init->rpc_ops->version); /* see if the client already exists */ do { spin_lock(&nfs_client_lock); - clp = __nfs_find_client(addr, nfsversion, 1); + clp = nfs_match_client(cl_init); if (clp) goto found_client; if (new) @@ -280,7 +364,7 @@ static struct nfs_client *nfs_get_client(const char *hostname, spin_unlock(&nfs_client_lock); - new = nfs_alloc_client(hostname, addr, nfsversion); + new = nfs_alloc_client(cl_init); } while (new); return ERR_PTR(-ENOMEM); @@ -302,7 +386,7 @@ found_client: if (new) nfs_free_client(new); - error = wait_event_interruptible(nfs_client_active_wq, + error = wait_event_killable(nfs_client_active_wq, clp->cl_cons_state != NFS_CS_INITING); if (error < 0) { nfs_put_client(clp); @@ -344,12 +428,16 @@ static void nfs_init_timeout_values(struct rpc_timeout *to, int proto, switch (proto) { case XPRT_TRANSPORT_TCP: case XPRT_TRANSPORT_RDMA: - if (!to->to_initval) + if (to->to_initval == 0) to->to_initval = 60 * HZ; if (to->to_initval > NFS_MAX_TCP_TIMEOUT) to->to_initval = NFS_MAX_TCP_TIMEOUT; to->to_increment = to->to_initval; to->to_maxval = to->to_initval + (to->to_increment * to->to_retries); + if (to->to_maxval > NFS_MAX_TCP_TIMEOUT) + to->to_maxval = NFS_MAX_TCP_TIMEOUT; + if (to->to_maxval < to->to_initval) + to->to_maxval = to->to_initval; to->to_exponential = 0; break; case XPRT_TRANSPORT_UDP: @@ -367,19 +455,17 @@ static void nfs_init_timeout_values(struct rpc_timeout *to, int proto, /* * Create an RPC client handle */ -static int nfs_create_rpc_client(struct nfs_client *clp, int proto, - unsigned int timeo, - unsigned int retrans, - rpc_authflavor_t flavor, - int flags) +static int nfs_create_rpc_client(struct nfs_client *clp, + const struct rpc_timeout *timeparms, + rpc_authflavor_t flavor, + int flags) { - struct rpc_timeout timeparms; struct rpc_clnt *clnt = NULL; struct rpc_create_args args = { - .protocol = proto, + .protocol = clp->cl_proto, .address = (struct sockaddr *)&clp->cl_addr, - .addrsize = sizeof(clp->cl_addr), - .timeout = &timeparms, + .addrsize = clp->cl_addrlen, + .timeout = timeparms, .servername = clp->cl_hostname, .program = &nfs_program, .version = clp->rpc_ops->version, @@ -390,10 +476,6 @@ static int nfs_create_rpc_client(struct nfs_client *clp, int proto, if (!IS_ERR(clp->cl_rpcclient)) return 0; - nfs_init_timeout_values(&timeparms, proto, timeo, retrans); - clp->retrans_timeo = timeparms.to_initval; - clp->retrans_count = timeparms.to_retries; - clnt = rpc_create(&args); if (IS_ERR(clnt)) { dprintk("%s: cannot create RPC client. Error = %ld\n", @@ -411,7 +493,7 @@ static int nfs_create_rpc_client(struct nfs_client *clp, int proto, static void nfs_destroy_server(struct nfs_server *server) { if (!(server->flags & NFS_MOUNT_NONLM)) - lockd_down(); /* release rpc.lockd */ + nlmclnt_done(server->nlm_host); } /* @@ -419,20 +501,29 @@ static void nfs_destroy_server(struct nfs_server *server) */ static int nfs_start_lockd(struct nfs_server *server) { - int error = 0; + struct nlm_host *host; + struct nfs_client *clp = server->nfs_client; + struct nlmclnt_initdata nlm_init = { + .hostname = clp->cl_hostname, + .address = (struct sockaddr *)&clp->cl_addr, + .addrlen = clp->cl_addrlen, + .protocol = server->flags & NFS_MOUNT_TCP ? + IPPROTO_TCP : IPPROTO_UDP, + .nfs_version = clp->rpc_ops->version, + }; - if (server->nfs_client->cl_nfsversion > 3) - goto out; + if (nlm_init.nfs_version > 3) + return 0; if (server->flags & NFS_MOUNT_NONLM) - goto out; - error = lockd_up((server->flags & NFS_MOUNT_TCP) ? - IPPROTO_TCP : IPPROTO_UDP); - if (error < 0) - server->flags |= NFS_MOUNT_NONLM; - else - server->destroy = nfs_destroy_server; -out: - return error; + return 0; + + host = nlmclnt_init(&nlm_init); + if (IS_ERR(host)) + return PTR_ERR(host); + + server->nlm_host = host; + server->destroy = nfs_destroy_server; + return 0; } /* @@ -441,7 +532,7 @@ out: #ifdef CONFIG_NFS_V3_ACL static void nfs_init_server_aclclient(struct nfs_server *server) { - if (server->nfs_client->cl_nfsversion != 3) + if (server->nfs_client->rpc_ops->version != 3) goto out_noacl; if (server->flags & NFS_MOUNT_NOACL) goto out_noacl; @@ -468,7 +559,9 @@ static inline void nfs_init_server_aclclient(struct nfs_server *server) /* * Create a general RPC client */ -static int nfs_init_server_rpcclient(struct nfs_server *server, rpc_authflavor_t pseudoflavour) +static int nfs_init_server_rpcclient(struct nfs_server *server, + const struct rpc_timeout *timeo, + rpc_authflavor_t pseudoflavour) { struct nfs_client *clp = server->nfs_client; @@ -478,6 +571,11 @@ static int nfs_init_server_rpcclient(struct nfs_server *server, rpc_authflavor_t return PTR_ERR(server->client); } + memcpy(&server->client->cl_timeout_default, + timeo, + sizeof(server->client->cl_timeout_default)); + server->client->cl_timeout = &server->client->cl_timeout_default; + if (pseudoflavour != clp->cl_rpcclient->cl_auth->au_flavor) { struct rpc_auth *auth; @@ -491,10 +589,6 @@ static int nfs_init_server_rpcclient(struct nfs_server *server, rpc_authflavor_t if (server->flags & NFS_MOUNT_SOFT) server->client->cl_softrtry = 1; - server->client->cl_intr = 0; - if (server->flags & NFS4_MOUNT_INTR) - server->client->cl_intr = 1; - return 0; } @@ -502,6 +596,7 @@ static int nfs_init_server_rpcclient(struct nfs_server *server, rpc_authflavor_t * Initialise an NFS2 or NFS3 client */ static int nfs_init_client(struct nfs_client *clp, + const struct rpc_timeout *timeparms, const struct nfs_parsed_mount_data *data) { int error; @@ -512,18 +607,11 @@ static int nfs_init_client(struct nfs_client *clp, return 0; } - /* Check NFS protocol revision and initialize RPC op vector */ - clp->rpc_ops = &nfs_v2_clientops; -#ifdef CONFIG_NFS_V3 - if (clp->cl_nfsversion == 3) - clp->rpc_ops = &nfs_v3_clientops; -#endif /* * Create a client RPC handle for doing FSSTAT with UNIX auth only * - RFC 2623, sec 2.3.2 */ - error = nfs_create_rpc_client(clp, data->nfs_server.protocol, - data->timeo, data->retrans, RPC_AUTH_UNIX, 0); + error = nfs_create_rpc_client(clp, timeparms, RPC_AUTH_UNIX, 0); if (error < 0) goto error; nfs_mark_client_ready(clp, NFS_CS_READY); @@ -541,25 +629,34 @@ error: static int nfs_init_server(struct nfs_server *server, const struct nfs_parsed_mount_data *data) { + struct nfs_client_initdata cl_init = { + .hostname = data->nfs_server.hostname, + .addr = (const struct sockaddr *)&data->nfs_server.address, + .addrlen = data->nfs_server.addrlen, + .rpc_ops = &nfs_v2_clientops, + .proto = data->nfs_server.protocol, + }; + struct rpc_timeout timeparms; struct nfs_client *clp; - int error, nfsvers = 2; + int error; dprintk("--> nfs_init_server()\n"); #ifdef CONFIG_NFS_V3 if (data->flags & NFS_MOUNT_VER3) - nfsvers = 3; + cl_init.rpc_ops = &nfs_v3_clientops; #endif /* Allocate or find a client reference we can use */ - clp = nfs_get_client(data->nfs_server.hostname, - &data->nfs_server.address, nfsvers); + clp = nfs_get_client(&cl_init); if (IS_ERR(clp)) { dprintk("<-- nfs_init_server() = error %ld\n", PTR_ERR(clp)); return PTR_ERR(clp); } - error = nfs_init_client(clp, data); + nfs_init_timeout_values(&timeparms, data->nfs_server.protocol, + data->timeo, data->retrans); + error = nfs_init_client(clp, &timeparms, data); if (error < 0) goto error; @@ -583,7 +680,7 @@ static int nfs_init_server(struct nfs_server *server, if (error < 0) goto error; - error = nfs_init_server_rpcclient(server, data->auth_flavors[0]); + error = nfs_init_server_rpcclient(server, &timeparms, data->auth_flavors[0]); if (error < 0) goto error; @@ -729,6 +826,9 @@ static struct nfs_server *nfs_alloc_server(void) INIT_LIST_HEAD(&server->client_link); INIT_LIST_HEAD(&server->master_link); + init_waitqueue_head(&server->active_wq); + atomic_set(&server->active, 0); + server->io_stats = nfs_alloc_iostats(); if (!server->io_stats) { kfree(server); @@ -840,7 +940,7 @@ error: * Initialise an NFS4 client record */ static int nfs4_init_client(struct nfs_client *clp, - int proto, int timeo, int retrans, + const struct rpc_timeout *timeparms, const char *ip_addr, rpc_authflavor_t authflavour) { @@ -855,7 +955,7 @@ static int nfs4_init_client(struct nfs_client *clp, /* Check NFS protocol revision and initialize RPC op vector */ clp->rpc_ops = &nfs_v4_clientops; - error = nfs_create_rpc_client(clp, proto, timeo, retrans, authflavour, + error = nfs_create_rpc_client(clp, timeparms, authflavour, RPC_CLNT_CREATE_DISCRTRY); if (error < 0) goto error; @@ -882,23 +982,32 @@ error: * Set up an NFS4 client */ static int nfs4_set_client(struct nfs_server *server, - const char *hostname, const struct sockaddr_in *addr, + const char *hostname, + const struct sockaddr *addr, + const size_t addrlen, const char *ip_addr, rpc_authflavor_t authflavour, - int proto, int timeo, int retrans) + int proto, const struct rpc_timeout *timeparms) { + struct nfs_client_initdata cl_init = { + .hostname = hostname, + .addr = addr, + .addrlen = addrlen, + .rpc_ops = &nfs_v4_clientops, + .proto = proto, + }; struct nfs_client *clp; int error; dprintk("--> nfs4_set_client()\n"); /* Allocate or find a client reference we can use */ - clp = nfs_get_client(hostname, addr, 4); + clp = nfs_get_client(&cl_init); if (IS_ERR(clp)) { error = PTR_ERR(clp); goto error; } - error = nfs4_init_client(clp, proto, timeo, retrans, ip_addr, authflavour); + error = nfs4_init_client(clp, timeparms, ip_addr, authflavour); if (error < 0) goto error_put; @@ -919,10 +1028,26 @@ error: static int nfs4_init_server(struct nfs_server *server, const struct nfs_parsed_mount_data *data) { + struct rpc_timeout timeparms; int error; dprintk("--> nfs4_init_server()\n"); + nfs_init_timeout_values(&timeparms, data->nfs_server.protocol, + data->timeo, data->retrans); + + /* Get a client record */ + error = nfs4_set_client(server, + data->nfs_server.hostname, + (const struct sockaddr *)&data->nfs_server.address, + data->nfs_server.addrlen, + data->client_address, + data->auth_flavors[0], + data->nfs_server.protocol, + &timeparms); + if (error < 0) + goto error; + /* Initialise the client representation from the mount data */ server->flags = data->flags & NFS_MOUNT_FLAGMASK; server->caps |= NFS_CAP_ATOMIC_OPEN; @@ -937,8 +1062,9 @@ static int nfs4_init_server(struct nfs_server *server, server->acdirmin = data->acdirmin * HZ; server->acdirmax = data->acdirmax * HZ; - error = nfs_init_server_rpcclient(server, data->auth_flavors[0]); + error = nfs_init_server_rpcclient(server, &timeparms, data->auth_flavors[0]); +error: /* Done */ dprintk("<-- nfs4_init_server() = %d\n", error); return error; @@ -961,17 +1087,6 @@ struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data, if (!server) return ERR_PTR(-ENOMEM); - /* Get a client record */ - error = nfs4_set_client(server, - data->nfs_server.hostname, - &data->nfs_server.address, - data->client_address, - data->auth_flavors[0], - data->nfs_server.protocol, - data->timeo, data->retrans); - if (error < 0) - goto error; - /* set up the general RPC client */ error = nfs4_init_server(server, data); if (error < 0) @@ -1039,12 +1154,13 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data, /* Get a client representation. * Note: NFSv4 always uses TCP, */ - error = nfs4_set_client(server, data->hostname, data->addr, - parent_client->cl_ipaddr, - data->authflavor, - parent_server->client->cl_xprt->prot, - parent_client->retrans_timeo, - parent_client->retrans_count); + error = nfs4_set_client(server, data->hostname, + data->addr, + data->addrlen, + parent_client->cl_ipaddr, + data->authflavor, + parent_server->client->cl_xprt->prot, + parent_server->client->cl_timeout); if (error < 0) goto error; @@ -1052,7 +1168,7 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data, nfs_server_copy_userdata(server, parent_server); server->caps |= NFS_CAP_ATOMIC_OPEN; - error = nfs_init_server_rpcclient(server, data->authflavor); + error = nfs_init_server_rpcclient(server, parent_server->client->cl_timeout, data->authflavor); if (error < 0) goto error; @@ -1121,7 +1237,9 @@ struct nfs_server *nfs_clone_server(struct nfs_server *source, server->fsid = fattr->fsid; - error = nfs_init_server_rpcclient(server, source->client->cl_auth->au_flavor); + error = nfs_init_server_rpcclient(server, + source->client->cl_timeout, + source->client->cl_auth->au_flavor); if (error < 0) goto out_free_server; if (!IS_ERR(source->client_acl)) @@ -1263,10 +1381,10 @@ static int nfs_server_list_show(struct seq_file *m, void *v) /* display one transport per line on subsequent lines */ clp = list_entry(v, struct nfs_client, cl_share_link); - seq_printf(m, "v%d %02x%02x%02x%02x %4hx %3d %s\n", - clp->cl_nfsversion, - NIPQUAD(clp->cl_addr.sin_addr), - ntohs(clp->cl_addr.sin_port), + seq_printf(m, "v%u %s %s %3d %s\n", + clp->rpc_ops->version, + rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_HEX_ADDR), + rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_HEX_PORT), atomic_read(&clp->cl_count), clp->cl_hostname); @@ -1342,10 +1460,10 @@ static int nfs_volume_list_show(struct seq_file *m, void *v) (unsigned long long) server->fsid.major, (unsigned long long) server->fsid.minor); - seq_printf(m, "v%d %02x%02x%02x%02x %4hx %-7s %-17s\n", - clp->cl_nfsversion, - NIPQUAD(clp->cl_addr.sin_addr), - ntohs(clp->cl_addr.sin_port), + seq_printf(m, "v%u %s %s %-7s %-17s\n", + clp->rpc_ops->version, + rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_HEX_ADDR), + rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_HEX_PORT), dev, fsid); diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index 11833f4caea..00a5e4405e1 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -49,7 +49,7 @@ static int nfs_delegation_claim_locks(struct nfs_open_context *ctx, struct nfs4_ struct file_lock *fl; int status; - for (fl = inode->i_flock; fl != 0; fl = fl->fl_next) { + for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) { if (!(fl->fl_flags & (FL_POSIX|FL_FLOCK))) continue; if (nfs_file_open_context(fl->fl_file) != ctx) @@ -125,6 +125,32 @@ void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, st put_rpccred(oldcred); } +static int nfs_do_return_delegation(struct inode *inode, struct nfs_delegation *delegation, int issync) +{ + int res = 0; + + res = nfs4_proc_delegreturn(inode, delegation->cred, &delegation->stateid, issync); + nfs_free_delegation(delegation); + return res; +} + +static struct nfs_delegation *nfs_detach_delegation_locked(struct nfs_inode *nfsi, const nfs4_stateid *stateid) +{ + struct nfs_delegation *delegation = rcu_dereference(nfsi->delegation); + + if (delegation == NULL) + goto nomatch; + if (stateid != NULL && memcmp(delegation->stateid.data, stateid->data, + sizeof(delegation->stateid.data)) != 0) + goto nomatch; + list_del_rcu(&delegation->super_list); + nfsi->delegation_state = 0; + rcu_assign_pointer(nfsi->delegation, NULL); + return delegation; +nomatch: + return NULL; +} + /* * Set up a delegation on an inode */ @@ -133,6 +159,7 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct struct nfs_client *clp = NFS_SERVER(inode)->nfs_client; struct nfs_inode *nfsi = NFS_I(inode); struct nfs_delegation *delegation; + struct nfs_delegation *freeme = NULL; int status = 0; delegation = kmalloc(sizeof(*delegation), GFP_KERNEL); @@ -147,41 +174,45 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct delegation->inode = inode; spin_lock(&clp->cl_lock); - if (rcu_dereference(nfsi->delegation) == NULL) { - list_add_rcu(&delegation->super_list, &clp->cl_delegations); - nfsi->delegation_state = delegation->type; - rcu_assign_pointer(nfsi->delegation, delegation); - delegation = NULL; - } else { + if (rcu_dereference(nfsi->delegation) != NULL) { if (memcmp(&delegation->stateid, &nfsi->delegation->stateid, - sizeof(delegation->stateid)) != 0 || - delegation->type != nfsi->delegation->type) { - printk("%s: server %u.%u.%u.%u, handed out a duplicate delegation!\n", - __FUNCTION__, NIPQUAD(clp->cl_addr.sin_addr)); - status = -EIO; + sizeof(delegation->stateid)) == 0 && + delegation->type == nfsi->delegation->type) { + goto out; + } + /* + * Deal with broken servers that hand out two + * delegations for the same file. + */ + dfprintk(FILE, "%s: server %s handed out " + "a duplicate delegation!\n", + __FUNCTION__, clp->cl_hostname); + if (delegation->type <= nfsi->delegation->type) { + freeme = delegation; + delegation = NULL; + goto out; } + freeme = nfs_detach_delegation_locked(nfsi, NULL); } + list_add_rcu(&delegation->super_list, &clp->cl_delegations); + nfsi->delegation_state = delegation->type; + rcu_assign_pointer(nfsi->delegation, delegation); + delegation = NULL; /* Ensure we revalidate the attributes and page cache! */ spin_lock(&inode->i_lock); nfsi->cache_validity |= NFS_INO_REVAL_FORCED; spin_unlock(&inode->i_lock); +out: spin_unlock(&clp->cl_lock); if (delegation != NULL) nfs_free_delegation(delegation); + if (freeme != NULL) + nfs_do_return_delegation(inode, freeme, 0); return status; } -static int nfs_do_return_delegation(struct inode *inode, struct nfs_delegation *delegation) -{ - int res = 0; - - res = nfs4_proc_delegreturn(inode, delegation->cred, &delegation->stateid); - nfs_free_delegation(delegation); - return res; -} - /* Sync all data to disk upon delegation return */ static void nfs_msync_inode(struct inode *inode) { @@ -207,24 +238,28 @@ static int __nfs_inode_return_delegation(struct inode *inode, struct nfs_delegat up_read(&clp->cl_sem); nfs_msync_inode(inode); - return nfs_do_return_delegation(inode, delegation); + return nfs_do_return_delegation(inode, delegation, 1); } -static struct nfs_delegation *nfs_detach_delegation_locked(struct nfs_inode *nfsi, const nfs4_stateid *stateid) +/* + * This function returns the delegation without reclaiming opens + * or protecting against delegation reclaims. + * It is therefore really only safe to be called from + * nfs4_clear_inode() + */ +void nfs_inode_return_delegation_noreclaim(struct inode *inode) { - struct nfs_delegation *delegation = rcu_dereference(nfsi->delegation); + struct nfs_client *clp = NFS_SERVER(inode)->nfs_client; + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs_delegation *delegation; - if (delegation == NULL) - goto nomatch; - if (stateid != NULL && memcmp(delegation->stateid.data, stateid->data, - sizeof(delegation->stateid.data)) != 0) - goto nomatch; - list_del_rcu(&delegation->super_list); - nfsi->delegation_state = 0; - rcu_assign_pointer(nfsi->delegation, NULL); - return delegation; -nomatch: - return NULL; + if (rcu_dereference(nfsi->delegation) != NULL) { + spin_lock(&clp->cl_lock); + delegation = nfs_detach_delegation_locked(nfsi, NULL); + spin_unlock(&clp->cl_lock); + if (delegation != NULL) + nfs_do_return_delegation(inode, delegation, 0); + } } int nfs_inode_return_delegation(struct inode *inode) @@ -314,8 +349,9 @@ void nfs_expire_all_delegations(struct nfs_client *clp) __module_get(THIS_MODULE); atomic_inc(&clp->cl_count); task = kthread_run(nfs_do_expire_all_delegations, clp, - "%u.%u.%u.%u-delegreturn", - NIPQUAD(clp->cl_addr.sin_addr)); + "%s-delegreturn", + rpc_peeraddr2str(clp->cl_rpcclient, + RPC_DISPLAY_ADDR)); if (!IS_ERR(task)) return; nfs_put_client(clp); @@ -386,7 +422,7 @@ static int recall_thread(void *data) nfs_msync_inode(inode); if (delegation != NULL) - nfs_do_return_delegation(inode, delegation); + nfs_do_return_delegation(inode, delegation, 1); iput(inode); module_put_and_exit(0); } diff --git a/fs/nfs/delegation.h b/fs/nfs/delegation.h index 5874ce7fdba..f1c5e2a5d88 100644 --- a/fs/nfs/delegation.h +++ b/fs/nfs/delegation.h @@ -29,6 +29,7 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res); int nfs_inode_return_delegation(struct inode *inode); int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid); +void nfs_inode_return_delegation_noreclaim(struct inode *inode); struct inode *nfs_delegation_find_inode(struct nfs_client *clp, const struct nfs_fh *fhandle); void nfs_return_all_delegations(struct super_block *sb); @@ -39,7 +40,7 @@ void nfs_delegation_mark_reclaim(struct nfs_client *clp); void nfs_delegation_reap_unclaimed(struct nfs_client *clp); /* NFSv4 delegation-related procedures */ -int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4_stateid *stateid); +int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4_stateid *stateid, int issync); int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state *state, const nfs4_stateid *stateid); int nfs4_lock_delegation_recall(struct nfs4_state *state, struct file_lock *fl); int nfs4_copy_delegation_stateid(nfs4_stateid *dst, struct inode *inode); diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index f697b5c74b7..6cea7479c5b 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -154,7 +154,6 @@ typedef struct { struct nfs_entry *entry; decode_dirent_t decode; int plus; - int error; unsigned long timestamp; int timestamp_valid; } nfs_readdir_descriptor_t; @@ -192,7 +191,7 @@ int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page *page) /* We requested READDIRPLUS, but the server doesn't grok it */ if (error == -ENOTSUPP && desc->plus) { NFS_SERVER(inode)->caps &= ~NFS_CAP_READDIRPLUS; - clear_bit(NFS_INO_ADVISE_RDPLUS, &NFS_FLAGS(inode)); + clear_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(inode)->flags); desc->plus = 0; goto again; } @@ -213,7 +212,6 @@ int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page *page) return 0; error: unlock_page(page); - desc->error = error; return -EIO; } @@ -483,13 +481,13 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent, goto out; } timestamp = jiffies; - desc->error = NFS_PROTO(inode)->readdir(file->f_path.dentry, cred, *desc->dir_cookie, - page, + status = NFS_PROTO(inode)->readdir(file->f_path.dentry, cred, + *desc->dir_cookie, page, NFS_SERVER(inode)->dtsize, desc->plus); desc->page = page; desc->ptr = kmap(page); /* matching kunmap in nfs_do_filldir */ - if (desc->error >= 0) { + if (status >= 0) { desc->timestamp = timestamp; desc->timestamp_valid = 1; if ((status = dir_decode(desc)) == 0) @@ -537,12 +535,6 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) lock_kernel(); - res = nfs_revalidate_mapping_nolock(inode, filp->f_mapping); - if (res < 0) { - unlock_kernel(); - return res; - } - /* * filp->f_pos points to the dirent entry number. * *desc->dir_cookie has the cookie for the next entry. We have @@ -564,6 +556,10 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) desc->entry = &my_entry; nfs_block_sillyrename(dentry); + res = nfs_revalidate_mapping_nolock(inode, filp->f_mapping); + if (res < 0) + goto out; + while(!desc->entry->eof) { res = readdir_search_pagecache(desc); @@ -579,7 +575,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) break; } if (res == -ETOOSMALL && desc->plus) { - clear_bit(NFS_INO_ADVISE_RDPLUS, &NFS_FLAGS(inode)); + clear_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(inode)->flags); nfs_zap_caches(inode); desc->plus = 0; desc->entry->eof = 0; @@ -594,6 +590,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) break; } } +out: nfs_unblock_sillyrename(dentry); unlock_kernel(); if (res > 0) @@ -639,6 +636,21 @@ static int nfs_fsync_dir(struct file *filp, struct dentry *dentry, int datasync) return 0; } +/** + * nfs_force_lookup_revalidate - Mark the directory as having changed + * @dir - pointer to directory inode + * + * This forces the revalidation code in nfs_lookup_revalidate() to do a + * full lookup on all child dentries of 'dir' whenever a change occurs + * on the server that might have invalidated our dcache. + * + * The caller should be holding dir->i_lock + */ +void nfs_force_lookup_revalidate(struct inode *dir) +{ + NFS_I(dir)->cache_change_attribute = jiffies; +} + /* * A check for whether or not the parent directory has changed. * In the case it has, we assume that the dentries are untrustworthy @@ -698,6 +710,8 @@ int nfs_lookup_verify_inode(struct inode *inode, struct nameidata *nd) { struct nfs_server *server = NFS_SERVER(inode); + if (test_bit(NFS_INO_MOUNTPOINT, &NFS_I(inode)->flags)) + return 0; if (nd != NULL) { /* VFS wants an on-the-wire revalidation */ if (nd->flags & LOOKUP_REVAL) @@ -827,6 +841,10 @@ static int nfs_dentry_delete(struct dentry *dentry) dentry->d_parent->d_name.name, dentry->d_name.name, dentry->d_flags); + /* Unhash any dentry with a stale inode */ + if (dentry->d_inode != NULL && NFS_STALE(dentry->d_inode)) + return 1; + if (dentry->d_flags & DCACHE_NFSFS_RENAMED) { /* Unhash it, so that ->d_iput() would be called */ return 1; @@ -846,7 +864,6 @@ static int nfs_dentry_delete(struct dentry *dentry) */ static void nfs_dentry_iput(struct dentry *dentry, struct inode *inode) { - nfs_inode_return_delegation(inode); if (S_ISDIR(inode->i_mode)) /* drop any readdir cache as it could easily be old */ NFS_I(inode)->cache_validity |= NFS_INO_INVALID_DATA; @@ -1268,6 +1285,12 @@ out_err: return error; } +static void nfs_dentry_handle_enoent(struct dentry *dentry) +{ + if (dentry->d_inode != NULL && !d_unhashed(dentry)) + d_delete(dentry); +} + static int nfs_rmdir(struct inode *dir, struct dentry *dentry) { int error; @@ -1280,6 +1303,8 @@ static int nfs_rmdir(struct inode *dir, struct dentry *dentry) /* Ensure the VFS deletes this inode */ if (error == 0 && dentry->d_inode != NULL) clear_nlink(dentry->d_inode); + else if (error == -ENOENT) + nfs_dentry_handle_enoent(dentry); unlock_kernel(); return error; @@ -1386,6 +1411,8 @@ static int nfs_safe_remove(struct dentry *dentry) nfs_mark_for_revalidate(inode); } else error = NFS_PROTO(dir)->remove(dir, &dentry->d_name); + if (error == -ENOENT) + nfs_dentry_handle_enoent(dentry); out: return error; } @@ -1422,7 +1449,7 @@ static int nfs_unlink(struct inode *dir, struct dentry *dentry) spin_unlock(&dentry->d_lock); spin_unlock(&dcache_lock); error = nfs_safe_remove(dentry); - if (!error) { + if (!error || error == -ENOENT) { nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); } else if (need_rehash) d_rehash(dentry); @@ -1635,7 +1662,8 @@ out: d_move(old_dentry, new_dentry); nfs_set_verifier(new_dentry, nfs_save_change_attribute(new_dir)); - } + } else if (error == -ENOENT) + nfs_dentry_handle_enoent(old_dentry); /* new dentry created? */ if (dentry) @@ -1666,13 +1694,19 @@ int nfs_access_cache_shrinker(int nr_to_scan, gfp_t gfp_mask) restart: spin_lock(&nfs_access_lru_lock); list_for_each_entry(nfsi, &nfs_access_lru_list, access_cache_inode_lru) { + struct rw_semaphore *s_umount; struct inode *inode; if (nr_to_scan-- == 0) break; + s_umount = &nfsi->vfs_inode.i_sb->s_umount; + if (!down_read_trylock(s_umount)) + continue; inode = igrab(&nfsi->vfs_inode); - if (inode == NULL) + if (inode == NULL) { + up_read(s_umount); continue; + } spin_lock(&inode->i_lock); if (list_empty(&nfsi->access_cache_entry_lru)) goto remove_lru_entry; @@ -1691,6 +1725,7 @@ remove_lru_entry: spin_unlock(&inode->i_lock); spin_unlock(&nfs_access_lru_lock); iput(inode); + up_read(s_umount); goto restart; } spin_unlock(&nfs_access_lru_lock); @@ -1731,7 +1766,7 @@ static void __nfs_access_zap_cache(struct inode *inode) void nfs_access_zap_cache(struct inode *inode) { /* Remove from global LRU init */ - if (test_and_clear_bit(NFS_INO_ACL_LRU_SET, &NFS_FLAGS(inode))) { + if (test_and_clear_bit(NFS_INO_ACL_LRU_SET, &NFS_I(inode)->flags)) { spin_lock(&nfs_access_lru_lock); list_del_init(&NFS_I(inode)->access_cache_inode_lru); spin_unlock(&nfs_access_lru_lock); @@ -1845,7 +1880,7 @@ static void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *s smp_mb__after_atomic_inc(); /* Add inode to global LRU list */ - if (!test_and_set_bit(NFS_INO_ACL_LRU_SET, &NFS_FLAGS(inode))) { + if (!test_and_set_bit(NFS_INO_ACL_LRU_SET, &NFS_I(inode)->flags)) { spin_lock(&nfs_access_lru_lock); list_add_tail(&NFS_I(inode)->access_cache_inode_lru, &nfs_access_lru_list); spin_unlock(&nfs_access_lru_lock); diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 3c9d16b4f80..16844f98f50 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -193,7 +193,7 @@ static ssize_t nfs_direct_wait(struct nfs_direct_req *dreq) if (dreq->iocb) goto out; - result = wait_for_completion_interruptible(&dreq->completion); + result = wait_for_completion_killable(&dreq->completion); if (!result) result = dreq->error; @@ -272,6 +272,16 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_direct_req *dreq, unsigned long user_addr = (unsigned long)iov->iov_base; size_t count = iov->iov_len; size_t rsize = NFS_SERVER(inode)->rsize; + struct rpc_task *task; + struct rpc_message msg = { + .rpc_cred = ctx->cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = NFS_CLIENT(inode), + .rpc_message = &msg, + .callback_ops = &nfs_read_direct_ops, + .flags = RPC_TASK_ASYNC, + }; unsigned int pgbase; int result; ssize_t started = 0; @@ -311,7 +321,7 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_direct_req *dreq, data->req = (struct nfs_page *) dreq; data->inode = inode; - data->cred = ctx->cred; + data->cred = msg.rpc_cred; data->args.fh = NFS_FH(inode); data->args.context = ctx; data->args.offset = pos; @@ -321,14 +331,16 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_direct_req *dreq, data->res.fattr = &data->fattr; data->res.eof = 0; data->res.count = bytes; + msg.rpc_argp = &data->args; + msg.rpc_resp = &data->res; - rpc_init_task(&data->task, NFS_CLIENT(inode), RPC_TASK_ASYNC, - &nfs_read_direct_ops, data); - NFS_PROTO(inode)->read_setup(data); + task_setup_data.task = &data->task; + task_setup_data.callback_data = data; + NFS_PROTO(inode)->read_setup(data, &msg); - data->task.tk_cookie = (unsigned long) inode; - - rpc_execute(&data->task); + task = rpc_run_task(&task_setup_data); + if (!IS_ERR(task)) + rpc_put_task(task); dprintk("NFS: %5u initiated direct read call " "(req %s/%Ld, %zu bytes @ offset %Lu)\n", @@ -391,9 +403,7 @@ static ssize_t nfs_direct_read(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) { ssize_t result = 0; - sigset_t oldset; struct inode *inode = iocb->ki_filp->f_mapping->host; - struct rpc_clnt *clnt = NFS_CLIENT(inode); struct nfs_direct_req *dreq; dreq = nfs_direct_req_alloc(); @@ -405,11 +415,9 @@ static ssize_t nfs_direct_read(struct kiocb *iocb, const struct iovec *iov, if (!is_sync_kiocb(iocb)) dreq->iocb = iocb; - rpc_clnt_sigmask(clnt, &oldset); result = nfs_direct_read_schedule_iovec(dreq, iov, nr_segs, pos); if (!result) result = nfs_direct_wait(dreq); - rpc_clnt_sigunmask(clnt, &oldset); nfs_direct_req_release(dreq); return result; @@ -431,6 +439,15 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq) struct inode *inode = dreq->inode; struct list_head *p; struct nfs_write_data *data; + struct rpc_task *task; + struct rpc_message msg = { + .rpc_cred = dreq->ctx->cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = NFS_CLIENT(inode), + .callback_ops = &nfs_write_direct_ops, + .flags = RPC_TASK_ASYNC, + }; dreq->count = 0; get_dreq(dreq); @@ -440,6 +457,9 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq) get_dreq(dreq); + /* Use stable writes */ + data->args.stable = NFS_FILE_SYNC; + /* * Reset data->res. */ @@ -451,17 +471,18 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq) * Reuse data->task; data->args should not have changed * since the original request was sent. */ - rpc_init_task(&data->task, NFS_CLIENT(inode), RPC_TASK_ASYNC, - &nfs_write_direct_ops, data); - NFS_PROTO(inode)->write_setup(data, FLUSH_STABLE); - - data->task.tk_priority = RPC_PRIORITY_NORMAL; - data->task.tk_cookie = (unsigned long) inode; + task_setup_data.task = &data->task; + task_setup_data.callback_data = data; + msg.rpc_argp = &data->args; + msg.rpc_resp = &data->res; + NFS_PROTO(inode)->write_setup(data, &msg); /* * We're called via an RPC callback, so BKL is already held. */ - rpc_execute(&data->task); + task = rpc_run_task(&task_setup_data); + if (!IS_ERR(task)) + rpc_put_task(task); dprintk("NFS: %5u rescheduled direct write call (req %s/%Ld, %u bytes @ offset %Lu)\n", data->task.tk_pid, @@ -504,9 +525,23 @@ static const struct rpc_call_ops nfs_commit_direct_ops = { static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq) { struct nfs_write_data *data = dreq->commit_data; + struct rpc_task *task; + struct rpc_message msg = { + .rpc_argp = &data->args, + .rpc_resp = &data->res, + .rpc_cred = dreq->ctx->cred, + }; + struct rpc_task_setup task_setup_data = { + .task = &data->task, + .rpc_client = NFS_CLIENT(dreq->inode), + .rpc_message = &msg, + .callback_ops = &nfs_commit_direct_ops, + .callback_data = data, + .flags = RPC_TASK_ASYNC, + }; data->inode = dreq->inode; - data->cred = dreq->ctx->cred; + data->cred = msg.rpc_cred; data->args.fh = NFS_FH(data->inode); data->args.offset = 0; @@ -515,18 +550,16 @@ static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq) data->res.fattr = &data->fattr; data->res.verf = &data->verf; - rpc_init_task(&data->task, NFS_CLIENT(dreq->inode), RPC_TASK_ASYNC, - &nfs_commit_direct_ops, data); - NFS_PROTO(data->inode)->commit_setup(data, 0); + NFS_PROTO(data->inode)->commit_setup(data, &msg); - data->task.tk_priority = RPC_PRIORITY_NORMAL; - data->task.tk_cookie = (unsigned long)data->inode; /* Note: task.tk_ops->rpc_release will free dreq->commit_data */ dreq->commit_data = NULL; dprintk("NFS: %5u initiated commit call\n", data->task.tk_pid); - rpc_execute(&data->task); + task = rpc_run_task(&task_setup_data); + if (!IS_ERR(task)) + rpc_put_task(task); } static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode *inode) @@ -641,6 +674,16 @@ static ssize_t nfs_direct_write_schedule_segment(struct nfs_direct_req *dreq, struct inode *inode = ctx->path.dentry->d_inode; unsigned long user_addr = (unsigned long)iov->iov_base; size_t count = iov->iov_len; + struct rpc_task *task; + struct rpc_message msg = { + .rpc_cred = ctx->cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = NFS_CLIENT(inode), + .rpc_message = &msg, + .callback_ops = &nfs_write_direct_ops, + .flags = RPC_TASK_ASYNC, + }; size_t wsize = NFS_SERVER(inode)->wsize; unsigned int pgbase; int result; @@ -683,25 +726,27 @@ static ssize_t nfs_direct_write_schedule_segment(struct nfs_direct_req *dreq, data->req = (struct nfs_page *) dreq; data->inode = inode; - data->cred = ctx->cred; + data->cred = msg.rpc_cred; data->args.fh = NFS_FH(inode); data->args.context = ctx; data->args.offset = pos; data->args.pgbase = pgbase; data->args.pages = data->pagevec; data->args.count = bytes; + data->args.stable = sync; data->res.fattr = &data->fattr; data->res.count = bytes; data->res.verf = &data->verf; - rpc_init_task(&data->task, NFS_CLIENT(inode), RPC_TASK_ASYNC, - &nfs_write_direct_ops, data); - NFS_PROTO(inode)->write_setup(data, sync); - - data->task.tk_priority = RPC_PRIORITY_NORMAL; - data->task.tk_cookie = (unsigned long) inode; + task_setup_data.task = &data->task; + task_setup_data.callback_data = data; + msg.rpc_argp = &data->args; + msg.rpc_resp = &data->res; + NFS_PROTO(inode)->write_setup(data, &msg); - rpc_execute(&data->task); + task = rpc_run_task(&task_setup_data); + if (!IS_ERR(task)) + rpc_put_task(task); dprintk("NFS: %5u initiated direct write call " "(req %s/%Ld, %zu bytes @ offset %Lu)\n", @@ -767,12 +812,10 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, const struct iovec *iov, size_t count) { ssize_t result = 0; - sigset_t oldset; struct inode *inode = iocb->ki_filp->f_mapping->host; - struct rpc_clnt *clnt = NFS_CLIENT(inode); struct nfs_direct_req *dreq; size_t wsize = NFS_SERVER(inode)->wsize; - int sync = 0; + int sync = NFS_UNSTABLE; dreq = nfs_direct_req_alloc(); if (!dreq) @@ -780,18 +823,16 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, const struct iovec *iov, nfs_alloc_commit_data(dreq); if (dreq->commit_data == NULL || count < wsize) - sync = FLUSH_STABLE; + sync = NFS_FILE_SYNC; dreq->inode = inode; dreq->ctx = get_nfs_open_context(nfs_file_open_context(iocb->ki_filp)); if (!is_sync_kiocb(iocb)) dreq->iocb = iocb; - rpc_clnt_sigmask(clnt, &oldset); result = nfs_direct_write_schedule_iovec(dreq, iov, nr_segs, pos, sync); if (!result) result = nfs_direct_wait(dreq); - rpc_clnt_sigunmask(clnt, &oldset); nfs_direct_req_release(dreq); return result; diff --git a/fs/nfs/file.c b/fs/nfs/file.c index b3bb89f7d5d..5d2e9d9a4e2 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -64,7 +64,11 @@ const struct file_operations nfs_file_operations = { .write = do_sync_write, .aio_read = nfs_file_read, .aio_write = nfs_file_write, +#ifdef CONFIG_MMU .mmap = nfs_file_mmap, +#else + .mmap = generic_file_mmap, +#endif .open = nfs_file_open, .flush = nfs_file_flush, .release = nfs_file_release, @@ -349,7 +353,9 @@ static int nfs_write_end(struct file *file, struct address_space *mapping, unlock_page(page); page_cache_release(page); - return status < 0 ? status : copied; + if (status < 0) + return status; + return copied; } static void nfs_invalidate_page(struct page *page, unsigned long offset) @@ -392,35 +398,27 @@ static int nfs_vm_page_mkwrite(struct vm_area_struct *vma, struct page *page) struct file *filp = vma->vm_file; unsigned pagelen; int ret = -EINVAL; - void *fsdata; struct address_space *mapping; - loff_t offset; lock_page(page); mapping = page->mapping; - if (mapping != vma->vm_file->f_path.dentry->d_inode->i_mapping) { - unlock_page(page); - return -EINVAL; - } + if (mapping != vma->vm_file->f_path.dentry->d_inode->i_mapping) + goto out_unlock; + + ret = 0; pagelen = nfs_page_length(page); - offset = (loff_t)page->index << PAGE_CACHE_SHIFT; - unlock_page(page); + if (pagelen == 0) + goto out_unlock; - /* - * we can use mapping after releasing the page lock, because: - * we hold mmap_sem on the fault path, which should pin the vma - * which should pin the file, which pins the dentry which should - * hold a reference on inode. - */ + ret = nfs_flush_incompatible(filp, page); + if (ret != 0) + goto out_unlock; - if (pagelen) { - struct page *page2 = NULL; - ret = nfs_write_begin(filp, mapping, offset, pagelen, - 0, &page2, &fsdata); - if (!ret) - ret = nfs_write_end(filp, mapping, offset, pagelen, - pagelen, page2, fsdata); - } + ret = nfs_updatepage(filp, page, 0, pagelen); + if (ret == 0) + ret = pagelen; +out_unlock: + unlock_page(page); return ret; } diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c index e6242cdbaf9..fae97196daa 100644 --- a/fs/nfs/getroot.c +++ b/fs/nfs/getroot.c @@ -96,7 +96,7 @@ struct dentry *nfs_get_root(struct super_block *sb, struct nfs_fh *mntfh) inode = nfs_fhget(sb, mntfh, fsinfo.fattr); if (IS_ERR(inode)) { dprintk("nfs_get_root: get root inode failed\n"); - return ERR_PTR(PTR_ERR(inode)); + return ERR_CAST(inode); } error = nfs_superblock_set_dummy_root(sb, inode); @@ -266,7 +266,7 @@ struct dentry *nfs4_get_root(struct super_block *sb, struct nfs_fh *mntfh) inode = nfs_fhget(sb, mntfh, &fattr); if (IS_ERR(inode)) { dprintk("nfs_get_root: get root inode failed\n"); - return ERR_PTR(PTR_ERR(inode)); + return ERR_CAST(inode); } error = nfs_superblock_set_dummy_root(sb, inode); diff --git a/fs/nfs/idmap.c b/fs/nfs/idmap.c index d11eb055265..86147b0ab2c 100644 --- a/fs/nfs/idmap.c +++ b/fs/nfs/idmap.c @@ -72,39 +72,39 @@ module_param_call(idmap_cache_timeout, param_set_idmap_timeout, param_get_int, &nfs_idmap_cache_timeout, 0644); struct idmap_hashent { - unsigned long ih_expires; - __u32 ih_id; - int ih_namelen; - char ih_name[IDMAP_NAMESZ]; + unsigned long ih_expires; + __u32 ih_id; + size_t ih_namelen; + char ih_name[IDMAP_NAMESZ]; }; struct idmap_hashtable { - __u8 h_type; - struct idmap_hashent h_entries[IDMAP_HASH_SZ]; + __u8 h_type; + struct idmap_hashent h_entries[IDMAP_HASH_SZ]; }; struct idmap { - struct dentry *idmap_dentry; - wait_queue_head_t idmap_wq; - struct idmap_msg idmap_im; - struct mutex idmap_lock; /* Serializes upcalls */ - struct mutex idmap_im_lock; /* Protects the hashtable */ - struct idmap_hashtable idmap_user_hash; - struct idmap_hashtable idmap_group_hash; + struct dentry *idmap_dentry; + wait_queue_head_t idmap_wq; + struct idmap_msg idmap_im; + struct mutex idmap_lock; /* Serializes upcalls */ + struct mutex idmap_im_lock; /* Protects the hashtable */ + struct idmap_hashtable idmap_user_hash; + struct idmap_hashtable idmap_group_hash; }; -static ssize_t idmap_pipe_upcall(struct file *, struct rpc_pipe_msg *, - char __user *, size_t); -static ssize_t idmap_pipe_downcall(struct file *, const char __user *, - size_t); -static void idmap_pipe_destroy_msg(struct rpc_pipe_msg *); +static ssize_t idmap_pipe_upcall(struct file *, struct rpc_pipe_msg *, + char __user *, size_t); +static ssize_t idmap_pipe_downcall(struct file *, const char __user *, + size_t); +static void idmap_pipe_destroy_msg(struct rpc_pipe_msg *); static unsigned int fnvhash32(const void *, size_t); static struct rpc_pipe_ops idmap_upcall_ops = { - .upcall = idmap_pipe_upcall, - .downcall = idmap_pipe_downcall, - .destroy_msg = idmap_pipe_destroy_msg, + .upcall = idmap_pipe_upcall, + .downcall = idmap_pipe_downcall, + .destroy_msg = idmap_pipe_destroy_msg, }; int @@ -115,19 +115,20 @@ nfs_idmap_new(struct nfs_client *clp) BUG_ON(clp->cl_idmap != NULL); - if ((idmap = kzalloc(sizeof(*idmap), GFP_KERNEL)) == NULL) - return -ENOMEM; + idmap = kzalloc(sizeof(*idmap), GFP_KERNEL); + if (idmap == NULL) + return -ENOMEM; - idmap->idmap_dentry = rpc_mkpipe(clp->cl_rpcclient->cl_dentry, "idmap", - idmap, &idmap_upcall_ops, 0); - if (IS_ERR(idmap->idmap_dentry)) { + idmap->idmap_dentry = rpc_mkpipe(clp->cl_rpcclient->cl_dentry, "idmap", + idmap, &idmap_upcall_ops, 0); + if (IS_ERR(idmap->idmap_dentry)) { error = PTR_ERR(idmap->idmap_dentry); kfree(idmap); return error; } - mutex_init(&idmap->idmap_lock); - mutex_init(&idmap->idmap_im_lock); + mutex_init(&idmap->idmap_lock); + mutex_init(&idmap->idmap_im_lock); init_waitqueue_head(&idmap->idmap_wq); idmap->idmap_user_hash.h_type = IDMAP_TYPE_USER; idmap->idmap_group_hash.h_type = IDMAP_TYPE_GROUP; @@ -192,7 +193,7 @@ idmap_lookup_id(struct idmap_hashtable *h, __u32 id) * pretty trivial. */ static inline struct idmap_hashent * -idmap_alloc_name(struct idmap_hashtable *h, char *name, unsigned len) +idmap_alloc_name(struct idmap_hashtable *h, char *name, size_t len) { return idmap_name_hash(h, name, len); } @@ -285,7 +286,7 @@ nfs_idmap_id(struct idmap *idmap, struct idmap_hashtable *h, memset(im, 0, sizeof(*im)); mutex_unlock(&idmap->idmap_im_lock); mutex_unlock(&idmap->idmap_lock); - return (ret); + return ret; } /* @@ -308,7 +309,7 @@ nfs_idmap_name(struct idmap *idmap, struct idmap_hashtable *h, mutex_lock(&idmap->idmap_im_lock); he = idmap_lookup_id(h, id); - if (he != 0) { + if (he) { memcpy(name, he->ih_name, he->ih_namelen); ret = he->ih_namelen; goto out; @@ -354,42 +355,40 @@ nfs_idmap_name(struct idmap *idmap, struct idmap_hashtable *h, /* RPC pipefs upcall/downcall routines */ static ssize_t idmap_pipe_upcall(struct file *filp, struct rpc_pipe_msg *msg, - char __user *dst, size_t buflen) + char __user *dst, size_t buflen) { - char *data = (char *)msg->data + msg->copied; - ssize_t mlen = msg->len - msg->copied; - ssize_t left; - - if (mlen > buflen) - mlen = buflen; - - left = copy_to_user(dst, data, mlen); - if (left < 0) { - msg->errno = left; - return left; + char *data = (char *)msg->data + msg->copied; + size_t mlen = min(msg->len, buflen); + unsigned long left; + + left = copy_to_user(dst, data, mlen); + if (left == mlen) { + msg->errno = -EFAULT; + return -EFAULT; } + mlen -= left; msg->copied += mlen; msg->errno = 0; - return mlen; + return mlen; } static ssize_t idmap_pipe_downcall(struct file *filp, const char __user *src, size_t mlen) { - struct rpc_inode *rpci = RPC_I(filp->f_path.dentry->d_inode); + struct rpc_inode *rpci = RPC_I(filp->f_path.dentry->d_inode); struct idmap *idmap = (struct idmap *)rpci->private; struct idmap_msg im_in, *im = &idmap->idmap_im; struct idmap_hashtable *h; struct idmap_hashent *he = NULL; - int namelen_in; + size_t namelen_in; int ret; - if (mlen != sizeof(im_in)) - return (-ENOSPC); + if (mlen != sizeof(im_in)) + return -ENOSPC; - if (copy_from_user(&im_in, src, mlen) != 0) - return (-EFAULT); + if (copy_from_user(&im_in, src, mlen) != 0) + return -EFAULT; mutex_lock(&idmap->idmap_im_lock); @@ -487,7 +486,7 @@ static unsigned int fnvhash32(const void *buf, size_t buflen) hash ^= (unsigned int)*p; } - return (hash); + return hash; } int nfs_map_name_to_uid(struct nfs_client *clp, const char *name, size_t namelen, __u32 *uid) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index db5d96dc610..6f88d7c77ac 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -192,7 +192,7 @@ void nfs_invalidate_atime(struct inode *inode) */ static void nfs_invalidate_inode(struct inode *inode) { - set_bit(NFS_INO_STALE, &NFS_FLAGS(inode)); + set_bit(NFS_INO_STALE, &NFS_I(inode)->flags); nfs_zap_caches_locked(inode); } @@ -229,7 +229,7 @@ nfs_init_locked(struct inode *inode, void *opaque) struct nfs_find_desc *desc = (struct nfs_find_desc *)opaque; struct nfs_fattr *fattr = desc->fattr; - NFS_FILEID(inode) = fattr->fileid; + set_nfs_fileid(inode, fattr->fileid); nfs_copy_fh(NFS_FH(inode), desc->fh); return 0; } @@ -291,7 +291,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) inode->i_fop = &nfs_dir_operations; if (nfs_server_capable(inode, NFS_CAP_READDIRPLUS) && fattr->size <= NFS_LIMIT_READDIRPLUS) - set_bit(NFS_INO_ADVISE_RDPLUS, &NFS_FLAGS(inode)); + set_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(inode)->flags); /* Deal with crossing mountpoints */ if (!nfs_fsid_equal(&NFS_SB(sb)->fsid, &fattr->fsid)) { if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) @@ -299,6 +299,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) else inode->i_op = &nfs_mountpoint_inode_operations; inode->i_fop = NULL; + set_bit(NFS_INO_MOUNTPOINT, &nfsi->flags); } } else if (S_ISLNK(inode->i_mode)) inode->i_op = &nfs_symlink_inode_operations; @@ -433,15 +434,11 @@ static int nfs_wait_schedule(void *word) */ static int nfs_wait_on_inode(struct inode *inode) { - struct rpc_clnt *clnt = NFS_CLIENT(inode); struct nfs_inode *nfsi = NFS_I(inode); - sigset_t oldmask; int error; - rpc_clnt_sigmask(clnt, &oldmask); error = wait_on_bit_lock(&nfsi->flags, NFS_INO_REVALIDATING, - nfs_wait_schedule, TASK_INTERRUPTIBLE); - rpc_clnt_sigunmask(clnt, &oldmask); + nfs_wait_schedule, TASK_KILLABLE); return error; } @@ -461,9 +458,18 @@ int nfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) int need_atime = NFS_I(inode)->cache_validity & NFS_INO_INVALID_ATIME; int err; - /* Flush out writes to the server in order to update c/mtime */ - if (S_ISREG(inode->i_mode)) + /* + * Flush out writes to the server in order to update c/mtime. + * + * Hold the i_mutex to suspend application writes temporarily; + * this prevents long-running writing applications from blocking + * nfs_wb_nocommit. + */ + if (S_ISREG(inode->i_mode)) { + mutex_lock(&inode->i_mutex); nfs_wb_nocommit(inode); + mutex_unlock(&inode->i_mutex); + } /* * We may force a getattr if the user cares about atime. @@ -500,6 +506,7 @@ static struct nfs_open_context *alloc_nfs_open_context(struct vfsmount *mnt, str ctx->cred = get_rpccred(cred); ctx->state = NULL; ctx->lockowner = current->files; + ctx->flags = 0; ctx->error = 0; ctx->dir_cookie = 0; atomic_set(&ctx->count, 1); @@ -659,7 +666,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) if (status == -ESTALE) { nfs_zap_caches(inode); if (!S_ISDIR(inode->i_mode)) - set_bit(NFS_INO_STALE, &NFS_FLAGS(inode)); + set_bit(NFS_INO_STALE, &NFS_I(inode)->flags); } goto out; } @@ -814,8 +821,9 @@ static void nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr *fattr) if (S_ISDIR(inode->i_mode)) nfsi->cache_validity |= NFS_INO_INVALID_DATA; } - if (inode->i_size == fattr->pre_size && nfsi->npages == 0) - inode->i_size = fattr->size; + if (inode->i_size == nfs_size_to_loff_t(fattr->pre_size) && + nfsi->npages == 0) + inode->i_size = nfs_size_to_loff_t(fattr->size); } } @@ -997,8 +1005,9 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) server = NFS_SERVER(inode); /* Update the fsid? */ - if (S_ISDIR(inode->i_mode) - && !nfs_fsid_equal(&server->fsid, &fattr->fsid)) + if (S_ISDIR(inode->i_mode) && + !nfs_fsid_equal(&server->fsid, &fattr->fsid) && + !test_bit(NFS_INO_MOUNTPOINT, &nfsi->flags)) server->fsid = fattr->fsid; /* @@ -1019,7 +1028,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) dprintk("NFS: mtime change on server for file %s/%ld\n", inode->i_sb->s_id, inode->i_ino); invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA; - nfsi->cache_change_attribute = now; + if (S_ISDIR(inode->i_mode)) + nfs_force_lookup_revalidate(inode); } /* If ctime has changed we should definitely clear access+acl caches */ if (!timespec_equal(&inode->i_ctime, &fattr->ctime)) @@ -1028,7 +1038,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) dprintk("NFS: change_attr change on server for file %s/%ld\n", inode->i_sb->s_id, inode->i_ino); invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; - nfsi->cache_change_attribute = now; + if (S_ISDIR(inode->i_mode)) + nfs_force_lookup_revalidate(inode); } /* Check if our cached file size is stale */ @@ -1133,7 +1144,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) void nfs4_clear_inode(struct inode *inode) { /* If we are holding a delegation, return it! */ - nfs_inode_return_delegation(inode); + nfs_inode_return_delegation_noreclaim(inode); /* First call standard NFS clear_inode() code */ nfs_clear_inode(inode); } diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index f3acf48412b..931992763e6 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -3,6 +3,7 @@ */ #include <linux/mount.h> +#include <linux/security.h> struct nfs_string; @@ -21,7 +22,8 @@ struct nfs_clone_mount { struct nfs_fattr *fattr; char *hostname; char *mnt_path; - struct sockaddr_in *addr; + struct sockaddr *addr; + size_t addrlen; rpc_authflavor_t authflavor; }; @@ -41,28 +43,31 @@ struct nfs_parsed_mount_data { char *client_address; struct { - struct sockaddr_in address; + struct sockaddr_storage address; + size_t addrlen; char *hostname; - unsigned int program; unsigned int version; unsigned short port; int protocol; } mount_server; struct { - struct sockaddr_in address; + struct sockaddr_storage address; + size_t addrlen; char *hostname; char *export_path; - unsigned int program; int protocol; } nfs_server; + + struct security_mnt_opts lsm_opts; }; /* client.c */ extern struct rpc_program nfs_program; extern void nfs_put_client(struct nfs_client *); -extern struct nfs_client *nfs_find_client(const struct sockaddr_in *, int); +extern struct nfs_client *nfs_find_client(const struct sockaddr *, u32); +extern struct nfs_client *nfs_find_client_next(struct nfs_client *); extern struct nfs_server *nfs_create_server( const struct nfs_parsed_mount_data *, struct nfs_fh *); @@ -160,6 +165,8 @@ extern struct rpc_stat nfs_rpcstat; extern int __init register_nfs_fs(void); extern void __exit unregister_nfs_fs(void); +extern void nfs_sb_active(struct nfs_server *server); +extern void nfs_sb_deactive(struct nfs_server *server); /* namespace.c */ extern char *nfs_path(const char *base, diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c index 8afd9f7e7a9..49c7cd0502c 100644 --- a/fs/nfs/mount_clnt.c +++ b/fs/nfs/mount_clnt.c @@ -56,7 +56,7 @@ int nfs_mount(struct sockaddr *addr, size_t len, char *hostname, char *path, .program = &mnt_program, .version = version, .authflavor = RPC_AUTH_UNIX, - .flags = RPC_CLNT_CREATE_INTR, + .flags = 0, }; struct rpc_clnt *mnt_clnt; int status; diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c index acfc56f9edc..607f6eb9cdb 100644 --- a/fs/nfs/namespace.c +++ b/fs/nfs/namespace.c @@ -107,38 +107,40 @@ static void * nfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd) BUG_ON(IS_ROOT(dentry)); dprintk("%s: enter\n", __FUNCTION__); - dput(nd->dentry); - nd->dentry = dget(dentry); + dput(nd->path.dentry); + nd->path.dentry = dget(dentry); /* Look it up again */ - parent = dget_parent(nd->dentry); + parent = dget_parent(nd->path.dentry); err = server->nfs_client->rpc_ops->lookup(parent->d_inode, - &nd->dentry->d_name, + &nd->path.dentry->d_name, &fh, &fattr); dput(parent); if (err != 0) goto out_err; if (fattr.valid & NFS_ATTR_FATTR_V4_REFERRAL) - mnt = nfs_do_refmount(nd->mnt, nd->dentry); + mnt = nfs_do_refmount(nd->path.mnt, nd->path.dentry); else - mnt = nfs_do_submount(nd->mnt, nd->dentry, &fh, &fattr); + mnt = nfs_do_submount(nd->path.mnt, nd->path.dentry, &fh, + &fattr); err = PTR_ERR(mnt); if (IS_ERR(mnt)) goto out_err; mntget(mnt); - err = do_add_mount(mnt, nd, nd->mnt->mnt_flags|MNT_SHRINKABLE, &nfs_automount_list); + err = do_add_mount(mnt, nd, nd->path.mnt->mnt_flags|MNT_SHRINKABLE, + &nfs_automount_list); if (err < 0) { mntput(mnt); if (err == -EBUSY) goto out_follow; goto out_err; } - mntput(nd->mnt); - dput(nd->dentry); - nd->mnt = mnt; - nd->dentry = dget(mnt->mnt_root); + mntput(nd->path.mnt); + dput(nd->path.dentry); + nd->path.mnt = mnt; + nd->path.dentry = dget(mnt->mnt_root); schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout); out: dprintk("%s: done, returned %d\n", __FUNCTION__, err); @@ -146,10 +148,11 @@ out: dprintk("<-- nfs_follow_mountpoint() = %d\n", err); return ERR_PTR(err); out_err: - path_release(nd); + path_put(&nd->path); goto out; out_follow: - while(d_mountpoint(nd->dentry) && follow_down(&nd->mnt, &nd->dentry)) + while (d_mountpoint(nd->path.dentry) && + follow_down(&nd->path.mnt, &nd->path.dentry)) ; err = 0; goto out; @@ -188,7 +191,7 @@ static struct vfsmount *nfs_do_clone_mount(struct nfs_server *server, { #ifdef CONFIG_NFS_V4 struct vfsmount *mnt = NULL; - switch (server->nfs_client->cl_nfsversion) { + switch (server->nfs_client->rpc_ops->version) { case 2: case 3: mnt = vfs_kern_mount(&nfs_xdev_fs_type, 0, devname, mountdata); diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c index 668ab96c7b5..1f7ea675e0c 100644 --- a/fs/nfs/nfs2xdr.c +++ b/fs/nfs/nfs2xdr.c @@ -262,7 +262,9 @@ static int nfs_xdr_readres(struct rpc_rqst *req, __be32 *p, struct nfs_readres *res) { struct kvec *iov = req->rq_rcv_buf.head; - int status, count, recvd, hdrlen; + size_t hdrlen; + u32 count, recvd; + int status; if ((status = ntohl(*p++))) return -nfs_stat_to_errno(status); @@ -273,7 +275,7 @@ nfs_xdr_readres(struct rpc_rqst *req, __be32 *p, struct nfs_readres *res) hdrlen = (u8 *) p - (u8 *) iov->iov_base; if (iov->iov_len < hdrlen) { dprintk("NFS: READ reply header overflowed:" - "length %d > %Zu\n", hdrlen, iov->iov_len); + "length %Zu > %Zu\n", hdrlen, iov->iov_len); return -errno_NFSERR_IO; } else if (iov->iov_len != hdrlen) { dprintk("NFS: READ header is short. iovec will be shifted.\n"); @@ -283,11 +285,11 @@ nfs_xdr_readres(struct rpc_rqst *req, __be32 *p, struct nfs_readres *res) recvd = req->rq_rcv_buf.len - hdrlen; if (count > recvd) { dprintk("NFS: server cheating in read reply: " - "count %d > recvd %d\n", count, recvd); + "count %u > recvd %u\n", count, recvd); count = recvd; } - dprintk("RPC: readres OK count %d\n", count); + dprintk("RPC: readres OK count %u\n", count); if (count < res->count) res->count = count; @@ -423,9 +425,10 @@ nfs_xdr_readdirres(struct rpc_rqst *req, __be32 *p, void *dummy) struct xdr_buf *rcvbuf = &req->rq_rcv_buf; struct kvec *iov = rcvbuf->head; struct page **page; - int hdrlen, recvd; + size_t hdrlen; + unsigned int pglen, recvd; + u32 len; int status, nr; - unsigned int len, pglen; __be32 *end, *entry, *kaddr; if ((status = ntohl(*p++))) @@ -434,7 +437,7 @@ nfs_xdr_readdirres(struct rpc_rqst *req, __be32 *p, void *dummy) hdrlen = (u8 *) p - (u8 *) iov->iov_base; if (iov->iov_len < hdrlen) { dprintk("NFS: READDIR reply header overflowed:" - "length %d > %Zu\n", hdrlen, iov->iov_len); + "length %Zu > %Zu\n", hdrlen, iov->iov_len); return -errno_NFSERR_IO; } else if (iov->iov_len != hdrlen) { dprintk("NFS: READDIR header is short. iovec will be shifted.\n"); @@ -576,7 +579,8 @@ nfs_xdr_readlinkres(struct rpc_rqst *req, __be32 *p, void *dummy) { struct xdr_buf *rcvbuf = &req->rq_rcv_buf; struct kvec *iov = rcvbuf->head; - int hdrlen, len, recvd; + size_t hdrlen; + u32 len, recvd; char *kaddr; int status; @@ -584,14 +588,14 @@ nfs_xdr_readlinkres(struct rpc_rqst *req, __be32 *p, void *dummy) return -nfs_stat_to_errno(status); /* Convert length of symlink */ len = ntohl(*p++); - if (len >= rcvbuf->page_len || len <= 0) { + if (len >= rcvbuf->page_len) { dprintk("nfs: server returned giant symlink!\n"); return -ENAMETOOLONG; } hdrlen = (u8 *) p - (u8 *) iov->iov_base; if (iov->iov_len < hdrlen) { dprintk("NFS: READLINK reply header overflowed:" - "length %d > %Zu\n", hdrlen, iov->iov_len); + "length %Zu > %Zu\n", hdrlen, iov->iov_len); return -errno_NFSERR_IO; } else if (iov->iov_len != hdrlen) { dprintk("NFS: READLINK header is short. iovec will be shifted.\n"); diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 4cdc2361a66..549dbce714a 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -27,17 +27,14 @@ static int nfs3_rpc_wrapper(struct rpc_clnt *clnt, struct rpc_message *msg, int flags) { - sigset_t oldset; int res; - rpc_clnt_sigmask(clnt, &oldset); do { res = rpc_call_sync(clnt, msg, flags); if (res != -EJUKEBOX) break; - schedule_timeout_interruptible(NFS_JUKEBOX_RETRY_TIME); + schedule_timeout_killable(NFS_JUKEBOX_RETRY_TIME); res = -ERESTARTSYS; - } while (!signalled()); - rpc_clnt_sigunmask(clnt, &oldset); + } while (!fatal_signal_pending(current)); return res; } @@ -732,16 +729,9 @@ static int nfs3_read_done(struct rpc_task *task, struct nfs_read_data *data) return 0; } -static void nfs3_proc_read_setup(struct nfs_read_data *data) +static void nfs3_proc_read_setup(struct nfs_read_data *data, struct rpc_message *msg) { - struct rpc_message msg = { - .rpc_proc = &nfs3_procedures[NFS3PROC_READ], - .rpc_argp = &data->args, - .rpc_resp = &data->res, - .rpc_cred = data->cred, - }; - - rpc_call_setup(&data->task, &msg, 0); + msg->rpc_proc = &nfs3_procedures[NFS3PROC_READ]; } static int nfs3_write_done(struct rpc_task *task, struct nfs_write_data *data) @@ -753,24 +743,9 @@ static int nfs3_write_done(struct rpc_task *task, struct nfs_write_data *data) return 0; } -static void nfs3_proc_write_setup(struct nfs_write_data *data, int how) +static void nfs3_proc_write_setup(struct nfs_write_data *data, struct rpc_message *msg) { - struct rpc_message msg = { - .rpc_proc = &nfs3_procedures[NFS3PROC_WRITE], - .rpc_argp = &data->args, - .rpc_resp = &data->res, - .rpc_cred = data->cred, - }; - - data->args.stable = NFS_UNSTABLE; - if (how & FLUSH_STABLE) { - data->args.stable = NFS_FILE_SYNC; - if (NFS_I(data->inode)->ncommit) - data->args.stable = NFS_DATA_SYNC; - } - - /* Finalize the task. */ - rpc_call_setup(&data->task, &msg, 0); + msg->rpc_proc = &nfs3_procedures[NFS3PROC_WRITE]; } static int nfs3_commit_done(struct rpc_task *task, struct nfs_write_data *data) @@ -781,22 +756,17 @@ static int nfs3_commit_done(struct rpc_task *task, struct nfs_write_data *data) return 0; } -static void nfs3_proc_commit_setup(struct nfs_write_data *data, int how) +static void nfs3_proc_commit_setup(struct nfs_write_data *data, struct rpc_message *msg) { - struct rpc_message msg = { - .rpc_proc = &nfs3_procedures[NFS3PROC_COMMIT], - .rpc_argp = &data->args, - .rpc_resp = &data->res, - .rpc_cred = data->cred, - }; - - rpc_call_setup(&data->task, &msg, 0); + msg->rpc_proc = &nfs3_procedures[NFS3PROC_COMMIT]; } static int nfs3_proc_lock(struct file *filp, int cmd, struct file_lock *fl) { - return nlmclnt_proc(filp->f_path.dentry->d_inode, cmd, fl); + struct inode *inode = filp->f_path.dentry->d_inode; + + return nlmclnt_proc(NFS_SERVER(inode)->nlm_host, cmd, fl); } const struct nfs_rpc_ops nfs_v3_clientops = { diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index 616d3267b7e..3917e2fa4e4 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -506,9 +506,9 @@ nfs3_xdr_readdirres(struct rpc_rqst *req, __be32 *p, struct nfs3_readdirres *res struct xdr_buf *rcvbuf = &req->rq_rcv_buf; struct kvec *iov = rcvbuf->head; struct page **page; - int hdrlen, recvd; + size_t hdrlen; + u32 len, recvd, pglen; int status, nr; - unsigned int len, pglen; __be32 *entry, *end, *kaddr; status = ntohl(*p++); @@ -527,7 +527,7 @@ nfs3_xdr_readdirres(struct rpc_rqst *req, __be32 *p, struct nfs3_readdirres *res hdrlen = (u8 *) p - (u8 *) iov->iov_base; if (iov->iov_len < hdrlen) { dprintk("NFS: READDIR reply header overflowed:" - "length %d > %Zu\n", hdrlen, iov->iov_len); + "length %Zu > %Zu\n", hdrlen, iov->iov_len); return -errno_NFSERR_IO; } else if (iov->iov_len != hdrlen) { dprintk("NFS: READDIR header is short. iovec will be shifted.\n"); @@ -549,7 +549,7 @@ nfs3_xdr_readdirres(struct rpc_rqst *req, __be32 *p, struct nfs3_readdirres *res len = ntohl(*p++); /* string length */ p += XDR_QUADLEN(len) + 2; /* name + cookie */ if (len > NFS3_MAXNAMLEN) { - dprintk("NFS: giant filename in readdir (len %x)!\n", + dprintk("NFS: giant filename in readdir (len 0x%x)!\n", len); goto err_unmap; } @@ -570,7 +570,7 @@ nfs3_xdr_readdirres(struct rpc_rqst *req, __be32 *p, struct nfs3_readdirres *res len = ntohl(*p++); if (len > NFS3_FHSIZE) { dprintk("NFS: giant filehandle in " - "readdir (len %x)!\n", len); + "readdir (len 0x%x)!\n", len); goto err_unmap; } p += XDR_QUADLEN(len); @@ -815,7 +815,8 @@ nfs3_xdr_readlinkres(struct rpc_rqst *req, __be32 *p, struct nfs_fattr *fattr) { struct xdr_buf *rcvbuf = &req->rq_rcv_buf; struct kvec *iov = rcvbuf->head; - int hdrlen, len, recvd; + size_t hdrlen; + u32 len, recvd; char *kaddr; int status; @@ -827,7 +828,7 @@ nfs3_xdr_readlinkres(struct rpc_rqst *req, __be32 *p, struct nfs_fattr *fattr) /* Convert length of symlink */ len = ntohl(*p++); - if (len >= rcvbuf->page_len || len <= 0) { + if (len >= rcvbuf->page_len) { dprintk("nfs: server returned giant symlink!\n"); return -ENAMETOOLONG; } @@ -835,7 +836,7 @@ nfs3_xdr_readlinkres(struct rpc_rqst *req, __be32 *p, struct nfs_fattr *fattr) hdrlen = (u8 *) p - (u8 *) iov->iov_base; if (iov->iov_len < hdrlen) { dprintk("NFS: READLINK reply header overflowed:" - "length %d > %Zu\n", hdrlen, iov->iov_len); + "length %Zu > %Zu\n", hdrlen, iov->iov_len); return -errno_NFSERR_IO; } else if (iov->iov_len != hdrlen) { dprintk("NFS: READLINK header is short. " @@ -863,7 +864,9 @@ static int nfs3_xdr_readres(struct rpc_rqst *req, __be32 *p, struct nfs_readres *res) { struct kvec *iov = req->rq_rcv_buf.head; - int status, count, ocount, recvd, hdrlen; + size_t hdrlen; + u32 count, ocount, recvd; + int status; status = ntohl(*p++); p = xdr_decode_post_op_attr(p, res->fattr); @@ -871,7 +874,7 @@ nfs3_xdr_readres(struct rpc_rqst *req, __be32 *p, struct nfs_readres *res) if (status != 0) return -nfs_stat_to_errno(status); - /* Decode reply could and EOF flag. NFSv3 is somewhat redundant + /* Decode reply count and EOF flag. NFSv3 is somewhat redundant * in that it puts the count both in the res struct and in the * opaque data count. */ count = ntohl(*p++); @@ -886,7 +889,7 @@ nfs3_xdr_readres(struct rpc_rqst *req, __be32 *p, struct nfs_readres *res) hdrlen = (u8 *) p - (u8 *) iov->iov_base; if (iov->iov_len < hdrlen) { dprintk("NFS: READ reply header overflowed:" - "length %d > %Zu\n", hdrlen, iov->iov_len); + "length %Zu > %Zu\n", hdrlen, iov->iov_len); return -errno_NFSERR_IO; } else if (iov->iov_len != hdrlen) { dprintk("NFS: READ header is short. iovec will be shifted.\n"); @@ -896,7 +899,7 @@ nfs3_xdr_readres(struct rpc_rqst *req, __be32 *p, struct nfs_readres *res) recvd = req->rq_rcv_buf.len - hdrlen; if (count > recvd) { dprintk("NFS: server cheating in read reply: " - "count %d > recvd %d\n", count, recvd); + "count %u > recvd %u\n", count, recvd); count = recvd; res->eof = 0; } diff --git a/fs/nfs/nfs4namespace.c b/fs/nfs/nfs4namespace.c index dd5fef20c70..5f9ba41ed5b 100644 --- a/fs/nfs/nfs4namespace.c +++ b/fs/nfs/nfs4namespace.c @@ -114,10 +114,7 @@ static inline int valid_ipaddr4(const char *buf) * nfs_follow_referral - set up mountpoint when hitting a referral on moved error * @mnt_parent - mountpoint of parent directory * @dentry - parent directory - * @fspath - fs path returned in fs_locations - * @mntpath - mount path to new server - * @hostname - hostname of new server - * @addr - host addr of new server + * @locations - array of NFSv4 server location information * */ static struct vfsmount *nfs_follow_referral(const struct vfsmount *mnt_parent, @@ -131,7 +128,8 @@ static struct vfsmount *nfs_follow_referral(const struct vfsmount *mnt_parent, .authflavor = NFS_SB(mnt_parent->mnt_sb)->client->cl_auth->au_flavor, }; char *page = NULL, *page2 = NULL; - int loc, s, error; + unsigned int s; + int loc, error; if (locations == NULL || locations->nlocations <= 0) goto out; @@ -174,7 +172,10 @@ static struct vfsmount *nfs_follow_referral(const struct vfsmount *mnt_parent, s = 0; while (s < location->nservers) { - struct sockaddr_in addr = {}; + struct sockaddr_in addr = { + .sin_family = AF_INET, + .sin_port = htons(NFS_PORT), + }; if (location->servers[s].len <= 0 || valid_ipaddr4(location->servers[s].data) < 0) { @@ -183,10 +184,9 @@ static struct vfsmount *nfs_follow_referral(const struct vfsmount *mnt_parent, } mountdata.hostname = location->servers[s].data; - addr.sin_addr.s_addr = in_aton(mountdata.hostname); - addr.sin_family = AF_INET; - addr.sin_port = htons(NFS_PORT); - mountdata.addr = &addr; + addr.sin_addr.s_addr = in_aton(mountdata.hostname), + mountdata.addr = (struct sockaddr *)&addr; + mountdata.addrlen = sizeof(addr); snprintf(page, PAGE_SIZE, "%s:%s", mountdata.hostname, diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 9e2e1c7291d..7ce07862c2f 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -210,7 +210,7 @@ static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo) spin_lock(&dir->i_lock); nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE|NFS_INO_INVALID_DATA; if (!cinfo->atomic || cinfo->before != nfsi->change_attr) - nfsi->cache_change_attribute = jiffies; + nfs_force_lookup_revalidate(dir); nfsi->change_attr = cinfo->after; spin_unlock(&dir->i_lock); } @@ -316,12 +316,9 @@ static void nfs4_opendata_put(struct nfs4_opendata *p) static int nfs4_wait_for_completion_rpc_task(struct rpc_task *task) { - sigset_t oldset; int ret; - rpc_clnt_sigmask(task->tk_client, &oldset); ret = rpc_wait_for_completion_task(task); - rpc_clnt_sigunmask(task->tk_client, &oldset); return ret; } @@ -718,19 +715,6 @@ int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state return err; } -static void nfs4_open_confirm_prepare(struct rpc_task *task, void *calldata) -{ - struct nfs4_opendata *data = calldata; - struct rpc_message msg = { - .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_CONFIRM], - .rpc_argp = &data->c_arg, - .rpc_resp = &data->c_res, - .rpc_cred = data->owner->so_cred, - }; - data->timestamp = jiffies; - rpc_call_setup(task, &msg, 0); -} - static void nfs4_open_confirm_done(struct rpc_task *task, void *calldata) { struct nfs4_opendata *data = calldata; @@ -767,7 +751,6 @@ out_free: } static const struct rpc_call_ops nfs4_open_confirm_ops = { - .rpc_call_prepare = nfs4_open_confirm_prepare, .rpc_call_done = nfs4_open_confirm_done, .rpc_release = nfs4_open_confirm_release, }; @@ -779,12 +762,26 @@ static int _nfs4_proc_open_confirm(struct nfs4_opendata *data) { struct nfs_server *server = NFS_SERVER(data->dir->d_inode); struct rpc_task *task; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_CONFIRM], + .rpc_argp = &data->c_arg, + .rpc_resp = &data->c_res, + .rpc_cred = data->owner->so_cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = server->client, + .rpc_message = &msg, + .callback_ops = &nfs4_open_confirm_ops, + .callback_data = data, + .flags = RPC_TASK_ASYNC, + }; int status; kref_get(&data->kref); data->rpc_done = 0; data->rpc_status = 0; - task = rpc_run_task(server->client, RPC_TASK_ASYNC, &nfs4_open_confirm_ops, data); + data->timestamp = jiffies; + task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) return PTR_ERR(task); status = nfs4_wait_for_completion_rpc_task(task); @@ -801,13 +798,7 @@ static void nfs4_open_prepare(struct rpc_task *task, void *calldata) { struct nfs4_opendata *data = calldata; struct nfs4_state_owner *sp = data->owner; - struct rpc_message msg = { - .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN], - .rpc_argp = &data->o_arg, - .rpc_resp = &data->o_res, - .rpc_cred = sp->so_cred, - }; - + if (nfs_wait_on_sequence(data->o_arg.seqid, task) != 0) return; /* @@ -832,11 +823,11 @@ static void nfs4_open_prepare(struct rpc_task *task, void *calldata) data->o_arg.id = sp->so_owner_id.id; data->o_arg.clientid = sp->so_client->cl_clientid; if (data->o_arg.claim == NFS4_OPEN_CLAIM_PREVIOUS) { - msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_NOATTR]; + task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_NOATTR]; nfs_copy_fh(&data->o_res.fh, data->o_arg.fh); } data->timestamp = jiffies; - rpc_call_setup(task, &msg, 0); + rpc_call_start(task); return; out_no_action: task->tk_action = NULL; @@ -908,13 +899,26 @@ static int _nfs4_proc_open(struct nfs4_opendata *data) struct nfs_openargs *o_arg = &data->o_arg; struct nfs_openres *o_res = &data->o_res; struct rpc_task *task; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN], + .rpc_argp = o_arg, + .rpc_resp = o_res, + .rpc_cred = data->owner->so_cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = server->client, + .rpc_message = &msg, + .callback_ops = &nfs4_open_ops, + .callback_data = data, + .flags = RPC_TASK_ASYNC, + }; int status; kref_get(&data->kref); data->rpc_done = 0; data->rpc_status = 0; data->cancelled = 0; - task = rpc_run_task(server->client, RPC_TASK_ASYNC, &nfs4_open_ops, data); + task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) return PTR_ERR(task); status = nfs4_wait_for_completion_rpc_task(task); @@ -1244,12 +1248,6 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data) { struct nfs4_closedata *calldata = data; struct nfs4_state *state = calldata->state; - struct rpc_message msg = { - .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CLOSE], - .rpc_argp = &calldata->arg, - .rpc_resp = &calldata->res, - .rpc_cred = state->owner->so_cred, - }; int clear_rd, clear_wr, clear_rdwr; if (nfs_wait_on_sequence(calldata->arg.seqid, task) != 0) @@ -1276,14 +1274,14 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data) } nfs_fattr_init(calldata->res.fattr); if (test_bit(NFS_O_RDONLY_STATE, &state->flags) != 0) { - msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE]; + task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE]; calldata->arg.open_flags = FMODE_READ; } else if (test_bit(NFS_O_WRONLY_STATE, &state->flags) != 0) { - msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE]; + task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE]; calldata->arg.open_flags = FMODE_WRITE; } calldata->timestamp = jiffies; - rpc_call_setup(task, &msg, 0); + rpc_call_start(task); } static const struct rpc_call_ops nfs4_close_ops = { @@ -1309,6 +1307,16 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait) struct nfs4_closedata *calldata; struct nfs4_state_owner *sp = state->owner; struct rpc_task *task; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CLOSE], + .rpc_cred = state->owner->so_cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = server->client, + .rpc_message = &msg, + .callback_ops = &nfs4_close_ops, + .flags = RPC_TASK_ASYNC, + }; int status = -ENOMEM; calldata = kmalloc(sizeof(*calldata), GFP_KERNEL); @@ -1328,7 +1336,10 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait) calldata->path.mnt = mntget(path->mnt); calldata->path.dentry = dget(path->dentry); - task = rpc_run_task(server->client, RPC_TASK_ASYNC, &nfs4_close_ops, calldata); + msg.rpc_argp = &calldata->arg, + msg.rpc_resp = &calldata->res, + task_setup_data.callback_data = calldata; + task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) return PTR_ERR(task); status = 0; @@ -1373,11 +1384,11 @@ out_close: struct dentry * nfs4_atomic_open(struct inode *dir, struct dentry *dentry, struct nameidata *nd) { - struct dentry *parent; struct path path = { - .mnt = nd->mnt, + .mnt = nd->path.mnt, .dentry = dentry, }; + struct dentry *parent; struct iattr attr; struct rpc_cred *cred; struct nfs4_state *state; @@ -1422,7 +1433,7 @@ int nfs4_open_revalidate(struct inode *dir, struct dentry *dentry, int openflags, struct nameidata *nd) { struct path path = { - .mnt = nd->mnt, + .mnt = nd->path.mnt, .dentry = dentry, }; struct rpc_cred *cred; @@ -1874,7 +1885,7 @@ nfs4_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, int flags, struct nameidata *nd) { struct path path = { - .mnt = nd->mnt, + .mnt = nd->path.mnt, .dentry = dentry, }; struct nfs4_state *state; @@ -2414,18 +2425,10 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data) return 0; } -static void nfs4_proc_read_setup(struct nfs_read_data *data) +static void nfs4_proc_read_setup(struct nfs_read_data *data, struct rpc_message *msg) { - struct rpc_message msg = { - .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ], - .rpc_argp = &data->args, - .rpc_resp = &data->res, - .rpc_cred = data->cred, - }; - data->timestamp = jiffies; - - rpc_call_setup(&data->task, &msg, 0); + msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ]; } static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data) @@ -2443,33 +2446,15 @@ static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data) return 0; } -static void nfs4_proc_write_setup(struct nfs_write_data *data, int how) +static void nfs4_proc_write_setup(struct nfs_write_data *data, struct rpc_message *msg) { - struct rpc_message msg = { - .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_WRITE], - .rpc_argp = &data->args, - .rpc_resp = &data->res, - .rpc_cred = data->cred, - }; - struct inode *inode = data->inode; - struct nfs_server *server = NFS_SERVER(inode); - int stable; - - if (how & FLUSH_STABLE) { - if (!NFS_I(inode)->ncommit) - stable = NFS_FILE_SYNC; - else - stable = NFS_DATA_SYNC; - } else - stable = NFS_UNSTABLE; - data->args.stable = stable; + struct nfs_server *server = NFS_SERVER(data->inode); + data->args.bitmask = server->attr_bitmask; data->res.server = server; - data->timestamp = jiffies; - /* Finalize the task. */ - rpc_call_setup(&data->task, &msg, 0); + msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_WRITE]; } static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data) @@ -2484,20 +2469,13 @@ static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data) return 0; } -static void nfs4_proc_commit_setup(struct nfs_write_data *data, int how) +static void nfs4_proc_commit_setup(struct nfs_write_data *data, struct rpc_message *msg) { - struct rpc_message msg = { - .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COMMIT], - .rpc_argp = &data->args, - .rpc_resp = &data->res, - .rpc_cred = data->cred, - }; struct nfs_server *server = NFS_SERVER(data->inode); data->args.bitmask = server->attr_bitmask; data->res.server = server; - - rpc_call_setup(&data->task, &msg, 0); + msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COMMIT]; } /* @@ -2804,9 +2782,9 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server) return 0; } -static int nfs4_wait_bit_interruptible(void *word) +static int nfs4_wait_bit_killable(void *word) { - if (signal_pending(current)) + if (fatal_signal_pending(current)) return -ERESTARTSYS; schedule(); return 0; @@ -2814,18 +2792,14 @@ static int nfs4_wait_bit_interruptible(void *word) static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs_client *clp) { - sigset_t oldset; int res; might_sleep(); rwsem_acquire(&clp->cl_sem.dep_map, 0, 0, _RET_IP_); - rpc_clnt_sigmask(clnt, &oldset); res = wait_on_bit(&clp->cl_state, NFS4CLNT_STATE_RECOVER, - nfs4_wait_bit_interruptible, - TASK_INTERRUPTIBLE); - rpc_clnt_sigunmask(clnt, &oldset); + nfs4_wait_bit_killable, TASK_KILLABLE); rwsem_release(&clp->cl_sem.dep_map, 1, _RET_IP_); return res; @@ -2833,7 +2807,6 @@ static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs_client *clp) static int nfs4_delay(struct rpc_clnt *clnt, long *timeout) { - sigset_t oldset; int res = 0; might_sleep(); @@ -2842,14 +2815,9 @@ static int nfs4_delay(struct rpc_clnt *clnt, long *timeout) *timeout = NFS4_POLL_RETRY_MIN; if (*timeout > NFS4_POLL_RETRY_MAX) *timeout = NFS4_POLL_RETRY_MAX; - rpc_clnt_sigmask(clnt, &oldset); - if (clnt->cl_intr) { - schedule_timeout_interruptible(*timeout); - if (signalled()) - res = -ERESTARTSYS; - } else - schedule_timeout_uninterruptible(*timeout); - rpc_clnt_sigunmask(clnt, &oldset); + schedule_timeout_killable(*timeout); + if (fatal_signal_pending(current)) + res = -ERESTARTSYS; *timeout <<= 1; return res; } @@ -2910,14 +2878,20 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, unsigned short po for(;;) { setclientid.sc_name_len = scnprintf(setclientid.sc_name, - sizeof(setclientid.sc_name), "%s/%u.%u.%u.%u %s %u", - clp->cl_ipaddr, NIPQUAD(clp->cl_addr.sin_addr), + sizeof(setclientid.sc_name), "%s/%s %s %s %u", + clp->cl_ipaddr, + rpc_peeraddr2str(clp->cl_rpcclient, + RPC_DISPLAY_ADDR), + rpc_peeraddr2str(clp->cl_rpcclient, + RPC_DISPLAY_PROTO), cred->cr_ops->cr_name, clp->cl_id_uniquifier); setclientid.sc_netid_len = scnprintf(setclientid.sc_netid, - sizeof(setclientid.sc_netid), "tcp"); + sizeof(setclientid.sc_netid), + rpc_peeraddr2str(clp->cl_rpcclient, + RPC_DISPLAY_NETID)); setclientid.sc_uaddr_len = scnprintf(setclientid.sc_uaddr, - sizeof(setclientid.sc_uaddr), "%s.%d.%d", + sizeof(setclientid.sc_uaddr), "%s.%u.%u", clp->cl_ipaddr, port >> 8, port & 255); status = rpc_call_sync(clp->cl_rpcclient, &msg, 0); @@ -2981,25 +2955,11 @@ struct nfs4_delegreturndata { struct nfs4_delegreturnres res; struct nfs_fh fh; nfs4_stateid stateid; - struct rpc_cred *cred; unsigned long timestamp; struct nfs_fattr fattr; int rpc_status; }; -static void nfs4_delegreturn_prepare(struct rpc_task *task, void *calldata) -{ - struct nfs4_delegreturndata *data = calldata; - struct rpc_message msg = { - .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_DELEGRETURN], - .rpc_argp = &data->args, - .rpc_resp = &data->res, - .rpc_cred = data->cred, - }; - nfs_fattr_init(data->res.fattr); - rpc_call_setup(task, &msg, 0); -} - static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata) { struct nfs4_delegreturndata *data = calldata; @@ -3010,24 +2970,30 @@ static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata) static void nfs4_delegreturn_release(void *calldata) { - struct nfs4_delegreturndata *data = calldata; - - put_rpccred(data->cred); kfree(calldata); } static const struct rpc_call_ops nfs4_delegreturn_ops = { - .rpc_call_prepare = nfs4_delegreturn_prepare, .rpc_call_done = nfs4_delegreturn_done, .rpc_release = nfs4_delegreturn_release, }; -static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4_stateid *stateid) +static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4_stateid *stateid, int issync) { struct nfs4_delegreturndata *data; struct nfs_server *server = NFS_SERVER(inode); struct rpc_task *task; - int status; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_DELEGRETURN], + .rpc_cred = cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = server->client, + .rpc_message = &msg, + .callback_ops = &nfs4_delegreturn_ops, + .flags = RPC_TASK_ASYNC, + }; + int status = 0; data = kmalloc(sizeof(*data), GFP_KERNEL); if (data == NULL) @@ -3039,30 +3005,37 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co memcpy(&data->stateid, stateid, sizeof(data->stateid)); data->res.fattr = &data->fattr; data->res.server = server; - data->cred = get_rpccred(cred); + nfs_fattr_init(data->res.fattr); data->timestamp = jiffies; data->rpc_status = 0; - task = rpc_run_task(NFS_CLIENT(inode), RPC_TASK_ASYNC, &nfs4_delegreturn_ops, data); + task_setup_data.callback_data = data; + msg.rpc_argp = &data->args, + msg.rpc_resp = &data->res, + task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) return PTR_ERR(task); + if (!issync) + goto out; status = nfs4_wait_for_completion_rpc_task(task); - if (status == 0) { - status = data->rpc_status; - if (status == 0) - nfs_refresh_inode(inode, &data->fattr); - } + if (status != 0) + goto out; + status = data->rpc_status; + if (status != 0) + goto out; + nfs_refresh_inode(inode, &data->fattr); +out: rpc_put_task(task); return status; } -int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4_stateid *stateid) +int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4_stateid *stateid, int issync) { struct nfs_server *server = NFS_SERVER(inode); struct nfs4_exception exception = { }; int err; do { - err = _nfs4_proc_delegreturn(inode, cred, stateid); + err = _nfs4_proc_delegreturn(inode, cred, stateid, issync); switch (err) { case -NFS4ERR_STALE_STATEID: case -NFS4ERR_EXPIRED: @@ -3083,7 +3056,7 @@ int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4 static unsigned long nfs4_set_lock_task_retry(unsigned long timeout) { - schedule_timeout_interruptible(timeout); + schedule_timeout_killable(timeout); timeout <<= 1; if (timeout > NFS4_LOCK_MAXTIMEOUT) return NFS4_LOCK_MAXTIMEOUT; @@ -3230,12 +3203,6 @@ static void nfs4_locku_done(struct rpc_task *task, void *data) static void nfs4_locku_prepare(struct rpc_task *task, void *data) { struct nfs4_unlockdata *calldata = data; - struct rpc_message msg = { - .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOCKU], - .rpc_argp = &calldata->arg, - .rpc_resp = &calldata->res, - .rpc_cred = calldata->lsp->ls_state->owner->so_cred, - }; if (nfs_wait_on_sequence(calldata->arg.seqid, task) != 0) return; @@ -3245,7 +3212,7 @@ static void nfs4_locku_prepare(struct rpc_task *task, void *data) return; } calldata->timestamp = jiffies; - rpc_call_setup(task, &msg, 0); + rpc_call_start(task); } static const struct rpc_call_ops nfs4_locku_ops = { @@ -3260,6 +3227,16 @@ static struct rpc_task *nfs4_do_unlck(struct file_lock *fl, struct nfs_seqid *seqid) { struct nfs4_unlockdata *data; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOCKU], + .rpc_cred = ctx->cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = NFS_CLIENT(lsp->ls_state->inode), + .rpc_message = &msg, + .callback_ops = &nfs4_locku_ops, + .flags = RPC_TASK_ASYNC, + }; /* Ensure this is an unlock - when canceling a lock, the * canceled lock is passed in, and it won't be an unlock. @@ -3272,7 +3249,10 @@ static struct rpc_task *nfs4_do_unlck(struct file_lock *fl, return ERR_PTR(-ENOMEM); } - return rpc_run_task(NFS_CLIENT(lsp->ls_state->inode), RPC_TASK_ASYNC, &nfs4_locku_ops, data); + msg.rpc_argp = &data->arg, + msg.rpc_resp = &data->res, + task_setup_data.callback_data = data; + return rpc_run_task(&task_setup_data); } static int nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock *request) @@ -3331,15 +3311,12 @@ static struct nfs4_lockdata *nfs4_alloc_lockdata(struct file_lock *fl, p->arg.fh = NFS_FH(inode); p->arg.fl = &p->fl; - if (!(lsp->ls_seqid.flags & NFS_SEQID_CONFIRMED)) { - p->arg.open_seqid = nfs_alloc_seqid(&lsp->ls_state->owner->so_seqid); - if (p->arg.open_seqid == NULL) - goto out_free; - - } + p->arg.open_seqid = nfs_alloc_seqid(&lsp->ls_state->owner->so_seqid); + if (p->arg.open_seqid == NULL) + goto out_free; p->arg.lock_seqid = nfs_alloc_seqid(&lsp->ls_seqid); if (p->arg.lock_seqid == NULL) - goto out_free; + goto out_free_seqid; p->arg.lock_stateid = &lsp->ls_stateid; p->arg.lock_owner.clientid = server->nfs_client->cl_clientid; p->arg.lock_owner.id = lsp->ls_id.id; @@ -3348,9 +3325,9 @@ static struct nfs4_lockdata *nfs4_alloc_lockdata(struct file_lock *fl, p->ctx = get_nfs_open_context(ctx); memcpy(&p->fl, fl, sizeof(p->fl)); return p; +out_free_seqid: + nfs_free_seqid(p->arg.open_seqid); out_free: - if (p->arg.open_seqid != NULL) - nfs_free_seqid(p->arg.open_seqid); kfree(p); return NULL; } @@ -3359,31 +3336,20 @@ static void nfs4_lock_prepare(struct rpc_task *task, void *calldata) { struct nfs4_lockdata *data = calldata; struct nfs4_state *state = data->lsp->ls_state; - struct nfs4_state_owner *sp = state->owner; - struct rpc_message msg = { - .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOCK], - .rpc_argp = &data->arg, - .rpc_resp = &data->res, - .rpc_cred = sp->so_cred, - }; dprintk("%s: begin!\n", __FUNCTION__); + if (nfs_wait_on_sequence(data->arg.lock_seqid, task) != 0) + return; /* Do we need to do an open_to_lock_owner? */ if (!(data->arg.lock_seqid->sequence->flags & NFS_SEQID_CONFIRMED)) { if (nfs_wait_on_sequence(data->arg.open_seqid, task) != 0) return; data->arg.open_stateid = &state->stateid; data->arg.new_lock_owner = 1; - /* Retest in case we raced... */ - if (!(data->arg.lock_seqid->sequence->flags & NFS_SEQID_CONFIRMED)) - goto do_rpc; - } - if (nfs_wait_on_sequence(data->arg.lock_seqid, task) != 0) - return; - data->arg.new_lock_owner = 0; -do_rpc: + } else + data->arg.new_lock_owner = 0; data->timestamp = jiffies; - rpc_call_setup(task, &msg, 0); + rpc_call_start(task); dprintk("%s: done!, ret = %d\n", __FUNCTION__, data->rpc_status); } @@ -3419,6 +3385,7 @@ static void nfs4_lock_release(void *calldata) struct nfs4_lockdata *data = calldata; dprintk("%s: begin!\n", __FUNCTION__); + nfs_free_seqid(data->arg.open_seqid); if (data->cancelled != 0) { struct rpc_task *task; task = nfs4_do_unlck(&data->fl, data->ctx, data->lsp, @@ -3428,8 +3395,6 @@ static void nfs4_lock_release(void *calldata) dprintk("%s: cancelling lock!\n", __FUNCTION__); } else nfs_free_seqid(data->arg.lock_seqid); - if (data->arg.open_seqid != NULL) - nfs_free_seqid(data->arg.open_seqid); nfs4_put_lock_state(data->lsp); put_nfs_open_context(data->ctx); kfree(data); @@ -3446,6 +3411,16 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f { struct nfs4_lockdata *data; struct rpc_task *task; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOCK], + .rpc_cred = state->owner->so_cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = NFS_CLIENT(state->inode), + .rpc_message = &msg, + .callback_ops = &nfs4_lock_ops, + .flags = RPC_TASK_ASYNC, + }; int ret; dprintk("%s: begin!\n", __FUNCTION__); @@ -3457,8 +3432,10 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f data->arg.block = 1; if (reclaim != 0) data->arg.reclaim = 1; - task = rpc_run_task(NFS_CLIENT(state->inode), RPC_TASK_ASYNC, - &nfs4_lock_ops, data); + msg.rpc_argp = &data->arg, + msg.rpc_resp = &data->res, + task_setup_data.callback_data = data; + task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) return PTR_ERR(task); ret = nfs4_wait_for_completion_rpc_task(task); @@ -3631,10 +3608,6 @@ int nfs4_setxattr(struct dentry *dentry, const char *key, const void *buf, if (strcmp(key, XATTR_NAME_NFSV4_ACL) != 0) return -EOPNOTSUPP; - if (!S_ISREG(inode->i_mode) && - (!S_ISDIR(inode->i_mode) || inode->i_mode & S_ISVTX)) - return -EPERM; - return nfs4_proc_set_acl(inode, buf, buflen); } diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 5a39c6f78ac..b962397004c 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -644,27 +644,26 @@ void nfs4_copy_stateid(nfs4_stateid *dst, struct nfs4_state *state, fl_owner_t f struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter) { - struct rpc_sequence *sequence = counter->sequence; struct nfs_seqid *new; new = kmalloc(sizeof(*new), GFP_KERNEL); if (new != NULL) { new->sequence = counter; - spin_lock(&sequence->lock); - list_add_tail(&new->list, &sequence->list); - spin_unlock(&sequence->lock); + INIT_LIST_HEAD(&new->list); } return new; } void nfs_free_seqid(struct nfs_seqid *seqid) { - struct rpc_sequence *sequence = seqid->sequence->sequence; + if (!list_empty(&seqid->list)) { + struct rpc_sequence *sequence = seqid->sequence->sequence; - spin_lock(&sequence->lock); - list_del(&seqid->list); - spin_unlock(&sequence->lock); - rpc_wake_up(&sequence->wait); + spin_lock(&sequence->lock); + list_del(&seqid->list); + spin_unlock(&sequence->lock); + rpc_wake_up(&sequence->wait); + } kfree(seqid); } @@ -675,6 +674,7 @@ void nfs_free_seqid(struct nfs_seqid *seqid) */ static void nfs_increment_seqid(int status, struct nfs_seqid *seqid) { + BUG_ON(list_first_entry(&seqid->sequence->sequence->list, struct nfs_seqid, list) != seqid); switch (status) { case 0: break; @@ -682,8 +682,8 @@ static void nfs_increment_seqid(int status, struct nfs_seqid *seqid) if (seqid->sequence->flags & NFS_SEQID_CONFIRMED) return; printk(KERN_WARNING "NFS: v4 server returned a bad" - "sequence-id error on an" - "unconfirmed sequence %p!\n", + " sequence-id error on an" + " unconfirmed sequence %p!\n", seqid->sequence); case -NFS4ERR_STALE_CLIENTID: case -NFS4ERR_STALE_STATEID: @@ -726,15 +726,15 @@ int nfs_wait_on_sequence(struct nfs_seqid *seqid, struct rpc_task *task) struct rpc_sequence *sequence = seqid->sequence->sequence; int status = 0; - if (sequence->list.next == &seqid->list) - goto out; spin_lock(&sequence->lock); - if (sequence->list.next != &seqid->list) { - rpc_sleep_on(&sequence->wait, task, NULL, NULL); - status = -EAGAIN; - } + if (list_empty(&seqid->list)) + list_add_tail(&seqid->list, &sequence->list); + if (list_first_entry(&sequence->list, struct nfs_seqid, list) == seqid) + goto unlock; + rpc_sleep_on(&sequence->wait, task, NULL, NULL); + status = -EAGAIN; +unlock: spin_unlock(&sequence->lock); -out: return status; } @@ -758,8 +758,9 @@ static void nfs4_recover_state(struct nfs_client *clp) __module_get(THIS_MODULE); atomic_inc(&clp->cl_count); - task = kthread_run(reclaimer, clp, "%u.%u.%u.%u-reclaim", - NIPQUAD(clp->cl_addr.sin_addr)); + task = kthread_run(reclaimer, clp, "%s-reclaim", + rpc_peeraddr2str(clp->cl_rpcclient, + RPC_DISPLAY_ADDR)); if (!IS_ERR(task)) return; nfs4_clear_recover_bit(clp); @@ -784,7 +785,7 @@ static int nfs4_reclaim_locks(struct nfs4_state_recovery_ops *ops, struct nfs4_s struct file_lock *fl; int status = 0; - for (fl = inode->i_flock; fl != 0; fl = fl->fl_next) { + for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) { if (!(fl->fl_flags & (FL_POSIX|FL_FLOCK))) continue; if (nfs_file_open_context(fl->fl_file)->state != state) @@ -970,8 +971,8 @@ out: module_put_and_exit(0); return 0; out_error: - printk(KERN_WARNING "Error: state recovery failed on NFSv4 server %u.%u.%u.%u with error %d\n", - NIPQUAD(clp->cl_addr.sin_addr), -status); + printk(KERN_WARNING "Error: state recovery failed on NFSv4 server %s" + " with error %d\n", clp->cl_hostname, -status); set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); goto out; } diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 51dd3804866..db1ed9c46ed 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -116,10 +116,12 @@ static int nfs4_stat_to_errno(int); #define decode_renew_maxsz (op_decode_hdr_maxsz) #define encode_setclientid_maxsz \ (op_encode_hdr_maxsz + \ - 4 /*server->ip_addr*/ + \ - 1 /*Netid*/ + \ - 6 /*uaddr*/ + \ - 6 + (NFS4_VERIFIER_SIZE >> 2)) + XDR_QUADLEN(NFS4_VERIFIER_SIZE) + \ + XDR_QUADLEN(NFS4_SETCLIENTID_NAMELEN) + \ + 1 /* sc_prog */ + \ + XDR_QUADLEN(RPCBIND_MAXNETIDLEN) + \ + XDR_QUADLEN(RPCBIND_MAXUADDRLEN) + \ + 1) /* sc_cb_ident */ #define decode_setclientid_maxsz \ (op_decode_hdr_maxsz + \ 2 + \ @@ -2515,14 +2517,12 @@ static int decode_attr_files_total(struct xdr_stream *xdr, uint32_t *bitmap, uin static int decode_pathname(struct xdr_stream *xdr, struct nfs4_pathname *path) { - int n; + u32 n; __be32 *p; int status = 0; READ_BUF(4); READ32(n); - if (n < 0) - goto out_eio; if (n == 0) goto root_path; dprintk("path "); @@ -2579,13 +2579,11 @@ static int decode_attr_fs_locations(struct xdr_stream *xdr, uint32_t *bitmap, st goto out_eio; res->nlocations = 0; while (res->nlocations < n) { - int m; + u32 m; struct nfs4_fs_location *loc = &res->locations[res->nlocations]; READ_BUF(4); READ32(m); - if (m <= 0) - goto out_eio; loc->nservers = 0; dprintk("%s: servers ", __FUNCTION__); @@ -2598,8 +2596,12 @@ static int decode_attr_fs_locations(struct xdr_stream *xdr, uint32_t *bitmap, st if (loc->nservers < NFS4_FS_LOCATION_MAXSERVERS) loc->nservers++; else { - int i; - dprintk("%s: using first %d of %d servers returned for location %d\n", __FUNCTION__, NFS4_FS_LOCATION_MAXSERVERS, m, res->nlocations); + unsigned int i; + dprintk("%s: using first %u of %u servers " + "returned for location %u\n", + __FUNCTION__, + NFS4_FS_LOCATION_MAXSERVERS, + m, res->nlocations); for (i = loc->nservers; i < m; i++) { unsigned int len; char *data; @@ -3476,10 +3478,11 @@ static int decode_readdir(struct xdr_stream *xdr, struct rpc_rqst *req, struct n struct xdr_buf *rcvbuf = &req->rq_rcv_buf; struct page *page = *rcvbuf->pages; struct kvec *iov = rcvbuf->head; - unsigned int nr, pglen = rcvbuf->page_len; + size_t hdrlen; + u32 recvd, pglen = rcvbuf->page_len; __be32 *end, *entry, *p, *kaddr; - uint32_t len, attrlen, xlen; - int hdrlen, recvd, status; + unsigned int nr; + int status; status = decode_op_hdr(xdr, OP_READDIR); if (status) @@ -3503,6 +3506,7 @@ static int decode_readdir(struct xdr_stream *xdr, struct rpc_rqst *req, struct n end = p + ((pglen + readdir->pgbase) >> 2); entry = p; for (nr = 0; *p++; nr++) { + u32 len, attrlen, xlen; if (end - p < 3) goto short_pkt; dprintk("cookie = %Lu, ", *((unsigned long long *)p)); @@ -3551,7 +3555,8 @@ static int decode_readlink(struct xdr_stream *xdr, struct rpc_rqst *req) { struct xdr_buf *rcvbuf = &req->rq_rcv_buf; struct kvec *iov = rcvbuf->head; - int hdrlen, len, recvd; + size_t hdrlen; + u32 len, recvd; __be32 *p; char *kaddr; int status; @@ -3646,7 +3651,8 @@ static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req, if (unlikely(bitmap[0] & (FATTR4_WORD0_ACL - 1U))) return -EIO; if (likely(bitmap[0] & FATTR4_WORD0_ACL)) { - int hdrlen, recvd; + size_t hdrlen; + u32 recvd; /* We ignore &savep and don't do consistency checks on * the attr length. Let userspace figure it out.... */ diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c index 4b0334590ee..531379d3682 100644 --- a/fs/nfs/nfsroot.c +++ b/fs/nfs/nfsroot.c @@ -228,10 +228,7 @@ static int __init root_nfs_parse(char *name, char *buf) nfs_data.flags &= ~NFS_MOUNT_SOFT; break; case Opt_intr: - nfs_data.flags |= NFS_MOUNT_INTR; - break; case Opt_nointr: - nfs_data.flags &= ~NFS_MOUNT_INTR; break; case Opt_posix: nfs_data.flags |= NFS_MOUNT_POSIX; diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 345bb9b4765..7f079209d70 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -58,7 +58,6 @@ nfs_create_request(struct nfs_open_context *ctx, struct inode *inode, struct page *page, unsigned int offset, unsigned int count) { - struct nfs_server *server = NFS_SERVER(inode); struct nfs_page *req; for (;;) { @@ -67,7 +66,7 @@ nfs_create_request(struct nfs_open_context *ctx, struct inode *inode, if (req != NULL) break; - if (signalled() && (server->flags & NFS_MOUNT_INTR)) + if (fatal_signal_pending(current)) return ERR_PTR(-ERESTARTSYS); yield(); } @@ -111,13 +110,14 @@ void nfs_unlock_request(struct nfs_page *req) * nfs_set_page_tag_locked - Tag a request as locked * @req: */ -static int nfs_set_page_tag_locked(struct nfs_page *req) +int nfs_set_page_tag_locked(struct nfs_page *req) { struct nfs_inode *nfsi = NFS_I(req->wb_context->path.dentry->d_inode); - if (!nfs_lock_request(req)) + if (!nfs_lock_request_dontget(req)) return 0; - radix_tree_tag_set(&nfsi->nfs_page_tree, req->wb_index, NFS_PAGE_TAG_LOCKED); + if (req->wb_page != NULL) + radix_tree_tag_set(&nfsi->nfs_page_tree, req->wb_index, NFS_PAGE_TAG_LOCKED); return 1; } @@ -132,9 +132,10 @@ void nfs_clear_page_tag_locked(struct nfs_page *req) if (req->wb_page != NULL) { spin_lock(&inode->i_lock); radix_tree_tag_clear(&nfsi->nfs_page_tree, req->wb_index, NFS_PAGE_TAG_LOCKED); + nfs_unlock_request(req); spin_unlock(&inode->i_lock); - } - nfs_unlock_request(req); + } else + nfs_unlock_request(req); } /** @@ -175,11 +176,11 @@ void nfs_release_request(struct nfs_page *req) kref_put(&req->wb_kref, nfs_free_request); } -static int nfs_wait_bit_interruptible(void *word) +static int nfs_wait_bit_killable(void *word) { int ret = 0; - if (signal_pending(current)) + if (fatal_signal_pending(current)) ret = -ERESTARTSYS; else schedule(); @@ -190,26 +191,18 @@ static int nfs_wait_bit_interruptible(void *word) * nfs_wait_on_request - Wait for a request to complete. * @req: request to wait upon. * - * Interruptible by signals only if mounted with intr flag. + * Interruptible by fatal signals only. * The user is responsible for holding a count on the request. */ int nfs_wait_on_request(struct nfs_page *req) { - struct rpc_clnt *clnt = NFS_CLIENT(req->wb_context->path.dentry->d_inode); - sigset_t oldmask; int ret = 0; if (!test_bit(PG_BUSY, &req->wb_flags)) goto out; - /* - * Note: the call to rpc_clnt_sigmask() suffices to ensure that we - * are not interrupted if intr flag is not set - */ - rpc_clnt_sigmask(clnt, &oldmask); ret = out_of_line_wait_on_bit(&req->wb_flags, PG_BUSY, - nfs_wait_bit_interruptible, TASK_INTERRUPTIBLE); - rpc_clnt_sigunmask(clnt, &oldmask); + nfs_wait_bit_killable, TASK_KILLABLE); out: return ret; } @@ -421,6 +414,7 @@ int nfs_scan_list(struct nfs_inode *nfsi, goto out; idx_start = req->wb_index + 1; if (nfs_set_page_tag_locked(req)) { + kref_get(&req->wb_kref); nfs_list_remove_request(req); radix_tree_tag_clear(&nfsi->nfs_page_tree, req->wb_index, tag); diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index 4f80d88e9fe..5ccf7faee19 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -565,16 +565,9 @@ static int nfs_read_done(struct rpc_task *task, struct nfs_read_data *data) return 0; } -static void nfs_proc_read_setup(struct nfs_read_data *data) +static void nfs_proc_read_setup(struct nfs_read_data *data, struct rpc_message *msg) { - struct rpc_message msg = { - .rpc_proc = &nfs_procedures[NFSPROC_READ], - .rpc_argp = &data->args, - .rpc_resp = &data->res, - .rpc_cred = data->cred, - }; - - rpc_call_setup(&data->task, &msg, 0); + msg->rpc_proc = &nfs_procedures[NFSPROC_READ]; } static int nfs_write_done(struct rpc_task *task, struct nfs_write_data *data) @@ -584,24 +577,15 @@ static int nfs_write_done(struct rpc_task *task, struct nfs_write_data *data) return 0; } -static void nfs_proc_write_setup(struct nfs_write_data *data, int how) +static void nfs_proc_write_setup(struct nfs_write_data *data, struct rpc_message *msg) { - struct rpc_message msg = { - .rpc_proc = &nfs_procedures[NFSPROC_WRITE], - .rpc_argp = &data->args, - .rpc_resp = &data->res, - .rpc_cred = data->cred, - }; - /* Note: NFSv2 ignores @stable and always uses NFS_FILE_SYNC */ data->args.stable = NFS_FILE_SYNC; - - /* Finalize the task. */ - rpc_call_setup(&data->task, &msg, 0); + msg->rpc_proc = &nfs_procedures[NFSPROC_WRITE]; } static void -nfs_proc_commit_setup(struct nfs_write_data *data, int how) +nfs_proc_commit_setup(struct nfs_write_data *data, struct rpc_message *msg) { BUG(); } @@ -609,7 +593,9 @@ nfs_proc_commit_setup(struct nfs_write_data *data, int how) static int nfs_proc_lock(struct file *filp, int cmd, struct file_lock *fl) { - return nlmclnt_proc(filp->f_path.dentry->d_inode, cmd, fl); + struct inode *inode = filp->f_path.dentry->d_inode; + + return nlmclnt_proc(NFS_SERVER(inode)->nlm_host, cmd, fl); } diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 4587a86adaa..5a70be589bb 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -79,7 +79,7 @@ void nfs_readdata_release(void *data) static int nfs_return_empty_page(struct page *page) { - zero_user_page(page, 0, PAGE_CACHE_SIZE, KM_USER0); + zero_user(page, 0, PAGE_CACHE_SIZE); SetPageUptodate(page); unlock_page(page); return 0; @@ -103,10 +103,10 @@ static void nfs_readpage_truncate_uninitialised_page(struct nfs_read_data *data) pglen = PAGE_CACHE_SIZE - base; for (;;) { if (remainder <= pglen) { - zero_user_page(*pages, base, remainder, KM_USER0); + zero_user(*pages, base, remainder); break; } - zero_user_page(*pages, base, pglen, KM_USER0); + zero_user(*pages, base, pglen); pages++; remainder -= pglen; pglen = PAGE_CACHE_SIZE; @@ -130,7 +130,7 @@ static int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode, return PTR_ERR(new); } if (len < PAGE_CACHE_SIZE) - zero_user_page(page, len, PAGE_CACHE_SIZE - len, KM_USER0); + zero_user_segment(page, len, PAGE_CACHE_SIZE); nfs_list_add_request(new, &one_request); if (NFS_SERVER(inode)->rsize < PAGE_CACHE_SIZE) @@ -160,12 +160,26 @@ static void nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data, const struct rpc_call_ops *call_ops, unsigned int count, unsigned int offset) { - struct inode *inode; - int flags; + struct inode *inode = req->wb_context->path.dentry->d_inode; + int swap_flags = IS_SWAPFILE(inode) ? NFS_RPC_SWAPFLAGS : 0; + struct rpc_task *task; + struct rpc_message msg = { + .rpc_argp = &data->args, + .rpc_resp = &data->res, + .rpc_cred = req->wb_context->cred, + }; + struct rpc_task_setup task_setup_data = { + .task = &data->task, + .rpc_client = NFS_CLIENT(inode), + .rpc_message = &msg, + .callback_ops = call_ops, + .callback_data = data, + .flags = RPC_TASK_ASYNC | swap_flags, + }; data->req = req; - data->inode = inode = req->wb_context->path.dentry->d_inode; - data->cred = req->wb_context->cred; + data->inode = inode; + data->cred = msg.rpc_cred; data->args.fh = NFS_FH(inode); data->args.offset = req_offset(req) + offset; @@ -180,11 +194,7 @@ static void nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data, nfs_fattr_init(&data->fattr); /* Set up the initial task struct. */ - flags = RPC_TASK_ASYNC | (IS_SWAPFILE(inode)? NFS_RPC_SWAPFLAGS : 0); - rpc_init_task(&data->task, NFS_CLIENT(inode), flags, call_ops, data); - NFS_PROTO(inode)->read_setup(data); - - data->task.tk_cookie = (unsigned long)inode; + NFS_PROTO(inode)->read_setup(data, &msg); dprintk("NFS: %5u initiated read call (req %s/%Ld, %u bytes @ offset %Lu)\n", data->task.tk_pid, @@ -192,6 +202,10 @@ static void nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data, (long long)NFS_FILEID(inode), count, (unsigned long long)data->args.offset); + + task = rpc_run_task(&task_setup_data); + if (!IS_ERR(task)) + rpc_put_task(task); } static void @@ -208,19 +222,6 @@ nfs_async_read_error(struct list_head *head) } /* - * Start an async read operation - */ -static void nfs_execute_read(struct nfs_read_data *data) -{ - struct rpc_clnt *clnt = NFS_CLIENT(data->inode); - sigset_t oldset; - - rpc_clnt_sigmask(clnt, &oldset); - rpc_execute(&data->task); - rpc_clnt_sigunmask(clnt, &oldset); -} - -/* * Generate multiple requests to fill a single page. * * We optimize to reduce the number of read operations on the wire. If we @@ -274,7 +275,6 @@ static int nfs_pagein_multi(struct inode *inode, struct list_head *head, unsigne rsize, offset); offset += rsize; nbytes -= rsize; - nfs_execute_read(data); } while (nbytes != 0); return 0; @@ -312,8 +312,6 @@ static int nfs_pagein_one(struct inode *inode, struct list_head *head, unsigned req = nfs_list_entry(data->pages.next); nfs_read_rpcsetup(req, data, &nfs_read_full_ops, count, 0); - - nfs_execute_read(data); return 0; out_bad: nfs_async_read_error(head); @@ -338,7 +336,7 @@ int nfs_readpage_result(struct rpc_task *task, struct nfs_read_data *data) nfs_add_stats(data->inode, NFSIOS_SERVERREADBYTES, data->res.count); if (task->tk_status == -ESTALE) { - set_bit(NFS_INO_STALE, &NFS_FLAGS(data->inode)); + set_bit(NFS_INO_STALE, &NFS_I(data->inode)->flags); nfs_mark_for_revalidate(data->inode); } return 0; @@ -534,8 +532,11 @@ readpage_async_filler(void *data, struct page *page) goto out_error; if (len < PAGE_CACHE_SIZE) - zero_user_page(page, len, PAGE_CACHE_SIZE - len, KM_USER0); - nfs_pageio_add_request(desc->pgio, new); + zero_user_segment(page, len, PAGE_CACHE_SIZE); + if (!nfs_pageio_add_request(desc->pgio, new)) { + error = desc->pgio->pg_error; + goto out_unlock; + } return 0; out_error: error = PTR_ERR(new); diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 0b0c72a072f..f9219024f31 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -45,6 +45,8 @@ #include <linux/nfs_idmap.h> #include <linux/vfs.h> #include <linux/inet.h> +#include <linux/in6.h> +#include <net/ipv6.h> #include <linux/nfs_xdr.h> #include <linux/magic.h> #include <linux/parser.h> @@ -83,11 +85,11 @@ enum { Opt_actimeo, Opt_namelen, Opt_mountport, - Opt_mountprog, Opt_mountvers, - Opt_nfsprog, Opt_nfsvers, + Opt_mountvers, + Opt_nfsvers, /* Mount options that take string arguments */ - Opt_sec, Opt_proto, Opt_mountproto, + Opt_sec, Opt_proto, Opt_mountproto, Opt_mounthost, Opt_addr, Opt_mountaddr, Opt_clientaddr, /* Mount options that are ignored */ @@ -137,9 +139,7 @@ static match_table_t nfs_mount_option_tokens = { { Opt_userspace, "retry=%u" }, { Opt_namelen, "namlen=%u" }, { Opt_mountport, "mountport=%u" }, - { Opt_mountprog, "mountprog=%u" }, { Opt_mountvers, "mountvers=%u" }, - { Opt_nfsprog, "nfsprog=%u" }, { Opt_nfsvers, "nfsvers=%u" }, { Opt_nfsvers, "vers=%u" }, @@ -148,7 +148,7 @@ static match_table_t nfs_mount_option_tokens = { { Opt_mountproto, "mountproto=%s" }, { Opt_addr, "addr=%s" }, { Opt_clientaddr, "clientaddr=%s" }, - { Opt_userspace, "mounthost=%s" }, + { Opt_mounthost, "mounthost=%s" }, { Opt_mountaddr, "mountaddr=%s" }, { Opt_err, NULL } @@ -190,6 +190,10 @@ static match_table_t nfs_secflavor_tokens = { { Opt_sec_lkeyi, "lkeyi" }, { Opt_sec_lkeyp, "lkeyp" }, + { Opt_sec_spkm, "spkm3" }, + { Opt_sec_spkmi, "spkm3i" }, + { Opt_sec_spkmp, "spkm3p" }, + { Opt_sec_err, NULL } }; @@ -202,6 +206,7 @@ static int nfs_get_sb(struct file_system_type *, int, const char *, void *, stru static int nfs_xdev_get_sb(struct file_system_type *fs_type, int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt); static void nfs_kill_super(struct super_block *); +static void nfs_put_super(struct super_block *); static struct file_system_type nfs_fs_type = { .owner = THIS_MODULE, @@ -223,6 +228,7 @@ static const struct super_operations nfs_sops = { .alloc_inode = nfs_alloc_inode, .destroy_inode = nfs_destroy_inode, .write_inode = nfs_write_inode, + .put_super = nfs_put_super, .statfs = nfs_statfs, .clear_inode = nfs_clear_inode, .umount_begin = nfs_umount_begin, @@ -325,6 +331,28 @@ void __exit unregister_nfs_fs(void) unregister_filesystem(&nfs_fs_type); } +void nfs_sb_active(struct nfs_server *server) +{ + atomic_inc(&server->active); +} + +void nfs_sb_deactive(struct nfs_server *server) +{ + if (atomic_dec_and_test(&server->active)) + wake_up(&server->active_wq); +} + +static void nfs_put_super(struct super_block *sb) +{ + struct nfs_server *server = NFS_SB(sb); + /* + * Make sure there are no outstanding ops to this server. + * If so, wait for them to finish before allowing the + * unmount to continue. + */ + wait_event(server->active_wq, atomic_read(&server->active) == 0); +} + /* * Deliver file system statistics to userspace */ @@ -424,7 +452,6 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss, const char *nostr; } nfs_info[] = { { NFS_MOUNT_SOFT, ",soft", ",hard" }, - { NFS_MOUNT_INTR, ",intr", ",nointr" }, { NFS_MOUNT_NOCTO, ",nocto", "" }, { NFS_MOUNT_NOAC, ",noac", "" }, { NFS_MOUNT_NONLM, ",nolock", "" }, @@ -455,8 +482,8 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss, } seq_printf(m, ",proto=%s", rpc_peeraddr2str(nfss->client, RPC_DISPLAY_PROTO)); - seq_printf(m, ",timeo=%lu", 10U * clp->retrans_timeo / HZ); - seq_printf(m, ",retrans=%u", clp->retrans_count); + seq_printf(m, ",timeo=%lu", 10U * nfss->client->cl_timeout->to_initval / HZ); + seq_printf(m, ",retrans=%u", nfss->client->cl_timeout->to_retries); seq_printf(m, ",sec=%s", nfs_pseudoflavour_to_name(nfss->client->cl_auth->au_flavor)); } @@ -469,8 +496,9 @@ static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt) nfs_show_mount_options(m, nfss, 0); - seq_printf(m, ",addr="NIPQUAD_FMT, - NIPQUAD(nfss->nfs_client->cl_addr.sin_addr)); + seq_printf(m, ",addr=%s", + rpc_peeraddr2str(nfss->nfs_client->cl_rpcclient, + RPC_DISPLAY_ADDR)); return 0; } @@ -507,7 +535,7 @@ static int nfs_show_stats(struct seq_file *m, struct vfsmount *mnt) seq_printf(m, ",namelen=%d", nfss->namelen); #ifdef CONFIG_NFS_V4 - if (nfss->nfs_client->cl_nfsversion == 4) { + if (nfss->nfs_client->rpc_ops->version == 4) { seq_printf(m, "\n\tnfsv4:\t"); seq_printf(m, "bm0=0x%x", nfss->attr_bitmask[0]); seq_printf(m, ",bm1=0x%x", nfss->attr_bitmask[1]); @@ -561,8 +589,6 @@ static void nfs_umount_begin(struct vfsmount *vfsmnt, int flags) struct nfs_server *server = NFS_SB(vfsmnt->mnt_sb); struct rpc_clnt *rpc; - shrink_submounts(vfsmnt, &nfs_automount_list); - if (!(flags & MNT_FORCE)) return; /* -EIO all pending I/O */ @@ -575,16 +601,40 @@ static void nfs_umount_begin(struct vfsmount *vfsmnt, int flags) } /* - * Sanity-check a server address provided by the mount command + * Set the port number in an address. Be agnostic about the address family. + */ +static void nfs_set_port(struct sockaddr *sap, unsigned short port) +{ + switch (sap->sa_family) { + case AF_INET: { + struct sockaddr_in *ap = (struct sockaddr_in *)sap; + ap->sin_port = htons(port); + break; + } + case AF_INET6: { + struct sockaddr_in6 *ap = (struct sockaddr_in6 *)sap; + ap->sin6_port = htons(port); + break; + } + } +} + +/* + * Sanity-check a server address provided by the mount command. + * + * Address family must be initialized, and address must not be + * the ANY address for that family. */ static int nfs_verify_server_address(struct sockaddr *addr) { switch (addr->sa_family) { case AF_INET: { - struct sockaddr_in *sa = (struct sockaddr_in *) addr; - if (sa->sin_addr.s_addr != INADDR_ANY) - return 1; - break; + struct sockaddr_in *sa = (struct sockaddr_in *)addr; + return sa->sin_addr.s_addr != htonl(INADDR_ANY); + } + case AF_INET6: { + struct in6_addr *sa = &((struct sockaddr_in6 *)addr)->sin6_addr; + return !ipv6_addr_any(sa); } } @@ -592,13 +642,49 @@ static int nfs_verify_server_address(struct sockaddr *addr) } /* + * Parse string addresses passed in via a mount option, + * and construct a sockaddr based on the result. + * + * If address parsing fails, set the sockaddr's address + * family to AF_UNSPEC to force nfs_verify_server_address() + * to punt the mount. + */ +static void nfs_parse_server_address(char *value, + struct sockaddr *sap, + size_t *len) +{ + if (strchr(value, ':')) { + struct sockaddr_in6 *ap = (struct sockaddr_in6 *)sap; + u8 *addr = (u8 *)&ap->sin6_addr.in6_u; + + ap->sin6_family = AF_INET6; + *len = sizeof(*ap); + if (in6_pton(value, -1, addr, '\0', NULL)) + return; + } else { + struct sockaddr_in *ap = (struct sockaddr_in *)sap; + u8 *addr = (u8 *)&ap->sin_addr.s_addr; + + ap->sin_family = AF_INET; + *len = sizeof(*ap); + if (in4_pton(value, -1, addr, '\0', NULL)) + return; + } + + sap->sa_family = AF_UNSPEC; + *len = 0; +} + +/* * Error-check and convert a string of mount options from user space into * a data structure */ static int nfs_parse_mount_options(char *raw, struct nfs_parsed_mount_data *mnt) { - char *p, *string; + char *p, *string, *secdata; + unsigned short port = 0; + int rc; if (!raw) { dfprintk(MOUNT, "NFS: mount options string was NULL.\n"); @@ -606,6 +692,20 @@ static int nfs_parse_mount_options(char *raw, } dfprintk(MOUNT, "NFS: nfs mount opts='%s'\n", raw); + secdata = alloc_secdata(); + if (!secdata) + goto out_nomem; + + rc = security_sb_copy_data(raw, secdata); + if (rc) + goto out_security_failure; + + rc = security_sb_parse_opts_str(secdata, &mnt->lsm_opts); + if (rc) + goto out_security_failure; + + free_secdata(secdata); + while ((p = strsep(&raw, ",")) != NULL) { substring_t args[MAX_OPT_ARGS]; int option, token; @@ -624,10 +724,7 @@ static int nfs_parse_mount_options(char *raw, mnt->flags &= ~NFS_MOUNT_SOFT; break; case Opt_intr: - mnt->flags |= NFS_MOUNT_INTR; - break; case Opt_nointr: - mnt->flags &= ~NFS_MOUNT_INTR; break; case Opt_posix: mnt->flags |= NFS_MOUNT_POSIX; @@ -701,7 +798,7 @@ static int nfs_parse_mount_options(char *raw, return 0; if (option < 0 || option > 65535) return 0; - mnt->nfs_server.address.sin_port = htons(option); + port = option; break; case Opt_rsize: if (match_int(args, &mnt->rsize)) @@ -763,13 +860,6 @@ static int nfs_parse_mount_options(char *raw, return 0; mnt->mount_server.port = option; break; - case Opt_mountprog: - if (match_int(args, &option)) - return 0; - if (option < 0) - return 0; - mnt->mount_server.program = option; - break; case Opt_mountvers: if (match_int(args, &option)) return 0; @@ -777,13 +867,6 @@ static int nfs_parse_mount_options(char *raw, return 0; mnt->mount_server.version = option; break; - case Opt_nfsprog: - if (match_int(args, &option)) - return 0; - if (option < 0) - return 0; - mnt->nfs_server.program = option; - break; case Opt_nfsvers: if (match_int(args, &option)) return 0; @@ -927,24 +1010,32 @@ static int nfs_parse_mount_options(char *raw, string = match_strdup(args); if (string == NULL) goto out_nomem; - mnt->nfs_server.address.sin_family = AF_INET; - mnt->nfs_server.address.sin_addr.s_addr = - in_aton(string); + nfs_parse_server_address(string, (struct sockaddr *) + &mnt->nfs_server.address, + &mnt->nfs_server.addrlen); kfree(string); break; case Opt_clientaddr: string = match_strdup(args); if (string == NULL) goto out_nomem; + kfree(mnt->client_address); mnt->client_address = string; break; + case Opt_mounthost: + string = match_strdup(args); + if (string == NULL) + goto out_nomem; + kfree(mnt->mount_server.hostname); + mnt->mount_server.hostname = string; + break; case Opt_mountaddr: string = match_strdup(args); if (string == NULL) goto out_nomem; - mnt->mount_server.address.sin_family = AF_INET; - mnt->mount_server.address.sin_addr.s_addr = - in_aton(string); + nfs_parse_server_address(string, (struct sockaddr *) + &mnt->mount_server.address, + &mnt->mount_server.addrlen); kfree(string); break; @@ -957,12 +1048,17 @@ static int nfs_parse_mount_options(char *raw, } } + nfs_set_port((struct sockaddr *)&mnt->nfs_server.address, port); + return 1; out_nomem: printk(KERN_INFO "NFS: not enough memory to parse option\n"); return 0; - +out_security_failure: + free_secdata(secdata); + printk(KERN_INFO "NFS: security options invalid: %d\n", rc); + return 0; out_unrec_vers: printk(KERN_INFO "NFS: unrecognized NFS version number\n"); return 0; @@ -987,7 +1083,8 @@ out_unknown: static int nfs_try_mount(struct nfs_parsed_mount_data *args, struct nfs_fh *root_fh) { - struct sockaddr_in sin; + struct sockaddr *sap = (struct sockaddr *)&args->mount_server.address; + char *hostname; int status; if (args->mount_server.version == 0) { @@ -997,25 +1094,32 @@ static int nfs_try_mount(struct nfs_parsed_mount_data *args, args->mount_server.version = NFS_MNT_VERSION; } + if (args->mount_server.hostname) + hostname = args->mount_server.hostname; + else + hostname = args->nfs_server.hostname; + /* * Construct the mount server's address. */ - if (args->mount_server.address.sin_addr.s_addr != INADDR_ANY) - sin = args->mount_server.address; - else - sin = args->nfs_server.address; + if (args->mount_server.address.ss_family == AF_UNSPEC) { + memcpy(sap, &args->nfs_server.address, + args->nfs_server.addrlen); + args->mount_server.addrlen = args->nfs_server.addrlen; + } + /* * autobind will be used if mount_server.port == 0 */ - sin.sin_port = htons(args->mount_server.port); + nfs_set_port(sap, args->mount_server.port); /* * Now ask the mount server to map our export path * to a file handle. */ - status = nfs_mount((struct sockaddr *) &sin, - sizeof(sin), - args->nfs_server.hostname, + status = nfs_mount(sap, + args->mount_server.addrlen, + hostname, args->nfs_server.export_path, args->mount_server.version, args->mount_server.protocol, @@ -1023,8 +1127,8 @@ static int nfs_try_mount(struct nfs_parsed_mount_data *args, if (status == 0) return 0; - dfprintk(MOUNT, "NFS: unable to mount server " NIPQUAD_FMT - ", error %d\n", NIPQUAD(sin.sin_addr.s_addr), status); + dfprintk(MOUNT, "NFS: unable to mount server %s, error %d", + hostname, status); return status; } @@ -1043,9 +1147,6 @@ static int nfs_try_mount(struct nfs_parsed_mount_data *args, * * + breaking back: trying proto=udp after proto=tcp, v2 after v3, * mountproto=tcp after mountproto=udp, and so on - * - * XXX: as far as I can tell, changing the NFS program number is not - * supported in the NFS client. */ static int nfs_validate_mount_data(void *options, struct nfs_parsed_mount_data *args, @@ -1069,9 +1170,7 @@ static int nfs_validate_mount_data(void *options, args->acdirmin = 30; args->acdirmax = 60; args->mount_server.protocol = XPRT_TRANSPORT_UDP; - args->mount_server.program = NFS_MNT_PROGRAM; args->nfs_server.protocol = XPRT_TRANSPORT_TCP; - args->nfs_server.program = NFS_PROGRAM; switch (data->version) { case 1: @@ -1102,9 +1201,6 @@ static int nfs_validate_mount_data(void *options, memset(mntfh->data + mntfh->size, 0, sizeof(mntfh->data) - mntfh->size); - if (!nfs_verify_server_address((struct sockaddr *) &data->addr)) - goto out_no_address; - /* * Translate to nfs_parsed_mount_data, which nfs_fill_super * can deal with. @@ -1119,7 +1215,14 @@ static int nfs_validate_mount_data(void *options, args->acregmax = data->acregmax; args->acdirmin = data->acdirmin; args->acdirmax = data->acdirmax; - args->nfs_server.address = data->addr; + + memcpy(&args->nfs_server.address, &data->addr, + sizeof(data->addr)); + args->nfs_server.addrlen = sizeof(data->addr); + if (!nfs_verify_server_address((struct sockaddr *) + &args->nfs_server.address)) + goto out_no_address; + if (!(data->flags & NFS_MOUNT_TCP)) args->nfs_server.protocol = XPRT_TRANSPORT_UDP; /* N.B. caller will free nfs_server.hostname in all cases */ @@ -1127,6 +1230,33 @@ static int nfs_validate_mount_data(void *options, args->namlen = data->namlen; args->bsize = data->bsize; args->auth_flavors[0] = data->pseudoflavor; + + /* + * The legacy version 6 binary mount data from userspace has a + * field used only to transport selinux information into the + * the kernel. To continue to support that functionality we + * have a touch of selinux knowledge here in the NFS code. The + * userspace code converted context=blah to just blah so we are + * converting back to the full string selinux understands. + */ + if (data->context[0]){ +#ifdef CONFIG_SECURITY_SELINUX + int rc; + char *opts_str = kmalloc(sizeof(data->context) + 8, GFP_KERNEL); + if (!opts_str) + return -ENOMEM; + strcpy(opts_str, "context="); + data->context[NFS_MAX_CONTEXT_LEN] = '\0'; + strcat(opts_str, &data->context[0]); + rc = security_sb_parse_opts_str(opts_str, &args->lsm_opts); + kfree(opts_str); + if (rc) + return rc; +#else + return -EINVAL; +#endif + } + break; default: { unsigned int len; @@ -1322,15 +1452,50 @@ static int nfs_set_super(struct super_block *s, void *data) return ret; } +static int nfs_compare_super_address(struct nfs_server *server1, + struct nfs_server *server2) +{ + struct sockaddr *sap1, *sap2; + + sap1 = (struct sockaddr *)&server1->nfs_client->cl_addr; + sap2 = (struct sockaddr *)&server2->nfs_client->cl_addr; + + if (sap1->sa_family != sap2->sa_family) + return 0; + + switch (sap1->sa_family) { + case AF_INET: { + struct sockaddr_in *sin1 = (struct sockaddr_in *)sap1; + struct sockaddr_in *sin2 = (struct sockaddr_in *)sap2; + if (sin1->sin_addr.s_addr != sin2->sin_addr.s_addr) + return 0; + if (sin1->sin_port != sin2->sin_port) + return 0; + break; + } + case AF_INET6: { + struct sockaddr_in6 *sin1 = (struct sockaddr_in6 *)sap1; + struct sockaddr_in6 *sin2 = (struct sockaddr_in6 *)sap2; + if (!ipv6_addr_equal(&sin1->sin6_addr, &sin2->sin6_addr)) + return 0; + if (sin1->sin6_port != sin2->sin6_port) + return 0; + break; + } + default: + return 0; + } + + return 1; +} + static int nfs_compare_super(struct super_block *sb, void *data) { struct nfs_sb_mountdata *sb_mntdata = data; struct nfs_server *server = sb_mntdata->server, *old = NFS_SB(sb); int mntflags = sb_mntdata->mntflags; - if (memcmp(&old->nfs_client->cl_addr, - &server->nfs_client->cl_addr, - sizeof(old->nfs_client->cl_addr)) != 0) + if (!nfs_compare_super_address(old, server)) return 0; /* Note: NFS_MOUNT_UNSHARED == NFS4_MOUNT_UNSHARED */ if (old->flags & NFS_MOUNT_UNSHARED) @@ -1354,6 +1519,8 @@ static int nfs_get_sb(struct file_system_type *fs_type, }; int error; + security_init_mnt_opts(&data.lsm_opts); + /* Validate the mount data */ error = nfs_validate_mount_data(raw_data, &data, &mntfh, dev_name); if (error < 0) @@ -1393,6 +1560,10 @@ static int nfs_get_sb(struct file_system_type *fs_type, goto error_splat_super; } + error = security_sb_set_mnt_opts(s, &data.lsm_opts); + if (error) + goto error_splat_root; + s->s_flags |= MS_ACTIVE; mnt->mnt_sb = s; mnt->mnt_root = mntroot; @@ -1400,12 +1571,16 @@ static int nfs_get_sb(struct file_system_type *fs_type, out: kfree(data.nfs_server.hostname); + kfree(data.mount_server.hostname); + security_free_mnt_opts(&data.lsm_opts); return error; out_err_nosb: nfs_free_server(server); goto out; +error_splat_root: + dput(mntroot); error_splat_super: up_write(&s->s_umount); deactivate_super(s); @@ -1485,6 +1660,9 @@ static int nfs_xdev_get_sb(struct file_system_type *fs_type, int flags, mnt->mnt_sb = s; mnt->mnt_root = mntroot; + /* clone any lsm security options from the parent to the new sb */ + security_sb_clone_mnt_opts(data->sb, s); + dprintk("<-- nfs_xdev_get_sb() = 0\n"); return 0; @@ -1528,12 +1706,35 @@ static void nfs4_fill_super(struct super_block *sb) } /* + * If the user didn't specify a port, set the port number to + * the NFS version 4 default port. + */ +static void nfs4_default_port(struct sockaddr *sap) +{ + switch (sap->sa_family) { + case AF_INET: { + struct sockaddr_in *ap = (struct sockaddr_in *)sap; + if (ap->sin_port == 0) + ap->sin_port = htons(NFS_PORT); + break; + } + case AF_INET6: { + struct sockaddr_in6 *ap = (struct sockaddr_in6 *)sap; + if (ap->sin6_port == 0) + ap->sin6_port = htons(NFS_PORT); + break; + } + } +} + +/* * Validate NFSv4 mount options */ static int nfs4_validate_mount_data(void *options, struct nfs_parsed_mount_data *args, const char *dev_name) { + struct sockaddr_in *ap; struct nfs4_mount_data *data = (struct nfs4_mount_data *)options; char *c; @@ -1554,18 +1755,21 @@ static int nfs4_validate_mount_data(void *options, switch (data->version) { case 1: - if (data->host_addrlen != sizeof(args->nfs_server.address)) + ap = (struct sockaddr_in *)&args->nfs_server.address; + if (data->host_addrlen > sizeof(args->nfs_server.address)) + goto out_no_address; + if (data->host_addrlen == 0) goto out_no_address; - if (copy_from_user(&args->nfs_server.address, - data->host_addr, - sizeof(args->nfs_server.address))) + args->nfs_server.addrlen = data->host_addrlen; + if (copy_from_user(ap, data->host_addr, data->host_addrlen)) return -EFAULT; - if (args->nfs_server.address.sin_port == 0) - args->nfs_server.address.sin_port = htons(NFS_PORT); if (!nfs_verify_server_address((struct sockaddr *) &args->nfs_server.address)) goto out_no_address; + nfs4_default_port((struct sockaddr *) + &args->nfs_server.address); + switch (data->auth_flavourlen) { case 0: args->auth_flavors[0] = RPC_AUTH_UNIX; @@ -1623,6 +1827,9 @@ static int nfs4_validate_mount_data(void *options, &args->nfs_server.address)) return -EINVAL; + nfs4_default_port((struct sockaddr *) + &args->nfs_server.address); + switch (args->auth_flavor_len) { case 0: args->auth_flavors[0] = RPC_AUTH_UNIX; @@ -1643,21 +1850,16 @@ static int nfs4_validate_mount_data(void *options, len = c - dev_name; if (len > NFS4_MAXNAMLEN) return -ENAMETOOLONG; - args->nfs_server.hostname = kzalloc(len, GFP_KERNEL); - if (args->nfs_server.hostname == NULL) - return -ENOMEM; - strncpy(args->nfs_server.hostname, dev_name, len - 1); + /* N.B. caller will free nfs_server.hostname in all cases */ + args->nfs_server.hostname = kstrndup(dev_name, len, GFP_KERNEL); c++; /* step over the ':' */ len = strlen(c); if (len > NFS4_MAXPATHLEN) return -ENAMETOOLONG; - args->nfs_server.export_path = kzalloc(len + 1, GFP_KERNEL); - if (args->nfs_server.export_path == NULL) - return -ENOMEM; - strncpy(args->nfs_server.export_path, c, len); + args->nfs_server.export_path = kstrndup(c, len, GFP_KERNEL); - dprintk("MNTPATH: %s\n", args->nfs_server.export_path); + dprintk("NFS: MNTPATH: '%s'\n", args->nfs_server.export_path); if (args->client_address == NULL) goto out_no_client_address; @@ -1703,6 +1905,8 @@ static int nfs4_get_sb(struct file_system_type *fs_type, }; int error; + security_init_mnt_opts(&data.lsm_opts); + /* Validate the mount data */ error = nfs4_validate_mount_data(raw_data, &data, dev_name); if (error < 0) @@ -1751,6 +1955,7 @@ out: kfree(data.client_address); kfree(data.nfs_server.export_path); kfree(data.nfs_server.hostname); + security_free_mnt_opts(&data.lsm_opts); return error; out_free: diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index 233ad38161f..75741536342 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -14,6 +14,8 @@ #include <linux/sched.h> #include <linux/wait.h> +#include "internal.h" + struct nfs_unlinkdata { struct hlist_node list; struct nfs_removeargs args; @@ -69,24 +71,6 @@ static void nfs_dec_sillycount(struct inode *dir) } /** - * nfs_async_unlink_init - Initialize the RPC info - * task: rpc_task of the sillydelete - */ -static void nfs_async_unlink_init(struct rpc_task *task, void *calldata) -{ - struct nfs_unlinkdata *data = calldata; - struct inode *dir = data->dir; - struct rpc_message msg = { - .rpc_argp = &data->args, - .rpc_resp = &data->res, - .rpc_cred = data->cred, - }; - - NFS_PROTO(dir)->unlink_setup(&msg, dir); - rpc_call_setup(task, &msg, 0); -} - -/** * nfs_async_unlink_done - Sillydelete post-processing * @task: rpc_task of the sillydelete * @@ -113,32 +97,45 @@ static void nfs_async_unlink_release(void *calldata) struct nfs_unlinkdata *data = calldata; nfs_dec_sillycount(data->dir); + nfs_sb_deactive(NFS_SERVER(data->dir)); nfs_free_unlinkdata(data); } static const struct rpc_call_ops nfs_unlink_ops = { - .rpc_call_prepare = nfs_async_unlink_init, .rpc_call_done = nfs_async_unlink_done, .rpc_release = nfs_async_unlink_release, }; static int nfs_do_call_unlink(struct dentry *parent, struct inode *dir, struct nfs_unlinkdata *data) { + struct rpc_message msg = { + .rpc_argp = &data->args, + .rpc_resp = &data->res, + .rpc_cred = data->cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_message = &msg, + .callback_ops = &nfs_unlink_ops, + .callback_data = data, + .flags = RPC_TASK_ASYNC, + }; struct rpc_task *task; struct dentry *alias; alias = d_lookup(parent, &data->args.name); if (alias != NULL) { int ret = 0; + /* * Hey, we raced with lookup... See if we need to transfer * the sillyrename information to the aliased dentry. */ nfs_free_dname(data); spin_lock(&alias->d_lock); - if (!(alias->d_flags & DCACHE_NFSFS_RENAMED)) { + if (alias->d_inode != NULL && + !(alias->d_flags & DCACHE_NFSFS_RENAMED)) { alias->d_fsdata = data; - alias->d_flags ^= DCACHE_NFSFS_RENAMED; + alias->d_flags |= DCACHE_NFSFS_RENAMED; ret = 1; } spin_unlock(&alias->d_lock); @@ -151,10 +148,14 @@ static int nfs_do_call_unlink(struct dentry *parent, struct inode *dir, struct n nfs_dec_sillycount(dir); return 0; } + nfs_sb_active(NFS_SERVER(dir)); data->args.fh = NFS_FH(dir); nfs_fattr_init(&data->res.dir_attr); - task = rpc_run_task(NFS_CLIENT(dir), RPC_TASK_ASYNC, &nfs_unlink_ops, data); + NFS_PROTO(dir)->unlink_setup(&msg, dir); + + task_setup_data.rpc_client = NFS_CLIENT(dir); + task = rpc_run_task(&task_setup_data); if (!IS_ERR(task)) rpc_put_task(task); return 1; diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 51cc1bd6a11..bed63416a55 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -39,6 +39,7 @@ static struct nfs_page * nfs_update_request(struct nfs_open_context*, unsigned int, unsigned int); static void nfs_pageio_init_write(struct nfs_pageio_descriptor *desc, struct inode *inode, int ioflags); +static void nfs_redirty_request(struct nfs_page *req); static const struct rpc_call_ops nfs_write_partial_ops; static const struct rpc_call_ops nfs_write_full_ops; static const struct rpc_call_ops nfs_commit_ops; @@ -196,7 +197,7 @@ static int nfs_writepage_setup(struct nfs_open_context *ctx, struct page *page, } /* Update file length */ nfs_grow_file(page, offset, count); - nfs_unlock_request(req); + nfs_clear_page_tag_locked(req); return 0; } @@ -252,7 +253,6 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio, struct page *page) { struct inode *inode = page->mapping->host; - struct nfs_inode *nfsi = NFS_I(inode); struct nfs_page *req; int ret; @@ -263,10 +263,10 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio, spin_unlock(&inode->i_lock); return 0; } - if (nfs_lock_request_dontget(req)) + if (nfs_set_page_tag_locked(req)) break; /* Note: If we hold the page lock, as is the case in nfs_writepage, - * then the call to nfs_lock_request_dontget() will always + * then the call to nfs_set_page_tag_locked() will always * succeed provided that someone hasn't already marked the * request as dirty (in which case we don't care). */ @@ -280,7 +280,7 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio, if (test_bit(PG_NEED_COMMIT, &req->wb_flags)) { /* This request is marked for commit */ spin_unlock(&inode->i_lock); - nfs_unlock_request(req); + nfs_clear_page_tag_locked(req); nfs_pageio_complete(pgio); return 0; } @@ -288,10 +288,13 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio, spin_unlock(&inode->i_lock); BUG(); } - radix_tree_tag_set(&nfsi->nfs_page_tree, req->wb_index, - NFS_PAGE_TAG_LOCKED); spin_unlock(&inode->i_lock); - nfs_pageio_add_request(pgio, req); + if (!nfs_pageio_add_request(pgio, req)) { + nfs_redirty_request(req); + nfs_end_page_writeback(page); + nfs_clear_page_tag_locked(req); + return pgio->pg_error; + } return 0; } @@ -381,6 +384,7 @@ static int nfs_inode_add_request(struct inode *inode, struct nfs_page *req) set_page_private(req->wb_page, (unsigned long)req); nfsi->npages++; kref_get(&req->wb_kref); + radix_tree_tag_set(&nfsi->nfs_page_tree, req->wb_index, NFS_PAGE_TAG_LOCKED); return 0; } @@ -490,7 +494,7 @@ int nfs_reschedule_unstable_write(struct nfs_page *req) /* * Wait for a request to complete. * - * Interruptible by signals only if mounted with intr flag. + * Interruptible by fatal signals only. */ static int nfs_wait_on_requests_locked(struct inode *inode, pgoff_t idx_start, unsigned int npages) { @@ -596,7 +600,7 @@ static struct nfs_page * nfs_update_request(struct nfs_open_context* ctx, spin_lock(&inode->i_lock); req = nfs_page_find_request_locked(page); if (req) { - if (!nfs_lock_request_dontget(req)) { + if (!nfs_set_page_tag_locked(req)) { int error; spin_unlock(&inode->i_lock); @@ -646,7 +650,7 @@ static struct nfs_page * nfs_update_request(struct nfs_open_context* ctx, || req->wb_page != page || !nfs_dirty_request(req) || offset > rqend || end < req->wb_offset) { - nfs_unlock_request(req); + nfs_clear_page_tag_locked(req); return ERR_PTR(-EBUSY); } @@ -667,9 +671,7 @@ zero_page: * then we need to zero any uninitalised data. */ if (req->wb_pgbase == 0 && req->wb_bytes != PAGE_CACHE_SIZE && !PageUptodate(req->wb_page)) - zero_user_page(req->wb_page, req->wb_bytes, - PAGE_CACHE_SIZE - req->wb_bytes, - KM_USER0); + zero_user_segment(req->wb_page, req->wb_bytes, PAGE_CACHE_SIZE); return req; } @@ -701,6 +703,17 @@ int nfs_flush_incompatible(struct file *file, struct page *page) } /* + * If the page cache is marked as unsafe or invalid, then we can't rely on + * the PageUptodate() flag. In this case, we will need to turn off + * write optimisations that depend on the page contents being correct. + */ +static int nfs_write_pageuptodate(struct page *page, struct inode *inode) +{ + return PageUptodate(page) && + !(NFS_I(inode)->cache_validity & (NFS_INO_REVAL_PAGECACHE|NFS_INO_INVALID_DATA)); +} + +/* * Update and possibly write a cached page of an NFS file. * * XXX: Keep an eye on generic_file_read to make sure it doesn't do bad @@ -721,10 +734,13 @@ int nfs_updatepage(struct file *file, struct page *page, (long long)(page_offset(page) +offset)); /* If we're not using byte range locks, and we know the page - * is entirely in cache, it may be more efficient to avoid - * fragmenting write requests. + * is up to date, it may be more efficient to extend the write + * to cover the entire page in order to avoid fragmentation + * inefficiencies. */ - if (PageUptodate(page) && inode->i_flock == NULL && !(file->f_mode & O_SYNC)) { + if (nfs_write_pageuptodate(page, inode) && + inode->i_flock == NULL && + !(file->f_flags & O_SYNC)) { count = max(count + offset, nfs_page_length(page)); offset = 0; } @@ -755,7 +771,7 @@ static void nfs_writepage_release(struct nfs_page *req) nfs_clear_page_tag_locked(req); } -static inline int flush_task_priority(int how) +static int flush_task_priority(int how) { switch (how & (FLUSH_HIGHPRI|FLUSH_LOWPRI)) { case FLUSH_HIGHPRI: @@ -775,15 +791,31 @@ static void nfs_write_rpcsetup(struct nfs_page *req, unsigned int count, unsigned int offset, int how) { - struct inode *inode; - int flags; + struct inode *inode = req->wb_context->path.dentry->d_inode; + int flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC; + int priority = flush_task_priority(how); + struct rpc_task *task; + struct rpc_message msg = { + .rpc_argp = &data->args, + .rpc_resp = &data->res, + .rpc_cred = req->wb_context->cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = NFS_CLIENT(inode), + .task = &data->task, + .rpc_message = &msg, + .callback_ops = call_ops, + .callback_data = data, + .flags = flags, + .priority = priority, + }; /* Set up the RPC argument and reply structs * NB: take care not to mess about with data->commit et al. */ data->req = req; data->inode = inode = req->wb_context->path.dentry->d_inode; - data->cred = req->wb_context->cred; + data->cred = msg.rpc_cred; data->args.fh = NFS_FH(inode); data->args.offset = req_offset(req) + offset; @@ -791,6 +823,12 @@ static void nfs_write_rpcsetup(struct nfs_page *req, data->args.pages = data->pagevec; data->args.count = count; data->args.context = req->wb_context; + data->args.stable = NFS_UNSTABLE; + if (how & FLUSH_STABLE) { + data->args.stable = NFS_DATA_SYNC; + if (!NFS_I(inode)->ncommit) + data->args.stable = NFS_FILE_SYNC; + } data->res.fattr = &data->fattr; data->res.count = count; @@ -798,12 +836,7 @@ static void nfs_write_rpcsetup(struct nfs_page *req, nfs_fattr_init(&data->fattr); /* Set up the initial task struct. */ - flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC; - rpc_init_task(&data->task, NFS_CLIENT(inode), flags, call_ops, data); - NFS_PROTO(inode)->write_setup(data, how); - - data->task.tk_priority = flush_task_priority(how); - data->task.tk_cookie = (unsigned long)inode; + NFS_PROTO(inode)->write_setup(data, &msg); dprintk("NFS: %5u initiated write call " "(req %s/%Ld, %u bytes @ offset %Lu)\n", @@ -812,16 +845,10 @@ static void nfs_write_rpcsetup(struct nfs_page *req, (long long)NFS_FILEID(inode), count, (unsigned long long)data->args.offset); -} - -static void nfs_execute_write(struct nfs_write_data *data) -{ - struct rpc_clnt *clnt = NFS_CLIENT(data->inode); - sigset_t oldset; - rpc_clnt_sigmask(clnt, &oldset); - rpc_execute(&data->task); - rpc_clnt_sigunmask(clnt, &oldset); + task = rpc_run_task(&task_setup_data); + if (!IS_ERR(task)) + rpc_put_task(task); } /* @@ -868,7 +895,6 @@ static int nfs_flush_multi(struct inode *inode, struct list_head *head, unsigned wsize, offset, how); offset += wsize; nbytes -= wsize; - nfs_execute_write(data); } while (nbytes != 0); return 0; @@ -916,7 +942,6 @@ static int nfs_flush_one(struct inode *inode, struct list_head *head, unsigned i /* Set up the argument struct */ nfs_write_rpcsetup(req, data, &nfs_write_full_ops, count, 0, how); - nfs_execute_write(data); return 0; out_bad: while (!list_empty(head)) { @@ -932,7 +957,7 @@ static int nfs_flush_one(struct inode *inode, struct list_head *head, unsigned i static void nfs_pageio_init_write(struct nfs_pageio_descriptor *pgio, struct inode *inode, int ioflags) { - int wsize = NFS_SERVER(inode)->wsize; + size_t wsize = NFS_SERVER(inode)->wsize; if (wsize < PAGE_CACHE_SIZE) nfs_pageio_init(pgio, inode, nfs_flush_multi, wsize, ioflags); @@ -1146,19 +1171,33 @@ static void nfs_commit_rpcsetup(struct list_head *head, struct nfs_write_data *data, int how) { - struct nfs_page *first; - struct inode *inode; - int flags; + struct nfs_page *first = nfs_list_entry(head->next); + struct inode *inode = first->wb_context->path.dentry->d_inode; + int flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC; + int priority = flush_task_priority(how); + struct rpc_task *task; + struct rpc_message msg = { + .rpc_argp = &data->args, + .rpc_resp = &data->res, + .rpc_cred = first->wb_context->cred, + }; + struct rpc_task_setup task_setup_data = { + .task = &data->task, + .rpc_client = NFS_CLIENT(inode), + .rpc_message = &msg, + .callback_ops = &nfs_commit_ops, + .callback_data = data, + .flags = flags, + .priority = priority, + }; /* Set up the RPC argument and reply structs * NB: take care not to mess about with data->commit et al. */ list_splice_init(head, &data->pages); - first = nfs_list_entry(data->pages.next); - inode = first->wb_context->path.dentry->d_inode; data->inode = inode; - data->cred = first->wb_context->cred; + data->cred = msg.rpc_cred; data->args.fh = NFS_FH(data->inode); /* Note: we always request a commit of the entire inode */ @@ -1170,14 +1209,13 @@ static void nfs_commit_rpcsetup(struct list_head *head, nfs_fattr_init(&data->fattr); /* Set up the initial task struct. */ - flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC; - rpc_init_task(&data->task, NFS_CLIENT(inode), flags, &nfs_commit_ops, data); - NFS_PROTO(inode)->commit_setup(data, how); + NFS_PROTO(inode)->commit_setup(data, &msg); - data->task.tk_priority = flush_task_priority(how); - data->task.tk_cookie = (unsigned long)inode; - dprintk("NFS: %5u initiated commit call\n", data->task.tk_pid); + + task = rpc_run_task(&task_setup_data); + if (!IS_ERR(task)) + rpc_put_task(task); } /* @@ -1197,7 +1235,6 @@ nfs_commit_list(struct inode *inode, struct list_head *head, int how) /* Set up the argument struct */ nfs_commit_rpcsetup(head, data, how); - nfs_execute_write(data); return 0; out_bad: while (!list_empty(head)) { diff --git a/fs/nfsctl.c b/fs/nfsctl.c index 51f1b31acbf..aed8145d908 100644 --- a/fs/nfsctl.c +++ b/fs/nfsctl.c @@ -41,9 +41,9 @@ static struct file *do_open(char *name, int flags) error = may_open(&nd, MAY_WRITE, FMODE_WRITE); if (!error) - return dentry_open(nd.dentry, nd.mnt, flags); + return dentry_open(nd.path.dentry, nd.path.mnt, flags); - path_release(&nd); + path_put(&nd.path); return ERR_PTR(error); } diff --git a/fs/nfsd/auth.c b/fs/nfsd/auth.c index 21928056e35..d13403e3362 100644 --- a/fs/nfsd/auth.c +++ b/fs/nfsd/auth.c @@ -11,8 +11,6 @@ #include <linux/nfsd/nfsd.h> #include <linux/nfsd/export.h> -#define CAP_NFSD_MASK (CAP_FS_MASK|CAP_TO_MASK(CAP_SYS_RESOURCE)) - int nfsexp_flags(struct svc_rqst *rqstp, struct svc_export *exp) { struct exp_flavor_info *f; @@ -69,10 +67,12 @@ int nfsd_setuser(struct svc_rqst *rqstp, struct svc_export *exp) ret = set_current_groups(cred.cr_group_info); put_group_info(cred.cr_group_info); if ((cred.cr_uid)) { - cap_t(current->cap_effective) &= ~CAP_NFSD_MASK; + current->cap_effective = + cap_drop_nfsd_set(current->cap_effective); } else { - cap_t(current->cap_effective) |= (CAP_NFSD_MASK & - current->cap_permitted); + current->cap_effective = + cap_raise_nfsd_set(current->cap_effective, + current->cap_permitted); } return ret; } diff --git a/fs/nfsd/auth.h b/fs/nfsd/auth.h new file mode 100644 index 00000000000..78b3c0e9382 --- /dev/null +++ b/fs/nfsd/auth.h @@ -0,0 +1,22 @@ +/* + * nfsd-specific authentication stuff. + * uid/gid mapping not yet implemented. + * + * Copyright (C) 1995, 1996 Olaf Kirch <okir@monad.swb.de> + */ + +#ifndef LINUX_NFSD_AUTH_H +#define LINUX_NFSD_AUTH_H + +#define nfsd_luid(rq, uid) ((u32)(uid)) +#define nfsd_lgid(rq, gid) ((u32)(gid)) +#define nfsd_ruid(rq, uid) ((u32)(uid)) +#define nfsd_rgid(rq, gid) ((u32)(gid)) + +/* + * Set the current process's fsuid/fsgid etc to those of the NFS + * client user + */ +int nfsd_setuser(struct svc_rqst *, struct svc_export *); + +#endif /* LINUX_NFSD_AUTH_H */ diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index 66d0aeb32a4..8a6f7c924c7 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -63,10 +63,8 @@ static void expkey_put(struct kref *ref) struct svc_expkey *key = container_of(ref, struct svc_expkey, h.ref); if (test_bit(CACHE_VALID, &key->h.flags) && - !test_bit(CACHE_NEGATIVE, &key->h.flags)) { - dput(key->ek_dentry); - mntput(key->ek_mnt); - } + !test_bit(CACHE_NEGATIVE, &key->h.flags)) + path_put(&key->ek_path); auth_domain_put(key->ek_client); kfree(key); } @@ -169,15 +167,14 @@ static int expkey_parse(struct cache_detail *cd, char *mesg, int mlen) goto out; dprintk("Found the path %s\n", buf); - key.ek_mnt = nd.mnt; - key.ek_dentry = nd.dentry; - + key.ek_path = nd.path; + ek = svc_expkey_update(&key, ek); if (ek) cache_put(&ek->h, &svc_expkey_cache); else err = -ENOMEM; - path_release(&nd); + path_put(&nd.path); } cache_flush(); out: @@ -206,7 +203,7 @@ static int expkey_show(struct seq_file *m, if (test_bit(CACHE_VALID, &h->flags) && !test_bit(CACHE_NEGATIVE, &h->flags)) { seq_printf(m, " "); - seq_path(m, ek->ek_mnt, ek->ek_dentry, "\\ \t\n"); + seq_path(m, &ek->ek_path, "\\ \t\n"); } seq_printf(m, "\n"); return 0; @@ -243,8 +240,8 @@ static inline void expkey_update(struct cache_head *cnew, struct svc_expkey *new = container_of(cnew, struct svc_expkey, h); struct svc_expkey *item = container_of(citem, struct svc_expkey, h); - new->ek_mnt = mntget(item->ek_mnt); - new->ek_dentry = dget(item->ek_dentry); + new->ek_path = item->ek_path; + path_get(&item->ek_path); } static struct cache_head *expkey_alloc(void) @@ -332,10 +329,9 @@ static void nfsd4_fslocs_free(struct nfsd4_fs_locations *fsloc) static void svc_export_put(struct kref *ref) { struct svc_export *exp = container_of(ref, struct svc_export, h.ref); - dput(exp->ex_dentry); - mntput(exp->ex_mnt); + path_put(&exp->ex_path); auth_domain_put(exp->ex_client); - kfree(exp->ex_path); + kfree(exp->ex_pathname); nfsd4_fslocs_free(&exp->ex_fslocs); kfree(exp); } @@ -349,7 +345,7 @@ static void svc_export_request(struct cache_detail *cd, char *pth; qword_add(bpp, blen, exp->ex_client->name); - pth = d_path(exp->ex_dentry, exp->ex_mnt, *bpp, *blen); + pth = d_path(&exp->ex_path, *bpp, *blen); if (IS_ERR(pth)) { /* is this correct? */ (*bpp)[0] = '\n'; @@ -507,8 +503,8 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen) struct svc_export exp, *expp; int an_int; - nd.dentry = NULL; - exp.ex_path = NULL; + nd.path.dentry = NULL; + exp.ex_pathname = NULL; /* fs locations */ exp.ex_fslocs.locations = NULL; @@ -547,11 +543,11 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen) exp.h.flags = 0; exp.ex_client = dom; - exp.ex_mnt = nd.mnt; - exp.ex_dentry = nd.dentry; - exp.ex_path = kstrdup(buf, GFP_KERNEL); + exp.ex_path.mnt = nd.path.mnt; + exp.ex_path.dentry = nd.path.dentry; + exp.ex_pathname = kstrdup(buf, GFP_KERNEL); err = -ENOMEM; - if (!exp.ex_path) + if (!exp.ex_pathname) goto out; /* expiry */ @@ -610,7 +606,7 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen) goto out; } - err = check_export(nd.dentry->d_inode, exp.ex_flags, + err = check_export(nd.path.dentry->d_inode, exp.ex_flags, exp.ex_uuid); if (err) goto out; } @@ -628,9 +624,9 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen) out: nfsd4_fslocs_free(&exp.ex_fslocs); kfree(exp.ex_uuid); - kfree(exp.ex_path); - if (nd.dentry) - path_release(&nd); + kfree(exp.ex_pathname); + if (nd.path.dentry) + path_put(&nd.path); out_no_path: if (dom) auth_domain_put(dom); @@ -653,7 +649,7 @@ static int svc_export_show(struct seq_file *m, return 0; } exp = container_of(h, struct svc_export, h); - seq_path(m, exp->ex_mnt, exp->ex_dentry, " \t\n\\"); + seq_path(m, &exp->ex_path, " \t\n\\"); seq_putc(m, '\t'); seq_escape(m, exp->ex_client->name, " \t\n\\"); seq_putc(m, '('); @@ -680,8 +676,8 @@ static int svc_export_match(struct cache_head *a, struct cache_head *b) struct svc_export *orig = container_of(a, struct svc_export, h); struct svc_export *new = container_of(b, struct svc_export, h); return orig->ex_client == new->ex_client && - orig->ex_dentry == new->ex_dentry && - orig->ex_mnt == new->ex_mnt; + orig->ex_path.dentry == new->ex_path.dentry && + orig->ex_path.mnt == new->ex_path.mnt; } static void svc_export_init(struct cache_head *cnew, struct cache_head *citem) @@ -691,9 +687,9 @@ static void svc_export_init(struct cache_head *cnew, struct cache_head *citem) kref_get(&item->ex_client->ref); new->ex_client = item->ex_client; - new->ex_dentry = dget(item->ex_dentry); - new->ex_mnt = mntget(item->ex_mnt); - new->ex_path = NULL; + new->ex_path.dentry = dget(item->ex_path.dentry); + new->ex_path.mnt = mntget(item->ex_path.mnt); + new->ex_pathname = NULL; new->ex_fslocs.locations = NULL; new->ex_fslocs.locations_count = 0; new->ex_fslocs.migrated = 0; @@ -711,8 +707,8 @@ static void export_update(struct cache_head *cnew, struct cache_head *citem) new->ex_fsid = item->ex_fsid; new->ex_uuid = item->ex_uuid; item->ex_uuid = NULL; - new->ex_path = item->ex_path; - item->ex_path = NULL; + new->ex_pathname = item->ex_pathname; + item->ex_pathname = NULL; new->ex_fslocs.locations = item->ex_fslocs.locations; item->ex_fslocs.locations = NULL; new->ex_fslocs.locations_count = item->ex_fslocs.locations_count; @@ -755,8 +751,8 @@ svc_export_lookup(struct svc_export *exp) struct cache_head *ch; int hash; hash = hash_ptr(exp->ex_client, EXPORT_HASHBITS); - hash ^= hash_ptr(exp->ex_dentry, EXPORT_HASHBITS); - hash ^= hash_ptr(exp->ex_mnt, EXPORT_HASHBITS); + hash ^= hash_ptr(exp->ex_path.dentry, EXPORT_HASHBITS); + hash ^= hash_ptr(exp->ex_path.mnt, EXPORT_HASHBITS); ch = sunrpc_cache_lookup(&svc_export_cache, &exp->h, hash); @@ -772,8 +768,8 @@ svc_export_update(struct svc_export *new, struct svc_export *old) struct cache_head *ch; int hash; hash = hash_ptr(old->ex_client, EXPORT_HASHBITS); - hash ^= hash_ptr(old->ex_dentry, EXPORT_HASHBITS); - hash ^= hash_ptr(old->ex_mnt, EXPORT_HASHBITS); + hash ^= hash_ptr(old->ex_path.dentry, EXPORT_HASHBITS); + hash ^= hash_ptr(old->ex_path.mnt, EXPORT_HASHBITS); ch = sunrpc_cache_update(&svc_export_cache, &new->h, &old->h, @@ -815,8 +811,7 @@ static int exp_set_key(svc_client *clp, int fsid_type, u32 *fsidv, key.ek_client = clp; key.ek_fsidtype = fsid_type; memcpy(key.ek_fsid, fsidv, key_len(fsid_type)); - key.ek_mnt = exp->ex_mnt; - key.ek_dentry = exp->ex_dentry; + key.ek_path = exp->ex_path; key.h.expiry_time = NEVER; key.h.flags = 0; @@ -865,13 +860,13 @@ static svc_export *exp_get_by_name(svc_client *clp, struct vfsmount *mnt, { struct svc_export *exp, key; int err; - + if (!clp) return ERR_PTR(-ENOENT); key.ex_client = clp; - key.ex_mnt = mnt; - key.ex_dentry = dentry; + key.ex_path.mnt = mnt; + key.ex_path.dentry = dentry; exp = svc_export_lookup(&key); if (exp == NULL) @@ -968,7 +963,7 @@ static int exp_fsid_hash(svc_client *clp, struct svc_export *exp) static int exp_hash(struct auth_domain *clp, struct svc_export *exp) { u32 fsid[2]; - struct inode *inode = exp->ex_dentry->d_inode; + struct inode *inode = exp->ex_path.dentry->d_inode; dev_t dev = inode->i_sb->s_dev; if (old_valid_dev(dev)) { @@ -982,7 +977,7 @@ static int exp_hash(struct auth_domain *clp, struct svc_export *exp) static void exp_unhash(struct svc_export *exp) { struct svc_expkey *ek; - struct inode *inode = exp->ex_dentry->d_inode; + struct inode *inode = exp->ex_path.dentry->d_inode; ek = exp_get_key(exp->ex_client, inode->i_sb->s_dev, inode->i_ino); if (!IS_ERR(ek)) { @@ -1030,15 +1025,16 @@ exp_export(struct nfsctl_export *nxp) goto out_unlock; err = -EINVAL; - exp = exp_get_by_name(clp, nd.mnt, nd.dentry, NULL); + exp = exp_get_by_name(clp, nd.path.mnt, nd.path.dentry, NULL); memset(&new, 0, sizeof(new)); /* must make sure there won't be an ex_fsid clash */ if ((nxp->ex_flags & NFSEXP_FSID) && (!IS_ERR(fsid_key = exp_get_fsid_key(clp, nxp->ex_dev))) && - fsid_key->ek_mnt && - (fsid_key->ek_mnt != nd.mnt || fsid_key->ek_dentry != nd.dentry) ) + fsid_key->ek_path.mnt && + (fsid_key->ek_path.mnt != nd.path.mnt || + fsid_key->ek_path.dentry != nd.path.dentry)) goto finish; if (!IS_ERR(exp)) { @@ -1054,7 +1050,7 @@ exp_export(struct nfsctl_export *nxp) goto finish; } - err = check_export(nd.dentry->d_inode, nxp->ex_flags, NULL); + err = check_export(nd.path.dentry->d_inode, nxp->ex_flags, NULL); if (err) goto finish; err = -ENOMEM; @@ -1063,12 +1059,11 @@ exp_export(struct nfsctl_export *nxp) new.h.expiry_time = NEVER; new.h.flags = 0; - new.ex_path = kstrdup(nxp->ex_path, GFP_KERNEL); - if (!new.ex_path) + new.ex_pathname = kstrdup(nxp->ex_path, GFP_KERNEL); + if (!new.ex_pathname) goto finish; new.ex_client = clp; - new.ex_mnt = nd.mnt; - new.ex_dentry = nd.dentry; + new.ex_path = nd.path; new.ex_flags = nxp->ex_flags; new.ex_anon_uid = nxp->ex_anon_uid; new.ex_anon_gid = nxp->ex_anon_gid; @@ -1089,15 +1084,14 @@ exp_export(struct nfsctl_export *nxp) } else err = 0; finish: - if (new.ex_path) - kfree(new.ex_path); + kfree(new.ex_pathname); if (exp) exp_put(exp); if (fsid_key && !IS_ERR(fsid_key)) cache_put(&fsid_key->h, &svc_expkey_cache); if (clp) auth_domain_put(clp); - path_release(&nd); + path_put(&nd.path); out_unlock: exp_writeunlock(); out: @@ -1148,8 +1142,8 @@ exp_unexport(struct nfsctl_export *nxp) goto out_domain; err = -EINVAL; - exp = exp_get_by_name(dom, nd.mnt, nd.dentry, NULL); - path_release(&nd); + exp = exp_get_by_name(dom, nd.path.mnt, nd.path.dentry, NULL); + path_put(&nd.path); if (IS_ERR(exp)) goto out_domain; @@ -1185,12 +1179,12 @@ exp_rootfh(svc_client *clp, char *path, struct knfsd_fh *f, int maxsize) printk("nfsd: exp_rootfh path not found %s", path); return err; } - inode = nd.dentry->d_inode; + inode = nd.path.dentry->d_inode; dprintk("nfsd: exp_rootfh(%s [%p] %s:%s/%ld)\n", - path, nd.dentry, clp->name, + path, nd.path.dentry, clp->name, inode->i_sb->s_id, inode->i_ino); - exp = exp_parent(clp, nd.mnt, nd.dentry, NULL); + exp = exp_parent(clp, nd.path.mnt, nd.path.dentry, NULL); if (IS_ERR(exp)) { err = PTR_ERR(exp); goto out; @@ -1200,7 +1194,7 @@ exp_rootfh(svc_client *clp, char *path, struct knfsd_fh *f, int maxsize) * fh must be initialized before calling fh_compose */ fh_init(&fh, maxsize); - if (fh_compose(&fh, exp, nd.dentry, NULL)) + if (fh_compose(&fh, exp, nd.path.dentry, NULL)) err = -EINVAL; else err = 0; @@ -1208,7 +1202,7 @@ exp_rootfh(svc_client *clp, char *path, struct knfsd_fh *f, int maxsize) fh_put(&fh); exp_put(exp); out: - path_release(&nd); + path_put(&nd.path); return err; } @@ -1218,13 +1212,13 @@ static struct svc_export *exp_find(struct auth_domain *clp, int fsid_type, struct svc_export *exp; struct svc_expkey *ek = exp_find_key(clp, fsid_type, fsidv, reqp); if (IS_ERR(ek)) - return ERR_PTR(PTR_ERR(ek)); + return ERR_CAST(ek); - exp = exp_get_by_name(clp, ek->ek_mnt, ek->ek_dentry, reqp); + exp = exp_get_by_name(clp, ek->ek_path.mnt, ek->ek_path.dentry, reqp); cache_put(&ek->h, &svc_expkey_cache); if (IS_ERR(exp)) - return ERR_PTR(PTR_ERR(exp)); + return ERR_CAST(exp); return exp; } @@ -1357,11 +1351,9 @@ exp_pseudoroot(struct svc_rqst *rqstp, struct svc_fh *fhp) mk_fsid(FSID_NUM, fsidv, 0, 0, 0, NULL); exp = rqst_exp_find(rqstp, FSID_NUM, fsidv); - if (PTR_ERR(exp) == -ENOENT) - return nfserr_perm; if (IS_ERR(exp)) return nfserrno(PTR_ERR(exp)); - rv = fh_compose(fhp, exp, exp->ex_dentry, NULL); + rv = fh_compose(fhp, exp, exp->ex_path.dentry, NULL); if (rv) goto out; rv = check_nfsd_access(exp, rqstp); @@ -1637,13 +1629,19 @@ exp_verify_string(char *cp, int max) /* * Initialize the exports module. */ -void +int nfsd_export_init(void) { + int rv; dprintk("nfsd: initializing export module.\n"); - cache_register(&svc_export_cache); - cache_register(&svc_expkey_cache); + rv = cache_register(&svc_export_cache); + if (rv) + return rv; + rv = cache_register(&svc_expkey_cache); + if (rv) + cache_unregister(&svc_export_cache); + return rv; } @@ -1670,10 +1668,8 @@ nfsd_export_shutdown(void) exp_writelock(); - if (cache_unregister(&svc_expkey_cache)) - printk(KERN_ERR "nfsd: failed to unregister expkey cache\n"); - if (cache_unregister(&svc_export_cache)) - printk(KERN_ERR "nfsd: failed to unregister export cache\n"); + cache_unregister(&svc_expkey_cache); + cache_unregister(&svc_export_cache); svcauth_unix_purge(); exp_writeunlock(); diff --git a/fs/nfsd/nfs2acl.c b/fs/nfsd/nfs2acl.c index 0e5fa11e6b4..1c3b7654e96 100644 --- a/fs/nfsd/nfs2acl.c +++ b/fs/nfsd/nfs2acl.c @@ -221,12 +221,17 @@ static int nfsaclsvc_encode_getaclres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclres *resp) { struct dentry *dentry = resp->fh.fh_dentry; - struct inode *inode = dentry->d_inode; + struct inode *inode; struct kvec *head = rqstp->rq_res.head; unsigned int base; int n; int w; + /* + * Since this is version 2, the check for nfserr in + * nfsd_dispatch actually ensures the following cannot happen. + * However, it seems fragile to depend on that. + */ if (dentry == NULL || dentry->d_inode == NULL) return 0; inode = dentry->d_inode; diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c index eac82830bfd..c721a1e6e9d 100644 --- a/fs/nfsd/nfs3proc.c +++ b/fs/nfsd/nfs3proc.c @@ -67,7 +67,7 @@ nfsd3_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle *argp, if (nfserr) RETURN_STATUS(nfserr); - err = vfs_getattr(resp->fh.fh_export->ex_mnt, + err = vfs_getattr(resp->fh.fh_export->ex_path.mnt, resp->fh.fh_dentry, &resp->stat); nfserr = nfserrno(err); diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index f917fd25858..17d0dd99720 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -21,6 +21,7 @@ #include <linux/sunrpc/svc.h> #include <linux/nfsd/nfsd.h> #include <linux/nfsd/xdr3.h> +#include "auth.h" #define NFSDDBG_FACILITY NFSDDBG_XDR @@ -88,10 +89,10 @@ encode_fh(__be32 *p, struct svc_fh *fhp) * no slashes or null bytes. */ static __be32 * -decode_filename(__be32 *p, char **namp, int *lenp) +decode_filename(__be32 *p, char **namp, unsigned int *lenp) { char *name; - int i; + unsigned int i; if ((p = xdr_decode_string_inplace(p, namp, lenp, NFS3_MAXNAMLEN)) != NULL) { for (i = 0, name = *namp; i < *lenp; i++, name++) { @@ -217,7 +218,7 @@ encode_post_op_attr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) int err; struct kstat stat; - err = vfs_getattr(fhp->fh_export->ex_mnt, dentry, &stat); + err = vfs_getattr(fhp->fh_export->ex_path.mnt, dentry, &stat); if (!err) { *p++ = xdr_one; /* attributes follow */ lease_get_mtime(dentry->d_inode, &stat.mtime); @@ -269,7 +270,7 @@ void fill_post_wcc(struct svc_fh *fhp) if (fhp->fh_post_saved) printk("nfsd: inode locked twice during operation.\n"); - err = vfs_getattr(fhp->fh_export->ex_mnt, fhp->fh_dentry, + err = vfs_getattr(fhp->fh_export->ex_path.mnt, fhp->fh_dentry, &fhp->fh_post_attr); if (err) fhp->fh_post_saved = 0; @@ -452,8 +453,7 @@ int nfs3svc_decode_symlinkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_symlinkargs *args) { - unsigned int len; - int avail; + unsigned int len, avail; char *old, *new; struct kvec *vec; @@ -486,7 +486,8 @@ nfs3svc_decode_symlinkargs(struct svc_rqst *rqstp, __be32 *p, /* now copy next page if there is one */ if (len && !avail && rqstp->rq_arg.page_len) { avail = rqstp->rq_arg.page_len; - if (avail > PAGE_SIZE) avail = PAGE_SIZE; + if (avail > PAGE_SIZE) + avail = PAGE_SIZE; old = page_address(rqstp->rq_arg.pages[0]); } while (len && avail && *old) { @@ -816,11 +817,11 @@ static __be32 * encode_entryplus_baggage(struct nfsd3_readdirres *cd, __be32 *p, struct svc_fh *fhp) { - p = encode_post_op_attr(cd->rqstp, p, fhp); - *p++ = xdr_one; /* yes, a file handle follows */ - p = encode_fh(p, fhp); - fh_put(fhp); - return p; + p = encode_post_op_attr(cd->rqstp, p, fhp); + *p++ = xdr_one; /* yes, a file handle follows */ + p = encode_fh(p, fhp); + fh_put(fhp); + return p; } static int diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index 9d536a8cb37..aae2b29ae2c 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -350,30 +350,6 @@ static struct rpc_version * nfs_cb_version[] = { static int do_probe_callback(void *data) { struct nfs4_client *clp = data; - struct nfs4_callback *cb = &clp->cl_callback; - struct rpc_message msg = { - .rpc_proc = &nfs4_cb_procedures[NFSPROC4_CLNT_CB_NULL], - .rpc_argp = clp, - }; - int status; - - status = rpc_call_sync(cb->cb_client, &msg, RPC_TASK_SOFT); - - if (status) { - rpc_shutdown_client(cb->cb_client); - cb->cb_client = NULL; - } else - atomic_set(&cb->cb_set, 1); - put_nfs4_client(clp); - return 0; -} - -/* - * Set up the callback client and put a NFSPROC4_CB_NULL on the wire... - */ -void -nfsd4_probe_callback(struct nfs4_client *clp) -{ struct sockaddr_in addr; struct nfs4_callback *cb = &clp->cl_callback; struct rpc_timeout timeparms = { @@ -390,13 +366,15 @@ nfsd4_probe_callback(struct nfs4_client *clp) .timeout = &timeparms, .program = program, .version = nfs_cb_version[1]->number, - .authflavor = RPC_AUTH_UNIX, /* XXX: need AUTH_GSS... */ + .authflavor = RPC_AUTH_UNIX, /* XXX: need AUTH_GSS... */ .flags = (RPC_CLNT_CREATE_NOPING), }; - struct task_struct *t; - - if (atomic_read(&cb->cb_set)) - return; + struct rpc_message msg = { + .rpc_proc = &nfs4_cb_procedures[NFSPROC4_CLNT_CB_NULL], + .rpc_argp = clp, + }; + struct rpc_clnt *client; + int status; /* Initialize address */ memset(&addr, 0, sizeof(addr)); @@ -416,29 +394,50 @@ nfsd4_probe_callback(struct nfs4_client *clp) program->stats->program = program; /* Create RPC client */ - cb->cb_client = rpc_create(&args); - if (IS_ERR(cb->cb_client)) { + client = rpc_create(&args); + if (IS_ERR(client)) { dprintk("NFSD: couldn't create callback client\n"); + status = PTR_ERR(client); goto out_err; } + status = rpc_call_sync(client, &msg, RPC_TASK_SOFT); + + if (status) + goto out_release_client; + + cb->cb_client = client; + atomic_set(&cb->cb_set, 1); + put_nfs4_client(clp); + return 0; +out_release_client: + rpc_shutdown_client(client); +out_err: + put_nfs4_client(clp); + dprintk("NFSD: warning: no callback path to client %.*s\n", + (int)clp->cl_name.len, clp->cl_name.data); + return status; +} + +/* + * Set up the callback client and put a NFSPROC4_CB_NULL on the wire... + */ +void +nfsd4_probe_callback(struct nfs4_client *clp) +{ + struct task_struct *t; + + BUG_ON(atomic_read(&clp->cl_callback.cb_set)); + /* the task holds a reference to the nfs4_client struct */ atomic_inc(&clp->cl_count); t = kthread_run(do_probe_callback, clp, "nfs4_cb_probe"); if (IS_ERR(t)) - goto out_release_clp; + atomic_dec(&clp->cl_count); return; - -out_release_clp: - atomic_dec(&clp->cl_count); - rpc_shutdown_client(cb->cb_client); -out_err: - cb->cb_client = NULL; - dprintk("NFSD: warning: no callback path to client %.*s\n", - (int)clp->cl_name.len, clp->cl_name.data); } /* @@ -458,9 +457,6 @@ nfsd4_cb_recall(struct nfs4_delegation *dp) int retries = 1; int status = 0; - if ((!atomic_read(&clp->cl_callback.cb_set)) || !clnt) - return; - cbr->cbr_trunc = 0; /* XXX need to implement truncate optimization */ cbr->cbr_dp = dp; @@ -469,6 +465,7 @@ nfsd4_cb_recall(struct nfs4_delegation *dp) switch (status) { case -EIO: /* Network partition? */ + atomic_set(&clp->cl_callback.cb_set, 0); case -EBADHANDLE: case -NFS4ERR_BAD_STATEID: /* Race: client probably got cb_recall @@ -481,11 +478,10 @@ nfsd4_cb_recall(struct nfs4_delegation *dp) status = rpc_call_sync(clnt, &msg, RPC_TASK_SOFT); } out_put_cred: - if (status == -EIO) - atomic_set(&clp->cl_callback.cb_set, 0); - /* Success or failure, now we're either waiting for lease expiration - * or deleg_return. */ - dprintk("NFSD: nfs4_cb_recall: dp %p dl_flock %p dl_count %d\n",dp, dp->dl_flock, atomic_read(&dp->dl_count)); + /* + * Success or failure, now we're either waiting for lease expiration + * or deleg_return. + */ put_nfs4_client(clp); nfs4_put_delegation(dp); return; diff --git a/fs/nfsd/nfs4idmap.c b/fs/nfsd/nfs4idmap.c index 4c0c683ce07..996bd88b75b 100644 --- a/fs/nfsd/nfs4idmap.c +++ b/fs/nfsd/nfs4idmap.c @@ -255,13 +255,10 @@ idtoname_parse(struct cache_detail *cd, char *buf, int buflen) goto out; if (len == 0) set_bit(CACHE_NEGATIVE, &ent.h.flags); - else { - if (error >= IDMAP_NAMESZ) { - error = -EINVAL; - goto out; - } + else if (len >= IDMAP_NAMESZ) + goto out; + else memcpy(ent.name, buf1, sizeof(ent.name)); - } error = -ENOMEM; res = idtoname_update(&ent, res); if (res == NULL) @@ -467,20 +464,25 @@ nametoid_update(struct ent *new, struct ent *old) * Exported API */ -void +int nfsd_idmap_init(void) { - cache_register(&idtoname_cache); - cache_register(&nametoid_cache); + int rv; + + rv = cache_register(&idtoname_cache); + if (rv) + return rv; + rv = cache_register(&nametoid_cache); + if (rv) + cache_unregister(&idtoname_cache); + return rv; } void nfsd_idmap_shutdown(void) { - if (cache_unregister(&idtoname_cache)) - printk(KERN_ERR "nfsd: failed to unregister idtoname cache\n"); - if (cache_unregister(&nametoid_cache)) - printk(KERN_ERR "nfsd: failed to unregister nametoid cache\n"); + cache_unregister(&idtoname_cache); + cache_unregister(&nametoid_cache); } /* diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 18ead1790bb..c593db047d8 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -750,7 +750,7 @@ _nfsd4_verify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, cstate->current_fh.fh_export, cstate->current_fh.fh_dentry, buf, &count, verify->ve_bmval, - rqstp); + rqstp, 0); /* this means that nfsd4_encode_fattr() ran out of space */ if (status == nfserr_resource && count == 0) diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 1602cd00dd4..1ff90625860 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -120,9 +120,9 @@ out_no_tfm: static void nfsd4_sync_rec_dir(void) { - mutex_lock(&rec_dir.dentry->d_inode->i_mutex); - nfsd_sync_dir(rec_dir.dentry); - mutex_unlock(&rec_dir.dentry->d_inode->i_mutex); + mutex_lock(&rec_dir.path.dentry->d_inode->i_mutex); + nfsd_sync_dir(rec_dir.path.dentry); + mutex_unlock(&rec_dir.path.dentry->d_inode->i_mutex); } int @@ -142,9 +142,9 @@ nfsd4_create_clid_dir(struct nfs4_client *clp) nfs4_save_user(&uid, &gid); /* lock the parent */ - mutex_lock(&rec_dir.dentry->d_inode->i_mutex); + mutex_lock(&rec_dir.path.dentry->d_inode->i_mutex); - dentry = lookup_one_len(dname, rec_dir.dentry, HEXDIR_LEN-1); + dentry = lookup_one_len(dname, rec_dir.path.dentry, HEXDIR_LEN-1); if (IS_ERR(dentry)) { status = PTR_ERR(dentry); goto out_unlock; @@ -154,11 +154,11 @@ nfsd4_create_clid_dir(struct nfs4_client *clp) dprintk("NFSD: nfsd4_create_clid_dir: DIRECTORY EXISTS\n"); goto out_put; } - status = vfs_mkdir(rec_dir.dentry->d_inode, dentry, S_IRWXU); + status = vfs_mkdir(rec_dir.path.dentry->d_inode, dentry, S_IRWXU); out_put: dput(dentry); out_unlock: - mutex_unlock(&rec_dir.dentry->d_inode->i_mutex); + mutex_unlock(&rec_dir.path.dentry->d_inode->i_mutex); if (status == 0) { clp->cl_firststate = 1; nfsd4_sync_rec_dir(); @@ -221,7 +221,7 @@ nfsd4_list_rec_dir(struct dentry *dir, recdir_func *f) nfs4_save_user(&uid, &gid); - filp = dentry_open(dget(dir), mntget(rec_dir.mnt), O_RDONLY); + filp = dentry_open(dget(dir), mntget(rec_dir.path.mnt), O_RDONLY); status = PTR_ERR(filp); if (IS_ERR(filp)) goto out; @@ -286,9 +286,9 @@ nfsd4_unlink_clid_dir(char *name, int namlen) dprintk("NFSD: nfsd4_unlink_clid_dir. name %.*s\n", namlen, name); - mutex_lock(&rec_dir.dentry->d_inode->i_mutex); - dentry = lookup_one_len(name, rec_dir.dentry, namlen); - mutex_unlock(&rec_dir.dentry->d_inode->i_mutex); + mutex_lock(&rec_dir.path.dentry->d_inode->i_mutex); + dentry = lookup_one_len(name, rec_dir.path.dentry, namlen); + mutex_unlock(&rec_dir.path.dentry->d_inode->i_mutex); if (IS_ERR(dentry)) { status = PTR_ERR(dentry); return status; @@ -297,7 +297,7 @@ nfsd4_unlink_clid_dir(char *name, int namlen) if (!dentry->d_inode) goto out; - status = nfsd4_clear_clid_dir(rec_dir.dentry, dentry); + status = nfsd4_clear_clid_dir(rec_dir.path.dentry, dentry); out: dput(dentry); return status; @@ -347,12 +347,12 @@ nfsd4_recdir_purge_old(void) { if (!rec_dir_init) return; - status = nfsd4_list_rec_dir(rec_dir.dentry, purge_old); + status = nfsd4_list_rec_dir(rec_dir.path.dentry, purge_old); if (status == 0) nfsd4_sync_rec_dir(); if (status) printk("nfsd4: failed to purge old clients from recovery" - " directory %s\n", rec_dir.dentry->d_name.name); + " directory %s\n", rec_dir.path.dentry->d_name.name); return; } @@ -373,10 +373,10 @@ int nfsd4_recdir_load(void) { int status; - status = nfsd4_list_rec_dir(rec_dir.dentry, load_recdir); + status = nfsd4_list_rec_dir(rec_dir.path.dentry, load_recdir); if (status) printk("nfsd4: failed loading clients from recovery" - " directory %s\n", rec_dir.dentry->d_name.name); + " directory %s\n", rec_dir.path.dentry->d_name.name); return status; } @@ -415,5 +415,5 @@ nfsd4_shutdown_recdir(void) if (!rec_dir_init) return; rec_dir_init = 0; - path_release(&rec_dir); + path_put(&rec_dir.path); } diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 31673cd251c..bcb97d8e8b8 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -61,7 +61,6 @@ static time_t lease_time = 90; /* default lease time */ static time_t user_lease_time = 90; static time_t boot_time; static int in_grace = 1; -static u32 current_clientid = 1; static u32 current_ownerid = 1; static u32 current_fileid = 1; static u32 current_delegid = 1; @@ -340,21 +339,20 @@ STALE_CLIENTID(clientid_t *clid) * This type of memory management is somewhat inefficient, but we use it * anyway since SETCLIENTID is not a common operation. */ -static inline struct nfs4_client * -alloc_client(struct xdr_netobj name) +static struct nfs4_client *alloc_client(struct xdr_netobj name) { struct nfs4_client *clp; - if ((clp = kzalloc(sizeof(struct nfs4_client), GFP_KERNEL))!= NULL) { - if ((clp->cl_name.data = kmalloc(name.len, GFP_KERNEL)) != NULL) { - memcpy(clp->cl_name.data, name.data, name.len); - clp->cl_name.len = name.len; - } - else { - kfree(clp); - clp = NULL; - } + clp = kzalloc(sizeof(struct nfs4_client), GFP_KERNEL); + if (clp == NULL) + return NULL; + clp->cl_name.data = kmalloc(name.len, GFP_KERNEL); + if (clp->cl_name.data == NULL) { + kfree(clp); + return NULL; } + memcpy(clp->cl_name.data, name.data, name.len); + clp->cl_name.len = name.len; return clp; } @@ -363,8 +361,11 @@ shutdown_callback_client(struct nfs4_client *clp) { struct rpc_clnt *clnt = clp->cl_callback.cb_client; - /* shutdown rpc client, ending any outstanding recall rpcs */ if (clnt) { + /* + * Callback threads take a reference on the client, so there + * should be no outstanding callbacks at this point. + */ clp->cl_callback.cb_client = NULL; rpc_shutdown_client(clnt); } @@ -422,12 +423,13 @@ expire_client(struct nfs4_client *clp) put_nfs4_client(clp); } -static struct nfs4_client * -create_client(struct xdr_netobj name, char *recdir) { +static struct nfs4_client *create_client(struct xdr_netobj name, char *recdir) +{ struct nfs4_client *clp; - if (!(clp = alloc_client(name))) - goto out; + clp = alloc_client(name); + if (clp == NULL) + return NULL; memcpy(clp->cl_recdir, recdir, HEXDIR_LEN); atomic_set(&clp->cl_count, 1); atomic_set(&clp->cl_callback.cb_set, 0); @@ -436,32 +438,30 @@ create_client(struct xdr_netobj name, char *recdir) { INIT_LIST_HEAD(&clp->cl_openowners); INIT_LIST_HEAD(&clp->cl_delegations); INIT_LIST_HEAD(&clp->cl_lru); -out: return clp; } -static void -copy_verf(struct nfs4_client *target, nfs4_verifier *source) { - memcpy(target->cl_verifier.data, source->data, sizeof(target->cl_verifier.data)); +static void copy_verf(struct nfs4_client *target, nfs4_verifier *source) +{ + memcpy(target->cl_verifier.data, source->data, + sizeof(target->cl_verifier.data)); } -static void -copy_clid(struct nfs4_client *target, struct nfs4_client *source) { +static void copy_clid(struct nfs4_client *target, struct nfs4_client *source) +{ target->cl_clientid.cl_boot = source->cl_clientid.cl_boot; target->cl_clientid.cl_id = source->cl_clientid.cl_id; } -static void -copy_cred(struct svc_cred *target, struct svc_cred *source) { - +static void copy_cred(struct svc_cred *target, struct svc_cred *source) +{ target->cr_uid = source->cr_uid; target->cr_gid = source->cr_gid; target->cr_group_info = source->cr_group_info; get_group_info(target->cr_group_info); } -static inline int -same_name(const char *n1, const char *n2) +static int same_name(const char *n1, const char *n2) { return 0 == memcmp(n1, n2, HEXDIR_LEN); } @@ -485,26 +485,26 @@ same_creds(struct svc_cred *cr1, struct svc_cred *cr2) return cr1->cr_uid == cr2->cr_uid; } -static void -gen_clid(struct nfs4_client *clp) { +static void gen_clid(struct nfs4_client *clp) +{ + static u32 current_clientid = 1; + clp->cl_clientid.cl_boot = boot_time; clp->cl_clientid.cl_id = current_clientid++; } -static void -gen_confirm(struct nfs4_client *clp) { - struct timespec tv; - u32 * p; +static void gen_confirm(struct nfs4_client *clp) +{ + static u32 i; + u32 *p; - tv = CURRENT_TIME; p = (u32 *)clp->cl_confirm.data; - *p++ = tv.tv_sec; - *p++ = tv.tv_nsec; + *p++ = get_seconds(); + *p++ = i++; } -static int -check_name(struct xdr_netobj name) { - +static int check_name(struct xdr_netobj name) +{ if (name.len == 0) return 0; if (name.len > NFS4_OPAQUE_LIMIT) { @@ -683,39 +683,6 @@ out_err: return; } -/* - * RFC 3010 has a complex implmentation description of processing a - * SETCLIENTID request consisting of 5 bullets, labeled as - * CASE0 - CASE4 below. - * - * NOTES: - * callback information will be processed in a future patch - * - * an unconfirmed record is added when: - * NORMAL (part of CASE 4): there is no confirmed nor unconfirmed record. - * CASE 1: confirmed record found with matching name, principal, - * verifier, and clientid. - * CASE 2: confirmed record found with matching name, principal, - * and there is no unconfirmed record with matching - * name and principal - * - * an unconfirmed record is replaced when: - * CASE 3: confirmed record found with matching name, principal, - * and an unconfirmed record is found with matching - * name, principal, and with clientid and - * confirm that does not match the confirmed record. - * CASE 4: there is no confirmed record with matching name and - * principal. there is an unconfirmed record with - * matching name, principal. - * - * an unconfirmed record is deleted when: - * CASE 1: an unconfirmed record that matches input name, verifier, - * and confirmed clientid. - * CASE 4: any unconfirmed records with matching name and principal - * that exist after an unconfirmed record has been replaced - * as described above. - * - */ __be32 nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_setclientid *setclid) @@ -748,11 +715,7 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, nfs4_lock_state(); conf = find_confirmed_client_by_str(dname, strhashval); if (conf) { - /* - * CASE 0: - * clname match, confirmed, different principal - * or different ip_address - */ + /* RFC 3530 14.2.33 CASE 0: */ status = nfserr_clid_inuse; if (!same_creds(&conf->cl_cred, &rqstp->rq_cred) || conf->cl_addr != sin->sin_addr.s_addr) { @@ -761,12 +724,17 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, goto out; } } + /* + * section 14.2.33 of RFC 3530 (under the heading "IMPLEMENTATION") + * has a description of SETCLIENTID request processing consisting + * of 5 bullet points, labeled as CASE0 - CASE4 below. + */ unconf = find_unconfirmed_client_by_str(dname, strhashval); status = nfserr_resource; if (!conf) { - /* - * CASE 4: - * placed first, because it is the normal case. + /* + * RFC 3530 14.2.33 CASE 4: + * placed first, because it is the normal case */ if (unconf) expire_client(unconf); @@ -776,17 +744,8 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, gen_clid(new); } else if (same_verf(&conf->cl_verifier, &clverifier)) { /* - * CASE 1: - * cl_name match, confirmed, principal match - * verifier match: probable callback update - * - * remove any unconfirmed nfs4_client with - * matching cl_name, cl_verifier, and cl_clientid - * - * create and insert an unconfirmed nfs4_client with same - * cl_name, cl_verifier, and cl_clientid as existing - * nfs4_client, but with the new callback info and a - * new cl_confirm + * RFC 3530 14.2.33 CASE 1: + * probable callback update */ if (unconf) { /* Note this is removing unconfirmed {*x***}, @@ -802,43 +761,25 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, copy_clid(new, conf); } else if (!unconf) { /* - * CASE 2: - * clname match, confirmed, principal match - * verfier does not match - * no unconfirmed. create a new unconfirmed nfs4_client - * using input clverifier, clname, and callback info - * and generate a new cl_clientid and cl_confirm. + * RFC 3530 14.2.33 CASE 2: + * probable client reboot; state will be removed if + * confirmed. */ new = create_client(clname, dname); if (new == NULL) goto out; gen_clid(new); - } else if (!same_verf(&conf->cl_confirm, &unconf->cl_confirm)) { - /* - * CASE3: - * confirmed found (name, principal match) - * confirmed verifier does not match input clverifier - * - * unconfirmed found (name match) - * confirmed->cl_confirm != unconfirmed->cl_confirm - * - * remove unconfirmed. - * - * create an unconfirmed nfs4_client - * with same cl_name as existing confirmed nfs4_client, - * but with new callback info, new cl_clientid, - * new cl_verifier and a new cl_confirm + } else { + /* + * RFC 3530 14.2.33 CASE 3: + * probable client reboot; state will be removed if + * confirmed. */ expire_client(unconf); new = create_client(clname, dname); if (new == NULL) goto out; gen_clid(new); - } else { - /* No cases hit !!! */ - status = nfserr_inval; - goto out; - } copy_verf(new, &clverifier); new->cl_addr = sin->sin_addr.s_addr; @@ -857,11 +798,9 @@ out: /* - * RFC 3010 has a complex implmentation description of processing a - * SETCLIENTID_CONFIRM request consisting of 4 bullets describing - * processing on a DRC miss, labeled as CASE1 - CASE4 below. - * - * NOTE: callback information will be processed here in a future patch + * Section 14.2.34 of RFC 3530 (under the heading "IMPLEMENTATION") has + * a description of SETCLIENTID_CONFIRM request processing consisting of 4 + * bullets, labeled as CASE1 - CASE4 below. */ __be32 nfsd4_setclientid_confirm(struct svc_rqst *rqstp, @@ -892,16 +831,16 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp, if (unconf && unconf->cl_addr != sin->sin_addr.s_addr) goto out; - if ((conf && unconf) && - (same_verf(&unconf->cl_confirm, &confirm)) && - (same_verf(&conf->cl_verifier, &unconf->cl_verifier)) && - (same_name(conf->cl_recdir,unconf->cl_recdir)) && - (!same_verf(&conf->cl_confirm, &unconf->cl_confirm))) { - /* CASE 1: - * unconf record that matches input clientid and input confirm. - * conf record that matches input clientid. - * conf and unconf records match names, verifiers - */ + /* + * section 14.2.34 of RFC 3530 has a description of + * SETCLIENTID_CONFIRM request processing consisting + * of 4 bullet points, labeled as CASE1 - CASE4 below. + */ + if (conf && unconf && same_verf(&confirm, &unconf->cl_confirm)) { + /* + * RFC 3530 14.2.34 CASE 1: + * callback update + */ if (!same_creds(&conf->cl_cred, &unconf->cl_cred)) status = nfserr_clid_inuse; else { @@ -914,15 +853,11 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp, status = nfs_ok; } - } else if ((conf && !unconf) || - ((conf && unconf) && - (!same_verf(&conf->cl_verifier, &unconf->cl_verifier) || - !same_name(conf->cl_recdir, unconf->cl_recdir)))) { - /* CASE 2: - * conf record that matches input clientid. - * if unconf record matches input clientid, then - * unconf->cl_name or unconf->cl_verifier don't match the - * conf record. + } else if (conf && !unconf) { + /* + * RFC 3530 14.2.34 CASE 2: + * probable retransmitted request; play it safe and + * do nothing. */ if (!same_creds(&conf->cl_cred, &rqstp->rq_cred)) status = nfserr_clid_inuse; @@ -930,10 +865,9 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp, status = nfs_ok; } else if (!conf && unconf && same_verf(&unconf->cl_confirm, &confirm)) { - /* CASE 3: - * conf record not found. - * unconf record found. - * unconf->cl_confirm matches input confirm + /* + * RFC 3530 14.2.34 CASE 3: + * Normal case; new or rebooted client: */ if (!same_creds(&unconf->cl_cred, &rqstp->rq_cred)) { status = nfserr_clid_inuse; @@ -948,16 +882,15 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp, } move_to_confirmed(unconf); conf = unconf; + nfsd4_probe_callback(conf); status = nfs_ok; } } else if ((!conf || (conf && !same_verf(&conf->cl_confirm, &confirm))) && (!unconf || (unconf && !same_verf(&unconf->cl_confirm, &confirm)))) { - /* CASE 4: - * conf record not found, or if conf, conf->cl_confirm does not - * match input confirm. - * unconf record not found, or if unconf, unconf->cl_confirm - * does not match input confirm. + /* + * RFC 3530 14.2.34 CASE 4: + * Client probably hasn't noticed that we rebooted yet. */ status = nfserr_stale_clientid; } else { @@ -965,8 +898,6 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp, status = nfserr_clid_inuse; } out: - if (!status) - nfsd4_probe_callback(conf); nfs4_unlock_state(); return status; } @@ -1226,14 +1157,19 @@ find_file(struct inode *ino) return NULL; } -static int access_valid(u32 x) +static inline int access_valid(u32 x) { - return (x > 0 && x < 4); + if (x < NFS4_SHARE_ACCESS_READ) + return 0; + if (x > NFS4_SHARE_ACCESS_BOTH) + return 0; + return 1; } -static int deny_valid(u32 x) +static inline int deny_valid(u32 x) { - return (x >= 0 && x < 5); + /* Note: unlike access bits, deny bits may be zero. */ + return x <= NFS4_SHARE_DENY_BOTH; } static void @@ -2162,8 +2098,10 @@ nfs4_preprocess_seqid_op(struct svc_fh *current_fh, u32 seqid, stateid_t *statei goto check_replay; } + *stpp = stp; + *sopp = sop = stp->st_stateowner; + if (lock) { - struct nfs4_stateowner *sop = stp->st_stateowner; clientid_t *lockclid = &lock->v.new.clientid; struct nfs4_client *clp = sop->so_client; int lkflg = 0; @@ -2193,9 +2131,6 @@ nfs4_preprocess_seqid_op(struct svc_fh *current_fh, u32 seqid, stateid_t *statei return nfserr_bad_stateid; } - *stpp = stp; - *sopp = sop = stp->st_stateowner; - /* * We now validate the seqid and stateid generation numbers. * For the moment, we ignore the possibility of @@ -3326,11 +3261,11 @@ nfs4_reset_recoverydir(char *recdir) if (status) return status; status = -ENOTDIR; - if (S_ISDIR(nd.dentry->d_inode->i_mode)) { + if (S_ISDIR(nd.path.dentry->d_inode->i_mode)) { nfs4_set_recdir(recdir); status = 0; } - path_release(&nd); + path_put(&nd.path); return status; } diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 57333944af7..0e6a179ecca 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -148,12 +148,12 @@ xdr_error: \ } \ } while (0) -static __be32 *read_buf(struct nfsd4_compoundargs *argp, int nbytes) +static __be32 *read_buf(struct nfsd4_compoundargs *argp, u32 nbytes) { /* We want more bytes than seem to be available. * Maybe we need a new page, maybe we have just run out */ - int avail = (char*)argp->end - (char*)argp->p; + unsigned int avail = (char *)argp->end - (char *)argp->p; __be32 *p; if (avail + argp->pagelen < nbytes) return NULL; @@ -169,6 +169,11 @@ static __be32 *read_buf(struct nfsd4_compoundargs *argp, int nbytes) return NULL; } + /* + * The following memcpy is safe because read_buf is always + * called with nbytes > avail, and the two cases above both + * guarantee p points to at least nbytes bytes. + */ memcpy(p, argp->p, avail); /* step to next page */ argp->p = page_address(argp->pagelist[0]); @@ -1325,9 +1330,9 @@ static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp, __be32 * *stat = exp_pseudoroot(rqstp, &tmp_fh); if (*stat) return NULL; - rootpath = tmp_fh.fh_export->ex_path; + rootpath = tmp_fh.fh_export->ex_pathname; - path = exp->ex_path; + path = exp->ex_pathname; if (strncmp(path, rootpath, strlen(rootpath))) { dprintk("nfsd: fs_locations failed;" @@ -1448,7 +1453,7 @@ static __be32 fattr_handle_absent_fs(u32 *bmval0, u32 *bmval1, u32 *rdattr_err) __be32 nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, __be32 *buffer, int *countp, u32 *bmval, - struct svc_rqst *rqstp) + struct svc_rqst *rqstp, int ignore_crossmnt) { u32 bmval0 = bmval[0]; u32 bmval1 = bmval[1]; @@ -1476,7 +1481,7 @@ nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, goto out; } - err = vfs_getattr(exp->ex_mnt, dentry, &stat); + err = vfs_getattr(exp->ex_path.mnt, dentry, &stat); if (err) goto out_nfserr; if ((bmval0 & (FATTR4_WORD0_FILES_FREE | FATTR4_WORD0_FILES_TOTAL | @@ -1828,9 +1833,14 @@ out_acl: if (bmval1 & FATTR4_WORD1_MOUNTED_ON_FILEID) { if ((buflen -= 8) < 0) goto out_resource; - if (exp->ex_mnt->mnt_root->d_inode == dentry->d_inode) { - err = vfs_getattr(exp->ex_mnt->mnt_parent, - exp->ex_mnt->mnt_mountpoint, &stat); + /* + * Get parent's attributes if not ignoring crossmount + * and this is the root of a cross-mounted filesystem. + */ + if (ignore_crossmnt == 0 && + exp->ex_path.mnt->mnt_root->d_inode == dentry->d_inode) { + err = vfs_getattr(exp->ex_path.mnt->mnt_parent, + exp->ex_path.mnt->mnt_mountpoint, &stat); if (err) goto out_nfserr; } @@ -1864,13 +1874,25 @@ nfsd4_encode_dirent_fattr(struct nfsd4_readdir *cd, struct svc_export *exp = cd->rd_fhp->fh_export; struct dentry *dentry; __be32 nfserr; + int ignore_crossmnt = 0; dentry = lookup_one_len(name, cd->rd_fhp->fh_dentry, namlen); if (IS_ERR(dentry)) return nfserrno(PTR_ERR(dentry)); exp_get(exp); - if (d_mountpoint(dentry)) { + /* + * In the case of a mountpoint, the client may be asking for + * attributes that are only properties of the underlying filesystem + * as opposed to the cross-mounted file system. In such a case, + * we will not follow the cross mount and will fill the attribtutes + * directly from the mountpoint dentry. + */ + if (d_mountpoint(dentry) && + (cd->rd_bmval[0] & ~FATTR4_WORD0_RDATTR_ERROR) == 0 && + (cd->rd_bmval[1] & ~FATTR4_WORD1_MOUNTED_ON_FILEID) == 0) + ignore_crossmnt = 1; + else if (d_mountpoint(dentry)) { int err; /* @@ -1889,7 +1911,7 @@ nfsd4_encode_dirent_fattr(struct nfsd4_readdir *cd, } nfserr = nfsd4_encode_fattr(NULL, exp, dentry, p, buflen, cd->rd_bmval, - cd->rd_rqstp); + cd->rd_rqstp, ignore_crossmnt); out_put: dput(dentry); exp_put(exp); @@ -2043,7 +2065,7 @@ nfsd4_encode_getattr(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4 buflen = resp->end - resp->p - (COMPOUND_ERR_SLACK_SPACE >> 2); nfserr = nfsd4_encode_fattr(fhp, fhp->fh_export, fhp->fh_dentry, resp->p, &buflen, getattr->ga_bmval, - resp->rqstp); + resp->rqstp, 0); if (!nfserr) resp->p += buflen; return nfserr; diff --git a/fs/nfsd/nfscache.c b/fs/nfsd/nfscache.c index 578f2c9d56b..5bfc2ac60d5 100644 --- a/fs/nfsd/nfscache.c +++ b/fs/nfsd/nfscache.c @@ -44,17 +44,17 @@ static int nfsd_cache_append(struct svc_rqst *rqstp, struct kvec *vec); */ static DEFINE_SPINLOCK(cache_lock); -void -nfsd_cache_init(void) +int nfsd_reply_cache_init(void) { struct svc_cacherep *rp; int i; INIT_LIST_HEAD(&lru_head); i = CACHESIZE; - while(i) { + while (i) { rp = kmalloc(sizeof(*rp), GFP_KERNEL); - if (!rp) break; + if (!rp) + goto out_nomem; list_add(&rp->c_lru, &lru_head); rp->c_state = RC_UNUSED; rp->c_type = RC_NOCACHE; @@ -62,23 +62,19 @@ nfsd_cache_init(void) i--; } - if (i) - printk (KERN_ERR "nfsd: cannot allocate all %d cache entries, only got %d\n", - CACHESIZE, CACHESIZE-i); - hash_list = kcalloc (HASHSIZE, sizeof(struct hlist_head), GFP_KERNEL); - if (!hash_list) { - nfsd_cache_shutdown(); - printk (KERN_ERR "nfsd: cannot allocate %Zd bytes for hash list\n", - HASHSIZE * sizeof(struct hlist_head)); - return; - } + if (!hash_list) + goto out_nomem; cache_disabled = 0; + return 0; +out_nomem: + printk(KERN_ERR "nfsd: failed to allocate reply cache\n"); + nfsd_reply_cache_shutdown(); + return -ENOMEM; } -void -nfsd_cache_shutdown(void) +void nfsd_reply_cache_shutdown(void) { struct svc_cacherep *rp; diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 77dc9893b7b..8516137cdbb 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -304,6 +304,9 @@ static ssize_t write_filehandle(struct file *file, char *buf, size_t size) struct auth_domain *dom; struct knfsd_fh fh; + if (size == 0) + return -EINVAL; + if (buf[size-1] != '\n') return -EINVAL; buf[size-1] = 0; @@ -503,7 +506,7 @@ static ssize_t write_ports(struct file *file, char *buf, size_t size) int len = 0; lock_kernel(); if (nfsd_serv) - len = svc_sock_names(buf, nfsd_serv, NULL); + len = svc_xprt_names(nfsd_serv, buf, 0); unlock_kernel(); return len; } @@ -540,7 +543,7 @@ static ssize_t write_ports(struct file *file, char *buf, size_t size) } return err < 0 ? err : 0; } - if (buf[0] == '-') { + if (buf[0] == '-' && isdigit(buf[1])) { char *toclose = kstrdup(buf+1, GFP_KERNEL); int len = 0; if (!toclose) @@ -554,6 +557,53 @@ static ssize_t write_ports(struct file *file, char *buf, size_t size) kfree(toclose); return len; } + /* + * Add a transport listener by writing it's transport name + */ + if (isalpha(buf[0])) { + int err; + char transport[16]; + int port; + if (sscanf(buf, "%15s %4d", transport, &port) == 2) { + err = nfsd_create_serv(); + if (!err) { + err = svc_create_xprt(nfsd_serv, + transport, port, + SVC_SOCK_ANONYMOUS); + if (err == -ENOENT) + /* Give a reasonable perror msg for + * bad transport string */ + err = -EPROTONOSUPPORT; + } + return err < 0 ? err : 0; + } + } + /* + * Remove a transport by writing it's transport name and port number + */ + if (buf[0] == '-' && isalpha(buf[1])) { + struct svc_xprt *xprt; + int err = -EINVAL; + char transport[16]; + int port; + if (sscanf(&buf[1], "%15s %4d", transport, &port) == 2) { + if (port == 0) + return -EINVAL; + lock_kernel(); + if (nfsd_serv) { + xprt = svc_find_xprt(nfsd_serv, transport, + AF_UNSPEC, port); + if (xprt) { + svc_close_xprt(xprt); + svc_xprt_put(xprt); + err = 0; + } else + err = -ENOTCONN; + } + unlock_kernel(); + return err < 0 ? err : 0; + } + } return -EINVAL; } @@ -616,7 +666,7 @@ static ssize_t write_recoverydir(struct file *file, char *buf, size_t size) char *recdir; int len, status; - if (size > PATH_MAX || buf[size-1] != '\n') + if (size == 0 || size > PATH_MAX || buf[size-1] != '\n') return -EINVAL; buf[size-1] = 0; @@ -674,6 +724,27 @@ static struct file_system_type nfsd_fs_type = { .kill_sb = kill_litter_super, }; +#ifdef CONFIG_PROC_FS +static int create_proc_exports_entry(void) +{ + struct proc_dir_entry *entry; + + entry = proc_mkdir("fs/nfs", NULL); + if (!entry) + return -ENOMEM; + entry = create_proc_entry("fs/nfs/exports", 0, NULL); + if (!entry) + return -ENOMEM; + entry->proc_fops = &exports_operations; + return 0; +} +#else /* CONFIG_PROC_FS */ +static int create_proc_exports_entry(void) +{ + return 0; +} +#endif + static int __init init_nfsd(void) { int retval; @@ -683,32 +754,43 @@ static int __init init_nfsd(void) if (retval) return retval; nfsd_stat_init(); /* Statistics */ - nfsd_cache_init(); /* RPC reply cache */ - nfsd_export_init(); /* Exports table */ + retval = nfsd_reply_cache_init(); + if (retval) + goto out_free_stat; + retval = nfsd_export_init(); + if (retval) + goto out_free_cache; nfsd_lockd_init(); /* lockd->nfsd callbacks */ - nfsd_idmap_init(); /* Name to ID mapping */ - if (proc_mkdir("fs/nfs", NULL)) { - struct proc_dir_entry *entry; - entry = create_proc_entry("fs/nfs/exports", 0, NULL); - if (entry) - entry->proc_fops = &exports_operations; - } + retval = nfsd_idmap_init(); + if (retval) + goto out_free_lockd; + retval = create_proc_exports_entry(); + if (retval) + goto out_free_idmap; retval = register_filesystem(&nfsd_fs_type); - if (retval) { - nfsd_export_shutdown(); - nfsd_cache_shutdown(); - remove_proc_entry("fs/nfs/exports", NULL); - remove_proc_entry("fs/nfs", NULL); - nfsd_stat_shutdown(); - nfsd_lockd_shutdown(); - } + if (retval) + goto out_free_all; + return 0; +out_free_all: + remove_proc_entry("fs/nfs/exports", NULL); + remove_proc_entry("fs/nfs", NULL); +out_free_idmap: + nfsd_idmap_shutdown(); +out_free_lockd: + nfsd_lockd_shutdown(); + nfsd_export_shutdown(); +out_free_cache: + nfsd_reply_cache_shutdown(); +out_free_stat: + nfsd_stat_shutdown(); + nfsd4_free_slabs(); return retval; } static void __exit exit_nfsd(void) { nfsd_export_shutdown(); - nfsd_cache_shutdown(); + nfsd_reply_cache_shutdown(); remove_proc_entry("fs/nfs/exports", NULL); remove_proc_entry("fs/nfs", NULL); nfsd_stat_shutdown(); diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index 468f17a7844..3e6b3f41ee1 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -22,6 +22,7 @@ #include <linux/sunrpc/svc.h> #include <linux/sunrpc/svcauth_gss.h> #include <linux/nfsd/nfsd.h> +#include "auth.h" #define NFSDDBG_FACILITY NFSDDBG_FH @@ -46,7 +47,7 @@ static int nfsd_acceptable(void *expv, struct dentry *dentry) return 1; tdentry = dget(dentry); - while (tdentry != exp->ex_dentry && ! IS_ROOT(tdentry)) { + while (tdentry != exp->ex_path.dentry && !IS_ROOT(tdentry)) { /* make sure parents give x permission to user */ int err; parent = dget_parent(tdentry); @@ -58,9 +59,9 @@ static int nfsd_acceptable(void *expv, struct dentry *dentry) dput(tdentry); tdentry = parent; } - if (tdentry != exp->ex_dentry) + if (tdentry != exp->ex_path.dentry) dprintk("nfsd_acceptable failed at %p %s\n", tdentry, tdentry->d_name.name); - rv = (tdentry == exp->ex_dentry); + rv = (tdentry == exp->ex_path.dentry); dput(tdentry); return rv; } @@ -100,7 +101,7 @@ static __be32 nfsd_setuser_and_check_port(struct svc_rqst *rqstp, { /* Check if the request originated from a secure port. */ if (!rqstp->rq_secure && EX_SECURE(exp)) { - char buf[RPC_MAX_ADDRBUFLEN]; + RPC_IFDEBUG(char buf[RPC_MAX_ADDRBUFLEN]); dprintk(KERN_WARNING "nfsd: request from insecure port %s!\n", svc_print_addr(rqstp, buf, sizeof(buf))); @@ -208,9 +209,9 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) fileid_type = fh->fh_fileid_type; if (fileid_type == FILEID_ROOT) - dentry = dget(exp->ex_dentry); + dentry = dget(exp->ex_path.dentry); else { - dentry = exportfs_decode_fh(exp->ex_mnt, fid, + dentry = exportfs_decode_fh(exp->ex_path.mnt, fid, data_left, fileid_type, nfsd_acceptable, exp); } @@ -231,6 +232,7 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) fhp->fh_dentry = dentry; fhp->fh_export = exp; nfsd_nr_verified++; + cache_get(&exp->h); } else { /* * just rechecking permissions @@ -240,6 +242,7 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) dprintk("nfsd: fh_verify - just checking\n"); dentry = fhp->fh_dentry; exp = fhp->fh_export; + cache_get(&exp->h); /* * Set user creds for this exportpoint; necessary even * in the "just checking" case because this may be a @@ -251,8 +254,6 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) if (error) goto out; } - cache_get(&exp->h); - error = nfsd_mode_check(rqstp, dentry->d_inode->i_mode, type); if (error) @@ -298,7 +299,7 @@ out: static void _fh_update(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry) { - if (dentry != exp->ex_dentry) { + if (dentry != exp->ex_path.dentry) { struct fid *fid = (struct fid *) (fhp->fh_handle.fh_auth + fhp->fh_handle.fh_size/4 - 1); int maxsize = (fhp->fh_maxsize - fhp->fh_handle.fh_size)/4; @@ -343,12 +344,12 @@ fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, struct inode * inode = dentry->d_inode; struct dentry *parent = dentry->d_parent; __u32 *datap; - dev_t ex_dev = exp->ex_dentry->d_inode->i_sb->s_dev; - int root_export = (exp->ex_dentry == exp->ex_dentry->d_sb->s_root); + dev_t ex_dev = exp->ex_path.dentry->d_inode->i_sb->s_dev; + int root_export = (exp->ex_path.dentry == exp->ex_path.dentry->d_sb->s_root); dprintk("nfsd: fh_compose(exp %02x:%02x/%ld %s/%s, ino=%ld)\n", MAJOR(ex_dev), MINOR(ex_dev), - (long) exp->ex_dentry->d_inode->i_ino, + (long) exp->ex_path.dentry->d_inode->i_ino, parent->d_name.name, dentry->d_name.name, (inode ? inode->i_ino : 0)); @@ -390,7 +391,7 @@ fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, /* FALL THROUGH */ case FSID_MAJOR_MINOR: case FSID_ENCODE_DEV: - if (!(exp->ex_dentry->d_inode->i_sb->s_type->fs_flags + if (!(exp->ex_path.dentry->d_inode->i_sb->s_type->fs_flags & FS_REQUIRES_DEV)) goto retry; break; @@ -453,7 +454,7 @@ fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, fhp->fh_handle.ofh_dev = old_encode_dev(ex_dev); fhp->fh_handle.ofh_xdev = fhp->fh_handle.ofh_dev; fhp->fh_handle.ofh_xino = - ino_t_to_u32(exp->ex_dentry->d_inode->i_ino); + ino_t_to_u32(exp->ex_path.dentry->d_inode->i_ino); fhp->fh_handle.ofh_dirino = ino_t_to_u32(parent_ino(dentry)); if (inode) _fh_update_old(dentry, exp, &fhp->fh_handle); @@ -464,7 +465,7 @@ fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, datap = fhp->fh_handle.fh_auth+0; fhp->fh_handle.fh_fsid_type = fsid_type; mk_fsid(fsid_type, datap, ex_dev, - exp->ex_dentry->d_inode->i_ino, + exp->ex_path.dentry->d_inode->i_ino, exp->ex_fsid, exp->ex_uuid); len = key_len(fsid_type); @@ -570,7 +571,7 @@ enum fsid_source fsid_source(struct svc_fh *fhp) case FSID_DEV: case FSID_ENCODE_DEV: case FSID_MAJOR_MINOR: - if (fhp->fh_export->ex_dentry->d_inode->i_sb->s_type->fs_flags + if (fhp->fh_export->ex_path.dentry->d_inode->i_sb->s_type->fs_flags & FS_REQUIRES_DEV) return FSIDSOURCE_DEV; break; diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c index 977a71f64e1..6cfc96a1248 100644 --- a/fs/nfsd/nfsproc.c +++ b/fs/nfsd/nfsproc.c @@ -41,7 +41,7 @@ static __be32 nfsd_return_attrs(__be32 err, struct nfsd_attrstat *resp) { if (err) return err; - return nfserrno(vfs_getattr(resp->fh.fh_export->ex_mnt, + return nfserrno(vfs_getattr(resp->fh.fh_export->ex_path.mnt, resp->fh.fh_dentry, &resp->stat)); } @@ -49,7 +49,7 @@ static __be32 nfsd_return_dirop(__be32 err, struct nfsd_diropres *resp) { if (err) return err; - return nfserrno(vfs_getattr(resp->fh.fh_export->ex_mnt, + return nfserrno(vfs_getattr(resp->fh.fh_export->ex_path.mnt, resp->fh.fh_dentry, &resp->stat)); } @@ -164,7 +164,7 @@ nfsd_proc_read(struct svc_rqst *rqstp, struct nfsd_readargs *argp, &resp->count); if (nfserr) return nfserr; - return nfserrno(vfs_getattr(resp->fh.fh_export->ex_mnt, + return nfserrno(vfs_getattr(resp->fh.fh_export->ex_path.mnt, resp->fh.fh_dentry, &resp->stat)); } diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 1190aeaa92b..9647b0f7bc0 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -155,8 +155,8 @@ static int killsig; /* signal that was used to kill last nfsd */ static void nfsd_last_thread(struct svc_serv *serv) { /* When last nfsd thread exits we need to do some clean-up */ - struct svc_sock *svsk; - list_for_each_entry(svsk, &serv->sv_permsocks, sk_list) + struct svc_xprt *xprt; + list_for_each_entry(xprt, &serv->sv_permsocks, xpt_list) lockd_down(); nfsd_serv = NULL; nfsd_racache_shutdown(); @@ -236,7 +236,7 @@ static int nfsd_init_socks(int port) error = lockd_up(IPPROTO_UDP); if (error >= 0) { - error = svc_makesock(nfsd_serv, IPPROTO_UDP, port, + error = svc_create_xprt(nfsd_serv, "udp", port, SVC_SOCK_DEFAULTS); if (error < 0) lockd_down(); @@ -247,7 +247,7 @@ static int nfsd_init_socks(int port) #ifdef CONFIG_NFSD_TCP error = lockd_up(IPPROTO_TCP); if (error >= 0) { - error = svc_makesock(nfsd_serv, IPPROTO_TCP, port, + error = svc_create_xprt(nfsd_serv, "tcp", port, SVC_SOCK_DEFAULTS); if (error < 0) lockd_down(); diff --git a/fs/nfsd/nfsxdr.c b/fs/nfsd/nfsxdr.c index b86e3658a0a..afd08e2c90a 100644 --- a/fs/nfsd/nfsxdr.c +++ b/fs/nfsd/nfsxdr.c @@ -15,6 +15,7 @@ #include <linux/nfsd/nfsd.h> #include <linux/nfsd/xdr.h> #include <linux/mm.h> +#include "auth.h" #define NFSDDBG_FACILITY NFSDDBG_XDR @@ -62,10 +63,10 @@ encode_fh(__be32 *p, struct svc_fh *fhp) * no slashes or null bytes. */ static __be32 * -decode_filename(__be32 *p, char **namp, int *lenp) +decode_filename(__be32 *p, char **namp, unsigned int *lenp) { char *name; - int i; + unsigned int i; if ((p = xdr_decode_string_inplace(p, namp, lenp, NFS_MAXNAMLEN)) != NULL) { for (i = 0, name = *namp; i < *lenp; i++, name++) { @@ -78,10 +79,10 @@ decode_filename(__be32 *p, char **namp, int *lenp) } static __be32 * -decode_pathname(__be32 *p, char **namp, int *lenp) +decode_pathname(__be32 *p, char **namp, unsigned int *lenp) { char *name; - int i; + unsigned int i; if ((p = xdr_decode_string_inplace(p, namp, lenp, NFS_MAXPATHLEN)) != NULL) { for (i = 0, name = *namp; i < *lenp; i++, name++) { @@ -206,7 +207,7 @@ encode_fattr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp, __be32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { struct kstat stat; - vfs_getattr(fhp->fh_export->ex_mnt, fhp->fh_dentry, &stat); + vfs_getattr(fhp->fh_export->ex_path.mnt, fhp->fh_dentry, &stat); return encode_fattr(rqstp, p, fhp, &stat); } diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index d0199189924..46f59d5365a 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -101,7 +101,7 @@ nfsd_cross_mnt(struct svc_rqst *rqstp, struct dentry **dpp, { struct svc_export *exp = *expp, *exp2 = NULL; struct dentry *dentry = *dpp; - struct vfsmount *mnt = mntget(exp->ex_mnt); + struct vfsmount *mnt = mntget(exp->ex_path.mnt); struct dentry *mounts = dget(dentry); int err = 0; @@ -132,7 +132,7 @@ out: __be32 nfsd_lookup_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp, - const char *name, int len, + const char *name, unsigned int len, struct svc_export **exp_ret, struct dentry **dentry_ret) { struct svc_export *exp; @@ -156,15 +156,15 @@ nfsd_lookup_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp, if (isdotent(name, len)) { if (len==1) dentry = dget(dparent); - else if (dparent != exp->ex_dentry) { + else if (dparent != exp->ex_path.dentry) dentry = dget_parent(dparent); - } else if (!EX_NOHIDE(exp)) + else if (!EX_NOHIDE(exp)) dentry = dget(dparent); /* .. == . just like at / */ else { /* checking mountpoint crossing is very different when stepping up */ struct svc_export *exp2 = NULL; struct dentry *dp; - struct vfsmount *mnt = mntget(exp->ex_mnt); + struct vfsmount *mnt = mntget(exp->ex_path.mnt); dentry = dget(dparent); while(dentry == mnt->mnt_root && follow_up(&mnt, &dentry)) ; @@ -226,7 +226,7 @@ out_nfserr: */ __be32 nfsd_lookup(struct svc_rqst *rqstp, struct svc_fh *fhp, const char *name, - int len, struct svc_fh *resfh) + unsigned int len, struct svc_fh *resfh) { struct svc_export *exp; struct dentry *dentry; @@ -721,7 +721,8 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, DQUOT_INIT(inode); } - *filp = dentry_open(dget(dentry), mntget(fhp->fh_export->ex_mnt), flags); + *filp = dentry_open(dget(dentry), mntget(fhp->fh_export->ex_path.mnt), + flags); if (IS_ERR(*filp)) host_err = PTR_ERR(*filp); out_nfserr: @@ -1151,6 +1152,26 @@ nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, } #endif /* CONFIG_NFSD_V3 */ +__be32 +nfsd_create_setattr(struct svc_rqst *rqstp, struct svc_fh *resfhp, + struct iattr *iap) +{ + /* + * Mode has already been set earlier in create: + */ + iap->ia_valid &= ~ATTR_MODE; + /* + * Setting uid/gid works only for root. Irix appears to + * send along the gid on create when it tries to implement + * setgid directories via NFS: + */ + if (current->fsuid != 0) + iap->ia_valid &= ~(ATTR_UID|ATTR_GID); + if (iap->ia_valid) + return nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); + return 0; +} + /* * Create a file (regular, directory, device, fifo); UNIX sockets * not yet implemented. @@ -1167,6 +1188,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, struct dentry *dentry, *dchild = NULL; struct inode *dirp; __be32 err; + __be32 err2; int host_err; err = nfserr_perm; @@ -1257,16 +1279,9 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, } - /* Set file attributes. Mode has already been set and - * setting uid/gid works only for root. Irix appears to - * send along the gid when it tries to implement setgid - * directories via NFS. - */ - if ((iap->ia_valid &= ~(ATTR_UID|ATTR_GID|ATTR_MODE)) != 0) { - __be32 err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); - if (err2) - err = err2; - } + err2 = nfsd_create_setattr(rqstp, resfhp, iap); + if (err2) + err = err2; /* * Update the file handle to get the new inode info. */ @@ -1295,6 +1310,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, struct dentry *dentry, *dchild = NULL; struct inode *dirp; __be32 err; + __be32 err2; int host_err; __u32 v_mtime=0, v_atime=0; @@ -1399,16 +1415,10 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, iap->ia_atime.tv_nsec = 0; } - /* Set file attributes. - * Irix appears to send along the gid when it tries to - * implement setgid directories via NFS. Clear out all that cruft. - */ set_attr: - if ((iap->ia_valid &= ~(ATTR_UID|ATTR_GID|ATTR_MODE)) != 0) { - __be32 err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); - if (err2) - err = err2; - } + err2 = nfsd_create_setattr(rqstp, resfhp, iap); + if (err2) + err = err2; /* * Update the filehandle to get the new inode info. @@ -1453,7 +1463,7 @@ nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp) if (!inode->i_op || !inode->i_op->readlink) goto out; - touch_atime(fhp->fh_export->ex_mnt, dentry); + touch_atime(fhp->fh_export->ex_path.mnt, dentry); /* N.B. Why does this call need a get_fs()?? * Remove the set_fs and watch the fireworks:-) --okir */ diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c index ad87cb01299..00e9ccde8e4 100644 --- a/fs/ntfs/aops.c +++ b/fs/ntfs/aops.c @@ -87,13 +87,17 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate) /* Check for the current buffer head overflowing. */ if (unlikely(file_ofs + bh->b_size > init_size)) { int ofs; + void *kaddr; ofs = 0; if (file_ofs < init_size) ofs = init_size - file_ofs; local_irq_save(flags); - zero_user_page(page, bh_offset(bh) + ofs, - bh->b_size - ofs, KM_BIO_SRC_IRQ); + kaddr = kmap_atomic(page, KM_BIO_SRC_IRQ); + memset(kaddr + bh_offset(bh) + ofs, 0, + bh->b_size - ofs); + flush_dcache_page(page); + kunmap_atomic(kaddr, KM_BIO_SRC_IRQ); local_irq_restore(flags); } } else { @@ -334,7 +338,7 @@ handle_hole: bh->b_blocknr = -1UL; clear_buffer_mapped(bh); handle_zblock: - zero_user_page(page, i * blocksize, blocksize, KM_USER0); + zero_user(page, i * blocksize, blocksize); if (likely(!err)) set_buffer_uptodate(bh); } while (i++, iblock++, (bh = bh->b_this_page) != head); @@ -410,7 +414,7 @@ retry_readpage: /* Is the page fully outside i_size? (truncate in progress) */ if (unlikely(page->index >= (i_size + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT)) { - zero_user_page(page, 0, PAGE_CACHE_SIZE, KM_USER0); + zero_user(page, 0, PAGE_CACHE_SIZE); ntfs_debug("Read outside i_size - truncated?"); goto done; } @@ -459,7 +463,7 @@ retry_readpage: * ok to ignore the compressed flag here. */ if (unlikely(page->index > 0)) { - zero_user_page(page, 0, PAGE_CACHE_SIZE, KM_USER0); + zero_user(page, 0, PAGE_CACHE_SIZE); goto done; } if (!NInoAttr(ni)) @@ -788,8 +792,7 @@ lock_retry_remap: if (err == -ENOENT || lcn == LCN_ENOENT) { bh->b_blocknr = -1; clear_buffer_dirty(bh); - zero_user_page(page, bh_offset(bh), blocksize, - KM_USER0); + zero_user(page, bh_offset(bh), blocksize); set_buffer_uptodate(bh); err = 0; continue; @@ -1414,8 +1417,7 @@ retry_writepage: if (page->index >= (i_size >> PAGE_CACHE_SHIFT)) { /* The page straddles i_size. */ unsigned int ofs = i_size & ~PAGE_CACHE_MASK; - zero_user_page(page, ofs, PAGE_CACHE_SIZE - ofs, - KM_USER0); + zero_user_segment(page, ofs, PAGE_CACHE_SIZE); } /* Handle mst protected attributes. */ if (NInoMstProtected(ni)) diff --git a/fs/ntfs/compress.c b/fs/ntfs/compress.c index d1619d05eb2..33ff314cc50 100644 --- a/fs/ntfs/compress.c +++ b/fs/ntfs/compress.c @@ -565,7 +565,7 @@ int ntfs_read_compressed_block(struct page *page) if (xpage >= max_page) { kfree(bhs); kfree(pages); - zero_user_page(page, 0, PAGE_CACHE_SIZE, KM_USER0); + zero_user(page, 0, PAGE_CACHE_SIZE); ntfs_debug("Compressed read outside i_size - truncated?"); SetPageUptodate(page); unlock_page(page); diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c index 6cd08dfdc2e..3c5550cd11d 100644 --- a/fs/ntfs/file.c +++ b/fs/ntfs/file.c @@ -607,8 +607,8 @@ do_next_page: ntfs_submit_bh_for_read(bh); *wait_bh++ = bh; } else { - zero_user_page(page, bh_offset(bh), - blocksize, KM_USER0); + zero_user(page, bh_offset(bh), + blocksize); set_buffer_uptodate(bh); } } @@ -683,9 +683,8 @@ map_buffer_cached: ntfs_submit_bh_for_read(bh); *wait_bh++ = bh; } else { - zero_user_page(page, - bh_offset(bh), - blocksize, KM_USER0); + zero_user(page, bh_offset(bh), + blocksize); set_buffer_uptodate(bh); } } @@ -703,8 +702,8 @@ map_buffer_cached: */ if (bh_end <= pos || bh_pos >= end) { if (!buffer_uptodate(bh)) { - zero_user_page(page, bh_offset(bh), - blocksize, KM_USER0); + zero_user(page, bh_offset(bh), + blocksize); set_buffer_uptodate(bh); } mark_buffer_dirty(bh); @@ -743,8 +742,7 @@ map_buffer_cached: if (!buffer_uptodate(bh)) set_buffer_uptodate(bh); } else if (!buffer_uptodate(bh)) { - zero_user_page(page, bh_offset(bh), blocksize, - KM_USER0); + zero_user(page, bh_offset(bh), blocksize); set_buffer_uptodate(bh); } continue; @@ -868,8 +866,8 @@ rl_not_mapped_enoent: if (!buffer_uptodate(bh)) set_buffer_uptodate(bh); } else if (!buffer_uptodate(bh)) { - zero_user_page(page, bh_offset(bh), - blocksize, KM_USER0); + zero_user(page, bh_offset(bh), + blocksize); set_buffer_uptodate(bh); } continue; @@ -1128,8 +1126,8 @@ rl_not_mapped_enoent: if (likely(bh_pos < initialized_size)) ofs = initialized_size - bh_pos; - zero_user_page(page, bh_offset(bh) + ofs, - blocksize - ofs, KM_USER0); + zero_user_segment(page, bh_offset(bh) + ofs, + blocksize); } } else /* if (unlikely(!buffer_uptodate(bh))) */ err = -EIO; @@ -1269,8 +1267,8 @@ rl_not_mapped_enoent: if (PageUptodate(page)) set_buffer_uptodate(bh); else { - zero_user_page(page, bh_offset(bh), - blocksize, KM_USER0); + zero_user(page, bh_offset(bh), + blocksize); set_buffer_uptodate(bh); } } @@ -1330,7 +1328,7 @@ err_out: len = PAGE_CACHE_SIZE; if (len > bytes) len = bytes; - zero_user_page(*pages, 0, len, KM_USER0); + zero_user(*pages, 0, len); } goto out; } @@ -1451,7 +1449,7 @@ err_out: len = PAGE_CACHE_SIZE; if (len > bytes) len = bytes; - zero_user_page(*pages, 0, len, KM_USER0); + zero_user(*pages, 0, len); } goto out; } diff --git a/fs/ntfs/malloc.h b/fs/ntfs/malloc.h index e38e402e410..cd0be3f5c3c 100644 --- a/fs/ntfs/malloc.h +++ b/fs/ntfs/malloc.h @@ -85,8 +85,7 @@ static inline void *ntfs_malloc_nofs_nofail(unsigned long size) static inline void ntfs_free(void *addr) { - if (likely(((unsigned long)addr < VMALLOC_START) || - ((unsigned long)addr >= VMALLOC_END ))) { + if (!is_vmalloc_addr(addr)) { kfree(addr); /* free_page((unsigned long)addr); */ return; diff --git a/fs/ocfs2/Makefile b/fs/ocfs2/Makefile index 9fb8132f19b..4d4ce48bb42 100644 --- a/fs/ocfs2/Makefile +++ b/fs/ocfs2/Makefile @@ -19,16 +19,17 @@ ocfs2-objs := \ ioctl.o \ journal.o \ localalloc.o \ + locks.o \ mmap.o \ namei.o \ + resize.o \ slot_map.o \ suballoc.o \ super.o \ symlink.o \ sysfile.o \ uptodate.o \ - ver.o \ - vote.o + ver.o obj-$(CONFIG_OCFS2_FS) += cluster/ obj-$(CONFIG_OCFS2_FS) += dlm/ diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index 23c8cda43f1..447206eb5c2 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -3338,7 +3338,7 @@ static int ocfs2_insert_path(struct inode *inode, if (insert->ins_split != SPLIT_NONE) { /* * We could call ocfs2_insert_at_leaf() for some types - * of splits, but it's easier to just let one seperate + * of splits, but it's easier to just let one separate * function sort it all out. */ ocfs2_split_record(inode, left_path, right_path, @@ -4731,7 +4731,7 @@ int __ocfs2_flush_truncate_log(struct ocfs2_super *osb) mutex_lock(&data_alloc_inode->i_mutex); - status = ocfs2_meta_lock(data_alloc_inode, &data_alloc_bh, 1); + status = ocfs2_inode_lock(data_alloc_inode, &data_alloc_bh, 1); if (status < 0) { mlog_errno(status); goto out_mutex; @@ -4753,7 +4753,7 @@ int __ocfs2_flush_truncate_log(struct ocfs2_super *osb) out_unlock: brelse(data_alloc_bh); - ocfs2_meta_unlock(data_alloc_inode, 1); + ocfs2_inode_unlock(data_alloc_inode, 1); out_mutex: mutex_unlock(&data_alloc_inode->i_mutex); @@ -5077,7 +5077,7 @@ static int ocfs2_free_cached_items(struct ocfs2_super *osb, mutex_lock(&inode->i_mutex); - ret = ocfs2_meta_lock(inode, &di_bh, 1); + ret = ocfs2_inode_lock(inode, &di_bh, 1); if (ret) { mlog_errno(ret); goto out_mutex; @@ -5118,7 +5118,7 @@ out_journal: ocfs2_commit_trans(osb, handle); out_unlock: - ocfs2_meta_unlock(inode, 1); + ocfs2_inode_unlock(inode, 1); brelse(di_bh); out_mutex: mutex_unlock(&inode->i_mutex); @@ -5670,7 +5670,7 @@ static void ocfs2_map_and_dirty_page(struct inode *inode, handle_t *handle, mlog_errno(ret); if (zero) - zero_user_page(page, from, to - from, KM_USER0); + zero_user_segment(page, from, to); /* * Need to set the buffers we zero'd into uptodate diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index 56f7790cad4..90383ed6100 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -26,6 +26,7 @@ #include <asm/byteorder.h> #include <linux/swap.h> #include <linux/pipe_fs_i.h> +#include <linux/mpage.h> #define MLOG_MASK_PREFIX ML_FILE_IO #include <cluster/masklog.h> @@ -139,7 +140,8 @@ static int ocfs2_get_block(struct inode *inode, sector_t iblock, { int err = 0; unsigned int ext_flags; - u64 p_blkno, past_eof; + u64 max_blocks = bh_result->b_size >> inode->i_blkbits; + u64 p_blkno, count, past_eof; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); mlog_entry("(0x%p, %llu, 0x%p, %d)\n", inode, @@ -155,7 +157,7 @@ static int ocfs2_get_block(struct inode *inode, sector_t iblock, goto bail; } - err = ocfs2_extent_map_get_blocks(inode, iblock, &p_blkno, NULL, + err = ocfs2_extent_map_get_blocks(inode, iblock, &p_blkno, &count, &ext_flags); if (err) { mlog(ML_ERROR, "Error %d from get_blocks(0x%p, %llu, 1, " @@ -164,6 +166,9 @@ static int ocfs2_get_block(struct inode *inode, sector_t iblock, goto bail; } + if (max_blocks < count) + count = max_blocks; + /* * ocfs2 never allocates in this function - the only time we * need to use BH_New is when we're extending i_size on a file @@ -178,6 +183,8 @@ static int ocfs2_get_block(struct inode *inode, sector_t iblock, if (p_blkno && !(ext_flags & OCFS2_EXT_UNWRITTEN)) map_bh(bh_result, inode->i_sb, p_blkno); + bh_result->b_size = count << inode->i_blkbits; + if (!ocfs2_sparse_alloc(osb)) { if (p_blkno == 0) { err = -EIO; @@ -210,7 +217,7 @@ int ocfs2_read_inline_data(struct inode *inode, struct page *page, struct buffer_head *di_bh) { void *kaddr; - unsigned int size; + loff_t size; struct ocfs2_dinode *di = (struct ocfs2_dinode *)di_bh->b_data; if (!(le16_to_cpu(di->i_dyn_features) & OCFS2_INLINE_DATA_FL)) { @@ -224,8 +231,9 @@ int ocfs2_read_inline_data(struct inode *inode, struct page *page, if (size > PAGE_CACHE_SIZE || size > ocfs2_max_inline_data(inode->i_sb)) { ocfs2_error(inode->i_sb, - "Inode %llu has with inline data has bad size: %u", - (unsigned long long)OCFS2_I(inode)->ip_blkno, size); + "Inode %llu has with inline data has bad size: %Lu", + (unsigned long long)OCFS2_I(inode)->ip_blkno, + (unsigned long long)size); return -EROFS; } @@ -249,7 +257,7 @@ static int ocfs2_readpage_inline(struct inode *inode, struct page *page) struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); BUG_ON(!PageLocked(page)); - BUG_ON(!OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL); + BUG_ON(!(OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL)); ret = ocfs2_read_block(osb, OCFS2_I(inode)->ip_blkno, &di_bh, OCFS2_BH_CACHED, inode); @@ -275,7 +283,7 @@ static int ocfs2_readpage(struct file *file, struct page *page) mlog_entry("(0x%p, %lu)\n", file, (page ? page->index : 0)); - ret = ocfs2_meta_lock_with_page(inode, NULL, 0, page); + ret = ocfs2_inode_lock_with_page(inode, NULL, 0, page); if (ret != 0) { if (ret == AOP_TRUNCATED_PAGE) unlock = 0; @@ -285,7 +293,7 @@ static int ocfs2_readpage(struct file *file, struct page *page) if (down_read_trylock(&oi->ip_alloc_sem) == 0) { ret = AOP_TRUNCATED_PAGE; - goto out_meta_unlock; + goto out_inode_unlock; } /* @@ -299,31 +307,22 @@ static int ocfs2_readpage(struct file *file, struct page *page) * XXX sys_readahead() seems to get that wrong? */ if (start >= i_size_read(inode)) { - zero_user_page(page, 0, PAGE_SIZE, KM_USER0); + zero_user(page, 0, PAGE_SIZE); SetPageUptodate(page); ret = 0; goto out_alloc; } - ret = ocfs2_data_lock_with_page(inode, 0, page); - if (ret != 0) { - if (ret == AOP_TRUNCATED_PAGE) - unlock = 0; - mlog_errno(ret); - goto out_alloc; - } - if (oi->ip_dyn_features & OCFS2_INLINE_DATA_FL) ret = ocfs2_readpage_inline(inode, page); else ret = block_read_full_page(page, ocfs2_get_block); unlock = 0; - ocfs2_data_unlock(inode, 0); out_alloc: up_read(&OCFS2_I(inode)->ip_alloc_sem); -out_meta_unlock: - ocfs2_meta_unlock(inode, 0); +out_inode_unlock: + ocfs2_inode_unlock(inode, 0); out: if (unlock) unlock_page(page); @@ -331,6 +330,62 @@ out: return ret; } +/* + * This is used only for read-ahead. Failures or difficult to handle + * situations are safe to ignore. + * + * Right now, we don't bother with BH_Boundary - in-inode extent lists + * are quite large (243 extents on 4k blocks), so most inodes don't + * grow out to a tree. If need be, detecting boundary extents could + * trivially be added in a future version of ocfs2_get_block(). + */ +static int ocfs2_readpages(struct file *filp, struct address_space *mapping, + struct list_head *pages, unsigned nr_pages) +{ + int ret, err = -EIO; + struct inode *inode = mapping->host; + struct ocfs2_inode_info *oi = OCFS2_I(inode); + loff_t start; + struct page *last; + + /* + * Use the nonblocking flag for the dlm code to avoid page + * lock inversion, but don't bother with retrying. + */ + ret = ocfs2_inode_lock_full(inode, NULL, 0, OCFS2_LOCK_NONBLOCK); + if (ret) + return err; + + if (down_read_trylock(&oi->ip_alloc_sem) == 0) { + ocfs2_inode_unlock(inode, 0); + return err; + } + + /* + * Don't bother with inline-data. There isn't anything + * to read-ahead in that case anyway... + */ + if (oi->ip_dyn_features & OCFS2_INLINE_DATA_FL) + goto out_unlock; + + /* + * Check whether a remote node truncated this file - we just + * drop out in that case as it's not worth handling here. + */ + last = list_entry(pages->prev, struct page, lru); + start = (loff_t)last->index << PAGE_CACHE_SHIFT; + if (start >= i_size_read(inode)) + goto out_unlock; + + err = mpage_readpages(mapping, pages, nr_pages, ocfs2_get_block); + +out_unlock: + up_read(&oi->ip_alloc_sem); + ocfs2_inode_unlock(inode, 0); + + return err; +} + /* Note: Because we don't support holes, our allocation has * already happened (allocation writes zeros to the file data) * so we don't have to worry about ordered writes in @@ -452,7 +507,7 @@ static sector_t ocfs2_bmap(struct address_space *mapping, sector_t block) * accessed concurrently from multiple nodes. */ if (!INODE_JOURNAL(inode)) { - err = ocfs2_meta_lock(inode, NULL, 0); + err = ocfs2_inode_lock(inode, NULL, 0); if (err) { if (err != -ENOENT) mlog_errno(err); @@ -467,7 +522,7 @@ static sector_t ocfs2_bmap(struct address_space *mapping, sector_t block) if (!INODE_JOURNAL(inode)) { up_read(&OCFS2_I(inode)->ip_alloc_sem); - ocfs2_meta_unlock(inode, 0); + ocfs2_inode_unlock(inode, 0); } if (err) { @@ -638,34 +693,12 @@ static ssize_t ocfs2_direct_IO(int rw, if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) return 0; - if (!ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb))) { - /* - * We get PR data locks even for O_DIRECT. This - * allows concurrent O_DIRECT I/O but doesn't let - * O_DIRECT with extending and buffered zeroing writes - * race. If they did race then the buffered zeroing - * could be written back after the O_DIRECT I/O. It's - * one thing to tell people not to mix buffered and - * O_DIRECT writes, but expecting them to understand - * that file extension is also an implicit buffered - * write is too much. By getting the PR we force - * writeback of the buffered zeroing before - * proceeding. - */ - ret = ocfs2_data_lock(inode, 0); - if (ret < 0) { - mlog_errno(ret); - goto out; - } - ocfs2_data_unlock(inode, 0); - } - ret = blockdev_direct_IO_no_locking(rw, iocb, inode, inode->i_sb->s_bdev, iov, offset, nr_segs, ocfs2_direct_IO_get_blocks, ocfs2_dio_end_io); -out: + mlog_exit(ret); return ret; } @@ -836,7 +869,7 @@ int ocfs2_map_page_blocks(struct page *page, u64 *p_blkno, if (block_start >= to) break; - zero_user_page(page, block_start, bh->b_size, KM_USER0); + zero_user(page, block_start, bh->b_size); set_buffer_uptodate(bh); mark_buffer_dirty(bh); @@ -1001,7 +1034,7 @@ static void ocfs2_zero_new_buffers(struct page *page, unsigned from, unsigned to start = max(from, block_start); end = min(to, block_end); - zero_user_page(page, start, end - start, KM_USER0); + zero_user_segment(page, start, end); set_buffer_uptodate(bh); } @@ -1754,7 +1787,7 @@ static int ocfs2_write_begin(struct file *file, struct address_space *mapping, struct buffer_head *di_bh = NULL; struct inode *inode = mapping->host; - ret = ocfs2_meta_lock(inode, &di_bh, 1); + ret = ocfs2_inode_lock(inode, &di_bh, 1); if (ret) { mlog_errno(ret); return ret; @@ -1769,30 +1802,22 @@ static int ocfs2_write_begin(struct file *file, struct address_space *mapping, */ down_write(&OCFS2_I(inode)->ip_alloc_sem); - ret = ocfs2_data_lock(inode, 1); - if (ret) { - mlog_errno(ret); - goto out_fail; - } - ret = ocfs2_write_begin_nolock(mapping, pos, len, flags, pagep, fsdata, di_bh, NULL); if (ret) { mlog_errno(ret); - goto out_fail_data; + goto out_fail; } brelse(di_bh); return 0; -out_fail_data: - ocfs2_data_unlock(inode, 1); out_fail: up_write(&OCFS2_I(inode)->ip_alloc_sem); brelse(di_bh); - ocfs2_meta_unlock(inode, 1); + ocfs2_inode_unlock(inode, 1); return ret; } @@ -1908,15 +1933,15 @@ static int ocfs2_write_end(struct file *file, struct address_space *mapping, ret = ocfs2_write_end_nolock(mapping, pos, len, copied, page, fsdata); - ocfs2_data_unlock(inode, 1); up_write(&OCFS2_I(inode)->ip_alloc_sem); - ocfs2_meta_unlock(inode, 1); + ocfs2_inode_unlock(inode, 1); return ret; } const struct address_space_operations ocfs2_aops = { .readpage = ocfs2_readpage, + .readpages = ocfs2_readpages, .writepage = ocfs2_writepage, .write_begin = ocfs2_write_begin, .write_end = ocfs2_write_end, diff --git a/fs/ocfs2/buffer_head_io.c b/fs/ocfs2/buffer_head_io.c index c9037414f4f..f136639f5b4 100644 --- a/fs/ocfs2/buffer_head_io.c +++ b/fs/ocfs2/buffer_head_io.c @@ -79,7 +79,7 @@ int ocfs2_write_block(struct ocfs2_super *osb, struct buffer_head *bh, * information for this bh as it's not marked locally * uptodate. */ ret = -EIO; - brelse(bh); + put_bh(bh); } mutex_unlock(&OCFS2_I(inode)->ip_io_mutex); @@ -256,7 +256,7 @@ int ocfs2_read_blocks(struct ocfs2_super *osb, u64 block, int nr, * for this bh as it's not marked locally * uptodate. */ status = -EIO; - brelse(bh); + put_bh(bh); bhs[i] = NULL; continue; } @@ -280,3 +280,64 @@ bail: mlog_exit(status); return status; } + +/* Check whether the blkno is the super block or one of the backups. */ +static void ocfs2_check_super_or_backup(struct super_block *sb, + sector_t blkno) +{ + int i; + u64 backup_blkno; + + if (blkno == OCFS2_SUPER_BLOCK_BLKNO) + return; + + for (i = 0; i < OCFS2_MAX_BACKUP_SUPERBLOCKS; i++) { + backup_blkno = ocfs2_backup_super_blkno(sb, i); + if (backup_blkno == blkno) + return; + } + + BUG(); +} + +/* + * Write super block and backups doesn't need to collaborate with journal, + * so we don't need to lock ip_io_mutex and inode doesn't need to bea passed + * into this function. + */ +int ocfs2_write_super_or_backup(struct ocfs2_super *osb, + struct buffer_head *bh) +{ + int ret = 0; + + mlog_entry_void(); + + BUG_ON(buffer_jbd(bh)); + ocfs2_check_super_or_backup(osb->sb, bh->b_blocknr); + + if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb)) { + ret = -EROFS; + goto out; + } + + lock_buffer(bh); + set_buffer_uptodate(bh); + + /* remove from dirty list before I/O. */ + clear_buffer_dirty(bh); + + get_bh(bh); /* for end_buffer_write_sync() */ + bh->b_end_io = end_buffer_write_sync; + submit_bh(WRITE, bh); + + wait_on_buffer(bh); + + if (!buffer_uptodate(bh)) { + ret = -EIO; + put_bh(bh); + } + +out: + mlog_exit(ret); + return ret; +} diff --git a/fs/ocfs2/buffer_head_io.h b/fs/ocfs2/buffer_head_io.h index 6cc20930fac..c2e78614c3e 100644 --- a/fs/ocfs2/buffer_head_io.h +++ b/fs/ocfs2/buffer_head_io.h @@ -47,6 +47,8 @@ int ocfs2_read_blocks(struct ocfs2_super *osb, int flags, struct inode *inode); +int ocfs2_write_super_or_backup(struct ocfs2_super *osb, + struct buffer_head *bh); #define OCFS2_BH_CACHED 1 #define OCFS2_BH_READAHEAD 8 diff --git a/fs/ocfs2/cluster/heartbeat.h b/fs/ocfs2/cluster/heartbeat.h index 35397dd5ecd..e511339886b 100644 --- a/fs/ocfs2/cluster/heartbeat.h +++ b/fs/ocfs2/cluster/heartbeat.h @@ -35,7 +35,7 @@ #define O2HB_LIVE_THRESHOLD 2 /* number of equal samples to be seen as dead */ extern unsigned int o2hb_dead_threshold; -#define O2HB_DEFAULT_DEAD_THRESHOLD 7 +#define O2HB_DEFAULT_DEAD_THRESHOLD 31 /* Otherwise MAX_WRITE_TIMEOUT will be zero... */ #define O2HB_MIN_DEAD_THRESHOLD 2 #define O2HB_MAX_WRITE_TIMEOUT_MS (O2HB_REGION_TIMEOUT_MS * (o2hb_dead_threshold - 1)) diff --git a/fs/ocfs2/cluster/masklog.c b/fs/ocfs2/cluster/masklog.c index a4882c8df94..23c732f2752 100644 --- a/fs/ocfs2/cluster/masklog.c +++ b/fs/ocfs2/cluster/masklog.c @@ -146,7 +146,7 @@ static struct kset mlog_kset = { .kobj = {.ktype = &mlog_ktype}, }; -int mlog_sys_init(struct kset *o2cb_subsys) +int mlog_sys_init(struct kset *o2cb_kset) { int i = 0; @@ -157,7 +157,7 @@ int mlog_sys_init(struct kset *o2cb_subsys) mlog_attr_ptrs[i] = NULL; kobject_set_name(&mlog_kset.kobj, "logmask"); - kobj_set_kset_s(&mlog_kset, *o2cb_subsys); + mlog_kset.kobj.kset = o2cb_kset; return kset_register(&mlog_kset); } diff --git a/fs/ocfs2/cluster/nodemanager.c b/fs/ocfs2/cluster/nodemanager.c index af2070da308..709fba25bf7 100644 --- a/fs/ocfs2/cluster/nodemanager.c +++ b/fs/ocfs2/cluster/nodemanager.c @@ -24,7 +24,6 @@ #include <linux/sysctl.h> #include <linux/configfs.h> -#include "endian.h" #include "tcp.h" #include "nodemanager.h" #include "heartbeat.h" diff --git a/fs/ocfs2/cluster/sys.c b/fs/ocfs2/cluster/sys.c index 64f6f378fd0..0c095ce7723 100644 --- a/fs/ocfs2/cluster/sys.c +++ b/fs/ocfs2/cluster/sys.c @@ -28,96 +28,55 @@ #include <linux/module.h> #include <linux/kobject.h> #include <linux/sysfs.h> +#include <linux/fs.h> #include "ocfs2_nodemanager.h" #include "masklog.h" #include "sys.h" -struct o2cb_attribute { - struct attribute attr; - ssize_t (*show)(char *buf); - ssize_t (*store)(const char *buf, size_t count); -}; - -#define O2CB_ATTR(_name, _mode, _show, _store) \ -struct o2cb_attribute o2cb_attr_##_name = __ATTR(_name, _mode, _show, _store) - -#define to_o2cb_attr(_attr) container_of(_attr, struct o2cb_attribute, attr) -static ssize_t o2cb_interface_revision_show(char *buf) +static ssize_t version_show(struct kobject *kobj, struct kobj_attribute *attr, + char *buf) { return snprintf(buf, PAGE_SIZE, "%u\n", O2NM_API_VERSION); } - -static O2CB_ATTR(interface_revision, S_IFREG | S_IRUGO, o2cb_interface_revision_show, NULL); +static struct kobj_attribute attr_version = + __ATTR(interface_revision, S_IFREG | S_IRUGO, version_show, NULL); static struct attribute *o2cb_attrs[] = { - &o2cb_attr_interface_revision.attr, + &attr_version.attr, NULL, }; -static ssize_t -o2cb_show(struct kobject * kobj, struct attribute * attr, char * buffer); -static ssize_t -o2cb_store(struct kobject * kobj, struct attribute * attr, - const char * buffer, size_t count); -static struct sysfs_ops o2cb_sysfs_ops = { - .show = o2cb_show, - .store = o2cb_store, +static struct attribute_group o2cb_attr_group = { + .attrs = o2cb_attrs, }; -static struct kobj_type o2cb_subsys_type = { - .default_attrs = o2cb_attrs, - .sysfs_ops = &o2cb_sysfs_ops, -}; - -/* gives us o2cb_subsys */ -static decl_subsys(o2cb, NULL, NULL); - -static ssize_t -o2cb_show(struct kobject * kobj, struct attribute * attr, char * buffer) -{ - struct o2cb_attribute *o2cb_attr = to_o2cb_attr(attr); - struct kset *sbs = to_kset(kobj); - - BUG_ON(sbs != &o2cb_subsys); - - if (o2cb_attr->show) - return o2cb_attr->show(buffer); - return -EIO; -} - -static ssize_t -o2cb_store(struct kobject * kobj, struct attribute * attr, - const char * buffer, size_t count) -{ - struct o2cb_attribute *o2cb_attr = to_o2cb_attr(attr); - struct kset *sbs = to_kset(kobj); - - BUG_ON(sbs != &o2cb_subsys); - - if (o2cb_attr->store) - return o2cb_attr->store(buffer, count); - return -EIO; -} +static struct kset *o2cb_kset; void o2cb_sys_shutdown(void) { mlog_sys_shutdown(); - subsystem_unregister(&o2cb_subsys); + kset_unregister(o2cb_kset); } int o2cb_sys_init(void) { int ret; - o2cb_subsys.kobj.ktype = &o2cb_subsys_type; - ret = subsystem_register(&o2cb_subsys); + o2cb_kset = kset_create_and_add("o2cb", NULL, NULL); + if (!o2cb_kset) + return -ENOMEM; + + ret = sysfs_create_group(&o2cb_kset->kobj, &o2cb_attr_group); if (ret) - return ret; + goto error; - ret = mlog_sys_init(&o2cb_subsys); + ret = mlog_sys_init(o2cb_kset); if (ret) - subsystem_unregister(&o2cb_subsys); + goto error; + return 0; +error: + kset_unregister(o2cb_kset); return ret; } diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c index ee50c9610e7..b8057c51b20 100644 --- a/fs/ocfs2/cluster/tcp.c +++ b/fs/ocfs2/cluster/tcp.c @@ -451,9 +451,9 @@ static void o2net_set_nn_state(struct o2net_node *nn, /* delay if we're withing a RECONNECT_DELAY of the * last attempt */ delay = (nn->nn_last_connect_attempt + - msecs_to_jiffies(o2net_reconnect_delay(sc->sc_node))) + msecs_to_jiffies(o2net_reconnect_delay(NULL))) - jiffies; - if (delay > msecs_to_jiffies(o2net_reconnect_delay(sc->sc_node))) + if (delay > msecs_to_jiffies(o2net_reconnect_delay(NULL))) delay = 0; mlog(ML_CONN, "queueing conn attempt in %lu jiffies\n", delay); queue_delayed_work(o2net_wq, &nn->nn_connect_work, delay); @@ -1552,12 +1552,11 @@ static void o2net_connect_expired(struct work_struct *work) spin_lock(&nn->nn_lock); if (!nn->nn_sc_valid) { - struct o2nm_node *node = nn->nn_sc->sc_node; mlog(ML_ERROR, "no connection established with node %u after " "%u.%u seconds, giving up and returning errors.\n", o2net_num_from_nn(nn), - o2net_idle_timeout(node) / 1000, - o2net_idle_timeout(node) % 1000); + o2net_idle_timeout(NULL) / 1000, + o2net_idle_timeout(NULL) % 1000); o2net_set_nn_state(nn, NULL, 0, -ENOTCONN); } diff --git a/fs/ocfs2/cluster/tcp.h b/fs/ocfs2/cluster/tcp.h index da880fc215f..f36f66aab3d 100644 --- a/fs/ocfs2/cluster/tcp.h +++ b/fs/ocfs2/cluster/tcp.h @@ -60,8 +60,8 @@ typedef void (o2net_post_msg_handler_func)(int status, void *data, /* same as hb delay, we're waiting for another node to recognize our hb */ #define O2NET_RECONNECT_DELAY_MS_DEFAULT 2000 -#define O2NET_KEEPALIVE_DELAY_MS_DEFAULT 5000 -#define O2NET_IDLE_TIMEOUT_MS_DEFAULT 10000 +#define O2NET_KEEPALIVE_DELAY_MS_DEFAULT 2000 +#define O2NET_IDLE_TIMEOUT_MS_DEFAULT 30000 /* TODO: figure this out.... */ diff --git a/fs/ocfs2/cluster/tcp_internal.h b/fs/ocfs2/cluster/tcp_internal.h index 9606111fe89..d25b9af2850 100644 --- a/fs/ocfs2/cluster/tcp_internal.h +++ b/fs/ocfs2/cluster/tcp_internal.h @@ -38,6 +38,21 @@ * locking semantics of the file system using the protocol. It should * be somewhere else, I'm sure, but right now it isn't. * + * With version 11, we separate out the filesystem locking portion. The + * filesystem now has a major.minor version it negotiates. Version 11 + * introduces this negotiation to the o2dlm protocol, and as such the + * version here in tcp_internal.h should not need to be bumped for + * filesystem locking changes. + * + * New in version 11 + * - Negotiation of filesystem locking in the dlm join. + * + * New in version 10: + * - Meta/data locks combined + * + * New in version 9: + * - All votes removed + * * New in version 8: * - Replace delete inode votes with a cluster lock * @@ -60,7 +75,7 @@ * - full 64 bit i_size in the metadata lock lvbs * - introduction of "rw" lock and pushing meta/data locking down */ -#define O2NET_PROTOCOL_VERSION 8ULL +#define O2NET_PROTOCOL_VERSION 11ULL struct o2net_handshake { __be64 protocol_version; __be64 connector_id; diff --git a/fs/ocfs2/cluster/ver.c b/fs/ocfs2/cluster/ver.c index 7286c48bb30..a56eee6abad 100644 --- a/fs/ocfs2/cluster/ver.c +++ b/fs/ocfs2/cluster/ver.c @@ -28,7 +28,7 @@ #include "ver.h" -#define CLUSTER_BUILD_VERSION "1.3.3" +#define CLUSTER_BUILD_VERSION "1.5.0" #define VERSION_STR "OCFS2 Node Manager " CLUSTER_BUILD_VERSION diff --git a/fs/ocfs2/dcache.c b/fs/ocfs2/dcache.c index 9923278ea6d..b1cc7c381e8 100644 --- a/fs/ocfs2/dcache.c +++ b/fs/ocfs2/dcache.c @@ -128,9 +128,9 @@ static int ocfs2_match_dentry(struct dentry *dentry, /* * Walk the inode alias list, and find a dentry which has a given * parent. ocfs2_dentry_attach_lock() wants to find _any_ alias as it - * is looking for a dentry_lock reference. The vote thread is looking - * to unhash aliases, so we allow it to skip any that already have - * that property. + * is looking for a dentry_lock reference. The downconvert thread is + * looking to unhash aliases, so we allow it to skip any that already + * have that property. */ struct dentry *ocfs2_find_local_alias(struct inode *inode, u64 parent_blkno, @@ -266,7 +266,7 @@ int ocfs2_dentry_attach_lock(struct dentry *dentry, dl->dl_count = 0; /* * Does this have to happen below, for all attaches, in case - * the struct inode gets blown away by votes? + * the struct inode gets blown away by the downconvert thread? */ dl->dl_inode = igrab(inode); dl->dl_parent_blkno = parent_blkno; diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c index 63b28fdceb4..8a187584808 100644 --- a/fs/ocfs2/dir.c +++ b/fs/ocfs2/dir.c @@ -390,9 +390,8 @@ static int __ocfs2_delete_entry(handle_t *handle, struct inode *dir, goto bail; } if (pde) - pde->rec_len = - cpu_to_le16(le16_to_cpu(pde->rec_len) + - le16_to_cpu(de->rec_len)); + le16_add_cpu(&pde->rec_len, + le16_to_cpu(de->rec_len)); else de->inode = 0; dir->i_version++; @@ -846,14 +845,14 @@ int ocfs2_readdir(struct file * filp, void * dirent, filldir_t filldir) mlog_entry("dirino=%llu\n", (unsigned long long)OCFS2_I(inode)->ip_blkno); - error = ocfs2_meta_lock_atime(inode, filp->f_vfsmnt, &lock_level); + error = ocfs2_inode_lock_atime(inode, filp->f_vfsmnt, &lock_level); if (lock_level && error >= 0) { /* We release EX lock which used to update atime * and get PR lock again to reduce contention * on commonly accessed directories. */ - ocfs2_meta_unlock(inode, 1); + ocfs2_inode_unlock(inode, 1); lock_level = 0; - error = ocfs2_meta_lock(inode, NULL, 0); + error = ocfs2_inode_lock(inode, NULL, 0); } if (error < 0) { if (error != -ENOENT) @@ -865,7 +864,7 @@ int ocfs2_readdir(struct file * filp, void * dirent, filldir_t filldir) error = ocfs2_dir_foreach_blk(inode, &filp->f_version, &filp->f_pos, dirent, filldir, NULL); - ocfs2_meta_unlock(inode, lock_level); + ocfs2_inode_unlock(inode, lock_level); bail_nolock: mlog_exit(error); @@ -1215,7 +1214,7 @@ static int ocfs2_expand_inline_dir(struct inode *dir, struct buffer_head *di_bh, down_write(&oi->ip_alloc_sem); /* - * Prepare for worst case allocation scenario of two seperate + * Prepare for worst case allocation scenario of two separate * extents. */ if (alloc == 2) diff --git a/fs/ocfs2/dlm/dlmapi.h b/fs/ocfs2/dlm/dlmapi.h index cfd5cb65cab..b5786a787fa 100644 --- a/fs/ocfs2/dlm/dlmapi.h +++ b/fs/ocfs2/dlm/dlmapi.h @@ -193,7 +193,12 @@ enum dlm_status dlmunlock(struct dlm_ctxt *dlm, dlm_astunlockfunc_t *unlockast, void *data); -struct dlm_ctxt * dlm_register_domain(const char *domain, u32 key); +struct dlm_protocol_version { + u8 pv_major; + u8 pv_minor; +}; +struct dlm_ctxt * dlm_register_domain(const char *domain, u32 key, + struct dlm_protocol_version *fs_proto); void dlm_unregister_domain(struct dlm_ctxt *dlm); diff --git a/fs/ocfs2/dlm/dlmast.c b/fs/ocfs2/dlm/dlmast.c index 2fd8bded38f..644bee55d8b 100644 --- a/fs/ocfs2/dlm/dlmast.c +++ b/fs/ocfs2/dlm/dlmast.c @@ -43,7 +43,6 @@ #include "cluster/heartbeat.h" #include "cluster/nodemanager.h" #include "cluster/tcp.h" -#include "cluster/endian.h" #include "dlmapi.h" #include "dlmcommon.h" diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h index e90b92f9ece..dc8ea666efd 100644 --- a/fs/ocfs2/dlm/dlmcommon.h +++ b/fs/ocfs2/dlm/dlmcommon.h @@ -142,6 +142,12 @@ struct dlm_ctxt spinlock_t work_lock; struct list_head dlm_domain_handlers; struct list_head dlm_eviction_callbacks; + + /* The filesystem specifies this at domain registration. We + * cache it here to know what to tell other nodes. */ + struct dlm_protocol_version fs_locking_proto; + /* This is the inter-dlm communication version */ + struct dlm_protocol_version dlm_locking_proto; }; static inline struct hlist_head *dlm_lockres_hash(struct dlm_ctxt *dlm, unsigned i) @@ -170,6 +176,7 @@ struct dlm_mig_lockres_priv { struct dlm_lock_resource *lockres; u8 real_master; + u8 extra_ref; }; struct dlm_assert_master_priv @@ -589,10 +596,26 @@ struct dlm_proxy_ast #define DLM_PROXY_AST_MAX_LEN (sizeof(struct dlm_proxy_ast)+DLM_LVB_LEN) #define DLM_MOD_KEY (0x666c6172) -enum dlm_query_join_response { +enum dlm_query_join_response_code { JOIN_DISALLOW = 0, JOIN_OK, JOIN_OK_NO_MAP, + JOIN_PROTOCOL_MISMATCH, +}; + +struct dlm_query_join_packet { + u8 code; /* Response code. dlm_minor and fs_minor + are only valid if this is JOIN_OK */ + u8 dlm_minor; /* The minor version of the protocol the + dlm is speaking. */ + u8 fs_minor; /* The minor version of the protocol the + filesystem is speaking. */ + u8 reserved; +}; + +union dlm_query_join_response { + u32 intval; + struct dlm_query_join_packet packet; }; struct dlm_lock_request @@ -633,6 +656,8 @@ struct dlm_query_join_request u8 node_idx; u8 pad1[2]; u8 name_len; + struct dlm_protocol_version dlm_proto; + struct dlm_protocol_version fs_proto; u8 domain[O2NM_MAX_NAME_LEN]; u8 node_map[BITS_TO_BYTES(O2NM_MAX_NODES)]; }; diff --git a/fs/ocfs2/dlm/dlmconvert.c b/fs/ocfs2/dlm/dlmconvert.c index ecb4d997221..75997b4deaf 100644 --- a/fs/ocfs2/dlm/dlmconvert.c +++ b/fs/ocfs2/dlm/dlmconvert.c @@ -487,7 +487,7 @@ int dlm_convert_lock_handler(struct o2net_msg *msg, u32 len, void *data, "cookie=%u:%llu\n", dlm_get_lock_cookie_node(be64_to_cpu(cnv->cookie)), dlm_get_lock_cookie_seq(be64_to_cpu(cnv->cookie))); - __dlm_print_one_lock_resource(res); + dlm_print_one_lock_resource(res); goto leave; } diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index 6954565b8cc..0879d86113e 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -123,6 +123,17 @@ DEFINE_SPINLOCK(dlm_domain_lock); LIST_HEAD(dlm_domains); static DECLARE_WAIT_QUEUE_HEAD(dlm_domain_events); +/* + * The supported protocol version for DLM communication. Running domains + * will have a negotiated version with the same major number and a minor + * number equal or smaller. The dlm_ctxt->dlm_locking_proto field should + * be used to determine what a running domain is actually using. + */ +static const struct dlm_protocol_version dlm_protocol = { + .pv_major = 1, + .pv_minor = 0, +}; + #define DLM_DOMAIN_BACKOFF_MS 200 static int dlm_query_join_handler(struct o2net_msg *msg, u32 len, void *data, @@ -133,6 +144,8 @@ static int dlm_cancel_join_handler(struct o2net_msg *msg, u32 len, void *data, void **ret_data); static int dlm_exit_domain_handler(struct o2net_msg *msg, u32 len, void *data, void **ret_data); +static int dlm_protocol_compare(struct dlm_protocol_version *existing, + struct dlm_protocol_version *request); static void dlm_unregister_domain_handlers(struct dlm_ctxt *dlm); @@ -668,12 +681,78 @@ void dlm_unregister_domain(struct dlm_ctxt *dlm) } EXPORT_SYMBOL_GPL(dlm_unregister_domain); +static int dlm_query_join_proto_check(char *proto_type, int node, + struct dlm_protocol_version *ours, + struct dlm_protocol_version *request) +{ + int rc; + struct dlm_protocol_version proto = *request; + + if (!dlm_protocol_compare(ours, &proto)) { + mlog(0, + "node %u wanted to join with %s locking protocol " + "%u.%u, we respond with %u.%u\n", + node, proto_type, + request->pv_major, + request->pv_minor, + proto.pv_major, proto.pv_minor); + request->pv_minor = proto.pv_minor; + rc = 0; + } else { + mlog(ML_NOTICE, + "Node %u wanted to join with %s locking " + "protocol %u.%u, but we have %u.%u, disallowing\n", + node, proto_type, + request->pv_major, + request->pv_minor, + ours->pv_major, + ours->pv_minor); + rc = 1; + } + + return rc; +} + +/* + * struct dlm_query_join_packet is made up of four one-byte fields. They + * are effectively in big-endian order already. However, little-endian + * machines swap them before putting the packet on the wire (because + * query_join's response is a status, and that status is treated as a u32 + * on the wire). Thus, a big-endian and little-endian machines will treat + * this structure differently. + * + * The solution is to have little-endian machines swap the structure when + * converting from the structure to the u32 representation. This will + * result in the structure having the correct format on the wire no matter + * the host endian format. + */ +static void dlm_query_join_packet_to_wire(struct dlm_query_join_packet *packet, + u32 *wire) +{ + union dlm_query_join_response response; + + response.packet = *packet; + *wire = cpu_to_be32(response.intval); +} + +static void dlm_query_join_wire_to_packet(u32 wire, + struct dlm_query_join_packet *packet) +{ + union dlm_query_join_response response; + + response.intval = cpu_to_be32(wire); + *packet = response.packet; +} + static int dlm_query_join_handler(struct o2net_msg *msg, u32 len, void *data, void **ret_data) { struct dlm_query_join_request *query; - enum dlm_query_join_response response; + struct dlm_query_join_packet packet = { + .code = JOIN_DISALLOW, + }; struct dlm_ctxt *dlm = NULL; + u32 response; u8 nodenum; query = (struct dlm_query_join_request *) msg->buf; @@ -690,11 +769,11 @@ static int dlm_query_join_handler(struct o2net_msg *msg, u32 len, void *data, mlog(0, "node %u is not in our live map yet\n", query->node_idx); - response = JOIN_DISALLOW; + packet.code = JOIN_DISALLOW; goto respond; } - response = JOIN_OK_NO_MAP; + packet.code = JOIN_OK_NO_MAP; spin_lock(&dlm_domain_lock); dlm = __dlm_lookup_domain_full(query->domain, query->name_len); @@ -713,7 +792,7 @@ static int dlm_query_join_handler(struct o2net_msg *msg, u32 len, void *data, mlog(0, "disallow join as node %u does not " "have node %u in its nodemap\n", query->node_idx, nodenum); - response = JOIN_DISALLOW; + packet.code = JOIN_DISALLOW; goto unlock_respond; } } @@ -733,30 +812,44 @@ static int dlm_query_join_handler(struct o2net_msg *msg, u32 len, void *data, /*If this is a brand new context and we * haven't started our join process yet, then * the other node won the race. */ - response = JOIN_OK_NO_MAP; + packet.code = JOIN_OK_NO_MAP; } else if (dlm->joining_node != DLM_LOCK_RES_OWNER_UNKNOWN) { /* Disallow parallel joins. */ - response = JOIN_DISALLOW; + packet.code = JOIN_DISALLOW; } else if (dlm->reco.state & DLM_RECO_STATE_ACTIVE) { mlog(0, "node %u trying to join, but recovery " "is ongoing.\n", bit); - response = JOIN_DISALLOW; + packet.code = JOIN_DISALLOW; } else if (test_bit(bit, dlm->recovery_map)) { mlog(0, "node %u trying to join, but it " "still needs recovery.\n", bit); - response = JOIN_DISALLOW; + packet.code = JOIN_DISALLOW; } else if (test_bit(bit, dlm->domain_map)) { mlog(0, "node %u trying to join, but it " "is still in the domain! needs recovery?\n", bit); - response = JOIN_DISALLOW; + packet.code = JOIN_DISALLOW; } else { /* Alright we're fully a part of this domain * so we keep some state as to who's joining * and indicate to him that needs to be fixed * up. */ - response = JOIN_OK; - __dlm_set_joining_node(dlm, query->node_idx); + + /* Make sure we speak compatible locking protocols. */ + if (dlm_query_join_proto_check("DLM", bit, + &dlm->dlm_locking_proto, + &query->dlm_proto)) { + packet.code = JOIN_PROTOCOL_MISMATCH; + } else if (dlm_query_join_proto_check("fs", bit, + &dlm->fs_locking_proto, + &query->fs_proto)) { + packet.code = JOIN_PROTOCOL_MISMATCH; + } else { + packet.dlm_minor = query->dlm_proto.pv_minor; + packet.fs_minor = query->fs_proto.pv_minor; + packet.code = JOIN_OK; + __dlm_set_joining_node(dlm, query->node_idx); + } } spin_unlock(&dlm->spinlock); @@ -765,8 +858,9 @@ unlock_respond: spin_unlock(&dlm_domain_lock); respond: - mlog(0, "We respond with %u\n", response); + mlog(0, "We respond with %u\n", packet.code); + dlm_query_join_packet_to_wire(&packet, &response); return response; } @@ -872,7 +966,7 @@ static int dlm_send_join_cancels(struct dlm_ctxt *dlm, sizeof(unsigned long))) { mlog(ML_ERROR, "map_size %u != BITS_TO_LONGS(O2NM_MAX_NODES) %u\n", - map_size, BITS_TO_LONGS(O2NM_MAX_NODES)); + map_size, (unsigned)BITS_TO_LONGS(O2NM_MAX_NODES)); return -EINVAL; } @@ -899,10 +993,12 @@ static int dlm_send_join_cancels(struct dlm_ctxt *dlm, static int dlm_request_join(struct dlm_ctxt *dlm, int node, - enum dlm_query_join_response *response) + enum dlm_query_join_response_code *response) { - int status, retval; + int status; struct dlm_query_join_request join_msg; + struct dlm_query_join_packet packet; + u32 join_resp; mlog(0, "querying node %d\n", node); @@ -910,16 +1006,20 @@ static int dlm_request_join(struct dlm_ctxt *dlm, join_msg.node_idx = dlm->node_num; join_msg.name_len = strlen(dlm->name); memcpy(join_msg.domain, dlm->name, join_msg.name_len); + join_msg.dlm_proto = dlm->dlm_locking_proto; + join_msg.fs_proto = dlm->fs_locking_proto; /* copy live node map to join message */ byte_copymap(join_msg.node_map, dlm->live_nodes_map, O2NM_MAX_NODES); status = o2net_send_message(DLM_QUERY_JOIN_MSG, DLM_MOD_KEY, &join_msg, - sizeof(join_msg), node, &retval); + sizeof(join_msg), node, + &join_resp); if (status < 0 && status != -ENOPROTOOPT) { mlog_errno(status); goto bail; } + dlm_query_join_wire_to_packet(join_resp, &packet); /* -ENOPROTOOPT from the net code means the other side isn't listening for our message type -- that's fine, it means @@ -928,18 +1028,43 @@ static int dlm_request_join(struct dlm_ctxt *dlm, if (status == -ENOPROTOOPT) { status = 0; *response = JOIN_OK_NO_MAP; - } else if (retval == JOIN_DISALLOW || - retval == JOIN_OK || - retval == JOIN_OK_NO_MAP) { - *response = retval; + } else if (packet.code == JOIN_DISALLOW || + packet.code == JOIN_OK_NO_MAP) { + *response = packet.code; + } else if (packet.code == JOIN_PROTOCOL_MISMATCH) { + mlog(ML_NOTICE, + "This node requested DLM locking protocol %u.%u and " + "filesystem locking protocol %u.%u. At least one of " + "the protocol versions on node %d is not compatible, " + "disconnecting\n", + dlm->dlm_locking_proto.pv_major, + dlm->dlm_locking_proto.pv_minor, + dlm->fs_locking_proto.pv_major, + dlm->fs_locking_proto.pv_minor, + node); + status = -EPROTO; + *response = packet.code; + } else if (packet.code == JOIN_OK) { + *response = packet.code; + /* Use the same locking protocol as the remote node */ + dlm->dlm_locking_proto.pv_minor = packet.dlm_minor; + dlm->fs_locking_proto.pv_minor = packet.fs_minor; + mlog(0, + "Node %d responds JOIN_OK with DLM locking protocol " + "%u.%u and fs locking protocol %u.%u\n", + node, + dlm->dlm_locking_proto.pv_major, + dlm->dlm_locking_proto.pv_minor, + dlm->fs_locking_proto.pv_major, + dlm->fs_locking_proto.pv_minor); } else { status = -EINVAL; - mlog(ML_ERROR, "invalid response %d from node %u\n", retval, - node); + mlog(ML_ERROR, "invalid response %d from node %u\n", + packet.code, node); } mlog(0, "status %d, node %d response is %d\n", status, node, - *response); + *response); bail: return status; @@ -1008,7 +1133,7 @@ struct domain_join_ctxt { static int dlm_should_restart_join(struct dlm_ctxt *dlm, struct domain_join_ctxt *ctxt, - enum dlm_query_join_response response) + enum dlm_query_join_response_code response) { int ret; @@ -1034,7 +1159,7 @@ static int dlm_try_to_join_domain(struct dlm_ctxt *dlm) { int status = 0, tmpstat, node; struct domain_join_ctxt *ctxt; - enum dlm_query_join_response response = JOIN_DISALLOW; + enum dlm_query_join_response_code response = JOIN_DISALLOW; mlog_entry("%p", dlm); @@ -1450,10 +1575,38 @@ leave: } /* - * dlm_register_domain: one-time setup per "domain" + * Compare a requested locking protocol version against the current one. + * + * If the major numbers are different, they are incompatible. + * If the current minor is greater than the request, they are incompatible. + * If the current minor is less than or equal to the request, they are + * compatible, and the requester should run at the current minor version. + */ +static int dlm_protocol_compare(struct dlm_protocol_version *existing, + struct dlm_protocol_version *request) +{ + if (existing->pv_major != request->pv_major) + return 1; + + if (existing->pv_minor > request->pv_minor) + return 1; + + if (existing->pv_minor < request->pv_minor) + request->pv_minor = existing->pv_minor; + + return 0; +} + +/* + * dlm_register_domain: one-time setup per "domain". + * + * The filesystem passes in the requested locking version via proto. + * If registration was successful, proto will contain the negotiated + * locking protocol. */ struct dlm_ctxt * dlm_register_domain(const char *domain, - u32 key) + u32 key, + struct dlm_protocol_version *fs_proto) { int ret; struct dlm_ctxt *dlm = NULL; @@ -1496,6 +1649,15 @@ retry: goto retry; } + if (dlm_protocol_compare(&dlm->fs_locking_proto, fs_proto)) { + mlog(ML_ERROR, + "Requested locking protocol version is not " + "compatible with already registered domain " + "\"%s\"\n", domain); + ret = -EPROTO; + goto leave; + } + __dlm_get(dlm); dlm->num_joins++; @@ -1526,6 +1688,13 @@ retry: list_add_tail(&dlm->list, &dlm_domains); spin_unlock(&dlm_domain_lock); + /* + * Pass the locking protocol version into the join. If the join + * succeeds, it will have the negotiated protocol set. + */ + dlm->dlm_locking_proto = dlm_protocol; + dlm->fs_locking_proto = *fs_proto; + ret = dlm_join_domain(dlm); if (ret) { mlog_errno(ret); @@ -1533,6 +1702,9 @@ retry: goto leave; } + /* Tell the caller what locking protocol we negotiated */ + *fs_proto = dlm->fs_locking_proto; + ret = 0; leave: if (new_ctxt) diff --git a/fs/ocfs2/dlm/dlmfs.c b/fs/ocfs2/dlm/dlmfs.c index 6639baab079..61a000f8524 100644 --- a/fs/ocfs2/dlm/dlmfs.c +++ b/fs/ocfs2/dlm/dlmfs.c @@ -60,6 +60,8 @@ #define MLOG_MASK_PREFIX ML_DLMFS #include "cluster/masklog.h" +#include "ocfs2_lockingver.h" + static const struct super_operations dlmfs_ops; static const struct file_operations dlmfs_file_operations; static const struct inode_operations dlmfs_dir_inode_operations; @@ -70,6 +72,16 @@ static struct kmem_cache *dlmfs_inode_cache; struct workqueue_struct *user_dlm_worker; /* + * This is the userdlmfs locking protocol version. + * + * See fs/ocfs2/dlmglue.c for more details on locking versions. + */ +static const struct dlm_protocol_version user_locking_protocol = { + .pv_major = OCFS2_LOCKING_PROTOCOL_MAJOR, + .pv_minor = OCFS2_LOCKING_PROTOCOL_MINOR, +}; + +/* * decodes a set of open flags into a valid lock level and a set of flags. * returns < 0 if we have invalid flags * flags which mean something to us: @@ -416,6 +428,7 @@ static int dlmfs_mkdir(struct inode * dir, struct qstr *domain = &dentry->d_name; struct dlmfs_inode_private *ip; struct dlm_ctxt *dlm; + struct dlm_protocol_version proto = user_locking_protocol; mlog(0, "mkdir %.*s\n", domain->len, domain->name); @@ -435,7 +448,7 @@ static int dlmfs_mkdir(struct inode * dir, ip = DLMFS_I(inode); - dlm = user_dlm_register_context(domain); + dlm = user_dlm_register_context(domain, &proto); if (IS_ERR(dlm)) { status = PTR_ERR(dlm); mlog(ML_ERROR, "Error %d could not register domain \"%.*s\"\n", diff --git a/fs/ocfs2/dlm/dlmfsver.c b/fs/ocfs2/dlm/dlmfsver.c index d2be3ad841f..a733b3321f8 100644 --- a/fs/ocfs2/dlm/dlmfsver.c +++ b/fs/ocfs2/dlm/dlmfsver.c @@ -28,7 +28,7 @@ #include "dlmfsver.h" -#define DLM_BUILD_VERSION "1.3.3" +#define DLM_BUILD_VERSION "1.5.0" #define VERSION_STR "OCFS2 DLMFS " DLM_BUILD_VERSION diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index a54d33d95ad..ea6b8957786 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -1663,7 +1663,12 @@ way_up_top: dlm_put_mle(tmpmle); } send_response: - + /* + * __dlm_lookup_lockres() grabbed a reference to this lockres. + * The reference is released by dlm_assert_master_worker() under + * the call to dlm_dispatch_assert_master(). If + * dlm_assert_master_worker() isn't called, we drop it here. + */ if (dispatch_assert) { if (response != DLM_MASTER_RESP_YES) mlog(ML_ERROR, "invalid response %d\n", response); @@ -1678,7 +1683,11 @@ send_response: if (ret < 0) { mlog(ML_ERROR, "failed to dispatch assert master work\n"); response = DLM_MASTER_RESP_ERROR; + dlm_lockres_put(res); } + } else { + if (res) + dlm_lockres_put(res); } dlm_put(dlm); @@ -1695,9 +1704,9 @@ send_response: * can periodically run all locks owned by this node * and re-assert across the cluster... */ -int dlm_do_assert_master(struct dlm_ctxt *dlm, - struct dlm_lock_resource *res, - void *nodemap, u32 flags) +static int dlm_do_assert_master(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res, + void *nodemap, u32 flags) { struct dlm_assert_master assert; int to, tmpret; @@ -2348,7 +2357,7 @@ int dlm_deref_lockres_handler(struct o2net_msg *msg, u32 len, void *data, mlog(ML_ERROR, "%s:%.*s: node %u trying to drop ref " "but it is already dropped!\n", dlm->name, res->lockname.len, res->lockname.name, node); - __dlm_print_one_lock_resource(res); + dlm_print_one_lock_resource(res); } ret = 0; goto done; @@ -2408,7 +2417,7 @@ static void dlm_deref_lockres_worker(struct dlm_work_item *item, void *data) mlog(ML_ERROR, "%s:%.*s: node %u trying to drop ref " "but it is already dropped!\n", dlm->name, res->lockname.len, res->lockname.name, node); - __dlm_print_one_lock_resource(res); + dlm_print_one_lock_resource(res); } dlm_lockres_put(res); @@ -2933,6 +2942,9 @@ static void dlm_remove_nonlocal_locks(struct dlm_ctxt *dlm, dlm_lockres_clear_refmap_bit(lock->ml.node, res); list_del_init(&lock->list); dlm_lock_put(lock); + /* In a normal unlock, we would have added a + * DLM_UNLOCK_FREE_LOCK action. Force it. */ + dlm_lock_put(lock); } } queue++; diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index 2fde7bf9143..bcb9260c373 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -519,9 +519,9 @@ static int dlm_do_recovery(struct dlm_ctxt *dlm) return 0; master_here: - mlog(0, "(%d) mastering recovery of %s:%u here(this=%u)!\n", - task_pid_nr(dlm->dlm_reco_thread_task), - dlm->name, dlm->reco.dead_node, dlm->node_num); + mlog(ML_NOTICE, "(%d) Node %u is the Recovery Master for the Dead Node " + "%u for Domain %s\n", task_pid_nr(dlm->dlm_reco_thread_task), + dlm->node_num, dlm->reco.dead_node, dlm->name); status = dlm_remaster_locks(dlm, dlm->reco.dead_node); if (status < 0) { @@ -1191,7 +1191,7 @@ static int dlm_add_lock_to_array(struct dlm_lock *lock, (ml->type == LKM_EXMODE || memcmp(mres->lvb, lock->lksb->lvb, DLM_LVB_LEN))) { mlog(ML_ERROR, "mismatched lvbs!\n"); - __dlm_print_one_lock_resource(lock->lockres); + dlm_print_one_lock_resource(lock->lockres); BUG(); } memcpy(mres->lvb, lock->lksb->lvb, DLM_LVB_LEN); @@ -1327,6 +1327,7 @@ int dlm_mig_lockres_handler(struct o2net_msg *msg, u32 len, void *data, (struct dlm_migratable_lockres *)msg->buf; int ret = 0; u8 real_master; + u8 extra_refs = 0; char *buf = NULL; struct dlm_work_item *item = NULL; struct dlm_lock_resource *res = NULL; @@ -1404,16 +1405,28 @@ int dlm_mig_lockres_handler(struct o2net_msg *msg, u32 len, void *data, __dlm_insert_lockres(dlm, res); spin_unlock(&dlm->spinlock); + /* Add an extra ref for this lock-less lockres lest the + * dlm_thread purges it before we get the chance to add + * locks to it */ + dlm_lockres_get(res); + + /* There are three refs that need to be put. + * 1. Taken above. + * 2. kref_init in dlm_new_lockres()->dlm_init_lockres(). + * 3. dlm_lookup_lockres() + * The first one is handled at the end of this function. The + * other two are handled in the worker thread after locks have + * been attached. Yes, we don't wait for purge time to match + * kref_init. The lockres will still have atleast one ref + * added because it is in the hash __dlm_insert_lockres() */ + extra_refs++; + /* now that the new lockres is inserted, * make it usable by other processes */ spin_lock(&res->spinlock); res->state &= ~DLM_LOCK_RES_IN_PROGRESS; spin_unlock(&res->spinlock); wake_up(&res->wq); - - /* add an extra ref for just-allocated lockres - * otherwise the lockres will be purged immediately */ - dlm_lockres_get(res); } /* at this point we have allocated everything we need, @@ -1443,12 +1456,17 @@ int dlm_mig_lockres_handler(struct o2net_msg *msg, u32 len, void *data, dlm_init_work_item(dlm, item, dlm_mig_lockres_worker, buf); item->u.ml.lockres = res; /* already have a ref */ item->u.ml.real_master = real_master; + item->u.ml.extra_ref = extra_refs; spin_lock(&dlm->work_lock); list_add_tail(&item->list, &dlm->work_list); spin_unlock(&dlm->work_lock); queue_work(dlm->dlm_worker, &dlm->dispatched_work); leave: + /* One extra ref taken needs to be put here */ + if (extra_refs) + dlm_lockres_put(res); + dlm_put(dlm); if (ret < 0) { if (buf) @@ -1464,17 +1482,19 @@ leave: static void dlm_mig_lockres_worker(struct dlm_work_item *item, void *data) { - struct dlm_ctxt *dlm = data; + struct dlm_ctxt *dlm; struct dlm_migratable_lockres *mres; int ret = 0; struct dlm_lock_resource *res; u8 real_master; + u8 extra_ref; dlm = item->dlm; mres = (struct dlm_migratable_lockres *)data; res = item->u.ml.lockres; real_master = item->u.ml.real_master; + extra_ref = item->u.ml.extra_ref; if (real_master == DLM_LOCK_RES_OWNER_UNKNOWN) { /* this case is super-rare. only occurs if @@ -1517,6 +1537,12 @@ again: } leave: + /* See comment in dlm_mig_lockres_handler() */ + if (res) { + if (extra_ref) + dlm_lockres_put(res); + dlm_lockres_put(res); + } kfree(data); mlog_exit(ret); } @@ -1644,7 +1670,8 @@ int dlm_master_requery_handler(struct o2net_msg *msg, u32 len, void *data, /* retry!? */ BUG(); } - } + } else /* put.. incase we are not the master */ + dlm_lockres_put(res); spin_unlock(&res->spinlock); } spin_unlock(&dlm->spinlock); @@ -1921,6 +1948,7 @@ void dlm_move_lockres_to_recovery_list(struct dlm_ctxt *dlm, "Recovering res %s:%.*s, is already on recovery list!\n", dlm->name, res->lockname.len, res->lockname.name); list_del_init(&res->recovering); + dlm_lockres_put(res); } /* We need to hold a reference while on the recovery list */ dlm_lockres_get(res); @@ -2130,11 +2158,16 @@ static void dlm_free_dead_locks(struct dlm_ctxt *dlm, assert_spin_locked(&dlm->spinlock); assert_spin_locked(&res->spinlock); + /* We do two dlm_lock_put(). One for removing from list and the other is + * to force the DLM_UNLOCK_FREE_LOCK action so as to free the locks */ + /* TODO: check pending_asts, pending_basts here */ list_for_each_entry_safe(lock, next, &res->granted, list) { if (lock->ml.node == dead_node) { list_del_init(&lock->list); dlm_lock_put(lock); + /* Can't schedule DLM_UNLOCK_FREE_LOCK - do manually */ + dlm_lock_put(lock); freed++; } } @@ -2142,6 +2175,8 @@ static void dlm_free_dead_locks(struct dlm_ctxt *dlm, if (lock->ml.node == dead_node) { list_del_init(&lock->list); dlm_lock_put(lock); + /* Can't schedule DLM_UNLOCK_FREE_LOCK - do manually */ + dlm_lock_put(lock); freed++; } } @@ -2149,6 +2184,8 @@ static void dlm_free_dead_locks(struct dlm_ctxt *dlm, if (lock->ml.node == dead_node) { list_del_init(&lock->list); dlm_lock_put(lock); + /* Can't schedule DLM_UNLOCK_FREE_LOCK - do manually */ + dlm_lock_put(lock); freed++; } } @@ -2270,6 +2307,12 @@ static void __dlm_hb_node_down(struct dlm_ctxt *dlm, int idx) } } + /* Clean up join state on node death. */ + if (dlm->joining_node == idx) { + mlog(0, "Clearing join state for node %u\n", idx); + __dlm_set_joining_node(dlm, DLM_LOCK_RES_OWNER_UNKNOWN); + } + /* check to see if the node is already considered dead */ if (!test_bit(idx, dlm->live_nodes_map)) { mlog(0, "for domain %s, node %d is already dead. " @@ -2288,12 +2331,6 @@ static void __dlm_hb_node_down(struct dlm_ctxt *dlm, int idx) clear_bit(idx, dlm->live_nodes_map); - /* Clean up join state on node death. */ - if (dlm->joining_node == idx) { - mlog(0, "Clearing join state for node %u\n", idx); - __dlm_set_joining_node(dlm, DLM_LOCK_RES_OWNER_UNKNOWN); - } - /* make sure local cleanup occurs before the heartbeat events */ if (!test_bit(idx, dlm->recovery_map)) dlm_do_local_recovery_cleanup(dlm, idx); @@ -2321,6 +2358,13 @@ void dlm_hb_node_down_cb(struct o2nm_node *node, int idx, void *data) if (!dlm_grab(dlm)) return; + /* + * This will notify any dlm users that a node in our domain + * went away without notifying us first. + */ + if (test_bit(idx, dlm->domain_map)) + dlm_fire_domain_eviction_callbacks(dlm, idx); + spin_lock(&dlm->spinlock); __dlm_hb_node_down(dlm, idx); spin_unlock(&dlm->spinlock); diff --git a/fs/ocfs2/dlm/dlmthread.c b/fs/ocfs2/dlm/dlmthread.c index cebd089f895..4060bb328bc 100644 --- a/fs/ocfs2/dlm/dlmthread.c +++ b/fs/ocfs2/dlm/dlmthread.c @@ -176,12 +176,14 @@ static int dlm_purge_lockres(struct dlm_ctxt *dlm, res->lockname.name, master); if (!master) { + /* drop spinlock... retake below */ + spin_unlock(&dlm->spinlock); + spin_lock(&res->spinlock); /* This ensures that clear refmap is sent after the set */ __dlm_wait_on_lockres_flags(res, DLM_LOCK_RES_SETREF_INPROG); spin_unlock(&res->spinlock); - /* drop spinlock to do messaging, retake below */ - spin_unlock(&dlm->spinlock); + /* clear our bit from the master's refmap, ignore errors */ ret = dlm_drop_lockres_ref(dlm, res); if (ret < 0) { diff --git a/fs/ocfs2/dlm/dlmver.c b/fs/ocfs2/dlm/dlmver.c index 7ef2653f8f4..dfc0da4d158 100644 --- a/fs/ocfs2/dlm/dlmver.c +++ b/fs/ocfs2/dlm/dlmver.c @@ -28,7 +28,7 @@ #include "dlmver.h" -#define DLM_BUILD_VERSION "1.3.3" +#define DLM_BUILD_VERSION "1.5.0" #define VERSION_STR "OCFS2 DLM " DLM_BUILD_VERSION diff --git a/fs/ocfs2/dlm/userdlm.c b/fs/ocfs2/dlm/userdlm.c index 7d2f578b267..4cb1d3dae25 100644 --- a/fs/ocfs2/dlm/userdlm.c +++ b/fs/ocfs2/dlm/userdlm.c @@ -645,7 +645,8 @@ bail: return status; } -struct dlm_ctxt *user_dlm_register_context(struct qstr *name) +struct dlm_ctxt *user_dlm_register_context(struct qstr *name, + struct dlm_protocol_version *proto) { struct dlm_ctxt *dlm; u32 dlm_key; @@ -661,7 +662,7 @@ struct dlm_ctxt *user_dlm_register_context(struct qstr *name) snprintf(domain, name->len + 1, "%.*s", name->len, name->name); - dlm = dlm_register_domain(domain, dlm_key); + dlm = dlm_register_domain(domain, dlm_key, proto); if (IS_ERR(dlm)) mlog_errno(PTR_ERR(dlm)); diff --git a/fs/ocfs2/dlm/userdlm.h b/fs/ocfs2/dlm/userdlm.h index c400e93bbf7..39ec2773849 100644 --- a/fs/ocfs2/dlm/userdlm.h +++ b/fs/ocfs2/dlm/userdlm.h @@ -83,7 +83,8 @@ void user_dlm_write_lvb(struct inode *inode, void user_dlm_read_lvb(struct inode *inode, char *val, unsigned int len); -struct dlm_ctxt *user_dlm_register_context(struct qstr *name); +struct dlm_ctxt *user_dlm_register_context(struct qstr *name, + struct dlm_protocol_version *proto); void user_dlm_unregister_context(struct dlm_ctxt *dlm); struct dlmfs_inode_private { diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c index 4e97dcceaf8..1f1873bf41f 100644 --- a/fs/ocfs2/dlmglue.c +++ b/fs/ocfs2/dlmglue.c @@ -43,6 +43,7 @@ #include <cluster/masklog.h> #include "ocfs2.h" +#include "ocfs2_lockingver.h" #include "alloc.h" #include "dcache.h" @@ -55,7 +56,6 @@ #include "slot_map.h" #include "super.h" #include "uptodate.h" -#include "vote.h" #include "buffer_head_io.h" @@ -69,6 +69,7 @@ struct ocfs2_mask_waiter { static struct ocfs2_super *ocfs2_get_dentry_osb(struct ocfs2_lock_res *lockres); static struct ocfs2_super *ocfs2_get_inode_osb(struct ocfs2_lock_res *lockres); +static struct ocfs2_super *ocfs2_get_file_osb(struct ocfs2_lock_res *lockres); /* * Return value from ->downconvert_worker functions. @@ -153,10 +154,10 @@ struct ocfs2_lock_res_ops { struct ocfs2_super * (*get_osb)(struct ocfs2_lock_res *); /* - * Optionally called in the downconvert (or "vote") thread - * after a successful downconvert. The lockres will not be - * referenced after this callback is called, so it is safe to - * free memory, etc. + * Optionally called in the downconvert thread after a + * successful downconvert. The lockres will not be referenced + * after this callback is called, so it is safe to free + * memory, etc. * * The exact semantics of when this is called are controlled * by ->downconvert_worker() @@ -225,17 +226,12 @@ static struct ocfs2_lock_res_ops ocfs2_inode_rw_lops = { .flags = 0, }; -static struct ocfs2_lock_res_ops ocfs2_inode_meta_lops = { +static struct ocfs2_lock_res_ops ocfs2_inode_inode_lops = { .get_osb = ocfs2_get_inode_osb, .check_downconvert = ocfs2_check_meta_downconvert, .set_lvb = ocfs2_set_meta_lvb, - .flags = LOCK_TYPE_REQUIRES_REFRESH|LOCK_TYPE_USES_LVB, -}; - -static struct ocfs2_lock_res_ops ocfs2_inode_data_lops = { - .get_osb = ocfs2_get_inode_osb, .downconvert_worker = ocfs2_data_convert_worker, - .flags = 0, + .flags = LOCK_TYPE_REQUIRES_REFRESH|LOCK_TYPE_USES_LVB, }; static struct ocfs2_lock_res_ops ocfs2_super_lops = { @@ -258,10 +254,39 @@ static struct ocfs2_lock_res_ops ocfs2_inode_open_lops = { .flags = 0, }; +static struct ocfs2_lock_res_ops ocfs2_flock_lops = { + .get_osb = ocfs2_get_file_osb, + .flags = 0, +}; + +/* + * This is the filesystem locking protocol version. + * + * Whenever the filesystem does new things with locks (adds or removes a + * lock, orders them differently, does different things underneath a lock), + * the version must be changed. The protocol is negotiated when joining + * the dlm domain. A node may join the domain if its major version is + * identical to all other nodes and its minor version is greater than + * or equal to all other nodes. When its minor version is greater than + * the other nodes, it will run at the minor version specified by the + * other nodes. + * + * If a locking change is made that will not be compatible with older + * versions, the major number must be increased and the minor version set + * to zero. If a change merely adds a behavior that can be disabled when + * speaking to older versions, the minor version must be increased. If a + * change adds a fully backwards compatible change (eg, LVB changes that + * are just ignored by older versions), the version does not need to be + * updated. + */ +const struct dlm_protocol_version ocfs2_locking_protocol = { + .pv_major = OCFS2_LOCKING_PROTOCOL_MAJOR, + .pv_minor = OCFS2_LOCKING_PROTOCOL_MINOR, +}; + static inline int ocfs2_is_inode_lock(struct ocfs2_lock_res *lockres) { return lockres->l_type == OCFS2_LOCK_TYPE_META || - lockres->l_type == OCFS2_LOCK_TYPE_DATA || lockres->l_type == OCFS2_LOCK_TYPE_RW || lockres->l_type == OCFS2_LOCK_TYPE_OPEN; } @@ -310,12 +335,24 @@ static inline void ocfs2_recover_from_dlm_error(struct ocfs2_lock_res *lockres, "resource %s: %s\n", dlm_errname(_stat), _func, \ _lockres->l_name, dlm_errmsg(_stat)); \ } while (0) -static void ocfs2_vote_on_unlock(struct ocfs2_super *osb, - struct ocfs2_lock_res *lockres); -static int ocfs2_meta_lock_update(struct inode *inode, +static int ocfs2_downconvert_thread(void *arg); +static void ocfs2_downconvert_on_unlock(struct ocfs2_super *osb, + struct ocfs2_lock_res *lockres); +static int ocfs2_inode_lock_update(struct inode *inode, struct buffer_head **bh); static void ocfs2_drop_osb_locks(struct ocfs2_super *osb); static inline int ocfs2_highest_compat_lock_level(int level); +static void ocfs2_prepare_downconvert(struct ocfs2_lock_res *lockres, + int new_level); +static int ocfs2_downconvert_lock(struct ocfs2_super *osb, + struct ocfs2_lock_res *lockres, + int new_level, + int lvb); +static int ocfs2_prepare_cancel_convert(struct ocfs2_super *osb, + struct ocfs2_lock_res *lockres); +static int ocfs2_cancel_convert(struct ocfs2_super *osb, + struct ocfs2_lock_res *lockres); + static void ocfs2_build_lock_name(enum ocfs2_lock_type type, u64 blkno, @@ -402,10 +439,7 @@ void ocfs2_inode_lock_res_init(struct ocfs2_lock_res *res, ops = &ocfs2_inode_rw_lops; break; case OCFS2_LOCK_TYPE_META: - ops = &ocfs2_inode_meta_lops; - break; - case OCFS2_LOCK_TYPE_DATA: - ops = &ocfs2_inode_data_lops; + ops = &ocfs2_inode_inode_lops; break; case OCFS2_LOCK_TYPE_OPEN: ops = &ocfs2_inode_open_lops; @@ -428,6 +462,13 @@ static struct ocfs2_super *ocfs2_get_inode_osb(struct ocfs2_lock_res *lockres) return OCFS2_SB(inode->i_sb); } +static struct ocfs2_super *ocfs2_get_file_osb(struct ocfs2_lock_res *lockres) +{ + struct ocfs2_file_private *fp = lockres->l_priv; + + return OCFS2_SB(fp->fp_file->f_mapping->host->i_sb); +} + static __u64 ocfs2_get_dentry_lock_ino(struct ocfs2_lock_res *lockres) { __be64 inode_blkno_be; @@ -508,6 +549,21 @@ static void ocfs2_rename_lock_res_init(struct ocfs2_lock_res *res, &ocfs2_rename_lops, osb); } +void ocfs2_file_lock_res_init(struct ocfs2_lock_res *lockres, + struct ocfs2_file_private *fp) +{ + struct inode *inode = fp->fp_file->f_mapping->host; + struct ocfs2_inode_info *oi = OCFS2_I(inode); + + ocfs2_lock_res_init_once(lockres); + ocfs2_build_lock_name(OCFS2_LOCK_TYPE_FLOCK, oi->ip_blkno, + inode->i_generation, lockres->l_name); + ocfs2_lock_res_init_common(OCFS2_SB(inode->i_sb), lockres, + OCFS2_LOCK_TYPE_FLOCK, &ocfs2_flock_lops, + fp); + lockres->l_flags |= OCFS2_LOCK_NOCACHE; +} + void ocfs2_lock_res_free(struct ocfs2_lock_res *res) { mlog_entry_void(); @@ -724,6 +780,13 @@ static void ocfs2_blocking_ast(void *opaque, int level) lockres->l_name, level, lockres->l_level, ocfs2_lock_type_string(lockres->l_type)); + /* + * We can skip the bast for locks which don't enable caching - + * they'll be dropped at the earliest possible time anyway. + */ + if (lockres->l_flags & OCFS2_LOCK_NOCACHE) + return; + spin_lock_irqsave(&lockres->l_lock, flags); needs_downconvert = ocfs2_generic_handle_bast(lockres, level); if (needs_downconvert) @@ -732,7 +795,7 @@ static void ocfs2_blocking_ast(void *opaque, int level) wake_up(&lockres->l_event); - ocfs2_kick_vote_thread(osb); + ocfs2_wake_downconvert_thread(osb); } static void ocfs2_locking_ast(void *opaque) @@ -935,6 +998,21 @@ static int lockres_remove_mask_waiter(struct ocfs2_lock_res *lockres, } +static int ocfs2_wait_for_mask_interruptible(struct ocfs2_mask_waiter *mw, + struct ocfs2_lock_res *lockres) +{ + int ret; + + ret = wait_for_completion_interruptible(&mw->mw_complete); + if (ret) + lockres_remove_mask_waiter(lockres, mw); + else + ret = mw->mw_status; + /* Re-arm the completion in case we want to wait on it again */ + INIT_COMPLETION(mw->mw_complete); + return ret; +} + static int ocfs2_cluster_lock(struct ocfs2_super *osb, struct ocfs2_lock_res *lockres, int level, @@ -1089,7 +1167,7 @@ static void ocfs2_cluster_unlock(struct ocfs2_super *osb, mlog_entry_void(); spin_lock_irqsave(&lockres->l_lock, flags); ocfs2_dec_holders(lockres, level); - ocfs2_vote_on_unlock(osb, lockres); + ocfs2_downconvert_on_unlock(osb, lockres); spin_unlock_irqrestore(&lockres->l_lock, flags); mlog_exit_void(); } @@ -1147,13 +1225,7 @@ int ocfs2_create_new_inode_locks(struct inode *inode) * We don't want to use LKM_LOCAL on a meta data lock as they * don't use a generation in their lock names. */ - ret = ocfs2_create_new_lock(osb, &OCFS2_I(inode)->ip_meta_lockres, 1, 0); - if (ret) { - mlog_errno(ret); - goto bail; - } - - ret = ocfs2_create_new_lock(osb, &OCFS2_I(inode)->ip_data_lockres, 1, 1); + ret = ocfs2_create_new_lock(osb, &OCFS2_I(inode)->ip_inode_lockres, 1, 0); if (ret) { mlog_errno(ret); goto bail; @@ -1311,76 +1383,221 @@ out: mlog_exit_void(); } -int ocfs2_data_lock_full(struct inode *inode, - int write, - int arg_flags) +static int ocfs2_flock_handle_signal(struct ocfs2_lock_res *lockres, + int level) { - int status = 0, level; - struct ocfs2_lock_res *lockres; - struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + int ret; + struct ocfs2_super *osb = ocfs2_get_lockres_osb(lockres); + unsigned long flags; + struct ocfs2_mask_waiter mw; - BUG_ON(!inode); + ocfs2_init_mask_waiter(&mw); - mlog_entry_void(); +retry_cancel: + spin_lock_irqsave(&lockres->l_lock, flags); + if (lockres->l_flags & OCFS2_LOCK_BUSY) { + ret = ocfs2_prepare_cancel_convert(osb, lockres); + if (ret) { + spin_unlock_irqrestore(&lockres->l_lock, flags); + ret = ocfs2_cancel_convert(osb, lockres); + if (ret < 0) { + mlog_errno(ret); + goto out; + } + goto retry_cancel; + } + lockres_add_mask_waiter(lockres, &mw, OCFS2_LOCK_BUSY, 0); + spin_unlock_irqrestore(&lockres->l_lock, flags); - mlog(0, "inode %llu take %s DATA lock\n", - (unsigned long long)OCFS2_I(inode)->ip_blkno, - write ? "EXMODE" : "PRMODE"); + ocfs2_wait_for_mask(&mw); + goto retry_cancel; + } - /* We'll allow faking a readonly data lock for - * rodevices. */ - if (ocfs2_is_hard_readonly(OCFS2_SB(inode->i_sb))) { - if (write) { - status = -EROFS; - mlog_errno(status); + ret = -ERESTARTSYS; + /* + * We may still have gotten the lock, in which case there's no + * point to restarting the syscall. + */ + if (lockres->l_level == level) + ret = 0; + + mlog(0, "Cancel returning %d. flags: 0x%lx, level: %d, act: %d\n", ret, + lockres->l_flags, lockres->l_level, lockres->l_action); + + spin_unlock_irqrestore(&lockres->l_lock, flags); + +out: + return ret; +} + +/* + * ocfs2_file_lock() and ocfs2_file_unlock() map to a single pair of + * flock() calls. The locking approach this requires is sufficiently + * different from all other cluster lock types that we implement a + * seperate path to the "low-level" dlm calls. In particular: + * + * - No optimization of lock levels is done - we take at exactly + * what's been requested. + * + * - No lock caching is employed. We immediately downconvert to + * no-lock at unlock time. This also means flock locks never go on + * the blocking list). + * + * - Since userspace can trivially deadlock itself with flock, we make + * sure to allow cancellation of a misbehaving applications flock() + * request. + * + * - Access to any flock lockres doesn't require concurrency, so we + * can simplify the code by requiring the caller to guarantee + * serialization of dlmglue flock calls. + */ +int ocfs2_file_lock(struct file *file, int ex, int trylock) +{ + int ret, level = ex ? LKM_EXMODE : LKM_PRMODE; + unsigned int lkm_flags = trylock ? LKM_NOQUEUE : 0; + unsigned long flags; + struct ocfs2_file_private *fp = file->private_data; + struct ocfs2_lock_res *lockres = &fp->fp_flock; + struct ocfs2_super *osb = OCFS2_SB(file->f_mapping->host->i_sb); + struct ocfs2_mask_waiter mw; + + ocfs2_init_mask_waiter(&mw); + + if ((lockres->l_flags & OCFS2_LOCK_BUSY) || + (lockres->l_level > LKM_NLMODE)) { + mlog(ML_ERROR, + "File lock \"%s\" has busy or locked state: flags: 0x%lx, " + "level: %u\n", lockres->l_name, lockres->l_flags, + lockres->l_level); + return -EINVAL; + } + + spin_lock_irqsave(&lockres->l_lock, flags); + if (!(lockres->l_flags & OCFS2_LOCK_ATTACHED)) { + lockres_add_mask_waiter(lockres, &mw, OCFS2_LOCK_BUSY, 0); + spin_unlock_irqrestore(&lockres->l_lock, flags); + + /* + * Get the lock at NLMODE to start - that way we + * can cancel the upconvert request if need be. + */ + ret = ocfs2_lock_create(osb, lockres, LKM_NLMODE, 0); + if (ret < 0) { + mlog_errno(ret); + goto out; } - goto out; + + ret = ocfs2_wait_for_mask(&mw); + if (ret) { + mlog_errno(ret); + goto out; + } + spin_lock_irqsave(&lockres->l_lock, flags); } - if (ocfs2_mount_local(osb)) - goto out; + lockres->l_action = OCFS2_AST_CONVERT; + lkm_flags |= LKM_CONVERT; + lockres->l_requested = level; + lockres_or_flags(lockres, OCFS2_LOCK_BUSY); - lockres = &OCFS2_I(inode)->ip_data_lockres; + lockres_add_mask_waiter(lockres, &mw, OCFS2_LOCK_BUSY, 0); + spin_unlock_irqrestore(&lockres->l_lock, flags); - level = write ? LKM_EXMODE : LKM_PRMODE; + ret = dlmlock(osb->dlm, level, &lockres->l_lksb, lkm_flags, + lockres->l_name, OCFS2_LOCK_ID_MAX_LEN - 1, + ocfs2_locking_ast, lockres, ocfs2_blocking_ast); + if (ret != DLM_NORMAL) { + if (trylock && ret == DLM_NOTQUEUED) + ret = -EAGAIN; + else { + ocfs2_log_dlm_error("dlmlock", ret, lockres); + ret = -EINVAL; + } - status = ocfs2_cluster_lock(OCFS2_SB(inode->i_sb), lockres, level, - 0, arg_flags); - if (status < 0 && status != -EAGAIN) - mlog_errno(status); + ocfs2_recover_from_dlm_error(lockres, 1); + lockres_remove_mask_waiter(lockres, &mw); + goto out; + } + + ret = ocfs2_wait_for_mask_interruptible(&mw, lockres); + if (ret == -ERESTARTSYS) { + /* + * Userspace can cause deadlock itself with + * flock(). Current behavior locally is to allow the + * deadlock, but abort the system call if a signal is + * received. We follow this example, otherwise a + * poorly written program could sit in kernel until + * reboot. + * + * Handling this is a bit more complicated for Ocfs2 + * though. We can't exit this function with an + * outstanding lock request, so a cancel convert is + * required. We intentionally overwrite 'ret' - if the + * cancel fails and the lock was granted, it's easier + * to just bubble sucess back up to the user. + */ + ret = ocfs2_flock_handle_signal(lockres, level); + } out: - mlog_exit(status); - return status; + + mlog(0, "Lock: \"%s\" ex: %d, trylock: %d, returns: %d\n", + lockres->l_name, ex, trylock, ret); + return ret; } -/* see ocfs2_meta_lock_with_page() */ -int ocfs2_data_lock_with_page(struct inode *inode, - int write, - struct page *page) +void ocfs2_file_unlock(struct file *file) { int ret; + unsigned long flags; + struct ocfs2_file_private *fp = file->private_data; + struct ocfs2_lock_res *lockres = &fp->fp_flock; + struct ocfs2_super *osb = OCFS2_SB(file->f_mapping->host->i_sb); + struct ocfs2_mask_waiter mw; - ret = ocfs2_data_lock_full(inode, write, OCFS2_LOCK_NONBLOCK); - if (ret == -EAGAIN) { - unlock_page(page); - if (ocfs2_data_lock(inode, write) == 0) - ocfs2_data_unlock(inode, write); - ret = AOP_TRUNCATED_PAGE; + ocfs2_init_mask_waiter(&mw); + + if (!(lockres->l_flags & OCFS2_LOCK_ATTACHED)) + return; + + if (lockres->l_level == LKM_NLMODE) + return; + + mlog(0, "Unlock: \"%s\" flags: 0x%lx, level: %d, act: %d\n", + lockres->l_name, lockres->l_flags, lockres->l_level, + lockres->l_action); + + spin_lock_irqsave(&lockres->l_lock, flags); + /* + * Fake a blocking ast for the downconvert code. + */ + lockres_or_flags(lockres, OCFS2_LOCK_BLOCKED); + lockres->l_blocking = LKM_EXMODE; + + ocfs2_prepare_downconvert(lockres, LKM_NLMODE); + lockres_add_mask_waiter(lockres, &mw, OCFS2_LOCK_BUSY, 0); + spin_unlock_irqrestore(&lockres->l_lock, flags); + + ret = ocfs2_downconvert_lock(osb, lockres, LKM_NLMODE, 0); + if (ret) { + mlog_errno(ret); + return; } - return ret; + ret = ocfs2_wait_for_mask(&mw); + if (ret) + mlog_errno(ret); } -static void ocfs2_vote_on_unlock(struct ocfs2_super *osb, - struct ocfs2_lock_res *lockres) +static void ocfs2_downconvert_on_unlock(struct ocfs2_super *osb, + struct ocfs2_lock_res *lockres) { int kick = 0; mlog_entry_void(); /* If we know that another node is waiting on our lock, kick - * the vote thread * pre-emptively when we reach a release + * the downconvert thread * pre-emptively when we reach a release * condition. */ if (lockres->l_flags & OCFS2_LOCK_BLOCKED) { switch(lockres->l_blocking) { @@ -1398,27 +1615,7 @@ static void ocfs2_vote_on_unlock(struct ocfs2_super *osb, } if (kick) - ocfs2_kick_vote_thread(osb); - - mlog_exit_void(); -} - -void ocfs2_data_unlock(struct inode *inode, - int write) -{ - int level = write ? LKM_EXMODE : LKM_PRMODE; - struct ocfs2_lock_res *lockres = &OCFS2_I(inode)->ip_data_lockres; - struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); - - mlog_entry_void(); - - mlog(0, "inode %llu drop %s DATA lock\n", - (unsigned long long)OCFS2_I(inode)->ip_blkno, - write ? "EXMODE" : "PRMODE"); - - if (!ocfs2_is_hard_readonly(OCFS2_SB(inode->i_sb)) && - !ocfs2_mount_local(osb)) - ocfs2_cluster_unlock(OCFS2_SB(inode->i_sb), lockres, level); + ocfs2_wake_downconvert_thread(osb); mlog_exit_void(); } @@ -1442,11 +1639,11 @@ static u64 ocfs2_pack_timespec(struct timespec *spec) /* Call this with the lockres locked. I am reasonably sure we don't * need ip_lock in this function as anyone who would be changing those - * values is supposed to be blocked in ocfs2_meta_lock right now. */ + * values is supposed to be blocked in ocfs2_inode_lock right now. */ static void __ocfs2_stuff_meta_lvb(struct inode *inode) { struct ocfs2_inode_info *oi = OCFS2_I(inode); - struct ocfs2_lock_res *lockres = &oi->ip_meta_lockres; + struct ocfs2_lock_res *lockres = &oi->ip_inode_lockres; struct ocfs2_meta_lvb *lvb; mlog_entry_void(); @@ -1496,7 +1693,7 @@ static void ocfs2_unpack_timespec(struct timespec *spec, static void ocfs2_refresh_inode_from_lvb(struct inode *inode) { struct ocfs2_inode_info *oi = OCFS2_I(inode); - struct ocfs2_lock_res *lockres = &oi->ip_meta_lockres; + struct ocfs2_lock_res *lockres = &oi->ip_inode_lockres; struct ocfs2_meta_lvb *lvb; mlog_entry_void(); @@ -1604,12 +1801,12 @@ static inline void ocfs2_complete_lock_res_refresh(struct ocfs2_lock_res *lockre } /* may or may not return a bh if it went to disk. */ -static int ocfs2_meta_lock_update(struct inode *inode, +static int ocfs2_inode_lock_update(struct inode *inode, struct buffer_head **bh) { int status = 0; struct ocfs2_inode_info *oi = OCFS2_I(inode); - struct ocfs2_lock_res *lockres = &oi->ip_meta_lockres; + struct ocfs2_lock_res *lockres = &oi->ip_inode_lockres; struct ocfs2_dinode *fe; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); @@ -1721,7 +1918,7 @@ static int ocfs2_assign_bh(struct inode *inode, * returns < 0 error if the callback will never be called, otherwise * the result of the lock will be communicated via the callback. */ -int ocfs2_meta_lock_full(struct inode *inode, +int ocfs2_inode_lock_full(struct inode *inode, struct buffer_head **ret_bh, int ex, int arg_flags) @@ -1756,7 +1953,7 @@ int ocfs2_meta_lock_full(struct inode *inode, wait_event(osb->recovery_event, ocfs2_node_map_is_empty(osb, &osb->recovery_map)); - lockres = &OCFS2_I(inode)->ip_meta_lockres; + lockres = &OCFS2_I(inode)->ip_inode_lockres; level = ex ? LKM_EXMODE : LKM_PRMODE; dlm_flags = 0; if (arg_flags & OCFS2_META_LOCK_NOQUEUE) @@ -1795,11 +1992,11 @@ local: } /* This is fun. The caller may want a bh back, or it may - * not. ocfs2_meta_lock_update definitely wants one in, but + * not. ocfs2_inode_lock_update definitely wants one in, but * may or may not read one, depending on what's in the * LVB. The result of all of this is that we've *only* gone to * disk if we have to, so the complexity is worthwhile. */ - status = ocfs2_meta_lock_update(inode, &local_bh); + status = ocfs2_inode_lock_update(inode, &local_bh); if (status < 0) { if (status != -ENOENT) mlog_errno(status); @@ -1821,7 +2018,7 @@ bail: *ret_bh = NULL; } if (acquired) - ocfs2_meta_unlock(inode, ex); + ocfs2_inode_unlock(inode, ex); } if (local_bh) @@ -1832,19 +2029,20 @@ bail: } /* - * This is working around a lock inversion between tasks acquiring DLM locks - * while holding a page lock and the vote thread which blocks dlm lock acquiry - * while acquiring page locks. + * This is working around a lock inversion between tasks acquiring DLM + * locks while holding a page lock and the downconvert thread which + * blocks dlm lock acquiry while acquiring page locks. * * ** These _with_page variantes are only intended to be called from aop * methods that hold page locks and return a very specific *positive* error * code that aop methods pass up to the VFS -- test for errors with != 0. ** * - * The DLM is called such that it returns -EAGAIN if it would have blocked - * waiting for the vote thread. In that case we unlock our page so the vote - * thread can make progress. Once we've done this we have to return - * AOP_TRUNCATED_PAGE so the aop method that called us can bubble that back up - * into the VFS who will then immediately retry the aop call. + * The DLM is called such that it returns -EAGAIN if it would have + * blocked waiting for the downconvert thread. In that case we unlock + * our page so the downconvert thread can make progress. Once we've + * done this we have to return AOP_TRUNCATED_PAGE so the aop method + * that called us can bubble that back up into the VFS who will then + * immediately retry the aop call. * * We do a blocking lock and immediate unlock before returning, though, so that * the lock has a great chance of being cached on this node by the time the VFS @@ -1852,32 +2050,32 @@ bail: * ping locks back and forth, but that's a risk we're willing to take to avoid * the lock inversion simply. */ -int ocfs2_meta_lock_with_page(struct inode *inode, +int ocfs2_inode_lock_with_page(struct inode *inode, struct buffer_head **ret_bh, int ex, struct page *page) { int ret; - ret = ocfs2_meta_lock_full(inode, ret_bh, ex, OCFS2_LOCK_NONBLOCK); + ret = ocfs2_inode_lock_full(inode, ret_bh, ex, OCFS2_LOCK_NONBLOCK); if (ret == -EAGAIN) { unlock_page(page); - if (ocfs2_meta_lock(inode, ret_bh, ex) == 0) - ocfs2_meta_unlock(inode, ex); + if (ocfs2_inode_lock(inode, ret_bh, ex) == 0) + ocfs2_inode_unlock(inode, ex); ret = AOP_TRUNCATED_PAGE; } return ret; } -int ocfs2_meta_lock_atime(struct inode *inode, +int ocfs2_inode_lock_atime(struct inode *inode, struct vfsmount *vfsmnt, int *level) { int ret; mlog_entry_void(); - ret = ocfs2_meta_lock(inode, NULL, 0); + ret = ocfs2_inode_lock(inode, NULL, 0); if (ret < 0) { mlog_errno(ret); return ret; @@ -1890,8 +2088,8 @@ int ocfs2_meta_lock_atime(struct inode *inode, if (ocfs2_should_update_atime(inode, vfsmnt)) { struct buffer_head *bh = NULL; - ocfs2_meta_unlock(inode, 0); - ret = ocfs2_meta_lock(inode, &bh, 1); + ocfs2_inode_unlock(inode, 0); + ret = ocfs2_inode_lock(inode, &bh, 1); if (ret < 0) { mlog_errno(ret); return ret; @@ -1908,11 +2106,11 @@ int ocfs2_meta_lock_atime(struct inode *inode, return ret; } -void ocfs2_meta_unlock(struct inode *inode, +void ocfs2_inode_unlock(struct inode *inode, int ex) { int level = ex ? LKM_EXMODE : LKM_PRMODE; - struct ocfs2_lock_res *lockres = &OCFS2_I(inode)->ip_meta_lockres; + struct ocfs2_lock_res *lockres = &OCFS2_I(inode)->ip_inode_lockres; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); mlog_entry_void(); @@ -2211,7 +2409,7 @@ static int ocfs2_dlm_seq_show(struct seq_file *m, void *v) return 0; } -static struct seq_operations ocfs2_dlm_seq_ops = { +static const struct seq_operations ocfs2_dlm_seq_ops = { .start = ocfs2_dlm_seq_start, .stop = ocfs2_dlm_seq_stop, .next = ocfs2_dlm_seq_next, @@ -2320,11 +2518,11 @@ int ocfs2_dlm_init(struct ocfs2_super *osb) goto bail; } - /* launch vote thread */ - osb->vote_task = kthread_run(ocfs2_vote_thread, osb, "ocfs2vote"); - if (IS_ERR(osb->vote_task)) { - status = PTR_ERR(osb->vote_task); - osb->vote_task = NULL; + /* launch downconvert thread */ + osb->dc_task = kthread_run(ocfs2_downconvert_thread, osb, "ocfs2dc"); + if (IS_ERR(osb->dc_task)) { + status = PTR_ERR(osb->dc_task); + osb->dc_task = NULL; mlog_errno(status); goto bail; } @@ -2334,7 +2532,8 @@ int ocfs2_dlm_init(struct ocfs2_super *osb) dlm_key = crc32_le(0, osb->uuid_str, strlen(osb->uuid_str)); /* for now, uuid == domain */ - dlm = dlm_register_domain(osb->uuid_str, dlm_key); + dlm = dlm_register_domain(osb->uuid_str, dlm_key, + &osb->osb_locking_proto); if (IS_ERR(dlm)) { status = PTR_ERR(dlm); mlog_errno(status); @@ -2353,8 +2552,8 @@ local: bail: if (status < 0) { ocfs2_dlm_shutdown_debug(osb); - if (osb->vote_task) - kthread_stop(osb->vote_task); + if (osb->dc_task) + kthread_stop(osb->dc_task); } mlog_exit(status); @@ -2369,9 +2568,9 @@ void ocfs2_dlm_shutdown(struct ocfs2_super *osb) ocfs2_drop_osb_locks(osb); - if (osb->vote_task) { - kthread_stop(osb->vote_task); - osb->vote_task = NULL; + if (osb->dc_task) { + kthread_stop(osb->dc_task); + osb->dc_task = NULL; } ocfs2_lock_res_free(&osb->osb_super_lockres); @@ -2527,7 +2726,7 @@ out: /* Mark the lockres as being dropped. It will no longer be * queued if blocking, but we still may have to wait on it - * being dequeued from the vote thread before we can consider + * being dequeued from the downconvert thread before we can consider * it safe to drop. * * You can *not* attempt to call cluster_lock on this lockres anymore. */ @@ -2590,14 +2789,7 @@ int ocfs2_drop_inode_locks(struct inode *inode) status = err; err = ocfs2_drop_lock(OCFS2_SB(inode->i_sb), - &OCFS2_I(inode)->ip_data_lockres); - if (err < 0) - mlog_errno(err); - if (err < 0 && !status) - status = err; - - err = ocfs2_drop_lock(OCFS2_SB(inode->i_sb), - &OCFS2_I(inode)->ip_meta_lockres); + &OCFS2_I(inode)->ip_inode_lockres); if (err < 0) mlog_errno(err); if (err < 0 && !status) @@ -2850,6 +3042,9 @@ static int ocfs2_data_convert_worker(struct ocfs2_lock_res *lockres, inode = ocfs2_lock_res_inode(lockres); mapping = inode->i_mapping; + if (!S_ISREG(inode->i_mode)) + goto out; + /* * We need this before the filemap_fdatawrite() so that it can * transfer the dirty bit from the PTE to the @@ -2875,6 +3070,7 @@ static int ocfs2_data_convert_worker(struct ocfs2_lock_res *lockres, filemap_fdatawait(mapping); } +out: return UNBLOCK_CONTINUE; } @@ -2903,7 +3099,7 @@ static void ocfs2_set_meta_lvb(struct ocfs2_lock_res *lockres) /* * Does the final reference drop on our dentry lock. Right now this - * happens in the vote thread, but we could choose to simplify the + * happens in the downconvert thread, but we could choose to simplify the * dlmglue API and push these off to the ocfs2_wq in the future. */ static void ocfs2_dentry_post_unlock(struct ocfs2_super *osb, @@ -3023,8 +3219,8 @@ static int ocfs2_dentry_convert_worker(struct ocfs2_lock_res *lockres, return UNBLOCK_CONTINUE_POST; } -void ocfs2_process_blocked_lock(struct ocfs2_super *osb, - struct ocfs2_lock_res *lockres) +static void ocfs2_process_blocked_lock(struct ocfs2_super *osb, + struct ocfs2_lock_res *lockres) { int status; struct ocfs2_unblock_ctl ctl = {0, 0,}; @@ -3042,7 +3238,7 @@ void ocfs2_process_blocked_lock(struct ocfs2_super *osb, mlog(0, "lockres %s blocked.\n", lockres->l_name); /* Detect whether a lock has been marked as going away while - * the vote thread was processing other things. A lock can + * the downconvert thread was processing other things. A lock can * still be marked with OCFS2_LOCK_FREEING after this check, * but short circuiting here will still save us some * performance. */ @@ -3091,13 +3287,104 @@ static void ocfs2_schedule_blocked_lock(struct ocfs2_super *osb, lockres_or_flags(lockres, OCFS2_LOCK_QUEUED); - spin_lock(&osb->vote_task_lock); + spin_lock(&osb->dc_task_lock); if (list_empty(&lockres->l_blocked_list)) { list_add_tail(&lockres->l_blocked_list, &osb->blocked_lock_list); osb->blocked_lock_count++; } - spin_unlock(&osb->vote_task_lock); + spin_unlock(&osb->dc_task_lock); mlog_exit_void(); } + +static void ocfs2_downconvert_thread_do_work(struct ocfs2_super *osb) +{ + unsigned long processed; + struct ocfs2_lock_res *lockres; + + mlog_entry_void(); + + spin_lock(&osb->dc_task_lock); + /* grab this early so we know to try again if a state change and + * wake happens part-way through our work */ + osb->dc_work_sequence = osb->dc_wake_sequence; + + processed = osb->blocked_lock_count; + while (processed) { + BUG_ON(list_empty(&osb->blocked_lock_list)); + + lockres = list_entry(osb->blocked_lock_list.next, + struct ocfs2_lock_res, l_blocked_list); + list_del_init(&lockres->l_blocked_list); + osb->blocked_lock_count--; + spin_unlock(&osb->dc_task_lock); + + BUG_ON(!processed); + processed--; + + ocfs2_process_blocked_lock(osb, lockres); + + spin_lock(&osb->dc_task_lock); + } + spin_unlock(&osb->dc_task_lock); + + mlog_exit_void(); +} + +static int ocfs2_downconvert_thread_lists_empty(struct ocfs2_super *osb) +{ + int empty = 0; + + spin_lock(&osb->dc_task_lock); + if (list_empty(&osb->blocked_lock_list)) + empty = 1; + + spin_unlock(&osb->dc_task_lock); + return empty; +} + +static int ocfs2_downconvert_thread_should_wake(struct ocfs2_super *osb) +{ + int should_wake = 0; + + spin_lock(&osb->dc_task_lock); + if (osb->dc_work_sequence != osb->dc_wake_sequence) + should_wake = 1; + spin_unlock(&osb->dc_task_lock); + + return should_wake; +} + +static int ocfs2_downconvert_thread(void *arg) +{ + int status = 0; + struct ocfs2_super *osb = arg; + + /* only quit once we've been asked to stop and there is no more + * work available */ + while (!(kthread_should_stop() && + ocfs2_downconvert_thread_lists_empty(osb))) { + + wait_event_interruptible(osb->dc_event, + ocfs2_downconvert_thread_should_wake(osb) || + kthread_should_stop()); + + mlog(0, "downconvert_thread: awoken\n"); + + ocfs2_downconvert_thread_do_work(osb); + } + + osb->dc_task = NULL; + return status; +} + +void ocfs2_wake_downconvert_thread(struct ocfs2_super *osb) +{ + spin_lock(&osb->dc_task_lock); + /* make sure the voting thread gets a swipe at whatever changes + * the caller may have made to the voting state */ + osb->dc_wake_sequence++; + spin_unlock(&osb->dc_task_lock); + wake_up(&osb->dc_event); +} diff --git a/fs/ocfs2/dlmglue.h b/fs/ocfs2/dlmglue.h index 87a785e4120..e3cf902404b 100644 --- a/fs/ocfs2/dlmglue.h +++ b/fs/ocfs2/dlmglue.h @@ -49,12 +49,12 @@ struct ocfs2_meta_lvb { __be32 lvb_reserved2; }; -/* ocfs2_meta_lock_full() and ocfs2_data_lock_full() 'arg_flags' flags */ +/* ocfs2_inode_lock_full() 'arg_flags' flags */ /* don't wait on recovery. */ #define OCFS2_META_LOCK_RECOVERY (0x01) /* Instruct the dlm not to queue ourselves on the other node. */ #define OCFS2_META_LOCK_NOQUEUE (0x02) -/* don't block waiting for the vote thread, instead return -EAGAIN */ +/* don't block waiting for the downconvert thread, instead return -EAGAIN */ #define OCFS2_LOCK_NONBLOCK (0x04) int ocfs2_dlm_init(struct ocfs2_super *osb); @@ -66,38 +66,32 @@ void ocfs2_inode_lock_res_init(struct ocfs2_lock_res *res, struct inode *inode); void ocfs2_dentry_lock_res_init(struct ocfs2_dentry_lock *dl, u64 parent, struct inode *inode); +struct ocfs2_file_private; +void ocfs2_file_lock_res_init(struct ocfs2_lock_res *lockres, + struct ocfs2_file_private *fp); void ocfs2_lock_res_free(struct ocfs2_lock_res *res); int ocfs2_create_new_inode_locks(struct inode *inode); int ocfs2_drop_inode_locks(struct inode *inode); -int ocfs2_data_lock_full(struct inode *inode, - int write, - int arg_flags); -#define ocfs2_data_lock(inode, write) ocfs2_data_lock_full(inode, write, 0) -int ocfs2_data_lock_with_page(struct inode *inode, - int write, - struct page *page); -void ocfs2_data_unlock(struct inode *inode, - int write); int ocfs2_rw_lock(struct inode *inode, int write); void ocfs2_rw_unlock(struct inode *inode, int write); int ocfs2_open_lock(struct inode *inode); int ocfs2_try_open_lock(struct inode *inode, int write); void ocfs2_open_unlock(struct inode *inode); -int ocfs2_meta_lock_atime(struct inode *inode, +int ocfs2_inode_lock_atime(struct inode *inode, struct vfsmount *vfsmnt, int *level); -int ocfs2_meta_lock_full(struct inode *inode, +int ocfs2_inode_lock_full(struct inode *inode, struct buffer_head **ret_bh, int ex, int arg_flags); -int ocfs2_meta_lock_with_page(struct inode *inode, +int ocfs2_inode_lock_with_page(struct inode *inode, struct buffer_head **ret_bh, int ex, struct page *page); /* 99% of the time we don't want to supply any additional flags -- * those are for very specific cases only. */ -#define ocfs2_meta_lock(i, b, e) ocfs2_meta_lock_full(i, b, e, 0) -void ocfs2_meta_unlock(struct inode *inode, +#define ocfs2_inode_lock(i, b, e) ocfs2_inode_lock_full(i, b, e, 0) +void ocfs2_inode_unlock(struct inode *inode, int ex); int ocfs2_super_lock(struct ocfs2_super *osb, int ex); @@ -107,16 +101,18 @@ int ocfs2_rename_lock(struct ocfs2_super *osb); void ocfs2_rename_unlock(struct ocfs2_super *osb); int ocfs2_dentry_lock(struct dentry *dentry, int ex); void ocfs2_dentry_unlock(struct dentry *dentry, int ex); +int ocfs2_file_lock(struct file *file, int ex, int trylock); +void ocfs2_file_unlock(struct file *file); void ocfs2_mark_lockres_freeing(struct ocfs2_lock_res *lockres); void ocfs2_simple_drop_lockres(struct ocfs2_super *osb, struct ocfs2_lock_res *lockres); -/* for the vote thread */ -void ocfs2_process_blocked_lock(struct ocfs2_super *osb, - struct ocfs2_lock_res *lockres); +/* for the downconvert thread */ +void ocfs2_wake_downconvert_thread(struct ocfs2_super *osb); struct ocfs2_dlm_debug *ocfs2_new_dlm_debug(void); void ocfs2_put_dlm_debug(struct ocfs2_dlm_debug *dlm_debug); +extern const struct dlm_protocol_version ocfs2_locking_protocol; #endif /* DLMGLUE_H */ diff --git a/fs/ocfs2/export.c b/fs/ocfs2/export.c index 535bfa9568a..67527cebf21 100644 --- a/fs/ocfs2/export.c +++ b/fs/ocfs2/export.c @@ -58,7 +58,7 @@ static struct dentry *ocfs2_get_dentry(struct super_block *sb, return ERR_PTR(-ESTALE); } - inode = ocfs2_iget(OCFS2_SB(sb), handle->ih_blkno, 0); + inode = ocfs2_iget(OCFS2_SB(sb), handle->ih_blkno, 0, 0); if (IS_ERR(inode)) return (void *)inode; @@ -95,7 +95,7 @@ static struct dentry *ocfs2_get_parent(struct dentry *child) mlog(0, "find parent of directory %llu\n", (unsigned long long)OCFS2_I(dir)->ip_blkno); - status = ocfs2_meta_lock(dir, NULL, 0); + status = ocfs2_inode_lock(dir, NULL, 0); if (status < 0) { if (status != -ENOENT) mlog_errno(status); @@ -109,7 +109,7 @@ static struct dentry *ocfs2_get_parent(struct dentry *child) goto bail_unlock; } - inode = ocfs2_iget(OCFS2_SB(dir->i_sb), blkno, 0); + inode = ocfs2_iget(OCFS2_SB(dir->i_sb), blkno, 0, 0); if (IS_ERR(inode)) { mlog(ML_ERROR, "Unable to create inode %llu\n", (unsigned long long)blkno); @@ -126,7 +126,7 @@ static struct dentry *ocfs2_get_parent(struct dentry *child) parent->d_op = &ocfs2_dentry_ops; bail_unlock: - ocfs2_meta_unlock(dir, 0); + ocfs2_inode_unlock(dir, 0); bail: mlog_exit_ptr(parent); diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index b75b2e1f0e4..ed5d5232e85 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -51,6 +51,7 @@ #include "inode.h" #include "ioctl.h" #include "journal.h" +#include "locks.h" #include "mmap.h" #include "suballoc.h" #include "super.h" @@ -63,6 +64,35 @@ static int ocfs2_sync_inode(struct inode *inode) return sync_mapping_buffers(inode->i_mapping); } +static int ocfs2_init_file_private(struct inode *inode, struct file *file) +{ + struct ocfs2_file_private *fp; + + fp = kzalloc(sizeof(struct ocfs2_file_private), GFP_KERNEL); + if (!fp) + return -ENOMEM; + + fp->fp_file = file; + mutex_init(&fp->fp_mutex); + ocfs2_file_lock_res_init(&fp->fp_flock, fp); + file->private_data = fp; + + return 0; +} + +static void ocfs2_free_file_private(struct inode *inode, struct file *file) +{ + struct ocfs2_file_private *fp = file->private_data; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + + if (fp) { + ocfs2_simple_drop_lockres(osb, &fp->fp_flock); + ocfs2_lock_res_free(&fp->fp_flock); + kfree(fp); + file->private_data = NULL; + } +} + static int ocfs2_file_open(struct inode *inode, struct file *file) { int status; @@ -89,7 +119,18 @@ static int ocfs2_file_open(struct inode *inode, struct file *file) oi->ip_open_count++; spin_unlock(&oi->ip_lock); - status = 0; + + status = ocfs2_init_file_private(inode, file); + if (status) { + /* + * We want to set open count back if we're failing the + * open. + */ + spin_lock(&oi->ip_lock); + oi->ip_open_count--; + spin_unlock(&oi->ip_lock); + } + leave: mlog_exit(status); return status; @@ -108,11 +149,24 @@ static int ocfs2_file_release(struct inode *inode, struct file *file) oi->ip_flags &= ~OCFS2_INODE_OPEN_DIRECT; spin_unlock(&oi->ip_lock); + ocfs2_free_file_private(inode, file); + mlog_exit(0); return 0; } +static int ocfs2_dir_open(struct inode *inode, struct file *file) +{ + return ocfs2_init_file_private(inode, file); +} + +static int ocfs2_dir_release(struct inode *inode, struct file *file) +{ + ocfs2_free_file_private(inode, file); + return 0; +} + static int ocfs2_sync_file(struct file *file, struct dentry *dentry, int datasync) @@ -382,18 +436,13 @@ static int ocfs2_truncate_file(struct inode *inode, down_write(&OCFS2_I(inode)->ip_alloc_sem); - /* This forces other nodes to sync and drop their pages. Do - * this even if we have a truncate without allocation change - - * ocfs2 cluster sizes can be much greater than page size, so - * we have to truncate them anyway. */ - status = ocfs2_data_lock(inode, 1); - if (status < 0) { - up_write(&OCFS2_I(inode)->ip_alloc_sem); - - mlog_errno(status); - goto bail; - } - + /* + * The inode lock forced other nodes to sync and drop their + * pages, which (correctly) happens even if we have a truncate + * without allocation change - ocfs2 cluster sizes can be much + * greater than page size, so we have to truncate them + * anyway. + */ unmap_mapping_range(inode->i_mapping, new_i_size + PAGE_SIZE - 1, 0, 1); truncate_inode_pages(inode->i_mapping, new_i_size); @@ -403,7 +452,7 @@ static int ocfs2_truncate_file(struct inode *inode, if (status) mlog_errno(status); - goto bail_unlock_data; + goto bail_unlock_sem; } /* alright, we're going to need to do a full blown alloc size @@ -413,25 +462,23 @@ static int ocfs2_truncate_file(struct inode *inode, status = ocfs2_orphan_for_truncate(osb, inode, di_bh, new_i_size); if (status < 0) { mlog_errno(status); - goto bail_unlock_data; + goto bail_unlock_sem; } status = ocfs2_prepare_truncate(osb, inode, di_bh, &tc); if (status < 0) { mlog_errno(status); - goto bail_unlock_data; + goto bail_unlock_sem; } status = ocfs2_commit_truncate(osb, inode, di_bh, tc); if (status < 0) { mlog_errno(status); - goto bail_unlock_data; + goto bail_unlock_sem; } /* TODO: orphan dir cleanup here. */ -bail_unlock_data: - ocfs2_data_unlock(inode, 1); - +bail_unlock_sem: up_write(&OCFS2_I(inode)->ip_alloc_sem); bail: @@ -579,7 +626,7 @@ int ocfs2_lock_allocators(struct inode *inode, struct ocfs2_dinode *di, mlog(0, "extend inode %llu, i_size = %lld, di->i_clusters = %u, " "clusters_to_add = %u, extents_to_split = %u\n", - (unsigned long long)OCFS2_I(inode)->ip_blkno, i_size_read(inode), + (unsigned long long)OCFS2_I(inode)->ip_blkno, (long long)i_size_read(inode), le32_to_cpu(di->i_clusters), clusters_to_add, extents_to_split); num_free_extents = ocfs2_num_free_extents(osb, inode, di); @@ -760,7 +807,7 @@ restarted_transaction: le32_to_cpu(fe->i_clusters), (unsigned long long)le64_to_cpu(fe->i_size)); mlog(0, "inode: ip_clusters=%u, i_size=%lld\n", - OCFS2_I(inode)->ip_clusters, i_size_read(inode)); + OCFS2_I(inode)->ip_clusters, (long long)i_size_read(inode)); leave: if (handle) { @@ -917,7 +964,7 @@ static int ocfs2_extend_file(struct inode *inode, struct buffer_head *di_bh, u64 new_i_size) { - int ret = 0, data_locked = 0; + int ret = 0; struct ocfs2_inode_info *oi = OCFS2_I(inode); BUG_ON(!di_bh); @@ -943,20 +990,6 @@ static int ocfs2_extend_file(struct inode *inode, && ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb))) goto out_update_size; - /* - * protect the pages that ocfs2_zero_extend is going to be - * pulling into the page cache.. we do this before the - * metadata extend so that we don't get into the situation - * where we've extended the metadata but can't get the data - * lock to zero. - */ - ret = ocfs2_data_lock(inode, 1); - if (ret < 0) { - mlog_errno(ret); - goto out; - } - data_locked = 1; - /* * The alloc sem blocks people in read/write from reading our * allocation until we're done changing it. We depend on @@ -980,7 +1013,7 @@ static int ocfs2_extend_file(struct inode *inode, up_write(&oi->ip_alloc_sem); mlog_errno(ret); - goto out_unlock; + goto out; } } @@ -991,7 +1024,7 @@ static int ocfs2_extend_file(struct inode *inode, if (ret < 0) { mlog_errno(ret); - goto out_unlock; + goto out; } out_update_size: @@ -999,10 +1032,6 @@ out_update_size: if (ret < 0) mlog_errno(ret); -out_unlock: - if (data_locked) - ocfs2_data_unlock(inode, 1); - out: return ret; } @@ -1050,7 +1079,7 @@ int ocfs2_setattr(struct dentry *dentry, struct iattr *attr) } } - status = ocfs2_meta_lock(inode, &bh, 1); + status = ocfs2_inode_lock(inode, &bh, 1); if (status < 0) { if (status != -ENOENT) mlog_errno(status); @@ -1102,7 +1131,7 @@ int ocfs2_setattr(struct dentry *dentry, struct iattr *attr) bail_commit: ocfs2_commit_trans(osb, handle); bail_unlock: - ocfs2_meta_unlock(inode, 1); + ocfs2_inode_unlock(inode, 1); bail_unlock_rw: if (size_change) ocfs2_rw_unlock(inode, 1); @@ -1149,7 +1178,7 @@ int ocfs2_permission(struct inode *inode, int mask, struct nameidata *nd) mlog_entry_void(); - ret = ocfs2_meta_lock(inode, NULL, 0); + ret = ocfs2_inode_lock(inode, NULL, 0); if (ret) { if (ret != -ENOENT) mlog_errno(ret); @@ -1158,7 +1187,7 @@ int ocfs2_permission(struct inode *inode, int mask, struct nameidata *nd) ret = generic_permission(inode, mask, NULL); - ocfs2_meta_unlock(inode, 0); + ocfs2_inode_unlock(inode, 0); out: mlog_exit(ret); return ret; @@ -1630,7 +1659,7 @@ static int __ocfs2_change_file_space(struct file *file, struct inode *inode, goto out; } - ret = ocfs2_meta_lock(inode, &di_bh, 1); + ret = ocfs2_inode_lock(inode, &di_bh, 1); if (ret) { mlog_errno(ret); goto out_rw_unlock; @@ -1638,7 +1667,7 @@ static int __ocfs2_change_file_space(struct file *file, struct inode *inode, if (inode->i_flags & (S_IMMUTABLE|S_APPEND)) { ret = -EPERM; - goto out_meta_unlock; + goto out_inode_unlock; } switch (sr->l_whence) { @@ -1652,7 +1681,7 @@ static int __ocfs2_change_file_space(struct file *file, struct inode *inode, break; default: ret = -EINVAL; - goto out_meta_unlock; + goto out_inode_unlock; } sr->l_whence = 0; @@ -1663,14 +1692,14 @@ static int __ocfs2_change_file_space(struct file *file, struct inode *inode, || (sr->l_start + llen) < 0 || (sr->l_start + llen) > max_off) { ret = -EINVAL; - goto out_meta_unlock; + goto out_inode_unlock; } size = sr->l_start + sr->l_len; if (cmd == OCFS2_IOC_RESVSP || cmd == OCFS2_IOC_RESVSP64) { if (sr->l_len <= 0) { ret = -EINVAL; - goto out_meta_unlock; + goto out_inode_unlock; } } @@ -1678,7 +1707,7 @@ static int __ocfs2_change_file_space(struct file *file, struct inode *inode, ret = __ocfs2_write_remove_suid(inode, di_bh); if (ret) { mlog_errno(ret); - goto out_meta_unlock; + goto out_inode_unlock; } } @@ -1704,7 +1733,7 @@ static int __ocfs2_change_file_space(struct file *file, struct inode *inode, up_write(&OCFS2_I(inode)->ip_alloc_sem); if (ret) { mlog_errno(ret); - goto out_meta_unlock; + goto out_inode_unlock; } /* @@ -1714,7 +1743,7 @@ static int __ocfs2_change_file_space(struct file *file, struct inode *inode, if (IS_ERR(handle)) { ret = PTR_ERR(handle); mlog_errno(ret); - goto out_meta_unlock; + goto out_inode_unlock; } if (change_size && i_size_read(inode) < size) @@ -1727,9 +1756,9 @@ static int __ocfs2_change_file_space(struct file *file, struct inode *inode, ocfs2_commit_trans(osb, handle); -out_meta_unlock: +out_inode_unlock: brelse(di_bh); - ocfs2_meta_unlock(inode, 1); + ocfs2_inode_unlock(inode, 1); out_rw_unlock: ocfs2_rw_unlock(inode, 1); @@ -1799,7 +1828,7 @@ static int ocfs2_prepare_inode_for_write(struct dentry *dentry, * if we need to make modifications here. */ for(;;) { - ret = ocfs2_meta_lock(inode, NULL, meta_level); + ret = ocfs2_inode_lock(inode, NULL, meta_level); if (ret < 0) { meta_level = -1; mlog_errno(ret); @@ -1817,7 +1846,7 @@ static int ocfs2_prepare_inode_for_write(struct dentry *dentry, * set inode->i_size at the end of a write. */ if (should_remove_suid(dentry)) { if (meta_level == 0) { - ocfs2_meta_unlock(inode, meta_level); + ocfs2_inode_unlock(inode, meta_level); meta_level = 1; continue; } @@ -1886,7 +1915,7 @@ static int ocfs2_prepare_inode_for_write(struct dentry *dentry, *ppos = saved_pos; out_unlock: - ocfs2_meta_unlock(inode, meta_level); + ocfs2_inode_unlock(inode, meta_level); out: return ret; @@ -2099,12 +2128,12 @@ static ssize_t ocfs2_file_splice_read(struct file *in, /* * See the comment in ocfs2_file_aio_read() */ - ret = ocfs2_meta_lock(inode, NULL, 0); + ret = ocfs2_inode_lock(inode, NULL, 0); if (ret < 0) { mlog_errno(ret); goto bail; } - ocfs2_meta_unlock(inode, 0); + ocfs2_inode_unlock(inode, 0); ret = generic_file_splice_read(in, ppos, pipe, len, flags); @@ -2160,12 +2189,12 @@ static ssize_t ocfs2_file_aio_read(struct kiocb *iocb, * like i_size. This allows the checks down below * generic_file_aio_read() a chance of actually working. */ - ret = ocfs2_meta_lock_atime(inode, filp->f_vfsmnt, &lock_level); + ret = ocfs2_inode_lock_atime(inode, filp->f_vfsmnt, &lock_level); if (ret < 0) { mlog_errno(ret); goto bail; } - ocfs2_meta_unlock(inode, lock_level); + ocfs2_inode_unlock(inode, lock_level); ret = generic_file_aio_read(iocb, iov, nr_segs, iocb->ki_pos); if (ret == -EINVAL) @@ -2204,6 +2233,7 @@ const struct inode_operations ocfs2_special_file_iops = { }; const struct file_operations ocfs2_fops = { + .llseek = generic_file_llseek, .read = do_sync_read, .write = do_sync_write, .mmap = ocfs2_mmap, @@ -2216,16 +2246,21 @@ const struct file_operations ocfs2_fops = { #ifdef CONFIG_COMPAT .compat_ioctl = ocfs2_compat_ioctl, #endif + .flock = ocfs2_flock, .splice_read = ocfs2_file_splice_read, .splice_write = ocfs2_file_splice_write, }; const struct file_operations ocfs2_dops = { + .llseek = generic_file_llseek, .read = generic_read_dir, .readdir = ocfs2_readdir, .fsync = ocfs2_sync_file, + .release = ocfs2_dir_release, + .open = ocfs2_dir_open, .ioctl = ocfs2_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = ocfs2_compat_ioctl, #endif + .flock = ocfs2_flock, }; diff --git a/fs/ocfs2/file.h b/fs/ocfs2/file.h index 066f14add3a..048ddcaf5c8 100644 --- a/fs/ocfs2/file.h +++ b/fs/ocfs2/file.h @@ -32,6 +32,12 @@ extern const struct inode_operations ocfs2_file_iops; extern const struct inode_operations ocfs2_special_file_iops; struct ocfs2_alloc_context; +struct ocfs2_file_private { + struct file *fp_file; + struct mutex fp_mutex; + struct ocfs2_lock_res fp_flock; +}; + enum ocfs2_alloc_restarted { RESTART_NONE = 0, RESTART_TRANS, diff --git a/fs/ocfs2/heartbeat.c b/fs/ocfs2/heartbeat.c index c4c36171240..0758daf64da 100644 --- a/fs/ocfs2/heartbeat.c +++ b/fs/ocfs2/heartbeat.c @@ -30,9 +30,6 @@ #include <linux/highmem.h> #include <linux/kmod.h> -#include <cluster/heartbeat.h> -#include <cluster/nodemanager.h> - #include <dlm/dlmapi.h> #define MLOG_MASK_PREFIX ML_SUPER @@ -44,29 +41,28 @@ #include "heartbeat.h" #include "inode.h" #include "journal.h" -#include "vote.h" #include "buffer_head_io.h" -#define OCFS2_HB_NODE_DOWN_PRI (0x0000002) -#define OCFS2_HB_NODE_UP_PRI OCFS2_HB_NODE_DOWN_PRI - static inline void __ocfs2_node_map_set_bit(struct ocfs2_node_map *map, int bit); static inline void __ocfs2_node_map_clear_bit(struct ocfs2_node_map *map, int bit); static inline int __ocfs2_node_map_is_empty(struct ocfs2_node_map *map); -static void __ocfs2_node_map_dup(struct ocfs2_node_map *target, - struct ocfs2_node_map *from); -static void __ocfs2_node_map_set(struct ocfs2_node_map *target, - struct ocfs2_node_map *from); + +/* special case -1 for now + * TODO: should *really* make sure the calling func never passes -1!! */ +static void ocfs2_node_map_init(struct ocfs2_node_map *map) +{ + map->num_nodes = OCFS2_NODE_MAP_MAX_NODES; + memset(map->map, 0, BITS_TO_LONGS(OCFS2_NODE_MAP_MAX_NODES) * + sizeof(unsigned long)); +} void ocfs2_init_node_maps(struct ocfs2_super *osb) { spin_lock_init(&osb->node_map_lock); - ocfs2_node_map_init(&osb->mounted_map); ocfs2_node_map_init(&osb->recovery_map); - ocfs2_node_map_init(&osb->umount_map); ocfs2_node_map_init(&osb->osb_recovering_orphan_dirs); } @@ -87,24 +83,7 @@ static void ocfs2_do_node_down(int node_num, return; } - if (ocfs2_node_map_test_bit(osb, &osb->umount_map, node_num)) { - /* If a node is in the umount map, then we've been - * expecting him to go down and we know ahead of time - * that recovery is not necessary. */ - ocfs2_node_map_clear_bit(osb, &osb->umount_map, node_num); - return; - } - ocfs2_recovery_thread(osb, node_num); - - ocfs2_remove_node_from_vote_queues(osb, node_num); -} - -static void ocfs2_hb_node_down_cb(struct o2nm_node *node, - int node_num, - void *data) -{ - ocfs2_do_node_down(node_num, (struct ocfs2_super *) data); } /* Called from the dlm when it's about to evict a node. We may also @@ -121,27 +100,8 @@ static void ocfs2_dlm_eviction_cb(int node_num, ocfs2_do_node_down(node_num, osb); } -static void ocfs2_hb_node_up_cb(struct o2nm_node *node, - int node_num, - void *data) -{ - struct ocfs2_super *osb = data; - - BUG_ON(osb->node_num == node_num); - - mlog(0, "node up event for %d\n", node_num); - ocfs2_node_map_clear_bit(osb, &osb->umount_map, node_num); -} - void ocfs2_setup_hb_callbacks(struct ocfs2_super *osb) { - o2hb_setup_callback(&osb->osb_hb_down, O2HB_NODE_DOWN_CB, - ocfs2_hb_node_down_cb, osb, - OCFS2_HB_NODE_DOWN_PRI); - - o2hb_setup_callback(&osb->osb_hb_up, O2HB_NODE_UP_CB, - ocfs2_hb_node_up_cb, osb, OCFS2_HB_NODE_UP_PRI); - /* Not exactly a heartbeat callback, but leads to essentially * the same path so we set it up here. */ dlm_setup_eviction_cb(&osb->osb_eviction_cb, @@ -149,39 +109,6 @@ void ocfs2_setup_hb_callbacks(struct ocfs2_super *osb) osb); } -/* Most functions here are just stubs for now... */ -int ocfs2_register_hb_callbacks(struct ocfs2_super *osb) -{ - int status; - - if (ocfs2_mount_local(osb)) - return 0; - - status = o2hb_register_callback(osb->uuid_str, &osb->osb_hb_down); - if (status < 0) { - mlog_errno(status); - goto bail; - } - - status = o2hb_register_callback(osb->uuid_str, &osb->osb_hb_up); - if (status < 0) { - mlog_errno(status); - o2hb_unregister_callback(osb->uuid_str, &osb->osb_hb_down); - } - -bail: - return status; -} - -void ocfs2_clear_hb_callbacks(struct ocfs2_super *osb) -{ - if (ocfs2_mount_local(osb)) - return; - - o2hb_unregister_callback(osb->uuid_str, &osb->osb_hb_down); - o2hb_unregister_callback(osb->uuid_str, &osb->osb_hb_up); -} - void ocfs2_stop_heartbeat(struct ocfs2_super *osb) { int ret; @@ -214,15 +141,6 @@ void ocfs2_stop_heartbeat(struct ocfs2_super *osb) mlog_errno(ret); } -/* special case -1 for now - * TODO: should *really* make sure the calling func never passes -1!! */ -void ocfs2_node_map_init(struct ocfs2_node_map *map) -{ - map->num_nodes = OCFS2_NODE_MAP_MAX_NODES; - memset(map->map, 0, BITS_TO_LONGS(OCFS2_NODE_MAP_MAX_NODES) * - sizeof(unsigned long)); -} - static inline void __ocfs2_node_map_set_bit(struct ocfs2_node_map *map, int bit) { @@ -294,6 +212,8 @@ int ocfs2_node_map_is_empty(struct ocfs2_super *osb, return ret; } +#if 0 + static void __ocfs2_node_map_dup(struct ocfs2_node_map *target, struct ocfs2_node_map *from) { @@ -332,6 +252,8 @@ static void __ocfs2_node_map_set(struct ocfs2_node_map *target, target->map[i] = from->map[i]; } +#endif /* 0 */ + /* Returns whether the recovery bit was actually set - it may not be * if a node is still marked as needing recovery */ int ocfs2_recovery_map_set(struct ocfs2_super *osb, @@ -341,8 +263,6 @@ int ocfs2_recovery_map_set(struct ocfs2_super *osb, spin_lock(&osb->node_map_lock); - __ocfs2_node_map_clear_bit(&osb->mounted_map, num); - if (!test_bit(num, osb->recovery_map.map)) { __ocfs2_node_map_set_bit(&osb->recovery_map, num); set = 1; diff --git a/fs/ocfs2/heartbeat.h b/fs/ocfs2/heartbeat.h index e8fb079122e..eac63aed761 100644 --- a/fs/ocfs2/heartbeat.h +++ b/fs/ocfs2/heartbeat.h @@ -29,13 +29,10 @@ void ocfs2_init_node_maps(struct ocfs2_super *osb); void ocfs2_setup_hb_callbacks(struct ocfs2_super *osb); -int ocfs2_register_hb_callbacks(struct ocfs2_super *osb); -void ocfs2_clear_hb_callbacks(struct ocfs2_super *osb); void ocfs2_stop_heartbeat(struct ocfs2_super *osb); /* node map functions - used to keep track of mounted and in-recovery * nodes. */ -void ocfs2_node_map_init(struct ocfs2_node_map *map); int ocfs2_node_map_is_empty(struct ocfs2_super *osb, struct ocfs2_node_map *map); void ocfs2_node_map_set_bit(struct ocfs2_super *osb, @@ -59,9 +56,5 @@ int ocfs2_recovery_map_set(struct ocfs2_super *osb, int num); void ocfs2_recovery_map_clear(struct ocfs2_super *osb, int num); -/* returns 1 if bit is the only bit set in target, 0 otherwise */ -int ocfs2_node_map_is_only(struct ocfs2_super *osb, - struct ocfs2_node_map *target, - int bit); #endif /* OCFS2_HEARTBEAT_H */ diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c index ebb2bbe30f3..7e9e4c79aec 100644 --- a/fs/ocfs2/inode.c +++ b/fs/ocfs2/inode.c @@ -49,7 +49,6 @@ #include "symlink.h" #include "sysfile.h" #include "uptodate.h" -#include "vote.h" #include "buffer_head_io.h" @@ -58,8 +57,11 @@ struct ocfs2_find_inode_args u64 fi_blkno; unsigned long fi_ino; unsigned int fi_flags; + unsigned int fi_sysfile_type; }; +static struct lock_class_key ocfs2_sysfile_lock_key[NUM_SYSTEM_INODES]; + static int ocfs2_read_locked_inode(struct inode *inode, struct ocfs2_find_inode_args *args); static int ocfs2_init_locked_inode(struct inode *inode, void *opaque); @@ -107,7 +109,8 @@ void ocfs2_get_inode_flags(struct ocfs2_inode_info *oi) oi->ip_attr |= OCFS2_DIRSYNC_FL; } -struct inode *ocfs2_iget(struct ocfs2_super *osb, u64 blkno, int flags) +struct inode *ocfs2_iget(struct ocfs2_super *osb, u64 blkno, unsigned flags, + int sysfile_type) { struct inode *inode = NULL; struct super_block *sb = osb->sb; @@ -127,6 +130,7 @@ struct inode *ocfs2_iget(struct ocfs2_super *osb, u64 blkno, int flags) args.fi_blkno = blkno; args.fi_flags = flags; args.fi_ino = ino_from_blkno(sb, blkno); + args.fi_sysfile_type = sysfile_type; inode = iget5_locked(sb, args.fi_ino, ocfs2_find_actor, ocfs2_init_locked_inode, &args); @@ -201,6 +205,9 @@ static int ocfs2_init_locked_inode(struct inode *inode, void *opaque) inode->i_ino = args->fi_ino; OCFS2_I(inode)->ip_blkno = args->fi_blkno; + if (args->fi_sysfile_type != 0) + lockdep_set_class(&inode->i_mutex, + &ocfs2_sysfile_lock_key[args->fi_sysfile_type]); mlog_exit(0); return 0; @@ -322,7 +329,7 @@ int ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe, */ BUG_ON(le32_to_cpu(fe->i_flags) & OCFS2_SYSTEM_FL); - ocfs2_inode_lock_res_init(&OCFS2_I(inode)->ip_meta_lockres, + ocfs2_inode_lock_res_init(&OCFS2_I(inode)->ip_inode_lockres, OCFS2_LOCK_TYPE_META, 0, inode); ocfs2_inode_lock_res_init(&OCFS2_I(inode)->ip_open_lockres, @@ -333,10 +340,6 @@ int ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe, OCFS2_LOCK_TYPE_RW, inode->i_generation, inode); - ocfs2_inode_lock_res_init(&OCFS2_I(inode)->ip_data_lockres, - OCFS2_LOCK_TYPE_DATA, inode->i_generation, - inode); - ocfs2_set_inode_flags(inode); status = 0; @@ -414,7 +417,7 @@ static int ocfs2_read_locked_inode(struct inode *inode, if (args->fi_flags & OCFS2_FI_FLAG_SYSFILE) generation = osb->fs_generation; - ocfs2_inode_lock_res_init(&OCFS2_I(inode)->ip_meta_lockres, + ocfs2_inode_lock_res_init(&OCFS2_I(inode)->ip_inode_lockres, OCFS2_LOCK_TYPE_META, generation, inode); @@ -429,7 +432,7 @@ static int ocfs2_read_locked_inode(struct inode *inode, mlog_errno(status); return status; } - status = ocfs2_meta_lock(inode, NULL, 0); + status = ocfs2_inode_lock(inode, NULL, 0); if (status) { make_bad_inode(inode); mlog_errno(status); @@ -484,7 +487,7 @@ static int ocfs2_read_locked_inode(struct inode *inode, bail: if (can_lock) - ocfs2_meta_unlock(inode, 0); + ocfs2_inode_unlock(inode, 0); if (status < 0) make_bad_inode(inode); @@ -586,7 +589,7 @@ static int ocfs2_remove_inode(struct inode *inode, } mutex_lock(&inode_alloc_inode->i_mutex); - status = ocfs2_meta_lock(inode_alloc_inode, &inode_alloc_bh, 1); + status = ocfs2_inode_lock(inode_alloc_inode, &inode_alloc_bh, 1); if (status < 0) { mutex_unlock(&inode_alloc_inode->i_mutex); @@ -617,7 +620,7 @@ static int ocfs2_remove_inode(struct inode *inode, } di->i_dtime = cpu_to_le64(CURRENT_TIME.tv_sec); - le32_and_cpu(&di->i_flags, ~(OCFS2_VALID_FL | OCFS2_ORPHANED_FL)); + di->i_flags &= cpu_to_le32(~(OCFS2_VALID_FL | OCFS2_ORPHANED_FL)); status = ocfs2_journal_dirty(handle, di_bh); if (status < 0) { @@ -635,7 +638,7 @@ static int ocfs2_remove_inode(struct inode *inode, bail_commit: ocfs2_commit_trans(osb, handle); bail_unlock: - ocfs2_meta_unlock(inode_alloc_inode, 1); + ocfs2_inode_unlock(inode_alloc_inode, 1); mutex_unlock(&inode_alloc_inode->i_mutex); brelse(inode_alloc_bh); bail: @@ -709,7 +712,7 @@ static int ocfs2_wipe_inode(struct inode *inode, * delete_inode operation. We do this now to avoid races with * recovery completion on other nodes. */ mutex_lock(&orphan_dir_inode->i_mutex); - status = ocfs2_meta_lock(orphan_dir_inode, &orphan_dir_bh, 1); + status = ocfs2_inode_lock(orphan_dir_inode, &orphan_dir_bh, 1); if (status < 0) { mutex_unlock(&orphan_dir_inode->i_mutex); @@ -718,8 +721,8 @@ static int ocfs2_wipe_inode(struct inode *inode, } /* we do this while holding the orphan dir lock because we - * don't want recovery being run from another node to vote for - * an inode delete on us -- this will result in two nodes + * don't want recovery being run from another node to try an + * inode delete underneath us -- this will result in two nodes * truncating the same file! */ status = ocfs2_truncate_for_delete(osb, inode, di_bh); if (status < 0) { @@ -733,7 +736,7 @@ static int ocfs2_wipe_inode(struct inode *inode, mlog_errno(status); bail_unlock_dir: - ocfs2_meta_unlock(orphan_dir_inode, 1); + ocfs2_inode_unlock(orphan_dir_inode, 1); mutex_unlock(&orphan_dir_inode->i_mutex); brelse(orphan_dir_bh); bail: @@ -744,7 +747,7 @@ bail: } /* There is a series of simple checks that should be done before a - * vote is even considered. Encapsulate those in this function. */ + * trylock is even considered. Encapsulate those in this function. */ static int ocfs2_inode_is_valid_to_delete(struct inode *inode) { int ret = 0; @@ -758,14 +761,14 @@ static int ocfs2_inode_is_valid_to_delete(struct inode *inode) goto bail; } - /* If we're coming from process_vote we can't go into our own + /* If we're coming from downconvert_thread we can't go into our own * voting [hello, deadlock city!], so unforuntately we just * have to skip deleting this guy. That's OK though because * the node who's doing the actual deleting should handle it * anyway. */ - if (current == osb->vote_task) { + if (current == osb->dc_task) { mlog(0, "Skipping delete of %lu because we're currently " - "in process_vote\n", inode->i_ino); + "in downconvert\n", inode->i_ino); goto bail; } @@ -779,10 +782,9 @@ static int ocfs2_inode_is_valid_to_delete(struct inode *inode) goto bail_unlock; } - /* If we have voted "yes" on the wipe of this inode for - * another node, it will be marked here so we can safely skip - * it. Recovery will cleanup any inodes we might inadvertantly - * skip here. */ + /* If we have allowd wipe of this inode for another node, it + * will be marked here so we can safely skip it. Recovery will + * cleanup any inodes we might inadvertantly skip here. */ if (oi->ip_flags & OCFS2_INODE_SKIP_DELETE) { mlog(0, "Skipping delete of %lu because another node " "has done this for us.\n", inode->i_ino); @@ -929,13 +931,13 @@ void ocfs2_delete_inode(struct inode *inode) /* Lock down the inode. This gives us an up to date view of * it's metadata (for verification), and allows us to - * serialize delete_inode votes. + * serialize delete_inode on multiple nodes. * * Even though we might be doing a truncate, we don't take the * allocation lock here as it won't be needed - nobody will * have the file open. */ - status = ocfs2_meta_lock(inode, &di_bh, 1); + status = ocfs2_inode_lock(inode, &di_bh, 1); if (status < 0) { if (status != -ENOENT) mlog_errno(status); @@ -947,15 +949,15 @@ void ocfs2_delete_inode(struct inode *inode) * before we go ahead and wipe the inode. */ status = ocfs2_query_inode_wipe(inode, di_bh, &wipe); if (!wipe || status < 0) { - /* Error and inode busy vote both mean we won't be + /* Error and remote inode busy both mean we won't be * removing the inode, so they take almost the same * path. */ if (status < 0) mlog_errno(status); - /* Someone in the cluster has voted to not wipe this - * inode, or it was never completely orphaned. Write - * out the pages and exit now. */ + /* Someone in the cluster has disallowed a wipe of + * this inode, or it was never completely + * orphaned. Write out the pages and exit now. */ ocfs2_cleanup_delete_inode(inode, 1); goto bail_unlock_inode; } @@ -981,7 +983,7 @@ void ocfs2_delete_inode(struct inode *inode) OCFS2_I(inode)->ip_flags |= OCFS2_INODE_DELETED; bail_unlock_inode: - ocfs2_meta_unlock(inode, 1); + ocfs2_inode_unlock(inode, 1); brelse(di_bh); bail_unblock: status = sigprocmask(SIG_SETMASK, &oldset, NULL); @@ -1008,15 +1010,14 @@ void ocfs2_clear_inode(struct inode *inode) mlog_bug_on_msg(OCFS2_SB(inode->i_sb) == NULL, "Inode=%lu\n", inode->i_ino); - /* For remove delete_inode vote, we hold open lock before, - * now it is time to unlock PR and EX open locks. */ + /* To preven remote deletes we hold open lock before, now it + * is time to unlock PR and EX open locks. */ ocfs2_open_unlock(inode); /* Do these before all the other work so that we don't bounce - * the vote thread while waiting to destroy the locks. */ + * the downconvert thread while waiting to destroy the locks. */ ocfs2_mark_lockres_freeing(&oi->ip_rw_lockres); - ocfs2_mark_lockres_freeing(&oi->ip_meta_lockres); - ocfs2_mark_lockres_freeing(&oi->ip_data_lockres); + ocfs2_mark_lockres_freeing(&oi->ip_inode_lockres); ocfs2_mark_lockres_freeing(&oi->ip_open_lockres); /* We very well may get a clear_inode before all an inodes @@ -1039,8 +1040,7 @@ void ocfs2_clear_inode(struct inode *inode) mlog_errno(status); ocfs2_lock_res_free(&oi->ip_rw_lockres); - ocfs2_lock_res_free(&oi->ip_meta_lockres); - ocfs2_lock_res_free(&oi->ip_data_lockres); + ocfs2_lock_res_free(&oi->ip_inode_lockres); ocfs2_lock_res_free(&oi->ip_open_lockres); ocfs2_metadata_cache_purge(inode); @@ -1184,15 +1184,15 @@ int ocfs2_inode_revalidate(struct dentry *dentry) } spin_unlock(&OCFS2_I(inode)->ip_lock); - /* Let ocfs2_meta_lock do the work of updating our struct + /* Let ocfs2_inode_lock do the work of updating our struct * inode for us. */ - status = ocfs2_meta_lock(inode, NULL, 0); + status = ocfs2_inode_lock(inode, NULL, 0); if (status < 0) { if (status != -ENOENT) mlog_errno(status); goto bail; } - ocfs2_meta_unlock(inode, 0); + ocfs2_inode_unlock(inode, 0); bail: mlog_exit(status); diff --git a/fs/ocfs2/inode.h b/fs/ocfs2/inode.h index 70e881c5553..390a85596aa 100644 --- a/fs/ocfs2/inode.h +++ b/fs/ocfs2/inode.h @@ -34,8 +34,7 @@ struct ocfs2_inode_info u64 ip_blkno; struct ocfs2_lock_res ip_rw_lockres; - struct ocfs2_lock_res ip_meta_lockres; - struct ocfs2_lock_res ip_data_lockres; + struct ocfs2_lock_res ip_inode_lockres; struct ocfs2_lock_res ip_open_lockres; /* protects allocation changes on this inode. */ @@ -121,9 +120,10 @@ void ocfs2_delete_inode(struct inode *inode); void ocfs2_drop_inode(struct inode *inode); /* Flags for ocfs2_iget() */ -#define OCFS2_FI_FLAG_SYSFILE 0x4 -#define OCFS2_FI_FLAG_ORPHAN_RECOVERY 0x8 -struct inode *ocfs2_iget(struct ocfs2_super *osb, u64 feoff, int flags); +#define OCFS2_FI_FLAG_SYSFILE 0x1 +#define OCFS2_FI_FLAG_ORPHAN_RECOVERY 0x2 +struct inode *ocfs2_iget(struct ocfs2_super *osb, u64 feoff, unsigned flags, + int sysfile_type); int ocfs2_inode_init_private(struct inode *inode); int ocfs2_inode_revalidate(struct dentry *dentry); int ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe, diff --git a/fs/ocfs2/ioctl.c b/fs/ocfs2/ioctl.c index 87dcece7e1b..5177fba5162 100644 --- a/fs/ocfs2/ioctl.c +++ b/fs/ocfs2/ioctl.c @@ -20,6 +20,7 @@ #include "ocfs2_fs.h" #include "ioctl.h" +#include "resize.h" #include <linux/ext2_fs.h> @@ -27,14 +28,14 @@ static int ocfs2_get_inode_attr(struct inode *inode, unsigned *flags) { int status; - status = ocfs2_meta_lock(inode, NULL, 0); + status = ocfs2_inode_lock(inode, NULL, 0); if (status < 0) { mlog_errno(status); return status; } ocfs2_get_inode_flags(OCFS2_I(inode)); *flags = OCFS2_I(inode)->ip_attr; - ocfs2_meta_unlock(inode, 0); + ocfs2_inode_unlock(inode, 0); mlog_exit(status); return status; @@ -52,7 +53,7 @@ static int ocfs2_set_inode_attr(struct inode *inode, unsigned flags, mutex_lock(&inode->i_mutex); - status = ocfs2_meta_lock(inode, &bh, 1); + status = ocfs2_inode_lock(inode, &bh, 1); if (status < 0) { mlog_errno(status); goto bail; @@ -100,7 +101,7 @@ static int ocfs2_set_inode_attr(struct inode *inode, unsigned flags, ocfs2_commit_trans(osb, handle); bail_unlock: - ocfs2_meta_unlock(inode, 1); + ocfs2_inode_unlock(inode, 1); bail: mutex_unlock(&inode->i_mutex); @@ -115,8 +116,10 @@ int ocfs2_ioctl(struct inode * inode, struct file * filp, unsigned int cmd, unsigned long arg) { unsigned int flags; + int new_clusters; int status; struct ocfs2_space_resv sr; + struct ocfs2_new_group_input input; switch (cmd) { case OCFS2_IOC_GETFLAGS: @@ -140,6 +143,23 @@ int ocfs2_ioctl(struct inode * inode, struct file * filp, return -EFAULT; return ocfs2_change_file_space(filp, cmd, &sr); + case OCFS2_IOC_GROUP_EXTEND: + if (!capable(CAP_SYS_RESOURCE)) + return -EPERM; + + if (get_user(new_clusters, (int __user *)arg)) + return -EFAULT; + + return ocfs2_group_extend(inode, new_clusters); + case OCFS2_IOC_GROUP_ADD: + case OCFS2_IOC_GROUP_ADD64: + if (!capable(CAP_SYS_RESOURCE)) + return -EPERM; + + if (copy_from_user(&input, (int __user *) arg, sizeof(input))) + return -EFAULT; + + return ocfs2_group_add(inode, &input); default: return -ENOTTY; } @@ -162,6 +182,9 @@ long ocfs2_compat_ioctl(struct file *file, unsigned cmd, unsigned long arg) case OCFS2_IOC_RESVSP64: case OCFS2_IOC_UNRESVSP: case OCFS2_IOC_UNRESVSP64: + case OCFS2_IOC_GROUP_EXTEND: + case OCFS2_IOC_GROUP_ADD: + case OCFS2_IOC_GROUP_ADD64: break; default: return -ENOIOCTLCMD; diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index 8d81f6c1b87..f31c7e8c19c 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -44,7 +44,6 @@ #include "localalloc.h" #include "slot_map.h" #include "super.h" -#include "vote.h" #include "sysfile.h" #include "buffer_head_io.h" @@ -103,7 +102,7 @@ static int ocfs2_commit_cache(struct ocfs2_super *osb) mlog(0, "commit_thread: flushed transaction %lu (%u handles)\n", journal->j_trans_id, flushed); - ocfs2_kick_vote_thread(osb); + ocfs2_wake_downconvert_thread(osb); wake_up(&journal->j_checkpointed); finally: mlog_exit(status); @@ -314,14 +313,18 @@ int ocfs2_journal_dirty_data(handle_t *handle, return err; } -#define OCFS2_DEFAULT_COMMIT_INTERVAL (HZ * 5) +#define OCFS2_DEFAULT_COMMIT_INTERVAL (HZ * JBD_DEFAULT_MAX_COMMIT_AGE) void ocfs2_set_journal_params(struct ocfs2_super *osb) { journal_t *journal = osb->journal->j_journal; + unsigned long commit_interval = OCFS2_DEFAULT_COMMIT_INTERVAL; + + if (osb->osb_commit_interval) + commit_interval = osb->osb_commit_interval; spin_lock(&journal->j_state_lock); - journal->j_commit_interval = OCFS2_DEFAULT_COMMIT_INTERVAL; + journal->j_commit_interval = commit_interval; if (osb->s_mount_opt & OCFS2_MOUNT_BARRIER) journal->j_flags |= JFS_BARRIER; else @@ -337,7 +340,7 @@ int ocfs2_journal_init(struct ocfs2_journal *journal, int *dirty) struct ocfs2_dinode *di = NULL; struct buffer_head *bh = NULL; struct ocfs2_super *osb; - int meta_lock = 0; + int inode_lock = 0; mlog_entry_void(); @@ -367,14 +370,14 @@ int ocfs2_journal_init(struct ocfs2_journal *journal, int *dirty) /* Skip recovery waits here - journal inode metadata never * changes in a live cluster so it can be considered an * exception to the rule. */ - status = ocfs2_meta_lock_full(inode, &bh, 1, OCFS2_META_LOCK_RECOVERY); + status = ocfs2_inode_lock_full(inode, &bh, 1, OCFS2_META_LOCK_RECOVERY); if (status < 0) { if (status != -ERESTARTSYS) mlog(ML_ERROR, "Could not get lock on journal!\n"); goto done; } - meta_lock = 1; + inode_lock = 1; di = (struct ocfs2_dinode *)bh->b_data; if (inode->i_size < OCFS2_MIN_JOURNAL_SIZE) { @@ -414,8 +417,8 @@ int ocfs2_journal_init(struct ocfs2_journal *journal, int *dirty) status = 0; done: if (status < 0) { - if (meta_lock) - ocfs2_meta_unlock(inode, 1); + if (inode_lock) + ocfs2_inode_unlock(inode, 1); if (bh != NULL) brelse(bh); if (inode) { @@ -544,7 +547,7 @@ void ocfs2_journal_shutdown(struct ocfs2_super *osb) OCFS2_I(inode)->ip_open_count--; /* unlock our journal */ - ocfs2_meta_unlock(inode, 1); + ocfs2_inode_unlock(inode, 1); brelse(journal->j_bh); journal->j_bh = NULL; @@ -883,8 +886,8 @@ restart: ocfs2_super_unlock(osb, 1); /* We always run recovery on our own orphan dir - the dead - * node(s) may have voted "no" on an inode delete earlier. A - * revote is therefore required. */ + * node(s) may have disallowd a previos inode delete. Re-processing + * is therefore required. */ ocfs2_queue_recovery_completion(osb->journal, osb->slot_num, NULL, NULL); @@ -973,9 +976,9 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb, } SET_INODE_JOURNAL(inode); - status = ocfs2_meta_lock_full(inode, &bh, 1, OCFS2_META_LOCK_RECOVERY); + status = ocfs2_inode_lock_full(inode, &bh, 1, OCFS2_META_LOCK_RECOVERY); if (status < 0) { - mlog(0, "status returned from ocfs2_meta_lock=%d\n", status); + mlog(0, "status returned from ocfs2_inode_lock=%d\n", status); if (status != -ERESTARTSYS) mlog(ML_ERROR, "Could not lock journal!\n"); goto done; @@ -1047,7 +1050,7 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb, done: /* drop the lock on this nodes journal */ if (got_lock) - ocfs2_meta_unlock(inode, 1); + ocfs2_inode_unlock(inode, 1); if (inode) iput(inode); @@ -1162,14 +1165,14 @@ static int ocfs2_trylock_journal(struct ocfs2_super *osb, SET_INODE_JOURNAL(inode); flags = OCFS2_META_LOCK_RECOVERY | OCFS2_META_LOCK_NOQUEUE; - status = ocfs2_meta_lock_full(inode, NULL, 1, flags); + status = ocfs2_inode_lock_full(inode, NULL, 1, flags); if (status < 0) { if (status != -EAGAIN) mlog_errno(status); goto bail; } - ocfs2_meta_unlock(inode, 1); + ocfs2_inode_unlock(inode, 1); bail: if (inode) iput(inode); @@ -1241,7 +1244,7 @@ static int ocfs2_orphan_filldir(void *priv, const char *name, int name_len, /* Skip bad inodes so that recovery can continue */ iter = ocfs2_iget(p->osb, ino, - OCFS2_FI_FLAG_ORPHAN_RECOVERY); + OCFS2_FI_FLAG_ORPHAN_RECOVERY, 0); if (IS_ERR(iter)) return 0; @@ -1277,7 +1280,7 @@ static int ocfs2_queue_orphans(struct ocfs2_super *osb, } mutex_lock(&orphan_dir_inode->i_mutex); - status = ocfs2_meta_lock(orphan_dir_inode, NULL, 0); + status = ocfs2_inode_lock(orphan_dir_inode, NULL, 0); if (status < 0) { mlog_errno(status); goto out; @@ -1293,7 +1296,7 @@ static int ocfs2_queue_orphans(struct ocfs2_super *osb, *head = priv.head; out_cluster: - ocfs2_meta_unlock(orphan_dir_inode, 0); + ocfs2_inode_unlock(orphan_dir_inode, 0); out: mutex_unlock(&orphan_dir_inode->i_mutex); iput(orphan_dir_inode); @@ -1380,10 +1383,10 @@ static int ocfs2_recover_orphans(struct ocfs2_super *osb, iter = oi->ip_next_orphan; spin_lock(&oi->ip_lock); - /* Delete voting may have set these on the assumption - * that the other node would wipe them successfully. - * If they are still in the node's orphan dir, we need - * to reset that state. */ + /* The remote delete code may have set these on the + * assumption that the other node would wipe them + * successfully. If they are still in the node's + * orphan dir, we need to reset that state. */ oi->ip_flags &= ~(OCFS2_INODE_DELETED|OCFS2_INODE_SKIP_DELETE); /* Set the proper information to get us going into diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h index 4b32e096156..220f3e818e7 100644 --- a/fs/ocfs2/journal.h +++ b/fs/ocfs2/journal.h @@ -278,6 +278,12 @@ int ocfs2_journal_dirty_data(handle_t *handle, /* simple file updates like chmod, etc. */ #define OCFS2_INODE_UPDATE_CREDITS 1 +/* group extend. inode update and last group update. */ +#define OCFS2_GROUP_EXTEND_CREDITS (OCFS2_INODE_UPDATE_CREDITS + 1) + +/* group add. inode update and the new group update. */ +#define OCFS2_GROUP_ADD_CREDITS (OCFS2_INODE_UPDATE_CREDITS + 1) + /* get one bit out of a suballocator: dinode + group descriptor + * prev. group desc. if we relink. */ #define OCFS2_SUBALLOC_ALLOC (3) diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c index 58ea88b5af3..ab83fd56242 100644 --- a/fs/ocfs2/localalloc.c +++ b/fs/ocfs2/localalloc.c @@ -75,18 +75,12 @@ static int ocfs2_local_alloc_new_window(struct ocfs2_super *osb, static int ocfs2_local_alloc_slide_window(struct ocfs2_super *osb, struct inode *local_alloc_inode); -/* - * Determine how large our local alloc window should be, in bits. - * - * These values (and the behavior in ocfs2_alloc_should_use_local) have - * been chosen so that most allocations, including new block groups go - * through local alloc. - */ static inline int ocfs2_local_alloc_window_bits(struct ocfs2_super *osb) { - BUG_ON(osb->s_clustersize_bits < 12); + BUG_ON(osb->s_clustersize_bits > 20); - return 2048 >> (osb->s_clustersize_bits - 12); + /* Size local alloc windows by the megabyte */ + return osb->local_alloc_size << (20 - osb->s_clustersize_bits); } /* @@ -96,18 +90,23 @@ static inline int ocfs2_local_alloc_window_bits(struct ocfs2_super *osb) int ocfs2_alloc_should_use_local(struct ocfs2_super *osb, u64 bits) { int la_bits = ocfs2_local_alloc_window_bits(osb); + int ret = 0; if (osb->local_alloc_state != OCFS2_LA_ENABLED) - return 0; + goto bail; /* la_bits should be at least twice the size (in clusters) of * a new block group. We want to be sure block group * allocations go through the local alloc, so allow an * allocation to take up to half the bitmap. */ if (bits > (la_bits / 2)) - return 0; + goto bail; - return 1; + ret = 1; +bail: + mlog(0, "state=%d, bits=%llu, la_bits=%d, ret=%d\n", + osb->local_alloc_state, (unsigned long long)bits, la_bits, ret); + return ret; } int ocfs2_load_local_alloc(struct ocfs2_super *osb) @@ -121,6 +120,16 @@ int ocfs2_load_local_alloc(struct ocfs2_super *osb) mlog_entry_void(); + if (osb->local_alloc_size == 0) + goto bail; + + if (ocfs2_local_alloc_window_bits(osb) >= osb->bitmap_cpg) { + mlog(ML_NOTICE, "Requested local alloc window %d is larger " + "than max possible %u. Using defaults.\n", + ocfs2_local_alloc_window_bits(osb), (osb->bitmap_cpg - 1)); + osb->local_alloc_size = OCFS2_DEFAULT_LOCAL_ALLOC_SIZE; + } + /* read the alloc off disk */ inode = ocfs2_get_system_file_inode(osb, LOCAL_ALLOC_SYSTEM_INODE, osb->slot_num); @@ -181,6 +190,9 @@ bail: if (inode) iput(inode); + mlog(0, "Local alloc window bits = %d\n", + ocfs2_local_alloc_window_bits(osb)); + mlog_exit(status); return status; } @@ -231,7 +243,7 @@ void ocfs2_shutdown_local_alloc(struct ocfs2_super *osb) mutex_lock(&main_bm_inode->i_mutex); - status = ocfs2_meta_lock(main_bm_inode, &main_bm_bh, 1); + status = ocfs2_inode_lock(main_bm_inode, &main_bm_bh, 1); if (status < 0) { mlog_errno(status); goto out_mutex; @@ -286,7 +298,7 @@ out_unlock: if (main_bm_bh) brelse(main_bm_bh); - ocfs2_meta_unlock(main_bm_inode, 1); + ocfs2_inode_unlock(main_bm_inode, 1); out_mutex: mutex_unlock(&main_bm_inode->i_mutex); @@ -399,7 +411,7 @@ int ocfs2_complete_local_alloc_recovery(struct ocfs2_super *osb, mutex_lock(&main_bm_inode->i_mutex); - status = ocfs2_meta_lock(main_bm_inode, &main_bm_bh, 1); + status = ocfs2_inode_lock(main_bm_inode, &main_bm_bh, 1); if (status < 0) { mlog_errno(status); goto out_mutex; @@ -424,7 +436,7 @@ int ocfs2_complete_local_alloc_recovery(struct ocfs2_super *osb, ocfs2_commit_trans(osb, handle); out_unlock: - ocfs2_meta_unlock(main_bm_inode, 1); + ocfs2_inode_unlock(main_bm_inode, 1); out_mutex: mutex_unlock(&main_bm_inode->i_mutex); @@ -521,6 +533,9 @@ bail: iput(local_alloc_inode); } + mlog(0, "bits=%d, slot=%d, ret=%d\n", bits_wanted, osb->slot_num, + status); + mlog_exit(status); return status; } @@ -570,8 +585,7 @@ int ocfs2_claim_local_alloc_bits(struct ocfs2_super *osb, while(bits_wanted--) ocfs2_set_bit(start++, bitmap); - alloc->id1.bitmap1.i_used = cpu_to_le32(*num_bits + - le32_to_cpu(alloc->id1.bitmap1.i_used)); + le32_add_cpu(&alloc->id1.bitmap1.i_used, *num_bits); status = ocfs2_journal_dirty(handle, osb->local_alloc_bh); if (status < 0) { diff --git a/fs/ocfs2/locks.c b/fs/ocfs2/locks.c new file mode 100644 index 00000000000..203f8714387 --- /dev/null +++ b/fs/ocfs2/locks.c @@ -0,0 +1,125 @@ +/* -*- mode: c; c-basic-offset: 8; -*- + * vim: noexpandtab sw=8 ts=8 sts=0: + * + * locks.c + * + * Userspace file locking support + * + * Copyright (C) 2007 Oracle. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public + * License as published by the Free Software Foundation; either + * version 2 of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public + * License along with this program; if not, write to the + * Free Software Foundation, Inc., 59 Temple Place - Suite 330, + * Boston, MA 021110-1307, USA. + */ + +#include <linux/fs.h> + +#define MLOG_MASK_PREFIX ML_INODE +#include <cluster/masklog.h> + +#include "ocfs2.h" + +#include "dlmglue.h" +#include "file.h" +#include "locks.h" + +static int ocfs2_do_flock(struct file *file, struct inode *inode, + int cmd, struct file_lock *fl) +{ + int ret = 0, level = 0, trylock = 0; + struct ocfs2_file_private *fp = file->private_data; + struct ocfs2_lock_res *lockres = &fp->fp_flock; + + if (fl->fl_type == F_WRLCK) + level = 1; + if (!IS_SETLKW(cmd)) + trylock = 1; + + mutex_lock(&fp->fp_mutex); + + if (lockres->l_flags & OCFS2_LOCK_ATTACHED && + lockres->l_level > LKM_NLMODE) { + int old_level = 0; + + if (lockres->l_level == LKM_EXMODE) + old_level = 1; + + if (level == old_level) + goto out; + + /* + * Converting an existing lock is not guaranteed to be + * atomic, so we can get away with simply unlocking + * here and allowing the lock code to try at the new + * level. + */ + + flock_lock_file_wait(file, + &(struct file_lock){.fl_type = F_UNLCK}); + + ocfs2_file_unlock(file); + } + + ret = ocfs2_file_lock(file, level, trylock); + if (ret) { + if (ret == -EAGAIN && trylock) + ret = -EWOULDBLOCK; + else + mlog_errno(ret); + goto out; + } + + ret = flock_lock_file_wait(file, fl); + +out: + mutex_unlock(&fp->fp_mutex); + + return ret; +} + +static int ocfs2_do_funlock(struct file *file, int cmd, struct file_lock *fl) +{ + int ret; + struct ocfs2_file_private *fp = file->private_data; + + mutex_lock(&fp->fp_mutex); + ocfs2_file_unlock(file); + ret = flock_lock_file_wait(file, fl); + mutex_unlock(&fp->fp_mutex); + + return ret; +} + +/* + * Overall flow of ocfs2_flock() was influenced by gfs2_flock(). + */ +int ocfs2_flock(struct file *file, int cmd, struct file_lock *fl) +{ + struct inode *inode = file->f_mapping->host; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + + if (!(fl->fl_flags & FL_FLOCK)) + return -ENOLCK; + if (__mandatory_lock(inode)) + return -ENOLCK; + + if ((osb->s_mount_opt & OCFS2_MOUNT_LOCALFLOCKS) || + ocfs2_mount_local(osb)) + return flock_lock_file_wait(file, fl); + + if (fl->fl_type == F_UNLCK) + return ocfs2_do_funlock(file, cmd, fl); + else + return ocfs2_do_flock(file, inode, cmd, fl); +} diff --git a/fs/ocfs2/cluster/endian.h b/fs/ocfs2/locks.h index 2df9082f4e3..9743ef2324e 100644 --- a/fs/ocfs2/cluster/endian.h +++ b/fs/ocfs2/locks.h @@ -1,7 +1,11 @@ /* -*- mode: c; c-basic-offset: 8; -*- * vim: noexpandtab sw=8 ts=8 sts=0: * - * Copyright (C) 2005 Oracle. All rights reserved. + * locks.h + * + * Function prototypes for Userspace file locking support + * + * Copyright (C) 2002, 2004 Oracle. All rights reserved. * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public @@ -19,12 +23,9 @@ * Boston, MA 021110-1307, USA. */ -#ifndef OCFS2_CLUSTER_ENDIAN_H -#define OCFS2_CLUSTER_ENDIAN_H +#ifndef OCFS2_LOCKS_H +#define OCFS2_LOCKS_H -static inline void be32_add_cpu(__be32 *var, u32 val) -{ - *var = cpu_to_be32(be32_to_cpu(*var) + val); -} +int ocfs2_flock(struct file *file, int cmd, struct file_lock *fl); -#endif /* OCFS2_CLUSTER_ENDIAN_H */ +#endif /* OCFS2_LOCKS_H */ diff --git a/fs/ocfs2/mmap.c b/fs/ocfs2/mmap.c index 98756156d29..3dc18d67557 100644 --- a/fs/ocfs2/mmap.c +++ b/fs/ocfs2/mmap.c @@ -168,7 +168,7 @@ static int ocfs2_page_mkwrite(struct vm_area_struct *vma, struct page *page) * node. Taking the data lock will also ensure that we don't * attempt page truncation as part of a downconvert. */ - ret = ocfs2_meta_lock(inode, &di_bh, 1); + ret = ocfs2_inode_lock(inode, &di_bh, 1); if (ret < 0) { mlog_errno(ret); goto out; @@ -181,21 +181,12 @@ static int ocfs2_page_mkwrite(struct vm_area_struct *vma, struct page *page) */ down_write(&OCFS2_I(inode)->ip_alloc_sem); - ret = ocfs2_data_lock(inode, 1); - if (ret < 0) { - mlog_errno(ret); - goto out_meta_unlock; - } - ret = __ocfs2_page_mkwrite(inode, di_bh, page); - ocfs2_data_unlock(inode, 1); - -out_meta_unlock: up_write(&OCFS2_I(inode)->ip_alloc_sem); brelse(di_bh); - ocfs2_meta_unlock(inode, 1); + ocfs2_inode_unlock(inode, 1); out: ret2 = ocfs2_vm_op_unblock_sigs(&oldset); @@ -214,13 +205,13 @@ int ocfs2_mmap(struct file *file, struct vm_area_struct *vma) { int ret = 0, lock_level = 0; - ret = ocfs2_meta_lock_atime(file->f_dentry->d_inode, + ret = ocfs2_inode_lock_atime(file->f_dentry->d_inode, file->f_vfsmnt, &lock_level); if (ret < 0) { mlog_errno(ret); goto out; } - ocfs2_meta_unlock(file->f_dentry->d_inode, lock_level); + ocfs2_inode_unlock(file->f_dentry->d_inode, lock_level); out: vma->vm_ops = &ocfs2_file_vm_ops; vma->vm_flags |= VM_CAN_NONLINEAR; diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index 989ac271858..ae9ad958751 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -60,7 +60,6 @@ #include "symlink.h" #include "sysfile.h" #include "uptodate.h" -#include "vote.h" #include "buffer_head_io.h" @@ -116,7 +115,7 @@ static struct dentry *ocfs2_lookup(struct inode *dir, struct dentry *dentry, mlog(0, "find name %.*s in directory %llu\n", dentry->d_name.len, dentry->d_name.name, (unsigned long long)OCFS2_I(dir)->ip_blkno); - status = ocfs2_meta_lock(dir, NULL, 0); + status = ocfs2_inode_lock(dir, NULL, 0); if (status < 0) { if (status != -ENOENT) mlog_errno(status); @@ -129,7 +128,7 @@ static struct dentry *ocfs2_lookup(struct inode *dir, struct dentry *dentry, if (status < 0) goto bail_add; - inode = ocfs2_iget(OCFS2_SB(dir->i_sb), blkno, 0); + inode = ocfs2_iget(OCFS2_SB(dir->i_sb), blkno, 0, 0); if (IS_ERR(inode)) { ret = ERR_PTR(-EACCES); goto bail_unlock; @@ -176,8 +175,8 @@ bail_unlock: /* Don't drop the cluster lock until *after* the d_add -- * unlink on another node will message us to remove that * dentry under this lock so otherwise we can race this with - * the vote thread and have a stale dentry. */ - ocfs2_meta_unlock(dir, 0); + * the downconvert thread and have a stale dentry. */ + ocfs2_inode_unlock(dir, 0); bail: @@ -209,7 +208,7 @@ static int ocfs2_mknod(struct inode *dir, /* get our super block */ osb = OCFS2_SB(dir->i_sb); - status = ocfs2_meta_lock(dir, &parent_fe_bh, 1); + status = ocfs2_inode_lock(dir, &parent_fe_bh, 1); if (status < 0) { if (status != -ENOENT) mlog_errno(status); @@ -323,7 +322,7 @@ leave: if (handle) ocfs2_commit_trans(osb, handle); - ocfs2_meta_unlock(dir, 1); + ocfs2_inode_unlock(dir, 1); if (status == -ENOSPC) mlog(0, "Disk is full\n"); @@ -553,7 +552,7 @@ static int ocfs2_link(struct dentry *old_dentry, if (S_ISDIR(inode->i_mode)) return -EPERM; - err = ocfs2_meta_lock(dir, &parent_fe_bh, 1); + err = ocfs2_inode_lock(dir, &parent_fe_bh, 1); if (err < 0) { if (err != -ENOENT) mlog_errno(err); @@ -578,7 +577,7 @@ static int ocfs2_link(struct dentry *old_dentry, goto out; } - err = ocfs2_meta_lock(inode, &fe_bh, 1); + err = ocfs2_inode_lock(inode, &fe_bh, 1); if (err < 0) { if (err != -ENOENT) mlog_errno(err); @@ -643,10 +642,10 @@ static int ocfs2_link(struct dentry *old_dentry, out_commit: ocfs2_commit_trans(osb, handle); out_unlock_inode: - ocfs2_meta_unlock(inode, 1); + ocfs2_inode_unlock(inode, 1); out: - ocfs2_meta_unlock(dir, 1); + ocfs2_inode_unlock(dir, 1); if (de_bh) brelse(de_bh); @@ -720,7 +719,7 @@ static int ocfs2_unlink(struct inode *dir, return -EPERM; } - status = ocfs2_meta_lock(dir, &parent_node_bh, 1); + status = ocfs2_inode_lock(dir, &parent_node_bh, 1); if (status < 0) { if (status != -ENOENT) mlog_errno(status); @@ -745,7 +744,7 @@ static int ocfs2_unlink(struct inode *dir, goto leave; } - status = ocfs2_meta_lock(inode, &fe_bh, 1); + status = ocfs2_inode_lock(inode, &fe_bh, 1); if (status < 0) { if (status != -ENOENT) mlog_errno(status); @@ -765,7 +764,7 @@ static int ocfs2_unlink(struct inode *dir, status = ocfs2_remote_dentry_delete(dentry); if (status < 0) { - /* This vote should succeed under all normal + /* This remote delete should succeed under all normal * circumstances. */ mlog_errno(status); goto leave; @@ -841,13 +840,13 @@ leave: ocfs2_commit_trans(osb, handle); if (child_locked) - ocfs2_meta_unlock(inode, 1); + ocfs2_inode_unlock(inode, 1); - ocfs2_meta_unlock(dir, 1); + ocfs2_inode_unlock(dir, 1); if (orphan_dir) { /* This was locked for us in ocfs2_prepare_orphan_dir() */ - ocfs2_meta_unlock(orphan_dir, 1); + ocfs2_inode_unlock(orphan_dir, 1); mutex_unlock(&orphan_dir->i_mutex); iput(orphan_dir); } @@ -908,7 +907,7 @@ static int ocfs2_double_lock(struct ocfs2_super *osb, inode1 = tmpinode; } /* lock id2 */ - status = ocfs2_meta_lock(inode2, bh2, 1); + status = ocfs2_inode_lock(inode2, bh2, 1); if (status < 0) { if (status != -ENOENT) mlog_errno(status); @@ -917,14 +916,14 @@ static int ocfs2_double_lock(struct ocfs2_super *osb, } /* lock id1 */ - status = ocfs2_meta_lock(inode1, bh1, 1); + status = ocfs2_inode_lock(inode1, bh1, 1); if (status < 0) { /* * An error return must mean that no cluster locks * were held on function exit. */ if (oi1->ip_blkno != oi2->ip_blkno) - ocfs2_meta_unlock(inode2, 1); + ocfs2_inode_unlock(inode2, 1); if (status != -ENOENT) mlog_errno(status); @@ -937,10 +936,10 @@ bail: static void ocfs2_double_unlock(struct inode *inode1, struct inode *inode2) { - ocfs2_meta_unlock(inode1, 1); + ocfs2_inode_unlock(inode1, 1); if (inode1 != inode2) - ocfs2_meta_unlock(inode2, 1); + ocfs2_inode_unlock(inode2, 1); } static int ocfs2_rename(struct inode *old_dir, @@ -1031,10 +1030,11 @@ static int ocfs2_rename(struct inode *old_dir, /* * Aside from allowing a meta data update, the locking here - * also ensures that the vote thread on other nodes won't have - * to concurrently downconvert the inode and the dentry locks. + * also ensures that the downconvert thread on other nodes + * won't have to concurrently downconvert the inode and the + * dentry locks. */ - status = ocfs2_meta_lock(old_inode, &old_inode_bh, 1); + status = ocfs2_inode_lock(old_inode, &old_inode_bh, 1); if (status < 0) { if (status != -ENOENT) mlog_errno(status); @@ -1143,7 +1143,7 @@ static int ocfs2_rename(struct inode *old_dir, goto bail; } - status = ocfs2_meta_lock(new_inode, &newfe_bh, 1); + status = ocfs2_inode_lock(new_inode, &newfe_bh, 1); if (status < 0) { if (status != -ENOENT) mlog_errno(status); @@ -1355,14 +1355,14 @@ bail: ocfs2_double_unlock(old_dir, new_dir); if (old_child_locked) - ocfs2_meta_unlock(old_inode, 1); + ocfs2_inode_unlock(old_inode, 1); if (new_child_locked) - ocfs2_meta_unlock(new_inode, 1); + ocfs2_inode_unlock(new_inode, 1); if (orphan_dir) { /* This was locked for us in ocfs2_prepare_orphan_dir() */ - ocfs2_meta_unlock(orphan_dir, 1); + ocfs2_inode_unlock(orphan_dir, 1); mutex_unlock(&orphan_dir->i_mutex); iput(orphan_dir); } @@ -1530,7 +1530,7 @@ static int ocfs2_symlink(struct inode *dir, credits = ocfs2_calc_symlink_credits(sb); /* lock the parent directory */ - status = ocfs2_meta_lock(dir, &parent_fe_bh, 1); + status = ocfs2_inode_lock(dir, &parent_fe_bh, 1); if (status < 0) { if (status != -ENOENT) mlog_errno(status); @@ -1657,7 +1657,7 @@ bail: if (handle) ocfs2_commit_trans(osb, handle); - ocfs2_meta_unlock(dir, 1); + ocfs2_inode_unlock(dir, 1); if (new_fe_bh) brelse(new_fe_bh); @@ -1735,7 +1735,7 @@ static int ocfs2_prepare_orphan_dir(struct ocfs2_super *osb, mutex_lock(&orphan_dir_inode->i_mutex); - status = ocfs2_meta_lock(orphan_dir_inode, &orphan_dir_bh, 1); + status = ocfs2_inode_lock(orphan_dir_inode, &orphan_dir_bh, 1); if (status < 0) { mlog_errno(status); goto leave; @@ -1745,7 +1745,7 @@ static int ocfs2_prepare_orphan_dir(struct ocfs2_super *osb, orphan_dir_bh, name, OCFS2_ORPHAN_NAMELEN, de_bh); if (status < 0) { - ocfs2_meta_unlock(orphan_dir_inode, 1); + ocfs2_inode_unlock(orphan_dir_inode, 1); mlog_errno(status); goto leave; diff --git a/fs/ocfs2/ocfs1_fs_compat.h b/fs/ocfs2/ocfs1_fs_compat.h index 0b499bccec5..dfb313bda5d 100644 --- a/fs/ocfs2/ocfs1_fs_compat.h +++ b/fs/ocfs2/ocfs1_fs_compat.h @@ -77,7 +77,7 @@ struct ocfs1_disk_lock { /*00*/ __u32 curr_master; __u8 file_lock; - __u8 compat_pad[3]; /* Not in orignal definition. Used to + __u8 compat_pad[3]; /* Not in original definition. Used to make the already existing alignment explicit */ __u64 last_write_time; diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index 60a23e1906b..6546cef212e 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -43,7 +43,6 @@ #include "dlm/dlmapi.h" #include "ocfs2_fs.h" -#include "endian.h" #include "ocfs2_lockid.h" /* Most user visible OCFS2 inodes will have very few pieces of @@ -101,6 +100,7 @@ enum ocfs2_unlock_action { * about to be * dropped. */ #define OCFS2_LOCK_QUEUED (0x00000100) /* queued for downconvert */ +#define OCFS2_LOCK_NOCACHE (0x00000200) /* don't use a holder count */ struct ocfs2_lock_res_ops; @@ -170,6 +170,7 @@ enum ocfs2_mount_options OCFS2_MOUNT_NOINTR = 1 << 2, /* Don't catch signals */ OCFS2_MOUNT_ERRORS_PANIC = 1 << 3, /* Panic on errors */ OCFS2_MOUNT_DATA_WRITEBACK = 1 << 4, /* No data ordering */ + OCFS2_MOUNT_LOCALFLOCKS = 1 << 5, /* No cluster aware user file locks */ }; #define OCFS2_OSB_SOFT_RO 0x0001 @@ -189,9 +190,7 @@ struct ocfs2_super struct ocfs2_slot_info *slot_info; spinlock_t node_map_lock; - struct ocfs2_node_map mounted_map; struct ocfs2_node_map recovery_map; - struct ocfs2_node_map umount_map; u64 root_blkno; u64 system_dir_blkno; @@ -231,7 +230,9 @@ struct ocfs2_super wait_queue_head_t checkpoint_event; atomic_t needs_checkpoint; struct ocfs2_journal *journal; + unsigned long osb_commit_interval; + int local_alloc_size; enum ocfs2_local_alloc_state local_alloc_state; struct buffer_head *local_alloc_bh; u64 la_last_gd; @@ -249,33 +250,27 @@ struct ocfs2_super struct ocfs2_lock_res osb_rename_lockres; struct dlm_eviction_cb osb_eviction_cb; struct ocfs2_dlm_debug *osb_dlm_debug; + struct dlm_protocol_version osb_locking_proto; struct dentry *osb_debug_root; wait_queue_head_t recovery_event; - spinlock_t vote_task_lock; - struct task_struct *vote_task; - wait_queue_head_t vote_event; - unsigned long vote_wake_sequence; - unsigned long vote_work_sequence; + spinlock_t dc_task_lock; + struct task_struct *dc_task; + wait_queue_head_t dc_event; + unsigned long dc_wake_sequence; + unsigned long dc_work_sequence; + /* + * Any thread can add locks to the list, but the downconvert + * thread is the only one allowed to remove locks. Any change + * to this rule requires updating + * ocfs2_downconvert_thread_do_work(). + */ struct list_head blocked_lock_list; unsigned long blocked_lock_count; - struct list_head vote_list; - int vote_count; - - u32 net_key; - spinlock_t net_response_lock; - unsigned int net_response_ids; - struct list_head net_response_list; - - struct o2hb_callback_func osb_hb_up; - struct o2hb_callback_func osb_hb_down; - - struct list_head osb_net_handlers; - wait_queue_head_t osb_mount_event; /* Truncate log info */ diff --git a/fs/ocfs2/ocfs2_fs.h b/fs/ocfs2/ocfs2_fs.h index 6ef876759a7..3633edd3982 100644 --- a/fs/ocfs2/ocfs2_fs.h +++ b/fs/ocfs2/ocfs2_fs.h @@ -231,6 +231,20 @@ struct ocfs2_space_resv { #define OCFS2_IOC_RESVSP64 _IOW ('X', 42, struct ocfs2_space_resv) #define OCFS2_IOC_UNRESVSP64 _IOW ('X', 43, struct ocfs2_space_resv) +/* Used to pass group descriptor data when online resize is done */ +struct ocfs2_new_group_input { + __u64 group; /* Group descriptor's blkno. */ + __u32 clusters; /* Total number of clusters in this group */ + __u32 frees; /* Total free clusters in this group */ + __u16 chain; /* Chain for this group */ + __u16 reserved1; + __u32 reserved2; +}; + +#define OCFS2_IOC_GROUP_EXTEND _IOW('o', 1, int) +#define OCFS2_IOC_GROUP_ADD _IOW('o', 2,struct ocfs2_new_group_input) +#define OCFS2_IOC_GROUP_ADD64 _IOW('o', 3,struct ocfs2_new_group_input) + /* * Journal Flags (ocfs2_dinode.id1.journal1.i_flags) */ @@ -256,6 +270,14 @@ struct ocfs2_space_resv { /* Journal limits (in bytes) */ #define OCFS2_MIN_JOURNAL_SIZE (4 * 1024 * 1024) +/* + * Default local alloc size (in megabytes) + * + * The value chosen should be such that most allocations, including new + * block groups, use local alloc. + */ +#define OCFS2_DEFAULT_LOCAL_ALLOC_SIZE 8 + struct ocfs2_system_inode_info { char *si_name; int si_iflags; diff --git a/fs/ocfs2/ocfs2_lockid.h b/fs/ocfs2/ocfs2_lockid.h index 4ca02b1c38a..86f3e3799c2 100644 --- a/fs/ocfs2/ocfs2_lockid.h +++ b/fs/ocfs2/ocfs2_lockid.h @@ -45,6 +45,7 @@ enum ocfs2_lock_type { OCFS2_LOCK_TYPE_RW, OCFS2_LOCK_TYPE_DENTRY, OCFS2_LOCK_TYPE_OPEN, + OCFS2_LOCK_TYPE_FLOCK, OCFS2_NUM_LOCK_TYPES }; @@ -73,6 +74,9 @@ static inline char ocfs2_lock_type_char(enum ocfs2_lock_type type) case OCFS2_LOCK_TYPE_OPEN: c = 'O'; break; + case OCFS2_LOCK_TYPE_FLOCK: + c = 'F'; + break; default: c = '\0'; } @@ -90,6 +94,7 @@ static char *ocfs2_lock_type_strings[] = { [OCFS2_LOCK_TYPE_RW] = "Write/Read", [OCFS2_LOCK_TYPE_DENTRY] = "Dentry", [OCFS2_LOCK_TYPE_OPEN] = "Open", + [OCFS2_LOCK_TYPE_FLOCK] = "Flock", }; static inline const char *ocfs2_lock_type_string(enum ocfs2_lock_type type) diff --git a/fs/ocfs2/ocfs2_lockingver.h b/fs/ocfs2/ocfs2_lockingver.h new file mode 100644 index 00000000000..82d5eeac0ff --- /dev/null +++ b/fs/ocfs2/ocfs2_lockingver.h @@ -0,0 +1,30 @@ +/* -*- mode: c; c-basic-offset: 8; -*- + * vim: noexpandtab sw=8 ts=8 sts=0: + * + * ocfs2_lockingver.h + * + * Defines OCFS2 Locking version values. + * + * Copyright (C) 2008 Oracle. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public + * License, version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + */ + +#ifndef OCFS2_LOCKINGVER_H +#define OCFS2_LOCKINGVER_H + +/* + * The protocol version for ocfs2 cluster locking. See dlmglue.c for + * more details. + */ +#define OCFS2_LOCKING_PROTOCOL_MAJOR 1 +#define OCFS2_LOCKING_PROTOCOL_MINOR 0 + +#endif /* OCFS2_LOCKINGVER_H */ diff --git a/fs/ocfs2/resize.c b/fs/ocfs2/resize.c new file mode 100644 index 00000000000..8166968e901 --- /dev/null +++ b/fs/ocfs2/resize.c @@ -0,0 +1,634 @@ +/* -*- mode: c; c-basic-offset: 8; -*- + * vim: noexpandtab sw=8 ts=8 sts=0: + * + * resize.c + * + * volume resize. + * Inspired by ext3/resize.c. + * + * Copyright (C) 2007 Oracle. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public + * License as published by the Free Software Foundation; either + * version 2 of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public + * License along with this program; if not, write to the + * Free Software Foundation, Inc., 59 Temple Place - Suite 330, + * Boston, MA 021110-1307, USA. + */ + +#include <linux/fs.h> +#include <linux/types.h> + +#define MLOG_MASK_PREFIX ML_DISK_ALLOC +#include <cluster/masklog.h> + +#include "ocfs2.h" + +#include "alloc.h" +#include "dlmglue.h" +#include "inode.h" +#include "journal.h" +#include "super.h" +#include "sysfile.h" +#include "uptodate.h" + +#include "buffer_head_io.h" +#include "suballoc.h" +#include "resize.h" + +/* + * Check whether there are new backup superblocks exist + * in the last group. If there are some, mark them or clear + * them in the bitmap. + * + * Return how many backups we find in the last group. + */ +static u16 ocfs2_calc_new_backup_super(struct inode *inode, + struct ocfs2_group_desc *gd, + int new_clusters, + u32 first_new_cluster, + u16 cl_cpg, + int set) +{ + int i; + u16 backups = 0; + u32 cluster; + u64 blkno, gd_blkno, lgd_blkno = le64_to_cpu(gd->bg_blkno); + + for (i = 0; i < OCFS2_MAX_BACKUP_SUPERBLOCKS; i++) { + blkno = ocfs2_backup_super_blkno(inode->i_sb, i); + cluster = ocfs2_blocks_to_clusters(inode->i_sb, blkno); + + gd_blkno = ocfs2_which_cluster_group(inode, cluster); + if (gd_blkno < lgd_blkno) + continue; + else if (gd_blkno > lgd_blkno) + break; + + if (set) + ocfs2_set_bit(cluster % cl_cpg, + (unsigned long *)gd->bg_bitmap); + else + ocfs2_clear_bit(cluster % cl_cpg, + (unsigned long *)gd->bg_bitmap); + backups++; + } + + mlog_exit_void(); + return backups; +} + +static int ocfs2_update_last_group_and_inode(handle_t *handle, + struct inode *bm_inode, + struct buffer_head *bm_bh, + struct buffer_head *group_bh, + u32 first_new_cluster, + int new_clusters) +{ + int ret = 0; + struct ocfs2_super *osb = OCFS2_SB(bm_inode->i_sb); + struct ocfs2_dinode *fe = (struct ocfs2_dinode *) bm_bh->b_data; + struct ocfs2_chain_list *cl = &fe->id2.i_chain; + struct ocfs2_chain_rec *cr; + struct ocfs2_group_desc *group; + u16 chain, num_bits, backups = 0; + u16 cl_bpc = le16_to_cpu(cl->cl_bpc); + u16 cl_cpg = le16_to_cpu(cl->cl_cpg); + + mlog_entry("(new_clusters=%d, first_new_cluster = %u)\n", + new_clusters, first_new_cluster); + + ret = ocfs2_journal_access(handle, bm_inode, group_bh, + OCFS2_JOURNAL_ACCESS_WRITE); + if (ret < 0) { + mlog_errno(ret); + goto out; + } + + group = (struct ocfs2_group_desc *)group_bh->b_data; + + /* update the group first. */ + num_bits = new_clusters * cl_bpc; + le16_add_cpu(&group->bg_bits, num_bits); + le16_add_cpu(&group->bg_free_bits_count, num_bits); + + /* + * check whether there are some new backup superblocks exist in + * this group and update the group bitmap accordingly. + */ + if (OCFS2_HAS_COMPAT_FEATURE(osb->sb, + OCFS2_FEATURE_COMPAT_BACKUP_SB)) { + backups = ocfs2_calc_new_backup_super(bm_inode, + group, + new_clusters, + first_new_cluster, + cl_cpg, 1); + le16_add_cpu(&group->bg_free_bits_count, -1 * backups); + } + + ret = ocfs2_journal_dirty(handle, group_bh); + if (ret < 0) { + mlog_errno(ret); + goto out_rollback; + } + + /* update the inode accordingly. */ + ret = ocfs2_journal_access(handle, bm_inode, bm_bh, + OCFS2_JOURNAL_ACCESS_WRITE); + if (ret < 0) { + mlog_errno(ret); + goto out_rollback; + } + + chain = le16_to_cpu(group->bg_chain); + cr = (&cl->cl_recs[chain]); + le32_add_cpu(&cr->c_total, num_bits); + le32_add_cpu(&cr->c_free, num_bits); + le32_add_cpu(&fe->id1.bitmap1.i_total, num_bits); + le32_add_cpu(&fe->i_clusters, new_clusters); + + if (backups) { + le32_add_cpu(&cr->c_free, -1 * backups); + le32_add_cpu(&fe->id1.bitmap1.i_used, backups); + } + + spin_lock(&OCFS2_I(bm_inode)->ip_lock); + OCFS2_I(bm_inode)->ip_clusters = le32_to_cpu(fe->i_clusters); + le64_add_cpu(&fe->i_size, new_clusters << osb->s_clustersize_bits); + spin_unlock(&OCFS2_I(bm_inode)->ip_lock); + i_size_write(bm_inode, le64_to_cpu(fe->i_size)); + + ocfs2_journal_dirty(handle, bm_bh); + +out_rollback: + if (ret < 0) { + ocfs2_calc_new_backup_super(bm_inode, + group, + new_clusters, + first_new_cluster, + cl_cpg, 0); + le16_add_cpu(&group->bg_free_bits_count, backups); + le16_add_cpu(&group->bg_bits, -1 * num_bits); + le16_add_cpu(&group->bg_free_bits_count, -1 * num_bits); + } +out: + mlog_exit(ret); + return ret; +} + +static int update_backups(struct inode * inode, u32 clusters, char *data) +{ + int i, ret = 0; + u32 cluster; + u64 blkno; + struct buffer_head *backup = NULL; + struct ocfs2_dinode *backup_di = NULL; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + + /* calculate the real backups we need to update. */ + for (i = 0; i < OCFS2_MAX_BACKUP_SUPERBLOCKS; i++) { + blkno = ocfs2_backup_super_blkno(inode->i_sb, i); + cluster = ocfs2_blocks_to_clusters(inode->i_sb, blkno); + if (cluster > clusters) + break; + + ret = ocfs2_read_block(osb, blkno, &backup, 0, NULL); + if (ret < 0) { + mlog_errno(ret); + break; + } + + memcpy(backup->b_data, data, inode->i_sb->s_blocksize); + + backup_di = (struct ocfs2_dinode *)backup->b_data; + backup_di->i_blkno = cpu_to_le64(blkno); + + ret = ocfs2_write_super_or_backup(osb, backup); + brelse(backup); + backup = NULL; + if (ret < 0) { + mlog_errno(ret); + break; + } + } + + return ret; +} + +static void ocfs2_update_super_and_backups(struct inode *inode, + int new_clusters) +{ + int ret; + u32 clusters = 0; + struct buffer_head *super_bh = NULL; + struct ocfs2_dinode *super_di = NULL; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + + /* + * update the superblock last. + * It doesn't matter if the write failed. + */ + ret = ocfs2_read_block(osb, OCFS2_SUPER_BLOCK_BLKNO, + &super_bh, 0, NULL); + if (ret < 0) { + mlog_errno(ret); + goto out; + } + + super_di = (struct ocfs2_dinode *)super_bh->b_data; + le32_add_cpu(&super_di->i_clusters, new_clusters); + clusters = le32_to_cpu(super_di->i_clusters); + + ret = ocfs2_write_super_or_backup(osb, super_bh); + if (ret < 0) { + mlog_errno(ret); + goto out; + } + + if (OCFS2_HAS_COMPAT_FEATURE(osb->sb, OCFS2_FEATURE_COMPAT_BACKUP_SB)) + ret = update_backups(inode, clusters, super_bh->b_data); + +out: + brelse(super_bh); + if (ret) + printk(KERN_WARNING "ocfs2: Failed to update super blocks on %s" + " during fs resize. This condition is not fatal," + " but fsck.ocfs2 should be run to fix it\n", + osb->dev_str); + return; +} + +/* + * Extend the filesystem to the new number of clusters specified. This entry + * point is only used to extend the current filesystem to the end of the last + * existing group. + */ +int ocfs2_group_extend(struct inode * inode, int new_clusters) +{ + int ret; + handle_t *handle; + struct buffer_head *main_bm_bh = NULL; + struct buffer_head *group_bh = NULL; + struct inode *main_bm_inode = NULL; + struct ocfs2_dinode *fe = NULL; + struct ocfs2_group_desc *group = NULL; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + u16 cl_bpc; + u32 first_new_cluster; + u64 lgd_blkno; + + mlog_entry_void(); + + if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb)) + return -EROFS; + + if (new_clusters < 0) + return -EINVAL; + else if (new_clusters == 0) + return 0; + + main_bm_inode = ocfs2_get_system_file_inode(osb, + GLOBAL_BITMAP_SYSTEM_INODE, + OCFS2_INVALID_SLOT); + if (!main_bm_inode) { + ret = -EINVAL; + mlog_errno(ret); + goto out; + } + + mutex_lock(&main_bm_inode->i_mutex); + + ret = ocfs2_inode_lock(main_bm_inode, &main_bm_bh, 1); + if (ret < 0) { + mlog_errno(ret); + goto out_mutex; + } + + fe = (struct ocfs2_dinode *)main_bm_bh->b_data; + + if (le16_to_cpu(fe->id2.i_chain.cl_cpg) != + ocfs2_group_bitmap_size(osb->sb) * 8) { + mlog(ML_ERROR, "The disk is too old and small. " + "Force to do offline resize."); + ret = -EINVAL; + goto out_unlock; + } + + if (!OCFS2_IS_VALID_DINODE(fe)) { + OCFS2_RO_ON_INVALID_DINODE(main_bm_inode->i_sb, fe); + ret = -EIO; + goto out_unlock; + } + + first_new_cluster = le32_to_cpu(fe->i_clusters); + lgd_blkno = ocfs2_which_cluster_group(main_bm_inode, + first_new_cluster - 1); + + ret = ocfs2_read_block(osb, lgd_blkno, &group_bh, OCFS2_BH_CACHED, + main_bm_inode); + if (ret < 0) { + mlog_errno(ret); + goto out_unlock; + } + + group = (struct ocfs2_group_desc *)group_bh->b_data; + + ret = ocfs2_check_group_descriptor(inode->i_sb, fe, group); + if (ret) { + mlog_errno(ret); + goto out_unlock; + } + + cl_bpc = le16_to_cpu(fe->id2.i_chain.cl_bpc); + if (le16_to_cpu(group->bg_bits) / cl_bpc + new_clusters > + le16_to_cpu(fe->id2.i_chain.cl_cpg)) { + ret = -EINVAL; + goto out_unlock; + } + + mlog(0, "extend the last group at %llu, new clusters = %d\n", + (unsigned long long)le64_to_cpu(group->bg_blkno), new_clusters); + + handle = ocfs2_start_trans(osb, OCFS2_GROUP_EXTEND_CREDITS); + if (IS_ERR(handle)) { + mlog_errno(PTR_ERR(handle)); + ret = -EINVAL; + goto out_unlock; + } + + /* update the last group descriptor and inode. */ + ret = ocfs2_update_last_group_and_inode(handle, main_bm_inode, + main_bm_bh, group_bh, + first_new_cluster, + new_clusters); + if (ret) { + mlog_errno(ret); + goto out_commit; + } + + ocfs2_update_super_and_backups(main_bm_inode, new_clusters); + +out_commit: + ocfs2_commit_trans(osb, handle); +out_unlock: + brelse(group_bh); + brelse(main_bm_bh); + + ocfs2_inode_unlock(main_bm_inode, 1); + +out_mutex: + mutex_unlock(&main_bm_inode->i_mutex); + iput(main_bm_inode); + +out: + mlog_exit_void(); + return ret; +} + +static int ocfs2_check_new_group(struct inode *inode, + struct ocfs2_dinode *di, + struct ocfs2_new_group_input *input, + struct buffer_head *group_bh) +{ + int ret; + struct ocfs2_group_desc *gd; + u16 cl_bpc = le16_to_cpu(di->id2.i_chain.cl_bpc); + unsigned int max_bits = le16_to_cpu(di->id2.i_chain.cl_cpg) * + le16_to_cpu(di->id2.i_chain.cl_bpc); + + + gd = (struct ocfs2_group_desc *)group_bh->b_data; + + ret = -EIO; + if (!OCFS2_IS_VALID_GROUP_DESC(gd)) + mlog(ML_ERROR, "Group descriptor # %llu isn't valid.\n", + (unsigned long long)le64_to_cpu(gd->bg_blkno)); + else if (di->i_blkno != gd->bg_parent_dinode) + mlog(ML_ERROR, "Group descriptor # %llu has bad parent " + "pointer (%llu, expected %llu)\n", + (unsigned long long)le64_to_cpu(gd->bg_blkno), + (unsigned long long)le64_to_cpu(gd->bg_parent_dinode), + (unsigned long long)le64_to_cpu(di->i_blkno)); + else if (le16_to_cpu(gd->bg_bits) > max_bits) + mlog(ML_ERROR, "Group descriptor # %llu has bit count of %u\n", + (unsigned long long)le64_to_cpu(gd->bg_blkno), + le16_to_cpu(gd->bg_bits)); + else if (le16_to_cpu(gd->bg_free_bits_count) > le16_to_cpu(gd->bg_bits)) + mlog(ML_ERROR, "Group descriptor # %llu has bit count %u but " + "claims that %u are free\n", + (unsigned long long)le64_to_cpu(gd->bg_blkno), + le16_to_cpu(gd->bg_bits), + le16_to_cpu(gd->bg_free_bits_count)); + else if (le16_to_cpu(gd->bg_bits) > (8 * le16_to_cpu(gd->bg_size))) + mlog(ML_ERROR, "Group descriptor # %llu has bit count %u but " + "max bitmap bits of %u\n", + (unsigned long long)le64_to_cpu(gd->bg_blkno), + le16_to_cpu(gd->bg_bits), + 8 * le16_to_cpu(gd->bg_size)); + else if (le16_to_cpu(gd->bg_chain) != input->chain) + mlog(ML_ERROR, "Group descriptor # %llu has bad chain %u " + "while input has %u set.\n", + (unsigned long long)le64_to_cpu(gd->bg_blkno), + le16_to_cpu(gd->bg_chain), input->chain); + else if (le16_to_cpu(gd->bg_bits) != input->clusters * cl_bpc) + mlog(ML_ERROR, "Group descriptor # %llu has bit count %u but " + "input has %u clusters set\n", + (unsigned long long)le64_to_cpu(gd->bg_blkno), + le16_to_cpu(gd->bg_bits), input->clusters); + else if (le16_to_cpu(gd->bg_free_bits_count) != input->frees * cl_bpc) + mlog(ML_ERROR, "Group descriptor # %llu has free bit count %u " + "but it should have %u set\n", + (unsigned long long)le64_to_cpu(gd->bg_blkno), + le16_to_cpu(gd->bg_bits), + input->frees * cl_bpc); + else + ret = 0; + + return ret; +} + +static int ocfs2_verify_group_and_input(struct inode *inode, + struct ocfs2_dinode *di, + struct ocfs2_new_group_input *input, + struct buffer_head *group_bh) +{ + u16 cl_count = le16_to_cpu(di->id2.i_chain.cl_count); + u16 cl_cpg = le16_to_cpu(di->id2.i_chain.cl_cpg); + u16 next_free = le16_to_cpu(di->id2.i_chain.cl_next_free_rec); + u32 cluster = ocfs2_blocks_to_clusters(inode->i_sb, input->group); + u32 total_clusters = le32_to_cpu(di->i_clusters); + int ret = -EINVAL; + + if (cluster < total_clusters) + mlog(ML_ERROR, "add a group which is in the current volume.\n"); + else if (input->chain >= cl_count) + mlog(ML_ERROR, "input chain exceeds the limit.\n"); + else if (next_free != cl_count && next_free != input->chain) + mlog(ML_ERROR, + "the add group should be in chain %u\n", next_free); + else if (total_clusters + input->clusters < total_clusters) + mlog(ML_ERROR, "add group's clusters overflow.\n"); + else if (input->clusters > cl_cpg) + mlog(ML_ERROR, "the cluster exceeds the maximum of a group\n"); + else if (input->frees > input->clusters) + mlog(ML_ERROR, "the free cluster exceeds the total clusters\n"); + else if (total_clusters % cl_cpg != 0) + mlog(ML_ERROR, + "the last group isn't full. Use group extend first.\n"); + else if (input->group != ocfs2_which_cluster_group(inode, cluster)) + mlog(ML_ERROR, "group blkno is invalid\n"); + else if ((ret = ocfs2_check_new_group(inode, di, input, group_bh))) + mlog(ML_ERROR, "group descriptor check failed.\n"); + else + ret = 0; + + return ret; +} + +/* Add a new group descriptor to global_bitmap. */ +int ocfs2_group_add(struct inode *inode, struct ocfs2_new_group_input *input) +{ + int ret; + handle_t *handle; + struct buffer_head *main_bm_bh = NULL; + struct inode *main_bm_inode = NULL; + struct ocfs2_dinode *fe = NULL; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + struct buffer_head *group_bh = NULL; + struct ocfs2_group_desc *group = NULL; + struct ocfs2_chain_list *cl; + struct ocfs2_chain_rec *cr; + u16 cl_bpc; + + mlog_entry_void(); + + if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb)) + return -EROFS; + + main_bm_inode = ocfs2_get_system_file_inode(osb, + GLOBAL_BITMAP_SYSTEM_INODE, + OCFS2_INVALID_SLOT); + if (!main_bm_inode) { + ret = -EINVAL; + mlog_errno(ret); + goto out; + } + + mutex_lock(&main_bm_inode->i_mutex); + + ret = ocfs2_inode_lock(main_bm_inode, &main_bm_bh, 1); + if (ret < 0) { + mlog_errno(ret); + goto out_mutex; + } + + fe = (struct ocfs2_dinode *)main_bm_bh->b_data; + + if (le16_to_cpu(fe->id2.i_chain.cl_cpg) != + ocfs2_group_bitmap_size(osb->sb) * 8) { + mlog(ML_ERROR, "The disk is too old and small." + " Force to do offline resize."); + ret = -EINVAL; + goto out_unlock; + } + + ret = ocfs2_read_block(osb, input->group, &group_bh, 0, NULL); + if (ret < 0) { + mlog(ML_ERROR, "Can't read the group descriptor # %llu " + "from the device.", (unsigned long long)input->group); + goto out_unlock; + } + + ocfs2_set_new_buffer_uptodate(inode, group_bh); + + ret = ocfs2_verify_group_and_input(main_bm_inode, fe, input, group_bh); + if (ret) { + mlog_errno(ret); + goto out_unlock; + } + + mlog(0, "Add a new group %llu in chain = %u, length = %u\n", + (unsigned long long)input->group, input->chain, input->clusters); + + handle = ocfs2_start_trans(osb, OCFS2_GROUP_ADD_CREDITS); + if (IS_ERR(handle)) { + mlog_errno(PTR_ERR(handle)); + ret = -EINVAL; + goto out_unlock; + } + + cl_bpc = le16_to_cpu(fe->id2.i_chain.cl_bpc); + cl = &fe->id2.i_chain; + cr = &cl->cl_recs[input->chain]; + + ret = ocfs2_journal_access(handle, main_bm_inode, group_bh, + OCFS2_JOURNAL_ACCESS_WRITE); + if (ret < 0) { + mlog_errno(ret); + goto out_commit; + } + + group = (struct ocfs2_group_desc *)group_bh->b_data; + group->bg_next_group = cr->c_blkno; + + ret = ocfs2_journal_dirty(handle, group_bh); + if (ret < 0) { + mlog_errno(ret); + goto out_commit; + } + + ret = ocfs2_journal_access(handle, main_bm_inode, main_bm_bh, + OCFS2_JOURNAL_ACCESS_WRITE); + if (ret < 0) { + mlog_errno(ret); + goto out_commit; + } + + if (input->chain == le16_to_cpu(cl->cl_next_free_rec)) { + le16_add_cpu(&cl->cl_next_free_rec, 1); + memset(cr, 0, sizeof(struct ocfs2_chain_rec)); + } + + cr->c_blkno = cpu_to_le64(input->group); + le32_add_cpu(&cr->c_total, input->clusters * cl_bpc); + le32_add_cpu(&cr->c_free, input->frees * cl_bpc); + + le32_add_cpu(&fe->id1.bitmap1.i_total, input->clusters *cl_bpc); + le32_add_cpu(&fe->id1.bitmap1.i_used, + (input->clusters - input->frees) * cl_bpc); + le32_add_cpu(&fe->i_clusters, input->clusters); + + ocfs2_journal_dirty(handle, main_bm_bh); + + spin_lock(&OCFS2_I(main_bm_inode)->ip_lock); + OCFS2_I(main_bm_inode)->ip_clusters = le32_to_cpu(fe->i_clusters); + le64_add_cpu(&fe->i_size, input->clusters << osb->s_clustersize_bits); + spin_unlock(&OCFS2_I(main_bm_inode)->ip_lock); + i_size_write(main_bm_inode, le64_to_cpu(fe->i_size)); + + ocfs2_update_super_and_backups(main_bm_inode, input->clusters); + +out_commit: + ocfs2_commit_trans(osb, handle); +out_unlock: + brelse(group_bh); + brelse(main_bm_bh); + + ocfs2_inode_unlock(main_bm_inode, 1); + +out_mutex: + mutex_unlock(&main_bm_inode->i_mutex); + iput(main_bm_inode); + +out: + mlog_exit_void(); + return ret; +} diff --git a/fs/ocfs2/endian.h b/fs/ocfs2/resize.h index ff257628af1..f38841abf10 100644 --- a/fs/ocfs2/endian.h +++ b/fs/ocfs2/resize.h @@ -1,7 +1,11 @@ /* -*- mode: c; c-basic-offset: 8; -*- * vim: noexpandtab sw=8 ts=8 sts=0: * - * Copyright (C) 2005 Oracle. All rights reserved. + * resize.h + * + * Function prototypes + * + * Copyright (C) 2007 Oracle. All rights reserved. * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public @@ -19,32 +23,10 @@ * Boston, MA 021110-1307, USA. */ -#ifndef OCFS2_ENDIAN_H -#define OCFS2_ENDIAN_H - -static inline void le16_add_cpu(__le16 *var, u16 val) -{ - *var = cpu_to_le16(le16_to_cpu(*var) + val); -} - -static inline void le32_add_cpu(__le32 *var, u32 val) -{ - *var = cpu_to_le32(le32_to_cpu(*var) + val); -} - -static inline void le64_add_cpu(__le64 *var, u64 val) -{ - *var = cpu_to_le64(le64_to_cpu(*var) + val); -} - -static inline void le32_and_cpu(__le32 *var, u32 val) -{ - *var = cpu_to_le32(le32_to_cpu(*var) & val); -} +#ifndef OCFS2_RESIZE_H +#define OCFS2_RESIZE_H -static inline void be32_add_cpu(__be32 *var, u32 val) -{ - *var = cpu_to_be32(be32_to_cpu(*var) + val); -} +int ocfs2_group_extend(struct inode * inode, int new_clusters); +int ocfs2_group_add(struct inode *inode, struct ocfs2_new_group_input *input); -#endif /* OCFS2_ENDIAN_H */ +#endif /* OCFS2_RESIZE_H */ diff --git a/fs/ocfs2/slot_map.c b/fs/ocfs2/slot_map.c index af4882b62cf..3a50ce555e6 100644 --- a/fs/ocfs2/slot_map.c +++ b/fs/ocfs2/slot_map.c @@ -48,25 +48,6 @@ static void __ocfs2_fill_slot(struct ocfs2_slot_info *si, s16 slot_num, s16 node_num); -/* Use the slot information we've collected to create a map of mounted - * nodes. Should be holding an EX on super block. assumes slot info is - * up to date. Note that we call this *after* we find a slot, so our - * own node should be set in the map too... */ -void ocfs2_populate_mounted_map(struct ocfs2_super *osb) -{ - int i; - struct ocfs2_slot_info *si = osb->slot_info; - - spin_lock(&si->si_lock); - - for (i = 0; i < si->si_size; i++) - if (si->si_global_node_nums[i] != OCFS2_INVALID_SLOT) - ocfs2_node_map_set_bit(osb, &osb->mounted_map, - si->si_global_node_nums[i]); - - spin_unlock(&si->si_lock); -} - /* post the slot information on disk into our slot_info struct. */ void ocfs2_update_slot_info(struct ocfs2_slot_info *si) { diff --git a/fs/ocfs2/slot_map.h b/fs/ocfs2/slot_map.h index d8c8ceed031..1025872aaad 100644 --- a/fs/ocfs2/slot_map.h +++ b/fs/ocfs2/slot_map.h @@ -52,8 +52,6 @@ s16 ocfs2_node_num_to_slot(struct ocfs2_slot_info *si, void ocfs2_clear_slot(struct ocfs2_slot_info *si, s16 slot_num); -void ocfs2_populate_mounted_map(struct ocfs2_super *osb); - static inline int ocfs2_is_empty_slot(struct ocfs2_slot_info *si, int slot_num) { diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index 8f09f5235e3..72c198a004d 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -101,8 +101,6 @@ static inline int ocfs2_block_group_reasonably_empty(struct ocfs2_group_desc *bg static inline u32 ocfs2_desc_bitmap_to_cluster_off(struct inode *inode, u64 bg_blkno, u16 bg_bit_off); -static inline u64 ocfs2_which_cluster_group(struct inode *inode, - u32 cluster); static inline void ocfs2_block_to_cluster_group(struct inode *inode, u64 data_blkno, u64 *bg_blkno, @@ -114,7 +112,7 @@ void ocfs2_free_alloc_context(struct ocfs2_alloc_context *ac) if (inode) { if (ac->ac_which != OCFS2_AC_USE_LOCAL) - ocfs2_meta_unlock(inode, 1); + ocfs2_inode_unlock(inode, 1); mutex_unlock(&inode->i_mutex); @@ -131,9 +129,9 @@ static u32 ocfs2_bits_per_group(struct ocfs2_chain_list *cl) } /* somewhat more expensive than our other checks, so use sparingly. */ -static int ocfs2_check_group_descriptor(struct super_block *sb, - struct ocfs2_dinode *di, - struct ocfs2_group_desc *gd) +int ocfs2_check_group_descriptor(struct super_block *sb, + struct ocfs2_dinode *di, + struct ocfs2_group_desc *gd) { unsigned int max_bits; @@ -412,7 +410,7 @@ static int ocfs2_reserve_suballoc_bits(struct ocfs2_super *osb, mutex_lock(&alloc_inode->i_mutex); - status = ocfs2_meta_lock(alloc_inode, &bh, 1); + status = ocfs2_inode_lock(alloc_inode, &bh, 1); if (status < 0) { mutex_unlock(&alloc_inode->i_mutex); iput(alloc_inode); @@ -648,7 +646,7 @@ bail: * sync-data inodes." * * Note: OCFS2 already does this differently for metadata vs data - * allocations, as those bitmaps are seperate and undo access is never + * allocations, as those bitmaps are separate and undo access is never * called on a metadata group descriptor. */ static int ocfs2_test_bg_bit_allocatable(struct buffer_head *bg_bh, @@ -1443,8 +1441,7 @@ static inline u32 ocfs2_desc_bitmap_to_cluster_off(struct inode *inode, /* given a cluster offset, calculate which block group it belongs to * and return that block offset. */ -static inline u64 ocfs2_which_cluster_group(struct inode *inode, - u32 cluster) +u64 ocfs2_which_cluster_group(struct inode *inode, u32 cluster) { struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); u32 group_no; @@ -1519,8 +1516,9 @@ int __ocfs2_claim_clusters(struct ocfs2_super *osb, if (min_clusters > (osb->bitmap_cpg - 1)) { /* The only paths asking for contiguousness * should know about this already. */ - mlog(ML_ERROR, "minimum allocation requested exceeds " - "group bitmap size!"); + mlog(ML_ERROR, "minimum allocation requested %u exceeds " + "group bitmap size %u!\n", min_clusters, + osb->bitmap_cpg); status = -ENOSPC; goto bail; } diff --git a/fs/ocfs2/suballoc.h b/fs/ocfs2/suballoc.h index cafe9370309..8799033bb45 100644 --- a/fs/ocfs2/suballoc.h +++ b/fs/ocfs2/suballoc.h @@ -147,4 +147,12 @@ static inline int ocfs2_is_cluster_bitmap(struct inode *inode) int ocfs2_reserve_cluster_bitmap_bits(struct ocfs2_super *osb, struct ocfs2_alloc_context *ac); +/* given a cluster offset, calculate which block group it belongs to + * and return that block offset. */ +u64 ocfs2_which_cluster_group(struct inode *inode, u32 cluster); + +/* somewhat more expensive than our other checks, so use sparingly. */ +int ocfs2_check_group_descriptor(struct super_block *sb, + struct ocfs2_dinode *di, + struct ocfs2_group_desc *gd); #endif /* _CHAINALLOC_H_ */ diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 5ee77542066..bec75aff3d9 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -65,7 +65,6 @@ #include "sysfile.h" #include "uptodate.h" #include "ver.h" -#include "vote.h" #include "buffer_head_io.h" @@ -84,9 +83,11 @@ MODULE_LICENSE("GPL"); struct mount_options { + unsigned long commit_interval; unsigned long mount_opt; unsigned int atime_quantum; signed short slot; + unsigned int localalloc_opt; }; static int ocfs2_parse_options(struct super_block *sb, char *options, @@ -150,6 +151,9 @@ enum { Opt_data_writeback, Opt_atime_quantum, Opt_slot, + Opt_commit, + Opt_localalloc, + Opt_localflocks, Opt_err, }; @@ -165,6 +169,9 @@ static match_table_t tokens = { {Opt_data_writeback, "data=writeback"}, {Opt_atime_quantum, "atime_quantum=%u"}, {Opt_slot, "preferred_slot=%u"}, + {Opt_commit, "commit=%u"}, + {Opt_localalloc, "localalloc=%d"}, + {Opt_localflocks, "localflocks"}, {Opt_err, NULL} }; @@ -213,7 +220,7 @@ static int ocfs2_init_global_system_inodes(struct ocfs2_super *osb) mlog_entry_void(); - new = ocfs2_iget(osb, osb->root_blkno, OCFS2_FI_FLAG_SYSFILE); + new = ocfs2_iget(osb, osb->root_blkno, OCFS2_FI_FLAG_SYSFILE, 0); if (IS_ERR(new)) { status = PTR_ERR(new); mlog_errno(status); @@ -221,7 +228,7 @@ static int ocfs2_init_global_system_inodes(struct ocfs2_super *osb) } osb->root_inode = new; - new = ocfs2_iget(osb, osb->system_dir_blkno, OCFS2_FI_FLAG_SYSFILE); + new = ocfs2_iget(osb, osb->system_dir_blkno, OCFS2_FI_FLAG_SYSFILE, 0); if (IS_ERR(new)) { status = PTR_ERR(new); mlog_errno(status); @@ -443,6 +450,8 @@ unlock_osb: osb->s_mount_opt = parsed_options.mount_opt; osb->s_atime_quantum = parsed_options.atime_quantum; osb->preferred_slot = parsed_options.slot; + if (parsed_options.commit_interval) + osb->osb_commit_interval = parsed_options.commit_interval; if (!ocfs2_is_hard_readonly(osb)) ocfs2_set_journal_params(osb); @@ -597,6 +606,8 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent) osb->s_mount_opt = parsed_options.mount_opt; osb->s_atime_quantum = parsed_options.atime_quantum; osb->preferred_slot = parsed_options.slot; + osb->osb_commit_interval = parsed_options.commit_interval; + osb->local_alloc_size = parsed_options.localalloc_opt; sb->s_magic = OCFS2_SUPER_MAGIC; @@ -747,9 +758,11 @@ static int ocfs2_parse_options(struct super_block *sb, mlog_entry("remount: %d, options: \"%s\"\n", is_remount, options ? options : "(none)"); + mopt->commit_interval = 0; mopt->mount_opt = 0; mopt->atime_quantum = OCFS2_DEFAULT_ATIME_QUANTUM; mopt->slot = OCFS2_INVALID_SLOT; + mopt->localalloc_opt = OCFS2_DEFAULT_LOCAL_ALLOC_SIZE; if (!options) { status = 1; @@ -816,6 +829,41 @@ static int ocfs2_parse_options(struct super_block *sb, if (option) mopt->slot = (s16)option; break; + case Opt_commit: + option = 0; + if (match_int(&args[0], &option)) { + status = 0; + goto bail; + } + if (option < 0) + return 0; + if (option == 0) + option = JBD_DEFAULT_MAX_COMMIT_AGE; + mopt->commit_interval = HZ * option; + break; + case Opt_localalloc: + option = 0; + if (match_int(&args[0], &option)) { + status = 0; + goto bail; + } + if (option >= 0 && (option <= ocfs2_local_alloc_size(sb) * 8)) + mopt->localalloc_opt = option; + break; + case Opt_localflocks: + /* + * Changing this during remount could race + * flock() requests, or "unbalance" existing + * ones (e.g., a lock is taken in one mode but + * dropped in the other). If users care enough + * to flip locking modes during remount, we + * could add a "local" flag to individual + * flock structures for proper tracking of + * state. + */ + if (!is_remount) + mopt->mount_opt |= OCFS2_MOUNT_LOCALFLOCKS; + break; default: mlog(ML_ERROR, "Unrecognized mount option \"%s\" " @@ -864,6 +912,16 @@ static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt) if (osb->s_atime_quantum != OCFS2_DEFAULT_ATIME_QUANTUM) seq_printf(s, ",atime_quantum=%u", osb->s_atime_quantum); + if (osb->osb_commit_interval) + seq_printf(s, ",commit=%u", + (unsigned) (osb->osb_commit_interval / HZ)); + + if (osb->local_alloc_size != OCFS2_DEFAULT_LOCAL_ALLOC_SIZE) + seq_printf(s, ",localalloc=%d", osb->local_alloc_size); + + if (opts & OCFS2_MOUNT_LOCALFLOCKS) + seq_printf(s, ",localflocks,"); + return 0; } @@ -965,7 +1023,7 @@ static int ocfs2_statfs(struct dentry *dentry, struct kstatfs *buf) goto bail; } - status = ocfs2_meta_lock(inode, &bh, 0); + status = ocfs2_inode_lock(inode, &bh, 0); if (status < 0) { mlog_errno(status); goto bail; @@ -989,7 +1047,7 @@ static int ocfs2_statfs(struct dentry *dentry, struct kstatfs *buf) brelse(bh); - ocfs2_meta_unlock(inode, 0); + ocfs2_inode_unlock(inode, 0); status = 0; bail: if (inode) @@ -1020,8 +1078,7 @@ static void ocfs2_inode_init_once(struct kmem_cache *cachep, void *data) oi->ip_clusters = 0; ocfs2_lock_res_init_once(&oi->ip_rw_lockres); - ocfs2_lock_res_init_once(&oi->ip_meta_lockres); - ocfs2_lock_res_init_once(&oi->ip_data_lockres); + ocfs2_lock_res_init_once(&oi->ip_inode_lockres); ocfs2_lock_res_init_once(&oi->ip_open_lockres); ocfs2_metadata_cache_init(&oi->vfs_inode); @@ -1117,25 +1174,12 @@ static int ocfs2_mount_volume(struct super_block *sb) goto leave; } - status = ocfs2_register_hb_callbacks(osb); - if (status < 0) { - mlog_errno(status); - goto leave; - } - status = ocfs2_dlm_init(osb); if (status < 0) { mlog_errno(status); goto leave; } - /* requires vote_thread to be running. */ - status = ocfs2_register_net_handlers(osb); - if (status < 0) { - mlog_errno(status); - goto leave; - } - status = ocfs2_super_lock(osb, 1); if (status < 0) { mlog_errno(status); @@ -1150,8 +1194,6 @@ static int ocfs2_mount_volume(struct super_block *sb) goto leave; } - ocfs2_populate_mounted_map(osb); - /* load all node-local system inodes */ status = ocfs2_init_local_system_inodes(osb); if (status < 0) { @@ -1174,15 +1216,6 @@ static int ocfs2_mount_volume(struct super_block *sb) if (ocfs2_mount_local(osb)) goto leave; - /* This should be sent *after* we recovered our journal as it - * will cause other nodes to unmark us as needing - * recovery. However, we need to send it *before* dropping the - * super block lock as otherwise their recovery threads might - * try to clean us up while we're live! */ - status = ocfs2_request_mount_vote(osb); - if (status < 0) - mlog_errno(status); - leave: if (unlock_super) ocfs2_super_unlock(osb, 1); @@ -1240,10 +1273,6 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err) mlog_errno(tmp); return; } - - tmp = ocfs2_request_umount_vote(osb); - if (tmp < 0) - mlog_errno(tmp); } if (osb->slot_num != OCFS2_INVALID_SLOT) @@ -1254,13 +1283,8 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err) ocfs2_release_system_inodes(osb); - if (osb->dlm) { - ocfs2_unregister_net_handlers(osb); - + if (osb->dlm) ocfs2_dlm_shutdown(osb); - } - - ocfs2_clear_hb_callbacks(osb); debugfs_remove(osb->osb_debug_root); @@ -1315,7 +1339,6 @@ static int ocfs2_initialize_super(struct super_block *sb, int i, cbits, bbits; struct ocfs2_dinode *di = (struct ocfs2_dinode *)bh->b_data; struct inode *inode = NULL; - struct buffer_head *bitmap_bh = NULL; struct ocfs2_journal *journal; __le32 uuid_net_key; struct ocfs2_super *osb; @@ -1332,6 +1355,7 @@ static int ocfs2_initialize_super(struct super_block *sb, sb->s_fs_info = osb; sb->s_op = &ocfs2_sops; sb->s_export_op = &ocfs2_export_ops; + osb->osb_locking_proto = ocfs2_locking_protocol; sb->s_time_gran = 1; sb->s_flags |= MS_NOATIME; /* this is needed to support O_LARGEFILE */ @@ -1344,19 +1368,13 @@ static int ocfs2_initialize_super(struct super_block *sb, osb->s_sectsize_bits = blksize_bits(sector_size); BUG_ON(!osb->s_sectsize_bits); - osb->net_response_ids = 0; - spin_lock_init(&osb->net_response_lock); - INIT_LIST_HEAD(&osb->net_response_list); - - INIT_LIST_HEAD(&osb->osb_net_handlers); init_waitqueue_head(&osb->recovery_event); - spin_lock_init(&osb->vote_task_lock); - init_waitqueue_head(&osb->vote_event); - osb->vote_work_sequence = 0; - osb->vote_wake_sequence = 0; + spin_lock_init(&osb->dc_task_lock); + init_waitqueue_head(&osb->dc_event); + osb->dc_work_sequence = 0; + osb->dc_wake_sequence = 0; INIT_LIST_HEAD(&osb->blocked_lock_list); osb->blocked_lock_count = 0; - INIT_LIST_HEAD(&osb->vote_list); spin_lock_init(&osb->osb_lock); atomic_set(&osb->alloc_stats.moves, 0); @@ -1496,7 +1514,6 @@ static int ocfs2_initialize_super(struct super_block *sb, } memcpy(&uuid_net_key, di->id2.i_super.s_uuid, sizeof(uuid_net_key)); - osb->net_key = le32_to_cpu(uuid_net_key); strncpy(osb->vol_label, di->id2.i_super.s_label, 63); osb->vol_label[63] = '\0'; @@ -1539,25 +1556,9 @@ static int ocfs2_initialize_super(struct super_block *sb, } osb->bitmap_blkno = OCFS2_I(inode)->ip_blkno; - - /* We don't have a cluster lock on the bitmap here because - * we're only interested in static information and the extra - * complexity at mount time isn't worht it. Don't pass the - * inode in to the read function though as we don't want it to - * be put in the cache. */ - status = ocfs2_read_block(osb, osb->bitmap_blkno, &bitmap_bh, 0, - NULL); iput(inode); - if (status < 0) { - mlog_errno(status); - goto bail; - } - di = (struct ocfs2_dinode *) bitmap_bh->b_data; - osb->bitmap_cpg = le16_to_cpu(di->id2.i_chain.cl_cpg); - brelse(bitmap_bh); - mlog(0, "cluster bitmap inode: %llu, clusters per group: %u\n", - (unsigned long long)osb->bitmap_blkno, osb->bitmap_cpg); + osb->bitmap_cpg = ocfs2_group_bitmap_size(sb) * 8; status = ocfs2_init_slot_info(osb); if (status < 0) { diff --git a/fs/ocfs2/sysfile.c b/fs/ocfs2/sysfile.c index fd2e846e3e6..ab713ebdd54 100644 --- a/fs/ocfs2/sysfile.c +++ b/fs/ocfs2/sysfile.c @@ -112,7 +112,7 @@ static struct inode * _ocfs2_get_system_file_inode(struct ocfs2_super *osb, goto bail; } - inode = ocfs2_iget(osb, blkno, OCFS2_FI_FLAG_SYSFILE); + inode = ocfs2_iget(osb, blkno, OCFS2_FI_FLAG_SYSFILE, type); if (IS_ERR(inode)) { mlog_errno(PTR_ERR(inode)); inode = NULL; diff --git a/fs/ocfs2/ver.c b/fs/ocfs2/ver.c index 5405ce121c9..e2488f4128a 100644 --- a/fs/ocfs2/ver.c +++ b/fs/ocfs2/ver.c @@ -29,7 +29,7 @@ #include "ver.h" -#define OCFS2_BUILD_VERSION "1.3.3" +#define OCFS2_BUILD_VERSION "1.5.0" #define VERSION_STR "OCFS2 " OCFS2_BUILD_VERSION diff --git a/fs/ocfs2/vote.c b/fs/ocfs2/vote.c deleted file mode 100644 index c05358538f2..00000000000 --- a/fs/ocfs2/vote.c +++ /dev/null @@ -1,756 +0,0 @@ -/* -*- mode: c; c-basic-offset: 8; -*- - * vim: noexpandtab sw=8 ts=8 sts=0: - * - * vote.c - * - * description here - * - * Copyright (C) 2003, 2004 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License as published by the Free Software Foundation; either - * version 2 of the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. - */ - -#include <linux/types.h> -#include <linux/slab.h> -#include <linux/highmem.h> -#include <linux/kthread.h> - -#include <cluster/heartbeat.h> -#include <cluster/nodemanager.h> -#include <cluster/tcp.h> - -#include <dlm/dlmapi.h> - -#define MLOG_MASK_PREFIX ML_VOTE -#include <cluster/masklog.h> - -#include "ocfs2.h" - -#include "alloc.h" -#include "dlmglue.h" -#include "extent_map.h" -#include "heartbeat.h" -#include "inode.h" -#include "journal.h" -#include "slot_map.h" -#include "vote.h" - -#include "buffer_head_io.h" - -#define OCFS2_MESSAGE_TYPE_VOTE (0x1) -#define OCFS2_MESSAGE_TYPE_RESPONSE (0x2) -struct ocfs2_msg_hdr -{ - __be32 h_response_id; /* used to lookup message handle on sending - * node. */ - __be32 h_request; - __be64 h_blkno; - __be32 h_generation; - __be32 h_node_num; /* node sending this particular message. */ -}; - -struct ocfs2_vote_msg -{ - struct ocfs2_msg_hdr v_hdr; - __be32 v_reserved1; -} __attribute__ ((packed)); - -/* Responses are given these values to maintain backwards - * compatibility with older ocfs2 versions */ -#define OCFS2_RESPONSE_OK (0) -#define OCFS2_RESPONSE_BUSY (-16) -#define OCFS2_RESPONSE_BAD_MSG (-22) - -struct ocfs2_response_msg -{ - struct ocfs2_msg_hdr r_hdr; - __be32 r_response; -} __attribute__ ((packed)); - -struct ocfs2_vote_work { - struct list_head w_list; - struct ocfs2_vote_msg w_msg; -}; - -enum ocfs2_vote_request { - OCFS2_VOTE_REQ_INVALID = 0, - OCFS2_VOTE_REQ_MOUNT, - OCFS2_VOTE_REQ_UMOUNT, - OCFS2_VOTE_REQ_LAST -}; - -static inline int ocfs2_is_valid_vote_request(int request) -{ - return OCFS2_VOTE_REQ_INVALID < request && - request < OCFS2_VOTE_REQ_LAST; -} - -typedef void (*ocfs2_net_response_callback)(void *priv, - struct ocfs2_response_msg *resp); -struct ocfs2_net_response_cb { - ocfs2_net_response_callback rc_cb; - void *rc_priv; -}; - -struct ocfs2_net_wait_ctxt { - struct list_head n_list; - u32 n_response_id; - wait_queue_head_t n_event; - struct ocfs2_node_map n_node_map; - int n_response; /* an agreggate response. 0 if - * all nodes are go, < 0 on any - * negative response from any - * node or network error. */ - struct ocfs2_net_response_cb *n_callback; -}; - -static void ocfs2_process_mount_request(struct ocfs2_super *osb, - unsigned int node_num) -{ - mlog(0, "MOUNT vote from node %u\n", node_num); - /* The other node only sends us this message when he has an EX - * on the superblock, so our recovery threads (if having been - * launched) are waiting on it.*/ - ocfs2_recovery_map_clear(osb, node_num); - ocfs2_node_map_set_bit(osb, &osb->mounted_map, node_num); - - /* We clear the umount map here because a node may have been - * previously mounted, safely unmounted but never stopped - * heartbeating - in which case we'd have a stale entry. */ - ocfs2_node_map_clear_bit(osb, &osb->umount_map, node_num); -} - -static void ocfs2_process_umount_request(struct ocfs2_super *osb, - unsigned int node_num) -{ - mlog(0, "UMOUNT vote from node %u\n", node_num); - ocfs2_node_map_clear_bit(osb, &osb->mounted_map, node_num); - ocfs2_node_map_set_bit(osb, &osb->umount_map, node_num); -} - -static void ocfs2_process_vote(struct ocfs2_super *osb, - struct ocfs2_vote_msg *msg) -{ - int net_status, vote_response; - unsigned int node_num; - u64 blkno; - enum ocfs2_vote_request request; - struct ocfs2_msg_hdr *hdr = &msg->v_hdr; - struct ocfs2_response_msg response; - - /* decode the network mumbo jumbo into local variables. */ - request = be32_to_cpu(hdr->h_request); - blkno = be64_to_cpu(hdr->h_blkno); - node_num = be32_to_cpu(hdr->h_node_num); - - mlog(0, "processing vote: request = %u, blkno = %llu, node_num = %u\n", - request, (unsigned long long)blkno, node_num); - - if (!ocfs2_is_valid_vote_request(request)) { - mlog(ML_ERROR, "Invalid vote request %d from node %u\n", - request, node_num); - vote_response = OCFS2_RESPONSE_BAD_MSG; - goto respond; - } - - vote_response = OCFS2_RESPONSE_OK; - - switch (request) { - case OCFS2_VOTE_REQ_UMOUNT: - ocfs2_process_umount_request(osb, node_num); - goto respond; - case OCFS2_VOTE_REQ_MOUNT: - ocfs2_process_mount_request(osb, node_num); - goto respond; - default: - /* avoids a gcc warning */ - break; - } - -respond: - /* Response struture is small so we just put it on the stack - * and stuff it inline. */ - memset(&response, 0, sizeof(struct ocfs2_response_msg)); - response.r_hdr.h_response_id = hdr->h_response_id; - response.r_hdr.h_blkno = hdr->h_blkno; - response.r_hdr.h_generation = hdr->h_generation; - response.r_hdr.h_node_num = cpu_to_be32(osb->node_num); - response.r_response = cpu_to_be32(vote_response); - - net_status = o2net_send_message(OCFS2_MESSAGE_TYPE_RESPONSE, - osb->net_key, - &response, - sizeof(struct ocfs2_response_msg), - node_num, - NULL); - /* We still want to error print for ENOPROTOOPT here. The - * sending node shouldn't have unregistered his net handler - * without sending an unmount vote 1st */ - if (net_status < 0 - && net_status != -ETIMEDOUT - && net_status != -ENOTCONN) - mlog(ML_ERROR, "message to node %u fails with error %d!\n", - node_num, net_status); -} - -static void ocfs2_vote_thread_do_work(struct ocfs2_super *osb) -{ - unsigned long processed; - struct ocfs2_lock_res *lockres; - struct ocfs2_vote_work *work; - - mlog_entry_void(); - - spin_lock(&osb->vote_task_lock); - /* grab this early so we know to try again if a state change and - * wake happens part-way through our work */ - osb->vote_work_sequence = osb->vote_wake_sequence; - - processed = osb->blocked_lock_count; - while (processed) { - BUG_ON(list_empty(&osb->blocked_lock_list)); - - lockres = list_entry(osb->blocked_lock_list.next, - struct ocfs2_lock_res, l_blocked_list); - list_del_init(&lockres->l_blocked_list); - osb->blocked_lock_count--; - spin_unlock(&osb->vote_task_lock); - - BUG_ON(!processed); - processed--; - - ocfs2_process_blocked_lock(osb, lockres); - - spin_lock(&osb->vote_task_lock); - } - - while (osb->vote_count) { - BUG_ON(list_empty(&osb->vote_list)); - work = list_entry(osb->vote_list.next, - struct ocfs2_vote_work, w_list); - list_del(&work->w_list); - osb->vote_count--; - spin_unlock(&osb->vote_task_lock); - - ocfs2_process_vote(osb, &work->w_msg); - kfree(work); - - spin_lock(&osb->vote_task_lock); - } - spin_unlock(&osb->vote_task_lock); - - mlog_exit_void(); -} - -static int ocfs2_vote_thread_lists_empty(struct ocfs2_super *osb) -{ - int empty = 0; - - spin_lock(&osb->vote_task_lock); - if (list_empty(&osb->blocked_lock_list) && - list_empty(&osb->vote_list)) - empty = 1; - - spin_unlock(&osb->vote_task_lock); - return empty; -} - -static int ocfs2_vote_thread_should_wake(struct ocfs2_super *osb) -{ - int should_wake = 0; - - spin_lock(&osb->vote_task_lock); - if (osb->vote_work_sequence != osb->vote_wake_sequence) - should_wake = 1; - spin_unlock(&osb->vote_task_lock); - - return should_wake; -} - -int ocfs2_vote_thread(void *arg) -{ - int status = 0; - struct ocfs2_super *osb = arg; - - /* only quit once we've been asked to stop and there is no more - * work available */ - while (!(kthread_should_stop() && - ocfs2_vote_thread_lists_empty(osb))) { - - wait_event_interruptible(osb->vote_event, - ocfs2_vote_thread_should_wake(osb) || - kthread_should_stop()); - - mlog(0, "vote_thread: awoken\n"); - - ocfs2_vote_thread_do_work(osb); - } - - osb->vote_task = NULL; - return status; -} - -static struct ocfs2_net_wait_ctxt *ocfs2_new_net_wait_ctxt(unsigned int response_id) -{ - struct ocfs2_net_wait_ctxt *w; - - w = kzalloc(sizeof(*w), GFP_NOFS); - if (!w) { - mlog_errno(-ENOMEM); - goto bail; - } - - INIT_LIST_HEAD(&w->n_list); - init_waitqueue_head(&w->n_event); - ocfs2_node_map_init(&w->n_node_map); - w->n_response_id = response_id; - w->n_callback = NULL; -bail: - return w; -} - -static unsigned int ocfs2_new_response_id(struct ocfs2_super *osb) -{ - unsigned int ret; - - spin_lock(&osb->net_response_lock); - ret = ++osb->net_response_ids; - spin_unlock(&osb->net_response_lock); - - return ret; -} - -static void ocfs2_dequeue_net_wait_ctxt(struct ocfs2_super *osb, - struct ocfs2_net_wait_ctxt *w) -{ - spin_lock(&osb->net_response_lock); - list_del(&w->n_list); - spin_unlock(&osb->net_response_lock); -} - -static void ocfs2_queue_net_wait_ctxt(struct ocfs2_super *osb, - struct ocfs2_net_wait_ctxt *w) -{ - spin_lock(&osb->net_response_lock); - list_add_tail(&w->n_list, - &osb->net_response_list); - spin_unlock(&osb->net_response_lock); -} - -static void __ocfs2_mark_node_responded(struct ocfs2_super *osb, - struct ocfs2_net_wait_ctxt *w, - int node_num) -{ - assert_spin_locked(&osb->net_response_lock); - - ocfs2_node_map_clear_bit(osb, &w->n_node_map, node_num); - if (ocfs2_node_map_is_empty(osb, &w->n_node_map)) - wake_up(&w->n_event); -} - -/* Intended to be called from the node down callback, we fake remove - * the node from all our response contexts */ -void ocfs2_remove_node_from_vote_queues(struct ocfs2_super *osb, - int node_num) -{ - struct list_head *p; - struct ocfs2_net_wait_ctxt *w = NULL; - - spin_lock(&osb->net_response_lock); - - list_for_each(p, &osb->net_response_list) { - w = list_entry(p, struct ocfs2_net_wait_ctxt, n_list); - - __ocfs2_mark_node_responded(osb, w, node_num); - } - - spin_unlock(&osb->net_response_lock); -} - -static int ocfs2_broadcast_vote(struct ocfs2_super *osb, - struct ocfs2_vote_msg *request, - unsigned int response_id, - int *response, - struct ocfs2_net_response_cb *callback) -{ - int status, i, remote_err; - struct ocfs2_net_wait_ctxt *w = NULL; - int dequeued = 0; - - mlog_entry_void(); - - w = ocfs2_new_net_wait_ctxt(response_id); - if (!w) { - status = -ENOMEM; - mlog_errno(status); - goto bail; - } - w->n_callback = callback; - - /* we're pretty much ready to go at this point, and this fills - * in n_response which we need anyway... */ - ocfs2_queue_net_wait_ctxt(osb, w); - - i = ocfs2_node_map_iterate(osb, &osb->mounted_map, 0); - - while (i != O2NM_INVALID_NODE_NUM) { - if (i != osb->node_num) { - mlog(0, "trying to send request to node %i\n", i); - ocfs2_node_map_set_bit(osb, &w->n_node_map, i); - - remote_err = 0; - status = o2net_send_message(OCFS2_MESSAGE_TYPE_VOTE, - osb->net_key, - request, - sizeof(*request), - i, - &remote_err); - if (status == -ETIMEDOUT) { - mlog(0, "remote node %d timed out!\n", i); - status = -EAGAIN; - goto bail; - } - if (remote_err < 0) { - status = remote_err; - mlog(0, "remote error %d on node %d!\n", - remote_err, i); - mlog_errno(status); - goto bail; - } - if (status < 0) { - mlog_errno(status); - goto bail; - } - } - i++; - i = ocfs2_node_map_iterate(osb, &osb->mounted_map, i); - mlog(0, "next is %d, i am %d\n", i, osb->node_num); - } - mlog(0, "done sending, now waiting on responses...\n"); - - wait_event(w->n_event, ocfs2_node_map_is_empty(osb, &w->n_node_map)); - - ocfs2_dequeue_net_wait_ctxt(osb, w); - dequeued = 1; - - *response = w->n_response; - status = 0; -bail: - if (w) { - if (!dequeued) - ocfs2_dequeue_net_wait_ctxt(osb, w); - kfree(w); - } - - mlog_exit(status); - return status; -} - -static struct ocfs2_vote_msg * ocfs2_new_vote_request(struct ocfs2_super *osb, - u64 blkno, - unsigned int generation, - enum ocfs2_vote_request type) -{ - struct ocfs2_vote_msg *request; - struct ocfs2_msg_hdr *hdr; - - BUG_ON(!ocfs2_is_valid_vote_request(type)); - - request = kzalloc(sizeof(*request), GFP_NOFS); - if (!request) { - mlog_errno(-ENOMEM); - } else { - hdr = &request->v_hdr; - hdr->h_node_num = cpu_to_be32(osb->node_num); - hdr->h_request = cpu_to_be32(type); - hdr->h_blkno = cpu_to_be64(blkno); - hdr->h_generation = cpu_to_be32(generation); - } - - return request; -} - -/* Complete the buildup of a new vote request and process the - * broadcast return value. */ -static int ocfs2_do_request_vote(struct ocfs2_super *osb, - struct ocfs2_vote_msg *request, - struct ocfs2_net_response_cb *callback) -{ - int status, response = -EBUSY; - unsigned int response_id; - struct ocfs2_msg_hdr *hdr; - - response_id = ocfs2_new_response_id(osb); - - hdr = &request->v_hdr; - hdr->h_response_id = cpu_to_be32(response_id); - - status = ocfs2_broadcast_vote(osb, request, response_id, &response, - callback); - if (status < 0) { - mlog_errno(status); - goto bail; - } - - status = response; -bail: - - return status; -} - -int ocfs2_request_mount_vote(struct ocfs2_super *osb) -{ - int status; - struct ocfs2_vote_msg *request = NULL; - - request = ocfs2_new_vote_request(osb, 0ULL, 0, OCFS2_VOTE_REQ_MOUNT); - if (!request) { - status = -ENOMEM; - goto bail; - } - - status = -EAGAIN; - while (status == -EAGAIN) { - if (!(osb->s_mount_opt & OCFS2_MOUNT_NOINTR) && - signal_pending(current)) { - status = -ERESTARTSYS; - goto bail; - } - - if (ocfs2_node_map_is_only(osb, &osb->mounted_map, - osb->node_num)) { - status = 0; - goto bail; - } - - status = ocfs2_do_request_vote(osb, request, NULL); - } - -bail: - kfree(request); - return status; -} - -int ocfs2_request_umount_vote(struct ocfs2_super *osb) -{ - int status; - struct ocfs2_vote_msg *request = NULL; - - request = ocfs2_new_vote_request(osb, 0ULL, 0, OCFS2_VOTE_REQ_UMOUNT); - if (!request) { - status = -ENOMEM; - goto bail; - } - - status = -EAGAIN; - while (status == -EAGAIN) { - /* Do not check signals on this vote... We really want - * this one to go all the way through. */ - - if (ocfs2_node_map_is_only(osb, &osb->mounted_map, - osb->node_num)) { - status = 0; - goto bail; - } - - status = ocfs2_do_request_vote(osb, request, NULL); - } - -bail: - kfree(request); - return status; -} - -/* TODO: This should eventually be a hash table! */ -static struct ocfs2_net_wait_ctxt * __ocfs2_find_net_wait_ctxt(struct ocfs2_super *osb, - u32 response_id) -{ - struct list_head *p; - struct ocfs2_net_wait_ctxt *w = NULL; - - list_for_each(p, &osb->net_response_list) { - w = list_entry(p, struct ocfs2_net_wait_ctxt, n_list); - if (response_id == w->n_response_id) - break; - w = NULL; - } - - return w; -} - -/* Translate response codes into local node errno values */ -static inline int ocfs2_translate_response(int response) -{ - int ret; - - switch (response) { - case OCFS2_RESPONSE_OK: - ret = 0; - break; - - case OCFS2_RESPONSE_BUSY: - ret = -EBUSY; - break; - - default: - ret = -EINVAL; - } - - return ret; -} - -static int ocfs2_handle_response_message(struct o2net_msg *msg, - u32 len, - void *data, void **ret_data) -{ - unsigned int response_id, node_num; - int response_status; - struct ocfs2_super *osb = data; - struct ocfs2_response_msg *resp; - struct ocfs2_net_wait_ctxt * w; - struct ocfs2_net_response_cb *resp_cb; - - resp = (struct ocfs2_response_msg *) msg->buf; - - response_id = be32_to_cpu(resp->r_hdr.h_response_id); - node_num = be32_to_cpu(resp->r_hdr.h_node_num); - response_status = - ocfs2_translate_response(be32_to_cpu(resp->r_response)); - - mlog(0, "received response message:\n"); - mlog(0, "h_response_id = %u\n", response_id); - mlog(0, "h_request = %u\n", be32_to_cpu(resp->r_hdr.h_request)); - mlog(0, "h_blkno = %llu\n", - (unsigned long long)be64_to_cpu(resp->r_hdr.h_blkno)); - mlog(0, "h_generation = %u\n", be32_to_cpu(resp->r_hdr.h_generation)); - mlog(0, "h_node_num = %u\n", node_num); - mlog(0, "r_response = %d\n", response_status); - - spin_lock(&osb->net_response_lock); - w = __ocfs2_find_net_wait_ctxt(osb, response_id); - if (!w) { - mlog(0, "request not found!\n"); - goto bail; - } - resp_cb = w->n_callback; - - if (response_status && (!w->n_response)) { - /* we only really need one negative response so don't - * set it twice. */ - w->n_response = response_status; - } - - if (resp_cb) { - spin_unlock(&osb->net_response_lock); - - resp_cb->rc_cb(resp_cb->rc_priv, resp); - - spin_lock(&osb->net_response_lock); - } - - __ocfs2_mark_node_responded(osb, w, node_num); -bail: - spin_unlock(&osb->net_response_lock); - - return 0; -} - -static int ocfs2_handle_vote_message(struct o2net_msg *msg, - u32 len, - void *data, void **ret_data) -{ - int status; - struct ocfs2_super *osb = data; - struct ocfs2_vote_work *work; - - work = kmalloc(sizeof(struct ocfs2_vote_work), GFP_NOFS); - if (!work) { - status = -ENOMEM; - mlog_errno(status); - goto bail; - } - - INIT_LIST_HEAD(&work->w_list); - memcpy(&work->w_msg, msg->buf, sizeof(struct ocfs2_vote_msg)); - - mlog(0, "scheduling vote request:\n"); - mlog(0, "h_response_id = %u\n", - be32_to_cpu(work->w_msg.v_hdr.h_response_id)); - mlog(0, "h_request = %u\n", be32_to_cpu(work->w_msg.v_hdr.h_request)); - mlog(0, "h_blkno = %llu\n", - (unsigned long long)be64_to_cpu(work->w_msg.v_hdr.h_blkno)); - mlog(0, "h_generation = %u\n", - be32_to_cpu(work->w_msg.v_hdr.h_generation)); - mlog(0, "h_node_num = %u\n", - be32_to_cpu(work->w_msg.v_hdr.h_node_num)); - - spin_lock(&osb->vote_task_lock); - list_add_tail(&work->w_list, &osb->vote_list); - osb->vote_count++; - spin_unlock(&osb->vote_task_lock); - - ocfs2_kick_vote_thread(osb); - - status = 0; -bail: - return status; -} - -void ocfs2_unregister_net_handlers(struct ocfs2_super *osb) -{ - if (!osb->net_key) - return; - - o2net_unregister_handler_list(&osb->osb_net_handlers); - - if (!list_empty(&osb->net_response_list)) - mlog(ML_ERROR, "net response list not empty!\n"); - - osb->net_key = 0; -} - -int ocfs2_register_net_handlers(struct ocfs2_super *osb) -{ - int status = 0; - - if (ocfs2_mount_local(osb)) - return 0; - - status = o2net_register_handler(OCFS2_MESSAGE_TYPE_RESPONSE, - osb->net_key, - sizeof(struct ocfs2_response_msg), - ocfs2_handle_response_message, - osb, NULL, &osb->osb_net_handlers); - if (status) { - mlog_errno(status); - goto bail; - } - - status = o2net_register_handler(OCFS2_MESSAGE_TYPE_VOTE, - osb->net_key, - sizeof(struct ocfs2_vote_msg), - ocfs2_handle_vote_message, - osb, NULL, &osb->osb_net_handlers); - if (status) { - mlog_errno(status); - goto bail; - } -bail: - if (status < 0) - ocfs2_unregister_net_handlers(osb); - - return status; -} diff --git a/fs/ocfs2/vote.h b/fs/ocfs2/vote.h deleted file mode 100644 index 9ea46f62de3..00000000000 --- a/fs/ocfs2/vote.h +++ /dev/null @@ -1,48 +0,0 @@ -/* -*- mode: c; c-basic-offset: 8; -*- - * vim: noexpandtab sw=8 ts=8 sts=0: - * - * vote.h - * - * description here - * - * Copyright (C) 2002, 2004 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License as published by the Free Software Foundation; either - * version 2 of the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. - */ - - -#ifndef VOTE_H -#define VOTE_H - -int ocfs2_vote_thread(void *arg); -static inline void ocfs2_kick_vote_thread(struct ocfs2_super *osb) -{ - spin_lock(&osb->vote_task_lock); - /* make sure the voting thread gets a swipe at whatever changes - * the caller may have made to the voting state */ - osb->vote_wake_sequence++; - spin_unlock(&osb->vote_task_lock); - wake_up(&osb->vote_event); -} - -int ocfs2_request_mount_vote(struct ocfs2_super *osb); -int ocfs2_request_umount_vote(struct ocfs2_super *osb); -int ocfs2_register_net_handlers(struct ocfs2_super *osb); -void ocfs2_unregister_net_handlers(struct ocfs2_super *osb); - -void ocfs2_remove_node_from_vote_queues(struct ocfs2_super *osb, - int node_num); -#endif diff --git a/fs/open.c b/fs/open.c index 4932b4d1da0..3fa4e4ffce4 100644 --- a/fs/open.c +++ b/fs/open.c @@ -127,10 +127,10 @@ asmlinkage long sys_statfs(const char __user * path, struct statfs __user * buf) error = user_path_walk(path, &nd); if (!error) { struct statfs tmp; - error = vfs_statfs_native(nd.dentry, &tmp); + error = vfs_statfs_native(nd.path.dentry, &tmp); if (!error && copy_to_user(buf, &tmp, sizeof(tmp))) error = -EFAULT; - path_release(&nd); + path_put(&nd.path); } return error; } @@ -146,10 +146,10 @@ asmlinkage long sys_statfs64(const char __user *path, size_t sz, struct statfs64 error = user_path_walk(path, &nd); if (!error) { struct statfs64 tmp; - error = vfs_statfs64(nd.dentry, &tmp); + error = vfs_statfs64(nd.path.dentry, &tmp); if (!error && copy_to_user(buf, &tmp, sizeof(tmp))) error = -EFAULT; - path_release(&nd); + path_put(&nd.path); } return error; } @@ -233,7 +233,7 @@ static long do_sys_truncate(const char __user * path, loff_t length) error = user_path_walk(path, &nd); if (error) goto out; - inode = nd.dentry->d_inode; + inode = nd.path.dentry->d_inode; /* For directories it's -EISDIR, for other non-regulars - -EINVAL */ error = -EISDIR; @@ -271,13 +271,13 @@ static long do_sys_truncate(const char __user * path, loff_t length) error = locks_verify_truncate(inode, NULL, length); if (!error) { DQUOT_INIT(inode); - error = do_truncate(nd.dentry, length, 0, NULL); + error = do_truncate(nd.path.dentry, length, 0, NULL); } put_write_and_out: put_write_access(inode); dput_and_out: - path_release(&nd); + path_put(&nd.path); out: return error; } @@ -335,7 +335,7 @@ asmlinkage long sys_ftruncate(unsigned int fd, unsigned long length) { long ret = do_sys_ftruncate(fd, length, 1); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(2, ret, fd, length); return ret; } @@ -350,7 +350,7 @@ asmlinkage long sys_ftruncate64(unsigned int fd, loff_t length) { long ret = do_sys_ftruncate(fd, length, 0); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(2, ret, fd, length); return ret; } #endif @@ -455,14 +455,14 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) res = vfs_permission(&nd, mode); /* SuS v2 requires we report a read only fs too */ if(res || !(mode & S_IWOTH) || - special_file(nd.dentry->d_inode->i_mode)) + special_file(nd.path.dentry->d_inode->i_mode)) goto out_path_release; - if(IS_RDONLY(nd.dentry->d_inode)) + if(IS_RDONLY(nd.path.dentry->d_inode)) res = -EROFS; out_path_release: - path_release(&nd); + path_put(&nd.path); out: current->fsuid = old_fsuid; current->fsgid = old_fsgid; @@ -490,10 +490,10 @@ asmlinkage long sys_chdir(const char __user * filename) if (error) goto dput_and_out; - set_fs_pwd(current->fs, nd.mnt, nd.dentry); + set_fs_pwd(current->fs, &nd.path); dput_and_out: - path_release(&nd); + path_put(&nd.path); out: return error; } @@ -501,9 +501,7 @@ out: asmlinkage long sys_fchdir(unsigned int fd) { struct file *file; - struct dentry *dentry; struct inode *inode; - struct vfsmount *mnt; int error; error = -EBADF; @@ -511,9 +509,7 @@ asmlinkage long sys_fchdir(unsigned int fd) if (!file) goto out; - dentry = file->f_path.dentry; - mnt = file->f_path.mnt; - inode = dentry->d_inode; + inode = file->f_path.dentry->d_inode; error = -ENOTDIR; if (!S_ISDIR(inode->i_mode)) @@ -521,7 +517,7 @@ asmlinkage long sys_fchdir(unsigned int fd) error = file_permission(file, MAY_EXEC); if (!error) - set_fs_pwd(current->fs, mnt, dentry); + set_fs_pwd(current->fs, &file->f_path); out_putf: fput(file); out: @@ -545,11 +541,11 @@ asmlinkage long sys_chroot(const char __user * filename) if (!capable(CAP_SYS_CHROOT)) goto dput_and_out; - set_fs_root(current->fs, nd.mnt, nd.dentry); + set_fs_root(current->fs, &nd.path); set_fs_altroot(); error = 0; dput_and_out: - path_release(&nd); + path_put(&nd.path); out: return error; } @@ -602,7 +598,7 @@ asmlinkage long sys_fchmodat(int dfd, const char __user *filename, error = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW, &nd); if (error) goto out; - inode = nd.dentry->d_inode; + inode = nd.path.dentry->d_inode; error = -EROFS; if (IS_RDONLY(inode)) @@ -617,11 +613,11 @@ asmlinkage long sys_fchmodat(int dfd, const char __user *filename, mode = inode->i_mode; newattrs.ia_mode = (mode & S_IALLUGO) | (inode->i_mode & ~S_IALLUGO); newattrs.ia_valid = ATTR_MODE | ATTR_CTIME; - error = notify_change(nd.dentry, &newattrs); + error = notify_change(nd.path.dentry, &newattrs); mutex_unlock(&inode->i_mutex); dput_and_out: - path_release(&nd); + path_put(&nd.path); out: return error; } @@ -675,8 +671,8 @@ asmlinkage long sys_chown(const char __user * filename, uid_t user, gid_t group) error = user_path_walk(filename, &nd); if (error) goto out; - error = chown_common(nd.dentry, user, group); - path_release(&nd); + error = chown_common(nd.path.dentry, user, group); + path_put(&nd.path); out: return error; } @@ -695,8 +691,8 @@ asmlinkage long sys_fchownat(int dfd, const char __user *filename, uid_t user, error = __user_walk_fd(dfd, filename, follow, &nd); if (error) goto out; - error = chown_common(nd.dentry, user, group); - path_release(&nd); + error = chown_common(nd.path.dentry, user, group); + path_put(&nd.path); out: return error; } @@ -709,8 +705,8 @@ asmlinkage long sys_lchown(const char __user * filename, uid_t user, gid_t group error = user_path_walk_link(filename, &nd); if (error) goto out; - error = chown_common(nd.dentry, user, group); - path_release(&nd); + error = chown_common(nd.path.dentry, user, group); + path_put(&nd.path); out: return error; } @@ -863,7 +859,7 @@ struct file *lookup_instantiate_filp(struct nameidata *nd, struct dentry *dentry goto out; if (IS_ERR(dentry)) goto out_err; - nd->intent.open.file = __dentry_open(dget(dentry), mntget(nd->mnt), + nd->intent.open.file = __dentry_open(dget(dentry), mntget(nd->path.mnt), nd->intent.open.flags - 1, nd->intent.open.file, open); @@ -891,9 +887,10 @@ struct file *nameidata_to_filp(struct nameidata *nd, int flags) filp = nd->intent.open.file; /* Has the filesystem initialised the file for us? */ if (filp->f_path.dentry == NULL) - filp = __dentry_open(nd->dentry, nd->mnt, flags, filp, NULL); + filp = __dentry_open(nd->path.dentry, nd->path.mnt, flags, filp, + NULL); else - path_release(nd); + path_put(&nd->path); return filp; } @@ -906,6 +903,18 @@ struct file *dentry_open(struct dentry *dentry, struct vfsmount *mnt, int flags) int error; struct file *f; + /* + * We must always pass in a valid mount pointer. Historically + * callers got away with not passing it, but we must enforce this at + * the earliest possible point now to avoid strange problems deep in the + * filesystem stack. + */ + if (!mnt) { + printk(KERN_WARNING "%s called with NULL vfsmount\n", __func__); + dump_stack(); + return ERR_PTR(-EINVAL); + } + error = -ENFILE; f = get_empty_filp(); if (f == NULL) { @@ -991,7 +1000,7 @@ static void __put_unused_fd(struct files_struct *files, unsigned int fd) files->next_fd = fd; } -void fastcall put_unused_fd(unsigned int fd) +void put_unused_fd(unsigned int fd) { struct files_struct *files = current->files; spin_lock(&files->file_lock); @@ -1014,7 +1023,7 @@ EXPORT_SYMBOL(put_unused_fd); * will follow. */ -void fastcall fd_install(unsigned int fd, struct file * file) +void fd_install(unsigned int fd, struct file *file) { struct files_struct *files = current->files; struct fdtable *fdt; @@ -1058,10 +1067,9 @@ asmlinkage long sys_open(const char __user *filename, int flags, int mode) ret = do_sys_open(AT_FDCWD, filename, flags, mode); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(3, ret, filename, flags, mode); return ret; } -EXPORT_UNUSED_SYMBOL_GPL(sys_open); /* To be deleted for 2.6.25 */ asmlinkage long sys_openat(int dfd, const char __user *filename, int flags, int mode) @@ -1073,7 +1081,7 @@ asmlinkage long sys_openat(int dfd, const char __user *filename, int flags, ret = do_sys_open(dfd, filename, flags, mode); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(4, ret, dfd, filename, flags, mode); return ret; } diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index 6b7ff161894..d17b4fd204e 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -38,6 +38,8 @@ struct op_inode_info { union op_inode_data u; }; +static struct inode *openprom_iget(struct super_block *sb, ino_t ino); + static inline struct op_inode_info *OP_I(struct inode *inode) { return container_of(inode, struct op_inode_info, vfs_inode); @@ -226,10 +228,10 @@ static struct dentry *openpromfs_lookup(struct inode *dir, struct dentry *dentry return ERR_PTR(-ENOENT); found: - inode = iget(dir->i_sb, ino); + inode = openprom_iget(dir->i_sb, ino); mutex_unlock(&op_mutex); - if (!inode) - return ERR_PTR(-EINVAL); + if (IS_ERR(inode)) + return ERR_CAST(inode); ent_oi = OP_I(inode); ent_oi->type = ent_type; ent_oi->u = ent_data; @@ -348,14 +350,23 @@ static void openprom_destroy_inode(struct inode *inode) kmem_cache_free(op_inode_cachep, OP_I(inode)); } -static void openprom_read_inode(struct inode * inode) +static struct inode *openprom_iget(struct super_block *sb, ino_t ino) { - inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; - if (inode->i_ino == OPENPROM_ROOT_INO) { - inode->i_op = &openprom_inode_operations; - inode->i_fop = &openprom_operations; - inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO; + struct inode *inode; + + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (inode->i_state & I_NEW) { + inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; + if (inode->i_ino == OPENPROM_ROOT_INO) { + inode->i_op = &openprom_inode_operations; + inode->i_fop = &openprom_operations; + inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO; + } + unlock_new_inode(inode); } + return inode; } static int openprom_remount(struct super_block *sb, int *flags, char *data) @@ -367,7 +378,6 @@ static int openprom_remount(struct super_block *sb, int *flags, char *data) static const struct super_operations openprom_sops = { .alloc_inode = openprom_alloc_inode, .destroy_inode = openprom_destroy_inode, - .read_inode = openprom_read_inode, .statfs = simple_statfs, .remount_fs = openprom_remount, }; @@ -376,6 +386,7 @@ static int openprom_fill_super(struct super_block *s, void *data, int silent) { struct inode *root_inode; struct op_inode_info *oi; + int ret; s->s_flags |= MS_NOATIME; s->s_blocksize = 1024; @@ -383,9 +394,11 @@ static int openprom_fill_super(struct super_block *s, void *data, int silent) s->s_magic = OPENPROM_SUPER_MAGIC; s->s_op = &openprom_sops; s->s_time_gran = 1; - root_inode = iget(s, OPENPROM_ROOT_INO); - if (!root_inode) + root_inode = openprom_iget(s, OPENPROM_ROOT_INO); + if (IS_ERR(root_inode)) { + ret = PTR_ERR(root_inode); goto out_no_root; + } oi = OP_I(root_inode); oi->type = op_inode_node; @@ -393,13 +406,15 @@ static int openprom_fill_super(struct super_block *s, void *data, int silent) s->s_root = d_alloc_root(root_inode); if (!s->s_root) - goto out_no_root; + goto out_no_root_dentry; return 0; +out_no_root_dentry: + iput(root_inode); + ret = -ENOMEM; out_no_root: printk("openprom_fill_super: get root inode failed\n"); - iput(root_inode); - return -ENOMEM; + return ret; } static int openprom_get_sb(struct file_system_type *fs_type, diff --git a/fs/partitions/Kconfig b/fs/partitions/Kconfig index a99acd8de35..cb5f0a3f1b0 100644 --- a/fs/partitions/Kconfig +++ b/fs/partitions/Kconfig @@ -198,7 +198,7 @@ config LDM_DEBUG config SGI_PARTITION bool "SGI partition support" if PARTITION_ADVANCED - default y if (SGI_IP22 || SGI_IP27 || ((MACH_JAZZ || SNI_RM) && !CPU_LITTLE_ENDIAN)) + default y if DEFAULT_SGI_PARTITION help Say Y here if you would like to be able to read the hard disk partition table format used by SGI machines. diff --git a/fs/partitions/check.c b/fs/partitions/check.c index 722e12e5acc..03f808c5b79 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -18,6 +18,7 @@ #include <linux/fs.h> #include <linux/kmod.h> #include <linux/ctype.h> +#include <linux/genhd.h> #include "check.h" @@ -195,96 +196,61 @@ check_partition(struct gendisk *hd, struct block_device *bdev) return ERR_PTR(res); } -/* - * sysfs bindings for partitions - */ - -struct part_attribute { - struct attribute attr; - ssize_t (*show)(struct hd_struct *,char *); - ssize_t (*store)(struct hd_struct *,const char *, size_t); -}; - -static ssize_t -part_attr_show(struct kobject * kobj, struct attribute * attr, char * page) +static ssize_t part_start_show(struct device *dev, + struct device_attribute *attr, char *buf) { - struct hd_struct * p = container_of(kobj,struct hd_struct,kobj); - struct part_attribute * part_attr = container_of(attr,struct part_attribute,attr); - ssize_t ret = 0; - if (part_attr->show) - ret = part_attr->show(p, page); - return ret; -} -static ssize_t -part_attr_store(struct kobject * kobj, struct attribute * attr, - const char *page, size_t count) -{ - struct hd_struct * p = container_of(kobj,struct hd_struct,kobj); - struct part_attribute * part_attr = container_of(attr,struct part_attribute,attr); - ssize_t ret = 0; + struct hd_struct *p = dev_to_part(dev); - if (part_attr->store) - ret = part_attr->store(p, page, count); - return ret; + return sprintf(buf, "%llu\n",(unsigned long long)p->start_sect); } -static struct sysfs_ops part_sysfs_ops = { - .show = part_attr_show, - .store = part_attr_store, -}; - -static ssize_t part_uevent_store(struct hd_struct * p, - const char *page, size_t count) +static ssize_t part_size_show(struct device *dev, + struct device_attribute *attr, char *buf) { - kobject_uevent(&p->kobj, KOBJ_ADD); - return count; + struct hd_struct *p = dev_to_part(dev); + return sprintf(buf, "%llu\n",(unsigned long long)p->nr_sects); } -static ssize_t part_dev_read(struct hd_struct * p, char *page) -{ - struct gendisk *disk = container_of(p->kobj.parent,struct gendisk,kobj); - dev_t dev = MKDEV(disk->major, disk->first_minor + p->partno); - return print_dev_t(page, dev); -} -static ssize_t part_start_read(struct hd_struct * p, char *page) -{ - return sprintf(page, "%llu\n",(unsigned long long)p->start_sect); -} -static ssize_t part_size_read(struct hd_struct * p, char *page) -{ - return sprintf(page, "%llu\n",(unsigned long long)p->nr_sects); -} -static ssize_t part_stat_read(struct hd_struct * p, char *page) + +static ssize_t part_stat_show(struct device *dev, + struct device_attribute *attr, char *buf) { - return sprintf(page, "%8u %8llu %8u %8llu\n", - p->ios[0], (unsigned long long)p->sectors[0], - p->ios[1], (unsigned long long)p->sectors[1]); + struct hd_struct *p = dev_to_part(dev); + + preempt_disable(); + part_round_stats(p); + preempt_enable(); + return sprintf(buf, + "%8lu %8lu %8llu %8u " + "%8lu %8lu %8llu %8u " + "%8u %8u %8u" + "\n", + part_stat_read(p, ios[READ]), + part_stat_read(p, merges[READ]), + (unsigned long long)part_stat_read(p, sectors[READ]), + jiffies_to_msecs(part_stat_read(p, ticks[READ])), + part_stat_read(p, ios[WRITE]), + part_stat_read(p, merges[WRITE]), + (unsigned long long)part_stat_read(p, sectors[WRITE]), + jiffies_to_msecs(part_stat_read(p, ticks[WRITE])), + p->in_flight, + jiffies_to_msecs(part_stat_read(p, io_ticks)), + jiffies_to_msecs(part_stat_read(p, time_in_queue))); } -static struct part_attribute part_attr_uevent = { - .attr = {.name = "uevent", .mode = S_IWUSR }, - .store = part_uevent_store -}; -static struct part_attribute part_attr_dev = { - .attr = {.name = "dev", .mode = S_IRUGO }, - .show = part_dev_read -}; -static struct part_attribute part_attr_start = { - .attr = {.name = "start", .mode = S_IRUGO }, - .show = part_start_read -}; -static struct part_attribute part_attr_size = { - .attr = {.name = "size", .mode = S_IRUGO }, - .show = part_size_read -}; -static struct part_attribute part_attr_stat = { - .attr = {.name = "stat", .mode = S_IRUGO }, - .show = part_stat_read -}; #ifdef CONFIG_FAIL_MAKE_REQUEST +static ssize_t part_fail_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct hd_struct *p = dev_to_part(dev); + + return sprintf(buf, "%d\n", p->make_it_fail); +} -static ssize_t part_fail_store(struct hd_struct * p, +static ssize_t part_fail_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) { + struct hd_struct *p = dev_to_part(dev); int i; if (count > 0 && sscanf(buf, "%d", &i) > 0) @@ -292,50 +258,54 @@ static ssize_t part_fail_store(struct hd_struct * p, return count; } -static ssize_t part_fail_read(struct hd_struct * p, char *page) -{ - return sprintf(page, "%d\n", p->make_it_fail); -} -static struct part_attribute part_attr_fail = { - .attr = {.name = "make-it-fail", .mode = S_IRUGO | S_IWUSR }, - .store = part_fail_store, - .show = part_fail_read -}; +#endif +static DEVICE_ATTR(start, S_IRUGO, part_start_show, NULL); +static DEVICE_ATTR(size, S_IRUGO, part_size_show, NULL); +static DEVICE_ATTR(stat, S_IRUGO, part_stat_show, NULL); +#ifdef CONFIG_FAIL_MAKE_REQUEST +static struct device_attribute dev_attr_fail = + __ATTR(make-it-fail, S_IRUGO|S_IWUSR, part_fail_show, part_fail_store); #endif -static struct attribute * default_attrs[] = { - &part_attr_uevent.attr, - &part_attr_dev.attr, - &part_attr_start.attr, - &part_attr_size.attr, - &part_attr_stat.attr, +static struct attribute *part_attrs[] = { + &dev_attr_start.attr, + &dev_attr_size.attr, + &dev_attr_stat.attr, #ifdef CONFIG_FAIL_MAKE_REQUEST - &part_attr_fail.attr, + &dev_attr_fail.attr, #endif - NULL, + NULL }; -extern struct kset block_subsys; +static struct attribute_group part_attr_group = { + .attrs = part_attrs, +}; -static void part_release(struct kobject *kobj) +static struct attribute_group *part_attr_groups[] = { + &part_attr_group, + NULL +}; + +static void part_release(struct device *dev) { - struct hd_struct * p = container_of(kobj,struct hd_struct,kobj); + struct hd_struct *p = dev_to_part(dev); + free_part_stats(p); kfree(p); } -struct kobj_type ktype_part = { +struct device_type part_type = { + .name = "partition", + .groups = part_attr_groups, .release = part_release, - .default_attrs = default_attrs, - .sysfs_ops = &part_sysfs_ops, }; static inline void partition_sysfs_add_subdir(struct hd_struct *p) { struct kobject *k; - k = kobject_get(&p->kobj); - p->holder_dir = kobject_add_dir(k, "holders"); + k = kobject_get(&p->dev.kobj); + p->holder_dir = kobject_create_and_add("holders", k); kobject_put(k); } @@ -343,15 +313,16 @@ static inline void disk_sysfs_add_subdirs(struct gendisk *disk) { struct kobject *k; - k = kobject_get(&disk->kobj); - disk->holder_dir = kobject_add_dir(k, "holders"); - disk->slave_dir = kobject_add_dir(k, "slaves"); + k = kobject_get(&disk->dev.kobj); + disk->holder_dir = kobject_create_and_add("holders", k); + disk->slave_dir = kobject_create_and_add("slaves", k); kobject_put(k); } void delete_partition(struct gendisk *disk, int part) { struct hd_struct *p = disk->part[part-1]; + if (!p) return; if (!p->nr_sects) @@ -359,115 +330,63 @@ void delete_partition(struct gendisk *disk, int part) disk->part[part-1] = NULL; p->start_sect = 0; p->nr_sects = 0; - p->ios[0] = p->ios[1] = 0; - p->sectors[0] = p->sectors[1] = 0; - sysfs_remove_link(&p->kobj, "subsystem"); - kobject_unregister(p->holder_dir); - kobject_uevent(&p->kobj, KOBJ_REMOVE); - kobject_del(&p->kobj); - kobject_put(&p->kobj); + part_stat_set_all(p, 0); + kobject_put(p->holder_dir); + device_del(&p->dev); + put_device(&p->dev); +} + +static ssize_t whole_disk_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return 0; } +static DEVICE_ATTR(whole_disk, S_IRUSR | S_IRGRP | S_IROTH, + whole_disk_show, NULL); void add_partition(struct gendisk *disk, int part, sector_t start, sector_t len, int flags) { struct hd_struct *p; + int err; p = kzalloc(sizeof(*p), GFP_KERNEL); if (!p) return; - + + if (!init_part_stats(p)) { + kfree(p); + return; + } p->start_sect = start; p->nr_sects = len; p->partno = part; p->policy = disk->policy; - if (isdigit(disk->kobj.k_name[strlen(disk->kobj.k_name)-1])) - kobject_set_name(&p->kobj, "%sp%d", - kobject_name(&disk->kobj), part); + if (isdigit(disk->dev.bus_id[strlen(disk->dev.bus_id)-1])) + snprintf(p->dev.bus_id, BUS_ID_SIZE, + "%sp%d", disk->dev.bus_id, part); else - kobject_set_name(&p->kobj, "%s%d", - kobject_name(&disk->kobj),part); - p->kobj.parent = &disk->kobj; - p->kobj.ktype = &ktype_part; - kobject_init(&p->kobj); - kobject_add(&p->kobj); - if (!disk->part_uevent_suppress) - kobject_uevent(&p->kobj, KOBJ_ADD); - sysfs_create_link(&p->kobj, &block_subsys.kobj, "subsystem"); - if (flags & ADDPART_FLAG_WHOLEDISK) { - static struct attribute addpartattr = { - .name = "whole_disk", - .mode = S_IRUSR | S_IRGRP | S_IROTH, - }; - - sysfs_create_file(&p->kobj, &addpartattr); - } - partition_sysfs_add_subdir(p); + snprintf(p->dev.bus_id, BUS_ID_SIZE, + "%s%d", disk->dev.bus_id, part); + + device_initialize(&p->dev); + p->dev.devt = MKDEV(disk->major, disk->first_minor + part); + p->dev.class = &block_class; + p->dev.type = &part_type; + p->dev.parent = &disk->dev; disk->part[part-1] = p; -} - -static char *make_block_name(struct gendisk *disk) -{ - char *name; - static char *block_str = "block:"; - int size; - char *s; - - size = strlen(block_str) + strlen(disk->disk_name) + 1; - name = kmalloc(size, GFP_KERNEL); - if (!name) - return NULL; - strcpy(name, block_str); - strcat(name, disk->disk_name); - /* ewww... some of these buggers have / in name... */ - s = strchr(name, '/'); - if (s) - *s = '!'; - return name; -} - -static int disk_sysfs_symlinks(struct gendisk *disk) -{ - struct device *target = get_device(disk->driverfs_dev); - int err; - char *disk_name = NULL; - - if (target) { - disk_name = make_block_name(disk); - if (!disk_name) { - err = -ENOMEM; - goto err_out; - } - - err = sysfs_create_link(&disk->kobj, &target->kobj, "device"); - if (err) - goto err_out_disk_name; - - err = sysfs_create_link(&target->kobj, &disk->kobj, disk_name); - if (err) - goto err_out_dev_link; - } - - err = sysfs_create_link(&disk->kobj, &block_subsys.kobj, - "subsystem"); - if (err) - goto err_out_disk_name_lnk; - kfree(disk_name); - - return 0; + /* delay uevent until 'holders' subdir is created */ + p->dev.uevent_suppress = 1; + device_add(&p->dev); + partition_sysfs_add_subdir(p); + p->dev.uevent_suppress = 0; + if (flags & ADDPART_FLAG_WHOLEDISK) + err = device_create_file(&p->dev, &dev_attr_whole_disk); -err_out_disk_name_lnk: - if (target) { - sysfs_remove_link(&target->kobj, disk_name); -err_out_dev_link: - sysfs_remove_link(&disk->kobj, "device"); -err_out_disk_name: - kfree(disk_name); -err_out: - put_device(target); - } - return err; + /* suppress uevent if the disk supresses it */ + if (!disk->dev.uevent_suppress) + kobject_uevent(&p->dev.kobj, KOBJ_ADD); } /* Not exported, helper to add_disk(). */ @@ -479,19 +398,29 @@ void register_disk(struct gendisk *disk) struct hd_struct *p; int err; - kobject_set_name(&disk->kobj, "%s", disk->disk_name); - /* ewww... some of these buggers have / in name... */ - s = strchr(disk->kobj.k_name, '/'); + disk->dev.parent = disk->driverfs_dev; + disk->dev.devt = MKDEV(disk->major, disk->first_minor); + + strlcpy(disk->dev.bus_id, disk->disk_name, KOBJ_NAME_LEN); + /* ewww... some of these buggers have / in the name... */ + s = strchr(disk->dev.bus_id, '/'); if (s) *s = '!'; - if ((err = kobject_add(&disk->kobj))) + + /* delay uevents, until we scanned partition table */ + disk->dev.uevent_suppress = 1; + + if (device_add(&disk->dev)) return; - err = disk_sysfs_symlinks(disk); +#ifndef CONFIG_SYSFS_DEPRECATED + err = sysfs_create_link(block_depr, &disk->dev.kobj, + kobject_name(&disk->dev.kobj)); if (err) { - kobject_del(&disk->kobj); + device_del(&disk->dev); return; } - disk_sysfs_add_subdirs(disk); +#endif + disk_sysfs_add_subdirs(disk); /* No minors to use for partitions */ if (disk->minors == 1) @@ -505,25 +434,23 @@ void register_disk(struct gendisk *disk) if (!bdev) goto exit; - /* scan partition table, but suppress uevents */ bdev->bd_invalidated = 1; - disk->part_uevent_suppress = 1; err = blkdev_get(bdev, FMODE_READ, 0); - disk->part_uevent_suppress = 0; if (err < 0) goto exit; blkdev_put(bdev); exit: - /* announce disk after possible partitions are already created */ - kobject_uevent(&disk->kobj, KOBJ_ADD); + /* announce disk after possible partitions are created */ + disk->dev.uevent_suppress = 0; + kobject_uevent(&disk->dev.kobj, KOBJ_ADD); /* announce possible partitions */ for (i = 1; i < disk->minors; i++) { p = disk->part[i-1]; if (!p || !p->nr_sects) continue; - kobject_uevent(&p->kobj, KOBJ_ADD); + kobject_uevent(&p->dev.kobj, KOBJ_ADD); } } @@ -602,19 +529,11 @@ void del_gendisk(struct gendisk *disk) disk_stat_set_all(disk, 0); disk->stamp = 0; - kobject_uevent(&disk->kobj, KOBJ_REMOVE); - kobject_unregister(disk->holder_dir); - kobject_unregister(disk->slave_dir); - if (disk->driverfs_dev) { - char *disk_name = make_block_name(disk); - sysfs_remove_link(&disk->kobj, "device"); - if (disk_name) { - sysfs_remove_link(&disk->driverfs_dev->kobj, disk_name); - kfree(disk_name); - } - put_device(disk->driverfs_dev); - disk->driverfs_dev = NULL; - } - sysfs_remove_link(&disk->kobj, "subsystem"); - kobject_del(&disk->kobj); + kobject_put(disk->holder_dir); + kobject_put(disk->slave_dir); + disk->driverfs_dev = NULL; +#ifndef CONFIG_SYSFS_DEPRECATED + sysfs_remove_link(block_depr, disk->dev.bus_id); +#endif + device_del(&disk->dev); } diff --git a/fs/pipe.c b/fs/pipe.c index e66ec48e95d..8be381bbcb5 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -171,7 +171,7 @@ static void anon_pipe_buf_release(struct pipe_inode_info *pipe, * * Description: * This function returns a kernel virtual address mapping for the - * passed in @pipe_buffer. If @atomic is set, an atomic map is provided + * pipe_buffer passed in @buf. If @atomic is set, an atomic map is provided * and the caller has to be careful not to fault before calling * the unmap function. * @@ -208,15 +208,15 @@ void generic_pipe_buf_unmap(struct pipe_inode_info *pipe, } /** - * generic_pipe_buf_steal - attempt to take ownership of a @pipe_buffer + * generic_pipe_buf_steal - attempt to take ownership of a &pipe_buffer * @pipe: the pipe that the buffer belongs to * @buf: the buffer to attempt to steal * * Description: - * This function attempts to steal the @struct page attached to + * This function attempts to steal the &struct page attached to * @buf. If successful, this function returns 0 and returns with * the page locked. The caller may then reuse the page for whatever - * he wishes, the typical use is insertion into a different file + * he wishes; the typical use is insertion into a different file * page cache. */ int generic_pipe_buf_steal(struct pipe_inode_info *pipe, @@ -238,7 +238,7 @@ int generic_pipe_buf_steal(struct pipe_inode_info *pipe, } /** - * generic_pipe_buf_get - get a reference to a @struct pipe_buffer + * generic_pipe_buf_get - get a reference to a &struct pipe_buffer * @pipe: the pipe that the buffer belongs to * @buf: the buffer to get a reference to * @@ -576,9 +576,7 @@ bad_pipe_w(struct file *filp, const char __user *buf, size_t count, return -EBADF; } -static int -pipe_ioctl(struct inode *pino, struct file *filp, - unsigned int cmd, unsigned long arg) +static long pipe_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { struct inode *inode = filp->f_path.dentry->d_inode; struct pipe_inode_info *pipe; @@ -785,7 +783,7 @@ const struct file_operations read_fifo_fops = { .aio_read = pipe_read, .write = bad_pipe_w, .poll = pipe_poll, - .ioctl = pipe_ioctl, + .unlocked_ioctl = pipe_ioctl, .open = pipe_read_open, .release = pipe_read_release, .fasync = pipe_read_fasync, @@ -797,7 +795,7 @@ const struct file_operations write_fifo_fops = { .write = do_sync_write, .aio_write = pipe_write, .poll = pipe_poll, - .ioctl = pipe_ioctl, + .unlocked_ioctl = pipe_ioctl, .open = pipe_write_open, .release = pipe_write_release, .fasync = pipe_write_fasync, @@ -810,7 +808,7 @@ const struct file_operations rdwr_fifo_fops = { .write = do_sync_write, .aio_write = pipe_write, .poll = pipe_poll, - .ioctl = pipe_ioctl, + .unlocked_ioctl = pipe_ioctl, .open = pipe_rdwr_open, .release = pipe_rdwr_release, .fasync = pipe_rdwr_fasync, @@ -822,7 +820,7 @@ static const struct file_operations read_pipe_fops = { .aio_read = pipe_read, .write = bad_pipe_w, .poll = pipe_poll, - .ioctl = pipe_ioctl, + .unlocked_ioctl = pipe_ioctl, .open = pipe_read_open, .release = pipe_read_release, .fasync = pipe_read_fasync, @@ -834,7 +832,7 @@ static const struct file_operations write_pipe_fops = { .write = do_sync_write, .aio_write = pipe_write, .poll = pipe_poll, - .ioctl = pipe_ioctl, + .unlocked_ioctl = pipe_ioctl, .open = pipe_write_open, .release = pipe_write_release, .fasync = pipe_write_fasync, @@ -847,7 +845,7 @@ static const struct file_operations rdwr_pipe_fops = { .write = do_sync_write, .aio_write = pipe_write, .poll = pipe_poll, - .ioctl = pipe_ioctl, + .unlocked_ioctl = pipe_ioctl, .open = pipe_rdwr_open, .release = pipe_rdwr_release, .fasync = pipe_rdwr_fasync, @@ -959,13 +957,10 @@ struct file *create_write_pipe(void) struct dentry *dentry; struct qstr name = { .name = "" }; - f = get_empty_filp(); - if (!f) - return ERR_PTR(-ENFILE); err = -ENFILE; inode = get_pipe_inode(); if (!inode) - goto err_file; + goto err; err = -ENOMEM; dentry = d_alloc(pipe_mnt->mnt_sb->s_root, &name); @@ -980,22 +975,24 @@ struct file *create_write_pipe(void) */ dentry->d_flags &= ~DCACHE_UNHASHED; d_instantiate(dentry, inode); - f->f_path.mnt = mntget(pipe_mnt); - f->f_path.dentry = dentry; + + err = -ENFILE; + f = alloc_file(pipe_mnt, dentry, FMODE_WRITE, &write_pipe_fops); + if (!f) + goto err_dentry; f->f_mapping = inode->i_mapping; f->f_flags = O_WRONLY; - f->f_op = &write_pipe_fops; - f->f_mode = FMODE_WRITE; f->f_version = 0; return f; + err_dentry: + dput(dentry); err_inode: free_pipe_info(inode); iput(inode); - err_file: - put_filp(f); + err: return ERR_PTR(err); } diff --git a/fs/pnode.c b/fs/pnode.c index 89940f243fc..1d8f5447f3f 100644 --- a/fs/pnode.c +++ b/fs/pnode.c @@ -83,6 +83,8 @@ void change_mnt_propagation(struct vfsmount *mnt, int type) mnt->mnt_master = NULL; if (type == MS_UNBINDABLE) mnt->mnt_flags |= MNT_UNBINDABLE; + else + mnt->mnt_flags &= ~MNT_UNBINDABLE; } } @@ -223,7 +225,7 @@ out: */ static inline int do_refcount_check(struct vfsmount *mnt, int count) { - int mycount = atomic_read(&mnt->mnt_count); + int mycount = atomic_read(&mnt->mnt_count) - mnt->mnt_ghosts; return (mycount > count); } diff --git a/fs/proc/array.c b/fs/proc/array.c index eb97f2897e2..07d6c4853fe 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -77,6 +77,7 @@ #include <linux/cpuset.h> #include <linux/rcupdate.h> #include <linux/delayacct.h> +#include <linux/seq_file.h> #include <linux/pid_namespace.h> #include <asm/pgtable.h> @@ -88,18 +89,21 @@ do { memcpy(buffer, string, strlen(string)); \ buffer += strlen(string); } while (0) -static inline char *task_name(struct task_struct *p, char *buf) +static inline void task_name(struct seq_file *m, struct task_struct *p) { int i; + char *buf, *end; char *name; char tcomm[sizeof(p->comm)]; get_task_comm(tcomm, p); - ADDBUF(buf, "Name:\t"); + seq_printf(m, "Name:\t"); + end = m->buf + m->size; + buf = m->buf + m->count; name = tcomm; i = sizeof(tcomm); - do { + while (i && (buf < end)) { unsigned char c = *name; name++; i--; @@ -107,20 +111,21 @@ static inline char *task_name(struct task_struct *p, char *buf) if (!c) break; if (c == '\\') { - buf[1] = c; - buf += 2; + buf++; + if (buf < end) + *buf++ = c; continue; } if (c == '\n') { - buf[0] = '\\'; - buf[1] = 'n'; - buf += 2; + *buf++ = '\\'; + if (buf < end) + *buf++ = 'n'; continue; } buf++; - } while (i); - *buf = '\n'; - return buf+1; + } + m->count = buf - m->buf; + seq_printf(m, "\n"); } /* @@ -141,12 +146,7 @@ static const char *task_state_array[] = { static inline const char *get_task_state(struct task_struct *tsk) { - unsigned int state = (tsk->state & (TASK_RUNNING | - TASK_INTERRUPTIBLE | - TASK_UNINTERRUPTIBLE | - TASK_STOPPED | - TASK_TRACED)) | - tsk->exit_state; + unsigned int state = (tsk->state & TASK_REPORT) | tsk->exit_state; const char **p = &task_state_array[0]; while (state) { @@ -156,21 +156,20 @@ static inline const char *get_task_state(struct task_struct *tsk) return *p; } -static inline char *task_state(struct task_struct *p, char *buffer) +static inline void task_state(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *p) { struct group_info *group_info; int g; struct fdtable *fdt = NULL; - struct pid_namespace *ns; pid_t ppid, tpid; - ns = current->nsproxy->pid_ns; rcu_read_lock(); ppid = pid_alive(p) ? task_tgid_nr_ns(rcu_dereference(p->real_parent), ns) : 0; tpid = pid_alive(p) && p->ptrace ? task_pid_nr_ns(rcu_dereference(p->parent), ns) : 0; - buffer += sprintf(buffer, + seq_printf(m, "State:\t%s\n" "Tgid:\t%d\n" "Pid:\t%d\n" @@ -180,7 +179,7 @@ static inline char *task_state(struct task_struct *p, char *buffer) "Gid:\t%d\t%d\t%d\t%d\n", get_task_state(p), task_tgid_nr_ns(p, ns), - task_pid_nr_ns(p, ns), + pid_nr_ns(pid, ns), ppid, tpid, p->uid, p->euid, p->suid, p->fsuid, p->gid, p->egid, p->sgid, p->fsgid); @@ -188,7 +187,7 @@ static inline char *task_state(struct task_struct *p, char *buffer) task_lock(p); if (p->files) fdt = files_fdtable(p->files); - buffer += sprintf(buffer, + seq_printf(m, "FDSize:\t%d\n" "Groups:\t", fdt ? fdt->max_fds : 0); @@ -199,20 +198,18 @@ static inline char *task_state(struct task_struct *p, char *buffer) task_unlock(p); for (g = 0; g < min(group_info->ngroups, NGROUPS_SMALL); g++) - buffer += sprintf(buffer, "%d ", GROUP_AT(group_info, g)); + seq_printf(m, "%d ", GROUP_AT(group_info, g)); put_group_info(group_info); - buffer += sprintf(buffer, "\n"); - return buffer; + seq_printf(m, "\n"); } -static char *render_sigset_t(const char *header, sigset_t *set, char *buffer) +static void render_sigset_t(struct seq_file *m, const char *header, + sigset_t *set) { - int i, len; + int i; - len = strlen(header); - memcpy(buffer, header, len); - buffer += len; + seq_printf(m, "%s", header); i = _NSIG; do { @@ -223,12 +220,10 @@ static char *render_sigset_t(const char *header, sigset_t *set, char *buffer) if (sigismember(set, i+2)) x |= 2; if (sigismember(set, i+3)) x |= 4; if (sigismember(set, i+4)) x |= 8; - *buffer++ = (x < 10 ? '0' : 'a' - 10) + x; + seq_printf(m, "%x", x); } while (i >= 4); - *buffer++ = '\n'; - *buffer = 0; - return buffer; + seq_printf(m, "\n"); } static void collect_sigign_sigcatch(struct task_struct *p, sigset_t *ign, @@ -246,7 +241,7 @@ static void collect_sigign_sigcatch(struct task_struct *p, sigset_t *ign, } } -static inline char *task_sig(struct task_struct *p, char *buffer) +static inline void task_sig(struct seq_file *m, struct task_struct *p) { unsigned long flags; sigset_t pending, shpending, blocked, ignored, caught; @@ -273,58 +268,66 @@ static inline char *task_sig(struct task_struct *p, char *buffer) } rcu_read_unlock(); - buffer += sprintf(buffer, "Threads:\t%d\n", num_threads); - buffer += sprintf(buffer, "SigQ:\t%lu/%lu\n", qsize, qlim); + seq_printf(m, "Threads:\t%d\n", num_threads); + seq_printf(m, "SigQ:\t%lu/%lu\n", qsize, qlim); /* render them all */ - buffer = render_sigset_t("SigPnd:\t", &pending, buffer); - buffer = render_sigset_t("ShdPnd:\t", &shpending, buffer); - buffer = render_sigset_t("SigBlk:\t", &blocked, buffer); - buffer = render_sigset_t("SigIgn:\t", &ignored, buffer); - buffer = render_sigset_t("SigCgt:\t", &caught, buffer); + render_sigset_t(m, "SigPnd:\t", &pending); + render_sigset_t(m, "ShdPnd:\t", &shpending); + render_sigset_t(m, "SigBlk:\t", &blocked); + render_sigset_t(m, "SigIgn:\t", &ignored); + render_sigset_t(m, "SigCgt:\t", &caught); +} - return buffer; +static void render_cap_t(struct seq_file *m, const char *header, + kernel_cap_t *a) +{ + unsigned __capi; + + seq_printf(m, "%s", header); + CAP_FOR_EACH_U32(__capi) { + seq_printf(m, "%08x", + a->cap[(_LINUX_CAPABILITY_U32S-1) - __capi]); + } + seq_printf(m, "\n"); } -static inline char *task_cap(struct task_struct *p, char *buffer) +static inline void task_cap(struct seq_file *m, struct task_struct *p) { - return buffer + sprintf(buffer, "CapInh:\t%016x\n" - "CapPrm:\t%016x\n" - "CapEff:\t%016x\n", - cap_t(p->cap_inheritable), - cap_t(p->cap_permitted), - cap_t(p->cap_effective)); + render_cap_t(m, "CapInh:\t", &p->cap_inheritable); + render_cap_t(m, "CapPrm:\t", &p->cap_permitted); + render_cap_t(m, "CapEff:\t", &p->cap_effective); } -static inline char *task_context_switch_counts(struct task_struct *p, - char *buffer) +static inline void task_context_switch_counts(struct seq_file *m, + struct task_struct *p) { - return buffer + sprintf(buffer, "voluntary_ctxt_switches:\t%lu\n" - "nonvoluntary_ctxt_switches:\t%lu\n", - p->nvcsw, - p->nivcsw); + seq_printf(m, "voluntary_ctxt_switches:\t%lu\n" + "nonvoluntary_ctxt_switches:\t%lu\n", + p->nvcsw, + p->nivcsw); } -int proc_pid_status(struct task_struct *task, char *buffer) +int proc_pid_status(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *task) { - char *orig = buffer; struct mm_struct *mm = get_task_mm(task); - buffer = task_name(task, buffer); - buffer = task_state(task, buffer); + task_name(m, task); + task_state(m, ns, pid, task); if (mm) { - buffer = task_mem(mm, buffer); + task_mem(m, mm); mmput(mm); } - buffer = task_sig(task, buffer); - buffer = task_cap(task, buffer); - buffer = cpuset_task_status_allowed(task, buffer); + task_sig(m, task); + task_cap(m, task); + cpuset_task_status_allowed(m, task); #if defined(CONFIG_S390) - buffer = task_show_regs(task, buffer); + task_show_regs(m, task); #endif - buffer = task_context_switch_counts(task, buffer); - return buffer - orig; + task_context_switch_counts(m, task); + return 0; } /* @@ -386,14 +389,14 @@ static cputime_t task_gtime(struct task_struct *p) return p->gtime; } -static int do_task_stat(struct task_struct *task, char *buffer, int whole) +static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *task, int whole) { unsigned long vsize, eip, esp, wchan = ~0UL; long priority, nice; int tty_pgrp = -1, tty_nr = 0; sigset_t sigign, sigcatch; char state; - int res; pid_t ppid = 0, pgid = -1, sid = -1; int num_threads = 0; struct mm_struct *mm; @@ -405,9 +408,6 @@ static int do_task_stat(struct task_struct *task, char *buffer, int whole) unsigned long rsslim = 0; char tcomm[sizeof(task->comm)]; unsigned long flags; - struct pid_namespace *ns; - - ns = current->nsproxy->pid_ns; state = *get_task_state(task); vsize = eip = esp = 0; @@ -494,10 +494,10 @@ static int do_task_stat(struct task_struct *task, char *buffer, int whole) /* convert nsec -> ticks */ start_time = nsec_to_clock_t(start_time); - res = sprintf(buffer, "%d (%s) %c %d %d %d %d %d %u %lu \ + seq_printf(m, "%d (%s) %c %d %d %d %d %d %u %lu \ %lu %lu %lu %lu %lu %ld %ld %ld %ld %d 0 %llu %lu %ld %lu %lu %lu %lu %lu \ %lu %lu %lu %lu %lu %lu %lu %lu %d %d %u %u %llu %lu %ld\n", - task_pid_nr_ns(task, ns), + pid_nr_ns(pid, ns), tcomm, state, ppid, @@ -546,20 +546,23 @@ static int do_task_stat(struct task_struct *task, char *buffer, int whole) cputime_to_clock_t(cgtime)); if (mm) mmput(mm); - return res; + return 0; } -int proc_tid_stat(struct task_struct *task, char *buffer) +int proc_tid_stat(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *task) { - return do_task_stat(task, buffer, 0); + return do_task_stat(m, ns, pid, task, 0); } -int proc_tgid_stat(struct task_struct *task, char *buffer) +int proc_tgid_stat(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *task) { - return do_task_stat(task, buffer, 1); + return do_task_stat(m, ns, pid, task, 1); } -int proc_pid_statm(struct task_struct *task, char *buffer) +int proc_pid_statm(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *task) { int size = 0, resident = 0, shared = 0, text = 0, lib = 0, data = 0; struct mm_struct *mm = get_task_mm(task); @@ -568,7 +571,8 @@ int proc_pid_statm(struct task_struct *task, char *buffer) size = task_statm(mm, &shared, &text, &data, &resident); mmput(mm); } + seq_printf(m, "%d %d %d %d %d %d %d\n", + size, resident, shared, text, lib, data, 0); - return sprintf(buffer, "%d %d %d %d %d %d %d\n", - size, resident, shared, text, lib, data, 0); + return 0; } diff --git a/fs/proc/base.c b/fs/proc/base.c index 7411bfb0b7c..81d7d145292 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -88,10 +88,6 @@ * in /proc for a task before it execs a suid executable. */ - -/* Worst case buffer size needed for holding an integer. */ -#define PROC_NUMBUF 13 - struct pid_entry { char *name; int len; @@ -125,6 +121,10 @@ struct pid_entry { NOD(NAME, (S_IFREG|(MODE)), \ NULL, &proc_info_file_operations, \ { .proc_read = &proc_##OTYPE } ) +#define ONE(NAME, MODE, OTYPE) \ + NOD(NAME, (S_IFREG|(MODE)), \ + NULL, &proc_single_file_operations, \ + { .proc_show = &proc_##OTYPE } ) int maps_protect; EXPORT_SYMBOL(maps_protect); @@ -153,7 +153,7 @@ static int get_nr_threads(struct task_struct *tsk) return count; } -static int proc_cwd_link(struct inode *inode, struct dentry **dentry, struct vfsmount **mnt) +static int proc_cwd_link(struct inode *inode, struct path *path) { struct task_struct *task = get_proc_task(inode); struct fs_struct *fs = NULL; @@ -165,8 +165,8 @@ static int proc_cwd_link(struct inode *inode, struct dentry **dentry, struct vfs } if (fs) { read_lock(&fs->lock); - *mnt = mntget(fs->pwdmnt); - *dentry = dget(fs->pwd); + *path = fs->pwd; + path_get(&fs->pwd); read_unlock(&fs->lock); result = 0; put_fs_struct(fs); @@ -174,7 +174,7 @@ static int proc_cwd_link(struct inode *inode, struct dentry **dentry, struct vfs return result; } -static int proc_root_link(struct inode *inode, struct dentry **dentry, struct vfsmount **mnt) +static int proc_root_link(struct inode *inode, struct path *path) { struct task_struct *task = get_proc_task(inode); struct fs_struct *fs = NULL; @@ -186,8 +186,8 @@ static int proc_root_link(struct inode *inode, struct dentry **dentry, struct vf } if (fs) { read_lock(&fs->lock); - *mnt = mntget(fs->rootmnt); - *dentry = dget(fs->root); + *path = fs->root; + path_get(&fs->root); read_unlock(&fs->lock); result = 0; put_fs_struct(fs); @@ -199,7 +199,7 @@ static int proc_root_link(struct inode *inode, struct dentry **dentry, struct vf (task == current || \ (task->parent == current && \ (task->ptrace & PT_PTRACED) && \ - (task->state == TASK_STOPPED || task->state == TASK_TRACED) && \ + (task_is_stopped_or_traced(task)) && \ security_ptrace(current,task) == 0)) struct mm_struct *mm_for_maps(struct task_struct *task) @@ -310,6 +310,72 @@ static int proc_pid_schedstat(struct task_struct *task, char *buffer) } #endif +#ifdef CONFIG_LATENCYTOP +static int lstats_show_proc(struct seq_file *m, void *v) +{ + int i; + struct inode *inode = m->private; + struct task_struct *task = get_proc_task(inode); + + if (!task) + return -ESRCH; + seq_puts(m, "Latency Top version : v0.1\n"); + for (i = 0; i < 32; i++) { + if (task->latency_record[i].backtrace[0]) { + int q; + seq_printf(m, "%i %li %li ", + task->latency_record[i].count, + task->latency_record[i].time, + task->latency_record[i].max); + for (q = 0; q < LT_BACKTRACEDEPTH; q++) { + char sym[KSYM_NAME_LEN]; + char *c; + if (!task->latency_record[i].backtrace[q]) + break; + if (task->latency_record[i].backtrace[q] == ULONG_MAX) + break; + sprint_symbol(sym, task->latency_record[i].backtrace[q]); + c = strchr(sym, '+'); + if (c) + *c = 0; + seq_printf(m, "%s ", sym); + } + seq_printf(m, "\n"); + } + + } + put_task_struct(task); + return 0; +} + +static int lstats_open(struct inode *inode, struct file *file) +{ + return single_open(file, lstats_show_proc, inode); +} + +static ssize_t lstats_write(struct file *file, const char __user *buf, + size_t count, loff_t *offs) +{ + struct task_struct *task = get_proc_task(file->f_dentry->d_inode); + + if (!task) + return -ESRCH; + clear_all_latency_tracing(task); + put_task_struct(task); + + return count; +} + +static const struct file_operations proc_lstats_operations = { + .open = lstats_open, + .read = seq_read, + .write = lstats_write, + .llseek = seq_lseek, + .release = single_release, +}; + +#endif + /* The badness from the OOM killer */ unsigned long badness(struct task_struct *p, unsigned long uptime); static int proc_oom_score(struct task_struct *task, char *buffer) @@ -345,6 +411,7 @@ static const struct limit_names lnames[RLIM_NLIMITS] = { [RLIMIT_MSGQUEUE] = {"Max msgqueue size", "bytes"}, [RLIMIT_NICE] = {"Max nice priority", NULL}, [RLIMIT_RTPRIO] = {"Max realtime priority", NULL}, + [RLIMIT_RTTIME] = {"Max realtime timeout", "us"}, }; /* Display limits for a process */ @@ -435,7 +502,7 @@ static const struct inode_operations proc_def_inode_operations = { .setattr = proc_setattr, }; -extern struct seq_operations mounts_op; +extern const struct seq_operations mounts_op; struct proc_mounts { struct seq_file m; int event; @@ -514,7 +581,7 @@ static const struct file_operations proc_mounts_operations = { .poll = mounts_poll, }; -extern struct seq_operations mountstats_op; +extern const struct seq_operations mountstats_op; static int mountstats_open(struct inode *inode, struct file *file) { int ret = seq_open(file, &mountstats_op); @@ -591,6 +658,45 @@ static const struct file_operations proc_info_file_operations = { .read = proc_info_read, }; +static int proc_single_show(struct seq_file *m, void *v) +{ + struct inode *inode = m->private; + struct pid_namespace *ns; + struct pid *pid; + struct task_struct *task; + int ret; + + ns = inode->i_sb->s_fs_info; + pid = proc_pid(inode); + task = get_pid_task(pid, PIDTYPE_PID); + if (!task) + return -ESRCH; + + ret = PROC_I(inode)->op.proc_show(m, ns, pid, task); + + put_task_struct(task); + return ret; +} + +static int proc_single_open(struct inode *inode, struct file *filp) +{ + int ret; + ret = single_open(filp, proc_single_show, NULL); + if (!ret) { + struct seq_file *m = filp->private_data; + + m->private = inode; + } + return ret; +} + +static const struct file_operations proc_single_file_operations = { + .open = proc_single_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + static int mem_open(struct inode* inode, struct file* file) { file->private_data = (void*)((long)current->self_exec_id); @@ -716,7 +822,7 @@ out_no_task: } #endif -static loff_t mem_lseek(struct file * file, loff_t offset, int orig) +loff_t mem_lseek(struct file *file, loff_t offset, int orig) { switch (orig) { case 0: @@ -864,42 +970,6 @@ static const struct file_operations proc_oom_adjust_operations = { .write = oom_adjust_write, }; -#ifdef CONFIG_MMU -static ssize_t clear_refs_write(struct file *file, const char __user *buf, - size_t count, loff_t *ppos) -{ - struct task_struct *task; - char buffer[PROC_NUMBUF], *end; - struct mm_struct *mm; - - memset(buffer, 0, sizeof(buffer)); - if (count > sizeof(buffer) - 1) - count = sizeof(buffer) - 1; - if (copy_from_user(buffer, buf, count)) - return -EFAULT; - if (!simple_strtol(buffer, &end, 0)) - return -EINVAL; - if (*end == '\n') - end++; - task = get_proc_task(file->f_path.dentry->d_inode); - if (!task) - return -ESRCH; - mm = get_task_mm(task); - if (mm) { - clear_refs_smap(mm); - mmput(mm); - } - put_task_struct(task); - if (end - buffer == 0) - return -EIO; - return end - buffer; -} - -static struct file_operations proc_clear_refs_operations = { - .write = clear_refs_write, -}; -#endif - #ifdef CONFIG_AUDITSYSCALL #define TMPBUFLEN 21 static ssize_t proc_loginuid_read(struct file * file, char __user * buf, @@ -913,7 +983,7 @@ static ssize_t proc_loginuid_read(struct file * file, char __user * buf, if (!task) return -ESRCH; length = scnprintf(tmpbuf, TMPBUFLEN, "%u", - audit_get_loginuid(task->audit_context)); + audit_get_loginuid(task)); put_task_struct(task); return simple_read_from_buffer(buf, count, ppos, tmpbuf, length); } @@ -966,6 +1036,26 @@ static const struct file_operations proc_loginuid_operations = { .read = proc_loginuid_read, .write = proc_loginuid_write, }; + +static ssize_t proc_sessionid_read(struct file * file, char __user * buf, + size_t count, loff_t *ppos) +{ + struct inode * inode = file->f_path.dentry->d_inode; + struct task_struct *task = get_proc_task(inode); + ssize_t length; + char tmpbuf[TMPBUFLEN]; + + if (!task) + return -ESRCH; + length = scnprintf(tmpbuf, TMPBUFLEN, "%u", + audit_get_sessionid(task)); + put_task_struct(task); + return simple_read_from_buffer(buf, count, ppos, tmpbuf, length); +} + +static const struct file_operations proc_sessionid_operations = { + .read = proc_sessionid_read, +}; #endif #ifdef CONFIG_FAULT_INJECTION @@ -1020,6 +1110,7 @@ static const struct file_operations proc_fault_inject_operations = { }; #endif + #ifdef CONFIG_SCHED_DEBUG /* * Print out various scheduling related per-task fields: @@ -1089,39 +1180,36 @@ static void *proc_pid_follow_link(struct dentry *dentry, struct nameidata *nd) int error = -EACCES; /* We don't need a base pointer in the /proc filesystem */ - path_release(nd); + path_put(&nd->path); /* Are we allowed to snoop on the tasks file descriptors? */ if (!proc_fd_access_allowed(inode)) goto out; - error = PROC_I(inode)->op.proc_get_link(inode, &nd->dentry, &nd->mnt); + error = PROC_I(inode)->op.proc_get_link(inode, &nd->path); nd->last_type = LAST_BIND; out: return ERR_PTR(error); } -static int do_proc_readlink(struct dentry *dentry, struct vfsmount *mnt, - char __user *buffer, int buflen) +static int do_proc_readlink(struct path *path, char __user *buffer, int buflen) { - struct inode * inode; char *tmp = (char*)__get_free_page(GFP_TEMPORARY); - char *path; + char *pathname; int len; if (!tmp) return -ENOMEM; - inode = dentry->d_inode; - path = d_path(dentry, mnt, tmp, PAGE_SIZE); - len = PTR_ERR(path); - if (IS_ERR(path)) + pathname = d_path(path, tmp, PAGE_SIZE); + len = PTR_ERR(pathname); + if (IS_ERR(pathname)) goto out; - len = tmp + PAGE_SIZE - 1 - path; + len = tmp + PAGE_SIZE - 1 - pathname; if (len > buflen) len = buflen; - if (copy_to_user(buffer, path, len)) + if (copy_to_user(buffer, pathname, len)) len = -EFAULT; out: free_page((unsigned long)tmp); @@ -1132,20 +1220,18 @@ static int proc_pid_readlink(struct dentry * dentry, char __user * buffer, int b { int error = -EACCES; struct inode *inode = dentry->d_inode; - struct dentry *de; - struct vfsmount *mnt = NULL; + struct path path; /* Are we allowed to snoop on the tasks file descriptors? */ if (!proc_fd_access_allowed(inode)) goto out; - error = PROC_I(inode)->op.proc_get_link(inode, &de, &mnt); + error = PROC_I(inode)->op.proc_get_link(inode, &path); if (error) goto out; - error = do_proc_readlink(de, mnt, buffer, buflen); - dput(de); - mntput(mnt); + error = do_proc_readlink(&path, buffer, buflen); + path_put(&path); out: return error; } @@ -1372,8 +1458,7 @@ out: #define PROC_FDINFO_MAX 64 -static int proc_fd_info(struct inode *inode, struct dentry **dentry, - struct vfsmount **mnt, char *info) +static int proc_fd_info(struct inode *inode, struct path *path, char *info) { struct task_struct *task = get_proc_task(inode); struct files_struct *files = NULL; @@ -1392,10 +1477,10 @@ static int proc_fd_info(struct inode *inode, struct dentry **dentry, spin_lock(&files->file_lock); file = fcheck_files(files, fd); if (file) { - if (mnt) - *mnt = mntget(file->f_path.mnt); - if (dentry) - *dentry = dget(file->f_path.dentry); + if (path) { + *path = file->f_path; + path_get(&file->f_path); + } if (info) snprintf(info, PROC_FDINFO_MAX, "pos:\t%lli\n" @@ -1412,10 +1497,9 @@ static int proc_fd_info(struct inode *inode, struct dentry **dentry, return -ENOENT; } -static int proc_fd_link(struct inode *inode, struct dentry **dentry, - struct vfsmount **mnt) +static int proc_fd_link(struct inode *inode, struct path *path) { - return proc_fd_info(inode, dentry, mnt, NULL); + return proc_fd_info(inode, path, NULL); } static int tid_fd_revalidate(struct dentry *dentry, struct nameidata *nd) @@ -1609,7 +1693,7 @@ static ssize_t proc_fdinfo_read(struct file *file, char __user *buf, size_t len, loff_t *ppos) { char tmp[PROC_FDINFO_MAX]; - int err = proc_fd_info(file->f_path.dentry->d_inode, NULL, NULL, tmp); + int err = proc_fd_info(file->f_path.dentry->d_inode, NULL, tmp); if (!err) err = simple_read_from_buffer(buf, len, ppos, tmp, strlen(tmp)); return err; @@ -2026,15 +2110,23 @@ static const struct file_operations proc_coredump_filter_operations = { static int proc_self_readlink(struct dentry *dentry, char __user *buffer, int buflen) { + struct pid_namespace *ns = dentry->d_sb->s_fs_info; + pid_t tgid = task_tgid_nr_ns(current, ns); char tmp[PROC_NUMBUF]; - sprintf(tmp, "%d", task_tgid_vnr(current)); + if (!tgid) + return -ENOENT; + sprintf(tmp, "%d", tgid); return vfs_readlink(dentry,buffer,buflen,tmp); } static void *proc_self_follow_link(struct dentry *dentry, struct nameidata *nd) { + struct pid_namespace *ns = dentry->d_sb->s_fs_info; + pid_t tgid = task_tgid_nr_ns(current, ns); char tmp[PROC_NUMBUF]; - sprintf(tmp, "%d", task_tgid_vnr(current)); + if (!tgid) + return ERR_PTR(-ENOENT); + sprintf(tmp, "%d", task_tgid_nr_ns(current, ns)); return ERR_PTR(vfs_follow_link(nd,tmp)); } @@ -2197,16 +2289,19 @@ static const struct pid_entry tgid_base_stuff[] = { DIR("task", S_IRUGO|S_IXUGO, task), DIR("fd", S_IRUSR|S_IXUSR, fd), DIR("fdinfo", S_IRUSR|S_IXUSR, fdinfo), +#ifdef CONFIG_NET + DIR("net", S_IRUGO|S_IXUGO, net), +#endif REG("environ", S_IRUSR, environ), INF("auxv", S_IRUSR, pid_auxv), - INF("status", S_IRUGO, pid_status), + ONE("status", S_IRUGO, pid_status), INF("limits", S_IRUSR, pid_limits), #ifdef CONFIG_SCHED_DEBUG REG("sched", S_IRUGO|S_IWUSR, pid_sched), #endif INF("cmdline", S_IRUGO, pid_cmdline), - INF("stat", S_IRUGO, tgid_stat), - INF("statm", S_IRUGO, pid_statm), + ONE("stat", S_IRUGO, tgid_stat), + ONE("statm", S_IRUGO, pid_statm), REG("maps", S_IRUGO, maps), #ifdef CONFIG_NUMA REG("numa_maps", S_IRUGO, numa_maps), @@ -2217,9 +2312,10 @@ static const struct pid_entry tgid_base_stuff[] = { LNK("exe", exe), REG("mounts", S_IRUGO, mounts), REG("mountstats", S_IRUSR, mountstats), -#ifdef CONFIG_MMU +#ifdef CONFIG_PROC_PAGE_MONITOR REG("clear_refs", S_IWUSR, clear_refs), REG("smaps", S_IRUGO, smaps), + REG("pagemap", S_IRUSR, pagemap), #endif #ifdef CONFIG_SECURITY DIR("attr", S_IRUGO|S_IXUGO, attr_dir), @@ -2230,6 +2326,9 @@ static const struct pid_entry tgid_base_stuff[] = { #ifdef CONFIG_SCHEDSTATS INF("schedstat", S_IRUGO, pid_schedstat), #endif +#ifdef CONFIG_LATENCYTOP + REG("latency", S_IRUGO, lstats), +#endif #ifdef CONFIG_PROC_PID_CPUSET REG("cpuset", S_IRUGO, cpuset), #endif @@ -2240,6 +2339,7 @@ static const struct pid_entry tgid_base_stuff[] = { REG("oom_adj", S_IRUGO|S_IWUSR, oom_adjust), #ifdef CONFIG_AUDITSYSCALL REG("loginuid", S_IWUSR|S_IRUGO, loginuid), + REG("sessionid", S_IRUSR, sessionid), #endif #ifdef CONFIG_FAULT_INJECTION REG("make-it-fail", S_IRUGO|S_IWUSR, fault_inject), @@ -2285,7 +2385,8 @@ static void proc_flush_task_mnt(struct vfsmount *mnt, pid_t pid, pid_t tgid) name.len = snprintf(buf, sizeof(buf), "%d", pid); dentry = d_hash_and_lookup(mnt->mnt_root, &name); if (dentry) { - shrink_dcache_parent(dentry); + if (!(current->flags & PF_EXITING)) + shrink_dcache_parent(dentry); d_drop(dentry); dput(dentry); } @@ -2525,14 +2626,14 @@ static const struct pid_entry tid_base_stuff[] = { DIR("fdinfo", S_IRUSR|S_IXUSR, fdinfo), REG("environ", S_IRUSR, environ), INF("auxv", S_IRUSR, pid_auxv), - INF("status", S_IRUGO, pid_status), + ONE("status", S_IRUGO, pid_status), INF("limits", S_IRUSR, pid_limits), #ifdef CONFIG_SCHED_DEBUG REG("sched", S_IRUGO|S_IWUSR, pid_sched), #endif INF("cmdline", S_IRUGO, pid_cmdline), - INF("stat", S_IRUGO, tid_stat), - INF("statm", S_IRUGO, pid_statm), + ONE("stat", S_IRUGO, tid_stat), + ONE("statm", S_IRUGO, pid_statm), REG("maps", S_IRUGO, maps), #ifdef CONFIG_NUMA REG("numa_maps", S_IRUGO, numa_maps), @@ -2542,9 +2643,10 @@ static const struct pid_entry tid_base_stuff[] = { LNK("root", root), LNK("exe", exe), REG("mounts", S_IRUGO, mounts), -#ifdef CONFIG_MMU +#ifdef CONFIG_PROC_PAGE_MONITOR REG("clear_refs", S_IWUSR, clear_refs), REG("smaps", S_IRUGO, smaps), + REG("pagemap", S_IRUSR, pagemap), #endif #ifdef CONFIG_SECURITY DIR("attr", S_IRUGO|S_IXUGO, attr_dir), @@ -2555,6 +2657,9 @@ static const struct pid_entry tid_base_stuff[] = { #ifdef CONFIG_SCHEDSTATS INF("schedstat", S_IRUGO, pid_schedstat), #endif +#ifdef CONFIG_LATENCYTOP + REG("latency", S_IRUGO, lstats), +#endif #ifdef CONFIG_PROC_PID_CPUSET REG("cpuset", S_IRUGO, cpuset), #endif @@ -2565,6 +2670,7 @@ static const struct pid_entry tid_base_stuff[] = { REG("oom_adj", S_IRUGO|S_IWUSR, oom_adjust), #ifdef CONFIG_AUDITSYSCALL REG("loginuid", S_IWUSR|S_IRUGO, loginuid), + REG("sessionid", S_IRUSR, sessionid), #endif #ifdef CONFIG_FAULT_INJECTION REG("make-it-fail", S_IRUGO|S_IWUSR, fault_inject), diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 6a2fe5187b6..a36ad3c75cf 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -25,12 +25,6 @@ #include "internal.h" -static ssize_t proc_file_read(struct file *file, char __user *buf, - size_t nbytes, loff_t *ppos); -static ssize_t proc_file_write(struct file *file, const char __user *buffer, - size_t count, loff_t *ppos); -static loff_t proc_file_lseek(struct file *, loff_t, int); - DEFINE_SPINLOCK(proc_subdir_lock); static int proc_match(int len, const char *name, struct proc_dir_entry *de) @@ -40,12 +34,6 @@ static int proc_match(int len, const char *name, struct proc_dir_entry *de) return !memcmp(name, de->name, len); } -static const struct file_operations proc_file_operations = { - .llseek = proc_file_lseek, - .read = proc_file_read, - .write = proc_file_write, -}; - /* buffer size is one page but our output routines use some slack for overruns */ #define PROC_BLOCK_SIZE (PAGE_SIZE - 1024) @@ -233,6 +221,12 @@ proc_file_lseek(struct file *file, loff_t offset, int orig) return retval; } +static const struct file_operations proc_file_operations = { + .llseek = proc_file_lseek, + .read = proc_file_read, + .write = proc_file_write, +}; + static int proc_notify_change(struct dentry *dentry, struct iattr *iattr) { struct inode *inode = dentry->d_inode; @@ -383,15 +377,14 @@ static struct dentry_operations proc_dentry_operations = * Don't create negative dentries here, return -ENOENT by hand * instead. */ -struct dentry *proc_lookup(struct inode * dir, struct dentry *dentry, struct nameidata *nd) +struct dentry *proc_lookup_de(struct proc_dir_entry *de, struct inode *dir, + struct dentry *dentry) { struct inode *inode = NULL; - struct proc_dir_entry * de; int error = -ENOENT; lock_kernel(); spin_lock(&proc_subdir_lock); - de = PDE(dir); if (de) { for (de = de->subdir; de ; de = de->next) { if (de->namelen != dentry->d_name.len) @@ -399,19 +392,17 @@ struct dentry *proc_lookup(struct inode * dir, struct dentry *dentry, struct nam if (!memcmp(dentry->d_name.name, de->name, de->namelen)) { unsigned int ino; - if (de->shadow_proc) - de = de->shadow_proc(current, de); ino = de->low_ino; de_get(de); spin_unlock(&proc_subdir_lock); error = -EINVAL; inode = proc_get_inode(dir->i_sb, ino, de); - spin_lock(&proc_subdir_lock); - break; + goto out_unlock; } } } spin_unlock(&proc_subdir_lock); +out_unlock: unlock_kernel(); if (inode) { @@ -423,6 +414,12 @@ struct dentry *proc_lookup(struct inode * dir, struct dentry *dentry, struct nam return ERR_PTR(error); } +struct dentry *proc_lookup(struct inode *dir, struct dentry *dentry, + struct nameidata *nd) +{ + return proc_lookup_de(PDE(dir), dir, dentry); +} + /* * This returns non-zero if at EOF, so that the /proc * root directory can use this and check if it should @@ -432,10 +429,9 @@ struct dentry *proc_lookup(struct inode * dir, struct dentry *dentry, struct nam * value of the readdir() call, as long as it's non-negative * for success.. */ -int proc_readdir(struct file * filp, - void * dirent, filldir_t filldir) +int proc_readdir_de(struct proc_dir_entry *de, struct file *filp, void *dirent, + filldir_t filldir) { - struct proc_dir_entry * de; unsigned int ino; int i; struct inode *inode = filp->f_path.dentry->d_inode; @@ -444,7 +440,6 @@ int proc_readdir(struct file * filp, lock_kernel(); ino = inode->i_ino; - de = PDE(inode); if (!de) { ret = -EINVAL; goto out; @@ -505,6 +500,13 @@ out: unlock_kernel(); return ret; } +int proc_readdir(struct file *filp, void *dirent, filldir_t filldir) +{ + struct inode *inode = filp->f_path.dentry->d_inode; + + return proc_readdir_de(PDE(inode), filp, dirent, filldir); +} + /* * These are the generic /proc directory operations. They * use the in-memory "struct proc_dir_entry" tree to parse @@ -527,6 +529,7 @@ static const struct inode_operations proc_dir_inode_operations = { static int proc_register(struct proc_dir_entry * dir, struct proc_dir_entry * dp) { unsigned int i; + struct proc_dir_entry *tmp; i = get_inode_number(); if (i == 0) @@ -550,6 +553,15 @@ static int proc_register(struct proc_dir_entry * dir, struct proc_dir_entry * dp } spin_lock(&proc_subdir_lock); + + for (tmp = dir->subdir; tmp; tmp = tmp->next) + if (strcmp(tmp->name, dp->name) == 0) { + printk(KERN_WARNING "proc_dir_entry '%s' already " + "registered\n", dp->name); + dump_stack(); + break; + } + dp->next = dir->subdir; dp->parent = dir; dir->subdir = dp; @@ -558,7 +570,7 @@ static int proc_register(struct proc_dir_entry * dir, struct proc_dir_entry * dp return 0; } -static struct proc_dir_entry *proc_create(struct proc_dir_entry **parent, +static struct proc_dir_entry *__proc_create(struct proc_dir_entry **parent, const char *name, mode_t mode, nlink_t nlink) @@ -601,7 +613,7 @@ struct proc_dir_entry *proc_symlink(const char *name, { struct proc_dir_entry *ent; - ent = proc_create(&parent,name, + ent = __proc_create(&parent, name, (S_IFLNK | S_IRUGO | S_IWUGO | S_IXUGO),1); if (ent) { @@ -626,7 +638,7 @@ struct proc_dir_entry *proc_mkdir_mode(const char *name, mode_t mode, { struct proc_dir_entry *ent; - ent = proc_create(&parent, name, S_IFDIR | mode, 2); + ent = __proc_create(&parent, name, S_IFDIR | mode, 2); if (ent) { if (proc_register(parent, ent) < 0) { kfree(ent); @@ -660,7 +672,7 @@ struct proc_dir_entry *create_proc_entry(const char *name, mode_t mode, nlink = 1; } - ent = proc_create(&parent,name,mode,nlink); + ent = __proc_create(&parent, name, mode, nlink); if (ent) { if (proc_register(parent, ent) < 0) { kfree(ent); @@ -670,6 +682,38 @@ struct proc_dir_entry *create_proc_entry(const char *name, mode_t mode, return ent; } +struct proc_dir_entry *proc_create(const char *name, mode_t mode, + struct proc_dir_entry *parent, + const struct file_operations *proc_fops) +{ + struct proc_dir_entry *pde; + nlink_t nlink; + + if (S_ISDIR(mode)) { + if ((mode & S_IALLUGO) == 0) + mode |= S_IRUGO | S_IXUGO; + nlink = 2; + } else { + if ((mode & S_IFMT) == 0) + mode |= S_IFREG; + if ((mode & S_IALLUGO) == 0) + mode |= S_IRUGO; + nlink = 1; + } + + pde = __proc_create(&parent, name, mode, nlink); + if (!pde) + goto out; + pde->proc_fops = proc_fops; + if (proc_register(parent, pde) < 0) + goto out_free; + return pde; +out_free: + kfree(pde); +out: + return NULL; +} + void free_proc_entry(struct proc_dir_entry *de) { unsigned int ino = de->low_ino; @@ -679,7 +723,7 @@ void free_proc_entry(struct proc_dir_entry *de) release_inode_number(ino); - if (S_ISLNK(de->mode) && de->data) + if (S_ISLNK(de->mode)) kfree(de->data); kfree(de); } diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 1a551d92e1d..82b3a1b5a70 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -73,11 +73,6 @@ static void proc_delete_inode(struct inode *inode) struct vfsmount *proc_mnt; -static void proc_read_inode(struct inode * inode) -{ - inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; -} - static struct kmem_cache * proc_inode_cachep; static struct inode *proc_alloc_inode(struct super_block *sb) @@ -128,7 +123,6 @@ static int proc_remount(struct super_block *sb, int *flags, char *data) static const struct super_operations proc_sops = { .alloc_inode = proc_alloc_inode, .destroy_inode = proc_destroy_inode, - .read_inode = proc_read_inode, .drop_inode = generic_delete_inode, .delete_inode = proc_delete_inode, .statfs = simple_statfs, @@ -401,39 +395,41 @@ struct inode *proc_get_inode(struct super_block *sb, unsigned int ino, if (de != NULL && !try_module_get(de->owner)) goto out_mod; - inode = iget(sb, ino); + inode = iget_locked(sb, ino); if (!inode) goto out_ino; - - PROC_I(inode)->fd = 0; - PROC_I(inode)->pde = de; - if (de) { - if (de->mode) { - inode->i_mode = de->mode; - inode->i_uid = de->uid; - inode->i_gid = de->gid; - } - if (de->size) - inode->i_size = de->size; - if (de->nlink) - inode->i_nlink = de->nlink; - if (de->proc_iops) - inode->i_op = de->proc_iops; - if (de->proc_fops) { - if (S_ISREG(inode->i_mode)) { + if (inode->i_state & I_NEW) { + inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; + PROC_I(inode)->fd = 0; + PROC_I(inode)->pde = de; + if (de) { + if (de->mode) { + inode->i_mode = de->mode; + inode->i_uid = de->uid; + inode->i_gid = de->gid; + } + if (de->size) + inode->i_size = de->size; + if (de->nlink) + inode->i_nlink = de->nlink; + if (de->proc_iops) + inode->i_op = de->proc_iops; + if (de->proc_fops) { + if (S_ISREG(inode->i_mode)) { #ifdef CONFIG_COMPAT - if (!de->proc_fops->compat_ioctl) - inode->i_fop = - &proc_reg_file_ops_no_compat; - else + if (!de->proc_fops->compat_ioctl) + inode->i_fop = + &proc_reg_file_ops_no_compat; + else #endif - inode->i_fop = &proc_reg_file_ops; + inode->i_fop = &proc_reg_file_ops; + } else { + inode->i_fop = de->proc_fops; + } } - else - inode->i_fop = de->proc_fops; } + unlock_new_inode(inode); } - return inode; out_ino: @@ -471,4 +467,3 @@ out_no_root: de_put(&proc_root); return -ENOMEM; } -MODULE_LICENSE("GPL"); diff --git a/fs/proc/internal.h b/fs/proc/internal.h index 05b3e900626..bc72f5c8c47 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -46,21 +46,26 @@ extern int nommu_vma_show(struct seq_file *, struct vm_area_struct *); extern int maps_protect; -extern void create_seq_entry(char *name, mode_t mode, const struct file_operations *f); -extern int proc_exe_link(struct inode *, struct dentry **, struct vfsmount **); -extern int proc_tid_stat(struct task_struct *, char *); -extern int proc_tgid_stat(struct task_struct *, char *); -extern int proc_pid_status(struct task_struct *, char *); -extern int proc_pid_statm(struct task_struct *, char *); +extern void create_seq_entry(char *name, mode_t mode, + const struct file_operations *f); +extern int proc_exe_link(struct inode *, struct path *); +extern int proc_tid_stat(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *task); +extern int proc_tgid_stat(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *task); +extern int proc_pid_status(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *task); +extern int proc_pid_statm(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *task); +extern loff_t mem_lseek(struct file *file, loff_t offset, int orig); extern const struct file_operations proc_maps_operations; extern const struct file_operations proc_numa_maps_operations; extern const struct file_operations proc_smaps_operations; - -extern const struct file_operations proc_maps_operations; -extern const struct file_operations proc_numa_maps_operations; -extern const struct file_operations proc_smaps_operations; - +extern const struct file_operations proc_clear_refs_operations; +extern const struct file_operations proc_pagemap_operations; +extern const struct file_operations proc_net_operations; +extern const struct inode_operations proc_net_inode_operations; void free_proc_entry(struct proc_dir_entry *de); @@ -80,3 +85,8 @@ static inline int proc_fd(struct inode *inode) { return PROC_I(inode)->fd; } + +struct dentry *proc_lookup_de(struct proc_dir_entry *de, struct inode *ino, + struct dentry *dentry); +int proc_readdir_de(struct proc_dir_entry *de, struct file *filp, void *dirent, + filldir_t filldir); diff --git a/fs/proc/kcore.c b/fs/proc/kcore.c index 1be73082edd..e78c81fcf54 100644 --- a/fs/proc/kcore.c +++ b/fs/proc/kcore.c @@ -12,7 +12,6 @@ #include <linux/mm.h> #include <linux/proc_fs.h> #include <linux/user.h> -#include <linux/a.out.h> #include <linux/capability.h> #include <linux/elf.h> #include <linux/elfcore.h> @@ -325,7 +324,7 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos) if (m == NULL) { if (clear_user(buffer, tsz)) return -EFAULT; - } else if ((start >= VMALLOC_START) && (start < VMALLOC_END)) { + } else if (is_vmalloc_addr((void *)start)) { char * elf_buf; struct vm_struct *m; unsigned long curstart = start; diff --git a/fs/proc/nommu.c b/fs/proc/nommu.c index 22f789de390..941e95114b5 100644 --- a/fs/proc/nommu.c +++ b/fs/proc/nommu.c @@ -67,7 +67,7 @@ int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma) if (len < 1) len = 1; seq_printf(m, "%*c", len, ' '); - seq_path(m, file->f_path.mnt, file->f_path.dentry, ""); + seq_path(m, &file->f_path, ""); } seq_putc(m, '\n'); @@ -116,7 +116,7 @@ static void *nommu_vma_list_next(struct seq_file *m, void *v, loff_t *pos) return rb_next((struct rb_node *) v); } -static struct seq_operations proc_nommu_vma_list_seqop = { +static const struct seq_operations proc_nommu_vma_list_seqop = { .start = nommu_vma_list_start, .next = nommu_vma_list_next, .stop = nommu_vma_list_stop, diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index 3462bfde89f..2d563979cb0 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -29,8 +29,10 @@ #include <linux/mm.h> #include <linux/mmzone.h> #include <linux/pagemap.h> +#include <linux/interrupt.h> #include <linux/swap.h> #include <linux/slab.h> +#include <linux/genhd.h> #include <linux/smp.h> #include <linux/signal.h> #include <linux/module.h> @@ -46,6 +48,7 @@ #include <linux/vmalloc.h> #include <linux/crash_dump.h> #include <linux/pid_namespace.h> +#include <linux/bootmem.h> #include <asm/uaccess.h> #include <asm/pgtable.h> #include <asm/io.h> @@ -63,7 +66,6 @@ */ extern int get_hardware_list(char *); extern int get_stram_list(char *); -extern int get_filesystem_list(char *); extern int get_exec_domain_list(char *); extern int get_dma_list(char *); @@ -83,10 +85,15 @@ static int loadavg_read_proc(char *page, char **start, off_t off, { int a, b, c; int len; + unsigned long seq; + + do { + seq = read_seqbegin(&xtime_lock); + a = avenrun[0] + (FIXED_1/200); + b = avenrun[1] + (FIXED_1/200); + c = avenrun[2] + (FIXED_1/200); + } while (read_seqretry(&xtime_lock, seq)); - a = avenrun[0] + (FIXED_1/200); - b = avenrun[1] + (FIXED_1/200); - c = avenrun[2] + (FIXED_1/200); len = sprintf(page,"%d.%02d %d.%02d %d.%02d %ld/%d %d\n", LOAD_INT(a), LOAD_FRAC(a), LOAD_INT(b), LOAD_FRAC(b), @@ -216,7 +223,7 @@ static int meminfo_read_proc(char *page, char **start, off_t off, #undef K } -extern struct seq_operations fragmentation_op; +extern const struct seq_operations fragmentation_op; static int fragmentation_open(struct inode *inode, struct file *file) { (void)inode; @@ -230,7 +237,7 @@ static const struct file_operations fragmentation_file_operations = { .release = seq_release, }; -extern struct seq_operations pagetypeinfo_op; +extern const struct seq_operations pagetypeinfo_op; static int pagetypeinfo_open(struct inode *inode, struct file *file) { return seq_open(file, &pagetypeinfo_op); @@ -243,7 +250,7 @@ static const struct file_operations pagetypeinfo_file_ops = { .release = seq_release, }; -extern struct seq_operations zoneinfo_op; +extern const struct seq_operations zoneinfo_op; static int zoneinfo_open(struct inode *inode, struct file *file) { return seq_open(file, &zoneinfo_op); @@ -268,7 +275,7 @@ static int version_read_proc(char *page, char **start, off_t off, return proc_calc_metrics(page, start, off, count, eof, len); } -extern struct seq_operations cpuinfo_op; +extern const struct seq_operations cpuinfo_op; static int cpuinfo_open(struct inode *inode, struct file *file) { return seq_open(file, &cpuinfo_op); @@ -321,7 +328,7 @@ static void devinfo_stop(struct seq_file *f, void *v) /* Nothing to do */ } -static struct seq_operations devinfo_ops = { +static const struct seq_operations devinfo_ops = { .start = devinfo_start, .next = devinfo_next, .stop = devinfo_stop, @@ -340,7 +347,7 @@ static const struct file_operations proc_devinfo_operations = { .release = seq_release, }; -extern struct seq_operations vmstat_op; +extern const struct seq_operations vmstat_op; static int vmstat_open(struct inode *inode, struct file *file) { return seq_open(file, &vmstat_op); @@ -371,7 +378,6 @@ static int stram_read_proc(char *page, char **start, off_t off, #endif #ifdef CONFIG_BLOCK -extern struct seq_operations partitions_op; static int partitions_open(struct inode *inode, struct file *file) { return seq_open(file, &partitions_op); @@ -383,7 +389,6 @@ static const struct file_operations proc_partitions_operations = { .release = seq_release, }; -extern struct seq_operations diskstats_op; static int diskstats_open(struct inode *inode, struct file *file) { return seq_open(file, &diskstats_op); @@ -397,7 +402,7 @@ static const struct file_operations proc_diskstats_operations = { #endif #ifdef CONFIG_MODULES -extern struct seq_operations modules_op; +extern const struct seq_operations modules_op; static int modules_open(struct inode *inode, struct file *file) { return seq_open(file, &modules_op); @@ -424,7 +429,7 @@ static const struct file_operations proc_slabinfo_operations = { }; #ifdef CONFIG_DEBUG_SLAB_LEAK -extern struct seq_operations slabstats_op; +extern const struct seq_operations slabstats_op; static int slabstats_open(struct inode *inode, struct file *file) { unsigned long *n = kzalloc(PAGE_SIZE, GFP_KERNEL); @@ -598,8 +603,7 @@ static void int_seq_stop(struct seq_file *f, void *v) } -extern int show_interrupts(struct seq_file *f, void *v); /* In arch code */ -static struct seq_operations int_seq_ops = { +static const struct seq_operations int_seq_ops = { .start = int_seq_start, .next = int_seq_next, .stop = int_seq_stop, @@ -675,6 +679,137 @@ static const struct file_operations proc_sysrq_trigger_operations = { }; #endif +#ifdef CONFIG_PROC_PAGE_MONITOR +#define KPMSIZE sizeof(u64) +#define KPMMASK (KPMSIZE - 1) +/* /proc/kpagecount - an array exposing page counts + * + * Each entry is a u64 representing the corresponding + * physical page count. + */ +static ssize_t kpagecount_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) +{ + u64 __user *out = (u64 __user *)buf; + struct page *ppage; + unsigned long src = *ppos; + unsigned long pfn; + ssize_t ret = 0; + u64 pcount; + + pfn = src / KPMSIZE; + count = min_t(size_t, count, (max_pfn * KPMSIZE) - src); + if (src & KPMMASK || count & KPMMASK) + return -EIO; + + while (count > 0) { + ppage = NULL; + if (pfn_valid(pfn)) + ppage = pfn_to_page(pfn); + pfn++; + if (!ppage) + pcount = 0; + else + pcount = atomic_read(&ppage->_count); + + if (put_user(pcount, out++)) { + ret = -EFAULT; + break; + } + + count -= KPMSIZE; + } + + *ppos += (char __user *)out - buf; + if (!ret) + ret = (char __user *)out - buf; + return ret; +} + +static struct file_operations proc_kpagecount_operations = { + .llseek = mem_lseek, + .read = kpagecount_read, +}; + +/* /proc/kpageflags - an array exposing page flags + * + * Each entry is a u64 representing the corresponding + * physical page flags. + */ + +/* These macros are used to decouple internal flags from exported ones */ + +#define KPF_LOCKED 0 +#define KPF_ERROR 1 +#define KPF_REFERENCED 2 +#define KPF_UPTODATE 3 +#define KPF_DIRTY 4 +#define KPF_LRU 5 +#define KPF_ACTIVE 6 +#define KPF_SLAB 7 +#define KPF_WRITEBACK 8 +#define KPF_RECLAIM 9 +#define KPF_BUDDY 10 + +#define kpf_copy_bit(flags, srcpos, dstpos) (((flags >> srcpos) & 1) << dstpos) + +static ssize_t kpageflags_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) +{ + u64 __user *out = (u64 __user *)buf; + struct page *ppage; + unsigned long src = *ppos; + unsigned long pfn; + ssize_t ret = 0; + u64 kflags, uflags; + + pfn = src / KPMSIZE; + count = min_t(unsigned long, count, (max_pfn * KPMSIZE) - src); + if (src & KPMMASK || count & KPMMASK) + return -EIO; + + while (count > 0) { + ppage = NULL; + if (pfn_valid(pfn)) + ppage = pfn_to_page(pfn); + pfn++; + if (!ppage) + kflags = 0; + else + kflags = ppage->flags; + + uflags = kpf_copy_bit(KPF_LOCKED, PG_locked, kflags) | + kpf_copy_bit(kflags, KPF_ERROR, PG_error) | + kpf_copy_bit(kflags, KPF_REFERENCED, PG_referenced) | + kpf_copy_bit(kflags, KPF_UPTODATE, PG_uptodate) | + kpf_copy_bit(kflags, KPF_DIRTY, PG_dirty) | + kpf_copy_bit(kflags, KPF_LRU, PG_lru) | + kpf_copy_bit(kflags, KPF_ACTIVE, PG_active) | + kpf_copy_bit(kflags, KPF_SLAB, PG_slab) | + kpf_copy_bit(kflags, KPF_WRITEBACK, PG_writeback) | + kpf_copy_bit(kflags, KPF_RECLAIM, PG_reclaim) | + kpf_copy_bit(kflags, KPF_BUDDY, PG_buddy); + + if (put_user(uflags, out++)) { + ret = -EFAULT; + break; + } + + count -= KPMSIZE; + } + + *ppos += (char __user *)out - buf; + if (!ret) + ret = (char __user *)out - buf; + return ret; +} + +static struct file_operations proc_kpageflags_operations = { + .llseek = mem_lseek, + .read = kpageflags_read, +}; +#endif /* CONFIG_PROC_PAGE_MONITOR */ + struct proc_dir_entry *proc_root_kcore; void create_seq_entry(char *name, mode_t mode, const struct file_operations *f) @@ -755,6 +890,10 @@ void __init proc_misc_init(void) (size_t)high_memory - PAGE_OFFSET + PAGE_SIZE; } #endif +#ifdef CONFIG_PROC_PAGE_MONITOR + create_seq_entry("kpagecount", S_IRUSR, &proc_kpagecount_operations); + create_seq_entry("kpageflags", S_IRUSR, &proc_kpageflags_operations); +#endif #ifdef CONFIG_PROC_VMCORE proc_vmcore = create_proc_entry("vmcore", S_IRUSR, NULL); if (proc_vmcore) diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c index 0afe21ee060..4caa5f774fb 100644 --- a/fs/proc/proc_net.c +++ b/fs/proc/proc_net.c @@ -22,19 +22,128 @@ #include <linux/mount.h> #include <linux/nsproxy.h> #include <net/net_namespace.h> +#include <linux/seq_file.h> #include "internal.h" +int seq_open_net(struct inode *ino, struct file *f, + const struct seq_operations *ops, int size) +{ + struct net *net; + struct seq_net_private *p; + + BUG_ON(size < sizeof(*p)); + + net = get_proc_net(ino); + if (net == NULL) + return -ENXIO; + + p = __seq_open_private(f, ops, size); + if (p == NULL) { + put_net(net); + return -ENOMEM; + } + p->net = net; + return 0; +} +EXPORT_SYMBOL_GPL(seq_open_net); + +int seq_release_net(struct inode *ino, struct file *f) +{ + struct seq_file *seq; + struct seq_net_private *p; + + seq = f->private_data; + p = seq->private; + + put_net(p->net); + seq_release_private(ino, f); + return 0; +} +EXPORT_SYMBOL_GPL(seq_release_net); + +static struct net *get_proc_task_net(struct inode *dir) +{ + struct task_struct *task; + struct nsproxy *ns; + struct net *net = NULL; + + rcu_read_lock(); + task = pid_task(proc_pid(dir), PIDTYPE_PID); + if (task != NULL) { + ns = task_nsproxy(task); + if (ns != NULL) + net = get_net(ns->net_ns); + } + rcu_read_unlock(); + + return net; +} + +static struct dentry *proc_tgid_net_lookup(struct inode *dir, + struct dentry *dentry, struct nameidata *nd) +{ + struct dentry *de; + struct net *net; + + de = ERR_PTR(-ENOENT); + net = get_proc_task_net(dir); + if (net != NULL) { + de = proc_lookup_de(net->proc_net, dir, dentry); + put_net(net); + } + return de; +} + +static int proc_tgid_net_getattr(struct vfsmount *mnt, struct dentry *dentry, + struct kstat *stat) +{ + struct inode *inode = dentry->d_inode; + struct net *net; + + net = get_proc_task_net(inode); + + generic_fillattr(inode, stat); + + if (net != NULL) { + stat->nlink = net->proc_net->nlink; + put_net(net); + } + + return 0; +} + +const struct inode_operations proc_net_inode_operations = { + .lookup = proc_tgid_net_lookup, + .getattr = proc_tgid_net_getattr, +}; + +static int proc_tgid_net_readdir(struct file *filp, void *dirent, + filldir_t filldir) +{ + int ret; + struct net *net; + + ret = -EINVAL; + net = get_proc_task_net(filp->f_path.dentry->d_inode); + if (net != NULL) { + ret = proc_readdir_de(net->proc_net, filp, dirent, filldir); + put_net(net); + } + return ret; +} + +const struct file_operations proc_net_operations = { + .read = generic_read_dir, + .readdir = proc_tgid_net_readdir, +}; + + struct proc_dir_entry *proc_net_fops_create(struct net *net, const char *name, mode_t mode, const struct file_operations *fops) { - struct proc_dir_entry *res; - - res = create_proc_entry(name, mode, net->proc_net); - if (res) - res->proc_fops = fops; - return res; + return proc_create(name, mode, net->proc_net, fops); } EXPORT_SYMBOL_GPL(proc_net_fops_create); @@ -50,57 +159,52 @@ struct net *get_proc_net(const struct inode *inode) } EXPORT_SYMBOL_GPL(get_proc_net); -static struct proc_dir_entry *shadow_pde; - -static struct proc_dir_entry *proc_net_shadow(struct task_struct *task, - struct proc_dir_entry *de) +struct proc_dir_entry *proc_net_mkdir(struct net *net, const char *name, + struct proc_dir_entry *parent) { - return task->nsproxy->net_ns->proc_net; + struct proc_dir_entry *pde; + pde = proc_mkdir_mode(name, S_IRUGO | S_IXUGO, parent); + if (pde != NULL) + pde->data = net; + return pde; } +EXPORT_SYMBOL_GPL(proc_net_mkdir); static __net_init int proc_net_ns_init(struct net *net) { - struct proc_dir_entry *root, *netd, *net_statd; + struct proc_dir_entry *netd, *net_statd; int err; err = -ENOMEM; - root = kzalloc(sizeof(*root), GFP_KERNEL); - if (!root) + netd = kzalloc(sizeof(*netd), GFP_KERNEL); + if (!netd) goto out; - err = -EEXIST; - netd = proc_mkdir("net", root); - if (!netd) - goto free_root; + netd->data = net; + netd->nlink = 2; + netd->name = "net"; + netd->namelen = 3; + netd->parent = &proc_root; err = -EEXIST; - net_statd = proc_mkdir("stat", netd); + net_statd = proc_net_mkdir(net, "stat", netd); if (!net_statd) goto free_net; - root->data = net; - netd->data = net; - net_statd->data = net; - - net->proc_net_root = root; net->proc_net = netd; net->proc_net_stat = net_statd; - err = 0; + return 0; +free_net: + kfree(netd); out: return err; -free_net: - remove_proc_entry("net", root); -free_root: - kfree(root); - goto out; } static __net_exit void proc_net_ns_exit(struct net *net) { remove_proc_entry("stat", net->proc_net); - remove_proc_entry("net", net->proc_net_root); - kfree(net->proc_net_root); + kfree(net->proc_net); } static struct pernet_operations __net_initdata proc_net_ns_ops = { @@ -110,8 +214,7 @@ static struct pernet_operations __net_initdata proc_net_ns_ops = { int __init proc_net_init(void) { - shadow_pde = proc_mkdir("net", NULL); - shadow_pde->shadow_proc = proc_net_shadow; + proc_symlink("net", NULL, "self/net"); return register_pernet_subsys(&proc_net_ns_ops); } diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index 4e57fcf8598..614c34b6d1c 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -9,7 +9,7 @@ static struct dentry_operations proc_sys_dentry_operations; static const struct file_operations proc_sys_file_operations; -static struct inode_operations proc_sys_inode_operations; +static const struct inode_operations proc_sys_inode_operations; static void proc_sys_refresh_inode(struct inode *inode, struct ctl_table *table) { @@ -407,7 +407,7 @@ static int proc_sys_permission(struct inode *inode, int mask, struct nameidata * if (!nd || !depth) goto out; - dentry = nd->dentry; + dentry = nd->path.dentry; table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head); /* If the entry does not exist deny permission */ @@ -446,7 +446,7 @@ static const struct file_operations proc_sys_file_operations = { .readdir = proc_sys_readdir, }; -static struct inode_operations proc_sys_inode_operations = { +static const struct inode_operations proc_sys_inode_operations = { .lookup = proc_sys_lookup, .permission = proc_sys_permission, .setattr = proc_sys_setattr, diff --git a/fs/proc/proc_tty.c b/fs/proc/proc_tty.c index 22846225acf..49816e00b51 100644 --- a/fs/proc/proc_tty.c +++ b/fs/proc/proc_tty.c @@ -15,9 +15,6 @@ #include <linux/seq_file.h> #include <linux/bitops.h> -static int tty_ldiscs_read_proc(char *page, char **start, off_t off, - int count, int *eof, void *data); - /* * The /proc/tty directory inodes... */ @@ -120,7 +117,7 @@ static void t_stop(struct seq_file *m, void *v) mutex_unlock(&tty_mutex); } -static struct seq_operations tty_drivers_op = { +static const struct seq_operations tty_drivers_op = { .start = t_start, .next = t_next, .stop = t_stop, diff --git a/fs/proc/root.c b/fs/proc/root.c index 81f99e691f9..ef0fb57fc9e 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -232,6 +232,7 @@ void pid_ns_release_proc(struct pid_namespace *ns) EXPORT_SYMBOL(proc_symlink); EXPORT_SYMBOL(proc_mkdir); EXPORT_SYMBOL(create_proc_entry); +EXPORT_SYMBOL(proc_create); EXPORT_SYMBOL(remove_proc_entry); EXPORT_SYMBOL(proc_root); EXPORT_SYMBOL(proc_root_fs); diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 8043a3eab52..9dfb5ff2420 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -5,14 +5,18 @@ #include <linux/highmem.h> #include <linux/ptrace.h> #include <linux/pagemap.h> +#include <linux/ptrace.h> #include <linux/mempolicy.h> +#include <linux/swap.h> +#include <linux/swapops.h> +#include <linux/seq_file.h> #include <asm/elf.h> #include <asm/uaccess.h> #include <asm/tlbflush.h> #include "internal.h" -char *task_mem(struct mm_struct *mm, char *buffer) +void task_mem(struct seq_file *m, struct mm_struct *mm) { unsigned long data, text, lib; unsigned long hiwater_vm, total_vm, hiwater_rss, total_rss; @@ -34,7 +38,7 @@ char *task_mem(struct mm_struct *mm, char *buffer) data = mm->total_vm - mm->shared_vm - mm->stack_vm; text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK)) >> 10; lib = (mm->exec_vm << (PAGE_SHIFT-10)) - text; - buffer += sprintf(buffer, + seq_printf(m, "VmPeak:\t%8lu kB\n" "VmSize:\t%8lu kB\n" "VmLck:\t%8lu kB\n" @@ -53,7 +57,6 @@ char *task_mem(struct mm_struct *mm, char *buffer) data << (PAGE_SHIFT-10), mm->stack_vm << (PAGE_SHIFT-10), text, lib, (PTRS_PER_PTE*sizeof(pte_t)*mm->nr_ptes) >> 10); - return buffer; } unsigned long task_vsize(struct mm_struct *mm) @@ -72,7 +75,7 @@ int task_statm(struct mm_struct *mm, int *shared, int *text, return mm->total_vm; } -int proc_exe_link(struct inode *inode, struct dentry **dentry, struct vfsmount **mnt) +int proc_exe_link(struct inode *inode, struct path *path) { struct vm_area_struct * vma; int result = -ENOENT; @@ -95,8 +98,8 @@ int proc_exe_link(struct inode *inode, struct dentry **dentry, struct vfsmount * } if (vma) { - *mnt = mntget(vma->vm_file->f_path.mnt); - *dentry = dget(vma->vm_file->f_path.dentry); + *path = vma->vm_file->f_path; + path_get(&vma->vm_file->f_path); result = 0; } @@ -114,24 +117,124 @@ static void pad_len_spaces(struct seq_file *m, int len) seq_printf(m, "%*c", len, ' '); } -struct mem_size_stats +static void vma_stop(struct proc_maps_private *priv, struct vm_area_struct *vma) { - unsigned long resident; - unsigned long shared_clean; - unsigned long shared_dirty; - unsigned long private_clean; - unsigned long private_dirty; - unsigned long referenced; -}; + if (vma && vma != priv->tail_vma) { + struct mm_struct *mm = vma->vm_mm; + up_read(&mm->mmap_sem); + mmput(mm); + } +} -struct pmd_walker { - struct vm_area_struct *vma; - void *private; - void (*action)(struct vm_area_struct *, pmd_t *, unsigned long, - unsigned long, void *); -}; +static void *m_start(struct seq_file *m, loff_t *pos) +{ + struct proc_maps_private *priv = m->private; + unsigned long last_addr = m->version; + struct mm_struct *mm; + struct vm_area_struct *vma, *tail_vma = NULL; + loff_t l = *pos; + + /* Clear the per syscall fields in priv */ + priv->task = NULL; + priv->tail_vma = NULL; + + /* + * We remember last_addr rather than next_addr to hit with + * mmap_cache most of the time. We have zero last_addr at + * the beginning and also after lseek. We will have -1 last_addr + * after the end of the vmas. + */ + + if (last_addr == -1UL) + return NULL; + + priv->task = get_pid_task(priv->pid, PIDTYPE_PID); + if (!priv->task) + return NULL; + + mm = mm_for_maps(priv->task); + if (!mm) + return NULL; + + tail_vma = get_gate_vma(priv->task); + priv->tail_vma = tail_vma; + + /* Start with last addr hint */ + vma = find_vma(mm, last_addr); + if (last_addr && vma) { + vma = vma->vm_next; + goto out; + } + + /* + * Check the vma index is within the range and do + * sequential scan until m_index. + */ + vma = NULL; + if ((unsigned long)l < mm->map_count) { + vma = mm->mmap; + while (l-- && vma) + vma = vma->vm_next; + goto out; + } + + if (l != mm->map_count) + tail_vma = NULL; /* After gate vma */ + +out: + if (vma) + return vma; + + /* End of vmas has been reached */ + m->version = (tail_vma != NULL)? 0: -1UL; + up_read(&mm->mmap_sem); + mmput(mm); + return tail_vma; +} + +static void *m_next(struct seq_file *m, void *v, loff_t *pos) +{ + struct proc_maps_private *priv = m->private; + struct vm_area_struct *vma = v; + struct vm_area_struct *tail_vma = priv->tail_vma; + + (*pos)++; + if (vma && (vma != tail_vma) && vma->vm_next) + return vma->vm_next; + vma_stop(priv, vma); + return (vma != tail_vma)? tail_vma: NULL; +} + +static void m_stop(struct seq_file *m, void *v) +{ + struct proc_maps_private *priv = m->private; + struct vm_area_struct *vma = v; + + vma_stop(priv, vma); + if (priv->task) + put_task_struct(priv->task); +} + +static int do_maps_open(struct inode *inode, struct file *file, + const struct seq_operations *ops) +{ + struct proc_maps_private *priv; + int ret = -ENOMEM; + priv = kzalloc(sizeof(*priv), GFP_KERNEL); + if (priv) { + priv->pid = proc_pid(inode); + ret = seq_open(file, ops); + if (!ret) { + struct seq_file *m = file->private_data; + m->private = priv; + } else { + kfree(priv); + } + } + return ret; +} -static int show_map_internal(struct seq_file *m, void *v, struct mem_size_stats *mss) +static int show_map(struct seq_file *m, void *v) { struct proc_maps_private *priv = m->private; struct task_struct *task = priv->task; @@ -168,7 +271,7 @@ static int show_map_internal(struct seq_file *m, void *v, struct mem_size_stats */ if (file) { pad_len_spaces(m, len); - seq_path(m, file->f_path.mnt, file->f_path.dentry, "\n"); + seq_path(m, &file->f_path, "\n"); } else { const char *name = arch_vma_name(vma); if (!name) { @@ -191,41 +294,71 @@ static int show_map_internal(struct seq_file *m, void *v, struct mem_size_stats } seq_putc(m, '\n'); - if (mss) - seq_printf(m, - "Size: %8lu kB\n" - "Rss: %8lu kB\n" - "Shared_Clean: %8lu kB\n" - "Shared_Dirty: %8lu kB\n" - "Private_Clean: %8lu kB\n" - "Private_Dirty: %8lu kB\n" - "Referenced: %8lu kB\n", - (vma->vm_end - vma->vm_start) >> 10, - mss->resident >> 10, - mss->shared_clean >> 10, - mss->shared_dirty >> 10, - mss->private_clean >> 10, - mss->private_dirty >> 10, - mss->referenced >> 10); - if (m->count < m->size) /* vma is copied successfully */ m->version = (vma != get_gate_vma(task))? vma->vm_start: 0; return 0; } -static int show_map(struct seq_file *m, void *v) +static const struct seq_operations proc_pid_maps_op = { + .start = m_start, + .next = m_next, + .stop = m_stop, + .show = show_map +}; + +static int maps_open(struct inode *inode, struct file *file) { - return show_map_internal(m, v, NULL); + return do_maps_open(inode, file, &proc_pid_maps_op); } -static void smaps_pte_range(struct vm_area_struct *vma, pmd_t *pmd, - unsigned long addr, unsigned long end, - void *private) +const struct file_operations proc_maps_operations = { + .open = maps_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; + +/* + * Proportional Set Size(PSS): my share of RSS. + * + * PSS of a process is the count of pages it has in memory, where each + * page is divided by the number of processes sharing it. So if a + * process has 1000 pages all to itself, and 1000 shared with one other + * process, its PSS will be 1500. + * + * To keep (accumulated) division errors low, we adopt a 64bit + * fixed-point pss counter to minimize division errors. So (pss >> + * PSS_SHIFT) would be the real byte count. + * + * A shift of 12 before division means (assuming 4K page size): + * - 1M 3-user-pages add up to 8KB errors; + * - supports mapcount up to 2^24, or 16M; + * - supports PSS up to 2^52 bytes, or 4PB. + */ +#define PSS_SHIFT 12 + +#ifdef CONFIG_PROC_PAGE_MONITOR +struct mem_size_stats +{ + struct vm_area_struct *vma; + unsigned long resident; + unsigned long shared_clean; + unsigned long shared_dirty; + unsigned long private_clean; + unsigned long private_dirty; + unsigned long referenced; + u64 pss; +}; + +static int smaps_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end, + void *private) { struct mem_size_stats *mss = private; + struct vm_area_struct *vma = mss->vma; pte_t *pte, ptent; spinlock_t *ptl; struct page *page; + int mapcount; pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl); for (; addr != end; pte++, addr += PAGE_SIZE) { @@ -242,26 +375,88 @@ static void smaps_pte_range(struct vm_area_struct *vma, pmd_t *pmd, /* Accumulate the size in pages that have been accessed. */ if (pte_young(ptent) || PageReferenced(page)) mss->referenced += PAGE_SIZE; - if (page_mapcount(page) >= 2) { + mapcount = page_mapcount(page); + if (mapcount >= 2) { if (pte_dirty(ptent)) mss->shared_dirty += PAGE_SIZE; else mss->shared_clean += PAGE_SIZE; + mss->pss += (PAGE_SIZE << PSS_SHIFT) / mapcount; } else { if (pte_dirty(ptent)) mss->private_dirty += PAGE_SIZE; else mss->private_clean += PAGE_SIZE; + mss->pss += (PAGE_SIZE << PSS_SHIFT); } } pte_unmap_unlock(pte - 1, ptl); cond_resched(); + return 0; } -static void clear_refs_pte_range(struct vm_area_struct *vma, pmd_t *pmd, - unsigned long addr, unsigned long end, - void *private) +static struct mm_walk smaps_walk = { .pmd_entry = smaps_pte_range }; + +static int show_smap(struct seq_file *m, void *v) { + struct vm_area_struct *vma = v; + struct mem_size_stats mss; + int ret; + + memset(&mss, 0, sizeof mss); + mss.vma = vma; + if (vma->vm_mm && !is_vm_hugetlb_page(vma)) + walk_page_range(vma->vm_mm, vma->vm_start, vma->vm_end, + &smaps_walk, &mss); + + ret = show_map(m, v); + if (ret) + return ret; + + seq_printf(m, + "Size: %8lu kB\n" + "Rss: %8lu kB\n" + "Pss: %8lu kB\n" + "Shared_Clean: %8lu kB\n" + "Shared_Dirty: %8lu kB\n" + "Private_Clean: %8lu kB\n" + "Private_Dirty: %8lu kB\n" + "Referenced: %8lu kB\n", + (vma->vm_end - vma->vm_start) >> 10, + mss.resident >> 10, + (unsigned long)(mss.pss >> (10 + PSS_SHIFT)), + mss.shared_clean >> 10, + mss.shared_dirty >> 10, + mss.private_clean >> 10, + mss.private_dirty >> 10, + mss.referenced >> 10); + + return ret; +} + +static const struct seq_operations proc_pid_smaps_op = { + .start = m_start, + .next = m_next, + .stop = m_stop, + .show = show_smap +}; + +static int smaps_open(struct inode *inode, struct file *file) +{ + return do_maps_open(inode, file, &proc_pid_smaps_op); +} + +const struct file_operations proc_smaps_operations = { + .open = smaps_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; + +static int clear_refs_pte_range(pmd_t *pmd, unsigned long addr, + unsigned long end, void *private) +{ + struct vm_area_struct *vma = private; pte_t *pte, ptent; spinlock_t *ptl; struct page *page; @@ -282,235 +477,274 @@ static void clear_refs_pte_range(struct vm_area_struct *vma, pmd_t *pmd, } pte_unmap_unlock(pte - 1, ptl); cond_resched(); + return 0; } -static inline void walk_pmd_range(struct pmd_walker *walker, pud_t *pud, - unsigned long addr, unsigned long end) +static struct mm_walk clear_refs_walk = { .pmd_entry = clear_refs_pte_range }; + +static ssize_t clear_refs_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) { - pmd_t *pmd; - unsigned long next; + struct task_struct *task; + char buffer[PROC_NUMBUF], *end; + struct mm_struct *mm; + struct vm_area_struct *vma; - for (pmd = pmd_offset(pud, addr); addr != end; - pmd++, addr = next) { - next = pmd_addr_end(addr, end); - if (pmd_none_or_clear_bad(pmd)) - continue; - walker->action(walker->vma, pmd, addr, next, walker->private); + memset(buffer, 0, sizeof(buffer)); + if (count > sizeof(buffer) - 1) + count = sizeof(buffer) - 1; + if (copy_from_user(buffer, buf, count)) + return -EFAULT; + if (!simple_strtol(buffer, &end, 0)) + return -EINVAL; + if (*end == '\n') + end++; + task = get_proc_task(file->f_path.dentry->d_inode); + if (!task) + return -ESRCH; + mm = get_task_mm(task); + if (mm) { + down_read(&mm->mmap_sem); + for (vma = mm->mmap; vma; vma = vma->vm_next) + if (!is_vm_hugetlb_page(vma)) + walk_page_range(mm, vma->vm_start, vma->vm_end, + &clear_refs_walk, vma); + flush_tlb_mm(mm); + up_read(&mm->mmap_sem); + mmput(mm); } + put_task_struct(task); + if (end - buffer == 0) + return -EIO; + return end - buffer; } -static inline void walk_pud_range(struct pmd_walker *walker, pgd_t *pgd, - unsigned long addr, unsigned long end) -{ - pud_t *pud; - unsigned long next; +const struct file_operations proc_clear_refs_operations = { + .write = clear_refs_write, +}; - for (pud = pud_offset(pgd, addr); addr != end; - pud++, addr = next) { - next = pud_addr_end(addr, end); - if (pud_none_or_clear_bad(pud)) - continue; - walk_pmd_range(walker, pud, addr, next); +struct pagemapread { + char __user *out, *end; +}; + +#define PM_ENTRY_BYTES sizeof(u64) +#define PM_STATUS_BITS 3 +#define PM_STATUS_OFFSET (64 - PM_STATUS_BITS) +#define PM_STATUS_MASK (((1LL << PM_STATUS_BITS) - 1) << PM_STATUS_OFFSET) +#define PM_STATUS(nr) (((nr) << PM_STATUS_OFFSET) & PM_STATUS_MASK) +#define PM_PSHIFT_BITS 6 +#define PM_PSHIFT_OFFSET (PM_STATUS_OFFSET - PM_PSHIFT_BITS) +#define PM_PSHIFT_MASK (((1LL << PM_PSHIFT_BITS) - 1) << PM_PSHIFT_OFFSET) +#define PM_PSHIFT(x) (((u64) (x) << PM_PSHIFT_OFFSET) & PM_PSHIFT_MASK) +#define PM_PFRAME_MASK ((1LL << PM_PSHIFT_OFFSET) - 1) +#define PM_PFRAME(x) ((x) & PM_PFRAME_MASK) + +#define PM_PRESENT PM_STATUS(4LL) +#define PM_SWAP PM_STATUS(2LL) +#define PM_NOT_PRESENT PM_PSHIFT(PAGE_SHIFT) +#define PM_END_OF_BUFFER 1 + +static int add_to_pagemap(unsigned long addr, u64 pfn, + struct pagemapread *pm) +{ + /* + * Make sure there's room in the buffer for an + * entire entry. Otherwise, only copy part of + * the pfn. + */ + if (pm->out + PM_ENTRY_BYTES >= pm->end) { + if (copy_to_user(pm->out, &pfn, pm->end - pm->out)) + return -EFAULT; + pm->out = pm->end; + return PM_END_OF_BUFFER; } + + if (put_user(pfn, pm->out)) + return -EFAULT; + pm->out += PM_ENTRY_BYTES; + return 0; } -/* - * walk_page_range - walk the page tables of a VMA with a callback - * @vma - VMA to walk - * @action - callback invoked for every bottom-level (PTE) page table - * @private - private data passed to the callback function - * - * Recursively walk the page table for the memory area in a VMA, calling - * a callback for every bottom-level (PTE) page table. - */ -static inline void walk_page_range(struct vm_area_struct *vma, - void (*action)(struct vm_area_struct *, - pmd_t *, unsigned long, - unsigned long, void *), - void *private) +static int pagemap_pte_hole(unsigned long start, unsigned long end, + void *private) { - unsigned long addr = vma->vm_start; - unsigned long end = vma->vm_end; - struct pmd_walker walker = { - .vma = vma, - .private = private, - .action = action, - }; - pgd_t *pgd; - unsigned long next; - - for (pgd = pgd_offset(vma->vm_mm, addr); addr != end; - pgd++, addr = next) { - next = pgd_addr_end(addr, end); - if (pgd_none_or_clear_bad(pgd)) - continue; - walk_pud_range(&walker, pgd, addr, next); + struct pagemapread *pm = private; + unsigned long addr; + int err = 0; + for (addr = start; addr < end; addr += PAGE_SIZE) { + err = add_to_pagemap(addr, PM_NOT_PRESENT, pm); + if (err) + break; } + return err; } -static int show_smap(struct seq_file *m, void *v) +u64 swap_pte_to_pagemap_entry(pte_t pte) { - struct vm_area_struct *vma = v; - struct mem_size_stats mss; - - memset(&mss, 0, sizeof mss); - if (vma->vm_mm && !is_vm_hugetlb_page(vma)) - walk_page_range(vma, smaps_pte_range, &mss); - return show_map_internal(m, v, &mss); + swp_entry_t e = pte_to_swp_entry(pte); + return swp_type(e) | (swp_offset(e) << MAX_SWAPFILES_SHIFT); } -void clear_refs_smap(struct mm_struct *mm) +static int pagemap_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end, + void *private) { - struct vm_area_struct *vma; + struct pagemapread *pm = private; + pte_t *pte; + int err = 0; + + for (; addr != end; addr += PAGE_SIZE) { + u64 pfn = PM_NOT_PRESENT; + pte = pte_offset_map(pmd, addr); + if (is_swap_pte(*pte)) + pfn = PM_PFRAME(swap_pte_to_pagemap_entry(*pte)) + | PM_PSHIFT(PAGE_SHIFT) | PM_SWAP; + else if (pte_present(*pte)) + pfn = PM_PFRAME(pte_pfn(*pte)) + | PM_PSHIFT(PAGE_SHIFT) | PM_PRESENT; + /* unmap so we're not in atomic when we copy to userspace */ + pte_unmap(pte); + err = add_to_pagemap(addr, pfn, pm); + if (err) + return err; + } - down_read(&mm->mmap_sem); - for (vma = mm->mmap; vma; vma = vma->vm_next) - if (vma->vm_mm && !is_vm_hugetlb_page(vma)) - walk_page_range(vma, clear_refs_pte_range, NULL); - flush_tlb_mm(mm); - up_read(&mm->mmap_sem); + cond_resched(); + + return err; } -static void *m_start(struct seq_file *m, loff_t *pos) +static struct mm_walk pagemap_walk = { + .pmd_entry = pagemap_pte_range, + .pte_hole = pagemap_pte_hole +}; + +/* + * /proc/pid/pagemap - an array mapping virtual pages to pfns + * + * For each page in the address space, this file contains one 64-bit entry + * consisting of the following: + * + * Bits 0-55 page frame number (PFN) if present + * Bits 0-4 swap type if swapped + * Bits 5-55 swap offset if swapped + * Bits 55-60 page shift (page size = 1<<page shift) + * Bit 61 reserved for future use + * Bit 62 page swapped + * Bit 63 page present + * + * If the page is not present but in swap, then the PFN contains an + * encoding of the swap file number and the page's offset into the + * swap. Unmapped pages return a null PFN. This allows determining + * precisely which pages are mapped (or in swap) and comparing mapped + * pages between processes. + * + * Efficient users of this interface will use /proc/pid/maps to + * determine which areas of memory are actually mapped and llseek to + * skip over unmapped regions. + */ +static ssize_t pagemap_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) { - struct proc_maps_private *priv = m->private; - unsigned long last_addr = m->version; + struct task_struct *task = get_proc_task(file->f_path.dentry->d_inode); + struct page **pages, *page; + unsigned long uaddr, uend; struct mm_struct *mm; - struct vm_area_struct *vma, *tail_vma = NULL; - loff_t l = *pos; + struct pagemapread pm; + int pagecount; + int ret = -ESRCH; - /* Clear the per syscall fields in priv */ - priv->task = NULL; - priv->tail_vma = NULL; - - /* - * We remember last_addr rather than next_addr to hit with - * mmap_cache most of the time. We have zero last_addr at - * the beginning and also after lseek. We will have -1 last_addr - * after the end of the vmas. - */ + if (!task) + goto out; - if (last_addr == -1UL) - return NULL; + ret = -EACCES; + if (!ptrace_may_attach(task)) + goto out_task; - priv->task = get_pid_task(priv->pid, PIDTYPE_PID); - if (!priv->task) - return NULL; + ret = -EINVAL; + /* file position must be aligned */ + if (*ppos % PM_ENTRY_BYTES) + goto out_task; - mm = mm_for_maps(priv->task); + ret = 0; + mm = get_task_mm(task); if (!mm) - return NULL; - - priv->tail_vma = tail_vma = get_gate_vma(priv->task); - - /* Start with last addr hint */ - if (last_addr && (vma = find_vma(mm, last_addr))) { - vma = vma->vm_next; - goto out; - } - - /* - * Check the vma index is within the range and do - * sequential scan until m_index. - */ - vma = NULL; - if ((unsigned long)l < mm->map_count) { - vma = mm->mmap; - while (l-- && vma) - vma = vma->vm_next; - goto out; + goto out_task; + + ret = -ENOMEM; + uaddr = (unsigned long)buf & PAGE_MASK; + uend = (unsigned long)(buf + count); + pagecount = (PAGE_ALIGN(uend) - uaddr) / PAGE_SIZE; + pages = kmalloc(pagecount * sizeof(struct page *), GFP_KERNEL); + if (!pages) + goto out_mm; + + down_read(¤t->mm->mmap_sem); + ret = get_user_pages(current, current->mm, uaddr, pagecount, + 1, 0, pages, NULL); + up_read(¤t->mm->mmap_sem); + + if (ret < 0) + goto out_free; + + if (ret != pagecount) { + pagecount = ret; + ret = -EFAULT; + goto out_pages; } - if (l != mm->map_count) - tail_vma = NULL; /* After gate vma */ + pm.out = buf; + pm.end = buf + count; -out: - if (vma) - return vma; - - /* End of vmas has been reached */ - m->version = (tail_vma != NULL)? 0: -1UL; - up_read(&mm->mmap_sem); - mmput(mm); - return tail_vma; -} - -static void vma_stop(struct proc_maps_private *priv, struct vm_area_struct *vma) -{ - if (vma && vma != priv->tail_vma) { - struct mm_struct *mm = vma->vm_mm; - up_read(&mm->mmap_sem); - mmput(mm); + if (!ptrace_may_attach(task)) { + ret = -EIO; + } else { + unsigned long src = *ppos; + unsigned long svpfn = src / PM_ENTRY_BYTES; + unsigned long start_vaddr = svpfn << PAGE_SHIFT; + unsigned long end_vaddr = TASK_SIZE_OF(task); + + /* watch out for wraparound */ + if (svpfn > TASK_SIZE_OF(task) >> PAGE_SHIFT) + start_vaddr = end_vaddr; + + /* + * The odds are that this will stop walking way + * before end_vaddr, because the length of the + * user buffer is tracked in "pm", and the walk + * will stop when we hit the end of the buffer. + */ + ret = walk_page_range(mm, start_vaddr, end_vaddr, + &pagemap_walk, &pm); + if (ret == PM_END_OF_BUFFER) + ret = 0; + /* don't need mmap_sem for these, but this looks cleaner */ + *ppos += pm.out - buf; + if (!ret) + ret = pm.out - buf; } -} - -static void *m_next(struct seq_file *m, void *v, loff_t *pos) -{ - struct proc_maps_private *priv = m->private; - struct vm_area_struct *vma = v; - struct vm_area_struct *tail_vma = priv->tail_vma; - - (*pos)++; - if (vma && (vma != tail_vma) && vma->vm_next) - return vma->vm_next; - vma_stop(priv, vma); - return (vma != tail_vma)? tail_vma: NULL; -} - -static void m_stop(struct seq_file *m, void *v) -{ - struct proc_maps_private *priv = m->private; - struct vm_area_struct *vma = v; - - vma_stop(priv, vma); - if (priv->task) - put_task_struct(priv->task); -} - -static struct seq_operations proc_pid_maps_op = { - .start = m_start, - .next = m_next, - .stop = m_stop, - .show = show_map -}; - -static struct seq_operations proc_pid_smaps_op = { - .start = m_start, - .next = m_next, - .stop = m_stop, - .show = show_smap -}; -static int do_maps_open(struct inode *inode, struct file *file, - struct seq_operations *ops) -{ - struct proc_maps_private *priv; - int ret = -ENOMEM; - priv = kzalloc(sizeof(*priv), GFP_KERNEL); - if (priv) { - priv->pid = proc_pid(inode); - ret = seq_open(file, ops); - if (!ret) { - struct seq_file *m = file->private_data; - m->private = priv; - } else { - kfree(priv); - } +out_pages: + for (; pagecount; pagecount--) { + page = pages[pagecount-1]; + if (!PageReserved(page)) + SetPageDirty(page); + page_cache_release(page); } +out_free: + kfree(pages); +out_mm: + mmput(mm); +out_task: + put_task_struct(task); +out: return ret; } -static int maps_open(struct inode *inode, struct file *file) -{ - return do_maps_open(inode, file, &proc_pid_maps_op); -} - -const struct file_operations proc_maps_operations = { - .open = maps_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release_private, +const struct file_operations proc_pagemap_operations = { + .llseek = mem_lseek, /* borrow this */ + .read = pagemap_read, }; +#endif /* CONFIG_PROC_PAGE_MONITOR */ #ifdef CONFIG_NUMA extern int show_numa_map(struct seq_file *m, void *v); @@ -526,7 +760,7 @@ static int show_numa_map_checked(struct seq_file *m, void *v) return show_numa_map(m, v); } -static struct seq_operations proc_pid_numa_maps_op = { +static const struct seq_operations proc_pid_numa_maps_op = { .start = m_start, .next = m_next, .stop = m_stop, @@ -545,15 +779,3 @@ const struct file_operations proc_numa_maps_operations = { .release = seq_release_private, }; #endif - -static int smaps_open(struct inode *inode, struct file *file) -{ - return do_maps_open(inode, file, &proc_pid_smaps_op); -} - -const struct file_operations proc_smaps_operations = { - .open = smaps_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release_private, -}; diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c index 1932c2ca345..8011528518b 100644 --- a/fs/proc/task_nommu.c +++ b/fs/proc/task_nommu.c @@ -12,7 +12,7 @@ * each process that owns it. Non-shared memory is counted * accurately. */ -char *task_mem(struct mm_struct *mm, char *buffer) +void task_mem(struct seq_file *m, struct mm_struct *mm) { struct vm_list_struct *vml; unsigned long bytes = 0, sbytes = 0, slack = 0; @@ -58,14 +58,13 @@ char *task_mem(struct mm_struct *mm, char *buffer) bytes += kobjsize(current); /* includes kernel stack */ - buffer += sprintf(buffer, + seq_printf(m, "Mem:\t%8lu bytes\n" "Slack:\t%8lu bytes\n" "Shared:\t%8lu bytes\n", bytes, slack, sbytes); up_read(&mm->mmap_sem); - return buffer; } unsigned long task_vsize(struct mm_struct *mm) @@ -104,7 +103,7 @@ int task_statm(struct mm_struct *mm, int *shared, int *text, return size; } -int proc_exe_link(struct inode *inode, struct dentry **dentry, struct vfsmount **mnt) +int proc_exe_link(struct inode *inode, struct path *path) { struct vm_list_struct *vml; struct vm_area_struct *vma; @@ -127,8 +126,8 @@ int proc_exe_link(struct inode *inode, struct dentry **dentry, struct vfsmount * } if (vma) { - *mnt = mntget(vma->vm_file->f_path.mnt); - *dentry = dget(vma->vm_file->f_path.dentry); + *path = vma->vm_file->f_path; + path_get(&vma->vm_file->f_path); result = 0; } @@ -199,7 +198,7 @@ static void *m_next(struct seq_file *m, void *_vml, loff_t *pos) return vml ? vml->next : NULL; } -static struct seq_operations proc_pid_maps_ops = { +static const struct seq_operations proc_pid_maps_ops = { .start = m_start, .next = m_next, .stop = m_stop, diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c index 523e1098ae8..9ac0f5e064e 100644 --- a/fs/proc/vmcore.c +++ b/fs/proc/vmcore.c @@ -10,7 +10,6 @@ #include <linux/mm.h> #include <linux/proc_fs.h> #include <linux/user.h> -#include <linux/a.out.h> #include <linux/elf.h> #include <linux/elfcore.h> #include <linux/highmem.h> diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c index 638bdb96321..b31ab78052b 100644 --- a/fs/qnx4/inode.c +++ b/fs/qnx4/inode.c @@ -125,7 +125,6 @@ static int qnx4_write_inode(struct inode *inode, int unused) static void qnx4_put_super(struct super_block *sb); static struct inode *qnx4_alloc_inode(struct super_block *sb); static void qnx4_destroy_inode(struct inode *inode); -static void qnx4_read_inode(struct inode *); static int qnx4_remount(struct super_block *sb, int *flags, char *data); static int qnx4_statfs(struct dentry *, struct kstatfs *); @@ -133,7 +132,6 @@ static const struct super_operations qnx4_sops = { .alloc_inode = qnx4_alloc_inode, .destroy_inode = qnx4_destroy_inode, - .read_inode = qnx4_read_inode, .put_super = qnx4_put_super, .statfs = qnx4_statfs, .remount_fs = qnx4_remount, @@ -357,6 +355,7 @@ static int qnx4_fill_super(struct super_block *s, void *data, int silent) struct inode *root; const char *errmsg; struct qnx4_sb_info *qs; + int ret = -EINVAL; qs = kzalloc(sizeof(struct qnx4_sb_info), GFP_KERNEL); if (!qs) @@ -396,12 +395,14 @@ static int qnx4_fill_super(struct super_block *s, void *data, int silent) } /* does root not have inode number QNX4_ROOT_INO ?? */ - root = iget(s, QNX4_ROOT_INO * QNX4_INODES_PER_BLOCK); - if (!root) { + root = qnx4_iget(s, QNX4_ROOT_INO * QNX4_INODES_PER_BLOCK); + if (IS_ERR(root)) { printk("qnx4: get inode failed\n"); + ret = PTR_ERR(root); goto out; } + ret = -ENOMEM; s->s_root = d_alloc_root(root); if (s->s_root == NULL) goto outi; @@ -417,7 +418,7 @@ static int qnx4_fill_super(struct super_block *s, void *data, int silent) outnobh: kfree(qs); s->s_fs_info = NULL; - return -EINVAL; + return ret; } static void qnx4_put_super(struct super_block *sb) @@ -462,29 +463,38 @@ static const struct address_space_operations qnx4_aops = { .bmap = qnx4_bmap }; -static void qnx4_read_inode(struct inode *inode) +struct inode *qnx4_iget(struct super_block *sb, unsigned long ino) { struct buffer_head *bh; struct qnx4_inode_entry *raw_inode; - int block, ino; - struct super_block *sb = inode->i_sb; - struct qnx4_inode_entry *qnx4_inode = qnx4_raw_inode(inode); + int block; + struct qnx4_inode_entry *qnx4_inode; + struct inode *inode; - ino = inode->i_ino; + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; + + qnx4_inode = qnx4_raw_inode(inode); inode->i_mode = 0; QNX4DEBUG(("Reading inode : [%d]\n", ino)); if (!ino) { - printk("qnx4: bad inode number on dev %s: %d is out of range\n", + printk(KERN_ERR "qnx4: bad inode number on dev %s: %lu is " + "out of range\n", sb->s_id, ino); - return; + iget_failed(inode); + return ERR_PTR(-EIO); } block = ino / QNX4_INODES_PER_BLOCK; if (!(bh = sb_bread(sb, block))) { printk("qnx4: major problem: unable to read inode from dev " "%s\n", sb->s_id); - return; + iget_failed(inode); + return ERR_PTR(-EIO); } raw_inode = ((struct qnx4_inode_entry *) bh->b_data) + (ino % QNX4_INODES_PER_BLOCK); @@ -515,9 +525,16 @@ static void qnx4_read_inode(struct inode *inode) inode->i_op = &page_symlink_inode_operations; inode->i_mapping->a_ops = &qnx4_aops; qnx4_i(inode)->mmu_private = inode->i_size; - } else - printk("qnx4: bad inode %d on dev %s\n",ino,sb->s_id); + } else { + printk(KERN_ERR "qnx4: bad inode %lu on dev %s\n", + ino, sb->s_id); + iget_failed(inode); + brelse(bh); + return ERR_PTR(-EIO); + } brelse(bh); + unlock_new_inode(inode); + return inode; } static struct kmem_cache *qnx4_inode_cachep; diff --git a/fs/qnx4/namei.c b/fs/qnx4/namei.c index 733cdf01d64..775eed3a408 100644 --- a/fs/qnx4/namei.c +++ b/fs/qnx4/namei.c @@ -128,10 +128,12 @@ struct dentry * qnx4_lookup(struct inode *dir, struct dentry *dentry, struct nam } brelse(bh); - if ((foundinode = iget(dir->i_sb, ino)) == NULL) { + foundinode = qnx4_iget(dir->i_sb, ino); + if (IS_ERR(foundinode)) { unlock_kernel(); - QNX4DEBUG(("qnx4: lookup->iget -> NULL\n")); - return ERR_PTR(-EACCES); + QNX4DEBUG(("qnx4: lookup->iget -> error %ld\n", + PTR_ERR(foundinode))); + return ERR_CAST(foundinode); } out: unlock_kernel(); diff --git a/fs/quota.c b/fs/quota.c index 99b24b52bfc..84f28dd7211 100644 --- a/fs/quota.c +++ b/fs/quota.c @@ -341,11 +341,11 @@ static inline struct super_block *quotactl_block(const char __user *special) char *tmp = getname(special); if (IS_ERR(tmp)) - return ERR_PTR(PTR_ERR(tmp)); + return ERR_CAST(tmp); bdev = lookup_bdev(tmp); putname(tmp); if (IS_ERR(bdev)) - return ERR_PTR(PTR_ERR(bdev)); + return ERR_CAST(bdev); sb = get_super(bdev); bdput(bdev); if (!sb) diff --git a/fs/read_write.c b/fs/read_write.c index ea1f94cc722..49a98718ecd 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -197,25 +197,27 @@ int rw_verify_area(int read_write, struct file *file, loff_t *ppos, size_t count { struct inode *inode; loff_t pos; + int retval = -EINVAL; inode = file->f_path.dentry->d_inode; if (unlikely((ssize_t) count < 0)) - goto Einval; + return retval; pos = *ppos; if (unlikely((pos < 0) || (loff_t) (pos + count) < 0)) - goto Einval; + return retval; if (unlikely(inode->i_flock && mandatory_lock(inode))) { - int retval = locks_mandatory_area( + retval = locks_mandatory_area( read_write == READ ? FLOCK_VERIFY_READ : FLOCK_VERIFY_WRITE, inode, file, pos, count); if (retval < 0) return retval; } + retval = security_file_permission(file, + read_write == READ ? MAY_READ : MAY_WRITE); + if (retval) + return retval; return count > MAX_RW_COUNT ? MAX_RW_COUNT : count; - -Einval: - return -EINVAL; } static void wait_on_retry_sync_kiocb(struct kiocb *iocb) @@ -267,18 +269,15 @@ ssize_t vfs_read(struct file *file, char __user *buf, size_t count, loff_t *pos) ret = rw_verify_area(READ, file, pos, count); if (ret >= 0) { count = ret; - ret = security_file_permission (file, MAY_READ); - if (!ret) { - if (file->f_op->read) - ret = file->f_op->read(file, buf, count, pos); - else - ret = do_sync_read(file, buf, count, pos); - if (ret > 0) { - fsnotify_access(file->f_path.dentry); - add_rchar(current, ret); - } - inc_syscr(current); + if (file->f_op->read) + ret = file->f_op->read(file, buf, count, pos); + else + ret = do_sync_read(file, buf, count, pos); + if (ret > 0) { + fsnotify_access(file->f_path.dentry); + add_rchar(current, ret); } + inc_syscr(current); } return ret; @@ -325,18 +324,15 @@ ssize_t vfs_write(struct file *file, const char __user *buf, size_t count, loff_ ret = rw_verify_area(WRITE, file, pos, count); if (ret >= 0) { count = ret; - ret = security_file_permission (file, MAY_WRITE); - if (!ret) { - if (file->f_op->write) - ret = file->f_op->write(file, buf, count, pos); - else - ret = do_sync_write(file, buf, count, pos); - if (ret > 0) { - fsnotify_modify(file->f_path.dentry); - add_wchar(current, ret); - } - inc_syscw(current); + if (file->f_op->write) + ret = file->f_op->write(file, buf, count, pos); + else + ret = do_sync_write(file, buf, count, pos); + if (ret > 0) { + fsnotify_modify(file->f_path.dentry); + add_wchar(current, ret); } + inc_syscw(current); } return ret; @@ -370,7 +366,6 @@ asmlinkage ssize_t sys_read(unsigned int fd, char __user * buf, size_t count) return ret; } -EXPORT_UNUSED_SYMBOL_GPL(sys_read); /* to be deleted for 2.6.25 */ asmlinkage ssize_t sys_write(unsigned int fd, const char __user * buf, size_t count) { @@ -450,6 +445,7 @@ unsigned long iov_shorten(struct iovec *iov, unsigned long nr_segs, size_t to) } return seg; } +EXPORT_SYMBOL(iov_shorten); ssize_t do_sync_readv_writev(struct file *filp, const struct iovec *iov, unsigned long nr_segs, size_t len, loff_t *ppos, iov_fn_t fn) @@ -603,9 +599,6 @@ static ssize_t do_readv_writev(int type, struct file *file, ret = rw_verify_area(type, file, pos, tot_len); if (ret < 0) goto out; - ret = security_file_permission(file, type == READ ? MAY_READ : MAY_WRITE); - if (ret) - goto out; fnv = NULL; if (type == READ) { @@ -737,10 +730,6 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, goto fput_in; count = retval; - retval = security_file_permission (in_file, MAY_READ); - if (retval) - goto fput_in; - /* * Get output file, and verify that it is ok.. */ @@ -759,10 +748,6 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, goto fput_out; count = retval; - retval = security_file_permission (out_file, MAY_WRITE); - if (retval) - goto fput_out; - if (!max) max = min(in_inode->i_sb->s_maxbytes, out_inode->i_sb->s_maxbytes); diff --git a/fs/readdir.c b/fs/readdir.c index efe52e67657..4e026e5407f 100644 --- a/fs/readdir.c +++ b/fs/readdir.c @@ -30,7 +30,10 @@ int vfs_readdir(struct file *file, filldir_t filler, void *buf) if (res) goto out; - mutex_lock(&inode->i_mutex); + res = mutex_lock_killable(&inode->i_mutex); + if (res) + goto out; + res = -ENOENT; if (!IS_DEADDIR(inode)) { res = file->f_op->readdir(file, buf, filler); diff --git a/fs/reiserfs/bitmap.c b/fs/reiserfs/bitmap.c index 16b331dd991..f491ceb5af0 100644 --- a/fs/reiserfs/bitmap.c +++ b/fs/reiserfs/bitmap.c @@ -272,7 +272,7 @@ static inline int block_group_used(struct super_block *s, u32 id) /* If we don't have cached information on this bitmap block, we're * going to have to load it later anyway. Loading it here allows us - * to make a better decision. This favors long-term performace gain + * to make a better decision. This favors long-term performance gain * with a better on-disk layout vs. a short term gain of skipping the * read and potentially having a bad placement. */ if (info->free_count == UINT_MAX) { @@ -663,7 +663,7 @@ static inline void new_hashed_relocation(reiserfs_blocknr_hint_t * hint) /* * Relocation based on dirid, hashing them into a given bitmap block - * files. Formatted nodes are unaffected, a seperate policy covers them + * files. Formatted nodes are unaffected, a separate policy covers them */ static void dirid_groups(reiserfs_blocknr_hint_t * hint) { @@ -688,7 +688,7 @@ static void dirid_groups(reiserfs_blocknr_hint_t * hint) /* * Relocation based on oid, hashing them into a given bitmap block - * files. Formatted nodes are unaffected, a seperate policy covers them + * files. Formatted nodes are unaffected, a separate policy covers them */ static void oid_groups(reiserfs_blocknr_hint_t * hint) { diff --git a/fs/reiserfs/do_balan.c b/fs/reiserfs/do_balan.c index f85c5cf4934..7ee4208793b 100644 --- a/fs/reiserfs/do_balan.c +++ b/fs/reiserfs/do_balan.c @@ -283,7 +283,7 @@ static int balance_leaf(struct tree_balance *tb, struct item_head *ih, /* item h return balance_leaf_when_delete(tb, flag); zeros_num = 0; - if (flag == M_INSERT && body == 0) + if (flag == M_INSERT && !body) zeros_num = ih_item_len(ih); pos_in_item = tb->tb_path->pos_in_item; @@ -1728,7 +1728,7 @@ struct buffer_head *get_FEB(struct tree_balance *tb) struct buffer_info bi; for (i = 0; i < MAX_FEB_SIZE; i++) - if (tb->FEB[i] != 0) + if (tb->FEB[i] != NULL) break; if (i == MAX_FEB_SIZE) @@ -1827,7 +1827,7 @@ int get_left_neighbor_position(struct tree_balance *tb, int h) { int Sh_position = PATH_H_POSITION(tb->tb_path, h + 1); - RFALSE(PATH_H_PPARENT(tb->tb_path, h) == 0 || tb->FL[h] == 0, + RFALSE(PATH_H_PPARENT(tb->tb_path, h) == NULL || tb->FL[h] == NULL, "vs-12325: FL[%d](%p) or F[%d](%p) does not exist", h, tb->FL[h], h, PATH_H_PPARENT(tb->tb_path, h)); @@ -1841,7 +1841,7 @@ int get_right_neighbor_position(struct tree_balance *tb, int h) { int Sh_position = PATH_H_POSITION(tb->tb_path, h + 1); - RFALSE(PATH_H_PPARENT(tb->tb_path, h) == 0 || tb->FR[h] == 0, + RFALSE(PATH_H_PPARENT(tb->tb_path, h) == NULL || tb->FR[h] == NULL, "vs-12330: F[%d](%p) or FR[%d](%p) does not exist", h, PATH_H_PPARENT(tb->tb_path, h), h, tb->FR[h]); diff --git a/fs/reiserfs/fix_node.c b/fs/reiserfs/fix_node.c index 0ee35c6c9b7..07d05e0842b 100644 --- a/fs/reiserfs/fix_node.c +++ b/fs/reiserfs/fix_node.c @@ -153,7 +153,7 @@ static void create_virtual_node(struct tree_balance *tb, int h) if (vn->vn_mode == M_INSERT) { struct virtual_item *vi = vn->vn_vi + vn->vn_affected_item_num; - RFALSE(vn->vn_ins_ih == 0, + RFALSE(vn->vn_ins_ih == NULL, "vs-8040: item header of inserted item is not specified"); vi->vi_item_len = tb->insert_size[0]; vi->vi_ih = vn->vn_ins_ih; @@ -857,7 +857,8 @@ static int get_lfree(struct tree_balance *tb, int h) struct buffer_head *l, *f; int order; - if ((f = PATH_H_PPARENT(tb->tb_path, h)) == 0 || (l = tb->FL[h]) == 0) + if ((f = PATH_H_PPARENT(tb->tb_path, h)) == NULL || + (l = tb->FL[h]) == NULL) return 0; if (f == l) @@ -878,7 +879,8 @@ static int get_rfree(struct tree_balance *tb, int h) struct buffer_head *r, *f; int order; - if ((f = PATH_H_PPARENT(tb->tb_path, h)) == 0 || (r = tb->FR[h]) == 0) + if ((f = PATH_H_PPARENT(tb->tb_path, h)) == NULL || + (r = tb->FR[h]) == NULL) return 0; if (f == r) diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 231fd5ccadc..57917932212 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -1536,7 +1536,7 @@ static struct dentry *reiserfs_get_dentry(struct super_block *sb, if (!inode) inode = ERR_PTR(-ESTALE); if (IS_ERR(inode)) - return ERR_PTR(PTR_ERR(inode)); + return ERR_CAST(inode); result = d_alloc_anon(inode); if (!result) { iput(inode); @@ -2143,7 +2143,7 @@ int reiserfs_truncate_file(struct inode *p_s_inode, int update_timestamps) /* if we are not on a block boundary */ if (length) { length = blocksize - length; - zero_user_page(page, offset, length, KM_USER0); + zero_user(page, offset, length); if (buffer_mapped(bh) && bh->b_blocknr != 0) { mark_buffer_dirty(bh); } @@ -2367,7 +2367,7 @@ static int reiserfs_write_full_page(struct page *page, unlock_page(page); return 0; } - zero_user_page(page, last_offset, PAGE_CACHE_SIZE - last_offset, KM_USER0); + zero_user_segment(page, last_offset, PAGE_CACHE_SIZE); } bh = head; block = page->index << (PAGE_CACHE_SHIFT - s->s_blocksize_bits); diff --git a/fs/reiserfs/lbalance.c b/fs/reiserfs/lbalance.c index 281f8061ac5..6de060a6aa7 100644 --- a/fs/reiserfs/lbalance.c +++ b/fs/reiserfs/lbalance.c @@ -626,7 +626,7 @@ static void leaf_define_dest_src_infos(int shift_mode, struct tree_balance *tb, "vs-10250: leaf_define_dest_src_infos: shift type is unknown (%d)", shift_mode); } - RFALSE(src_bi->bi_bh == 0 || dest_bi->bi_bh == 0, + RFALSE(!src_bi->bi_bh || !dest_bi->bi_bh, "vs-10260: mode==%d, source (%p) or dest (%p) buffer is initialized incorrectly", shift_mode, src_bi->bi_bh, dest_bi->bi_bh); } diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c index b378eea332c..8867533cb72 100644 --- a/fs/reiserfs/namei.c +++ b/fs/reiserfs/namei.c @@ -452,7 +452,7 @@ static int reiserfs_add_entry(struct reiserfs_transaction_handle *th, buflen = DEH_SIZE + ROUND_UP(namelen); if (buflen > sizeof(small_buf)) { buffer = kmalloc(buflen, GFP_NOFS); - if (buffer == 0) + if (!buffer) return -ENOMEM; } else buffer = small_buf; diff --git a/fs/reiserfs/prints.c b/fs/reiserfs/prints.c index 5e7388b32d0..740bb8c0c1a 100644 --- a/fs/reiserfs/prints.c +++ b/fs/reiserfs/prints.c @@ -575,6 +575,8 @@ void print_block(struct buffer_head *bh, ...) //int print_mode, int first, int l printk ("Block %llu contains unformatted data\n", (unsigned long long)bh->b_blocknr); + + va_end(args); } static char print_tb_buf[2048]; diff --git a/fs/reiserfs/procfs.c b/fs/reiserfs/procfs.c index 00114462167..8f86c52b30d 100644 --- a/fs/reiserfs/procfs.c +++ b/fs/reiserfs/procfs.c @@ -444,7 +444,7 @@ static int r_show(struct seq_file *m, void *v) return show(m, v); } -static struct seq_operations r_ops = { +static const struct seq_operations r_ops = { .start = r_start, .next = r_next, .stop = r_stop, diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 5cd85fe5df5..393cc22c171 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -617,6 +617,7 @@ static const struct super_operations reiserfs_sops = { .unlockfs = reiserfs_unlockfs, .statfs = reiserfs_statfs, .remount_fs = reiserfs_remount, + .show_options = generic_show_options, #ifdef CONFIG_QUOTA .quota_read = reiserfs_quota_read, .quota_write = reiserfs_quota_write, @@ -1138,6 +1139,7 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg) unsigned long safe_mask = 0; unsigned int commit_max_age = (unsigned int)-1; struct reiserfs_journal *journal = SB_JOURNAL(s); + char *new_opts = kstrdup(arg, GFP_KERNEL); int err; #ifdef CONFIG_QUOTA int i; @@ -1153,7 +1155,8 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg) REISERFS_SB(s)->s_qf_names[i] = NULL; } #endif - return -EINVAL; + err = -EINVAL; + goto out_err; } handle_attrs(s); @@ -1191,9 +1194,9 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg) } if (blocks) { - int rc = reiserfs_resize(s, blocks); - if (rc != 0) - return rc; + err = reiserfs_resize(s, blocks); + if (err != 0) + goto out_err; } if (*mount_flags & MS_RDONLY) { @@ -1201,16 +1204,16 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg) /* remount read-only */ if (s->s_flags & MS_RDONLY) /* it is read-only already */ - return 0; + goto out_ok; /* try to remount file system with read-only permissions */ if (sb_umount_state(rs) == REISERFS_VALID_FS || REISERFS_SB(s)->s_mount_state != REISERFS_VALID_FS) { - return 0; + goto out_ok; } err = journal_begin(&th, s, 10); if (err) - return err; + goto out_err; /* Mounting a rw partition read-only. */ reiserfs_prepare_for_journal(s, SB_BUFFER_WITH_SB(s), 1); @@ -1220,11 +1223,13 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg) /* remount read-write */ if (!(s->s_flags & MS_RDONLY)) { reiserfs_xattr_init(s, *mount_flags); - return 0; /* We are read-write already */ + goto out_ok; /* We are read-write already */ } - if (reiserfs_is_journal_aborted(journal)) - return journal->j_errno; + if (reiserfs_is_journal_aborted(journal)) { + err = journal->j_errno; + goto out_err; + } handle_data_mode(s, mount_options); handle_barrier_mode(s, mount_options); @@ -1232,7 +1237,7 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg) s->s_flags &= ~MS_RDONLY; /* now it is safe to call journal_begin */ err = journal_begin(&th, s, 10); if (err) - return err; + goto out_err; /* Mount a partition which is read-only, read-write */ reiserfs_prepare_for_journal(s, SB_BUFFER_WITH_SB(s), 1); @@ -1247,7 +1252,7 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg) SB_JOURNAL(s)->j_must_wait = 1; err = journal_end(&th, s, 10); if (err) - return err; + goto out_err; s->s_dirt = 0; if (!(*mount_flags & MS_RDONLY)) { @@ -1255,7 +1260,14 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg) reiserfs_xattr_init(s, *mount_flags); } +out_ok: + kfree(s->s_options); + s->s_options = new_opts; return 0; + +out_err: + kfree(new_opts); + return err; } static int read_super_block(struct super_block *s, int offset) @@ -1559,6 +1571,8 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent) struct reiserfs_sb_info *sbi; int errval = -EINVAL; + save_mount_options(s, data); + sbi = kzalloc(sizeof(struct reiserfs_sb_info), GFP_KERNEL); if (!sbi) { errval = -ENOMEM; @@ -2012,29 +2026,29 @@ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id, if (err) return err; /* Quotafile not on the same filesystem? */ - if (nd.mnt->mnt_sb != sb) { - path_release(&nd); + if (nd.path.mnt->mnt_sb != sb) { + path_put(&nd.path); return -EXDEV; } /* We must not pack tails for quota files on reiserfs for quota IO to work */ - if (!REISERFS_I(nd.dentry->d_inode)->i_flags & i_nopack_mask) { + if (!(REISERFS_I(nd.path.dentry->d_inode)->i_flags & i_nopack_mask)) { reiserfs_warning(sb, "reiserfs: Quota file must have tail packing disabled."); - path_release(&nd); + path_put(&nd.path); return -EINVAL; } /* Not journalling quota? No more tests needed... */ if (!REISERFS_SB(sb)->s_qf_names[USRQUOTA] && !REISERFS_SB(sb)->s_qf_names[GRPQUOTA]) { - path_release(&nd); + path_put(&nd.path); return vfs_quota_on(sb, type, format_id, path); } /* Quotafile not of fs root? */ - if (nd.dentry->d_parent->d_inode != sb->s_root->d_inode) + if (nd.path.dentry->d_parent->d_inode != sb->s_root->d_inode) reiserfs_warning(sb, "reiserfs: Quota file not on filesystem root. " "Journalled quota will not work."); - path_release(&nd); + path_put(&nd.path); return vfs_quota_on(sb, type, format_id, path); } diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index 1597f6b649e..344b9b96cc5 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -155,7 +155,7 @@ static struct dentry *get_xa_file_dentry(const struct inode *inode, xadir = open_xa_dir(inode, flags); if (IS_ERR(xadir)) { - return ERR_PTR(PTR_ERR(xadir)); + return ERR_CAST(xadir); } else if (xadir && !xadir->d_inode) { dput(xadir); return ERR_PTR(-ENODATA); @@ -164,7 +164,7 @@ static struct dentry *get_xa_file_dentry(const struct inode *inode, xafile = lookup_one_len(name, xadir, strlen(name)); if (IS_ERR(xafile)) { dput(xadir); - return ERR_PTR(PTR_ERR(xafile)); + return ERR_CAST(xafile); } if (xafile->d_inode) { /* file exists */ @@ -191,28 +191,11 @@ static struct dentry *get_xa_file_dentry(const struct inode *inode, dput(xadir); if (err) xafile = ERR_PTR(err); - return xafile; -} - -/* Opens a file pointer to the attribute associated with inode */ -static struct file *open_xa_file(const struct inode *inode, const char *name, - int flags) -{ - struct dentry *xafile; - struct file *fp; - - xafile = get_xa_file_dentry(inode, name, flags); - if (IS_ERR(xafile)) - return ERR_PTR(PTR_ERR(xafile)); else if (!xafile->d_inode) { dput(xafile); - return ERR_PTR(-ENODATA); + xafile = ERR_PTR(-ENODATA); } - - fp = dentry_open(xafile, NULL, O_RDWR); - /* dentry_open dputs the dentry if it fails */ - - return fp; + return xafile; } /* @@ -228,9 +211,8 @@ static struct file *open_xa_file(const struct inode *inode, const char *name, * we're called with i_mutex held, so there are no worries about the directory * changing underneath us. */ -static int __xattr_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int __xattr_readdir(struct inode *inode, void *dirent, filldir_t filldir) { - struct inode *inode = filp->f_path.dentry->d_inode; struct cpu_key pos_key; /* key of current position in the directory (key of directory entry) */ INITIALIZE_PATH(path_to_entry); struct buffer_head *bh; @@ -374,23 +356,16 @@ static int __xattr_readdir(struct file *filp, void *dirent, filldir_t filldir) * */ static -int xattr_readdir(struct file *file, filldir_t filler, void *buf) +int xattr_readdir(struct inode *inode, filldir_t filler, void *buf) { - struct inode *inode = file->f_path.dentry->d_inode; - int res = -ENOTDIR; - if (!file->f_op || !file->f_op->readdir) - goto out; + int res = -ENOENT; mutex_lock_nested(&inode->i_mutex, I_MUTEX_XATTR); -// down(&inode->i_zombie); - res = -ENOENT; if (!IS_DEADDIR(inode)) { lock_kernel(); - res = __xattr_readdir(file, buf, filler); + res = __xattr_readdir(inode, buf, filler); unlock_kernel(); } -// up(&inode->i_zombie); mutex_unlock(&inode->i_mutex); - out: return res; } @@ -442,7 +417,7 @@ reiserfs_xattr_set(struct inode *inode, const char *name, const void *buffer, size_t buffer_size, int flags) { int err = 0; - struct file *fp; + struct dentry *dentry; struct page *page; char *data; struct address_space *mapping; @@ -460,18 +435,18 @@ reiserfs_xattr_set(struct inode *inode, const char *name, const void *buffer, xahash = xattr_hash(buffer, buffer_size); open_file: - fp = open_xa_file(inode, name, flags); - if (IS_ERR(fp)) { - err = PTR_ERR(fp); + dentry = get_xa_file_dentry(inode, name, flags); + if (IS_ERR(dentry)) { + err = PTR_ERR(dentry); goto out; } - xinode = fp->f_path.dentry->d_inode; + xinode = dentry->d_inode; REISERFS_I(inode)->i_flags |= i_has_xattr_dir; /* we need to copy it off.. */ if (xinode->i_nlink > 1) { - fput(fp); + dput(dentry); err = reiserfs_xattr_del(inode, name); if (err < 0) goto out; @@ -485,7 +460,7 @@ reiserfs_xattr_set(struct inode *inode, const char *name, const void *buffer, newattrs.ia_size = buffer_size; newattrs.ia_valid = ATTR_SIZE | ATTR_CTIME; mutex_lock_nested(&xinode->i_mutex, I_MUTEX_XATTR); - err = notify_change(fp->f_path.dentry, &newattrs); + err = notify_change(dentry, &newattrs); if (err) goto out_filp; @@ -518,15 +493,14 @@ reiserfs_xattr_set(struct inode *inode, const char *name, const void *buffer, rxh->h_hash = cpu_to_le32(xahash); } - err = reiserfs_prepare_write(fp, page, page_offset, + err = reiserfs_prepare_write(NULL, page, page_offset, page_offset + chunk + skip); if (!err) { if (buffer) memcpy(data + skip, buffer + buffer_pos, chunk); - err = - reiserfs_commit_write(fp, page, page_offset, - page_offset + chunk + - skip); + err = reiserfs_commit_write(NULL, page, page_offset, + page_offset + chunk + + skip); } unlock_page(page); reiserfs_put_page(page); @@ -548,7 +522,7 @@ reiserfs_xattr_set(struct inode *inode, const char *name, const void *buffer, out_filp: mutex_unlock(&xinode->i_mutex); - fput(fp); + dput(dentry); out: return err; @@ -562,7 +536,7 @@ reiserfs_xattr_get(const struct inode *inode, const char *name, void *buffer, size_t buffer_size) { ssize_t err = 0; - struct file *fp; + struct dentry *dentry; size_t isize; size_t file_pos = 0; size_t buffer_pos = 0; @@ -578,13 +552,13 @@ reiserfs_xattr_get(const struct inode *inode, const char *name, void *buffer, if (get_inode_sd_version(inode) == STAT_DATA_V1) return -EOPNOTSUPP; - fp = open_xa_file(inode, name, FL_READONLY); - if (IS_ERR(fp)) { - err = PTR_ERR(fp); + dentry = get_xa_file_dentry(inode, name, FL_READONLY); + if (IS_ERR(dentry)) { + err = PTR_ERR(dentry); goto out; } - xinode = fp->f_path.dentry->d_inode; + xinode = dentry->d_inode; isize = xinode->i_size; REISERFS_I(inode)->i_flags |= i_has_xattr_dir; @@ -652,7 +626,7 @@ reiserfs_xattr_get(const struct inode *inode, const char *name, void *buffer, } out_dput: - fput(fp); + dput(dentry); out: return err; @@ -742,7 +716,6 @@ reiserfs_delete_xattrs_filler(void *buf, const char *name, int namelen, /* This is called w/ inode->i_mutex downed */ int reiserfs_delete_xattrs(struct inode *inode) { - struct file *fp; struct dentry *dir, *root; int err = 0; @@ -763,15 +736,8 @@ int reiserfs_delete_xattrs(struct inode *inode) return 0; } - fp = dentry_open(dir, NULL, O_RDWR); - if (IS_ERR(fp)) { - err = PTR_ERR(fp); - /* dentry_open dputs the dentry if it fails */ - goto out; - } - lock_kernel(); - err = xattr_readdir(fp, reiserfs_delete_xattrs_filler, dir); + err = xattr_readdir(dir->d_inode, reiserfs_delete_xattrs_filler, dir); if (err) { unlock_kernel(); goto out_dir; @@ -791,7 +757,7 @@ int reiserfs_delete_xattrs(struct inode *inode) unlock_kernel(); out_dir: - fput(fp); + dput(dir); out: if (!err) @@ -833,7 +799,6 @@ reiserfs_chown_xattrs_filler(void *buf, const char *name, int namelen, int reiserfs_chown_xattrs(struct inode *inode, struct iattr *attrs) { - struct file *fp; struct dentry *dir; int err = 0; struct reiserfs_chown_buf buf; @@ -857,13 +822,6 @@ int reiserfs_chown_xattrs(struct inode *inode, struct iattr *attrs) goto out; } - fp = dentry_open(dir, NULL, O_RDWR); - if (IS_ERR(fp)) { - err = PTR_ERR(fp); - /* dentry_open dputs the dentry if it fails */ - goto out; - } - lock_kernel(); attrs->ia_valid &= (ATTR_UID | ATTR_GID | ATTR_CTIME); @@ -871,7 +829,7 @@ int reiserfs_chown_xattrs(struct inode *inode, struct iattr *attrs) buf.attrs = attrs; buf.inode = inode; - err = xattr_readdir(fp, reiserfs_chown_xattrs_filler, &buf); + err = xattr_readdir(dir->d_inode, reiserfs_chown_xattrs_filler, &buf); if (err) { unlock_kernel(); goto out_dir; @@ -881,7 +839,7 @@ int reiserfs_chown_xattrs(struct inode *inode, struct iattr *attrs) unlock_kernel(); out_dir: - fput(fp); + dput(dir); out: attrs->ia_valid = ia_valid; @@ -1029,7 +987,6 @@ reiserfs_listxattr_filler(void *buf, const char *name, int namelen, */ ssize_t reiserfs_listxattr(struct dentry * dentry, char *buffer, size_t size) { - struct file *fp; struct dentry *dir; int err = 0; struct reiserfs_listxattr_buf buf; @@ -1052,13 +1009,6 @@ ssize_t reiserfs_listxattr(struct dentry * dentry, char *buffer, size_t size) goto out; } - fp = dentry_open(dir, NULL, O_RDWR); - if (IS_ERR(fp)) { - err = PTR_ERR(fp); - /* dentry_open dputs the dentry if it fails */ - goto out; - } - buf.r_buf = buffer; buf.r_size = buffer ? size : 0; buf.r_pos = 0; @@ -1066,7 +1016,7 @@ ssize_t reiserfs_listxattr(struct dentry * dentry, char *buffer, size_t size) REISERFS_I(dentry->d_inode)->i_flags |= i_has_xattr_dir; - err = xattr_readdir(fp, reiserfs_listxattr_filler, &buf); + err = xattr_readdir(dir->d_inode, reiserfs_listxattr_filler, &buf); if (err) goto out_dir; @@ -1076,7 +1026,7 @@ ssize_t reiserfs_listxattr(struct dentry * dentry, char *buffer, size_t size) err = buf.r_pos; out_dir: - fput(fp); + dput(dir); out: reiserfs_read_unlock_xattr_i(dentry->d_inode); @@ -1084,7 +1034,7 @@ ssize_t reiserfs_listxattr(struct dentry * dentry, char *buffer, size_t size) } /* This is the implementation for the xattr plugin infrastructure */ -static struct list_head xattr_handlers = LIST_HEAD_INIT(xattr_handlers); +static LIST_HEAD(xattr_handlers); static DEFINE_RWLOCK(handler_lock); static struct reiserfs_xattr_handler *find_xattr_handler_prefix(const char diff --git a/fs/romfs/inode.c b/fs/romfs/inode.c index a49cf5b9a19..3f13d491c7c 100644 --- a/fs/romfs/inode.c +++ b/fs/romfs/inode.c @@ -84,6 +84,8 @@ struct romfs_inode_info { struct inode vfs_inode; }; +static struct inode *romfs_iget(struct super_block *, unsigned long); + /* instead of private superblock data */ static inline unsigned long romfs_maxsize(struct super_block *sb) { @@ -117,7 +119,7 @@ static int romfs_fill_super(struct super_block *s, void *data, int silent) struct buffer_head *bh; struct romfs_super_block *rsb; struct inode *root; - int sz; + int sz, ret = -EINVAL; /* I would parse the options here, but there are none.. :) */ @@ -157,10 +159,13 @@ static int romfs_fill_super(struct super_block *s, void *data, int silent) & ROMFH_MASK; s->s_op = &romfs_ops; - root = iget(s, sz); - if (!root) + root = romfs_iget(s, sz); + if (IS_ERR(root)) { + ret = PTR_ERR(root); goto out; + } + ret = -ENOMEM; s->s_root = d_alloc_root(root); if (!s->s_root) goto outiput; @@ -173,7 +178,7 @@ outiput: out: brelse(bh); outnobh: - return -EINVAL; + return ret; } /* That's simple too. */ @@ -335,8 +340,9 @@ static struct dentry * romfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) { unsigned long offset, maxoff; - int fslen, res; - struct inode *inode; + long res; + int fslen; + struct inode *inode = NULL; char fsname[ROMFS_MAXFN]; /* XXX dynamic? */ struct romfs_inode ri; const char *name; /* got from dentry */ @@ -346,7 +352,7 @@ romfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) offset = dir->i_ino & ROMFH_MASK; lock_kernel(); if (romfs_copyfrom(dir, &ri, offset, ROMFH_SIZE) <= 0) - goto out; + goto error; maxoff = romfs_maxsize(dir->i_sb); offset = be32_to_cpu(ri.spec) & ROMFH_MASK; @@ -359,9 +365,9 @@ romfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) for(;;) { if (!offset || offset >= maxoff) - goto out0; + goto success; /* negative success */ if (romfs_copyfrom(dir, &ri, offset, ROMFH_SIZE) <= 0) - goto out; + goto error; /* try to match the first 16 bytes of name */ fslen = romfs_strnlen(dir, offset+ROMFH_SIZE, ROMFH_SIZE); @@ -389,23 +395,17 @@ romfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) if ((be32_to_cpu(ri.next) & ROMFH_TYPE) == ROMFH_HRD) offset = be32_to_cpu(ri.spec) & ROMFH_MASK; - if ((inode = iget(dir->i_sb, offset))) - goto outi; - - /* - * it's a bit funky, _lookup needs to return an error code - * (negative) or a NULL, both as a dentry. ENOENT should not - * be returned, instead we need to create a negative dentry by - * d_add(dentry, NULL); and return 0 as no error. - * (Although as I see, it only matters on writable file - * systems). - */ - -out0: inode = NULL; -outi: res = 0; - d_add (dentry, inode); + inode = romfs_iget(dir->i_sb, offset); + if (IS_ERR(inode)) { + res = PTR_ERR(inode); + goto error; + } -out: unlock_kernel(); +success: + d_add(dentry, inode); + res = 0; +error: + unlock_kernel(); return ERR_PTR(res); } @@ -478,20 +478,29 @@ static mode_t romfs_modemap[] = S_IFBLK+0600, S_IFCHR+0600, S_IFSOCK+0644, S_IFIFO+0644 }; -static void -romfs_read_inode(struct inode *i) +static struct inode * +romfs_iget(struct super_block *sb, unsigned long ino) { - int nextfh, ino; + int nextfh; struct romfs_inode ri; + struct inode *i; + + ino &= ROMFH_MASK; + i = iget_locked(sb, ino); + if (!i) + return ERR_PTR(-ENOMEM); + if (!(i->i_state & I_NEW)) + return i; - ino = i->i_ino & ROMFH_MASK; i->i_mode = 0; /* Loop for finding the real hard link */ for(;;) { if (romfs_copyfrom(i, &ri, ino, ROMFH_SIZE) <= 0) { - printk("romfs: read error for inode 0x%x\n", ino); - return; + printk(KERN_ERR "romfs: read error for inode 0x%lx\n", + ino); + iget_failed(i); + return ERR_PTR(-EIO); } /* XXX: do romfs_checksum here too (with name) */ @@ -548,6 +557,8 @@ romfs_read_inode(struct inode *i) init_special_inode(i, ino, MKDEV(nextfh>>16,nextfh&0xffff)); } + unlock_new_inode(i); + return i; } static struct kmem_cache * romfs_inode_cachep; @@ -599,7 +610,6 @@ static int romfs_remount(struct super_block *sb, int *flags, char *data) static const struct super_operations romfs_ops = { .alloc_inode = romfs_alloc_inode, .destroy_inode = romfs_destroy_inode, - .read_inode = romfs_read_inode, .statfs = romfs_statfs, .remount_fs = romfs_remount, }; diff --git a/fs/select.c b/fs/select.c index 47f47925aea..5633fe98078 100644 --- a/fs/select.c +++ b/fs/select.c @@ -739,7 +739,7 @@ asmlinkage long sys_poll(struct pollfd __user *ufds, unsigned int nfds, timeout_jiffies = -1; else #endif - timeout_jiffies = msecs_to_jiffies(timeout_msecs); + timeout_jiffies = msecs_to_jiffies(timeout_msecs) + 1; } else { /* Infinite (< 0) or no (0) timeout */ timeout_jiffies = timeout_msecs; diff --git a/fs/seq_file.c b/fs/seq_file.c index ca71c115bda..853770274f2 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -342,13 +342,11 @@ int seq_printf(struct seq_file *m, const char *f, ...) } EXPORT_SYMBOL(seq_printf); -int seq_path(struct seq_file *m, - struct vfsmount *mnt, struct dentry *dentry, - char *esc) +int seq_path(struct seq_file *m, struct path *path, char *esc) { if (m->count < m->size) { char *s = m->buf + m->count; - char *p = d_path(dentry, mnt, s, m->size - m->count); + char *p = d_path(path, s, m->size - m->count); if (!IS_ERR(p)) { while (s <= p) { char c = *p++; diff --git a/fs/signalfd.c b/fs/signalfd.c index fb7f7e8034d..8ead0db3593 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -27,6 +27,7 @@ #include <linux/list.h> #include <linux/anon_inodes.h> #include <linux/signalfd.h> +#include <linux/syscalls.h> struct signalfd_ctx { sigset_t sigmask; @@ -66,7 +67,7 @@ static int signalfd_copyinfo(struct signalfd_siginfo __user *uinfo, BUILD_BUG_ON(sizeof(struct signalfd_siginfo) != 128); /* - * Unused memebers should be zero ... + * Unused members should be zero ... */ err = __clear_user(uinfo, sizeof(*uinfo)); @@ -110,9 +111,14 @@ static int signalfd_copyinfo(struct signalfd_siginfo __user *uinfo, err |= __put_user(kinfo->si_uid, &uinfo->ssi_uid); err |= __put_user((long) kinfo->si_ptr, &uinfo->ssi_ptr); break; - default: /* this is just in case for now ... */ + default: + /* + * This case catches also the signals queued by sigqueue(). + */ err |= __put_user(kinfo->si_pid, &uinfo->ssi_pid); err |= __put_user(kinfo->si_uid, &uinfo->ssi_uid); + err |= __put_user((long) kinfo->si_ptr, &uinfo->ssi_ptr); + err |= __put_user(kinfo->si_int, &uinfo->ssi_int); break; } diff --git a/fs/smbfs/Makefile b/fs/smbfs/Makefile index 6673ee82cb4..4faf8c4722c 100644 --- a/fs/smbfs/Makefile +++ b/fs/smbfs/Makefile @@ -16,23 +16,3 @@ EXTRA_CFLAGS += -DSMBFS_PARANOIA #EXTRA_CFLAGS += -DDEBUG_SMB_TIMESTAMP #EXTRA_CFLAGS += -Werror -# -# Maintainer rules -# - -# getopt.c not included. It is intentionally separate -SRC = proc.c dir.c cache.c sock.c inode.c file.c ioctl.c smbiod.c request.c \ - symlink.c - -proto: - -rm -f proto.h - @echo > proto2.h "/*" - @echo >> proto2.h " * Autogenerated with cproto on: " `date` - @echo >> proto2.h " */" - @echo >> proto2.h "" - @echo >> proto2.h "struct smb_request;" - @echo >> proto2.h "struct sock;" - @echo >> proto2.h "struct statfs;" - @echo >> proto2.h "" - cproto -E "gcc -E" -e -v -I $(TOPDIR)/include -DMAKING_PROTO -D__KERNEL__ $(SRC) >> proto2.h - mv proto2.h proto.h diff --git a/fs/smbfs/inode.c b/fs/smbfs/inode.c index 9416ead0c7a..376ef3ee6ed 100644 --- a/fs/smbfs/inode.c +++ b/fs/smbfs/inode.c @@ -500,6 +500,13 @@ static int smb_fill_super(struct super_block *sb, void *raw_data, int silent) struct smb_fattr root; int ver; void *mem; + static int warn_count; + + if (warn_count < 5) { + warn_count++; + printk(KERN_EMERG "smbfs is deprecated and will be removed" + " from the 2.6.27 kernel. Please migrate to cifs\n"); + } if (!raw_data) goto out_no_data; diff --git a/fs/smbfs/request.c b/fs/smbfs/request.c index ca4b2d59c0c..45f45933e86 100644 --- a/fs/smbfs/request.c +++ b/fs/smbfs/request.c @@ -105,7 +105,7 @@ struct smb_request *smb_alloc_request(struct smb_sb_info *server, int bufsize) if (nfs_try_to_free_pages(server)) continue; - if (signalled() && (server->flags & NFS_MOUNT_INTR)) + if (fatal_signal_pending(current)) return ERR_PTR(-ERESTARTSYS); current->policy = SCHED_YIELD; schedule(); diff --git a/fs/smbfs/smbiod.c b/fs/smbfs/smbiod.c index fae8e85af0e..6bd9b691a46 100644 --- a/fs/smbfs/smbiod.c +++ b/fs/smbfs/smbiod.c @@ -206,7 +206,7 @@ int smbiod_retry(struct smb_sb_info *server) smb_close_socket(server); - if (pid == 0) { + if (!pid) { /* FIXME: this is fatal, umount? */ printk(KERN_ERR "smb_retry: no connection process\n"); server->state = CONN_RETRIED; diff --git a/fs/smbfs/sock.c b/fs/smbfs/sock.c index e48bd8235a8..e37fe4deebd 100644 --- a/fs/smbfs/sock.c +++ b/fs/smbfs/sock.c @@ -329,9 +329,8 @@ smb_receive(struct smb_sb_info *server, struct smb_request *req) msg.msg_control = NULL; /* Dont repeat bytes and count available bufferspace */ - rlen = smb_move_iov(&p, &num, iov, req->rq_bytes_recvd); - if (req->rq_rlen < rlen) - rlen = req->rq_rlen; + rlen = min_t(int, smb_move_iov(&p, &num, iov, req->rq_bytes_recvd), + (req->rq_rlen - req->rq_bytes_recvd)); result = kernel_recvmsg(sock, &msg, p, num, rlen, flags); diff --git a/fs/splice.c b/fs/splice.c index 6bdcb6107bc..eeb1a86a701 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -254,11 +254,16 @@ ssize_t splice_to_pipe(struct pipe_inode_info *pipe, } while (page_nr < spd_pages) - page_cache_release(spd->pages[page_nr++]); + spd->spd_release(spd, page_nr++); return ret; } +static void spd_release_page(struct splice_pipe_desc *spd, unsigned int i) +{ + page_cache_release(spd->pages[i]); +} + static int __generic_file_splice_read(struct file *in, loff_t *ppos, struct pipe_inode_info *pipe, size_t len, @@ -277,6 +282,7 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, .partial = partial, .flags = flags, .ops = &page_cache_pipe_buf_ops, + .spd_release = spd_release_page, }; index = *ppos >> PAGE_CACHE_SHIFT; @@ -314,7 +320,7 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, break; error = add_to_page_cache_lru(page, mapping, index, - GFP_KERNEL); + mapping_gfp_mask(mapping)); if (unlikely(error)) { page_cache_release(page); if (error == -EEXIST) @@ -364,8 +370,10 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, * for an in-flight io page */ if (flags & SPLICE_F_NONBLOCK) { - if (TestSetPageLocked(page)) + if (TestSetPageLocked(page)) { + error = -EAGAIN; break; + } } else lock_page(page); @@ -473,9 +481,8 @@ ssize_t generic_file_splice_read(struct file *in, loff_t *ppos, struct pipe_inode_info *pipe, size_t len, unsigned int flags) { - ssize_t spliced; - int ret; loff_t isize, left; + int ret; isize = i_size_read(in->f_mapping->host); if (unlikely(*ppos >= isize)) @@ -485,29 +492,9 @@ ssize_t generic_file_splice_read(struct file *in, loff_t *ppos, if (unlikely(left < len)) len = left; - ret = 0; - spliced = 0; - while (len && !spliced) { - ret = __generic_file_splice_read(in, ppos, pipe, len, flags); - - if (ret < 0) - break; - else if (!ret) { - if (spliced) - break; - if (flags & SPLICE_F_NONBLOCK) { - ret = -EAGAIN; - break; - } - } - + ret = __generic_file_splice_read(in, ppos, pipe, len, flags); + if (ret > 0) *ppos += ret; - len -= ret; - spliced += ret; - } - - if (spliced) - return spliced; return ret; } @@ -908,10 +895,6 @@ static long do_splice_from(struct pipe_inode_info *pipe, struct file *out, if (unlikely(ret < 0)) return ret; - ret = security_file_permission(out, MAY_WRITE); - if (unlikely(ret < 0)) - return ret; - return out->f_op->splice_write(pipe, out, ppos, len, flags); } @@ -934,10 +917,6 @@ static long do_splice_to(struct file *in, loff_t *ppos, if (unlikely(ret < 0)) return ret; - ret = security_file_permission(in, MAY_READ); - if (unlikely(ret < 0)) - return ret; - return in->f_op->splice_read(in, ppos, pipe, len, flags); } @@ -1033,7 +1012,9 @@ ssize_t splice_direct_to_actor(struct file *in, struct splice_desc *sd, goto out_release; } +done: pipe->nrbufs = pipe->curbuf = 0; + file_accessed(in); return bytes; out_release: @@ -1049,16 +1030,11 @@ out_release: buf->ops = NULL; } } - pipe->nrbufs = pipe->curbuf = 0; - /* - * If we transferred some data, return the number of bytes: - */ - if (bytes > 0) - return bytes; - - return ret; + if (!bytes) + bytes = ret; + goto done; } EXPORT_SYMBOL(splice_direct_to_actor); @@ -1184,6 +1160,9 @@ static int copy_from_user_mmap_sem(void *dst, const void __user *src, size_t n) { int partial; + if (!access_ok(VERIFY_READ, src, n)) + return -EFAULT; + pagefault_disable(); partial = __copy_from_user_inatomic(dst, src, n); pagefault_enable(); @@ -1236,7 +1215,7 @@ static int get_iovec_page_array(const struct iovec __user *iov, if (unlikely(!len)) break; error = -EFAULT; - if (unlikely(!base)) + if (!access_ok(VERIFY_READ, base, len)) break; /* @@ -1392,6 +1371,11 @@ static long vmsplice_to_user(struct file *file, const struct iovec __user *iov, break; } + if (unlikely(!access_ok(VERIFY_WRITE, base, len))) { + error = -EFAULT; + break; + } + sd.len = 0; sd.total_len = len; sd.flags = flags; @@ -1440,6 +1424,7 @@ static long vmsplice_to_pipe(struct file *file, const struct iovec __user *iov, .partial = partial, .flags = flags, .ops = &user_page_pipe_buf_ops, + .spd_release = spd_release_page, }; pipe = pipe_info(file->f_path.dentry->d_inode); @@ -1665,6 +1650,13 @@ static int link_pipe(struct pipe_inode_info *ipipe, i++; } while (len); + /* + * return EAGAIN if we have the potential of some data in the + * future, otherwise just return 0 + */ + if (!ret && ipipe->waiting_writers && (flags & SPLICE_F_NONBLOCK)) + ret = -EAGAIN; + inode_double_unlock(ipipe->inode, opipe->inode); /* @@ -1705,11 +1697,8 @@ static long do_tee(struct file *in, struct file *out, size_t len, ret = link_ipipe_prep(ipipe, flags); if (!ret) { ret = link_opipe_prep(opipe, flags); - if (!ret) { + if (!ret) ret = link_pipe(ipipe, opipe, len, flags); - if (!ret && (flags & SPLICE_F_NONBLOCK)) - ret = -EAGAIN; - } } } diff --git a/fs/stat.c b/fs/stat.c index 68510068a64..9cf41f719d5 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -62,8 +62,8 @@ int vfs_stat_fd(int dfd, char __user *name, struct kstat *stat) error = __user_walk_fd(dfd, name, LOOKUP_FOLLOW, &nd); if (!error) { - error = vfs_getattr(nd.mnt, nd.dentry, stat); - path_release(&nd); + error = vfs_getattr(nd.path.mnt, nd.path.dentry, stat); + path_put(&nd.path); } return error; } @@ -82,8 +82,8 @@ int vfs_lstat_fd(int dfd, char __user *name, struct kstat *stat) error = __user_walk_fd(dfd, name, 0, &nd); if (!error) { - error = vfs_getattr(nd.mnt, nd.dentry, stat); - path_release(&nd); + error = vfs_getattr(nd.path.mnt, nd.path.dentry, stat); + path_put(&nd.path); } return error; } @@ -302,17 +302,18 @@ asmlinkage long sys_readlinkat(int dfd, const char __user *path, error = __user_walk_fd(dfd, path, 0, &nd); if (!error) { - struct inode * inode = nd.dentry->d_inode; + struct inode *inode = nd.path.dentry->d_inode; error = -EINVAL; if (inode->i_op && inode->i_op->readlink) { - error = security_inode_readlink(nd.dentry); + error = security_inode_readlink(nd.path.dentry); if (!error) { - touch_atime(nd.mnt, nd.dentry); - error = inode->i_op->readlink(nd.dentry, buf, bufsiz); + touch_atime(nd.path.mnt, nd.path.dentry); + error = inode->i_op->readlink(nd.path.dentry, + buf, bufsiz); } } - path_release(&nd); + path_put(&nd.path); } return error; } diff --git a/fs/super.c b/fs/super.c index ceaf2e3d594..09008dbd264 100644 --- a/fs/super.c +++ b/fs/super.c @@ -105,6 +105,7 @@ static inline void destroy_super(struct super_block *s) { security_sb_free(s); kfree(s->s_subtype); + kfree(s->s_options); kfree(s); } @@ -555,11 +556,11 @@ out: } /** - * mark_files_ro + * mark_files_ro - mark all files read-only * @sb: superblock in question * - * All files are marked read/only. We don't care about pending - * delete files so this should be used in 'force' mode only + * All files are marked read-only. We don't care about pending + * delete files so this should be used in 'force' mode only. */ static void mark_files_ro(struct super_block *sb) @@ -603,6 +604,7 @@ int do_remount_sb(struct super_block *sb, int flags, void *data, int force) mark_files_ro(sb); else if (!fs_may_remount_ro(sb)) return -EBUSY; + DQUOT_OFF(sb); } if (sb->s_op->remount_fs) { @@ -868,12 +870,12 @@ vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void if (!mnt) goto out; - if (data) { + if (data && !(type->fs_flags & FS_BINARY_MOUNTDATA)) { secdata = alloc_secdata(); if (!secdata) goto out_mnt; - error = security_sb_copy_data(type, data, secdata); + error = security_sb_copy_data(data, secdata); if (error) goto out_free_secdata; } @@ -943,6 +945,7 @@ do_kern_mount(const char *fstype, int flags, const char *name, void *data) put_filesystem(type); return mnt; } +EXPORT_SYMBOL_GPL(do_kern_mount); struct vfsmount *kern_mount_data(struct file_system_type *type, void *data) { diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index f281cc6584b..4948d9bc405 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -440,7 +440,7 @@ int sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd) /** * sysfs_remove_one - remove sysfs_dirent from parent * @acxt: addrm context to use - * @sd: sysfs_dirent to be added + * @sd: sysfs_dirent to be removed * * Mark @sd removed and drop nlink of parent inode if @sd is a * directory. @sd is unlinked from the children list. diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 4045bdcc4b3..baa663e6938 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -12,6 +12,7 @@ #include <linux/module.h> #include <linux/kobject.h> +#include <linux/kallsyms.h> #include <linux/namei.h> #include <linux/poll.h> #include <linux/list.h> @@ -20,43 +21,6 @@ #include "sysfs.h" -#define to_sattr(a) container_of(a,struct subsys_attribute, attr) - -/* - * Subsystem file operations. - * These operations allow subsystems to have files that can be - * read/written. - */ -static ssize_t -subsys_attr_show(struct kobject * kobj, struct attribute * attr, char * page) -{ - struct kset *kset = to_kset(kobj); - struct subsys_attribute * sattr = to_sattr(attr); - ssize_t ret = -EIO; - - if (sattr->show) - ret = sattr->show(kset, page); - return ret; -} - -static ssize_t -subsys_attr_store(struct kobject * kobj, struct attribute * attr, - const char * page, size_t count) -{ - struct kset *kset = to_kset(kobj); - struct subsys_attribute * sattr = to_sattr(attr); - ssize_t ret = -EIO; - - if (sattr->store) - ret = sattr->store(kset, page, count); - return ret; -} - -static struct sysfs_ops subsys_sysfs_ops = { - .show = subsys_attr_show, - .store = subsys_attr_store, -}; - /* * There's one sysfs_buffer for each open file and one * sysfs_open_dirent for each sysfs_dirent with one or more open @@ -66,7 +30,7 @@ static struct sysfs_ops subsys_sysfs_ops = { * sysfs_dirent->s_attr.open points to sysfs_open_dirent. s_attr.open * is protected by sysfs_open_dirent_lock. */ -static spinlock_t sysfs_open_dirent_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(sysfs_open_dirent_lock); struct sysfs_open_dirent { atomic_t refcnt; @@ -123,7 +87,12 @@ static int fill_read_buffer(struct dentry * dentry, struct sysfs_buffer * buffer * The code works fine with PAGE_SIZE return but it's likely to * indicate truncated result or overflow in normal use cases. */ - BUG_ON(count >= (ssize_t)PAGE_SIZE); + if (count >= (ssize_t)PAGE_SIZE) { + print_symbol("fill_read_buffer: %s returned bad count\n", + (unsigned long)ops->show); + /* Try to struggle along */ + count = PAGE_SIZE - 1; + } if (count >= 0) { buffer->needs_read_fill = 0; buffer->count = count; @@ -354,31 +323,23 @@ static int sysfs_open_file(struct inode *inode, struct file *file) { struct sysfs_dirent *attr_sd = file->f_path.dentry->d_fsdata; struct kobject *kobj = attr_sd->s_parent->s_dir.kobj; - struct sysfs_buffer * buffer; - struct sysfs_ops * ops = NULL; - int error; + struct sysfs_buffer *buffer; + struct sysfs_ops *ops; + int error = -EACCES; /* need attr_sd for attr and ops, its parent for kobj */ if (!sysfs_get_active_two(attr_sd)) return -ENODEV; - /* if the kobject has no ktype, then we assume that it is a subsystem - * itself, and use ops for it. - */ - if (kobj->kset && kobj->kset->ktype) - ops = kobj->kset->ktype->sysfs_ops; - else if (kobj->ktype) + /* every kobject with an attribute needs a ktype assigned */ + if (kobj->ktype && kobj->ktype->sysfs_ops) ops = kobj->ktype->sysfs_ops; - else - ops = &subsys_sysfs_ops; - - error = -EACCES; - - /* No sysfs operations, either from having no subsystem, - * or the subsystem have no operations. - */ - if (!ops) + else { + printk(KERN_ERR "missing sysfs attribute operations for " + "kobject: %s\n", kobject_name(kobj)); + WARN_ON(1); goto err_out; + } /* File needs write support. * The inode's perms must say it's ok, @@ -568,7 +529,11 @@ int sysfs_add_file_to_group(struct kobject *kobj, struct sysfs_dirent *dir_sd; int error; - dir_sd = sysfs_get_dirent(kobj->sd, group); + if (group) + dir_sd = sysfs_get_dirent(kobj->sd, group); + else + dir_sd = sysfs_get(kobj->sd); + if (!dir_sd) return -ENOENT; @@ -656,7 +621,10 @@ void sysfs_remove_file_from_group(struct kobject *kobj, { struct sysfs_dirent *dir_sd; - dir_sd = sysfs_get_dirent(kobj->sd, group); + if (group) + dir_sd = sysfs_get_dirent(kobj->sd, group); + else + dir_sd = sysfs_get(kobj->sd); if (dir_sd) { sysfs_hash_and_remove(dir_sd, attr->name); sysfs_put(dir_sd); diff --git a/fs/sysfs/group.c b/fs/sysfs/group.c index d1972374655..47790491503 100644 --- a/fs/sysfs/group.c +++ b/fs/sysfs/group.c @@ -16,25 +16,31 @@ #include "sysfs.h" -static void remove_files(struct sysfs_dirent *dir_sd, +static void remove_files(struct sysfs_dirent *dir_sd, struct kobject *kobj, const struct attribute_group *grp) { struct attribute *const* attr; + int i; - for (attr = grp->attrs; *attr; attr++) - sysfs_hash_and_remove(dir_sd, (*attr)->name); + for (i = 0, attr = grp->attrs; *attr; i++, attr++) + if (!grp->is_visible || + grp->is_visible(kobj, *attr, i)) + sysfs_hash_and_remove(dir_sd, (*attr)->name); } -static int create_files(struct sysfs_dirent *dir_sd, +static int create_files(struct sysfs_dirent *dir_sd, struct kobject *kobj, const struct attribute_group *grp) { struct attribute *const* attr; - int error = 0; + int error = 0, i; - for (attr = grp->attrs; *attr && !error; attr++) - error = sysfs_add_file(dir_sd, *attr, SYSFS_KOBJ_ATTR); + for (i = 0, attr = grp->attrs; *attr && !error; i++, attr++) + if (!grp->is_visible || + grp->is_visible(kobj, *attr, i)) + error |= + sysfs_add_file(dir_sd, *attr, SYSFS_KOBJ_ATTR); if (error) - remove_files(dir_sd, grp); + remove_files(dir_sd, kobj, grp); return error; } @@ -54,7 +60,7 @@ int sysfs_create_group(struct kobject * kobj, } else sd = kobj->sd; sysfs_get(sd); - error = create_files(sd, grp); + error = create_files(sd, kobj, grp); if (error) { if (grp->name) sysfs_remove_subdir(sd); @@ -71,11 +77,16 @@ void sysfs_remove_group(struct kobject * kobj, if (grp->name) { sd = sysfs_get_dirent(dir_sd, grp->name); - BUG_ON(!sd); + if (!sd) { + printk(KERN_WARNING "sysfs group %p not found for " + "kobject '%s'\n", grp, kobject_name(kobj)); + WARN_ON(!sd); + return; + } } else sd = sysfs_get(dir_sd); - remove_files(sd, grp); + remove_files(sd, kobj, grp); if (grp->name) sysfs_remove_subdir(sd); diff --git a/fs/sysfs/symlink.c b/fs/sysfs/symlink.c index 3eac20c63c4..5f66c446615 100644 --- a/fs/sysfs/symlink.c +++ b/fs/sysfs/symlink.c @@ -19,39 +19,6 @@ #include "sysfs.h" -static int object_depth(struct sysfs_dirent *sd) -{ - int depth = 0; - - for (; sd->s_parent; sd = sd->s_parent) - depth++; - - return depth; -} - -static int object_path_length(struct sysfs_dirent * sd) -{ - int length = 1; - - for (; sd->s_parent; sd = sd->s_parent) - length += strlen(sd->s_name) + 1; - - return length; -} - -static void fill_object_path(struct sysfs_dirent *sd, char *buffer, int length) -{ - --length; - for (; sd->s_parent; sd = sd->s_parent) { - int cur = strlen(sd->s_name); - - /* back up enough to print this bus id with '/' */ - length -= cur; - strncpy(buffer + length, sd->s_name, cur); - *(buffer + --length) = '/'; - } -} - /** * sysfs_create_link - create symlink between two objects. * @kobj: object whose directory we're creating the link in. @@ -112,7 +79,6 @@ int sysfs_create_link(struct kobject * kobj, struct kobject * target, const char return error; } - /** * sysfs_remove_link - remove symlink in object's directory. * @kobj: object we're acting for. @@ -124,24 +90,54 @@ void sysfs_remove_link(struct kobject * kobj, const char * name) sysfs_hash_and_remove(kobj->sd, name); } -static int sysfs_get_target_path(struct sysfs_dirent * parent_sd, - struct sysfs_dirent * target_sd, char *path) +static int sysfs_get_target_path(struct sysfs_dirent *parent_sd, + struct sysfs_dirent *target_sd, char *path) { - char * s; - int depth, size; + struct sysfs_dirent *base, *sd; + char *s = path; + int len = 0; + + /* go up to the root, stop at the base */ + base = parent_sd; + while (base->s_parent) { + sd = target_sd->s_parent; + while (sd->s_parent && base != sd) + sd = sd->s_parent; + + if (base == sd) + break; + + strcpy(s, "../"); + s += 3; + base = base->s_parent; + } + + /* determine end of target string for reverse fillup */ + sd = target_sd; + while (sd->s_parent && sd != base) { + len += strlen(sd->s_name) + 1; + sd = sd->s_parent; + } - depth = object_depth(parent_sd); - size = object_path_length(target_sd) + depth * 3 - 1; - if (size > PATH_MAX) + /* check limits */ + if (len < 2) + return -EINVAL; + len--; + if ((s - path) + len > PATH_MAX) return -ENAMETOOLONG; - pr_debug("%s: depth = %d, size = %d\n", __FUNCTION__, depth, size); + /* reverse fillup of target string from target to base */ + sd = target_sd; + while (sd->s_parent && sd != base) { + int slen = strlen(sd->s_name); - for (s = path; depth--; s += 3) - strcpy(s,"../"); + len -= slen; + strncpy(s + len, sd->s_name, slen); + if (len) + s[--len] = '/'; - fill_object_path(target_sd, path, size); - pr_debug("%s: path = '%s'\n", __FUNCTION__, path); + sd = sd->s_parent; + } return 0; } diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c index 81ec6c548c0..c5d60de0658 100644 --- a/fs/sysv/inode.c +++ b/fs/sysv/inode.c @@ -169,20 +169,27 @@ void sysv_set_inode(struct inode *inode, dev_t rdev) init_special_inode(inode, inode->i_mode, rdev); } -static void sysv_read_inode(struct inode *inode) +struct inode *sysv_iget(struct super_block *sb, unsigned int ino) { - struct super_block * sb = inode->i_sb; struct sysv_sb_info * sbi = SYSV_SB(sb); struct buffer_head * bh; struct sysv_inode * raw_inode; struct sysv_inode_info * si; - unsigned int block, ino = inode->i_ino; + struct inode *inode; + unsigned int block; if (!ino || ino > sbi->s_ninodes) { printk("Bad inode number on dev %s: %d is out of range\n", - inode->i_sb->s_id, ino); - goto bad_inode; + sb->s_id, ino); + return ERR_PTR(-EIO); } + + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; + raw_inode = sysv_raw_inode(sb, ino, &bh); if (!raw_inode) { printk("Major problem: unable to read inode from dev %s\n", @@ -214,11 +221,12 @@ static void sysv_read_inode(struct inode *inode) old_decode_dev(fs32_to_cpu(sbi, si->i_data[0]))); else sysv_set_inode(inode, 0); - return; + unlock_new_inode(inode); + return inode; bad_inode: - make_bad_inode(inode); - return; + iget_failed(inode); + return ERR_PTR(-EIO); } static struct buffer_head * sysv_update_inode(struct inode * inode) @@ -328,7 +336,6 @@ static void init_once(struct kmem_cache *cachep, void *p) const struct super_operations sysv_sops = { .alloc_inode = sysv_alloc_inode, .destroy_inode = sysv_destroy_inode, - .read_inode = sysv_read_inode, .write_inode = sysv_write_inode, .delete_inode = sysv_delete_inode, .put_super = sysv_put_super, diff --git a/fs/sysv/namei.c b/fs/sysv/namei.c index 6bd850b7641..a1f1ef33e81 100644 --- a/fs/sysv/namei.c +++ b/fs/sysv/namei.c @@ -53,9 +53,9 @@ static struct dentry *sysv_lookup(struct inode * dir, struct dentry * dentry, st ino = sysv_inode_by_name(dentry); if (ino) { - inode = iget(dir->i_sb, ino); - if (!inode) - return ERR_PTR(-EACCES); + inode = sysv_iget(dir->i_sb, ino); + if (IS_ERR(inode)) + return ERR_CAST(inode); } d_add(dentry, inode); return NULL; diff --git a/fs/sysv/super.c b/fs/sysv/super.c index 6f9707a1b95..5a903da5455 100644 --- a/fs/sysv/super.c +++ b/fs/sysv/super.c @@ -332,8 +332,8 @@ static int complete_read_super(struct super_block *sb, int silent, int size) sb->s_magic = SYSV_MAGIC_BASE + sbi->s_type; /* set up enough so that it can read an inode */ sb->s_op = &sysv_sops; - root_inode = iget(sb,SYSV_ROOT_INO); - if (!root_inode || is_bad_inode(root_inode)) { + root_inode = sysv_iget(sb, SYSV_ROOT_INO); + if (IS_ERR(root_inode)) { printk("SysV FS: get root inode failed\n"); return 0; } diff --git a/fs/sysv/sysv.h b/fs/sysv/sysv.h index 64c03bdf06a..42d51d1c05c 100644 --- a/fs/sysv/sysv.h +++ b/fs/sysv/sysv.h @@ -141,6 +141,7 @@ extern int __sysv_write_begin(struct file *file, struct address_space *mapping, struct page **pagep, void **fsdata); /* inode.c */ +extern struct inode *sysv_iget(struct super_block *, unsigned int); extern int sysv_write_inode(struct inode *, int); extern int sysv_sync_inode(struct inode *); extern int sysv_sync_file(struct file *, struct dentry *, int); diff --git a/fs/timerfd.c b/fs/timerfd.c index 61983f3b107..10c80b59ec4 100644 --- a/fs/timerfd.c +++ b/fs/timerfd.c @@ -25,13 +25,15 @@ struct timerfd_ctx { struct hrtimer tmr; ktime_t tintv; wait_queue_head_t wqh; + u64 ticks; int expired; + int clockid; }; /* * This gets called when the timer event triggers. We set the "expired" * flag, but we do not re-arm the timer (in case it's necessary, - * tintv.tv64 != 0) until the timer is read. + * tintv.tv64 != 0) until the timer is accessed. */ static enum hrtimer_restart timerfd_tmrproc(struct hrtimer *htmr) { @@ -40,13 +42,24 @@ static enum hrtimer_restart timerfd_tmrproc(struct hrtimer *htmr) spin_lock_irqsave(&ctx->wqh.lock, flags); ctx->expired = 1; + ctx->ticks++; wake_up_locked(&ctx->wqh); spin_unlock_irqrestore(&ctx->wqh.lock, flags); return HRTIMER_NORESTART; } -static void timerfd_setup(struct timerfd_ctx *ctx, int clockid, int flags, +static ktime_t timerfd_get_remaining(struct timerfd_ctx *ctx) +{ + ktime_t now, remaining; + + now = ctx->tmr.base->get_time(); + remaining = ktime_sub(ctx->tmr.expires, now); + + return remaining.tv64 < 0 ? ktime_set(0, 0): remaining; +} + +static void timerfd_setup(struct timerfd_ctx *ctx, int flags, const struct itimerspec *ktmr) { enum hrtimer_mode htmode; @@ -57,8 +70,9 @@ static void timerfd_setup(struct timerfd_ctx *ctx, int clockid, int flags, texp = timespec_to_ktime(ktmr->it_value); ctx->expired = 0; + ctx->ticks = 0; ctx->tintv = timespec_to_ktime(ktmr->it_interval); - hrtimer_init(&ctx->tmr, clockid, htmode); + hrtimer_init(&ctx->tmr, ctx->clockid, htmode); ctx->tmr.expires = texp; ctx->tmr.function = timerfd_tmrproc; if (texp.tv64 != 0) @@ -83,7 +97,7 @@ static unsigned int timerfd_poll(struct file *file, poll_table *wait) poll_wait(file, &ctx->wqh, wait); spin_lock_irqsave(&ctx->wqh.lock, flags); - if (ctx->expired) + if (ctx->ticks) events |= POLLIN; spin_unlock_irqrestore(&ctx->wqh.lock, flags); @@ -102,11 +116,11 @@ static ssize_t timerfd_read(struct file *file, char __user *buf, size_t count, return -EINVAL; spin_lock_irq(&ctx->wqh.lock); res = -EAGAIN; - if (!ctx->expired && !(file->f_flags & O_NONBLOCK)) { + if (!ctx->ticks && !(file->f_flags & O_NONBLOCK)) { __add_wait_queue(&ctx->wqh, &wait); for (res = 0;;) { set_current_state(TASK_INTERRUPTIBLE); - if (ctx->expired) { + if (ctx->ticks) { res = 0; break; } @@ -121,22 +135,21 @@ static ssize_t timerfd_read(struct file *file, char __user *buf, size_t count, __remove_wait_queue(&ctx->wqh, &wait); __set_current_state(TASK_RUNNING); } - if (ctx->expired) { - ctx->expired = 0; - if (ctx->tintv.tv64 != 0) { + if (ctx->ticks) { + ticks = ctx->ticks; + if (ctx->expired && ctx->tintv.tv64) { /* * If tintv.tv64 != 0, this is a periodic timer that * needs to be re-armed. We avoid doing it in the timer * callback to avoid DoS attacks specifying a very * short timer period. */ - ticks = (u64) - hrtimer_forward(&ctx->tmr, - hrtimer_cb_get_time(&ctx->tmr), - ctx->tintv); + ticks += hrtimer_forward_now(&ctx->tmr, + ctx->tintv) - 1; hrtimer_restart(&ctx->tmr); - } else - ticks = 1; + } + ctx->expired = 0; + ctx->ticks = 0; } spin_unlock_irq(&ctx->wqh.lock); if (ticks) @@ -150,76 +163,132 @@ static const struct file_operations timerfd_fops = { .read = timerfd_read, }; -asmlinkage long sys_timerfd(int ufd, int clockid, int flags, - const struct itimerspec __user *utmr) +static struct file *timerfd_fget(int fd) +{ + struct file *file; + + file = fget(fd); + if (!file) + return ERR_PTR(-EBADF); + if (file->f_op != &timerfd_fops) { + fput(file); + return ERR_PTR(-EINVAL); + } + + return file; +} + +asmlinkage long sys_timerfd_create(int clockid, int flags) { - int error; + int error, ufd; struct timerfd_ctx *ctx; struct file *file; struct inode *inode; - struct itimerspec ktmr; - - if (copy_from_user(&ktmr, utmr, sizeof(ktmr))) - return -EFAULT; + if (flags) + return -EINVAL; if (clockid != CLOCK_MONOTONIC && clockid != CLOCK_REALTIME) return -EINVAL; + + ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + + init_waitqueue_head(&ctx->wqh); + ctx->clockid = clockid; + hrtimer_init(&ctx->tmr, clockid, HRTIMER_MODE_ABS); + + error = anon_inode_getfd(&ufd, &inode, &file, "[timerfd]", + &timerfd_fops, ctx); + if (error) { + kfree(ctx); + return error; + } + + return ufd; +} + +asmlinkage long sys_timerfd_settime(int ufd, int flags, + const struct itimerspec __user *utmr, + struct itimerspec __user *otmr) +{ + struct file *file; + struct timerfd_ctx *ctx; + struct itimerspec ktmr, kotmr; + + if (copy_from_user(&ktmr, utmr, sizeof(ktmr))) + return -EFAULT; + if (!timespec_valid(&ktmr.it_value) || !timespec_valid(&ktmr.it_interval)) return -EINVAL; - if (ufd == -1) { - ctx = kmalloc(sizeof(*ctx), GFP_KERNEL); - if (!ctx) - return -ENOMEM; - - init_waitqueue_head(&ctx->wqh); - - timerfd_setup(ctx, clockid, flags, &ktmr); - - /* - * When we call this, the initialization must be complete, since - * anon_inode_getfd() will install the fd. - */ - error = anon_inode_getfd(&ufd, &inode, &file, "[timerfd]", - &timerfd_fops, ctx); - if (error) - goto err_tmrcancel; - } else { - file = fget(ufd); - if (!file) - return -EBADF; - ctx = file->private_data; - if (file->f_op != &timerfd_fops) { - fput(file); - return -EINVAL; - } - /* - * We need to stop the existing timer before reprogramming - * it to the new values. - */ - for (;;) { - spin_lock_irq(&ctx->wqh.lock); - if (hrtimer_try_to_cancel(&ctx->tmr) >= 0) - break; - spin_unlock_irq(&ctx->wqh.lock); - cpu_relax(); - } - /* - * Re-program the timer to the new value ... - */ - timerfd_setup(ctx, clockid, flags, &ktmr); + file = timerfd_fget(ufd); + if (IS_ERR(file)) + return PTR_ERR(file); + ctx = file->private_data; + /* + * We need to stop the existing timer before reprogramming + * it to the new values. + */ + for (;;) { + spin_lock_irq(&ctx->wqh.lock); + if (hrtimer_try_to_cancel(&ctx->tmr) >= 0) + break; spin_unlock_irq(&ctx->wqh.lock); - fput(file); + cpu_relax(); } - return ufd; + /* + * If the timer is expired and it's periodic, we need to advance it + * because the caller may want to know the previous expiration time. + * We do not update "ticks" and "expired" since the timer will be + * re-programmed again in the following timerfd_setup() call. + */ + if (ctx->expired && ctx->tintv.tv64) + hrtimer_forward_now(&ctx->tmr, ctx->tintv); -err_tmrcancel: - hrtimer_cancel(&ctx->tmr); - kfree(ctx); - return error; + kotmr.it_value = ktime_to_timespec(timerfd_get_remaining(ctx)); + kotmr.it_interval = ktime_to_timespec(ctx->tintv); + + /* + * Re-program the timer to the new value ... + */ + timerfd_setup(ctx, flags, &ktmr); + + spin_unlock_irq(&ctx->wqh.lock); + fput(file); + if (otmr && copy_to_user(otmr, &kotmr, sizeof(kotmr))) + return -EFAULT; + + return 0; +} + +asmlinkage long sys_timerfd_gettime(int ufd, struct itimerspec __user *otmr) +{ + struct file *file; + struct timerfd_ctx *ctx; + struct itimerspec kotmr; + + file = timerfd_fget(ufd); + if (IS_ERR(file)) + return PTR_ERR(file); + ctx = file->private_data; + + spin_lock_irq(&ctx->wqh.lock); + if (ctx->expired && ctx->tintv.tv64) { + ctx->expired = 0; + ctx->ticks += + hrtimer_forward_now(&ctx->tmr, ctx->tintv) - 1; + hrtimer_restart(&ctx->tmr); + } + kotmr.it_value = ktime_to_timespec(timerfd_get_remaining(ctx)); + kotmr.it_interval = ktime_to_timespec(ctx->tintv); + spin_unlock_irq(&ctx->wqh.lock); + fput(file); + + return copy_to_user(otmr, &kotmr, sizeof(kotmr)) ? -EFAULT: 0; } diff --git a/fs/udf/balloc.c b/fs/udf/balloc.c index ab26176f6b9..f855dcbbdfb 100644 --- a/fs/udf/balloc.c +++ b/fs/udf/balloc.c @@ -28,15 +28,16 @@ #include "udf_i.h" #include "udf_sb.h" -#define udf_clear_bit(nr,addr) ext2_clear_bit(nr,addr) -#define udf_set_bit(nr,addr) ext2_set_bit(nr,addr) +#define udf_clear_bit(nr, addr) ext2_clear_bit(nr, addr) +#define udf_set_bit(nr, addr) ext2_set_bit(nr, addr) #define udf_test_bit(nr, addr) ext2_test_bit(nr, addr) #define udf_find_first_one_bit(addr, size) find_first_one_bit(addr, size) -#define udf_find_next_one_bit(addr, size, offset) find_next_one_bit(addr, size, offset) +#define udf_find_next_one_bit(addr, size, offset) \ + find_next_one_bit(addr, size, offset) #define leBPL_to_cpup(x) leNUM_to_cpup(BITS_PER_LONG, x) -#define leNUM_to_cpup(x,y) xleNUM_to_cpup(x,y) -#define xleNUM_to_cpup(x,y) (le ## x ## _to_cpup(y)) +#define leNUM_to_cpup(x, y) xleNUM_to_cpup(x, y) +#define xleNUM_to_cpup(x, y) (le ## x ## _to_cpup(y)) #define uintBPL_t uint(BITS_PER_LONG) #define uint(x) xuint(x) #define xuint(x) __le ## x @@ -62,7 +63,8 @@ static inline int find_next_one_bit(void *addr, int size, int offset) result += BITS_PER_LONG; } while (size & ~(BITS_PER_LONG - 1)) { - if ((tmp = leBPL_to_cpup(p++))) + tmp = leBPL_to_cpup(p++); + if (tmp) goto found_middle; result += BITS_PER_LONG; size -= BITS_PER_LONG; @@ -88,12 +90,12 @@ static int read_block_bitmap(struct super_block *sb, kernel_lb_addr loc; loc.logicalBlockNum = bitmap->s_extPosition; - loc.partitionReferenceNum = UDF_SB_PARTITION(sb); + loc.partitionReferenceNum = UDF_SB(sb)->s_partition; bh = udf_tread(sb, udf_get_lb_pblock(sb, loc, block)); - if (!bh) { + if (!bh) retval = -EIO; - } + bitmap->s_block_bitmap[bitmap_nr] = bh; return retval; } @@ -138,6 +140,20 @@ static inline int load_block_bitmap(struct super_block *sb, return slot; } +static bool udf_add_free_space(struct udf_sb_info *sbi, + u16 partition, u32 cnt) +{ + struct logicalVolIntegrityDesc *lvid; + + if (sbi->s_lvid_bh == NULL) + return false; + + lvid = (struct logicalVolIntegrityDesc *)sbi->s_lvid_bh->b_data; + lvid->freeSpaceTable[partition] = cpu_to_le32(le32_to_cpu( + lvid->freeSpaceTable[partition]) + cnt); + return true; +} + static void udf_bitmap_free_blocks(struct super_block *sb, struct inode *inode, struct udf_bitmap *bitmap, @@ -155,57 +171,58 @@ static void udf_bitmap_free_blocks(struct super_block *sb, mutex_lock(&sbi->s_alloc_mutex); if (bloc.logicalBlockNum < 0 || - (bloc.logicalBlockNum + count) > UDF_SB_PARTLEN(sb, bloc.partitionReferenceNum)) { + (bloc.logicalBlockNum + count) > + sbi->s_partmaps[bloc.partitionReferenceNum].s_partition_len) { udf_debug("%d < %d || %d + %d > %d\n", bloc.logicalBlockNum, 0, bloc.logicalBlockNum, count, - UDF_SB_PARTLEN(sb, bloc.partitionReferenceNum)); + sbi->s_partmaps[bloc.partitionReferenceNum]. + s_partition_len); goto error_return; } - block = bloc.logicalBlockNum + offset + (sizeof(struct spaceBitmapDesc) << 3); + block = bloc.logicalBlockNum + offset + + (sizeof(struct spaceBitmapDesc) << 3); -do_more: - overflow = 0; - block_group = block >> (sb->s_blocksize_bits + 3); - bit = block % (sb->s_blocksize << 3); + do { + overflow = 0; + block_group = block >> (sb->s_blocksize_bits + 3); + bit = block % (sb->s_blocksize << 3); - /* - * Check to see if we are freeing blocks across a group boundary. - */ - if (bit + count > (sb->s_blocksize << 3)) { - overflow = bit + count - (sb->s_blocksize << 3); - count -= overflow; - } - bitmap_nr = load_block_bitmap(sb, bitmap, block_group); - if (bitmap_nr < 0) - goto error_return; + /* + * Check to see if we are freeing blocks across a group boundary. + */ + if (bit + count > (sb->s_blocksize << 3)) { + overflow = bit + count - (sb->s_blocksize << 3); + count -= overflow; + } + bitmap_nr = load_block_bitmap(sb, bitmap, block_group); + if (bitmap_nr < 0) + goto error_return; - bh = bitmap->s_block_bitmap[bitmap_nr]; - for (i = 0; i < count; i++) { - if (udf_set_bit(bit + i, bh->b_data)) { - udf_debug("bit %ld already set\n", bit + i); - udf_debug("byte=%2x\n", ((char *)bh->b_data)[(bit + i) >> 3]); - } else { - if (inode) - DQUOT_FREE_BLOCK(inode, 1); - if (UDF_SB_LVIDBH(sb)) { - UDF_SB_LVID(sb)->freeSpaceTable[UDF_SB_PARTITION(sb)] = - cpu_to_le32(le32_to_cpu(UDF_SB_LVID(sb)->freeSpaceTable[UDF_SB_PARTITION(sb)]) + 1); + bh = bitmap->s_block_bitmap[bitmap_nr]; + for (i = 0; i < count; i++) { + if (udf_set_bit(bit + i, bh->b_data)) { + udf_debug("bit %ld already set\n", bit + i); + udf_debug("byte=%2x\n", + ((char *)bh->b_data)[(bit + i) >> 3]); + } else { + if (inode) + DQUOT_FREE_BLOCK(inode, 1); + udf_add_free_space(sbi, sbi->s_partition, 1); } } - } - mark_buffer_dirty(bh); - if (overflow) { - block += count; - count = overflow; - goto do_more; - } + mark_buffer_dirty(bh); + if (overflow) { + block += count; + count = overflow; + } + } while (overflow); + error_return: sb->s_dirt = 1; - if (UDF_SB_LVIDBH(sb)) - mark_buffer_dirty(UDF_SB_LVIDBH(sb)); + if (sbi->s_lvid_bh) + mark_buffer_dirty(sbi->s_lvid_bh); mutex_unlock(&sbi->s_alloc_mutex); - return; } static int udf_bitmap_prealloc_blocks(struct super_block *sb, @@ -219,53 +236,50 @@ static int udf_bitmap_prealloc_blocks(struct super_block *sb, int bit, block, block_group, group_start; int nr_groups, bitmap_nr; struct buffer_head *bh; + __u32 part_len; mutex_lock(&sbi->s_alloc_mutex); - if (first_block < 0 || first_block >= UDF_SB_PARTLEN(sb, partition)) + part_len = sbi->s_partmaps[partition].s_partition_len; + if (first_block < 0 || first_block >= part_len) goto out; - if (first_block + block_count > UDF_SB_PARTLEN(sb, partition)) - block_count = UDF_SB_PARTLEN(sb, partition) - first_block; + if (first_block + block_count > part_len) + block_count = part_len - first_block; -repeat: - nr_groups = (UDF_SB_PARTLEN(sb, partition) + - (sizeof(struct spaceBitmapDesc) << 3) + - (sb->s_blocksize * 8) - 1) / (sb->s_blocksize * 8); - block = first_block + (sizeof(struct spaceBitmapDesc) << 3); - block_group = block >> (sb->s_blocksize_bits + 3); - group_start = block_group ? 0 : sizeof(struct spaceBitmapDesc); + do { + nr_groups = udf_compute_nr_groups(sb, partition); + block = first_block + (sizeof(struct spaceBitmapDesc) << 3); + block_group = block >> (sb->s_blocksize_bits + 3); + group_start = block_group ? 0 : sizeof(struct spaceBitmapDesc); - bitmap_nr = load_block_bitmap(sb, bitmap, block_group); - if (bitmap_nr < 0) - goto out; - bh = bitmap->s_block_bitmap[bitmap_nr]; + bitmap_nr = load_block_bitmap(sb, bitmap, block_group); + if (bitmap_nr < 0) + goto out; + bh = bitmap->s_block_bitmap[bitmap_nr]; - bit = block % (sb->s_blocksize << 3); + bit = block % (sb->s_blocksize << 3); - while (bit < (sb->s_blocksize << 3) && block_count > 0) { - if (!udf_test_bit(bit, bh->b_data)) { - goto out; - } else if (DQUOT_PREALLOC_BLOCK(inode, 1)) { - goto out; - } else if (!udf_clear_bit(bit, bh->b_data)) { - udf_debug("bit already cleared for block %d\n", bit); - DQUOT_FREE_BLOCK(inode, 1); - goto out; + while (bit < (sb->s_blocksize << 3) && block_count > 0) { + if (!udf_test_bit(bit, bh->b_data)) + goto out; + else if (DQUOT_PREALLOC_BLOCK(inode, 1)) + goto out; + else if (!udf_clear_bit(bit, bh->b_data)) { + udf_debug("bit already cleared for block %d\n", bit); + DQUOT_FREE_BLOCK(inode, 1); + goto out; + } + block_count--; + alloc_count++; + bit++; + block++; } - block_count--; - alloc_count++; - bit++; - block++; - } - mark_buffer_dirty(bh); - if (block_count > 0) - goto repeat; + mark_buffer_dirty(bh); + } while (block_count > 0); + out: - if (UDF_SB_LVIDBH(sb)) { - UDF_SB_LVID(sb)->freeSpaceTable[partition] = - cpu_to_le32(le32_to_cpu(UDF_SB_LVID(sb)->freeSpaceTable[partition]) - alloc_count); - mark_buffer_dirty(UDF_SB_LVIDBH(sb)); - } + if (udf_add_free_space(sbi, partition, -alloc_count)) + mark_buffer_dirty(sbi->s_lvid_bh); sb->s_dirt = 1; mutex_unlock(&sbi->s_alloc_mutex); return alloc_count; @@ -287,7 +301,7 @@ static int udf_bitmap_new_block(struct super_block *sb, mutex_lock(&sbi->s_alloc_mutex); repeat: - if (goal < 0 || goal >= UDF_SB_PARTLEN(sb, partition)) + if (goal < 0 || goal >= sbi->s_partmaps[partition].s_partition_len) goal = 0; nr_groups = bitmap->s_nr_groups; @@ -312,14 +326,16 @@ repeat: if (bit < end_goal) goto got_block; - ptr = memscan((char *)bh->b_data + (bit >> 3), 0xFF, sb->s_blocksize - ((bit + 7) >> 3)); + ptr = memscan((char *)bh->b_data + (bit >> 3), 0xFF, + sb->s_blocksize - ((bit + 7) >> 3)); newbit = (ptr - ((char *)bh->b_data)) << 3; if (newbit < sb->s_blocksize << 3) { bit = newbit; goto search_back; } - newbit = udf_find_next_one_bit(bh->b_data, sb->s_blocksize << 3, bit); + newbit = udf_find_next_one_bit(bh->b_data, + sb->s_blocksize << 3, bit); if (newbit < sb->s_blocksize << 3) { bit = newbit; goto got_block; @@ -358,15 +374,20 @@ repeat: if (bit < sb->s_blocksize << 3) goto search_back; else - bit = udf_find_next_one_bit(bh->b_data, sb->s_blocksize << 3, group_start << 3); + bit = udf_find_next_one_bit(bh->b_data, sb->s_blocksize << 3, + group_start << 3); if (bit >= sb->s_blocksize << 3) { mutex_unlock(&sbi->s_alloc_mutex); return 0; } search_back: - for (i = 0; i < 7 && bit > (group_start << 3) && udf_test_bit(bit - 1, bh->b_data); i++, bit--) - ; /* empty loop */ + i = 0; + while (i < 7 && bit > (group_start << 3) && + udf_test_bit(bit - 1, bh->b_data)) { + ++i; + --bit; + } got_block: @@ -389,11 +410,8 @@ got_block: mark_buffer_dirty(bh); - if (UDF_SB_LVIDBH(sb)) { - UDF_SB_LVID(sb)->freeSpaceTable[partition] = - cpu_to_le32(le32_to_cpu(UDF_SB_LVID(sb)->freeSpaceTable[partition]) - 1); - mark_buffer_dirty(UDF_SB_LVIDBH(sb)); - } + if (udf_add_free_space(sbi, partition, -1)) + mark_buffer_dirty(sbi->s_lvid_bh); sb->s_dirt = 1; mutex_unlock(&sbi->s_alloc_mutex); *err = 0; @@ -418,56 +436,70 @@ static void udf_table_free_blocks(struct super_block *sb, struct extent_position oepos, epos; int8_t etype; int i; + struct udf_inode_info *iinfo; mutex_lock(&sbi->s_alloc_mutex); if (bloc.logicalBlockNum < 0 || - (bloc.logicalBlockNum + count) > UDF_SB_PARTLEN(sb, bloc.partitionReferenceNum)) { + (bloc.logicalBlockNum + count) > + sbi->s_partmaps[bloc.partitionReferenceNum].s_partition_len) { udf_debug("%d < %d || %d + %d > %d\n", bloc.logicalBlockNum, 0, bloc.logicalBlockNum, count, - UDF_SB_PARTLEN(sb, bloc.partitionReferenceNum)); + sbi->s_partmaps[bloc.partitionReferenceNum]. + s_partition_len); goto error_return; } - /* We do this up front - There are some error conditions that could occure, - but.. oh well */ + iinfo = UDF_I(table); + /* We do this up front - There are some error conditions that + could occure, but.. oh well */ if (inode) DQUOT_FREE_BLOCK(inode, count); - if (UDF_SB_LVIDBH(sb)) { - UDF_SB_LVID(sb)->freeSpaceTable[UDF_SB_PARTITION(sb)] = - cpu_to_le32(le32_to_cpu(UDF_SB_LVID(sb)->freeSpaceTable[UDF_SB_PARTITION(sb)]) + count); - mark_buffer_dirty(UDF_SB_LVIDBH(sb)); - } + if (udf_add_free_space(sbi, sbi->s_partition, count)) + mark_buffer_dirty(sbi->s_lvid_bh); start = bloc.logicalBlockNum + offset; end = bloc.logicalBlockNum + offset + count - 1; epos.offset = oepos.offset = sizeof(struct unallocSpaceEntry); elen = 0; - epos.block = oepos.block = UDF_I_LOCATION(table); + epos.block = oepos.block = iinfo->i_location; epos.bh = oepos.bh = NULL; while (count && (etype = udf_next_aext(table, &epos, &eloc, &elen, 1)) != -1) { - if (((eloc.logicalBlockNum + (elen >> sb->s_blocksize_bits)) == start)) { - if ((0x3FFFFFFF - elen) < (count << sb->s_blocksize_bits)) { - count -= ((0x3FFFFFFF - elen) >> sb->s_blocksize_bits); - start += ((0x3FFFFFFF - elen) >> sb->s_blocksize_bits); - elen = (etype << 30) | (0x40000000 - sb->s_blocksize); + if (((eloc.logicalBlockNum + + (elen >> sb->s_blocksize_bits)) == start)) { + if ((0x3FFFFFFF - elen) < + (count << sb->s_blocksize_bits)) { + uint32_t tmp = ((0x3FFFFFFF - elen) >> + sb->s_blocksize_bits); + count -= tmp; + start += tmp; + elen = (etype << 30) | + (0x40000000 - sb->s_blocksize); } else { - elen = (etype << 30) | (elen + (count << sb->s_blocksize_bits)); + elen = (etype << 30) | + (elen + + (count << sb->s_blocksize_bits)); start += count; count = 0; } udf_write_aext(table, &oepos, eloc, elen, 1); } else if (eloc.logicalBlockNum == (end + 1)) { - if ((0x3FFFFFFF - elen) < (count << sb->s_blocksize_bits)) { - count -= ((0x3FFFFFFF - elen) >> sb->s_blocksize_bits); - end -= ((0x3FFFFFFF - elen) >> sb->s_blocksize_bits); - eloc.logicalBlockNum -= ((0x3FFFFFFF - elen) >> sb->s_blocksize_bits); - elen = (etype << 30) | (0x40000000 - sb->s_blocksize); + if ((0x3FFFFFFF - elen) < + (count << sb->s_blocksize_bits)) { + uint32_t tmp = ((0x3FFFFFFF - elen) >> + sb->s_blocksize_bits); + count -= tmp; + end -= tmp; + eloc.logicalBlockNum -= tmp; + elen = (etype << 30) | + (0x40000000 - sb->s_blocksize); } else { eloc.logicalBlockNum = start; - elen = (etype << 30) | (elen + (count << sb->s_blocksize_bits)); + elen = (etype << 30) | + (elen + + (count << sb->s_blocksize_bits)); end -= count; count = 0; } @@ -488,9 +520,9 @@ static void udf_table_free_blocks(struct super_block *sb, if (count) { /* - * NOTE: we CANNOT use udf_add_aext here, as it can try to allocate - * a new block, and since we hold the super block lock already - * very bad things would happen :) + * NOTE: we CANNOT use udf_add_aext here, as it can try to + * allocate a new block, and since we hold the super block + * lock already very bad things would happen :) * * We copy the behavior of udf_add_aext, but instead of * trying to allocate a new block close to the existing one, @@ -509,11 +541,11 @@ static void udf_table_free_blocks(struct super_block *sb, elen = EXT_RECORDED_ALLOCATED | (count << sb->s_blocksize_bits); - if (UDF_I_ALLOCTYPE(table) == ICBTAG_FLAG_AD_SHORT) { + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT) adsize = sizeof(short_ad); - } else if (UDF_I_ALLOCTYPE(table) == ICBTAG_FLAG_AD_LONG) { + else if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_LONG) adsize = sizeof(long_ad); - } else { + else { brelse(oepos.bh); brelse(epos.bh); goto error_return; @@ -531,56 +563,70 @@ static void udf_table_free_blocks(struct super_block *sb, eloc.logicalBlockNum++; elen -= sb->s_blocksize; - if (!(epos.bh = udf_tread(sb, udf_get_lb_pblock(sb, epos.block, 0)))) { + epos.bh = udf_tread(sb, + udf_get_lb_pblock(sb, epos.block, 0)); + if (!epos.bh) { brelse(oepos.bh); goto error_return; } aed = (struct allocExtDesc *)(epos.bh->b_data); - aed->previousAllocExtLocation = cpu_to_le32(oepos.block.logicalBlockNum); + aed->previousAllocExtLocation = + cpu_to_le32(oepos.block.logicalBlockNum); if (epos.offset + adsize > sb->s_blocksize) { loffset = epos.offset; aed->lengthAllocDescs = cpu_to_le32(adsize); - sptr = UDF_I_DATA(table) + epos.offset - adsize; - dptr = epos.bh->b_data + sizeof(struct allocExtDesc); + sptr = iinfo->i_ext.i_data + epos.offset + - adsize; + dptr = epos.bh->b_data + + sizeof(struct allocExtDesc); memcpy(dptr, sptr, adsize); - epos.offset = sizeof(struct allocExtDesc) + adsize; + epos.offset = sizeof(struct allocExtDesc) + + adsize; } else { loffset = epos.offset + adsize; aed->lengthAllocDescs = cpu_to_le32(0); if (oepos.bh) { sptr = oepos.bh->b_data + epos.offset; - aed = (struct allocExtDesc *)oepos.bh->b_data; + aed = (struct allocExtDesc *) + oepos.bh->b_data; aed->lengthAllocDescs = - cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) + adsize); + cpu_to_le32(le32_to_cpu( + aed->lengthAllocDescs) + + adsize); } else { - sptr = UDF_I_DATA(table) + epos.offset; - UDF_I_LENALLOC(table) += adsize; + sptr = iinfo->i_ext.i_data + + epos.offset; + iinfo->i_lenAlloc += adsize; mark_inode_dirty(table); } epos.offset = sizeof(struct allocExtDesc); } - if (UDF_SB_UDFREV(sb) >= 0x0200) - udf_new_tag(epos.bh->b_data, TAG_IDENT_AED, 3, 1, - epos.block.logicalBlockNum, sizeof(tag)); + if (sbi->s_udfrev >= 0x0200) + udf_new_tag(epos.bh->b_data, TAG_IDENT_AED, + 3, 1, epos.block.logicalBlockNum, + sizeof(tag)); else - udf_new_tag(epos.bh->b_data, TAG_IDENT_AED, 2, 1, - epos.block.logicalBlockNum, sizeof(tag)); - - switch (UDF_I_ALLOCTYPE(table)) { - case ICBTAG_FLAG_AD_SHORT: - sad = (short_ad *)sptr; - sad->extLength = cpu_to_le32( - EXT_NEXT_EXTENT_ALLOCDECS | - sb->s_blocksize); - sad->extPosition = cpu_to_le32(epos.block.logicalBlockNum); - break; - case ICBTAG_FLAG_AD_LONG: - lad = (long_ad *)sptr; - lad->extLength = cpu_to_le32( - EXT_NEXT_EXTENT_ALLOCDECS | - sb->s_blocksize); - lad->extLocation = cpu_to_lelb(epos.block); - break; + udf_new_tag(epos.bh->b_data, TAG_IDENT_AED, + 2, 1, epos.block.logicalBlockNum, + sizeof(tag)); + + switch (iinfo->i_alloc_type) { + case ICBTAG_FLAG_AD_SHORT: + sad = (short_ad *)sptr; + sad->extLength = cpu_to_le32( + EXT_NEXT_EXTENT_ALLOCDECS | + sb->s_blocksize); + sad->extPosition = + cpu_to_le32(epos.block.logicalBlockNum); + break; + case ICBTAG_FLAG_AD_LONG: + lad = (long_ad *)sptr; + lad->extLength = cpu_to_le32( + EXT_NEXT_EXTENT_ALLOCDECS | + sb->s_blocksize); + lad->extLocation = + cpu_to_lelb(epos.block); + break; } if (oepos.bh) { udf_update_tag(oepos.bh->b_data, loffset); @@ -590,16 +636,18 @@ static void udf_table_free_blocks(struct super_block *sb, } } - if (elen) { /* It's possible that stealing the block emptied the extent */ + /* It's possible that stealing the block emptied the extent */ + if (elen) { udf_write_aext(table, &epos, eloc, elen, 1); if (!epos.bh) { - UDF_I_LENALLOC(table) += adsize; + iinfo->i_lenAlloc += adsize; mark_inode_dirty(table); } else { aed = (struct allocExtDesc *)epos.bh->b_data; aed->lengthAllocDescs = - cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) + adsize); + cpu_to_le32(le32_to_cpu( + aed->lengthAllocDescs) + adsize); udf_update_tag(epos.bh->b_data, epos.offset); mark_buffer_dirty(epos.bh); } @@ -626,20 +674,23 @@ static int udf_table_prealloc_blocks(struct super_block *sb, kernel_lb_addr eloc; struct extent_position epos; int8_t etype = -1; + struct udf_inode_info *iinfo; - if (first_block < 0 || first_block >= UDF_SB_PARTLEN(sb, partition)) + if (first_block < 0 || + first_block >= sbi->s_partmaps[partition].s_partition_len) return 0; - if (UDF_I_ALLOCTYPE(table) == ICBTAG_FLAG_AD_SHORT) + iinfo = UDF_I(table); + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT) adsize = sizeof(short_ad); - else if (UDF_I_ALLOCTYPE(table) == ICBTAG_FLAG_AD_LONG) + else if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_LONG) adsize = sizeof(long_ad); else return 0; mutex_lock(&sbi->s_alloc_mutex); epos.offset = sizeof(struct unallocSpaceEntry); - epos.block = UDF_I_LOCATION(table); + epos.block = iinfo->i_location; epos.bh = NULL; eloc.logicalBlockNum = 0xFFFFFFFF; @@ -654,26 +705,26 @@ static int udf_table_prealloc_blocks(struct super_block *sb, epos.offset -= adsize; alloc_count = (elen >> sb->s_blocksize_bits); - if (inode && DQUOT_PREALLOC_BLOCK(inode, alloc_count > block_count ? block_count : alloc_count)) { + if (inode && DQUOT_PREALLOC_BLOCK(inode, + alloc_count > block_count ? block_count : alloc_count)) alloc_count = 0; - } else if (alloc_count > block_count) { + else if (alloc_count > block_count) { alloc_count = block_count; eloc.logicalBlockNum += alloc_count; elen -= (alloc_count << sb->s_blocksize_bits); - udf_write_aext(table, &epos, eloc, (etype << 30) | elen, 1); - } else { - udf_delete_aext(table, epos, eloc, (etype << 30) | elen); - } + udf_write_aext(table, &epos, eloc, + (etype << 30) | elen, 1); + } else + udf_delete_aext(table, epos, eloc, + (etype << 30) | elen); } else { alloc_count = 0; } brelse(epos.bh); - if (alloc_count && UDF_SB_LVIDBH(sb)) { - UDF_SB_LVID(sb)->freeSpaceTable[partition] = - cpu_to_le32(le32_to_cpu(UDF_SB_LVID(sb)->freeSpaceTable[partition]) - alloc_count); - mark_buffer_dirty(UDF_SB_LVIDBH(sb)); + if (alloc_count && udf_add_free_space(sbi, partition, -alloc_count)) { + mark_buffer_dirty(sbi->s_lvid_bh); sb->s_dirt = 1; } mutex_unlock(&sbi->s_alloc_mutex); @@ -692,33 +743,35 @@ static int udf_table_new_block(struct super_block *sb, kernel_lb_addr eloc, uninitialized_var(goal_eloc); struct extent_position epos, goal_epos; int8_t etype; + struct udf_inode_info *iinfo = UDF_I(table); *err = -ENOSPC; - if (UDF_I_ALLOCTYPE(table) == ICBTAG_FLAG_AD_SHORT) + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT) adsize = sizeof(short_ad); - else if (UDF_I_ALLOCTYPE(table) == ICBTAG_FLAG_AD_LONG) + else if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_LONG) adsize = sizeof(long_ad); else return newblock; mutex_lock(&sbi->s_alloc_mutex); - if (goal < 0 || goal >= UDF_SB_PARTLEN(sb, partition)) + if (goal < 0 || goal >= sbi->s_partmaps[partition].s_partition_len) goal = 0; - /* We search for the closest matching block to goal. If we find a exact hit, - we stop. Otherwise we keep going till we run out of extents. - We store the buffer_head, bloc, and extoffset of the current closest - match and use that when we are done. + /* We search for the closest matching block to goal. If we find + a exact hit, we stop. Otherwise we keep going till we run out + of extents. We store the buffer_head, bloc, and extoffset + of the current closest match and use that when we are done. */ epos.offset = sizeof(struct unallocSpaceEntry); - epos.block = UDF_I_LOCATION(table); + epos.block = iinfo->i_location; epos.bh = goal_epos.bh = NULL; while (spread && (etype = udf_next_aext(table, &epos, &eloc, &elen, 1)) != -1) { if (goal >= eloc.logicalBlockNum) { - if (goal < eloc.logicalBlockNum + (elen >> sb->s_blocksize_bits)) + if (goal < eloc.logicalBlockNum + + (elen >> sb->s_blocksize_bits)) nspread = 0; else nspread = goal - eloc.logicalBlockNum - @@ -771,11 +824,8 @@ static int udf_table_new_block(struct super_block *sb, udf_delete_aext(table, goal_epos, goal_eloc, goal_elen); brelse(goal_epos.bh); - if (UDF_SB_LVIDBH(sb)) { - UDF_SB_LVID(sb)->freeSpaceTable[partition] = - cpu_to_le32(le32_to_cpu(UDF_SB_LVID(sb)->freeSpaceTable[partition]) - 1); - mark_buffer_dirty(UDF_SB_LVIDBH(sb)); - } + if (udf_add_free_space(sbi, partition, -1)) + mark_buffer_dirty(sbi->s_lvid_bh); sb->s_dirt = 1; mutex_unlock(&sbi->s_alloc_mutex); @@ -789,22 +839,23 @@ inline void udf_free_blocks(struct super_block *sb, uint32_t count) { uint16_t partition = bloc.partitionReferenceNum; + struct udf_part_map *map = &UDF_SB(sb)->s_partmaps[partition]; - if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_UNALLOC_BITMAP) { + if (map->s_partition_flags & UDF_PART_FLAG_UNALLOC_BITMAP) { return udf_bitmap_free_blocks(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_uspace.s_bitmap, + map->s_uspace.s_bitmap, bloc, offset, count); - } else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_UNALLOC_TABLE) { + } else if (map->s_partition_flags & UDF_PART_FLAG_UNALLOC_TABLE) { return udf_table_free_blocks(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_uspace.s_table, + map->s_uspace.s_table, bloc, offset, count); - } else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_BITMAP) { + } else if (map->s_partition_flags & UDF_PART_FLAG_FREED_BITMAP) { return udf_bitmap_free_blocks(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_fspace.s_bitmap, + map->s_fspace.s_bitmap, bloc, offset, count); - } else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_TABLE) { + } else if (map->s_partition_flags & UDF_PART_FLAG_FREED_TABLE) { return udf_table_free_blocks(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_fspace.s_table, + map->s_fspace.s_table, bloc, offset, count); } else { return; @@ -816,51 +867,55 @@ inline int udf_prealloc_blocks(struct super_block *sb, uint16_t partition, uint32_t first_block, uint32_t block_count) { - if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_UNALLOC_BITMAP) { + struct udf_part_map *map = &UDF_SB(sb)->s_partmaps[partition]; + + if (map->s_partition_flags & UDF_PART_FLAG_UNALLOC_BITMAP) return udf_bitmap_prealloc_blocks(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_uspace.s_bitmap, - partition, first_block, block_count); - } else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_UNALLOC_TABLE) { + map->s_uspace.s_bitmap, + partition, first_block, + block_count); + else if (map->s_partition_flags & UDF_PART_FLAG_UNALLOC_TABLE) return udf_table_prealloc_blocks(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_uspace.s_table, - partition, first_block, block_count); - } else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_BITMAP) { + map->s_uspace.s_table, + partition, first_block, + block_count); + else if (map->s_partition_flags & UDF_PART_FLAG_FREED_BITMAP) return udf_bitmap_prealloc_blocks(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_fspace.s_bitmap, - partition, first_block, block_count); - } else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_TABLE) { + map->s_fspace.s_bitmap, + partition, first_block, + block_count); + else if (map->s_partition_flags & UDF_PART_FLAG_FREED_TABLE) return udf_table_prealloc_blocks(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_fspace.s_table, - partition, first_block, block_count); - } else { + map->s_fspace.s_table, + partition, first_block, + block_count); + else return 0; - } } inline int udf_new_block(struct super_block *sb, struct inode *inode, uint16_t partition, uint32_t goal, int *err) { - int ret; + struct udf_part_map *map = &UDF_SB(sb)->s_partmaps[partition]; - if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_UNALLOC_BITMAP) { - ret = udf_bitmap_new_block(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_uspace.s_bitmap, + if (map->s_partition_flags & UDF_PART_FLAG_UNALLOC_BITMAP) + return udf_bitmap_new_block(sb, inode, + map->s_uspace.s_bitmap, partition, goal, err); - return ret; - } else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_UNALLOC_TABLE) { + else if (map->s_partition_flags & UDF_PART_FLAG_UNALLOC_TABLE) return udf_table_new_block(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_uspace.s_table, + map->s_uspace.s_table, partition, goal, err); - } else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_BITMAP) { + else if (map->s_partition_flags & UDF_PART_FLAG_FREED_BITMAP) return udf_bitmap_new_block(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_fspace.s_bitmap, + map->s_fspace.s_bitmap, partition, goal, err); - } else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_TABLE) { + else if (map->s_partition_flags & UDF_PART_FLAG_FREED_TABLE) return udf_table_new_block(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_fspace.s_table, + map->s_fspace.s_table, partition, goal, err); - } else { + else { *err = -EIO; return 0; } diff --git a/fs/udf/crc.c b/fs/udf/crc.c index 85aaee5fab2..b1661296e78 100644 --- a/fs/udf/crc.c +++ b/fs/udf/crc.c @@ -79,7 +79,7 @@ static uint16_t crc_table[256] = { * July 21, 1997 - Andrew E. Mileski * Adapted from OSTA-UDF(tm) 1.50 standard. */ -uint16_t udf_crc(uint8_t * data, uint32_t size, uint16_t crc) +uint16_t udf_crc(uint8_t *data, uint32_t size, uint16_t crc) { while (size--) crc = crc_table[(crc >> 8 ^ *(data++)) & 0xffU] ^ (crc << 8); diff --git a/fs/udf/dir.c b/fs/udf/dir.c index 9e3b9f97ddb..8d8643ada19 100644 --- a/fs/udf/dir.c +++ b/fs/udf/dir.c @@ -36,80 +36,20 @@ #include "udf_i.h" #include "udf_sb.h" -/* Prototypes for file operations */ -static int udf_readdir(struct file *, void *, filldir_t); -static int do_udf_readdir(struct inode *, struct file *, filldir_t, void *); - -/* readdir and lookup functions */ - -const struct file_operations udf_dir_operations = { - .read = generic_read_dir, - .readdir = udf_readdir, - .ioctl = udf_ioctl, - .fsync = udf_fsync_file, -}; - -/* - * udf_readdir - * - * PURPOSE - * Read a directory entry. - * - * DESCRIPTION - * Optional - sys_getdents() will return -ENOTDIR if this routine is not - * available. - * - * Refer to sys_getdents() in fs/readdir.c - * sys_getdents() -> . - * - * PRE-CONDITIONS - * filp Pointer to directory file. - * buf Pointer to directory entry buffer. - * filldir Pointer to filldir function. - * - * POST-CONDITIONS - * <return> >=0 on success. - * - * HISTORY - * July 1, 1997 - Andrew E. Mileski - * Written, tested, and released. - */ - -int udf_readdir(struct file *filp, void *dirent, filldir_t filldir) -{ - struct inode *dir = filp->f_path.dentry->d_inode; - int result; - - lock_kernel(); - - if (filp->f_pos == 0) { - if (filldir(dirent, ".", 1, filp->f_pos, dir->i_ino, DT_DIR) < 0) { - unlock_kernel(); - return 0; - } - filp->f_pos++; - } - - result = do_udf_readdir(dir, filp, filldir, dirent); - unlock_kernel(); - return result; -} - -static int -do_udf_readdir(struct inode *dir, struct file *filp, filldir_t filldir, - void *dirent) +static int do_udf_readdir(struct inode *dir, struct file *filp, + filldir_t filldir, void *dirent) { struct udf_fileident_bh fibh; struct fileIdentDesc *fi = NULL; struct fileIdentDesc cfi; int block, iblock; - loff_t nf_pos = filp->f_pos - 1; + loff_t nf_pos = (filp->f_pos - 1) << 2; int flen; char fname[UDF_NAME_LEN]; char *nameptr; uint16_t liu; uint8_t lfi; - loff_t size = (udf_ext0_offset(dir) + dir->i_size) >> 2; + loff_t size = udf_ext0_offset(dir) + dir->i_size; struct buffer_head *tmp, *bha[16]; kernel_lb_addr eloc; uint32_t elen; @@ -117,23 +57,26 @@ do_udf_readdir(struct inode *dir, struct file *filp, filldir_t filldir, int i, num; unsigned int dt_type; struct extent_position epos = { NULL, 0, {0, 0} }; + struct udf_inode_info *iinfo; if (nf_pos >= size) return 0; if (nf_pos == 0) - nf_pos = (udf_ext0_offset(dir) >> 2); + nf_pos = udf_ext0_offset(dir); - fibh.soffset = fibh.eoffset = (nf_pos & ((dir->i_sb->s_blocksize - 1) >> 2)) << 2; - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) { + fibh.soffset = fibh.eoffset = nf_pos & (dir->i_sb->s_blocksize - 1); + iinfo = UDF_I(dir); + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) { fibh.sbh = fibh.ebh = NULL; - } else if (inode_bmap(dir, nf_pos >> (dir->i_sb->s_blocksize_bits - 2), + } else if (inode_bmap(dir, nf_pos >> dir->i_sb->s_blocksize_bits, &epos, &eloc, &elen, &offset) == (EXT_RECORDED_ALLOCATED >> 30)) { block = udf_get_lb_pblock(dir->i_sb, eloc, offset); if ((++offset << dir->i_sb->s_blocksize_bits) < elen) { - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_SHORT) + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT) epos.offset -= sizeof(short_ad); - else if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_LONG) + else if (iinfo->i_alloc_type == + ICBTAG_FLAG_AD_LONG) epos.offset -= sizeof(long_ad); } else { offset = 0; @@ -168,7 +111,7 @@ do_udf_readdir(struct inode *dir, struct file *filp, filldir_t filldir, } while (nf_pos < size) { - filp->f_pos = nf_pos + 1; + filp->f_pos = (nf_pos >> 2) + 1; fi = udf_fileident_read(dir, &nf_pos, &fibh, &cfi, &epos, &eloc, &elen, &offset); @@ -235,7 +178,7 @@ do_udf_readdir(struct inode *dir, struct file *filp, filldir_t filldir, } } /* end while */ - filp->f_pos = nf_pos + 1; + filp->f_pos = (nf_pos >> 2) + 1; if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); @@ -244,3 +187,57 @@ do_udf_readdir(struct inode *dir, struct file *filp, filldir_t filldir, return 0; } + +/* + * udf_readdir + * + * PURPOSE + * Read a directory entry. + * + * DESCRIPTION + * Optional - sys_getdents() will return -ENOTDIR if this routine is not + * available. + * + * Refer to sys_getdents() in fs/readdir.c + * sys_getdents() -> . + * + * PRE-CONDITIONS + * filp Pointer to directory file. + * buf Pointer to directory entry buffer. + * filldir Pointer to filldir function. + * + * POST-CONDITIONS + * <return> >=0 on success. + * + * HISTORY + * July 1, 1997 - Andrew E. Mileski + * Written, tested, and released. + */ + +static int udf_readdir(struct file *filp, void *dirent, filldir_t filldir) +{ + struct inode *dir = filp->f_path.dentry->d_inode; + int result; + + lock_kernel(); + + if (filp->f_pos == 0) { + if (filldir(dirent, ".", 1, filp->f_pos, dir->i_ino, DT_DIR) < 0) { + unlock_kernel(); + return 0; + } + filp->f_pos++; + } + + result = do_udf_readdir(dir, filp, filldir, dirent); + unlock_kernel(); + return result; +} + +/* readdir and lookup functions */ +const struct file_operations udf_dir_operations = { + .read = generic_read_dir, + .readdir = udf_readdir, + .ioctl = udf_ioctl, + .fsync = udf_fsync_file, +}; diff --git a/fs/udf/directory.c b/fs/udf/directory.c index ff8c08fd7bf..2820f8fcf4c 100644 --- a/fs/udf/directory.c +++ b/fs/udf/directory.c @@ -19,7 +19,7 @@ #include <linux/buffer_head.h> #if 0 -static uint8_t *udf_filead_read(struct inode *dir, uint8_t * tmpad, +static uint8_t *udf_filead_read(struct inode *dir, uint8_t *tmpad, uint8_t ad_size, kernel_lb_addr fe_loc, int *pos, int *offset, struct buffer_head **bh, int *error) @@ -45,7 +45,8 @@ static uint8_t *udf_filead_read(struct inode *dir, uint8_t * tmpad, block = udf_get_lb_pblock(dir->i_sb, fe_loc, ++*pos); if (!block) return NULL; - if (!(*bh = udf_tread(dir->i_sb, block))) + *bh = udf_tread(dir->i_sb, block); + if (!*bh) return NULL; } else if (*offset > dir->i_sb->s_blocksize) { ad = tmpad; @@ -57,10 +58,12 @@ static uint8_t *udf_filead_read(struct inode *dir, uint8_t * tmpad, block = udf_get_lb_pblock(dir->i_sb, fe_loc, ++*pos); if (!block) return NULL; - if (!((*bh) = udf_tread(dir->i_sb, block))) + (*bh) = udf_tread(dir->i_sb, block); + if (!*bh) return NULL; - memcpy((uint8_t *)ad + remainder, (*bh)->b_data, ad_size - remainder); + memcpy((uint8_t *)ad + remainder, (*bh)->b_data, + ad_size - remainder); *offset = ad_size - remainder; } @@ -68,29 +71,31 @@ static uint8_t *udf_filead_read(struct inode *dir, uint8_t * tmpad, } #endif -struct fileIdentDesc *udf_fileident_read(struct inode *dir, loff_t * nf_pos, +struct fileIdentDesc *udf_fileident_read(struct inode *dir, loff_t *nf_pos, struct udf_fileident_bh *fibh, struct fileIdentDesc *cfi, struct extent_position *epos, - kernel_lb_addr * eloc, uint32_t * elen, - sector_t * offset) + kernel_lb_addr *eloc, uint32_t *elen, + sector_t *offset) { struct fileIdentDesc *fi; int i, num, block; struct buffer_head *tmp, *bha[16]; + struct udf_inode_info *iinfo = UDF_I(dir); fibh->soffset = fibh->eoffset; - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) { - fi = udf_get_fileident(UDF_I_DATA(dir) - - (UDF_I_EFE(dir) ? + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) { + fi = udf_get_fileident(iinfo->i_ext.i_data - + (iinfo->i_efe ? sizeof(struct extendedFileEntry) : sizeof(struct fileEntry)), - dir->i_sb->s_blocksize, &(fibh->eoffset)); + dir->i_sb->s_blocksize, + &(fibh->eoffset)); if (!fi) return NULL; - *nf_pos += ((fibh->eoffset - fibh->soffset) >> 2); + *nf_pos += fibh->eoffset - fibh->soffset; memcpy((uint8_t *)cfi, (uint8_t *)fi, sizeof(struct fileIdentDesc)); @@ -100,6 +105,7 @@ struct fileIdentDesc *udf_fileident_read(struct inode *dir, loff_t * nf_pos, if (fibh->eoffset == dir->i_sb->s_blocksize) { int lextoffset = epos->offset; + unsigned char blocksize_bits = dir->i_sb->s_blocksize_bits; if (udf_next_aext(dir, epos, eloc, elen, 1) != (EXT_RECORDED_ALLOCATED >> 30)) @@ -109,24 +115,27 @@ struct fileIdentDesc *udf_fileident_read(struct inode *dir, loff_t * nf_pos, (*offset)++; - if ((*offset << dir->i_sb->s_blocksize_bits) >= *elen) + if ((*offset << blocksize_bits) >= *elen) *offset = 0; else epos->offset = lextoffset; brelse(fibh->sbh); - if (!(fibh->sbh = fibh->ebh = udf_tread(dir->i_sb, block))) + fibh->sbh = fibh->ebh = udf_tread(dir->i_sb, block); + if (!fibh->sbh) return NULL; fibh->soffset = fibh->eoffset = 0; - if (!(*offset & ((16 >> (dir->i_sb->s_blocksize_bits - 9)) - 1))) { - i = 16 >> (dir->i_sb->s_blocksize_bits - 9); - if (i + *offset > (*elen >> dir->i_sb->s_blocksize_bits)) - i = (*elen >> dir->i_sb->s_blocksize_bits)-*offset; + if (!(*offset & ((16 >> (blocksize_bits - 9)) - 1))) { + i = 16 >> (blocksize_bits - 9); + if (i + *offset > (*elen >> blocksize_bits)) + i = (*elen >> blocksize_bits)-*offset; for (num = 0; i > 0; i--) { - block = udf_get_lb_pblock(dir->i_sb, *eloc, *offset + i); + block = udf_get_lb_pblock(dir->i_sb, *eloc, + *offset + i); tmp = udf_tgetblk(dir->i_sb, block); - if (tmp && !buffer_uptodate(tmp) && !buffer_locked(tmp)) + if (tmp && !buffer_uptodate(tmp) && + !buffer_locked(tmp)) bha[num++] = tmp; else brelse(tmp); @@ -148,7 +157,7 @@ struct fileIdentDesc *udf_fileident_read(struct inode *dir, loff_t * nf_pos, if (!fi) return NULL; - *nf_pos += ((fibh->eoffset - fibh->soffset) >> 2); + *nf_pos += fibh->eoffset - fibh->soffset; if (fibh->eoffset <= dir->i_sb->s_blocksize) { memcpy((uint8_t *)cfi, (uint8_t *)fi, @@ -172,20 +181,23 @@ struct fileIdentDesc *udf_fileident_read(struct inode *dir, loff_t * nf_pos, fibh->soffset -= dir->i_sb->s_blocksize; fibh->eoffset -= dir->i_sb->s_blocksize; - if (!(fibh->ebh = udf_tread(dir->i_sb, block))) + fibh->ebh = udf_tread(dir->i_sb, block); + if (!fibh->ebh) return NULL; if (sizeof(struct fileIdentDesc) > -fibh->soffset) { int fi_len; memcpy((uint8_t *)cfi, (uint8_t *)fi, -fibh->soffset); - memcpy((uint8_t *)cfi - fibh->soffset, fibh->ebh->b_data, + memcpy((uint8_t *)cfi - fibh->soffset, + fibh->ebh->b_data, sizeof(struct fileIdentDesc) + fibh->soffset); - fi_len = (sizeof(struct fileIdentDesc) + cfi->lengthFileIdent + + fi_len = (sizeof(struct fileIdentDesc) + + cfi->lengthFileIdent + le16_to_cpu(cfi->lengthOfImpUse) + 3) & ~3; - *nf_pos += ((fi_len - (fibh->eoffset - fibh->soffset)) >> 2); + *nf_pos += fi_len - (fibh->eoffset - fibh->soffset); fibh->eoffset = fibh->soffset + fi_len; } else { memcpy((uint8_t *)cfi, (uint8_t *)fi, @@ -210,11 +222,10 @@ struct fileIdentDesc *udf_get_fileident(void *buffer, int bufsize, int *offset) ptr = buffer; - if ((*offset > 0) && (*offset < bufsize)) { + if ((*offset > 0) && (*offset < bufsize)) ptr += *offset; - } fi = (struct fileIdentDesc *)ptr; - if (le16_to_cpu(fi->descTag.tagIdent) != TAG_IDENT_FID) { + if (fi->descTag.tagIdent != cpu_to_le16(TAG_IDENT_FID)) { udf_debug("0x%x != TAG_IDENT_FID\n", le16_to_cpu(fi->descTag.tagIdent)); udf_debug("offset: %u sizeof: %lu bufsize: %u\n", @@ -222,12 +233,11 @@ struct fileIdentDesc *udf_get_fileident(void *buffer, int bufsize, int *offset) bufsize); return NULL; } - if ((*offset + sizeof(struct fileIdentDesc)) > bufsize) { + if ((*offset + sizeof(struct fileIdentDesc)) > bufsize) lengthThisIdent = sizeof(struct fileIdentDesc); - } else { + else lengthThisIdent = sizeof(struct fileIdentDesc) + fi->lengthFileIdent + le16_to_cpu(fi->lengthOfImpUse); - } /* we need to figure padding, too! */ padlen = lengthThisIdent % UDF_NAME_PAD; @@ -252,17 +262,17 @@ static extent_ad *udf_get_fileextent(void *buffer, int bufsize, int *offset) fe = (struct fileEntry *)buffer; - if (le16_to_cpu(fe->descTag.tagIdent) != TAG_IDENT_FE) { + if (fe->descTag.tagIdent != cpu_to_le16(TAG_IDENT_FE)) { udf_debug("0x%x != TAG_IDENT_FE\n", le16_to_cpu(fe->descTag.tagIdent)); return NULL; } - ptr = (uint8_t *)(fe->extendedAttr) + le32_to_cpu(fe->lengthExtendedAttr); + ptr = (uint8_t *)(fe->extendedAttr) + + le32_to_cpu(fe->lengthExtendedAttr); - if ((*offset > 0) && (*offset < le32_to_cpu(fe->lengthAllocDescs))) { + if ((*offset > 0) && (*offset < le32_to_cpu(fe->lengthAllocDescs))) ptr += *offset; - } ext = (extent_ad *)ptr; @@ -271,7 +281,7 @@ static extent_ad *udf_get_fileextent(void *buffer, int bufsize, int *offset) } #endif -short_ad *udf_get_fileshortad(uint8_t *ptr, int maxoffset, int *offset, +short_ad *udf_get_fileshortad(uint8_t *ptr, int maxoffset, uint32_t *offset, int inc) { short_ad *sa; @@ -281,17 +291,20 @@ short_ad *udf_get_fileshortad(uint8_t *ptr, int maxoffset, int *offset, return NULL; } - if ((*offset < 0) || ((*offset + sizeof(short_ad)) > maxoffset)) - return NULL; - else if ((sa = (short_ad *)ptr)->extLength == 0) + if ((*offset + sizeof(short_ad)) > maxoffset) return NULL; + else { + sa = (short_ad *)ptr; + if (sa->extLength == 0) + return NULL; + } if (inc) *offset += sizeof(short_ad); return sa; } -long_ad *udf_get_filelongad(uint8_t *ptr, int maxoffset, int *offset, int inc) +long_ad *udf_get_filelongad(uint8_t *ptr, int maxoffset, uint32_t *offset, int inc) { long_ad *la; @@ -300,10 +313,13 @@ long_ad *udf_get_filelongad(uint8_t *ptr, int maxoffset, int *offset, int inc) return NULL; } - if ((*offset < 0) || ((*offset + sizeof(long_ad)) > maxoffset)) - return NULL; - else if ((la = (long_ad *)ptr)->extLength == 0) + if ((*offset + sizeof(long_ad)) > maxoffset) return NULL; + else { + la = (long_ad *)ptr; + if (la->extLength == 0) + return NULL; + } if (inc) *offset += sizeof(long_ad); diff --git a/fs/udf/file.c b/fs/udf/file.c index 7c7a1b39d56..97c71ae7c68 100644 --- a/fs/udf/file.c +++ b/fs/udf/file.c @@ -45,12 +45,13 @@ static int udf_adinicb_readpage(struct file *file, struct page *page) { struct inode *inode = page->mapping->host; char *kaddr; + struct udf_inode_info *iinfo = UDF_I(inode); BUG_ON(!PageLocked(page)); kaddr = kmap(page); memset(kaddr, 0, PAGE_CACHE_SIZE); - memcpy(kaddr, UDF_I_DATA(inode) + UDF_I_LENEATTR(inode), inode->i_size); + memcpy(kaddr, iinfo->i_ext.i_data + iinfo->i_lenEAttr, inode->i_size); flush_dcache_page(page); SetPageUptodate(page); kunmap(page); @@ -59,15 +60,17 @@ static int udf_adinicb_readpage(struct file *file, struct page *page) return 0; } -static int udf_adinicb_writepage(struct page *page, struct writeback_control *wbc) +static int udf_adinicb_writepage(struct page *page, + struct writeback_control *wbc) { struct inode *inode = page->mapping->host; char *kaddr; + struct udf_inode_info *iinfo = UDF_I(inode); BUG_ON(!PageLocked(page)); kaddr = kmap(page); - memcpy(UDF_I_DATA(inode) + UDF_I_LENEATTR(inode), kaddr, inode->i_size); + memcpy(iinfo->i_ext.i_data + iinfo->i_lenEAttr, kaddr, inode->i_size); mark_inode_dirty(inode); SetPageUptodate(page); kunmap(page); @@ -84,9 +87,10 @@ static int udf_adinicb_write_end(struct file *file, struct inode *inode = mapping->host; unsigned offset = pos & (PAGE_CACHE_SIZE - 1); char *kaddr; + struct udf_inode_info *iinfo = UDF_I(inode); kaddr = kmap_atomic(page, KM_USER0); - memcpy(UDF_I_DATA(inode) + UDF_I_LENEATTR(inode) + offset, + memcpy(iinfo->i_ext.i_data + iinfo->i_lenEAttr + offset, kaddr + offset, copied); kunmap_atomic(kaddr, KM_USER0); @@ -109,25 +113,27 @@ static ssize_t udf_file_aio_write(struct kiocb *iocb, const struct iovec *iov, struct inode *inode = file->f_path.dentry->d_inode; int err, pos; size_t count = iocb->ki_left; + struct udf_inode_info *iinfo = UDF_I(inode); - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) { + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) { if (file->f_flags & O_APPEND) pos = inode->i_size; else pos = ppos; - if (inode->i_sb->s_blocksize < (udf_file_entry_alloc_offset(inode) + + if (inode->i_sb->s_blocksize < + (udf_file_entry_alloc_offset(inode) + pos + count)) { udf_expand_file_adinicb(inode, pos + count, &err); - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) { + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) { udf_debug("udf_expand_adinicb: err=%d\n", err); return err; } } else { if (pos + count > inode->i_size) - UDF_I_LENALLOC(inode) = pos + count; + iinfo->i_lenAlloc = pos + count; else - UDF_I_LENALLOC(inode) = inode->i_size; + iinfo->i_lenAlloc = inode->i_size; } } @@ -191,23 +197,28 @@ int udf_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, switch (cmd) { case UDF_GETVOLIDENT: - return copy_to_user((char __user *)arg, - UDF_SB_VOLIDENT(inode->i_sb), 32) ? -EFAULT : 0; + if (copy_to_user((char __user *)arg, + UDF_SB(inode->i_sb)->s_volume_ident, 32)) + return -EFAULT; + else + return 0; case UDF_RELOCATE_BLOCKS: if (!capable(CAP_SYS_ADMIN)) return -EACCES; if (get_user(old_block, (long __user *)arg)) return -EFAULT; - if ((result = udf_relocate_blocks(inode->i_sb, - old_block, &new_block)) == 0) + result = udf_relocate_blocks(inode->i_sb, + old_block, &new_block); + if (result == 0) result = put_user(new_block, (long __user *)arg); return result; case UDF_GETEASIZE: - result = put_user(UDF_I_LENEATTR(inode), (int __user *)arg); + result = put_user(UDF_I(inode)->i_lenEAttr, (int __user *)arg); break; case UDF_GETEABLOCK: - result = copy_to_user((char __user *)arg, UDF_I_DATA(inode), - UDF_I_LENEATTR(inode)) ? -EFAULT : 0; + result = copy_to_user((char __user *)arg, + UDF_I(inode)->i_ext.i_data, + UDF_I(inode)->i_lenEAttr) ? -EFAULT : 0; break; } diff --git a/fs/udf/ialloc.c b/fs/udf/ialloc.c index 636d8f61392..84360315aca 100644 --- a/fs/udf/ialloc.c +++ b/fs/udf/ialloc.c @@ -43,19 +43,21 @@ void udf_free_inode(struct inode *inode) clear_inode(inode); mutex_lock(&sbi->s_alloc_mutex); - if (sbi->s_lvidbh) { + if (sbi->s_lvid_bh) { + struct logicalVolIntegrityDescImpUse *lvidiu = + udf_sb_lvidiu(sbi); if (S_ISDIR(inode->i_mode)) - UDF_SB_LVIDIU(sb)->numDirs = - cpu_to_le32(le32_to_cpu(UDF_SB_LVIDIU(sb)->numDirs) - 1); + lvidiu->numDirs = + cpu_to_le32(le32_to_cpu(lvidiu->numDirs) - 1); else - UDF_SB_LVIDIU(sb)->numFiles = - cpu_to_le32(le32_to_cpu(UDF_SB_LVIDIU(sb)->numFiles) - 1); + lvidiu->numFiles = + cpu_to_le32(le32_to_cpu(lvidiu->numFiles) - 1); - mark_buffer_dirty(sbi->s_lvidbh); + mark_buffer_dirty(sbi->s_lvid_bh); } mutex_unlock(&sbi->s_alloc_mutex); - udf_free_blocks(sb, NULL, UDF_I_LOCATION(inode), 0, 1); + udf_free_blocks(sb, NULL, UDF_I(inode)->i_location, 0, 1); } struct inode *udf_new_inode(struct inode *dir, int mode, int *err) @@ -64,7 +66,9 @@ struct inode *udf_new_inode(struct inode *dir, int mode, int *err) struct udf_sb_info *sbi = UDF_SB(sb); struct inode *inode; int block; - uint32_t start = UDF_I_LOCATION(dir).logicalBlockNum; + uint32_t start = UDF_I(dir)->i_location.logicalBlockNum; + struct udf_inode_info *iinfo; + struct udf_inode_info *dinfo = UDF_I(dir); inode = new_inode(sb); @@ -74,13 +78,15 @@ struct inode *udf_new_inode(struct inode *dir, int mode, int *err) } *err = -ENOSPC; - UDF_I_UNIQUE(inode) = 0; - UDF_I_LENEXTENTS(inode) = 0; - UDF_I_NEXT_ALLOC_BLOCK(inode) = 0; - UDF_I_NEXT_ALLOC_GOAL(inode) = 0; - UDF_I_STRAT4096(inode) = 0; + iinfo = UDF_I(inode); + iinfo->i_unique = 0; + iinfo->i_lenExtents = 0; + iinfo->i_next_alloc_block = 0; + iinfo->i_next_alloc_goal = 0; + iinfo->i_strat4096 = 0; - block = udf_new_block(dir->i_sb, NULL, UDF_I_LOCATION(dir).partitionReferenceNum, + block = udf_new_block(dir->i_sb, NULL, + dinfo->i_location.partitionReferenceNum, start, err); if (*err) { iput(inode); @@ -88,21 +94,27 @@ struct inode *udf_new_inode(struct inode *dir, int mode, int *err) } mutex_lock(&sbi->s_alloc_mutex); - if (UDF_SB_LVIDBH(sb)) { + if (sbi->s_lvid_bh) { + struct logicalVolIntegrityDesc *lvid = + (struct logicalVolIntegrityDesc *) + sbi->s_lvid_bh->b_data; + struct logicalVolIntegrityDescImpUse *lvidiu = + udf_sb_lvidiu(sbi); struct logicalVolHeaderDesc *lvhd; uint64_t uniqueID; - lvhd = (struct logicalVolHeaderDesc *)(UDF_SB_LVID(sb)->logicalVolContentsUse); + lvhd = (struct logicalVolHeaderDesc *) + (lvid->logicalVolContentsUse); if (S_ISDIR(mode)) - UDF_SB_LVIDIU(sb)->numDirs = - cpu_to_le32(le32_to_cpu(UDF_SB_LVIDIU(sb)->numDirs) + 1); + lvidiu->numDirs = + cpu_to_le32(le32_to_cpu(lvidiu->numDirs) + 1); else - UDF_SB_LVIDIU(sb)->numFiles = - cpu_to_le32(le32_to_cpu(UDF_SB_LVIDIU(sb)->numFiles) + 1); - UDF_I_UNIQUE(inode) = uniqueID = le64_to_cpu(lvhd->uniqueID); + lvidiu->numFiles = + cpu_to_le32(le32_to_cpu(lvidiu->numFiles) + 1); + iinfo->i_unique = uniqueID = le64_to_cpu(lvhd->uniqueID); if (!(++uniqueID & 0x00000000FFFFFFFFUL)) uniqueID += 16; lvhd->uniqueID = cpu_to_le64(uniqueID); - mark_buffer_dirty(UDF_SB_LVIDBH(sb)); + mark_buffer_dirty(sbi->s_lvid_bh); } inode->i_mode = mode; inode->i_uid = current->fsuid; @@ -114,35 +126,41 @@ struct inode *udf_new_inode(struct inode *dir, int mode, int *err) inode->i_gid = current->fsgid; } - UDF_I_LOCATION(inode).logicalBlockNum = block; - UDF_I_LOCATION(inode).partitionReferenceNum = UDF_I_LOCATION(dir).partitionReferenceNum; - inode->i_ino = udf_get_lb_pblock(sb, UDF_I_LOCATION(inode), 0); + iinfo->i_location.logicalBlockNum = block; + iinfo->i_location.partitionReferenceNum = + dinfo->i_location.partitionReferenceNum; + inode->i_ino = udf_get_lb_pblock(sb, iinfo->i_location, 0); inode->i_blocks = 0; - UDF_I_LENEATTR(inode) = 0; - UDF_I_LENALLOC(inode) = 0; - UDF_I_USE(inode) = 0; + iinfo->i_lenEAttr = 0; + iinfo->i_lenAlloc = 0; + iinfo->i_use = 0; if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_EXTENDED_FE)) { - UDF_I_EFE(inode) = 1; - UDF_UPDATE_UDFREV(inode->i_sb, UDF_VERS_USE_EXTENDED_FE); - UDF_I_DATA(inode) = kzalloc(inode->i_sb->s_blocksize - sizeof(struct extendedFileEntry), GFP_KERNEL); + iinfo->i_efe = 1; + if (UDF_VERS_USE_EXTENDED_FE > sbi->s_udfrev) + sbi->s_udfrev = UDF_VERS_USE_EXTENDED_FE; + iinfo->i_ext.i_data = kzalloc(inode->i_sb->s_blocksize - + sizeof(struct extendedFileEntry), + GFP_KERNEL); } else { - UDF_I_EFE(inode) = 0; - UDF_I_DATA(inode) = kzalloc(inode->i_sb->s_blocksize - sizeof(struct fileEntry), GFP_KERNEL); + iinfo->i_efe = 0; + iinfo->i_ext.i_data = kzalloc(inode->i_sb->s_blocksize - + sizeof(struct fileEntry), + GFP_KERNEL); } - if (!UDF_I_DATA(inode)) { + if (!iinfo->i_ext.i_data) { iput(inode); *err = -ENOMEM; mutex_unlock(&sbi->s_alloc_mutex); return NULL; } if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_AD_IN_ICB)) - UDF_I_ALLOCTYPE(inode) = ICBTAG_FLAG_AD_IN_ICB; + iinfo->i_alloc_type = ICBTAG_FLAG_AD_IN_ICB; else if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_SHORT_AD)) - UDF_I_ALLOCTYPE(inode) = ICBTAG_FLAG_AD_SHORT; + iinfo->i_alloc_type = ICBTAG_FLAG_AD_SHORT; else - UDF_I_ALLOCTYPE(inode) = ICBTAG_FLAG_AD_LONG; + iinfo->i_alloc_type = ICBTAG_FLAG_AD_LONG; inode->i_mtime = inode->i_atime = inode->i_ctime = - UDF_I_CRTIME(inode) = current_fs_time(inode->i_sb); + iinfo->i_crtime = current_fs_time(inode->i_sb); insert_inode_hash(inode); mark_inode_dirty(inode); mutex_unlock(&sbi->s_alloc_mutex); diff --git a/fs/udf/inode.c b/fs/udf/inode.c index 6ff8151984c..24cfa55d0fd 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -19,7 +19,8 @@ * 10/04/98 dgb Added rudimentary directory functions * 10/07/98 Fully working udf_block_map! It works! * 11/25/98 bmap altered to better support extents - * 12/06/98 blf partition support in udf_iget, udf_block_map and udf_read_inode + * 12/06/98 blf partition support in udf_iget, udf_block_map + * and udf_read_inode * 12/12/98 rewrote udf_block_map to handle next extents and descs across * block boundaries (which is not actually allowed) * 12/20/98 added support for strategy 4096 @@ -51,7 +52,7 @@ static int udf_update_inode(struct inode *, int); static void udf_fill_inode(struct inode *, struct buffer_head *); static int udf_alloc_i_data(struct inode *inode, size_t size); static struct buffer_head *inode_getblk(struct inode *, sector_t, int *, - long *, int *); + sector_t *, int *); static int8_t udf_insert_aext(struct inode *, struct extent_position, kernel_lb_addr, uint32_t); static void udf_split_extents(struct inode *, int *, int, int, @@ -111,16 +112,18 @@ no_delete: */ void udf_clear_inode(struct inode *inode) { + struct udf_inode_info *iinfo; if (!(inode->i_sb->s_flags & MS_RDONLY)) { lock_kernel(); /* Discard preallocation for directories, symlinks, etc. */ udf_discard_prealloc(inode); udf_truncate_tail_extent(inode); unlock_kernel(); - write_inode_now(inode, 1); + write_inode_now(inode, 0); } - kfree(UDF_I_DATA(inode)); - UDF_I_DATA(inode) = NULL; + iinfo = UDF_I(inode); + kfree(iinfo->i_ext.i_data); + iinfo->i_ext.i_data = NULL; } static int udf_writepage(struct page *page, struct writeback_control *wbc) @@ -160,6 +163,7 @@ void udf_expand_file_adinicb(struct inode *inode, int newsize, int *err) { struct page *page; char *kaddr; + struct udf_inode_info *iinfo = UDF_I(inode); struct writeback_control udf_wbc = { .sync_mode = WB_SYNC_NONE, .nr_to_write = 1, @@ -168,11 +172,11 @@ void udf_expand_file_adinicb(struct inode *inode, int newsize, int *err) /* from now on we have normal address_space methods */ inode->i_data.a_ops = &udf_aops; - if (!UDF_I_LENALLOC(inode)) { + if (!iinfo->i_lenAlloc) { if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_SHORT_AD)) - UDF_I_ALLOCTYPE(inode) = ICBTAG_FLAG_AD_SHORT; + iinfo->i_alloc_type = ICBTAG_FLAG_AD_SHORT; else - UDF_I_ALLOCTYPE(inode) = ICBTAG_FLAG_AD_LONG; + iinfo->i_alloc_type = ICBTAG_FLAG_AD_LONG; mark_inode_dirty(inode); return; } @@ -182,21 +186,21 @@ void udf_expand_file_adinicb(struct inode *inode, int newsize, int *err) if (!PageUptodate(page)) { kaddr = kmap(page); - memset(kaddr + UDF_I_LENALLOC(inode), 0x00, - PAGE_CACHE_SIZE - UDF_I_LENALLOC(inode)); - memcpy(kaddr, UDF_I_DATA(inode) + UDF_I_LENEATTR(inode), - UDF_I_LENALLOC(inode)); + memset(kaddr + iinfo->i_lenAlloc, 0x00, + PAGE_CACHE_SIZE - iinfo->i_lenAlloc); + memcpy(kaddr, iinfo->i_ext.i_data + iinfo->i_lenEAttr, + iinfo->i_lenAlloc); flush_dcache_page(page); SetPageUptodate(page); kunmap(page); } - memset(UDF_I_DATA(inode) + UDF_I_LENEATTR(inode), 0x00, - UDF_I_LENALLOC(inode)); - UDF_I_LENALLOC(inode) = 0; + memset(iinfo->i_ext.i_data + iinfo->i_lenEAttr, 0x00, + iinfo->i_lenAlloc); + iinfo->i_lenAlloc = 0; if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_SHORT_AD)) - UDF_I_ALLOCTYPE(inode) = ICBTAG_FLAG_AD_SHORT; + iinfo->i_alloc_type = ICBTAG_FLAG_AD_SHORT; else - UDF_I_ALLOCTYPE(inode) = ICBTAG_FLAG_AD_LONG; + iinfo->i_alloc_type = ICBTAG_FLAG_AD_LONG; inode->i_data.a_ops->writepage(page, &udf_wbc); page_cache_release(page); @@ -215,9 +219,10 @@ struct buffer_head *udf_expand_dir_adinicb(struct inode *inode, int *block, struct extent_position epos; struct udf_fileident_bh sfibh, dfibh; - loff_t f_pos = udf_ext0_offset(inode) >> 2; - int size = (udf_ext0_offset(inode) + inode->i_size) >> 2; + loff_t f_pos = udf_ext0_offset(inode); + int size = udf_ext0_offset(inode) + inode->i_size; struct fileIdentDesc cfi, *sfi, *dfi; + struct udf_inode_info *iinfo = UDF_I(inode); if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_SHORT_AD)) alloctype = ICBTAG_FLAG_AD_SHORT; @@ -225,19 +230,20 @@ struct buffer_head *udf_expand_dir_adinicb(struct inode *inode, int *block, alloctype = ICBTAG_FLAG_AD_LONG; if (!inode->i_size) { - UDF_I_ALLOCTYPE(inode) = alloctype; + iinfo->i_alloc_type = alloctype; mark_inode_dirty(inode); return NULL; } /* alloc block, and copy data to it */ *block = udf_new_block(inode->i_sb, inode, - UDF_I_LOCATION(inode).partitionReferenceNum, - UDF_I_LOCATION(inode).logicalBlockNum, err); + iinfo->i_location.partitionReferenceNum, + iinfo->i_location.logicalBlockNum, err); if (!(*block)) return NULL; newblock = udf_get_pblock(inode->i_sb, *block, - UDF_I_LOCATION(inode).partitionReferenceNum, 0); + iinfo->i_location.partitionReferenceNum, + 0); if (!newblock) return NULL; dbh = udf_tgetblk(inode->i_sb, newblock); @@ -249,39 +255,44 @@ struct buffer_head *udf_expand_dir_adinicb(struct inode *inode, int *block, unlock_buffer(dbh); mark_buffer_dirty_inode(dbh, inode); - sfibh.soffset = sfibh.eoffset = (f_pos & ((inode->i_sb->s_blocksize - 1) >> 2)) << 2; + sfibh.soffset = sfibh.eoffset = + f_pos & (inode->i_sb->s_blocksize - 1); sfibh.sbh = sfibh.ebh = NULL; dfibh.soffset = dfibh.eoffset = 0; dfibh.sbh = dfibh.ebh = dbh; - while ((f_pos < size)) { - UDF_I_ALLOCTYPE(inode) = ICBTAG_FLAG_AD_IN_ICB; - sfi = udf_fileident_read(inode, &f_pos, &sfibh, &cfi, NULL, NULL, NULL, NULL); + while (f_pos < size) { + iinfo->i_alloc_type = ICBTAG_FLAG_AD_IN_ICB; + sfi = udf_fileident_read(inode, &f_pos, &sfibh, &cfi, NULL, + NULL, NULL, NULL); if (!sfi) { brelse(dbh); return NULL; } - UDF_I_ALLOCTYPE(inode) = alloctype; + iinfo->i_alloc_type = alloctype; sfi->descTag.tagLocation = cpu_to_le32(*block); dfibh.soffset = dfibh.eoffset; dfibh.eoffset += (sfibh.eoffset - sfibh.soffset); dfi = (struct fileIdentDesc *)(dbh->b_data + dfibh.soffset); if (udf_write_fi(inode, sfi, dfi, &dfibh, sfi->impUse, - sfi->fileIdent + le16_to_cpu(sfi->lengthOfImpUse))) { - UDF_I_ALLOCTYPE(inode) = ICBTAG_FLAG_AD_IN_ICB; + sfi->fileIdent + + le16_to_cpu(sfi->lengthOfImpUse))) { + iinfo->i_alloc_type = ICBTAG_FLAG_AD_IN_ICB; brelse(dbh); return NULL; } } mark_buffer_dirty_inode(dbh, inode); - memset(UDF_I_DATA(inode) + UDF_I_LENEATTR(inode), 0, UDF_I_LENALLOC(inode)); - UDF_I_LENALLOC(inode) = 0; + memset(iinfo->i_ext.i_data + iinfo->i_lenEAttr, 0, + iinfo->i_lenAlloc); + iinfo->i_lenAlloc = 0; eloc.logicalBlockNum = *block; - eloc.partitionReferenceNum = UDF_I_LOCATION(inode).partitionReferenceNum; - elen = inode->i_size; - UDF_I_LENEXTENTS(inode) = elen; + eloc.partitionReferenceNum = + iinfo->i_location.partitionReferenceNum; + elen = inode->i_sb->s_blocksize; + iinfo->i_lenExtents = elen; epos.bh = NULL; - epos.block = UDF_I_LOCATION(inode); + epos.block = iinfo->i_location; epos.offset = udf_file_entry_alloc_offset(inode); udf_add_aext(inode, &epos, eloc, elen, 0); /* UniqueID stuff */ @@ -296,7 +307,8 @@ static int udf_get_block(struct inode *inode, sector_t block, { int err, new; struct buffer_head *bh; - unsigned long phys; + sector_t phys = 0; + struct udf_inode_info *iinfo; if (!create) { phys = udf_block_map(inode, block); @@ -314,9 +326,10 @@ static int udf_get_block(struct inode *inode, sector_t block, if (block < 0) goto abort_negative; - if (block == UDF_I_NEXT_ALLOC_BLOCK(inode) + 1) { - UDF_I_NEXT_ALLOC_BLOCK(inode)++; - UDF_I_NEXT_ALLOC_GOAL(inode)++; + iinfo = UDF_I(inode); + if (block == iinfo->i_next_alloc_block + 1) { + iinfo->i_next_alloc_block++; + iinfo->i_next_alloc_goal++; } err = 0; @@ -366,32 +379,35 @@ static struct buffer_head *udf_getblk(struct inode *inode, long block, /* Extend the file by 'blocks' blocks, return the number of extents added */ int udf_extend_file(struct inode *inode, struct extent_position *last_pos, - kernel_long_ad * last_ext, sector_t blocks) + kernel_long_ad *last_ext, sector_t blocks) { sector_t add; int count = 0, fake = !(last_ext->extLength & UDF_EXTENT_LENGTH_MASK); struct super_block *sb = inode->i_sb; kernel_lb_addr prealloc_loc = {}; int prealloc_len = 0; + struct udf_inode_info *iinfo; /* The previous extent is fake and we should not extend by anything * - there's nothing to do... */ if (!blocks && fake) return 0; + iinfo = UDF_I(inode); /* Round the last extent up to a multiple of block size */ if (last_ext->extLength & (sb->s_blocksize - 1)) { last_ext->extLength = (last_ext->extLength & UDF_EXTENT_FLAG_MASK) | (((last_ext->extLength & UDF_EXTENT_LENGTH_MASK) + sb->s_blocksize - 1) & ~(sb->s_blocksize - 1)); - UDF_I_LENEXTENTS(inode) = - (UDF_I_LENEXTENTS(inode) + sb->s_blocksize - 1) & + iinfo->i_lenExtents = + (iinfo->i_lenExtents + sb->s_blocksize - 1) & ~(sb->s_blocksize - 1); } /* Last extent are just preallocated blocks? */ - if ((last_ext->extLength & UDF_EXTENT_FLAG_MASK) == EXT_NOT_RECORDED_ALLOCATED) { + if ((last_ext->extLength & UDF_EXTENT_FLAG_MASK) == + EXT_NOT_RECORDED_ALLOCATED) { /* Save the extent so that we can reattach it to the end */ prealloc_loc = last_ext->extLocation; prealloc_len = last_ext->extLength; @@ -399,13 +415,15 @@ int udf_extend_file(struct inode *inode, struct extent_position *last_pos, last_ext->extLength = EXT_NOT_RECORDED_NOT_ALLOCATED | (last_ext->extLength & UDF_EXTENT_LENGTH_MASK); last_ext->extLocation.logicalBlockNum = 0; - last_ext->extLocation.partitionReferenceNum = 0; + last_ext->extLocation.partitionReferenceNum = 0; } /* Can we merge with the previous extent? */ - if ((last_ext->extLength & UDF_EXTENT_FLAG_MASK) == EXT_NOT_RECORDED_NOT_ALLOCATED) { - add = ((1 << 30) - sb->s_blocksize - (last_ext->extLength & - UDF_EXTENT_LENGTH_MASK)) >> sb->s_blocksize_bits; + if ((last_ext->extLength & UDF_EXTENT_FLAG_MASK) == + EXT_NOT_RECORDED_NOT_ALLOCATED) { + add = ((1 << 30) - sb->s_blocksize - + (last_ext->extLength & UDF_EXTENT_LENGTH_MASK)) >> + sb->s_blocksize_bits; if (add > blocks) add = blocks; blocks -= add; @@ -416,9 +434,9 @@ int udf_extend_file(struct inode *inode, struct extent_position *last_pos, udf_add_aext(inode, last_pos, last_ext->extLocation, last_ext->extLength, 1); count++; - } else { - udf_write_aext(inode, last_pos, last_ext->extLocation, last_ext->extLength, 1); - } + } else + udf_write_aext(inode, last_pos, last_ext->extLocation, + last_ext->extLength, 1); /* Managed to do everything necessary? */ if (!blocks) @@ -426,9 +444,10 @@ int udf_extend_file(struct inode *inode, struct extent_position *last_pos, /* All further extents will be NOT_RECORDED_NOT_ALLOCATED */ last_ext->extLocation.logicalBlockNum = 0; - last_ext->extLocation.partitionReferenceNum = 0; + last_ext->extLocation.partitionReferenceNum = 0; add = (1 << (30-sb->s_blocksize_bits)) - 1; - last_ext->extLength = EXT_NOT_RECORDED_NOT_ALLOCATED | (add << sb->s_blocksize_bits); + last_ext->extLength = EXT_NOT_RECORDED_NOT_ALLOCATED | + (add << sb->s_blocksize_bits); /* Create enough extents to cover the whole hole */ while (blocks > add) { @@ -450,7 +469,8 @@ int udf_extend_file(struct inode *inode, struct extent_position *last_pos, out: /* Do we have some preallocated blocks saved? */ if (prealloc_len) { - if (udf_add_aext(inode, last_pos, prealloc_loc, prealloc_len, 1) == -1) + if (udf_add_aext(inode, last_pos, prealloc_loc, + prealloc_len, 1) == -1) return -1; last_ext->extLocation = prealloc_loc; last_ext->extLength = prealloc_len; @@ -458,9 +478,9 @@ out: } /* last_pos should point to the last written extent... */ - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_SHORT) + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT) last_pos->offset -= sizeof(short_ad); - else if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_LONG) + else if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_LONG) last_pos->offset -= sizeof(long_ad); else return -1; @@ -469,7 +489,7 @@ out: } static struct buffer_head *inode_getblk(struct inode *inode, sector_t block, - int *err, long *phys, int *new) + int *err, sector_t *phys, int *new) { static sector_t last_block; struct buffer_head *result = NULL; @@ -483,11 +503,12 @@ static struct buffer_head *inode_getblk(struct inode *inode, sector_t block, uint32_t newblocknum, newblock; sector_t offset = 0; int8_t etype; - int goal = 0, pgoal = UDF_I_LOCATION(inode).logicalBlockNum; + struct udf_inode_info *iinfo = UDF_I(inode); + int goal = 0, pgoal = iinfo->i_location.logicalBlockNum; int lastblock = 0; prev_epos.offset = udf_file_entry_alloc_offset(inode); - prev_epos.block = UDF_I_LOCATION(inode); + prev_epos.block = iinfo->i_location; prev_epos.bh = NULL; cur_epos = next_epos = prev_epos; b_off = (loff_t)block << inode->i_sb->s_blocksize_bits; @@ -515,7 +536,8 @@ static struct buffer_head *inode_getblk(struct inode *inode, sector_t block, prev_epos.offset = cur_epos.offset; cur_epos.offset = next_epos.offset; - if ((etype = udf_next_aext(inode, &next_epos, &eloc, &elen, 1)) == -1) + etype = udf_next_aext(inode, &next_epos, &eloc, &elen, 1); + if (etype == -1) break; c = !c; @@ -569,9 +591,11 @@ static struct buffer_head *inode_getblk(struct inode *inode, sector_t block, startnum = 1; } else { /* Create a fake extent when there's not one */ - memset(&laarr[0].extLocation, 0x00, sizeof(kernel_lb_addr)); + memset(&laarr[0].extLocation, 0x00, + sizeof(kernel_lb_addr)); laarr[0].extLength = EXT_NOT_RECORDED_NOT_ALLOCATED; - /* Will udf_extend_file() create real extent from a fake one? */ + /* Will udf_extend_file() create real extent from + a fake one? */ startnum = (offset > 0); } /* Create extents for the hole between EOF and offset */ @@ -589,14 +613,16 @@ static struct buffer_head *inode_getblk(struct inode *inode, sector_t block, offset = 0; count += ret; /* We are not covered by a preallocated extent? */ - if ((laarr[0].extLength & UDF_EXTENT_FLAG_MASK) != EXT_NOT_RECORDED_ALLOCATED) { + if ((laarr[0].extLength & UDF_EXTENT_FLAG_MASK) != + EXT_NOT_RECORDED_ALLOCATED) { /* Is there any real extent? - otherwise we overwrite * the fake one... */ if (count) c = !c; laarr[c].extLength = EXT_NOT_RECORDED_NOT_ALLOCATED | inode->i_sb->s_blocksize; - memset(&laarr[c].extLocation, 0x00, sizeof(kernel_lb_addr)); + memset(&laarr[c].extLocation, 0x00, + sizeof(kernel_lb_addr)); count++; endnum++; } @@ -605,7 +631,8 @@ static struct buffer_head *inode_getblk(struct inode *inode, sector_t block, } else { endnum = startnum = ((count > 2) ? 2 : count); - /* if the current extent is in position 0, swap it with the previous */ + /* if the current extent is in position 0, + swap it with the previous */ if (!c && count != 1) { laarr[2] = laarr[0]; laarr[0] = laarr[1]; @@ -613,44 +640,47 @@ static struct buffer_head *inode_getblk(struct inode *inode, sector_t block, c = 1; } - /* if the current block is located in an extent, read the next extent */ - if ((etype = udf_next_aext(inode, &next_epos, &eloc, &elen, 0)) != -1) { + /* if the current block is located in an extent, + read the next extent */ + etype = udf_next_aext(inode, &next_epos, &eloc, &elen, 0); + if (etype != -1) { laarr[c + 1].extLength = (etype << 30) | elen; laarr[c + 1].extLocation = eloc; count++; startnum++; endnum++; - } else { + } else lastblock = 1; - } } /* if the current extent is not recorded but allocated, get the * block in the extent corresponding to the requested block */ - if ((laarr[c].extLength >> 30) == (EXT_NOT_RECORDED_ALLOCATED >> 30)) { + if ((laarr[c].extLength >> 30) == (EXT_NOT_RECORDED_ALLOCATED >> 30)) newblocknum = laarr[c].extLocation.logicalBlockNum + offset; - } else { /* otherwise, allocate a new block */ - if (UDF_I_NEXT_ALLOC_BLOCK(inode) == block) - goal = UDF_I_NEXT_ALLOC_GOAL(inode); + else { /* otherwise, allocate a new block */ + if (iinfo->i_next_alloc_block == block) + goal = iinfo->i_next_alloc_goal; if (!goal) { - if (!(goal = pgoal)) - goal = UDF_I_LOCATION(inode).logicalBlockNum + 1; + if (!(goal = pgoal)) /* XXX: what was intended here? */ + goal = iinfo->i_location.logicalBlockNum + 1; } - if (!(newblocknum = udf_new_block(inode->i_sb, inode, - UDF_I_LOCATION(inode).partitionReferenceNum, - goal, err))) { + newblocknum = udf_new_block(inode->i_sb, inode, + iinfo->i_location.partitionReferenceNum, + goal, err); + if (!newblocknum) { brelse(prev_epos.bh); *err = -ENOSPC; return NULL; } - UDF_I_LENEXTENTS(inode) += inode->i_sb->s_blocksize; + iinfo->i_lenExtents += inode->i_sb->s_blocksize; } - /* if the extent the requsted block is located in contains multiple blocks, - * split the extent into at most three extents. blocks prior to requested - * block, requested block, and blocks after requested block */ + /* if the extent the requsted block is located in contains multiple + * blocks, split the extent into at most three extents. blocks prior + * to requested block, requested block, and blocks after requested + * block */ udf_split_extents(inode, &c, offset, newblocknum, laarr, &endnum); #ifdef UDF_PREALLOCATE @@ -668,15 +698,15 @@ static struct buffer_head *inode_getblk(struct inode *inode, sector_t block, brelse(prev_epos.bh); - if (!(newblock = udf_get_pblock(inode->i_sb, newblocknum, - UDF_I_LOCATION(inode).partitionReferenceNum, 0))) { + newblock = udf_get_pblock(inode->i_sb, newblocknum, + iinfo->i_location.partitionReferenceNum, 0); + if (!newblock) return NULL; - } *phys = newblock; *err = 0; *new = 1; - UDF_I_NEXT_ALLOC_BLOCK(inode) = block; - UDF_I_NEXT_ALLOC_GOAL(inode) = newblocknum; + iinfo->i_next_alloc_block = block; + iinfo->i_next_alloc_goal = newblocknum; inode->i_ctime = current_fs_time(inode->i_sb); if (IS_SYNC(inode)) @@ -692,16 +722,20 @@ static void udf_split_extents(struct inode *inode, int *c, int offset, kernel_long_ad laarr[EXTENT_MERGE_SIZE], int *endnum) { + unsigned long blocksize = inode->i_sb->s_blocksize; + unsigned char blocksize_bits = inode->i_sb->s_blocksize_bits; + if ((laarr[*c].extLength >> 30) == (EXT_NOT_RECORDED_ALLOCATED >> 30) || - (laarr[*c].extLength >> 30) == (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) { + (laarr[*c].extLength >> 30) == + (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) { int curr = *c; int blen = ((laarr[curr].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits; + blocksize - 1) >> blocksize_bits; int8_t etype = (laarr[curr].extLength >> 30); - if (blen == 1) { + if (blen == 1) ; - } else if (!offset || blen == offset + 1) { + else if (!offset || blen == offset + 1) { laarr[curr + 2] = laarr[curr + 1]; laarr[curr + 1] = laarr[curr]; } else { @@ -711,15 +745,18 @@ static void udf_split_extents(struct inode *inode, int *c, int offset, if (offset) { if (etype == (EXT_NOT_RECORDED_ALLOCATED >> 30)) { - udf_free_blocks(inode->i_sb, inode, laarr[curr].extLocation, 0, offset); - laarr[curr].extLength = EXT_NOT_RECORDED_NOT_ALLOCATED | - (offset << inode->i_sb->s_blocksize_bits); + udf_free_blocks(inode->i_sb, inode, + laarr[curr].extLocation, + 0, offset); + laarr[curr].extLength = + EXT_NOT_RECORDED_NOT_ALLOCATED | + (offset << blocksize_bits); laarr[curr].extLocation.logicalBlockNum = 0; - laarr[curr].extLocation.partitionReferenceNum = 0; - } else { + laarr[curr].extLocation. + partitionReferenceNum = 0; + } else laarr[curr].extLength = (etype << 30) | - (offset << inode->i_sb->s_blocksize_bits); - } + (offset << blocksize_bits); curr++; (*c)++; (*endnum)++; @@ -728,16 +765,17 @@ static void udf_split_extents(struct inode *inode, int *c, int offset, laarr[curr].extLocation.logicalBlockNum = newblocknum; if (etype == (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) laarr[curr].extLocation.partitionReferenceNum = - UDF_I_LOCATION(inode).partitionReferenceNum; + UDF_I(inode)->i_location.partitionReferenceNum; laarr[curr].extLength = EXT_RECORDED_ALLOCATED | - inode->i_sb->s_blocksize; + blocksize; curr++; if (blen != offset + 1) { if (etype == (EXT_NOT_RECORDED_ALLOCATED >> 30)) - laarr[curr].extLocation.logicalBlockNum += (offset + 1); + laarr[curr].extLocation.logicalBlockNum += + offset + 1; laarr[curr].extLength = (etype << 30) | - ((blen - (offset + 1)) << inode->i_sb->s_blocksize_bits); + ((blen - (offset + 1)) << blocksize_bits); curr++; (*endnum)++; } @@ -756,69 +794,86 @@ static void udf_prealloc_extents(struct inode *inode, int c, int lastblock, else start = c; } else { - if ((laarr[c + 1].extLength >> 30) == (EXT_NOT_RECORDED_ALLOCATED >> 30)) { + if ((laarr[c + 1].extLength >> 30) == + (EXT_NOT_RECORDED_ALLOCATED >> 30)) { start = c + 1; - length = currlength = (((laarr[c + 1].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits); - } else { + length = currlength = + (((laarr[c + 1].extLength & + UDF_EXTENT_LENGTH_MASK) + + inode->i_sb->s_blocksize - 1) >> + inode->i_sb->s_blocksize_bits); + } else start = c; - } } for (i = start + 1; i <= *endnum; i++) { if (i == *endnum) { if (lastblock) length += UDF_DEFAULT_PREALLOC_BLOCKS; - } else if ((laarr[i].extLength >> 30) == (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) { - length += (((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits); - } else { + } else if ((laarr[i].extLength >> 30) == + (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) { + length += (((laarr[i].extLength & + UDF_EXTENT_LENGTH_MASK) + + inode->i_sb->s_blocksize - 1) >> + inode->i_sb->s_blocksize_bits); + } else break; - } } if (length) { int next = laarr[start].extLocation.logicalBlockNum + (((laarr[start].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits); + inode->i_sb->s_blocksize - 1) >> + inode->i_sb->s_blocksize_bits); int numalloc = udf_prealloc_blocks(inode->i_sb, inode, - laarr[start].extLocation.partitionReferenceNum, - next, (UDF_DEFAULT_PREALLOC_BLOCKS > length ? length : - UDF_DEFAULT_PREALLOC_BLOCKS) - currlength); + laarr[start].extLocation.partitionReferenceNum, + next, (UDF_DEFAULT_PREALLOC_BLOCKS > length ? + length : UDF_DEFAULT_PREALLOC_BLOCKS) - + currlength); if (numalloc) { - if (start == (c + 1)) { + if (start == (c + 1)) laarr[start].extLength += - (numalloc << inode->i_sb->s_blocksize_bits); - } else { + (numalloc << + inode->i_sb->s_blocksize_bits); + else { memmove(&laarr[c + 2], &laarr[c + 1], sizeof(long_ad) * (*endnum - (c + 1))); (*endnum)++; laarr[c + 1].extLocation.logicalBlockNum = next; laarr[c + 1].extLocation.partitionReferenceNum = - laarr[c].extLocation.partitionReferenceNum; - laarr[c + 1].extLength = EXT_NOT_RECORDED_ALLOCATED | - (numalloc << inode->i_sb->s_blocksize_bits); + laarr[c].extLocation. + partitionReferenceNum; + laarr[c + 1].extLength = + EXT_NOT_RECORDED_ALLOCATED | + (numalloc << + inode->i_sb->s_blocksize_bits); start = c + 1; } for (i = start + 1; numalloc && i < *endnum; i++) { - int elen = ((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits; + int elen = ((laarr[i].extLength & + UDF_EXTENT_LENGTH_MASK) + + inode->i_sb->s_blocksize - 1) >> + inode->i_sb->s_blocksize_bits; if (elen > numalloc) { laarr[i].extLength -= - (numalloc << inode->i_sb->s_blocksize_bits); + (numalloc << + inode->i_sb->s_blocksize_bits); numalloc = 0; } else { numalloc -= elen; if (*endnum > (i + 1)) - memmove(&laarr[i], &laarr[i + 1], - sizeof(long_ad) * (*endnum - (i + 1))); + memmove(&laarr[i], + &laarr[i + 1], + sizeof(long_ad) * + (*endnum - (i + 1))); i--; (*endnum)--; } } - UDF_I_LENEXTENTS(inode) += numalloc << inode->i_sb->s_blocksize_bits; + UDF_I(inode)->i_lenExtents += + numalloc << inode->i_sb->s_blocksize_bits; } } } @@ -828,70 +883,97 @@ static void udf_merge_extents(struct inode *inode, int *endnum) { int i; + unsigned long blocksize = inode->i_sb->s_blocksize; + unsigned char blocksize_bits = inode->i_sb->s_blocksize_bits; for (i = 0; i < (*endnum - 1); i++) { - if ((laarr[i].extLength >> 30) == (laarr[i + 1].extLength >> 30)) { - if (((laarr[i].extLength >> 30) == (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) || - ((laarr[i + 1].extLocation.logicalBlockNum - laarr[i].extLocation.logicalBlockNum) == - (((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits))) { - if (((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - (laarr[i + 1].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) & ~UDF_EXTENT_LENGTH_MASK) { - laarr[i + 1].extLength = (laarr[i + 1].extLength - - (laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - UDF_EXTENT_LENGTH_MASK) & ~(inode->i_sb->s_blocksize - 1); - laarr[i].extLength = (laarr[i].extLength & UDF_EXTENT_FLAG_MASK) + - (UDF_EXTENT_LENGTH_MASK + 1) - inode->i_sb->s_blocksize; - laarr[i + 1].extLocation.logicalBlockNum = - laarr[i].extLocation.logicalBlockNum + - ((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) >> - inode->i_sb->s_blocksize_bits); - } else { - laarr[i].extLength = laarr[i + 1].extLength + - (((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) & ~(inode->i_sb->s_blocksize - 1)); - if (*endnum > (i + 2)) - memmove(&laarr[i + 1], &laarr[i + 2], - sizeof(long_ad) * (*endnum - (i + 2))); - i--; - (*endnum)--; - } + kernel_long_ad *li /*l[i]*/ = &laarr[i]; + kernel_long_ad *lip1 /*l[i plus 1]*/ = &laarr[i + 1]; + + if (((li->extLength >> 30) == (lip1->extLength >> 30)) && + (((li->extLength >> 30) == + (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) || + ((lip1->extLocation.logicalBlockNum - + li->extLocation.logicalBlockNum) == + (((li->extLength & UDF_EXTENT_LENGTH_MASK) + + blocksize - 1) >> blocksize_bits)))) { + + if (((li->extLength & UDF_EXTENT_LENGTH_MASK) + + (lip1->extLength & UDF_EXTENT_LENGTH_MASK) + + blocksize - 1) & ~UDF_EXTENT_LENGTH_MASK) { + lip1->extLength = (lip1->extLength - + (li->extLength & + UDF_EXTENT_LENGTH_MASK) + + UDF_EXTENT_LENGTH_MASK) & + ~(blocksize - 1); + li->extLength = (li->extLength & + UDF_EXTENT_FLAG_MASK) + + (UDF_EXTENT_LENGTH_MASK + 1) - + blocksize; + lip1->extLocation.logicalBlockNum = + li->extLocation.logicalBlockNum + + ((li->extLength & + UDF_EXTENT_LENGTH_MASK) >> + blocksize_bits); + } else { + li->extLength = lip1->extLength + + (((li->extLength & + UDF_EXTENT_LENGTH_MASK) + + blocksize - 1) & ~(blocksize - 1)); + if (*endnum > (i + 2)) + memmove(&laarr[i + 1], &laarr[i + 2], + sizeof(long_ad) * + (*endnum - (i + 2))); + i--; + (*endnum)--; } - } else if (((laarr[i].extLength >> 30) == (EXT_NOT_RECORDED_ALLOCATED >> 30)) && - ((laarr[i + 1].extLength >> 30) == (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30))) { - udf_free_blocks(inode->i_sb, inode, laarr[i].extLocation, 0, - ((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits); - laarr[i].extLocation.logicalBlockNum = 0; - laarr[i].extLocation.partitionReferenceNum = 0; - - if (((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - (laarr[i + 1].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) & ~UDF_EXTENT_LENGTH_MASK) { - laarr[i + 1].extLength = (laarr[i + 1].extLength - - (laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - UDF_EXTENT_LENGTH_MASK) & ~(inode->i_sb->s_blocksize - 1); - laarr[i].extLength = (laarr[i].extLength & UDF_EXTENT_FLAG_MASK) + - (UDF_EXTENT_LENGTH_MASK + 1) - inode->i_sb->s_blocksize; + } else if (((li->extLength >> 30) == + (EXT_NOT_RECORDED_ALLOCATED >> 30)) && + ((lip1->extLength >> 30) == + (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30))) { + udf_free_blocks(inode->i_sb, inode, li->extLocation, 0, + ((li->extLength & + UDF_EXTENT_LENGTH_MASK) + + blocksize - 1) >> blocksize_bits); + li->extLocation.logicalBlockNum = 0; + li->extLocation.partitionReferenceNum = 0; + + if (((li->extLength & UDF_EXTENT_LENGTH_MASK) + + (lip1->extLength & UDF_EXTENT_LENGTH_MASK) + + blocksize - 1) & ~UDF_EXTENT_LENGTH_MASK) { + lip1->extLength = (lip1->extLength - + (li->extLength & + UDF_EXTENT_LENGTH_MASK) + + UDF_EXTENT_LENGTH_MASK) & + ~(blocksize - 1); + li->extLength = (li->extLength & + UDF_EXTENT_FLAG_MASK) + + (UDF_EXTENT_LENGTH_MASK + 1) - + blocksize; } else { - laarr[i].extLength = laarr[i + 1].extLength + - (((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) & ~(inode->i_sb->s_blocksize - 1)); + li->extLength = lip1->extLength + + (((li->extLength & + UDF_EXTENT_LENGTH_MASK) + + blocksize - 1) & ~(blocksize - 1)); if (*endnum > (i + 2)) memmove(&laarr[i + 1], &laarr[i + 2], - sizeof(long_ad) * (*endnum - (i + 2))); + sizeof(long_ad) * + (*endnum - (i + 2))); i--; (*endnum)--; } - } else if ((laarr[i].extLength >> 30) == (EXT_NOT_RECORDED_ALLOCATED >> 30)) { - udf_free_blocks(inode->i_sb, inode, laarr[i].extLocation, 0, - ((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits); - laarr[i].extLocation.logicalBlockNum = 0; - laarr[i].extLocation.partitionReferenceNum = 0; - laarr[i].extLength = (laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) | - EXT_NOT_RECORDED_NOT_ALLOCATED; + } else if ((li->extLength >> 30) == + (EXT_NOT_RECORDED_ALLOCATED >> 30)) { + udf_free_blocks(inode->i_sb, inode, + li->extLocation, 0, + ((li->extLength & + UDF_EXTENT_LENGTH_MASK) + + blocksize - 1) >> blocksize_bits); + li->extLocation.logicalBlockNum = 0; + li->extLocation.partitionReferenceNum = 0; + li->extLength = (li->extLength & + UDF_EXTENT_LENGTH_MASK) | + EXT_NOT_RECORDED_NOT_ALLOCATED; } } } @@ -953,6 +1035,7 @@ void udf_truncate(struct inode *inode) { int offset; int err; + struct udf_inode_info *iinfo; if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))) @@ -961,25 +1044,28 @@ void udf_truncate(struct inode *inode) return; lock_kernel(); - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) { - if (inode->i_sb->s_blocksize < (udf_file_entry_alloc_offset(inode) + - inode->i_size)) { + iinfo = UDF_I(inode); + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) { + if (inode->i_sb->s_blocksize < + (udf_file_entry_alloc_offset(inode) + + inode->i_size)) { udf_expand_file_adinicb(inode, inode->i_size, &err); - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) { - inode->i_size = UDF_I_LENALLOC(inode); + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) { + inode->i_size = iinfo->i_lenAlloc; unlock_kernel(); return; - } else { + } else udf_truncate_extents(inode); - } } else { offset = inode->i_size & (inode->i_sb->s_blocksize - 1); - memset(UDF_I_DATA(inode) + UDF_I_LENEATTR(inode) + offset, 0x00, - inode->i_sb->s_blocksize - offset - udf_file_entry_alloc_offset(inode)); - UDF_I_LENALLOC(inode) = inode->i_size; + memset(iinfo->i_ext.i_data + iinfo->i_lenEAttr + offset, + 0x00, inode->i_sb->s_blocksize - + offset - udf_file_entry_alloc_offset(inode)); + iinfo->i_lenAlloc = inode->i_size; } } else { - block_truncate_page(inode->i_mapping, inode->i_size, udf_get_block); + block_truncate_page(inode->i_mapping, inode->i_size, + udf_get_block); udf_truncate_extents(inode); } @@ -996,6 +1082,7 @@ static void __udf_read_inode(struct inode *inode) struct buffer_head *bh = NULL; struct fileEntry *fe; uint16_t ident; + struct udf_inode_info *iinfo = UDF_I(inode); /* * Set defaults, but the inode is still incomplete! @@ -1009,7 +1096,7 @@ static void __udf_read_inode(struct inode *inode) * i_nlink = 1 * i_op = NULL; */ - bh = udf_read_ptagged(inode->i_sb, UDF_I_LOCATION(inode), 0, &ident); + bh = udf_read_ptagged(inode->i_sb, iinfo->i_location, 0, &ident); if (!bh) { printk(KERN_ERR "udf: udf_read_inode(ino %ld) failed !bh\n", inode->i_ino); @@ -1019,8 +1106,8 @@ static void __udf_read_inode(struct inode *inode) if (ident != TAG_IDENT_FE && ident != TAG_IDENT_EFE && ident != TAG_IDENT_USE) { - printk(KERN_ERR "udf: udf_read_inode(ino %ld) failed ident=%d\n", - inode->i_ino, ident); + printk(KERN_ERR "udf: udf_read_inode(ino %ld) " + "failed ident=%d\n", inode->i_ino, ident); brelse(bh); make_bad_inode(inode); return; @@ -1028,11 +1115,12 @@ static void __udf_read_inode(struct inode *inode) fe = (struct fileEntry *)bh->b_data; - if (le16_to_cpu(fe->icbTag.strategyType) == 4096) { + if (fe->icbTag.strategyType == cpu_to_le16(4096)) { struct buffer_head *ibh = NULL, *nbh = NULL; struct indirectEntry *ie; - ibh = udf_read_ptagged(inode->i_sb, UDF_I_LOCATION(inode), 1, &ident); + ibh = udf_read_ptagged(inode->i_sb, iinfo->i_location, 1, + &ident); if (ident == TAG_IDENT_IE) { if (ibh) { kernel_lb_addr loc; @@ -1041,10 +1129,12 @@ static void __udf_read_inode(struct inode *inode) loc = lelb_to_cpu(ie->indirectICB.extLocation); if (ie->indirectICB.extLength && - (nbh = udf_read_ptagged(inode->i_sb, loc, 0, &ident))) { + (nbh = udf_read_ptagged(inode->i_sb, loc, 0, + &ident))) { if (ident == TAG_IDENT_FE || ident == TAG_IDENT_EFE) { - memcpy(&UDF_I_LOCATION(inode), &loc, + memcpy(&iinfo->i_location, + &loc, sizeof(kernel_lb_addr)); brelse(bh); brelse(ibh); @@ -1062,7 +1152,7 @@ static void __udf_read_inode(struct inode *inode) } else { brelse(ibh); } - } else if (le16_to_cpu(fe->icbTag.strategyType) != 4) { + } else if (fe->icbTag.strategyType != cpu_to_le16(4)) { printk(KERN_ERR "udf: unsupported strategy type: %d\n", le16_to_cpu(fe->icbTag.strategyType)); brelse(bh); @@ -1081,51 +1171,63 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh) time_t convtime; long convtime_usec; int offset; + struct udf_sb_info *sbi = UDF_SB(inode->i_sb); + struct udf_inode_info *iinfo = UDF_I(inode); fe = (struct fileEntry *)bh->b_data; efe = (struct extendedFileEntry *)bh->b_data; - if (le16_to_cpu(fe->icbTag.strategyType) == 4) - UDF_I_STRAT4096(inode) = 0; - else /* if (le16_to_cpu(fe->icbTag.strategyType) == 4096) */ - UDF_I_STRAT4096(inode) = 1; - - UDF_I_ALLOCTYPE(inode) = le16_to_cpu(fe->icbTag.flags) & ICBTAG_FLAG_AD_MASK; - UDF_I_UNIQUE(inode) = 0; - UDF_I_LENEATTR(inode) = 0; - UDF_I_LENEXTENTS(inode) = 0; - UDF_I_LENALLOC(inode) = 0; - UDF_I_NEXT_ALLOC_BLOCK(inode) = 0; - UDF_I_NEXT_ALLOC_GOAL(inode) = 0; - if (le16_to_cpu(fe->descTag.tagIdent) == TAG_IDENT_EFE) { - UDF_I_EFE(inode) = 1; - UDF_I_USE(inode) = 0; - if (udf_alloc_i_data(inode, inode->i_sb->s_blocksize - sizeof(struct extendedFileEntry))) { + if (fe->icbTag.strategyType == cpu_to_le16(4)) + iinfo->i_strat4096 = 0; + else /* if (fe->icbTag.strategyType == cpu_to_le16(4096)) */ + iinfo->i_strat4096 = 1; + + iinfo->i_alloc_type = le16_to_cpu(fe->icbTag.flags) & + ICBTAG_FLAG_AD_MASK; + iinfo->i_unique = 0; + iinfo->i_lenEAttr = 0; + iinfo->i_lenExtents = 0; + iinfo->i_lenAlloc = 0; + iinfo->i_next_alloc_block = 0; + iinfo->i_next_alloc_goal = 0; + if (fe->descTag.tagIdent == cpu_to_le16(TAG_IDENT_EFE)) { + iinfo->i_efe = 1; + iinfo->i_use = 0; + if (udf_alloc_i_data(inode, inode->i_sb->s_blocksize - + sizeof(struct extendedFileEntry))) { make_bad_inode(inode); return; } - memcpy(UDF_I_DATA(inode), bh->b_data + sizeof(struct extendedFileEntry), - inode->i_sb->s_blocksize - sizeof(struct extendedFileEntry)); - } else if (le16_to_cpu(fe->descTag.tagIdent) == TAG_IDENT_FE) { - UDF_I_EFE(inode) = 0; - UDF_I_USE(inode) = 0; - if (udf_alloc_i_data(inode, inode->i_sb->s_blocksize - sizeof(struct fileEntry))) { + memcpy(iinfo->i_ext.i_data, + bh->b_data + sizeof(struct extendedFileEntry), + inode->i_sb->s_blocksize - + sizeof(struct extendedFileEntry)); + } else if (fe->descTag.tagIdent == cpu_to_le16(TAG_IDENT_FE)) { + iinfo->i_efe = 0; + iinfo->i_use = 0; + if (udf_alloc_i_data(inode, inode->i_sb->s_blocksize - + sizeof(struct fileEntry))) { make_bad_inode(inode); return; } - memcpy(UDF_I_DATA(inode), bh->b_data + sizeof(struct fileEntry), + memcpy(iinfo->i_ext.i_data, + bh->b_data + sizeof(struct fileEntry), inode->i_sb->s_blocksize - sizeof(struct fileEntry)); - } else if (le16_to_cpu(fe->descTag.tagIdent) == TAG_IDENT_USE) { - UDF_I_EFE(inode) = 0; - UDF_I_USE(inode) = 1; - UDF_I_LENALLOC(inode) = - le32_to_cpu(((struct unallocSpaceEntry *)bh->b_data)->lengthAllocDescs); - if (udf_alloc_i_data(inode, inode->i_sb->s_blocksize - sizeof(struct unallocSpaceEntry))) { + } else if (fe->descTag.tagIdent == cpu_to_le16(TAG_IDENT_USE)) { + iinfo->i_efe = 0; + iinfo->i_use = 1; + iinfo->i_lenAlloc = le32_to_cpu( + ((struct unallocSpaceEntry *)bh->b_data)-> + lengthAllocDescs); + if (udf_alloc_i_data(inode, inode->i_sb->s_blocksize - + sizeof(struct unallocSpaceEntry))) { make_bad_inode(inode); return; } - memcpy(UDF_I_DATA(inode), bh->b_data + sizeof(struct unallocSpaceEntry), - inode->i_sb->s_blocksize - sizeof(struct unallocSpaceEntry)); + memcpy(iinfo->i_ext.i_data, + bh->b_data + sizeof(struct unallocSpaceEntry), + inode->i_sb->s_blocksize - + sizeof(struct unallocSpaceEntry)); return; } @@ -1146,12 +1248,12 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh) inode->i_nlink = 1; inode->i_size = le64_to_cpu(fe->informationLength); - UDF_I_LENEXTENTS(inode) = inode->i_size; + iinfo->i_lenExtents = inode->i_size; inode->i_mode = udf_convert_permissions(fe); inode->i_mode &= ~UDF_SB(inode->i_sb)->s_umask; - if (UDF_I_EFE(inode) == 0) { + if (iinfo->i_efe == 0) { inode->i_blocks = le64_to_cpu(fe->logicalBlocksRecorded) << (inode->i_sb->s_blocksize_bits - 9); @@ -1160,7 +1262,7 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh) inode->i_atime.tv_sec = convtime; inode->i_atime.tv_nsec = convtime_usec * 1000; } else { - inode->i_atime = UDF_SB_RECORDTIME(inode->i_sb); + inode->i_atime = sbi->s_record_time; } if (udf_stamp_to_time(&convtime, &convtime_usec, @@ -1168,7 +1270,7 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh) inode->i_mtime.tv_sec = convtime; inode->i_mtime.tv_nsec = convtime_usec * 1000; } else { - inode->i_mtime = UDF_SB_RECORDTIME(inode->i_sb); + inode->i_mtime = sbi->s_record_time; } if (udf_stamp_to_time(&convtime, &convtime_usec, @@ -1176,13 +1278,13 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh) inode->i_ctime.tv_sec = convtime; inode->i_ctime.tv_nsec = convtime_usec * 1000; } else { - inode->i_ctime = UDF_SB_RECORDTIME(inode->i_sb); + inode->i_ctime = sbi->s_record_time; } - UDF_I_UNIQUE(inode) = le64_to_cpu(fe->uniqueID); - UDF_I_LENEATTR(inode) = le32_to_cpu(fe->lengthExtendedAttr); - UDF_I_LENALLOC(inode) = le32_to_cpu(fe->lengthAllocDescs); - offset = sizeof(struct fileEntry) + UDF_I_LENEATTR(inode); + iinfo->i_unique = le64_to_cpu(fe->uniqueID); + iinfo->i_lenEAttr = le32_to_cpu(fe->lengthExtendedAttr); + iinfo->i_lenAlloc = le32_to_cpu(fe->lengthAllocDescs); + offset = sizeof(struct fileEntry) + iinfo->i_lenEAttr; } else { inode->i_blocks = le64_to_cpu(efe->logicalBlocksRecorded) << (inode->i_sb->s_blocksize_bits - 9); @@ -1192,7 +1294,7 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh) inode->i_atime.tv_sec = convtime; inode->i_atime.tv_nsec = convtime_usec * 1000; } else { - inode->i_atime = UDF_SB_RECORDTIME(inode->i_sb); + inode->i_atime = sbi->s_record_time; } if (udf_stamp_to_time(&convtime, &convtime_usec, @@ -1200,15 +1302,15 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh) inode->i_mtime.tv_sec = convtime; inode->i_mtime.tv_nsec = convtime_usec * 1000; } else { - inode->i_mtime = UDF_SB_RECORDTIME(inode->i_sb); + inode->i_mtime = sbi->s_record_time; } if (udf_stamp_to_time(&convtime, &convtime_usec, lets_to_cpu(efe->createTime))) { - UDF_I_CRTIME(inode).tv_sec = convtime; - UDF_I_CRTIME(inode).tv_nsec = convtime_usec * 1000; + iinfo->i_crtime.tv_sec = convtime; + iinfo->i_crtime.tv_nsec = convtime_usec * 1000; } else { - UDF_I_CRTIME(inode) = UDF_SB_RECORDTIME(inode->i_sb); + iinfo->i_crtime = sbi->s_record_time; } if (udf_stamp_to_time(&convtime, &convtime_usec, @@ -1216,13 +1318,14 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh) inode->i_ctime.tv_sec = convtime; inode->i_ctime.tv_nsec = convtime_usec * 1000; } else { - inode->i_ctime = UDF_SB_RECORDTIME(inode->i_sb); + inode->i_ctime = sbi->s_record_time; } - UDF_I_UNIQUE(inode) = le64_to_cpu(efe->uniqueID); - UDF_I_LENEATTR(inode) = le32_to_cpu(efe->lengthExtendedAttr); - UDF_I_LENALLOC(inode) = le32_to_cpu(efe->lengthAllocDescs); - offset = sizeof(struct extendedFileEntry) + UDF_I_LENEATTR(inode); + iinfo->i_unique = le64_to_cpu(efe->uniqueID); + iinfo->i_lenEAttr = le32_to_cpu(efe->lengthExtendedAttr); + iinfo->i_lenAlloc = le32_to_cpu(efe->lengthAllocDescs); + offset = sizeof(struct extendedFileEntry) + + iinfo->i_lenEAttr; } switch (fe->icbTag.fileType) { @@ -1235,7 +1338,7 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh) case ICBTAG_FILE_TYPE_REALTIME: case ICBTAG_FILE_TYPE_REGULAR: case ICBTAG_FILE_TYPE_UNDEF: - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) inode->i_data.a_ops = &udf_adinicb_aops; else inode->i_data.a_ops = &udf_aops; @@ -1261,31 +1364,33 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh) inode->i_mode = S_IFLNK | S_IRWXUGO; break; default: - printk(KERN_ERR "udf: udf_fill_inode(ino %ld) failed unknown file type=%d\n", - inode->i_ino, fe->icbTag.fileType); + printk(KERN_ERR "udf: udf_fill_inode(ino %ld) failed unknown " + "file type=%d\n", inode->i_ino, + fe->icbTag.fileType); make_bad_inode(inode); return; } if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) { - struct deviceSpec *dsea = (struct deviceSpec *)udf_get_extendedattr(inode, 12, 1); + struct deviceSpec *dsea = + (struct deviceSpec *)udf_get_extendedattr(inode, 12, 1); if (dsea) { init_special_inode(inode, inode->i_mode, - MKDEV(le32_to_cpu(dsea->majorDeviceIdent), - le32_to_cpu(dsea->minorDeviceIdent))); + MKDEV(le32_to_cpu(dsea->majorDeviceIdent), + le32_to_cpu(dsea->minorDeviceIdent))); /* Developer ID ??? */ - } else { + } else make_bad_inode(inode); - } } } static int udf_alloc_i_data(struct inode *inode, size_t size) { - UDF_I_DATA(inode) = kmalloc(size, GFP_KERNEL); + struct udf_inode_info *iinfo = UDF_I(inode); + iinfo->i_ext.i_data = kmalloc(size, GFP_KERNEL); - if (!UDF_I_DATA(inode)) { - printk(KERN_ERR "udf:udf_alloc_i_data (ino %ld) no free memory\n", - inode->i_ino); + if (!iinfo->i_ext.i_data) { + printk(KERN_ERR "udf:udf_alloc_i_data (ino %ld) " + "no free memory\n", inode->i_ino); return -ENOMEM; } @@ -1301,12 +1406,12 @@ static mode_t udf_convert_permissions(struct fileEntry *fe) permissions = le32_to_cpu(fe->permissions); flags = le16_to_cpu(fe->icbTag.flags); - mode = (( permissions ) & S_IRWXO) | - (( permissions >> 2 ) & S_IRWXG) | - (( permissions >> 4 ) & S_IRWXU) | - (( flags & ICBTAG_FLAG_SETUID) ? S_ISUID : 0) | - (( flags & ICBTAG_FLAG_SETGID) ? S_ISGID : 0) | - (( flags & ICBTAG_FLAG_STICKY) ? S_ISVTX : 0); + mode = ((permissions) & S_IRWXO) | + ((permissions >> 2) & S_IRWXG) | + ((permissions >> 4) & S_IRWXU) | + ((flags & ICBTAG_FLAG_SETUID) ? S_ISUID : 0) | + ((flags & ICBTAG_FLAG_SETGID) ? S_ISGID : 0) | + ((flags & ICBTAG_FLAG_STICKY) ? S_ISVTX : 0); return mode; } @@ -1350,11 +1455,15 @@ static int udf_update_inode(struct inode *inode, int do_sync) uint32_t udfperms; uint16_t icbflags; uint16_t crclen; - int i; kernel_timestamp cpu_time; int err = 0; + struct udf_sb_info *sbi = UDF_SB(inode->i_sb); + unsigned char blocksize_bits = inode->i_sb->s_blocksize_bits; + struct udf_inode_info *iinfo = UDF_I(inode); - bh = udf_tread(inode->i_sb, udf_get_lb_pblock(inode->i_sb, UDF_I_LOCATION(inode), 0)); + bh = udf_tread(inode->i_sb, + udf_get_lb_pblock(inode->i_sb, + iinfo->i_location, 0)); if (!bh) { udf_debug("bread failure\n"); return -EIO; @@ -1365,23 +1474,24 @@ static int udf_update_inode(struct inode *inode, int do_sync) fe = (struct fileEntry *)bh->b_data; efe = (struct extendedFileEntry *)bh->b_data; - if (le16_to_cpu(fe->descTag.tagIdent) == TAG_IDENT_USE) { + if (fe->descTag.tagIdent == cpu_to_le16(TAG_IDENT_USE)) { struct unallocSpaceEntry *use = (struct unallocSpaceEntry *)bh->b_data; - use->lengthAllocDescs = cpu_to_le32(UDF_I_LENALLOC(inode)); - memcpy(bh->b_data + sizeof(struct unallocSpaceEntry), UDF_I_DATA(inode), - inode->i_sb->s_blocksize - sizeof(struct unallocSpaceEntry)); - crclen = sizeof(struct unallocSpaceEntry) + UDF_I_LENALLOC(inode) - sizeof(tag); - use->descTag.tagLocation = cpu_to_le32(UDF_I_LOCATION(inode).logicalBlockNum); + use->lengthAllocDescs = cpu_to_le32(iinfo->i_lenAlloc); + memcpy(bh->b_data + sizeof(struct unallocSpaceEntry), + iinfo->i_ext.i_data, inode->i_sb->s_blocksize - + sizeof(struct unallocSpaceEntry)); + crclen = sizeof(struct unallocSpaceEntry) + + iinfo->i_lenAlloc - sizeof(tag); + use->descTag.tagLocation = cpu_to_le32( + iinfo->i_location. + logicalBlockNum); use->descTag.descCRCLength = cpu_to_le16(crclen); - use->descTag.descCRC = cpu_to_le16(udf_crc((char *)use + sizeof(tag), crclen, 0)); - - use->descTag.tagChecksum = 0; - for (i = 0; i < 16; i++) { - if (i != 4) - use->descTag.tagChecksum += ((uint8_t *)&(use->descTag))[i]; - } + use->descTag.descCRC = cpu_to_le16(udf_crc((char *)use + + sizeof(tag), crclen, + 0)); + use->descTag.tagChecksum = udf_tag_checksum(&use->descTag); mark_buffer_dirty(bh); brelse(bh); @@ -1398,14 +1508,14 @@ static int udf_update_inode(struct inode *inode, int do_sync) else fe->gid = cpu_to_le32(inode->i_gid); - udfperms = ((inode->i_mode & S_IRWXO) ) | - ((inode->i_mode & S_IRWXG) << 2) | - ((inode->i_mode & S_IRWXU) << 4); + udfperms = ((inode->i_mode & S_IRWXO)) | + ((inode->i_mode & S_IRWXG) << 2) | + ((inode->i_mode & S_IRWXU) << 4); - udfperms |= (le32_to_cpu(fe->permissions) & - (FE_PERM_O_DELETE | FE_PERM_O_CHATTR | - FE_PERM_G_DELETE | FE_PERM_G_CHATTR | - FE_PERM_U_DELETE | FE_PERM_U_CHATTR)); + udfperms |= (le32_to_cpu(fe->permissions) & + (FE_PERM_O_DELETE | FE_PERM_O_CHATTR | + FE_PERM_G_DELETE | FE_PERM_G_CHATTR | + FE_PERM_U_DELETE | FE_PERM_U_CHATTR)); fe->permissions = cpu_to_le32(udfperms); if (S_ISDIR(inode->i_mode)) @@ -1426,8 +1536,9 @@ static int udf_update_inode(struct inode *inode, int do_sync) sizeof(regid), 12, 0x3); dsea->attrType = cpu_to_le32(12); dsea->attrSubtype = 1; - dsea->attrLength = cpu_to_le32(sizeof(struct deviceSpec) + - sizeof(regid)); + dsea->attrLength = cpu_to_le32( + sizeof(struct deviceSpec) + + sizeof(regid)); dsea->impUseLength = cpu_to_le32(sizeof(regid)); } eid = (regid *)dsea->impUse; @@ -1439,12 +1550,13 @@ static int udf_update_inode(struct inode *inode, int do_sync) dsea->minorDeviceIdent = cpu_to_le32(iminor(inode)); } - if (UDF_I_EFE(inode) == 0) { - memcpy(bh->b_data + sizeof(struct fileEntry), UDF_I_DATA(inode), + if (iinfo->i_efe == 0) { + memcpy(bh->b_data + sizeof(struct fileEntry), + iinfo->i_ext.i_data, inode->i_sb->s_blocksize - sizeof(struct fileEntry)); fe->logicalBlocksRecorded = cpu_to_le64( - (inode->i_blocks + (1 << (inode->i_sb->s_blocksize_bits - 9)) - 1) >> - (inode->i_sb->s_blocksize_bits - 9)); + (inode->i_blocks + (1 << (blocksize_bits - 9)) - 1) >> + (blocksize_bits - 9)); if (udf_time_to_stamp(&cpu_time, inode->i_atime)) fe->accessTime = cpu_to_lets(cpu_time); @@ -1456,40 +1568,41 @@ static int udf_update_inode(struct inode *inode, int do_sync) strcpy(fe->impIdent.ident, UDF_ID_DEVELOPER); fe->impIdent.identSuffix[0] = UDF_OS_CLASS_UNIX; fe->impIdent.identSuffix[1] = UDF_OS_ID_LINUX; - fe->uniqueID = cpu_to_le64(UDF_I_UNIQUE(inode)); - fe->lengthExtendedAttr = cpu_to_le32(UDF_I_LENEATTR(inode)); - fe->lengthAllocDescs = cpu_to_le32(UDF_I_LENALLOC(inode)); + fe->uniqueID = cpu_to_le64(iinfo->i_unique); + fe->lengthExtendedAttr = cpu_to_le32(iinfo->i_lenEAttr); + fe->lengthAllocDescs = cpu_to_le32(iinfo->i_lenAlloc); fe->descTag.tagIdent = cpu_to_le16(TAG_IDENT_FE); crclen = sizeof(struct fileEntry); } else { - memcpy(bh->b_data + sizeof(struct extendedFileEntry), UDF_I_DATA(inode), - inode->i_sb->s_blocksize - sizeof(struct extendedFileEntry)); + memcpy(bh->b_data + sizeof(struct extendedFileEntry), + iinfo->i_ext.i_data, + inode->i_sb->s_blocksize - + sizeof(struct extendedFileEntry)); efe->objectSize = cpu_to_le64(inode->i_size); efe->logicalBlocksRecorded = cpu_to_le64( - (inode->i_blocks + (1 << (inode->i_sb->s_blocksize_bits - 9)) - 1) >> - (inode->i_sb->s_blocksize_bits - 9)); + (inode->i_blocks + (1 << (blocksize_bits - 9)) - 1) >> + (blocksize_bits - 9)); - if (UDF_I_CRTIME(inode).tv_sec > inode->i_atime.tv_sec || - (UDF_I_CRTIME(inode).tv_sec == inode->i_atime.tv_sec && - UDF_I_CRTIME(inode).tv_nsec > inode->i_atime.tv_nsec)) { - UDF_I_CRTIME(inode) = inode->i_atime; - } - if (UDF_I_CRTIME(inode).tv_sec > inode->i_mtime.tv_sec || - (UDF_I_CRTIME(inode).tv_sec == inode->i_mtime.tv_sec && - UDF_I_CRTIME(inode).tv_nsec > inode->i_mtime.tv_nsec)) { - UDF_I_CRTIME(inode) = inode->i_mtime; - } - if (UDF_I_CRTIME(inode).tv_sec > inode->i_ctime.tv_sec || - (UDF_I_CRTIME(inode).tv_sec == inode->i_ctime.tv_sec && - UDF_I_CRTIME(inode).tv_nsec > inode->i_ctime.tv_nsec)) { - UDF_I_CRTIME(inode) = inode->i_ctime; - } + if (iinfo->i_crtime.tv_sec > inode->i_atime.tv_sec || + (iinfo->i_crtime.tv_sec == inode->i_atime.tv_sec && + iinfo->i_crtime.tv_nsec > inode->i_atime.tv_nsec)) + iinfo->i_crtime = inode->i_atime; + + if (iinfo->i_crtime.tv_sec > inode->i_mtime.tv_sec || + (iinfo->i_crtime.tv_sec == inode->i_mtime.tv_sec && + iinfo->i_crtime.tv_nsec > inode->i_mtime.tv_nsec)) + iinfo->i_crtime = inode->i_mtime; + + if (iinfo->i_crtime.tv_sec > inode->i_ctime.tv_sec || + (iinfo->i_crtime.tv_sec == inode->i_ctime.tv_sec && + iinfo->i_crtime.tv_nsec > inode->i_ctime.tv_nsec)) + iinfo->i_crtime = inode->i_ctime; if (udf_time_to_stamp(&cpu_time, inode->i_atime)) efe->accessTime = cpu_to_lets(cpu_time); if (udf_time_to_stamp(&cpu_time, inode->i_mtime)) efe->modificationTime = cpu_to_lets(cpu_time); - if (udf_time_to_stamp(&cpu_time, UDF_I_CRTIME(inode))) + if (udf_time_to_stamp(&cpu_time, iinfo->i_crtime)) efe->createTime = cpu_to_lets(cpu_time); if (udf_time_to_stamp(&cpu_time, inode->i_ctime)) efe->attrTime = cpu_to_lets(cpu_time); @@ -1498,13 +1611,13 @@ static int udf_update_inode(struct inode *inode, int do_sync) strcpy(efe->impIdent.ident, UDF_ID_DEVELOPER); efe->impIdent.identSuffix[0] = UDF_OS_CLASS_UNIX; efe->impIdent.identSuffix[1] = UDF_OS_ID_LINUX; - efe->uniqueID = cpu_to_le64(UDF_I_UNIQUE(inode)); - efe->lengthExtendedAttr = cpu_to_le32(UDF_I_LENEATTR(inode)); - efe->lengthAllocDescs = cpu_to_le32(UDF_I_LENALLOC(inode)); + efe->uniqueID = cpu_to_le64(iinfo->i_unique); + efe->lengthExtendedAttr = cpu_to_le32(iinfo->i_lenEAttr); + efe->lengthAllocDescs = cpu_to_le32(iinfo->i_lenAlloc); efe->descTag.tagIdent = cpu_to_le16(TAG_IDENT_EFE); crclen = sizeof(struct extendedFileEntry); } - if (UDF_I_STRAT4096(inode)) { + if (iinfo->i_strat4096) { fe->icbTag.strategyType = cpu_to_le16(4096); fe->icbTag.strategyParameter = cpu_to_le16(1); fe->icbTag.numEntries = cpu_to_le16(2); @@ -1528,7 +1641,7 @@ static int udf_update_inode(struct inode *inode, int do_sync) else if (S_ISSOCK(inode->i_mode)) fe->icbTag.fileType = ICBTAG_FILE_TYPE_SOCKET; - icbflags = UDF_I_ALLOCTYPE(inode) | + icbflags = iinfo->i_alloc_type | ((inode->i_mode & S_ISUID) ? ICBTAG_FLAG_SETUID : 0) | ((inode->i_mode & S_ISGID) ? ICBTAG_FLAG_SETGID : 0) | ((inode->i_mode & S_ISVTX) ? ICBTAG_FLAG_STICKY : 0) | @@ -1537,29 +1650,28 @@ static int udf_update_inode(struct inode *inode, int do_sync) ICBTAG_FLAG_SETGID | ICBTAG_FLAG_STICKY)); fe->icbTag.flags = cpu_to_le16(icbflags); - if (UDF_SB_UDFREV(inode->i_sb) >= 0x0200) + if (sbi->s_udfrev >= 0x0200) fe->descTag.descVersion = cpu_to_le16(3); else fe->descTag.descVersion = cpu_to_le16(2); - fe->descTag.tagSerialNum = cpu_to_le16(UDF_SB_SERIALNUM(inode->i_sb)); - fe->descTag.tagLocation = cpu_to_le32(UDF_I_LOCATION(inode).logicalBlockNum); - crclen += UDF_I_LENEATTR(inode) + UDF_I_LENALLOC(inode) - sizeof(tag); + fe->descTag.tagSerialNum = cpu_to_le16(sbi->s_serial_number); + fe->descTag.tagLocation = cpu_to_le32( + iinfo->i_location.logicalBlockNum); + crclen += iinfo->i_lenEAttr + iinfo->i_lenAlloc - + sizeof(tag); fe->descTag.descCRCLength = cpu_to_le16(crclen); - fe->descTag.descCRC = cpu_to_le16(udf_crc((char *)fe + sizeof(tag), crclen, 0)); - - fe->descTag.tagChecksum = 0; - for (i = 0; i < 16; i++) { - if (i != 4) - fe->descTag.tagChecksum += ((uint8_t *)&(fe->descTag))[i]; - } + fe->descTag.descCRC = cpu_to_le16(udf_crc((char *)fe + sizeof(tag), + crclen, 0)); + fe->descTag.tagChecksum = udf_tag_checksum(&fe->descTag); /* write the data blocks */ mark_buffer_dirty(bh); if (do_sync) { sync_dirty_buffer(bh); if (buffer_req(bh) && !buffer_uptodate(bh)) { - printk("IO error syncing udf inode [%s:%08lx]\n", - inode->i_sb->s_id, inode->i_ino); + printk(KERN_WARNING "IO error syncing udf inode " + "[%s:%08lx]\n", inode->i_sb->s_id, + inode->i_ino); err = -EIO; } } @@ -1577,7 +1689,7 @@ struct inode *udf_iget(struct super_block *sb, kernel_lb_addr ino) return NULL; if (inode->i_state & I_NEW) { - memcpy(&UDF_I_LOCATION(inode), &ino, sizeof(kernel_lb_addr)); + memcpy(&UDF_I(inode)->i_location, &ino, sizeof(kernel_lb_addr)); __udf_read_inode(inode); unlock_new_inode(inode); } @@ -1585,7 +1697,8 @@ struct inode *udf_iget(struct super_block *sb, kernel_lb_addr ino) if (is_bad_inode(inode)) goto out_iput; - if (ino.logicalBlockNum >= UDF_SB_PARTLEN(sb, ino.partitionReferenceNum)) { + if (ino.logicalBlockNum >= UDF_SB(sb)-> + s_partmaps[ino.partitionReferenceNum].s_partition_len) { udf_debug("block=%d, partition=%d out of range\n", ino.logicalBlockNum, ino.partitionReferenceNum); make_bad_inode(inode); @@ -1599,7 +1712,7 @@ struct inode *udf_iget(struct super_block *sb, kernel_lb_addr ino) return NULL; } -int8_t udf_add_aext(struct inode * inode, struct extent_position * epos, +int8_t udf_add_aext(struct inode *inode, struct extent_position *epos, kernel_lb_addr eloc, uint32_t elen, int inc) { int adsize; @@ -1608,15 +1721,18 @@ int8_t udf_add_aext(struct inode * inode, struct extent_position * epos, struct allocExtDesc *aed; int8_t etype; uint8_t *ptr; + struct udf_inode_info *iinfo = UDF_I(inode); if (!epos->bh) - ptr = UDF_I_DATA(inode) + epos->offset - udf_file_entry_alloc_offset(inode) + UDF_I_LENEATTR(inode); + ptr = iinfo->i_ext.i_data + epos->offset - + udf_file_entry_alloc_offset(inode) + + iinfo->i_lenEAttr; else ptr = epos->bh->b_data + epos->offset; - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_SHORT) + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT) adsize = sizeof(short_ad); - else if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_LONG) + else if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_LONG) adsize = sizeof(long_ad); else return -1; @@ -1627,15 +1743,16 @@ int8_t udf_add_aext(struct inode * inode, struct extent_position * epos, int err, loffset; kernel_lb_addr obloc = epos->block; - if (!(epos->block.logicalBlockNum = udf_new_block(inode->i_sb, NULL, - obloc.partitionReferenceNum, - obloc.logicalBlockNum, &err))) { + epos->block.logicalBlockNum = udf_new_block(inode->i_sb, NULL, + obloc.partitionReferenceNum, + obloc.logicalBlockNum, &err); + if (!epos->block.logicalBlockNum) return -1; - } - if (!(nbh = udf_tgetblk(inode->i_sb, udf_get_lb_pblock(inode->i_sb, - epos->block, 0)))) { + nbh = udf_tgetblk(inode->i_sb, udf_get_lb_pblock(inode->i_sb, + epos->block, + 0)); + if (!nbh) return -1; - } lock_buffer(nbh); memset(nbh->b_data, 0x00, inode->i_sb->s_blocksize); set_buffer_uptodate(nbh); @@ -1644,7 +1761,8 @@ int8_t udf_add_aext(struct inode * inode, struct extent_position * epos, aed = (struct allocExtDesc *)(nbh->b_data); if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT)) - aed->previousAllocExtLocation = cpu_to_le32(obloc.logicalBlockNum); + aed->previousAllocExtLocation = + cpu_to_le32(obloc.logicalBlockNum); if (epos->offset + adsize > inode->i_sb->s_blocksize) { loffset = epos->offset; aed->lengthAllocDescs = cpu_to_le32(adsize); @@ -1661,24 +1779,26 @@ int8_t udf_add_aext(struct inode * inode, struct extent_position * epos, if (epos->bh) { aed = (struct allocExtDesc *)epos->bh->b_data; aed->lengthAllocDescs = - cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) + adsize); + cpu_to_le32(le32_to_cpu( + aed->lengthAllocDescs) + adsize); } else { - UDF_I_LENALLOC(inode) += adsize; + iinfo->i_lenAlloc += adsize; mark_inode_dirty(inode); } } - if (UDF_SB_UDFREV(inode->i_sb) >= 0x0200) + if (UDF_SB(inode->i_sb)->s_udfrev >= 0x0200) udf_new_tag(nbh->b_data, TAG_IDENT_AED, 3, 1, epos->block.logicalBlockNum, sizeof(tag)); else udf_new_tag(nbh->b_data, TAG_IDENT_AED, 2, 1, epos->block.logicalBlockNum, sizeof(tag)); - switch (UDF_I_ALLOCTYPE(inode)) { + switch (iinfo->i_alloc_type) { case ICBTAG_FLAG_AD_SHORT: sad = (short_ad *)sptr; sad->extLength = cpu_to_le32(EXT_NEXT_EXTENT_ALLOCDECS | inode->i_sb->s_blocksize); - sad->extPosition = cpu_to_le32(epos->block.logicalBlockNum); + sad->extPosition = + cpu_to_le32(epos->block.logicalBlockNum); break; case ICBTAG_FLAG_AD_LONG: lad = (long_ad *)sptr; @@ -1690,10 +1810,11 @@ int8_t udf_add_aext(struct inode * inode, struct extent_position * epos, } if (epos->bh) { if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) || - UDF_SB_UDFREV(inode->i_sb) >= 0x0201) + UDF_SB(inode->i_sb)->s_udfrev >= 0x0201) udf_update_tag(epos->bh->b_data, loffset); else - udf_update_tag(epos->bh->b_data, sizeof(struct allocExtDesc)); + udf_update_tag(epos->bh->b_data, + sizeof(struct allocExtDesc)); mark_buffer_dirty_inode(epos->bh, inode); brelse(epos->bh); } else { @@ -1705,36 +1826,43 @@ int8_t udf_add_aext(struct inode * inode, struct extent_position * epos, etype = udf_write_aext(inode, epos, eloc, elen, inc); if (!epos->bh) { - UDF_I_LENALLOC(inode) += adsize; + iinfo->i_lenAlloc += adsize; mark_inode_dirty(inode); } else { aed = (struct allocExtDesc *)epos->bh->b_data; aed->lengthAllocDescs = - cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) + adsize); - if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) || UDF_SB_UDFREV(inode->i_sb) >= 0x0201) - udf_update_tag(epos->bh->b_data, epos->offset + (inc ? 0 : adsize)); + cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) + + adsize); + if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) || + UDF_SB(inode->i_sb)->s_udfrev >= 0x0201) + udf_update_tag(epos->bh->b_data, + epos->offset + (inc ? 0 : adsize)); else - udf_update_tag(epos->bh->b_data, sizeof(struct allocExtDesc)); + udf_update_tag(epos->bh->b_data, + sizeof(struct allocExtDesc)); mark_buffer_dirty_inode(epos->bh, inode); } return etype; } -int8_t udf_write_aext(struct inode * inode, struct extent_position * epos, +int8_t udf_write_aext(struct inode *inode, struct extent_position *epos, kernel_lb_addr eloc, uint32_t elen, int inc) { int adsize; uint8_t *ptr; short_ad *sad; long_ad *lad; + struct udf_inode_info *iinfo = UDF_I(inode); if (!epos->bh) - ptr = UDF_I_DATA(inode) + epos->offset - udf_file_entry_alloc_offset(inode) + UDF_I_LENEATTR(inode); + ptr = iinfo->i_ext.i_data + epos->offset - + udf_file_entry_alloc_offset(inode) + + iinfo->i_lenEAttr; else ptr = epos->bh->b_data + epos->offset; - switch (UDF_I_ALLOCTYPE(inode)) { + switch (iinfo->i_alloc_type) { case ICBTAG_FLAG_AD_SHORT: sad = (short_ad *)ptr; sad->extLength = cpu_to_le32(elen); @@ -1754,10 +1882,12 @@ int8_t udf_write_aext(struct inode * inode, struct extent_position * epos, if (epos->bh) { if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) || - UDF_SB_UDFREV(inode->i_sb) >= 0x0201) { - struct allocExtDesc *aed = (struct allocExtDesc *)epos->bh->b_data; + UDF_SB(inode->i_sb)->s_udfrev >= 0x0201) { + struct allocExtDesc *aed = + (struct allocExtDesc *)epos->bh->b_data; udf_update_tag(epos->bh->b_data, - le32_to_cpu(aed->lengthAllocDescs) + sizeof(struct allocExtDesc)); + le32_to_cpu(aed->lengthAllocDescs) + + sizeof(struct allocExtDesc)); } mark_buffer_dirty_inode(epos->bh, inode); } else { @@ -1770,19 +1900,21 @@ int8_t udf_write_aext(struct inode * inode, struct extent_position * epos, return (elen >> 30); } -int8_t udf_next_aext(struct inode * inode, struct extent_position * epos, - kernel_lb_addr * eloc, uint32_t * elen, int inc) +int8_t udf_next_aext(struct inode *inode, struct extent_position *epos, + kernel_lb_addr *eloc, uint32_t *elen, int inc) { int8_t etype; while ((etype = udf_current_aext(inode, epos, eloc, elen, inc)) == (EXT_NEXT_EXTENT_ALLOCDECS >> 30)) { + int block; epos->block = *eloc; epos->offset = sizeof(struct allocExtDesc); brelse(epos->bh); - if (!(epos->bh = udf_tread(inode->i_sb, udf_get_lb_pblock(inode->i_sb, epos->block, 0)))) { - udf_debug("reading block %d failed!\n", - udf_get_lb_pblock(inode->i_sb, epos->block, 0)); + block = udf_get_lb_pblock(inode->i_sb, epos->block, 0); + epos->bh = udf_tread(inode->i_sb, block); + if (!epos->bh) { + udf_debug("reading block %d failed!\n", block); return -1; } } @@ -1790,47 +1922,55 @@ int8_t udf_next_aext(struct inode * inode, struct extent_position * epos, return etype; } -int8_t udf_current_aext(struct inode * inode, struct extent_position * epos, - kernel_lb_addr * eloc, uint32_t * elen, int inc) +int8_t udf_current_aext(struct inode *inode, struct extent_position *epos, + kernel_lb_addr *eloc, uint32_t *elen, int inc) { int alen; int8_t etype; uint8_t *ptr; short_ad *sad; long_ad *lad; - + struct udf_inode_info *iinfo = UDF_I(inode); if (!epos->bh) { if (!epos->offset) epos->offset = udf_file_entry_alloc_offset(inode); - ptr = UDF_I_DATA(inode) + epos->offset - udf_file_entry_alloc_offset(inode) + UDF_I_LENEATTR(inode); - alen = udf_file_entry_alloc_offset(inode) + UDF_I_LENALLOC(inode); + ptr = iinfo->i_ext.i_data + epos->offset - + udf_file_entry_alloc_offset(inode) + + iinfo->i_lenEAttr; + alen = udf_file_entry_alloc_offset(inode) + + iinfo->i_lenAlloc; } else { if (!epos->offset) epos->offset = sizeof(struct allocExtDesc); ptr = epos->bh->b_data + epos->offset; alen = sizeof(struct allocExtDesc) + - le32_to_cpu(((struct allocExtDesc *)epos->bh->b_data)->lengthAllocDescs); + le32_to_cpu(((struct allocExtDesc *)epos->bh->b_data)-> + lengthAllocDescs); } - switch (UDF_I_ALLOCTYPE(inode)) { + switch (iinfo->i_alloc_type) { case ICBTAG_FLAG_AD_SHORT: - if (!(sad = udf_get_fileshortad(ptr, alen, &epos->offset, inc))) + sad = udf_get_fileshortad(ptr, alen, &epos->offset, inc); + if (!sad) return -1; etype = le32_to_cpu(sad->extLength) >> 30; eloc->logicalBlockNum = le32_to_cpu(sad->extPosition); - eloc->partitionReferenceNum = UDF_I_LOCATION(inode).partitionReferenceNum; + eloc->partitionReferenceNum = + iinfo->i_location.partitionReferenceNum; *elen = le32_to_cpu(sad->extLength) & UDF_EXTENT_LENGTH_MASK; break; case ICBTAG_FLAG_AD_LONG: - if (!(lad = udf_get_filelongad(ptr, alen, &epos->offset, inc))) + lad = udf_get_filelongad(ptr, alen, &epos->offset, inc); + if (!lad) return -1; etype = le32_to_cpu(lad->extLength) >> 30; *eloc = lelb_to_cpu(lad->extLocation); *elen = le32_to_cpu(lad->extLength) & UDF_EXTENT_LENGTH_MASK; break; default: - udf_debug("alloc_type = %d unsupported\n", UDF_I_ALLOCTYPE(inode)); + udf_debug("alloc_type = %d unsupported\n", + iinfo->i_alloc_type); return -1; } @@ -1858,22 +1998,24 @@ static int8_t udf_insert_aext(struct inode *inode, struct extent_position epos, return (nelen >> 30); } -int8_t udf_delete_aext(struct inode * inode, struct extent_position epos, +int8_t udf_delete_aext(struct inode *inode, struct extent_position epos, kernel_lb_addr eloc, uint32_t elen) { struct extent_position oepos; int adsize; int8_t etype; struct allocExtDesc *aed; + struct udf_inode_info *iinfo; if (epos.bh) { get_bh(epos.bh); get_bh(epos.bh); } - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_SHORT) + iinfo = UDF_I(inode); + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT) adsize = sizeof(short_ad); - else if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_LONG) + else if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_LONG) adsize = sizeof(long_ad); else adsize = 0; @@ -1900,33 +2042,39 @@ int8_t udf_delete_aext(struct inode * inode, struct extent_position epos, udf_write_aext(inode, &oepos, eloc, elen, 1); udf_write_aext(inode, &oepos, eloc, elen, 1); if (!oepos.bh) { - UDF_I_LENALLOC(inode) -= (adsize * 2); + iinfo->i_lenAlloc -= (adsize * 2); mark_inode_dirty(inode); } else { aed = (struct allocExtDesc *)oepos.bh->b_data; aed->lengthAllocDescs = - cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) - (2 * adsize)); + cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) - + (2 * adsize)); if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) || - UDF_SB_UDFREV(inode->i_sb) >= 0x0201) - udf_update_tag(oepos.bh->b_data, oepos.offset - (2 * adsize)); + UDF_SB(inode->i_sb)->s_udfrev >= 0x0201) + udf_update_tag(oepos.bh->b_data, + oepos.offset - (2 * adsize)); else - udf_update_tag(oepos.bh->b_data, sizeof(struct allocExtDesc)); + udf_update_tag(oepos.bh->b_data, + sizeof(struct allocExtDesc)); mark_buffer_dirty_inode(oepos.bh, inode); } } else { udf_write_aext(inode, &oepos, eloc, elen, 1); if (!oepos.bh) { - UDF_I_LENALLOC(inode) -= adsize; + iinfo->i_lenAlloc -= adsize; mark_inode_dirty(inode); } else { aed = (struct allocExtDesc *)oepos.bh->b_data; aed->lengthAllocDescs = - cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) - adsize); + cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) - + adsize); if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) || - UDF_SB_UDFREV(inode->i_sb) >= 0x0201) - udf_update_tag(oepos.bh->b_data, epos.offset - adsize); + UDF_SB(inode->i_sb)->s_udfrev >= 0x0201) + udf_update_tag(oepos.bh->b_data, + epos.offset - adsize); else - udf_update_tag(oepos.bh->b_data, sizeof(struct allocExtDesc)); + udf_update_tag(oepos.bh->b_data, + sizeof(struct allocExtDesc)); mark_buffer_dirty_inode(oepos.bh, inode); } } @@ -1937,34 +2085,38 @@ int8_t udf_delete_aext(struct inode * inode, struct extent_position epos, return (elen >> 30); } -int8_t inode_bmap(struct inode * inode, sector_t block, - struct extent_position * pos, kernel_lb_addr * eloc, - uint32_t * elen, sector_t * offset) +int8_t inode_bmap(struct inode *inode, sector_t block, + struct extent_position *pos, kernel_lb_addr *eloc, + uint32_t *elen, sector_t *offset) { + unsigned char blocksize_bits = inode->i_sb->s_blocksize_bits; loff_t lbcount = 0, bcount = - (loff_t) block << inode->i_sb->s_blocksize_bits; + (loff_t) block << blocksize_bits; int8_t etype; + struct udf_inode_info *iinfo; if (block < 0) { printk(KERN_ERR "udf: inode_bmap: block < 0\n"); return -1; } + iinfo = UDF_I(inode); pos->offset = 0; - pos->block = UDF_I_LOCATION(inode); + pos->block = iinfo->i_location; pos->bh = NULL; *elen = 0; do { - if ((etype = udf_next_aext(inode, pos, eloc, elen, 1)) == -1) { - *offset = (bcount - lbcount) >> inode->i_sb->s_blocksize_bits; - UDF_I_LENEXTENTS(inode) = lbcount; + etype = udf_next_aext(inode, pos, eloc, elen, 1); + if (etype == -1) { + *offset = (bcount - lbcount) >> blocksize_bits; + iinfo->i_lenExtents = lbcount; return -1; } lbcount += *elen; } while (lbcount <= bcount); - *offset = (bcount + *elen - lbcount) >> inode->i_sb->s_blocksize_bits; + *offset = (bcount + *elen - lbcount) >> blocksize_bits; return etype; } @@ -1979,7 +2131,8 @@ long udf_block_map(struct inode *inode, sector_t block) lock_kernel(); - if (inode_bmap(inode, block, &epos, &eloc, &elen, &offset) == (EXT_RECORDED_ALLOCATED >> 30)) + if (inode_bmap(inode, block, &epos, &eloc, &elen, &offset) == + (EXT_RECORDED_ALLOCATED >> 30)) ret = udf_get_lb_pblock(inode->i_sb, eloc, offset); else ret = 0; diff --git a/fs/udf/misc.c b/fs/udf/misc.c index 15297deb505..a1d6da0caf7 100644 --- a/fs/udf/misc.c +++ b/fs/udf/misc.c @@ -51,18 +51,18 @@ struct genericFormat *udf_add_extendedattr(struct inode *inode, uint32_t size, uint8_t *ea = NULL, *ad = NULL; int offset; uint16_t crclen; - int i; + struct udf_inode_info *iinfo = UDF_I(inode); - ea = UDF_I_DATA(inode); - if (UDF_I_LENEATTR(inode)) { - ad = UDF_I_DATA(inode) + UDF_I_LENEATTR(inode); + ea = iinfo->i_ext.i_data; + if (iinfo->i_lenEAttr) { + ad = iinfo->i_ext.i_data + iinfo->i_lenEAttr; } else { ad = ea; size += sizeof(struct extendedAttrHeaderDesc); } offset = inode->i_sb->s_blocksize - udf_file_entry_alloc_offset(inode) - - UDF_I_LENALLOC(inode); + iinfo->i_lenAlloc; /* TODO - Check for FreeEASpace */ @@ -70,69 +70,80 @@ struct genericFormat *udf_add_extendedattr(struct inode *inode, uint32_t size, struct extendedAttrHeaderDesc *eahd; eahd = (struct extendedAttrHeaderDesc *)ea; - if (UDF_I_LENALLOC(inode)) { - memmove(&ad[size], ad, UDF_I_LENALLOC(inode)); - } + if (iinfo->i_lenAlloc) + memmove(&ad[size], ad, iinfo->i_lenAlloc); - if (UDF_I_LENEATTR(inode)) { + if (iinfo->i_lenEAttr) { /* check checksum/crc */ - if (le16_to_cpu(eahd->descTag.tagIdent) != TAG_IDENT_EAHD || - le32_to_cpu(eahd->descTag.tagLocation) != UDF_I_LOCATION(inode).logicalBlockNum) { + if (eahd->descTag.tagIdent != + cpu_to_le16(TAG_IDENT_EAHD) || + le32_to_cpu(eahd->descTag.tagLocation) != + iinfo->i_location.logicalBlockNum) return NULL; - } } else { + struct udf_sb_info *sbi = UDF_SB(inode->i_sb); + size -= sizeof(struct extendedAttrHeaderDesc); - UDF_I_LENEATTR(inode) += sizeof(struct extendedAttrHeaderDesc); + iinfo->i_lenEAttr += + sizeof(struct extendedAttrHeaderDesc); eahd->descTag.tagIdent = cpu_to_le16(TAG_IDENT_EAHD); - if (UDF_SB_UDFREV(inode->i_sb) >= 0x0200) + if (sbi->s_udfrev >= 0x0200) eahd->descTag.descVersion = cpu_to_le16(3); else eahd->descTag.descVersion = cpu_to_le16(2); - eahd->descTag.tagSerialNum = cpu_to_le16(UDF_SB_SERIALNUM(inode->i_sb)); - eahd->descTag.tagLocation = cpu_to_le32(UDF_I_LOCATION(inode).logicalBlockNum); + eahd->descTag.tagSerialNum = + cpu_to_le16(sbi->s_serial_number); + eahd->descTag.tagLocation = cpu_to_le32( + iinfo->i_location.logicalBlockNum); eahd->impAttrLocation = cpu_to_le32(0xFFFFFFFF); eahd->appAttrLocation = cpu_to_le32(0xFFFFFFFF); } - offset = UDF_I_LENEATTR(inode); + offset = iinfo->i_lenEAttr; if (type < 2048) { - if (le32_to_cpu(eahd->appAttrLocation) < UDF_I_LENEATTR(inode)) { - uint32_t aal = le32_to_cpu(eahd->appAttrLocation); + if (le32_to_cpu(eahd->appAttrLocation) < + iinfo->i_lenEAttr) { + uint32_t aal = + le32_to_cpu(eahd->appAttrLocation); memmove(&ea[offset - aal + size], &ea[aal], offset - aal); offset -= aal; - eahd->appAttrLocation = cpu_to_le32(aal + size); + eahd->appAttrLocation = + cpu_to_le32(aal + size); } - if (le32_to_cpu(eahd->impAttrLocation) < UDF_I_LENEATTR(inode)) { - uint32_t ial = le32_to_cpu(eahd->impAttrLocation); + if (le32_to_cpu(eahd->impAttrLocation) < + iinfo->i_lenEAttr) { + uint32_t ial = + le32_to_cpu(eahd->impAttrLocation); memmove(&ea[offset - ial + size], &ea[ial], offset - ial); offset -= ial; - eahd->impAttrLocation = cpu_to_le32(ial + size); + eahd->impAttrLocation = + cpu_to_le32(ial + size); } } else if (type < 65536) { - if (le32_to_cpu(eahd->appAttrLocation) < UDF_I_LENEATTR(inode)) { - uint32_t aal = le32_to_cpu(eahd->appAttrLocation); + if (le32_to_cpu(eahd->appAttrLocation) < + iinfo->i_lenEAttr) { + uint32_t aal = + le32_to_cpu(eahd->appAttrLocation); memmove(&ea[offset - aal + size], &ea[aal], offset - aal); offset -= aal; - eahd->appAttrLocation = cpu_to_le32(aal + size); + eahd->appAttrLocation = + cpu_to_le32(aal + size); } } /* rewrite CRC + checksum of eahd */ crclen = sizeof(struct extendedAttrHeaderDesc) - sizeof(tag); eahd->descTag.descCRCLength = cpu_to_le16(crclen); eahd->descTag.descCRC = cpu_to_le16(udf_crc((char *)eahd + - sizeof(tag), crclen, 0)); - eahd->descTag.tagChecksum = 0; - for (i = 0; i < 16; i++) - if (i != 4) - eahd->descTag.tagChecksum += ((uint8_t *)&(eahd->descTag))[i]; - UDF_I_LENEATTR(inode) += size; + sizeof(tag), crclen, 0)); + eahd->descTag.tagChecksum = udf_tag_checksum(&eahd->descTag); + iinfo->i_lenEAttr += size; return (struct genericFormat *)&ea[offset]; } - if (loc & 0x02) { - } + if (loc & 0x02) + ; return NULL; } @@ -143,18 +154,20 @@ struct genericFormat *udf_get_extendedattr(struct inode *inode, uint32_t type, struct genericFormat *gaf; uint8_t *ea = NULL; uint32_t offset; + struct udf_inode_info *iinfo = UDF_I(inode); - ea = UDF_I_DATA(inode); + ea = iinfo->i_ext.i_data; - if (UDF_I_LENEATTR(inode)) { + if (iinfo->i_lenEAttr) { struct extendedAttrHeaderDesc *eahd; eahd = (struct extendedAttrHeaderDesc *)ea; /* check checksum/crc */ - if (le16_to_cpu(eahd->descTag.tagIdent) != TAG_IDENT_EAHD || - le32_to_cpu(eahd->descTag.tagLocation) != UDF_I_LOCATION(inode).logicalBlockNum) { + if (eahd->descTag.tagIdent != + cpu_to_le16(TAG_IDENT_EAHD) || + le32_to_cpu(eahd->descTag.tagLocation) != + iinfo->i_location.logicalBlockNum) return NULL; - } if (type < 2048) offset = sizeof(struct extendedAttrHeaderDesc); @@ -163,9 +176,10 @@ struct genericFormat *udf_get_extendedattr(struct inode *inode, uint32_t type, else offset = le32_to_cpu(eahd->appAttrLocation); - while (offset < UDF_I_LENEATTR(inode)) { + while (offset < iinfo->i_lenEAttr) { gaf = (struct genericFormat *)&ea[offset]; - if (le32_to_cpu(gaf->attrType) == type && gaf->attrSubtype == subtype) + if (le32_to_cpu(gaf->attrType) == type && + gaf->attrSubtype == subtype) return gaf; else offset += le32_to_cpu(gaf->attrLength); @@ -186,21 +200,20 @@ struct genericFormat *udf_get_extendedattr(struct inode *inode, uint32_t type, * Written, tested, and released. */ struct buffer_head *udf_read_tagged(struct super_block *sb, uint32_t block, - uint32_t location, uint16_t * ident) + uint32_t location, uint16_t *ident) { tag *tag_p; struct buffer_head *bh = NULL; - register uint8_t checksum; - register int i; + struct udf_sb_info *sbi = UDF_SB(sb); /* Read the block */ if (block == 0xFFFFFFFF) return NULL; - bh = udf_tread(sb, block + UDF_SB_SESSION(sb)); + bh = udf_tread(sb, block + sbi->s_session); if (!bh) { udf_debug("block=%d, location=%d: read failed\n", - block + UDF_SB_SESSION(sb), location); + block + sbi->s_session, location); return NULL; } @@ -210,24 +223,20 @@ struct buffer_head *udf_read_tagged(struct super_block *sb, uint32_t block, if (location != le32_to_cpu(tag_p->tagLocation)) { udf_debug("location mismatch block %u, tag %u != %u\n", - block + UDF_SB_SESSION(sb), le32_to_cpu(tag_p->tagLocation), location); + block + sbi->s_session, + le32_to_cpu(tag_p->tagLocation), location); goto error_out; } /* Verify the tag checksum */ - checksum = 0U; - for (i = 0; i < 4; i++) - checksum += (uint8_t)(bh->b_data[i]); - for (i = 5; i < 16; i++) - checksum += (uint8_t)(bh->b_data[i]); - if (checksum != tag_p->tagChecksum) { + if (udf_tag_checksum(tag_p) != tag_p->tagChecksum) { printk(KERN_ERR "udf: tag checksum failed block %d\n", block); goto error_out; } /* Verify the tag version */ - if (le16_to_cpu(tag_p->descVersion) != 0x0002U && - le16_to_cpu(tag_p->descVersion) != 0x0003U) { + if (tag_p->descVersion != cpu_to_le16(0x0002U) && + tag_p->descVersion != cpu_to_le16(0x0003U)) { udf_debug("tag version 0x%04x != 0x0002 || 0x0003 block %d\n", le16_to_cpu(tag_p->descVersion), block); goto error_out; @@ -236,11 +245,11 @@ struct buffer_head *udf_read_tagged(struct super_block *sb, uint32_t block, /* Verify the descriptor CRC */ if (le16_to_cpu(tag_p->descCRCLength) + sizeof(tag) > sb->s_blocksize || le16_to_cpu(tag_p->descCRC) == udf_crc(bh->b_data + sizeof(tag), - le16_to_cpu(tag_p->descCRCLength), 0)) { + le16_to_cpu(tag_p->descCRCLength), 0)) return bh; - } + udf_debug("Crc failure block %d: crc = %d, crclen = %d\n", - block + UDF_SB_SESSION(sb), le16_to_cpu(tag_p->descCRC), + block + sbi->s_session, le16_to_cpu(tag_p->descCRC), le16_to_cpu(tag_p->descCRCLength)); error_out: @@ -249,7 +258,7 @@ error_out: } struct buffer_head *udf_read_ptagged(struct super_block *sb, kernel_lb_addr loc, - uint32_t offset, uint16_t * ident) + uint32_t offset, uint16_t *ident) { return udf_read_tagged(sb, udf_get_lb_pblock(sb, loc, offset), loc.logicalBlockNum + offset, ident); @@ -258,17 +267,11 @@ struct buffer_head *udf_read_ptagged(struct super_block *sb, kernel_lb_addr loc, void udf_update_tag(char *data, int length) { tag *tptr = (tag *)data; - int i; - length -= sizeof(tag); - tptr->tagChecksum = 0; tptr->descCRCLength = cpu_to_le16(length); tptr->descCRC = cpu_to_le16(udf_crc(data + sizeof(tag), length, 0)); - - for (i = 0; i < 16; i++) - if (i != 4) - tptr->tagChecksum += (uint8_t)(data[i]); + tptr->tagChecksum = udf_tag_checksum(tptr); } void udf_new_tag(char *data, uint16_t ident, uint16_t version, uint16_t snum, @@ -281,3 +284,14 @@ void udf_new_tag(char *data, uint16_t ident, uint16_t version, uint16_t snum, tptr->tagLocation = cpu_to_le32(loc); udf_update_tag(data, length); } + +u8 udf_tag_checksum(const tag *t) +{ + u8 *data = (u8 *)t; + u8 checksum = 0; + int i; + for (i = 0; i < sizeof(tag); ++i) + if (i != 4) /* position of checksum */ + checksum += data[i]; + return checksum; +} diff --git a/fs/udf/namei.c b/fs/udf/namei.c index bec96a6b334..112a5fb0b27 100644 --- a/fs/udf/namei.c +++ b/fs/udf/namei.c @@ -43,12 +43,10 @@ static inline int udf_match(int len1, const char *name1, int len2, int udf_write_fi(struct inode *inode, struct fileIdentDesc *cfi, struct fileIdentDesc *sfi, struct udf_fileident_bh *fibh, - uint8_t * impuse, uint8_t * fileident) + uint8_t *impuse, uint8_t *fileident) { uint16_t crclen = fibh->eoffset - fibh->soffset - sizeof(tag); uint16_t crc; - uint8_t checksum = 0; - int i; int offset; uint16_t liu = le16_to_cpu(cfi->lengthOfImpUse); uint8_t lfi = cfi->lengthFileIdent; @@ -56,7 +54,7 @@ int udf_write_fi(struct inode *inode, struct fileIdentDesc *cfi, sizeof(struct fileIdentDesc); int adinicb = 0; - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) + if (UDF_I(inode)->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) adinicb = 1; offset = fibh->soffset + sizeof(struct fileIdentDesc); @@ -68,7 +66,8 @@ int udf_write_fi(struct inode *inode, struct fileIdentDesc *cfi, memcpy(fibh->ebh->b_data + offset, impuse, liu); } else { memcpy((uint8_t *)sfi->impUse, impuse, -offset); - memcpy(fibh->ebh->b_data, impuse - offset, liu + offset); + memcpy(fibh->ebh->b_data, impuse - offset, + liu + offset); } } @@ -80,8 +79,10 @@ int udf_write_fi(struct inode *inode, struct fileIdentDesc *cfi, } else if (offset >= 0) { memcpy(fibh->ebh->b_data + offset, fileident, lfi); } else { - memcpy((uint8_t *)sfi->fileIdent + liu, fileident, -offset); - memcpy(fibh->ebh->b_data, fileident - offset, lfi + offset); + memcpy((uint8_t *)sfi->fileIdent + liu, fileident, + -offset); + memcpy(fibh->ebh->b_data, fileident - offset, + lfi + offset); } } @@ -101,27 +102,29 @@ int udf_write_fi(struct inode *inode, struct fileIdentDesc *cfi, if (fibh->sbh == fibh->ebh) { crc = udf_crc((uint8_t *)sfi->impUse, - crclen + sizeof(tag) - sizeof(struct fileIdentDesc), crc); + crclen + sizeof(tag) - + sizeof(struct fileIdentDesc), crc); } else if (sizeof(struct fileIdentDesc) >= -fibh->soffset) { - crc = udf_crc(fibh->ebh->b_data + sizeof(struct fileIdentDesc) + fibh->soffset, - crclen + sizeof(tag) - sizeof(struct fileIdentDesc), crc); + crc = udf_crc(fibh->ebh->b_data + + sizeof(struct fileIdentDesc) + + fibh->soffset, + crclen + sizeof(tag) - + sizeof(struct fileIdentDesc), + crc); } else { crc = udf_crc((uint8_t *)sfi->impUse, - -fibh->soffset - sizeof(struct fileIdentDesc), crc); + -fibh->soffset - sizeof(struct fileIdentDesc), + crc); crc = udf_crc(fibh->ebh->b_data, fibh->eoffset, crc); } cfi->descTag.descCRC = cpu_to_le16(crc); cfi->descTag.descCRCLength = cpu_to_le16(crclen); + cfi->descTag.tagChecksum = udf_tag_checksum(&cfi->descTag); - for (i = 0; i < 16; i++) { - if (i != 4) - checksum += ((uint8_t *)&cfi->descTag)[i]; - } - - cfi->descTag.tagChecksum = checksum; if (adinicb || (sizeof(struct fileIdentDesc) <= -fibh->soffset)) { - memcpy((uint8_t *)sfi, (uint8_t *)cfi, sizeof(struct fileIdentDesc)); + memcpy((uint8_t *)sfi, (uint8_t *)cfi, + sizeof(struct fileIdentDesc)); } else { memcpy((uint8_t *)sfi, (uint8_t *)cfi, -fibh->soffset); memcpy(fibh->ebh->b_data, (uint8_t *)cfi - fibh->soffset, @@ -155,26 +158,28 @@ static struct fileIdentDesc *udf_find_entry(struct inode *dir, uint32_t elen; sector_t offset; struct extent_position epos = {}; + struct udf_inode_info *dinfo = UDF_I(dir); - size = (udf_ext0_offset(dir) + dir->i_size) >> 2; - f_pos = (udf_ext0_offset(dir) >> 2); + size = udf_ext0_offset(dir) + dir->i_size; + f_pos = udf_ext0_offset(dir); - fibh->soffset = fibh->eoffset = (f_pos & ((dir->i_sb->s_blocksize - 1) >> 2)) << 2; - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) { + fibh->soffset = fibh->eoffset = f_pos & (dir->i_sb->s_blocksize - 1); + if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) fibh->sbh = fibh->ebh = NULL; - } else if (inode_bmap(dir, f_pos >> (dir->i_sb->s_blocksize_bits - 2), - &epos, &eloc, &elen, &offset) == (EXT_RECORDED_ALLOCATED >> 30)) { + else if (inode_bmap(dir, f_pos >> dir->i_sb->s_blocksize_bits, + &epos, &eloc, &elen, &offset) == + (EXT_RECORDED_ALLOCATED >> 30)) { block = udf_get_lb_pblock(dir->i_sb, eloc, offset); if ((++offset << dir->i_sb->s_blocksize_bits) < elen) { - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_SHORT) + if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT) epos.offset -= sizeof(short_ad); - else if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_LONG) + else if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_LONG) epos.offset -= sizeof(long_ad); - } else { + } else offset = 0; - } - if (!(fibh->sbh = fibh->ebh = udf_tread(dir->i_sb, block))) { + fibh->sbh = fibh->ebh = udf_tread(dir->i_sb, block); + if (!fibh->sbh) { brelse(epos.bh); return NULL; } @@ -183,7 +188,7 @@ static struct fileIdentDesc *udf_find_entry(struct inode *dir, return NULL; } - while ((f_pos < size)) { + while (f_pos < size) { fi = udf_fileident_read(dir, &f_pos, fibh, cfi, &epos, &eloc, &elen, &offset); if (!fi) { @@ -202,14 +207,18 @@ static struct fileIdentDesc *udf_find_entry(struct inode *dir, } else { int poffset; /* Unpaded ending offset */ - poffset = fibh->soffset + sizeof(struct fileIdentDesc) + liu + lfi; + poffset = fibh->soffset + sizeof(struct fileIdentDesc) + + liu + lfi; - if (poffset >= lfi) { - nameptr = (uint8_t *)(fibh->ebh->b_data + poffset - lfi); - } else { + if (poffset >= lfi) + nameptr = (uint8_t *)(fibh->ebh->b_data + + poffset - lfi); + else { nameptr = fname; - memcpy(nameptr, fi->fileIdent + liu, lfi - poffset); - memcpy(nameptr + lfi - poffset, fibh->ebh->b_data, poffset); + memcpy(nameptr, fi->fileIdent + liu, + lfi - poffset); + memcpy(nameptr + lfi - poffset, + fibh->ebh->b_data, poffset); } } @@ -226,11 +235,11 @@ static struct fileIdentDesc *udf_find_entry(struct inode *dir, if (!lfi) continue; - if ((flen = udf_get_filename(dir->i_sb, nameptr, fname, lfi))) { - if (udf_match(flen, fname, dentry->d_name.len, dentry->d_name.name)) { - brelse(epos.bh); - return fi; - } + flen = udf_get_filename(dir->i_sb, nameptr, fname, lfi); + if (flen && udf_match(flen, fname, dentry->d_name.len, + dentry->d_name.name)) { + brelse(epos.bh); + return fi; } } @@ -291,16 +300,16 @@ static struct dentry *udf_lookup(struct inode *dir, struct dentry *dentry, if (!strncmp(dentry->d_name.name, ".B=", 3)) { kernel_lb_addr lb = { .logicalBlockNum = 0, - .partitionReferenceNum = simple_strtoul(dentry->d_name.name + 3, - NULL, 0), + .partitionReferenceNum = + simple_strtoul(dentry->d_name.name + 3, + NULL, 0), }; inode = udf_iget(dir->i_sb, lb); if (!inode) { unlock_kernel(); return ERR_PTR(-EACCES); } - } - else + } else #endif /* UDF_RECOVERY */ if (udf_find_entry(dir, dentry, &fibh, &cfi)) { @@ -325,14 +334,14 @@ static struct fileIdentDesc *udf_add_entry(struct inode *dir, struct udf_fileident_bh *fibh, struct fileIdentDesc *cfi, int *err) { - struct super_block *sb; + struct super_block *sb = dir->i_sb; struct fileIdentDesc *fi = NULL; char name[UDF_NAME_LEN], fname[UDF_NAME_LEN]; int namelen; loff_t f_pos; int flen; char *nameptr; - loff_t size = (udf_ext0_offset(dir) + dir->i_size) >> 2; + loff_t size = udf_ext0_offset(dir) + dir->i_size; int nfidlen; uint8_t lfi; uint16_t liu; @@ -341,16 +350,16 @@ static struct fileIdentDesc *udf_add_entry(struct inode *dir, uint32_t elen; sector_t offset; struct extent_position epos = {}; - - sb = dir->i_sb; + struct udf_inode_info *dinfo; if (dentry) { if (!dentry->d_name.len) { *err = -EINVAL; return NULL; } - if (!(namelen = udf_put_filename(sb, dentry->d_name.name, name, - dentry->d_name.len))) { + namelen = udf_put_filename(sb, dentry->d_name.name, name, + dentry->d_name.len); + if (!namelen) { *err = -ENAMETOOLONG; return NULL; } @@ -360,39 +369,40 @@ static struct fileIdentDesc *udf_add_entry(struct inode *dir, nfidlen = (sizeof(struct fileIdentDesc) + namelen + 3) & ~3; - f_pos = (udf_ext0_offset(dir) >> 2); + f_pos = udf_ext0_offset(dir); - fibh->soffset = fibh->eoffset = (f_pos & ((dir->i_sb->s_blocksize - 1) >> 2)) << 2; - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) { + fibh->soffset = fibh->eoffset = f_pos & (dir->i_sb->s_blocksize - 1); + dinfo = UDF_I(dir); + if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) fibh->sbh = fibh->ebh = NULL; - } else if (inode_bmap(dir, f_pos >> (dir->i_sb->s_blocksize_bits - 2), - &epos, &eloc, &elen, &offset) == (EXT_RECORDED_ALLOCATED >> 30)) { + else if (inode_bmap(dir, f_pos >> dir->i_sb->s_blocksize_bits, + &epos, &eloc, &elen, &offset) == + (EXT_RECORDED_ALLOCATED >> 30)) { block = udf_get_lb_pblock(dir->i_sb, eloc, offset); if ((++offset << dir->i_sb->s_blocksize_bits) < elen) { - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_SHORT) + if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT) epos.offset -= sizeof(short_ad); - else if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_LONG) + else if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_LONG) epos.offset -= sizeof(long_ad); - } else { + } else offset = 0; - } - if (!(fibh->sbh = fibh->ebh = udf_tread(dir->i_sb, block))) { + fibh->sbh = fibh->ebh = udf_tread(dir->i_sb, block); + if (!fibh->sbh) { brelse(epos.bh); *err = -EIO; return NULL; } - block = UDF_I_LOCATION(dir).logicalBlockNum; - + block = dinfo->i_location.logicalBlockNum; } else { - block = udf_get_lb_pblock(dir->i_sb, UDF_I_LOCATION(dir), 0); + block = udf_get_lb_pblock(dir->i_sb, dinfo->i_location, 0); fibh->sbh = fibh->ebh = NULL; fibh->soffset = fibh->eoffset = sb->s_blocksize; goto add; } - while ((f_pos < size)) { + while (f_pos < size) { fi = udf_fileident_read(dir, &f_pos, fibh, cfi, &epos, &eloc, &elen, &offset); @@ -408,33 +418,39 @@ static struct fileIdentDesc *udf_add_entry(struct inode *dir, liu = le16_to_cpu(cfi->lengthOfImpUse); lfi = cfi->lengthFileIdent; - if (fibh->sbh == fibh->ebh) { + if (fibh->sbh == fibh->ebh) nameptr = fi->fileIdent + liu; - } else { + else { int poffset; /* Unpaded ending offset */ - poffset = fibh->soffset + sizeof(struct fileIdentDesc) + liu + lfi; + poffset = fibh->soffset + sizeof(struct fileIdentDesc) + + liu + lfi; - if (poffset >= lfi) { - nameptr = (char *)(fibh->ebh->b_data + poffset - lfi); - } else { + if (poffset >= lfi) + nameptr = (char *)(fibh->ebh->b_data + + poffset - lfi); + else { nameptr = fname; - memcpy(nameptr, fi->fileIdent + liu, lfi - poffset); - memcpy(nameptr + lfi - poffset, fibh->ebh->b_data, poffset); + memcpy(nameptr, fi->fileIdent + liu, + lfi - poffset); + memcpy(nameptr + lfi - poffset, + fibh->ebh->b_data, poffset); } } if ((cfi->fileCharacteristics & FID_FILE_CHAR_DELETED) != 0) { - if (((sizeof(struct fileIdentDesc) + liu + lfi + 3) & ~3) == nfidlen) { + if (((sizeof(struct fileIdentDesc) + + liu + lfi + 3) & ~3) == nfidlen) { brelse(epos.bh); cfi->descTag.tagSerialNum = cpu_to_le16(1); cfi->fileVersionNum = cpu_to_le16(1); cfi->fileCharacteristics = 0; cfi->lengthFileIdent = namelen; cfi->lengthOfImpUse = cpu_to_le16(0); - if (!udf_write_fi(dir, cfi, fi, fibh, NULL, name)) { + if (!udf_write_fi(dir, cfi, fi, fibh, NULL, + name)) return fi; - } else { + else { *err = -EIO; return NULL; } @@ -444,8 +460,9 @@ static struct fileIdentDesc *udf_add_entry(struct inode *dir, if (!lfi || !dentry) continue; - if ((flen = udf_get_filename(dir->i_sb, nameptr, fname, lfi)) && - udf_match(flen, fname, dentry->d_name.len, dentry->d_name.name)) { + flen = udf_get_filename(dir->i_sb, nameptr, fname, lfi); + if (flen && udf_match(flen, fname, dentry->d_name.len, + dentry->d_name.name)) { if (fibh->sbh != fibh->ebh) brelse(fibh->ebh); brelse(fibh->sbh); @@ -456,29 +473,34 @@ static struct fileIdentDesc *udf_add_entry(struct inode *dir, } add: + if (dinfo->i_alloc_type != ICBTAG_FLAG_AD_IN_ICB) { + elen = (elen + sb->s_blocksize - 1) & ~(sb->s_blocksize - 1); + if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT) + epos.offset -= sizeof(short_ad); + else if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_LONG) + epos.offset -= sizeof(long_ad); + udf_write_aext(dir, &epos, eloc, elen, 1); + } f_pos += nfidlen; - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB && + if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB && sb->s_blocksize - fibh->eoffset < nfidlen) { brelse(epos.bh); epos.bh = NULL; fibh->soffset -= udf_ext0_offset(dir); fibh->eoffset -= udf_ext0_offset(dir); - f_pos -= (udf_ext0_offset(dir) >> 2); + f_pos -= udf_ext0_offset(dir); if (fibh->sbh != fibh->ebh) brelse(fibh->ebh); brelse(fibh->sbh); - if (!(fibh->sbh = fibh->ebh = udf_expand_dir_adinicb(dir, &block, err))) + fibh->sbh = fibh->ebh = + udf_expand_dir_adinicb(dir, &block, err); + if (!fibh->sbh) return NULL; - epos.block = UDF_I_LOCATION(dir); - eloc.logicalBlockNum = block; - eloc.partitionReferenceNum = UDF_I_LOCATION(dir).partitionReferenceNum; - elen = dir->i_sb->s_blocksize; + epos.block = dinfo->i_location; epos.offset = udf_file_entry_alloc_offset(dir); - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_SHORT) - epos.offset += sizeof(short_ad); - else if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_LONG) - epos.offset += sizeof(long_ad); + /* Load extent udf_expand_dir_adinicb() has created */ + udf_current_aext(dir, &epos, &eloc, &elen, 1); } if (sb->s_blocksize - fibh->eoffset >= nfidlen) { @@ -489,15 +511,19 @@ add: fibh->sbh = fibh->ebh; } - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) { - block = UDF_I_LOCATION(dir).logicalBlockNum; - fi = (struct fileIdentDesc *)(UDF_I_DATA(dir) + fibh->soffset - - udf_ext0_offset(dir) + - UDF_I_LENEATTR(dir)); + if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) { + block = dinfo->i_location.logicalBlockNum; + fi = (struct fileIdentDesc *) + (dinfo->i_ext.i_data + + fibh->soffset - + udf_ext0_offset(dir) + + dinfo->i_lenEAttr); } else { - block = eloc.logicalBlockNum + ((elen - 1) >> - dir->i_sb->s_blocksize_bits); - fi = (struct fileIdentDesc *)(fibh->sbh->b_data + fibh->soffset); + block = eloc.logicalBlockNum + + ((elen - 1) >> + dir->i_sb->s_blocksize_bits); + fi = (struct fileIdentDesc *) + (fibh->sbh->b_data + fibh->soffset); } } else { fibh->soffset = fibh->eoffset - sb->s_blocksize; @@ -509,7 +535,8 @@ add: block = eloc.logicalBlockNum + ((elen - 1) >> dir->i_sb->s_blocksize_bits); - fibh->ebh = udf_bread(dir, f_pos >> (dir->i_sb->s_blocksize_bits - 2), 1, err); + fibh->ebh = udf_bread(dir, + f_pos >> dir->i_sb->s_blocksize_bits, 1, err); if (!fibh->ebh) { brelse(epos.bh); brelse(fibh->sbh); @@ -521,32 +548,34 @@ add: (EXT_RECORDED_ALLOCATED >> 30)) { block = eloc.logicalBlockNum + ((elen - 1) >> dir->i_sb->s_blocksize_bits); - } else { + } else block++; - } brelse(fibh->sbh); fibh->sbh = fibh->ebh; fi = (struct fileIdentDesc *)(fibh->sbh->b_data); } else { fi = (struct fileIdentDesc *) - (fibh->sbh->b_data + sb->s_blocksize + fibh->soffset); + (fibh->sbh->b_data + sb->s_blocksize + + fibh->soffset); } } memset(cfi, 0, sizeof(struct fileIdentDesc)); - if (UDF_SB_UDFREV(sb) >= 0x0200) - udf_new_tag((char *)cfi, TAG_IDENT_FID, 3, 1, block, sizeof(tag)); + if (UDF_SB(sb)->s_udfrev >= 0x0200) + udf_new_tag((char *)cfi, TAG_IDENT_FID, 3, 1, block, + sizeof(tag)); else - udf_new_tag((char *)cfi, TAG_IDENT_FID, 2, 1, block, sizeof(tag)); + udf_new_tag((char *)cfi, TAG_IDENT_FID, 2, 1, block, + sizeof(tag)); cfi->fileVersionNum = cpu_to_le16(1); cfi->lengthFileIdent = namelen; cfi->lengthOfImpUse = cpu_to_le16(0); if (!udf_write_fi(dir, cfi, fi, fibh, NULL, name)) { brelse(epos.bh); dir->i_size += nfidlen; - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) - UDF_I_LENALLOC(dir) += nfidlen; + if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) + dinfo->i_lenAlloc += nfidlen; mark_inode_dirty(dir); return fi; } else { @@ -578,6 +607,7 @@ static int udf_create(struct inode *dir, struct dentry *dentry, int mode, struct inode *inode; struct fileIdentDesc cfi, *fi; int err; + struct udf_inode_info *iinfo; lock_kernel(); inode = udf_new_inode(dir, mode, &err); @@ -586,7 +616,8 @@ static int udf_create(struct inode *dir, struct dentry *dentry, int mode, return err; } - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) + iinfo = UDF_I(inode); + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) inode->i_data.a_ops = &udf_adinicb_aops; else inode->i_data.a_ops = &udf_aops; @@ -595,7 +626,8 @@ static int udf_create(struct inode *dir, struct dentry *dentry, int mode, inode->i_mode = mode; mark_inode_dirty(inode); - if (!(fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err))) { + fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err); + if (!fi) { inode->i_nlink--; mark_inode_dirty(inode); iput(inode); @@ -603,13 +635,12 @@ static int udf_create(struct inode *dir, struct dentry *dentry, int mode, return err; } cfi.icb.extLength = cpu_to_le32(inode->i_sb->s_blocksize); - cfi.icb.extLocation = cpu_to_lelb(UDF_I_LOCATION(inode)); + cfi.icb.extLocation = cpu_to_lelb(iinfo->i_location); *(__le32 *)((struct allocDescImpUse *)cfi.icb.impUse)->impUse = - cpu_to_le32(UDF_I_UNIQUE(inode) & 0x00000000FFFFFFFFUL); + cpu_to_le32(iinfo->i_unique & 0x00000000FFFFFFFFUL); udf_write_fi(dir, &cfi, fi, &fibh, NULL, NULL); - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) { + if (UDF_I(dir)->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) mark_inode_dirty(dir); - } if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); brelse(fibh.sbh); @@ -626,6 +657,7 @@ static int udf_mknod(struct inode *dir, struct dentry *dentry, int mode, struct udf_fileident_bh fibh; struct fileIdentDesc cfi, *fi; int err; + struct udf_inode_info *iinfo; if (!old_valid_dev(rdev)) return -EINVAL; @@ -636,9 +668,11 @@ static int udf_mknod(struct inode *dir, struct dentry *dentry, int mode, if (!inode) goto out; + iinfo = UDF_I(inode); inode->i_uid = current->fsuid; init_special_inode(inode, mode, rdev); - if (!(fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err))) { + fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err); + if (!fi) { inode->i_nlink--; mark_inode_dirty(inode); iput(inode); @@ -646,13 +680,12 @@ static int udf_mknod(struct inode *dir, struct dentry *dentry, int mode, return err; } cfi.icb.extLength = cpu_to_le32(inode->i_sb->s_blocksize); - cfi.icb.extLocation = cpu_to_lelb(UDF_I_LOCATION(inode)); + cfi.icb.extLocation = cpu_to_lelb(iinfo->i_location); *(__le32 *)((struct allocDescImpUse *)cfi.icb.impUse)->impUse = - cpu_to_le32(UDF_I_UNIQUE(inode) & 0x00000000FFFFFFFFUL); + cpu_to_le32(iinfo->i_unique & 0x00000000FFFFFFFFUL); udf_write_fi(dir, &cfi, fi, &fibh, NULL, NULL); - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) { + if (UDF_I(dir)->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) mark_inode_dirty(dir); - } mark_inode_dirty(inode); if (fibh.sbh != fibh.ebh) @@ -672,6 +705,8 @@ static int udf_mkdir(struct inode *dir, struct dentry *dentry, int mode) struct udf_fileident_bh fibh; struct fileIdentDesc cfi, *fi; int err; + struct udf_inode_info *dinfo = UDF_I(dir); + struct udf_inode_info *iinfo; lock_kernel(); err = -EMLINK; @@ -683,9 +718,11 @@ static int udf_mkdir(struct inode *dir, struct dentry *dentry, int mode) if (!inode) goto out; + iinfo = UDF_I(inode); inode->i_op = &udf_dir_inode_operations; inode->i_fop = &udf_dir_operations; - if (!(fi = udf_add_entry(inode, NULL, &fibh, &cfi, &err))) { + fi = udf_add_entry(inode, NULL, &fibh, &cfi, &err); + if (!fi) { inode->i_nlink--; mark_inode_dirty(inode); iput(inode); @@ -693,10 +730,11 @@ static int udf_mkdir(struct inode *dir, struct dentry *dentry, int mode) } inode->i_nlink = 2; cfi.icb.extLength = cpu_to_le32(inode->i_sb->s_blocksize); - cfi.icb.extLocation = cpu_to_lelb(UDF_I_LOCATION(dir)); + cfi.icb.extLocation = cpu_to_lelb(dinfo->i_location); *(__le32 *)((struct allocDescImpUse *)cfi.icb.impUse)->impUse = - cpu_to_le32(UDF_I_UNIQUE(dir) & 0x00000000FFFFFFFFUL); - cfi.fileCharacteristics = FID_FILE_CHAR_DIRECTORY | FID_FILE_CHAR_PARENT; + cpu_to_le32(dinfo->i_unique & 0x00000000FFFFFFFFUL); + cfi.fileCharacteristics = + FID_FILE_CHAR_DIRECTORY | FID_FILE_CHAR_PARENT; udf_write_fi(inode, &cfi, fi, &fibh, NULL, NULL); brelse(fibh.sbh); inode->i_mode = S_IFDIR | mode; @@ -704,16 +742,17 @@ static int udf_mkdir(struct inode *dir, struct dentry *dentry, int mode) inode->i_mode |= S_ISGID; mark_inode_dirty(inode); - if (!(fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err))) { + fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err); + if (!fi) { inode->i_nlink = 0; mark_inode_dirty(inode); iput(inode); goto out; } cfi.icb.extLength = cpu_to_le32(inode->i_sb->s_blocksize); - cfi.icb.extLocation = cpu_to_lelb(UDF_I_LOCATION(inode)); + cfi.icb.extLocation = cpu_to_lelb(iinfo->i_location); *(__le32 *)((struct allocDescImpUse *)cfi.icb.impUse)->impUse = - cpu_to_le32(UDF_I_UNIQUE(inode) & 0x00000000FFFFFFFFUL); + cpu_to_le32(iinfo->i_unique & 0x00000000FFFFFFFFUL); cfi.fileCharacteristics |= FID_FILE_CHAR_DIRECTORY; udf_write_fi(dir, &cfi, fi, &fibh, NULL, NULL); inc_nlink(dir); @@ -734,32 +773,33 @@ static int empty_dir(struct inode *dir) struct fileIdentDesc *fi, cfi; struct udf_fileident_bh fibh; loff_t f_pos; - loff_t size = (udf_ext0_offset(dir) + dir->i_size) >> 2; + loff_t size = udf_ext0_offset(dir) + dir->i_size; int block; kernel_lb_addr eloc; uint32_t elen; sector_t offset; struct extent_position epos = {}; + struct udf_inode_info *dinfo = UDF_I(dir); - f_pos = (udf_ext0_offset(dir) >> 2); + f_pos = udf_ext0_offset(dir); + fibh.soffset = fibh.eoffset = f_pos & (dir->i_sb->s_blocksize - 1); - fibh.soffset = fibh.eoffset = (f_pos & ((dir->i_sb->s_blocksize - 1) >> 2)) << 2; - - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) { + if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) fibh.sbh = fibh.ebh = NULL; - } else if (inode_bmap(dir, f_pos >> (dir->i_sb->s_blocksize_bits - 2), - &epos, &eloc, &elen, &offset) == (EXT_RECORDED_ALLOCATED >> 30)) { + else if (inode_bmap(dir, f_pos >> dir->i_sb->s_blocksize_bits, + &epos, &eloc, &elen, &offset) == + (EXT_RECORDED_ALLOCATED >> 30)) { block = udf_get_lb_pblock(dir->i_sb, eloc, offset); if ((++offset << dir->i_sb->s_blocksize_bits) < elen) { - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_SHORT) + if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT) epos.offset -= sizeof(short_ad); - else if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_LONG) + else if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_LONG) epos.offset -= sizeof(long_ad); - } else { + } else offset = 0; - } - if (!(fibh.sbh = fibh.ebh = udf_tread(dir->i_sb, block))) { + fibh.sbh = fibh.ebh = udf_tread(dir->i_sb, block); + if (!fibh.sbh) { brelse(epos.bh); return 0; } @@ -768,7 +808,7 @@ static int empty_dir(struct inode *dir) return 0; } - while ((f_pos < size)) { + while (f_pos < size) { fi = udf_fileident_read(dir, &f_pos, &fibh, &cfi, &epos, &eloc, &elen, &offset); if (!fi) { @@ -828,7 +868,8 @@ static int udf_rmdir(struct inode *dir, struct dentry *dentry) clear_nlink(inode); inode->i_size = 0; inode_dec_link_count(dir); - inode->i_ctime = dir->i_ctime = dir->i_mtime = current_fs_time(dir->i_sb); + inode->i_ctime = dir->i_ctime = dir->i_mtime = + current_fs_time(dir->i_sb); mark_inode_dirty(dir); end_rmdir: @@ -901,36 +942,42 @@ static int udf_symlink(struct inode *dir, struct dentry *dentry, int block; char name[UDF_NAME_LEN]; int namelen; + struct buffer_head *bh; + struct udf_inode_info *iinfo; lock_kernel(); - if (!(inode = udf_new_inode(dir, S_IFLNK, &err))) + inode = udf_new_inode(dir, S_IFLNK, &err); + if (!inode) goto out; + iinfo = UDF_I(inode); inode->i_mode = S_IFLNK | S_IRWXUGO; inode->i_data.a_ops = &udf_symlink_aops; inode->i_op = &page_symlink_inode_operations; - if (UDF_I_ALLOCTYPE(inode) != ICBTAG_FLAG_AD_IN_ICB) { + if (iinfo->i_alloc_type != ICBTAG_FLAG_AD_IN_ICB) { kernel_lb_addr eloc; uint32_t elen; block = udf_new_block(inode->i_sb, inode, - UDF_I_LOCATION(inode).partitionReferenceNum, - UDF_I_LOCATION(inode).logicalBlockNum, &err); + iinfo->i_location.partitionReferenceNum, + iinfo->i_location.logicalBlockNum, &err); if (!block) goto out_no_entry; - epos.block = UDF_I_LOCATION(inode); + epos.block = iinfo->i_location; epos.offset = udf_file_entry_alloc_offset(inode); epos.bh = NULL; eloc.logicalBlockNum = block; - eloc.partitionReferenceNum = UDF_I_LOCATION(inode).partitionReferenceNum; + eloc.partitionReferenceNum = + iinfo->i_location.partitionReferenceNum; elen = inode->i_sb->s_blocksize; - UDF_I_LENEXTENTS(inode) = elen; + iinfo->i_lenExtents = elen; udf_add_aext(inode, &epos, eloc, elen, 0); brelse(epos.bh); block = udf_get_pblock(inode->i_sb, block, - UDF_I_LOCATION(inode).partitionReferenceNum, 0); + iinfo->i_location.partitionReferenceNum, + 0); epos.bh = udf_tread(inode->i_sb, block); lock_buffer(epos.bh); memset(epos.bh->b_data, 0x00, inode->i_sb->s_blocksize); @@ -938,9 +985,8 @@ static int udf_symlink(struct inode *dir, struct dentry *dentry, unlock_buffer(epos.bh); mark_buffer_dirty_inode(epos.bh, inode); ea = epos.bh->b_data + udf_ext0_offset(inode); - } else { - ea = UDF_I_DATA(inode) + UDF_I_LENEATTR(inode); - } + } else + ea = iinfo->i_ext.i_data + iinfo->i_lenEAttr; eoffset = inode->i_sb->s_blocksize - udf_ext0_offset(inode); pc = (struct pathComponent *)ea; @@ -977,7 +1023,8 @@ static int udf_symlink(struct inode *dir, struct dentry *dentry, if (compstart[0] == '.') { if ((symname - compstart) == 1) pc->componentType = 4; - else if ((symname - compstart) == 2 && compstart[1] == '.') + else if ((symname - compstart) == 2 && + compstart[1] == '.') pc->componentType = 3; } @@ -987,7 +1034,8 @@ static int udf_symlink(struct inode *dir, struct dentry *dentry, if (!namelen) goto out_no_entry; - if (elen + sizeof(struct pathComponent) + namelen > eoffset) + if (elen + sizeof(struct pathComponent) + namelen > + eoffset) goto out_no_entry; else pc->lengthComponentIdent = namelen; @@ -1006,30 +1054,34 @@ static int udf_symlink(struct inode *dir, struct dentry *dentry, brelse(epos.bh); inode->i_size = elen; - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) - UDF_I_LENALLOC(inode) = inode->i_size; + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) + iinfo->i_lenAlloc = inode->i_size; mark_inode_dirty(inode); - if (!(fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err))) + fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err); + if (!fi) goto out_no_entry; cfi.icb.extLength = cpu_to_le32(inode->i_sb->s_blocksize); - cfi.icb.extLocation = cpu_to_lelb(UDF_I_LOCATION(inode)); - if (UDF_SB_LVIDBH(inode->i_sb)) { + cfi.icb.extLocation = cpu_to_lelb(iinfo->i_location); + bh = UDF_SB(inode->i_sb)->s_lvid_bh; + if (bh) { + struct logicalVolIntegrityDesc *lvid = + (struct logicalVolIntegrityDesc *)bh->b_data; struct logicalVolHeaderDesc *lvhd; uint64_t uniqueID; - lvhd = (struct logicalVolHeaderDesc *)(UDF_SB_LVID(inode->i_sb)->logicalVolContentsUse); + lvhd = (struct logicalVolHeaderDesc *) + lvid->logicalVolContentsUse; uniqueID = le64_to_cpu(lvhd->uniqueID); *(__le32 *)((struct allocDescImpUse *)cfi.icb.impUse)->impUse = cpu_to_le32(uniqueID & 0x00000000FFFFFFFFUL); if (!(++uniqueID & 0x00000000FFFFFFFFUL)) uniqueID += 16; lvhd->uniqueID = cpu_to_le64(uniqueID); - mark_buffer_dirty(UDF_SB_LVIDBH(inode->i_sb)); + mark_buffer_dirty(bh); } udf_write_fi(dir, &cfi, fi, &fibh, NULL, NULL); - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) { + if (UDF_I(dir)->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) mark_inode_dirty(dir); - } if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); brelse(fibh.sbh); @@ -1053,6 +1105,7 @@ static int udf_link(struct dentry *old_dentry, struct inode *dir, struct udf_fileident_bh fibh; struct fileIdentDesc cfi, *fi; int err; + struct buffer_head *bh; lock_kernel(); if (inode->i_nlink >= (256 << sizeof(inode->i_nlink)) - 1) { @@ -1060,28 +1113,32 @@ static int udf_link(struct dentry *old_dentry, struct inode *dir, return -EMLINK; } - if (!(fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err))) { + fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err); + if (!fi) { unlock_kernel(); return err; } cfi.icb.extLength = cpu_to_le32(inode->i_sb->s_blocksize); - cfi.icb.extLocation = cpu_to_lelb(UDF_I_LOCATION(inode)); - if (UDF_SB_LVIDBH(inode->i_sb)) { + cfi.icb.extLocation = cpu_to_lelb(UDF_I(inode)->i_location); + bh = UDF_SB(inode->i_sb)->s_lvid_bh; + if (bh) { + struct logicalVolIntegrityDesc *lvid = + (struct logicalVolIntegrityDesc *)bh->b_data; struct logicalVolHeaderDesc *lvhd; uint64_t uniqueID; - lvhd = (struct logicalVolHeaderDesc *)(UDF_SB_LVID(inode->i_sb)->logicalVolContentsUse); + lvhd = (struct logicalVolHeaderDesc *) + (lvid->logicalVolContentsUse); uniqueID = le64_to_cpu(lvhd->uniqueID); *(__le32 *)((struct allocDescImpUse *)cfi.icb.impUse)->impUse = cpu_to_le32(uniqueID & 0x00000000FFFFFFFFUL); if (!(++uniqueID & 0x00000000FFFFFFFFUL)) uniqueID += 16; lvhd->uniqueID = cpu_to_le64(uniqueID); - mark_buffer_dirty(UDF_SB_LVIDBH(inode->i_sb)); + mark_buffer_dirty(bh); } udf_write_fi(dir, &cfi, fi, &fibh, NULL, NULL); - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) { + if (UDF_I(dir)->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) mark_inode_dirty(dir); - } if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); @@ -1105,13 +1162,16 @@ static int udf_rename(struct inode *old_dir, struct dentry *old_dentry, struct inode *old_inode = old_dentry->d_inode; struct inode *new_inode = new_dentry->d_inode; struct udf_fileident_bh ofibh, nfibh; - struct fileIdentDesc *ofi = NULL, *nfi = NULL, *dir_fi = NULL, ocfi, ncfi; + struct fileIdentDesc *ofi = NULL, *nfi = NULL, *dir_fi = NULL; + struct fileIdentDesc ocfi, ncfi; struct buffer_head *dir_bh = NULL; int retval = -ENOENT; kernel_lb_addr tloc; + struct udf_inode_info *old_iinfo = UDF_I(old_inode); lock_kernel(); - if ((ofi = udf_find_entry(old_dir, old_dentry, &ofibh, &ocfi))) { + ofi = udf_find_entry(old_dir, old_dentry, &ofibh, &ocfi); + if (ofi) { if (ofibh.sbh != ofibh.ebh) brelse(ofibh.ebh); brelse(ofibh.sbh); @@ -1131,7 +1191,7 @@ static int udf_rename(struct inode *old_dir, struct dentry *old_dentry, } } if (S_ISDIR(old_inode->i_mode)) { - uint32_t offset = udf_ext0_offset(old_inode); + int offset = udf_ext0_offset(old_inode); if (new_inode) { retval = -ENOTEMPTY; @@ -1139,30 +1199,36 @@ static int udf_rename(struct inode *old_dir, struct dentry *old_dentry, goto end_rename; } retval = -EIO; - if (UDF_I_ALLOCTYPE(old_inode) == ICBTAG_FLAG_AD_IN_ICB) { - dir_fi = udf_get_fileident(UDF_I_DATA(old_inode) - - (UDF_I_EFE(old_inode) ? - sizeof(struct extendedFileEntry) : - sizeof(struct fileEntry)), - old_inode->i_sb->s_blocksize, &offset); + if (old_iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) { + dir_fi = udf_get_fileident( + old_iinfo->i_ext.i_data - + (old_iinfo->i_efe ? + sizeof(struct extendedFileEntry) : + sizeof(struct fileEntry)), + old_inode->i_sb->s_blocksize, &offset); } else { dir_bh = udf_bread(old_inode, 0, 0, &retval); if (!dir_bh) goto end_rename; - dir_fi = udf_get_fileident(dir_bh->b_data, old_inode->i_sb->s_blocksize, &offset); + dir_fi = udf_get_fileident(dir_bh->b_data, + old_inode->i_sb->s_blocksize, &offset); } if (!dir_fi) goto end_rename; tloc = lelb_to_cpu(dir_fi->icb.extLocation); - if (udf_get_lb_pblock(old_inode->i_sb, tloc, 0) != old_dir->i_ino) + if (udf_get_lb_pblock(old_inode->i_sb, tloc, 0) != + old_dir->i_ino) goto end_rename; retval = -EMLINK; - if (!new_inode && new_dir->i_nlink >= (256 << sizeof(new_dir->i_nlink)) - 1) + if (!new_inode && + new_dir->i_nlink >= + (256 << sizeof(new_dir->i_nlink)) - 1) goto end_rename; } if (!nfi) { - nfi = udf_add_entry(new_dir, new_dentry, &nfibh, &ncfi, &retval); + nfi = udf_add_entry(new_dir, new_dentry, &nfibh, &ncfi, + &retval); if (!nfi) goto end_rename; } @@ -1194,18 +1260,19 @@ static int udf_rename(struct inode *old_dir, struct dentry *old_dentry, mark_inode_dirty(old_dir); if (dir_fi) { - dir_fi->icb.extLocation = cpu_to_lelb(UDF_I_LOCATION(new_dir)); - udf_update_tag((char *)dir_fi, (sizeof(struct fileIdentDesc) + - le16_to_cpu(dir_fi->lengthOfImpUse) + 3) & ~3); - if (UDF_I_ALLOCTYPE(old_inode) == ICBTAG_FLAG_AD_IN_ICB) { + dir_fi->icb.extLocation = cpu_to_lelb(UDF_I(new_dir)->i_location); + udf_update_tag((char *)dir_fi, + (sizeof(struct fileIdentDesc) + + le16_to_cpu(dir_fi->lengthOfImpUse) + 3) & ~3); + if (old_iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) mark_inode_dirty(old_inode); - } else { + else mark_buffer_dirty_inode(dir_bh, old_inode); - } + inode_dec_link_count(old_dir); - if (new_inode) { + if (new_inode) inode_dec_link_count(new_inode); - } else { + else { inc_nlink(new_dir); mark_inode_dirty(new_dir); } diff --git a/fs/udf/partition.c b/fs/udf/partition.c index aaab24c8c49..fc533345ab8 100644 --- a/fs/udf/partition.c +++ b/fs/udf/partition.c @@ -31,15 +31,18 @@ inline uint32_t udf_get_pblock(struct super_block *sb, uint32_t block, uint16_t partition, uint32_t offset) { - if (partition >= UDF_SB_NUMPARTS(sb)) { - udf_debug("block=%d, partition=%d, offset=%d: invalid partition\n", - block, partition, offset); + struct udf_sb_info *sbi = UDF_SB(sb); + struct udf_part_map *map; + if (partition >= sbi->s_partitions) { + udf_debug("block=%d, partition=%d, offset=%d: " + "invalid partition\n", block, partition, offset); return 0xFFFFFFFF; } - if (UDF_SB_PARTFUNC(sb, partition)) - return UDF_SB_PARTFUNC(sb, partition)(sb, block, partition, offset); + map = &sbi->s_partmaps[partition]; + if (map->s_partition_func) + return map->s_partition_func(sb, block, partition, offset); else - return UDF_SB_PARTROOT(sb, partition) + block + offset; + return map->s_partition_root + block + offset; } uint32_t udf_get_pblock_virt15(struct super_block *sb, uint32_t block, @@ -49,12 +52,18 @@ uint32_t udf_get_pblock_virt15(struct super_block *sb, uint32_t block, uint32_t newblock; uint32_t index; uint32_t loc; + struct udf_sb_info *sbi = UDF_SB(sb); + struct udf_part_map *map; + struct udf_virtual_data *vdata; + struct udf_inode_info *iinfo; - index = (sb->s_blocksize - UDF_SB_TYPEVIRT(sb,partition).s_start_offset) / sizeof(uint32_t); + map = &sbi->s_partmaps[partition]; + vdata = &map->s_type_specific.s_virtual; + index = (sb->s_blocksize - vdata->s_start_offset) / sizeof(uint32_t); - if (block > UDF_SB_TYPEVIRT(sb,partition).s_num_entries) { - udf_debug("Trying to access block beyond end of VAT (%d max %d)\n", - block, UDF_SB_TYPEVIRT(sb,partition).s_num_entries); + if (block > vdata->s_num_entries) { + udf_debug("Trying to access block beyond end of VAT " + "(%d max %d)\n", block, vdata->s_num_entries); return 0xFFFFFFFF; } @@ -64,12 +73,13 @@ uint32_t udf_get_pblock_virt15(struct super_block *sb, uint32_t block, index = block % (sb->s_blocksize / sizeof(uint32_t)); } else { newblock = 0; - index = UDF_SB_TYPEVIRT(sb,partition).s_start_offset / sizeof(uint32_t) + block; + index = vdata->s_start_offset / sizeof(uint32_t) + block; } - loc = udf_block_map(UDF_SB_VAT(sb), newblock); + loc = udf_block_map(sbi->s_vat_inode, newblock); - if (!(bh = sb_bread(sb, loc))) { + bh = sb_bread(sb, loc); + if (!bh) { udf_debug("get_pblock(UDF_VIRTUAL_MAP:%p,%d,%d) VAT: %d[%d]\n", sb, block, partition, loc, index); return 0xFFFFFFFF; @@ -79,50 +89,61 @@ uint32_t udf_get_pblock_virt15(struct super_block *sb, uint32_t block, brelse(bh); - if (UDF_I_LOCATION(UDF_SB_VAT(sb)).partitionReferenceNum == partition) { + iinfo = UDF_I(sbi->s_vat_inode); + if (iinfo->i_location.partitionReferenceNum == partition) { udf_debug("recursive call to udf_get_pblock!\n"); return 0xFFFFFFFF; } return udf_get_pblock(sb, loc, - UDF_I_LOCATION(UDF_SB_VAT(sb)).partitionReferenceNum, + iinfo->i_location.partitionReferenceNum, offset); } -inline uint32_t udf_get_pblock_virt20(struct super_block * sb, uint32_t block, +inline uint32_t udf_get_pblock_virt20(struct super_block *sb, uint32_t block, uint16_t partition, uint32_t offset) { return udf_get_pblock_virt15(sb, block, partition, offset); } -uint32_t udf_get_pblock_spar15(struct super_block * sb, uint32_t block, +uint32_t udf_get_pblock_spar15(struct super_block *sb, uint32_t block, uint16_t partition, uint32_t offset) { int i; struct sparingTable *st = NULL; - uint32_t packet = (block + offset) & ~(UDF_SB_TYPESPAR(sb,partition).s_packet_len - 1); + struct udf_sb_info *sbi = UDF_SB(sb); + struct udf_part_map *map; + uint32_t packet; + struct udf_sparing_data *sdata; + + map = &sbi->s_partmaps[partition]; + sdata = &map->s_type_specific.s_sparing; + packet = (block + offset) & ~(sdata->s_packet_len - 1); for (i = 0; i < 4; i++) { - if (UDF_SB_TYPESPAR(sb,partition).s_spar_map[i] != NULL) { - st = (struct sparingTable *)UDF_SB_TYPESPAR(sb,partition).s_spar_map[i]->b_data; + if (sdata->s_spar_map[i] != NULL) { + st = (struct sparingTable *) + sdata->s_spar_map[i]->b_data; break; } } if (st) { for (i = 0; i < le16_to_cpu(st->reallocationTableLen); i++) { - if (le32_to_cpu(st->mapEntry[i].origLocation) >= 0xFFFFFFF0) { + struct sparingEntry *entry = &st->mapEntry[i]; + u32 origLoc = le32_to_cpu(entry->origLocation); + if (origLoc >= 0xFFFFFFF0) break; - } else if (le32_to_cpu(st->mapEntry[i].origLocation) == packet) { - return le32_to_cpu(st->mapEntry[i].mappedLocation) + - ((block + offset) & (UDF_SB_TYPESPAR(sb,partition).s_packet_len - 1)); - } else if (le32_to_cpu(st->mapEntry[i].origLocation) > packet) { + else if (origLoc == packet) + return le32_to_cpu(entry->mappedLocation) + + ((block + offset) & + (sdata->s_packet_len - 1)); + else if (origLoc > packet) break; - } } } - return UDF_SB_PARTROOT(sb,partition) + block + offset; + return map->s_partition_root + block + offset; } int udf_relocate_blocks(struct super_block *sb, long old_block, long *new_block) @@ -132,69 +153,109 @@ int udf_relocate_blocks(struct super_block *sb, long old_block, long *new_block) struct sparingEntry mapEntry; uint32_t packet; int i, j, k, l; + struct udf_sb_info *sbi = UDF_SB(sb); + u16 reallocationTableLen; + struct buffer_head *bh; - for (i = 0; i < UDF_SB_NUMPARTS(sb); i++) { - if (old_block > UDF_SB_PARTROOT(sb,i) && - old_block < UDF_SB_PARTROOT(sb,i) + UDF_SB_PARTLEN(sb,i)) { - sdata = &UDF_SB_TYPESPAR(sb,i); - packet = (old_block - UDF_SB_PARTROOT(sb,i)) & ~(sdata->s_packet_len - 1); + for (i = 0; i < sbi->s_partitions; i++) { + struct udf_part_map *map = &sbi->s_partmaps[i]; + if (old_block > map->s_partition_root && + old_block < map->s_partition_root + map->s_partition_len) { + sdata = &map->s_type_specific.s_sparing; + packet = (old_block - map->s_partition_root) & + ~(sdata->s_packet_len - 1); - for (j = 0; j < 4; j++) { - if (UDF_SB_TYPESPAR(sb,i).s_spar_map[j] != NULL) { - st = (struct sparingTable *)sdata->s_spar_map[j]->b_data; + for (j = 0; j < 4; j++) + if (sdata->s_spar_map[j] != NULL) { + st = (struct sparingTable *) + sdata->s_spar_map[j]->b_data; break; } - } if (!st) return 1; - for (k = 0; k < le16_to_cpu(st->reallocationTableLen); k++) { - if (le32_to_cpu(st->mapEntry[k].origLocation) == 0xFFFFFFFF) { + reallocationTableLen = + le16_to_cpu(st->reallocationTableLen); + for (k = 0; k < reallocationTableLen; k++) { + struct sparingEntry *entry = &st->mapEntry[k]; + u32 origLoc = le32_to_cpu(entry->origLocation); + + if (origLoc == 0xFFFFFFFF) { for (; j < 4; j++) { - if (sdata->s_spar_map[j]) { - st = (struct sparingTable *)sdata->s_spar_map[j]->b_data; - st->mapEntry[k].origLocation = cpu_to_le32(packet); - udf_update_tag((char *)st, sizeof(struct sparingTable) + le16_to_cpu(st->reallocationTableLen) * sizeof(struct sparingEntry)); - mark_buffer_dirty(sdata->s_spar_map[j]); - } + int len; + bh = sdata->s_spar_map[j]; + if (!bh) + continue; + + st = (struct sparingTable *) + bh->b_data; + entry->origLocation = + cpu_to_le32(packet); + len = + sizeof(struct sparingTable) + + reallocationTableLen * + sizeof(struct sparingEntry); + udf_update_tag((char *)st, len); + mark_buffer_dirty(bh); } - *new_block = le32_to_cpu(st->mapEntry[k].mappedLocation) + - ((old_block - UDF_SB_PARTROOT(sb,i)) & (sdata->s_packet_len - 1)); + *new_block = le32_to_cpu( + entry->mappedLocation) + + ((old_block - + map->s_partition_root) & + (sdata->s_packet_len - 1)); return 0; - } else if (le32_to_cpu(st->mapEntry[k].origLocation) == packet) { - *new_block = le32_to_cpu(st->mapEntry[k].mappedLocation) + - ((old_block - UDF_SB_PARTROOT(sb,i)) & (sdata->s_packet_len - 1)); + } else if (origLoc == packet) { + *new_block = le32_to_cpu( + entry->mappedLocation) + + ((old_block - + map->s_partition_root) & + (sdata->s_packet_len - 1)); return 0; - } else if (le32_to_cpu(st->mapEntry[k].origLocation) > packet) { + } else if (origLoc > packet) break; - } } - for (l = k; l < le16_to_cpu(st->reallocationTableLen); l++) { - if (le32_to_cpu(st->mapEntry[l].origLocation) == 0xFFFFFFFF) { - for (; j < 4; j++) { - if (sdata->s_spar_map[j]) { - st = (struct sparingTable *)sdata->s_spar_map[j]->b_data; - mapEntry = st->mapEntry[l]; - mapEntry.origLocation = cpu_to_le32(packet); - memmove(&st->mapEntry[k + 1], &st->mapEntry[k], (l - k) * sizeof(struct sparingEntry)); - st->mapEntry[k] = mapEntry; - udf_update_tag((char *)st, sizeof(struct sparingTable) + le16_to_cpu(st->reallocationTableLen) * sizeof(struct sparingEntry)); - mark_buffer_dirty(sdata->s_spar_map[j]); - } - } - *new_block = le32_to_cpu(st->mapEntry[k].mappedLocation) + - ((old_block - UDF_SB_PARTROOT(sb,i)) & (sdata->s_packet_len - 1)); - return 0; + for (l = k; l < reallocationTableLen; l++) { + struct sparingEntry *entry = &st->mapEntry[l]; + u32 origLoc = le32_to_cpu(entry->origLocation); + + if (origLoc != 0xFFFFFFFF) + continue; + + for (; j < 4; j++) { + bh = sdata->s_spar_map[j]; + if (!bh) + continue; + + st = (struct sparingTable *)bh->b_data; + mapEntry = st->mapEntry[l]; + mapEntry.origLocation = + cpu_to_le32(packet); + memmove(&st->mapEntry[k + 1], + &st->mapEntry[k], + (l - k) * + sizeof(struct sparingEntry)); + st->mapEntry[k] = mapEntry; + udf_update_tag((char *)st, + sizeof(struct sparingTable) + + reallocationTableLen * + sizeof(struct sparingEntry)); + mark_buffer_dirty(bh); } + *new_block = + le32_to_cpu( + st->mapEntry[k].mappedLocation) + + ((old_block - map->s_partition_root) & + (sdata->s_packet_len - 1)); + return 0; } return 1; } /* if old_block */ } - if (i == UDF_SB_NUMPARTS(sb)) { + if (i == sbi->s_partitions) { /* outside of partitions */ /* for now, fail =) */ return 1; diff --git a/fs/udf/super.c b/fs/udf/super.c index 4360c7a0574..f3ac4abfc94 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -33,8 +33,8 @@ * 10/17/98 added freespace count for "df" * 11/11/98 gr added novrs option * 11/26/98 dgb added fileset,anchor mount options - * 12/06/98 blf really hosed things royally. vat/sparing support. sequenced vol descs - * rewrote option handling based on isofs + * 12/06/98 blf really hosed things royally. vat/sparing support. sequenced + * vol descs. rewrote option handling based on isofs * 12/20/98 find the free space bitmap (if it exists) */ @@ -52,6 +52,9 @@ #include <linux/buffer_head.h> #include <linux/vfs.h> #include <linux/vmalloc.h> +#include <linux/errno.h> +#include <linux/mount.h> +#include <linux/seq_file.h> #include <asm/byteorder.h> #include <linux/udf_fs.h> @@ -70,6 +73,8 @@ #define VDS_POS_TERMINATING_DESC 6 #define VDS_POS_LENGTH 7 +#define UDF_DEFAULT_BLOCKSIZE 2048 + static char error_buf[1024]; /* These are the "meat" - everything else is stuffing */ @@ -94,6 +99,17 @@ static void udf_open_lvid(struct super_block *); static void udf_close_lvid(struct super_block *); static unsigned int udf_count_free(struct super_block *); static int udf_statfs(struct dentry *, struct kstatfs *); +static int udf_show_options(struct seq_file *, struct vfsmount *); + +struct logicalVolIntegrityDescImpUse *udf_sb_lvidiu(struct udf_sb_info *sbi) +{ + struct logicalVolIntegrityDesc *lvid = + (struct logicalVolIntegrityDesc *)sbi->s_lvid_bh->b_data; + __u32 number_of_partitions = le32_to_cpu(lvid->numOfPartitions); + __u32 offset = number_of_partitions * 2 * + sizeof(uint32_t)/sizeof(uint8_t); + return (struct logicalVolIntegrityDescImpUse *)&(lvid->impUse[offset]); +} /* UDF filesystem type */ static int udf_get_sb(struct file_system_type *fs_type, @@ -116,7 +132,7 @@ static struct kmem_cache *udf_inode_cachep; static struct inode *udf_alloc_inode(struct super_block *sb) { struct udf_inode_info *ei; - ei = (struct udf_inode_info *)kmem_cache_alloc(udf_inode_cachep, GFP_KERNEL); + ei = kmem_cache_alloc(udf_inode_cachep, GFP_KERNEL); if (!ei) return NULL; @@ -170,6 +186,7 @@ static const struct super_operations udf_sb_ops = { .write_super = udf_write_super, .statfs = udf_statfs, .remount_fs = udf_remount_fs, + .show_options = udf_show_options, }; struct udf_options { @@ -218,6 +235,79 @@ static void __exit exit_udf_fs(void) module_init(init_udf_fs) module_exit(exit_udf_fs) +static int udf_sb_alloc_partition_maps(struct super_block *sb, u32 count) +{ + struct udf_sb_info *sbi = UDF_SB(sb); + + sbi->s_partmaps = kcalloc(count, sizeof(struct udf_part_map), + GFP_KERNEL); + if (!sbi->s_partmaps) { + udf_error(sb, __FUNCTION__, + "Unable to allocate space for %d partition maps", + count); + sbi->s_partitions = 0; + return -ENOMEM; + } + + sbi->s_partitions = count; + return 0; +} + +static int udf_show_options(struct seq_file *seq, struct vfsmount *mnt) +{ + struct super_block *sb = mnt->mnt_sb; + struct udf_sb_info *sbi = UDF_SB(sb); + + if (!UDF_QUERY_FLAG(sb, UDF_FLAG_STRICT)) + seq_puts(seq, ",nostrict"); + if (sb->s_blocksize != UDF_DEFAULT_BLOCKSIZE) + seq_printf(seq, ",bs=%lu", sb->s_blocksize); + if (UDF_QUERY_FLAG(sb, UDF_FLAG_UNHIDE)) + seq_puts(seq, ",unhide"); + if (UDF_QUERY_FLAG(sb, UDF_FLAG_UNDELETE)) + seq_puts(seq, ",undelete"); + if (!UDF_QUERY_FLAG(sb, UDF_FLAG_USE_AD_IN_ICB)) + seq_puts(seq, ",noadinicb"); + if (UDF_QUERY_FLAG(sb, UDF_FLAG_USE_SHORT_AD)) + seq_puts(seq, ",shortad"); + if (UDF_QUERY_FLAG(sb, UDF_FLAG_UID_FORGET)) + seq_puts(seq, ",uid=forget"); + if (UDF_QUERY_FLAG(sb, UDF_FLAG_UID_IGNORE)) + seq_puts(seq, ",uid=ignore"); + if (UDF_QUERY_FLAG(sb, UDF_FLAG_GID_FORGET)) + seq_puts(seq, ",gid=forget"); + if (UDF_QUERY_FLAG(sb, UDF_FLAG_GID_IGNORE)) + seq_puts(seq, ",gid=ignore"); + if (UDF_QUERY_FLAG(sb, UDF_FLAG_UID_SET)) + seq_printf(seq, ",uid=%u", sbi->s_uid); + if (UDF_QUERY_FLAG(sb, UDF_FLAG_GID_SET)) + seq_printf(seq, ",gid=%u", sbi->s_gid); + if (sbi->s_umask != 0) + seq_printf(seq, ",umask=%o", sbi->s_umask); + if (UDF_QUERY_FLAG(sb, UDF_FLAG_SESSION_SET)) + seq_printf(seq, ",session=%u", sbi->s_session); + if (UDF_QUERY_FLAG(sb, UDF_FLAG_LASTBLOCK_SET)) + seq_printf(seq, ",lastblock=%u", sbi->s_last_block); + /* + * s_anchor[2] could be zeroed out in case there is no anchor + * in the specified block, but then the "anchor=N" option + * originally given by the user wasn't effective, so it's OK + * if we don't show it. + */ + if (sbi->s_anchor[2] != 0) + seq_printf(seq, ",anchor=%u", sbi->s_anchor[2]); + /* + * volume, partition, fileset and rootdir seem to be ignored + * currently + */ + if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) + seq_puts(seq, ",utf8"); + if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP) && sbi->s_nls_map) + seq_printf(seq, ",iocharset=%s", sbi->s_nls_map->charset); + + return 0; +} + /* * udf_parse_options * @@ -310,13 +400,14 @@ static match_table_t tokens = { {Opt_err, NULL} }; -static int udf_parse_options(char *options, struct udf_options *uopt) +static int udf_parse_options(char *options, struct udf_options *uopt, + bool remount) { char *p; int option; uopt->novrs = 0; - uopt->blocksize = 2048; + uopt->blocksize = UDF_DEFAULT_BLOCKSIZE; uopt->partition = 0xFFFF; uopt->session = 0xFFFFFFFF; uopt->lastblock = 0; @@ -386,11 +477,15 @@ static int udf_parse_options(char *options, struct udf_options *uopt) if (match_int(args, &option)) return 0; uopt->session = option; + if (!remount) + uopt->flags |= (1 << UDF_FLAG_SESSION_SET); break; case Opt_lastblock: if (match_int(args, &option)) return 0; uopt->lastblock = option; + if (!remount) + uopt->flags |= (1 << UDF_FLAG_LASTBLOCK_SET); break; case Opt_anchor: if (match_int(args, &option)) @@ -447,7 +542,7 @@ static int udf_parse_options(char *options, struct udf_options *uopt) return 1; } -void udf_write_super(struct super_block *sb) +static void udf_write_super(struct super_block *sb) { lock_kernel(); @@ -461,22 +556,23 @@ void udf_write_super(struct super_block *sb) static int udf_remount_fs(struct super_block *sb, int *flags, char *options) { struct udf_options uopt; + struct udf_sb_info *sbi = UDF_SB(sb); - uopt.flags = UDF_SB(sb)->s_flags; - uopt.uid = UDF_SB(sb)->s_uid; - uopt.gid = UDF_SB(sb)->s_gid; - uopt.umask = UDF_SB(sb)->s_umask; + uopt.flags = sbi->s_flags; + uopt.uid = sbi->s_uid; + uopt.gid = sbi->s_gid; + uopt.umask = sbi->s_umask; - if (!udf_parse_options(options, &uopt)) + if (!udf_parse_options(options, &uopt, true)) return -EINVAL; - UDF_SB(sb)->s_flags = uopt.flags; - UDF_SB(sb)->s_uid = uopt.uid; - UDF_SB(sb)->s_gid = uopt.gid; - UDF_SB(sb)->s_umask = uopt.umask; + sbi->s_flags = uopt.flags; + sbi->s_uid = uopt.uid; + sbi->s_gid = uopt.gid; + sbi->s_umask = uopt.umask; - if (UDF_SB_LVIDBH(sb)) { - int write_rev = le16_to_cpu(UDF_SB_LVIDIU(sb)->minUDFWriteRev); + if (sbi->s_lvid_bh) { + int write_rev = le16_to_cpu(udf_sb_lvidiu(sbi)->minUDFWriteRev); if (write_rev > UDF_MAX_WRITE_VERSION) *flags |= MS_RDONLY; } @@ -538,17 +634,19 @@ static int udf_vrs(struct super_block *sb, int silent) int iso9660 = 0; int nsr02 = 0; int nsr03 = 0; + struct udf_sb_info *sbi; /* Block size must be a multiple of 512 */ if (sb->s_blocksize & 511) return 0; + sbi = UDF_SB(sb); if (sb->s_blocksize < sizeof(struct volStructDesc)) sectorsize = sizeof(struct volStructDesc); else sectorsize = sb->s_blocksize; - sector += (UDF_SB_SESSION(sb) << sb->s_blocksize_bits); + sector += (sbi->s_session << sb->s_blocksize_bits); udf_debug("Starting at sector %u (%ld byte sectors)\n", (sector >> sb->s_blocksize_bits), sb->s_blocksize); @@ -561,47 +659,52 @@ static int udf_vrs(struct super_block *sb, int silent) /* Look for ISO descriptors */ vsd = (struct volStructDesc *)(bh->b_data + - (sector & (sb->s_blocksize - 1))); + (sector & (sb->s_blocksize - 1))); if (vsd->stdIdent[0] == 0) { brelse(bh); break; - } else if (!strncmp(vsd->stdIdent, VSD_STD_ID_CD001, VSD_STD_ID_LEN)) { + } else if (!strncmp(vsd->stdIdent, VSD_STD_ID_CD001, + VSD_STD_ID_LEN)) { iso9660 = sector; switch (vsd->structType) { case 0: udf_debug("ISO9660 Boot Record found\n"); break; case 1: - udf_debug - ("ISO9660 Primary Volume Descriptor found\n"); + udf_debug("ISO9660 Primary Volume Descriptor " + "found\n"); break; case 2: - udf_debug - ("ISO9660 Supplementary Volume Descriptor found\n"); + udf_debug("ISO9660 Supplementary Volume " + "Descriptor found\n"); break; case 3: - udf_debug - ("ISO9660 Volume Partition Descriptor found\n"); + udf_debug("ISO9660 Volume Partition Descriptor " + "found\n"); break; case 255: - udf_debug - ("ISO9660 Volume Descriptor Set Terminator found\n"); + udf_debug("ISO9660 Volume Descriptor Set " + "Terminator found\n"); break; default: udf_debug("ISO9660 VRS (%u) found\n", vsd->structType); break; } - } else if (!strncmp(vsd->stdIdent, VSD_STD_ID_BEA01, VSD_STD_ID_LEN)) { - } else if (!strncmp(vsd->stdIdent, VSD_STD_ID_TEA01, VSD_STD_ID_LEN)) { + } else if (!strncmp(vsd->stdIdent, VSD_STD_ID_BEA01, + VSD_STD_ID_LEN)) + ; /* nothing */ + else if (!strncmp(vsd->stdIdent, VSD_STD_ID_TEA01, + VSD_STD_ID_LEN)) { brelse(bh); break; - } else if (!strncmp(vsd->stdIdent, VSD_STD_ID_NSR02, VSD_STD_ID_LEN)) { + } else if (!strncmp(vsd->stdIdent, VSD_STD_ID_NSR02, + VSD_STD_ID_LEN)) nsr02 = sector; - } else if (!strncmp(vsd->stdIdent, VSD_STD_ID_NSR03, VSD_STD_ID_LEN)) { + else if (!strncmp(vsd->stdIdent, VSD_STD_ID_NSR03, + VSD_STD_ID_LEN)) nsr03 = sector; - } brelse(bh); } @@ -609,7 +712,7 @@ static int udf_vrs(struct super_block *sb, int silent) return nsr03; else if (nsr02) return nsr02; - else if (sector - (UDF_SB_SESSION(sb) << sb->s_blocksize_bits) == 32768) + else if (sector - (sbi->s_session << sb->s_blocksize_bits) == 32768) return -1; else return 0; @@ -634,11 +737,15 @@ static int udf_vrs(struct super_block *sb, int silent) */ static void udf_find_anchor(struct super_block *sb) { - int lastblock = UDF_SB_LASTBLOCK(sb); + int lastblock; struct buffer_head *bh = NULL; uint16_t ident; uint32_t location; int i; + struct udf_sb_info *sbi; + + sbi = UDF_SB(sb); + lastblock = sbi->s_last_block; if (lastblock) { int varlastblock = udf_variable_to_fixed(lastblock); @@ -658,57 +765,83 @@ static void udf_find_anchor(struct super_block *sb) * however, if the disc isn't closed, it could be 512 */ for (i = 0; !lastblock && i < ARRAY_SIZE(last); i++) { - if (last[i] < 0 || !(bh = sb_bread(sb, last[i]))) { - ident = location = 0; - } else { - ident = le16_to_cpu(((tag *)bh->b_data)->tagIdent); - location = le32_to_cpu(((tag *)bh->b_data)->tagLocation); - brelse(bh); + ident = location = 0; + if (last[i] >= 0) { + bh = sb_bread(sb, last[i]); + if (bh) { + tag *t = (tag *)bh->b_data; + ident = le16_to_cpu(t->tagIdent); + location = le32_to_cpu(t->tagLocation); + brelse(bh); + } } if (ident == TAG_IDENT_AVDP) { - if (location == last[i] - UDF_SB_SESSION(sb)) { - lastblock = UDF_SB_ANCHOR(sb)[0] = last[i] - UDF_SB_SESSION(sb); - UDF_SB_ANCHOR(sb)[1] = last[i] - 256 - UDF_SB_SESSION(sb); - } else if (location == udf_variable_to_fixed(last[i]) - UDF_SB_SESSION(sb)) { + if (location == last[i] - sbi->s_session) { + lastblock = last[i] - sbi->s_session; + sbi->s_anchor[0] = lastblock; + sbi->s_anchor[1] = lastblock - 256; + } else if (location == + udf_variable_to_fixed(last[i]) - + sbi->s_session) { UDF_SET_FLAG(sb, UDF_FLAG_VARCONV); - lastblock = UDF_SB_ANCHOR(sb)[0] = udf_variable_to_fixed(last[i]) - UDF_SB_SESSION(sb); - UDF_SB_ANCHOR(sb)[1] = lastblock - 256 - UDF_SB_SESSION(sb); + lastblock = + udf_variable_to_fixed(last[i]) - + sbi->s_session; + sbi->s_anchor[0] = lastblock; + sbi->s_anchor[1] = lastblock - 256 - + sbi->s_session; } else { - udf_debug("Anchor found at block %d, location mismatch %d.\n", + udf_debug("Anchor found at block %d, " + "location mismatch %d.\n", last[i], location); } - } else if (ident == TAG_IDENT_FE || ident == TAG_IDENT_EFE) { + } else if (ident == TAG_IDENT_FE || + ident == TAG_IDENT_EFE) { lastblock = last[i]; - UDF_SB_ANCHOR(sb)[3] = 512; + sbi->s_anchor[3] = 512; } else { - if (last[i] < 256 || !(bh = sb_bread(sb, last[i] - 256))) { - ident = location = 0; - } else { - ident = le16_to_cpu(((tag *)bh->b_data)->tagIdent); - location = le32_to_cpu(((tag *)bh->b_data)->tagLocation); - brelse(bh); + ident = location = 0; + if (last[i] >= 256) { + bh = sb_bread(sb, last[i] - 256); + if (bh) { + tag *t = (tag *)bh->b_data; + ident = le16_to_cpu( + t->tagIdent); + location = le32_to_cpu( + t->tagLocation); + brelse(bh); + } } if (ident == TAG_IDENT_AVDP && - location == last[i] - 256 - UDF_SB_SESSION(sb)) { + location == last[i] - 256 - + sbi->s_session) { lastblock = last[i]; - UDF_SB_ANCHOR(sb)[1] = last[i] - 256; + sbi->s_anchor[1] = last[i] - 256; } else { - if (last[i] < 312 + UDF_SB_SESSION(sb) || - !(bh = sb_bread(sb, last[i] - 312 - UDF_SB_SESSION(sb)))) { - ident = location = 0; - } else { - ident = le16_to_cpu(((tag *)bh->b_data)->tagIdent); - location = le32_to_cpu(((tag *)bh->b_data)->tagLocation); - brelse(bh); + ident = location = 0; + if (last[i] >= 312 + sbi->s_session) { + bh = sb_bread(sb, + last[i] - 312 - + sbi->s_session); + if (bh) { + tag *t = (tag *) + bh->b_data; + ident = le16_to_cpu( + t->tagIdent); + location = le32_to_cpu( + t->tagLocation); + brelse(bh); + } } if (ident == TAG_IDENT_AVDP && location == udf_variable_to_fixed(last[i]) - 256) { - UDF_SET_FLAG(sb, UDF_FLAG_VARCONV); + UDF_SET_FLAG(sb, + UDF_FLAG_VARCONV); lastblock = udf_variable_to_fixed(last[i]); - UDF_SB_ANCHOR(sb)[1] = lastblock - 256; + sbi->s_anchor[1] = lastblock - 256; } } } @@ -716,10 +849,12 @@ static void udf_find_anchor(struct super_block *sb) } if (!lastblock) { - /* We havn't found the lastblock. check 312 */ - if ((bh = sb_bread(sb, 312 + UDF_SB_SESSION(sb)))) { - ident = le16_to_cpu(((tag *)bh->b_data)->tagIdent); - location = le32_to_cpu(((tag *)bh->b_data)->tagLocation); + /* We haven't found the lastblock. check 312 */ + bh = sb_bread(sb, 312 + sbi->s_session); + if (bh) { + tag *t = (tag *)bh->b_data; + ident = le16_to_cpu(t->tagIdent); + location = le32_to_cpu(t->tagLocation); brelse(bh); if (ident == TAG_IDENT_AVDP && location == 256) @@ -727,29 +862,33 @@ static void udf_find_anchor(struct super_block *sb) } } - for (i = 0; i < ARRAY_SIZE(UDF_SB_ANCHOR(sb)); i++) { - if (UDF_SB_ANCHOR(sb)[i]) { - if (!(bh = udf_read_tagged(sb, UDF_SB_ANCHOR(sb)[i], - UDF_SB_ANCHOR(sb)[i], &ident))) { - UDF_SB_ANCHOR(sb)[i] = 0; - } else { + for (i = 0; i < ARRAY_SIZE(sbi->s_anchor); i++) { + if (sbi->s_anchor[i]) { + bh = udf_read_tagged(sb, sbi->s_anchor[i], + sbi->s_anchor[i], &ident); + if (!bh) + sbi->s_anchor[i] = 0; + else { brelse(bh); if ((ident != TAG_IDENT_AVDP) && - (i || (ident != TAG_IDENT_FE && ident != TAG_IDENT_EFE))) { - UDF_SB_ANCHOR(sb)[i] = 0; - } + (i || (ident != TAG_IDENT_FE && + ident != TAG_IDENT_EFE))) + sbi->s_anchor[i] = 0; } } } - UDF_SB_LASTBLOCK(sb) = lastblock; + sbi->s_last_block = lastblock; } -static int udf_find_fileset(struct super_block *sb, kernel_lb_addr *fileset, kernel_lb_addr *root) +static int udf_find_fileset(struct super_block *sb, + kernel_lb_addr *fileset, + kernel_lb_addr *root) { struct buffer_head *bh = NULL; long lastblock; uint16_t ident; + struct udf_sb_info *sbi; if (fileset->logicalBlockNum != 0xFFFFFFFF || fileset->partitionReferenceNum != 0xFFFF) { @@ -764,22 +903,27 @@ static int udf_find_fileset(struct super_block *sb, kernel_lb_addr *fileset, ker } - if (!bh) { /* Search backwards through the partitions */ + sbi = UDF_SB(sb); + if (!bh) { + /* Search backwards through the partitions */ kernel_lb_addr newfileset; /* --> cvg: FIXME - is it reasonable? */ return 1; - for (newfileset.partitionReferenceNum = UDF_SB_NUMPARTS(sb) - 1; + for (newfileset.partitionReferenceNum = sbi->s_partitions - 1; (newfileset.partitionReferenceNum != 0xFFFF && fileset->logicalBlockNum == 0xFFFFFFFF && fileset->partitionReferenceNum == 0xFFFF); newfileset.partitionReferenceNum--) { - lastblock = UDF_SB_PARTLEN(sb, newfileset.partitionReferenceNum); + lastblock = sbi->s_partmaps + [newfileset.partitionReferenceNum] + .s_partition_len; newfileset.logicalBlockNum = 0; do { - bh = udf_read_ptagged(sb, newfileset, 0, &ident); + bh = udf_read_ptagged(sb, newfileset, 0, + &ident); if (!bh) { newfileset.logicalBlockNum++; continue; @@ -789,11 +933,12 @@ static int udf_find_fileset(struct super_block *sb, kernel_lb_addr *fileset, ker case TAG_IDENT_SBD: { struct spaceBitmapDesc *sp; - sp = (struct spaceBitmapDesc *)bh->b_data; + sp = (struct spaceBitmapDesc *) + bh->b_data; newfileset.logicalBlockNum += 1 + ((le32_to_cpu(sp->numOfBytes) + - sizeof(struct spaceBitmapDesc) - 1) - >> sb->s_blocksize_bits); + sizeof(struct spaceBitmapDesc) + - 1) >> sb->s_blocksize_bits); brelse(bh); break; } @@ -818,7 +963,7 @@ static int udf_find_fileset(struct super_block *sb, kernel_lb_addr *fileset, ker fileset->logicalBlockNum, fileset->partitionReferenceNum); - UDF_SB_PARTITION(sb) = fileset->partitionReferenceNum; + sbi->s_partition = fileset->partitionReferenceNum; udf_load_fileset(sb, bh, root); brelse(bh); return 0; @@ -840,26 +985,26 @@ static void udf_load_pvoldesc(struct super_block *sb, struct buffer_head *bh) lets_to_cpu(pvoldesc->recordingDateAndTime))) { kernel_timestamp ts; ts = lets_to_cpu(pvoldesc->recordingDateAndTime); - udf_debug("recording time %ld/%ld, %04u/%02u/%02u %02u:%02u (%x)\n", + udf_debug("recording time %ld/%ld, %04u/%02u/%02u" + " %02u:%02u (%x)\n", recording, recording_usec, ts.year, ts.month, ts.day, ts.hour, ts.minute, ts.typeAndTimezone); - UDF_SB_RECORDTIME(sb).tv_sec = recording; - UDF_SB_RECORDTIME(sb).tv_nsec = recording_usec * 1000; + UDF_SB(sb)->s_record_time.tv_sec = recording; + UDF_SB(sb)->s_record_time.tv_nsec = recording_usec * 1000; } - if (!udf_build_ustr(&instr, pvoldesc->volIdent, 32)) { + if (!udf_build_ustr(&instr, pvoldesc->volIdent, 32)) if (udf_CS0toUTF8(&outstr, &instr)) { - strncpy(UDF_SB_VOLIDENT(sb), outstr.u_name, + strncpy(UDF_SB(sb)->s_volume_ident, outstr.u_name, outstr.u_len > 31 ? 31 : outstr.u_len); - udf_debug("volIdent[] = '%s'\n", UDF_SB_VOLIDENT(sb)); + udf_debug("volIdent[] = '%s'\n", + UDF_SB(sb)->s_volume_ident); } - } - if (!udf_build_ustr(&instr, pvoldesc->volSetIdent, 128)) { + if (!udf_build_ustr(&instr, pvoldesc->volSetIdent, 128)) if (udf_CS0toUTF8(&outstr, &instr)) udf_debug("volSetIdent[] = '%s'\n", outstr.u_name); - } } static void udf_load_fileset(struct super_block *sb, struct buffer_head *bh, @@ -871,65 +1016,124 @@ static void udf_load_fileset(struct super_block *sb, struct buffer_head *bh, *root = lelb_to_cpu(fset->rootDirectoryICB.extLocation); - UDF_SB_SERIALNUM(sb) = le16_to_cpu(fset->descTag.tagSerialNum); + UDF_SB(sb)->s_serial_number = le16_to_cpu(fset->descTag.tagSerialNum); udf_debug("Rootdir at block=%d, partition=%d\n", root->logicalBlockNum, root->partitionReferenceNum); } +int udf_compute_nr_groups(struct super_block *sb, u32 partition) +{ + struct udf_part_map *map = &UDF_SB(sb)->s_partmaps[partition]; + return (map->s_partition_len + + (sizeof(struct spaceBitmapDesc) << 3) + + (sb->s_blocksize * 8) - 1) / + (sb->s_blocksize * 8); +} + +static struct udf_bitmap *udf_sb_alloc_bitmap(struct super_block *sb, u32 index) +{ + struct udf_bitmap *bitmap; + int nr_groups; + int size; + + nr_groups = udf_compute_nr_groups(sb, index); + size = sizeof(struct udf_bitmap) + + (sizeof(struct buffer_head *) * nr_groups); + + if (size <= PAGE_SIZE) + bitmap = kmalloc(size, GFP_KERNEL); + else + bitmap = vmalloc(size); /* TODO: get rid of vmalloc */ + + if (bitmap == NULL) { + udf_error(sb, __FUNCTION__, + "Unable to allocate space for bitmap " + "and %d buffer_head pointers", nr_groups); + return NULL; + } + + memset(bitmap, 0x00, size); + bitmap->s_block_bitmap = (struct buffer_head **)(bitmap + 1); + bitmap->s_nr_groups = nr_groups; + return bitmap; +} + static int udf_load_partdesc(struct super_block *sb, struct buffer_head *bh) { struct partitionDesc *p; int i; + struct udf_part_map *map; + struct udf_sb_info *sbi; p = (struct partitionDesc *)bh->b_data; + sbi = UDF_SB(sb); - for (i = 0; i < UDF_SB_NUMPARTS(sb); i++) { + for (i = 0; i < sbi->s_partitions; i++) { + map = &sbi->s_partmaps[i]; udf_debug("Searching map: (%d == %d)\n", - UDF_SB_PARTMAPS(sb)[i].s_partition_num, le16_to_cpu(p->partitionNumber)); - if (UDF_SB_PARTMAPS(sb)[i].s_partition_num == le16_to_cpu(p->partitionNumber)) { - UDF_SB_PARTLEN(sb,i) = le32_to_cpu(p->partitionLength); /* blocks */ - UDF_SB_PARTROOT(sb,i) = le32_to_cpu(p->partitionStartingLocation); - if (le32_to_cpu(p->accessType) == PD_ACCESS_TYPE_READ_ONLY) - UDF_SB_PARTFLAGS(sb,i) |= UDF_PART_FLAG_READ_ONLY; - if (le32_to_cpu(p->accessType) == PD_ACCESS_TYPE_WRITE_ONCE) - UDF_SB_PARTFLAGS(sb,i) |= UDF_PART_FLAG_WRITE_ONCE; - if (le32_to_cpu(p->accessType) == PD_ACCESS_TYPE_REWRITABLE) - UDF_SB_PARTFLAGS(sb,i) |= UDF_PART_FLAG_REWRITABLE; - if (le32_to_cpu(p->accessType) == PD_ACCESS_TYPE_OVERWRITABLE) - UDF_SB_PARTFLAGS(sb,i) |= UDF_PART_FLAG_OVERWRITABLE; - - if (!strcmp(p->partitionContents.ident, PD_PARTITION_CONTENTS_NSR02) || - !strcmp(p->partitionContents.ident, PD_PARTITION_CONTENTS_NSR03)) { + map->s_partition_num, + le16_to_cpu(p->partitionNumber)); + if (map->s_partition_num == + le16_to_cpu(p->partitionNumber)) { + map->s_partition_len = + le32_to_cpu(p->partitionLength); /* blocks */ + map->s_partition_root = + le32_to_cpu(p->partitionStartingLocation); + if (p->accessType == + cpu_to_le32(PD_ACCESS_TYPE_READ_ONLY)) + map->s_partition_flags |= + UDF_PART_FLAG_READ_ONLY; + if (p->accessType == + cpu_to_le32(PD_ACCESS_TYPE_WRITE_ONCE)) + map->s_partition_flags |= + UDF_PART_FLAG_WRITE_ONCE; + if (p->accessType == + cpu_to_le32(PD_ACCESS_TYPE_REWRITABLE)) + map->s_partition_flags |= + UDF_PART_FLAG_REWRITABLE; + if (p->accessType == + cpu_to_le32(PD_ACCESS_TYPE_OVERWRITABLE)) + map->s_partition_flags |= + UDF_PART_FLAG_OVERWRITABLE; + + if (!strcmp(p->partitionContents.ident, + PD_PARTITION_CONTENTS_NSR02) || + !strcmp(p->partitionContents.ident, + PD_PARTITION_CONTENTS_NSR03)) { struct partitionHeaderDesc *phd; - phd = (struct partitionHeaderDesc *)(p->partitionContentsUse); + phd = (struct partitionHeaderDesc *) + (p->partitionContentsUse); if (phd->unallocSpaceTable.extLength) { kernel_lb_addr loc = { .logicalBlockNum = le32_to_cpu(phd->unallocSpaceTable.extPosition), .partitionReferenceNum = i, }; - UDF_SB_PARTMAPS(sb)[i].s_uspace.s_table = + map->s_uspace.s_table = udf_iget(sb, loc); - if (!UDF_SB_PARTMAPS(sb)[i].s_uspace.s_table) { + if (!map->s_uspace.s_table) { udf_debug("cannot load unallocSpaceTable (part %d)\n", i); return 1; } - UDF_SB_PARTFLAGS(sb,i) |= UDF_PART_FLAG_UNALLOC_TABLE; + map->s_partition_flags |= + UDF_PART_FLAG_UNALLOC_TABLE; udf_debug("unallocSpaceTable (part %d) @ %ld\n", - i, UDF_SB_PARTMAPS(sb)[i].s_uspace.s_table->i_ino); + i, map->s_uspace.s_table->i_ino); } if (phd->unallocSpaceBitmap.extLength) { - UDF_SB_ALLOC_BITMAP(sb, i, s_uspace); - if (UDF_SB_PARTMAPS(sb)[i].s_uspace.s_bitmap != NULL) { - UDF_SB_PARTMAPS(sb)[i].s_uspace.s_bitmap->s_extLength = + struct udf_bitmap *bitmap = + udf_sb_alloc_bitmap(sb, i); + map->s_uspace.s_bitmap = bitmap; + if (bitmap != NULL) { + bitmap->s_extLength = le32_to_cpu(phd->unallocSpaceBitmap.extLength); - UDF_SB_PARTMAPS(sb)[i].s_uspace.s_bitmap->s_extPosition = + bitmap->s_extPosition = le32_to_cpu(phd->unallocSpaceBitmap.extPosition); - UDF_SB_PARTFLAGS(sb,i) |= UDF_PART_FLAG_UNALLOC_BITMAP; + map->s_partition_flags |= UDF_PART_FLAG_UNALLOC_BITMAP; udf_debug("unallocSpaceBitmap (part %d) @ %d\n", - i, UDF_SB_PARTMAPS(sb)[i].s_uspace.s_bitmap->s_extPosition); + i, bitmap->s_extPosition); } } if (phd->partitionIntegrityTable.extLength) @@ -940,40 +1144,45 @@ static int udf_load_partdesc(struct super_block *sb, struct buffer_head *bh) .partitionReferenceNum = i, }; - UDF_SB_PARTMAPS(sb)[i].s_fspace.s_table = + map->s_fspace.s_table = udf_iget(sb, loc); - if (!UDF_SB_PARTMAPS(sb)[i].s_fspace.s_table) { + if (!map->s_fspace.s_table) { udf_debug("cannot load freedSpaceTable (part %d)\n", i); return 1; } - UDF_SB_PARTFLAGS(sb,i) |= UDF_PART_FLAG_FREED_TABLE; + map->s_partition_flags |= + UDF_PART_FLAG_FREED_TABLE; udf_debug("freedSpaceTable (part %d) @ %ld\n", - i, UDF_SB_PARTMAPS(sb)[i].s_fspace.s_table->i_ino); + i, map->s_fspace.s_table->i_ino); } if (phd->freedSpaceBitmap.extLength) { - UDF_SB_ALLOC_BITMAP(sb, i, s_fspace); - if (UDF_SB_PARTMAPS(sb)[i].s_fspace.s_bitmap != NULL) { - UDF_SB_PARTMAPS(sb)[i].s_fspace.s_bitmap->s_extLength = + struct udf_bitmap *bitmap = + udf_sb_alloc_bitmap(sb, i); + map->s_fspace.s_bitmap = bitmap; + if (bitmap != NULL) { + bitmap->s_extLength = le32_to_cpu(phd->freedSpaceBitmap.extLength); - UDF_SB_PARTMAPS(sb)[i].s_fspace.s_bitmap->s_extPosition = + bitmap->s_extPosition = le32_to_cpu(phd->freedSpaceBitmap.extPosition); - UDF_SB_PARTFLAGS(sb,i) |= UDF_PART_FLAG_FREED_BITMAP; + map->s_partition_flags |= UDF_PART_FLAG_FREED_BITMAP; udf_debug("freedSpaceBitmap (part %d) @ %d\n", - i, UDF_SB_PARTMAPS(sb)[i].s_fspace.s_bitmap->s_extPosition); + i, bitmap->s_extPosition); } } } break; } } - if (i == UDF_SB_NUMPARTS(sb)) { + if (i == sbi->s_partitions) udf_debug("Partition (%d) not found in partition map\n", le16_to_cpu(p->partitionNumber)); - } else { - udf_debug("Partition (%d:%d type %x) starts at physical %d, block length %d\n", - le16_to_cpu(p->partitionNumber), i, UDF_SB_PARTTYPE(sb,i), - UDF_SB_PARTROOT(sb,i), UDF_SB_PARTLEN(sb,i)); - } + else + udf_debug("Partition (%d:%d type %x) starts at physical %d, " + "block length %d\n", + le16_to_cpu(p->partitionNumber), i, + map->s_partition_type, + map->s_partition_root, + map->s_partition_len); return 0; } @@ -983,70 +1192,105 @@ static int udf_load_logicalvol(struct super_block *sb, struct buffer_head *bh, struct logicalVolDesc *lvd; int i, j, offset; uint8_t type; + struct udf_sb_info *sbi = UDF_SB(sb); + struct genericPartitionMap *gpm; lvd = (struct logicalVolDesc *)bh->b_data; - UDF_SB_ALLOC_PARTMAPS(sb, le32_to_cpu(lvd->numPartitionMaps)); + i = udf_sb_alloc_partition_maps(sb, le32_to_cpu(lvd->numPartitionMaps)); + if (i != 0) + return i; for (i = 0, offset = 0; - i < UDF_SB_NUMPARTS(sb) && offset < le32_to_cpu(lvd->mapTableLength); - i++, offset += ((struct genericPartitionMap *)&(lvd->partitionMaps[offset]))->partitionMapLength) { - type = ((struct genericPartitionMap *)&(lvd->partitionMaps[offset]))->partitionMapType; + i < sbi->s_partitions && offset < le32_to_cpu(lvd->mapTableLength); + i++, offset += gpm->partitionMapLength) { + struct udf_part_map *map = &sbi->s_partmaps[i]; + gpm = (struct genericPartitionMap *) + &(lvd->partitionMaps[offset]); + type = gpm->partitionMapType; if (type == 1) { - struct genericPartitionMap1 *gpm1 = (struct genericPartitionMap1 *)&(lvd->partitionMaps[offset]); - UDF_SB_PARTTYPE(sb,i) = UDF_TYPE1_MAP15; - UDF_SB_PARTVSN(sb,i) = le16_to_cpu(gpm1->volSeqNum); - UDF_SB_PARTNUM(sb,i) = le16_to_cpu(gpm1->partitionNum); - UDF_SB_PARTFUNC(sb,i) = NULL; + struct genericPartitionMap1 *gpm1 = + (struct genericPartitionMap1 *)gpm; + map->s_partition_type = UDF_TYPE1_MAP15; + map->s_volumeseqnum = le16_to_cpu(gpm1->volSeqNum); + map->s_partition_num = le16_to_cpu(gpm1->partitionNum); + map->s_partition_func = NULL; } else if (type == 2) { - struct udfPartitionMap2 *upm2 = (struct udfPartitionMap2 *)&(lvd->partitionMaps[offset]); - if (!strncmp(upm2->partIdent.ident, UDF_ID_VIRTUAL, strlen(UDF_ID_VIRTUAL))) { - if (le16_to_cpu(((__le16 *)upm2->partIdent.identSuffix)[0]) == 0x0150) { - UDF_SB_PARTTYPE(sb,i) = UDF_VIRTUAL_MAP15; - UDF_SB_PARTFUNC(sb,i) = udf_get_pblock_virt15; - } else if (le16_to_cpu(((__le16 *)upm2->partIdent.identSuffix)[0]) == 0x0200) { - UDF_SB_PARTTYPE(sb,i) = UDF_VIRTUAL_MAP20; - UDF_SB_PARTFUNC(sb,i) = udf_get_pblock_virt20; + struct udfPartitionMap2 *upm2 = + (struct udfPartitionMap2 *)gpm; + if (!strncmp(upm2->partIdent.ident, UDF_ID_VIRTUAL, + strlen(UDF_ID_VIRTUAL))) { + u16 suf = + le16_to_cpu(((__le16 *)upm2->partIdent. + identSuffix)[0]); + if (suf == 0x0150) { + map->s_partition_type = + UDF_VIRTUAL_MAP15; + map->s_partition_func = + udf_get_pblock_virt15; + } else if (suf == 0x0200) { + map->s_partition_type = + UDF_VIRTUAL_MAP20; + map->s_partition_func = + udf_get_pblock_virt20; } - } else if (!strncmp(upm2->partIdent.ident, UDF_ID_SPARABLE, strlen(UDF_ID_SPARABLE))) { + } else if (!strncmp(upm2->partIdent.ident, + UDF_ID_SPARABLE, + strlen(UDF_ID_SPARABLE))) { uint32_t loc; uint16_t ident; struct sparingTable *st; - struct sparablePartitionMap *spm = (struct sparablePartitionMap *)&(lvd->partitionMaps[offset]); + struct sparablePartitionMap *spm = + (struct sparablePartitionMap *)gpm; - UDF_SB_PARTTYPE(sb,i) = UDF_SPARABLE_MAP15; - UDF_SB_TYPESPAR(sb,i).s_packet_len = le16_to_cpu(spm->packetLength); + map->s_partition_type = UDF_SPARABLE_MAP15; + map->s_type_specific.s_sparing.s_packet_len = + le16_to_cpu(spm->packetLength); for (j = 0; j < spm->numSparingTables; j++) { - loc = le32_to_cpu(spm->locSparingTable[j]); - UDF_SB_TYPESPAR(sb,i).s_spar_map[j] = - udf_read_tagged(sb, loc, loc, &ident); - if (UDF_SB_TYPESPAR(sb,i).s_spar_map[j] != NULL) { - st = (struct sparingTable *)UDF_SB_TYPESPAR(sb,i).s_spar_map[j]->b_data; - if (ident != 0 || - strncmp(st->sparingIdent.ident, UDF_ID_SPARING, strlen(UDF_ID_SPARING))) { - brelse(UDF_SB_TYPESPAR(sb,i).s_spar_map[j]); - UDF_SB_TYPESPAR(sb,i).s_spar_map[j] = NULL; + struct buffer_head *bh2; + + loc = le32_to_cpu( + spm->locSparingTable[j]); + bh2 = udf_read_tagged(sb, loc, loc, + &ident); + map->s_type_specific.s_sparing. + s_spar_map[j] = bh2; + + if (bh2 != NULL) { + st = (struct sparingTable *) + bh2->b_data; + if (ident != 0 || strncmp( + st->sparingIdent.ident, + UDF_ID_SPARING, + strlen(UDF_ID_SPARING))) { + brelse(bh2); + map->s_type_specific. + s_sparing. + s_spar_map[j] = + NULL; } } } - UDF_SB_PARTFUNC(sb,i) = udf_get_pblock_spar15; + map->s_partition_func = udf_get_pblock_spar15; } else { - udf_debug("Unknown ident: %s\n", upm2->partIdent.ident); + udf_debug("Unknown ident: %s\n", + upm2->partIdent.ident); continue; } - UDF_SB_PARTVSN(sb,i) = le16_to_cpu(upm2->volSeqNum); - UDF_SB_PARTNUM(sb,i) = le16_to_cpu(upm2->partitionNum); + map->s_volumeseqnum = le16_to_cpu(upm2->volSeqNum); + map->s_partition_num = le16_to_cpu(upm2->partitionNum); } udf_debug("Partition (%d:%d) type %d on volume %d\n", - i, UDF_SB_PARTNUM(sb,i), type, UDF_SB_PARTVSN(sb,i)); + i, map->s_partition_num, type, + map->s_volumeseqnum); } if (fileset) { long_ad *la = (long_ad *)&(lvd->logicalVolContentsUse[0]); *fileset = lelb_to_cpu(la->extLocation); - udf_debug("FileSet found in LogicalVolDesc at block=%d, partition=%d\n", - fileset->logicalBlockNum, + udf_debug("FileSet found in LogicalVolDesc at block=%d, " + "partition=%d\n", fileset->logicalBlockNum, fileset->partitionReferenceNum); } if (lvd->integritySeqExt.extLength) @@ -1063,22 +1307,26 @@ static void udf_load_logicalvolint(struct super_block *sb, kernel_extent_ad loc) { struct buffer_head *bh = NULL; uint16_t ident; + struct udf_sb_info *sbi = UDF_SB(sb); + struct logicalVolIntegrityDesc *lvid; while (loc.extLength > 0 && (bh = udf_read_tagged(sb, loc.extLocation, loc.extLocation, &ident)) && ident == TAG_IDENT_LVID) { - UDF_SB_LVIDBH(sb) = bh; + sbi->s_lvid_bh = bh; + lvid = (struct logicalVolIntegrityDesc *)bh->b_data; - if (UDF_SB_LVID(sb)->nextIntegrityExt.extLength) - udf_load_logicalvolint(sb, leea_to_cpu(UDF_SB_LVID(sb)->nextIntegrityExt)); + if (lvid->nextIntegrityExt.extLength) + udf_load_logicalvolint(sb, + leea_to_cpu(lvid->nextIntegrityExt)); - if (UDF_SB_LVIDBH(sb) != bh) + if (sbi->s_lvid_bh != bh) brelse(bh); loc.extLength -= sb->s_blocksize; loc.extLocation++; } - if (UDF_SB_LVIDBH(sb) != bh) + if (sbi->s_lvid_bh != bh) brelse(bh); } @@ -1097,11 +1345,12 @@ static void udf_load_logicalvolint(struct super_block *sb, kernel_extent_ad loc) * July 1, 1997 - Andrew E. Mileski * Written, tested, and released. */ -static int udf_process_sequence(struct super_block *sb, long block, long lastblock, - kernel_lb_addr *fileset) +static int udf_process_sequence(struct super_block *sb, long block, + long lastblock, kernel_lb_addr *fileset) { struct buffer_head *bh = NULL; struct udf_vds_record vds[VDS_POS_LENGTH]; + struct udf_vds_record *curr; struct generic_desc *gd; struct volDescPtr *vdp; int done = 0; @@ -1124,43 +1373,51 @@ static int udf_process_sequence(struct super_block *sb, long block, long lastblo vdsn = le32_to_cpu(gd->volDescSeqNum); switch (ident) { case TAG_IDENT_PVD: /* ISO 13346 3/10.1 */ - if (vdsn >= vds[VDS_POS_PRIMARY_VOL_DESC].volDescSeqNum) { - vds[VDS_POS_PRIMARY_VOL_DESC].volDescSeqNum = vdsn; - vds[VDS_POS_PRIMARY_VOL_DESC].block = block; + curr = &vds[VDS_POS_PRIMARY_VOL_DESC]; + if (vdsn >= curr->volDescSeqNum) { + curr->volDescSeqNum = vdsn; + curr->block = block; } break; case TAG_IDENT_VDP: /* ISO 13346 3/10.3 */ - if (vdsn >= vds[VDS_POS_VOL_DESC_PTR].volDescSeqNum) { - vds[VDS_POS_VOL_DESC_PTR].volDescSeqNum = vdsn; - vds[VDS_POS_VOL_DESC_PTR].block = block; + curr = &vds[VDS_POS_VOL_DESC_PTR]; + if (vdsn >= curr->volDescSeqNum) { + curr->volDescSeqNum = vdsn; + curr->block = block; vdp = (struct volDescPtr *)bh->b_data; - next_s = le32_to_cpu(vdp->nextVolDescSeqExt.extLocation); - next_e = le32_to_cpu(vdp->nextVolDescSeqExt.extLength); + next_s = le32_to_cpu( + vdp->nextVolDescSeqExt.extLocation); + next_e = le32_to_cpu( + vdp->nextVolDescSeqExt.extLength); next_e = next_e >> sb->s_blocksize_bits; next_e += next_s; } break; case TAG_IDENT_IUVD: /* ISO 13346 3/10.4 */ - if (vdsn >= vds[VDS_POS_IMP_USE_VOL_DESC].volDescSeqNum) { - vds[VDS_POS_IMP_USE_VOL_DESC].volDescSeqNum = vdsn; - vds[VDS_POS_IMP_USE_VOL_DESC].block = block; + curr = &vds[VDS_POS_IMP_USE_VOL_DESC]; + if (vdsn >= curr->volDescSeqNum) { + curr->volDescSeqNum = vdsn; + curr->block = block; } break; case TAG_IDENT_PD: /* ISO 13346 3/10.5 */ - if (!vds[VDS_POS_PARTITION_DESC].block) - vds[VDS_POS_PARTITION_DESC].block = block; + curr = &vds[VDS_POS_PARTITION_DESC]; + if (!curr->block) + curr->block = block; break; case TAG_IDENT_LVD: /* ISO 13346 3/10.6 */ - if (vdsn >= vds[VDS_POS_LOGICAL_VOL_DESC].volDescSeqNum) { - vds[VDS_POS_LOGICAL_VOL_DESC].volDescSeqNum = vdsn; - vds[VDS_POS_LOGICAL_VOL_DESC].block = block; + curr = &vds[VDS_POS_LOGICAL_VOL_DESC]; + if (vdsn >= curr->volDescSeqNum) { + curr->volDescSeqNum = vdsn; + curr->block = block; } break; case TAG_IDENT_USD: /* ISO 13346 3/10.8 */ - if (vdsn >= vds[VDS_POS_UNALLOC_SPACE_DESC].volDescSeqNum) { - vds[VDS_POS_UNALLOC_SPACE_DESC].volDescSeqNum = vdsn; - vds[VDS_POS_UNALLOC_SPACE_DESC].block = block; + curr = &vds[VDS_POS_UNALLOC_SPACE_DESC]; + if (vdsn >= curr->volDescSeqNum) { + curr->volDescSeqNum = vdsn; + curr->block = block; } break; case TAG_IDENT_TD: /* ISO 13346 3/10.9 */ @@ -1169,32 +1426,38 @@ static int udf_process_sequence(struct super_block *sb, long block, long lastblo block = next_s; lastblock = next_e; next_s = next_e = 0; - } else { + } else done = 1; - } break; } brelse(bh); } for (i = 0; i < VDS_POS_LENGTH; i++) { if (vds[i].block) { - bh = udf_read_tagged(sb, vds[i].block, vds[i].block, &ident); + bh = udf_read_tagged(sb, vds[i].block, vds[i].block, + &ident); if (i == VDS_POS_PRIMARY_VOL_DESC) { udf_load_pvoldesc(sb, bh); } else if (i == VDS_POS_LOGICAL_VOL_DESC) { - udf_load_logicalvol(sb, bh, fileset); + if (udf_load_logicalvol(sb, bh, fileset)) { + brelse(bh); + return 1; + } } else if (i == VDS_POS_PARTITION_DESC) { struct buffer_head *bh2 = NULL; if (udf_load_partdesc(sb, bh)) { brelse(bh); return 1; } - for (j = vds[i].block + 1; j < vds[VDS_POS_TERMINATING_DESC].block; j++) { + for (j = vds[i].block + 1; + j < vds[VDS_POS_TERMINATING_DESC].block; + j++) { bh2 = udf_read_tagged(sb, j, j, &ident); gd = (struct generic_desc *)bh2->b_data; if (ident == TAG_IDENT_PD) - if (udf_load_partdesc(sb, bh2)) { + if (udf_load_partdesc(sb, + bh2)) { brelse(bh); brelse(bh2); return 1; @@ -1222,14 +1485,17 @@ static int udf_check_valid(struct super_block *sb, int novrs, int silent) } /* Check that it is NSR02 compliant */ /* Process any "CD-ROM Volume Descriptor Set" (ECMA 167 2/8.3.1) */ - else if ((block = udf_vrs(sb, silent)) == -1) { - udf_debug("Failed to read byte 32768. Assuming open disc. " - "Skipping validity check\n"); - if (!UDF_SB_LASTBLOCK(sb)) - UDF_SB_LASTBLOCK(sb) = udf_get_last_block(sb); - return 0; - } else { - return !block; + else { + block = udf_vrs(sb, silent); + if (block == -1) { + struct udf_sb_info *sbi = UDF_SB(sb); + udf_debug("Failed to read byte 32768. Assuming open " + "disc. Skipping validity check\n"); + if (!sbi->s_last_block) + sbi->s_last_block = udf_get_last_block(sb); + return 0; + } else + return !block; } } @@ -1240,100 +1506,121 @@ static int udf_load_partition(struct super_block *sb, kernel_lb_addr *fileset) struct buffer_head *bh; long main_s, main_e, reserve_s, reserve_e; int i, j; + struct udf_sb_info *sbi; if (!sb) return 1; + sbi = UDF_SB(sb); - for (i = 0; i < ARRAY_SIZE(UDF_SB_ANCHOR(sb)); i++) { - if (UDF_SB_ANCHOR(sb)[i] && - (bh = udf_read_tagged(sb, UDF_SB_ANCHOR(sb)[i], - UDF_SB_ANCHOR(sb)[i], &ident))) { - anchor = (struct anchorVolDescPtr *)bh->b_data; + for (i = 0; i < ARRAY_SIZE(sbi->s_anchor); i++) { + if (!sbi->s_anchor[i]) + continue; + bh = udf_read_tagged(sb, sbi->s_anchor[i], sbi->s_anchor[i], + &ident); + if (!bh) + continue; - /* Locate the main sequence */ - main_s = le32_to_cpu(anchor->mainVolDescSeqExt.extLocation); - main_e = le32_to_cpu(anchor->mainVolDescSeqExt.extLength ); - main_e = main_e >> sb->s_blocksize_bits; - main_e += main_s; + anchor = (struct anchorVolDescPtr *)bh->b_data; - /* Locate the reserve sequence */ - reserve_s = le32_to_cpu(anchor->reserveVolDescSeqExt.extLocation); - reserve_e = le32_to_cpu(anchor->reserveVolDescSeqExt.extLength); - reserve_e = reserve_e >> sb->s_blocksize_bits; - reserve_e += reserve_s; + /* Locate the main sequence */ + main_s = le32_to_cpu(anchor->mainVolDescSeqExt.extLocation); + main_e = le32_to_cpu(anchor->mainVolDescSeqExt.extLength); + main_e = main_e >> sb->s_blocksize_bits; + main_e += main_s; - brelse(bh); + /* Locate the reserve sequence */ + reserve_s = le32_to_cpu( + anchor->reserveVolDescSeqExt.extLocation); + reserve_e = le32_to_cpu( + anchor->reserveVolDescSeqExt.extLength); + reserve_e = reserve_e >> sb->s_blocksize_bits; + reserve_e += reserve_s; - /* Process the main & reserve sequences */ - /* responsible for finding the PartitionDesc(s) */ - if (!(udf_process_sequence(sb, main_s, main_e, fileset) && - udf_process_sequence(sb, reserve_s, reserve_e, fileset))) { - break; - } - } + brelse(bh); + + /* Process the main & reserve sequences */ + /* responsible for finding the PartitionDesc(s) */ + if (!(udf_process_sequence(sb, main_s, main_e, + fileset) && + udf_process_sequence(sb, reserve_s, reserve_e, + fileset))) + break; } - if (i == ARRAY_SIZE(UDF_SB_ANCHOR(sb))) { + if (i == ARRAY_SIZE(sbi->s_anchor)) { udf_debug("No Anchor block found\n"); return 1; - } else - udf_debug("Using anchor in block %d\n", UDF_SB_ANCHOR(sb)[i]); + } + udf_debug("Using anchor in block %d\n", sbi->s_anchor[i]); - for (i = 0; i < UDF_SB_NUMPARTS(sb); i++) { + for (i = 0; i < sbi->s_partitions; i++) { kernel_lb_addr uninitialized_var(ino); - switch (UDF_SB_PARTTYPE(sb, i)) { + struct udf_part_map *map = &sbi->s_partmaps[i]; + switch (map->s_partition_type) { case UDF_VIRTUAL_MAP15: case UDF_VIRTUAL_MAP20: - if (!UDF_SB_LASTBLOCK(sb)) { - UDF_SB_LASTBLOCK(sb) = udf_get_last_block(sb); + if (!sbi->s_last_block) { + sbi->s_last_block = udf_get_last_block(sb); udf_find_anchor(sb); } - if (!UDF_SB_LASTBLOCK(sb)) { + if (!sbi->s_last_block) { udf_debug("Unable to determine Lastblock (For " "Virtual Partition)\n"); return 1; } - for (j = 0; j < UDF_SB_NUMPARTS(sb); j++) { + for (j = 0; j < sbi->s_partitions; j++) { + struct udf_part_map *map2 = &sbi->s_partmaps[j]; if (j != i && - UDF_SB_PARTVSN(sb, i) == UDF_SB_PARTVSN(sb, j) && - UDF_SB_PARTNUM(sb, i) == UDF_SB_PARTNUM(sb, j)) { + map->s_volumeseqnum == + map2->s_volumeseqnum && + map->s_partition_num == + map2->s_partition_num) { ino.partitionReferenceNum = j; - ino.logicalBlockNum = UDF_SB_LASTBLOCK(sb) - UDF_SB_PARTROOT(sb, j); + ino.logicalBlockNum = + sbi->s_last_block - + map2->s_partition_root; break; } } - if (j == UDF_SB_NUMPARTS(sb)) + if (j == sbi->s_partitions) return 1; - if (!(UDF_SB_VAT(sb) = udf_iget(sb, ino))) + sbi->s_vat_inode = udf_iget(sb, ino); + if (!sbi->s_vat_inode) return 1; - if (UDF_SB_PARTTYPE(sb, i) == UDF_VIRTUAL_MAP15) { - UDF_SB_TYPEVIRT(sb, i).s_start_offset = - udf_ext0_offset(UDF_SB_VAT(sb)); - UDF_SB_TYPEVIRT(sb, i).s_num_entries = - (UDF_SB_VAT(sb)->i_size - 36) >> 2; - } else if (UDF_SB_PARTTYPE(sb, i) == UDF_VIRTUAL_MAP20) { - struct buffer_head *bh = NULL; + if (map->s_partition_type == UDF_VIRTUAL_MAP15) { + map->s_type_specific.s_virtual.s_start_offset = + udf_ext0_offset(sbi->s_vat_inode); + map->s_type_specific.s_virtual.s_num_entries = + (sbi->s_vat_inode->i_size - 36) >> 2; + } else if (map->s_partition_type == UDF_VIRTUAL_MAP20) { uint32_t pos; + struct virtualAllocationTable20 *vat20; - pos = udf_block_map(UDF_SB_VAT(sb), 0); + pos = udf_block_map(sbi->s_vat_inode, 0); bh = sb_bread(sb, pos); if (!bh) return 1; - UDF_SB_TYPEVIRT(sb, i).s_start_offset = - le16_to_cpu(((struct virtualAllocationTable20 *)bh->b_data + - udf_ext0_offset(UDF_SB_VAT(sb)))->lengthHeader) + - udf_ext0_offset(UDF_SB_VAT(sb)); - UDF_SB_TYPEVIRT(sb, i).s_num_entries = (UDF_SB_VAT(sb)->i_size - - UDF_SB_TYPEVIRT(sb, i).s_start_offset) >> 2; + vat20 = (struct virtualAllocationTable20 *) + bh->b_data + + udf_ext0_offset(sbi->s_vat_inode); + map->s_type_specific.s_virtual.s_start_offset = + le16_to_cpu(vat20->lengthHeader) + + udf_ext0_offset(sbi->s_vat_inode); + map->s_type_specific.s_virtual.s_num_entries = + (sbi->s_vat_inode->i_size - + map->s_type_specific.s_virtual. + s_start_offset) >> 2; brelse(bh); } - UDF_SB_PARTROOT(sb, i) = udf_get_pblock(sb, 0, i, 0); - UDF_SB_PARTLEN(sb, i) = UDF_SB_PARTLEN(sb, ino.partitionReferenceNum); + map->s_partition_root = udf_get_pblock(sb, 0, i, 0); + map->s_partition_len = + sbi->s_partmaps[ino.partitionReferenceNum]. + s_partition_len; } } return 0; @@ -1341,62 +1628,86 @@ static int udf_load_partition(struct super_block *sb, kernel_lb_addr *fileset) static void udf_open_lvid(struct super_block *sb) { - if (UDF_SB_LVIDBH(sb)) { - int i; + struct udf_sb_info *sbi = UDF_SB(sb); + struct buffer_head *bh = sbi->s_lvid_bh; + if (bh) { kernel_timestamp cpu_time; + struct logicalVolIntegrityDesc *lvid = + (struct logicalVolIntegrityDesc *)bh->b_data; + struct logicalVolIntegrityDescImpUse *lvidiu = + udf_sb_lvidiu(sbi); - UDF_SB_LVIDIU(sb)->impIdent.identSuffix[0] = UDF_OS_CLASS_UNIX; - UDF_SB_LVIDIU(sb)->impIdent.identSuffix[1] = UDF_OS_ID_LINUX; + lvidiu->impIdent.identSuffix[0] = UDF_OS_CLASS_UNIX; + lvidiu->impIdent.identSuffix[1] = UDF_OS_ID_LINUX; if (udf_time_to_stamp(&cpu_time, CURRENT_TIME)) - UDF_SB_LVID(sb)->recordingDateAndTime = cpu_to_lets(cpu_time); - UDF_SB_LVID(sb)->integrityType = LVID_INTEGRITY_TYPE_OPEN; - - UDF_SB_LVID(sb)->descTag.descCRC = cpu_to_le16(udf_crc((char *)UDF_SB_LVID(sb) + sizeof(tag), - le16_to_cpu(UDF_SB_LVID(sb)->descTag.descCRCLength), 0)); + lvid->recordingDateAndTime = cpu_to_lets(cpu_time); + lvid->integrityType = LVID_INTEGRITY_TYPE_OPEN; - UDF_SB_LVID(sb)->descTag.tagChecksum = 0; - for (i = 0; i < 16; i++) - if (i != 4) - UDF_SB_LVID(sb)->descTag.tagChecksum += - ((uint8_t *) &(UDF_SB_LVID(sb)->descTag))[i]; + lvid->descTag.descCRC = cpu_to_le16( + udf_crc((char *)lvid + sizeof(tag), + le16_to_cpu(lvid->descTag.descCRCLength), + 0)); - mark_buffer_dirty(UDF_SB_LVIDBH(sb)); + lvid->descTag.tagChecksum = udf_tag_checksum(&lvid->descTag); + mark_buffer_dirty(bh); } } static void udf_close_lvid(struct super_block *sb) { kernel_timestamp cpu_time; - int i; + struct udf_sb_info *sbi = UDF_SB(sb); + struct buffer_head *bh = sbi->s_lvid_bh; + struct logicalVolIntegrityDesc *lvid; + + if (!bh) + return; + + lvid = (struct logicalVolIntegrityDesc *)bh->b_data; - if (UDF_SB_LVIDBH(sb) && - UDF_SB_LVID(sb)->integrityType == LVID_INTEGRITY_TYPE_OPEN) { - UDF_SB_LVIDIU(sb)->impIdent.identSuffix[0] = UDF_OS_CLASS_UNIX; - UDF_SB_LVIDIU(sb)->impIdent.identSuffix[1] = UDF_OS_ID_LINUX; + if (lvid->integrityType == LVID_INTEGRITY_TYPE_OPEN) { + struct logicalVolIntegrityDescImpUse *lvidiu = + udf_sb_lvidiu(sbi); + lvidiu->impIdent.identSuffix[0] = UDF_OS_CLASS_UNIX; + lvidiu->impIdent.identSuffix[1] = UDF_OS_ID_LINUX; if (udf_time_to_stamp(&cpu_time, CURRENT_TIME)) - UDF_SB_LVID(sb)->recordingDateAndTime = cpu_to_lets(cpu_time); - if (UDF_MAX_WRITE_VERSION > le16_to_cpu(UDF_SB_LVIDIU(sb)->maxUDFWriteRev)) - UDF_SB_LVIDIU(sb)->maxUDFWriteRev = cpu_to_le16(UDF_MAX_WRITE_VERSION); - if (UDF_SB_UDFREV(sb) > le16_to_cpu(UDF_SB_LVIDIU(sb)->minUDFReadRev)) - UDF_SB_LVIDIU(sb)->minUDFReadRev = cpu_to_le16(UDF_SB_UDFREV(sb)); - if (UDF_SB_UDFREV(sb) > le16_to_cpu(UDF_SB_LVIDIU(sb)->minUDFWriteRev)) - UDF_SB_LVIDIU(sb)->minUDFWriteRev = cpu_to_le16(UDF_SB_UDFREV(sb)); - UDF_SB_LVID(sb)->integrityType = cpu_to_le32(LVID_INTEGRITY_TYPE_CLOSE); - - UDF_SB_LVID(sb)->descTag.descCRC = - cpu_to_le16(udf_crc((char *)UDF_SB_LVID(sb) + sizeof(tag), - le16_to_cpu(UDF_SB_LVID(sb)->descTag.descCRCLength), 0)); - - UDF_SB_LVID(sb)->descTag.tagChecksum = 0; - for (i = 0; i < 16; i++) - if (i != 4) - UDF_SB_LVID(sb)->descTag.tagChecksum += - ((uint8_t *)&(UDF_SB_LVID(sb)->descTag))[i]; - - mark_buffer_dirty(UDF_SB_LVIDBH(sb)); + lvid->recordingDateAndTime = cpu_to_lets(cpu_time); + if (UDF_MAX_WRITE_VERSION > le16_to_cpu(lvidiu->maxUDFWriteRev)) + lvidiu->maxUDFWriteRev = + cpu_to_le16(UDF_MAX_WRITE_VERSION); + if (sbi->s_udfrev > le16_to_cpu(lvidiu->minUDFReadRev)) + lvidiu->minUDFReadRev = cpu_to_le16(sbi->s_udfrev); + if (sbi->s_udfrev > le16_to_cpu(lvidiu->minUDFWriteRev)) + lvidiu->minUDFWriteRev = cpu_to_le16(sbi->s_udfrev); + lvid->integrityType = cpu_to_le32(LVID_INTEGRITY_TYPE_CLOSE); + + lvid->descTag.descCRC = cpu_to_le16( + udf_crc((char *)lvid + sizeof(tag), + le16_to_cpu(lvid->descTag.descCRCLength), + 0)); + + lvid->descTag.tagChecksum = udf_tag_checksum(&lvid->descTag); + mark_buffer_dirty(bh); } } +static void udf_sb_free_bitmap(struct udf_bitmap *bitmap) +{ + int i; + int nr_groups = bitmap->s_nr_groups; + int size = sizeof(struct udf_bitmap) + (sizeof(struct buffer_head *) * + nr_groups); + + for (i = 0; i < nr_groups; i++) + if (bitmap->s_block_bitmap[i]) + brelse(bitmap->s_block_bitmap[i]); + + if (size <= PAGE_SIZE) + kfree(bitmap); + else + vfree(bitmap); +} + /* * udf_read_super * @@ -1426,16 +1737,15 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) uopt.gid = -1; uopt.umask = 0; - sbi = kmalloc(sizeof(struct udf_sb_info), GFP_KERNEL); + sbi = kzalloc(sizeof(struct udf_sb_info), GFP_KERNEL); if (!sbi) return -ENOMEM; sb->s_fs_info = sbi; - memset(UDF_SB(sb), 0x00, sizeof(struct udf_sb_info)); mutex_init(&sbi->s_alloc_mutex); - if (!udf_parse_options((char *)options, &uopt)) + if (!udf_parse_options((char *)options, &uopt, false)) goto error_out; if (uopt.flags & (1 << UDF_FLAG_UTF8) && @@ -1459,30 +1769,31 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) fileset.logicalBlockNum = 0xFFFFFFFF; fileset.partitionReferenceNum = 0xFFFF; - UDF_SB(sb)->s_flags = uopt.flags; - UDF_SB(sb)->s_uid = uopt.uid; - UDF_SB(sb)->s_gid = uopt.gid; - UDF_SB(sb)->s_umask = uopt.umask; - UDF_SB(sb)->s_nls_map = uopt.nls_map; + sbi->s_flags = uopt.flags; + sbi->s_uid = uopt.uid; + sbi->s_gid = uopt.gid; + sbi->s_umask = uopt.umask; + sbi->s_nls_map = uopt.nls_map; /* Set the block size for all transfers */ if (!udf_set_blocksize(sb, uopt.blocksize)) goto error_out; if (uopt.session == 0xFFFFFFFF) - UDF_SB_SESSION(sb) = udf_get_last_session(sb); + sbi->s_session = udf_get_last_session(sb); else - UDF_SB_SESSION(sb) = uopt.session; + sbi->s_session = uopt.session; - udf_debug("Multi-session=%d\n", UDF_SB_SESSION(sb)); + udf_debug("Multi-session=%d\n", sbi->s_session); - UDF_SB_LASTBLOCK(sb) = uopt.lastblock; - UDF_SB_ANCHOR(sb)[0] = UDF_SB_ANCHOR(sb)[1] = 0; - UDF_SB_ANCHOR(sb)[2] = uopt.anchor; - UDF_SB_ANCHOR(sb)[3] = 256; + sbi->s_last_block = uopt.lastblock; + sbi->s_anchor[0] = sbi->s_anchor[1] = 0; + sbi->s_anchor[2] = uopt.anchor; + sbi->s_anchor[3] = 256; - if (udf_check_valid(sb, uopt.novrs, silent)) { /* read volume recognition sequences */ - printk("UDF-fs: No VRS found\n"); + if (udf_check_valid(sb, uopt.novrs, silent)) { + /* read volume recognition sequences */ + printk(KERN_WARNING "UDF-fs: No VRS found\n"); goto error_out; } @@ -1496,27 +1807,30 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) sb->s_time_gran = 1000; if (udf_load_partition(sb, &fileset)) { - printk("UDF-fs: No partition found (1)\n"); + printk(KERN_WARNING "UDF-fs: No partition found (1)\n"); goto error_out; } - udf_debug("Lastblock=%d\n", UDF_SB_LASTBLOCK(sb)); + udf_debug("Lastblock=%d\n", sbi->s_last_block); - if (UDF_SB_LVIDBH(sb)) { - uint16_t minUDFReadRev = le16_to_cpu(UDF_SB_LVIDIU(sb)->minUDFReadRev); - uint16_t minUDFWriteRev = le16_to_cpu(UDF_SB_LVIDIU(sb)->minUDFWriteRev); - /* uint16_t maxUDFWriteRev = le16_to_cpu(UDF_SB_LVIDIU(sb)->maxUDFWriteRev); */ + if (sbi->s_lvid_bh) { + struct logicalVolIntegrityDescImpUse *lvidiu = + udf_sb_lvidiu(sbi); + uint16_t minUDFReadRev = le16_to_cpu(lvidiu->minUDFReadRev); + uint16_t minUDFWriteRev = le16_to_cpu(lvidiu->minUDFWriteRev); + /* uint16_t maxUDFWriteRev = + le16_to_cpu(lvidiu->maxUDFWriteRev); */ if (minUDFReadRev > UDF_MAX_READ_VERSION) { - printk("UDF-fs: minUDFReadRev=%x (max is %x)\n", - le16_to_cpu(UDF_SB_LVIDIU(sb)->minUDFReadRev), + printk(KERN_ERR "UDF-fs: minUDFReadRev=%x " + "(max is %x)\n", + le16_to_cpu(lvidiu->minUDFReadRev), UDF_MAX_READ_VERSION); goto error_out; - } else if (minUDFWriteRev > UDF_MAX_WRITE_VERSION) { + } else if (minUDFWriteRev > UDF_MAX_WRITE_VERSION) sb->s_flags |= MS_RDONLY; - } - UDF_SB_UDFREV(sb) = minUDFWriteRev; + sbi->s_udfrev = minUDFWriteRev; if (minUDFReadRev >= UDF_VERS_USE_EXTENDED_FE) UDF_SET_FLAG(sb, UDF_FLAG_USE_EXTENDED_FE); @@ -1524,29 +1838,30 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) UDF_SET_FLAG(sb, UDF_FLAG_USE_STREAMS); } - if (!UDF_SB_NUMPARTS(sb)) { - printk("UDF-fs: No partition found (2)\n"); + if (!sbi->s_partitions) { + printk(KERN_WARNING "UDF-fs: No partition found (2)\n"); goto error_out; } - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_READ_ONLY) { - printk("UDF-fs: Partition marked readonly; forcing readonly mount\n"); + if (sbi->s_partmaps[sbi->s_partition].s_partition_flags & + UDF_PART_FLAG_READ_ONLY) { + printk(KERN_NOTICE "UDF-fs: Partition marked readonly; " + "forcing readonly mount\n"); sb->s_flags |= MS_RDONLY; } if (udf_find_fileset(sb, &fileset, &rootdir)) { - printk("UDF-fs: No fileset found\n"); + printk(KERN_WARNING "UDF-fs: No fileset found\n"); goto error_out; } if (!silent) { kernel_timestamp ts; - udf_time_to_stamp(&ts, UDF_SB_RECORDTIME(sb)); - udf_info("UDF %s (%s) Mounting volume '%s', " + udf_time_to_stamp(&ts, sbi->s_record_time); + udf_info("UDF: Mounting volume '%s', " "timestamp %04u/%02u/%02u %02u:%02u (%x)\n", - UDFFS_VERSION, UDFFS_DATE, - UDF_SB_VOLIDENT(sb), ts.year, ts.month, ts.day, ts.hour, ts.minute, - ts.typeAndTimezone); + sbi->s_volume_ident, ts.year, ts.month, ts.day, + ts.hour, ts.minute, ts.typeAndTimezone); } if (!(sb->s_flags & MS_RDONLY)) udf_open_lvid(sb); @@ -1556,7 +1871,8 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) /* perhaps it's not extensible enough, but for now ... */ inode = udf_iget(sb, rootdir); if (!inode) { - printk("UDF-fs: Error in udf_iget, block=%d, partition=%d\n", + printk(KERN_ERR "UDF-fs: Error in udf_iget, block=%d, " + "partition=%d\n", rootdir.logicalBlockNum, rootdir.partitionReferenceNum); goto error_out; } @@ -1564,7 +1880,7 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) /* Allocate a dentry for the root inode */ sb->s_root = d_alloc_root(inode); if (!sb->s_root) { - printk("UDF-fs: Couldn't allocate root dentry\n"); + printk(KERN_ERR "UDF-fs: Couldn't allocate root dentry\n"); iput(inode); goto error_out; } @@ -1572,30 +1888,32 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) return 0; error_out: - if (UDF_SB_VAT(sb)) - iput(UDF_SB_VAT(sb)); - if (UDF_SB_NUMPARTS(sb)) { - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_UNALLOC_TABLE) - iput(UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_uspace.s_table); - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_FREED_TABLE) - iput(UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_fspace.s_table); - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_UNALLOC_BITMAP) - UDF_SB_FREE_BITMAP(sb,UDF_SB_PARTITION(sb), s_uspace); - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_FREED_BITMAP) - UDF_SB_FREE_BITMAP(sb,UDF_SB_PARTITION(sb), s_fspace); - if (UDF_SB_PARTTYPE(sb, UDF_SB_PARTITION(sb)) == UDF_SPARABLE_MAP15) { + if (sbi->s_vat_inode) + iput(sbi->s_vat_inode); + if (sbi->s_partitions) { + struct udf_part_map *map = &sbi->s_partmaps[sbi->s_partition]; + if (map->s_partition_flags & UDF_PART_FLAG_UNALLOC_TABLE) + iput(map->s_uspace.s_table); + if (map->s_partition_flags & UDF_PART_FLAG_FREED_TABLE) + iput(map->s_fspace.s_table); + if (map->s_partition_flags & UDF_PART_FLAG_UNALLOC_BITMAP) + udf_sb_free_bitmap(map->s_uspace.s_bitmap); + if (map->s_partition_flags & UDF_PART_FLAG_FREED_BITMAP) + udf_sb_free_bitmap(map->s_fspace.s_bitmap); + if (map->s_partition_type == UDF_SPARABLE_MAP15) for (i = 0; i < 4; i++) - brelse(UDF_SB_TYPESPAR(sb, UDF_SB_PARTITION(sb)).s_spar_map[i]); - } + brelse(map->s_type_specific.s_sparing. + s_spar_map[i]); } #ifdef CONFIG_UDF_NLS if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) - unload_nls(UDF_SB(sb)->s_nls_map); + unload_nls(sbi->s_nls_map); #endif if (!(sb->s_flags & MS_RDONLY)) udf_close_lvid(sb); - brelse(UDF_SB_LVIDBH(sb)); - UDF_SB_FREE(sb); + brelse(sbi->s_lvid_bh); + + kfree(sbi->s_partmaps); kfree(sbi); sb->s_fs_info = NULL; @@ -1614,7 +1932,7 @@ void udf_error(struct super_block *sb, const char *function, va_start(args, fmt); vsnprintf(error_buf, sizeof(error_buf), fmt, args); va_end(args); - printk (KERN_CRIT "UDF-fs error (device %s): %s: %s\n", + printk(KERN_CRIT "UDF-fs error (device %s): %s: %s\n", sb->s_id, function, error_buf); } @@ -1646,31 +1964,34 @@ void udf_warning(struct super_block *sb, const char *function, static void udf_put_super(struct super_block *sb) { int i; + struct udf_sb_info *sbi; - if (UDF_SB_VAT(sb)) - iput(UDF_SB_VAT(sb)); - if (UDF_SB_NUMPARTS(sb)) { - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_UNALLOC_TABLE) - iput(UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_uspace.s_table); - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_FREED_TABLE) - iput(UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_fspace.s_table); - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_UNALLOC_BITMAP) - UDF_SB_FREE_BITMAP(sb,UDF_SB_PARTITION(sb), s_uspace); - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_FREED_BITMAP) - UDF_SB_FREE_BITMAP(sb,UDF_SB_PARTITION(sb), s_fspace); - if (UDF_SB_PARTTYPE(sb, UDF_SB_PARTITION(sb)) == UDF_SPARABLE_MAP15) { + sbi = UDF_SB(sb); + if (sbi->s_vat_inode) + iput(sbi->s_vat_inode); + if (sbi->s_partitions) { + struct udf_part_map *map = &sbi->s_partmaps[sbi->s_partition]; + if (map->s_partition_flags & UDF_PART_FLAG_UNALLOC_TABLE) + iput(map->s_uspace.s_table); + if (map->s_partition_flags & UDF_PART_FLAG_FREED_TABLE) + iput(map->s_fspace.s_table); + if (map->s_partition_flags & UDF_PART_FLAG_UNALLOC_BITMAP) + udf_sb_free_bitmap(map->s_uspace.s_bitmap); + if (map->s_partition_flags & UDF_PART_FLAG_FREED_BITMAP) + udf_sb_free_bitmap(map->s_fspace.s_bitmap); + if (map->s_partition_type == UDF_SPARABLE_MAP15) for (i = 0; i < 4; i++) - brelse(UDF_SB_TYPESPAR(sb, UDF_SB_PARTITION(sb)).s_spar_map[i]); - } + brelse(map->s_type_specific.s_sparing. + s_spar_map[i]); } #ifdef CONFIG_UDF_NLS if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) - unload_nls(UDF_SB(sb)->s_nls_map); + unload_nls(sbi->s_nls_map); #endif if (!(sb->s_flags & MS_RDONLY)) udf_close_lvid(sb); - brelse(UDF_SB_LVIDBH(sb)); - UDF_SB_FREE(sb); + brelse(sbi->s_lvid_bh); + kfree(sbi->s_partmaps); kfree(sb->s_fs_info); sb->s_fs_info = NULL; } @@ -1691,15 +2012,22 @@ static void udf_put_super(struct super_block *sb) static int udf_statfs(struct dentry *dentry, struct kstatfs *buf) { struct super_block *sb = dentry->d_sb; + struct udf_sb_info *sbi = UDF_SB(sb); + struct logicalVolIntegrityDescImpUse *lvidiu; + + if (sbi->s_lvid_bh != NULL) + lvidiu = udf_sb_lvidiu(sbi); + else + lvidiu = NULL; buf->f_type = UDF_SUPER_MAGIC; buf->f_bsize = sb->s_blocksize; - buf->f_blocks = UDF_SB_PARTLEN(sb, UDF_SB_PARTITION(sb)); + buf->f_blocks = sbi->s_partmaps[sbi->s_partition].s_partition_len; buf->f_bfree = udf_count_free(sb); buf->f_bavail = buf->f_bfree; - buf->f_files = (UDF_SB_LVIDBH(sb) ? - (le32_to_cpu(UDF_SB_LVIDIU(sb)->numFiles) + - le32_to_cpu(UDF_SB_LVIDIU(sb)->numDirs)) : 0) + buf->f_bfree; + buf->f_files = (lvidiu != NULL ? (le32_to_cpu(lvidiu->numFiles) + + le32_to_cpu(lvidiu->numDirs)) : 0) + + buf->f_bfree; buf->f_ffree = buf->f_bfree; /* __kernel_fsid_t f_fsid */ buf->f_namelen = UDF_NAME_LEN - 2; @@ -1711,7 +2039,8 @@ static unsigned char udf_bitmap_lookup[16] = { 0, 1, 1, 2, 1, 2, 2, 3, 1, 2, 2, 3, 2, 3, 3, 4 }; -static unsigned int udf_count_free_bitmap(struct super_block *sb, struct udf_bitmap *bitmap) +static unsigned int udf_count_free_bitmap(struct super_block *sb, + struct udf_bitmap *bitmap) { struct buffer_head *bh = NULL; unsigned int accum = 0; @@ -1727,7 +2056,7 @@ static unsigned int udf_count_free_bitmap(struct super_block *sb, struct udf_bit lock_kernel(); loc.logicalBlockNum = bitmap->s_extPosition; - loc.partitionReferenceNum = UDF_SB_PARTITION(sb); + loc.partitionReferenceNum = UDF_SB(sb)->s_partition; bh = udf_read_ptagged(sb, loc, 0, &ident); if (!bh) { @@ -1772,7 +2101,8 @@ out: return accum; } -static unsigned int udf_count_free_table(struct super_block *sb, struct inode *table) +static unsigned int udf_count_free_table(struct super_block *sb, + struct inode *table) { unsigned int accum = 0; uint32_t elen; @@ -1782,13 +2112,13 @@ static unsigned int udf_count_free_table(struct super_block *sb, struct inode *t lock_kernel(); - epos.block = UDF_I_LOCATION(table); + epos.block = UDF_I(table)->i_location; epos.offset = sizeof(struct unallocSpaceEntry); epos.bh = NULL; - while ((etype = udf_next_aext(table, &epos, &eloc, &elen, 1)) != -1) { + while ((etype = udf_next_aext(table, &epos, &eloc, &elen, 1)) != -1) accum += (elen >> table->i_sb->s_blocksize_bits); - } + brelse(epos.bh); unlock_kernel(); @@ -1799,10 +2129,17 @@ static unsigned int udf_count_free_table(struct super_block *sb, struct inode *t static unsigned int udf_count_free(struct super_block *sb) { unsigned int accum = 0; - - if (UDF_SB_LVIDBH(sb)) { - if (le32_to_cpu(UDF_SB_LVID(sb)->numOfPartitions) > UDF_SB_PARTITION(sb)) { - accum = le32_to_cpu(UDF_SB_LVID(sb)->freeSpaceTable[UDF_SB_PARTITION(sb)]); + struct udf_sb_info *sbi; + struct udf_part_map *map; + + sbi = UDF_SB(sb); + if (sbi->s_lvid_bh) { + struct logicalVolIntegrityDesc *lvid = + (struct logicalVolIntegrityDesc *) + sbi->s_lvid_bh->b_data; + if (le32_to_cpu(lvid->numOfPartitions) > sbi->s_partition) { + accum = le32_to_cpu( + lvid->freeSpaceTable[sbi->s_partition]); if (accum == 0xFFFFFFFF) accum = 0; } @@ -1811,24 +2148,25 @@ static unsigned int udf_count_free(struct super_block *sb) if (accum) return accum; - if (UDF_SB_PARTFLAGS(sb,UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_UNALLOC_BITMAP) { + map = &sbi->s_partmaps[sbi->s_partition]; + if (map->s_partition_flags & UDF_PART_FLAG_UNALLOC_BITMAP) { accum += udf_count_free_bitmap(sb, - UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_uspace.s_bitmap); + map->s_uspace.s_bitmap); } - if (UDF_SB_PARTFLAGS(sb,UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_FREED_BITMAP) { + if (map->s_partition_flags & UDF_PART_FLAG_FREED_BITMAP) { accum += udf_count_free_bitmap(sb, - UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_fspace.s_bitmap); + map->s_fspace.s_bitmap); } if (accum) return accum; - if (UDF_SB_PARTFLAGS(sb,UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_UNALLOC_TABLE) { + if (map->s_partition_flags & UDF_PART_FLAG_UNALLOC_TABLE) { accum += udf_count_free_table(sb, - UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_uspace.s_table); + map->s_uspace.s_table); } - if (UDF_SB_PARTFLAGS(sb,UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_FREED_TABLE) { + if (map->s_partition_flags & UDF_PART_FLAG_FREED_TABLE) { accum += udf_count_free_table(sb, - UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_fspace.s_table); + map->s_fspace.s_table); } return accum; diff --git a/fs/udf/symlink.c b/fs/udf/symlink.c index e6f933dd6a7..6ec99221e50 100644 --- a/fs/udf/symlink.c +++ b/fs/udf/symlink.c @@ -33,7 +33,8 @@ #include <linux/buffer_head.h> #include "udf_i.h" -static void udf_pc_to_char(struct super_block *sb, char *from, int fromlen, char *to) +static void udf_pc_to_char(struct super_block *sb, char *from, int fromlen, + char *to) { struct pathComponent *pc; int elen = 0; @@ -78,10 +79,12 @@ static int udf_symlink_filler(struct file *file, struct page *page) char *symlink; int err = -EIO; char *p = kmap(page); + struct udf_inode_info *iinfo; lock_kernel(); - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) { - symlink = UDF_I_DATA(inode) + UDF_I_LENEATTR(inode); + iinfo = UDF_I(inode); + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) { + symlink = iinfo->i_ext.i_data + iinfo->i_lenEAttr; } else { bh = sb_bread(inode->i_sb, udf_block_map(inode, 0)); diff --git a/fs/udf/truncate.c b/fs/udf/truncate.c index 7fc3912885a..fe61be17cda 100644 --- a/fs/udf/truncate.c +++ b/fs/udf/truncate.c @@ -74,17 +74,18 @@ void udf_truncate_tail_extent(struct inode *inode) uint64_t lbcount = 0; int8_t etype = -1, netype; int adsize; + struct udf_inode_info *iinfo = UDF_I(inode); - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB || - inode->i_size == UDF_I_LENEXTENTS(inode)) + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB || + inode->i_size == iinfo->i_lenExtents) return; /* Are we going to delete the file anyway? */ if (inode->i_nlink == 0) return; - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_SHORT) + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT) adsize = sizeof(short_ad); - else if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_LONG) + else if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_LONG) adsize = sizeof(long_ad); else BUG(); @@ -117,7 +118,7 @@ void udf_truncate_tail_extent(struct inode *inode) } /* This inode entry is in-memory only and thus we don't have to mark * the inode dirty */ - UDF_I_LENEXTENTS(inode) = inode->i_size; + iinfo->i_lenExtents = inode->i_size; brelse(epos.bh); } @@ -129,19 +130,20 @@ void udf_discard_prealloc(struct inode *inode) uint64_t lbcount = 0; int8_t etype = -1, netype; int adsize; + struct udf_inode_info *iinfo = UDF_I(inode); - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB || - inode->i_size == UDF_I_LENEXTENTS(inode)) + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB || + inode->i_size == iinfo->i_lenExtents) return; - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_SHORT) + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT) adsize = sizeof(short_ad); - else if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_LONG) + else if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_LONG) adsize = sizeof(long_ad); else adsize = 0; - epos.block = UDF_I_LOCATION(inode); + epos.block = iinfo->i_location; /* Find the last extent in the file */ while ((netype = udf_next_aext(inode, &epos, &eloc, &elen, 1)) != -1) { @@ -153,8 +155,9 @@ void udf_discard_prealloc(struct inode *inode) lbcount -= elen; extent_trunc(inode, &epos, eloc, etype, elen, 0); if (!epos.bh) { - UDF_I_LENALLOC(inode) = - epos.offset - udf_file_entry_alloc_offset(inode); + iinfo->i_lenAlloc = + epos.offset - + udf_file_entry_alloc_offset(inode); mark_inode_dirty(inode); } else { struct allocExtDesc *aed = @@ -163,7 +166,7 @@ void udf_discard_prealloc(struct inode *inode) cpu_to_le32(epos.offset - sizeof(struct allocExtDesc)); if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) || - UDF_SB_UDFREV(inode->i_sb) >= 0x0201) + UDF_SB(inode->i_sb)->s_udfrev >= 0x0201) udf_update_tag(epos.bh->b_data, epos.offset); else udf_update_tag(epos.bh->b_data, @@ -173,7 +176,7 @@ void udf_discard_prealloc(struct inode *inode) } /* This inode entry is in-memory only and thus we don't have to mark * the inode dirty */ - UDF_I_LENEXTENTS(inode) = lbcount; + iinfo->i_lenExtents = lbcount; brelse(epos.bh); } @@ -184,13 +187,15 @@ void udf_truncate_extents(struct inode *inode) uint32_t elen, nelen = 0, indirect_ext_len = 0, lenalloc; int8_t etype; struct super_block *sb = inode->i_sb; + struct udf_sb_info *sbi = UDF_SB(sb); sector_t first_block = inode->i_size >> sb->s_blocksize_bits, offset; loff_t byte_offset; int adsize; + struct udf_inode_info *iinfo = UDF_I(inode); - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_SHORT) + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT) adsize = sizeof(short_ad); - else if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_LONG) + else if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_LONG) adsize = sizeof(long_ad); else BUG(); @@ -212,7 +217,8 @@ void udf_truncate_extents(struct inode *inode) else lenalloc -= sizeof(struct allocExtDesc); - while ((etype = udf_current_aext(inode, &epos, &eloc, &elen, 0)) != -1) { + while ((etype = udf_current_aext(inode, &epos, &eloc, + &elen, 0)) != -1) { if (etype == (EXT_NEXT_EXTENT_ALLOCDECS >> 30)) { udf_write_aext(inode, &epos, neloc, nelen, 0); if (indirect_ext_len) { @@ -224,35 +230,43 @@ void udf_truncate_extents(struct inode *inode) 0, indirect_ext_len); } else { if (!epos.bh) { - UDF_I_LENALLOC(inode) = lenalloc; + iinfo->i_lenAlloc = + lenalloc; mark_inode_dirty(inode); } else { struct allocExtDesc *aed = - (struct allocExtDesc *)(epos.bh->b_data); + (struct allocExtDesc *) + (epos.bh->b_data); + int len = + sizeof(struct allocExtDesc); + aed->lengthAllocDescs = cpu_to_le32(lenalloc); - if (!UDF_QUERY_FLAG(sb, UDF_FLAG_STRICT) || - UDF_SB_UDFREV(sb) >= 0x0201) - udf_update_tag(epos.bh->b_data, - lenalloc + - sizeof(struct allocExtDesc)); - else - udf_update_tag(epos.bh->b_data, - sizeof(struct allocExtDesc)); - mark_buffer_dirty_inode(epos.bh, inode); + if (!UDF_QUERY_FLAG(sb, + UDF_FLAG_STRICT) || + sbi->s_udfrev >= 0x0201) + len += lenalloc; + + udf_update_tag(epos.bh->b_data, + len); + mark_buffer_dirty_inode( + epos.bh, inode); } } brelse(epos.bh); epos.offset = sizeof(struct allocExtDesc); epos.block = eloc; - epos.bh = udf_tread(sb, udf_get_lb_pblock(sb, eloc, 0)); + epos.bh = udf_tread(sb, + udf_get_lb_pblock(sb, eloc, 0)); if (elen) - indirect_ext_len = (elen + sb->s_blocksize -1) >> + indirect_ext_len = + (elen + sb->s_blocksize - 1) >> sb->s_blocksize_bits; else indirect_ext_len = 1; } else { - extent_trunc(inode, &epos, eloc, etype, elen, 0); + extent_trunc(inode, &epos, eloc, etype, + elen, 0); epos.offset += adsize; } } @@ -264,19 +278,20 @@ void udf_truncate_extents(struct inode *inode) indirect_ext_len); } else { if (!epos.bh) { - UDF_I_LENALLOC(inode) = lenalloc; + iinfo->i_lenAlloc = lenalloc; mark_inode_dirty(inode); } else { struct allocExtDesc *aed = (struct allocExtDesc *)(epos.bh->b_data); aed->lengthAllocDescs = cpu_to_le32(lenalloc); if (!UDF_QUERY_FLAG(sb, UDF_FLAG_STRICT) || - UDF_SB_UDFREV(sb) >= 0x0201) + sbi->s_udfrev >= 0x0201) udf_update_tag(epos.bh->b_data, - lenalloc + sizeof(struct allocExtDesc)); + lenalloc + + sizeof(struct allocExtDesc)); else udf_update_tag(epos.bh->b_data, - sizeof(struct allocExtDesc)); + sizeof(struct allocExtDesc)); mark_buffer_dirty_inode(epos.bh, inode); } } @@ -290,13 +305,16 @@ void udf_truncate_extents(struct inode *inode) * extending the file by 'offset' blocks. */ if ((!epos.bh && - epos.offset == udf_file_entry_alloc_offset(inode)) || - (epos.bh && epos.offset == sizeof(struct allocExtDesc))) { + epos.offset == + udf_file_entry_alloc_offset(inode)) || + (epos.bh && epos.offset == + sizeof(struct allocExtDesc))) { /* File has no extents at all or has empty last * indirect extent! Create a fake extent... */ extent.extLocation.logicalBlockNum = 0; extent.extLocation.partitionReferenceNum = 0; - extent.extLength = EXT_NOT_RECORDED_NOT_ALLOCATED; + extent.extLength = + EXT_NOT_RECORDED_NOT_ALLOCATED; } else { epos.offset -= adsize; etype = udf_next_aext(inode, &epos, @@ -305,10 +323,12 @@ void udf_truncate_extents(struct inode *inode) extent.extLength |= etype << 30; } udf_extend_file(inode, &epos, &extent, - offset + ((inode->i_size & (sb->s_blocksize - 1)) != 0)); + offset + + ((inode->i_size & + (sb->s_blocksize - 1)) != 0)); } } - UDF_I_LENEXTENTS(inode) = inode->i_size; + iinfo->i_lenExtents = inode->i_size; brelse(epos.bh); } diff --git a/fs/udf/udf_i.h b/fs/udf/udf_i.h index d7dbe6f3ba0..ccc52f16bf7 100644 --- a/fs/udf/udf_i.h +++ b/fs/udf/udf_i.h @@ -7,20 +7,4 @@ static inline struct udf_inode_info *UDF_I(struct inode *inode) return list_entry(inode, struct udf_inode_info, vfs_inode); } -#define UDF_I_LOCATION(X) ( UDF_I(X)->i_location ) -#define UDF_I_LENEATTR(X) ( UDF_I(X)->i_lenEAttr ) -#define UDF_I_LENALLOC(X) ( UDF_I(X)->i_lenAlloc ) -#define UDF_I_LENEXTENTS(X) ( UDF_I(X)->i_lenExtents ) -#define UDF_I_UNIQUE(X) ( UDF_I(X)->i_unique ) -#define UDF_I_ALLOCTYPE(X) ( UDF_I(X)->i_alloc_type ) -#define UDF_I_EFE(X) ( UDF_I(X)->i_efe ) -#define UDF_I_USE(X) ( UDF_I(X)->i_use ) -#define UDF_I_STRAT4096(X) ( UDF_I(X)->i_strat4096 ) -#define UDF_I_NEXT_ALLOC_BLOCK(X) ( UDF_I(X)->i_next_alloc_block ) -#define UDF_I_NEXT_ALLOC_GOAL(X) ( UDF_I(X)->i_next_alloc_goal ) -#define UDF_I_CRTIME(X) ( UDF_I(X)->i_crtime ) -#define UDF_I_SAD(X) ( UDF_I(X)->i_ext.i_sad ) -#define UDF_I_LAD(X) ( UDF_I(X)->i_ext.i_lad ) -#define UDF_I_DATA(X) ( UDF_I(X)->i_ext.i_data ) - #endif /* !defined(_LINUX_UDF_I_H) */ diff --git a/fs/udf/udf_sb.h b/fs/udf/udf_sb.h index 3c2982017c6..737d1c604ee 100644 --- a/fs/udf/udf_sb.h +++ b/fs/udf/udf_sb.h @@ -26,6 +26,8 @@ #define UDF_FLAG_GID_IGNORE 14 #define UDF_FLAG_UID_SET 15 #define UDF_FLAG_GID_SET 16 +#define UDF_FLAG_SESSION_SET 17 +#define UDF_FLAG_LASTBLOCK_SET 18 #define UDF_PART_FLAG_UNALLOC_BITMAP 0x0001 #define UDF_PART_FLAG_UNALLOC_TABLE 0x0002 @@ -41,96 +43,12 @@ static inline struct udf_sb_info *UDF_SB(struct super_block *sb) return sb->s_fs_info; } -#define UDF_SB_FREE(X)\ -{\ - if (UDF_SB(X)) {\ - kfree(UDF_SB_PARTMAPS(X));\ - UDF_SB_PARTMAPS(X) = NULL;\ - }\ -} - -#define UDF_SB_ALLOC_PARTMAPS(X,Y)\ -{\ - UDF_SB_PARTMAPS(X) = kmalloc(sizeof(struct udf_part_map) * Y, GFP_KERNEL);\ - if (UDF_SB_PARTMAPS(X) != NULL) {\ - UDF_SB_NUMPARTS(X) = Y;\ - memset(UDF_SB_PARTMAPS(X), 0x00, sizeof(struct udf_part_map) * Y);\ - } else {\ - UDF_SB_NUMPARTS(X) = 0;\ - udf_error(X, __FUNCTION__, "Unable to allocate space for %d partition maps", Y);\ - }\ -} - -#define UDF_SB_ALLOC_BITMAP(X,Y,Z)\ -{\ - int nr_groups = ((UDF_SB_PARTLEN((X),(Y)) + (sizeof(struct spaceBitmapDesc) << 3) +\ - ((X)->s_blocksize * 8) - 1) / ((X)->s_blocksize * 8));\ - int size = sizeof(struct udf_bitmap) + (sizeof(struct buffer_head *) * nr_groups);\ - if (size <= PAGE_SIZE)\ - UDF_SB_PARTMAPS(X)[(Y)].Z.s_bitmap = kmalloc(size, GFP_KERNEL);\ - else\ - UDF_SB_PARTMAPS(X)[(Y)].Z.s_bitmap = vmalloc(size);\ - if (UDF_SB_PARTMAPS(X)[(Y)].Z.s_bitmap != NULL) {\ - memset(UDF_SB_PARTMAPS(X)[(Y)].Z.s_bitmap, 0x00, size);\ - UDF_SB_PARTMAPS(X)[(Y)].Z.s_bitmap->s_block_bitmap =\ - (struct buffer_head **)(UDF_SB_PARTMAPS(X)[(Y)].Z.s_bitmap + 1);\ - UDF_SB_PARTMAPS(X)[(Y)].Z.s_bitmap->s_nr_groups = nr_groups;\ - } else {\ - udf_error(X, __FUNCTION__, "Unable to allocate space for bitmap and %d buffer_head pointers", nr_groups);\ - }\ -} +struct logicalVolIntegrityDescImpUse *udf_sb_lvidiu(struct udf_sb_info *sbi); -#define UDF_SB_FREE_BITMAP(X,Y,Z)\ -{\ - int i;\ - int nr_groups = UDF_SB_BITMAP_NR_GROUPS(X,Y,Z);\ - int size = sizeof(struct udf_bitmap) + (sizeof(struct buffer_head *) * nr_groups);\ - for (i = 0; i < nr_groups; i++) {\ - if (UDF_SB_BITMAP(X,Y,Z,i))\ - brelse(UDF_SB_BITMAP(X,Y,Z,i));\ - }\ - if (size <= PAGE_SIZE)\ - kfree(UDF_SB_PARTMAPS(X)[Y].Z.s_bitmap);\ - else\ - vfree(UDF_SB_PARTMAPS(X)[Y].Z.s_bitmap);\ -} +int udf_compute_nr_groups(struct super_block *sb, u32 partition); #define UDF_QUERY_FLAG(X,Y) ( UDF_SB(X)->s_flags & ( 1 << (Y) ) ) #define UDF_SET_FLAG(X,Y) ( UDF_SB(X)->s_flags |= ( 1 << (Y) ) ) #define UDF_CLEAR_FLAG(X,Y) ( UDF_SB(X)->s_flags &= ~( 1 << (Y) ) ) -#define UDF_UPDATE_UDFREV(X,Y) ( ((Y) > UDF_SB_UDFREV(X)) ? UDF_SB_UDFREV(X) = (Y) : UDF_SB_UDFREV(X) ) - -#define UDF_SB_PARTMAPS(X) ( UDF_SB(X)->s_partmaps ) -#define UDF_SB_PARTTYPE(X,Y) ( UDF_SB_PARTMAPS(X)[(Y)].s_partition_type ) -#define UDF_SB_PARTROOT(X,Y) ( UDF_SB_PARTMAPS(X)[(Y)].s_partition_root ) -#define UDF_SB_PARTLEN(X,Y) ( UDF_SB_PARTMAPS(X)[(Y)].s_partition_len ) -#define UDF_SB_PARTVSN(X,Y) ( UDF_SB_PARTMAPS(X)[(Y)].s_volumeseqnum ) -#define UDF_SB_PARTNUM(X,Y) ( UDF_SB_PARTMAPS(X)[(Y)].s_partition_num ) -#define UDF_SB_TYPESPAR(X,Y) ( UDF_SB_PARTMAPS(X)[(Y)].s_type_specific.s_sparing ) -#define UDF_SB_TYPEVIRT(X,Y) ( UDF_SB_PARTMAPS(X)[(Y)].s_type_specific.s_virtual ) -#define UDF_SB_PARTFUNC(X,Y) ( UDF_SB_PARTMAPS(X)[(Y)].s_partition_func ) -#define UDF_SB_PARTFLAGS(X,Y) ( UDF_SB_PARTMAPS(X)[(Y)].s_partition_flags ) -#define UDF_SB_BITMAP(X,Y,Z,I) ( UDF_SB_PARTMAPS(X)[(Y)].Z.s_bitmap->s_block_bitmap[I] ) -#define UDF_SB_BITMAP_NR_GROUPS(X,Y,Z) ( UDF_SB_PARTMAPS(X)[(Y)].Z.s_bitmap->s_nr_groups ) - -#define UDF_SB_VOLIDENT(X) ( UDF_SB(X)->s_volident ) -#define UDF_SB_NUMPARTS(X) ( UDF_SB(X)->s_partitions ) -#define UDF_SB_PARTITION(X) ( UDF_SB(X)->s_partition ) -#define UDF_SB_SESSION(X) ( UDF_SB(X)->s_session ) -#define UDF_SB_ANCHOR(X) ( UDF_SB(X)->s_anchor ) -#define UDF_SB_LASTBLOCK(X) ( UDF_SB(X)->s_lastblock ) -#define UDF_SB_LVIDBH(X) ( UDF_SB(X)->s_lvidbh ) -#define UDF_SB_LVID(X) ( (struct logicalVolIntegrityDesc *)UDF_SB_LVIDBH(X)->b_data ) -#define UDF_SB_LVIDIU(X) ( (struct logicalVolIntegrityDescImpUse *)&(UDF_SB_LVID(X)->impUse[le32_to_cpu(UDF_SB_LVID(X)->numOfPartitions) * 2 * sizeof(uint32_t)/sizeof(uint8_t)]) ) - -#define UDF_SB_UMASK(X) ( UDF_SB(X)->s_umask ) -#define UDF_SB_GID(X) ( UDF_SB(X)->s_gid ) -#define UDF_SB_UID(X) ( UDF_SB(X)->s_uid ) -#define UDF_SB_RECORDTIME(X) ( UDF_SB(X)->s_recordtime ) -#define UDF_SB_SERIALNUM(X) ( UDF_SB(X)->s_serialnum ) -#define UDF_SB_UDFREV(X) ( UDF_SB(X)->s_udfrev ) -#define UDF_SB_FLAGS(X) ( UDF_SB(X)->s_flags ) -#define UDF_SB_VAT(X) ( UDF_SB(X)->s_vat ) - #endif /* __LINUX_UDF_SB_H */ diff --git a/fs/udf/udfdecl.h b/fs/udf/udfdecl.h index c8016cc9e7e..681dc2b66cd 100644 --- a/fs/udf/udfdecl.h +++ b/fs/udf/udfdecl.h @@ -24,18 +24,21 @@ #define UDF_PATH_LEN 1023 #define udf_file_entry_alloc_offset(inode)\ - (UDF_I_USE(inode) ?\ + (UDF_I(inode)->i_use ?\ sizeof(struct unallocSpaceEntry) :\ - ((UDF_I_EFE(inode) ?\ + ((UDF_I(inode)->i_efe ?\ sizeof(struct extendedFileEntry) :\ - sizeof(struct fileEntry)) + UDF_I_LENEATTR(inode))) + sizeof(struct fileEntry)) + UDF_I(inode)->i_lenEAttr)) #define udf_ext0_offset(inode)\ - (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB ?\ + (UDF_I(inode)->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB ?\ udf_file_entry_alloc_offset(inode) : 0) #define udf_get_lb_pblock(sb,loc,offset) udf_get_pblock((sb), (loc).logicalBlockNum, (loc).partitionReferenceNum, (offset)) +/* computes tag checksum */ +u8 udf_tag_checksum(const tag *t); + struct dentry; struct inode; struct task_struct; @@ -185,8 +188,8 @@ extern struct fileIdentDesc *udf_fileident_read(struct inode *, loff_t *, sector_t *); extern struct fileIdentDesc *udf_get_fileident(void *buffer, int bufsize, int *offset); -extern long_ad *udf_get_filelongad(uint8_t *, int, int *, int); -extern short_ad *udf_get_fileshortad(uint8_t *, int, int *, int); +extern long_ad *udf_get_filelongad(uint8_t *, int, uint32_t *, int); +extern short_ad *udf_get_fileshortad(uint8_t *, int, uint32_t *, int); /* crc.c */ extern uint16_t udf_crc(uint8_t *, uint32_t, uint16_t); diff --git a/fs/udf/udftime.c b/fs/udf/udftime.c index adcb87c2da7..ce595732ba6 100644 --- a/fs/udf/udftime.c +++ b/fs/udf/udftime.c @@ -18,8 +18,10 @@ Boston, MA 02111-1307, USA. */ /* - * dgb 10/02/98: ripped this from glibc source to help convert timestamps to unix time - * 10/04/98: added new table-based lookup after seeing how ugly the gnu code is + * dgb 10/02/98: ripped this from glibc source to help convert timestamps + * to unix time + * 10/04/98: added new table-based lookup after seeing how ugly + * the gnu code is * blf 09/27/99: ripped out all the old code and inserted new table from * John Brockmeyer (without leap second corrections) * rewrote udf_stamp_to_time and fixed timezone accounting in @@ -55,27 +57,27 @@ static const unsigned short int __mon_yday[2][13] = { #define MAX_YEAR_SECONDS 69 #define SPD 0x15180 /*3600*24 */ -#define SPY(y,l,s) (SPD * (365*y+l)+s) - -static time_t year_seconds[MAX_YEAR_SECONDS]= { -/*1970*/ SPY( 0, 0,0), SPY( 1, 0,0), SPY( 2, 0,0), SPY( 3, 1,0), -/*1974*/ SPY( 4, 1,0), SPY( 5, 1,0), SPY( 6, 1,0), SPY( 7, 2,0), -/*1978*/ SPY( 8, 2,0), SPY( 9, 2,0), SPY(10, 2,0), SPY(11, 3,0), -/*1982*/ SPY(12, 3,0), SPY(13, 3,0), SPY(14, 3,0), SPY(15, 4,0), -/*1986*/ SPY(16, 4,0), SPY(17, 4,0), SPY(18, 4,0), SPY(19, 5,0), -/*1990*/ SPY(20, 5,0), SPY(21, 5,0), SPY(22, 5,0), SPY(23, 6,0), -/*1994*/ SPY(24, 6,0), SPY(25, 6,0), SPY(26, 6,0), SPY(27, 7,0), -/*1998*/ SPY(28, 7,0), SPY(29, 7,0), SPY(30, 7,0), SPY(31, 8,0), -/*2002*/ SPY(32, 8,0), SPY(33, 8,0), SPY(34, 8,0), SPY(35, 9,0), -/*2006*/ SPY(36, 9,0), SPY(37, 9,0), SPY(38, 9,0), SPY(39,10,0), -/*2010*/ SPY(40,10,0), SPY(41,10,0), SPY(42,10,0), SPY(43,11,0), -/*2014*/ SPY(44,11,0), SPY(45,11,0), SPY(46,11,0), SPY(47,12,0), -/*2018*/ SPY(48,12,0), SPY(49,12,0), SPY(50,12,0), SPY(51,13,0), -/*2022*/ SPY(52,13,0), SPY(53,13,0), SPY(54,13,0), SPY(55,14,0), -/*2026*/ SPY(56,14,0), SPY(57,14,0), SPY(58,14,0), SPY(59,15,0), -/*2030*/ SPY(60,15,0), SPY(61,15,0), SPY(62,15,0), SPY(63,16,0), -/*2034*/ SPY(64,16,0), SPY(65,16,0), SPY(66,16,0), SPY(67,17,0), -/*2038*/ SPY(68,17,0) +#define SPY(y, l, s) (SPD * (365 * y + l) + s) + +static time_t year_seconds[MAX_YEAR_SECONDS] = { +/*1970*/ SPY(0, 0, 0), SPY(1, 0, 0), SPY(2, 0, 0), SPY(3, 1, 0), +/*1974*/ SPY(4, 1, 0), SPY(5, 1, 0), SPY(6, 1, 0), SPY(7, 2, 0), +/*1978*/ SPY(8, 2, 0), SPY(9, 2, 0), SPY(10, 2, 0), SPY(11, 3, 0), +/*1982*/ SPY(12, 3, 0), SPY(13, 3, 0), SPY(14, 3, 0), SPY(15, 4, 0), +/*1986*/ SPY(16, 4, 0), SPY(17, 4, 0), SPY(18, 4, 0), SPY(19, 5, 0), +/*1990*/ SPY(20, 5, 0), SPY(21, 5, 0), SPY(22, 5, 0), SPY(23, 6, 0), +/*1994*/ SPY(24, 6, 0), SPY(25, 6, 0), SPY(26, 6, 0), SPY(27, 7, 0), +/*1998*/ SPY(28, 7, 0), SPY(29, 7, 0), SPY(30, 7, 0), SPY(31, 8, 0), +/*2002*/ SPY(32, 8, 0), SPY(33, 8, 0), SPY(34, 8, 0), SPY(35, 9, 0), +/*2006*/ SPY(36, 9, 0), SPY(37, 9, 0), SPY(38, 9, 0), SPY(39, 10, 0), +/*2010*/ SPY(40, 10, 0), SPY(41, 10, 0), SPY(42, 10, 0), SPY(43, 11, 0), +/*2014*/ SPY(44, 11, 0), SPY(45, 11, 0), SPY(46, 11, 0), SPY(47, 12, 0), +/*2018*/ SPY(48, 12, 0), SPY(49, 12, 0), SPY(50, 12, 0), SPY(51, 13, 0), +/*2022*/ SPY(52, 13, 0), SPY(53, 13, 0), SPY(54, 13, 0), SPY(55, 14, 0), +/*2026*/ SPY(56, 14, 0), SPY(57, 14, 0), SPY(58, 14, 0), SPY(59, 15, 0), +/*2030*/ SPY(60, 15, 0), SPY(61, 15, 0), SPY(62, 15, 0), SPY(63, 16, 0), +/*2034*/ SPY(64, 16, 0), SPY(65, 16, 0), SPY(66, 16, 0), SPY(67, 17, 0), +/*2038*/ SPY(68, 17, 0) }; extern struct timezone sys_tz; @@ -115,7 +117,7 @@ time_t *udf_stamp_to_time(time_t *dest, long *dest_usec, kernel_timestamp src) return dest; } -kernel_timestamp *udf_time_to_stamp(kernel_timestamp * dest, struct timespec ts) +kernel_timestamp *udf_time_to_stamp(kernel_timestamp *dest, struct timespec ts) { long int days, rem, y; const unsigned short int *ip; @@ -137,7 +139,7 @@ kernel_timestamp *udf_time_to_stamp(kernel_timestamp * dest, struct timespec ts) dest->second = rem % 60; y = 1970; -#define DIV(a,b) ((a) / (b) - ((a) % (b) < 0)) +#define DIV(a, b) ((a) / (b) - ((a) % (b) < 0)) #define LEAPS_THRU_END_OF(y) (DIV (y, 4) - DIV (y, 100) + DIV (y, 400)) while (days < 0 || days >= (__isleap(y) ? 366 : 365)) { @@ -145,8 +147,8 @@ kernel_timestamp *udf_time_to_stamp(kernel_timestamp * dest, struct timespec ts) /* Adjust DAYS and Y to match the guessed year. */ days -= ((yg - y) * 365 - + LEAPS_THRU_END_OF (yg - 1) - - LEAPS_THRU_END_OF (y - 1)); + + LEAPS_THRU_END_OF(yg - 1) + - LEAPS_THRU_END_OF(y - 1)); y = yg; } dest->year = y; @@ -158,7 +160,8 @@ kernel_timestamp *udf_time_to_stamp(kernel_timestamp * dest, struct timespec ts) dest->day = days + 1; dest->centiseconds = ts.tv_nsec / 10000000; - dest->hundredsOfMicroseconds = (ts.tv_nsec / 1000 - dest->centiseconds * 10000) / 100; + dest->hundredsOfMicroseconds = (ts.tv_nsec / 1000 - + dest->centiseconds * 10000) / 100; dest->microseconds = (ts.tv_nsec / 1000 - dest->centiseconds * 10000 - dest->hundredsOfMicroseconds * 100); return dest; diff --git a/fs/udf/unicode.c b/fs/udf/unicode.c index 9e6099c26c2..e533b11703b 100644 --- a/fs/udf/unicode.c +++ b/fs/udf/unicode.c @@ -136,12 +136,18 @@ int udf_CS0toUTF8(struct ustr *utf_o, struct ustr *ocu_i) if (c < 0x80U) { utf_o->u_name[utf_o->u_len++] = (uint8_t)c; } else if (c < 0x800U) { - utf_o->u_name[utf_o->u_len++] = (uint8_t)(0xc0 | (c >> 6)); - utf_o->u_name[utf_o->u_len++] = (uint8_t)(0x80 | (c & 0x3f)); + utf_o->u_name[utf_o->u_len++] = + (uint8_t)(0xc0 | (c >> 6)); + utf_o->u_name[utf_o->u_len++] = + (uint8_t)(0x80 | (c & 0x3f)); } else { - utf_o->u_name[utf_o->u_len++] = (uint8_t)(0xe0 | (c >> 12)); - utf_o->u_name[utf_o->u_len++] = (uint8_t)(0x80 | ((c >> 6) & 0x3f)); - utf_o->u_name[utf_o->u_len++] = (uint8_t)(0x80 | (c & 0x3f)); + utf_o->u_name[utf_o->u_len++] = + (uint8_t)(0xe0 | (c >> 12)); + utf_o->u_name[utf_o->u_len++] = + (uint8_t)(0x80 | + ((c >> 6) & 0x3f)); + utf_o->u_name[utf_o->u_len++] = + (uint8_t)(0x80 | (c & 0x3f)); } } utf_o->u_cmpID = 8; @@ -232,9 +238,8 @@ try_again: goto error_out; } - if (max_val == 0xffffU) { + if (max_val == 0xffffU) ocu[++u_len] = (uint8_t)(utf_char >> 8); - } ocu[++u_len] = (uint8_t)(utf_char & 0xffU); } @@ -330,29 +335,29 @@ int udf_get_filename(struct super_block *sb, uint8_t *sname, uint8_t *dname, struct ustr filename, unifilename; int len; - if (udf_build_ustr_exact(&unifilename, sname, flen)) { + if (udf_build_ustr_exact(&unifilename, sname, flen)) return 0; - } if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) { if (!udf_CS0toUTF8(&filename, &unifilename)) { - udf_debug("Failed in udf_get_filename: sname = %s\n", sname); + udf_debug("Failed in udf_get_filename: sname = %s\n", + sname); return 0; } } else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) { - if (!udf_CS0toNLS(UDF_SB(sb)->s_nls_map, &filename, &unifilename)) { - udf_debug("Failed in udf_get_filename: sname = %s\n", sname); + if (!udf_CS0toNLS(UDF_SB(sb)->s_nls_map, &filename, + &unifilename)) { + udf_debug("Failed in udf_get_filename: sname = %s\n", + sname); return 0; } - } else { + } else return 0; - } len = udf_translate_to_linux(dname, filename.u_name, filename.u_len, unifilename.u_name, unifilename.u_len); - if (len) { + if (len) return len; - } return 0; } @@ -363,23 +368,20 @@ int udf_put_filename(struct super_block *sb, const uint8_t *sname, struct ustr unifilename; int namelen; - if (!(udf_char_to_ustr(&unifilename, sname, flen))) { + if (!udf_char_to_ustr(&unifilename, sname, flen)) return 0; - } if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) { namelen = udf_UTF8toCS0(dname, &unifilename, UDF_NAME_LEN); - if (!namelen) { + if (!namelen) return 0; - } } else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) { - namelen = udf_NLStoCS0(UDF_SB(sb)->s_nls_map, dname, &unifilename, UDF_NAME_LEN); - if (!namelen) { + namelen = udf_NLStoCS0(UDF_SB(sb)->s_nls_map, dname, + &unifilename, UDF_NAME_LEN); + if (!namelen) return 0; - } - } else { + } else return 0; - } return namelen; } @@ -389,8 +391,9 @@ int udf_put_filename(struct super_block *sb, const uint8_t *sname, #define CRC_MARK '#' #define EXT_SIZE 5 -static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName, int udfLen, - uint8_t *fidName, int fidNameLen) +static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName, + int udfLen, uint8_t *fidName, + int fidNameLen) { int index, newIndex = 0, needsCRC = 0; int extIndex = 0, newExtIndex = 0, hasExt = 0; @@ -409,13 +412,16 @@ static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName, int udfLen if (curr == '/' || curr == 0) { needsCRC = 1; curr = ILLEGAL_CHAR_MARK; - while (index + 1 < udfLen && (udfName[index + 1] == '/' || - udfName[index + 1] == 0)) + while (index + 1 < udfLen && + (udfName[index + 1] == '/' || + udfName[index + 1] == 0)) index++; - } if (curr == EXT_MARK && (udfLen - index - 1) <= EXT_SIZE) { - if (udfLen == index + 1) { + } + if (curr == EXT_MARK && + (udfLen - index - 1) <= EXT_SIZE) { + if (udfLen == index + 1) hasExt = 0; - } else { + else { hasExt = 1; extIndex = index; newExtIndex = newIndex; @@ -433,16 +439,18 @@ static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName, int udfLen if (hasExt) { int maxFilenameLen; - for(index = 0; index < EXT_SIZE && extIndex + index + 1 < udfLen; index++) { + for (index = 0; + index < EXT_SIZE && extIndex + index + 1 < udfLen; + index++) { curr = udfName[extIndex + index + 1]; if (curr == '/' || curr == 0) { needsCRC = 1; curr = ILLEGAL_CHAR_MARK; - while(extIndex + index + 2 < udfLen && - (index + 1 < EXT_SIZE - && (udfName[extIndex + index + 2] == '/' || - udfName[extIndex + index + 2] == 0))) + while (extIndex + index + 2 < udfLen && + (index + 1 < EXT_SIZE && + (udfName[extIndex + index + 2] == '/' || + udfName[extIndex + index + 2] == 0))) index++; } ext[localExtIndex++] = curr; @@ -452,9 +460,8 @@ static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName, int udfLen newIndex = maxFilenameLen; else newIndex = newExtIndex; - } else if (newIndex > 250) { + } else if (newIndex > 250) newIndex = 250; - } newName[newIndex++] = CRC_MARK; valueCRC = udf_crc(fidName, fidNameLen, 0); newName[newIndex++] = hexChar[(valueCRC & 0xf000) >> 12]; diff --git a/fs/ufs/balloc.c b/fs/ufs/balloc.c index f63a09ce868..1e7598fb978 100644 --- a/fs/ufs/balloc.c +++ b/fs/ufs/balloc.c @@ -9,7 +9,6 @@ */ #include <linux/fs.h> -#include <linux/ufs_fs.h> #include <linux/stat.h> #include <linux/time.h> #include <linux/string.h> @@ -19,6 +18,7 @@ #include <linux/bitops.h> #include <asm/byteorder.h> +#include "ufs_fs.h" #include "ufs.h" #include "swab.h" #include "util.h" @@ -315,8 +315,8 @@ static void ufs_change_blocknr(struct inode *inode, sector_t beg, } UFSD(" change from %llu to %llu, pos %u\n", - (unsigned long long)pos + oldb, - (unsigned long long)pos + newb, pos); + (unsigned long long)(pos + oldb), + (unsigned long long)(pos + newb), pos); bh->b_blocknr = newb + pos; unmap_underlying_metadata(bh->b_bdev, diff --git a/fs/ufs/cylinder.c b/fs/ufs/cylinder.c index 2a815665644..b4676322ddb 100644 --- a/fs/ufs/cylinder.c +++ b/fs/ufs/cylinder.c @@ -9,7 +9,6 @@ */ #include <linux/fs.h> -#include <linux/ufs_fs.h> #include <linux/time.h> #include <linux/stat.h> #include <linux/string.h> @@ -17,6 +16,7 @@ #include <asm/byteorder.h> +#include "ufs_fs.h" #include "ufs.h" #include "swab.h" #include "util.h" diff --git a/fs/ufs/dir.c b/fs/ufs/dir.c index aaf2878305c..ef563fc8d72 100644 --- a/fs/ufs/dir.c +++ b/fs/ufs/dir.c @@ -18,9 +18,9 @@ #include <linux/time.h> #include <linux/fs.h> -#include <linux/ufs_fs.h> #include <linux/swap.h> +#include "ufs_fs.h" #include "ufs.h" #include "swab.h" #include "util.h" diff --git a/fs/ufs/file.c b/fs/ufs/file.c index a46c97bf023..625ef17c6f8 100644 --- a/fs/ufs/file.c +++ b/fs/ufs/file.c @@ -24,9 +24,9 @@ */ #include <linux/fs.h> -#include <linux/ufs_fs.h> #include <linux/buffer_head.h> /* for sync_mapping_buffers() */ +#include "ufs_fs.h" #include "ufs.h" diff --git a/fs/ufs/ialloc.c b/fs/ufs/ialloc.c index 7e260bc0d94..ac181f6806a 100644 --- a/fs/ufs/ialloc.c +++ b/fs/ufs/ialloc.c @@ -24,7 +24,6 @@ */ #include <linux/fs.h> -#include <linux/ufs_fs.h> #include <linux/time.h> #include <linux/stat.h> #include <linux/string.h> @@ -34,6 +33,7 @@ #include <linux/bitops.h> #include <asm/byteorder.h> +#include "ufs_fs.h" #include "ufs.h" #include "swab.h" #include "util.h" diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c index 4320782761a..5446b888fc8 100644 --- a/fs/ufs/inode.c +++ b/fs/ufs/inode.c @@ -30,7 +30,6 @@ #include <linux/errno.h> #include <linux/fs.h> -#include <linux/ufs_fs.h> #include <linux/time.h> #include <linux/stat.h> #include <linux/string.h> @@ -38,6 +37,7 @@ #include <linux/smp_lock.h> #include <linux/buffer_head.h> +#include "ufs_fs.h" #include "ufs.h" #include "swab.h" #include "util.h" @@ -714,26 +714,30 @@ static int ufs2_read_inode(struct inode *inode, struct ufs2_inode *ufs2_inode) return 0; } -void ufs_read_inode(struct inode * inode) +struct inode *ufs_iget(struct super_block *sb, unsigned long ino) { - struct ufs_inode_info *ufsi = UFS_I(inode); - struct super_block * sb; - struct ufs_sb_private_info * uspi; + struct ufs_inode_info *ufsi; + struct ufs_sb_private_info *uspi = UFS_SB(sb)->s_uspi; struct buffer_head * bh; + struct inode *inode; int err; - UFSD("ENTER, ino %lu\n", inode->i_ino); - - sb = inode->i_sb; - uspi = UFS_SB(sb)->s_uspi; + UFSD("ENTER, ino %lu\n", ino); - if (inode->i_ino < UFS_ROOTINO || - inode->i_ino > (uspi->s_ncg * uspi->s_ipg)) { + if (ino < UFS_ROOTINO || ino > (uspi->s_ncg * uspi->s_ipg)) { ufs_warning(sb, "ufs_read_inode", "bad inode number (%lu)\n", - inode->i_ino); - goto bad_inode; + ino); + return ERR_PTR(-EIO); } + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; + + ufsi = UFS_I(inode); + bh = sb_bread(sb, uspi->s_sbbase + ufs_inotofsba(inode->i_ino)); if (!bh) { ufs_warning(sb, "ufs_read_inode", "unable to read inode %lu\n", @@ -765,10 +769,12 @@ void ufs_read_inode(struct inode * inode) brelse(bh); UFSD("EXIT\n"); - return; + unlock_new_inode(inode); + return inode; bad_inode: - make_bad_inode(inode); + iget_failed(inode); + return ERR_PTR(-EIO); } static void ufs1_update_inode(struct inode *inode, struct ufs_inode *ufs_inode) diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c index d8bfbee2fe2..e3a9b1fac75 100644 --- a/fs/ufs/namei.c +++ b/fs/ufs/namei.c @@ -29,8 +29,9 @@ #include <linux/time.h> #include <linux/fs.h> -#include <linux/ufs_fs.h> #include <linux/smp_lock.h> + +#include "ufs_fs.h" #include "ufs.h" #include "util.h" @@ -57,10 +58,10 @@ static struct dentry *ufs_lookup(struct inode * dir, struct dentry *dentry, stru lock_kernel(); ino = ufs_inode_by_name(dir, dentry); if (ino) { - inode = iget(dir->i_sb, ino); - if (!inode) { + inode = ufs_iget(dir->i_sb, ino); + if (IS_ERR(inode)) { unlock_kernel(); - return ERR_PTR(-EACCES); + return ERR_CAST(inode); } } unlock_kernel(); diff --git a/fs/ufs/super.c b/fs/ufs/super.c index 0072cb33ebe..85b22b5977f 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -76,7 +76,6 @@ #include <linux/errno.h> #include <linux/fs.h> -#include <linux/ufs_fs.h> #include <linux/slab.h> #include <linux/time.h> #include <linux/stat.h> @@ -91,6 +90,7 @@ #include <linux/mount.h> #include <linux/seq_file.h> +#include "ufs_fs.h" #include "ufs.h" #include "swab.h" #include "util.h" @@ -131,6 +131,8 @@ static void ufs_print_super_stuff(struct super_block *sb, printk(KERN_INFO" cs_nffree(Num of free frags): %llu\n", (unsigned long long) fs64_to_cpu(sb, usb3->fs_un1.fs_u2.cs_nffree)); + printk(KERN_INFO" fs_maxsymlinklen: %u\n", + fs32_to_cpu(sb, usb3->fs_un2.fs_44.fs_maxsymlinklen)); } else { printk(" sblkno: %u\n", fs32_to_cpu(sb, usb1->fs_sblkno)); printk(" cblkno: %u\n", fs32_to_cpu(sb, usb1->fs_cblkno)); @@ -633,6 +635,7 @@ static int ufs_fill_super(struct super_block *sb, void *data, int silent) unsigned block_size, super_block_size; unsigned flags; unsigned super_block_offset; + int ret = -EINVAL; uspi = NULL; ubh = NULL; @@ -1060,17 +1063,21 @@ magic_found: uspi->s_bpf = uspi->s_fsize << 3; uspi->s_bpfshift = uspi->s_fshift + 3; uspi->s_bpfmask = uspi->s_bpf - 1; - if ((sbi->s_mount_opt & UFS_MOUNT_UFSTYPE) == - UFS_MOUNT_UFSTYPE_44BSD) + if ((sbi->s_mount_opt & UFS_MOUNT_UFSTYPE) == UFS_MOUNT_UFSTYPE_44BSD || + (sbi->s_mount_opt & UFS_MOUNT_UFSTYPE) == UFS_MOUNT_UFSTYPE_UFS2) uspi->s_maxsymlinklen = fs32_to_cpu(sb, usb3->fs_un2.fs_44.fs_maxsymlinklen); - inode = iget(sb, UFS_ROOTINO); - if (!inode || is_bad_inode(inode)) + inode = ufs_iget(sb, UFS_ROOTINO); + if (IS_ERR(inode)) { + ret = PTR_ERR(inode); goto failed; + } sb->s_root = d_alloc_root(inode); - if (!sb->s_root) + if (!sb->s_root) { + ret = -ENOMEM; goto dalloc_failed; + } ufs_setup_cstotal(sb); /* @@ -1092,7 +1099,7 @@ failed: kfree(sbi); sb->s_fs_info = NULL; UFSD("EXIT (FAILED)\n"); - return -EINVAL; + return ret; failed_nomem: UFSD("EXIT (NOMEM)\n"); @@ -1326,7 +1333,6 @@ static ssize_t ufs_quota_write(struct super_block *, int, const char *, size_t, static const struct super_operations ufs_super_ops = { .alloc_inode = ufs_alloc_inode, .destroy_inode = ufs_destroy_inode, - .read_inode = ufs_read_inode, .write_inode = ufs_write_inode, .delete_inode = ufs_delete_inode, .put_super = ufs_put_super, diff --git a/fs/ufs/symlink.c b/fs/ufs/symlink.c index 43ac10e75a4..c0156eda44b 100644 --- a/fs/ufs/symlink.c +++ b/fs/ufs/symlink.c @@ -27,7 +27,8 @@ #include <linux/fs.h> #include <linux/namei.h> -#include <linux/ufs_fs.h> + +#include "ufs_fs.h" #include "ufs.h" diff --git a/fs/ufs/truncate.c b/fs/ufs/truncate.c index 311ded34c2b..41dd431ce22 100644 --- a/fs/ufs/truncate.c +++ b/fs/ufs/truncate.c @@ -36,7 +36,6 @@ #include <linux/errno.h> #include <linux/fs.h> -#include <linux/ufs_fs.h> #include <linux/fcntl.h> #include <linux/time.h> #include <linux/stat.h> @@ -46,6 +45,7 @@ #include <linux/blkdev.h> #include <linux/sched.h> +#include "ufs_fs.h" #include "ufs.h" #include "swab.h" #include "util.h" diff --git a/fs/ufs/ufs.h b/fs/ufs/ufs.h index 7faa4cd71a2..fcb9231bb9e 100644 --- a/fs/ufs/ufs.h +++ b/fs/ufs/ufs.h @@ -106,7 +106,7 @@ extern void ufs_free_inode (struct inode *inode); extern struct inode * ufs_new_inode (struct inode *, int); /* inode.c */ -extern void ufs_read_inode (struct inode *); +extern struct inode *ufs_iget(struct super_block *, unsigned long); extern void ufs_put_inode (struct inode *); extern int ufs_write_inode (struct inode *, int); extern int ufs_sync_inode (struct inode *); diff --git a/fs/ufs/ufs_fs.h b/fs/ufs/ufs_fs.h new file mode 100644 index 00000000000..54bde1895a8 --- /dev/null +++ b/fs/ufs/ufs_fs.h @@ -0,0 +1,947 @@ +/* + * linux/include/linux/ufs_fs.h + * + * Copyright (C) 1996 + * Adrian Rodriguez (adrian@franklins-tower.rutgers.edu) + * Laboratory for Computer Science Research Computing Facility + * Rutgers, The State University of New Jersey + * + * Clean swab support by Fare <fare@tunes.org> + * just hope no one is using NNUUXXI on __?64 structure elements + * 64-bit clean thanks to Maciej W. Rozycki <macro@ds2.pg.gda.pl> + * + * 4.4BSD (FreeBSD) support added on February 1st 1998 by + * Niels Kristian Bech Jensen <nkbj@image.dk> partially based + * on code by Martin von Loewis <martin@mira.isdn.cs.tu-berlin.de>. + * + * NeXTstep support added on February 5th 1998 by + * Niels Kristian Bech Jensen <nkbj@image.dk>. + * + * Write support by Daniel Pirkl <daniel.pirkl@email.cz> + * + * HP/UX hfs filesystem support added by + * Martin K. Petersen <mkp@mkp.net>, August 1999 + * + * UFS2 (of FreeBSD 5.x) support added by + * Niraj Kumar <niraj17@iitbombay.org> , Jan 2004 + * + */ + +#ifndef __LINUX_UFS_FS_H +#define __LINUX_UFS_FS_H + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/stat.h> +#include <linux/fs.h> + +#include <asm/div64.h> +typedef __u64 __bitwise __fs64; +typedef __u32 __bitwise __fs32; +typedef __u16 __bitwise __fs16; + +#define UFS_BBLOCK 0 +#define UFS_BBSIZE 8192 +#define UFS_SBLOCK 8192 +#define UFS_SBSIZE 8192 + +#define UFS_SECTOR_SIZE 512 +#define UFS_SECTOR_BITS 9 +#define UFS_MAGIC 0x00011954 +#define UFS2_MAGIC 0x19540119 +#define UFS_CIGAM 0x54190100 /* byteswapped MAGIC */ + +/* Copied from FreeBSD */ +/* + * Each disk drive contains some number of filesystems. + * A filesystem consists of a number of cylinder groups. + * Each cylinder group has inodes and data. + * + * A filesystem is described by its super-block, which in turn + * describes the cylinder groups. The super-block is critical + * data and is replicated in each cylinder group to protect against + * catastrophic loss. This is done at `newfs' time and the critical + * super-block data does not change, so the copies need not be + * referenced further unless disaster strikes. + * + * For filesystem fs, the offsets of the various blocks of interest + * are given in the super block as: + * [fs->fs_sblkno] Super-block + * [fs->fs_cblkno] Cylinder group block + * [fs->fs_iblkno] Inode blocks + * [fs->fs_dblkno] Data blocks + * The beginning of cylinder group cg in fs, is given by + * the ``cgbase(fs, cg)'' macro. + * + * Depending on the architecture and the media, the superblock may + * reside in any one of four places. For tiny media where every block + * counts, it is placed at the very front of the partition. Historically, + * UFS1 placed it 8K from the front to leave room for the disk label and + * a small bootstrap. For UFS2 it got moved to 64K from the front to leave + * room for the disk label and a bigger bootstrap, and for really piggy + * systems we check at 256K from the front if the first three fail. In + * all cases the size of the superblock will be SBLOCKSIZE. All values are + * given in byte-offset form, so they do not imply a sector size. The + * SBLOCKSEARCH specifies the order in which the locations should be searched. + */ +#define SBLOCK_FLOPPY 0 +#define SBLOCK_UFS1 8192 +#define SBLOCK_UFS2 65536 +#define SBLOCK_PIGGY 262144 +#define SBLOCKSIZE 8192 +#define SBLOCKSEARCH \ + { SBLOCK_UFS2, SBLOCK_UFS1, SBLOCK_FLOPPY, SBLOCK_PIGGY, -1 } + + +/* HP specific MAGIC values */ + +#define UFS_MAGIC_LFN 0x00095014 /* fs supports filenames > 14 chars */ +#define UFS_CIGAM_LFN 0x14500900 /* srahc 41 < semanelif stroppus sf */ + +#define UFS_MAGIC_SEC 0x00612195 /* B1 security fs */ +#define UFS_CIGAM_SEC 0x95216100 + +#define UFS_MAGIC_FEA 0x00195612 /* fs_featurebits supported */ +#define UFS_CIGAM_FEA 0x12561900 + +#define UFS_MAGIC_4GB 0x05231994 /* fs > 4 GB && fs_featurebits */ +#define UFS_CIGAM_4GB 0x94192305 + +/* Seems somebody at HP goofed here. B1 and lfs are both 0x2 !?! */ +#define UFS_FSF_LFN 0x00000001 /* long file names */ +#define UFS_FSF_B1 0x00000002 /* B1 security */ +#define UFS_FSF_LFS 0x00000002 /* large files */ +#define UFS_FSF_LUID 0x00000004 /* large UIDs */ + +/* End of HP stuff */ + + +#define UFS_BSIZE 8192 +#define UFS_MINBSIZE 4096 +#define UFS_FSIZE 1024 +#define UFS_MAXFRAG (UFS_BSIZE / UFS_FSIZE) + +#define UFS_NDADDR 12 +#define UFS_NINDIR 3 + +#define UFS_IND_BLOCK (UFS_NDADDR + 0) +#define UFS_DIND_BLOCK (UFS_NDADDR + 1) +#define UFS_TIND_BLOCK (UFS_NDADDR + 2) + +#define UFS_NDIR_FRAGMENT (UFS_NDADDR << uspi->s_fpbshift) +#define UFS_IND_FRAGMENT (UFS_IND_BLOCK << uspi->s_fpbshift) +#define UFS_DIND_FRAGMENT (UFS_DIND_BLOCK << uspi->s_fpbshift) +#define UFS_TIND_FRAGMENT (UFS_TIND_BLOCK << uspi->s_fpbshift) + +#define UFS_ROOTINO 2 +#define UFS_FIRST_INO (UFS_ROOTINO + 1) + +#define UFS_USEEFT ((__u16)65535) + +#define UFS_FSOK 0x7c269d38 +#define UFS_FSACTIVE ((__s8)0x00) +#define UFS_FSCLEAN ((__s8)0x01) +#define UFS_FSSTABLE ((__s8)0x02) +#define UFS_FSOSF1 ((__s8)0x03) /* is this correct for DEC OSF/1? */ +#define UFS_FSBAD ((__s8)0xff) + +/* From here to next blank line, s_flags for ufs_sb_info */ +/* directory entry encoding */ +#define UFS_DE_MASK 0x00000010 /* mask for the following */ +#define UFS_DE_OLD 0x00000000 +#define UFS_DE_44BSD 0x00000010 +/* uid encoding */ +#define UFS_UID_MASK 0x00000060 /* mask for the following */ +#define UFS_UID_OLD 0x00000000 +#define UFS_UID_44BSD 0x00000020 +#define UFS_UID_EFT 0x00000040 +/* superblock state encoding */ +#define UFS_ST_MASK 0x00000700 /* mask for the following */ +#define UFS_ST_OLD 0x00000000 +#define UFS_ST_44BSD 0x00000100 +#define UFS_ST_SUN 0x00000200 /* Solaris */ +#define UFS_ST_SUNOS 0x00000300 +#define UFS_ST_SUNx86 0x00000400 /* Solaris x86 */ +/*cylinder group encoding */ +#define UFS_CG_MASK 0x00003000 /* mask for the following */ +#define UFS_CG_OLD 0x00000000 +#define UFS_CG_44BSD 0x00002000 +#define UFS_CG_SUN 0x00001000 +/* filesystem type encoding */ +#define UFS_TYPE_MASK 0x00010000 /* mask for the following */ +#define UFS_TYPE_UFS1 0x00000000 +#define UFS_TYPE_UFS2 0x00010000 + + +/* fs_inodefmt options */ +#define UFS_42INODEFMT -1 +#define UFS_44INODEFMT 2 + +/* + * MINFREE gives the minimum acceptable percentage of file system + * blocks which may be free. If the freelist drops below this level + * only the superuser may continue to allocate blocks. This may + * be set to 0 if no reserve of free blocks is deemed necessary, + * however throughput drops by fifty percent if the file system + * is run at between 95% and 100% full; thus the minimum default + * value of fs_minfree is 5%. However, to get good clustering + * performance, 10% is a better choice. hence we use 10% as our + * default value. With 10% free space, fragmentation is not a + * problem, so we choose to optimize for time. + */ +#define UFS_MINFREE 5 +#define UFS_DEFAULTOPT UFS_OPTTIME + +/* + * Turn file system block numbers into disk block addresses. + * This maps file system blocks to device size blocks. + */ +#define ufs_fsbtodb(uspi, b) ((b) << (uspi)->s_fsbtodb) +#define ufs_dbtofsb(uspi, b) ((b) >> (uspi)->s_fsbtodb) + +/* + * Cylinder group macros to locate things in cylinder groups. + * They calc file system addresses of cylinder group data structures. + */ +#define ufs_cgbase(c) (uspi->s_fpg * (c)) +#define ufs_cgstart(c) ((uspi)->fs_magic == UFS2_MAGIC ? ufs_cgbase(c) : \ + (ufs_cgbase(c) + uspi->s_cgoffset * ((c) & ~uspi->s_cgmask))) +#define ufs_cgsblock(c) (ufs_cgstart(c) + uspi->s_sblkno) /* super blk */ +#define ufs_cgcmin(c) (ufs_cgstart(c) + uspi->s_cblkno) /* cg block */ +#define ufs_cgimin(c) (ufs_cgstart(c) + uspi->s_iblkno) /* inode blk */ +#define ufs_cgdmin(c) (ufs_cgstart(c) + uspi->s_dblkno) /* 1st data */ + +/* + * Macros for handling inode numbers: + * inode number to file system block offset. + * inode number to cylinder group number. + * inode number to file system block address. + */ +#define ufs_inotocg(x) ((x) / uspi->s_ipg) +#define ufs_inotocgoff(x) ((x) % uspi->s_ipg) +#define ufs_inotofsba(x) (((u64)ufs_cgimin(ufs_inotocg(x))) + ufs_inotocgoff(x) / uspi->s_inopf) +#define ufs_inotofsbo(x) ((x) % uspi->s_inopf) + +/* + * Compute the cylinder and rotational position of a cyl block addr. + */ +#define ufs_cbtocylno(bno) \ + ((bno) * uspi->s_nspf / uspi->s_spc) +#define ufs_cbtorpos(bno) \ + ((((bno) * uspi->s_nspf % uspi->s_spc / uspi->s_nsect \ + * uspi->s_trackskew + (bno) * uspi->s_nspf % uspi->s_spc \ + % uspi->s_nsect * uspi->s_interleave) % uspi->s_nsect \ + * uspi->s_nrpos) / uspi->s_npsect) + +/* + * The following macros optimize certain frequently calculated + * quantities by using shifts and masks in place of divisions + * modulos and multiplications. + */ +#define ufs_blkoff(loc) ((loc) & uspi->s_qbmask) +#define ufs_fragoff(loc) ((loc) & uspi->s_qfmask) +#define ufs_lblktosize(blk) ((blk) << uspi->s_bshift) +#define ufs_lblkno(loc) ((loc) >> uspi->s_bshift) +#define ufs_numfrags(loc) ((loc) >> uspi->s_fshift) +#define ufs_blkroundup(size) (((size) + uspi->s_qbmask) & uspi->s_bmask) +#define ufs_fragroundup(size) (((size) + uspi->s_qfmask) & uspi->s_fmask) +#define ufs_fragstoblks(frags) ((frags) >> uspi->s_fpbshift) +#define ufs_blkstofrags(blks) ((blks) << uspi->s_fpbshift) +#define ufs_fragnum(fsb) ((fsb) & uspi->s_fpbmask) +#define ufs_blknum(fsb) ((fsb) & ~uspi->s_fpbmask) + +#define UFS_MAXNAMLEN 255 +#define UFS_MAXMNTLEN 512 +#define UFS2_MAXMNTLEN 468 +#define UFS2_MAXVOLLEN 32 +#define UFS_MAXCSBUFS 31 +#define UFS_LINK_MAX 32000 +/* +#define UFS2_NOCSPTRS ((128 / sizeof(void *)) - 4) +*/ +#define UFS2_NOCSPTRS 28 + +/* + * UFS_DIR_PAD defines the directory entries boundaries + * (must be a multiple of 4) + */ +#define UFS_DIR_PAD 4 +#define UFS_DIR_ROUND (UFS_DIR_PAD - 1) +#define UFS_DIR_REC_LEN(name_len) (((name_len) + 1 + 8 + UFS_DIR_ROUND) & ~UFS_DIR_ROUND) + +struct ufs_timeval { + __fs32 tv_sec; + __fs32 tv_usec; +}; + +struct ufs_dir_entry { + __fs32 d_ino; /* inode number of this entry */ + __fs16 d_reclen; /* length of this entry */ + union { + __fs16 d_namlen; /* actual length of d_name */ + struct { + __u8 d_type; /* file type */ + __u8 d_namlen; /* length of string in d_name */ + } d_44; + } d_u; + __u8 d_name[UFS_MAXNAMLEN + 1]; /* file name */ +}; + +struct ufs_csum { + __fs32 cs_ndir; /* number of directories */ + __fs32 cs_nbfree; /* number of free blocks */ + __fs32 cs_nifree; /* number of free inodes */ + __fs32 cs_nffree; /* number of free frags */ +}; +struct ufs2_csum_total { + __fs64 cs_ndir; /* number of directories */ + __fs64 cs_nbfree; /* number of free blocks */ + __fs64 cs_nifree; /* number of free inodes */ + __fs64 cs_nffree; /* number of free frags */ + __fs64 cs_numclusters; /* number of free clusters */ + __fs64 cs_spare[3]; /* future expansion */ +}; + +struct ufs_csum_core { + __u64 cs_ndir; /* number of directories */ + __u64 cs_nbfree; /* number of free blocks */ + __u64 cs_nifree; /* number of free inodes */ + __u64 cs_nffree; /* number of free frags */ + __u64 cs_numclusters; /* number of free clusters */ +}; + +/* + * File system flags + */ +#define UFS_UNCLEAN 0x01 /* file system not clean at mount (unused) */ +#define UFS_DOSOFTDEP 0x02 /* file system using soft dependencies */ +#define UFS_NEEDSFSCK 0x04 /* needs sync fsck (FreeBSD compat, unused) */ +#define UFS_INDEXDIRS 0x08 /* kernel supports indexed directories */ +#define UFS_ACLS 0x10 /* file system has ACLs enabled */ +#define UFS_MULTILABEL 0x20 /* file system is MAC multi-label */ +#define UFS_FLAGS_UPDATED 0x80 /* flags have been moved to new location */ + +#if 0 +/* + * This is the actual superblock, as it is laid out on the disk. + * Do NOT use this structure, because of sizeof(ufs_super_block) > 512 and + * it may occupy several blocks, use + * struct ufs_super_block_(first,second,third) instead. + */ +struct ufs_super_block { + union { + struct { + __fs32 fs_link; /* UNUSED */ + } fs_42; + struct { + __fs32 fs_state; /* file system state flag */ + } fs_sun; + } fs_u0; + __fs32 fs_rlink; /* UNUSED */ + __fs32 fs_sblkno; /* addr of super-block in filesys */ + __fs32 fs_cblkno; /* offset of cyl-block in filesys */ + __fs32 fs_iblkno; /* offset of inode-blocks in filesys */ + __fs32 fs_dblkno; /* offset of first data after cg */ + __fs32 fs_cgoffset; /* cylinder group offset in cylinder */ + __fs32 fs_cgmask; /* used to calc mod fs_ntrak */ + __fs32 fs_time; /* last time written -- time_t */ + __fs32 fs_size; /* number of blocks in fs */ + __fs32 fs_dsize; /* number of data blocks in fs */ + __fs32 fs_ncg; /* number of cylinder groups */ + __fs32 fs_bsize; /* size of basic blocks in fs */ + __fs32 fs_fsize; /* size of frag blocks in fs */ + __fs32 fs_frag; /* number of frags in a block in fs */ +/* these are configuration parameters */ + __fs32 fs_minfree; /* minimum percentage of free blocks */ + __fs32 fs_rotdelay; /* num of ms for optimal next block */ + __fs32 fs_rps; /* disk revolutions per second */ +/* these fields can be computed from the others */ + __fs32 fs_bmask; /* ``blkoff'' calc of blk offsets */ + __fs32 fs_fmask; /* ``fragoff'' calc of frag offsets */ + __fs32 fs_bshift; /* ``lblkno'' calc of logical blkno */ + __fs32 fs_fshift; /* ``numfrags'' calc number of frags */ +/* these are configuration parameters */ + __fs32 fs_maxcontig; /* max number of contiguous blks */ + __fs32 fs_maxbpg; /* max number of blks per cyl group */ +/* these fields can be computed from the others */ + __fs32 fs_fragshift; /* block to frag shift */ + __fs32 fs_fsbtodb; /* fsbtodb and dbtofsb shift constant */ + __fs32 fs_sbsize; /* actual size of super block */ + __fs32 fs_csmask; /* csum block offset */ + __fs32 fs_csshift; /* csum block number */ + __fs32 fs_nindir; /* value of NINDIR */ + __fs32 fs_inopb; /* value of INOPB */ + __fs32 fs_nspf; /* value of NSPF */ +/* yet another configuration parameter */ + __fs32 fs_optim; /* optimization preference, see below */ +/* these fields are derived from the hardware */ + union { + struct { + __fs32 fs_npsect; /* # sectors/track including spares */ + } fs_sun; + struct { + __fs32 fs_state; /* file system state time stamp */ + } fs_sunx86; + } fs_u1; + __fs32 fs_interleave; /* hardware sector interleave */ + __fs32 fs_trackskew; /* sector 0 skew, per track */ +/* a unique id for this filesystem (currently unused and unmaintained) */ +/* In 4.3 Tahoe this space is used by fs_headswitch and fs_trkseek */ +/* Neither of those fields is used in the Tahoe code right now but */ +/* there could be problems if they are. */ + __fs32 fs_id[2]; /* file system id */ +/* sizes determined by number of cylinder groups and their sizes */ + __fs32 fs_csaddr; /* blk addr of cyl grp summary area */ + __fs32 fs_cssize; /* size of cyl grp summary area */ + __fs32 fs_cgsize; /* cylinder group size */ +/* these fields are derived from the hardware */ + __fs32 fs_ntrak; /* tracks per cylinder */ + __fs32 fs_nsect; /* sectors per track */ + __fs32 fs_spc; /* sectors per cylinder */ +/* this comes from the disk driver partitioning */ + __fs32 fs_ncyl; /* cylinders in file system */ +/* these fields can be computed from the others */ + __fs32 fs_cpg; /* cylinders per group */ + __fs32 fs_ipg; /* inodes per cylinder group */ + __fs32 fs_fpg; /* blocks per group * fs_frag */ +/* this data must be re-computed after crashes */ + struct ufs_csum fs_cstotal; /* cylinder summary information */ +/* these fields are cleared at mount time */ + __s8 fs_fmod; /* super block modified flag */ + __s8 fs_clean; /* file system is clean flag */ + __s8 fs_ronly; /* mounted read-only flag */ + __s8 fs_flags; + union { + struct { + __s8 fs_fsmnt[UFS_MAXMNTLEN];/* name mounted on */ + __fs32 fs_cgrotor; /* last cg searched */ + __fs32 fs_csp[UFS_MAXCSBUFS];/*list of fs_cs info buffers */ + __fs32 fs_maxcluster; + __fs32 fs_cpc; /* cyl per cycle in postbl */ + __fs16 fs_opostbl[16][8]; /* old rotation block list head */ + } fs_u1; + struct { + __s8 fs_fsmnt[UFS2_MAXMNTLEN]; /* name mounted on */ + __u8 fs_volname[UFS2_MAXVOLLEN]; /* volume name */ + __fs64 fs_swuid; /* system-wide uid */ + __fs32 fs_pad; /* due to alignment of fs_swuid */ + __fs32 fs_cgrotor; /* last cg searched */ + __fs32 fs_ocsp[UFS2_NOCSPTRS]; /*list of fs_cs info buffers */ + __fs32 fs_contigdirs;/*# of contiguously allocated dirs */ + __fs32 fs_csp; /* cg summary info buffer for fs_cs */ + __fs32 fs_maxcluster; + __fs32 fs_active;/* used by snapshots to track fs */ + __fs32 fs_old_cpc; /* cyl per cycle in postbl */ + __fs32 fs_maxbsize;/*maximum blocking factor permitted */ + __fs64 fs_sparecon64[17];/*old rotation block list head */ + __fs64 fs_sblockloc; /* byte offset of standard superblock */ + struct ufs2_csum_total fs_cstotal;/*cylinder summary information*/ + struct ufs_timeval fs_time; /* last time written */ + __fs64 fs_size; /* number of blocks in fs */ + __fs64 fs_dsize; /* number of data blocks in fs */ + __fs64 fs_csaddr; /* blk addr of cyl grp summary area */ + __fs64 fs_pendingblocks;/* blocks in process of being freed */ + __fs32 fs_pendinginodes;/*inodes in process of being freed */ + } fs_u2; + } fs_u11; + union { + struct { + __fs32 fs_sparecon[53];/* reserved for future constants */ + __fs32 fs_reclaim; + __fs32 fs_sparecon2[1]; + __fs32 fs_state; /* file system state time stamp */ + __fs32 fs_qbmask[2]; /* ~usb_bmask */ + __fs32 fs_qfmask[2]; /* ~usb_fmask */ + } fs_sun; + struct { + __fs32 fs_sparecon[53];/* reserved for future constants */ + __fs32 fs_reclaim; + __fs32 fs_sparecon2[1]; + __fs32 fs_npsect; /* # sectors/track including spares */ + __fs32 fs_qbmask[2]; /* ~usb_bmask */ + __fs32 fs_qfmask[2]; /* ~usb_fmask */ + } fs_sunx86; + struct { + __fs32 fs_sparecon[50];/* reserved for future constants */ + __fs32 fs_contigsumsize;/* size of cluster summary array */ + __fs32 fs_maxsymlinklen;/* max length of an internal symlink */ + __fs32 fs_inodefmt; /* format of on-disk inodes */ + __fs32 fs_maxfilesize[2]; /* max representable file size */ + __fs32 fs_qbmask[2]; /* ~usb_bmask */ + __fs32 fs_qfmask[2]; /* ~usb_fmask */ + __fs32 fs_state; /* file system state time stamp */ + } fs_44; + } fs_u2; + __fs32 fs_postblformat; /* format of positional layout tables */ + __fs32 fs_nrpos; /* number of rotational positions */ + __fs32 fs_postbloff; /* (__s16) rotation block list head */ + __fs32 fs_rotbloff; /* (__u8) blocks for each rotation */ + __fs32 fs_magic; /* magic number */ + __u8 fs_space[1]; /* list of blocks for each rotation */ +}; +#endif/*struct ufs_super_block*/ + +/* + * Preference for optimization. + */ +#define UFS_OPTTIME 0 /* minimize allocation time */ +#define UFS_OPTSPACE 1 /* minimize disk fragmentation */ + +/* + * Rotational layout table format types + */ +#define UFS_42POSTBLFMT -1 /* 4.2BSD rotational table format */ +#define UFS_DYNAMICPOSTBLFMT 1 /* dynamic rotational table format */ + +/* + * Convert cylinder group to base address of its global summary info. + */ +#define fs_cs(indx) s_csp[(indx)] + +/* + * Cylinder group block for a file system. + * + * Writable fields in the cylinder group are protected by the associated + * super block lock fs->fs_lock. + */ +#define CG_MAGIC 0x090255 +#define ufs_cg_chkmagic(sb, ucg) \ + (fs32_to_cpu((sb), (ucg)->cg_magic) == CG_MAGIC) +/* + * Macros for access to old cylinder group array structures + */ +#define ufs_ocg_blktot(sb, ucg) fs32_to_cpu((sb), ((struct ufs_old_cylinder_group *)(ucg))->cg_btot) +#define ufs_ocg_blks(sb, ucg, cylno) fs32_to_cpu((sb), ((struct ufs_old_cylinder_group *)(ucg))->cg_b[cylno]) +#define ufs_ocg_inosused(sb, ucg) fs32_to_cpu((sb), ((struct ufs_old_cylinder_group *)(ucg))->cg_iused) +#define ufs_ocg_blksfree(sb, ucg) fs32_to_cpu((sb), ((struct ufs_old_cylinder_group *)(ucg))->cg_free) +#define ufs_ocg_chkmagic(sb, ucg) \ + (fs32_to_cpu((sb), ((struct ufs_old_cylinder_group *)(ucg))->cg_magic) == CG_MAGIC) + +/* + * size of this structure is 172 B + */ +struct ufs_cylinder_group { + __fs32 cg_link; /* linked list of cyl groups */ + __fs32 cg_magic; /* magic number */ + __fs32 cg_time; /* time last written */ + __fs32 cg_cgx; /* we are the cgx'th cylinder group */ + __fs16 cg_ncyl; /* number of cyl's this cg */ + __fs16 cg_niblk; /* number of inode blocks this cg */ + __fs32 cg_ndblk; /* number of data blocks this cg */ + struct ufs_csum cg_cs; /* cylinder summary information */ + __fs32 cg_rotor; /* position of last used block */ + __fs32 cg_frotor; /* position of last used frag */ + __fs32 cg_irotor; /* position of last used inode */ + __fs32 cg_frsum[UFS_MAXFRAG]; /* counts of available frags */ + __fs32 cg_btotoff; /* (__u32) block totals per cylinder */ + __fs32 cg_boff; /* (short) free block positions */ + __fs32 cg_iusedoff; /* (char) used inode map */ + __fs32 cg_freeoff; /* (u_char) free block map */ + __fs32 cg_nextfreeoff; /* (u_char) next available space */ + union { + struct { + __fs32 cg_clustersumoff; /* (u_int32) counts of avail clusters */ + __fs32 cg_clusteroff; /* (u_int8) free cluster map */ + __fs32 cg_nclusterblks; /* number of clusters this cg */ + __fs32 cg_sparecon[13]; /* reserved for future use */ + } cg_44; + struct { + __fs32 cg_clustersumoff;/* (u_int32) counts of avail clusters */ + __fs32 cg_clusteroff; /* (u_int8) free cluster map */ + __fs32 cg_nclusterblks;/* number of clusters this cg */ + __fs32 cg_niblk; /* number of inode blocks this cg */ + __fs32 cg_initediblk; /* last initialized inode */ + __fs32 cg_sparecon32[3];/* reserved for future use */ + __fs64 cg_time; /* time last written */ + __fs64 cg_sparecon[3]; /* reserved for future use */ + } cg_u2; + __fs32 cg_sparecon[16]; /* reserved for future use */ + } cg_u; + __u8 cg_space[1]; /* space for cylinder group maps */ +/* actually longer */ +}; + +/* Historic Cylinder group info */ +struct ufs_old_cylinder_group { + __fs32 cg_link; /* linked list of cyl groups */ + __fs32 cg_rlink; /* for incore cyl groups */ + __fs32 cg_time; /* time last written */ + __fs32 cg_cgx; /* we are the cgx'th cylinder group */ + __fs16 cg_ncyl; /* number of cyl's this cg */ + __fs16 cg_niblk; /* number of inode blocks this cg */ + __fs32 cg_ndblk; /* number of data blocks this cg */ + struct ufs_csum cg_cs; /* cylinder summary information */ + __fs32 cg_rotor; /* position of last used block */ + __fs32 cg_frotor; /* position of last used frag */ + __fs32 cg_irotor; /* position of last used inode */ + __fs32 cg_frsum[8]; /* counts of available frags */ + __fs32 cg_btot[32]; /* block totals per cylinder */ + __fs16 cg_b[32][8]; /* positions of free blocks */ + __u8 cg_iused[256]; /* used inode map */ + __fs32 cg_magic; /* magic number */ + __u8 cg_free[1]; /* free block map */ +/* actually longer */ +}; + +/* + * structure of an on-disk inode + */ +struct ufs_inode { + __fs16 ui_mode; /* 0x0 */ + __fs16 ui_nlink; /* 0x2 */ + union { + struct { + __fs16 ui_suid; /* 0x4 */ + __fs16 ui_sgid; /* 0x6 */ + } oldids; + __fs32 ui_inumber; /* 0x4 lsf: inode number */ + __fs32 ui_author; /* 0x4 GNU HURD: author */ + } ui_u1; + __fs64 ui_size; /* 0x8 */ + struct ufs_timeval ui_atime; /* 0x10 access */ + struct ufs_timeval ui_mtime; /* 0x18 modification */ + struct ufs_timeval ui_ctime; /* 0x20 creation */ + union { + struct { + __fs32 ui_db[UFS_NDADDR];/* 0x28 data blocks */ + __fs32 ui_ib[UFS_NINDIR];/* 0x58 indirect blocks */ + } ui_addr; + __u8 ui_symlink[4*(UFS_NDADDR+UFS_NINDIR)];/* 0x28 fast symlink */ + } ui_u2; + __fs32 ui_flags; /* 0x64 immutable, append-only... */ + __fs32 ui_blocks; /* 0x68 blocks in use */ + __fs32 ui_gen; /* 0x6c like ext2 i_version, for NFS support */ + union { + struct { + __fs32 ui_shadow; /* 0x70 shadow inode with security data */ + __fs32 ui_uid; /* 0x74 long EFT version of uid */ + __fs32 ui_gid; /* 0x78 long EFT version of gid */ + __fs32 ui_oeftflag; /* 0x7c reserved */ + } ui_sun; + struct { + __fs32 ui_uid; /* 0x70 File owner */ + __fs32 ui_gid; /* 0x74 File group */ + __fs32 ui_spare[2]; /* 0x78 reserved */ + } ui_44; + struct { + __fs32 ui_uid; /* 0x70 */ + __fs32 ui_gid; /* 0x74 */ + __fs16 ui_modeh; /* 0x78 mode high bits */ + __fs16 ui_spare; /* 0x7A unused */ + __fs32 ui_trans; /* 0x7c filesystem translator */ + } ui_hurd; + } ui_u3; +}; + +#define UFS_NXADDR 2 /* External addresses in inode. */ +struct ufs2_inode { + __fs16 ui_mode; /* 0: IFMT, permissions; see below. */ + __fs16 ui_nlink; /* 2: File link count. */ + __fs32 ui_uid; /* 4: File owner. */ + __fs32 ui_gid; /* 8: File group. */ + __fs32 ui_blksize; /* 12: Inode blocksize. */ + __fs64 ui_size; /* 16: File byte count. */ + __fs64 ui_blocks; /* 24: Bytes actually held. */ + __fs64 ui_atime; /* 32: Last access time. */ + __fs64 ui_mtime; /* 40: Last modified time. */ + __fs64 ui_ctime; /* 48: Last inode change time. */ + __fs64 ui_birthtime; /* 56: Inode creation time. */ + __fs32 ui_mtimensec; /* 64: Last modified time. */ + __fs32 ui_atimensec; /* 68: Last access time. */ + __fs32 ui_ctimensec; /* 72: Last inode change time. */ + __fs32 ui_birthnsec; /* 76: Inode creation time. */ + __fs32 ui_gen; /* 80: Generation number. */ + __fs32 ui_kernflags; /* 84: Kernel flags. */ + __fs32 ui_flags; /* 88: Status flags (chflags). */ + __fs32 ui_extsize; /* 92: External attributes block. */ + __fs64 ui_extb[UFS_NXADDR];/* 96: External attributes block. */ + union { + struct { + __fs64 ui_db[UFS_NDADDR]; /* 112: Direct disk blocks. */ + __fs64 ui_ib[UFS_NINDIR];/* 208: Indirect disk blocks.*/ + } ui_addr; + __u8 ui_symlink[2*4*(UFS_NDADDR+UFS_NINDIR)];/* 0x28 fast symlink */ + } ui_u2; + __fs64 ui_spare[3]; /* 232: Reserved; currently unused */ +}; + + +/* FreeBSD has these in sys/stat.h */ +/* ui_flags that can be set by a file owner */ +#define UFS_UF_SETTABLE 0x0000ffff +#define UFS_UF_NODUMP 0x00000001 /* do not dump */ +#define UFS_UF_IMMUTABLE 0x00000002 /* immutable (can't "change") */ +#define UFS_UF_APPEND 0x00000004 /* append-only */ +#define UFS_UF_OPAQUE 0x00000008 /* directory is opaque (unionfs) */ +#define UFS_UF_NOUNLINK 0x00000010 /* can't be removed or renamed */ +/* ui_flags that only root can set */ +#define UFS_SF_SETTABLE 0xffff0000 +#define UFS_SF_ARCHIVED 0x00010000 /* archived */ +#define UFS_SF_IMMUTABLE 0x00020000 /* immutable (can't "change") */ +#define UFS_SF_APPEND 0x00040000 /* append-only */ +#define UFS_SF_NOUNLINK 0x00100000 /* can't be removed or renamed */ + +/* + * This structure is used for reading disk structures larger + * than the size of fragment. + */ +struct ufs_buffer_head { + __u64 fragment; /* first fragment */ + __u64 count; /* number of fragments */ + struct buffer_head * bh[UFS_MAXFRAG]; /* buffers */ +}; + +struct ufs_cg_private_info { + struct ufs_buffer_head c_ubh; + __u32 c_cgx; /* number of cylidner group */ + __u16 c_ncyl; /* number of cyl's this cg */ + __u16 c_niblk; /* number of inode blocks this cg */ + __u32 c_ndblk; /* number of data blocks this cg */ + __u32 c_rotor; /* position of last used block */ + __u32 c_frotor; /* position of last used frag */ + __u32 c_irotor; /* position of last used inode */ + __u32 c_btotoff; /* (__u32) block totals per cylinder */ + __u32 c_boff; /* (short) free block positions */ + __u32 c_iusedoff; /* (char) used inode map */ + __u32 c_freeoff; /* (u_char) free block map */ + __u32 c_nextfreeoff; /* (u_char) next available space */ + __u32 c_clustersumoff;/* (u_int32) counts of avail clusters */ + __u32 c_clusteroff; /* (u_int8) free cluster map */ + __u32 c_nclusterblks; /* number of clusters this cg */ +}; + + +struct ufs_sb_private_info { + struct ufs_buffer_head s_ubh; /* buffer containing super block */ + struct ufs_csum_core cs_total; + __u32 s_sblkno; /* offset of super-blocks in filesys */ + __u32 s_cblkno; /* offset of cg-block in filesys */ + __u32 s_iblkno; /* offset of inode-blocks in filesys */ + __u32 s_dblkno; /* offset of first data after cg */ + __u32 s_cgoffset; /* cylinder group offset in cylinder */ + __u32 s_cgmask; /* used to calc mod fs_ntrak */ + __u32 s_size; /* number of blocks (fragments) in fs */ + __u32 s_dsize; /* number of data blocks in fs */ + __u64 s_u2_size; /* ufs2: number of blocks (fragments) in fs */ + __u64 s_u2_dsize; /*ufs2: number of data blocks in fs */ + __u32 s_ncg; /* number of cylinder groups */ + __u32 s_bsize; /* size of basic blocks */ + __u32 s_fsize; /* size of fragments */ + __u32 s_fpb; /* fragments per block */ + __u32 s_minfree; /* minimum percentage of free blocks */ + __u32 s_bmask; /* `blkoff'' calc of blk offsets */ + __u32 s_fmask; /* s_fsize mask */ + __u32 s_bshift; /* `lblkno'' calc of logical blkno */ + __u32 s_fshift; /* s_fsize shift */ + __u32 s_fpbshift; /* fragments per block shift */ + __u32 s_fsbtodb; /* fsbtodb and dbtofsb shift constant */ + __u32 s_sbsize; /* actual size of super block */ + __u32 s_csmask; /* csum block offset */ + __u32 s_csshift; /* csum block number */ + __u32 s_nindir; /* value of NINDIR */ + __u32 s_inopb; /* value of INOPB */ + __u32 s_nspf; /* value of NSPF */ + __u32 s_npsect; /* # sectors/track including spares */ + __u32 s_interleave; /* hardware sector interleave */ + __u32 s_trackskew; /* sector 0 skew, per track */ + __u64 s_csaddr; /* blk addr of cyl grp summary area */ + __u32 s_cssize; /* size of cyl grp summary area */ + __u32 s_cgsize; /* cylinder group size */ + __u32 s_ntrak; /* tracks per cylinder */ + __u32 s_nsect; /* sectors per track */ + __u32 s_spc; /* sectors per cylinder */ + __u32 s_ipg; /* inodes per cylinder group */ + __u32 s_fpg; /* fragments per group */ + __u32 s_cpc; /* cyl per cycle in postbl */ + __s32 s_contigsumsize;/* size of cluster summary array, 44bsd */ + __s64 s_qbmask; /* ~usb_bmask */ + __s64 s_qfmask; /* ~usb_fmask */ + __s32 s_postblformat; /* format of positional layout tables */ + __s32 s_nrpos; /* number of rotational positions */ + __s32 s_postbloff; /* (__s16) rotation block list head */ + __s32 s_rotbloff; /* (__u8) blocks for each rotation */ + + __u32 s_fpbmask; /* fragments per block mask */ + __u32 s_apb; /* address per block */ + __u32 s_2apb; /* address per block^2 */ + __u32 s_3apb; /* address per block^3 */ + __u32 s_apbmask; /* address per block mask */ + __u32 s_apbshift; /* address per block shift */ + __u32 s_2apbshift; /* address per block shift * 2 */ + __u32 s_3apbshift; /* address per block shift * 3 */ + __u32 s_nspfshift; /* number of sector per fragment shift */ + __u32 s_nspb; /* number of sector per block */ + __u32 s_inopf; /* inodes per fragment */ + __u32 s_sbbase; /* offset of NeXTstep superblock */ + __u32 s_bpf; /* bits per fragment */ + __u32 s_bpfshift; /* bits per fragment shift*/ + __u32 s_bpfmask; /* bits per fragment mask */ + + __u32 s_maxsymlinklen;/* upper limit on fast symlinks' size */ + __s32 fs_magic; /* filesystem magic */ + unsigned int s_dirblksize; +}; + +/* + * Sizes of this structures are: + * ufs_super_block_first 512 + * ufs_super_block_second 512 + * ufs_super_block_third 356 + */ +struct ufs_super_block_first { + union { + struct { + __fs32 fs_link; /* UNUSED */ + } fs_42; + struct { + __fs32 fs_state; /* file system state flag */ + } fs_sun; + } fs_u0; + __fs32 fs_rlink; + __fs32 fs_sblkno; + __fs32 fs_cblkno; + __fs32 fs_iblkno; + __fs32 fs_dblkno; + __fs32 fs_cgoffset; + __fs32 fs_cgmask; + __fs32 fs_time; + __fs32 fs_size; + __fs32 fs_dsize; + __fs32 fs_ncg; + __fs32 fs_bsize; + __fs32 fs_fsize; + __fs32 fs_frag; + __fs32 fs_minfree; + __fs32 fs_rotdelay; + __fs32 fs_rps; + __fs32 fs_bmask; + __fs32 fs_fmask; + __fs32 fs_bshift; + __fs32 fs_fshift; + __fs32 fs_maxcontig; + __fs32 fs_maxbpg; + __fs32 fs_fragshift; + __fs32 fs_fsbtodb; + __fs32 fs_sbsize; + __fs32 fs_csmask; + __fs32 fs_csshift; + __fs32 fs_nindir; + __fs32 fs_inopb; + __fs32 fs_nspf; + __fs32 fs_optim; + union { + struct { + __fs32 fs_npsect; + } fs_sun; + struct { + __fs32 fs_state; + } fs_sunx86; + } fs_u1; + __fs32 fs_interleave; + __fs32 fs_trackskew; + __fs32 fs_id[2]; + __fs32 fs_csaddr; + __fs32 fs_cssize; + __fs32 fs_cgsize; + __fs32 fs_ntrak; + __fs32 fs_nsect; + __fs32 fs_spc; + __fs32 fs_ncyl; + __fs32 fs_cpg; + __fs32 fs_ipg; + __fs32 fs_fpg; + struct ufs_csum fs_cstotal; + __s8 fs_fmod; + __s8 fs_clean; + __s8 fs_ronly; + __s8 fs_flags; + __s8 fs_fsmnt[UFS_MAXMNTLEN - 212]; + +}; + +struct ufs_super_block_second { + union { + struct { + __s8 fs_fsmnt[212]; + __fs32 fs_cgrotor; + __fs32 fs_csp[UFS_MAXCSBUFS]; + __fs32 fs_maxcluster; + __fs32 fs_cpc; + __fs16 fs_opostbl[82]; + } fs_u1; + struct { + __s8 fs_fsmnt[UFS2_MAXMNTLEN - UFS_MAXMNTLEN + 212]; + __u8 fs_volname[UFS2_MAXVOLLEN]; + __fs64 fs_swuid; + __fs32 fs_pad; + __fs32 fs_cgrotor; + __fs32 fs_ocsp[UFS2_NOCSPTRS]; + __fs32 fs_contigdirs; + __fs32 fs_csp; + __fs32 fs_maxcluster; + __fs32 fs_active; + __fs32 fs_old_cpc; + __fs32 fs_maxbsize; + __fs64 fs_sparecon64[17]; + __fs64 fs_sblockloc; + __fs64 cs_ndir; + __fs64 cs_nbfree; + } fs_u2; + } fs_un; +}; + +struct ufs_super_block_third { + union { + struct { + __fs16 fs_opostbl[46]; + } fs_u1; + struct { + __fs64 cs_nifree; /* number of free inodes */ + __fs64 cs_nffree; /* number of free frags */ + __fs64 cs_numclusters; /* number of free clusters */ + __fs64 cs_spare[3]; /* future expansion */ + struct ufs_timeval fs_time; /* last time written */ + __fs64 fs_size; /* number of blocks in fs */ + __fs64 fs_dsize; /* number of data blocks in fs */ + __fs64 fs_csaddr; /* blk addr of cyl grp summary area */ + __fs64 fs_pendingblocks;/* blocks in process of being freed */ + __fs32 fs_pendinginodes;/*inodes in process of being freed */ + } __attribute__ ((packed)) fs_u2; + } fs_un1; + union { + struct { + __fs32 fs_sparecon[53];/* reserved for future constants */ + __fs32 fs_reclaim; + __fs32 fs_sparecon2[1]; + __fs32 fs_state; /* file system state time stamp */ + __fs32 fs_qbmask[2]; /* ~usb_bmask */ + __fs32 fs_qfmask[2]; /* ~usb_fmask */ + } fs_sun; + struct { + __fs32 fs_sparecon[53];/* reserved for future constants */ + __fs32 fs_reclaim; + __fs32 fs_sparecon2[1]; + __fs32 fs_npsect; /* # sectors/track including spares */ + __fs32 fs_qbmask[2]; /* ~usb_bmask */ + __fs32 fs_qfmask[2]; /* ~usb_fmask */ + } fs_sunx86; + struct { + __fs32 fs_sparecon[50];/* reserved for future constants */ + __fs32 fs_contigsumsize;/* size of cluster summary array */ + __fs32 fs_maxsymlinklen;/* max length of an internal symlink */ + __fs32 fs_inodefmt; /* format of on-disk inodes */ + __fs32 fs_maxfilesize[2]; /* max representable file size */ + __fs32 fs_qbmask[2]; /* ~usb_bmask */ + __fs32 fs_qfmask[2]; /* ~usb_fmask */ + __fs32 fs_state; /* file system state time stamp */ + } fs_44; + } fs_un2; + __fs32 fs_postblformat; + __fs32 fs_nrpos; + __fs32 fs_postbloff; + __fs32 fs_rotbloff; + __fs32 fs_magic; + __u8 fs_space[1]; +}; + +#endif /* __LINUX_UFS_FS_H */ diff --git a/fs/ufs/util.c b/fs/ufs/util.c index 410084dae38..85a7fc9e4a4 100644 --- a/fs/ufs/util.c +++ b/fs/ufs/util.c @@ -8,9 +8,9 @@ #include <linux/string.h> #include <linux/slab.h> -#include <linux/ufs_fs.h> #include <linux/buffer_head.h> +#include "ufs_fs.h" #include "ufs.h" #include "swab.h" #include "util.h" diff --git a/fs/ufs/util.h b/fs/ufs/util.h index b26fc4dec1e..23ceed8c8fb 100644 --- a/fs/ufs/util.h +++ b/fs/ufs/util.h @@ -58,7 +58,7 @@ ufs_set_fs_state(struct super_block *sb, struct ufs_super_block_first *usb1, { switch (UFS_SB(sb)->s_flags & UFS_ST_MASK) { case UFS_ST_SUNOS: - if (fs32_to_cpu(sb, usb3->fs_postblformat == UFS_42POSTBLFMT)) { + if (fs32_to_cpu(sb, usb3->fs_postblformat) == UFS_42POSTBLFMT) { usb1->fs_u0.fs_sun.fs_state = cpu_to_fs32(sb, value); break; } diff --git a/fs/utimes.c b/fs/utimes.c index b9912ecbee2..b18da9c0b97 100644 --- a/fs/utimes.c +++ b/fs/utimes.c @@ -6,6 +6,7 @@ #include <linux/sched.h> #include <linux/stat.h> #include <linux/utime.h> +#include <linux/syscalls.h> #include <asm/uaccess.h> #include <asm/unistd.h> @@ -83,7 +84,7 @@ long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags if (error) goto out; - dentry = nd.dentry; + dentry = nd.path.dentry; } inode = dentry->d_inode; @@ -137,7 +138,7 @@ dput_and_out: if (f) fput(f); else - path_release(&nd); + path_put(&nd.path); out: return error; } diff --git a/fs/vfat/namei.c b/fs/vfat/namei.c index c28add2fbe9..cd450bea9f1 100644 --- a/fs/vfat/namei.c +++ b/fs/vfat/namei.c @@ -705,7 +705,7 @@ static struct dentry *vfat_lookup(struct inode *dir, struct dentry *dentry, brelse(sinfo.bh); if (IS_ERR(inode)) { unlock_kernel(); - return ERR_PTR(PTR_ERR(inode)); + return ERR_CAST(inode); } alias = d_find_alias(inode); if (alias) { diff --git a/fs/xattr.c b/fs/xattr.c index 6645b7313b3..3acab161546 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -105,6 +105,33 @@ out: EXPORT_SYMBOL_GPL(vfs_setxattr); ssize_t +xattr_getsecurity(struct inode *inode, const char *name, void *value, + size_t size) +{ + void *buffer = NULL; + ssize_t len; + + if (!value || !size) { + len = security_inode_getsecurity(inode, name, &buffer, false); + goto out_noalloc; + } + + len = security_inode_getsecurity(inode, name, &buffer, true); + if (len < 0) + return len; + if (size < len) { + len = -ERANGE; + goto out; + } + memcpy(value, buffer, len); +out: + security_release_secctx(buffer, len); +out_noalloc: + return len; +} +EXPORT_SYMBOL_GPL(xattr_getsecurity); + +ssize_t vfs_getxattr(struct dentry *dentry, char *name, void *value, size_t size) { struct inode *inode = dentry->d_inode; @@ -118,23 +145,23 @@ vfs_getxattr(struct dentry *dentry, char *name, void *value, size_t size) if (error) return error; - if (inode->i_op->getxattr) - error = inode->i_op->getxattr(dentry, name, value, size); - else - error = -EOPNOTSUPP; - if (!strncmp(name, XATTR_SECURITY_PREFIX, XATTR_SECURITY_PREFIX_LEN)) { const char *suffix = name + XATTR_SECURITY_PREFIX_LEN; - int ret = security_inode_getsecurity(inode, suffix, value, - size, error); + int ret = xattr_getsecurity(inode, suffix, value, size); /* * Only overwrite the return value if a security module * is actually active. */ - if (ret != -EOPNOTSUPP) - error = ret; + if (ret == -EOPNOTSUPP) + goto nolsm; + return ret; } +nolsm: + if (inode->i_op->getxattr) + error = inode->i_op->getxattr(dentry, name, value, size); + else + error = -EOPNOTSUPP; return error; } @@ -235,8 +262,8 @@ sys_setxattr(char __user *path, char __user *name, void __user *value, error = user_path_walk(path, &nd); if (error) return error; - error = setxattr(nd.dentry, name, value, size, flags); - path_release(&nd); + error = setxattr(nd.path.dentry, name, value, size, flags); + path_put(&nd.path); return error; } @@ -250,8 +277,8 @@ sys_lsetxattr(char __user *path, char __user *name, void __user *value, error = user_path_walk_link(path, &nd); if (error) return error; - error = setxattr(nd.dentry, name, value, size, flags); - path_release(&nd); + error = setxattr(nd.path.dentry, name, value, size, flags); + path_put(&nd.path); return error; } @@ -320,8 +347,8 @@ sys_getxattr(char __user *path, char __user *name, void __user *value, error = user_path_walk(path, &nd); if (error) return error; - error = getxattr(nd.dentry, name, value, size); - path_release(&nd); + error = getxattr(nd.path.dentry, name, value, size); + path_put(&nd.path); return error; } @@ -335,8 +362,8 @@ sys_lgetxattr(char __user *path, char __user *name, void __user *value, error = user_path_walk_link(path, &nd); if (error) return error; - error = getxattr(nd.dentry, name, value, size); - path_release(&nd); + error = getxattr(nd.path.dentry, name, value, size); + path_put(&nd.path); return error; } @@ -394,8 +421,8 @@ sys_listxattr(char __user *path, char __user *list, size_t size) error = user_path_walk(path, &nd); if (error) return error; - error = listxattr(nd.dentry, list, size); - path_release(&nd); + error = listxattr(nd.path.dentry, list, size); + path_put(&nd.path); return error; } @@ -408,8 +435,8 @@ sys_llistxattr(char __user *path, char __user *list, size_t size) error = user_path_walk_link(path, &nd); if (error) return error; - error = listxattr(nd.dentry, list, size); - path_release(&nd); + error = listxattr(nd.path.dentry, list, size); + path_put(&nd.path); return error; } @@ -455,8 +482,8 @@ sys_removexattr(char __user *path, char __user *name) error = user_path_walk(path, &nd); if (error) return error; - error = removexattr(nd.dentry, name); - path_release(&nd); + error = removexattr(nd.path.dentry, name); + path_put(&nd.path); return error; } @@ -469,8 +496,8 @@ sys_lremovexattr(char __user *path, char __user *name) error = user_path_walk_link(path, &nd); if (error) return error; - error = removexattr(nd.dentry, name); - path_release(&nd); + error = removexattr(nd.path.dentry, name); + path_put(&nd.path); return error; } diff --git a/fs/xfs/Kbuild b/fs/xfs/Kbuild deleted file mode 100644 index 2566e96706f..00000000000 --- a/fs/xfs/Kbuild +++ /dev/null @@ -1,6 +0,0 @@ -# -# The xfs people like to share Makefile with 2.6 and 2.4. -# Utilise file named Kbuild file which has precedence over Makefile. -# - -include $(srctree)/$(obj)/Makefile-linux-2.6 diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 49e3e7e5e3d..36ec614e699 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -1 +1,117 @@ -include $(TOPDIR)/fs/xfs/Makefile-linux-$(VERSION).$(PATCHLEVEL) +# +# Copyright (c) 2000-2005 Silicon Graphics, Inc. +# All Rights Reserved. +# +# This program is free software; you can redistribute it and/or +# modify it under the terms of the GNU General Public License as +# published by the Free Software Foundation. +# +# This program is distributed in the hope that it would be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program; if not, write the Free Software Foundation, +# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA +# + +EXTRA_CFLAGS += -I$(src) -I$(src)/linux-2.6 -funsigned-char + +XFS_LINUX := linux-2.6 + +ifeq ($(CONFIG_XFS_DEBUG),y) + EXTRA_CFLAGS += -g +endif + +obj-$(CONFIG_XFS_FS) += xfs.o + +xfs-$(CONFIG_XFS_QUOTA) += $(addprefix quota/, \ + xfs_dquot.o \ + xfs_dquot_item.o \ + xfs_trans_dquot.o \ + xfs_qm_syscalls.o \ + xfs_qm_bhv.o \ + xfs_qm.o) + +ifeq ($(CONFIG_XFS_QUOTA),y) +xfs-$(CONFIG_PROC_FS) += quota/xfs_qm_stats.o +endif + +xfs-$(CONFIG_XFS_RT) += xfs_rtalloc.o +xfs-$(CONFIG_XFS_POSIX_ACL) += xfs_acl.o +xfs-$(CONFIG_PROC_FS) += $(XFS_LINUX)/xfs_stats.o +xfs-$(CONFIG_SYSCTL) += $(XFS_LINUX)/xfs_sysctl.o +xfs-$(CONFIG_COMPAT) += $(XFS_LINUX)/xfs_ioctl32.o + + +xfs-y += xfs_alloc.o \ + xfs_alloc_btree.o \ + xfs_attr.o \ + xfs_attr_leaf.o \ + xfs_bit.o \ + xfs_bmap.o \ + xfs_bmap_btree.o \ + xfs_btree.o \ + xfs_buf_item.o \ + xfs_da_btree.o \ + xfs_dir2.o \ + xfs_dir2_block.o \ + xfs_dir2_data.o \ + xfs_dir2_leaf.o \ + xfs_dir2_node.o \ + xfs_dir2_sf.o \ + xfs_error.o \ + xfs_extfree_item.o \ + xfs_filestream.o \ + xfs_fsops.o \ + xfs_ialloc.o \ + xfs_ialloc_btree.o \ + xfs_iget.o \ + xfs_inode.o \ + xfs_inode_item.o \ + xfs_iomap.o \ + xfs_itable.o \ + xfs_dfrag.o \ + xfs_log.o \ + xfs_log_recover.o \ + xfs_mount.o \ + xfs_mru_cache.o \ + xfs_rename.o \ + xfs_trans.o \ + xfs_trans_ail.o \ + xfs_trans_buf.o \ + xfs_trans_extfree.o \ + xfs_trans_inode.o \ + xfs_trans_item.o \ + xfs_utils.o \ + xfs_vfsops.o \ + xfs_vnodeops.o \ + xfs_rw.o \ + xfs_dmops.o \ + xfs_qmops.o + +xfs-$(CONFIG_XFS_TRACE) += xfs_dir2_trace.o + +# Objects in linux/ +xfs-y += $(addprefix $(XFS_LINUX)/, \ + kmem.o \ + xfs_aops.o \ + xfs_buf.o \ + xfs_export.o \ + xfs_file.o \ + xfs_fs_subr.o \ + xfs_globals.o \ + xfs_ioctl.o \ + xfs_iops.o \ + xfs_lrw.o \ + xfs_super.o \ + xfs_vnode.o) + +# Objects in support/ +xfs-y += $(addprefix support/, \ + debug.o \ + uuid.o) + +xfs-$(CONFIG_XFS_TRACE) += support/ktrace.o + diff --git a/fs/xfs/Makefile-linux-2.6 b/fs/xfs/Makefile-linux-2.6 deleted file mode 100644 index d1491aa7a0e..00000000000 --- a/fs/xfs/Makefile-linux-2.6 +++ /dev/null @@ -1,118 +0,0 @@ -# -# Copyright (c) 2000-2005 Silicon Graphics, Inc. -# All Rights Reserved. -# -# This program is free software; you can redistribute it and/or -# modify it under the terms of the GNU General Public License as -# published by the Free Software Foundation. -# -# This program is distributed in the hope that it would be useful, -# but WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -# GNU General Public License for more details. -# -# You should have received a copy of the GNU General Public License -# along with this program; if not, write the Free Software Foundation, -# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA -# - -EXTRA_CFLAGS += -Ifs/xfs -Ifs/xfs/linux-2.6 -funsigned-char - -XFS_LINUX := linux-2.6 - -ifeq ($(CONFIG_XFS_DEBUG),y) - EXTRA_CFLAGS += -g -endif - -obj-$(CONFIG_XFS_FS) += xfs.o - -xfs-$(CONFIG_XFS_QUOTA) += $(addprefix quota/, \ - xfs_dquot.o \ - xfs_dquot_item.o \ - xfs_trans_dquot.o \ - xfs_qm_syscalls.o \ - xfs_qm_bhv.o \ - xfs_qm.o) - -ifeq ($(CONFIG_XFS_QUOTA),y) -xfs-$(CONFIG_PROC_FS) += quota/xfs_qm_stats.o -endif - -xfs-$(CONFIG_XFS_RT) += xfs_rtalloc.o -xfs-$(CONFIG_XFS_POSIX_ACL) += xfs_acl.o -xfs-$(CONFIG_PROC_FS) += $(XFS_LINUX)/xfs_stats.o -xfs-$(CONFIG_SYSCTL) += $(XFS_LINUX)/xfs_sysctl.o -xfs-$(CONFIG_COMPAT) += $(XFS_LINUX)/xfs_ioctl32.o - - -xfs-y += xfs_alloc.o \ - xfs_alloc_btree.o \ - xfs_attr.o \ - xfs_attr_leaf.o \ - xfs_bit.o \ - xfs_bmap.o \ - xfs_bmap_btree.o \ - xfs_btree.o \ - xfs_buf_item.o \ - xfs_da_btree.o \ - xfs_dir2.o \ - xfs_dir2_block.o \ - xfs_dir2_data.o \ - xfs_dir2_leaf.o \ - xfs_dir2_node.o \ - xfs_dir2_sf.o \ - xfs_error.o \ - xfs_extfree_item.o \ - xfs_filestream.o \ - xfs_fsops.o \ - xfs_ialloc.o \ - xfs_ialloc_btree.o \ - xfs_iget.o \ - xfs_inode.o \ - xfs_inode_item.o \ - xfs_iocore.o \ - xfs_iomap.o \ - xfs_itable.o \ - xfs_dfrag.o \ - xfs_log.o \ - xfs_log_recover.o \ - xfs_mount.o \ - xfs_mru_cache.o \ - xfs_rename.o \ - xfs_trans.o \ - xfs_trans_ail.o \ - xfs_trans_buf.o \ - xfs_trans_extfree.o \ - xfs_trans_inode.o \ - xfs_trans_item.o \ - xfs_utils.o \ - xfs_vfsops.o \ - xfs_vnodeops.o \ - xfs_rw.o \ - xfs_dmops.o \ - xfs_qmops.o - -xfs-$(CONFIG_XFS_TRACE) += xfs_dir2_trace.o - -# Objects in linux/ -xfs-y += $(addprefix $(XFS_LINUX)/, \ - kmem.o \ - xfs_aops.o \ - xfs_buf.o \ - xfs_export.o \ - xfs_file.o \ - xfs_fs_subr.o \ - xfs_globals.o \ - xfs_ioctl.o \ - xfs_iops.o \ - xfs_lrw.o \ - xfs_super.o \ - xfs_vnode.o) - -# Objects in support/ -xfs-y += $(addprefix support/, \ - debug.o \ - uuid.o) - -xfs-$(CONFIG_XFS_TRACE) += support/ktrace.o - diff --git a/fs/xfs/linux-2.6/kmem.c b/fs/xfs/linux-2.6/kmem.c index ed2b16dff91..e040f1ce1b6 100644 --- a/fs/xfs/linux-2.6/kmem.c +++ b/fs/xfs/linux-2.6/kmem.c @@ -92,8 +92,7 @@ kmem_zalloc_greedy(size_t *size, size_t minsize, size_t maxsize, void kmem_free(void *ptr, size_t size) { - if (((unsigned long)ptr < VMALLOC_START) || - ((unsigned long)ptr >= VMALLOC_END)) { + if (!is_vmalloc_addr(ptr)) { kfree(ptr); } else { vfree(ptr); diff --git a/fs/xfs/linux-2.6/spin.h b/fs/xfs/linux-2.6/spin.h deleted file mode 100644 index 50a6191178f..00000000000 --- a/fs/xfs/linux-2.6/spin.h +++ /dev/null @@ -1,45 +0,0 @@ -/* - * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_SUPPORT_SPIN_H__ -#define __XFS_SUPPORT_SPIN_H__ - -#include <linux/sched.h> /* preempt needs this */ -#include <linux/spinlock.h> - -/* - * Map lock_t from IRIX to Linux spinlocks. - * - * We do not make use of lock_t from interrupt context, so we do not - * have to worry about disabling interrupts at all (unlike IRIX). - */ - -typedef spinlock_t lock_t; - -#define SPLDECL(s) unsigned long s -#ifndef DEFINE_SPINLOCK -#define DEFINE_SPINLOCK(s) spinlock_t s = SPIN_LOCK_UNLOCKED -#endif - -#define spinlock_init(lock, name) spin_lock_init(lock) -#define spinlock_destroy(lock) -#define mutex_spinlock(lock) ({ spin_lock(lock); 0; }) -#define mutex_spinunlock(lock, s) do { spin_unlock(lock); (void)s; } while (0) -#define nested_spinlock(lock) spin_lock(lock) -#define nested_spinunlock(lock) spin_unlock(lock) - -#endif /* __XFS_SUPPORT_SPIN_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index 2e34b104107..e0519529c26 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -107,6 +107,18 @@ xfs_page_trace( #define xfs_page_trace(tag, inode, page, pgoff) #endif +STATIC struct block_device * +xfs_find_bdev_for_inode( + struct xfs_inode *ip) +{ + struct xfs_mount *mp = ip->i_mount; + + if (XFS_IS_REALTIME_INODE(ip)) + return mp->m_rtdev_targp->bt_bdev; + else + return mp->m_ddev_targp->bt_bdev; +} + /* * Schedule IO completion handling on a xfsdatad if this was * the final hold on this ioend. If we are asked to wait, @@ -151,7 +163,7 @@ xfs_destroy_ioend( /* * Update on-disk file size now that data has been written to disk. * The current in-memory file size is i_size. If a write is beyond - * eof io_new_size will be the intended file size until i_size is + * eof i_new_size will be the intended file size until i_size is * updated. If this write does not extend all the way to the valid * file size then restrict this update to the end of the write. */ @@ -173,7 +185,7 @@ xfs_setfilesize( xfs_ilock(ip, XFS_ILOCK_EXCL); - isize = MAX(ip->i_size, ip->i_iocore.io_new_size); + isize = MAX(ip->i_size, ip->i_new_size); isize = MIN(isize, bsize); if (ip->i_d.di_size < isize) { @@ -226,12 +238,13 @@ xfs_end_bio_unwritten( { xfs_ioend_t *ioend = container_of(work, xfs_ioend_t, io_work); + struct xfs_inode *ip = XFS_I(ioend->io_inode); xfs_off_t offset = ioend->io_offset; size_t size = ioend->io_size; if (likely(!ioend->io_error)) { - xfs_bmap(XFS_I(ioend->io_inode), offset, size, - BMAPI_UNWRITTEN, NULL, NULL); + if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) + xfs_iomap_write_unwritten(ip, offset, size); xfs_setfilesize(ioend); } xfs_destroy_ioend(ioend); @@ -304,7 +317,7 @@ xfs_map_blocks( xfs_inode_t *ip = XFS_I(inode); int error, nmaps = 1; - error = xfs_bmap(ip, offset, count, + error = xfs_iomap(ip, offset, count, flags, mapp, &nmaps); if (!error && (flags & (BMAPI_WRITE|BMAPI_ALLOCATE))) xfs_iflags_set(ip, XFS_IMODIFIED); @@ -1323,7 +1336,7 @@ __xfs_get_blocks( offset = (xfs_off_t)iblock << inode->i_blkbits; ASSERT(bh_result->b_size >= (1 << inode->i_blkbits)); size = bh_result->b_size; - error = xfs_bmap(XFS_I(inode), offset, size, + error = xfs_iomap(XFS_I(inode), offset, size, create ? flags : BMAPI_READ, &iomap, &niomap); if (error) return -error; @@ -1471,28 +1484,21 @@ xfs_vm_direct_IO( { struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; - xfs_iomap_t iomap; - int maps = 1; - int error; + struct block_device *bdev; ssize_t ret; - error = xfs_bmap(XFS_I(inode), offset, 0, - BMAPI_DEVICE, &iomap, &maps); - if (error) - return -error; + bdev = xfs_find_bdev_for_inode(XFS_I(inode)); if (rw == WRITE) { iocb->private = xfs_alloc_ioend(inode, IOMAP_UNWRITTEN); ret = blockdev_direct_IO_own_locking(rw, iocb, inode, - iomap.iomap_target->bt_bdev, - iov, offset, nr_segs, + bdev, iov, offset, nr_segs, xfs_get_blocks_direct, xfs_end_io_direct); } else { iocb->private = xfs_alloc_ioend(inode, IOMAP_READ); ret = blockdev_direct_IO_no_locking(rw, iocb, inode, - iomap.iomap_target->bt_bdev, - iov, offset, nr_segs, + bdev, iov, offset, nr_segs, xfs_get_blocks_direct, xfs_end_io_direct); } @@ -1525,8 +1531,7 @@ xfs_vm_bmap( struct inode *inode = (struct inode *)mapping->host; struct xfs_inode *ip = XFS_I(inode); - vn_trace_entry(XFS_I(inode), __FUNCTION__, - (inst_t *)__return_address); + xfs_itrace_entry(XFS_I(inode)); xfs_rwlock(ip, VRWLOCK_READ); xfs_flush_pages(ip, (xfs_off_t)0, -1, 0, FI_REMAPF); xfs_rwunlock(ip, VRWLOCK_READ); diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index a49dd8d4b06..e347bfd47c9 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -387,8 +387,6 @@ _xfs_buf_lookup_pages( if (unlikely(page == NULL)) { if (flags & XBF_READ_AHEAD) { bp->b_page_count = i; - for (i = 0; i < bp->b_page_count; i++) - unlock_page(bp->b_pages[i]); return -ENOMEM; } @@ -418,24 +416,17 @@ _xfs_buf_lookup_pages( ASSERT(!PagePrivate(page)); if (!PageUptodate(page)) { page_count--; - if (blocksize >= PAGE_CACHE_SIZE) { - if (flags & XBF_READ) - bp->b_locked = 1; - } else if (!PagePrivate(page)) { + if (blocksize < PAGE_CACHE_SIZE && !PagePrivate(page)) { if (test_page_region(page, offset, nbytes)) page_count++; } } + unlock_page(page); bp->b_pages[i] = page; offset = 0; } - if (!bp->b_locked) { - for (i = 0; i < bp->b_page_count; i++) - unlock_page(bp->b_pages[i]); - } - if (page_count == bp->b_page_count) bp->b_flags |= XBF_DONE; @@ -709,8 +700,7 @@ static inline struct page * mem_to_page( void *addr) { - if (((unsigned long)addr < VMALLOC_START) || - ((unsigned long)addr >= VMALLOC_END)) { + if ((!is_vmalloc_addr(addr))) { return virt_to_page(addr); } else { return vmalloc_to_page(addr); @@ -752,7 +742,6 @@ xfs_buf_associate_memory( bp->b_pages[i] = mem_to_page((void *)pageaddr); pageaddr += PAGE_CACHE_SIZE; } - bp->b_locked = 0; bp->b_count_desired = len; bp->b_buffer_length = buflen; @@ -1099,25 +1088,13 @@ xfs_buf_iostart( return status; } -STATIC_INLINE int -_xfs_buf_iolocked( - xfs_buf_t *bp) -{ - ASSERT(bp->b_flags & (XBF_READ | XBF_WRITE)); - if (bp->b_flags & XBF_READ) - return bp->b_locked; - return 0; -} - STATIC_INLINE void _xfs_buf_ioend( xfs_buf_t *bp, int schedule) { - if (atomic_dec_and_test(&bp->b_io_remaining) == 1) { - bp->b_locked = 0; + if (atomic_dec_and_test(&bp->b_io_remaining) == 1) xfs_buf_ioend(bp, schedule); - } } STATIC void @@ -1148,10 +1125,6 @@ xfs_buf_bio_end_io( if (--bvec >= bio->bi_io_vec) prefetchw(&bvec->bv_page->flags); - - if (_xfs_buf_iolocked(bp)) { - unlock_page(page); - } } while (bvec >= bio->bi_io_vec); _xfs_buf_ioend(bp, 1); @@ -1162,13 +1135,12 @@ STATIC void _xfs_buf_ioapply( xfs_buf_t *bp) { - int i, rw, map_i, total_nr_pages, nr_pages; + int rw, map_i, total_nr_pages, nr_pages; struct bio *bio; int offset = bp->b_offset; int size = bp->b_count_desired; sector_t sector = bp->b_bn; unsigned int blocksize = bp->b_target->bt_bsize; - int locking = _xfs_buf_iolocked(bp); total_nr_pages = bp->b_page_count; map_i = 0; @@ -1191,7 +1163,7 @@ _xfs_buf_ioapply( * filesystem block size is not smaller than the page size. */ if ((bp->b_buffer_length < PAGE_CACHE_SIZE) && - (bp->b_flags & XBF_READ) && locking && + (bp->b_flags & XBF_READ) && (blocksize >= PAGE_CACHE_SIZE)) { bio = bio_alloc(GFP_NOIO, 1); @@ -1208,24 +1180,6 @@ _xfs_buf_ioapply( goto submit_io; } - /* Lock down the pages which we need to for the request */ - if (locking && (bp->b_flags & XBF_WRITE) && (bp->b_locked == 0)) { - for (i = 0; size; i++) { - int nbytes = PAGE_CACHE_SIZE - offset; - struct page *page = bp->b_pages[i]; - - if (nbytes > size) - nbytes = size; - - lock_page(page); - - size -= nbytes; - offset = 0; - } - offset = bp->b_offset; - size = bp->b_count_desired; - } - next_chunk: atomic_inc(&bp->b_io_remaining); nr_pages = BIO_MAX_SECTORS >> (PAGE_SHIFT - BBSHIFT); @@ -1572,7 +1526,7 @@ xfs_alloc_delwrite_queue( INIT_LIST_HEAD(&btp->bt_list); INIT_LIST_HEAD(&btp->bt_delwrite_queue); - spinlock_init(&btp->bt_delwrite_lock, "delwri_lock"); + spin_lock_init(&btp->bt_delwrite_lock); btp->bt_flags = 0; btp->bt_task = kthread_run(xfsbufd, btp, "xfsbufd"); if (IS_ERR(btp->bt_task)) { diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h index b5908a34b15..a3d207de48b 100644 --- a/fs/xfs/linux-2.6/xfs_buf.h +++ b/fs/xfs/linux-2.6/xfs_buf.h @@ -143,7 +143,6 @@ typedef struct xfs_buf { void *b_fspriv2; void *b_fspriv3; unsigned short b_error; /* error code on I/O */ - unsigned short b_locked; /* page array is locked */ unsigned int b_page_count; /* size of page array */ unsigned int b_offset; /* page offset in first page */ struct page **b_pages; /* array of page pointers */ diff --git a/fs/xfs/linux-2.6/xfs_export.c b/fs/xfs/linux-2.6/xfs_export.c index 15bd4948832..ca4f66c4de1 100644 --- a/fs/xfs/linux-2.6/xfs_export.c +++ b/fs/xfs/linux-2.6/xfs_export.c @@ -118,20 +118,29 @@ xfs_nfs_get_inode( u64 ino, u32 generation) { - xfs_fid_t xfid; - bhv_vnode_t *vp; + xfs_mount_t *mp = XFS_M(sb); + xfs_inode_t *ip; int error; - xfid.fid_len = sizeof(xfs_fid_t) - sizeof(xfid.fid_len); - xfid.fid_pad = 0; - xfid.fid_ino = ino; - xfid.fid_gen = generation; + /* + * NFS can sometimes send requests for ino 0. Fail them gracefully. + */ + if (ino == 0) + return ERR_PTR(-ESTALE); - error = xfs_vget(XFS_M(sb), &vp, &xfid); + error = xfs_iget(mp, NULL, ino, 0, XFS_ILOCK_SHARED, &ip, 0); if (error) return ERR_PTR(-error); + if (!ip) + return ERR_PTR(-EIO); + + if (!ip->i_d.di_mode || ip->i_d.di_gen != generation) { + xfs_iput_new(ip, XFS_ILOCK_SHARED); + return ERR_PTR(-ENOENT); + } - return vp ? vn_to_inode(vp) : NULL; + xfs_iunlock(ip, XFS_ILOCK_SHARED); + return ip->i_vnode; } STATIC struct dentry * diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c index 21a1c2b1c5f..edab1ffbb16 100644 --- a/fs/xfs/linux-2.6/xfs_file.c +++ b/fs/xfs/linux-2.6/xfs_file.c @@ -350,8 +350,8 @@ xfs_file_readdir( size = buf.used; de = (struct hack_dirent *)buf.dirent; - curr_offset = de->offset /* & 0x7fffffff */; while (size > 0) { + curr_offset = de->offset /* & 0x7fffffff */; if (filldir(dirent, de->name, de->namlen, curr_offset & 0x7fffffff, de->ino, de->d_type)) { @@ -362,7 +362,6 @@ xfs_file_readdir( sizeof(u64)); size -= reclen; de = (struct hack_dirent *)((char *)de + reclen); - curr_offset = de->offset /* & 0x7fffffff */; } } diff --git a/fs/xfs/linux-2.6/xfs_globals.c b/fs/xfs/linux-2.6/xfs_globals.c index 9febf9dc999..ef90e64641e 100644 --- a/fs/xfs/linux-2.6/xfs_globals.c +++ b/fs/xfs/linux-2.6/xfs_globals.c @@ -47,5 +47,6 @@ xfs_param_t xfs_params = { /* * Global system credential structure. */ -cred_t sys_cred_val, *sys_cred = &sys_cred_val; +static cred_t sys_cred_val; +cred_t *sys_cred = &sys_cred_val; diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index 98a56568bb2..f34bd010eb5 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -75,7 +75,6 @@ xfs_find_handle( xfs_handle_t handle; xfs_fsop_handlereq_t hreq; struct inode *inode; - bhv_vnode_t *vp; if (copy_from_user(&hreq, arg, sizeof(hreq))) return -XFS_ERROR(EFAULT); @@ -92,10 +91,10 @@ xfs_find_handle( if (error) return error; - ASSERT(nd.dentry); - ASSERT(nd.dentry->d_inode); - inode = igrab(nd.dentry->d_inode); - path_release(&nd); + ASSERT(nd.path.dentry); + ASSERT(nd.path.dentry->d_inode); + inode = igrab(nd.path.dentry->d_inode); + path_put(&nd.path); break; } @@ -134,21 +133,16 @@ xfs_find_handle( return -XFS_ERROR(EBADF); } - /* we need the vnode */ - vp = vn_from_inode(inode); - /* now we can grab the fsid */ memcpy(&handle.ha_fsid, XFS_I(inode)->i_mount->m_fixedfsid, sizeof(xfs_fsid_t)); hsize = sizeof(xfs_fsid_t); if (cmd != XFS_IOC_PATH_TO_FSHANDLE) { - xfs_inode_t *ip; + xfs_inode_t *ip = XFS_I(inode); int lock_mode; /* need to get access to the xfs_inode to read the generation */ - ip = xfs_vtoi(vp); - ASSERT(ip); lock_mode = xfs_ilock_map_shared(ip); /* fill in fid section of handle from inode */ @@ -176,21 +170,19 @@ xfs_find_handle( /* - * Convert userspace handle data into vnode (and inode). - * We [ab]use the fact that all the fsop_handlereq ioctl calls - * have a data structure argument whose first component is always - * a xfs_fsop_handlereq_t, so we can cast to and from this type. - * This allows us to optimise the copy_from_user calls and gives - * a handy, shared routine. + * Convert userspace handle data into inode. + * + * We use the fact that all the fsop_handlereq ioctl calls have a data + * structure argument whose first component is always a xfs_fsop_handlereq_t, + * so we can pass that sub structure into this handy, shared routine. * - * If no error, caller must always VN_RELE the returned vp. + * If no error, caller must always iput the returned inode. */ STATIC int xfs_vget_fsop_handlereq( xfs_mount_t *mp, struct inode *parinode, /* parent inode pointer */ xfs_fsop_handlereq_t *hreq, - bhv_vnode_t **vp, struct inode **inode) { void __user *hanp; @@ -199,8 +191,6 @@ xfs_vget_fsop_handlereq( xfs_handle_t *handlep; xfs_handle_t handle; xfs_inode_t *ip; - struct inode *inodep; - bhv_vnode_t *vpp; xfs_ino_t ino; __u32 igen; int error; @@ -241,7 +231,7 @@ xfs_vget_fsop_handlereq( } /* - * Get the XFS inode, building a vnode to go with it. + * Get the XFS inode, building a Linux inode to go with it. */ error = xfs_iget(mp, NULL, ino, 0, XFS_ILOCK_SHARED, &ip, 0); if (error) @@ -253,12 +243,9 @@ xfs_vget_fsop_handlereq( return XFS_ERROR(ENOENT); } - vpp = XFS_ITOV(ip); - inodep = vn_to_inode(vpp); xfs_iunlock(ip, XFS_ILOCK_SHARED); - *vp = vpp; - *inode = inodep; + *inode = XFS_ITOV(ip); return 0; } @@ -275,7 +262,6 @@ xfs_open_by_handle( struct file *filp; struct inode *inode; struct dentry *dentry; - bhv_vnode_t *vp; xfs_fsop_handlereq_t hreq; if (!capable(CAP_SYS_ADMIN)) @@ -283,7 +269,7 @@ xfs_open_by_handle( if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t))) return -XFS_ERROR(EFAULT); - error = xfs_vget_fsop_handlereq(mp, parinode, &hreq, &vp, &inode); + error = xfs_vget_fsop_handlereq(mp, parinode, &hreq, &inode); if (error) return -error; @@ -385,7 +371,6 @@ xfs_readlink_by_handle( { struct inode *inode; xfs_fsop_handlereq_t hreq; - bhv_vnode_t *vp; __u32 olen; void *link; int error; @@ -395,7 +380,7 @@ xfs_readlink_by_handle( if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t))) return -XFS_ERROR(EFAULT); - error = xfs_vget_fsop_handlereq(mp, parinode, &hreq, &vp, &inode); + error = xfs_vget_fsop_handlereq(mp, parinode, &hreq, &inode); if (error) return -error; @@ -438,34 +423,32 @@ xfs_fssetdm_by_handle( struct fsdmidata fsd; xfs_fsop_setdm_handlereq_t dmhreq; struct inode *inode; - bhv_vnode_t *vp; if (!capable(CAP_MKNOD)) return -XFS_ERROR(EPERM); if (copy_from_user(&dmhreq, arg, sizeof(xfs_fsop_setdm_handlereq_t))) return -XFS_ERROR(EFAULT); - error = xfs_vget_fsop_handlereq(mp, parinode, &dmhreq.hreq, &vp, &inode); + error = xfs_vget_fsop_handlereq(mp, parinode, &dmhreq.hreq, &inode); if (error) return -error; if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) { - VN_RELE(vp); - return -XFS_ERROR(EPERM); + error = -XFS_ERROR(EPERM); + goto out; } if (copy_from_user(&fsd, dmhreq.data, sizeof(fsd))) { - VN_RELE(vp); - return -XFS_ERROR(EFAULT); + error = -XFS_ERROR(EFAULT); + goto out; } - error = xfs_set_dmattrs(xfs_vtoi(vp), - fsd.fsd_dmevmask, fsd.fsd_dmstate); + error = -xfs_set_dmattrs(XFS_I(inode), fsd.fsd_dmevmask, + fsd.fsd_dmstate); - VN_RELE(vp); - if (error) - return -error; - return 0; + out: + iput(inode); + return error; } STATIC int @@ -478,7 +461,6 @@ xfs_attrlist_by_handle( attrlist_cursor_kern_t *cursor; xfs_fsop_attrlist_handlereq_t al_hreq; struct inode *inode; - bhv_vnode_t *vp; char *kbuf; if (!capable(CAP_SYS_ADMIN)) @@ -488,8 +470,7 @@ xfs_attrlist_by_handle( if (al_hreq.buflen > XATTR_LIST_MAX) return -XFS_ERROR(EINVAL); - error = xfs_vget_fsop_handlereq(mp, parinode, &al_hreq.hreq, - &vp, &inode); + error = xfs_vget_fsop_handlereq(mp, parinode, &al_hreq.hreq, &inode); if (error) goto out; @@ -509,7 +490,7 @@ xfs_attrlist_by_handle( out_kfree: kfree(kbuf); out_vn_rele: - VN_RELE(vp); + iput(inode); out: return -error; } @@ -531,7 +512,7 @@ xfs_attrmulti_attr_get( if (!kbuf) return ENOMEM; - error = xfs_attr_get(XFS_I(inode), name, kbuf, len, flags, NULL); + error = xfs_attr_get(XFS_I(inode), name, kbuf, (int *)len, flags, NULL); if (error) goto out_kfree; @@ -598,7 +579,6 @@ xfs_attrmulti_by_handle( xfs_attr_multiop_t *ops; xfs_fsop_attrmulti_handlereq_t am_hreq; struct inode *inode; - bhv_vnode_t *vp; unsigned int i, size; char *attr_name; @@ -607,7 +587,7 @@ xfs_attrmulti_by_handle( if (copy_from_user(&am_hreq, arg, sizeof(xfs_fsop_attrmulti_handlereq_t))) return -XFS_ERROR(EFAULT); - error = xfs_vget_fsop_handlereq(mp, parinode, &am_hreq.hreq, &vp, &inode); + error = xfs_vget_fsop_handlereq(mp, parinode, &am_hreq.hreq, &inode); if (error) goto out; @@ -666,7 +646,7 @@ xfs_attrmulti_by_handle( out_kfree_ops: kfree(ops); out_vn_rele: - VN_RELE(vp); + iput(inode); out: return -error; } @@ -702,7 +682,6 @@ xfs_ioc_fsgeometry( STATIC int xfs_ioc_xattr( - bhv_vnode_t *vp, xfs_inode_t *ip, struct file *filp, unsigned int cmd, @@ -735,12 +714,10 @@ xfs_ioctl( void __user *arg) { struct inode *inode = filp->f_path.dentry->d_inode; - bhv_vnode_t *vp = vn_from_inode(inode); xfs_mount_t *mp = ip->i_mount; int error; - vn_trace_entry(XFS_I(inode), "xfs_ioctl", (inst_t *)__return_address); - + xfs_itrace_entry(XFS_I(inode)); switch (cmd) { case XFS_IOC_ALLOCSP: @@ -755,7 +732,7 @@ xfs_ioctl( * Only allow the sys admin to reserve space unless * unwritten extents are enabled. */ - if (!XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb) && + if (!xfs_sb_version_hasextflgbit(&mp->m_sb) && !capable(CAP_SYS_ADMIN)) return -EPERM; @@ -764,7 +741,7 @@ xfs_ioctl( case XFS_IOC_DIOINFO: { struct dioattr da; xfs_buftarg_t *target = - (ip->i_d.di_flags & XFS_DIFLAG_REALTIME) ? + XFS_IS_REALTIME_INODE(ip) ? mp->m_rtdev_targp : mp->m_ddev_targp; da.d_mem = da.d_miniosz = 1 << target->bt_sshift; @@ -796,7 +773,7 @@ xfs_ioctl( case XFS_IOC_GETXFLAGS: case XFS_IOC_SETXFLAGS: case XFS_IOC_FSSETXATTR: - return xfs_ioc_xattr(vp, ip, filp, cmd, arg); + return xfs_ioc_xattr(ip, filp, cmd, arg); case XFS_IOC_FSSETDM: { struct fsdmidata dmi; @@ -1203,7 +1180,6 @@ xfs_ioc_fsgetxattr( STATIC int xfs_ioc_xattr( - bhv_vnode_t *vp, xfs_inode_t *ip, struct file *filp, unsigned int cmd, @@ -1237,7 +1213,7 @@ xfs_ioc_xattr( error = xfs_setattr(ip, vattr, attr_flags, NULL); if (likely(!error)) - __vn_revalidate(vp, vattr); /* update flags */ + vn_revalidate(XFS_ITOV(ip)); /* update flags */ error = -error; break; } @@ -1272,7 +1248,7 @@ xfs_ioc_xattr( error = xfs_setattr(ip, vattr, attr_flags, NULL); if (likely(!error)) - __vn_revalidate(vp, vattr); /* update flags */ + vn_revalidate(XFS_ITOV(ip)); /* update flags */ error = -error; break; } diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.c b/fs/xfs/linux-2.6/xfs_ioctl32.c index bf2a956b63c..a4b254eb43b 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl32.c +++ b/fs/xfs/linux-2.6/xfs_ioctl32.c @@ -44,6 +44,7 @@ #include "xfs_error.h" #include "xfs_dfrag.h" #include "xfs_vnodeops.h" +#include "xfs_ioctl32.h" #define _NATIVE_IOC(cmd, type) \ _IOC(_IOC_DIR(cmd), _IOC_TYPE(cmd), _IOC_NR(cmd), sizeof(type)) @@ -379,9 +380,6 @@ xfs_compat_ioctl( switch (cmd) { case XFS_IOC_DIOINFO: case XFS_IOC_FSGEOMETRY: - case XFS_IOC_GETVERSION: - case XFS_IOC_GETXFLAGS: - case XFS_IOC_SETXFLAGS: case XFS_IOC_FSGETXATTR: case XFS_IOC_FSSETXATTR: case XFS_IOC_FSGETXATTRA: @@ -407,6 +405,11 @@ xfs_compat_ioctl( case XFS_IOC_ERROR_CLEARALL: break; + case XFS_IOC32_GETXFLAGS: + case XFS_IOC32_SETXFLAGS: + case XFS_IOC32_GETVERSION: + cmd = _NATIVE_IOC(cmd, long); + break; #ifdef BROKEN_X86_ALIGNMENT /* xfs_flock_t has wrong u32 vs u64 alignment */ case XFS_IOC_ALLOCSP_32: diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c index 5e8bb7f71b5..cc4abd3daa4 100644 --- a/fs/xfs/linux-2.6/xfs_iops.c +++ b/fs/xfs/linux-2.6/xfs_iops.c @@ -52,6 +52,7 @@ #include <linux/xattr.h> #include <linux/namei.h> #include <linux/security.h> +#include <linux/falloc.h> /* * Bring the atime in the XFS inode uptodate. @@ -71,6 +72,22 @@ xfs_synchronize_atime( } /* + * If the linux inode exists, mark it dirty. + * Used when commiting a dirty inode into a transaction so that + * the inode will get written back by the linux code + */ +void +xfs_mark_inode_dirty_sync( + xfs_inode_t *ip) +{ + bhv_vnode_t *vp; + + vp = XFS_ITOV_NULL(ip); + if (vp) + mark_inode_dirty_sync(vn_to_inode(vp)); +} + +/* * Change the requested timestamp in the given inode. * We don't lock across timestamp updates, and we don't log them but * we do record the fact that there is dirty information in core. @@ -184,10 +201,6 @@ xfs_validate_fields( struct xfs_inode *ip = XFS_I(inode); loff_t size; - inode->i_nlink = ip->i_d.di_nlink; - inode->i_blocks = - XFS_FSB_TO_BB(ip->i_mount, ip->i_d.di_nblocks + - ip->i_delayed_blks); /* we're under i_sem so i_size can't change under us */ size = XFS_ISIZE(ip); if (i_size_read(inode) != size) @@ -542,12 +555,31 @@ xfs_vn_put_link( #ifdef CONFIG_XFS_POSIX_ACL STATIC int +xfs_check_acl( + struct inode *inode, + int mask) +{ + struct xfs_inode *ip = XFS_I(inode); + int error; + + xfs_itrace_entry(ip); + + if (XFS_IFORK_Q(ip)) { + error = xfs_acl_iaccess(ip, mask, NULL); + if (error != -1) + return -error; + } + + return -EAGAIN; +} + +STATIC int xfs_vn_permission( - struct inode *inode, - int mode, - struct nameidata *nd) + struct inode *inode, + int mask, + struct nameidata *nd) { - return -xfs_access(XFS_I(inode), mode << 6, NULL); + return generic_permission(inode, mask, xfs_check_acl); } #else #define xfs_vn_permission NULL @@ -555,33 +587,61 @@ xfs_vn_permission( STATIC int xfs_vn_getattr( - struct vfsmount *mnt, - struct dentry *dentry, - struct kstat *stat) + struct vfsmount *mnt, + struct dentry *dentry, + struct kstat *stat) { - struct inode *inode = dentry->d_inode; - bhv_vattr_t vattr = { .va_mask = XFS_AT_STAT }; - int error; - - error = xfs_getattr(XFS_I(inode), &vattr, ATTR_LAZY); - if (likely(!error)) { - stat->size = i_size_read(inode); - stat->dev = inode->i_sb->s_dev; - stat->rdev = (vattr.va_rdev == 0) ? 0 : - MKDEV(sysv_major(vattr.va_rdev) & 0x1ff, - sysv_minor(vattr.va_rdev)); - stat->mode = vattr.va_mode; - stat->nlink = vattr.va_nlink; - stat->uid = vattr.va_uid; - stat->gid = vattr.va_gid; - stat->ino = vattr.va_nodeid; - stat->atime = vattr.va_atime; - stat->mtime = vattr.va_mtime; - stat->ctime = vattr.va_ctime; - stat->blocks = vattr.va_nblocks; - stat->blksize = vattr.va_blocksize; + struct inode *inode = dentry->d_inode; + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + + xfs_itrace_entry(ip); + + if (XFS_FORCED_SHUTDOWN(mp)) + return XFS_ERROR(EIO); + + stat->size = XFS_ISIZE(ip); + stat->dev = inode->i_sb->s_dev; + stat->mode = ip->i_d.di_mode; + stat->nlink = ip->i_d.di_nlink; + stat->uid = ip->i_d.di_uid; + stat->gid = ip->i_d.di_gid; + stat->ino = ip->i_ino; +#if XFS_BIG_INUMS + stat->ino += mp->m_inoadd; +#endif + stat->atime = inode->i_atime; + stat->mtime.tv_sec = ip->i_d.di_mtime.t_sec; + stat->mtime.tv_nsec = ip->i_d.di_mtime.t_nsec; + stat->ctime.tv_sec = ip->i_d.di_ctime.t_sec; + stat->ctime.tv_nsec = ip->i_d.di_ctime.t_nsec; + stat->blocks = + XFS_FSB_TO_BB(mp, ip->i_d.di_nblocks + ip->i_delayed_blks); + + + switch (inode->i_mode & S_IFMT) { + case S_IFBLK: + case S_IFCHR: + stat->blksize = BLKDEV_IOSIZE; + stat->rdev = MKDEV(sysv_major(ip->i_df.if_u2.if_rdev) & 0x1ff, + sysv_minor(ip->i_df.if_u2.if_rdev)); + break; + default: + if (XFS_IS_REALTIME_INODE(ip)) { + /* + * If the file blocks are being allocated from a + * realtime volume, then return the inode's realtime + * extent size or the realtime volume's extent size. + */ + stat->blksize = + xfs_get_extsz_hint(ip) << mp->m_sb.sb_blocklog; + } else + stat->blksize = xfs_preferred_iosize(mp); + stat->rdev = 0; + break; } - return -error; + + return 0; } STATIC int @@ -636,7 +696,7 @@ xfs_vn_setattr( error = xfs_setattr(XFS_I(inode), &vattr, flags, NULL); if (likely(!error)) - __vn_revalidate(vn_from_inode(inode), &vattr); + vn_revalidate(vn_from_inode(inode)); return -error; } @@ -750,6 +810,47 @@ xfs_vn_removexattr( return namesp->attr_remove(vp, attr, xflags); } +STATIC long +xfs_vn_fallocate( + struct inode *inode, + int mode, + loff_t offset, + loff_t len) +{ + long error; + loff_t new_size = 0; + xfs_flock64_t bf; + xfs_inode_t *ip = XFS_I(inode); + + /* preallocation on directories not yet supported */ + error = -ENODEV; + if (S_ISDIR(inode->i_mode)) + goto out_error; + + bf.l_whence = 0; + bf.l_start = offset; + bf.l_len = len; + + xfs_ilock(ip, XFS_IOLOCK_EXCL); + error = xfs_change_file_space(ip, XFS_IOC_RESVSP, &bf, + 0, NULL, ATTR_NOLOCK); + if (!error && !(mode & FALLOC_FL_KEEP_SIZE) && + offset + len > i_size_read(inode)) + new_size = offset + len; + + /* Change file size if needed */ + if (new_size) { + bhv_vattr_t va; + + va.va_mask = XFS_AT_SIZE; + va.va_size = new_size; + error = xfs_setattr(ip, &va, ATTR_NOLOCK, NULL); + } + + xfs_iunlock(ip, XFS_IOLOCK_EXCL); +out_error: + return error; +} const struct inode_operations xfs_inode_operations = { .permission = xfs_vn_permission, @@ -760,6 +861,7 @@ const struct inode_operations xfs_inode_operations = { .getxattr = xfs_vn_getxattr, .listxattr = xfs_vn_listxattr, .removexattr = xfs_vn_removexattr, + .fallocate = xfs_vn_fallocate, }; const struct inode_operations xfs_dir_inode_operations = { diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h index dc3752de22d..3ca39c4e5d2 100644 --- a/fs/xfs/linux-2.6/xfs_linux.h +++ b/fs/xfs/linux-2.6/xfs_linux.h @@ -43,7 +43,6 @@ #include <kmem.h> #include <mrlock.h> -#include <spin.h> #include <sv.h> #include <mutex.h> #include <sema.h> @@ -75,6 +74,7 @@ #include <linux/notifier.h> #include <linux/delay.h> #include <linux/log2.h> +#include <linux/spinlock.h> #include <asm/page.h> #include <asm/div64.h> @@ -136,43 +136,19 @@ #define current_restore_flags_nested(sp, f) \ (current->flags = ((current->flags & ~(f)) | (*(sp) & (f)))) -#define NBPP PAGE_SIZE -#define NDPP (1 << (PAGE_SHIFT - 9)) +#define spinlock_destroy(lock) #define NBBY 8 /* number of bits per byte */ -#define NBPC PAGE_SIZE /* Number of bytes per click */ -#define BPCSHIFT PAGE_SHIFT /* LOG2(NBPC) if exact */ /* * Size of block device i/o is parameterized here. * Currently the system supports page-sized i/o. */ -#define BLKDEV_IOSHIFT BPCSHIFT +#define BLKDEV_IOSHIFT PAGE_CACHE_SHIFT #define BLKDEV_IOSIZE (1<<BLKDEV_IOSHIFT) /* number of BB's per block device block */ #define BLKDEV_BB BTOBB(BLKDEV_IOSIZE) -/* bytes to clicks */ -#define btoc(x) (((__psunsigned_t)(x)+(NBPC-1))>>BPCSHIFT) -#define btoct(x) ((__psunsigned_t)(x)>>BPCSHIFT) -#define btoc64(x) (((__uint64_t)(x)+(NBPC-1))>>BPCSHIFT) -#define btoct64(x) ((__uint64_t)(x)>>BPCSHIFT) - -/* off_t bytes to clicks */ -#define offtoc(x) (((__uint64_t)(x)+(NBPC-1))>>BPCSHIFT) -#define offtoct(x) ((xfs_off_t)(x)>>BPCSHIFT) - -/* clicks to off_t bytes */ -#define ctooff(x) ((xfs_off_t)(x)<<BPCSHIFT) - -/* clicks to bytes */ -#define ctob(x) ((__psunsigned_t)(x)<<BPCSHIFT) -#define btoct(x) ((__psunsigned_t)(x)>>BPCSHIFT) -#define ctob64(x) ((__uint64_t)(x)<<BPCSHIFT) - -/* bytes to clicks */ -#define btoc(x) (((__psunsigned_t)(x)+(NBPC-1))>>BPCSHIFT) - #define ENOATTR ENODATA /* Attribute not found */ #define EWRONGFS EINVAL /* Mount with wrong filesystem type */ #define EFSCORRUPTED EUCLEAN /* Filesystem is corrupted */ @@ -205,10 +181,6 @@ #define xfs_stack_trace() dump_stack() #define xfs_itruncate_data(ip, off) \ (-vmtruncate(vn_to_inode(XFS_ITOV(ip)), (off))) -#define xfs_statvfs_fsid(statp, mp) \ - ({ u64 id = huge_encode_dev((mp)->m_ddev_targp->bt_dev); \ - __kernel_fsid_t *fsid = &(statp)->f_fsid; \ - (fsid->val[0] = (u32)id, fsid->val[1] = (u32)(id >> 32)); }) /* Move the kernel do_div definition off to one side */ diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c index d6a8dddb226..16635338849 100644 --- a/fs/xfs/linux-2.6/xfs_lrw.c +++ b/fs/xfs/linux-2.6/xfs_lrw.c @@ -58,14 +58,12 @@ void xfs_rw_enter_trace( int tag, - xfs_iocore_t *io, + xfs_inode_t *ip, void *data, size_t segs, loff_t offset, int ioflags) { - xfs_inode_t *ip = XFS_IO_INODE(io); - if (ip->i_rwtrace == NULL) return; ktrace_enter(ip->i_rwtrace, @@ -78,8 +76,8 @@ xfs_rw_enter_trace( (void *)((unsigned long)((offset >> 32) & 0xffffffff)), (void *)((unsigned long)(offset & 0xffffffff)), (void *)((unsigned long)ioflags), - (void *)((unsigned long)((io->io_new_size >> 32) & 0xffffffff)), - (void *)((unsigned long)(io->io_new_size & 0xffffffff)), + (void *)((unsigned long)((ip->i_new_size >> 32) & 0xffffffff)), + (void *)((unsigned long)(ip->i_new_size & 0xffffffff)), (void *)((unsigned long)current_pid()), (void *)NULL, (void *)NULL, @@ -89,13 +87,12 @@ xfs_rw_enter_trace( void xfs_inval_cached_trace( - xfs_iocore_t *io, + xfs_inode_t *ip, xfs_off_t offset, xfs_off_t len, xfs_off_t first, xfs_off_t last) { - xfs_inode_t *ip = XFS_IO_INODE(io); if (ip->i_rwtrace == NULL) return; @@ -131,7 +128,7 @@ xfs_inval_cached_trace( */ STATIC int xfs_iozero( - struct inode *ip, /* inode */ + struct xfs_inode *ip, /* inode */ loff_t pos, /* offset in file */ size_t count) /* size of data to zero */ { @@ -139,7 +136,7 @@ xfs_iozero( struct address_space *mapping; int status; - mapping = ip->i_mapping; + mapping = ip->i_vnode->i_mapping; do { unsigned offset, bytes; void *fsdata; @@ -155,7 +152,7 @@ xfs_iozero( if (status) break; - zero_user_page(page, offset, bytes, KM_USER0); + zero_user(page, offset, bytes); status = pagecache_write_end(NULL, mapping, pos, bytes, bytes, page, fsdata); @@ -205,7 +202,7 @@ xfs_read( if (unlikely(ioflags & IO_ISDIRECT)) { xfs_buftarg_t *target = - (ip->i_d.di_flags & XFS_DIFLAG_REALTIME) ? + XFS_IS_REALTIME_INODE(ip) ? mp->m_rtdev_targp : mp->m_ddev_targp; if ((*offset & target->bt_smask) || (size & target->bt_smask)) { @@ -246,9 +243,8 @@ xfs_read( if (unlikely(ioflags & IO_ISDIRECT)) { if (VN_CACHED(vp)) - ret = xfs_flushinval_pages(ip, - ctooff(offtoct(*offset)), - -1, FI_REMAPF_LOCKED); + ret = xfs_flushinval_pages(ip, (*offset & PAGE_CACHE_MASK), + -1, FI_REMAPF_LOCKED); mutex_unlock(&inode->i_mutex); if (ret) { xfs_iunlock(ip, XFS_IOLOCK_SHARED); @@ -256,7 +252,7 @@ xfs_read( } } - xfs_rw_enter_trace(XFS_READ_ENTER, &ip->i_iocore, + xfs_rw_enter_trace(XFS_READ_ENTER, ip, (void *)iovp, segs, *offset, ioflags); iocb->ki_pos = *offset; @@ -301,7 +297,7 @@ xfs_splice_read( return -error; } } - xfs_rw_enter_trace(XFS_SPLICE_READ_ENTER, &ip->i_iocore, + xfs_rw_enter_trace(XFS_SPLICE_READ_ENTER, ip, pipe, count, *ppos, ioflags); ret = generic_file_splice_read(infilp, ppos, pipe, count, flags); if (ret > 0) @@ -323,7 +319,6 @@ xfs_splice_write( { bhv_vnode_t *vp = XFS_ITOV(ip); xfs_mount_t *mp = ip->i_mount; - xfs_iocore_t *io = &ip->i_iocore; ssize_t ret; struct inode *inode = outfilp->f_mapping->host; xfs_fsize_t isize, new_size; @@ -350,10 +345,10 @@ xfs_splice_write( xfs_ilock(ip, XFS_ILOCK_EXCL); if (new_size > ip->i_size) - io->io_new_size = new_size; + ip->i_new_size = new_size; xfs_iunlock(ip, XFS_ILOCK_EXCL); - xfs_rw_enter_trace(XFS_SPLICE_WRITE_ENTER, &ip->i_iocore, + xfs_rw_enter_trace(XFS_SPLICE_WRITE_ENTER, ip, pipe, count, *ppos, ioflags); ret = generic_file_splice_write(pipe, outfilp, ppos, count, flags); if (ret > 0) @@ -370,9 +365,9 @@ xfs_splice_write( xfs_iunlock(ip, XFS_ILOCK_EXCL); } - if (io->io_new_size) { + if (ip->i_new_size) { xfs_ilock(ip, XFS_ILOCK_EXCL); - io->io_new_size = 0; + ip->i_new_size = 0; if (ip->i_d.di_size > ip->i_size) ip->i_d.di_size = ip->i_size; xfs_iunlock(ip, XFS_ILOCK_EXCL); @@ -389,20 +384,19 @@ xfs_splice_write( */ STATIC int /* error (positive) */ xfs_zero_last_block( - struct inode *ip, - xfs_iocore_t *io, + xfs_inode_t *ip, xfs_fsize_t offset, xfs_fsize_t isize) { xfs_fileoff_t last_fsb; - xfs_mount_t *mp = io->io_mount; + xfs_mount_t *mp = ip->i_mount; int nimaps; int zero_offset; int zero_len; int error = 0; xfs_bmbt_irec_t imap; - ASSERT(ismrlocked(io->io_lock, MR_UPDATE) != 0); + ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE) != 0); zero_offset = XFS_B_FSB_OFFSET(mp, isize); if (zero_offset == 0) { @@ -415,7 +409,7 @@ xfs_zero_last_block( last_fsb = XFS_B_TO_FSBT(mp, isize); nimaps = 1; - error = XFS_BMAPI(mp, NULL, io, last_fsb, 1, 0, NULL, 0, &imap, + error = xfs_bmapi(NULL, ip, last_fsb, 1, 0, NULL, 0, &imap, &nimaps, NULL, NULL); if (error) { return error; @@ -433,14 +427,14 @@ xfs_zero_last_block( * out sync. We need to drop the ilock while we do this so we * don't deadlock when the buffer cache calls back to us. */ - XFS_IUNLOCK(mp, io, XFS_ILOCK_EXCL| XFS_EXTSIZE_RD); + xfs_iunlock(ip, XFS_ILOCK_EXCL| XFS_EXTSIZE_RD); zero_len = mp->m_sb.sb_blocksize - zero_offset; if (isize + zero_len > offset) zero_len = offset - isize; error = xfs_iozero(ip, isize, zero_len); - XFS_ILOCK(mp, io, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD); + xfs_ilock(ip, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD); ASSERT(error >= 0); return error; } @@ -458,35 +452,33 @@ xfs_zero_last_block( int /* error (positive) */ xfs_zero_eof( - bhv_vnode_t *vp, - xfs_iocore_t *io, + xfs_inode_t *ip, xfs_off_t offset, /* starting I/O offset */ xfs_fsize_t isize) /* current inode size */ { - struct inode *ip = vn_to_inode(vp); + xfs_mount_t *mp = ip->i_mount; xfs_fileoff_t start_zero_fsb; xfs_fileoff_t end_zero_fsb; xfs_fileoff_t zero_count_fsb; xfs_fileoff_t last_fsb; xfs_fileoff_t zero_off; xfs_fsize_t zero_len; - xfs_mount_t *mp = io->io_mount; int nimaps; int error = 0; xfs_bmbt_irec_t imap; - ASSERT(ismrlocked(io->io_lock, MR_UPDATE)); - ASSERT(ismrlocked(io->io_iolock, MR_UPDATE)); + ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE)); + ASSERT(ismrlocked(&ip->i_iolock, MR_UPDATE)); ASSERT(offset > isize); /* * First handle zeroing the block on which isize resides. * We only zero a part of that block so it is handled specially. */ - error = xfs_zero_last_block(ip, io, offset, isize); + error = xfs_zero_last_block(ip, offset, isize); if (error) { - ASSERT(ismrlocked(io->io_lock, MR_UPDATE)); - ASSERT(ismrlocked(io->io_iolock, MR_UPDATE)); + ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE)); + ASSERT(ismrlocked(&ip->i_iolock, MR_UPDATE)); return error; } @@ -514,11 +506,11 @@ xfs_zero_eof( while (start_zero_fsb <= end_zero_fsb) { nimaps = 1; zero_count_fsb = end_zero_fsb - start_zero_fsb + 1; - error = XFS_BMAPI(mp, NULL, io, start_zero_fsb, zero_count_fsb, + error = xfs_bmapi(NULL, ip, start_zero_fsb, zero_count_fsb, 0, NULL, 0, &imap, &nimaps, NULL, NULL); if (error) { - ASSERT(ismrlocked(io->io_lock, MR_UPDATE)); - ASSERT(ismrlocked(io->io_iolock, MR_UPDATE)); + ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE)); + ASSERT(ismrlocked(&ip->i_iolock, MR_UPDATE)); return error; } ASSERT(nimaps > 0); @@ -542,7 +534,7 @@ xfs_zero_eof( * Drop the inode lock while we're doing the I/O. * We'll still have the iolock to protect us. */ - XFS_IUNLOCK(mp, io, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD); + xfs_iunlock(ip, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD); zero_off = XFS_FSB_TO_B(mp, start_zero_fsb); zero_len = XFS_FSB_TO_B(mp, imap.br_blockcount); @@ -558,14 +550,13 @@ xfs_zero_eof( start_zero_fsb = imap.br_startoff + imap.br_blockcount; ASSERT(start_zero_fsb <= (end_zero_fsb + 1)); - XFS_ILOCK(mp, io, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD); + xfs_ilock(ip, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD); } return 0; out_lock: - - XFS_ILOCK(mp, io, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD); + xfs_ilock(ip, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD); ASSERT(error >= 0); return error; } @@ -587,7 +578,6 @@ xfs_write( xfs_mount_t *mp; ssize_t ret = 0, error = 0; xfs_fsize_t isize, new_size; - xfs_iocore_t *io; int iolock; int eventsent = 0; bhv_vrwlock_t locktype; @@ -607,8 +597,7 @@ xfs_write( if (count == 0) return 0; - io = &xip->i_iocore; - mp = io->io_mount; + mp = xip->i_mount; xfs_wait_for_freeze(mp, SB_FREEZE_WRITE); @@ -667,7 +656,7 @@ start: if (ioflags & IO_ISDIRECT) { xfs_buftarg_t *target = - (xip->i_d.di_flags & XFS_DIFLAG_REALTIME) ? + XFS_IS_REALTIME_INODE(xip) ? mp->m_rtdev_targp : mp->m_ddev_targp; if ((pos & target->bt_smask) || (count & target->bt_smask)) { @@ -688,7 +677,7 @@ start: new_size = pos + count; if (new_size > xip->i_size) - io->io_new_size = new_size; + xip->i_new_size = new_size; if (likely(!(ioflags & IO_INVIS))) { file_update_time(file); @@ -706,7 +695,7 @@ start: */ if (pos > xip->i_size) { - error = xfs_zero_eof(vp, io, pos, xip->i_size); + error = xfs_zero_eof(xip, pos, xip->i_size); if (error) { xfs_iunlock(xip, XFS_ILOCK_EXCL); goto out_unlock_internal; @@ -740,10 +729,10 @@ retry: if ((ioflags & IO_ISDIRECT)) { if (VN_CACHED(vp)) { WARN_ON(need_i_mutex == 0); - xfs_inval_cached_trace(io, pos, -1, - ctooff(offtoct(pos)), -1); + xfs_inval_cached_trace(xip, pos, -1, + (pos & PAGE_CACHE_MASK), -1); error = xfs_flushinval_pages(xip, - ctooff(offtoct(pos)), + (pos & PAGE_CACHE_MASK), -1, FI_REMAPF_LOCKED); if (error) goto out_unlock_internal; @@ -751,7 +740,7 @@ retry: if (need_i_mutex) { /* demote the lock now the cached pages are gone */ - XFS_ILOCK_DEMOTE(mp, io, XFS_IOLOCK_EXCL); + xfs_ilock_demote(xip, XFS_IOLOCK_EXCL); mutex_unlock(&inode->i_mutex); iolock = XFS_IOLOCK_SHARED; @@ -759,7 +748,7 @@ retry: need_i_mutex = 0; } - xfs_rw_enter_trace(XFS_DIOWR_ENTER, io, (void *)iovp, segs, + xfs_rw_enter_trace(XFS_DIOWR_ENTER, xip, (void *)iovp, segs, *offset, ioflags); ret = generic_file_direct_write(iocb, iovp, &segs, pos, offset, count, ocount); @@ -779,7 +768,7 @@ retry: goto relock; } } else { - xfs_rw_enter_trace(XFS_WRITE_ENTER, io, (void *)iovp, segs, + xfs_rw_enter_trace(XFS_WRITE_ENTER, xip, (void *)iovp, segs, *offset, ioflags); ret = generic_file_buffered_write(iocb, iovp, segs, pos, offset, count, ret); @@ -843,9 +832,9 @@ retry: } out_unlock_internal: - if (io->io_new_size) { + if (xip->i_new_size) { xfs_ilock(xip, XFS_ILOCK_EXCL); - io->io_new_size = 0; + xip->i_new_size = 0; /* * If this was a direct or synchronous I/O that failed (such * as ENOSPC) then part of the I/O may have been written to @@ -894,25 +883,6 @@ xfs_bdstrat_cb(struct xfs_buf *bp) } } - -int -xfs_bmap( - xfs_inode_t *ip, - xfs_off_t offset, - ssize_t count, - int flags, - xfs_iomap_t *iomapp, - int *niomaps) -{ - xfs_iocore_t *io = &ip->i_iocore; - - ASSERT((ip->i_d.di_mode & S_IFMT) == S_IFREG); - ASSERT(((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) != 0) == - ((ip->i_iocore.io_flags & XFS_IOCORE_RT) != 0)); - - return xfs_iomap(io, offset, count, flags, iomapp, niomaps); -} - /* * Wrapper around bdstrat so that we can stop data * from going to disk in case we are shutting down the filesystem. diff --git a/fs/xfs/linux-2.6/xfs_lrw.h b/fs/xfs/linux-2.6/xfs_lrw.h index 4b7747a828d..e200253139c 100644 --- a/fs/xfs/linux-2.6/xfs_lrw.h +++ b/fs/xfs/linux-2.6/xfs_lrw.h @@ -19,7 +19,6 @@ #define __XFS_LRW_H__ struct xfs_mount; -struct xfs_iocore; struct xfs_inode; struct xfs_bmbt_irec; struct xfs_buf; @@ -60,20 +59,19 @@ struct xfs_iomap; #define XFS_IOMAP_UNWRITTEN 27 #define XFS_SPLICE_READ_ENTER 28 #define XFS_SPLICE_WRITE_ENTER 29 -extern void xfs_rw_enter_trace(int, struct xfs_iocore *, - void *, size_t, loff_t, int); -extern void xfs_inval_cached_trace(struct xfs_iocore *, - xfs_off_t, xfs_off_t, xfs_off_t, xfs_off_t); +extern void xfs_rw_enter_trace(int, struct xfs_inode *, + void *, size_t, loff_t, int); +extern void xfs_inval_cached_trace(struct xfs_inode *, + xfs_off_t, xfs_off_t, xfs_off_t, xfs_off_t); #else -#define xfs_rw_enter_trace(tag, io, data, size, offset, ioflags) -#define xfs_inval_cached_trace(io, offset, len, first, last) +#define xfs_rw_enter_trace(tag, ip, data, size, offset, ioflags) +#define xfs_inval_cached_trace(ip, offset, len, first, last) #endif extern int xfsbdstrat(struct xfs_mount *, struct xfs_buf *); extern int xfs_bdstrat_cb(struct xfs_buf *); extern int xfs_dev_is_read_only(struct xfs_mount *, char *); -extern int xfs_zero_eof(struct inode *, struct xfs_iocore *, xfs_off_t, - xfs_fsize_t); +extern int xfs_zero_eof(struct xfs_inode *, xfs_off_t, xfs_fsize_t); #endif /* __XFS_LRW_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 8cb63c60c04..8831d951879 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -41,6 +41,7 @@ #include "xfs_rtalloc.h" #include "xfs_error.h" #include "xfs_itable.h" +#include "xfs_fsops.h" #include "xfs_rw.h" #include "xfs_acl.h" #include "xfs_attr.h" @@ -49,6 +50,8 @@ #include "xfs_vnodeops.h" #include "xfs_vfsops.h" #include "xfs_version.h" +#include "xfs_log_priv.h" +#include "xfs_trans_priv.h" #include <linux/namei.h> #include <linux/init.h> @@ -87,6 +90,435 @@ xfs_args_allocate( return args; } +#define MNTOPT_LOGBUFS "logbufs" /* number of XFS log buffers */ +#define MNTOPT_LOGBSIZE "logbsize" /* size of XFS log buffers */ +#define MNTOPT_LOGDEV "logdev" /* log device */ +#define MNTOPT_RTDEV "rtdev" /* realtime I/O device */ +#define MNTOPT_BIOSIZE "biosize" /* log2 of preferred buffered io size */ +#define MNTOPT_WSYNC "wsync" /* safe-mode nfs compatible mount */ +#define MNTOPT_INO64 "ino64" /* force inodes into 64-bit range */ +#define MNTOPT_NOALIGN "noalign" /* turn off stripe alignment */ +#define MNTOPT_SWALLOC "swalloc" /* turn on stripe width allocation */ +#define MNTOPT_SUNIT "sunit" /* data volume stripe unit */ +#define MNTOPT_SWIDTH "swidth" /* data volume stripe width */ +#define MNTOPT_NOUUID "nouuid" /* ignore filesystem UUID */ +#define MNTOPT_MTPT "mtpt" /* filesystem mount point */ +#define MNTOPT_GRPID "grpid" /* group-ID from parent directory */ +#define MNTOPT_NOGRPID "nogrpid" /* group-ID from current process */ +#define MNTOPT_BSDGROUPS "bsdgroups" /* group-ID from parent directory */ +#define MNTOPT_SYSVGROUPS "sysvgroups" /* group-ID from current process */ +#define MNTOPT_ALLOCSIZE "allocsize" /* preferred allocation size */ +#define MNTOPT_NORECOVERY "norecovery" /* don't run XFS recovery */ +#define MNTOPT_BARRIER "barrier" /* use writer barriers for log write and + * unwritten extent conversion */ +#define MNTOPT_NOBARRIER "nobarrier" /* .. disable */ +#define MNTOPT_OSYNCISOSYNC "osyncisosync" /* o_sync is REALLY o_sync */ +#define MNTOPT_64BITINODE "inode64" /* inodes can be allocated anywhere */ +#define MNTOPT_IKEEP "ikeep" /* do not free empty inode clusters */ +#define MNTOPT_NOIKEEP "noikeep" /* free empty inode clusters */ +#define MNTOPT_LARGEIO "largeio" /* report large I/O sizes in stat() */ +#define MNTOPT_NOLARGEIO "nolargeio" /* do not report large I/O sizes + * in stat(). */ +#define MNTOPT_ATTR2 "attr2" /* do use attr2 attribute format */ +#define MNTOPT_NOATTR2 "noattr2" /* do not use attr2 attribute format */ +#define MNTOPT_FILESTREAM "filestreams" /* use filestreams allocator */ +#define MNTOPT_QUOTA "quota" /* disk quotas (user) */ +#define MNTOPT_NOQUOTA "noquota" /* no quotas */ +#define MNTOPT_USRQUOTA "usrquota" /* user quota enabled */ +#define MNTOPT_GRPQUOTA "grpquota" /* group quota enabled */ +#define MNTOPT_PRJQUOTA "prjquota" /* project quota enabled */ +#define MNTOPT_UQUOTA "uquota" /* user quota (IRIX variant) */ +#define MNTOPT_GQUOTA "gquota" /* group quota (IRIX variant) */ +#define MNTOPT_PQUOTA "pquota" /* project quota (IRIX variant) */ +#define MNTOPT_UQUOTANOENF "uqnoenforce"/* user quota limit enforcement */ +#define MNTOPT_GQUOTANOENF "gqnoenforce"/* group quota limit enforcement */ +#define MNTOPT_PQUOTANOENF "pqnoenforce"/* project quota limit enforcement */ +#define MNTOPT_QUOTANOENF "qnoenforce" /* same as uqnoenforce */ +#define MNTOPT_DMAPI "dmapi" /* DMI enabled (DMAPI / XDSM) */ +#define MNTOPT_XDSM "xdsm" /* DMI enabled (DMAPI / XDSM) */ +#define MNTOPT_DMI "dmi" /* DMI enabled (DMAPI / XDSM) */ + +STATIC unsigned long +suffix_strtoul(char *s, char **endp, unsigned int base) +{ + int last, shift_left_factor = 0; + char *value = s; + + last = strlen(value) - 1; + if (value[last] == 'K' || value[last] == 'k') { + shift_left_factor = 10; + value[last] = '\0'; + } + if (value[last] == 'M' || value[last] == 'm') { + shift_left_factor = 20; + value[last] = '\0'; + } + if (value[last] == 'G' || value[last] == 'g') { + shift_left_factor = 30; + value[last] = '\0'; + } + + return simple_strtoul((const char *)s, endp, base) << shift_left_factor; +} + +STATIC int +xfs_parseargs( + struct xfs_mount *mp, + char *options, + struct xfs_mount_args *args, + int update) +{ + char *this_char, *value, *eov; + int dsunit, dswidth, vol_dsunit, vol_dswidth; + int iosize; + int dmapi_implies_ikeep = 1; + + args->flags |= XFSMNT_BARRIER; + args->flags2 |= XFSMNT2_COMPAT_IOSIZE; + + if (!options) + goto done; + + iosize = dsunit = dswidth = vol_dsunit = vol_dswidth = 0; + + while ((this_char = strsep(&options, ",")) != NULL) { + if (!*this_char) + continue; + if ((value = strchr(this_char, '=')) != NULL) + *value++ = 0; + + if (!strcmp(this_char, MNTOPT_LOGBUFS)) { + if (!value || !*value) { + cmn_err(CE_WARN, + "XFS: %s option requires an argument", + this_char); + return EINVAL; + } + args->logbufs = simple_strtoul(value, &eov, 10); + } else if (!strcmp(this_char, MNTOPT_LOGBSIZE)) { + if (!value || !*value) { + cmn_err(CE_WARN, + "XFS: %s option requires an argument", + this_char); + return EINVAL; + } + args->logbufsize = suffix_strtoul(value, &eov, 10); + } else if (!strcmp(this_char, MNTOPT_LOGDEV)) { + if (!value || !*value) { + cmn_err(CE_WARN, + "XFS: %s option requires an argument", + this_char); + return EINVAL; + } + strncpy(args->logname, value, MAXNAMELEN); + } else if (!strcmp(this_char, MNTOPT_MTPT)) { + if (!value || !*value) { + cmn_err(CE_WARN, + "XFS: %s option requires an argument", + this_char); + return EINVAL; + } + strncpy(args->mtpt, value, MAXNAMELEN); + } else if (!strcmp(this_char, MNTOPT_RTDEV)) { + if (!value || !*value) { + cmn_err(CE_WARN, + "XFS: %s option requires an argument", + this_char); + return EINVAL; + } + strncpy(args->rtname, value, MAXNAMELEN); + } else if (!strcmp(this_char, MNTOPT_BIOSIZE)) { + if (!value || !*value) { + cmn_err(CE_WARN, + "XFS: %s option requires an argument", + this_char); + return EINVAL; + } + iosize = simple_strtoul(value, &eov, 10); + args->flags |= XFSMNT_IOSIZE; + args->iosizelog = (uint8_t) iosize; + } else if (!strcmp(this_char, MNTOPT_ALLOCSIZE)) { + if (!value || !*value) { + cmn_err(CE_WARN, + "XFS: %s option requires an argument", + this_char); + return EINVAL; + } + iosize = suffix_strtoul(value, &eov, 10); + args->flags |= XFSMNT_IOSIZE; + args->iosizelog = ffs(iosize) - 1; + } else if (!strcmp(this_char, MNTOPT_GRPID) || + !strcmp(this_char, MNTOPT_BSDGROUPS)) { + mp->m_flags |= XFS_MOUNT_GRPID; + } else if (!strcmp(this_char, MNTOPT_NOGRPID) || + !strcmp(this_char, MNTOPT_SYSVGROUPS)) { + mp->m_flags &= ~XFS_MOUNT_GRPID; + } else if (!strcmp(this_char, MNTOPT_WSYNC)) { + args->flags |= XFSMNT_WSYNC; + } else if (!strcmp(this_char, MNTOPT_OSYNCISOSYNC)) { + args->flags |= XFSMNT_OSYNCISOSYNC; + } else if (!strcmp(this_char, MNTOPT_NORECOVERY)) { + args->flags |= XFSMNT_NORECOVERY; + } else if (!strcmp(this_char, MNTOPT_INO64)) { + args->flags |= XFSMNT_INO64; +#if !XFS_BIG_INUMS + cmn_err(CE_WARN, + "XFS: %s option not allowed on this system", + this_char); + return EINVAL; +#endif + } else if (!strcmp(this_char, MNTOPT_NOALIGN)) { + args->flags |= XFSMNT_NOALIGN; + } else if (!strcmp(this_char, MNTOPT_SWALLOC)) { + args->flags |= XFSMNT_SWALLOC; + } else if (!strcmp(this_char, MNTOPT_SUNIT)) { + if (!value || !*value) { + cmn_err(CE_WARN, + "XFS: %s option requires an argument", + this_char); + return EINVAL; + } + dsunit = simple_strtoul(value, &eov, 10); + } else if (!strcmp(this_char, MNTOPT_SWIDTH)) { + if (!value || !*value) { + cmn_err(CE_WARN, + "XFS: %s option requires an argument", + this_char); + return EINVAL; + } + dswidth = simple_strtoul(value, &eov, 10); + } else if (!strcmp(this_char, MNTOPT_64BITINODE)) { + args->flags &= ~XFSMNT_32BITINODES; +#if !XFS_BIG_INUMS + cmn_err(CE_WARN, + "XFS: %s option not allowed on this system", + this_char); + return EINVAL; +#endif + } else if (!strcmp(this_char, MNTOPT_NOUUID)) { + args->flags |= XFSMNT_NOUUID; + } else if (!strcmp(this_char, MNTOPT_BARRIER)) { + args->flags |= XFSMNT_BARRIER; + } else if (!strcmp(this_char, MNTOPT_NOBARRIER)) { + args->flags &= ~XFSMNT_BARRIER; + } else if (!strcmp(this_char, MNTOPT_IKEEP)) { + args->flags |= XFSMNT_IKEEP; + } else if (!strcmp(this_char, MNTOPT_NOIKEEP)) { + dmapi_implies_ikeep = 0; + args->flags &= ~XFSMNT_IKEEP; + } else if (!strcmp(this_char, MNTOPT_LARGEIO)) { + args->flags2 &= ~XFSMNT2_COMPAT_IOSIZE; + } else if (!strcmp(this_char, MNTOPT_NOLARGEIO)) { + args->flags2 |= XFSMNT2_COMPAT_IOSIZE; + } else if (!strcmp(this_char, MNTOPT_ATTR2)) { + args->flags |= XFSMNT_ATTR2; + } else if (!strcmp(this_char, MNTOPT_NOATTR2)) { + args->flags &= ~XFSMNT_ATTR2; + } else if (!strcmp(this_char, MNTOPT_FILESTREAM)) { + args->flags2 |= XFSMNT2_FILESTREAMS; + } else if (!strcmp(this_char, MNTOPT_NOQUOTA)) { + args->flags &= ~(XFSMNT_UQUOTAENF|XFSMNT_UQUOTA); + args->flags &= ~(XFSMNT_GQUOTAENF|XFSMNT_GQUOTA); + } else if (!strcmp(this_char, MNTOPT_QUOTA) || + !strcmp(this_char, MNTOPT_UQUOTA) || + !strcmp(this_char, MNTOPT_USRQUOTA)) { + args->flags |= XFSMNT_UQUOTA | XFSMNT_UQUOTAENF; + } else if (!strcmp(this_char, MNTOPT_QUOTANOENF) || + !strcmp(this_char, MNTOPT_UQUOTANOENF)) { + args->flags |= XFSMNT_UQUOTA; + args->flags &= ~XFSMNT_UQUOTAENF; + } else if (!strcmp(this_char, MNTOPT_PQUOTA) || + !strcmp(this_char, MNTOPT_PRJQUOTA)) { + args->flags |= XFSMNT_PQUOTA | XFSMNT_PQUOTAENF; + } else if (!strcmp(this_char, MNTOPT_PQUOTANOENF)) { + args->flags |= XFSMNT_PQUOTA; + args->flags &= ~XFSMNT_PQUOTAENF; + } else if (!strcmp(this_char, MNTOPT_GQUOTA) || + !strcmp(this_char, MNTOPT_GRPQUOTA)) { + args->flags |= XFSMNT_GQUOTA | XFSMNT_GQUOTAENF; + } else if (!strcmp(this_char, MNTOPT_GQUOTANOENF)) { + args->flags |= XFSMNT_GQUOTA; + args->flags &= ~XFSMNT_GQUOTAENF; + } else if (!strcmp(this_char, MNTOPT_DMAPI)) { + args->flags |= XFSMNT_DMAPI; + } else if (!strcmp(this_char, MNTOPT_XDSM)) { + args->flags |= XFSMNT_DMAPI; + } else if (!strcmp(this_char, MNTOPT_DMI)) { + args->flags |= XFSMNT_DMAPI; + } else if (!strcmp(this_char, "ihashsize")) { + cmn_err(CE_WARN, + "XFS: ihashsize no longer used, option is deprecated."); + } else if (!strcmp(this_char, "osyncisdsync")) { + /* no-op, this is now the default */ + cmn_err(CE_WARN, + "XFS: osyncisdsync is now the default, option is deprecated."); + } else if (!strcmp(this_char, "irixsgid")) { + cmn_err(CE_WARN, + "XFS: irixsgid is now a sysctl(2) variable, option is deprecated."); + } else { + cmn_err(CE_WARN, + "XFS: unknown mount option [%s].", this_char); + return EINVAL; + } + } + + if (args->flags & XFSMNT_NORECOVERY) { + if ((mp->m_flags & XFS_MOUNT_RDONLY) == 0) { + cmn_err(CE_WARN, + "XFS: no-recovery mounts must be read-only."); + return EINVAL; + } + } + + if ((args->flags & XFSMNT_NOALIGN) && (dsunit || dswidth)) { + cmn_err(CE_WARN, + "XFS: sunit and swidth options incompatible with the noalign option"); + return EINVAL; + } + + if ((args->flags & XFSMNT_GQUOTA) && (args->flags & XFSMNT_PQUOTA)) { + cmn_err(CE_WARN, + "XFS: cannot mount with both project and group quota"); + return EINVAL; + } + + if ((args->flags & XFSMNT_DMAPI) && *args->mtpt == '\0') { + printk("XFS: %s option needs the mount point option as well\n", + MNTOPT_DMAPI); + return EINVAL; + } + + if ((dsunit && !dswidth) || (!dsunit && dswidth)) { + cmn_err(CE_WARN, + "XFS: sunit and swidth must be specified together"); + return EINVAL; + } + + if (dsunit && (dswidth % dsunit != 0)) { + cmn_err(CE_WARN, + "XFS: stripe width (%d) must be a multiple of the stripe unit (%d)", + dswidth, dsunit); + return EINVAL; + } + + /* + * Applications using DMI filesystems often expect the + * inode generation number to be monotonically increasing. + * If we delete inode chunks we break this assumption, so + * keep unused inode chunks on disk for DMI filesystems + * until we come up with a better solution. + * Note that if "ikeep" or "noikeep" mount options are + * supplied, then they are honored. + */ + if ((args->flags & XFSMNT_DMAPI) && dmapi_implies_ikeep) + args->flags |= XFSMNT_IKEEP; + + if ((args->flags & XFSMNT_NOALIGN) != XFSMNT_NOALIGN) { + if (dsunit) { + args->sunit = dsunit; + args->flags |= XFSMNT_RETERR; + } else { + args->sunit = vol_dsunit; + } + dswidth ? (args->swidth = dswidth) : + (args->swidth = vol_dswidth); + } else { + args->sunit = args->swidth = 0; + } + +done: + if (args->flags & XFSMNT_32BITINODES) + mp->m_flags |= XFS_MOUNT_SMALL_INUMS; + if (args->flags2) + args->flags |= XFSMNT_FLAGS2; + return 0; +} + +struct proc_xfs_info { + int flag; + char *str; +}; + +STATIC int +xfs_showargs( + struct xfs_mount *mp, + struct seq_file *m) +{ + static struct proc_xfs_info xfs_info_set[] = { + /* the few simple ones we can get from the mount struct */ + { XFS_MOUNT_IKEEP, "," MNTOPT_IKEEP }, + { XFS_MOUNT_WSYNC, "," MNTOPT_WSYNC }, + { XFS_MOUNT_INO64, "," MNTOPT_INO64 }, + { XFS_MOUNT_NOALIGN, "," MNTOPT_NOALIGN }, + { XFS_MOUNT_SWALLOC, "," MNTOPT_SWALLOC }, + { XFS_MOUNT_NOUUID, "," MNTOPT_NOUUID }, + { XFS_MOUNT_NORECOVERY, "," MNTOPT_NORECOVERY }, + { XFS_MOUNT_OSYNCISOSYNC, "," MNTOPT_OSYNCISOSYNC }, + { XFS_MOUNT_ATTR2, "," MNTOPT_ATTR2 }, + { XFS_MOUNT_FILESTREAMS, "," MNTOPT_FILESTREAM }, + { XFS_MOUNT_DMAPI, "," MNTOPT_DMAPI }, + { XFS_MOUNT_GRPID, "," MNTOPT_GRPID }, + { 0, NULL } + }; + static struct proc_xfs_info xfs_info_unset[] = { + /* the few simple ones we can get from the mount struct */ + { XFS_MOUNT_COMPAT_IOSIZE, "," MNTOPT_LARGEIO }, + { XFS_MOUNT_BARRIER, "," MNTOPT_NOBARRIER }, + { XFS_MOUNT_SMALL_INUMS, "," MNTOPT_64BITINODE }, + { 0, NULL } + }; + struct proc_xfs_info *xfs_infop; + + for (xfs_infop = xfs_info_set; xfs_infop->flag; xfs_infop++) { + if (mp->m_flags & xfs_infop->flag) + seq_puts(m, xfs_infop->str); + } + for (xfs_infop = xfs_info_unset; xfs_infop->flag; xfs_infop++) { + if (!(mp->m_flags & xfs_infop->flag)) + seq_puts(m, xfs_infop->str); + } + + if (mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) + seq_printf(m, "," MNTOPT_ALLOCSIZE "=%dk", + (int)(1 << mp->m_writeio_log) >> 10); + + if (mp->m_logbufs > 0) + seq_printf(m, "," MNTOPT_LOGBUFS "=%d", mp->m_logbufs); + if (mp->m_logbsize > 0) + seq_printf(m, "," MNTOPT_LOGBSIZE "=%dk", mp->m_logbsize >> 10); + + if (mp->m_logname) + seq_printf(m, "," MNTOPT_LOGDEV "=%s", mp->m_logname); + if (mp->m_rtname) + seq_printf(m, "," MNTOPT_RTDEV "=%s", mp->m_rtname); + + if (mp->m_dalign > 0) + seq_printf(m, "," MNTOPT_SUNIT "=%d", + (int)XFS_FSB_TO_BB(mp, mp->m_dalign)); + if (mp->m_swidth > 0) + seq_printf(m, "," MNTOPT_SWIDTH "=%d", + (int)XFS_FSB_TO_BB(mp, mp->m_swidth)); + + if (mp->m_qflags & (XFS_UQUOTA_ACCT|XFS_UQUOTA_ENFD)) + seq_puts(m, "," MNTOPT_USRQUOTA); + else if (mp->m_qflags & XFS_UQUOTA_ACCT) + seq_puts(m, "," MNTOPT_UQUOTANOENF); + + if (mp->m_qflags & (XFS_PQUOTA_ACCT|XFS_OQUOTA_ENFD)) + seq_puts(m, "," MNTOPT_PRJQUOTA); + else if (mp->m_qflags & XFS_PQUOTA_ACCT) + seq_puts(m, "," MNTOPT_PQUOTANOENF); + + if (mp->m_qflags & (XFS_GQUOTA_ACCT|XFS_OQUOTA_ENFD)) + seq_puts(m, "," MNTOPT_GRPQUOTA); + else if (mp->m_qflags & XFS_GQUOTA_ACCT) + seq_puts(m, "," MNTOPT_GQUOTANOENF); + + if (!(mp->m_qflags & XFS_ALL_QUOTA_ACCT)) + seq_puts(m, "," MNTOPT_NOQUOTA); + + return 0; +} __uint64_t xfs_max_file_offset( unsigned int blockshift) @@ -137,7 +569,7 @@ xfs_set_inodeops( break; case S_IFLNK: inode->i_op = &xfs_symlink_inode_operations; - if (inode->i_blocks) + if (!(XFS_I(inode)->i_df.if_flags & XFS_IFINLINE)) inode->i_mapping->a_ops = &xfs_address_space_operations; break; default: @@ -174,8 +606,6 @@ xfs_revalidate_inode( inode->i_generation = ip->i_d.di_gen; i_size_write(inode, ip->i_d.di_size); - inode->i_blocks = - XFS_FSB_TO_BB(mp, ip->i_d.di_nblocks + ip->i_delayed_blks); inode->i_atime.tv_sec = ip->i_d.di_atime.t_sec; inode->i_atime.tv_nsec = ip->i_d.di_atime.t_nsec; inode->i_mtime.tv_sec = ip->i_d.di_mtime.t_sec; @@ -334,6 +764,64 @@ xfs_blkdev_issue_flush( blkdev_issue_flush(buftarg->bt_bdev, NULL); } +/* + * XFS AIL push thread support + */ +void +xfsaild_wakeup( + xfs_mount_t *mp, + xfs_lsn_t threshold_lsn) +{ + mp->m_ail.xa_target = threshold_lsn; + wake_up_process(mp->m_ail.xa_task); +} + +int +xfsaild( + void *data) +{ + xfs_mount_t *mp = (xfs_mount_t *)data; + xfs_lsn_t last_pushed_lsn = 0; + long tout = 0; + + while (!kthread_should_stop()) { + if (tout) + schedule_timeout_interruptible(msecs_to_jiffies(tout)); + tout = 1000; + + /* swsusp */ + try_to_freeze(); + + ASSERT(mp->m_log); + if (XFS_FORCED_SHUTDOWN(mp)) + continue; + + tout = xfsaild_push(mp, &last_pushed_lsn); + } + + return 0; +} /* xfsaild */ + +int +xfsaild_start( + xfs_mount_t *mp) +{ + mp->m_ail.xa_target = 0; + mp->m_ail.xa_task = kthread_run(xfsaild, mp, "xfsaild"); + if (IS_ERR(mp->m_ail.xa_task)) + return -PTR_ERR(mp->m_ail.xa_task); + return 0; +} + +void +xfsaild_stop( + xfs_mount_t *mp) +{ + kthread_stop(mp->m_ail.xa_task); +} + + + STATIC struct inode * xfs_fs_alloc_inode( struct super_block *sb) @@ -361,7 +849,7 @@ xfs_fs_inode_init_once( inode_init_once(vn_to_inode((bhv_vnode_t *)vnode)); } -STATIC int +STATIC int __init xfs_init_zones(void) { xfs_vnode_zone = kmem_zone_init_flags(sizeof(bhv_vnode_t), "xfs_vnode", @@ -410,8 +898,7 @@ xfs_fs_write_inode( { int error = 0, flags = FLUSH_INODE; - vn_trace_entry(XFS_I(inode), __FUNCTION__, - (inst_t *)__return_address); + xfs_itrace_entry(XFS_I(inode)); if (sync) { filemap_fdatawait(inode->i_mapping); flags |= FLUSH_SYNC; @@ -438,8 +925,7 @@ xfs_fs_clear_inode( * find an inode with di_mode == 0 but without IGET_CREATE set. */ if (ip) { - vn_trace_entry(ip, __FUNCTION__, (inst_t *)__return_address); - + xfs_itrace_entry(ip); XFS_STATS_INC(vn_rele); XFS_STATS_INC(vn_remove); XFS_STATS_INC(vn_reclaim); @@ -683,8 +1169,44 @@ xfs_fs_statfs( struct dentry *dentry, struct kstatfs *statp) { - return -xfs_statvfs(XFS_M(dentry->d_sb), statp, - vn_from_inode(dentry->d_inode)); + struct xfs_mount *mp = XFS_M(dentry->d_sb); + xfs_sb_t *sbp = &mp->m_sb; + __uint64_t fakeinos, id; + xfs_extlen_t lsize; + + statp->f_type = XFS_SB_MAGIC; + statp->f_namelen = MAXNAMELEN - 1; + + id = huge_encode_dev(mp->m_ddev_targp->bt_dev); + statp->f_fsid.val[0] = (u32)id; + statp->f_fsid.val[1] = (u32)(id >> 32); + + xfs_icsb_sync_counters_flags(mp, XFS_ICSB_LAZY_COUNT); + + spin_lock(&mp->m_sb_lock); + statp->f_bsize = sbp->sb_blocksize; + lsize = sbp->sb_logstart ? sbp->sb_logblocks : 0; + statp->f_blocks = sbp->sb_dblocks - lsize; + statp->f_bfree = statp->f_bavail = + sbp->sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp); + fakeinos = statp->f_bfree << sbp->sb_inopblog; +#if XFS_BIG_INUMS + fakeinos += mp->m_inoadd; +#endif + statp->f_files = + MIN(sbp->sb_icount + fakeinos, (__uint64_t)XFS_MAXINUMBER); + if (mp->m_maxicount) +#if XFS_BIG_INUMS + if (!mp->m_inoadd) +#endif + statp->f_files = min_t(typeof(statp->f_files), + statp->f_files, + mp->m_maxicount); + statp->f_ffree = statp->f_files - (sbp->sb_icount - sbp->sb_ifree); + spin_unlock(&mp->m_sb_lock); + + XFS_QM_DQSTATVFS(XFS_I(dentry->d_inode), statp); + return 0; } STATIC int @@ -704,11 +1226,19 @@ xfs_fs_remount( return -error; } +/* + * Second stage of a freeze. The data is already frozen so we only + * need to take care of themetadata. Once that's done write a dummy + * record to dirty the log in case of a crash while frozen. + */ STATIC void xfs_fs_lockfs( struct super_block *sb) { - xfs_freeze(XFS_M(sb)); + struct xfs_mount *mp = XFS_M(sb); + + xfs_attr_quiesce(mp); + xfs_fs_log_dummy(mp); } STATIC int @@ -779,7 +1309,6 @@ xfs_fs_fill_super( struct inode *rootvp; struct xfs_mount *mp = NULL; struct xfs_mount_args *args = xfs_args_allocate(sb, silent); - struct kstatfs statvfs; int error; mp = xfs_mount_init(); @@ -807,21 +1336,19 @@ xfs_fs_fill_super( if (error) goto fail_vfsop; - error = xfs_statvfs(mp, &statvfs, NULL); - if (error) - goto fail_unmount; - sb->s_dirt = 1; - sb->s_magic = statvfs.f_type; - sb->s_blocksize = statvfs.f_bsize; - sb->s_blocksize_bits = ffs(statvfs.f_bsize) - 1; + sb->s_magic = XFS_SB_MAGIC; + sb->s_blocksize = mp->m_sb.sb_blocksize; + sb->s_blocksize_bits = ffs(sb->s_blocksize) - 1; sb->s_maxbytes = xfs_max_file_offset(sb->s_blocksize_bits); sb->s_time_gran = 1; set_posix_acl_flag(sb); - error = xfs_root(mp, &rootvp); - if (error) + rootvp = igrab(mp->m_rootip->i_vnode); + if (!rootvp) { + error = ENOENT; goto fail_unmount; + } sb->s_root = d_alloc_root(vn_to_inode(rootvp)); if (!sb->s_root) { @@ -841,8 +1368,7 @@ xfs_fs_fill_super( goto fail_vnrele; } - vn_trace_exit(XFS_I(sb->s_root->d_inode), __FUNCTION__, - (inst_t *)__return_address); + xfs_itrace_exit(XFS_I(sb->s_root->d_inode)); kmem_free(args, sizeof(*args)); return 0; diff --git a/fs/xfs/linux-2.6/xfs_vnode.c b/fs/xfs/linux-2.6/xfs_vnode.c index 814169fd7e1..bc7afe00733 100644 --- a/fs/xfs/linux-2.6/xfs_vnode.c +++ b/fs/xfs/linux-2.6/xfs_vnode.c @@ -40,7 +40,7 @@ #define vptosync(v) (&vsync[((unsigned long)v) % NVSYNC]) static wait_queue_head_t vsync[NVSYNC]; -void +void __init vn_init(void) { int i; @@ -82,84 +82,55 @@ vn_ioerror( xfs_do_force_shutdown(ip->i_mount, SHUTDOWN_DEVICE_REQ, f, l); } -bhv_vnode_t * -vn_initialize( - struct inode *inode) -{ - bhv_vnode_t *vp = vn_from_inode(inode); - - XFS_STATS_INC(vn_active); - XFS_STATS_INC(vn_alloc); - - ASSERT(VN_CACHED(vp) == 0); - - return vp; -} - /* - * Revalidate the Linux inode from the vattr. + * Revalidate the Linux inode from the XFS inode. * Note: i_size _not_ updated; we must hold the inode * semaphore when doing that - callers responsibility. */ -void -vn_revalidate_core( - bhv_vnode_t *vp, - bhv_vattr_t *vap) +int +vn_revalidate( + bhv_vnode_t *vp) { - struct inode *inode = vn_to_inode(vp); - - inode->i_mode = vap->va_mode; - inode->i_nlink = vap->va_nlink; - inode->i_uid = vap->va_uid; - inode->i_gid = vap->va_gid; - inode->i_blocks = vap->va_nblocks; - inode->i_mtime = vap->va_mtime; - inode->i_ctime = vap->va_ctime; - if (vap->va_xflags & XFS_XFLAG_IMMUTABLE) + struct inode *inode = vn_to_inode(vp); + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + unsigned long xflags; + + xfs_itrace_entry(ip); + + if (XFS_FORCED_SHUTDOWN(mp)) + return -EIO; + + xfs_ilock(ip, XFS_ILOCK_SHARED); + inode->i_mode = ip->i_d.di_mode; + inode->i_uid = ip->i_d.di_uid; + inode->i_gid = ip->i_d.di_gid; + inode->i_mtime.tv_sec = ip->i_d.di_mtime.t_sec; + inode->i_mtime.tv_nsec = ip->i_d.di_mtime.t_nsec; + inode->i_ctime.tv_sec = ip->i_d.di_ctime.t_sec; + inode->i_ctime.tv_nsec = ip->i_d.di_ctime.t_nsec; + + xflags = xfs_ip2xflags(ip); + if (xflags & XFS_XFLAG_IMMUTABLE) inode->i_flags |= S_IMMUTABLE; else inode->i_flags &= ~S_IMMUTABLE; - if (vap->va_xflags & XFS_XFLAG_APPEND) + if (xflags & XFS_XFLAG_APPEND) inode->i_flags |= S_APPEND; else inode->i_flags &= ~S_APPEND; - if (vap->va_xflags & XFS_XFLAG_SYNC) + if (xflags & XFS_XFLAG_SYNC) inode->i_flags |= S_SYNC; else inode->i_flags &= ~S_SYNC; - if (vap->va_xflags & XFS_XFLAG_NOATIME) + if (xflags & XFS_XFLAG_NOATIME) inode->i_flags |= S_NOATIME; else inode->i_flags &= ~S_NOATIME; -} - -/* - * Revalidate the Linux inode from the vnode. - */ -int -__vn_revalidate( - bhv_vnode_t *vp, - bhv_vattr_t *vattr) -{ - int error; - - vn_trace_entry(xfs_vtoi(vp), __FUNCTION__, (inst_t *)__return_address); - vattr->va_mask = XFS_AT_STAT | XFS_AT_XFLAGS; - error = xfs_getattr(xfs_vtoi(vp), vattr, 0); - if (likely(!error)) { - vn_revalidate_core(vp, vattr); - xfs_iflags_clear(xfs_vtoi(vp), XFS_IMODIFIED); - } - return -error; -} - -int -vn_revalidate( - bhv_vnode_t *vp) -{ - bhv_vattr_t vattr; + xfs_iunlock(ip, XFS_ILOCK_SHARED); - return __vn_revalidate(vp, &vattr); + xfs_iflags_clear(ip, XFS_IMODIFIED); + return 0; } /* @@ -179,7 +150,7 @@ vn_hold( return vp; } -#ifdef XFS_VNODE_TRACE +#ifdef XFS_INODE_TRACE /* * Reference count of Linux inode if present, -1 if the xfs_inode @@ -211,32 +182,32 @@ static inline int xfs_icount(struct xfs_inode *ip) * Vnode tracing code. */ void -vn_trace_entry(xfs_inode_t *ip, const char *func, inst_t *ra) +_xfs_itrace_entry(xfs_inode_t *ip, const char *func, inst_t *ra) { - KTRACE_ENTER(ip, VNODE_KTRACE_ENTRY, func, 0, ra); + KTRACE_ENTER(ip, INODE_KTRACE_ENTRY, func, 0, ra); } void -vn_trace_exit(xfs_inode_t *ip, const char *func, inst_t *ra) +_xfs_itrace_exit(xfs_inode_t *ip, const char *func, inst_t *ra) { - KTRACE_ENTER(ip, VNODE_KTRACE_EXIT, func, 0, ra); + KTRACE_ENTER(ip, INODE_KTRACE_EXIT, func, 0, ra); } void -vn_trace_hold(xfs_inode_t *ip, char *file, int line, inst_t *ra) +xfs_itrace_hold(xfs_inode_t *ip, char *file, int line, inst_t *ra) { - KTRACE_ENTER(ip, VNODE_KTRACE_HOLD, file, line, ra); + KTRACE_ENTER(ip, INODE_KTRACE_HOLD, file, line, ra); } void -vn_trace_ref(xfs_inode_t *ip, char *file, int line, inst_t *ra) +_xfs_itrace_ref(xfs_inode_t *ip, char *file, int line, inst_t *ra) { - KTRACE_ENTER(ip, VNODE_KTRACE_REF, file, line, ra); + KTRACE_ENTER(ip, INODE_KTRACE_REF, file, line, ra); } void -vn_trace_rele(xfs_inode_t *ip, char *file, int line, inst_t *ra) +xfs_itrace_rele(xfs_inode_t *ip, char *file, int line, inst_t *ra) { - KTRACE_ENTER(ip, VNODE_KTRACE_RELE, file, line, ra); + KTRACE_ENTER(ip, INODE_KTRACE_RELE, file, line, ra); } -#endif /* XFS_VNODE_TRACE */ +#endif /* XFS_INODE_TRACE */ diff --git a/fs/xfs/linux-2.6/xfs_vnode.h b/fs/xfs/linux-2.6/xfs_vnode.h index 55fb4694858..b5ea418693b 100644 --- a/fs/xfs/linux-2.6/xfs_vnode.h +++ b/fs/xfs/linux-2.6/xfs_vnode.h @@ -187,10 +187,7 @@ typedef struct bhv_vattr { (VN_ISREG(vp) && ((mode) & (VSGID|(VEXEC>>3))) == VSGID) extern void vn_init(void); -extern bhv_vnode_t *vn_initialize(struct inode *); extern int vn_revalidate(bhv_vnode_t *); -extern int __vn_revalidate(bhv_vnode_t *, bhv_vattr_t *); -extern void vn_revalidate_core(bhv_vnode_t *, bhv_vattr_t *); /* * Yeah, these don't take vnode anymore at all, all this should be @@ -210,12 +207,12 @@ static inline int vn_count(bhv_vnode_t *vp) */ extern bhv_vnode_t *vn_hold(bhv_vnode_t *); -#if defined(XFS_VNODE_TRACE) +#if defined(XFS_INODE_TRACE) #define VN_HOLD(vp) \ ((void)vn_hold(vp), \ - vn_trace_hold(xfs_vtoi(vp), __FILE__, __LINE__, (inst_t *)__return_address)) + xfs_itrace_hold(xfs_vtoi(vp), __FILE__, __LINE__, (inst_t *)__return_address)) #define VN_RELE(vp) \ - (vn_trace_rele(xfs_vtoi(vp), __FILE__, __LINE__, (inst_t *)__return_address), \ + (xfs_itrace_rele(xfs_vtoi(vp), __FILE__, __LINE__, (inst_t *)__return_address), \ iput(vn_to_inode(vp))) #else #define VN_HOLD(vp) ((void)vn_hold(vp)) @@ -238,11 +235,6 @@ static inline bhv_vnode_t *vn_grab(bhv_vnode_t *vp) /* * Dealing with bad inodes */ -static inline void vn_mark_bad(bhv_vnode_t *vp) -{ - make_bad_inode(vn_to_inode(vp)); -} - static inline int VN_BAD(bhv_vnode_t *vp) { return is_bad_inode(vn_to_inode(vp)); @@ -296,26 +288,36 @@ static inline void vn_atime_to_time_t(bhv_vnode_t *vp, time_t *tt) /* * Tracking vnode activity. */ -#if defined(XFS_VNODE_TRACE) - -#define VNODE_TRACE_SIZE 16 /* number of trace entries */ -#define VNODE_KTRACE_ENTRY 1 -#define VNODE_KTRACE_EXIT 2 -#define VNODE_KTRACE_HOLD 3 -#define VNODE_KTRACE_REF 4 -#define VNODE_KTRACE_RELE 5 - -extern void vn_trace_entry(struct xfs_inode *, const char *, inst_t *); -extern void vn_trace_exit(struct xfs_inode *, const char *, inst_t *); -extern void vn_trace_hold(struct xfs_inode *, char *, int, inst_t *); -extern void vn_trace_ref(struct xfs_inode *, char *, int, inst_t *); -extern void vn_trace_rele(struct xfs_inode *, char *, int, inst_t *); +#if defined(XFS_INODE_TRACE) + +#define INODE_TRACE_SIZE 16 /* number of trace entries */ +#define INODE_KTRACE_ENTRY 1 +#define INODE_KTRACE_EXIT 2 +#define INODE_KTRACE_HOLD 3 +#define INODE_KTRACE_REF 4 +#define INODE_KTRACE_RELE 5 + +extern void _xfs_itrace_entry(struct xfs_inode *, const char *, inst_t *); +extern void _xfs_itrace_exit(struct xfs_inode *, const char *, inst_t *); +extern void xfs_itrace_hold(struct xfs_inode *, char *, int, inst_t *); +extern void _xfs_itrace_ref(struct xfs_inode *, char *, int, inst_t *); +extern void xfs_itrace_rele(struct xfs_inode *, char *, int, inst_t *); +#define xfs_itrace_entry(ip) \ + _xfs_itrace_entry(ip, __FUNCTION__, (inst_t *)__return_address) +#define xfs_itrace_exit(ip) \ + _xfs_itrace_exit(ip, __FUNCTION__, (inst_t *)__return_address) +#define xfs_itrace_exit_tag(ip, tag) \ + _xfs_itrace_exit(ip, tag, (inst_t *)__return_address) +#define xfs_itrace_ref(ip) \ + _xfs_itrace_ref(ip, __FILE__, __LINE__, (inst_t *)__return_address) + #else -#define vn_trace_entry(a,b,c) -#define vn_trace_exit(a,b,c) -#define vn_trace_hold(a,b,c,d) -#define vn_trace_ref(a,b,c,d) -#define vn_trace_rele(a,b,c,d) +#define xfs_itrace_entry(a) +#define xfs_itrace_exit(a) +#define xfs_itrace_exit_tag(a, b) +#define xfs_itrace_hold(a, b, c, d) +#define xfs_itrace_ref(a) +#define xfs_itrace_rele(a, b, c, d) #endif #endif /* __XFS_VNODE_H__ */ diff --git a/fs/xfs/quota/xfs_dquot.c b/fs/xfs/quota/xfs_dquot.c index cfdd35ee9f7..665babcca6a 100644 --- a/fs/xfs/quota/xfs_dquot.c +++ b/fs/xfs/quota/xfs_dquot.c @@ -1209,7 +1209,6 @@ xfs_qm_dqflush( xfs_buf_t *bp; xfs_disk_dquot_t *ddqp; int error; - SPLDECL(s); ASSERT(XFS_DQ_IS_LOCKED(dqp)); ASSERT(XFS_DQ_IS_FLUSH_LOCKED(dqp)); @@ -1270,9 +1269,9 @@ xfs_qm_dqflush( mp = dqp->q_mount; /* lsn is 64 bits */ - AIL_LOCK(mp, s); + spin_lock(&mp->m_ail_lock); dqp->q_logitem.qli_flush_lsn = dqp->q_logitem.qli_item.li_lsn; - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); /* * Attach an iodone routine so that we can remove this dquot from the @@ -1318,7 +1317,6 @@ xfs_qm_dqflush_done( xfs_dq_logitem_t *qip) { xfs_dquot_t *dqp; - SPLDECL(s); dqp = qip->qli_dquot; @@ -1333,15 +1331,15 @@ xfs_qm_dqflush_done( if ((qip->qli_item.li_flags & XFS_LI_IN_AIL) && qip->qli_item.li_lsn == qip->qli_flush_lsn) { - AIL_LOCK(dqp->q_mount, s); + spin_lock(&dqp->q_mount->m_ail_lock); /* * xfs_trans_delete_ail() drops the AIL lock. */ if (qip->qli_item.li_lsn == qip->qli_flush_lsn) xfs_trans_delete_ail(dqp->q_mount, - (xfs_log_item_t*)qip, s); + (xfs_log_item_t*)qip); else - AIL_UNLOCK(dqp->q_mount, s); + spin_unlock(&dqp->q_mount->m_ail_lock); } /* diff --git a/fs/xfs/quota/xfs_dquot.h b/fs/xfs/quota/xfs_dquot.h index 78d3ab95c5f..5c371a92e3e 100644 --- a/fs/xfs/quota/xfs_dquot.h +++ b/fs/xfs/quota/xfs_dquot.h @@ -123,11 +123,6 @@ XFS_DQ_IS_LOCKED(xfs_dquot_t *dqp) vsema(&((dqp)->q_flock)); \ (dqp)->dq_flags &= ~(XFS_DQ_FLOCKED); } -#define XFS_DQ_PINLOCK(dqp) mutex_spinlock( \ - &(XFS_DQ_TO_QINF(dqp)->qi_pinlock)) -#define XFS_DQ_PINUNLOCK(dqp, s) mutex_spinunlock( \ - &(XFS_DQ_TO_QINF(dqp)->qi_pinlock), s) - #define XFS_DQ_IS_FLUSH_LOCKED(dqp) (issemalocked(&((dqp)->q_flock))) #define XFS_DQ_IS_ON_FREELIST(dqp) ((dqp)->dq_flnext != (dqp)) #define XFS_DQ_IS_DIRTY(dqp) ((dqp)->dq_flags & XFS_DQ_DIRTY) diff --git a/fs/xfs/quota/xfs_dquot_item.c b/fs/xfs/quota/xfs_dquot_item.c index ddb61fe22a5..1800e8d1f64 100644 --- a/fs/xfs/quota/xfs_dquot_item.c +++ b/fs/xfs/quota/xfs_dquot_item.c @@ -94,14 +94,13 @@ STATIC void xfs_qm_dquot_logitem_pin( xfs_dq_logitem_t *logitem) { - unsigned long s; xfs_dquot_t *dqp; dqp = logitem->qli_dquot; ASSERT(XFS_DQ_IS_LOCKED(dqp)); - s = XFS_DQ_PINLOCK(dqp); + spin_lock(&(XFS_DQ_TO_QINF(dqp)->qi_pinlock)); dqp->q_pincount++; - XFS_DQ_PINUNLOCK(dqp, s); + spin_unlock(&(XFS_DQ_TO_QINF(dqp)->qi_pinlock)); } /* @@ -115,17 +114,16 @@ xfs_qm_dquot_logitem_unpin( xfs_dq_logitem_t *logitem, int stale) { - unsigned long s; xfs_dquot_t *dqp; dqp = logitem->qli_dquot; ASSERT(dqp->q_pincount > 0); - s = XFS_DQ_PINLOCK(dqp); + spin_lock(&(XFS_DQ_TO_QINF(dqp)->qi_pinlock)); dqp->q_pincount--; if (dqp->q_pincount == 0) { sv_broadcast(&dqp->q_pinwait); } - XFS_DQ_PINUNLOCK(dqp, s); + spin_unlock(&(XFS_DQ_TO_QINF(dqp)->qi_pinlock)); } /* ARGSUSED */ @@ -189,8 +187,6 @@ void xfs_qm_dqunpin_wait( xfs_dquot_t *dqp) { - SPLDECL(s); - ASSERT(XFS_DQ_IS_LOCKED(dqp)); if (dqp->q_pincount == 0) { return; @@ -200,9 +196,9 @@ xfs_qm_dqunpin_wait( * Give the log a push so we don't wait here too long. */ xfs_log_force(dqp->q_mount, (xfs_lsn_t)0, XFS_LOG_FORCE); - s = XFS_DQ_PINLOCK(dqp); + spin_lock(&(XFS_DQ_TO_QINF(dqp)->qi_pinlock)); if (dqp->q_pincount == 0) { - XFS_DQ_PINUNLOCK(dqp, s); + spin_unlock(&(XFS_DQ_TO_QINF(dqp)->qi_pinlock)); return; } sv_wait(&(dqp->q_pinwait), PINOD, @@ -216,8 +212,8 @@ xfs_qm_dqunpin_wait( * If so, we want to push it out to help us take this item off the AIL as soon * as possible. * - * We must not be holding the AIL_LOCK at this point. Calling incore() to - * search the buffer cache can be a time consuming thing, and AIL_LOCK is a + * We must not be holding the AIL lock at this point. Calling incore() to + * search the buffer cache can be a time consuming thing, and AIL lock is a * spinlock. */ STATIC void @@ -322,7 +318,7 @@ xfs_qm_dquot_logitem_trylock( * want to do that now since we might sleep in the device * strategy routine. We also don't want to grab the buffer lock * here because we'd like not to call into the buffer cache - * while holding the AIL_LOCK. + * while holding the AIL lock. * Make sure to only return PUSHBUF if we set pushbuf_flag * ourselves. If someone else is doing it then we don't * want to go to the push routine and duplicate their efforts. @@ -562,15 +558,14 @@ xfs_qm_qoffend_logitem_committed( xfs_lsn_t lsn) { xfs_qoff_logitem_t *qfs; - SPLDECL(s); qfs = qfe->qql_start_lip; - AIL_LOCK(qfs->qql_item.li_mountp,s); + spin_lock(&qfs->qql_item.li_mountp->m_ail_lock); /* * Delete the qoff-start logitem from the AIL. * xfs_trans_delete_ail() drops the AIL lock. */ - xfs_trans_delete_ail(qfs->qql_item.li_mountp, (xfs_log_item_t *)qfs, s); + xfs_trans_delete_ail(qfs->qql_item.li_mountp, (xfs_log_item_t *)qfs); kmem_free(qfs, sizeof(xfs_qoff_logitem_t)); kmem_free(qfe, sizeof(xfs_qoff_logitem_t)); return (xfs_lsn_t)-1; diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c index d488645f833..8e9c5ae6504 100644 --- a/fs/xfs/quota/xfs_qm.c +++ b/fs/xfs/quota/xfs_qm.c @@ -310,7 +310,6 @@ xfs_qm_mount_quotas( xfs_mount_t *mp, int mfsi_flags) { - unsigned long s; int error = 0; uint sbf; @@ -367,13 +366,13 @@ xfs_qm_mount_quotas( write_changes: /* - * We actually don't have to acquire the SB_LOCK at all. + * We actually don't have to acquire the m_sb_lock at all. * This can only be called from mount, and that's single threaded. XXX */ - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); sbf = mp->m_sb.sb_qflags; mp->m_sb.sb_qflags = mp->m_qflags & XFS_MOUNT_QUOTA_ALL; - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); if (sbf != (mp->m_qflags & XFS_MOUNT_QUOTA_ALL)) { if (xfs_qm_write_sb_changes(mp, XFS_SB_QFLAGS)) { @@ -1139,7 +1138,7 @@ xfs_qm_init_quotainfo( return error; } - spinlock_init(&qinf->qi_pinlock, "xfs_qinf_pin"); + spin_lock_init(&qinf->qi_pinlock); xfs_qm_list_init(&qinf->qi_dqlist, "mpdqlist", 0); qinf->qi_dqreclaims = 0; @@ -1370,7 +1369,6 @@ xfs_qm_qino_alloc( { xfs_trans_t *tp; int error; - unsigned long s; int committed; tp = xfs_trans_alloc(mp, XFS_TRANS_QM_QINOCREATE); @@ -1402,18 +1400,18 @@ xfs_qm_qino_alloc( * sbfields arg may contain fields other than *QUOTINO; * VERSIONNUM for example. */ - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); if (flags & XFS_QMOPT_SBVERSION) { #if defined(DEBUG) && defined(XFS_LOUD_RECOVERY) unsigned oldv = mp->m_sb.sb_versionnum; #endif - ASSERT(!XFS_SB_VERSION_HASQUOTA(&mp->m_sb)); + ASSERT(!xfs_sb_version_hasquota(&mp->m_sb)); ASSERT((sbfields & (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | XFS_SB_GQUOTINO | XFS_SB_QFLAGS)) == (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | XFS_SB_GQUOTINO | XFS_SB_QFLAGS)); - XFS_SB_VERSION_ADDQUOTA(&mp->m_sb); + xfs_sb_version_addquota(&mp->m_sb); mp->m_sb.sb_uquotino = NULLFSINO; mp->m_sb.sb_gquotino = NULLFSINO; @@ -1429,7 +1427,7 @@ xfs_qm_qino_alloc( mp->m_sb.sb_uquotino = (*ip)->i_ino; else mp->m_sb.sb_gquotino = (*ip)->i_ino; - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); xfs_mod_sb(tp, sbfields); if ((error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES))) { @@ -1650,14 +1648,14 @@ xfs_qm_quotacheck_dqadjust( * Adjust the inode count and the block count to reflect this inode's * resource usage. */ - be64_add(&dqp->q_core.d_icount, 1); + be64_add_cpu(&dqp->q_core.d_icount, 1); dqp->q_res_icount++; if (nblks) { - be64_add(&dqp->q_core.d_bcount, nblks); + be64_add_cpu(&dqp->q_core.d_bcount, nblks); dqp->q_res_bcount += nblks; } if (rtblks) { - be64_add(&dqp->q_core.d_rtbcount, rtblks); + be64_add_cpu(&dqp->q_core.d_rtbcount, rtblks); dqp->q_res_rtbcount += rtblks; } @@ -1956,7 +1954,7 @@ xfs_qm_init_quotainos( /* * Get the uquota and gquota inodes */ - if (XFS_SB_VERSION_HASQUOTA(&mp->m_sb)) { + if (xfs_sb_version_hasquota(&mp->m_sb)) { if (XFS_IS_UQUOTA_ON(mp) && mp->m_sb.sb_uquotino != NULLFSINO) { ASSERT(mp->m_sb.sb_uquotino > 0); diff --git a/fs/xfs/quota/xfs_qm.h b/fs/xfs/quota/xfs_qm.h index 23ccaa5fcea..baf537c1c17 100644 --- a/fs/xfs/quota/xfs_qm.h +++ b/fs/xfs/quota/xfs_qm.h @@ -52,8 +52,8 @@ extern kmem_zone_t *qm_dqtrxzone; /* * Dquot hashtable constants/threshold values. */ -#define XFS_QM_HASHSIZE_LOW (NBPP / sizeof(xfs_dqhash_t)) -#define XFS_QM_HASHSIZE_HIGH ((NBPP * 4) / sizeof(xfs_dqhash_t)) +#define XFS_QM_HASHSIZE_LOW (PAGE_SIZE / sizeof(xfs_dqhash_t)) +#define XFS_QM_HASHSIZE_HIGH ((PAGE_SIZE * 4) / sizeof(xfs_dqhash_t)) /* * This defines the unit of allocation of dquots. @@ -106,7 +106,7 @@ typedef struct xfs_qm { typedef struct xfs_quotainfo { xfs_inode_t *qi_uquotaip; /* user quota inode */ xfs_inode_t *qi_gquotaip; /* group quota inode */ - lock_t qi_pinlock; /* dquot pinning mutex */ + spinlock_t qi_pinlock; /* dquot pinning lock */ xfs_dqlist_t qi_dqlist; /* all dquots in filesys */ int qi_dqreclaims; /* a change here indicates a removal in the dqlist */ diff --git a/fs/xfs/quota/xfs_qm_bhv.c b/fs/xfs/quota/xfs_qm_bhv.c index 97bb3293758..f4f6c4c861d 100644 --- a/fs/xfs/quota/xfs_qm_bhv.c +++ b/fs/xfs/quota/xfs_qm_bhv.c @@ -118,7 +118,7 @@ xfs_qm_newmount( *quotaflags = 0; *needquotamount = B_FALSE; - quotaondisk = XFS_SB_VERSION_HASQUOTA(&mp->m_sb) && + quotaondisk = xfs_sb_version_hasquota(&mp->m_sb) && (mp->m_sb.sb_qflags & XFS_ALL_QUOTA_ACCT); if (quotaondisk) { diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c index ad5579d4eac..d2b8be7e75f 100644 --- a/fs/xfs/quota/xfs_qm_syscalls.c +++ b/fs/xfs/quota/xfs_qm_syscalls.c @@ -200,7 +200,6 @@ xfs_qm_scall_quotaoff( boolean_t force) { uint dqtype; - unsigned long s; int error; uint inactivate_flags; xfs_qoff_logitem_t *qoffstart; @@ -237,9 +236,9 @@ xfs_qm_scall_quotaoff( if ((flags & XFS_ALL_QUOTA_ACCT) == 0) { mp->m_qflags &= ~(flags); - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); mp->m_sb.sb_qflags = mp->m_qflags; - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); mutex_unlock(&(XFS_QI_QOFFLOCK(mp))); /* XXX what to do if error ? Revert back to old vals incore ? */ @@ -378,7 +377,7 @@ xfs_qm_scall_trunc_qfiles( if (!capable(CAP_SYS_ADMIN)) return XFS_ERROR(EPERM); error = 0; - if (!XFS_SB_VERSION_HASQUOTA(&mp->m_sb) || flags == 0) { + if (!xfs_sb_version_hasquota(&mp->m_sb) || flags == 0) { qdprintk("qtrunc flags=%x m_qflags=%x\n", flags, mp->m_qflags); return XFS_ERROR(EINVAL); } @@ -415,7 +414,6 @@ xfs_qm_scall_quotaon( uint flags) { int error; - unsigned long s; uint qf; uint accflags; __int64_t sbflags; @@ -468,10 +466,10 @@ xfs_qm_scall_quotaon( * Change sb_qflags on disk but not incore mp->qflags * if this is the root filesystem. */ - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); qf = mp->m_sb.sb_qflags; mp->m_sb.sb_qflags = qf | flags; - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); /* * There's nothing to change if it's the same. @@ -524,7 +522,7 @@ xfs_qm_scall_getqstat( memset(out, 0, sizeof(fs_quota_stat_t)); out->qs_version = FS_QSTAT_VERSION; - if (! XFS_SB_VERSION_HASQUOTA(&mp->m_sb)) { + if (!xfs_sb_version_hasquota(&mp->m_sb)) { out->qs_uquota.qfs_ino = NULLFSINO; out->qs_gquota.qfs_ino = NULLFSINO; return (0); @@ -815,7 +813,6 @@ xfs_qm_log_quotaoff( { xfs_trans_t *tp; int error; - unsigned long s; xfs_qoff_logitem_t *qoffi=NULL; uint oldsbqflag=0; @@ -832,10 +829,10 @@ xfs_qm_log_quotaoff( qoffi = xfs_trans_get_qoff_item(tp, NULL, flags & XFS_ALL_QUOTA_ACCT); xfs_trans_log_quotaoff_item(tp, qoffi); - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); oldsbqflag = mp->m_sb.sb_qflags; mp->m_sb.sb_qflags = (mp->m_qflags & ~(flags)) & XFS_MOUNT_QUOTA_ALL; - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); xfs_mod_sb(tp, XFS_SB_QFLAGS); @@ -854,9 +851,9 @@ error0: * No one else is modifying sb_qflags, so this is OK. * We still hold the quotaofflock. */ - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); mp->m_sb.sb_qflags = oldsbqflag; - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); } *qoffstartp = qoffi; return (error); diff --git a/fs/xfs/quota/xfs_trans_dquot.c b/fs/xfs/quota/xfs_trans_dquot.c index 7de6874bf1b..f441f836ca8 100644 --- a/fs/xfs/quota/xfs_trans_dquot.c +++ b/fs/xfs/quota/xfs_trans_dquot.c @@ -421,13 +421,13 @@ xfs_trans_apply_dquot_deltas( (xfs_qcnt_t) -qtrx->qt_icount_delta); #endif if (totalbdelta) - be64_add(&d->d_bcount, (xfs_qcnt_t)totalbdelta); + be64_add_cpu(&d->d_bcount, (xfs_qcnt_t)totalbdelta); if (qtrx->qt_icount_delta) - be64_add(&d->d_icount, (xfs_qcnt_t)qtrx->qt_icount_delta); + be64_add_cpu(&d->d_icount, (xfs_qcnt_t)qtrx->qt_icount_delta); if (totalrtbdelta) - be64_add(&d->d_rtbcount, (xfs_qcnt_t)totalrtbdelta); + be64_add_cpu(&d->d_rtbcount, (xfs_qcnt_t)totalrtbdelta); /* * Get any default limits in use. diff --git a/fs/xfs/support/debug.c b/fs/xfs/support/debug.c index f45a49ffd3a..c27abef7b84 100644 --- a/fs/xfs/support/debug.c +++ b/fs/xfs/support/debug.c @@ -17,7 +17,6 @@ */ #include <xfs.h> #include "debug.h" -#include "spin.h" static char message[1024]; /* keep it off the stack */ static DEFINE_SPINLOCK(xfs_err_lock); @@ -81,3 +80,9 @@ assfail(char *expr, char *file, int line) printk("Assertion failed: %s, file: %s, line: %d\n", expr, file, line); BUG(); } + +void +xfs_hex_dump(void *p, int length) +{ + print_hex_dump(KERN_ALERT, "", DUMP_PREFIX_OFFSET, 16, 1, p, length, 1); +} diff --git a/fs/xfs/support/ktrace.c b/fs/xfs/support/ktrace.c index 5cf2e86caa7..129067cfcb8 100644 --- a/fs/xfs/support/ktrace.c +++ b/fs/xfs/support/ktrace.c @@ -21,7 +21,7 @@ static kmem_zone_t *ktrace_hdr_zone; static kmem_zone_t *ktrace_ent_zone; static int ktrace_zentries; -void +void __init ktrace_init(int zentries) { ktrace_zentries = zentries; @@ -36,7 +36,7 @@ ktrace_init(int zentries) ASSERT(ktrace_ent_zone); } -void +void __exit ktrace_uninit(void) { kmem_zone_destroy(ktrace_hdr_zone); @@ -90,8 +90,6 @@ ktrace_alloc(int nentries, unsigned int __nocast sleep) return NULL; } - spinlock_init(&(ktp->kt_lock), "kt_lock"); - ktp->kt_entries = ktep; ktp->kt_nentries = nentries; ktp->kt_index = 0; @@ -114,8 +112,6 @@ ktrace_free(ktrace_t *ktp) if (ktp == (ktrace_t *)NULL) return; - spinlock_destroy(&ktp->kt_lock); - /* * Special treatment for the Vnode trace buffer. */ diff --git a/fs/xfs/support/ktrace.h b/fs/xfs/support/ktrace.h index 0d73216287c..56e72b40a85 100644 --- a/fs/xfs/support/ktrace.h +++ b/fs/xfs/support/ktrace.h @@ -18,8 +18,6 @@ #ifndef __XFS_SUPPORT_KTRACE_H__ #define __XFS_SUPPORT_KTRACE_H__ -#include <spin.h> - /* * Trace buffer entry structure. */ @@ -31,7 +29,6 @@ typedef struct ktrace_entry { * Trace buffer header structure. */ typedef struct ktrace { - lock_t kt_lock; /* mutex to guard counters */ int kt_nentries; /* number of entries in trace buf */ int kt_index; /* current index in entries */ int kt_rollover; diff --git a/fs/xfs/support/uuid.c b/fs/xfs/support/uuid.c index e157015c70f..493a6ecf859 100644 --- a/fs/xfs/support/uuid.c +++ b/fs/xfs/support/uuid.c @@ -133,7 +133,7 @@ uuid_table_remove(uuid_t *uuid) mutex_unlock(&uuid_monitor); } -void +void __init uuid_init(void) { mutex_init(&uuid_monitor); diff --git a/fs/xfs/xfs.h b/fs/xfs/xfs.h index b5a7d92c684..540e4c98982 100644 --- a/fs/xfs/xfs.h +++ b/fs/xfs/xfs.h @@ -37,7 +37,7 @@ #define XFS_LOG_TRACE 1 #define XFS_RW_TRACE 1 #define XFS_BUF_TRACE 1 -#define XFS_VNODE_TRACE 1 +#define XFS_INODE_TRACE 1 #define XFS_FILESTREAMS_TRACE 1 #endif diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c index 5bfb66f33ca..7272fe39a92 100644 --- a/fs/xfs/xfs_acl.c +++ b/fs/xfs/xfs_acl.c @@ -392,32 +392,6 @@ xfs_acl_allow_set( } /* - * The access control process to determine the access permission: - * if uid == file owner id, use the file owner bits. - * if gid == file owner group id, use the file group bits. - * scan ACL for a matching user or group, and use matched entry - * permission. Use total permissions of all matching group entries, - * until all acl entries are exhausted. The final permission produced - * by matching acl entry or entries needs to be & with group permission. - * if not owner, owning group, or matching entry in ACL, use file - * other bits. - */ -STATIC int -xfs_acl_capability_check( - mode_t mode, - cred_t *cr) -{ - if ((mode & ACL_READ) && !capable_cred(cr, CAP_DAC_READ_SEARCH)) - return EACCES; - if ((mode & ACL_WRITE) && !capable_cred(cr, CAP_DAC_OVERRIDE)) - return EACCES; - if ((mode & ACL_EXECUTE) && !capable_cred(cr, CAP_DAC_OVERRIDE)) - return EACCES; - - return 0; -} - -/* * Note: cr is only used here for the capability check if the ACL test fails. * It is not used to find out the credentials uid or groups etc, as was * done in IRIX. It is assumed that the uid and groups for the current @@ -438,7 +412,6 @@ xfs_acl_access( matched.ae_tag = 0; /* Invalid type */ matched.ae_perm = 0; - md >>= 6; /* Normalize the bits for comparison */ for (i = 0; i < fap->acl_cnt; i++) { /* @@ -520,7 +493,8 @@ xfs_acl_access( break; } - return xfs_acl_capability_check(md, cr); + /* EACCES tells generic_permission to check for capability overrides */ + return EACCES; } /* diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h index 34b7d339129..332a772461c 100644 --- a/fs/xfs/xfs_acl.h +++ b/fs/xfs/xfs_acl.h @@ -75,7 +75,6 @@ extern int xfs_acl_vremove(bhv_vnode_t *, int); #define _ACL_GET_DEFAULT(pv,pd) (xfs_acl_vtoacl(pv,NULL,pd) == 0) #define _ACL_ACCESS_EXISTS xfs_acl_vhasacl_access #define _ACL_DEFAULT_EXISTS xfs_acl_vhasacl_default -#define _ACL_XFS_IACCESS(i,m,c) (XFS_IFORK_Q(i) ? xfs_acl_iaccess(i,m,c) : -1) #define _ACL_ALLOC(a) ((a) = kmem_zone_alloc(xfs_acl_zone, KM_SLEEP)) #define _ACL_FREE(a) ((a)? kmem_zone_free(xfs_acl_zone, (a)):(void)0) @@ -95,7 +94,6 @@ extern int xfs_acl_vremove(bhv_vnode_t *, int); #define _ACL_GET_DEFAULT(pv,pd) (0) #define _ACL_ACCESS_EXISTS (NULL) #define _ACL_DEFAULT_EXISTS (NULL) -#define _ACL_XFS_IACCESS(i,m,c) (-1) #endif #endif /* __XFS_ACL_H__ */ diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/xfs_ag.h index 9381b0360c4..61b292a9fb4 100644 --- a/fs/xfs/xfs_ag.h +++ b/fs/xfs/xfs_ag.h @@ -193,7 +193,7 @@ typedef struct xfs_perag xfs_agino_t pagi_count; /* number of allocated inodes */ int pagb_count; /* pagb slots in use */ #ifdef __KERNEL__ - lock_t pagb_lock; /* lock for pagb_list */ + spinlock_t pagb_lock; /* lock for pagb_list */ #endif xfs_perag_busy_t *pagb_list; /* unstable blocks */ atomic_t pagf_fstrms; /* # of filestreams active in this AG */ diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c index 012a649a19c..bdbfbbee495 100644 --- a/fs/xfs/xfs_alloc.c +++ b/fs/xfs/xfs_alloc.c @@ -592,7 +592,7 @@ xfs_alloc_ag_vextent( if (!(args->wasfromfl)) { agf = XFS_BUF_TO_AGF(args->agbp); - be32_add(&agf->agf_freeblks, -(args->len)); + be32_add_cpu(&agf->agf_freeblks, -(args->len)); xfs_trans_agblocks_delta(args->tp, -((long)(args->len))); args->pag->pagf_freeblks -= args->len; @@ -1720,7 +1720,7 @@ xfs_free_ag_extent( agf = XFS_BUF_TO_AGF(agbp); pag = &mp->m_perag[agno]; - be32_add(&agf->agf_freeblks, len); + be32_add_cpu(&agf->agf_freeblks, len); xfs_trans_agblocks_delta(tp, len); pag->pagf_freeblks += len; XFS_WANT_CORRUPTED_GOTO( @@ -2008,18 +2008,18 @@ xfs_alloc_get_freelist( * Get the block number and update the data structures. */ bno = be32_to_cpu(agfl->agfl_bno[be32_to_cpu(agf->agf_flfirst)]); - be32_add(&agf->agf_flfirst, 1); + be32_add_cpu(&agf->agf_flfirst, 1); xfs_trans_brelse(tp, agflbp); if (be32_to_cpu(agf->agf_flfirst) == XFS_AGFL_SIZE(mp)) agf->agf_flfirst = 0; pag = &mp->m_perag[be32_to_cpu(agf->agf_seqno)]; - be32_add(&agf->agf_flcount, -1); + be32_add_cpu(&agf->agf_flcount, -1); xfs_trans_agflist_delta(tp, -1); pag->pagf_flcount--; logflags = XFS_AGF_FLFIRST | XFS_AGF_FLCOUNT; if (btreeblk) { - be32_add(&agf->agf_btreeblks, 1); + be32_add_cpu(&agf->agf_btreeblks, 1); pag->pagf_btreeblks++; logflags |= XFS_AGF_BTREEBLKS; } @@ -2117,17 +2117,17 @@ xfs_alloc_put_freelist( be32_to_cpu(agf->agf_seqno), &agflbp))) return error; agfl = XFS_BUF_TO_AGFL(agflbp); - be32_add(&agf->agf_fllast, 1); + be32_add_cpu(&agf->agf_fllast, 1); if (be32_to_cpu(agf->agf_fllast) == XFS_AGFL_SIZE(mp)) agf->agf_fllast = 0; pag = &mp->m_perag[be32_to_cpu(agf->agf_seqno)]; - be32_add(&agf->agf_flcount, 1); + be32_add_cpu(&agf->agf_flcount, 1); xfs_trans_agflist_delta(tp, 1); pag->pagf_flcount++; logflags = XFS_AGF_FLLAST | XFS_AGF_FLCOUNT; if (btreeblk) { - be32_add(&agf->agf_btreeblks, -1); + be32_add_cpu(&agf->agf_btreeblks, -1); pag->pagf_btreeblks--; logflags |= XFS_AGF_BTREEBLKS; } @@ -2206,7 +2206,7 @@ xfs_alloc_read_agf( be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNOi]); pag->pagf_levels[XFS_BTNUM_CNTi] = be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNTi]); - spinlock_init(&pag->pagb_lock, "xfspagb"); + spin_lock_init(&pag->pagb_lock); pag->pagb_list = kmem_zalloc(XFS_PAGB_NUM_SLOTS * sizeof(xfs_perag_busy_t), KM_SLEEP); pag->pagf_init = 1; @@ -2500,10 +2500,9 @@ xfs_alloc_mark_busy(xfs_trans_t *tp, xfs_mount_t *mp; xfs_perag_busy_t *bsy; int n; - SPLDECL(s); mp = tp->t_mountp; - s = mutex_spinlock(&mp->m_perag[agno].pagb_lock); + spin_lock(&mp->m_perag[agno].pagb_lock); /* search pagb_list for an open slot */ for (bsy = mp->m_perag[agno].pagb_list, n = 0; @@ -2533,7 +2532,7 @@ xfs_alloc_mark_busy(xfs_trans_t *tp, xfs_trans_set_sync(tp); } - mutex_spinunlock(&mp->m_perag[agno].pagb_lock, s); + spin_unlock(&mp->m_perag[agno].pagb_lock); } void @@ -2543,11 +2542,10 @@ xfs_alloc_clear_busy(xfs_trans_t *tp, { xfs_mount_t *mp; xfs_perag_busy_t *list; - SPLDECL(s); mp = tp->t_mountp; - s = mutex_spinlock(&mp->m_perag[agno].pagb_lock); + spin_lock(&mp->m_perag[agno].pagb_lock); list = mp->m_perag[agno].pagb_list; ASSERT(idx < XFS_PAGB_NUM_SLOTS); @@ -2559,7 +2557,7 @@ xfs_alloc_clear_busy(xfs_trans_t *tp, TRACE_UNBUSY("xfs_alloc_clear_busy", "missing", agno, idx, tp); } - mutex_spinunlock(&mp->m_perag[agno].pagb_lock, s); + spin_unlock(&mp->m_perag[agno].pagb_lock); } @@ -2578,11 +2576,10 @@ xfs_alloc_search_busy(xfs_trans_t *tp, xfs_agblock_t uend, bend; xfs_lsn_t lsn; int cnt; - SPLDECL(s); mp = tp->t_mountp; - s = mutex_spinlock(&mp->m_perag[agno].pagb_lock); + spin_lock(&mp->m_perag[agno].pagb_lock); cnt = mp->m_perag[agno].pagb_count; uend = bno + len - 1; @@ -2615,12 +2612,12 @@ xfs_alloc_search_busy(xfs_trans_t *tp, if (cnt) { TRACE_BUSYSEARCH("xfs_alloc_search_busy", "found", agno, bno, len, n, tp); lsn = bsy->busy_tp->t_commit_lsn; - mutex_spinunlock(&mp->m_perag[agno].pagb_lock, s); + spin_unlock(&mp->m_perag[agno].pagb_lock); xfs_log_force(mp, lsn, XFS_LOG_FORCE|XFS_LOG_SYNC); } else { TRACE_BUSYSEARCH("xfs_alloc_search_busy", "not-found", agno, bno, len, n, tp); n = -1; - mutex_spinunlock(&mp->m_perag[agno].pagb_lock, s); + spin_unlock(&mp->m_perag[agno].pagb_lock); } return n; diff --git a/fs/xfs/xfs_alloc_btree.c b/fs/xfs/xfs_alloc_btree.c index 1603ce59585..3ce2645508a 100644 --- a/fs/xfs/xfs_alloc_btree.c +++ b/fs/xfs/xfs_alloc_btree.c @@ -221,7 +221,7 @@ xfs_alloc_delrec( */ bno = be32_to_cpu(agf->agf_roots[cur->bc_btnum]); agf->agf_roots[cur->bc_btnum] = *lpp; - be32_add(&agf->agf_levels[cur->bc_btnum], -1); + be32_add_cpu(&agf->agf_levels[cur->bc_btnum], -1); mp->m_perag[be32_to_cpu(agf->agf_seqno)].pagf_levels[cur->bc_btnum]--; /* * Put this buffer/block on the ag's freelist. @@ -1256,9 +1256,9 @@ xfs_alloc_lshift( /* * Bump and log left's numrecs, decrement and log right's numrecs. */ - be16_add(&left->bb_numrecs, 1); + be16_add_cpu(&left->bb_numrecs, 1); xfs_alloc_log_block(cur->bc_tp, lbp, XFS_BB_NUMRECS); - be16_add(&right->bb_numrecs, -1); + be16_add_cpu(&right->bb_numrecs, -1); xfs_alloc_log_block(cur->bc_tp, rbp, XFS_BB_NUMRECS); /* * Slide the contents of right down one entry. @@ -1346,7 +1346,7 @@ xfs_alloc_newroot( agf = XFS_BUF_TO_AGF(cur->bc_private.a.agbp); agf->agf_roots[cur->bc_btnum] = cpu_to_be32(nbno); - be32_add(&agf->agf_levels[cur->bc_btnum], 1); + be32_add_cpu(&agf->agf_levels[cur->bc_btnum], 1); seqno = be32_to_cpu(agf->agf_seqno); mp->m_perag[seqno].pagf_levels[cur->bc_btnum]++; xfs_alloc_log_agf(cur->bc_tp, cur->bc_private.a.agbp, @@ -1558,9 +1558,9 @@ xfs_alloc_rshift( /* * Decrement and log left's numrecs, bump and log right's numrecs. */ - be16_add(&left->bb_numrecs, -1); + be16_add_cpu(&left->bb_numrecs, -1); xfs_alloc_log_block(cur->bc_tp, lbp, XFS_BB_NUMRECS); - be16_add(&right->bb_numrecs, 1); + be16_add_cpu(&right->bb_numrecs, 1); xfs_alloc_log_block(cur->bc_tp, rbp, XFS_BB_NUMRECS); /* * Using a temporary cursor, update the parent key values of the @@ -1643,7 +1643,7 @@ xfs_alloc_split( */ if ((be16_to_cpu(left->bb_numrecs) & 1) && cur->bc_ptrs[level] <= be16_to_cpu(right->bb_numrecs) + 1) - be16_add(&right->bb_numrecs, 1); + be16_add_cpu(&right->bb_numrecs, 1); i = be16_to_cpu(left->bb_numrecs) - be16_to_cpu(right->bb_numrecs) + 1; /* * For non-leaf blocks, copy keys and addresses over to the new block. @@ -1689,7 +1689,7 @@ xfs_alloc_split( * Adjust numrecs, sibling pointers. */ lbno = XFS_DADDR_TO_AGBNO(cur->bc_mp, XFS_BUF_ADDR(lbp)); - be16_add(&left->bb_numrecs, -(be16_to_cpu(right->bb_numrecs))); + be16_add_cpu(&left->bb_numrecs, -(be16_to_cpu(right->bb_numrecs))); right->bb_rightsib = left->bb_rightsib; left->bb_rightsib = cpu_to_be32(rbno); right->bb_leftsib = cpu_to_be32(lbno); diff --git a/fs/xfs/xfs_arch.h b/fs/xfs/xfs_arch.h index c4836890b72..f9472a2076d 100644 --- a/fs/xfs/xfs_arch.h +++ b/fs/xfs/xfs_arch.h @@ -170,21 +170,6 @@ } \ } -static inline void be16_add(__be16 *a, __s16 b) -{ - *a = cpu_to_be16(be16_to_cpu(*a) + b); -} - -static inline void be32_add(__be32 *a, __s32 b) -{ - *a = cpu_to_be32(be32_to_cpu(*a) + b); -} - -static inline void be64_add(__be64 *a, __s64 b) -{ - *a = cpu_to_be64(be64_to_cpu(*a) + b); -} - /* * In directories inode numbers are stored as unaligned arrays of unsigned * 8bit integers on disk. diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c index 93fa64dd1be..e58f321fdae 100644 --- a/fs/xfs/xfs_attr.c +++ b/fs/xfs/xfs_attr.c @@ -929,7 +929,7 @@ xfs_attr_shortform_addname(xfs_da_args_t *args) * This leaf block cannot have a "remote" value, we only call this routine * if bmap_one_block() says there is only one block (ie: no remote blks). */ -int +STATIC int xfs_attr_leaf_addname(xfs_da_args_t *args) { xfs_inode_t *dp; diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index 81f45dae1c5..96ba6aa4ed8 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -226,17 +226,15 @@ xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes) STATIC void xfs_sbversion_add_attr2(xfs_mount_t *mp, xfs_trans_t *tp) { - unsigned long s; - if ((mp->m_flags & XFS_MOUNT_ATTR2) && - !(XFS_SB_VERSION_HASATTR2(&mp->m_sb))) { - s = XFS_SB_LOCK(mp); - if (!XFS_SB_VERSION_HASATTR2(&mp->m_sb)) { - XFS_SB_VERSION_ADDATTR2(&mp->m_sb); - XFS_SB_UNLOCK(mp, s); + !(xfs_sb_version_hasattr2(&mp->m_sb))) { + spin_lock(&mp->m_sb_lock); + if (!xfs_sb_version_hasattr2(&mp->m_sb)) { + xfs_sb_version_addattr2(&mp->m_sb); + spin_unlock(&mp->m_sb_lock); xfs_mod_sb(tp, XFS_SB_VERSIONNUM | XFS_SB_FEATURES2); } else - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); } } @@ -319,7 +317,7 @@ xfs_attr_shortform_add(xfs_da_args_t *args, int forkoff) memcpy(sfe->nameval, args->name, args->namelen); memcpy(&sfe->nameval[args->namelen], args->value, args->valuelen); sf->hdr.count++; - be16_add(&sf->hdr.totsize, size); + be16_add_cpu(&sf->hdr.totsize, size); xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_ADATA); xfs_sbversion_add_attr2(mp, args->trans); @@ -365,7 +363,7 @@ xfs_attr_shortform_remove(xfs_da_args_t *args) if (end != totsize) memmove(&((char *)sf)[base], &((char *)sf)[end], totsize - end); sf->hdr.count--; - be16_add(&sf->hdr.totsize, -size); + be16_add_cpu(&sf->hdr.totsize, -size); /* * Fix up the start offset of the attribute fork @@ -1135,7 +1133,7 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex) xfs_da_log_buf(args->trans, bp, XFS_DA_LOGRANGE(leaf, entry, tmp + sizeof(*entry))); } - be16_add(&hdr->count, 1); + be16_add_cpu(&hdr->count, 1); /* * Allocate space for the new string (at the end of the run). @@ -1149,7 +1147,7 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex) mp->m_sb.sb_blocksize, NULL)); ASSERT(be16_to_cpu(map->size) < XFS_LBSIZE(mp)); ASSERT((be16_to_cpu(map->size) & 0x3) == 0); - be16_add(&map->size, + be16_add_cpu(&map->size, -xfs_attr_leaf_newentsize(args->namelen, args->valuelen, mp->m_sb.sb_blocksize, &tmp)); entry->nameidx = cpu_to_be16(be16_to_cpu(map->base) + @@ -1216,12 +1214,12 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex) map = &hdr->freemap[0]; for (i = 0; i < XFS_ATTR_LEAF_MAPSIZE; map++, i++) { if (be16_to_cpu(map->base) == tmp) { - be16_add(&map->base, sizeof(xfs_attr_leaf_entry_t)); - be16_add(&map->size, + be16_add_cpu(&map->base, sizeof(xfs_attr_leaf_entry_t)); + be16_add_cpu(&map->size, -((int)sizeof(xfs_attr_leaf_entry_t))); } } - be16_add(&hdr->usedbytes, xfs_attr_leaf_entsize(leaf, args->index)); + be16_add_cpu(&hdr->usedbytes, xfs_attr_leaf_entsize(leaf, args->index)); xfs_da_log_buf(args->trans, bp, XFS_DA_LOGRANGE(leaf, hdr, sizeof(*hdr))); return(0); @@ -1729,9 +1727,9 @@ xfs_attr_leaf_remove(xfs_dabuf_t *bp, xfs_da_args_t *args) ASSERT(be16_to_cpu(map->base) < XFS_LBSIZE(mp)); ASSERT(be16_to_cpu(map->size) < XFS_LBSIZE(mp)); if (be16_to_cpu(map->base) == tablesize) { - be16_add(&map->base, + be16_add_cpu(&map->base, -((int)sizeof(xfs_attr_leaf_entry_t))); - be16_add(&map->size, sizeof(xfs_attr_leaf_entry_t)); + be16_add_cpu(&map->size, sizeof(xfs_attr_leaf_entry_t)); } if ((be16_to_cpu(map->base) + be16_to_cpu(map->size)) @@ -1753,19 +1751,19 @@ xfs_attr_leaf_remove(xfs_dabuf_t *bp, xfs_da_args_t *args) if ((before >= 0) || (after >= 0)) { if ((before >= 0) && (after >= 0)) { map = &hdr->freemap[before]; - be16_add(&map->size, entsize); - be16_add(&map->size, + be16_add_cpu(&map->size, entsize); + be16_add_cpu(&map->size, be16_to_cpu(hdr->freemap[after].size)); hdr->freemap[after].base = 0; hdr->freemap[after].size = 0; } else if (before >= 0) { map = &hdr->freemap[before]; - be16_add(&map->size, entsize); + be16_add_cpu(&map->size, entsize); } else { map = &hdr->freemap[after]; /* both on-disk, don't endian flip twice */ map->base = entry->nameidx; - be16_add(&map->size, entsize); + be16_add_cpu(&map->size, entsize); } } else { /* @@ -1790,7 +1788,7 @@ xfs_attr_leaf_remove(xfs_dabuf_t *bp, xfs_da_args_t *args) * Compress the remaining entries and zero out the removed stuff. */ memset(XFS_ATTR_LEAF_NAME(leaf, args->index), 0, entsize); - be16_add(&hdr->usedbytes, -entsize); + be16_add_cpu(&hdr->usedbytes, -entsize); xfs_da_log_buf(args->trans, bp, XFS_DA_LOGRANGE(leaf, XFS_ATTR_LEAF_NAME(leaf, args->index), entsize)); @@ -1798,7 +1796,7 @@ xfs_attr_leaf_remove(xfs_dabuf_t *bp, xfs_da_args_t *args) tmp = (be16_to_cpu(hdr->count) - args->index) * sizeof(xfs_attr_leaf_entry_t); memmove((char *)entry, (char *)(entry+1), tmp); - be16_add(&hdr->count, -1); + be16_add_cpu(&hdr->count, -1); xfs_da_log_buf(args->trans, bp, XFS_DA_LOGRANGE(leaf, entry, tmp + sizeof(*entry))); entry = &leaf->entries[be16_to_cpu(hdr->count)]; @@ -2184,15 +2182,15 @@ xfs_attr_leaf_moveents(xfs_attr_leafblock_t *leaf_s, int start_s, */ if (entry_s->flags & XFS_ATTR_INCOMPLETE) { /* skip partials? */ memset(XFS_ATTR_LEAF_NAME(leaf_s, start_s + i), 0, tmp); - be16_add(&hdr_s->usedbytes, -tmp); - be16_add(&hdr_s->count, -1); + be16_add_cpu(&hdr_s->usedbytes, -tmp); + be16_add_cpu(&hdr_s->count, -1); entry_d--; /* to compensate for ++ in loop hdr */ desti--; if ((start_s + i) < offset) result++; /* insertion index adjustment */ } else { #endif /* GROT */ - be16_add(&hdr_d->firstused, -tmp); + be16_add_cpu(&hdr_d->firstused, -tmp); /* both on-disk, don't endian flip twice */ entry_d->hashval = entry_s->hashval; /* both on-disk, don't endian flip twice */ @@ -2205,10 +2203,10 @@ xfs_attr_leaf_moveents(xfs_attr_leafblock_t *leaf_s, int start_s, ASSERT(be16_to_cpu(entry_s->nameidx) + tmp <= XFS_LBSIZE(mp)); memset(XFS_ATTR_LEAF_NAME(leaf_s, start_s + i), 0, tmp); - be16_add(&hdr_s->usedbytes, -tmp); - be16_add(&hdr_d->usedbytes, tmp); - be16_add(&hdr_s->count, -1); - be16_add(&hdr_d->count, 1); + be16_add_cpu(&hdr_s->usedbytes, -tmp); + be16_add_cpu(&hdr_d->usedbytes, tmp); + be16_add_cpu(&hdr_s->count, -1); + be16_add_cpu(&hdr_d->count, 1); tmp = be16_to_cpu(hdr_d->count) * sizeof(xfs_attr_leaf_entry_t) + sizeof(xfs_attr_leaf_hdr_t); @@ -2249,7 +2247,7 @@ xfs_attr_leaf_moveents(xfs_attr_leafblock_t *leaf_s, int start_s, * Fill in the freemap information */ hdr_d->freemap[0].base = cpu_to_be16(sizeof(xfs_attr_leaf_hdr_t)); - be16_add(&hdr_d->freemap[0].base, be16_to_cpu(hdr_d->count) * + be16_add_cpu(&hdr_d->freemap[0].base, be16_to_cpu(hdr_d->count) * sizeof(xfs_attr_leaf_entry_t)); hdr_d->freemap[0].size = cpu_to_be16(be16_to_cpu(hdr_d->firstused) - be16_to_cpu(hdr_d->freemap[0].base)); diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 2e9b34b7344..2def273855a 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -2830,11 +2830,11 @@ xfs_bmap_btalloc( args.prod = align; if ((args.mod = (xfs_extlen_t)do_mod(ap->off, args.prod))) args.mod = (xfs_extlen_t)(args.prod - args.mod); - } else if (mp->m_sb.sb_blocksize >= NBPP) { + } else if (mp->m_sb.sb_blocksize >= PAGE_CACHE_SIZE) { args.prod = 1; args.mod = 0; } else { - args.prod = NBPP >> mp->m_sb.sb_blocklog; + args.prod = PAGE_CACHE_SIZE >> mp->m_sb.sb_blocklog; if ((args.mod = (xfs_extlen_t)(do_mod(ap->off, args.prod)))) args.mod = (xfs_extlen_t)(args.prod - args.mod); } @@ -2969,7 +2969,7 @@ STATIC int xfs_bmap_alloc( xfs_bmalloca_t *ap) /* bmap alloc argument struct */ { - if ((ap->ip->i_d.di_flags & XFS_DIFLAG_REALTIME) && ap->userdata) + if (XFS_IS_REALTIME_INODE(ap->ip) && ap->userdata) return xfs_bmap_rtalloc(ap); return xfs_bmap_btalloc(ap); } @@ -3096,8 +3096,7 @@ xfs_bmap_del_extent( /* * Realtime allocation. Free it and record di_nblocks update. */ - if (whichfork == XFS_DATA_FORK && - (ip->i_d.di_flags & XFS_DIFLAG_REALTIME)) { + if (whichfork == XFS_DATA_FORK && XFS_IS_REALTIME_INODE(ip)) { xfs_fsblock_t bno; xfs_filblks_t len; @@ -3956,7 +3955,6 @@ xfs_bmap_add_attrfork( xfs_bmap_free_t flist; /* freed extent records */ xfs_mount_t *mp; /* mount structure */ xfs_trans_t *tp; /* transaction pointer */ - unsigned long s; /* spinlock spl value */ int blks; /* space reservation */ int version = 1; /* superblock attr version */ int committed; /* xaction was committed */ @@ -4049,24 +4047,24 @@ xfs_bmap_add_attrfork( xfs_trans_log_inode(tp, ip, logflags); if (error) goto error2; - if (!XFS_SB_VERSION_HASATTR(&mp->m_sb) || - (!XFS_SB_VERSION_HASATTR2(&mp->m_sb) && version == 2)) { + if (!xfs_sb_version_hasattr(&mp->m_sb) || + (!xfs_sb_version_hasattr2(&mp->m_sb) && version == 2)) { __int64_t sbfields = 0; - s = XFS_SB_LOCK(mp); - if (!XFS_SB_VERSION_HASATTR(&mp->m_sb)) { - XFS_SB_VERSION_ADDATTR(&mp->m_sb); + spin_lock(&mp->m_sb_lock); + if (!xfs_sb_version_hasattr(&mp->m_sb)) { + xfs_sb_version_addattr(&mp->m_sb); sbfields |= XFS_SB_VERSIONNUM; } - if (!XFS_SB_VERSION_HASATTR2(&mp->m_sb) && version == 2) { - XFS_SB_VERSION_ADDATTR2(&mp->m_sb); + if (!xfs_sb_version_hasattr2(&mp->m_sb) && version == 2) { + xfs_sb_version_addattr2(&mp->m_sb); sbfields |= (XFS_SB_VERSIONNUM | XFS_SB_FEATURES2); } if (sbfields) { - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); xfs_mod_sb(tp, sbfields); } else - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); } if ((error = xfs_bmap_finish(&tp, &flist, &committed))) goto error2; @@ -5045,7 +5043,7 @@ xfs_bmapi( * A wasdelay extent has been initialized, so * shouldn't be flagged as unwritten. */ - if (wr && XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb)) { + if (wr && xfs_sb_version_hasextflgbit(&mp->m_sb)) { if (!wasdelay && (flags & XFS_BMAPI_PREALLOC)) got.br_state = XFS_EXT_UNWRITTEN; } @@ -5485,7 +5483,7 @@ xfs_bunmapi( * get rid of part of a realtime extent. */ if (del.br_state == XFS_EXT_UNWRITTEN || - !XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb)) { + !xfs_sb_version_hasextflgbit(&mp->m_sb)) { /* * This piece is unwritten, or we're not * using unwritten extents. Skip over it. @@ -5537,7 +5535,7 @@ xfs_bunmapi( } else if ((del.br_startoff == start && (del.br_state == XFS_EXT_UNWRITTEN || xfs_trans_get_block_res(tp) == 0)) || - !XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb)) { + !xfs_sb_version_hasextflgbit(&mp->m_sb)) { /* * Can't make it unwritten. There isn't * a full extent here so just skip it. @@ -6394,7 +6392,7 @@ xfs_bmap_count_blocks( * Recursively walks each level of a btree * to count total fsblocks is use. */ -int /* error */ +STATIC int /* error */ xfs_bmap_count_tree( xfs_mount_t *mp, /* file system mount point */ xfs_trans_t *tp, /* transaction pointer */ @@ -6470,7 +6468,7 @@ xfs_bmap_count_tree( /* * Count leaf blocks given a range of extent records. */ -int +STATIC int xfs_bmap_count_leaves( xfs_ifork_t *ifp, xfs_extnum_t idx, @@ -6490,7 +6488,7 @@ xfs_bmap_count_leaves( * Count leaf blocks given a range of extent records originally * in btree format. */ -int +STATIC int xfs_bmap_disk_count_leaves( xfs_extnum_t idx, xfs_bmbt_block_t *block, diff --git a/fs/xfs/xfs_bmap.h b/fs/xfs/xfs_bmap.h index 68267d75ff1..87224b7d798 100644 --- a/fs/xfs/xfs_bmap.h +++ b/fs/xfs/xfs_bmap.h @@ -25,6 +25,8 @@ struct xfs_inode; struct xfs_mount; struct xfs_trans; +extern kmem_zone_t *xfs_bmap_free_item_zone; + /* * DELTA: describe a change to the in-core extent list. * diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/xfs_bmap_btree.c index 32b49ec00fb..bd18987326a 100644 --- a/fs/xfs/xfs_bmap_btree.c +++ b/fs/xfs/xfs_bmap_btree.c @@ -631,7 +631,7 @@ xfs_bmbt_delrec( memcpy(lrp, rrp, numrrecs * sizeof(*lrp)); xfs_bmbt_log_recs(cur, lbp, numlrecs + 1, numlrecs + numrrecs); } - be16_add(&left->bb_numrecs, numrrecs); + be16_add_cpu(&left->bb_numrecs, numrrecs); left->bb_rightsib = right->bb_rightsib; xfs_bmbt_log_block(cur, lbp, XFS_BB_RIGHTSIB | XFS_BB_NUMRECS); if (be64_to_cpu(left->bb_rightsib) != NULLDFSBNO) { @@ -924,7 +924,7 @@ xfs_bmbt_killroot( xfs_iroot_realloc(ip, i, cur->bc_private.b.whichfork); block = ifp->if_broot; } - be16_add(&block->bb_numrecs, i); + be16_add_cpu(&block->bb_numrecs, i); ASSERT(block->bb_numrecs == cblock->bb_numrecs); kp = XFS_BMAP_KEY_IADDR(block, 1, cur); ckp = XFS_BMAP_KEY_IADDR(cblock, 1, cur); @@ -947,7 +947,7 @@ xfs_bmbt_killroot( XFS_TRANS_DQ_BCOUNT, -1L); xfs_trans_binval(cur->bc_tp, cbp); cur->bc_bufs[level - 1] = NULL; - be16_add(&block->bb_level, -1); + be16_add_cpu(&block->bb_level, -1); xfs_trans_log_inode(cur->bc_tp, ip, XFS_ILOG_CORE | XFS_ILOG_FBROOT(cur->bc_private.b.whichfork)); cur->bc_nlevels--; @@ -1401,9 +1401,9 @@ xfs_bmbt_rshift( key.br_startoff = cpu_to_be64(xfs_bmbt_disk_get_startoff(rrp)); rkp = &key; } - be16_add(&left->bb_numrecs, -1); + be16_add_cpu(&left->bb_numrecs, -1); xfs_bmbt_log_block(cur, lbp, XFS_BB_NUMRECS); - be16_add(&right->bb_numrecs, 1); + be16_add_cpu(&right->bb_numrecs, 1); #ifdef DEBUG if (level > 0) xfs_btree_check_key(XFS_BTNUM_BMAP, rkp, rkp + 1); @@ -1535,7 +1535,7 @@ xfs_bmbt_split( right->bb_numrecs = cpu_to_be16(be16_to_cpu(left->bb_numrecs) / 2); if ((be16_to_cpu(left->bb_numrecs) & 1) && cur->bc_ptrs[level] <= be16_to_cpu(right->bb_numrecs) + 1) - be16_add(&right->bb_numrecs, 1); + be16_add_cpu(&right->bb_numrecs, 1); i = be16_to_cpu(left->bb_numrecs) - be16_to_cpu(right->bb_numrecs) + 1; if (level > 0) { lkp = XFS_BMAP_KEY_IADDR(left, i, cur); @@ -1562,7 +1562,7 @@ xfs_bmbt_split( xfs_bmbt_log_recs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs)); *startoff = xfs_bmbt_disk_get_startoff(rrp); } - be16_add(&left->bb_numrecs, -(be16_to_cpu(right->bb_numrecs))); + be16_add_cpu(&left->bb_numrecs, -(be16_to_cpu(right->bb_numrecs))); right->bb_rightsib = left->bb_rightsib; left->bb_rightsib = cpu_to_be64(args.fsbno); right->bb_leftsib = cpu_to_be64(lbno); @@ -2062,8 +2062,7 @@ xfs_bmbt_insert( pcur->bc_private.b.allocated; pcur->bc_private.b.allocated = 0; ASSERT((cur->bc_private.b.firstblock != NULLFSBLOCK) || - (cur->bc_private.b.ip->i_d.di_flags & - XFS_DIFLAG_REALTIME)); + XFS_IS_REALTIME_INODE(cur->bc_private.b.ip)); cur->bc_private.b.firstblock = pcur->bc_private.b.firstblock; ASSERT(cur->bc_private.b.flist == @@ -2241,7 +2240,7 @@ xfs_bmbt_newroot( bp = xfs_btree_get_bufl(args.mp, cur->bc_tp, args.fsbno, 0); cblock = XFS_BUF_TO_BMBT_BLOCK(bp); *cblock = *block; - be16_add(&block->bb_level, 1); + be16_add_cpu(&block->bb_level, 1); block->bb_numrecs = cpu_to_be16(1); cur->bc_nlevels++; cur->bc_ptrs[level + 1] = 1; diff --git a/fs/xfs/xfs_bmap_btree.h b/fs/xfs/xfs_bmap_btree.h index 2d950e97591..cd0d4b4bb81 100644 --- a/fs/xfs/xfs_bmap_btree.h +++ b/fs/xfs/xfs_bmap_btree.h @@ -120,7 +120,7 @@ typedef enum { * Extent state and extent format macros. */ #define XFS_EXTFMT_INODE(x) \ - (XFS_SB_VERSION_HASEXTFLGBIT(&((x)->i_mount->m_sb)) ? \ + (xfs_sb_version_hasextflgbit(&((x)->i_mount->m_sb)) ? \ XFS_EXTFMT_HASSTATE : XFS_EXTFMT_NOSTATE) #define ISUNWRITTEN(x) ((x)->br_state == XFS_EXT_UNWRITTEN) diff --git a/fs/xfs/xfs_btree.h b/fs/xfs/xfs_btree.h index 6e40a0a198f..7440b78f9ce 100644 --- a/fs/xfs/xfs_btree.h +++ b/fs/xfs/xfs_btree.h @@ -24,6 +24,8 @@ struct xfs_inode; struct xfs_mount; struct xfs_trans; +extern kmem_zone_t *xfs_btree_cur_zone; + /* * This nonsense is to make -wlint happy. */ diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c index c8f2c2886fe..63debd147eb 100644 --- a/fs/xfs/xfs_buf_item.c +++ b/fs/xfs/xfs_buf_item.c @@ -378,7 +378,6 @@ xfs_buf_item_unpin( xfs_mount_t *mp; xfs_buf_t *bp; int freed; - SPLDECL(s); bp = bip->bli_buf; ASSERT(bp != NULL); @@ -409,8 +408,8 @@ xfs_buf_item_unpin( XFS_BUF_SET_FSPRIVATE(bp, NULL); XFS_BUF_CLR_IODONE_FUNC(bp); } else { - AIL_LOCK(mp,s); - xfs_trans_delete_ail(mp, (xfs_log_item_t *)bip, s); + spin_lock(&mp->m_ail_lock); + xfs_trans_delete_ail(mp, (xfs_log_item_t *)bip); xfs_buf_item_relse(bp); ASSERT(XFS_BUF_FSPRIVATE(bp, void *) == NULL); } @@ -1113,7 +1112,6 @@ xfs_buf_iodone( xfs_buf_log_item_t *bip) { struct xfs_mount *mp; - SPLDECL(s); ASSERT(bip->bli_buf == bp); @@ -1128,11 +1126,11 @@ xfs_buf_iodone( * * Either way, AIL is useless if we're forcing a shutdown. */ - AIL_LOCK(mp,s); + spin_lock(&mp->m_ail_lock); /* * xfs_trans_delete_ail() drops the AIL lock. */ - xfs_trans_delete_ail(mp, (xfs_log_item_t *)bip, s); + xfs_trans_delete_ail(mp, (xfs_log_item_t *)bip); #ifdef XFS_TRANS_DEBUG kmem_free(bip->bli_orig, XFS_BUF_COUNT(bp)); diff --git a/fs/xfs/xfs_buf_item.h b/fs/xfs/xfs_buf_item.h index d7e13614306..5a41c348bb1 100644 --- a/fs/xfs/xfs_buf_item.h +++ b/fs/xfs/xfs_buf_item.h @@ -18,6 +18,8 @@ #ifndef __XFS_BUF_ITEM_H__ #define __XFS_BUF_ITEM_H__ +extern kmem_zone_t *xfs_buf_item_zone; + /* * This is the structure used to lay out a buf log item in the * log. The data map describes which 128 byte chunks of the buffer diff --git a/fs/xfs/xfs_clnt.h b/fs/xfs/xfs_clnt.h index d16c1b97107..d5d1e60ee22 100644 --- a/fs/xfs/xfs_clnt.h +++ b/fs/xfs/xfs_clnt.h @@ -86,7 +86,7 @@ struct xfs_mount_args { #define XFSMNT_NOUUID 0x01000000 /* Ignore fs uuid */ #define XFSMNT_DMAPI 0x02000000 /* enable dmapi/xdsm */ #define XFSMNT_BARRIER 0x04000000 /* use write barriers */ -#define XFSMNT_IDELETE 0x08000000 /* inode cluster delete */ +#define XFSMNT_IKEEP 0x08000000 /* inode cluster delete */ #define XFSMNT_SWALLOC 0x10000000 /* turn on stripe width * allocation */ #define XFSMNT_DIRSYNC 0x40000000 /* sync creat,link,unlink,rename diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c index 26d09e2e1a7..021a8f7e563 100644 --- a/fs/xfs/xfs_da_btree.c +++ b/fs/xfs/xfs_da_btree.c @@ -511,12 +511,12 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1, * Move the req'd B-tree elements from high in node1 to * low in node2. */ - be16_add(&node2->hdr.count, count); + be16_add_cpu(&node2->hdr.count, count); tmp = count * (uint)sizeof(xfs_da_node_entry_t); btree_s = &node1->btree[be16_to_cpu(node1->hdr.count) - count]; btree_d = &node2->btree[0]; memcpy(btree_d, btree_s, tmp); - be16_add(&node1->hdr.count, -count); + be16_add_cpu(&node1->hdr.count, -count); } else { /* * Move the req'd B-tree elements from low in node2 to @@ -527,7 +527,7 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1, btree_s = &node2->btree[0]; btree_d = &node1->btree[be16_to_cpu(node1->hdr.count)]; memcpy(btree_d, btree_s, tmp); - be16_add(&node1->hdr.count, count); + be16_add_cpu(&node1->hdr.count, count); xfs_da_log_buf(tp, blk1->bp, XFS_DA_LOGRANGE(node1, btree_d, tmp)); @@ -539,7 +539,7 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1, btree_s = &node2->btree[count]; btree_d = &node2->btree[0]; memmove(btree_d, btree_s, tmp); - be16_add(&node2->hdr.count, -count); + be16_add_cpu(&node2->hdr.count, -count); } /* @@ -604,7 +604,7 @@ xfs_da_node_add(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk, btree->before = cpu_to_be32(newblk->blkno); xfs_da_log_buf(state->args->trans, oldblk->bp, XFS_DA_LOGRANGE(node, btree, tmp + sizeof(*btree))); - be16_add(&node->hdr.count, 1); + be16_add_cpu(&node->hdr.count, 1); xfs_da_log_buf(state->args->trans, oldblk->bp, XFS_DA_LOGRANGE(node, &node->hdr, sizeof(node->hdr))); @@ -959,7 +959,7 @@ xfs_da_node_remove(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk) memset((char *)btree, 0, sizeof(xfs_da_node_entry_t)); xfs_da_log_buf(state->args->trans, drop_blk->bp, XFS_DA_LOGRANGE(node, btree, sizeof(*btree))); - be16_add(&node->hdr.count, -1); + be16_add_cpu(&node->hdr.count, -1); xfs_da_log_buf(state->args->trans, drop_blk->bp, XFS_DA_LOGRANGE(node, &node->hdr, sizeof(node->hdr))); @@ -1018,7 +1018,7 @@ xfs_da_node_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk, */ tmp = be16_to_cpu(drop_node->hdr.count) * (uint)sizeof(xfs_da_node_entry_t); memcpy(btree, &drop_node->btree[0], tmp); - be16_add(&save_node->hdr.count, be16_to_cpu(drop_node->hdr.count)); + be16_add_cpu(&save_node->hdr.count, be16_to_cpu(drop_node->hdr.count)); xfs_da_log_buf(tp, save_blk->bp, XFS_DA_LOGRANGE(save_node, &save_node->hdr, @@ -2218,7 +2218,7 @@ xfs_da_state_free(xfs_da_state_t *state) #ifdef XFS_DABUF_DEBUG xfs_dabuf_t *xfs_dabuf_global_list; -lock_t xfs_dabuf_global_lock; +spinlock_t xfs_dabuf_global_lock; #endif /* @@ -2264,10 +2264,9 @@ xfs_da_buf_make(int nbuf, xfs_buf_t **bps, inst_t *ra) } #ifdef XFS_DABUF_DEBUG { - SPLDECL(s); xfs_dabuf_t *p; - s = mutex_spinlock(&xfs_dabuf_global_lock); + spin_lock(&xfs_dabuf_global_lock); for (p = xfs_dabuf_global_list; p; p = p->next) { ASSERT(p->blkno != dabuf->blkno || p->target != dabuf->target); @@ -2277,7 +2276,7 @@ xfs_da_buf_make(int nbuf, xfs_buf_t **bps, inst_t *ra) xfs_dabuf_global_list->prev = dabuf; dabuf->next = xfs_dabuf_global_list; xfs_dabuf_global_list = dabuf; - mutex_spinunlock(&xfs_dabuf_global_lock, s); + spin_unlock(&xfs_dabuf_global_lock); } #endif return dabuf; @@ -2319,16 +2318,14 @@ xfs_da_buf_done(xfs_dabuf_t *dabuf) kmem_free(dabuf->data, BBTOB(dabuf->bbcount)); #ifdef XFS_DABUF_DEBUG { - SPLDECL(s); - - s = mutex_spinlock(&xfs_dabuf_global_lock); + spin_lock(&xfs_dabuf_global_lock); if (dabuf->prev) dabuf->prev->next = dabuf->next; else xfs_dabuf_global_list = dabuf->next; if (dabuf->next) dabuf->next->prev = dabuf->prev; - mutex_spinunlock(&xfs_dabuf_global_lock, s); + spin_unlock(&xfs_dabuf_global_lock); } memset(dabuf, 0, XFS_DA_BUF_SIZE(dabuf->nbuf)); #endif diff --git a/fs/xfs/xfs_da_btree.h b/fs/xfs/xfs_da_btree.h index 44dabf02f2a..7facf86f74f 100644 --- a/fs/xfs/xfs_da_btree.h +++ b/fs/xfs/xfs_da_btree.h @@ -260,6 +260,7 @@ void xfs_da_binval(struct xfs_trans *tp, xfs_dabuf_t *dabuf); xfs_daddr_t xfs_da_blkno(xfs_dabuf_t *dabuf); extern struct kmem_zone *xfs_da_state_zone; +extern struct kmem_zone *xfs_dabuf_zone; #endif /* __KERNEL__ */ #endif /* __XFS_DA_BTREE_H__ */ diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c index 584f1ae85cd..3f53fad356a 100644 --- a/fs/xfs/xfs_dfrag.c +++ b/fs/xfs/xfs_dfrag.c @@ -52,76 +52,72 @@ xfs_swapext( xfs_swapext_t __user *sxu) { xfs_swapext_t *sxp; - xfs_inode_t *ip=NULL, *tip=NULL; - xfs_mount_t *mp; - struct file *fp = NULL, *tfp = NULL; - bhv_vnode_t *vp, *tvp; + xfs_inode_t *ip, *tip; + struct file *file, *target_file; int error = 0; sxp = kmem_alloc(sizeof(xfs_swapext_t), KM_MAYFAIL); if (!sxp) { error = XFS_ERROR(ENOMEM); - goto error0; + goto out; } if (copy_from_user(sxp, sxu, sizeof(xfs_swapext_t))) { error = XFS_ERROR(EFAULT); - goto error0; + goto out_free_sxp; } /* Pull information for the target fd */ - if (((fp = fget((int)sxp->sx_fdtarget)) == NULL) || - ((vp = vn_from_inode(fp->f_path.dentry->d_inode)) == NULL)) { + file = fget((int)sxp->sx_fdtarget); + if (!file) { error = XFS_ERROR(EINVAL); - goto error0; + goto out_free_sxp; } - ip = xfs_vtoi(vp); - if (ip == NULL) { + if (!(file->f_mode & FMODE_WRITE) || (file->f_flags & O_APPEND)) { error = XFS_ERROR(EBADF); - goto error0; + goto out_put_file; } - if (((tfp = fget((int)sxp->sx_fdtmp)) == NULL) || - ((tvp = vn_from_inode(tfp->f_path.dentry->d_inode)) == NULL)) { + target_file = fget((int)sxp->sx_fdtmp); + if (!target_file) { error = XFS_ERROR(EINVAL); - goto error0; + goto out_put_file; } - tip = xfs_vtoi(tvp); - if (tip == NULL) { + if (!(target_file->f_mode & FMODE_WRITE) || + (target_file->f_flags & O_APPEND)) { error = XFS_ERROR(EBADF); - goto error0; + goto out_put_target_file; } + ip = XFS_I(file->f_path.dentry->d_inode); + tip = XFS_I(target_file->f_path.dentry->d_inode); + if (ip->i_mount != tip->i_mount) { - error = XFS_ERROR(EINVAL); - goto error0; + error = XFS_ERROR(EINVAL); + goto out_put_target_file; } if (ip->i_ino == tip->i_ino) { - error = XFS_ERROR(EINVAL); - goto error0; + error = XFS_ERROR(EINVAL); + goto out_put_target_file; } - mp = ip->i_mount; - - if (XFS_FORCED_SHUTDOWN(mp)) { - error = XFS_ERROR(EIO); - goto error0; + if (XFS_FORCED_SHUTDOWN(ip->i_mount)) { + error = XFS_ERROR(EIO); + goto out_put_target_file; } - error = XFS_SWAP_EXTENTS(mp, &ip->i_iocore, &tip->i_iocore, sxp); - - error0: - if (fp != NULL) - fput(fp); - if (tfp != NULL) - fput(tfp); - - if (sxp != NULL) - kmem_free(sxp, sizeof(xfs_swapext_t)); + error = xfs_swap_extents(ip, tip, sxp); + out_put_target_file: + fput(target_file); + out_put_file: + fput(file); + out_free_sxp: + kmem_free(sxp, sizeof(xfs_swapext_t)); + out: return error; } @@ -169,15 +165,6 @@ xfs_swap_extents( xfs_lock_inodes(ips, 2, 0, lock_flags); locked = 1; - /* Check permissions */ - error = xfs_iaccess(ip, S_IWUSR, NULL); - if (error) - goto error0; - - error = xfs_iaccess(tip, S_IWUSR, NULL); - if (error) - goto error0; - /* Verify that both files have the same format */ if ((ip->i_d.di_mode & S_IFMT) != (tip->i_d.di_mode & S_IFMT)) { error = XFS_ERROR(EINVAL); @@ -185,8 +172,7 @@ xfs_swap_extents( } /* Verify both files are either real-time or non-realtime */ - if ((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) != - (tip->i_d.di_flags & XFS_DIFLAG_REALTIME)) { + if (XFS_IS_REALTIME_INODE(ip) != XFS_IS_REALTIME_INODE(tip)) { error = XFS_ERROR(EINVAL); goto error0; } @@ -199,7 +185,7 @@ xfs_swap_extents( } if (VN_CACHED(tvp) != 0) { - xfs_inval_cached_trace(&tip->i_iocore, 0, -1, 0, -1); + xfs_inval_cached_trace(tip, 0, -1, 0, -1); error = xfs_flushinval_pages(tip, 0, -1, FI_REMAPF_LOCKED); if (error) diff --git a/fs/xfs/xfs_dinode.h b/fs/xfs/xfs_dinode.h index dedd713574e..c9065eaf2a4 100644 --- a/fs/xfs/xfs_dinode.h +++ b/fs/xfs/xfs_dinode.h @@ -171,69 +171,35 @@ typedef enum xfs_dinode_fmt /* * Inode data & attribute fork sizes, per inode. */ -#define XFS_CFORK_Q(dcp) ((dcp)->di_forkoff != 0) -#define XFS_CFORK_Q_DISK(dcp) ((dcp)->di_forkoff != 0) - -#define XFS_CFORK_BOFF(dcp) ((int)((dcp)->di_forkoff << 3)) -#define XFS_CFORK_BOFF_DISK(dcp) ((int)((dcp)->di_forkoff << 3)) - -#define XFS_CFORK_DSIZE_DISK(dcp,mp) \ - (XFS_CFORK_Q_DISK(dcp) ? XFS_CFORK_BOFF_DISK(dcp) : XFS_LITINO(mp)) -#define XFS_CFORK_DSIZE(dcp,mp) \ - (XFS_CFORK_Q(dcp) ? XFS_CFORK_BOFF(dcp) : XFS_LITINO(mp)) - -#define XFS_CFORK_ASIZE_DISK(dcp,mp) \ - (XFS_CFORK_Q_DISK(dcp) ? XFS_LITINO(mp) - XFS_CFORK_BOFF_DISK(dcp) : 0) -#define XFS_CFORK_ASIZE(dcp,mp) \ - (XFS_CFORK_Q(dcp) ? XFS_LITINO(mp) - XFS_CFORK_BOFF(dcp) : 0) - -#define XFS_CFORK_SIZE_DISK(dcp,mp,w) \ - ((w) == XFS_DATA_FORK ? \ - XFS_CFORK_DSIZE_DISK(dcp, mp) : \ - XFS_CFORK_ASIZE_DISK(dcp, mp)) -#define XFS_CFORK_SIZE(dcp,mp,w) \ - ((w) == XFS_DATA_FORK ? \ - XFS_CFORK_DSIZE(dcp, mp) : XFS_CFORK_ASIZE(dcp, mp)) +#define XFS_DFORK_Q(dip) ((dip)->di_core.di_forkoff != 0) +#define XFS_DFORK_BOFF(dip) ((int)((dip)->di_core.di_forkoff << 3)) #define XFS_DFORK_DSIZE(dip,mp) \ - XFS_CFORK_DSIZE_DISK(&(dip)->di_core, mp) -#define XFS_DFORK_DSIZE_HOST(dip,mp) \ - XFS_CFORK_DSIZE(&(dip)->di_core, mp) + (XFS_DFORK_Q(dip) ? \ + XFS_DFORK_BOFF(dip) : \ + XFS_LITINO(mp)) #define XFS_DFORK_ASIZE(dip,mp) \ - XFS_CFORK_ASIZE_DISK(&(dip)->di_core, mp) -#define XFS_DFORK_ASIZE_HOST(dip,mp) \ - XFS_CFORK_ASIZE(&(dip)->di_core, mp) -#define XFS_DFORK_SIZE(dip,mp,w) \ - XFS_CFORK_SIZE_DISK(&(dip)->di_core, mp, w) -#define XFS_DFORK_SIZE_HOST(dip,mp,w) \ - XFS_CFORK_SIZE(&(dip)->di_core, mp, w) + (XFS_DFORK_Q(dip) ? \ + XFS_LITINO(mp) - XFS_DFORK_BOFF(dip) : \ + 0) +#define XFS_DFORK_SIZE(dip,mp,w) \ + ((w) == XFS_DATA_FORK ? \ + XFS_DFORK_DSIZE(dip, mp) : \ + XFS_DFORK_ASIZE(dip, mp)) -#define XFS_DFORK_Q(dip) XFS_CFORK_Q_DISK(&(dip)->di_core) -#define XFS_DFORK_BOFF(dip) XFS_CFORK_BOFF_DISK(&(dip)->di_core) -#define XFS_DFORK_DPTR(dip) ((dip)->di_u.di_c) -#define XFS_DFORK_APTR(dip) \ +#define XFS_DFORK_DPTR(dip) ((dip)->di_u.di_c) +#define XFS_DFORK_APTR(dip) \ ((dip)->di_u.di_c + XFS_DFORK_BOFF(dip)) -#define XFS_DFORK_PTR(dip,w) \ +#define XFS_DFORK_PTR(dip,w) \ ((w) == XFS_DATA_FORK ? XFS_DFORK_DPTR(dip) : XFS_DFORK_APTR(dip)) -#define XFS_CFORK_FORMAT(dcp,w) \ - ((w) == XFS_DATA_FORK ? (dcp)->di_format : (dcp)->di_aformat) -#define XFS_CFORK_FMT_SET(dcp,w,n) \ +#define XFS_DFORK_FORMAT(dip,w) \ ((w) == XFS_DATA_FORK ? \ - ((dcp)->di_format = (n)) : ((dcp)->di_aformat = (n))) -#define XFS_DFORK_FORMAT(dip,w) XFS_CFORK_FORMAT(&(dip)->di_core, w) - -#define XFS_CFORK_NEXTENTS_DISK(dcp,w) \ + (dip)->di_core.di_format : \ + (dip)->di_core.di_aformat) +#define XFS_DFORK_NEXTENTS(dip,w) \ ((w) == XFS_DATA_FORK ? \ - be32_to_cpu((dcp)->di_nextents) : \ - be16_to_cpu((dcp)->di_anextents)) -#define XFS_CFORK_NEXTENTS(dcp,w) \ - ((w) == XFS_DATA_FORK ? (dcp)->di_nextents : (dcp)->di_anextents) -#define XFS_DFORK_NEXTENTS(dip,w) XFS_CFORK_NEXTENTS_DISK(&(dip)->di_core, w) -#define XFS_DFORK_NEXTENTS_HOST(dip,w) XFS_CFORK_NEXTENTS(&(dip)->di_core, w) - -#define XFS_CFORK_NEXT_SET(dcp,w,n) \ - ((w) == XFS_DATA_FORK ? \ - ((dcp)->di_nextents = (n)) : ((dcp)->di_anextents = (n))) + be32_to_cpu((dip)->di_core.di_nextents) : \ + be16_to_cpu((dip)->di_core.di_anextents)) #define XFS_BUF_TO_DINODE(bp) ((xfs_dinode_t *)XFS_BUF_PTR(bp)) @@ -273,6 +239,12 @@ typedef enum xfs_dinode_fmt #define XFS_DIFLAG_NODEFRAG (1 << XFS_DIFLAG_NODEFRAG_BIT) #define XFS_DIFLAG_FILESTREAM (1 << XFS_DIFLAG_FILESTREAM_BIT) +#ifdef CONFIG_XFS_RT +#define XFS_IS_REALTIME_INODE(ip) ((ip)->i_d.di_flags & XFS_DIFLAG_REALTIME) +#else +#define XFS_IS_REALTIME_INODE(ip) (0) +#endif + #define XFS_DIFLAG_ANY \ (XFS_DIFLAG_REALTIME | XFS_DIFLAG_PREALLOC | XFS_DIFLAG_NEWRTBM | \ XFS_DIFLAG_IMMUTABLE | XFS_DIFLAG_APPEND | XFS_DIFLAG_SYNC | \ diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c index b0f1ee8fcb9..e92e73f0e6a 100644 --- a/fs/xfs/xfs_dir2.c +++ b/fs/xfs/xfs_dir2.c @@ -42,13 +42,14 @@ #include "xfs_dir2_node.h" #include "xfs_dir2_trace.h" #include "xfs_error.h" +#include "xfs_vnodeops.h" void xfs_dir_mount( xfs_mount_t *mp) { - ASSERT(XFS_SB_VERSION_HASDIRV2(&mp->m_sb)); + ASSERT(xfs_sb_version_hasdirv2(&mp->m_sb)); ASSERT((1 << (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog)) <= XFS_MAX_BLOCKSIZE); mp->m_dirblksize = 1 << (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog); @@ -301,7 +302,7 @@ xfs_readdir( int rval; /* return value */ int v; /* type-checking value */ - vn_trace_entry(dp, __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(dp); if (XFS_FORCED_SHUTDOWN(dp->i_mount)) return XFS_ERROR(EIO); diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/xfs_dir2_block.c index a5f4f4fb886..fb5a556725b 100644 --- a/fs/xfs/xfs_dir2_block.c +++ b/fs/xfs/xfs_dir2_block.c @@ -271,7 +271,7 @@ xfs_dir2_block_addname( } lfloglow = toidx + 1 - (be32_to_cpu(btp->stale) - 1); lfloghigh -= be32_to_cpu(btp->stale) - 1; - be32_add(&btp->count, -(be32_to_cpu(btp->stale) - 1)); + be32_add_cpu(&btp->count, -(be32_to_cpu(btp->stale) - 1)); xfs_dir2_data_make_free(tp, bp, (xfs_dir2_data_aoff_t)((char *)blp - (char *)block), (xfs_dir2_data_aoff_t)((be32_to_cpu(btp->stale) - 1) * sizeof(*blp)), @@ -326,7 +326,7 @@ xfs_dir2_block_addname( /* * Update the tail (entry count). */ - be32_add(&btp->count, 1); + be32_add_cpu(&btp->count, 1); /* * If we now need to rebuild the bestfree map, do so. * This needs to happen before the next call to use_free. @@ -387,7 +387,7 @@ xfs_dir2_block_addname( lfloglow = MIN(mid, lfloglow); lfloghigh = MAX(highstale, lfloghigh); } - be32_add(&btp->stale, -1); + be32_add_cpu(&btp->stale, -1); } /* * Point to the new data entry. @@ -767,7 +767,7 @@ xfs_dir2_block_removename( /* * Fix up the block tail. */ - be32_add(&btp->stale, 1); + be32_add_cpu(&btp->stale, 1); xfs_dir2_block_log_tail(tp, bp); /* * Remove the leaf entry by marking it stale. diff --git a/fs/xfs/xfs_dir2_data.c b/fs/xfs/xfs_dir2_data.c index d2452699e9b..fb8c9e08b23 100644 --- a/fs/xfs/xfs_dir2_data.c +++ b/fs/xfs/xfs_dir2_data.c @@ -587,7 +587,7 @@ xfs_dir2_data_make_free( /* * Fix up the new big freespace. */ - be16_add(&prevdup->length, len + be16_to_cpu(postdup->length)); + be16_add_cpu(&prevdup->length, len + be16_to_cpu(postdup->length)); *xfs_dir2_data_unused_tag_p(prevdup) = cpu_to_be16((char *)prevdup - (char *)d); xfs_dir2_data_log_unused(tp, bp, prevdup); @@ -621,7 +621,7 @@ xfs_dir2_data_make_free( */ else if (prevdup) { dfp = xfs_dir2_data_freefind(d, prevdup); - be16_add(&prevdup->length, len); + be16_add_cpu(&prevdup->length, len); *xfs_dir2_data_unused_tag_p(prevdup) = cpu_to_be16((char *)prevdup - (char *)d); xfs_dir2_data_log_unused(tp, bp, prevdup); diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/xfs_dir2_leaf.c index 0ca0020ba09..bc52b803d79 100644 --- a/fs/xfs/xfs_dir2_leaf.c +++ b/fs/xfs/xfs_dir2_leaf.c @@ -359,7 +359,7 @@ xfs_dir2_leaf_addname( bestsp--; memmove(&bestsp[0], &bestsp[1], be32_to_cpu(ltp->bestcount) * sizeof(bestsp[0])); - be32_add(<p->bestcount, 1); + be32_add_cpu(<p->bestcount, 1); xfs_dir2_leaf_log_tail(tp, lbp); xfs_dir2_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1); } @@ -445,7 +445,7 @@ xfs_dir2_leaf_addname( */ lfloglow = index; lfloghigh = be16_to_cpu(leaf->hdr.count); - be16_add(&leaf->hdr.count, 1); + be16_add_cpu(&leaf->hdr.count, 1); } /* * There are stale entries. @@ -523,7 +523,7 @@ xfs_dir2_leaf_addname( lfloglow = MIN(index, lfloglow); lfloghigh = MAX(highstale, lfloghigh); } - be16_add(&leaf->hdr.stale, -1); + be16_add_cpu(&leaf->hdr.stale, -1); } /* * Fill in the new leaf entry. @@ -626,7 +626,7 @@ xfs_dir2_leaf_compact( * Update and log the header, log the leaf entries. */ ASSERT(be16_to_cpu(leaf->hdr.stale) == from - to); - be16_add(&leaf->hdr.count, -(be16_to_cpu(leaf->hdr.stale))); + be16_add_cpu(&leaf->hdr.count, -(be16_to_cpu(leaf->hdr.stale))); leaf->hdr.stale = 0; xfs_dir2_leaf_log_header(args->trans, bp); if (loglow != -1) @@ -728,7 +728,7 @@ xfs_dir2_leaf_compact_x1( /* * Adjust the leaf header values. */ - be16_add(&leaf->hdr.count, -(from - to)); + be16_add_cpu(&leaf->hdr.count, -(from - to)); leaf->hdr.stale = cpu_to_be16(1); /* * Remember the low/high stale value only in the "right" @@ -1470,7 +1470,7 @@ xfs_dir2_leaf_removename( /* * We just mark the leaf entry stale by putting a null in it. */ - be16_add(&leaf->hdr.stale, 1); + be16_add_cpu(&leaf->hdr.stale, 1); xfs_dir2_leaf_log_header(tp, lbp); lep->address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR); xfs_dir2_leaf_log_ents(tp, lbp, index, index); @@ -1531,7 +1531,7 @@ xfs_dir2_leaf_removename( */ memmove(&bestsp[db - i], bestsp, (be32_to_cpu(ltp->bestcount) - (db - i)) * sizeof(*bestsp)); - be32_add(<p->bestcount, -(db - i)); + be32_add_cpu(<p->bestcount, -(db - i)); xfs_dir2_leaf_log_tail(tp, lbp); xfs_dir2_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1); } else @@ -1712,7 +1712,7 @@ xfs_dir2_leaf_trim_data( * Eliminate the last bests entry from the table. */ bestsp = xfs_dir2_leaf_bests_p(ltp); - be32_add(<p->bestcount, -1); + be32_add_cpu(<p->bestcount, -1); memmove(&bestsp[1], &bestsp[0], be32_to_cpu(ltp->bestcount) * sizeof(*bestsp)); xfs_dir2_leaf_log_tail(tp, lbp); xfs_dir2_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1); diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/xfs_dir2_node.c index eb18e399e83..8dade711f09 100644 --- a/fs/xfs/xfs_dir2_node.c +++ b/fs/xfs/xfs_dir2_node.c @@ -254,7 +254,7 @@ xfs_dir2_leafn_add( (be16_to_cpu(leaf->hdr.count) - index) * sizeof(*lep)); lfloglow = index; lfloghigh = be16_to_cpu(leaf->hdr.count); - be16_add(&leaf->hdr.count, 1); + be16_add_cpu(&leaf->hdr.count, 1); } /* * There are stale entries. We'll use one for the new entry. @@ -322,7 +322,7 @@ xfs_dir2_leafn_add( lfloglow = MIN(index, lfloglow); lfloghigh = MAX(highstale, lfloghigh); } - be16_add(&leaf->hdr.stale, -1); + be16_add_cpu(&leaf->hdr.stale, -1); } /* * Insert the new entry, log everything. @@ -697,10 +697,10 @@ xfs_dir2_leafn_moveents( /* * Update the headers and log them. */ - be16_add(&leaf_s->hdr.count, -(count)); - be16_add(&leaf_s->hdr.stale, -(stale)); - be16_add(&leaf_d->hdr.count, count); - be16_add(&leaf_d->hdr.stale, stale); + be16_add_cpu(&leaf_s->hdr.count, -(count)); + be16_add_cpu(&leaf_s->hdr.stale, -(stale)); + be16_add_cpu(&leaf_d->hdr.count, count); + be16_add_cpu(&leaf_d->hdr.stale, stale); xfs_dir2_leaf_log_header(tp, bp_s); xfs_dir2_leaf_log_header(tp, bp_d); xfs_dir2_leafn_check(args->dp, bp_s); @@ -885,7 +885,7 @@ xfs_dir2_leafn_remove( * Kill the leaf entry by marking it stale. * Log the leaf block changes. */ - be16_add(&leaf->hdr.stale, 1); + be16_add_cpu(&leaf->hdr.stale, 1); xfs_dir2_leaf_log_header(tp, bp); lep->address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR); xfs_dir2_leaf_log_ents(tp, bp, index, index); @@ -971,7 +971,7 @@ xfs_dir2_leafn_remove( /* * One less used entry in the free table. */ - be32_add(&free->hdr.nused, -1); + be32_add_cpu(&free->hdr.nused, -1); xfs_dir2_free_log_header(tp, fbp); /* * If this was the last entry in the table, we can @@ -1642,7 +1642,7 @@ xfs_dir2_node_addname_int( * (this should always be true) then update the header. */ if (be16_to_cpu(free->bests[findex]) == NULLDATAOFF) { - be32_add(&free->hdr.nused, 1); + be32_add_cpu(&free->hdr.nused, 1); xfs_dir2_free_log_header(tp, fbp); } /* diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c index a4634d94e56..05e5365d3c3 100644 --- a/fs/xfs/xfs_error.c +++ b/fs/xfs/xfs_error.c @@ -230,37 +230,6 @@ xfs_error_report( } } -STATIC void -xfs_hex_dump(void *p, int length) -{ - __uint8_t *uip = (__uint8_t*)p; - int i; - char sbuf[128], *s; - - s = sbuf; - *s = '\0'; - for (i=0; i<length; i++, uip++) { - if ((i % 16) == 0) { - if (*s != '\0') - cmn_err(CE_ALERT, "%s\n", sbuf); - s = sbuf; - sprintf(s, "0x%x: ", i); - while( *s != '\0') - s++; - } - sprintf(s, "%02x ", *uip); - - /* - * the kernel sprintf is a void; user sprintf returns - * the sprintf'ed string's length. Find the new end- - * of-string - */ - while( *s != '\0') - s++; - } - cmn_err(CE_ALERT, "%s\n", sbuf); -} - void xfs_corruption_error( char *tag, diff --git a/fs/xfs/xfs_error.h b/fs/xfs/xfs_error.h index 10e9d9619ae..6490d2a9f8e 100644 --- a/fs/xfs/xfs_error.h +++ b/fs/xfs/xfs_error.h @@ -174,6 +174,8 @@ extern void xfs_cmn_err(int panic_tag, int level, struct xfs_mount *mp, /* PRINTFLIKE3 */ extern void xfs_fs_cmn_err(int level, struct xfs_mount *mp, char *fmt, ...); +extern void xfs_hex_dump(void *p, int length); + #define xfs_fs_repair_cmn_err(level, mp, fmt, args...) \ xfs_fs_cmn_err(level, mp, fmt " Unmount and run xfs_repair.", ## args) diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c index f938a51be81..132bd07b9bb 100644 --- a/fs/xfs/xfs_extfree_item.c +++ b/fs/xfs/xfs_extfree_item.c @@ -110,19 +110,18 @@ STATIC void xfs_efi_item_unpin(xfs_efi_log_item_t *efip, int stale) { xfs_mount_t *mp; - SPLDECL(s); mp = efip->efi_item.li_mountp; - AIL_LOCK(mp, s); + spin_lock(&mp->m_ail_lock); if (efip->efi_flags & XFS_EFI_CANCELED) { /* * xfs_trans_delete_ail() drops the AIL lock. */ - xfs_trans_delete_ail(mp, (xfs_log_item_t *)efip, s); + xfs_trans_delete_ail(mp, (xfs_log_item_t *)efip); xfs_efi_item_free(efip); } else { efip->efi_flags |= XFS_EFI_COMMITTED; - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); } } @@ -138,10 +137,9 @@ xfs_efi_item_unpin_remove(xfs_efi_log_item_t *efip, xfs_trans_t *tp) { xfs_mount_t *mp; xfs_log_item_desc_t *lidp; - SPLDECL(s); mp = efip->efi_item.li_mountp; - AIL_LOCK(mp, s); + spin_lock(&mp->m_ail_lock); if (efip->efi_flags & XFS_EFI_CANCELED) { /* * free the xaction descriptor pointing to this item @@ -152,11 +150,11 @@ xfs_efi_item_unpin_remove(xfs_efi_log_item_t *efip, xfs_trans_t *tp) * pull the item off the AIL. * xfs_trans_delete_ail() drops the AIL lock. */ - xfs_trans_delete_ail(mp, (xfs_log_item_t *)efip, s); + xfs_trans_delete_ail(mp, (xfs_log_item_t *)efip); xfs_efi_item_free(efip); } else { efip->efi_flags |= XFS_EFI_COMMITTED; - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); } } @@ -350,13 +348,12 @@ xfs_efi_release(xfs_efi_log_item_t *efip, { xfs_mount_t *mp; int extents_left; - SPLDECL(s); mp = efip->efi_item.li_mountp; ASSERT(efip->efi_next_extent > 0); ASSERT(efip->efi_flags & XFS_EFI_COMMITTED); - AIL_LOCK(mp, s); + spin_lock(&mp->m_ail_lock); ASSERT(efip->efi_next_extent >= nextents); efip->efi_next_extent -= nextents; extents_left = efip->efi_next_extent; @@ -364,10 +361,10 @@ xfs_efi_release(xfs_efi_log_item_t *efip, /* * xfs_trans_delete_ail() drops the AIL lock. */ - xfs_trans_delete_ail(mp, (xfs_log_item_t *)efip, s); + xfs_trans_delete_ail(mp, (xfs_log_item_t *)efip); xfs_efi_item_free(efip); } else { - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); } } diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 36d8f6aa11a..eb03eab5ca5 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -348,7 +348,7 @@ _xfs_filestream_update_ag( } /* xfs_fstrm_free_func(): callback for freeing cached stream items. */ -void +STATIC void xfs_fstrm_free_func( unsigned long ino, void *data) diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h index aab96627651..3bed6433d05 100644 --- a/fs/xfs/xfs_fs.h +++ b/fs/xfs/xfs_fs.h @@ -419,9 +419,13 @@ typedef struct xfs_handle { /* * ioctl commands that are used by Linux filesystems */ -#define XFS_IOC_GETXFLAGS _IOR('f', 1, long) -#define XFS_IOC_SETXFLAGS _IOW('f', 2, long) -#define XFS_IOC_GETVERSION _IOR('v', 1, long) +#define XFS_IOC_GETXFLAGS FS_IOC_GETFLAGS +#define XFS_IOC_SETXFLAGS FS_IOC_SETFLAGS +#define XFS_IOC_GETVERSION FS_IOC_GETVERSION +/* 32-bit compat counterparts */ +#define XFS_IOC32_GETXFLAGS FS_IOC32_GETFLAGS +#define XFS_IOC32_SETXFLAGS FS_IOC32_SETFLAGS +#define XFS_IOC32_GETVERSION FS_IOC32_GETVERSION /* * ioctl commands that replace IRIX fcntl()'s diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index c92d5b82102..d3a0f538d6a 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -77,36 +77,36 @@ xfs_fs_geometry( if (new_version >= 3) { geo->version = XFS_FSOP_GEOM_VERSION; geo->flags = - (XFS_SB_VERSION_HASATTR(&mp->m_sb) ? + (xfs_sb_version_hasattr(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_ATTR : 0) | - (XFS_SB_VERSION_HASNLINK(&mp->m_sb) ? + (xfs_sb_version_hasnlink(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_NLINK : 0) | - (XFS_SB_VERSION_HASQUOTA(&mp->m_sb) ? + (xfs_sb_version_hasquota(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_QUOTA : 0) | - (XFS_SB_VERSION_HASALIGN(&mp->m_sb) ? + (xfs_sb_version_hasalign(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_IALIGN : 0) | - (XFS_SB_VERSION_HASDALIGN(&mp->m_sb) ? + (xfs_sb_version_hasdalign(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_DALIGN : 0) | - (XFS_SB_VERSION_HASSHARED(&mp->m_sb) ? + (xfs_sb_version_hasshared(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_SHARED : 0) | - (XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb) ? + (xfs_sb_version_hasextflgbit(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_EXTFLG : 0) | - (XFS_SB_VERSION_HASDIRV2(&mp->m_sb) ? + (xfs_sb_version_hasdirv2(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_DIRV2 : 0) | - (XFS_SB_VERSION_HASSECTOR(&mp->m_sb) ? + (xfs_sb_version_hassector(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_SECTOR : 0) | (xfs_sb_version_haslazysbcount(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_LAZYSB : 0) | - (XFS_SB_VERSION_HASATTR2(&mp->m_sb) ? + (xfs_sb_version_hasattr2(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_ATTR2 : 0); - geo->logsectsize = XFS_SB_VERSION_HASSECTOR(&mp->m_sb) ? + geo->logsectsize = xfs_sb_version_hassector(&mp->m_sb) ? mp->m_sb.sb_logsectsize : BBSIZE; geo->rtsectsize = mp->m_sb.sb_blocksize; geo->dirblocksize = mp->m_dirblksize; } if (new_version >= 4) { geo->flags |= - (XFS_SB_VERSION_HASLOGV2(&mp->m_sb) ? + (xfs_sb_version_haslogv2(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_LOGV2 : 0); geo->logsunit = mp->m_sb.sb_logsunit; } @@ -318,7 +318,7 @@ xfs_growfs_data_private( } ASSERT(bp); agi = XFS_BUF_TO_AGI(bp); - be32_add(&agi->agi_length, new); + be32_add_cpu(&agi->agi_length, new); ASSERT(nagcount == oagcount || be32_to_cpu(agi->agi_length) == mp->m_sb.sb_agblocks); xfs_ialloc_log_agi(tp, bp, XFS_AGI_LENGTH); @@ -331,7 +331,7 @@ xfs_growfs_data_private( } ASSERT(bp); agf = XFS_BUF_TO_AGF(bp); - be32_add(&agf->agf_length, new); + be32_add_cpu(&agf->agf_length, new); ASSERT(be32_to_cpu(agf->agf_length) == be32_to_cpu(agi->agi_length)); xfs_alloc_log_agf(tp, bp, XFS_AGF_LENGTH); @@ -462,15 +462,13 @@ xfs_fs_counts( xfs_mount_t *mp, xfs_fsop_counts_t *cnt) { - unsigned long s; - xfs_icsb_sync_counters_flags(mp, XFS_ICSB_LAZY_COUNT); - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); cnt->freedata = mp->m_sb.sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp); cnt->freertx = mp->m_sb.sb_frextents; cnt->freeino = mp->m_sb.sb_ifree; cnt->allocino = mp->m_sb.sb_icount; - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); return 0; } @@ -497,7 +495,6 @@ xfs_reserve_blocks( { __int64_t lcounter, delta, fdblks_delta; __uint64_t request; - unsigned long s; /* If inval is null, report current values and return */ if (inval == (__uint64_t *)NULL) { @@ -515,7 +512,7 @@ xfs_reserve_blocks( * problem. we needto work out if we are freeing or allocation * blocks first, then we can do the modification as necessary. * - * We do this under the XFS_SB_LOCK so that if we are near + * We do this under the m_sb_lock so that if we are near * ENOSPC, we will hold out any changes while we work out * what to do. This means that the amount of free space can * change while we do this, so we need to retry if we end up @@ -526,7 +523,7 @@ xfs_reserve_blocks( * enabled, disabled or even compiled in.... */ retry: - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); xfs_icsb_sync_counters_flags(mp, XFS_ICSB_SB_LOCKED); /* @@ -569,7 +566,7 @@ out: outval->resblks = mp->m_resblks; outval->resblks_avail = mp->m_resblks_avail; } - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); if (fdblks_delta) { /* diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c index 1409c2d61c1..5a146cb2298 100644 --- a/fs/xfs/xfs_ialloc.c +++ b/fs/xfs/xfs_ialloc.c @@ -191,7 +191,7 @@ xfs_ialloc_ag_alloc( ASSERT(!(args.mp->m_flags & XFS_MOUNT_NOALIGN)); args.alignment = args.mp->m_dalign; isaligned = 1; - } else if (XFS_SB_VERSION_HASALIGN(&args.mp->m_sb) && + } else if (xfs_sb_version_hasalign(&args.mp->m_sb) && args.mp->m_sb.sb_inoalignmt >= XFS_B_TO_FSBT(args.mp, XFS_INODE_CLUSTER_SIZE(args.mp))) @@ -230,7 +230,7 @@ xfs_ialloc_ag_alloc( args.agbno = be32_to_cpu(agi->agi_root); args.fsbno = XFS_AGB_TO_FSB(args.mp, be32_to_cpu(agi->agi_seqno), args.agbno); - if (XFS_SB_VERSION_HASALIGN(&args.mp->m_sb) && + if (xfs_sb_version_hasalign(&args.mp->m_sb) && args.mp->m_sb.sb_inoalignmt >= XFS_B_TO_FSBT(args.mp, XFS_INODE_CLUSTER_SIZE(args.mp))) args.alignment = args.mp->m_sb.sb_inoalignmt; @@ -271,7 +271,7 @@ xfs_ialloc_ag_alloc( * use the old version so that old kernels will continue to be * able to use the file system. */ - if (XFS_SB_VERSION_HASNLINK(&args.mp->m_sb)) + if (xfs_sb_version_hasnlink(&args.mp->m_sb)) version = XFS_DINODE_VERSION_2; else version = XFS_DINODE_VERSION_1; @@ -301,8 +301,8 @@ xfs_ialloc_ag_alloc( } xfs_trans_inode_alloc_buf(tp, fbuf); } - be32_add(&agi->agi_count, newlen); - be32_add(&agi->agi_freecount, newlen); + be32_add_cpu(&agi->agi_count, newlen); + be32_add_cpu(&agi->agi_freecount, newlen); agno = be32_to_cpu(agi->agi_seqno); down_read(&args.mp->m_peraglock); args.mp->m_perag[agno].pagi_freecount += newlen; @@ -885,7 +885,7 @@ nextag: if ((error = xfs_inobt_update(cur, rec.ir_startino, rec.ir_freecount, rec.ir_free))) goto error0; - be32_add(&agi->agi_freecount, -1); + be32_add_cpu(&agi->agi_freecount, -1); xfs_ialloc_log_agi(tp, agbp, XFS_AGI_FREECOUNT); down_read(&mp->m_peraglock); mp->m_perag[tagno].pagi_freecount--; @@ -1053,7 +1053,7 @@ xfs_difree( /* * When an inode cluster is free, it becomes eligible for removal */ - if ((mp->m_flags & XFS_MOUNT_IDELETE) && + if (!(mp->m_flags & XFS_MOUNT_IKEEP) && (rec.ir_freecount == XFS_IALLOC_INODES(mp))) { *delete = 1; @@ -1065,8 +1065,8 @@ xfs_difree( * to be freed when the transaction is committed. */ ilen = XFS_IALLOC_INODES(mp); - be32_add(&agi->agi_count, -ilen); - be32_add(&agi->agi_freecount, -(ilen - 1)); + be32_add_cpu(&agi->agi_count, -ilen); + be32_add_cpu(&agi->agi_freecount, -(ilen - 1)); xfs_ialloc_log_agi(tp, agbp, XFS_AGI_COUNT | XFS_AGI_FREECOUNT); down_read(&mp->m_peraglock); mp->m_perag[agno].pagi_freecount -= ilen - 1; @@ -1095,7 +1095,7 @@ xfs_difree( /* * Change the inode free counts and log the ag/sb changes. */ - be32_add(&agi->agi_freecount, 1); + be32_add_cpu(&agi->agi_freecount, 1); xfs_ialloc_log_agi(tp, agbp, XFS_AGI_FREECOUNT); down_read(&mp->m_peraglock); mp->m_perag[agno].pagi_freecount++; diff --git a/fs/xfs/xfs_ialloc_btree.c b/fs/xfs/xfs_ialloc_btree.c index 8cdeeaf8632..e5310c90e50 100644 --- a/fs/xfs/xfs_ialloc_btree.c +++ b/fs/xfs/xfs_ialloc_btree.c @@ -189,7 +189,7 @@ xfs_inobt_delrec( */ bno = be32_to_cpu(agi->agi_root); agi->agi_root = *pp; - be32_add(&agi->agi_level, -1); + be32_add_cpu(&agi->agi_level, -1); /* * Free the block. */ @@ -1132,7 +1132,7 @@ xfs_inobt_lshift( /* * Bump and log left's numrecs, decrement and log right's numrecs. */ - be16_add(&left->bb_numrecs, 1); + be16_add_cpu(&left->bb_numrecs, 1); xfs_inobt_log_block(cur->bc_tp, lbp, XFS_BB_NUMRECS); #ifdef DEBUG if (level > 0) @@ -1140,7 +1140,7 @@ xfs_inobt_lshift( else xfs_btree_check_rec(cur->bc_btnum, lrp - 1, lrp); #endif - be16_add(&right->bb_numrecs, -1); + be16_add_cpu(&right->bb_numrecs, -1); xfs_inobt_log_block(cur->bc_tp, rbp, XFS_BB_NUMRECS); /* * Slide the contents of right down one entry. @@ -1232,7 +1232,7 @@ xfs_inobt_newroot( * Set the root data in the a.g. inode structure. */ agi->agi_root = cpu_to_be32(args.agbno); - be32_add(&agi->agi_level, 1); + be32_add_cpu(&agi->agi_level, 1); xfs_ialloc_log_agi(args.tp, cur->bc_private.i.agbp, XFS_AGI_ROOT | XFS_AGI_LEVEL); /* @@ -1426,9 +1426,9 @@ xfs_inobt_rshift( /* * Decrement and log left's numrecs, bump and log right's numrecs. */ - be16_add(&left->bb_numrecs, -1); + be16_add_cpu(&left->bb_numrecs, -1); xfs_inobt_log_block(cur->bc_tp, lbp, XFS_BB_NUMRECS); - be16_add(&right->bb_numrecs, 1); + be16_add_cpu(&right->bb_numrecs, 1); #ifdef DEBUG if (level > 0) xfs_btree_check_key(cur->bc_btnum, rkp, rkp + 1); @@ -1529,7 +1529,7 @@ xfs_inobt_split( */ if ((be16_to_cpu(left->bb_numrecs) & 1) && cur->bc_ptrs[level] <= be16_to_cpu(right->bb_numrecs) + 1) - be16_add(&right->bb_numrecs, 1); + be16_add_cpu(&right->bb_numrecs, 1); i = be16_to_cpu(left->bb_numrecs) - be16_to_cpu(right->bb_numrecs) + 1; /* * For non-leaf blocks, copy keys and addresses over to the new block. @@ -1565,7 +1565,7 @@ xfs_inobt_split( * Find the left block number by looking in the buffer. * Adjust numrecs, sibling pointers. */ - be16_add(&left->bb_numrecs, -(be16_to_cpu(right->bb_numrecs))); + be16_add_cpu(&left->bb_numrecs, -(be16_to_cpu(right->bb_numrecs))); right->bb_rightsib = left->bb_rightsib; left->bb_rightsib = cpu_to_be32(args.agbno); right->bb_leftsib = cpu_to_be32(lbno); diff --git a/fs/xfs/xfs_ialloc_btree.h b/fs/xfs/xfs_ialloc_btree.h index bf8e9aff272..8efc4a5b8b9 100644 --- a/fs/xfs/xfs_ialloc_btree.h +++ b/fs/xfs/xfs_ialloc_btree.h @@ -81,8 +81,6 @@ typedef struct xfs_btree_sblock xfs_inobt_block_t; #define XFS_INOBT_MASK(i) ((xfs_inofree_t)1 << (i)) #define XFS_INOBT_IS_FREE(rp,i) \ (((rp)->ir_free & XFS_INOBT_MASK(i)) != 0) -#define XFS_INOBT_IS_FREE_DISK(rp,i) \ - ((be64_to_cpu((rp)->ir_free) & XFS_INOBT_MASK(i)) != 0) #define XFS_INOBT_SET_FREE(rp,i) ((rp)->ir_free |= XFS_INOBT_MASK(i)) #define XFS_INOBT_CLR_FREE(rp,i) ((rp)->ir_free &= ~XFS_INOBT_MASK(i)) diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index fb69ef180b2..8e09b71f410 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -65,7 +65,7 @@ */ STATIC int xfs_iget_core( - bhv_vnode_t *vp, + struct inode *inode, xfs_mount_t *mp, xfs_trans_t *tp, xfs_ino_t ino, @@ -74,9 +74,9 @@ xfs_iget_core( xfs_inode_t **ipp, xfs_daddr_t bno) { + struct inode *old_inode; xfs_inode_t *ip; xfs_inode_t *iq; - bhv_vnode_t *inode_vp; int error; xfs_icluster_t *icl, *new_icl = NULL; unsigned long first_index, mask; @@ -111,8 +111,8 @@ again: goto again; } - inode_vp = XFS_ITOV_NULL(ip); - if (inode_vp == NULL) { + old_inode = ip->i_vnode; + if (old_inode == NULL) { /* * If IRECLAIM is set this inode is * on its way out of the system, @@ -140,28 +140,9 @@ again: return ENOENT; } - /* - * There may be transactions sitting in the - * incore log buffers or being flushed to disk - * at this time. We can't clear the - * XFS_IRECLAIMABLE flag until these - * transactions have hit the disk, otherwise we - * will void the guarantee the flag provides - * xfs_iunpin() - */ - if (xfs_ipincount(ip)) { - read_unlock(&pag->pag_ici_lock); - xfs_log_force(mp, 0, - XFS_LOG_FORCE|XFS_LOG_SYNC); - XFS_STATS_INC(xs_ig_frecycle); - goto again; - } - - vn_trace_exit(ip, "xfs_iget.alloc", - (inst_t *)__return_address); + xfs_itrace_exit_tag(ip, "xfs_iget.alloc"); XFS_STATS_INC(xs_ig_found); - xfs_iflags_clear(ip, XFS_IRECLAIMABLE); read_unlock(&pag->pag_ici_lock); @@ -171,13 +152,11 @@ again: goto finish_inode; - } else if (vp != inode_vp) { - struct inode *inode = vn_to_inode(inode_vp); - + } else if (inode != old_inode) { /* The inode is being torn down, pause and * try again. */ - if (inode->i_state & (I_FREEING | I_CLEAR)) { + if (old_inode->i_state & (I_FREEING | I_CLEAR)) { read_unlock(&pag->pag_ici_lock); delay(1); XFS_STATS_INC(xs_ig_frecycle); @@ -190,7 +169,7 @@ again: */ cmn_err(CE_PANIC, "xfs_iget_core: ambiguous vns: vp/0x%p, invp/0x%p", - inode_vp, vp); + old_inode, inode); } /* @@ -200,20 +179,16 @@ again: XFS_STATS_INC(xs_ig_found); finish_inode: - if (ip->i_d.di_mode == 0) { - if (!(flags & XFS_IGET_CREATE)) { - xfs_put_perag(mp, pag); - return ENOENT; - } - xfs_iocore_inode_reinit(ip); + if (ip->i_d.di_mode == 0 && !(flags & XFS_IGET_CREATE)) { + xfs_put_perag(mp, pag); + return ENOENT; } if (lock_flags != 0) xfs_ilock(ip, lock_flags); xfs_iflags_clear(ip, XFS_ISTALE); - vn_trace_exit(ip, "xfs_iget.found", - (inst_t *)__return_address); + xfs_itrace_exit_tag(ip, "xfs_iget.found"); goto return_ip; } @@ -234,10 +209,16 @@ finish_inode: return error; } - vn_trace_exit(ip, "xfs_iget.alloc", (inst_t *)__return_address); + xfs_itrace_exit_tag(ip, "xfs_iget.alloc"); + + + mrlock_init(&ip->i_lock, MRLOCK_ALLOW_EQUAL_PRI|MRLOCK_BARRIER, + "xfsino", ip->i_ino); + mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino); + init_waitqueue_head(&ip->i_ipin_wait); + atomic_set(&ip->i_pincount, 0); + initnsema(&ip->i_flock, 1, "xfsfino"); - xfs_inode_lock_init(ip, vp); - xfs_iocore_inode_init(ip); if (lock_flags) xfs_ilock(ip, lock_flags); @@ -254,6 +235,7 @@ finish_inode: */ new_icl = kmem_zone_alloc(xfs_icluster_zone, KM_SLEEP); if (radix_tree_preload(GFP_KERNEL)) { + xfs_idestroy(ip); delay(1); goto again; } @@ -333,9 +315,6 @@ finish_inode: ASSERT(ip->i_df.if_ext_max == XFS_IFORK_DSIZE(ip) / sizeof(xfs_bmbt_rec_t)); - ASSERT(((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) != 0) == - ((ip->i_iocore.io_flags & XFS_IOCORE_RT) != 0)); - xfs_iflags_set(ip, XFS_IMODIFIED); *ipp = ip; @@ -343,7 +322,7 @@ finish_inode: * If we have a real type for an on-disk inode, we can set ops(&unlock) * now. If it's a new inode being created, xfs_ialloc will handle it. */ - xfs_initialize_vnode(mp, vp, ip); + xfs_initialize_vnode(mp, inode, ip); return 0; } @@ -363,69 +342,58 @@ xfs_iget( xfs_daddr_t bno) { struct inode *inode; - bhv_vnode_t *vp = NULL; + xfs_inode_t *ip; int error; XFS_STATS_INC(xs_ig_attempts); retry: inode = iget_locked(mp->m_super, ino); - if (inode) { - xfs_inode_t *ip; - - vp = vn_from_inode(inode); - if (inode->i_state & I_NEW) { - vn_initialize(inode); - error = xfs_iget_core(vp, mp, tp, ino, flags, - lock_flags, ipp, bno); - if (error) { - vn_mark_bad(vp); - if (inode->i_state & I_NEW) - unlock_new_inode(inode); - iput(inode); - } - } else { - /* - * If the inode is not fully constructed due to - * filehandle mismatches wait for the inode to go - * away and try again. - * - * iget_locked will call __wait_on_freeing_inode - * to wait for the inode to go away. - */ - if (is_bad_inode(inode) || - ((ip = xfs_vtoi(vp)) == NULL)) { - iput(inode); - delay(1); - goto retry; - } - - if (lock_flags != 0) - xfs_ilock(ip, lock_flags); - XFS_STATS_INC(xs_ig_found); - *ipp = ip; - error = 0; + if (!inode) + /* If we got no inode we are out of memory */ + return ENOMEM; + + if (inode->i_state & I_NEW) { + XFS_STATS_INC(vn_active); + XFS_STATS_INC(vn_alloc); + + error = xfs_iget_core(inode, mp, tp, ino, flags, + lock_flags, ipp, bno); + if (error) { + make_bad_inode(inode); + if (inode->i_state & I_NEW) + unlock_new_inode(inode); + iput(inode); } - } else - error = ENOMEM; /* If we got no inode we are out of memory */ + return error; + } - return error; -} + /* + * If the inode is not fully constructed due to + * filehandle mismatches wait for the inode to go + * away and try again. + * + * iget_locked will call __wait_on_freeing_inode + * to wait for the inode to go away. + */ + if (is_bad_inode(inode)) { + iput(inode); + delay(1); + goto retry; + } -/* - * Do the setup for the various locks within the incore inode. - */ -void -xfs_inode_lock_init( - xfs_inode_t *ip, - bhv_vnode_t *vp) -{ - mrlock_init(&ip->i_lock, MRLOCK_ALLOW_EQUAL_PRI|MRLOCK_BARRIER, - "xfsino", ip->i_ino); - mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino); - init_waitqueue_head(&ip->i_ipin_wait); - atomic_set(&ip->i_pincount, 0); - initnsema(&ip->i_flock, 1, "xfsfino"); + ip = XFS_I(inode); + if (!ip) { + iput(inode); + delay(1); + goto retry; + } + + if (lock_flags != 0) + xfs_ilock(ip, lock_flags); + XFS_STATS_INC(xs_ig_found); + *ipp = ip; + return 0; } /* @@ -465,11 +433,9 @@ void xfs_iput(xfs_inode_t *ip, uint lock_flags) { - bhv_vnode_t *vp = XFS_ITOV(ip); - - vn_trace_entry(ip, "xfs_iput", (inst_t *)__return_address); + xfs_itrace_entry(ip); xfs_iunlock(ip, lock_flags); - VN_RELE(vp); + IRELE(ip); } /* @@ -479,20 +445,19 @@ void xfs_iput_new(xfs_inode_t *ip, uint lock_flags) { - bhv_vnode_t *vp = XFS_ITOV(ip); - struct inode *inode = vn_to_inode(vp); + struct inode *inode = ip->i_vnode; - vn_trace_entry(ip, "xfs_iput_new", (inst_t *)__return_address); + xfs_itrace_entry(ip); if ((ip->i_d.di_mode == 0)) { ASSERT(!xfs_iflags_test(ip, XFS_IRECLAIMABLE)); - vn_mark_bad(vp); + make_bad_inode(inode); } if (inode->i_state & I_NEW) unlock_new_inode(inode); if (lock_flags) xfs_iunlock(ip, lock_flags); - VN_RELE(vp); + IRELE(ip); } @@ -505,8 +470,6 @@ xfs_iput_new(xfs_inode_t *ip, void xfs_ireclaim(xfs_inode_t *ip) { - bhv_vnode_t *vp; - /* * Remove from old hash list and mount list. */ @@ -535,9 +498,8 @@ xfs_ireclaim(xfs_inode_t *ip) /* * Pull our behavior descriptor from the vnode chain. */ - vp = XFS_ITOV_NULL(ip); - if (vp) { - vn_to_inode(vp)->i_private = NULL; + if (ip->i_vnode) { + ip->i_vnode->i_private = NULL; ip->i_vnode = NULL; } diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 34494808281..f43a6e01d68 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -15,6 +15,8 @@ * along with this program; if not, write the Free Software Foundation, * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ +#include <linux/log2.h> + #include "xfs.h" #include "xfs_fs.h" #include "xfs_types.h" @@ -826,15 +828,17 @@ xfs_ip2xflags( xfs_icdinode_t *dic = &ip->i_d; return _xfs_dic2xflags(dic->di_flags) | - (XFS_CFORK_Q(dic) ? XFS_XFLAG_HASATTR : 0); + (XFS_IFORK_Q(ip) ? XFS_XFLAG_HASATTR : 0); } uint xfs_dic2xflags( - xfs_dinode_core_t *dic) + xfs_dinode_t *dip) { + xfs_dinode_core_t *dic = &dip->di_core; + return _xfs_dic2xflags(be16_to_cpu(dic->di_flags)) | - (XFS_CFORK_Q_DISK(dic) ? XFS_XFLAG_HASATTR : 0); + (XFS_DFORK_Q(dip) ? XFS_XFLAG_HASATTR : 0); } /* @@ -884,8 +888,8 @@ xfs_iread( * Initialize inode's trace buffers. * Do this before xfs_iformat in case it adds entries. */ -#ifdef XFS_VNODE_TRACE - ip->i_trace = ktrace_alloc(VNODE_TRACE_SIZE, KM_SLEEP); +#ifdef XFS_INODE_TRACE + ip->i_trace = ktrace_alloc(INODE_TRACE_SIZE, KM_SLEEP); #endif #ifdef XFS_BMAP_TRACE ip->i_xtrace = ktrace_alloc(XFS_BMAP_KTRACE_SIZE, KM_SLEEP); @@ -1143,7 +1147,7 @@ xfs_ialloc( * the inode version number now. This way we only do the conversion * here rather than here and in the flush/logging code. */ - if (XFS_SB_VERSION_HASNLINK(&tp->t_mountp->m_sb) && + if (xfs_sb_version_hasnlink(&tp->t_mountp->m_sb) && ip->i_d.di_version == XFS_DINODE_VERSION_1) { ip->i_d.di_version = XFS_DINODE_VERSION_2; /* @@ -1220,10 +1224,8 @@ xfs_ialloc( ip->i_d.di_extsize = pip->i_d.di_extsize; } } else if ((mode & S_IFMT) == S_IFREG) { - if (pip->i_d.di_flags & XFS_DIFLAG_RTINHERIT) { + if (pip->i_d.di_flags & XFS_DIFLAG_RTINHERIT) di_flags |= XFS_DIFLAG_REALTIME; - ip->i_iocore.io_flags |= XFS_IOCORE_RT; - } if (pip->i_d.di_flags & XFS_DIFLAG_EXTSZINHERIT) { di_flags |= XFS_DIFLAG_EXTSIZE; ip->i_d.di_extsize = pip->i_d.di_extsize; @@ -1298,7 +1300,10 @@ xfs_isize_check( if ((ip->i_d.di_mode & S_IFMT) != S_IFREG) return; - if (ip->i_d.di_flags & (XFS_DIFLAG_REALTIME | XFS_DIFLAG_EXTSIZE)) + if (XFS_IS_REALTIME_INODE(ip)) + return; + + if (ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE) return; nimaps = 2; @@ -1711,7 +1716,7 @@ xfs_itruncate_finish( * runs. */ XFS_BMAP_INIT(&free_list, &first_block); - error = XFS_BUNMAPI(mp, ntp, &ip->i_iocore, + error = xfs_bunmapi(ntp, ip, first_unmap_block, unmap_len, XFS_BMAPI_AFLAG(fork) | (sync ? 0 : XFS_BMAPI_ASYNC), @@ -1844,8 +1849,6 @@ xfs_igrow_start( xfs_fsize_t new_size, cred_t *credp) { - int error; - ASSERT(ismrlocked(&(ip->i_lock), MR_UPDATE) != 0); ASSERT(ismrlocked(&(ip->i_iolock), MR_UPDATE) != 0); ASSERT(new_size > ip->i_size); @@ -1855,9 +1858,7 @@ xfs_igrow_start( * xfs_write_file() beyond the end of the file * and any blocks between the old and new file sizes. */ - error = xfs_zero_eof(XFS_ITOV(ip), &ip->i_iocore, new_size, - ip->i_size); - return error; + return xfs_zero_eof(ip, new_size, ip->i_size); } /* @@ -1959,24 +1960,6 @@ xfs_iunlink( ASSERT(agi->agi_unlinked[bucket_index]); ASSERT(be32_to_cpu(agi->agi_unlinked[bucket_index]) != agino); - error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0, 0); - if (error) - return error; - - /* - * Clear the on-disk di_nlink. This is to prevent xfs_bulkstat - * from picking up this inode when it is reclaimed (its incore state - * initialzed but not flushed to disk yet). The in-core di_nlink is - * already cleared in xfs_droplink() and a corresponding transaction - * logged. The hack here just synchronizes the in-core to on-disk - * di_nlink value in advance before the actual inode sync to disk. - * This is OK because the inode is already unlinked and would never - * change its di_nlink again for this inode generation. - * This is a temporary hack that would require a proper fix - * in the future. - */ - dip->di_core.di_nlink = 0; - if (be32_to_cpu(agi->agi_unlinked[bucket_index]) != NULLAGINO) { /* * There is already another inode in the bucket we need @@ -1984,6 +1967,10 @@ xfs_iunlink( * Here we put the head pointer into our next pointer, * and then we fall through to point the head at us. */ + error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0, 0); + if (error) + return error; + ASSERT(be32_to_cpu(dip->di_next_unlinked) == NULLAGINO); /* both on-disk, don't endian flip twice */ dip->di_next_unlinked = agi->agi_unlinked[bucket_index]; @@ -2209,7 +2196,6 @@ xfs_ifree_cluster( xfs_inode_log_item_t *iip; xfs_log_item_t *lip; xfs_perag_t *pag = xfs_get_perag(mp, inum); - SPLDECL(s); if (mp->m_sb.sb_blocksize >= XFS_INODE_CLUSTER_SIZE(mp)) { blks_per_cluster = 1; @@ -2311,9 +2297,9 @@ xfs_ifree_cluster( iip = (xfs_inode_log_item_t *)lip; ASSERT(iip->ili_logged == 1); lip->li_cb = (void(*)(xfs_buf_t*,xfs_log_item_t*)) xfs_istale_done; - AIL_LOCK(mp,s); + spin_lock(&mp->m_ail_lock); iip->ili_flush_lsn = iip->ili_item.li_lsn; - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); xfs_iflags_set(iip->ili_inode, XFS_ISTALE); pre_flushed++; } @@ -2334,9 +2320,9 @@ xfs_ifree_cluster( iip->ili_last_fields = iip->ili_format.ilf_fields; iip->ili_format.ilf_fields = 0; iip->ili_logged = 1; - AIL_LOCK(mp,s); + spin_lock(&mp->m_ail_lock); iip->ili_flush_lsn = iip->ili_item.li_lsn; - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); xfs_buf_attach_iodone(bp, (void(*)(xfs_buf_t*,xfs_log_item_t*)) @@ -2374,6 +2360,8 @@ xfs_ifree( int error; int delete; xfs_ino_t first_ino; + xfs_dinode_t *dip; + xfs_buf_t *ibp; ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE)); ASSERT(ip->i_transp == tp); @@ -2409,8 +2397,27 @@ xfs_ifree( * by reincarnations of this inode. */ ip->i_d.di_gen++; + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + error = xfs_itobp(ip->i_mount, tp, ip, &dip, &ibp, 0, 0); + if (error) + return error; + + /* + * Clear the on-disk di_mode. This is to prevent xfs_bulkstat + * from picking up this inode when it is reclaimed (its incore state + * initialzed but not flushed to disk yet). The in-core di_mode is + * already cleared and a corresponding transaction logged. + * The hack here just synchronizes the in-core to on-disk + * di_mode value in advance before the actual inode sync to disk. + * This is OK because the inode is already unlinked and would never + * change its di_mode again for this inode generation. + * This is a temporary hack that would require a proper fix + * in the future. + */ + dip->di_core.di_mode = 0; + if (delete) { xfs_ifree_cluster(ip, tp, first_ino); } @@ -2735,7 +2742,6 @@ void xfs_idestroy( xfs_inode_t *ip) { - switch (ip->i_d.di_mode & S_IFMT) { case S_IFREG: case S_IFDIR: @@ -2749,7 +2755,7 @@ xfs_idestroy( mrfree(&ip->i_iolock); freesema(&ip->i_flock); -#ifdef XFS_VNODE_TRACE +#ifdef XFS_INODE_TRACE ktrace_free(ip->i_trace); #endif #ifdef XFS_BMAP_TRACE @@ -2775,16 +2781,15 @@ xfs_idestroy( */ xfs_mount_t *mp = ip->i_mount; xfs_log_item_t *lip = &ip->i_itemp->ili_item; - int s; ASSERT(((lip->li_flags & XFS_LI_IN_AIL) == 0) || XFS_FORCED_SHUTDOWN(ip->i_mount)); if (lip->li_flags & XFS_LI_IN_AIL) { - AIL_LOCK(mp, s); + spin_lock(&mp->m_ail_lock); if (lip->li_flags & XFS_LI_IN_AIL) - xfs_trans_delete_ail(mp, lip, s); + xfs_trans_delete_ail(mp, lip); else - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); } xfs_inode_item_destroy(ip); } @@ -2816,40 +2821,8 @@ xfs_iunpin( { ASSERT(atomic_read(&ip->i_pincount) > 0); - if (atomic_dec_and_lock(&ip->i_pincount, &ip->i_flags_lock)) { - - /* - * If the inode is currently being reclaimed, the link between - * the bhv_vnode and the xfs_inode will be broken after the - * XFS_IRECLAIM* flag is set. Hence, if these flags are not - * set, then we can move forward and mark the linux inode dirty - * knowing that it is still valid as it won't freed until after - * the bhv_vnode<->xfs_inode link is broken in xfs_reclaim. The - * i_flags_lock is used to synchronise the setting of the - * XFS_IRECLAIM* flags and the breaking of the link, and so we - * can execute atomically w.r.t to reclaim by holding this lock - * here. - * - * However, we still need to issue the unpin wakeup call as the - * inode reclaim may be blocked waiting for the inode to become - * unpinned. - */ - - if (!__xfs_iflags_test(ip, XFS_IRECLAIM|XFS_IRECLAIMABLE)) { - bhv_vnode_t *vp = XFS_ITOV_NULL(ip); - struct inode *inode = NULL; - - BUG_ON(vp == NULL); - inode = vn_to_inode(vp); - BUG_ON(inode->i_state & I_CLEAR); - - /* make sync come back and flush this inode */ - if (!(inode->i_state & (I_NEW|I_FREEING))) - mark_inode_dirty_sync(inode); - } - spin_unlock(&ip->i_flags_lock); + if (atomic_dec_and_test(&ip->i_pincount)) wake_up(&ip->i_ipin_wait); - } } /* @@ -3338,7 +3311,6 @@ xfs_iflush_int( #ifdef XFS_TRANS_DEBUG int first; #endif - SPLDECL(s); ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE|MR_ACCESS)); ASSERT(issemalocked(&(ip->i_flock))); @@ -3462,9 +3434,9 @@ xfs_iflush_int( * has been updated, then make the conversion permanent. */ ASSERT(ip->i_d.di_version == XFS_DINODE_VERSION_1 || - XFS_SB_VERSION_HASNLINK(&mp->m_sb)); + xfs_sb_version_hasnlink(&mp->m_sb)); if (ip->i_d.di_version == XFS_DINODE_VERSION_1) { - if (!XFS_SB_VERSION_HASNLINK(&mp->m_sb)) { + if (!xfs_sb_version_hasnlink(&mp->m_sb)) { /* * Convert it back. */ @@ -3533,9 +3505,9 @@ xfs_iflush_int( iip->ili_logged = 1; ASSERT(sizeof(xfs_lsn_t) == 8); /* don't lock if it shrinks */ - AIL_LOCK(mp,s); + spin_lock(&mp->m_ail_lock); iip->ili_flush_lsn = iip->ili_item.li_lsn; - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); /* * Attach the function xfs_iflush_done to the inode's @@ -3611,95 +3583,6 @@ xfs_iflush_all( XFS_MOUNT_IUNLOCK(mp); } -/* - * xfs_iaccess: check accessibility of inode for mode. - */ -int -xfs_iaccess( - xfs_inode_t *ip, - mode_t mode, - cred_t *cr) -{ - int error; - mode_t orgmode = mode; - struct inode *inode = vn_to_inode(XFS_ITOV(ip)); - - if (mode & S_IWUSR) { - umode_t imode = inode->i_mode; - - if (IS_RDONLY(inode) && - (S_ISREG(imode) || S_ISDIR(imode) || S_ISLNK(imode))) - return XFS_ERROR(EROFS); - - if (IS_IMMUTABLE(inode)) - return XFS_ERROR(EACCES); - } - - /* - * If there's an Access Control List it's used instead of - * the mode bits. - */ - if ((error = _ACL_XFS_IACCESS(ip, mode, cr)) != -1) - return error ? XFS_ERROR(error) : 0; - - if (current_fsuid(cr) != ip->i_d.di_uid) { - mode >>= 3; - if (!in_group_p((gid_t)ip->i_d.di_gid)) - mode >>= 3; - } - - /* - * If the DACs are ok we don't need any capability check. - */ - if ((ip->i_d.di_mode & mode) == mode) - return 0; - /* - * Read/write DACs are always overridable. - * Executable DACs are overridable if at least one exec bit is set. - */ - if (!(orgmode & S_IXUSR) || - (inode->i_mode & S_IXUGO) || S_ISDIR(inode->i_mode)) - if (capable_cred(cr, CAP_DAC_OVERRIDE)) - return 0; - - if ((orgmode == S_IRUSR) || - (S_ISDIR(inode->i_mode) && (!(orgmode & S_IWUSR)))) { - if (capable_cred(cr, CAP_DAC_READ_SEARCH)) - return 0; -#ifdef NOISE - cmn_err(CE_NOTE, "Ick: mode=%o, orgmode=%o", mode, orgmode); -#endif /* NOISE */ - return XFS_ERROR(EACCES); - } - return XFS_ERROR(EACCES); -} - -/* - * xfs_iroundup: round up argument to next power of two - */ -uint -xfs_iroundup( - uint v) -{ - int i; - uint m; - - if ((v & (v - 1)) == 0) - return v; - ASSERT((v & 0x80000000) == 0); - if ((v & (v + 1)) == 0) - return v + 1; - for (i = 0, m = 1; i < 31; i++, m <<= 1) { - if (v & m) - continue; - v |= m; - if ((v & (v + 1)) == 0) - return v + 1; - } - ASSERT(0); - return( 0 ); -} - #ifdef XFS_ILOCK_TRACE ktrace_t *xfs_ilock_trace_buf; @@ -4206,7 +4089,7 @@ xfs_iext_realloc_direct( return; } if (!is_power_of_2(new_size)){ - rnew_size = xfs_iroundup(new_size); + rnew_size = roundup_pow_of_two(new_size); } if (rnew_size != ifp->if_real_bytes) { ifp->if_u1.if_extents = @@ -4229,7 +4112,7 @@ xfs_iext_realloc_direct( else { new_size += ifp->if_bytes; if (!is_power_of_2(new_size)) { - rnew_size = xfs_iroundup(new_size); + rnew_size = roundup_pow_of_two(new_size); } xfs_iext_inline_to_direct(ifp, rnew_size); } diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index e5aff929cc6..bfcd72cbaee 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -132,45 +132,6 @@ typedef struct dm_attrs_s { __uint16_t da_pad; /* DMIG extra padding */ } dm_attrs_t; -typedef struct xfs_iocore { - void *io_obj; /* pointer to container - * inode or dcxvn structure */ - struct xfs_mount *io_mount; /* fs mount struct ptr */ -#ifdef DEBUG - mrlock_t *io_lock; /* inode IO lock */ - mrlock_t *io_iolock; /* inode IO lock */ -#endif - - /* I/O state */ - xfs_fsize_t io_new_size; /* sz when write completes */ - - /* Miscellaneous state. */ - unsigned int io_flags; /* IO related flags */ - - /* DMAPI state */ - dm_attrs_t io_dmattrs; - -} xfs_iocore_t; - -#define io_dmevmask io_dmattrs.da_dmevmask -#define io_dmstate io_dmattrs.da_dmstate - -#define XFS_IO_INODE(io) ((xfs_inode_t *) ((io)->io_obj)) -#define XFS_IO_DCXVN(io) ((dcxvn_t *) ((io)->io_obj)) - -/* - * Flags in the flags field - */ - -#define XFS_IOCORE_RT 0x1 - -/* - * xfs_iocore prototypes - */ - -extern void xfs_iocore_inode_init(struct xfs_inode *); -extern void xfs_iocore_inode_reinit(struct xfs_inode *); - /* * This is the xfs inode cluster structure. This structure is used by * xfs_iflush to find inodes that share a cluster and can be flushed to disk at @@ -181,7 +142,7 @@ typedef struct xfs_icluster { xfs_daddr_t icl_blkno; /* starting block number of * the cluster */ struct xfs_buf *icl_buf; /* the inode buffer */ - lock_t icl_lock; /* inode list lock */ + spinlock_t icl_lock; /* inode list lock */ } xfs_icluster_t; /* @@ -283,9 +244,6 @@ typedef struct xfs_inode { struct xfs_inode **i_refcache; /* ptr to entry in ref cache */ struct xfs_inode *i_release; /* inode to unref */ #endif - /* I/O state */ - xfs_iocore_t i_iocore; /* I/O core */ - /* Miscellaneous state. */ unsigned short i_flags; /* see defined flags below */ unsigned char i_update_core; /* timestamps/size is dirty */ @@ -298,9 +256,10 @@ typedef struct xfs_inode { struct hlist_node i_cnode; /* cluster link node */ xfs_fsize_t i_size; /* in-memory size */ + xfs_fsize_t i_new_size; /* size when write completes */ atomic_t i_iocount; /* outstanding I/O count */ /* Trace buffers per inode. */ -#ifdef XFS_VNODE_TRACE +#ifdef XFS_INODE_TRACE struct ktrace *i_trace; /* general inode trace */ #endif #ifdef XFS_BMAP_TRACE @@ -382,17 +341,42 @@ xfs_iflags_test_and_clear(xfs_inode_t *ip, unsigned short flags) /* * Fork handling. */ -#define XFS_IFORK_PTR(ip,w) \ - ((w) == XFS_DATA_FORK ? &(ip)->i_df : (ip)->i_afp) -#define XFS_IFORK_Q(ip) XFS_CFORK_Q(&(ip)->i_d) -#define XFS_IFORK_DSIZE(ip) XFS_CFORK_DSIZE(&ip->i_d, ip->i_mount) -#define XFS_IFORK_ASIZE(ip) XFS_CFORK_ASIZE(&ip->i_d, ip->i_mount) -#define XFS_IFORK_SIZE(ip,w) XFS_CFORK_SIZE(&ip->i_d, ip->i_mount, w) -#define XFS_IFORK_FORMAT(ip,w) XFS_CFORK_FORMAT(&ip->i_d, w) -#define XFS_IFORK_FMT_SET(ip,w,n) XFS_CFORK_FMT_SET(&ip->i_d, w, n) -#define XFS_IFORK_NEXTENTS(ip,w) XFS_CFORK_NEXTENTS(&ip->i_d, w) -#define XFS_IFORK_NEXT_SET(ip,w,n) XFS_CFORK_NEXT_SET(&ip->i_d, w, n) +#define XFS_IFORK_Q(ip) ((ip)->i_d.di_forkoff != 0) +#define XFS_IFORK_BOFF(ip) ((int)((ip)->i_d.di_forkoff << 3)) + +#define XFS_IFORK_PTR(ip,w) \ + ((w) == XFS_DATA_FORK ? \ + &(ip)->i_df : \ + (ip)->i_afp) +#define XFS_IFORK_DSIZE(ip) \ + (XFS_IFORK_Q(ip) ? \ + XFS_IFORK_BOFF(ip) : \ + XFS_LITINO((ip)->i_mount)) +#define XFS_IFORK_ASIZE(ip) \ + (XFS_IFORK_Q(ip) ? \ + XFS_LITINO((ip)->i_mount) - XFS_IFORK_BOFF(ip) : \ + 0) +#define XFS_IFORK_SIZE(ip,w) \ + ((w) == XFS_DATA_FORK ? \ + XFS_IFORK_DSIZE(ip) : \ + XFS_IFORK_ASIZE(ip)) +#define XFS_IFORK_FORMAT(ip,w) \ + ((w) == XFS_DATA_FORK ? \ + (ip)->i_d.di_format : \ + (ip)->i_d.di_aformat) +#define XFS_IFORK_FMT_SET(ip,w,n) \ + ((w) == XFS_DATA_FORK ? \ + ((ip)->i_d.di_format = (n)) : \ + ((ip)->i_d.di_aformat = (n))) +#define XFS_IFORK_NEXTENTS(ip,w) \ + ((w) == XFS_DATA_FORK ? \ + (ip)->i_d.di_nextents : \ + (ip)->i_d.di_anextents) +#define XFS_IFORK_NEXT_SET(ip,w,n) \ + ((w) == XFS_DATA_FORK ? \ + ((ip)->i_d.di_nextents = (n)) : \ + ((ip)->i_d.di_anextents = (n))) #ifdef __KERNEL__ @@ -509,7 +493,6 @@ void xfs_ihash_init(struct xfs_mount *); void xfs_ihash_free(struct xfs_mount *); xfs_inode_t *xfs_inode_incore(struct xfs_mount *, xfs_ino_t, struct xfs_trans *); -void xfs_inode_lock_init(xfs_inode_t *, bhv_vnode_t *); int xfs_iget(struct xfs_mount *, struct xfs_trans *, xfs_ino_t, uint, uint, xfs_inode_t **, xfs_daddr_t); void xfs_iput(xfs_inode_t *, uint); @@ -545,7 +528,7 @@ void xfs_dinode_to_disk(struct xfs_dinode_core *, struct xfs_icdinode *); uint xfs_ip2xflags(struct xfs_inode *); -uint xfs_dic2xflags(struct xfs_dinode_core *); +uint xfs_dic2xflags(struct xfs_dinode *); int xfs_ifree(struct xfs_trans *, xfs_inode_t *, struct xfs_bmap_free *); int xfs_itruncate_start(xfs_inode_t *, uint, xfs_fsize_t); @@ -567,13 +550,12 @@ void xfs_iunpin(xfs_inode_t *); int xfs_iextents_copy(xfs_inode_t *, xfs_bmbt_rec_t *, int); int xfs_iflush(xfs_inode_t *, uint); void xfs_iflush_all(struct xfs_mount *); -int xfs_iaccess(xfs_inode_t *, mode_t, cred_t *); -uint xfs_iroundup(uint); void xfs_ichgtime(xfs_inode_t *, int); xfs_fsize_t xfs_file_last_byte(xfs_inode_t *); void xfs_lock_inodes(xfs_inode_t **, int, int, uint); void xfs_synchronize_atime(xfs_inode_t *); +void xfs_mark_inode_dirty_sync(xfs_inode_t *); xfs_bmbt_rec_host_t *xfs_iext_get_ext(xfs_ifork_t *, xfs_extnum_t); void xfs_iext_insert(xfs_ifork_t *, xfs_extnum_t, xfs_extnum_t, diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index 565d470a6b4..2c775b4ae9e 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -274,6 +274,11 @@ xfs_inode_item_format( */ xfs_synchronize_atime(ip); + /* + * make sure the linux inode is dirty + */ + xfs_mark_inode_dirty_sync(ip); + vecp->i_addr = (xfs_caddr_t)&ip->i_d; vecp->i_len = sizeof(xfs_dinode_core_t); XLOG_VEC_SET_TYPE(vecp, XLOG_REG_TYPE_ICORE); @@ -291,9 +296,9 @@ xfs_inode_item_format( */ mp = ip->i_mount; ASSERT(ip->i_d.di_version == XFS_DINODE_VERSION_1 || - XFS_SB_VERSION_HASNLINK(&mp->m_sb)); + xfs_sb_version_hasnlink(&mp->m_sb)); if (ip->i_d.di_version == XFS_DINODE_VERSION_1) { - if (!XFS_SB_VERSION_HASNLINK(&mp->m_sb)) { + if (!xfs_sb_version_hasnlink(&mp->m_sb)) { /* * Convert it back. */ @@ -615,7 +620,7 @@ xfs_inode_item_trylock( return XFS_ITEM_PUSHBUF; } else { /* - * We hold the AIL_LOCK, so we must specify the + * We hold the AIL lock, so we must specify the * NONOTIFY flag so that we won't double trip. */ xfs_iunlock(ip, XFS_ILOCK_SHARED|XFS_IUNLOCK_NONOTIFY); @@ -749,7 +754,7 @@ xfs_inode_item_committed( * marked delayed write. If that's the case, we'll initiate a bawrite on that * buffer to expedite the process. * - * We aren't holding the AIL_LOCK (or the flush lock) when this gets called, + * We aren't holding the AIL lock (or the flush lock) when this gets called, * so it is inherently race-y. */ STATIC void @@ -792,7 +797,7 @@ xfs_inode_item_pushbuf( if (XFS_BUF_ISDELAYWRITE(bp)) { /* * We were racing with iflush because we don't hold - * the AIL_LOCK or the flush lock. However, at this point, + * the AIL lock or the flush lock. However, at this point, * we have the buffer, and we know that it's dirty. * So, it's possible that iflush raced with us, and * this item is already taken off the AIL. @@ -968,7 +973,6 @@ xfs_iflush_done( xfs_inode_log_item_t *iip) { xfs_inode_t *ip; - SPLDECL(s); ip = iip->ili_inode; @@ -983,15 +987,15 @@ xfs_iflush_done( */ if (iip->ili_logged && (iip->ili_item.li_lsn == iip->ili_flush_lsn)) { - AIL_LOCK(ip->i_mount, s); + spin_lock(&ip->i_mount->m_ail_lock); if (iip->ili_item.li_lsn == iip->ili_flush_lsn) { /* * xfs_trans_delete_ail() drops the AIL lock. */ xfs_trans_delete_ail(ip->i_mount, - (xfs_log_item_t*)iip, s); + (xfs_log_item_t*)iip); } else { - AIL_UNLOCK(ip->i_mount, s); + spin_unlock(&ip->i_mount->m_ail_lock); } } @@ -1025,21 +1029,19 @@ xfs_iflush_abort( { xfs_inode_log_item_t *iip; xfs_mount_t *mp; - SPLDECL(s); iip = ip->i_itemp; mp = ip->i_mount; if (iip) { if (iip->ili_item.li_flags & XFS_LI_IN_AIL) { - AIL_LOCK(mp, s); + spin_lock(&mp->m_ail_lock); if (iip->ili_item.li_flags & XFS_LI_IN_AIL) { /* * xfs_trans_delete_ail() drops the AIL lock. */ - xfs_trans_delete_ail(mp, (xfs_log_item_t *)iip, - s); + xfs_trans_delete_ail(mp, (xfs_log_item_t *)iip); } else - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); } iip->ili_logged = 0; /* diff --git a/fs/xfs/xfs_iocore.c b/fs/xfs/xfs_iocore.c deleted file mode 100644 index b27b5d5be84..00000000000 --- a/fs/xfs/xfs_iocore.c +++ /dev/null @@ -1,119 +0,0 @@ -/* - * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_types.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_dir2.h" -#include "xfs_dfrag.h" -#include "xfs_dmapi.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_alloc_btree.h" -#include "xfs_ialloc_btree.h" -#include "xfs_dir2_sf.h" -#include "xfs_attr_sf.h" -#include "xfs_dinode.h" -#include "xfs_inode.h" -#include "xfs_inode_item.h" -#include "xfs_itable.h" -#include "xfs_btree.h" -#include "xfs_alloc.h" -#include "xfs_ialloc.h" -#include "xfs_bmap.h" -#include "xfs_error.h" -#include "xfs_rw.h" -#include "xfs_quota.h" -#include "xfs_trans_space.h" -#include "xfs_iomap.h" - - -STATIC xfs_fsize_t -xfs_size_fn( - xfs_inode_t *ip) -{ - return XFS_ISIZE(ip); -} - -STATIC int -xfs_ioinit( - struct xfs_mount *mp, - struct xfs_mount_args *mntargs, - int flags) -{ - return xfs_mountfs(mp, flags); -} - -xfs_ioops_t xfs_iocore_xfs = { - .xfs_ioinit = (xfs_ioinit_t) xfs_ioinit, - .xfs_bmapi_func = (xfs_bmapi_t) xfs_bmapi, - .xfs_bunmapi_func = (xfs_bunmapi_t) xfs_bunmapi, - .xfs_bmap_eof_func = (xfs_bmap_eof_t) xfs_bmap_eof, - .xfs_iomap_write_direct = - (xfs_iomap_write_direct_t) xfs_iomap_write_direct, - .xfs_iomap_write_delay = - (xfs_iomap_write_delay_t) xfs_iomap_write_delay, - .xfs_iomap_write_allocate = - (xfs_iomap_write_allocate_t) xfs_iomap_write_allocate, - .xfs_iomap_write_unwritten = - (xfs_iomap_write_unwritten_t) xfs_iomap_write_unwritten, - .xfs_ilock = (xfs_lock_t) xfs_ilock, - .xfs_lck_map_shared = (xfs_lck_map_shared_t) xfs_ilock_map_shared, - .xfs_ilock_demote = (xfs_lock_demote_t) xfs_ilock_demote, - .xfs_ilock_nowait = (xfs_lock_nowait_t) xfs_ilock_nowait, - .xfs_unlock = (xfs_unlk_t) xfs_iunlock, - .xfs_size_func = (xfs_size_t) xfs_size_fn, - .xfs_iodone = (xfs_iodone_t) fs_noerr, - .xfs_swap_extents_func = (xfs_swap_extents_t) xfs_swap_extents, -}; - -void -xfs_iocore_inode_reinit( - xfs_inode_t *ip) -{ - xfs_iocore_t *io = &ip->i_iocore; - - io->io_flags = 0; - if (ip->i_d.di_flags & XFS_DIFLAG_REALTIME) - io->io_flags |= XFS_IOCORE_RT; - io->io_dmevmask = ip->i_d.di_dmevmask; - io->io_dmstate = ip->i_d.di_dmstate; -} - -void -xfs_iocore_inode_init( - xfs_inode_t *ip) -{ - xfs_iocore_t *io = &ip->i_iocore; - xfs_mount_t *mp = ip->i_mount; - - io->io_mount = mp; -#ifdef DEBUG - io->io_lock = &ip->i_lock; - io->io_iolock = &ip->i_iolock; -#endif - - io->io_obj = (void *)ip; - - xfs_iocore_inode_reinit(ip); -} diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 72786e356d5..fde37f87d52 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -53,12 +53,10 @@ void xfs_iomap_enter_trace( int tag, - xfs_iocore_t *io, + xfs_inode_t *ip, xfs_off_t offset, ssize_t count) { - xfs_inode_t *ip = XFS_IO_INODE(io); - if (!ip->i_rwtrace) return; @@ -70,8 +68,8 @@ xfs_iomap_enter_trace( (void *)((unsigned long)((offset >> 32) & 0xffffffff)), (void *)((unsigned long)(offset & 0xffffffff)), (void *)((unsigned long)count), - (void *)((unsigned long)((io->io_new_size >> 32) & 0xffffffff)), - (void *)((unsigned long)(io->io_new_size & 0xffffffff)), + (void *)((unsigned long)((ip->i_new_size >> 32) & 0xffffffff)), + (void *)((unsigned long)(ip->i_new_size & 0xffffffff)), (void *)((unsigned long)current_pid()), (void *)NULL, (void *)NULL, @@ -84,15 +82,13 @@ xfs_iomap_enter_trace( void xfs_iomap_map_trace( int tag, - xfs_iocore_t *io, + xfs_inode_t *ip, xfs_off_t offset, ssize_t count, xfs_iomap_t *iomapp, xfs_bmbt_irec_t *imapp, int flags) { - xfs_inode_t *ip = XFS_IO_INODE(io); - if (!ip->i_rwtrace) return; @@ -126,7 +122,7 @@ xfs_iomap_map_trace( STATIC int xfs_imap_to_bmap( - xfs_iocore_t *io, + xfs_inode_t *ip, xfs_off_t offset, xfs_bmbt_irec_t *imap, xfs_iomap_t *iomapp, @@ -134,11 +130,10 @@ xfs_imap_to_bmap( int iomaps, /* Number of iomap entries */ int flags) { - xfs_mount_t *mp; + xfs_mount_t *mp = ip->i_mount; int pbm; xfs_fsblock_t start_block; - mp = io->io_mount; for (pbm = 0; imaps && pbm < iomaps; imaps--, iomapp++, imap++, pbm++) { iomapp->iomap_offset = XFS_FSB_TO_B(mp, imap->br_startoff); @@ -146,7 +141,7 @@ xfs_imap_to_bmap( iomapp->iomap_bsize = XFS_FSB_TO_B(mp, imap->br_blockcount); iomapp->iomap_flags = flags; - if (io->io_flags & XFS_IOCORE_RT) { + if (XFS_IS_REALTIME_INODE(ip)) { iomapp->iomap_flags |= IOMAP_REALTIME; iomapp->iomap_target = mp->m_rtdev_targp; } else { @@ -160,7 +155,7 @@ xfs_imap_to_bmap( iomapp->iomap_bn = IOMAP_DADDR_NULL; iomapp->iomap_flags |= IOMAP_DELAY; } else { - iomapp->iomap_bn = XFS_FSB_TO_DB_IO(io, start_block); + iomapp->iomap_bn = XFS_FSB_TO_DB(ip, start_block); if (ISUNWRITTEN(imap)) iomapp->iomap_flags |= IOMAP_UNWRITTEN; } @@ -172,14 +167,14 @@ xfs_imap_to_bmap( int xfs_iomap( - xfs_iocore_t *io, + xfs_inode_t *ip, xfs_off_t offset, ssize_t count, int flags, xfs_iomap_t *iomapp, int *niomaps) { - xfs_mount_t *mp = io->io_mount; + xfs_mount_t *mp = ip->i_mount; xfs_fileoff_t offset_fsb, end_fsb; int error = 0; int lockmode = 0; @@ -188,45 +183,37 @@ xfs_iomap( int bmapi_flags = 0; int iomap_flags = 0; + ASSERT((ip->i_d.di_mode & S_IFMT) == S_IFREG); + if (XFS_FORCED_SHUTDOWN(mp)) return XFS_ERROR(EIO); - switch (flags & - (BMAPI_READ | BMAPI_WRITE | BMAPI_ALLOCATE | - BMAPI_UNWRITTEN | BMAPI_DEVICE)) { + switch (flags & (BMAPI_READ | BMAPI_WRITE | BMAPI_ALLOCATE)) { case BMAPI_READ: - xfs_iomap_enter_trace(XFS_IOMAP_READ_ENTER, io, offset, count); - lockmode = XFS_LCK_MAP_SHARED(mp, io); + xfs_iomap_enter_trace(XFS_IOMAP_READ_ENTER, ip, offset, count); + lockmode = xfs_ilock_map_shared(ip); bmapi_flags = XFS_BMAPI_ENTIRE; break; case BMAPI_WRITE: - xfs_iomap_enter_trace(XFS_IOMAP_WRITE_ENTER, io, offset, count); + xfs_iomap_enter_trace(XFS_IOMAP_WRITE_ENTER, ip, offset, count); lockmode = XFS_ILOCK_EXCL|XFS_EXTSIZE_WR; if (flags & BMAPI_IGNSTATE) bmapi_flags |= XFS_BMAPI_IGSTATE|XFS_BMAPI_ENTIRE; - XFS_ILOCK(mp, io, lockmode); + xfs_ilock(ip, lockmode); break; case BMAPI_ALLOCATE: - xfs_iomap_enter_trace(XFS_IOMAP_ALLOC_ENTER, io, offset, count); + xfs_iomap_enter_trace(XFS_IOMAP_ALLOC_ENTER, ip, offset, count); lockmode = XFS_ILOCK_SHARED|XFS_EXTSIZE_RD; bmapi_flags = XFS_BMAPI_ENTIRE; + /* Attempt non-blocking lock */ if (flags & BMAPI_TRYLOCK) { - if (!XFS_ILOCK_NOWAIT(mp, io, lockmode)) + if (!xfs_ilock_nowait(ip, lockmode)) return XFS_ERROR(EAGAIN); } else { - XFS_ILOCK(mp, io, lockmode); + xfs_ilock(ip, lockmode); } break; - case BMAPI_UNWRITTEN: - goto phase2; - case BMAPI_DEVICE: - lockmode = XFS_LCK_MAP_SHARED(mp, io); - iomapp->iomap_target = io->io_flags & XFS_IOCORE_RT ? - mp->m_rtdev_targp : mp->m_ddev_targp; - error = 0; - *niomaps = 1; - goto out; default: BUG(); } @@ -237,7 +224,7 @@ xfs_iomap( end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count); offset_fsb = XFS_B_TO_FSBT(mp, offset); - error = XFS_BMAPI(mp, NULL, io, offset_fsb, + error = xfs_bmapi(NULL, ip, offset_fsb, (xfs_filblks_t)(end_fsb - offset_fsb), bmapi_flags, NULL, 0, &imap, &nimaps, NULL, NULL); @@ -245,54 +232,48 @@ xfs_iomap( if (error) goto out; -phase2: - switch (flags & (BMAPI_WRITE|BMAPI_ALLOCATE|BMAPI_UNWRITTEN)) { + switch (flags & (BMAPI_WRITE|BMAPI_ALLOCATE)) { case BMAPI_WRITE: /* If we found an extent, return it */ if (nimaps && (imap.br_startblock != HOLESTARTBLOCK) && (imap.br_startblock != DELAYSTARTBLOCK)) { - xfs_iomap_map_trace(XFS_IOMAP_WRITE_MAP, io, + xfs_iomap_map_trace(XFS_IOMAP_WRITE_MAP, ip, offset, count, iomapp, &imap, flags); break; } if (flags & (BMAPI_DIRECT|BMAPI_MMAP)) { - error = XFS_IOMAP_WRITE_DIRECT(mp, io, offset, - count, flags, &imap, &nimaps, nimaps); + error = xfs_iomap_write_direct(ip, offset, count, flags, + &imap, &nimaps, nimaps); } else { - error = XFS_IOMAP_WRITE_DELAY(mp, io, offset, count, - flags, &imap, &nimaps); + error = xfs_iomap_write_delay(ip, offset, count, flags, + &imap, &nimaps); } if (!error) { - xfs_iomap_map_trace(XFS_IOMAP_ALLOC_MAP, io, + xfs_iomap_map_trace(XFS_IOMAP_ALLOC_MAP, ip, offset, count, iomapp, &imap, flags); } iomap_flags = IOMAP_NEW; break; case BMAPI_ALLOCATE: /* If we found an extent, return it */ - XFS_IUNLOCK(mp, io, lockmode); + xfs_iunlock(ip, lockmode); lockmode = 0; if (nimaps && !ISNULLSTARTBLOCK(imap.br_startblock)) { - xfs_iomap_map_trace(XFS_IOMAP_WRITE_MAP, io, + xfs_iomap_map_trace(XFS_IOMAP_WRITE_MAP, ip, offset, count, iomapp, &imap, flags); break; } - error = XFS_IOMAP_WRITE_ALLOCATE(mp, io, offset, count, + error = xfs_iomap_write_allocate(ip, offset, count, &imap, &nimaps); break; - case BMAPI_UNWRITTEN: - lockmode = 0; - error = XFS_IOMAP_WRITE_UNWRITTEN(mp, io, offset, count); - nimaps = 0; - break; } if (nimaps) { - *niomaps = xfs_imap_to_bmap(io, offset, &imap, + *niomaps = xfs_imap_to_bmap(ip, offset, &imap, iomapp, nimaps, *niomaps, iomap_flags); } else if (niomaps) { *niomaps = 0; @@ -300,14 +281,15 @@ phase2: out: if (lockmode) - XFS_IUNLOCK(mp, io, lockmode); + xfs_iunlock(ip, lockmode); return XFS_ERROR(error); } + STATIC int xfs_iomap_eof_align_last_fsb( xfs_mount_t *mp, - xfs_iocore_t *io, + xfs_inode_t *ip, xfs_fsize_t isize, xfs_extlen_t extsize, xfs_fileoff_t *last_fsb) @@ -316,7 +298,7 @@ xfs_iomap_eof_align_last_fsb( xfs_extlen_t align; int eof, error; - if (io->io_flags & XFS_IOCORE_RT) + if (XFS_IS_REALTIME_INODE(ip)) ; /* * If mounted with the "-o swalloc" option, roundup the allocation @@ -347,7 +329,7 @@ xfs_iomap_eof_align_last_fsb( } if (new_last_fsb) { - error = XFS_BMAP_EOF(mp, io, new_last_fsb, XFS_DATA_FORK, &eof); + error = xfs_bmap_eof(ip, new_last_fsb, XFS_DATA_FORK, &eof); if (error) return error; if (eof) @@ -416,7 +398,6 @@ xfs_iomap_write_direct( int found) { xfs_mount_t *mp = ip->i_mount; - xfs_iocore_t *io = &ip->i_iocore; xfs_fileoff_t offset_fsb; xfs_fileoff_t last_fsb; xfs_filblks_t count_fsb, resaligned; @@ -446,13 +427,13 @@ xfs_iomap_write_direct( extsz = xfs_get_extsz_hint(ip); isize = ip->i_size; - if (io->io_new_size > isize) - isize = io->io_new_size; + if (ip->i_new_size > isize) + isize = ip->i_new_size; offset_fsb = XFS_B_TO_FSBT(mp, offset); last_fsb = XFS_B_TO_FSB(mp, ((xfs_ufsize_t)(offset + count))); if ((offset + count) > isize) { - error = xfs_iomap_eof_align_last_fsb(mp, io, isize, extsz, + error = xfs_iomap_eof_align_last_fsb(mp, ip, isize, extsz, &last_fsb); if (error) goto error_out; @@ -519,7 +500,7 @@ xfs_iomap_write_direct( */ XFS_BMAP_INIT(&free_list, &firstfsb); nimaps = 1; - error = XFS_BMAPI(mp, tp, io, offset_fsb, count_fsb, bmapi_flag, + error = xfs_bmapi(tp, ip, offset_fsb, count_fsb, bmapi_flag, &firstfsb, 0, &imap, &nimaps, &free_list, NULL); if (error) goto error0; @@ -542,7 +523,8 @@ xfs_iomap_write_direct( goto error_out; } - if (unlikely(!imap.br_startblock && !(io->io_flags & XFS_IOCORE_RT))) { + if (unlikely(!imap.br_startblock && + !(XFS_IS_REALTIME_INODE(ip)))) { error = xfs_cmn_err_fsblock_zero(ip, &imap); goto error_out; } @@ -577,7 +559,7 @@ error_out: STATIC int xfs_iomap_eof_want_preallocate( xfs_mount_t *mp, - xfs_iocore_t *io, + xfs_inode_t *ip, xfs_fsize_t isize, xfs_off_t offset, size_t count, @@ -604,7 +586,7 @@ xfs_iomap_eof_want_preallocate( while (count_fsb > 0) { imaps = nimaps; firstblock = NULLFSBLOCK; - error = XFS_BMAPI(mp, NULL, io, start_fsb, count_fsb, 0, + error = xfs_bmapi(NULL, ip, start_fsb, count_fsb, 0, &firstblock, 0, imap, &imaps, NULL, NULL); if (error) return error; @@ -630,7 +612,6 @@ xfs_iomap_write_delay( int *nmaps) { xfs_mount_t *mp = ip->i_mount; - xfs_iocore_t *io = &ip->i_iocore; xfs_fileoff_t offset_fsb; xfs_fileoff_t last_fsb; xfs_off_t aligned_offset; @@ -658,10 +639,10 @@ xfs_iomap_write_delay( retry: isize = ip->i_size; - if (io->io_new_size > isize) - isize = io->io_new_size; + if (ip->i_new_size > isize) + isize = ip->i_new_size; - error = xfs_iomap_eof_want_preallocate(mp, io, isize, offset, count, + error = xfs_iomap_eof_want_preallocate(mp, ip, isize, offset, count, ioflag, imap, XFS_WRITE_IMAPS, &prealloc); if (error) return error; @@ -675,7 +656,7 @@ retry: } if (prealloc || extsz) { - error = xfs_iomap_eof_align_last_fsb(mp, io, isize, extsz, + error = xfs_iomap_eof_align_last_fsb(mp, ip, isize, extsz, &last_fsb); if (error) return error; @@ -683,7 +664,7 @@ retry: nimaps = XFS_WRITE_IMAPS; firstblock = NULLFSBLOCK; - error = XFS_BMAPI(mp, NULL, io, offset_fsb, + error = xfs_bmapi(NULL, ip, offset_fsb, (xfs_filblks_t)(last_fsb - offset_fsb), XFS_BMAPI_DELAY | XFS_BMAPI_WRITE | XFS_BMAPI_ENTIRE, &firstblock, 1, imap, @@ -697,7 +678,7 @@ retry: */ if (nimaps == 0) { xfs_iomap_enter_trace(XFS_IOMAP_WRITE_NOSPACE, - io, offset, count); + ip, offset, count); if (xfs_flush_space(ip, &fsynced, &ioflag)) return XFS_ERROR(ENOSPC); @@ -705,7 +686,8 @@ retry: goto retry; } - if (unlikely(!imap[0].br_startblock && !(io->io_flags & XFS_IOCORE_RT))) + if (unlikely(!imap[0].br_startblock && + !(XFS_IS_REALTIME_INODE(ip)))) return xfs_cmn_err_fsblock_zero(ip, &imap[0]); *ret_imap = imap[0]; @@ -720,6 +702,9 @@ retry: * the originating callers request. * * Called without a lock on the inode. + * + * We no longer bother to look at the incoming map - all we have to + * guarantee is that whatever we allocate fills the required range. */ int xfs_iomap_write_allocate( @@ -730,15 +715,14 @@ xfs_iomap_write_allocate( int *retmap) { xfs_mount_t *mp = ip->i_mount; - xfs_iocore_t *io = &ip->i_iocore; xfs_fileoff_t offset_fsb, last_block; xfs_fileoff_t end_fsb, map_start_fsb; xfs_fsblock_t first_block; xfs_bmap_free_t free_list; xfs_filblks_t count_fsb; - xfs_bmbt_irec_t imap[XFS_STRAT_WRITE_IMAPS]; + xfs_bmbt_irec_t imap; xfs_trans_t *tp; - int i, nimaps, committed; + int nimaps, committed; int error = 0; int nres; @@ -785,13 +769,38 @@ xfs_iomap_write_allocate( XFS_BMAP_INIT(&free_list, &first_block); - nimaps = XFS_STRAT_WRITE_IMAPS; /* - * Ensure we don't go beyond eof - it is possible - * the extents changed since we did the read call, - * we dropped the ilock in the interim. + * it is possible that the extents have changed since + * we did the read call as we dropped the ilock for a + * while. We have to be careful about truncates or hole + * punchs here - we are not allowed to allocate + * non-delalloc blocks here. + * + * The only protection against truncation is the pages + * for the range we are being asked to convert are + * locked and hence a truncate will block on them + * first. + * + * As a result, if we go beyond the range we really + * need and hit an delalloc extent boundary followed by + * a hole while we have excess blocks in the map, we + * will fill the hole incorrectly and overrun the + * transaction reservation. + * + * Using a single map prevents this as we are forced to + * check each map we look for overlap with the desired + * range and abort as soon as we find it. Also, given + * that we only return a single map, having one beyond + * what we can return is probably a bit silly. + * + * We also need to check that we don't go beyond EOF; + * this is a truncate optimisation as a truncate sets + * the new file size before block on the pages we + * currently have locked under writeback. Because they + * are about to be tossed, we don't need to write them + * back.... */ - + nimaps = 1; end_fsb = XFS_B_TO_FSB(mp, ip->i_size); xfs_bmap_last_offset(NULL, ip, &last_block, XFS_DATA_FORK); @@ -805,9 +814,9 @@ xfs_iomap_write_allocate( } /* Go get the actual blocks */ - error = XFS_BMAPI(mp, tp, io, map_start_fsb, count_fsb, + error = xfs_bmapi(tp, ip, map_start_fsb, count_fsb, XFS_BMAPI_WRITE, &first_block, 1, - imap, &nimaps, &free_list, NULL); + &imap, &nimaps, &free_list, NULL); if (error) goto trans_cancel; @@ -826,27 +835,24 @@ xfs_iomap_write_allocate( * See if we were able to allocate an extent that * covers at least part of the callers request */ - for (i = 0; i < nimaps; i++) { - if (unlikely(!imap[i].br_startblock && - !(io->io_flags & XFS_IOCORE_RT))) - return xfs_cmn_err_fsblock_zero(ip, &imap[i]); - if ((offset_fsb >= imap[i].br_startoff) && - (offset_fsb < (imap[i].br_startoff + - imap[i].br_blockcount))) { - *map = imap[i]; - *retmap = 1; - XFS_STATS_INC(xs_xstrat_quick); - return 0; - } - count_fsb -= imap[i].br_blockcount; + if (unlikely(!imap.br_startblock && + XFS_IS_REALTIME_INODE(ip))) + return xfs_cmn_err_fsblock_zero(ip, &imap); + if ((offset_fsb >= imap.br_startoff) && + (offset_fsb < (imap.br_startoff + + imap.br_blockcount))) { + *map = imap; + *retmap = 1; + XFS_STATS_INC(xs_xstrat_quick); + return 0; } - /* So far we have not mapped the requested part of the + /* + * So far we have not mapped the requested part of the * file, just surrounding data, try again. */ - nimaps--; - map_start_fsb = imap[nimaps].br_startoff + - imap[nimaps].br_blockcount; + count_fsb -= imap.br_blockcount; + map_start_fsb = imap.br_startoff + imap.br_blockcount; } trans_cancel: @@ -864,7 +870,6 @@ xfs_iomap_write_unwritten( size_t count) { xfs_mount_t *mp = ip->i_mount; - xfs_iocore_t *io = &ip->i_iocore; xfs_fileoff_t offset_fsb; xfs_filblks_t count_fsb; xfs_filblks_t numblks_fsb; @@ -877,8 +882,7 @@ xfs_iomap_write_unwritten( int committed; int error; - xfs_iomap_enter_trace(XFS_IOMAP_UNWRITTEN, - &ip->i_iocore, offset, count); + xfs_iomap_enter_trace(XFS_IOMAP_UNWRITTEN, ip, offset, count); offset_fsb = XFS_B_TO_FSBT(mp, offset); count_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count); @@ -912,7 +916,7 @@ xfs_iomap_write_unwritten( */ XFS_BMAP_INIT(&free_list, &firstfsb); nimaps = 1; - error = XFS_BMAPI(mp, tp, io, offset_fsb, count_fsb, + error = xfs_bmapi(tp, ip, offset_fsb, count_fsb, XFS_BMAPI_WRITE|XFS_BMAPI_CONVERT, &firstfsb, 1, &imap, &nimaps, &free_list, NULL); if (error) @@ -928,7 +932,7 @@ xfs_iomap_write_unwritten( return XFS_ERROR(error); if (unlikely(!imap.br_startblock && - !(io->io_flags & XFS_IOCORE_RT))) + !(XFS_IS_REALTIME_INODE(ip)))) return xfs_cmn_err_fsblock_zero(ip, &imap); if ((numblks_fsb = imap.br_blockcount) == 0) { diff --git a/fs/xfs/xfs_iomap.h b/fs/xfs/xfs_iomap.h index f5c09887fe9..ee1a0c134cc 100644 --- a/fs/xfs/xfs_iomap.h +++ b/fs/xfs/xfs_iomap.h @@ -36,14 +36,12 @@ typedef enum { BMAPI_READ = (1 << 0), /* read extents */ BMAPI_WRITE = (1 << 1), /* create extents */ BMAPI_ALLOCATE = (1 << 2), /* delayed allocate to real extents */ - BMAPI_UNWRITTEN = (1 << 3), /* unwritten extents to real extents */ /* modifiers */ BMAPI_IGNSTATE = (1 << 4), /* ignore unwritten state on read */ BMAPI_DIRECT = (1 << 5), /* direct instead of buffered write */ BMAPI_MMAP = (1 << 6), /* allocate for mmap write */ BMAPI_SYNC = (1 << 7), /* sync write to flush delalloc space */ BMAPI_TRYLOCK = (1 << 8), /* non-blocking request */ - BMAPI_DEVICE = (1 << 9), /* we only want to know the device */ } bmapi_flags_t; @@ -73,11 +71,10 @@ typedef struct xfs_iomap { iomap_flags_t iomap_flags; } xfs_iomap_t; -struct xfs_iocore; struct xfs_inode; struct xfs_bmbt_irec; -extern int xfs_iomap(struct xfs_iocore *, xfs_off_t, ssize_t, int, +extern int xfs_iomap(struct xfs_inode *, xfs_off_t, ssize_t, int, struct xfs_iomap *, int *); extern int xfs_iomap_write_direct(struct xfs_inode *, xfs_off_t, size_t, int, struct xfs_bmbt_irec *, int *, int); diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index 9fc4c288652..f615e04364f 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -45,7 +45,7 @@ xfs_internal_inum( xfs_ino_t ino) { return (ino == mp->m_sb.sb_rbmino || ino == mp->m_sb.sb_rsumino || - (XFS_SB_VERSION_HASQUOTA(&mp->m_sb) && + (xfs_sb_version_hasquota(&mp->m_sb) && (ino == mp->m_sb.sb_uquotino || ino == mp->m_sb.sb_gquotino))); } @@ -170,7 +170,7 @@ xfs_bulkstat_one_dinode( buf->bs_mtime.tv_nsec = be32_to_cpu(dic->di_mtime.t_nsec); buf->bs_ctime.tv_sec = be32_to_cpu(dic->di_ctime.t_sec); buf->bs_ctime.tv_nsec = be32_to_cpu(dic->di_ctime.t_nsec); - buf->bs_xflags = xfs_dic2xflags(dic); + buf->bs_xflags = xfs_dic2xflags(dip); buf->bs_extsize = be32_to_cpu(dic->di_extsize) << mp->m_sb.sb_blocklog; buf->bs_extents = be32_to_cpu(dic->di_nextents); buf->bs_gen = be32_to_cpu(dic->di_gen); @@ -291,7 +291,7 @@ xfs_bulkstat_use_dinode( dip = (xfs_dinode_t *) xfs_buf_offset(bp, clustidx << mp->m_sb.sb_inodelog); /* - * Check the buffer containing the on-disk inode for di_nlink == 0. + * Check the buffer containing the on-disk inode for di_mode == 0. * This is to prevent xfs_bulkstat from picking up just reclaimed * inodes that have their in-core state initialized but not flushed * to disk yet. This is a temporary hack that would require a proper @@ -299,7 +299,7 @@ xfs_bulkstat_use_dinode( */ if (be16_to_cpu(dip->di_core.di_magic) != XFS_DINODE_MAGIC || !XFS_DINODE_GOOD_VERSION(dip->di_core.di_version) || - !dip->di_core.di_nlink) + !dip->di_core.di_mode) return 0; if (flags & BULKSTAT_FG_QUICK) { *dipp = dip; @@ -307,7 +307,7 @@ xfs_bulkstat_use_dinode( } /* BULKSTAT_FG_INLINE: if attr fork is local, or not there, use it */ aformat = dip->di_core.di_aformat; - if ((XFS_CFORK_Q(&dip->di_core) == 0) || + if ((XFS_DFORK_Q(dip) == 0) || (aformat == XFS_DINODE_FMT_LOCAL) || (aformat == XFS_DINODE_FMT_EXTENTS && !dip->di_core.di_anextents)) { *dipp = dip; @@ -399,7 +399,7 @@ xfs_bulkstat( (XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_inodelog); nimask = ~(nicluster - 1); nbcluster = nicluster >> mp->m_sb.sb_inopblog; - irbuf = kmem_zalloc_greedy(&irbsize, NBPC, NBPC * 4, + irbuf = kmem_zalloc_greedy(&irbsize, PAGE_SIZE, PAGE_SIZE * 4, KM_SLEEP | KM_MAYFAIL | KM_LARGE); nirbuf = irbsize / sizeof(*irbuf); @@ -830,7 +830,7 @@ xfs_inumbers( agino = XFS_INO_TO_AGINO(mp, ino); left = *count; *count = 0; - bcount = MIN(left, (int)(NBPP / sizeof(*buffer))); + bcount = MIN(left, (int)(PAGE_SIZE / sizeof(*buffer))); buffer = kmem_alloc(bcount * sizeof(*buffer), KM_SLEEP); error = bufidx = 0; cur = NULL; diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 77c12715a7d..31f2b04f2c9 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -399,10 +399,10 @@ xfs_log_notify(xfs_mount_t *mp, /* mount of partition */ { xlog_t *log = mp->m_log; xlog_in_core_t *iclog = (xlog_in_core_t *)iclog_hndl; - int abortflg, spl; + int abortflg; cb->cb_next = NULL; - spl = LOG_LOCK(log); + spin_lock(&log->l_icloglock); abortflg = (iclog->ic_state & XLOG_STATE_IOERROR); if (!abortflg) { ASSERT_ALWAYS((iclog->ic_state == XLOG_STATE_ACTIVE) || @@ -411,7 +411,7 @@ xfs_log_notify(xfs_mount_t *mp, /* mount of partition */ *(iclog->ic_callback_tail) = cb; iclog->ic_callback_tail = &(cb->cb_next); } - LOG_UNLOCK(log, spl); + spin_unlock(&log->l_icloglock); return abortflg; } /* xfs_log_notify */ @@ -498,11 +498,14 @@ xfs_log_reserve(xfs_mount_t *mp, * Return error or zero. */ int -xfs_log_mount(xfs_mount_t *mp, - xfs_buftarg_t *log_target, - xfs_daddr_t blk_offset, - int num_bblks) +xfs_log_mount( + xfs_mount_t *mp, + xfs_buftarg_t *log_target, + xfs_daddr_t blk_offset, + int num_bblks) { + int error; + if (!(mp->m_flags & XFS_MOUNT_NORECOVERY)) cmn_err(CE_NOTE, "XFS mounting filesystem %s", mp->m_fsname); else { @@ -515,11 +518,21 @@ xfs_log_mount(xfs_mount_t *mp, mp->m_log = xlog_alloc_log(mp, log_target, blk_offset, num_bblks); /* + * Initialize the AIL now we have a log. + */ + spin_lock_init(&mp->m_ail_lock); + error = xfs_trans_ail_init(mp); + if (error) { + cmn_err(CE_WARN, "XFS: AIL initialisation failed: error %d", error); + goto error; + } + + /* * skip log recovery on a norecovery mount. pretend it all * just worked. */ if (!(mp->m_flags & XFS_MOUNT_NORECOVERY)) { - int error, readonly = (mp->m_flags & XFS_MOUNT_RDONLY); + int readonly = (mp->m_flags & XFS_MOUNT_RDONLY); if (readonly) mp->m_flags &= ~XFS_MOUNT_RDONLY; @@ -530,8 +543,7 @@ xfs_log_mount(xfs_mount_t *mp, mp->m_flags |= XFS_MOUNT_RDONLY; if (error) { cmn_err(CE_WARN, "XFS: log mount/recovery failed: error %d", error); - xlog_dealloc_log(mp->m_log); - return error; + goto error; } } @@ -540,6 +552,9 @@ xfs_log_mount(xfs_mount_t *mp, /* End mounting message in xfs_log_mount_finish */ return 0; +error: + xfs_log_unmount_dealloc(mp); + return error; } /* xfs_log_mount */ /* @@ -606,7 +621,6 @@ xfs_log_unmount_write(xfs_mount_t *mp) xfs_log_ticket_t tic = NULL; xfs_lsn_t lsn; int error; - SPLDECL(s); /* the data section must be 32 bit size aligned */ struct { @@ -659,24 +673,24 @@ xfs_log_unmount_write(xfs_mount_t *mp) } - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); iclog = log->l_iclog; iclog->ic_refcnt++; - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); xlog_state_want_sync(log, iclog); (void) xlog_state_release_iclog(log, iclog); - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); if (!(iclog->ic_state == XLOG_STATE_ACTIVE || iclog->ic_state == XLOG_STATE_DIRTY)) { if (!XLOG_FORCED_SHUTDOWN(log)) { sv_wait(&iclog->ic_forcesema, PMEM, &log->l_icloglock, s); } else { - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); } } else { - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); } if (tic) { xlog_trace_loggrant(log, tic, "unmount rec"); @@ -697,15 +711,15 @@ xfs_log_unmount_write(xfs_mount_t *mp) * a file system that went into forced_shutdown as * the result of an unmount.. */ - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); iclog = log->l_iclog; iclog->ic_refcnt++; - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); xlog_state_want_sync(log, iclog); (void) xlog_state_release_iclog(log, iclog); - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); if ( ! ( iclog->ic_state == XLOG_STATE_ACTIVE || iclog->ic_state == XLOG_STATE_DIRTY @@ -714,7 +728,7 @@ xfs_log_unmount_write(xfs_mount_t *mp) sv_wait(&iclog->ic_forcesema, PMEM, &log->l_icloglock, s); } else { - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); } } @@ -723,10 +737,14 @@ xfs_log_unmount_write(xfs_mount_t *mp) /* * Deallocate log structures for unmount/relocation. + * + * We need to stop the aild from running before we destroy + * and deallocate the log as the aild references the log. */ void xfs_log_unmount_dealloc(xfs_mount_t *mp) { + xfs_trans_ail_destroy(mp); xlog_dealloc_log(mp->m_log); } @@ -762,20 +780,18 @@ xfs_log_move_tail(xfs_mount_t *mp, xlog_ticket_t *tic; xlog_t *log = mp->m_log; int need_bytes, free_bytes, cycle, bytes; - SPLDECL(s); if (XLOG_FORCED_SHUTDOWN(log)) return; - ASSERT(!XFS_FORCED_SHUTDOWN(mp)); if (tail_lsn == 0) { /* needed since sync_lsn is 64 bits */ - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); tail_lsn = log->l_last_sync_lsn; - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); } - s = GRANT_LOCK(log); + spin_lock(&log->l_grant_lock); /* Also an invalid lsn. 1 implies that we aren't passing in a valid * tail_lsn. @@ -824,7 +840,7 @@ xfs_log_move_tail(xfs_mount_t *mp, tic = tic->t_next; } while (tic != log->l_reserve_headq); } - GRANT_UNLOCK(log, s); + spin_unlock(&log->l_grant_lock); } /* xfs_log_move_tail */ /* @@ -836,14 +852,13 @@ xfs_log_move_tail(xfs_mount_t *mp, int xfs_log_need_covered(xfs_mount_t *mp) { - SPLDECL(s); int needed = 0, gen; xlog_t *log = mp->m_log; if (!xfs_fs_writable(mp)) return 0; - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); if (((log->l_covered_state == XLOG_STATE_COVER_NEED) || (log->l_covered_state == XLOG_STATE_COVER_NEED2)) && !xfs_trans_first_ail(mp, &gen) @@ -856,7 +871,7 @@ xfs_log_need_covered(xfs_mount_t *mp) } needed = 1; } - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); return needed; } @@ -881,17 +896,16 @@ xfs_lsn_t xlog_assign_tail_lsn(xfs_mount_t *mp) { xfs_lsn_t tail_lsn; - SPLDECL(s); xlog_t *log = mp->m_log; tail_lsn = xfs_trans_tail_ail(mp); - s = GRANT_LOCK(log); + spin_lock(&log->l_grant_lock); if (tail_lsn != 0) { log->l_tail_lsn = tail_lsn; } else { tail_lsn = log->l_tail_lsn = log->l_last_sync_lsn; } - GRANT_UNLOCK(log, s); + spin_unlock(&log->l_grant_lock); return tail_lsn; } /* xlog_assign_tail_lsn */ @@ -911,7 +925,7 @@ xlog_assign_tail_lsn(xfs_mount_t *mp) * the tail. The details of this case are described below, but the end * result is that we return the size of the log as the amount of space left. */ -int +STATIC int xlog_space_left(xlog_t *log, int cycle, int bytes) { int free_bytes; @@ -1076,7 +1090,7 @@ xlog_get_iclog_buffer_size(xfs_mount_t *mp, size >>= 1; } - if (XFS_SB_VERSION_HASLOGV2(&mp->m_sb)) { + if (xfs_sb_version_haslogv2(&mp->m_sb)) { /* # headers = size / 32K * one header holds cycles from 32K of data */ @@ -1165,20 +1179,20 @@ xlog_alloc_log(xfs_mount_t *mp, log->l_flags |= XLOG_ACTIVE_RECOVERY; log->l_prev_block = -1; - ASSIGN_ANY_LSN_HOST(log->l_tail_lsn, 1, 0); + log->l_tail_lsn = xlog_assign_lsn(1, 0); /* log->l_tail_lsn = 0x100000000LL; cycle = 1; current block = 0 */ log->l_last_sync_lsn = log->l_tail_lsn; log->l_curr_cycle = 1; /* 0 is bad since this is initial value */ log->l_grant_reserve_cycle = 1; log->l_grant_write_cycle = 1; - if (XFS_SB_VERSION_HASSECTOR(&mp->m_sb)) { + if (xfs_sb_version_hassector(&mp->m_sb)) { log->l_sectbb_log = mp->m_sb.sb_logsectlog - BBSHIFT; ASSERT(log->l_sectbb_log <= mp->m_sectbb_log); /* for larger sector sizes, must have v2 or external log */ ASSERT(log->l_sectbb_log == 0 || log->l_logBBstart == 0 || - XFS_SB_VERSION_HASLOGV2(&mp->m_sb)); + xfs_sb_version_haslogv2(&mp->m_sb)); ASSERT(mp->m_sb.sb_logsectlog >= BBSHIFT); } log->l_sectbb_mask = (1 << log->l_sectbb_log) - 1; @@ -1193,8 +1207,8 @@ xlog_alloc_log(xfs_mount_t *mp, ASSERT(XFS_BUF_VALUSEMA(bp) <= 0); log->l_xbuf = bp; - spinlock_init(&log->l_icloglock, "iclog"); - spinlock_init(&log->l_grant_lock, "grhead_iclog"); + spin_lock_init(&log->l_icloglock); + spin_lock_init(&log->l_grant_lock); initnsema(&log->l_flushsema, 0, "ic-flush"); xlog_state_ticket_alloc(log); /* wait until after icloglock inited */ @@ -1231,12 +1245,12 @@ xlog_alloc_log(xfs_mount_t *mp, head = &iclog->ic_header; memset(head, 0, sizeof(xlog_rec_header_t)); - INT_SET(head->h_magicno, ARCH_CONVERT, XLOG_HEADER_MAGIC_NUM); - INT_SET(head->h_version, ARCH_CONVERT, - XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) ? 2 : 1); - INT_SET(head->h_size, ARCH_CONVERT, log->l_iclog_size); + head->h_magicno = cpu_to_be32(XLOG_HEADER_MAGIC_NUM); + head->h_version = cpu_to_be32( + xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? 2 : 1); + head->h_size = cpu_to_be32(log->l_iclog_size); /* new fields */ - INT_SET(head->h_fmt, ARCH_CONVERT, XLOG_FMT); + head->h_fmt = cpu_to_be32(XLOG_FMT); memcpy(&head->h_fs_uuid, &mp->m_sb.sb_uuid, sizeof(uuid_t)); @@ -1293,7 +1307,7 @@ xlog_commit_record(xfs_mount_t *mp, * pushes on an lsn which is further along in the log once we reach the high * water mark. In this manner, we would be creating a low water mark. */ -void +STATIC void xlog_grant_push_ail(xfs_mount_t *mp, int need_bytes) { @@ -1305,11 +1319,10 @@ xlog_grant_push_ail(xfs_mount_t *mp, int threshold_block; /* block in lsn we'd like to be at */ int threshold_cycle; /* lsn cycle we'd like to be at */ int free_threshold; - SPLDECL(s); ASSERT(BTOBB(need_bytes) < log->l_logBBsize); - s = GRANT_LOCK(log); + spin_lock(&log->l_grant_lock); free_bytes = xlog_space_left(log, log->l_grant_reserve_cycle, log->l_grant_reserve_bytes); @@ -1331,8 +1344,7 @@ xlog_grant_push_ail(xfs_mount_t *mp, threshold_block -= log->l_logBBsize; threshold_cycle += 1; } - ASSIGN_ANY_LSN_HOST(threshold_lsn, threshold_cycle, - threshold_block); + threshold_lsn = xlog_assign_lsn(threshold_cycle, threshold_block); /* Don't pass in an lsn greater than the lsn of the last * log record known to be on disk. @@ -1340,7 +1352,7 @@ xlog_grant_push_ail(xfs_mount_t *mp, if (XFS_LSN_CMP(threshold_lsn, log->l_last_sync_lsn) > 0) threshold_lsn = log->l_last_sync_lsn; } - GRANT_UNLOCK(log, s); + spin_unlock(&log->l_grant_lock); /* * Get the transaction layer to kick the dirty buffers out to @@ -1378,20 +1390,19 @@ xlog_grant_push_ail(xfs_mount_t *mp, * is added immediately before calling bwrite(). */ -int +STATIC int xlog_sync(xlog_t *log, xlog_in_core_t *iclog) { xfs_caddr_t dptr; /* pointer to byte sized element */ xfs_buf_t *bp; - int i, ops; + int i; uint count; /* byte count of bwrite */ uint count_init; /* initial count before roundup */ int roundoff; /* roundoff to BB or stripe */ int split = 0; /* split write into two regions */ int error; - SPLDECL(s); - int v2 = XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb); + int v2 = xfs_sb_version_haslogv2(&log->l_mp->m_sb); XFS_STATS_INC(xs_log_writes); ASSERT(iclog->ic_refcnt == 0); @@ -1415,30 +1426,26 @@ xlog_sync(xlog_t *log, roundoff < BBTOB(1))); /* move grant heads by roundoff in sync */ - s = GRANT_LOCK(log); + spin_lock(&log->l_grant_lock); xlog_grant_add_space(log, roundoff); - GRANT_UNLOCK(log, s); + spin_unlock(&log->l_grant_lock); /* put cycle number in every block */ xlog_pack_data(log, iclog, roundoff); /* real byte length */ if (v2) { - INT_SET(iclog->ic_header.h_len, - ARCH_CONVERT, - iclog->ic_offset + roundoff); + iclog->ic_header.h_len = + cpu_to_be32(iclog->ic_offset + roundoff); } else { - INT_SET(iclog->ic_header.h_len, ARCH_CONVERT, iclog->ic_offset); + iclog->ic_header.h_len = + cpu_to_be32(iclog->ic_offset); } - /* put ops count in correct order */ - ops = iclog->ic_header.h_num_logops; - INT_SET(iclog->ic_header.h_num_logops, ARCH_CONVERT, ops); - bp = iclog->ic_bp; ASSERT(XFS_BUF_FSPRIVATE2(bp, unsigned long) == (unsigned long)1); XFS_BUF_SET_FSPRIVATE2(bp, (unsigned long)2); - XFS_BUF_SET_ADDR(bp, BLOCK_LSN(INT_GET(iclog->ic_header.h_lsn, ARCH_CONVERT))); + XFS_BUF_SET_ADDR(bp, BLOCK_LSN(be64_to_cpu(iclog->ic_header.h_lsn))); XFS_STATS_ADD(xs_log_blocks, BTOBB(count)); @@ -1501,10 +1508,10 @@ xlog_sync(xlog_t *log, * a new cycle. Watch out for the header magic number * case, though. */ - for (i=0; i<split; i += BBSIZE) { - INT_MOD(*(uint *)dptr, ARCH_CONVERT, +1); - if (INT_GET(*(uint *)dptr, ARCH_CONVERT) == XLOG_HEADER_MAGIC_NUM) - INT_MOD(*(uint *)dptr, ARCH_CONVERT, +1); + for (i = 0; i < split; i += BBSIZE) { + be32_add_cpu((__be32 *)dptr, 1); + if (be32_to_cpu(*(__be32 *)dptr) == XLOG_HEADER_MAGIC_NUM) + be32_add_cpu((__be32 *)dptr, 1); dptr += BBSIZE; } @@ -1527,14 +1534,13 @@ xlog_sync(xlog_t *log, /* * Deallocate a log structure */ -void +STATIC void xlog_dealloc_log(xlog_t *log) { xlog_in_core_t *iclog, *next_iclog; xlog_ticket_t *tic, *next_tic; int i; - iclog = log->l_iclog; for (i=0; i<log->l_iclog_bufs; i++) { sv_destroy(&iclog->ic_forcesema); @@ -1565,7 +1571,7 @@ xlog_dealloc_log(xlog_t *log) tic = log->l_unmount_free; while (tic) { next_tic = tic->t_next; - kmem_free(tic, NBPP); + kmem_free(tic, PAGE_SIZE); tic = next_tic; } } @@ -1592,14 +1598,12 @@ xlog_state_finish_copy(xlog_t *log, int record_cnt, int copy_bytes) { - SPLDECL(s); + spin_lock(&log->l_icloglock); - s = LOG_LOCK(log); - - iclog->ic_header.h_num_logops += record_cnt; + be32_add_cpu(&iclog->ic_header.h_num_logops, record_cnt); iclog->ic_offset += copy_bytes; - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); } /* xlog_state_finish_copy */ @@ -1752,7 +1756,7 @@ xlog_print_tic_res(xfs_mount_t *mp, xlog_ticket_t *ticket) * we don't update ic_offset until the end when we know exactly how many * bytes have been written out. */ -int +STATIC int xlog_write(xfs_mount_t * mp, xfs_log_iovec_t reg[], int nentries, @@ -1823,7 +1827,7 @@ xlog_write(xfs_mount_t * mp, /* start_lsn is the first lsn written to. That's all we need. */ if (! *start_lsn) - *start_lsn = INT_GET(iclog->ic_header.h_lsn, ARCH_CONVERT); + *start_lsn = be64_to_cpu(iclog->ic_header.h_lsn); /* This loop writes out as many regions as can fit in the amount * of space which was allocated by xlog_state_get_iclog_space(). @@ -1839,7 +1843,7 @@ xlog_write(xfs_mount_t * mp, */ if (ticket->t_flags & XLOG_TIC_INITED) { logop_head = (xlog_op_header_t *)ptr; - INT_SET(logop_head->oh_tid, ARCH_CONVERT, ticket->t_tid); + logop_head->oh_tid = cpu_to_be32(ticket->t_tid); logop_head->oh_clientid = ticket->t_clientid; logop_head->oh_len = 0; logop_head->oh_flags = XLOG_START_TRANS; @@ -1853,7 +1857,7 @@ xlog_write(xfs_mount_t * mp, /* Copy log operation header directly into data section */ logop_head = (xlog_op_header_t *)ptr; - INT_SET(logop_head->oh_tid, ARCH_CONVERT, ticket->t_tid); + logop_head->oh_tid = cpu_to_be32(ticket->t_tid); logop_head->oh_clientid = ticket->t_clientid; logop_head->oh_res2 = 0; @@ -1888,13 +1892,14 @@ xlog_write(xfs_mount_t * mp, copy_off = partial_copy_len; if (need_copy <= iclog->ic_size - log_offset) { /*complete write */ - INT_SET(logop_head->oh_len, ARCH_CONVERT, copy_len = need_copy); + copy_len = need_copy; + logop_head->oh_len = cpu_to_be32(copy_len); if (partial_copy) logop_head->oh_flags|= (XLOG_END_TRANS|XLOG_WAS_CONT_TRANS); partial_copy_len = partial_copy = 0; } else { /* partial write */ copy_len = iclog->ic_size - log_offset; - INT_SET(logop_head->oh_len, ARCH_CONVERT, copy_len); + logop_head->oh_len = cpu_to_be32(copy_len); logop_head->oh_flags |= XLOG_CONTINUE_TRANS; if (partial_copy) logop_head->oh_flags |= XLOG_WAS_CONT_TRANS; @@ -1992,7 +1997,8 @@ xlog_state_clean_log(xlog_t *log) * We don't need to cover the dummy. */ if (!changed && - (INT_GET(iclog->ic_header.h_num_logops, ARCH_CONVERT) == XLOG_COVER_OPS)) { + (be32_to_cpu(iclog->ic_header.h_num_logops) == + XLOG_COVER_OPS)) { changed = 1; } else { /* @@ -2060,7 +2066,7 @@ xlog_get_lowest_lsn( lowest_lsn = 0; do { if (!(lsn_log->ic_state & (XLOG_STATE_ACTIVE|XLOG_STATE_DIRTY))) { - lsn = INT_GET(lsn_log->ic_header.h_lsn, ARCH_CONVERT); + lsn = be64_to_cpu(lsn_log->ic_header.h_lsn); if ((lsn && !lowest_lsn) || (XFS_LSN_CMP(lsn, lowest_lsn) < 0)) { lowest_lsn = lsn; @@ -2089,9 +2095,8 @@ xlog_state_do_callback( int funcdidcallbacks; /* flag: function did callbacks */ int repeats; /* for issuing console warnings if * looping too many times */ - SPLDECL(s); - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); first_iclog = iclog = log->l_iclog; ioerrors = 0; funcdidcallbacks = 0; @@ -2136,7 +2141,7 @@ xlog_state_do_callback( * to DO_CALLBACK, we will not process it when * we retry since a previous iclog is in the * CALLBACK and the state cannot change since - * we are holding the LOG_LOCK. + * we are holding the l_icloglock. */ if (!(iclog->ic_state & (XLOG_STATE_DONE_SYNC | @@ -2162,11 +2167,9 @@ xlog_state_do_callback( */ lowest_lsn = xlog_get_lowest_lsn(log); - if (lowest_lsn && ( - XFS_LSN_CMP( - lowest_lsn, - INT_GET(iclog->ic_header.h_lsn, ARCH_CONVERT) - )<0)) { + if (lowest_lsn && + XFS_LSN_CMP(lowest_lsn, + be64_to_cpu(iclog->ic_header.h_lsn)) < 0) { iclog = iclog->ic_next; continue; /* Leave this iclog for * another thread */ @@ -2174,19 +2177,18 @@ xlog_state_do_callback( iclog->ic_state = XLOG_STATE_CALLBACK; - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); /* l_last_sync_lsn field protected by - * GRANT_LOCK. Don't worry about iclog's lsn. + * l_grant_lock. Don't worry about iclog's lsn. * No one else can be here except us. */ - s = GRANT_LOCK(log); - ASSERT(XFS_LSN_CMP( - log->l_last_sync_lsn, - INT_GET(iclog->ic_header.h_lsn, ARCH_CONVERT) - )<=0); - log->l_last_sync_lsn = INT_GET(iclog->ic_header.h_lsn, ARCH_CONVERT); - GRANT_UNLOCK(log, s); + spin_lock(&log->l_grant_lock); + ASSERT(XFS_LSN_CMP(log->l_last_sync_lsn, + be64_to_cpu(iclog->ic_header.h_lsn)) <= 0); + log->l_last_sync_lsn = + be64_to_cpu(iclog->ic_header.h_lsn); + spin_unlock(&log->l_grant_lock); /* * Keep processing entries in the callback list @@ -2195,7 +2197,7 @@ xlog_state_do_callback( * empty and change the state to DIRTY so that * we don't miss any more callbacks being added. */ - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); } else { ioerrors++; } @@ -2204,14 +2206,14 @@ xlog_state_do_callback( while (cb) { iclog->ic_callback_tail = &(iclog->ic_callback); iclog->ic_callback = NULL; - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); /* perform callbacks in the order given */ for (; cb; cb = cb_next) { cb_next = cb->cb_next; cb->cb_func(cb->cb_arg, aborted); } - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); cb = iclog->ic_callback; } @@ -2258,7 +2260,7 @@ xlog_state_do_callback( * * SYNCING - i/o completion will go through logs * DONE_SYNC - interrupt thread should be waiting for - * LOG_LOCK + * l_icloglock * IOERROR - give up hope all ye who enter here */ if (iclog->ic_state == XLOG_STATE_WANT_SYNC || @@ -2276,7 +2278,7 @@ xlog_state_do_callback( flushcnt = log->l_flushcnt; log->l_flushcnt = 0; } - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); while (flushcnt--) vsema(&log->l_flushsema); } /* xlog_state_do_callback */ @@ -2296,15 +2298,14 @@ xlog_state_do_callback( * global state machine log lock. Assume that the calls to cvsema won't * take a long time. At least we know it won't sleep. */ -void +STATIC void xlog_state_done_syncing( xlog_in_core_t *iclog, int aborted) { xlog_t *log = iclog->ic_log; - SPLDECL(s); - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); ASSERT(iclog->ic_state == XLOG_STATE_SYNCING || iclog->ic_state == XLOG_STATE_IOERROR); @@ -2320,7 +2321,7 @@ xlog_state_done_syncing( */ if (iclog->ic_state != XLOG_STATE_IOERROR) { if (--iclog->ic_bwritecnt == 1) { - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); return; } iclog->ic_state = XLOG_STATE_DONE_SYNC; @@ -2332,7 +2333,7 @@ xlog_state_done_syncing( * I/O, the others get to wait for the result. */ sv_broadcast(&iclog->ic_writesema); - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); xlog_state_do_callback(log, aborted, iclog); /* also cleans log */ } /* xlog_state_done_syncing */ @@ -2357,7 +2358,7 @@ xlog_state_done_syncing( * needs to be incremented, depending on the amount of data which * is copied. */ -int +STATIC int xlog_state_get_iclog_space(xlog_t *log, int len, xlog_in_core_t **iclogp, @@ -2365,23 +2366,22 @@ xlog_state_get_iclog_space(xlog_t *log, int *continued_write, int *logoffsetp) { - SPLDECL(s); int log_offset; xlog_rec_header_t *head; xlog_in_core_t *iclog; int error; restart: - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); if (XLOG_FORCED_SHUTDOWN(log)) { - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); return XFS_ERROR(EIO); } iclog = log->l_iclog; if (! (iclog->ic_state == XLOG_STATE_ACTIVE)) { log->l_flushcnt++; - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); xlog_trace_iclog(iclog, XLOG_TRACE_SLEEP_FLUSH); XFS_STATS_INC(xs_log_noiclogs); /* Ensure that log writes happen */ @@ -2404,8 +2404,9 @@ restart: xlog_tic_add_region(ticket, log->l_iclog_hsize, XLOG_REG_TYPE_LRHEADER); - INT_SET(head->h_cycle, ARCH_CONVERT, log->l_curr_cycle); - ASSIGN_LSN(head->h_lsn, log); + head->h_cycle = cpu_to_be32(log->l_curr_cycle); + head->h_lsn = cpu_to_be64( + xlog_assign_lsn(log->l_curr_cycle, log->l_curr_block)); ASSERT(log->l_curr_block >= 0); } @@ -2423,12 +2424,12 @@ restart: /* If I'm the only one writing to this iclog, sync it to disk */ if (iclog->ic_refcnt == 1) { - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); if ((error = xlog_state_release_iclog(log, iclog))) return error; } else { iclog->ic_refcnt--; - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); } goto restart; } @@ -2449,7 +2450,7 @@ restart: *iclogp = iclog; ASSERT(iclog->ic_offset <= iclog->ic_size); - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); *logoffsetp = log_offset; return 0; @@ -2467,7 +2468,6 @@ xlog_grant_log_space(xlog_t *log, { int free_bytes; int need_bytes; - SPLDECL(s); #ifdef DEBUG xfs_lsn_t tail_lsn; #endif @@ -2479,7 +2479,7 @@ xlog_grant_log_space(xlog_t *log, #endif /* Is there space or do we need to sleep? */ - s = GRANT_LOCK(log); + spin_lock(&log->l_grant_lock); xlog_trace_loggrant(log, tic, "xlog_grant_log_space: enter"); /* something is already sleeping; insert new transaction at end */ @@ -2502,7 +2502,7 @@ xlog_grant_log_space(xlog_t *log, */ xlog_trace_loggrant(log, tic, "xlog_grant_log_space: wake 1"); - s = GRANT_LOCK(log); + spin_lock(&log->l_grant_lock); } if (tic->t_flags & XFS_LOG_PERM_RESERV) need_bytes = tic->t_unit_res*tic->t_ocnt; @@ -2524,14 +2524,14 @@ redo: sv_wait(&tic->t_sema, PINOD|PLTWAIT, &log->l_grant_lock, s); if (XLOG_FORCED_SHUTDOWN(log)) { - s = GRANT_LOCK(log); + spin_lock(&log->l_grant_lock); goto error_return; } xlog_trace_loggrant(log, tic, "xlog_grant_log_space: wake 2"); xlog_grant_push_ail(log->l_mp, need_bytes); - s = GRANT_LOCK(log); + spin_lock(&log->l_grant_lock); goto redo; } else if (tic->t_flags & XLOG_TIC_IN_Q) xlog_del_ticketq(&log->l_reserve_headq, tic); @@ -2553,7 +2553,7 @@ redo: #endif xlog_trace_loggrant(log, tic, "xlog_grant_log_space: exit"); xlog_verify_grant_head(log, 1); - GRANT_UNLOCK(log, s); + spin_unlock(&log->l_grant_lock); return 0; error_return: @@ -2567,7 +2567,7 @@ redo: */ tic->t_curr_res = 0; tic->t_cnt = 0; /* ungrant will give back unit_res * t_cnt. */ - GRANT_UNLOCK(log, s); + spin_unlock(&log->l_grant_lock); return XFS_ERROR(EIO); } /* xlog_grant_log_space */ @@ -2581,7 +2581,6 @@ STATIC int xlog_regrant_write_log_space(xlog_t *log, xlog_ticket_t *tic) { - SPLDECL(s); int free_bytes, need_bytes; xlog_ticket_t *ntic; #ifdef DEBUG @@ -2599,7 +2598,7 @@ xlog_regrant_write_log_space(xlog_t *log, panic("regrant Recovery problem"); #endif - s = GRANT_LOCK(log); + spin_lock(&log->l_grant_lock); xlog_trace_loggrant(log, tic, "xlog_regrant_write_log_space: enter"); if (XLOG_FORCED_SHUTDOWN(log)) @@ -2638,14 +2637,14 @@ xlog_regrant_write_log_space(xlog_t *log, /* If we're shutting down, this tic is already * off the queue */ if (XLOG_FORCED_SHUTDOWN(log)) { - s = GRANT_LOCK(log); + spin_lock(&log->l_grant_lock); goto error_return; } xlog_trace_loggrant(log, tic, "xlog_regrant_write_log_space: wake 1"); xlog_grant_push_ail(log->l_mp, tic->t_unit_res); - s = GRANT_LOCK(log); + spin_lock(&log->l_grant_lock); } } @@ -2665,14 +2664,14 @@ redo: /* If we're shutting down, this tic is already off the queue */ if (XLOG_FORCED_SHUTDOWN(log)) { - s = GRANT_LOCK(log); + spin_lock(&log->l_grant_lock); goto error_return; } xlog_trace_loggrant(log, tic, "xlog_regrant_write_log_space: wake 2"); xlog_grant_push_ail(log->l_mp, need_bytes); - s = GRANT_LOCK(log); + spin_lock(&log->l_grant_lock); goto redo; } else if (tic->t_flags & XLOG_TIC_IN_Q) xlog_del_ticketq(&log->l_write_headq, tic); @@ -2689,7 +2688,7 @@ redo: xlog_trace_loggrant(log, tic, "xlog_regrant_write_log_space: exit"); xlog_verify_grant_head(log, 1); - GRANT_UNLOCK(log, s); + spin_unlock(&log->l_grant_lock); return 0; @@ -2704,7 +2703,7 @@ redo: */ tic->t_curr_res = 0; tic->t_cnt = 0; /* ungrant will give back unit_res * t_cnt. */ - GRANT_UNLOCK(log, s); + spin_unlock(&log->l_grant_lock); return XFS_ERROR(EIO); } /* xlog_regrant_write_log_space */ @@ -2720,14 +2719,12 @@ STATIC void xlog_regrant_reserve_log_space(xlog_t *log, xlog_ticket_t *ticket) { - SPLDECL(s); - xlog_trace_loggrant(log, ticket, "xlog_regrant_reserve_log_space: enter"); if (ticket->t_cnt > 0) ticket->t_cnt--; - s = GRANT_LOCK(log); + spin_lock(&log->l_grant_lock); xlog_grant_sub_space(log, ticket->t_curr_res); ticket->t_curr_res = ticket->t_unit_res; xlog_tic_reset_res(ticket); @@ -2737,7 +2734,7 @@ xlog_regrant_reserve_log_space(xlog_t *log, /* just return if we still have some of the pre-reserved space */ if (ticket->t_cnt > 0) { - GRANT_UNLOCK(log, s); + spin_unlock(&log->l_grant_lock); return; } @@ -2745,7 +2742,7 @@ xlog_regrant_reserve_log_space(xlog_t *log, xlog_trace_loggrant(log, ticket, "xlog_regrant_reserve_log_space: exit"); xlog_verify_grant_head(log, 0); - GRANT_UNLOCK(log, s); + spin_unlock(&log->l_grant_lock); ticket->t_curr_res = ticket->t_unit_res; xlog_tic_reset_res(ticket); } /* xlog_regrant_reserve_log_space */ @@ -2769,12 +2766,10 @@ STATIC void xlog_ungrant_log_space(xlog_t *log, xlog_ticket_t *ticket) { - SPLDECL(s); - if (ticket->t_cnt > 0) ticket->t_cnt--; - s = GRANT_LOCK(log); + spin_lock(&log->l_grant_lock); xlog_trace_loggrant(log, ticket, "xlog_ungrant_log_space: enter"); xlog_grant_sub_space(log, ticket->t_curr_res); @@ -2791,7 +2786,7 @@ xlog_ungrant_log_space(xlog_t *log, xlog_trace_loggrant(log, ticket, "xlog_ungrant_log_space: exit"); xlog_verify_grant_head(log, 1); - GRANT_UNLOCK(log, s); + spin_unlock(&log->l_grant_lock); xfs_log_move_tail(log->l_mp, 1); } /* xlog_ungrant_log_space */ @@ -2799,15 +2794,13 @@ xlog_ungrant_log_space(xlog_t *log, /* * Atomically put back used ticket. */ -void +STATIC void xlog_state_put_ticket(xlog_t *log, xlog_ticket_t *tic) { - unsigned long s; - - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); xlog_ticket_put(log, tic); - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); } /* xlog_state_put_ticket */ /* @@ -2819,19 +2812,18 @@ xlog_state_put_ticket(xlog_t *log, * * */ -int +STATIC int xlog_state_release_iclog(xlog_t *log, xlog_in_core_t *iclog) { - SPLDECL(s); int sync = 0; /* do we sync? */ xlog_assign_tail_lsn(log->l_mp); - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); if (iclog->ic_state & XLOG_STATE_IOERROR) { - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); return XFS_ERROR(EIO); } @@ -2843,12 +2835,12 @@ xlog_state_release_iclog(xlog_t *log, iclog->ic_state == XLOG_STATE_WANT_SYNC) { sync++; iclog->ic_state = XLOG_STATE_SYNCING; - INT_SET(iclog->ic_header.h_tail_lsn, ARCH_CONVERT, log->l_tail_lsn); + iclog->ic_header.h_tail_lsn = cpu_to_be64(log->l_tail_lsn); xlog_verify_tail_lsn(log, iclog, log->l_tail_lsn); /* cycle incremented when incrementing curr_block */ } - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); /* * We let the log lock go, so it's possible that we hit a log I/O @@ -2881,7 +2873,7 @@ xlog_state_switch_iclogs(xlog_t *log, if (!eventual_size) eventual_size = iclog->ic_offset; iclog->ic_state = XLOG_STATE_WANT_SYNC; - INT_SET(iclog->ic_header.h_prev_block, ARCH_CONVERT, log->l_prev_block); + iclog->ic_header.h_prev_block = cpu_to_be32(log->l_prev_block); log->l_prev_block = log->l_curr_block; log->l_prev_cycle = log->l_curr_cycle; @@ -2889,7 +2881,7 @@ xlog_state_switch_iclogs(xlog_t *log, log->l_curr_block += BTOBB(eventual_size)+BTOBB(log->l_iclog_hsize); /* Round up to next log-sunit */ - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) && + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb) && log->l_mp->m_sb.sb_logsunit > 1) { __uint32_t sunit_bb = BTOBB(log->l_mp->m_sb.sb_logsunit); log->l_curr_block = roundup(log->l_curr_block, sunit_bb); @@ -2939,13 +2931,12 @@ xlog_state_sync_all(xlog_t *log, uint flags, int *log_flushed) { xlog_in_core_t *iclog; xfs_lsn_t lsn; - SPLDECL(s); - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); iclog = log->l_iclog; if (iclog->ic_state & XLOG_STATE_IOERROR) { - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); return XFS_ERROR(EIO); } @@ -2978,15 +2969,15 @@ xlog_state_sync_all(xlog_t *log, uint flags, int *log_flushed) * the previous sync. */ iclog->ic_refcnt++; - lsn = INT_GET(iclog->ic_header.h_lsn, ARCH_CONVERT); + lsn = be64_to_cpu(iclog->ic_header.h_lsn); xlog_state_switch_iclogs(log, iclog, 0); - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); if (xlog_state_release_iclog(log, iclog)) return XFS_ERROR(EIO); *log_flushed = 1; - s = LOG_LOCK(log); - if (INT_GET(iclog->ic_header.h_lsn, ARCH_CONVERT) == lsn && + spin_lock(&log->l_icloglock); + if (be64_to_cpu(iclog->ic_header.h_lsn) == lsn && iclog->ic_state != XLOG_STATE_DIRTY) goto maybe_sleep; else @@ -3011,12 +3002,12 @@ maybe_sleep: if (flags & XFS_LOG_SYNC) { /* * We must check if we're shutting down here, before - * we wait, while we're holding the LOG_LOCK. + * we wait, while we're holding the l_icloglock. * Then we check again after waking up, in case our * sleep was disturbed by a bad news. */ if (iclog->ic_state & XLOG_STATE_IOERROR) { - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); return XFS_ERROR(EIO); } XFS_STATS_INC(xs_log_force_sleep); @@ -3033,7 +3024,7 @@ maybe_sleep: } else { no_sleep: - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); } return 0; } /* xlog_state_sync_all */ @@ -3051,7 +3042,7 @@ no_sleep: * If filesystem activity goes to zero, the iclog will get flushed only by * bdflush(). */ -int +STATIC int xlog_state_sync(xlog_t *log, xfs_lsn_t lsn, uint flags, @@ -3059,26 +3050,24 @@ xlog_state_sync(xlog_t *log, { xlog_in_core_t *iclog; int already_slept = 0; - SPLDECL(s); - try_again: - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); iclog = log->l_iclog; if (iclog->ic_state & XLOG_STATE_IOERROR) { - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); return XFS_ERROR(EIO); } do { - if (INT_GET(iclog->ic_header.h_lsn, ARCH_CONVERT) != lsn) { - iclog = iclog->ic_next; - continue; + if (be64_to_cpu(iclog->ic_header.h_lsn) != lsn) { + iclog = iclog->ic_next; + continue; } if (iclog->ic_state == XLOG_STATE_DIRTY) { - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); return 0; } @@ -3113,11 +3102,11 @@ try_again: } else { iclog->ic_refcnt++; xlog_state_switch_iclogs(log, iclog, 0); - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); if (xlog_state_release_iclog(log, iclog)) return XFS_ERROR(EIO); *log_flushed = 1; - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); } } @@ -3129,7 +3118,7 @@ try_again: * gotten a log write error. */ if (iclog->ic_state & XLOG_STATE_IOERROR) { - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); return XFS_ERROR(EIO); } XFS_STATS_INC(xs_log_force_sleep); @@ -3143,13 +3132,13 @@ try_again: return XFS_ERROR(EIO); *log_flushed = 1; } else { /* just return */ - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); } return 0; } while (iclog != log->l_iclog); - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); return 0; } /* xlog_state_sync */ @@ -3158,12 +3147,10 @@ try_again: * Called when we want to mark the current iclog as being ready to sync to * disk. */ -void +STATIC void xlog_state_want_sync(xlog_t *log, xlog_in_core_t *iclog) { - SPLDECL(s); - - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); if (iclog->ic_state == XLOG_STATE_ACTIVE) { xlog_state_switch_iclogs(log, iclog, 0); @@ -3172,7 +3159,7 @@ xlog_state_want_sync(xlog_t *log, xlog_in_core_t *iclog) (XLOG_STATE_WANT_SYNC|XLOG_STATE_IOERROR)); } - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); } /* xlog_state_want_sync */ @@ -3193,16 +3180,15 @@ xlog_state_ticket_alloc(xlog_t *log) xlog_ticket_t *t_list; xlog_ticket_t *next; xfs_caddr_t buf; - uint i = (NBPP / sizeof(xlog_ticket_t)) - 2; - SPLDECL(s); + uint i = (PAGE_SIZE / sizeof(xlog_ticket_t)) - 2; /* * The kmem_zalloc may sleep, so we shouldn't be holding the * global lock. XXXmiken: may want to use zone allocator. */ - buf = (xfs_caddr_t) kmem_zalloc(NBPP, KM_SLEEP); + buf = (xfs_caddr_t) kmem_zalloc(PAGE_SIZE, KM_SLEEP); - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); /* Attach 1st ticket to Q, so we can keep track of allocated memory */ t_list = (xlog_ticket_t *)buf; @@ -3231,7 +3217,7 @@ xlog_state_ticket_alloc(xlog_t *log) } t_list->t_next = NULL; log->l_tail = t_list; - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); } /* xlog_state_ticket_alloc */ @@ -3273,7 +3259,7 @@ xlog_ticket_put(xlog_t *log, /* * Grab ticket off freelist or allocation some more */ -xlog_ticket_t * +STATIC xlog_ticket_t * xlog_ticket_get(xlog_t *log, int unit_bytes, int cnt, @@ -3282,15 +3268,14 @@ xlog_ticket_get(xlog_t *log, { xlog_ticket_t *tic; uint num_headers; - SPLDECL(s); alloc: if (log->l_freelist == NULL) xlog_state_ticket_alloc(log); /* potentially sleep */ - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); if (log->l_freelist == NULL) { - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); goto alloc; } tic = log->l_freelist; @@ -3298,7 +3283,7 @@ xlog_ticket_get(xlog_t *log, if (log->l_freelist == NULL) log->l_tail = NULL; log->l_ticket_cnt--; - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); /* * Permanent reservations have up to 'cnt'-1 active log operations @@ -3349,7 +3334,7 @@ xlog_ticket_get(xlog_t *log, unit_bytes += sizeof(xlog_op_header_t) * num_headers; /* for roundoff padding for transaction data and one for commit record */ - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) && + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb) && log->l_mp->m_sb.sb_logsunit > 1) { /* log su roundoff */ unit_bytes += 2*log->l_mp->m_sb.sb_logsunit; @@ -3473,10 +3458,9 @@ xlog_verify_iclog(xlog_t *log, __uint8_t clientid; int len, i, j, k, op_len; int idx; - SPLDECL(s); /* check validity of iclog pointers */ - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); icptr = log->l_iclog; for (i=0; i < log->l_iclog_bufs; i++) { if (icptr == NULL) @@ -3485,21 +3469,21 @@ xlog_verify_iclog(xlog_t *log, } if (icptr != log->l_iclog) xlog_panic("xlog_verify_iclog: corrupt iclog ring"); - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); /* check log magic numbers */ - ptr = (xfs_caddr_t) &(iclog->ic_header); - if (INT_GET(*(uint *)ptr, ARCH_CONVERT) != XLOG_HEADER_MAGIC_NUM) + if (be32_to_cpu(iclog->ic_header.h_magicno) != XLOG_HEADER_MAGIC_NUM) xlog_panic("xlog_verify_iclog: invalid magic num"); - for (ptr += BBSIZE; ptr < ((xfs_caddr_t)&(iclog->ic_header))+count; + ptr = (xfs_caddr_t) &iclog->ic_header; + for (ptr += BBSIZE; ptr < ((xfs_caddr_t)&iclog->ic_header) + count; ptr += BBSIZE) { - if (INT_GET(*(uint *)ptr, ARCH_CONVERT) == XLOG_HEADER_MAGIC_NUM) + if (be32_to_cpu(*(__be32 *)ptr) == XLOG_HEADER_MAGIC_NUM) xlog_panic("xlog_verify_iclog: unexpected magic num"); } /* check fields */ - len = INT_GET(iclog->ic_header.h_num_logops, ARCH_CONVERT); + len = be32_to_cpu(iclog->ic_header.h_num_logops); ptr = iclog->ic_datap; base_ptr = ptr; ophead = (xlog_op_header_t *)ptr; @@ -3517,9 +3501,11 @@ xlog_verify_iclog(xlog_t *log, if (idx >= (XLOG_HEADER_CYCLE_SIZE / BBSIZE)) { j = idx / (XLOG_HEADER_CYCLE_SIZE / BBSIZE); k = idx % (XLOG_HEADER_CYCLE_SIZE / BBSIZE); - clientid = GET_CLIENT_ID(xhdr[j].hic_xheader.xh_cycle_data[k], ARCH_CONVERT); + clientid = xlog_get_client_id( + xhdr[j].hic_xheader.xh_cycle_data[k]); } else { - clientid = GET_CLIENT_ID(iclog->ic_header.h_cycle_data[idx], ARCH_CONVERT); + clientid = xlog_get_client_id( + iclog->ic_header.h_cycle_data[idx]); } } if (clientid != XFS_TRANSACTION && clientid != XFS_LOG) @@ -3531,16 +3517,16 @@ xlog_verify_iclog(xlog_t *log, field_offset = (__psint_t) ((xfs_caddr_t)&(ophead->oh_len) - base_ptr); if (syncing == B_FALSE || (field_offset & 0x1ff)) { - op_len = INT_GET(ophead->oh_len, ARCH_CONVERT); + op_len = be32_to_cpu(ophead->oh_len); } else { idx = BTOBBT((__psint_t)&ophead->oh_len - (__psint_t)iclog->ic_datap); if (idx >= (XLOG_HEADER_CYCLE_SIZE / BBSIZE)) { j = idx / (XLOG_HEADER_CYCLE_SIZE / BBSIZE); k = idx % (XLOG_HEADER_CYCLE_SIZE / BBSIZE); - op_len = INT_GET(xhdr[j].hic_xheader.xh_cycle_data[k], ARCH_CONVERT); + op_len = be32_to_cpu(xhdr[j].hic_xheader.xh_cycle_data[k]); } else { - op_len = INT_GET(iclog->ic_header.h_cycle_data[idx], ARCH_CONVERT); + op_len = be32_to_cpu(iclog->ic_header.h_cycle_data[idx]); } } ptr += sizeof(xlog_op_header_t) + op_len; @@ -3549,7 +3535,7 @@ xlog_verify_iclog(xlog_t *log, #endif /* - * Mark all iclogs IOERROR. LOG_LOCK is held by the caller. + * Mark all iclogs IOERROR. l_icloglock is held by the caller. */ STATIC int xlog_state_ioerror( @@ -3597,8 +3583,6 @@ xfs_log_force_umount( xlog_t *log; int retval; int dummy; - SPLDECL(s); - SPLDECL(s2); log = mp->m_log; @@ -3627,8 +3611,8 @@ xfs_log_force_umount( * before we mark the filesystem SHUTDOWN and wake * everybody up to tell the bad news. */ - s = GRANT_LOCK(log); - s2 = LOG_LOCK(log); + spin_lock(&log->l_grant_lock); + spin_lock(&log->l_icloglock); mp->m_flags |= XFS_MOUNT_FS_SHUTDOWN; XFS_BUF_DONE(mp->m_sb_bp); /* @@ -3644,7 +3628,7 @@ xfs_log_force_umount( */ if (logerror) retval = xlog_state_ioerror(log); - LOG_UNLOCK(log, s2); + spin_unlock(&log->l_icloglock); /* * We don't want anybody waiting for log reservations @@ -3667,7 +3651,7 @@ xfs_log_force_umount( tic = tic->t_next; } while (tic != log->l_write_headq); } - GRANT_UNLOCK(log, s); + spin_unlock(&log->l_grant_lock); if (! (log->l_iclog->ic_state & XLOG_STATE_IOERROR)) { ASSERT(!logerror); @@ -3676,9 +3660,9 @@ xfs_log_force_umount( * log down completely. */ xlog_state_sync_all(log, XFS_LOG_FORCE|XFS_LOG_SYNC, &dummy); - s2 = LOG_LOCK(log); + spin_lock(&log->l_icloglock); retval = xlog_state_ioerror(log); - LOG_UNLOCK(log, s2); + spin_unlock(&log->l_icloglock); } /* * Wake up everybody waiting on xfs_log_force. @@ -3691,13 +3675,13 @@ xfs_log_force_umount( { xlog_in_core_t *iclog; - s = LOG_LOCK(log); + spin_lock(&log->l_icloglock); iclog = log->l_iclog; do { ASSERT(iclog->ic_callback == 0); iclog = iclog->ic_next; } while (iclog != log->l_iclog); - LOG_UNLOCK(log, s); + spin_unlock(&log->l_icloglock); } #endif /* return non-zero if log IOERROR transition had already happened */ diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h index ebbe93f4f97..4cdac048df5 100644 --- a/fs/xfs/xfs_log.h +++ b/fs/xfs/xfs_log.h @@ -22,8 +22,9 @@ #define CYCLE_LSN(lsn) ((uint)((lsn)>>32)) #define BLOCK_LSN(lsn) ((uint)(lsn)) + /* this is used in a spot where we might otherwise double-endian-flip */ -#define CYCLE_LSN_DISK(lsn) (((uint *)&(lsn))[0]) +#define CYCLE_LSN_DISK(lsn) (((__be32 *)&(lsn))[0]) #ifdef __KERNEL__ /* diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index 752f964b369..c6244cc733c 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -49,38 +49,27 @@ struct xfs_mount; #define XLOG_HEADER_SIZE 512 #define XLOG_REC_SHIFT(log) \ - BTOBB(1 << (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) ? \ + BTOBB(1 << (xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? \ XLOG_MAX_RECORD_BSHIFT : XLOG_BIG_RECORD_BSHIFT)) #define XLOG_TOTAL_REC_SHIFT(log) \ - BTOBB(XLOG_MAX_ICLOGS << (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) ? \ + BTOBB(XLOG_MAX_ICLOGS << (xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? \ XLOG_MAX_RECORD_BSHIFT : XLOG_BIG_RECORD_BSHIFT)) -/* - * set lsns - */ -#define ASSIGN_ANY_LSN_HOST(lsn,cycle,block) \ - { \ - (lsn) = ((xfs_lsn_t)(cycle)<<32)|(block); \ - } -#define ASSIGN_ANY_LSN_DISK(lsn,cycle,block) \ - { \ - INT_SET(((uint *)&(lsn))[0], ARCH_CONVERT, (cycle)); \ - INT_SET(((uint *)&(lsn))[1], ARCH_CONVERT, (block)); \ - } -#define ASSIGN_LSN(lsn,log) \ - ASSIGN_ANY_LSN_DISK(lsn,(log)->l_curr_cycle,(log)->l_curr_block); - -#define XLOG_SET(f,b) (((f) & (b)) == (b)) - -#define GET_CYCLE(ptr, arch) \ - (INT_GET(*(uint *)(ptr), arch) == XLOG_HEADER_MAGIC_NUM ? \ - INT_GET(*((uint *)(ptr)+1), arch) : \ - INT_GET(*(uint *)(ptr), arch) \ - ) +static inline xfs_lsn_t xlog_assign_lsn(uint cycle, uint block) +{ + return ((xfs_lsn_t)cycle << 32) | block; +} -#define BLK_AVG(blk1, blk2) ((blk1+blk2) >> 1) +static inline uint xlog_get_cycle(char *ptr) +{ + if (be32_to_cpu(*(__be32 *)ptr) == XLOG_HEADER_MAGIC_NUM) + return be32_to_cpu(*((__be32 *)ptr + 1)); + else + return be32_to_cpu(*(__be32 *)ptr); +} +#define BLK_AVG(blk1, blk2) ((blk1+blk2) >> 1) #ifdef __KERNEL__ @@ -96,19 +85,10 @@ struct xfs_mount; * * this has endian issues, of course. */ - -#ifndef XFS_NATIVE_HOST -#define GET_CLIENT_ID(i,arch) \ - ((i) & 0xff) -#else -#define GET_CLIENT_ID(i,arch) \ - ((i) >> 24) -#endif - -#define GRANT_LOCK(log) mutex_spinlock(&(log)->l_grant_lock) -#define GRANT_UNLOCK(log, s) mutex_spinunlock(&(log)->l_grant_lock, s) -#define LOG_LOCK(log) mutex_spinlock(&(log)->l_icloglock) -#define LOG_UNLOCK(log, s) mutex_spinunlock(&(log)->l_icloglock, s) +static inline uint xlog_get_client_id(__be32 i) +{ + return be32_to_cpu(i) >> 24; +} #define xlog_panic(args...) cmn_err(CE_PANIC, ## args) #define xlog_exit(args...) cmn_err(CE_PANIC, ## args) @@ -285,11 +265,11 @@ typedef struct xlog_ticket { typedef struct xlog_op_header { - xlog_tid_t oh_tid; /* transaction id of operation : 4 b */ - int oh_len; /* bytes in data region : 4 b */ - __uint8_t oh_clientid; /* who sent me this : 1 b */ - __uint8_t oh_flags; /* : 1 b */ - ushort oh_res2; /* 32 bit align : 2 b */ + __be32 oh_tid; /* transaction id of operation : 4 b */ + __be32 oh_len; /* bytes in data region : 4 b */ + __u8 oh_clientid; /* who sent me this : 1 b */ + __u8 oh_flags; /* : 1 b */ + __u16 oh_res2; /* 32 bit align : 2 b */ } xlog_op_header_t; @@ -307,25 +287,25 @@ typedef struct xlog_op_header { #endif typedef struct xlog_rec_header { - uint h_magicno; /* log record (LR) identifier : 4 */ - uint h_cycle; /* write cycle of log : 4 */ - int h_version; /* LR version : 4 */ - int h_len; /* len in bytes; should be 64-bit aligned: 4 */ - xfs_lsn_t h_lsn; /* lsn of this LR : 8 */ - xfs_lsn_t h_tail_lsn; /* lsn of 1st LR w/ buffers not committed: 8 */ - uint h_chksum; /* may not be used; non-zero if used : 4 */ - int h_prev_block; /* block number to previous LR : 4 */ - int h_num_logops; /* number of log operations in this LR : 4 */ - uint h_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; + __be32 h_magicno; /* log record (LR) identifier : 4 */ + __be32 h_cycle; /* write cycle of log : 4 */ + __be32 h_version; /* LR version : 4 */ + __be32 h_len; /* len in bytes; should be 64-bit aligned: 4 */ + __be64 h_lsn; /* lsn of this LR : 8 */ + __be64 h_tail_lsn; /* lsn of 1st LR w/ buffers not committed: 8 */ + __be32 h_chksum; /* may not be used; non-zero if used : 4 */ + __be32 h_prev_block; /* block number to previous LR : 4 */ + __be32 h_num_logops; /* number of log operations in this LR : 4 */ + __be32 h_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; /* new fields */ - int h_fmt; /* format of log record : 4 */ - uuid_t h_fs_uuid; /* uuid of FS : 16 */ - int h_size; /* iclog size : 4 */ + __be32 h_fmt; /* format of log record : 4 */ + uuid_t h_fs_uuid; /* uuid of FS : 16 */ + __be32 h_size; /* iclog size : 4 */ } xlog_rec_header_t; typedef struct xlog_rec_ext_header { - uint xh_cycle; /* write cycle of log : 4 */ - uint xh_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; /* : 256 */ + __be32 xh_cycle; /* write cycle of log : 4 */ + __be32 xh_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; /* : 256 */ } xlog_rec_ext_header_t; #ifdef __KERNEL__ @@ -415,7 +395,7 @@ typedef struct log { xlog_ticket_t *l_unmount_free;/* kmem_free these addresses */ xlog_ticket_t *l_tail; /* free list of tickets */ xlog_in_core_t *l_iclog; /* head log queue */ - lock_t l_icloglock; /* grab to change iclog state */ + spinlock_t l_icloglock; /* grab to change iclog state */ xfs_lsn_t l_tail_lsn; /* lsn of 1st LR with unflushed * buffers */ xfs_lsn_t l_last_sync_lsn;/* lsn of last LR on disk */ @@ -439,7 +419,7 @@ typedef struct log { char *l_iclog_bak[XLOG_MAX_ICLOGS]; /* The following block of fields are changed while holding grant_lock */ - lock_t l_grant_lock; + spinlock_t l_grant_lock; xlog_ticket_t *l_reserve_headq; xlog_ticket_t *l_write_headq; int l_grant_reserve_cycle; diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 851eca8a715..b2b70eba282 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -198,7 +198,7 @@ xlog_header_check_dump( cmn_err(CE_DEBUG, " log : uuid = "); for (b = 0; b < 16; b++) cmn_err(CE_DEBUG, "%02x",((uchar_t *)&head->h_fs_uuid)[b]); - cmn_err(CE_DEBUG, ", fmt = %d\n", INT_GET(head->h_fmt, ARCH_CONVERT)); + cmn_err(CE_DEBUG, ", fmt = %d\n", be32_to_cpu(head->h_fmt)); } #else #define xlog_header_check_dump(mp, head) @@ -212,14 +212,14 @@ xlog_header_check_recover( xfs_mount_t *mp, xlog_rec_header_t *head) { - ASSERT(INT_GET(head->h_magicno, ARCH_CONVERT) == XLOG_HEADER_MAGIC_NUM); + ASSERT(be32_to_cpu(head->h_magicno) == XLOG_HEADER_MAGIC_NUM); /* * IRIX doesn't write the h_fmt field and leaves it zeroed * (XLOG_FMT_UNKNOWN). This stops us from trying to recover * a dirty log created in IRIX. */ - if (unlikely(INT_GET(head->h_fmt, ARCH_CONVERT) != XLOG_FMT)) { + if (unlikely(be32_to_cpu(head->h_fmt) != XLOG_FMT)) { xlog_warn( "XFS: dirty log written in incompatible format - can't recover"); xlog_header_check_dump(mp, head); @@ -245,7 +245,7 @@ xlog_header_check_mount( xfs_mount_t *mp, xlog_rec_header_t *head) { - ASSERT(INT_GET(head->h_magicno, ARCH_CONVERT) == XLOG_HEADER_MAGIC_NUM); + ASSERT(be32_to_cpu(head->h_magicno) == XLOG_HEADER_MAGIC_NUM); if (uuid_is_nil(&head->h_fs_uuid)) { /* @@ -293,7 +293,7 @@ xlog_recover_iodone( * Note that the algorithm can not be perfect because the disk will not * necessarily be perfect. */ -int +STATIC int xlog_find_cycle_start( xlog_t *log, xfs_buf_t *bp, @@ -311,7 +311,7 @@ xlog_find_cycle_start( if ((error = xlog_bread(log, mid_blk, 1, bp))) return error; offset = xlog_align(log, mid_blk, 1, bp); - mid_cycle = GET_CYCLE(offset, ARCH_CONVERT); + mid_cycle = xlog_get_cycle(offset); if (mid_cycle == cycle) { *last_blk = mid_blk; /* last_half_cycle == mid_cycle */ @@ -371,7 +371,7 @@ xlog_find_verify_cycle( buf = xlog_align(log, i, bcount, bp); for (j = 0; j < bcount; j++) { - cycle = GET_CYCLE(buf, ARCH_CONVERT); + cycle = xlog_get_cycle(buf); if (cycle == stop_on_cycle_no) { *new_blk = i+j; goto out; @@ -447,8 +447,7 @@ xlog_find_verify_log_record( head = (xlog_rec_header_t *)offset; - if (XLOG_HEADER_MAGIC_NUM == - INT_GET(head->h_magicno, ARCH_CONVERT)) + if (XLOG_HEADER_MAGIC_NUM == be32_to_cpu(head->h_magicno)) break; if (!smallmem) @@ -479,8 +478,8 @@ xlog_find_verify_log_record( * reset last_blk. Only when last_blk points in the middle of a log * record do we update last_blk. */ - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { - uint h_size = INT_GET(head->h_size, ARCH_CONVERT); + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { + uint h_size = be32_to_cpu(head->h_size); xhdrs = h_size / XLOG_HEADER_CYCLE_SIZE; if (h_size % XLOG_HEADER_CYCLE_SIZE) @@ -489,8 +488,8 @@ xlog_find_verify_log_record( xhdrs = 1; } - if (*last_blk - i + extra_bblks - != BTOBB(INT_GET(head->h_len, ARCH_CONVERT)) + xhdrs) + if (*last_blk - i + extra_bblks != + BTOBB(be32_to_cpu(head->h_len)) + xhdrs) *last_blk = i; out: @@ -550,13 +549,13 @@ xlog_find_head( if ((error = xlog_bread(log, 0, 1, bp))) goto bp_err; offset = xlog_align(log, 0, 1, bp); - first_half_cycle = GET_CYCLE(offset, ARCH_CONVERT); + first_half_cycle = xlog_get_cycle(offset); last_blk = head_blk = log_bbnum - 1; /* get cycle # of last block */ if ((error = xlog_bread(log, last_blk, 1, bp))) goto bp_err; offset = xlog_align(log, last_blk, 1, bp); - last_half_cycle = GET_CYCLE(offset, ARCH_CONVERT); + last_half_cycle = xlog_get_cycle(offset); ASSERT(last_half_cycle != 0); /* @@ -808,7 +807,7 @@ xlog_find_tail( if ((error = xlog_bread(log, 0, 1, bp))) goto bread_err; offset = xlog_align(log, 0, 1, bp); - if (GET_CYCLE(offset, ARCH_CONVERT) == 0) { + if (xlog_get_cycle(offset) == 0) { *tail_blk = 0; /* leave all other log inited values alone */ goto exit; @@ -823,8 +822,7 @@ xlog_find_tail( if ((error = xlog_bread(log, i, 1, bp))) goto bread_err; offset = xlog_align(log, i, 1, bp); - if (XLOG_HEADER_MAGIC_NUM == - INT_GET(*(uint *)offset, ARCH_CONVERT)) { + if (XLOG_HEADER_MAGIC_NUM == be32_to_cpu(*(__be32 *)offset)) { found = 1; break; } @@ -841,7 +839,7 @@ xlog_find_tail( goto bread_err; offset = xlog_align(log, i, 1, bp); if (XLOG_HEADER_MAGIC_NUM == - INT_GET(*(uint*)offset, ARCH_CONVERT)) { + be32_to_cpu(*(__be32 *)offset)) { found = 2; break; } @@ -855,7 +853,7 @@ xlog_find_tail( /* find blk_no of tail of log */ rhead = (xlog_rec_header_t *)offset; - *tail_blk = BLOCK_LSN(INT_GET(rhead->h_tail_lsn, ARCH_CONVERT)); + *tail_blk = BLOCK_LSN(be64_to_cpu(rhead->h_tail_lsn)); /* * Reset log values according to the state of the log when we @@ -869,11 +867,11 @@ xlog_find_tail( */ log->l_prev_block = i; log->l_curr_block = (int)*head_blk; - log->l_curr_cycle = INT_GET(rhead->h_cycle, ARCH_CONVERT); + log->l_curr_cycle = be32_to_cpu(rhead->h_cycle); if (found == 2) log->l_curr_cycle++; - log->l_tail_lsn = INT_GET(rhead->h_tail_lsn, ARCH_CONVERT); - log->l_last_sync_lsn = INT_GET(rhead->h_lsn, ARCH_CONVERT); + log->l_tail_lsn = be64_to_cpu(rhead->h_tail_lsn); + log->l_last_sync_lsn = be64_to_cpu(rhead->h_lsn); log->l_grant_reserve_cycle = log->l_curr_cycle; log->l_grant_reserve_bytes = BBTOB(log->l_curr_block); log->l_grant_write_cycle = log->l_curr_cycle; @@ -890,9 +888,9 @@ xlog_find_tail( * unmount record if there is one, so we pass the lsn of the * unmount record rather than the block after it. */ - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { - int h_size = INT_GET(rhead->h_size, ARCH_CONVERT); - int h_version = INT_GET(rhead->h_version, ARCH_CONVERT); + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { + int h_size = be32_to_cpu(rhead->h_size); + int h_version = be32_to_cpu(rhead->h_version); if ((h_version & XLOG_VERSION_2) && (h_size > XLOG_HEADER_CYCLE_SIZE)) { @@ -906,10 +904,10 @@ xlog_find_tail( hblks = 1; } after_umount_blk = (i + hblks + (int) - BTOBB(INT_GET(rhead->h_len, ARCH_CONVERT))) % log->l_logBBsize; + BTOBB(be32_to_cpu(rhead->h_len))) % log->l_logBBsize; tail_lsn = log->l_tail_lsn; if (*head_blk == after_umount_blk && - INT_GET(rhead->h_num_logops, ARCH_CONVERT) == 1) { + be32_to_cpu(rhead->h_num_logops) == 1) { umount_data_blk = (i + hblks) % log->l_logBBsize; if ((error = xlog_bread(log, umount_data_blk, 1, bp))) { goto bread_err; @@ -922,10 +920,12 @@ xlog_find_tail( * log records will point recovery to after the * current unmount record. */ - ASSIGN_ANY_LSN_HOST(log->l_tail_lsn, log->l_curr_cycle, - after_umount_blk); - ASSIGN_ANY_LSN_HOST(log->l_last_sync_lsn, log->l_curr_cycle, - after_umount_blk); + log->l_tail_lsn = + xlog_assign_lsn(log->l_curr_cycle, + after_umount_blk); + log->l_last_sync_lsn = + xlog_assign_lsn(log->l_curr_cycle, + after_umount_blk); *tail_blk = after_umount_blk; /* @@ -986,7 +986,7 @@ exit: * -1 => use *blk_no as the first block of the log * >0 => error has occurred */ -int +STATIC int xlog_find_zeroed( xlog_t *log, xfs_daddr_t *blk_no) @@ -1007,7 +1007,7 @@ xlog_find_zeroed( if ((error = xlog_bread(log, 0, 1, bp))) goto bp_err; offset = xlog_align(log, 0, 1, bp); - first_cycle = GET_CYCLE(offset, ARCH_CONVERT); + first_cycle = xlog_get_cycle(offset); if (first_cycle == 0) { /* completely zeroed log */ *blk_no = 0; xlog_put_bp(bp); @@ -1018,7 +1018,7 @@ xlog_find_zeroed( if ((error = xlog_bread(log, log_bbnum-1, 1, bp))) goto bp_err; offset = xlog_align(log, log_bbnum-1, 1, bp); - last_cycle = GET_CYCLE(offset, ARCH_CONVERT); + last_cycle = xlog_get_cycle(offset); if (last_cycle != 0) { /* log completely written to */ xlog_put_bp(bp); return 0; @@ -1098,13 +1098,13 @@ xlog_add_record( xlog_rec_header_t *recp = (xlog_rec_header_t *)buf; memset(buf, 0, BBSIZE); - INT_SET(recp->h_magicno, ARCH_CONVERT, XLOG_HEADER_MAGIC_NUM); - INT_SET(recp->h_cycle, ARCH_CONVERT, cycle); - INT_SET(recp->h_version, ARCH_CONVERT, - XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) ? 2 : 1); - ASSIGN_ANY_LSN_DISK(recp->h_lsn, cycle, block); - ASSIGN_ANY_LSN_DISK(recp->h_tail_lsn, tail_cycle, tail_block); - INT_SET(recp->h_fmt, ARCH_CONVERT, XLOG_FMT); + recp->h_magicno = cpu_to_be32(XLOG_HEADER_MAGIC_NUM); + recp->h_cycle = cpu_to_be32(cycle); + recp->h_version = cpu_to_be32( + xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? 2 : 1); + recp->h_lsn = cpu_to_be64(xlog_assign_lsn(cycle, block)); + recp->h_tail_lsn = cpu_to_be64(xlog_assign_lsn(tail_cycle, tail_block)); + recp->h_fmt = cpu_to_be32(XLOG_FMT); memcpy(&recp->h_fs_uuid, &log->l_mp->m_sb.sb_uuid, sizeof(uuid_t)); } @@ -2211,7 +2211,7 @@ xlog_recover_do_buffer_trans( * overlap with future reads of those inodes. */ if (XFS_DINODE_MAGIC == - INT_GET(*((__uint16_t *)(xfs_buf_offset(bp, 0))), ARCH_CONVERT) && + be16_to_cpu(*((__be16 *)xfs_buf_offset(bp, 0))) && (XFS_BUF_COUNT(bp) != MAX(log->l_mp->m_sb.sb_blocksize, (__uint32_t)XFS_INODE_CLUSTER_SIZE(log->l_mp)))) { XFS_BUF_STALE(bp); @@ -2581,8 +2581,7 @@ xlog_recover_do_dquot_trans( /* * This type of quotas was turned off, so ignore this record. */ - type = INT_GET(recddq->d_flags, ARCH_CONVERT) & - (XFS_DQ_USER | XFS_DQ_PROJ | XFS_DQ_GROUP); + type = recddq->d_flags & (XFS_DQ_USER | XFS_DQ_PROJ | XFS_DQ_GROUP); ASSERT(type); if (log->l_quotaoffs_flag & type) return (0); @@ -2660,7 +2659,6 @@ xlog_recover_do_efi_trans( xfs_mount_t *mp; xfs_efi_log_item_t *efip; xfs_efi_log_format_t *efi_formatp; - SPLDECL(s); if (pass == XLOG_RECOVER_PASS1) { return 0; @@ -2678,11 +2676,11 @@ xlog_recover_do_efi_trans( efip->efi_next_extent = efi_formatp->efi_nextents; efip->efi_flags |= XFS_EFI_COMMITTED; - AIL_LOCK(mp,s); + spin_lock(&mp->m_ail_lock); /* * xfs_trans_update_ail() drops the AIL lock. */ - xfs_trans_update_ail(mp, (xfs_log_item_t *)efip, lsn, s); + xfs_trans_update_ail(mp, (xfs_log_item_t *)efip, lsn); return 0; } @@ -2707,7 +2705,6 @@ xlog_recover_do_efd_trans( xfs_log_item_t *lip; int gen; __uint64_t efi_id; - SPLDECL(s); if (pass == XLOG_RECOVER_PASS1) { return; @@ -2725,7 +2722,7 @@ xlog_recover_do_efd_trans( * in the AIL. */ mp = log->l_mp; - AIL_LOCK(mp,s); + spin_lock(&mp->m_ail_lock); lip = xfs_trans_first_ail(mp, &gen); while (lip != NULL) { if (lip->li_type == XFS_LI_EFI) { @@ -2735,22 +2732,14 @@ xlog_recover_do_efd_trans( * xfs_trans_delete_ail() drops the * AIL lock. */ - xfs_trans_delete_ail(mp, lip, s); - break; + xfs_trans_delete_ail(mp, lip); + xfs_efi_item_free(efip); + return; } } lip = xfs_trans_next_ail(mp, lip, &gen, NULL); } - - /* - * If we found it, then free it up. If it wasn't there, it - * must have been overwritten in the log. Oh well. - */ - if (lip != NULL) { - xfs_efi_item_free(efip); - } else { - AIL_UNLOCK(mp, s); - } + spin_unlock(&mp->m_ail_lock); } /* @@ -2897,8 +2886,8 @@ xlog_recover_process_data( unsigned long hash; uint flags; - lp = dp + INT_GET(rhead->h_len, ARCH_CONVERT); - num_logops = INT_GET(rhead->h_num_logops, ARCH_CONVERT); + lp = dp + be32_to_cpu(rhead->h_len); + num_logops = be32_to_cpu(rhead->h_num_logops); /* check the log format matches our own - else we can't recover */ if (xlog_header_check_recover(log->l_mp, rhead)) @@ -2915,15 +2904,20 @@ xlog_recover_process_data( ASSERT(0); return (XFS_ERROR(EIO)); } - tid = INT_GET(ohead->oh_tid, ARCH_CONVERT); + tid = be32_to_cpu(ohead->oh_tid); hash = XLOG_RHASH(tid); trans = xlog_recover_find_tid(rhash[hash], tid); if (trans == NULL) { /* not found; add new tid */ if (ohead->oh_flags & XLOG_START_TRANS) xlog_recover_new_tid(&rhash[hash], tid, - INT_GET(rhead->h_lsn, ARCH_CONVERT)); + be64_to_cpu(rhead->h_lsn)); } else { - ASSERT(dp+INT_GET(ohead->oh_len, ARCH_CONVERT) <= lp); + if (dp + be32_to_cpu(ohead->oh_len) > lp) { + xlog_warn( + "XFS: xlog_recover_process_data: bad length"); + WARN_ON(1); + return (XFS_ERROR(EIO)); + } flags = ohead->oh_flags & ~XLOG_END_TRANS; if (flags & XLOG_WAS_CONT_TRANS) flags &= ~XLOG_CONTINUE_TRANS; @@ -2937,8 +2931,7 @@ xlog_recover_process_data( break; case XLOG_WAS_CONT_TRANS: error = xlog_recover_add_to_cont_trans(trans, - dp, INT_GET(ohead->oh_len, - ARCH_CONVERT)); + dp, be32_to_cpu(ohead->oh_len)); break; case XLOG_START_TRANS: xlog_warn( @@ -2949,8 +2942,7 @@ xlog_recover_process_data( case 0: case XLOG_CONTINUE_TRANS: error = xlog_recover_add_to_trans(trans, - dp, INT_GET(ohead->oh_len, - ARCH_CONVERT)); + dp, be32_to_cpu(ohead->oh_len)); break; default: xlog_warn( @@ -2962,7 +2954,7 @@ xlog_recover_process_data( if (error) return error; } - dp += INT_GET(ohead->oh_len, ARCH_CONVERT); + dp += be32_to_cpu(ohead->oh_len); num_logops--; } return 0; @@ -3075,10 +3067,9 @@ xlog_recover_process_efis( xfs_efi_log_item_t *efip; int gen; xfs_mount_t *mp; - SPLDECL(s); mp = log->l_mp; - AIL_LOCK(mp,s); + spin_lock(&mp->m_ail_lock); lip = xfs_trans_first_ail(mp, &gen); while (lip != NULL) { @@ -3099,12 +3090,12 @@ xlog_recover_process_efis( continue; } - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); xlog_recover_process_efi(mp, efip); - AIL_LOCK(mp,s); + spin_lock(&mp->m_ail_lock); lip = xfs_trans_next_ail(mp, lip, &gen, NULL); } - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); } /* @@ -3315,16 +3306,16 @@ xlog_pack_data_checksum( int size) { int i; - uint *up; + __be32 *up; uint chksum = 0; - up = (uint *)iclog->ic_datap; + up = (__be32 *)iclog->ic_datap; /* divide length by 4 to get # words */ for (i = 0; i < (size >> 2); i++) { - chksum ^= INT_GET(*up, ARCH_CONVERT); + chksum ^= be32_to_cpu(*up); up++; } - INT_SET(iclog->ic_header.h_chksum, ARCH_CONVERT, chksum); + iclog->ic_header.h_chksum = cpu_to_be32(chksum); } #else #define xlog_pack_data_checksum(log, iclog, size) @@ -3341,7 +3332,7 @@ xlog_pack_data( { int i, j, k; int size = iclog->ic_offset + roundoff; - uint cycle_lsn; + __be32 cycle_lsn; xfs_caddr_t dp; xlog_in_core_2_t *xhdr; @@ -3352,18 +3343,18 @@ xlog_pack_data( dp = iclog->ic_datap; for (i = 0; i < BTOBB(size) && i < (XLOG_HEADER_CYCLE_SIZE / BBSIZE); i++) { - iclog->ic_header.h_cycle_data[i] = *(uint *)dp; - *(uint *)dp = cycle_lsn; + iclog->ic_header.h_cycle_data[i] = *(__be32 *)dp; + *(__be32 *)dp = cycle_lsn; dp += BBSIZE; } - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { xhdr = (xlog_in_core_2_t *)&iclog->ic_header; for ( ; i < BTOBB(size); i++) { j = i / (XLOG_HEADER_CYCLE_SIZE / BBSIZE); k = i % (XLOG_HEADER_CYCLE_SIZE / BBSIZE); - xhdr[j].hic_xheader.xh_cycle_data[k] = *(uint *)dp; - *(uint *)dp = cycle_lsn; + xhdr[j].hic_xheader.xh_cycle_data[k] = *(__be32 *)dp; + *(__be32 *)dp = cycle_lsn; dp += BBSIZE; } @@ -3380,24 +3371,24 @@ xlog_unpack_data_checksum( xfs_caddr_t dp, xlog_t *log) { - uint *up = (uint *)dp; + __be32 *up = (__be32 *)dp; uint chksum = 0; int i; /* divide length by 4 to get # words */ - for (i=0; i < INT_GET(rhead->h_len, ARCH_CONVERT) >> 2; i++) { - chksum ^= INT_GET(*up, ARCH_CONVERT); + for (i=0; i < be32_to_cpu(rhead->h_len) >> 2; i++) { + chksum ^= be32_to_cpu(*up); up++; } - if (chksum != INT_GET(rhead->h_chksum, ARCH_CONVERT)) { + if (chksum != be32_to_cpu(rhead->h_chksum)) { if (rhead->h_chksum || ((log->l_flags & XLOG_CHKSUM_MISMATCH) == 0)) { cmn_err(CE_DEBUG, "XFS: LogR chksum mismatch: was (0x%x) is (0x%x)\n", - INT_GET(rhead->h_chksum, ARCH_CONVERT), chksum); + be32_to_cpu(rhead->h_chksum), chksum); cmn_err(CE_DEBUG, "XFS: Disregard message if filesystem was created with non-DEBUG kernel"); - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { cmn_err(CE_DEBUG, "XFS: LogR this is a LogV2 filesystem\n"); } @@ -3418,18 +3409,18 @@ xlog_unpack_data( int i, j, k; xlog_in_core_2_t *xhdr; - for (i = 0; i < BTOBB(INT_GET(rhead->h_len, ARCH_CONVERT)) && + for (i = 0; i < BTOBB(be32_to_cpu(rhead->h_len)) && i < (XLOG_HEADER_CYCLE_SIZE / BBSIZE); i++) { - *(uint *)dp = *(uint *)&rhead->h_cycle_data[i]; + *(__be32 *)dp = *(__be32 *)&rhead->h_cycle_data[i]; dp += BBSIZE; } - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { xhdr = (xlog_in_core_2_t *)rhead; - for ( ; i < BTOBB(INT_GET(rhead->h_len, ARCH_CONVERT)); i++) { + for ( ; i < BTOBB(be32_to_cpu(rhead->h_len)); i++) { j = i / (XLOG_HEADER_CYCLE_SIZE / BBSIZE); k = i % (XLOG_HEADER_CYCLE_SIZE / BBSIZE); - *(uint *)dp = xhdr[j].hic_xheader.xh_cycle_data[k]; + *(__be32 *)dp = xhdr[j].hic_xheader.xh_cycle_data[k]; dp += BBSIZE; } } @@ -3445,24 +3436,21 @@ xlog_valid_rec_header( { int hlen; - if (unlikely( - (INT_GET(rhead->h_magicno, ARCH_CONVERT) != - XLOG_HEADER_MAGIC_NUM))) { + if (unlikely(be32_to_cpu(rhead->h_magicno) != XLOG_HEADER_MAGIC_NUM)) { XFS_ERROR_REPORT("xlog_valid_rec_header(1)", XFS_ERRLEVEL_LOW, log->l_mp); return XFS_ERROR(EFSCORRUPTED); } if (unlikely( (!rhead->h_version || - (INT_GET(rhead->h_version, ARCH_CONVERT) & - (~XLOG_VERSION_OKBITS)) != 0))) { + (be32_to_cpu(rhead->h_version) & (~XLOG_VERSION_OKBITS))))) { xlog_warn("XFS: %s: unrecognised log version (%d).", - __FUNCTION__, INT_GET(rhead->h_version, ARCH_CONVERT)); + __FUNCTION__, be32_to_cpu(rhead->h_version)); return XFS_ERROR(EIO); } /* LR body must have data or it wouldn't have been written */ - hlen = INT_GET(rhead->h_len, ARCH_CONVERT); + hlen = be32_to_cpu(rhead->h_len); if (unlikely( hlen <= 0 || hlen > INT_MAX )) { XFS_ERROR_REPORT("xlog_valid_rec_header(2)", XFS_ERRLEVEL_LOW, log->l_mp); @@ -3506,7 +3494,7 @@ xlog_do_recovery_pass( * Read the header of the tail block and get the iclog buffer size from * h_size. Use this to tell how many sectors make up the log header. */ - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { /* * When using variable length iclogs, read first sector of * iclog header and extract the header size from it. Get a @@ -3522,9 +3510,8 @@ xlog_do_recovery_pass( error = xlog_valid_rec_header(log, rhead, tail_blk); if (error) goto bread_err1; - h_size = INT_GET(rhead->h_size, ARCH_CONVERT); - if ((INT_GET(rhead->h_version, ARCH_CONVERT) - & XLOG_VERSION_2) && + h_size = be32_to_cpu(rhead->h_size); + if ((be32_to_cpu(rhead->h_version) & XLOG_VERSION_2) && (h_size > XLOG_HEADER_CYCLE_SIZE)) { hblks = h_size / XLOG_HEADER_CYCLE_SIZE; if (h_size % XLOG_HEADER_CYCLE_SIZE) @@ -3561,7 +3548,7 @@ xlog_do_recovery_pass( goto bread_err2; /* blocks in data section */ - bblks = (int)BTOBB(INT_GET(rhead->h_len, ARCH_CONVERT)); + bblks = (int)BTOBB(be32_to_cpu(rhead->h_len)); error = xlog_bread(log, blk_no + hblks, bblks, dbp); if (error) goto bread_err2; @@ -3636,7 +3623,7 @@ xlog_do_recovery_pass( if (error) goto bread_err2; - bblks = (int)BTOBB(INT_GET(rhead->h_len, ARCH_CONVERT)); + bblks = (int)BTOBB(be32_to_cpu(rhead->h_len)); blk_no += hblks; /* Read in data for log record */ @@ -3707,7 +3694,7 @@ xlog_do_recovery_pass( error = xlog_valid_rec_header(log, rhead, blk_no); if (error) goto bread_err2; - bblks = (int)BTOBB(INT_GET(rhead->h_len, ARCH_CONVERT)); + bblks = (int)BTOBB(be32_to_cpu(rhead->h_len)); if ((error = xlog_bread(log, blk_no+hblks, bblks, dbp))) goto bread_err2; offset = xlog_align(log, blk_no+hblks, bblks, dbp); @@ -3851,7 +3838,7 @@ xlog_do_recover( sbp = &log->l_mp->m_sb; xfs_sb_from_disk(sbp, XFS_BUF_TO_SBP(bp)); ASSERT(sbp->sb_magicnum == XFS_SB_MAGIC); - ASSERT(XFS_SB_GOOD_VERSION(sbp)); + ASSERT(xfs_sb_good_version(sbp)); xfs_buf_relse(bp); /* We've re-read the superblock so re-initialize per-cpu counters */ diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index ebdb76da527..8ed164eb954 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -44,7 +44,7 @@ #include "xfs_quota.h" #include "xfs_fsops.h" -STATIC void xfs_mount_log_sbunit(xfs_mount_t *, __int64_t); +STATIC void xfs_mount_log_sb(xfs_mount_t *, __int64_t); STATIC int xfs_uuid_mount(xfs_mount_t *); STATIC void xfs_uuid_unmount(xfs_mount_t *mp); STATIC void xfs_unmountfs_wait(xfs_mount_t *); @@ -119,6 +119,7 @@ static const struct { { offsetof(xfs_sb_t, sb_logsectsize),0 }, { offsetof(xfs_sb_t, sb_logsunit), 0 }, { offsetof(xfs_sb_t, sb_features2), 0 }, + { offsetof(xfs_sb_t, sb_bad_features2), 0 }, { sizeof(xfs_sb_t), 0 } }; @@ -136,15 +137,9 @@ xfs_mount_init(void) mp->m_flags |= XFS_MOUNT_NO_PERCPU_SB; } - AIL_LOCKINIT(&mp->m_ail_lock, "xfs_ail"); - spinlock_init(&mp->m_sb_lock, "xfs_sb"); + spin_lock_init(&mp->m_sb_lock); mutex_init(&mp->m_ilock); mutex_init(&mp->m_growlock); - /* - * Initialize the AIL. - */ - xfs_trans_ail_init(mp); - atomic_set(&mp->m_active_trans, 0); return mp; @@ -171,7 +166,7 @@ xfs_mount_free( sizeof(xfs_perag_t) * mp->m_sb.sb_agcount); } - AIL_LOCK_DESTROY(&mp->m_ail_lock); + spinlock_destroy(&mp->m_ail_lock); spinlock_destroy(&mp->m_sb_lock); mutex_destroy(&mp->m_ilock); mutex_destroy(&mp->m_growlock); @@ -231,7 +226,7 @@ xfs_mount_validate_sb( return XFS_ERROR(EWRONGFS); } - if (!XFS_SB_GOOD_VERSION(sbp)) { + if (!xfs_sb_good_version(sbp)) { xfs_fs_mount_cmn_err(flags, "bad version"); return XFS_ERROR(EWRONGFS); } @@ -306,7 +301,7 @@ xfs_mount_validate_sb( /* * Version 1 directory format has never worked on Linux. */ - if (unlikely(!XFS_SB_VERSION_HASDIRV2(sbp))) { + if (unlikely(!xfs_sb_version_hasdirv2(sbp))) { xfs_fs_mount_cmn_err(flags, "file system using version 1 directory format"); return XFS_ERROR(ENOSYS); @@ -455,6 +450,7 @@ xfs_sb_from_disk( to->sb_logsectsize = be16_to_cpu(from->sb_logsectsize); to->sb_logsunit = be32_to_cpu(from->sb_logsunit); to->sb_features2 = be32_to_cpu(from->sb_features2); + to->sb_bad_features2 = be32_to_cpu(from->sb_bad_features2); } /* @@ -616,7 +612,7 @@ xfs_mount_common(xfs_mount_t *mp, xfs_sb_t *sbp) int i; mp->m_agfrotor = mp->m_agirotor = 0; - spinlock_init(&mp->m_agirotor_lock, "m_agirotor_lock"); + spin_lock_init(&mp->m_agirotor_lock); mp->m_maxagi = mp->m_sb.sb_agcount; mp->m_blkbit_log = sbp->sb_blocklog + XFS_NBBYLOG; mp->m_blkbb_log = sbp->sb_blocklog - BBSHIFT; @@ -696,7 +692,6 @@ xfs_initialize_perag_data(xfs_mount_t *mp, xfs_agnumber_t agcount) uint64_t bfreelst = 0; uint64_t btree = 0; int error; - int s; for (index = 0; index < agcount; index++) { /* @@ -721,11 +716,11 @@ xfs_initialize_perag_data(xfs_mount_t *mp, xfs_agnumber_t agcount) /* * Overwrite incore superblock counters with just-read data */ - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); sbp->sb_ifree = ifree; sbp->sb_icount = ialloc; sbp->sb_fdblocks = bfree + bfreelst + btree; - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); /* Fixup the per-cpu counters as well. */ xfs_icsb_reinit_counters(mp); @@ -734,49 +729,13 @@ xfs_initialize_perag_data(xfs_mount_t *mp, xfs_agnumber_t agcount) } /* - * xfs_mountfs - * - * This function does the following on an initial mount of a file system: - * - reads the superblock from disk and init the mount struct - * - if we're a 32-bit kernel, do a size check on the superblock - * so we don't mount terabyte filesystems - * - init mount struct realtime fields - * - allocate inode hash table for fs - * - init directory manager - * - perform recovery and init the log manager + * Update alignment values based on mount options and sb values */ -int -xfs_mountfs( - xfs_mount_t *mp, - int mfsi_flags) +STATIC int +xfs_update_alignment(xfs_mount_t *mp, int mfsi_flags, __uint64_t *update_flags) { - xfs_buf_t *bp; xfs_sb_t *sbp = &(mp->m_sb); - xfs_inode_t *rip; - bhv_vnode_t *rvp = NULL; - int readio_log, writeio_log; - xfs_daddr_t d; - __uint64_t resblks; - __int64_t update_flags; - uint quotamount, quotaflags; - int agno; - int uuid_mounted = 0; - int error = 0; - - if (mp->m_sb_bp == NULL) { - if ((error = xfs_readsb(mp, mfsi_flags))) { - return error; - } - } - xfs_mount_common(mp, sbp); - /* - * Check if sb_agblocks is aligned at stripe boundary - * If sb_agblocks is NOT aligned turn off m_dalign since - * allocator alignment is within an ag, therefore ag has - * to be aligned at stripe boundary. - */ - update_flags = 0LL; if (mp->m_dalign && !(mfsi_flags & XFS_MFSI_SECOND)) { /* * If stripe unit and stripe width are not multiples @@ -787,8 +746,7 @@ xfs_mountfs( if (mp->m_flags & XFS_MOUNT_RETERR) { cmn_err(CE_WARN, "XFS: alignment check 1 failed"); - error = XFS_ERROR(EINVAL); - goto error1; + return XFS_ERROR(EINVAL); } mp->m_dalign = mp->m_swidth = 0; } else { @@ -798,8 +756,7 @@ xfs_mountfs( mp->m_dalign = XFS_BB_TO_FSBT(mp, mp->m_dalign); if (mp->m_dalign && (sbp->sb_agblocks % mp->m_dalign)) { if (mp->m_flags & XFS_MOUNT_RETERR) { - error = XFS_ERROR(EINVAL); - goto error1; + return XFS_ERROR(EINVAL); } xfs_fs_cmn_err(CE_WARN, mp, "stripe alignment turned off: sunit(%d)/swidth(%d) incompatible with agsize(%d)", @@ -816,8 +773,7 @@ xfs_mountfs( "stripe alignment turned off: sunit(%d) less than bsize(%d)", mp->m_dalign, mp->m_blockmask +1); - error = XFS_ERROR(EINVAL); - goto error1; + return XFS_ERROR(EINVAL); } mp->m_swidth = 0; } @@ -827,65 +783,61 @@ xfs_mountfs( * Update superblock with new values * and log changes */ - if (XFS_SB_VERSION_HASDALIGN(sbp)) { + if (xfs_sb_version_hasdalign(sbp)) { if (sbp->sb_unit != mp->m_dalign) { sbp->sb_unit = mp->m_dalign; - update_flags |= XFS_SB_UNIT; + *update_flags |= XFS_SB_UNIT; } if (sbp->sb_width != mp->m_swidth) { sbp->sb_width = mp->m_swidth; - update_flags |= XFS_SB_WIDTH; + *update_flags |= XFS_SB_WIDTH; } } } else if ((mp->m_flags & XFS_MOUNT_NOALIGN) != XFS_MOUNT_NOALIGN && - XFS_SB_VERSION_HASDALIGN(&mp->m_sb)) { + xfs_sb_version_hasdalign(&mp->m_sb)) { mp->m_dalign = sbp->sb_unit; mp->m_swidth = sbp->sb_width; } - xfs_alloc_compute_maxlevels(mp); - xfs_bmap_compute_maxlevels(mp, XFS_DATA_FORK); - xfs_bmap_compute_maxlevels(mp, XFS_ATTR_FORK); - xfs_ialloc_compute_maxlevels(mp); + return 0; +} - if (sbp->sb_imax_pct) { - __uint64_t icount; +/* + * Set the maximum inode count for this filesystem + */ +STATIC void +xfs_set_maxicount(xfs_mount_t *mp) +{ + xfs_sb_t *sbp = &(mp->m_sb); + __uint64_t icount; - /* Make sure the maximum inode count is a multiple of the - * units we allocate inodes in. + if (sbp->sb_imax_pct) { + /* + * Make sure the maximum inode count is a multiple + * of the units we allocate inodes in. */ - icount = sbp->sb_dblocks * sbp->sb_imax_pct; do_div(icount, 100); do_div(icount, mp->m_ialloc_blks); mp->m_maxicount = (icount * mp->m_ialloc_blks) << sbp->sb_inopblog; - } else + } else { mp->m_maxicount = 0; - - mp->m_maxioffset = xfs_max_file_offset(sbp->sb_blocklog); - - /* - * XFS uses the uuid from the superblock as the unique - * identifier for fsid. We can not use the uuid from the volume - * since a single partition filesystem is identical to a single - * partition volume/filesystem. - */ - if ((mfsi_flags & XFS_MFSI_SECOND) == 0 && - (mp->m_flags & XFS_MOUNT_NOUUID) == 0) { - if (xfs_uuid_mount(mp)) { - error = XFS_ERROR(EINVAL); - goto error1; - } - uuid_mounted=1; } +} + +/* + * Set the default minimum read and write sizes unless + * already specified in a mount option. + * We use smaller I/O sizes when the file system + * is being used for NFS service (wsync mount option). + */ +STATIC void +xfs_set_rw_sizes(xfs_mount_t *mp) +{ + xfs_sb_t *sbp = &(mp->m_sb); + int readio_log, writeio_log; - /* - * Set the default minimum read and write sizes unless - * already specified in a mount option. - * We use smaller I/O sizes when the file system - * is being used for NFS service (wsync mount option). - */ if (!(mp->m_flags & XFS_MOUNT_DFLT_IOSIZE)) { if (mp->m_flags & XFS_MOUNT_WSYNC) { readio_log = XFS_WSYNC_READIO_LOG; @@ -911,18 +863,15 @@ xfs_mountfs( mp->m_writeio_log = writeio_log; } mp->m_writeio_blocks = 1 << (mp->m_writeio_log - sbp->sb_blocklog); +} - /* - * Set the inode cluster size. - * This may still be overridden by the file system - * block size if it is larger than the chosen cluster size. - */ - mp->m_inode_cluster_size = XFS_INODE_BIG_CLUSTER_SIZE; - - /* - * Set whether we're using inode alignment. - */ - if (XFS_SB_VERSION_HASALIGN(&mp->m_sb) && +/* + * Set whether we're using inode alignment. + */ +STATIC void +xfs_set_inoalignment(xfs_mount_t *mp) +{ + if (xfs_sb_version_hasalign(&mp->m_sb) && mp->m_sb.sb_inoalignmt >= XFS_B_TO_FSBT(mp, mp->m_inode_cluster_size)) mp->m_inoalign_mask = mp->m_sb.sb_inoalignmt - 1; @@ -937,14 +886,22 @@ xfs_mountfs( mp->m_sinoalign = mp->m_dalign; else mp->m_sinoalign = 0; - /* - * Check that the data (and log if separate) are an ok size. - */ +} + +/* + * Check that the data (and log if separate) are an ok size. + */ +STATIC int +xfs_check_sizes(xfs_mount_t *mp, int mfsi_flags) +{ + xfs_buf_t *bp; + xfs_daddr_t d; + int error; + d = (xfs_daddr_t)XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks); if (XFS_BB_TO_FSB(mp, d) != mp->m_sb.sb_dblocks) { cmn_err(CE_WARN, "XFS: size check 1 failed"); - error = XFS_ERROR(E2BIG); - goto error1; + return XFS_ERROR(E2BIG); } error = xfs_read_buf(mp, mp->m_ddev_targp, d - XFS_FSS_TO_BB(mp, 1), @@ -953,10 +910,9 @@ xfs_mountfs( xfs_buf_relse(bp); } else { cmn_err(CE_WARN, "XFS: size check 2 failed"); - if (error == ENOSPC) { + if (error == ENOSPC) error = XFS_ERROR(E2BIG); - } - goto error1; + return error; } if (((mfsi_flags & XFS_MFSI_CLIENT) == 0) && @@ -964,8 +920,7 @@ xfs_mountfs( d = (xfs_daddr_t)XFS_FSB_TO_BB(mp, mp->m_sb.sb_logblocks); if (XFS_BB_TO_FSB(mp, d) != mp->m_sb.sb_logblocks) { cmn_err(CE_WARN, "XFS: size check 3 failed"); - error = XFS_ERROR(E2BIG); - goto error1; + return XFS_ERROR(E2BIG); } error = xfs_read_buf(mp, mp->m_logdev_targp, d - XFS_FSB_TO_BB(mp, 1), @@ -974,17 +929,143 @@ xfs_mountfs( xfs_buf_relse(bp); } else { cmn_err(CE_WARN, "XFS: size check 3 failed"); - if (error == ENOSPC) { + if (error == ENOSPC) error = XFS_ERROR(E2BIG); - } + return error; + } + } + return 0; +} + +/* + * xfs_mountfs + * + * This function does the following on an initial mount of a file system: + * - reads the superblock from disk and init the mount struct + * - if we're a 32-bit kernel, do a size check on the superblock + * so we don't mount terabyte filesystems + * - init mount struct realtime fields + * - allocate inode hash table for fs + * - init directory manager + * - perform recovery and init the log manager + */ +int +xfs_mountfs( + xfs_mount_t *mp, + int mfsi_flags) +{ + xfs_sb_t *sbp = &(mp->m_sb); + xfs_inode_t *rip; + bhv_vnode_t *rvp = NULL; + __uint64_t resblks; + __int64_t update_flags = 0LL; + uint quotamount, quotaflags; + int agno; + int uuid_mounted = 0; + int error = 0; + + if (mp->m_sb_bp == NULL) { + error = xfs_readsb(mp, mfsi_flags); + if (error) + return error; + } + xfs_mount_common(mp, sbp); + + /* + * Check for a mismatched features2 values. Older kernels + * read & wrote into the wrong sb offset for sb_features2 + * on some platforms due to xfs_sb_t not being 64bit size aligned + * when sb_features2 was added, which made older superblock + * reading/writing routines swap it as a 64-bit value. + * + * For backwards compatibility, we make both slots equal. + * + * If we detect a mismatched field, we OR the set bits into the + * existing features2 field in case it has already been modified; we + * don't want to lose any features. We then update the bad location + * with the ORed value so that older kernels will see any features2 + * flags, and mark the two fields as needing updates once the + * transaction subsystem is online. + */ + if (xfs_sb_has_mismatched_features2(sbp)) { + cmn_err(CE_WARN, + "XFS: correcting sb_features alignment problem"); + sbp->sb_features2 |= sbp->sb_bad_features2; + sbp->sb_bad_features2 = sbp->sb_features2; + update_flags |= XFS_SB_FEATURES2 | XFS_SB_BAD_FEATURES2; + + /* + * Re-check for ATTR2 in case it was found in bad_features2 + * slot. + */ + if (xfs_sb_version_hasattr2(&mp->m_sb)) + mp->m_flags |= XFS_MOUNT_ATTR2; + + } + + /* + * Check if sb_agblocks is aligned at stripe boundary + * If sb_agblocks is NOT aligned turn off m_dalign since + * allocator alignment is within an ag, therefore ag has + * to be aligned at stripe boundary. + */ + error = xfs_update_alignment(mp, mfsi_flags, &update_flags); + if (error) + goto error1; + + xfs_alloc_compute_maxlevels(mp); + xfs_bmap_compute_maxlevels(mp, XFS_DATA_FORK); + xfs_bmap_compute_maxlevels(mp, XFS_ATTR_FORK); + xfs_ialloc_compute_maxlevels(mp); + + xfs_set_maxicount(mp); + + mp->m_maxioffset = xfs_max_file_offset(sbp->sb_blocklog); + + /* + * XFS uses the uuid from the superblock as the unique + * identifier for fsid. We can not use the uuid from the volume + * since a single partition filesystem is identical to a single + * partition volume/filesystem. + */ + if ((mfsi_flags & XFS_MFSI_SECOND) == 0 && + (mp->m_flags & XFS_MOUNT_NOUUID) == 0) { + if (xfs_uuid_mount(mp)) { + error = XFS_ERROR(EINVAL); goto error1; } + uuid_mounted=1; } /* + * Set the minimum read and write sizes + */ + xfs_set_rw_sizes(mp); + + /* + * Set the inode cluster size. + * This may still be overridden by the file system + * block size if it is larger than the chosen cluster size. + */ + mp->m_inode_cluster_size = XFS_INODE_BIG_CLUSTER_SIZE; + + /* + * Set inode alignment fields + */ + xfs_set_inoalignment(mp); + + /* + * Check that the data (and log if separate) are an ok size. + */ + error = xfs_check_sizes(mp, mfsi_flags); + if (error) + goto error1; + + /* * Initialize realtime fields in the mount structure */ - if ((error = xfs_rtmount_init(mp))) { + error = xfs_rtmount_init(mp); + if (error) { cmn_err(CE_WARN, "XFS: RT mount failed"); goto error1; } @@ -1102,7 +1183,8 @@ xfs_mountfs( /* * Initialize realtime inode pointers in the mount structure */ - if ((error = xfs_rtmount_inodes(mp))) { + error = xfs_rtmount_inodes(mp); + if (error) { /* * Free up the root inode. */ @@ -1111,16 +1193,16 @@ xfs_mountfs( } /* - * If fs is not mounted readonly, then update the superblock - * unit and width changes. + * If fs is not mounted readonly, then update the superblock changes. */ if (update_flags && !(mp->m_flags & XFS_MOUNT_RDONLY)) - xfs_mount_log_sbunit(mp, update_flags); + xfs_mount_log_sb(mp, update_flags); /* * Initialise the XFS quota management subsystem for this mount */ - if ((error = XFS_QM_INIT(mp, "amount, "aflags))) + error = XFS_QM_INIT(mp, "amount, "aflags); + if (error) goto error4; /* @@ -1137,7 +1219,8 @@ xfs_mountfs( /* * Complete the quota initialisation, post-log-replay component. */ - if ((error = XFS_QM_MOUNT(mp, quotamount, quotaflags, mfsi_flags))) + error = XFS_QM_MOUNT(mp, quotamount, quotaflags, mfsi_flags); + if (error) goto error4; /* @@ -1255,7 +1338,6 @@ xfs_unmountfs(xfs_mount_t *mp, struct cred *cr) #if defined(DEBUG) || defined(INDUCE_IO_ERROR) xfs_errortag_clearall(mp, 0); #endif - XFS_IODONE(mp); xfs_mount_free(mp); return 0; } @@ -1441,7 +1523,7 @@ xfs_mod_sb(xfs_trans_t *tp, __int64_t fields) * Fields are not allowed to dip below zero, so if the delta would * do this do not apply it and return EINVAL. * - * The SB_LOCK must be held when this routine is called. + * The m_sb_lock must be held when this routine is called. */ int xfs_mod_incore_sb_unlocked( @@ -1606,7 +1688,7 @@ xfs_mod_incore_sb_unlocked( /* * xfs_mod_incore_sb() is used to change a field in the in-core * superblock structure by the specified delta. This modification - * is protected by the SB_LOCK. Just use the xfs_mod_incore_sb_unlocked() + * is protected by the m_sb_lock. Just use the xfs_mod_incore_sb_unlocked() * routine to do the work. */ int @@ -1616,7 +1698,6 @@ xfs_mod_incore_sb( int64_t delta, int rsvd) { - unsigned long s; int status; /* check for per-cpu counters */ @@ -1633,9 +1714,9 @@ xfs_mod_incore_sb( /* FALLTHROUGH */ #endif default: - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); status = xfs_mod_incore_sb_unlocked(mp, field, delta, rsvd); - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); break; } @@ -1656,7 +1737,6 @@ xfs_mod_incore_sb( int xfs_mod_incore_sb_batch(xfs_mount_t *mp, xfs_mod_sb_t *msb, uint nmsb, int rsvd) { - unsigned long s; int status=0; xfs_mod_sb_t *msbp; @@ -1664,10 +1744,10 @@ xfs_mod_incore_sb_batch(xfs_mount_t *mp, xfs_mod_sb_t *msb, uint nmsb, int rsvd) * Loop through the array of mod structures and apply each * individually. If any fail, then back out all those * which have already been applied. Do all of this within - * the scope of the SB_LOCK so that all of the changes will + * the scope of the m_sb_lock so that all of the changes will * be atomic. */ - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); msbp = &msb[0]; for (msbp = &msbp[0]; msbp < (msb + nmsb); msbp++) { /* @@ -1681,11 +1761,11 @@ xfs_mod_incore_sb_batch(xfs_mount_t *mp, xfs_mod_sb_t *msb, uint nmsb, int rsvd) case XFS_SBS_IFREE: case XFS_SBS_FDBLOCKS: if (!(mp->m_flags & XFS_MOUNT_NO_PERCPU_SB)) { - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); status = xfs_icsb_modify_counters(mp, msbp->msb_field, msbp->msb_delta, rsvd); - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); break; } /* FALLTHROUGH */ @@ -1719,12 +1799,12 @@ xfs_mod_incore_sb_batch(xfs_mount_t *mp, xfs_mod_sb_t *msb, uint nmsb, int rsvd) case XFS_SBS_IFREE: case XFS_SBS_FDBLOCKS: if (!(mp->m_flags & XFS_MOUNT_NO_PERCPU_SB)) { - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); status = xfs_icsb_modify_counters(mp, msbp->msb_field, -(msbp->msb_delta), rsvd); - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); break; } /* FALLTHROUGH */ @@ -1740,7 +1820,7 @@ xfs_mod_incore_sb_batch(xfs_mount_t *mp, xfs_mod_sb_t *msb, uint nmsb, int rsvd) msbp--; } } - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); return status; } @@ -1828,16 +1908,18 @@ xfs_uuid_unmount( /* * Used to log changes to the superblock unit and width fields which could - * be altered by the mount options. Only the first superblock is updated. + * be altered by the mount options, as well as any potential sb_features2 + * fixup. Only the first superblock is updated. */ STATIC void -xfs_mount_log_sbunit( +xfs_mount_log_sb( xfs_mount_t *mp, __int64_t fields) { xfs_trans_t *tp; - ASSERT(fields & (XFS_SB_UNIT|XFS_SB_WIDTH|XFS_SB_UUID)); + ASSERT(fields & (XFS_SB_UNIT | XFS_SB_WIDTH | XFS_SB_UUID | + XFS_SB_FEATURES2 | XFS_SB_BAD_FEATURES2)); tp = xfs_trans_alloc(mp, XFS_TRANS_SB_UNIT); if (xfs_trans_reserve(tp, 0, mp->m_sb.sb_sectsize + 128, 0, 0, @@ -1888,12 +1970,12 @@ xfs_mount_log_sbunit( * * Locking rules: * - * 1. XFS_SB_LOCK() before picking up per-cpu locks + * 1. m_sb_lock before picking up per-cpu locks * 2. per-cpu locks always picked up via for_each_online_cpu() order - * 3. accurate counter sync requires XFS_SB_LOCK + per cpu locks + * 3. accurate counter sync requires m_sb_lock + per cpu locks * 4. modifying per-cpu counters requires holding per-cpu lock - * 5. modifying global counters requires holding XFS_SB_LOCK - * 6. enabling or disabling a counter requires holding the XFS_SB_LOCK + * 5. modifying global counters requires holding m_sb_lock + * 6. enabling or disabling a counter requires holding the m_sb_lock * and _none_ of the per-cpu locks. * * Disabled counters are only ever re-enabled by a balance operation @@ -1920,7 +2002,6 @@ xfs_icsb_cpu_notify( { xfs_icsb_cnts_t *cntp; xfs_mount_t *mp; - int s; mp = (xfs_mount_t *)container_of(nfb, xfs_mount_t, m_icsb_notifier); cntp = (xfs_icsb_cnts_t *) @@ -1946,7 +2027,7 @@ xfs_icsb_cpu_notify( * count into the total on the global superblock and * re-enable the counters. */ xfs_icsb_lock(mp); - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); xfs_icsb_disable_counter(mp, XFS_SBS_ICOUNT); xfs_icsb_disable_counter(mp, XFS_SBS_IFREE); xfs_icsb_disable_counter(mp, XFS_SBS_FDBLOCKS); @@ -1963,7 +2044,7 @@ xfs_icsb_cpu_notify( XFS_ICSB_SB_LOCKED, 0); xfs_icsb_balance_counter(mp, XFS_SBS_FDBLOCKS, XFS_ICSB_SB_LOCKED, 0); - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); xfs_icsb_unlock(mp); break; } @@ -2194,11 +2275,10 @@ xfs_icsb_sync_counters_flags( int flags) { xfs_icsb_cnts_t cnt; - int s; /* Pass 1: lock all counters */ if ((flags & XFS_ICSB_SB_LOCKED) == 0) - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); xfs_icsb_count(mp, &cnt, flags); @@ -2211,7 +2291,7 @@ xfs_icsb_sync_counters_flags( mp->m_sb.sb_fdblocks = cnt.icsb_fdblocks; if ((flags & XFS_ICSB_SB_LOCKED) == 0) - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); } /* @@ -2252,11 +2332,10 @@ xfs_icsb_balance_counter( { uint64_t count, resid; int weight = num_online_cpus(); - int s; uint64_t min = (uint64_t)min_per_cpu; if (!(flags & XFS_ICSB_SB_LOCKED)) - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); /* disable counter and sync counter */ xfs_icsb_disable_counter(mp, field); @@ -2290,10 +2369,10 @@ xfs_icsb_balance_counter( xfs_icsb_enable_counter(mp, field, count, resid); out: if (!(flags & XFS_ICSB_SB_LOCKED)) - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); } -int +STATIC int xfs_icsb_modify_counters( xfs_mount_t *mp, xfs_sb_field_t field, @@ -2302,7 +2381,7 @@ xfs_icsb_modify_counters( { xfs_icsb_cnts_t *icsbp; long long lcounter; /* long counter for 64 bit fields */ - int cpu, ret = 0, s; + int cpu, ret = 0; might_sleep(); again: @@ -2380,15 +2459,15 @@ slow_path: * running atomically here, we know a rebalance cannot * be in progress. Hence we can go straight to operating * on the global superblock. We do not call xfs_mod_incore_sb() - * here even though we need to get the SB_LOCK. Doing so + * here even though we need to get the m_sb_lock. Doing so * will cause us to re-enter this function and deadlock. - * Hence we get the SB_LOCK ourselves and then call + * Hence we get the m_sb_lock ourselves and then call * xfs_mod_incore_sb_unlocked() as the unlocked path operates * directly on the global counters. */ - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); ret = xfs_mod_incore_sb_unlocked(mp, field, delta, rsvd); - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); /* * Now that we've modified the global superblock, we diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index c618f7cb5f0..1d8a4728d84 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -56,20 +56,12 @@ struct cred; struct log; struct xfs_mount_args; struct xfs_inode; -struct xfs_iocore; struct xfs_bmbt_irec; struct xfs_bmap_free; struct xfs_extdelta; struct xfs_swapext; struct xfs_mru_cache; -#define AIL_LOCK_T lock_t -#define AIL_LOCKINIT(x,y) spinlock_init(x,y) -#define AIL_LOCK_DESTROY(x) spinlock_destroy(x) -#define AIL_LOCK(mp,s) s=mutex_spinlock(&(mp)->m_ail_lock) -#define AIL_UNLOCK(mp,s) mutex_spinunlock(&(mp)->m_ail_lock, s) - - /* * Prototypes and functions for the Data Migration subsystem. */ @@ -196,105 +188,6 @@ typedef struct xfs_qmops { #define XFS_QM_QUOTACTL(mp, cmd, id, addr) \ (*(mp)->m_qm_ops->xfs_quotactl)(mp, cmd, id, addr) - -/* - * Prototypes and functions for I/O core modularization. - */ - -typedef int (*xfs_ioinit_t)(struct xfs_mount *, - struct xfs_mount_args *, int); -typedef int (*xfs_bmapi_t)(struct xfs_trans *, void *, - xfs_fileoff_t, xfs_filblks_t, int, - xfs_fsblock_t *, xfs_extlen_t, - struct xfs_bmbt_irec *, int *, - struct xfs_bmap_free *, struct xfs_extdelta *); -typedef int (*xfs_bunmapi_t)(struct xfs_trans *, - void *, xfs_fileoff_t, - xfs_filblks_t, int, xfs_extnum_t, - xfs_fsblock_t *, struct xfs_bmap_free *, - struct xfs_extdelta *, int *); -typedef int (*xfs_bmap_eof_t)(void *, xfs_fileoff_t, int, int *); -typedef int (*xfs_iomap_write_direct_t)( - void *, xfs_off_t, size_t, int, - struct xfs_bmbt_irec *, int *, int); -typedef int (*xfs_iomap_write_delay_t)( - void *, xfs_off_t, size_t, int, - struct xfs_bmbt_irec *, int *); -typedef int (*xfs_iomap_write_allocate_t)( - void *, xfs_off_t, size_t, - struct xfs_bmbt_irec *, int *); -typedef int (*xfs_iomap_write_unwritten_t)( - void *, xfs_off_t, size_t); -typedef uint (*xfs_lck_map_shared_t)(void *); -typedef void (*xfs_lock_t)(void *, uint); -typedef void (*xfs_lock_demote_t)(void *, uint); -typedef int (*xfs_lock_nowait_t)(void *, uint); -typedef void (*xfs_unlk_t)(void *, unsigned int); -typedef xfs_fsize_t (*xfs_size_t)(void *); -typedef xfs_fsize_t (*xfs_iodone_t)(struct xfs_mount *); -typedef int (*xfs_swap_extents_t)(void *, void *, - struct xfs_swapext*); - -typedef struct xfs_ioops { - xfs_ioinit_t xfs_ioinit; - xfs_bmapi_t xfs_bmapi_func; - xfs_bunmapi_t xfs_bunmapi_func; - xfs_bmap_eof_t xfs_bmap_eof_func; - xfs_iomap_write_direct_t xfs_iomap_write_direct; - xfs_iomap_write_delay_t xfs_iomap_write_delay; - xfs_iomap_write_allocate_t xfs_iomap_write_allocate; - xfs_iomap_write_unwritten_t xfs_iomap_write_unwritten; - xfs_lock_t xfs_ilock; - xfs_lck_map_shared_t xfs_lck_map_shared; - xfs_lock_demote_t xfs_ilock_demote; - xfs_lock_nowait_t xfs_ilock_nowait; - xfs_unlk_t xfs_unlock; - xfs_size_t xfs_size_func; - xfs_iodone_t xfs_iodone; - xfs_swap_extents_t xfs_swap_extents_func; -} xfs_ioops_t; - -#define XFS_IOINIT(mp, args, flags) \ - (*(mp)->m_io_ops.xfs_ioinit)(mp, args, flags) -#define XFS_BMAPI(mp, trans,io,bno,len,f,first,tot,mval,nmap,flist,delta) \ - (*(mp)->m_io_ops.xfs_bmapi_func) \ - (trans,(io)->io_obj,bno,len,f,first,tot,mval,nmap,flist,delta) -#define XFS_BUNMAPI(mp, trans,io,bno,len,f,nexts,first,flist,delta,done) \ - (*(mp)->m_io_ops.xfs_bunmapi_func) \ - (trans,(io)->io_obj,bno,len,f,nexts,first,flist,delta,done) -#define XFS_BMAP_EOF(mp, io, endoff, whichfork, eof) \ - (*(mp)->m_io_ops.xfs_bmap_eof_func) \ - ((io)->io_obj, endoff, whichfork, eof) -#define XFS_IOMAP_WRITE_DIRECT(mp, io, offset, count, flags, mval, nmap, found)\ - (*(mp)->m_io_ops.xfs_iomap_write_direct) \ - ((io)->io_obj, offset, count, flags, mval, nmap, found) -#define XFS_IOMAP_WRITE_DELAY(mp, io, offset, count, flags, mval, nmap) \ - (*(mp)->m_io_ops.xfs_iomap_write_delay) \ - ((io)->io_obj, offset, count, flags, mval, nmap) -#define XFS_IOMAP_WRITE_ALLOCATE(mp, io, offset, count, mval, nmap) \ - (*(mp)->m_io_ops.xfs_iomap_write_allocate) \ - ((io)->io_obj, offset, count, mval, nmap) -#define XFS_IOMAP_WRITE_UNWRITTEN(mp, io, offset, count) \ - (*(mp)->m_io_ops.xfs_iomap_write_unwritten) \ - ((io)->io_obj, offset, count) -#define XFS_LCK_MAP_SHARED(mp, io) \ - (*(mp)->m_io_ops.xfs_lck_map_shared)((io)->io_obj) -#define XFS_ILOCK(mp, io, mode) \ - (*(mp)->m_io_ops.xfs_ilock)((io)->io_obj, mode) -#define XFS_ILOCK_NOWAIT(mp, io, mode) \ - (*(mp)->m_io_ops.xfs_ilock_nowait)((io)->io_obj, mode) -#define XFS_IUNLOCK(mp, io, mode) \ - (*(mp)->m_io_ops.xfs_unlock)((io)->io_obj, mode) -#define XFS_ILOCK_DEMOTE(mp, io, mode) \ - (*(mp)->m_io_ops.xfs_ilock_demote)((io)->io_obj, mode) -#define XFS_SIZE(mp, io) \ - (*(mp)->m_io_ops.xfs_size_func)((io)->io_obj) -#define XFS_IODONE(mp) \ - (*(mp)->m_io_ops.xfs_iodone)(mp) -#define XFS_SWAP_EXTENTS(mp, io, tio, sxp) \ - (*(mp)->m_io_ops.xfs_swap_extents_func) \ - ((io)->io_obj, (tio)->io_obj, sxp) - #ifdef HAVE_PERCPU_SB /* @@ -326,14 +219,20 @@ extern void xfs_icsb_sync_counters_flags(struct xfs_mount *, int); #define xfs_icsb_sync_counters_flags(mp, flags) do { } while (0) #endif +typedef struct xfs_ail { + xfs_ail_entry_t xa_ail; + uint xa_gen; + struct task_struct *xa_task; + xfs_lsn_t xa_target; +} xfs_ail_t; + typedef struct xfs_mount { struct super_block *m_super; xfs_tid_t m_tid; /* next unused tid for fs */ - AIL_LOCK_T m_ail_lock; /* fs AIL mutex */ - xfs_ail_entry_t m_ail; /* fs active log item list */ - uint m_ail_gen; /* fs AIL generation count */ + spinlock_t m_ail_lock; /* fs AIL mutex */ + xfs_ail_t m_ail; /* fs active log item list */ xfs_sb_t m_sb; /* copy of fs superblock */ - lock_t m_sb_lock; /* sb counter mutex */ + spinlock_t m_sb_lock; /* sb counter lock */ struct xfs_buf *m_sb_bp; /* buffer for superblock */ char *m_fsname; /* filesystem name */ int m_fsname_len; /* strlen of fs name */ @@ -342,7 +241,7 @@ typedef struct xfs_mount { int m_bsize; /* fs logical block size */ xfs_agnumber_t m_agfrotor; /* last ag where space found */ xfs_agnumber_t m_agirotor; /* last ag dir inode alloced */ - lock_t m_agirotor_lock;/* .. and lock protecting it */ + spinlock_t m_agirotor_lock;/* .. and lock protecting it */ xfs_agnumber_t m_maxagi; /* highest inode alloc group */ struct xfs_inode *m_inodes; /* active inode list */ struct list_head m_del_inodes; /* inodes to reclaim */ @@ -423,7 +322,6 @@ typedef struct xfs_mount { * hash table */ struct xfs_dmops *m_dm_ops; /* vector of DMI ops */ struct xfs_qmops *m_qm_ops; /* vector of XQM ops */ - struct xfs_ioops m_io_ops; /* vector of I/O ops */ atomic_t m_active_trans; /* number trans frozen */ #ifdef HAVE_PERCPU_SB xfs_icsb_cnts_t *m_sb_cnts; /* per-cpu superblock counters */ @@ -468,7 +366,7 @@ typedef struct xfs_mount { #define XFS_MOUNT_SMALL_INUMS (1ULL << 15) /* users wants 32bit inodes */ #define XFS_MOUNT_NOUUID (1ULL << 16) /* ignore uuid during mount */ #define XFS_MOUNT_BARRIER (1ULL << 17) -#define XFS_MOUNT_IDELETE (1ULL << 18) /* delete empty inode clusters*/ +#define XFS_MOUNT_IKEEP (1ULL << 18) /* keep empty inode clusters*/ #define XFS_MOUNT_SWALLOC (1ULL << 19) /* turn on stripe width * allocation */ #define XFS_MOUNT_RDONLY (1ULL << 20) /* read-only fs */ @@ -610,8 +508,6 @@ typedef struct xfs_mod_sb { #define XFS_MOUNT_ILOCK(mp) mutex_lock(&((mp)->m_ilock)) #define XFS_MOUNT_IUNLOCK(mp) mutex_unlock(&((mp)->m_ilock)) -#define XFS_SB_LOCK(mp) mutex_spinlock(&(mp)->m_sb_lock) -#define XFS_SB_UNLOCK(mp,s) mutex_spinunlock(&(mp)->m_sb_lock,(s)) extern xfs_mount_t *xfs_mount_init(void); extern void xfs_mod_sb(xfs_trans_t *, __int64_t); @@ -646,7 +542,6 @@ extern int xfs_qmops_get(struct xfs_mount *, struct xfs_mount_args *); extern void xfs_qmops_put(struct xfs_mount *); extern struct xfs_dmops xfs_dmcore_xfs; -extern struct xfs_ioops xfs_iocore_xfs; extern int xfs_init(void); extern void xfs_cleanup(void); diff --git a/fs/xfs/xfs_mru_cache.c b/fs/xfs/xfs_mru_cache.c index e0b358c1c53..a0b2c0a2589 100644 --- a/fs/xfs/xfs_mru_cache.c +++ b/fs/xfs/xfs_mru_cache.c @@ -225,10 +225,14 @@ _xfs_mru_cache_list_insert( * list need to be deleted. For each element this involves removing it from the * data store, removing it from the reap list, calling the client's free * function and deleting the element from the element zone. + * + * We get called holding the mru->lock, which we drop and then reacquire. + * Sparse need special help with this to tell it we know what we are doing. */ STATIC void _xfs_mru_cache_clear_reap_list( - xfs_mru_cache_t *mru) + xfs_mru_cache_t *mru) __releases(mru->lock) __acquires(mru->lock) + { xfs_mru_cache_elem_t *elem, *next; struct list_head tmp; @@ -245,7 +249,7 @@ _xfs_mru_cache_clear_reap_list( */ list_move(&elem->list_node, &tmp); } - mutex_spinunlock(&mru->lock, 0); + spin_unlock(&mru->lock); list_for_each_entry_safe(elem, next, &tmp, list_node) { @@ -259,7 +263,7 @@ _xfs_mru_cache_clear_reap_list( kmem_zone_free(xfs_mru_elem_zone, elem); } - mutex_spinlock(&mru->lock); + spin_lock(&mru->lock); } /* @@ -280,7 +284,7 @@ _xfs_mru_cache_reap( if (!mru || !mru->lists) return; - mutex_spinlock(&mru->lock); + spin_lock(&mru->lock); next = _xfs_mru_cache_migrate(mru, jiffies); _xfs_mru_cache_clear_reap_list(mru); @@ -294,7 +298,7 @@ _xfs_mru_cache_reap( queue_delayed_work(xfs_mru_reap_wq, &mru->work, next); } - mutex_spinunlock(&mru->lock, 0); + spin_unlock(&mru->lock); } int @@ -368,7 +372,7 @@ xfs_mru_cache_create( */ INIT_RADIX_TREE(&mru->store, GFP_ATOMIC); INIT_LIST_HEAD(&mru->reap_list); - spinlock_init(&mru->lock, "xfs_mru_cache"); + spin_lock_init(&mru->lock); INIT_DELAYED_WORK(&mru->work, _xfs_mru_cache_reap); mru->grp_time = grp_time; @@ -398,17 +402,17 @@ xfs_mru_cache_flush( if (!mru || !mru->lists) return; - mutex_spinlock(&mru->lock); + spin_lock(&mru->lock); if (mru->queued) { - mutex_spinunlock(&mru->lock, 0); + spin_unlock(&mru->lock); cancel_rearming_delayed_workqueue(xfs_mru_reap_wq, &mru->work); - mutex_spinlock(&mru->lock); + spin_lock(&mru->lock); } _xfs_mru_cache_migrate(mru, jiffies + mru->grp_count * mru->grp_time); _xfs_mru_cache_clear_reap_list(mru); - mutex_spinunlock(&mru->lock, 0); + spin_unlock(&mru->lock); } void @@ -454,13 +458,13 @@ xfs_mru_cache_insert( elem->key = key; elem->value = value; - mutex_spinlock(&mru->lock); + spin_lock(&mru->lock); radix_tree_insert(&mru->store, key, elem); radix_tree_preload_end(); _xfs_mru_cache_list_insert(mru, elem); - mutex_spinunlock(&mru->lock, 0); + spin_unlock(&mru->lock); return 0; } @@ -483,14 +487,14 @@ xfs_mru_cache_remove( if (!mru || !mru->lists) return NULL; - mutex_spinlock(&mru->lock); + spin_lock(&mru->lock); elem = radix_tree_delete(&mru->store, key); if (elem) { value = elem->value; list_del(&elem->list_node); } - mutex_spinunlock(&mru->lock, 0); + spin_unlock(&mru->lock); if (elem) kmem_zone_free(xfs_mru_elem_zone, elem); @@ -528,6 +532,10 @@ xfs_mru_cache_delete( * * If the element isn't found, this function returns NULL and the spinlock is * released. xfs_mru_cache_done() should NOT be called when this occurs. + * + * Because sparse isn't smart enough to know about conditional lock return + * status, we need to help it get it right by annotating the path that does + * not release the lock. */ void * xfs_mru_cache_lookup( @@ -540,14 +548,14 @@ xfs_mru_cache_lookup( if (!mru || !mru->lists) return NULL; - mutex_spinlock(&mru->lock); + spin_lock(&mru->lock); elem = radix_tree_lookup(&mru->store, key); if (elem) { list_del(&elem->list_node); _xfs_mru_cache_list_insert(mru, elem); - } - else - mutex_spinunlock(&mru->lock, 0); + __release(mru_lock); /* help sparse not be stupid */ + } else + spin_unlock(&mru->lock); return elem ? elem->value : NULL; } @@ -571,10 +579,12 @@ xfs_mru_cache_peek( if (!mru || !mru->lists) return NULL; - mutex_spinlock(&mru->lock); + spin_lock(&mru->lock); elem = radix_tree_lookup(&mru->store, key); if (!elem) - mutex_spinunlock(&mru->lock, 0); + spin_unlock(&mru->lock); + else + __release(mru_lock); /* help sparse not be stupid */ return elem ? elem->value : NULL; } @@ -586,7 +596,7 @@ xfs_mru_cache_peek( */ void xfs_mru_cache_done( - xfs_mru_cache_t *mru) + xfs_mru_cache_t *mru) __releases(mru->lock) { - mutex_spinunlock(&mru->lock, 0); + spin_unlock(&mru->lock); } diff --git a/fs/xfs/xfs_qmops.c b/fs/xfs/xfs_qmops.c index 2ec1d8a2735..a294e58db8d 100644 --- a/fs/xfs/xfs_qmops.c +++ b/fs/xfs/xfs_qmops.c @@ -49,18 +49,17 @@ xfs_mount_reset_sbqflags(xfs_mount_t *mp) { int error; xfs_trans_t *tp; - unsigned long s; mp->m_qflags = 0; /* * It is OK to look at sb_qflags here in mount path, - * without SB_LOCK. + * without m_sb_lock. */ if (mp->m_sb.sb_qflags == 0) return 0; - s = XFS_SB_LOCK(mp); + spin_lock(&mp->m_sb_lock); mp->m_sb.sb_qflags = 0; - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); /* * if the fs is readonly, let the incore superblock run diff --git a/fs/xfs/xfs_rename.c b/fs/xfs/xfs_rename.c index 44ea0ba3647..7eb157a59f9 100644 --- a/fs/xfs/xfs_rename.c +++ b/fs/xfs/xfs_rename.c @@ -39,6 +39,7 @@ #include "xfs_refcache.h" #include "xfs_utils.h" #include "xfs_trans_space.h" +#include "xfs_vnodeops.h" /* @@ -118,7 +119,7 @@ xfs_lock_for_rename( inum1 = ip1->i_ino; ASSERT(ip1); - ITRACE(ip1); + xfs_itrace_ref(ip1); /* * Unlock dp1 and lock dp2 if they are different. @@ -141,7 +142,7 @@ xfs_lock_for_rename( IRELE (ip1); return error; } else { - ITRACE(ip2); + xfs_itrace_ref(ip2); } /* @@ -247,8 +248,8 @@ xfs_rename( int src_namelen = VNAMELEN(src_vname); int target_namelen = VNAMELEN(target_vname); - vn_trace_entry(src_dp, "xfs_rename", (inst_t *)__return_address); - vn_trace_entry(xfs_vtoi(target_dir_vp), "xfs_rename", (inst_t *)__return_address); + xfs_itrace_entry(src_dp); + xfs_itrace_entry(xfs_vtoi(target_dir_vp)); /* * Find the XFS behavior descriptor for the target directory diff --git a/fs/xfs/xfs_rtalloc.h b/fs/xfs/xfs_rtalloc.h index 799c1f87126..8d8dcd21571 100644 --- a/fs/xfs/xfs_rtalloc.h +++ b/fs/xfs/xfs_rtalloc.h @@ -21,8 +21,6 @@ struct xfs_mount; struct xfs_trans; -#define XFS_IS_REALTIME_INODE(ip) ((ip)->i_d.di_flags & XFS_DIFLAG_REALTIME) - /* Min and max rt extent sizes, specified in bytes */ #define XFS_MAX_RTEXTSIZE (1024 * 1024 * 1024) /* 1GB */ #define XFS_DFL_RTEXTSIZE (64 * 1024) /* 64KB */ diff --git a/fs/xfs/xfs_rw.h b/fs/xfs/xfs_rw.h index 49875e1d129..f87db5344ce 100644 --- a/fs/xfs/xfs_rw.h +++ b/fs/xfs/xfs_rw.h @@ -32,18 +32,10 @@ struct xfs_mount; static inline xfs_daddr_t xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb) { - return (((ip)->i_d.di_flags & XFS_DIFLAG_REALTIME) ? \ + return (XFS_IS_REALTIME_INODE(ip) ? \ (xfs_daddr_t)XFS_FSB_TO_BB((ip)->i_mount, (fsb)) : \ XFS_FSB_TO_DADDR((ip)->i_mount, (fsb))); } -#define XFS_FSB_TO_DB_IO(io,fsb) xfs_fsb_to_db_io(io,fsb) -static inline xfs_daddr_t -xfs_fsb_to_db_io(struct xfs_iocore *io, xfs_fsblock_t fsb) -{ - return (((io)->io_flags & XFS_IOCORE_RT) ? \ - XFS_FSB_TO_BB((io)->io_mount, (fsb)) : \ - XFS_FSB_TO_DADDR((io)->io_mount, (fsb))); -} /* * Flags for xfs_free_eofblocks @@ -61,7 +53,7 @@ xfs_get_extsz_hint( { xfs_extlen_t extsz; - if (unlikely(ip->i_d.di_flags & XFS_DIFLAG_REALTIME)) { + if (unlikely(XFS_IS_REALTIME_INODE(ip))) { extsz = (ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE) ? ip->i_d.di_extsize : ip->i_mount->m_sb.sb_rextsize; diff --git a/fs/xfs/xfs_sb.h b/fs/xfs/xfs_sb.h index 94660b1a6cc..d904efe7f87 100644 --- a/fs/xfs/xfs_sb.h +++ b/fs/xfs/xfs_sb.h @@ -89,6 +89,7 @@ struct xfs_mount; /* * Superblock - in core version. Must match the ondisk version below. + * Must be padded to 64 bit alignment. */ typedef struct xfs_sb { __uint32_t sb_magicnum; /* magic number == XFS_SB_MAGIC */ @@ -145,10 +146,21 @@ typedef struct xfs_sb { __uint16_t sb_logsectsize; /* sector size for the log, bytes */ __uint32_t sb_logsunit; /* stripe unit size for the log */ __uint32_t sb_features2; /* additional feature bits */ + + /* + * bad features2 field as a result of failing to pad the sb + * structure to 64 bits. Some machines will be using this field + * for features2 bits. Easiest just to mark it bad and not use + * it for anything else. + */ + __uint32_t sb_bad_features2; + + /* must be padded to 64 bit alignment */ } xfs_sb_t; /* - * Superblock - on disk version. Must match the in core version below. + * Superblock - on disk version. Must match the in core version above. + * Must be padded to 64 bit alignment. */ typedef struct xfs_dsb { __be32 sb_magicnum; /* magic number == XFS_SB_MAGIC */ @@ -205,6 +217,15 @@ typedef struct xfs_dsb { __be16 sb_logsectsize; /* sector size for the log, bytes */ __be32 sb_logsunit; /* stripe unit size for the log */ __be32 sb_features2; /* additional feature bits */ + /* + * bad features2 field as a result of failing to pad the sb + * structure to 64 bits. Some machines will be using this field + * for features2 bits. Easiest just to mark it bad and not use + * it for anything else. + */ + __be32 sb_bad_features2; + + /* must be padded to 64 bit alignment */ } xfs_dsb_t; /* @@ -223,7 +244,7 @@ typedef enum { XFS_SBS_GQUOTINO, XFS_SBS_QFLAGS, XFS_SBS_FLAGS, XFS_SBS_SHARED_VN, XFS_SBS_INOALIGNMT, XFS_SBS_UNIT, XFS_SBS_WIDTH, XFS_SBS_DIRBLKLOG, XFS_SBS_LOGSECTLOG, XFS_SBS_LOGSECTSIZE, XFS_SBS_LOGSUNIT, - XFS_SBS_FEATURES2, + XFS_SBS_FEATURES2, XFS_SBS_BAD_FEATURES2, XFS_SBS_FIELDCOUNT } xfs_sb_field_t; @@ -248,13 +269,15 @@ typedef enum { #define XFS_SB_IFREE XFS_SB_MVAL(IFREE) #define XFS_SB_FDBLOCKS XFS_SB_MVAL(FDBLOCKS) #define XFS_SB_FEATURES2 XFS_SB_MVAL(FEATURES2) +#define XFS_SB_BAD_FEATURES2 XFS_SB_MVAL(BAD_FEATURES2) #define XFS_SB_NUM_BITS ((int)XFS_SBS_FIELDCOUNT) #define XFS_SB_ALL_BITS ((1LL << XFS_SB_NUM_BITS) - 1) #define XFS_SB_MOD_BITS \ (XFS_SB_UUID | XFS_SB_ROOTINO | XFS_SB_RBMINO | XFS_SB_RSUMINO | \ XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | XFS_SB_GQUOTINO | \ XFS_SB_QFLAGS | XFS_SB_SHARED_VN | XFS_SB_UNIT | XFS_SB_WIDTH | \ - XFS_SB_ICOUNT | XFS_SB_IFREE | XFS_SB_FDBLOCKS | XFS_SB_FEATURES2) + XFS_SB_ICOUNT | XFS_SB_IFREE | XFS_SB_FDBLOCKS | XFS_SB_FEATURES2 | \ + XFS_SB_BAD_FEATURES2) /* @@ -271,7 +294,6 @@ typedef enum { #define XFS_SB_VERSION_NUM(sbp) ((sbp)->sb_versionnum & XFS_SB_VERSION_NUMBITS) -#define XFS_SB_GOOD_VERSION(sbp) xfs_sb_good_version(sbp) #ifdef __KERNEL__ static inline int xfs_sb_good_version(xfs_sb_t *sbp) { @@ -297,7 +319,15 @@ static inline int xfs_sb_good_version(xfs_sb_t *sbp) } #endif /* __KERNEL__ */ -#define XFS_SB_VERSION_TONEW(v) xfs_sb_version_tonew(v) +/* + * Detect a mismatched features2 field. Older kernels read/wrote + * this into the wrong slot, so to be safe we keep them in sync. + */ +static inline int xfs_sb_has_mismatched_features2(xfs_sb_t *sbp) +{ + return (sbp->sb_bad_features2 != sbp->sb_features2); +} + static inline unsigned xfs_sb_version_tonew(unsigned v) { return ((((v) == XFS_SB_VERSION_1) ? \ @@ -308,7 +338,6 @@ static inline unsigned xfs_sb_version_tonew(unsigned v) XFS_SB_VERSION_4); } -#define XFS_SB_VERSION_TOOLD(v) xfs_sb_version_toold(v) static inline unsigned xfs_sb_version_toold(unsigned v) { return (((v) & (XFS_SB_VERSION_QUOTABIT | XFS_SB_VERSION_ALIGNBIT)) ? \ @@ -320,7 +349,6 @@ static inline unsigned xfs_sb_version_toold(unsigned v) XFS_SB_VERSION_1))); } -#define XFS_SB_VERSION_HASATTR(sbp) xfs_sb_version_hasattr(sbp) static inline int xfs_sb_version_hasattr(xfs_sb_t *sbp) { return ((sbp)->sb_versionnum == XFS_SB_VERSION_2) || \ @@ -329,7 +357,6 @@ static inline int xfs_sb_version_hasattr(xfs_sb_t *sbp) ((sbp)->sb_versionnum & XFS_SB_VERSION_ATTRBIT)); } -#define XFS_SB_VERSION_ADDATTR(sbp) xfs_sb_version_addattr(sbp) static inline void xfs_sb_version_addattr(xfs_sb_t *sbp) { (sbp)->sb_versionnum = (((sbp)->sb_versionnum == XFS_SB_VERSION_1) ? \ @@ -339,7 +366,6 @@ static inline void xfs_sb_version_addattr(xfs_sb_t *sbp) (XFS_SB_VERSION_4 | XFS_SB_VERSION_ATTRBIT))); } -#define XFS_SB_VERSION_HASNLINK(sbp) xfs_sb_version_hasnlink(sbp) static inline int xfs_sb_version_hasnlink(xfs_sb_t *sbp) { return ((sbp)->sb_versionnum == XFS_SB_VERSION_3) || \ @@ -347,7 +373,6 @@ static inline int xfs_sb_version_hasnlink(xfs_sb_t *sbp) ((sbp)->sb_versionnum & XFS_SB_VERSION_NLINKBIT)); } -#define XFS_SB_VERSION_ADDNLINK(sbp) xfs_sb_version_addnlink(sbp) static inline void xfs_sb_version_addnlink(xfs_sb_t *sbp) { (sbp)->sb_versionnum = ((sbp)->sb_versionnum <= XFS_SB_VERSION_2 ? \ @@ -355,115 +380,63 @@ static inline void xfs_sb_version_addnlink(xfs_sb_t *sbp) ((sbp)->sb_versionnum | XFS_SB_VERSION_NLINKBIT)); } -#define XFS_SB_VERSION_HASQUOTA(sbp) xfs_sb_version_hasquota(sbp) static inline int xfs_sb_version_hasquota(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_QUOTABIT); } -#define XFS_SB_VERSION_ADDQUOTA(sbp) xfs_sb_version_addquota(sbp) static inline void xfs_sb_version_addquota(xfs_sb_t *sbp) { (sbp)->sb_versionnum = \ (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 ? \ ((sbp)->sb_versionnum | XFS_SB_VERSION_QUOTABIT) : \ - (XFS_SB_VERSION_TONEW((sbp)->sb_versionnum) | \ + (xfs_sb_version_tonew((sbp)->sb_versionnum) | \ XFS_SB_VERSION_QUOTABIT)); } -#define XFS_SB_VERSION_HASALIGN(sbp) xfs_sb_version_hasalign(sbp) static inline int xfs_sb_version_hasalign(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_ALIGNBIT); } -#define XFS_SB_VERSION_SUBALIGN(sbp) xfs_sb_version_subalign(sbp) -static inline void xfs_sb_version_subalign(xfs_sb_t *sbp) -{ - (sbp)->sb_versionnum = \ - XFS_SB_VERSION_TOOLD((sbp)->sb_versionnum & ~XFS_SB_VERSION_ALIGNBIT); -} - -#define XFS_SB_VERSION_HASDALIGN(sbp) xfs_sb_version_hasdalign(sbp) static inline int xfs_sb_version_hasdalign(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_DALIGNBIT); } -#define XFS_SB_VERSION_ADDDALIGN(sbp) xfs_sb_version_adddalign(sbp) -static inline int xfs_sb_version_adddalign(xfs_sb_t *sbp) -{ - return (sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum | XFS_SB_VERSION_DALIGNBIT); -} - -#define XFS_SB_VERSION_HASSHARED(sbp) xfs_sb_version_hasshared(sbp) static inline int xfs_sb_version_hasshared(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_SHAREDBIT); } -#define XFS_SB_VERSION_ADDSHARED(sbp) xfs_sb_version_addshared(sbp) -static inline int xfs_sb_version_addshared(xfs_sb_t *sbp) -{ - return (sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum | XFS_SB_VERSION_SHAREDBIT); -} - -#define XFS_SB_VERSION_SUBSHARED(sbp) xfs_sb_version_subshared(sbp) -static inline int xfs_sb_version_subshared(xfs_sb_t *sbp) -{ - return (sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum & ~XFS_SB_VERSION_SHAREDBIT); -} - -#define XFS_SB_VERSION_HASDIRV2(sbp) xfs_sb_version_hasdirv2(sbp) static inline int xfs_sb_version_hasdirv2(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_DIRV2BIT); } -#define XFS_SB_VERSION_HASLOGV2(sbp) xfs_sb_version_haslogv2(sbp) static inline int xfs_sb_version_haslogv2(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_LOGV2BIT); } -#define XFS_SB_VERSION_HASEXTFLGBIT(sbp) xfs_sb_version_hasextflgbit(sbp) static inline int xfs_sb_version_hasextflgbit(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_EXTFLGBIT); } -#define XFS_SB_VERSION_ADDEXTFLGBIT(sbp) xfs_sb_version_addextflgbit(sbp) -static inline int xfs_sb_version_addextflgbit(xfs_sb_t *sbp) -{ - return (sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum | XFS_SB_VERSION_EXTFLGBIT); -} - -#define XFS_SB_VERSION_SUBEXTFLGBIT(sbp) xfs_sb_version_subextflgbit(sbp) -static inline int xfs_sb_version_subextflgbit(xfs_sb_t *sbp) -{ - return (sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum & ~XFS_SB_VERSION_EXTFLGBIT); -} - -#define XFS_SB_VERSION_HASSECTOR(sbp) xfs_sb_version_hassector(sbp) static inline int xfs_sb_version_hassector(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_SECTORBIT); } -#define XFS_SB_VERSION_HASMOREBITS(sbp) xfs_sb_version_hasmorebits(sbp) static inline int xfs_sb_version_hasmorebits(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ @@ -476,24 +449,22 @@ static inline int xfs_sb_version_hasmorebits(xfs_sb_t *sbp) * For example, for a bit defined as XFS_SB_VERSION2_FUNBIT, has a macro: * * SB_VERSION_HASFUNBIT(xfs_sb_t *sbp) - * ((XFS_SB_VERSION_HASMOREBITS(sbp) && + * ((xfs_sb_version_hasmorebits(sbp) && * ((sbp)->sb_features2 & XFS_SB_VERSION2_FUNBIT) */ static inline int xfs_sb_version_haslazysbcount(xfs_sb_t *sbp) { - return (XFS_SB_VERSION_HASMOREBITS(sbp) && \ + return (xfs_sb_version_hasmorebits(sbp) && \ ((sbp)->sb_features2 & XFS_SB_VERSION2_LAZYSBCOUNTBIT)); } -#define XFS_SB_VERSION_HASATTR2(sbp) xfs_sb_version_hasattr2(sbp) static inline int xfs_sb_version_hasattr2(xfs_sb_t *sbp) { - return (XFS_SB_VERSION_HASMOREBITS(sbp)) && \ + return (xfs_sb_version_hasmorebits(sbp)) && \ ((sbp)->sb_features2 & XFS_SB_VERSION2_ATTR2BIT); } -#define XFS_SB_VERSION_ADDATTR2(sbp) xfs_sb_version_addattr2(sbp) static inline void xfs_sb_version_addattr2(xfs_sb_t *sbp) { ((sbp)->sb_versionnum = \ diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index 8878322ee79..140386434aa 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -567,26 +567,26 @@ xfs_trans_apply_sb_deltas( */ if (!xfs_sb_version_haslazysbcount(&(tp->t_mountp->m_sb))) { if (tp->t_icount_delta) - be64_add(&sbp->sb_icount, tp->t_icount_delta); + be64_add_cpu(&sbp->sb_icount, tp->t_icount_delta); if (tp->t_ifree_delta) - be64_add(&sbp->sb_ifree, tp->t_ifree_delta); + be64_add_cpu(&sbp->sb_ifree, tp->t_ifree_delta); if (tp->t_fdblocks_delta) - be64_add(&sbp->sb_fdblocks, tp->t_fdblocks_delta); + be64_add_cpu(&sbp->sb_fdblocks, tp->t_fdblocks_delta); if (tp->t_res_fdblocks_delta) - be64_add(&sbp->sb_fdblocks, tp->t_res_fdblocks_delta); + be64_add_cpu(&sbp->sb_fdblocks, tp->t_res_fdblocks_delta); } if (tp->t_frextents_delta) - be64_add(&sbp->sb_frextents, tp->t_frextents_delta); + be64_add_cpu(&sbp->sb_frextents, tp->t_frextents_delta); if (tp->t_res_frextents_delta) - be64_add(&sbp->sb_frextents, tp->t_res_frextents_delta); + be64_add_cpu(&sbp->sb_frextents, tp->t_res_frextents_delta); if (tp->t_dblocks_delta) { - be64_add(&sbp->sb_dblocks, tp->t_dblocks_delta); + be64_add_cpu(&sbp->sb_dblocks, tp->t_dblocks_delta); whole = 1; } if (tp->t_agcount_delta) { - be32_add(&sbp->sb_agcount, tp->t_agcount_delta); + be32_add_cpu(&sbp->sb_agcount, tp->t_agcount_delta); whole = 1; } if (tp->t_imaxpct_delta) { @@ -594,19 +594,19 @@ xfs_trans_apply_sb_deltas( whole = 1; } if (tp->t_rextsize_delta) { - be32_add(&sbp->sb_rextsize, tp->t_rextsize_delta); + be32_add_cpu(&sbp->sb_rextsize, tp->t_rextsize_delta); whole = 1; } if (tp->t_rbmblocks_delta) { - be32_add(&sbp->sb_rbmblocks, tp->t_rbmblocks_delta); + be32_add_cpu(&sbp->sb_rbmblocks, tp->t_rbmblocks_delta); whole = 1; } if (tp->t_rblocks_delta) { - be64_add(&sbp->sb_rblocks, tp->t_rblocks_delta); + be64_add_cpu(&sbp->sb_rblocks, tp->t_rblocks_delta); whole = 1; } if (tp->t_rextents_delta) { - be64_add(&sbp->sb_rextents, tp->t_rextents_delta); + be64_add_cpu(&sbp->sb_rextents, tp->t_rextents_delta); whole = 1; } if (tp->t_rextslog_delta) { @@ -1322,7 +1322,6 @@ xfs_trans_chunk_committed( xfs_lsn_t item_lsn; struct xfs_mount *mp; int i; - SPLDECL(s); lidp = licp->lic_descs; for (i = 0; i < licp->lic_unused; i++, lidp++) { @@ -1363,7 +1362,7 @@ xfs_trans_chunk_committed( * the test below. */ mp = lip->li_mountp; - AIL_LOCK(mp,s); + spin_lock(&mp->m_ail_lock); if (XFS_LSN_CMP(item_lsn, lip->li_lsn) > 0) { /* * This will set the item's lsn to item_lsn @@ -1372,9 +1371,9 @@ xfs_trans_chunk_committed( * * xfs_trans_update_ail() drops the AIL lock. */ - xfs_trans_update_ail(mp, lip, item_lsn, s); + xfs_trans_update_ail(mp, lip, item_lsn); } else { - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); } /* diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h index 0e26e729023..7f40628d85c 100644 --- a/fs/xfs/xfs_trans.h +++ b/fs/xfs/xfs_trans.h @@ -992,8 +992,9 @@ int _xfs_trans_commit(xfs_trans_t *, int *); #define xfs_trans_commit(tp, flags) _xfs_trans_commit(tp, flags, NULL) void xfs_trans_cancel(xfs_trans_t *, int); -void xfs_trans_ail_init(struct xfs_mount *); -xfs_lsn_t xfs_trans_push_ail(struct xfs_mount *, xfs_lsn_t); +int xfs_trans_ail_init(struct xfs_mount *); +void xfs_trans_ail_destroy(struct xfs_mount *); +void xfs_trans_push_ail(struct xfs_mount *, xfs_lsn_t); xfs_lsn_t xfs_trans_tail_ail(struct xfs_mount *); void xfs_trans_unlocked_item(struct xfs_mount *, xfs_log_item_t *); @@ -1001,6 +1002,8 @@ xfs_log_busy_slot_t *xfs_trans_add_busy(xfs_trans_t *tp, xfs_agnumber_t ag, xfs_extlen_t idx); +extern kmem_zone_t *xfs_trans_zone; + #endif /* __KERNEL__ */ #endif /* __XFS_TRANS_H__ */ diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c index 5b2ff59f19c..76d470d8a1e 100644 --- a/fs/xfs/xfs_trans_ail.c +++ b/fs/xfs/xfs_trans_ail.c @@ -34,9 +34,9 @@ STATIC xfs_log_item_t * xfs_ail_min(xfs_ail_entry_t *); STATIC xfs_log_item_t * xfs_ail_next(xfs_ail_entry_t *, xfs_log_item_t *); #ifdef DEBUG -STATIC void xfs_ail_check(xfs_ail_entry_t *); +STATIC void xfs_ail_check(xfs_ail_entry_t *, xfs_log_item_t *); #else -#define xfs_ail_check(a) +#define xfs_ail_check(a,l) #endif /* DEBUG */ @@ -55,16 +55,15 @@ xfs_trans_tail_ail( { xfs_lsn_t lsn; xfs_log_item_t *lip; - SPLDECL(s); - AIL_LOCK(mp,s); - lip = xfs_ail_min(&(mp->m_ail)); + spin_lock(&mp->m_ail_lock); + lip = xfs_ail_min(&(mp->m_ail.xa_ail)); if (lip == NULL) { lsn = (xfs_lsn_t)0; } else { lsn = lip->li_lsn; } - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); return lsn; } @@ -72,120 +71,185 @@ xfs_trans_tail_ail( /* * xfs_trans_push_ail * - * This routine is called to move the tail of the AIL - * forward. It does this by trying to flush items in the AIL - * whose lsns are below the given threshold_lsn. + * This routine is called to move the tail of the AIL forward. It does this by + * trying to flush items in the AIL whose lsns are below the given + * threshold_lsn. * - * The routine returns the lsn of the tail of the log. + * the push is run asynchronously in a separate thread, so we return the tail + * of the log right now instead of the tail after the push. This means we will + * either continue right away, or we will sleep waiting on the async thread to + * do it's work. + * + * We do this unlocked - we only need to know whether there is anything in the + * AIL at the time we are called. We don't need to access the contents of + * any of the objects, so the lock is not needed. */ -xfs_lsn_t +void xfs_trans_push_ail( xfs_mount_t *mp, xfs_lsn_t threshold_lsn) { - xfs_lsn_t lsn; xfs_log_item_t *lip; - int gen; - int restarts; - int lock_result; - int flush_log; - SPLDECL(s); -#define XFS_TRANS_PUSH_AIL_RESTARTS 1000 + lip = xfs_ail_min(&mp->m_ail.xa_ail); + if (lip && !XFS_FORCED_SHUTDOWN(mp)) { + if (XFS_LSN_CMP(threshold_lsn, mp->m_ail.xa_target) > 0) + xfsaild_wakeup(mp, threshold_lsn); + } +} + +/* + * Return the item in the AIL with the current lsn. + * Return the current tree generation number for use + * in calls to xfs_trans_next_ail(). + */ +STATIC xfs_log_item_t * +xfs_trans_first_push_ail( + xfs_mount_t *mp, + int *gen, + xfs_lsn_t lsn) +{ + xfs_log_item_t *lip; + + lip = xfs_ail_min(&(mp->m_ail.xa_ail)); + *gen = (int)mp->m_ail.xa_gen; + if (lsn == 0) + return lip; + + while (lip && (XFS_LSN_CMP(lip->li_lsn, lsn) < 0)) + lip = lip->li_ail.ail_forw; - AIL_LOCK(mp,s); - lip = xfs_trans_first_ail(mp, &gen); - if (lip == NULL || XFS_FORCED_SHUTDOWN(mp)) { + return lip; +} + +/* + * Function that does the work of pushing on the AIL + */ +long +xfsaild_push( + xfs_mount_t *mp, + xfs_lsn_t *last_lsn) +{ + long tout = 1000; /* milliseconds */ + xfs_lsn_t last_pushed_lsn = *last_lsn; + xfs_lsn_t target = mp->m_ail.xa_target; + xfs_lsn_t lsn; + xfs_log_item_t *lip; + int gen; + int restarts; + int flush_log, count, stuck; + +#define XFS_TRANS_PUSH_AIL_RESTARTS 10 + + spin_lock(&mp->m_ail_lock); + lip = xfs_trans_first_push_ail(mp, &gen, *last_lsn); + if (!lip || XFS_FORCED_SHUTDOWN(mp)) { /* - * Just return if the AIL is empty. + * AIL is empty or our push has reached the end. */ - AIL_UNLOCK(mp, s); - return (xfs_lsn_t)0; + spin_unlock(&mp->m_ail_lock); + last_pushed_lsn = 0; + goto out; } XFS_STATS_INC(xs_push_ail); /* * While the item we are looking at is below the given threshold - * try to flush it out. Make sure to limit the number of times - * we allow xfs_trans_next_ail() to restart scanning from the - * beginning of the list. We'd like not to stop until we've at least + * try to flush it out. We'd like not to stop until we've at least * tried to push on everything in the AIL with an LSN less than - * the given threshold. However, we may give up before that if - * we realize that we've been holding the AIL_LOCK for 'too long', - * blocking interrupts. Currently, too long is < 500us roughly. + * the given threshold. + * + * However, we will stop after a certain number of pushes and wait + * for a reduced timeout to fire before pushing further. This + * prevents use from spinning when we can't do anything or there is + * lots of contention on the AIL lists. */ - flush_log = 0; - restarts = 0; - while (((restarts < XFS_TRANS_PUSH_AIL_RESTARTS) && - (XFS_LSN_CMP(lip->li_lsn, threshold_lsn) < 0))) { + tout = 10; + lsn = lip->li_lsn; + flush_log = stuck = count = restarts = 0; + while ((XFS_LSN_CMP(lip->li_lsn, target) < 0)) { + int lock_result; /* - * If we can lock the item without sleeping, unlock - * the AIL lock and flush the item. Then re-grab the - * AIL lock so we can look for the next item on the - * AIL. Since we unlock the AIL while we flush the - * item, the next routine may start over again at the - * the beginning of the list if anything has changed. - * That is what the generation count is for. + * If we can lock the item without sleeping, unlock the AIL + * lock and flush the item. Then re-grab the AIL lock so we + * can look for the next item on the AIL. List changes are + * handled by the AIL lookup functions internally * - * If we can't lock the item, either its holder will flush - * it or it is already being flushed or it is being relogged. - * In any of these case it is being taken care of and we - * can just skip to the next item in the list. + * If we can't lock the item, either its holder will flush it + * or it is already being flushed or it is being relogged. In + * any of these case it is being taken care of and we can just + * skip to the next item in the list. */ lock_result = IOP_TRYLOCK(lip); + spin_unlock(&mp->m_ail_lock); switch (lock_result) { - case XFS_ITEM_SUCCESS: - AIL_UNLOCK(mp, s); + case XFS_ITEM_SUCCESS: XFS_STATS_INC(xs_push_ail_success); IOP_PUSH(lip); - AIL_LOCK(mp,s); + last_pushed_lsn = lsn; break; - case XFS_ITEM_PUSHBUF: - AIL_UNLOCK(mp, s); + case XFS_ITEM_PUSHBUF: XFS_STATS_INC(xs_push_ail_pushbuf); -#ifdef XFSRACEDEBUG - delay_for_intr(); - delay(300); -#endif - ASSERT(lip->li_ops->iop_pushbuf); - ASSERT(lip); IOP_PUSHBUF(lip); - AIL_LOCK(mp,s); + last_pushed_lsn = lsn; break; - case XFS_ITEM_PINNED: + case XFS_ITEM_PINNED: XFS_STATS_INC(xs_push_ail_pinned); + stuck++; flush_log = 1; break; - case XFS_ITEM_LOCKED: + case XFS_ITEM_LOCKED: XFS_STATS_INC(xs_push_ail_locked); + last_pushed_lsn = lsn; + stuck++; break; - case XFS_ITEM_FLUSHING: + case XFS_ITEM_FLUSHING: XFS_STATS_INC(xs_push_ail_flushing); + last_pushed_lsn = lsn; + stuck++; break; - default: + default: ASSERT(0); break; } - lip = xfs_trans_next_ail(mp, lip, &gen, &restarts); - if (lip == NULL) { + spin_lock(&mp->m_ail_lock); + /* should we bother continuing? */ + if (XFS_FORCED_SHUTDOWN(mp)) + break; + ASSERT(mp->m_log); + + count++; + + /* + * Are there too many items we can't do anything with? + * If we we are skipping too many items because we can't flush + * them or they are already being flushed, we back off and + * given them time to complete whatever operation is being + * done. i.e. remove pressure from the AIL while we can't make + * progress so traversals don't slow down further inserts and + * removals to/from the AIL. + * + * The value of 100 is an arbitrary magic number based on + * observation. + */ + if (stuck > 100) break; - } - if (XFS_FORCED_SHUTDOWN(mp)) { - /* - * Just return if we shut down during the last try. - */ - AIL_UNLOCK(mp, s); - return (xfs_lsn_t)0; - } + lip = xfs_trans_next_ail(mp, lip, &gen, &restarts); + if (lip == NULL) + break; + if (restarts > XFS_TRANS_PUSH_AIL_RESTARTS) + break; + lsn = lip->li_lsn; } + spin_unlock(&mp->m_ail_lock); if (flush_log) { /* @@ -193,22 +257,38 @@ xfs_trans_push_ail( * push out the log so it will become unpinned and * move forward in the AIL. */ - AIL_UNLOCK(mp, s); XFS_STATS_INC(xs_push_ail_flush); xfs_log_force(mp, (xfs_lsn_t)0, XFS_LOG_FORCE); - AIL_LOCK(mp, s); } - lip = xfs_ail_min(&(mp->m_ail)); - if (lip == NULL) { - lsn = (xfs_lsn_t)0; - } else { - lsn = lip->li_lsn; + if (!count) { + /* We're past our target or empty, so idle */ + tout = 1000; + } else if (XFS_LSN_CMP(lsn, target) >= 0) { + /* + * We reached the target so wait a bit longer for I/O to + * complete and remove pushed items from the AIL before we + * start the next scan from the start of the AIL. + */ + tout += 20; + last_pushed_lsn = 0; + } else if ((restarts > XFS_TRANS_PUSH_AIL_RESTARTS) || + ((stuck * 100) / count > 90)) { + /* + * Either there is a lot of contention on the AIL or we + * are stuck due to operations in progress. "Stuck" in this + * case is defined as >90% of the items we tried to push + * were stuck. + * + * Backoff a bit more to allow some I/O to complete before + * continuing from where we were. + */ + tout += 10; } - - AIL_UNLOCK(mp, s); - return lsn; -} /* xfs_trans_push_ail */ +out: + *last_lsn = last_pushed_lsn; + return tout; +} /* xfsaild_push */ /* @@ -249,7 +329,7 @@ xfs_trans_unlocked_item( * the call to xfs_log_move_tail() doesn't do anything if there's * not enough free space to wake people up so we're safe calling it. */ - min_lip = xfs_ail_min(&mp->m_ail); + min_lip = xfs_ail_min(&mp->m_ail.xa_ail); if (min_lip == lip) xfs_log_move_tail(mp, 1); @@ -269,21 +349,19 @@ xfs_trans_unlocked_item( * has changed. * * This function must be called with the AIL lock held. The lock - * is dropped before returning, so the caller must pass in the - * cookie returned by AIL_LOCK. + * is dropped before returning. */ void xfs_trans_update_ail( xfs_mount_t *mp, xfs_log_item_t *lip, - xfs_lsn_t lsn, - unsigned long s) __releases(mp->m_ail_lock) + xfs_lsn_t lsn) __releases(mp->m_ail_lock) { xfs_ail_entry_t *ailp; xfs_log_item_t *dlip=NULL; xfs_log_item_t *mlip; /* ptr to minimum lip */ - ailp = &(mp->m_ail); + ailp = &(mp->m_ail.xa_ail); mlip = xfs_ail_min(ailp); if (lip->li_flags & XFS_LI_IN_AIL) { @@ -296,14 +374,14 @@ xfs_trans_update_ail( lip->li_lsn = lsn; xfs_ail_insert(ailp, lip); - mp->m_ail_gen++; + mp->m_ail.xa_gen++; if (mlip == dlip) { - mlip = xfs_ail_min(&(mp->m_ail)); - AIL_UNLOCK(mp, s); + mlip = xfs_ail_min(&(mp->m_ail.xa_ail)); + spin_unlock(&mp->m_ail_lock); xfs_log_move_tail(mp, mlip->li_lsn); } else { - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); } @@ -322,21 +400,19 @@ xfs_trans_update_ail( * has changed. * * This function must be called with the AIL lock held. The lock - * is dropped before returning, so the caller must pass in the - * cookie returned by AIL_LOCK. + * is dropped before returning. */ void xfs_trans_delete_ail( xfs_mount_t *mp, - xfs_log_item_t *lip, - unsigned long s) __releases(mp->m_ail_lock) + xfs_log_item_t *lip) __releases(mp->m_ail_lock) { xfs_ail_entry_t *ailp; xfs_log_item_t *dlip; xfs_log_item_t *mlip; if (lip->li_flags & XFS_LI_IN_AIL) { - ailp = &(mp->m_ail); + ailp = &(mp->m_ail.xa_ail); mlip = xfs_ail_min(ailp); dlip = xfs_ail_delete(ailp, lip); ASSERT(dlip == lip); @@ -344,14 +420,14 @@ xfs_trans_delete_ail( lip->li_flags &= ~XFS_LI_IN_AIL; lip->li_lsn = 0; - mp->m_ail_gen++; + mp->m_ail.xa_gen++; if (mlip == dlip) { - mlip = xfs_ail_min(&(mp->m_ail)); - AIL_UNLOCK(mp, s); + mlip = xfs_ail_min(&(mp->m_ail.xa_ail)); + spin_unlock(&mp->m_ail_lock); xfs_log_move_tail(mp, (mlip ? mlip->li_lsn : 0)); } else { - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); } } else { @@ -360,12 +436,12 @@ xfs_trans_delete_ail( * serious trouble if we get to this stage. */ if (XFS_FORCED_SHUTDOWN(mp)) - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); else { xfs_cmn_err(XFS_PTAG_AILDELETE, CE_ALERT, mp, "%s: attempting to delete a log item that is not in the AIL", __FUNCTION__); - AIL_UNLOCK(mp, s); + spin_unlock(&mp->m_ail_lock); xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); } } @@ -385,10 +461,10 @@ xfs_trans_first_ail( { xfs_log_item_t *lip; - lip = xfs_ail_min(&(mp->m_ail)); - *gen = (int)mp->m_ail_gen; + lip = xfs_ail_min(&(mp->m_ail.xa_ail)); + *gen = (int)mp->m_ail.xa_gen; - return (lip); + return lip; } /* @@ -408,11 +484,11 @@ xfs_trans_next_ail( xfs_log_item_t *nlip; ASSERT(mp && lip && gen); - if (mp->m_ail_gen == *gen) { - nlip = xfs_ail_next(&(mp->m_ail), lip); + if (mp->m_ail.xa_gen == *gen) { + nlip = xfs_ail_next(&(mp->m_ail.xa_ail), lip); } else { - nlip = xfs_ail_min(&(mp->m_ail)); - *gen = (int)mp->m_ail_gen; + nlip = xfs_ail_min(&(mp->m_ail).xa_ail); + *gen = (int)mp->m_ail.xa_gen; if (restarts != NULL) { XFS_STATS_INC(xs_push_ail_restarts); (*restarts)++; @@ -437,12 +513,20 @@ xfs_trans_next_ail( /* * Initialize the doubly linked list to point only to itself. */ -void +int xfs_trans_ail_init( xfs_mount_t *mp) { - mp->m_ail.ail_forw = (xfs_log_item_t*)&(mp->m_ail); - mp->m_ail.ail_back = (xfs_log_item_t*)&(mp->m_ail); + mp->m_ail.xa_ail.ail_forw = (xfs_log_item_t*)&mp->m_ail.xa_ail; + mp->m_ail.xa_ail.ail_back = (xfs_log_item_t*)&mp->m_ail.xa_ail; + return xfsaild_start(mp); +} + +void +xfs_trans_ail_destroy( + xfs_mount_t *mp) +{ + xfsaild_stop(mp); } /* @@ -482,7 +566,7 @@ xfs_ail_insert( next_lip->li_ail.ail_forw = lip; lip->li_ail.ail_forw->li_ail.ail_back = lip; - xfs_ail_check(base); + xfs_ail_check(base, lip); return; } @@ -496,12 +580,12 @@ xfs_ail_delete( xfs_log_item_t *lip) /* ARGSUSED */ { + xfs_ail_check(base, lip); lip->li_ail.ail_forw->li_ail.ail_back = lip->li_ail.ail_back; lip->li_ail.ail_back->li_ail.ail_forw = lip->li_ail.ail_forw; lip->li_ail.ail_forw = NULL; lip->li_ail.ail_back = NULL; - xfs_ail_check(base); return lip; } @@ -545,13 +629,13 @@ xfs_ail_next( */ STATIC void xfs_ail_check( - xfs_ail_entry_t *base) + xfs_ail_entry_t *base, + xfs_log_item_t *lip) { - xfs_log_item_t *lip; xfs_log_item_t *prev_lip; - lip = base->ail_forw; - if (lip == (xfs_log_item_t*)base) { + prev_lip = base->ail_forw; + if (prev_lip == (xfs_log_item_t*)base) { /* * Make sure the pointers are correct when the list * is empty. @@ -561,9 +645,27 @@ xfs_ail_check( } /* + * Check the next and previous entries are valid. + */ + ASSERT((lip->li_flags & XFS_LI_IN_AIL) != 0); + prev_lip = lip->li_ail.ail_back; + if (prev_lip != (xfs_log_item_t*)base) { + ASSERT(prev_lip->li_ail.ail_forw == lip); + ASSERT(XFS_LSN_CMP(prev_lip->li_lsn, lip->li_lsn) <= 0); + } + prev_lip = lip->li_ail.ail_forw; + if (prev_lip != (xfs_log_item_t*)base) { + ASSERT(prev_lip->li_ail.ail_back == lip); + ASSERT(XFS_LSN_CMP(prev_lip->li_lsn, lip->li_lsn) >= 0); + } + + +#ifdef XFS_TRANS_DEBUG + /* * Walk the list checking forward and backward pointers, * lsn ordering, and that every entry has the XFS_LI_IN_AIL - * flag set. + * flag set. This is really expensive, so only do it when + * specifically debugging the transaction subsystem. */ prev_lip = (xfs_log_item_t*)base; while (lip != (xfs_log_item_t*)base) { @@ -578,5 +680,6 @@ xfs_ail_check( } ASSERT(lip == (xfs_log_item_t*)base); ASSERT(base->ail_back == prev_lip); +#endif /* XFS_TRANS_DEBUG */ } #endif /* DEBUG */ diff --git a/fs/xfs/xfs_trans_item.c b/fs/xfs/xfs_trans_item.c index 2912aac07c7..66a09f0d894 100644 --- a/fs/xfs/xfs_trans_item.c +++ b/fs/xfs/xfs_trans_item.c @@ -21,6 +21,7 @@ #include "xfs_log.h" #include "xfs_inum.h" #include "xfs_trans.h" +#include "xfs_trans_priv.h" STATIC int xfs_trans_unlock_chunk(xfs_log_item_chunk_t *, int, int, xfs_lsn_t); diff --git a/fs/xfs/xfs_trans_priv.h b/fs/xfs/xfs_trans_priv.h index 447ac4308c9..3c748c456ed 100644 --- a/fs/xfs/xfs_trans_priv.h +++ b/fs/xfs/xfs_trans_priv.h @@ -47,15 +47,22 @@ xfs_log_busy_slot_t *xfs_trans_add_busy(xfs_trans_t *tp, * From xfs_trans_ail.c */ void xfs_trans_update_ail(struct xfs_mount *mp, - struct xfs_log_item *lip, xfs_lsn_t lsn, - unsigned long s) + struct xfs_log_item *lip, xfs_lsn_t lsn) __releases(mp->m_ail_lock); void xfs_trans_delete_ail(struct xfs_mount *mp, - struct xfs_log_item *lip, unsigned long s) + struct xfs_log_item *lip) __releases(mp->m_ail_lock); struct xfs_log_item *xfs_trans_first_ail(struct xfs_mount *, int *); struct xfs_log_item *xfs_trans_next_ail(struct xfs_mount *, struct xfs_log_item *, int *, int *); +/* + * AIL push thread support + */ +long xfsaild_push(struct xfs_mount *, xfs_lsn_t *); +void xfsaild_wakeup(struct xfs_mount *, xfs_lsn_t); +int xfsaild_start(struct xfs_mount *); +void xfsaild_stop(struct xfs_mount *); + #endif /* __XFS_TRANS_PRIV_H__ */ diff --git a/fs/xfs/xfs_utils.c b/fs/xfs/xfs_utils.c index 673b405eaa3..18a85e74668 100644 --- a/fs/xfs/xfs_utils.c +++ b/fs/xfs/xfs_utils.c @@ -73,7 +73,7 @@ xfs_dir_lookup_int( { int error; - vn_trace_entry(dp, __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(dp); error = xfs_dir_lookup(NULL, dp, VNAME(dentry), VNAMELEN(dentry), inum); if (!error) { @@ -302,6 +302,7 @@ xfs_droplink( ASSERT (ip->i_d.di_nlink > 0); ip->i_d.di_nlink--; + drop_nlink(ip->i_vnode); xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); error = 0; @@ -330,7 +331,6 @@ xfs_bump_ino_vers2( xfs_inode_t *ip) { xfs_mount_t *mp; - unsigned long s; ASSERT(ismrlocked (&ip->i_lock, MR_UPDATE)); ASSERT(ip->i_d.di_version == XFS_DINODE_VERSION_1); @@ -339,14 +339,14 @@ xfs_bump_ino_vers2( ip->i_d.di_onlink = 0; memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad)); mp = tp->t_mountp; - if (!XFS_SB_VERSION_HASNLINK(&mp->m_sb)) { - s = XFS_SB_LOCK(mp); - if (!XFS_SB_VERSION_HASNLINK(&mp->m_sb)) { - XFS_SB_VERSION_ADDNLINK(&mp->m_sb); - XFS_SB_UNLOCK(mp, s); + if (!xfs_sb_version_hasnlink(&mp->m_sb)) { + spin_lock(&mp->m_sb_lock); + if (!xfs_sb_version_hasnlink(&mp->m_sb)) { + xfs_sb_version_addnlink(&mp->m_sb); + spin_unlock(&mp->m_sb_lock); xfs_mod_sb(tp, XFS_SB_VERSIONNUM); } else { - XFS_SB_UNLOCK(mp, s); + spin_unlock(&mp->m_sb_lock); } } /* Caller must log the inode */ @@ -366,6 +366,7 @@ xfs_bumplink( ASSERT(ip->i_d.di_nlink > 0); ip->i_d.di_nlink++; + inc_nlink(ip->i_vnode); if ((ip->i_d.di_version == XFS_DINODE_VERSION_1) && (ip->i_d.di_nlink > XFS_MAXLINK_1)) { /* diff --git a/fs/xfs/xfs_utils.h b/fs/xfs/xfs_utils.h index a00b26d8840..f857fcccb72 100644 --- a/fs/xfs/xfs_utils.h +++ b/fs/xfs/xfs_utils.h @@ -20,8 +20,6 @@ #define IRELE(ip) VN_RELE(XFS_ITOV(ip)) #define IHOLD(ip) VN_HOLD(XFS_ITOV(ip)) -#define ITRACE(ip) vn_trace_ref(ip, __FILE__, __LINE__, \ - (inst_t *)__return_address) extern int xfs_get_dir_entry (bhv_vname_t *, xfs_inode_t **); extern int xfs_dir_lookup_int (xfs_inode_t *, uint, bhv_vname_t *, xfs_ino_t *, diff --git a/fs/xfs/xfs_vfsops.c b/fs/xfs/xfs_vfsops.c index a1544597bcd..7094caff13c 100644 --- a/fs/xfs/xfs_vfsops.c +++ b/fs/xfs/xfs_vfsops.c @@ -58,17 +58,12 @@ #include "xfs_vfsops.h" -int +int __init xfs_init(void) { - extern kmem_zone_t *xfs_bmap_free_item_zone; - extern kmem_zone_t *xfs_btree_cur_zone; - extern kmem_zone_t *xfs_trans_zone; - extern kmem_zone_t *xfs_buf_item_zone; - extern kmem_zone_t *xfs_dabuf_zone; #ifdef XFS_DABUF_DEBUG - extern lock_t xfs_dabuf_global_lock; - spinlock_init(&xfs_dabuf_global_lock, "xfsda"); + extern spinlock_t xfs_dabuf_global_lock; + spin_lock_init(&xfs_dabuf_global_lock); #endif /* @@ -152,18 +147,12 @@ xfs_init(void) return 0; } -void +void __exit xfs_cleanup(void) { - extern kmem_zone_t *xfs_bmap_free_item_zone; - extern kmem_zone_t *xfs_btree_cur_zone; extern kmem_zone_t *xfs_inode_zone; - extern kmem_zone_t *xfs_trans_zone; - extern kmem_zone_t *xfs_da_state_zone; - extern kmem_zone_t *xfs_dabuf_zone; extern kmem_zone_t *xfs_efd_zone; extern kmem_zone_t *xfs_efi_zone; - extern kmem_zone_t *xfs_buf_item_zone; extern kmem_zone_t *xfs_icluster_zone; xfs_cleanup_procfs(); @@ -292,8 +281,8 @@ xfs_start_flags( mp->m_readio_log = mp->m_writeio_log = ap->iosizelog; } - if (ap->flags & XFSMNT_IDELETE) - mp->m_flags |= XFS_MOUNT_IDELETE; + if (ap->flags & XFSMNT_IKEEP) + mp->m_flags |= XFS_MOUNT_IKEEP; if (ap->flags & XFSMNT_DIRSYNC) mp->m_flags |= XFS_MOUNT_DIRSYNC; if (ap->flags & XFSMNT_ATTR2) @@ -341,7 +330,7 @@ xfs_finish_flags( int ronly = (mp->m_flags & XFS_MOUNT_RDONLY); /* Fail a mount where the logbuf is smaller then the log stripe */ - if (XFS_SB_VERSION_HASLOGV2(&mp->m_sb)) { + if (xfs_sb_version_haslogv2(&mp->m_sb)) { if ((ap->logbufsize <= 0) && (mp->m_sb.sb_logsunit > XLOG_BIG_RECORD_BSIZE)) { mp->m_logbsize = mp->m_sb.sb_logsunit; @@ -360,9 +349,8 @@ xfs_finish_flags( } } - if (XFS_SB_VERSION_HASATTR2(&mp->m_sb)) { + if (xfs_sb_version_hasattr2(&mp->m_sb)) mp->m_flags |= XFS_MOUNT_ATTR2; - } /* * prohibit r/w mounts of read-only filesystems @@ -377,7 +365,7 @@ xfs_finish_flags( * check for shared mount. */ if (ap->flags & XFSMNT_SHARED) { - if (!XFS_SB_VERSION_HASSHARED(&mp->m_sb)) + if (!xfs_sb_version_hasshared(&mp->m_sb)) return XFS_ERROR(EINVAL); /* @@ -449,8 +437,6 @@ xfs_mount( if (error) return error; - mp->m_io_ops = xfs_iocore_xfs; - if (args->flags & XFSMNT_QUIET) flags |= XFS_MFSI_QUIET; @@ -525,7 +511,7 @@ xfs_mount( if (!error && logdev && logdev != ddev) { unsigned int log_sector_size = BBSIZE; - if (XFS_SB_VERSION_HASSECTOR(&mp->m_sb)) + if (xfs_sb_version_hassector(&mp->m_sb)) log_sector_size = mp->m_sb.sb_logsectsize; error = xfs_setsize_buftarg(mp->m_logdev_targp, mp->m_sb.sb_blocksize, @@ -544,7 +530,7 @@ xfs_mount( if ((error = xfs_filestream_mount(mp))) goto error2; - error = XFS_IOINIT(mp, args, flags); + error = xfs_mountfs(mp, flags); if (error) goto error2; @@ -694,7 +680,7 @@ xfs_quiesce_fs( * care of the metadata. New transactions are already blocked, so we need to * wait for any remaining transactions to drain out before proceding. */ -STATIC void +void xfs_attr_quiesce( xfs_mount_t *mp) { @@ -821,80 +807,6 @@ fscorrupt_out2: } /* - * xfs_root extracts the root vnode from a vfs. - * - * vfsp -- the vfs struct for the desired file system - * vpp -- address of the caller's vnode pointer which should be - * set to the desired fs root vnode - */ -int -xfs_root( - xfs_mount_t *mp, - bhv_vnode_t **vpp) -{ - bhv_vnode_t *vp; - - vp = XFS_ITOV(mp->m_rootip); - VN_HOLD(vp); - *vpp = vp; - return 0; -} - -/* - * xfs_statvfs - * - * Fill in the statvfs structure for the given file system. We use - * the superblock lock in the mount structure to ensure a consistent - * snapshot of the counters returned. - */ -int -xfs_statvfs( - xfs_mount_t *mp, - bhv_statvfs_t *statp, - bhv_vnode_t *vp) -{ - __uint64_t fakeinos; - xfs_extlen_t lsize; - xfs_sb_t *sbp; - unsigned long s; - - sbp = &(mp->m_sb); - - statp->f_type = XFS_SB_MAGIC; - - xfs_icsb_sync_counters_flags(mp, XFS_ICSB_LAZY_COUNT); - s = XFS_SB_LOCK(mp); - statp->f_bsize = sbp->sb_blocksize; - lsize = sbp->sb_logstart ? sbp->sb_logblocks : 0; - statp->f_blocks = sbp->sb_dblocks - lsize; - statp->f_bfree = statp->f_bavail = - sbp->sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp); - fakeinos = statp->f_bfree << sbp->sb_inopblog; -#if XFS_BIG_INUMS - fakeinos += mp->m_inoadd; -#endif - statp->f_files = - MIN(sbp->sb_icount + fakeinos, (__uint64_t)XFS_MAXINUMBER); - if (mp->m_maxicount) -#if XFS_BIG_INUMS - if (!mp->m_inoadd) -#endif - statp->f_files = min_t(typeof(statp->f_files), - statp->f_files, - mp->m_maxicount); - statp->f_ffree = statp->f_files - (sbp->sb_icount - sbp->sb_ifree); - XFS_SB_UNLOCK(mp, s); - - xfs_statvfs_fsid(statp, mp); - statp->f_namelen = MAXNAMELEN - 1; - - if (vp) - XFS_QM_DQSTATVFS(xfs_vtoi(vp), statp); - return 0; -} - - -/* * xfs_sync flushes any pending I/O to file system vfsp. * * This routine is called by vfs_sync() to make sure that things make it @@ -981,8 +893,6 @@ xfs_sync_inodes( int *bypassed) { xfs_inode_t *ip = NULL; - xfs_inode_t *ip_next; - xfs_buf_t *bp; bhv_vnode_t *vp = NULL; int error; int last_error; @@ -992,7 +902,6 @@ xfs_sync_inodes( boolean_t mount_locked; boolean_t vnode_refed; int preempt; - xfs_dinode_t *dip; xfs_iptr_t *ipointer; #ifdef DEBUG boolean_t ipointer_in = B_FALSE; @@ -1045,6 +954,8 @@ xfs_sync_inodes( #define XFS_PREEMPT_MASK 0x7f + ASSERT(!(flags & SYNC_BDFLUSH)); + if (bypassed) *bypassed = 0; if (mp->m_flags & XFS_MOUNT_RDONLY) @@ -1057,7 +968,7 @@ xfs_sync_inodes( ipointer = (xfs_iptr_t *)kmem_zalloc(sizeof(xfs_iptr_t), KM_SLEEP); fflag = XFS_B_ASYNC; /* default is don't wait */ - if (flags & (SYNC_BDFLUSH | SYNC_DELWRI)) + if (flags & SYNC_DELWRI) fflag = XFS_B_DELWRI; if (flags & SYNC_WAIT) fflag = 0; /* synchronous overrides all */ @@ -1147,24 +1058,6 @@ xfs_sync_inodes( } /* - * If this is just vfs_sync() or pflushd() calling - * then we can skip inodes for which it looks like - * there is nothing to do. Since we don't have the - * inode locked this is racy, but these are periodic - * calls so it doesn't matter. For the others we want - * to know for sure, so we at least try to lock them. - */ - if (flags & SYNC_BDFLUSH) { - if (((ip->i_itemp == NULL) || - !(ip->i_itemp->ili_format.ilf_fields & - XFS_ILOG_ALL)) && - (ip->i_update_core == 0)) { - ip = ip->i_mnext; - continue; - } - } - - /* * Try to lock without sleeping. We're out of order with * the inode list lock here, so if we fail we need to drop * the mount lock and try again. If we're called from @@ -1181,7 +1074,7 @@ xfs_sync_inodes( * it. */ if (xfs_ilock_nowait(ip, lock_flags) == 0) { - if ((flags & SYNC_BDFLUSH) || (vp == NULL)) { + if (vp == NULL) { ip = ip->i_mnext; continue; } @@ -1242,160 +1135,27 @@ xfs_sync_inodes( xfs_ilock(ip, XFS_ILOCK_SHARED); } - if (flags & SYNC_BDFLUSH) { - if ((flags & SYNC_ATTR) && - ((ip->i_update_core) || - ((ip->i_itemp != NULL) && - (ip->i_itemp->ili_format.ilf_fields != 0)))) { - - /* Insert marker and drop lock if not already - * done. - */ - if (mount_locked) { - IPOINTER_INSERT(ip, mp); - } - - /* - * We don't want the periodic flushing of the - * inodes by vfs_sync() to interfere with - * I/O to the file, especially read I/O - * where it is only the access time stamp - * that is being flushed out. To prevent - * long periods where we have both inode - * locks held shared here while reading the - * inode's buffer in from disk, we drop the - * inode lock while reading in the inode - * buffer. We have to release the buffer - * and reacquire the inode lock so that they - * are acquired in the proper order (inode - * locks first). The buffer will go at the - * end of the lru chain, though, so we can - * expect it to still be there when we go - * for it again in xfs_iflush(). - */ - if ((xfs_ipincount(ip) == 0) && - xfs_iflock_nowait(ip)) { - - xfs_ifunlock(ip); - xfs_iunlock(ip, XFS_ILOCK_SHARED); - - error = xfs_itobp(mp, NULL, ip, - &dip, &bp, 0, 0); - if (!error) { - xfs_buf_relse(bp); - } else { - /* Bailing out, remove the - * marker and free it. - */ - XFS_MOUNT_ILOCK(mp); - IPOINTER_REMOVE(ip, mp); - XFS_MOUNT_IUNLOCK(mp); - - ASSERT(!(lock_flags & - XFS_IOLOCK_SHARED)); - - kmem_free(ipointer, - sizeof(xfs_iptr_t)); - return (0); - } - - /* - * Since we dropped the inode lock, - * the inode may have been reclaimed. - * Therefore, we reacquire the mount - * lock and check to see if we were the - * inode reclaimed. If this happened - * then the ipointer marker will no - * longer point back at us. In this - * case, move ip along to the inode - * after the marker, remove the marker - * and continue. - */ - XFS_MOUNT_ILOCK(mp); - mount_locked = B_TRUE; - - if (ip != ipointer->ip_mprev) { - IPOINTER_REMOVE(ip, mp); - - ASSERT(!vnode_refed); - ASSERT(!(lock_flags & - XFS_IOLOCK_SHARED)); - continue; - } - - ASSERT(ip->i_mount == mp); - - if (xfs_ilock_nowait(ip, - XFS_ILOCK_SHARED) == 0) { - ASSERT(ip->i_mount == mp); - /* - * We failed to reacquire - * the inode lock without - * sleeping, so just skip - * the inode for now. We - * clear the ILOCK bit from - * the lock_flags so that we - * won't try to drop a lock - * we don't hold below. - */ - lock_flags &= ~XFS_ILOCK_SHARED; - IPOINTER_REMOVE(ip_next, mp); - } else if ((xfs_ipincount(ip) == 0) && - xfs_iflock_nowait(ip)) { - ASSERT(ip->i_mount == mp); - /* - * Since this is vfs_sync() - * calling we only flush the - * inode out if we can lock - * it without sleeping and - * it is not pinned. Drop - * the mount lock here so - * that we don't hold it for - * too long. We already have - * a marker in the list here. - */ - XFS_MOUNT_IUNLOCK(mp); - mount_locked = B_FALSE; - error = xfs_iflush(ip, - XFS_IFLUSH_DELWRI); - } else { - ASSERT(ip->i_mount == mp); - IPOINTER_REMOVE(ip_next, mp); - } - } - - } + if ((flags & SYNC_ATTR) && + (ip->i_update_core || + (ip->i_itemp && ip->i_itemp->ili_format.ilf_fields))) { + if (mount_locked) + IPOINTER_INSERT(ip, mp); - } else { - if ((flags & SYNC_ATTR) && - ((ip->i_update_core) || - ((ip->i_itemp != NULL) && - (ip->i_itemp->ili_format.ilf_fields != 0)))) { - if (mount_locked) { - IPOINTER_INSERT(ip, mp); - } + if (flags & SYNC_WAIT) { + xfs_iflock(ip); + error = xfs_iflush(ip, XFS_IFLUSH_SYNC); - if (flags & SYNC_WAIT) { - xfs_iflock(ip); - error = xfs_iflush(ip, - XFS_IFLUSH_SYNC); - } else { - /* - * If we can't acquire the flush - * lock, then the inode is already - * being flushed so don't bother - * waiting. If we can lock it then - * do a delwri flush so we can - * combine multiple inode flushes - * in each disk write. - */ - if (xfs_iflock_nowait(ip)) { - error = xfs_iflush(ip, - XFS_IFLUSH_DELWRI); - } - else if (bypassed) - (*bypassed)++; - } + /* + * If we can't acquire the flush lock, then the inode + * is already being flushed so don't bother waiting. + * + * If we can lock it then do a delwri flush so we can + * combine multiple inode flushes in each disk write. + */ + } else if (xfs_iflock_nowait(ip)) { + error = xfs_iflush(ip, XFS_IFLUSH_DELWRI); + } else if (bypassed) { + (*bypassed)++; } } @@ -1627,499 +1387,3 @@ xfs_syncsub( return XFS_ERROR(last_error); } - -/* - * xfs_vget - called by DMAPI and NFSD to get vnode from file handle - */ -int -xfs_vget( - xfs_mount_t *mp, - bhv_vnode_t **vpp, - xfs_fid_t *xfid) -{ - xfs_inode_t *ip; - int error; - xfs_ino_t ino; - unsigned int igen; - - /* - * Invalid. Since handles can be created in user space and passed in - * via gethandle(), this is not cause for a panic. - */ - if (xfid->fid_len != sizeof(*xfid) - sizeof(xfid->fid_len)) - return XFS_ERROR(EINVAL); - - ino = xfid->fid_ino; - igen = xfid->fid_gen; - - /* - * NFS can sometimes send requests for ino 0. Fail them gracefully. - */ - if (ino == 0) - return XFS_ERROR(ESTALE); - - error = xfs_iget(mp, NULL, ino, 0, XFS_ILOCK_SHARED, &ip, 0); - if (error) { - *vpp = NULL; - return error; - } - - if (ip == NULL) { - *vpp = NULL; - return XFS_ERROR(EIO); - } - - if (ip->i_d.di_mode == 0 || ip->i_d.di_gen != igen) { - xfs_iput_new(ip, XFS_ILOCK_SHARED); - *vpp = NULL; - return XFS_ERROR(ENOENT); - } - - *vpp = XFS_ITOV(ip); - xfs_iunlock(ip, XFS_ILOCK_SHARED); - return 0; -} - - -#define MNTOPT_LOGBUFS "logbufs" /* number of XFS log buffers */ -#define MNTOPT_LOGBSIZE "logbsize" /* size of XFS log buffers */ -#define MNTOPT_LOGDEV "logdev" /* log device */ -#define MNTOPT_RTDEV "rtdev" /* realtime I/O device */ -#define MNTOPT_BIOSIZE "biosize" /* log2 of preferred buffered io size */ -#define MNTOPT_WSYNC "wsync" /* safe-mode nfs compatible mount */ -#define MNTOPT_INO64 "ino64" /* force inodes into 64-bit range */ -#define MNTOPT_NOALIGN "noalign" /* turn off stripe alignment */ -#define MNTOPT_SWALLOC "swalloc" /* turn on stripe width allocation */ -#define MNTOPT_SUNIT "sunit" /* data volume stripe unit */ -#define MNTOPT_SWIDTH "swidth" /* data volume stripe width */ -#define MNTOPT_NOUUID "nouuid" /* ignore filesystem UUID */ -#define MNTOPT_MTPT "mtpt" /* filesystem mount point */ -#define MNTOPT_GRPID "grpid" /* group-ID from parent directory */ -#define MNTOPT_NOGRPID "nogrpid" /* group-ID from current process */ -#define MNTOPT_BSDGROUPS "bsdgroups" /* group-ID from parent directory */ -#define MNTOPT_SYSVGROUPS "sysvgroups" /* group-ID from current process */ -#define MNTOPT_ALLOCSIZE "allocsize" /* preferred allocation size */ -#define MNTOPT_NORECOVERY "norecovery" /* don't run XFS recovery */ -#define MNTOPT_BARRIER "barrier" /* use writer barriers for log write and - * unwritten extent conversion */ -#define MNTOPT_NOBARRIER "nobarrier" /* .. disable */ -#define MNTOPT_OSYNCISOSYNC "osyncisosync" /* o_sync is REALLY o_sync */ -#define MNTOPT_64BITINODE "inode64" /* inodes can be allocated anywhere */ -#define MNTOPT_IKEEP "ikeep" /* do not free empty inode clusters */ -#define MNTOPT_NOIKEEP "noikeep" /* free empty inode clusters */ -#define MNTOPT_LARGEIO "largeio" /* report large I/O sizes in stat() */ -#define MNTOPT_NOLARGEIO "nolargeio" /* do not report large I/O sizes - * in stat(). */ -#define MNTOPT_ATTR2 "attr2" /* do use attr2 attribute format */ -#define MNTOPT_NOATTR2 "noattr2" /* do not use attr2 attribute format */ -#define MNTOPT_FILESTREAM "filestreams" /* use filestreams allocator */ -#define MNTOPT_QUOTA "quota" /* disk quotas (user) */ -#define MNTOPT_NOQUOTA "noquota" /* no quotas */ -#define MNTOPT_USRQUOTA "usrquota" /* user quota enabled */ -#define MNTOPT_GRPQUOTA "grpquota" /* group quota enabled */ -#define MNTOPT_PRJQUOTA "prjquota" /* project quota enabled */ -#define MNTOPT_UQUOTA "uquota" /* user quota (IRIX variant) */ -#define MNTOPT_GQUOTA "gquota" /* group quota (IRIX variant) */ -#define MNTOPT_PQUOTA "pquota" /* project quota (IRIX variant) */ -#define MNTOPT_UQUOTANOENF "uqnoenforce"/* user quota limit enforcement */ -#define MNTOPT_GQUOTANOENF "gqnoenforce"/* group quota limit enforcement */ -#define MNTOPT_PQUOTANOENF "pqnoenforce"/* project quota limit enforcement */ -#define MNTOPT_QUOTANOENF "qnoenforce" /* same as uqnoenforce */ -#define MNTOPT_DMAPI "dmapi" /* DMI enabled (DMAPI / XDSM) */ -#define MNTOPT_XDSM "xdsm" /* DMI enabled (DMAPI / XDSM) */ -#define MNTOPT_DMI "dmi" /* DMI enabled (DMAPI / XDSM) */ - -STATIC unsigned long -suffix_strtoul(char *s, char **endp, unsigned int base) -{ - int last, shift_left_factor = 0; - char *value = s; - - last = strlen(value) - 1; - if (value[last] == 'K' || value[last] == 'k') { - shift_left_factor = 10; - value[last] = '\0'; - } - if (value[last] == 'M' || value[last] == 'm') { - shift_left_factor = 20; - value[last] = '\0'; - } - if (value[last] == 'G' || value[last] == 'g') { - shift_left_factor = 30; - value[last] = '\0'; - } - - return simple_strtoul((const char *)s, endp, base) << shift_left_factor; -} - -int -xfs_parseargs( - struct xfs_mount *mp, - char *options, - struct xfs_mount_args *args, - int update) -{ - char *this_char, *value, *eov; - int dsunit, dswidth, vol_dsunit, vol_dswidth; - int iosize; - int ikeep = 0; - - args->flags |= XFSMNT_BARRIER; - args->flags2 |= XFSMNT2_COMPAT_IOSIZE; - - if (!options) - goto done; - - iosize = dsunit = dswidth = vol_dsunit = vol_dswidth = 0; - - while ((this_char = strsep(&options, ",")) != NULL) { - if (!*this_char) - continue; - if ((value = strchr(this_char, '=')) != NULL) - *value++ = 0; - - if (!strcmp(this_char, MNTOPT_LOGBUFS)) { - if (!value || !*value) { - cmn_err(CE_WARN, - "XFS: %s option requires an argument", - this_char); - return EINVAL; - } - args->logbufs = simple_strtoul(value, &eov, 10); - } else if (!strcmp(this_char, MNTOPT_LOGBSIZE)) { - if (!value || !*value) { - cmn_err(CE_WARN, - "XFS: %s option requires an argument", - this_char); - return EINVAL; - } - args->logbufsize = suffix_strtoul(value, &eov, 10); - } else if (!strcmp(this_char, MNTOPT_LOGDEV)) { - if (!value || !*value) { - cmn_err(CE_WARN, - "XFS: %s option requires an argument", - this_char); - return EINVAL; - } - strncpy(args->logname, value, MAXNAMELEN); - } else if (!strcmp(this_char, MNTOPT_MTPT)) { - if (!value || !*value) { - cmn_err(CE_WARN, - "XFS: %s option requires an argument", - this_char); - return EINVAL; - } - strncpy(args->mtpt, value, MAXNAMELEN); - } else if (!strcmp(this_char, MNTOPT_RTDEV)) { - if (!value || !*value) { - cmn_err(CE_WARN, - "XFS: %s option requires an argument", - this_char); - return EINVAL; - } - strncpy(args->rtname, value, MAXNAMELEN); - } else if (!strcmp(this_char, MNTOPT_BIOSIZE)) { - if (!value || !*value) { - cmn_err(CE_WARN, - "XFS: %s option requires an argument", - this_char); - return EINVAL; - } - iosize = simple_strtoul(value, &eov, 10); - args->flags |= XFSMNT_IOSIZE; - args->iosizelog = (uint8_t) iosize; - } else if (!strcmp(this_char, MNTOPT_ALLOCSIZE)) { - if (!value || !*value) { - cmn_err(CE_WARN, - "XFS: %s option requires an argument", - this_char); - return EINVAL; - } - iosize = suffix_strtoul(value, &eov, 10); - args->flags |= XFSMNT_IOSIZE; - args->iosizelog = ffs(iosize) - 1; - } else if (!strcmp(this_char, MNTOPT_GRPID) || - !strcmp(this_char, MNTOPT_BSDGROUPS)) { - mp->m_flags |= XFS_MOUNT_GRPID; - } else if (!strcmp(this_char, MNTOPT_NOGRPID) || - !strcmp(this_char, MNTOPT_SYSVGROUPS)) { - mp->m_flags &= ~XFS_MOUNT_GRPID; - } else if (!strcmp(this_char, MNTOPT_WSYNC)) { - args->flags |= XFSMNT_WSYNC; - } else if (!strcmp(this_char, MNTOPT_OSYNCISOSYNC)) { - args->flags |= XFSMNT_OSYNCISOSYNC; - } else if (!strcmp(this_char, MNTOPT_NORECOVERY)) { - args->flags |= XFSMNT_NORECOVERY; - } else if (!strcmp(this_char, MNTOPT_INO64)) { - args->flags |= XFSMNT_INO64; -#if !XFS_BIG_INUMS - cmn_err(CE_WARN, - "XFS: %s option not allowed on this system", - this_char); - return EINVAL; -#endif - } else if (!strcmp(this_char, MNTOPT_NOALIGN)) { - args->flags |= XFSMNT_NOALIGN; - } else if (!strcmp(this_char, MNTOPT_SWALLOC)) { - args->flags |= XFSMNT_SWALLOC; - } else if (!strcmp(this_char, MNTOPT_SUNIT)) { - if (!value || !*value) { - cmn_err(CE_WARN, - "XFS: %s option requires an argument", - this_char); - return EINVAL; - } - dsunit = simple_strtoul(value, &eov, 10); - } else if (!strcmp(this_char, MNTOPT_SWIDTH)) { - if (!value || !*value) { - cmn_err(CE_WARN, - "XFS: %s option requires an argument", - this_char); - return EINVAL; - } - dswidth = simple_strtoul(value, &eov, 10); - } else if (!strcmp(this_char, MNTOPT_64BITINODE)) { - args->flags &= ~XFSMNT_32BITINODES; -#if !XFS_BIG_INUMS - cmn_err(CE_WARN, - "XFS: %s option not allowed on this system", - this_char); - return EINVAL; -#endif - } else if (!strcmp(this_char, MNTOPT_NOUUID)) { - args->flags |= XFSMNT_NOUUID; - } else if (!strcmp(this_char, MNTOPT_BARRIER)) { - args->flags |= XFSMNT_BARRIER; - } else if (!strcmp(this_char, MNTOPT_NOBARRIER)) { - args->flags &= ~XFSMNT_BARRIER; - } else if (!strcmp(this_char, MNTOPT_IKEEP)) { - ikeep = 1; - args->flags &= ~XFSMNT_IDELETE; - } else if (!strcmp(this_char, MNTOPT_NOIKEEP)) { - args->flags |= XFSMNT_IDELETE; - } else if (!strcmp(this_char, MNTOPT_LARGEIO)) { - args->flags2 &= ~XFSMNT2_COMPAT_IOSIZE; - } else if (!strcmp(this_char, MNTOPT_NOLARGEIO)) { - args->flags2 |= XFSMNT2_COMPAT_IOSIZE; - } else if (!strcmp(this_char, MNTOPT_ATTR2)) { - args->flags |= XFSMNT_ATTR2; - } else if (!strcmp(this_char, MNTOPT_NOATTR2)) { - args->flags &= ~XFSMNT_ATTR2; - } else if (!strcmp(this_char, MNTOPT_FILESTREAM)) { - args->flags2 |= XFSMNT2_FILESTREAMS; - } else if (!strcmp(this_char, MNTOPT_NOQUOTA)) { - args->flags &= ~(XFSMNT_UQUOTAENF|XFSMNT_UQUOTA); - args->flags &= ~(XFSMNT_GQUOTAENF|XFSMNT_GQUOTA); - } else if (!strcmp(this_char, MNTOPT_QUOTA) || - !strcmp(this_char, MNTOPT_UQUOTA) || - !strcmp(this_char, MNTOPT_USRQUOTA)) { - args->flags |= XFSMNT_UQUOTA | XFSMNT_UQUOTAENF; - } else if (!strcmp(this_char, MNTOPT_QUOTANOENF) || - !strcmp(this_char, MNTOPT_UQUOTANOENF)) { - args->flags |= XFSMNT_UQUOTA; - args->flags &= ~XFSMNT_UQUOTAENF; - } else if (!strcmp(this_char, MNTOPT_PQUOTA) || - !strcmp(this_char, MNTOPT_PRJQUOTA)) { - args->flags |= XFSMNT_PQUOTA | XFSMNT_PQUOTAENF; - } else if (!strcmp(this_char, MNTOPT_PQUOTANOENF)) { - args->flags |= XFSMNT_PQUOTA; - args->flags &= ~XFSMNT_PQUOTAENF; - } else if (!strcmp(this_char, MNTOPT_GQUOTA) || - !strcmp(this_char, MNTOPT_GRPQUOTA)) { - args->flags |= XFSMNT_GQUOTA | XFSMNT_GQUOTAENF; - } else if (!strcmp(this_char, MNTOPT_GQUOTANOENF)) { - args->flags |= XFSMNT_GQUOTA; - args->flags &= ~XFSMNT_GQUOTAENF; - } else if (!strcmp(this_char, MNTOPT_DMAPI)) { - args->flags |= XFSMNT_DMAPI; - } else if (!strcmp(this_char, MNTOPT_XDSM)) { - args->flags |= XFSMNT_DMAPI; - } else if (!strcmp(this_char, MNTOPT_DMI)) { - args->flags |= XFSMNT_DMAPI; - } else if (!strcmp(this_char, "ihashsize")) { - cmn_err(CE_WARN, - "XFS: ihashsize no longer used, option is deprecated."); - } else if (!strcmp(this_char, "osyncisdsync")) { - /* no-op, this is now the default */ - cmn_err(CE_WARN, - "XFS: osyncisdsync is now the default, option is deprecated."); - } else if (!strcmp(this_char, "irixsgid")) { - cmn_err(CE_WARN, - "XFS: irixsgid is now a sysctl(2) variable, option is deprecated."); - } else { - cmn_err(CE_WARN, - "XFS: unknown mount option [%s].", this_char); - return EINVAL; - } - } - - if (args->flags & XFSMNT_NORECOVERY) { - if ((mp->m_flags & XFS_MOUNT_RDONLY) == 0) { - cmn_err(CE_WARN, - "XFS: no-recovery mounts must be read-only."); - return EINVAL; - } - } - - if ((args->flags & XFSMNT_NOALIGN) && (dsunit || dswidth)) { - cmn_err(CE_WARN, - "XFS: sunit and swidth options incompatible with the noalign option"); - return EINVAL; - } - - if ((args->flags & XFSMNT_GQUOTA) && (args->flags & XFSMNT_PQUOTA)) { - cmn_err(CE_WARN, - "XFS: cannot mount with both project and group quota"); - return EINVAL; - } - - if ((args->flags & XFSMNT_DMAPI) && *args->mtpt == '\0') { - printk("XFS: %s option needs the mount point option as well\n", - MNTOPT_DMAPI); - return EINVAL; - } - - if ((dsunit && !dswidth) || (!dsunit && dswidth)) { - cmn_err(CE_WARN, - "XFS: sunit and swidth must be specified together"); - return EINVAL; - } - - if (dsunit && (dswidth % dsunit != 0)) { - cmn_err(CE_WARN, - "XFS: stripe width (%d) must be a multiple of the stripe unit (%d)", - dswidth, dsunit); - return EINVAL; - } - - /* - * Applications using DMI filesystems often expect the - * inode generation number to be monotonically increasing. - * If we delete inode chunks we break this assumption, so - * keep unused inode chunks on disk for DMI filesystems - * until we come up with a better solution. - * Note that if "ikeep" or "noikeep" mount options are - * supplied, then they are honored. - */ - if (!(args->flags & XFSMNT_DMAPI) && !ikeep) - args->flags |= XFSMNT_IDELETE; - - if ((args->flags & XFSMNT_NOALIGN) != XFSMNT_NOALIGN) { - if (dsunit) { - args->sunit = dsunit; - args->flags |= XFSMNT_RETERR; - } else { - args->sunit = vol_dsunit; - } - dswidth ? (args->swidth = dswidth) : - (args->swidth = vol_dswidth); - } else { - args->sunit = args->swidth = 0; - } - -done: - if (args->flags & XFSMNT_32BITINODES) - mp->m_flags |= XFS_MOUNT_SMALL_INUMS; - if (args->flags2) - args->flags |= XFSMNT_FLAGS2; - return 0; -} - -int -xfs_showargs( - struct xfs_mount *mp, - struct seq_file *m) -{ - static struct proc_xfs_info { - int flag; - char *str; - } xfs_info[] = { - /* the few simple ones we can get from the mount struct */ - { XFS_MOUNT_WSYNC, "," MNTOPT_WSYNC }, - { XFS_MOUNT_INO64, "," MNTOPT_INO64 }, - { XFS_MOUNT_NOALIGN, "," MNTOPT_NOALIGN }, - { XFS_MOUNT_SWALLOC, "," MNTOPT_SWALLOC }, - { XFS_MOUNT_NOUUID, "," MNTOPT_NOUUID }, - { XFS_MOUNT_NORECOVERY, "," MNTOPT_NORECOVERY }, - { XFS_MOUNT_OSYNCISOSYNC, "," MNTOPT_OSYNCISOSYNC }, - { 0, NULL } - }; - struct proc_xfs_info *xfs_infop; - - for (xfs_infop = xfs_info; xfs_infop->flag; xfs_infop++) { - if (mp->m_flags & xfs_infop->flag) - seq_puts(m, xfs_infop->str); - } - - if (mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) - seq_printf(m, "," MNTOPT_ALLOCSIZE "=%dk", - (int)(1 << mp->m_writeio_log) >> 10); - - if (mp->m_logbufs > 0) - seq_printf(m, "," MNTOPT_LOGBUFS "=%d", mp->m_logbufs); - if (mp->m_logbsize > 0) - seq_printf(m, "," MNTOPT_LOGBSIZE "=%dk", mp->m_logbsize >> 10); - - if (mp->m_logname) - seq_printf(m, "," MNTOPT_LOGDEV "=%s", mp->m_logname); - if (mp->m_rtname) - seq_printf(m, "," MNTOPT_RTDEV "=%s", mp->m_rtname); - - if (mp->m_dalign > 0) - seq_printf(m, "," MNTOPT_SUNIT "=%d", - (int)XFS_FSB_TO_BB(mp, mp->m_dalign)); - if (mp->m_swidth > 0) - seq_printf(m, "," MNTOPT_SWIDTH "=%d", - (int)XFS_FSB_TO_BB(mp, mp->m_swidth)); - - if (!(mp->m_flags & XFS_MOUNT_IDELETE)) - seq_printf(m, "," MNTOPT_IKEEP); - if (!(mp->m_flags & XFS_MOUNT_COMPAT_IOSIZE)) - seq_printf(m, "," MNTOPT_LARGEIO); - - if (!(mp->m_flags & XFS_MOUNT_SMALL_INUMS)) - seq_printf(m, "," MNTOPT_64BITINODE); - if (mp->m_flags & XFS_MOUNT_GRPID) - seq_printf(m, "," MNTOPT_GRPID); - - if (mp->m_qflags & XFS_UQUOTA_ACCT) { - if (mp->m_qflags & XFS_UQUOTA_ENFD) - seq_puts(m, "," MNTOPT_USRQUOTA); - else - seq_puts(m, "," MNTOPT_UQUOTANOENF); - } - - if (mp->m_qflags & XFS_PQUOTA_ACCT) { - if (mp->m_qflags & XFS_OQUOTA_ENFD) - seq_puts(m, "," MNTOPT_PRJQUOTA); - else - seq_puts(m, "," MNTOPT_PQUOTANOENF); - } - - if (mp->m_qflags & XFS_GQUOTA_ACCT) { - if (mp->m_qflags & XFS_OQUOTA_ENFD) - seq_puts(m, "," MNTOPT_GRPQUOTA); - else - seq_puts(m, "," MNTOPT_GQUOTANOENF); - } - - if (!(mp->m_qflags & XFS_ALL_QUOTA_ACCT)) - seq_puts(m, "," MNTOPT_NOQUOTA); - - if (mp->m_flags & XFS_MOUNT_DMAPI) - seq_puts(m, "," MNTOPT_DMAPI); - return 0; -} - -/* - * Second stage of a freeze. The data is already frozen so we only - * need to take care of themetadata. Once that's done write a dummy - * record to dirty the log in case of a crash while frozen. - */ -void -xfs_freeze( - xfs_mount_t *mp) -{ - xfs_attr_quiesce(mp); - xfs_fs_log_dummy(mp); -} diff --git a/fs/xfs/xfs_vfsops.h b/fs/xfs/xfs_vfsops.h index a592fe02a33..1688817c55e 100644 --- a/fs/xfs/xfs_vfsops.h +++ b/fs/xfs/xfs_vfsops.h @@ -13,16 +13,9 @@ int xfs_mount(struct xfs_mount *mp, struct xfs_mount_args *args, int xfs_unmount(struct xfs_mount *mp, int flags, struct cred *credp); int xfs_mntupdate(struct xfs_mount *mp, int *flags, struct xfs_mount_args *args); -int xfs_root(struct xfs_mount *mp, bhv_vnode_t **vpp); -int xfs_statvfs(struct xfs_mount *mp, struct kstatfs *statp, - bhv_vnode_t *vp); int xfs_sync(struct xfs_mount *mp, int flags); -int xfs_vget(struct xfs_mount *mp, bhv_vnode_t **vpp, struct xfs_fid *xfid); -int xfs_parseargs(struct xfs_mount *mp, char *options, - struct xfs_mount_args *args, int update); -int xfs_showargs(struct xfs_mount *mp, struct seq_file *m); -void xfs_freeze(struct xfs_mount *mp); void xfs_do_force_shutdown(struct xfs_mount *mp, int flags, char *fname, int lnnum); +void xfs_attr_quiesce(struct xfs_mount *mp); #endif /* _XFS_VFSOPS_H */ diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index efd5aff9eaf..64c5953feca 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -88,7 +88,7 @@ xfs_getattr( bhv_vnode_t *vp = XFS_ITOV(ip); xfs_mount_t *mp = ip->i_mount; - vn_trace_entry(ip, __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(ip); if (XFS_FORCED_SHUTDOWN(mp)) return XFS_ERROR(EIO); @@ -136,7 +136,7 @@ xfs_getattr( default: vap->va_rdev = 0; - if (!(ip->i_d.di_flags & XFS_DIFLAG_REALTIME)) { + if (!(XFS_IS_REALTIME_INODE(ip))) { vap->va_blocksize = xfs_preferred_iosize(mp); } else { @@ -228,7 +228,7 @@ xfs_setattr( int file_owner; int need_iolock = 1; - vn_trace_entry(ip, __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(ip); if (mp->m_flags & XFS_MOUNT_RDONLY) return XFS_ERROR(EROFS); @@ -508,7 +508,7 @@ xfs_setattr( */ if ((ip->i_d.di_nextents || ip->i_delayed_blks) && (mask & XFS_AT_XFLAGS) && - (ip->i_d.di_flags & XFS_DIFLAG_REALTIME) != + (XFS_IS_REALTIME_INODE(ip)) != (vap->va_xflags & XFS_XFLAG_REALTIME)) { code = XFS_ERROR(EINVAL); /* EFBIG? */ goto error_return; @@ -520,7 +520,7 @@ xfs_setattr( if ((mask & XFS_AT_EXTSIZE) && vap->va_extsize != 0) { xfs_extlen_t size; - if ((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) || + if (XFS_IS_REALTIME_INODE(ip) || ((mask & XFS_AT_XFLAGS) && (vap->va_xflags & XFS_XFLAG_REALTIME))) { size = mp->m_sb.sb_rextsize << @@ -804,12 +804,8 @@ xfs_setattr( if (vap->va_xflags & XFS_XFLAG_EXTSZINHERIT) di_flags |= XFS_DIFLAG_EXTSZINHERIT; } else if ((ip->i_d.di_mode & S_IFMT) == S_IFREG) { - if (vap->va_xflags & XFS_XFLAG_REALTIME) { + if (vap->va_xflags & XFS_XFLAG_REALTIME) di_flags |= XFS_DIFLAG_REALTIME; - ip->i_iocore.io_flags |= XFS_IOCORE_RT; - } else { - ip->i_iocore.io_flags &= ~XFS_IOCORE_RT; - } if (vap->va_xflags & XFS_XFLAG_EXTSIZE) di_flags |= XFS_DIFLAG_EXTSIZE; } @@ -902,28 +898,6 @@ xfs_setattr( return code; } - -/* - * xfs_access - * Null conversion from vnode mode bits to inode mode bits, as in efs. - */ -int -xfs_access( - xfs_inode_t *ip, - int mode, - cred_t *credp) -{ - int error; - - vn_trace_entry(ip, __FUNCTION__, (inst_t *)__return_address); - - xfs_ilock(ip, XFS_ILOCK_SHARED); - error = xfs_iaccess(ip, mode, credp); - xfs_iunlock(ip, XFS_ILOCK_SHARED); - return error; -} - - /* * The maximum pathlen is 1024 bytes. Since the minimum file system * blocksize is 512 bytes, we can get a max of 2 extents back from @@ -987,7 +961,7 @@ xfs_readlink( int pathlen; int error = 0; - vn_trace_entry(ip, __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(ip); if (XFS_FORCED_SHUTDOWN(mp)) return XFS_ERROR(EIO); @@ -1033,7 +1007,7 @@ xfs_fsync( int error; int log_flushed = 0, changed = 1; - vn_trace_entry(ip, __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(ip); ASSERT(start >= 0 && stop >= -1); @@ -1149,7 +1123,7 @@ xfs_fsync( * If this inode is on the RT dev we need to flush that * cache as well. */ - if (ip->i_d.di_flags & XFS_DIFLAG_REALTIME) + if (XFS_IS_REALTIME_INODE(ip)) xfs_blkdev_issue_flush(ip->i_mount->m_rtdev_targp); } @@ -1188,7 +1162,7 @@ xfs_free_eofblocks( nimaps = 1; xfs_ilock(ip, XFS_ILOCK_SHARED); - error = XFS_BMAPI(mp, NULL, &ip->i_iocore, end_fsb, map_len, 0, + error = xfs_bmapi(NULL, ip, end_fsb, map_len, 0, NULL, 0, &imap, &nimaps, NULL, NULL); xfs_iunlock(ip, XFS_ILOCK_SHARED); @@ -1562,9 +1536,6 @@ xfs_release( error = xfs_free_eofblocks(mp, ip, XFS_FREE_EOF_LOCK); if (error) return error; - /* Update linux inode block count after free above */ - vn_to_inode(vp)->i_blocks = XFS_FSB_TO_BB(mp, - ip->i_d.di_nblocks + ip->i_delayed_blks); } } @@ -1592,7 +1563,7 @@ xfs_inactive( int error; int truncate; - vn_trace_entry(ip, __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(ip); /* * If the inode is already free, then there can be nothing @@ -1638,9 +1609,6 @@ xfs_inactive( error = xfs_free_eofblocks(mp, ip, XFS_FREE_EOF_LOCK); if (error) return VN_INACTIVE_CACHE; - /* Update linux inode block count after free above */ - vn_to_inode(vp)->i_blocks = XFS_FSB_TO_BB(mp, - ip->i_d.di_nblocks + ip->i_delayed_blks); } goto out; } @@ -1805,7 +1773,7 @@ xfs_lookup( int error; uint lock_mode; - vn_trace_entry(dp, __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(dp); if (XFS_FORCED_SHUTDOWN(dp->i_mount)) return XFS_ERROR(EIO); @@ -1814,7 +1782,7 @@ xfs_lookup( error = xfs_dir_lookup_int(dp, lock_mode, dentry, &e_inum, &ip); if (!error) { *vpp = XFS_ITOV(ip); - ITRACE(ip); + xfs_itrace_ref(ip); } xfs_iunlock_map_shared(dp, lock_mode); return error; @@ -1848,7 +1816,7 @@ xfs_create( int namelen; ASSERT(!*vpp); - vn_trace_entry(dp, __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(dp); namelen = VNAMELEN(dentry); @@ -1930,7 +1898,7 @@ xfs_create( goto error_return; goto abort_return; } - ITRACE(ip); + xfs_itrace_ref(ip); /* * At this point, we've gotten a newly allocated inode. @@ -2098,7 +2066,7 @@ again: e_inum = ip->i_ino; - ITRACE(ip); + xfs_itrace_ref(ip); /* * We want to lock in increasing inum. Since we've already @@ -2321,7 +2289,7 @@ xfs_remove( uint resblks; int namelen; - vn_trace_entry(dp, __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(dp); if (XFS_FORCED_SHUTDOWN(mp)) return XFS_ERROR(EIO); @@ -2364,9 +2332,8 @@ xfs_remove( dm_di_mode = ip->i_d.di_mode; - vn_trace_entry(ip, __FUNCTION__, (inst_t *)__return_address); - - ITRACE(ip); + xfs_itrace_entry(ip); + xfs_itrace_ref(ip); error = XFS_QM_DQATTACH(mp, dp, 0); if (!error && dp != ip) @@ -2498,8 +2465,7 @@ xfs_remove( if (link_zero && xfs_inode_is_filestream(ip)) xfs_filestream_deassociate(ip); - vn_trace_exit(ip, __FUNCTION__, (inst_t *)__return_address); - + xfs_itrace_exit(ip); IRELE(ip); /* Fall through to std_return with error = 0 */ @@ -2562,8 +2528,8 @@ xfs_link( char *target_name = VNAME(dentry); int target_namelen; - vn_trace_entry(tdp, __FUNCTION__, (inst_t *)__return_address); - vn_trace_entry(xfs_vtoi(src_vp), __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(tdp); + xfs_itrace_entry(xfs_vtoi(src_vp)); target_namelen = VNAMELEN(dentry); ASSERT(!VN_ISDIR(src_vp)); @@ -2744,7 +2710,7 @@ xfs_mkdir( /* Return through std_return after this point. */ - vn_trace_entry(dp, __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(dp); mp = dp->i_mount; udqp = gdqp = NULL; @@ -2810,7 +2776,7 @@ xfs_mkdir( goto error_return; goto abort_return; } - ITRACE(cdp); + xfs_itrace_ref(cdp); /* * Now we add the directory inode to the transaction. @@ -2936,7 +2902,7 @@ xfs_rmdir( int last_cdp_link; uint resblks; - vn_trace_entry(dp, __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(dp); if (XFS_FORCED_SHUTDOWN(mp)) return XFS_ERROR(EIO); @@ -3041,7 +3007,7 @@ xfs_rmdir( VN_HOLD(dir_vp); } - ITRACE(cdp); + xfs_itrace_ref(cdp); xfs_trans_ijoin(tp, cdp, XFS_ILOCK_EXCL); ASSERT(cdp->i_d.di_nlink >= 2); @@ -3189,8 +3155,7 @@ xfs_symlink( ip = NULL; tp = NULL; - vn_trace_entry(dp, __FUNCTION__, (inst_t *)__return_address); - + xfs_itrace_entry(dp); if (XFS_FORCED_SHUTDOWN(mp)) return XFS_ERROR(EIO); @@ -3317,7 +3282,7 @@ xfs_symlink( goto error_return; goto error1; } - ITRACE(ip); + xfs_itrace_ref(ip); /* * An error after we've joined dp to the transaction will result in the @@ -3465,27 +3430,6 @@ std_return: goto std_return; } - -int -xfs_fid2( - xfs_inode_t *ip, - xfs_fid_t *xfid) -{ - vn_trace_entry(ip, __FUNCTION__, (inst_t *)__return_address); - - xfid->fid_len = sizeof(xfs_fid_t) - sizeof(xfid->fid_len); - xfid->fid_pad = 0; - /* - * use memcpy because the inode is a long long and there's no - * assurance that xfid->fid_ino is properly aligned. - */ - memcpy(&xfid->fid_ino, &ip->i_ino, sizeof(xfid->fid_ino)); - xfid->fid_gen = ip->i_d.di_gen; - - return 0; -} - - int xfs_rwlock( xfs_inode_t *ip, @@ -3558,11 +3502,11 @@ xfs_inode_flush( if (iip && iip->ili_last_lsn) { xlog_t *log = mp->m_log; xfs_lsn_t sync_lsn; - int s, log_flags = XFS_LOG_FORCE; + int log_flags = XFS_LOG_FORCE; - s = GRANT_LOCK(log); + spin_lock(&log->l_grant_lock); sync_lsn = log->l_last_sync_lsn; - GRANT_UNLOCK(log, s); + spin_unlock(&log->l_grant_lock); if ((XFS_LSN_CMP(iip->ili_last_lsn, sync_lsn) > 0)) { if (flags & FLUSH_SYNC) @@ -3637,8 +3581,8 @@ xfs_set_dmattrs( xfs_ilock(ip, XFS_ILOCK_EXCL); xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); - ip->i_iocore.io_dmevmask = ip->i_d.di_dmevmask = evmask; - ip->i_iocore.io_dmstate = ip->i_d.di_dmstate = state; + ip->i_d.di_dmevmask = evmask; + ip->i_d.di_dmstate = state; xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); IHOLD(ip); @@ -3653,7 +3597,7 @@ xfs_reclaim( { bhv_vnode_t *vp = XFS_ITOV(ip); - vn_trace_entry(ip, __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(ip); ASSERT(!VN_MAPPED(vp)); @@ -3871,7 +3815,7 @@ xfs_alloc_file_space( int committed; int error; - vn_trace_entry(ip, __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(ip); if (XFS_FORCED_SHUTDOWN(mp)) return XFS_ERROR(EIO); @@ -3976,7 +3920,7 @@ retry: * Issue the xfs_bmapi() call to allocate the blocks */ XFS_BMAP_INIT(&free_list, &firstfsb); - error = XFS_BMAPI(mp, tp, &ip->i_iocore, startoffset_fsb, + error = xfs_bmapi(tp, ip, startoffset_fsb, allocatesize_fsb, bmapi_flag, &firstfsb, 0, imapp, &nimaps, &free_list, NULL); @@ -4052,13 +3996,13 @@ xfs_zero_remaining_bytes( int error = 0; bp = xfs_buf_get_noaddr(mp->m_sb.sb_blocksize, - ip->i_d.di_flags & XFS_DIFLAG_REALTIME ? + XFS_IS_REALTIME_INODE(ip) ? mp->m_rtdev_targp : mp->m_ddev_targp); for (offset = startoff; offset <= endoff; offset = lastoffset + 1) { offset_fsb = XFS_B_TO_FSBT(mp, offset); nimap = 1; - error = XFS_BMAPI(mp, NULL, &ip->i_iocore, offset_fsb, 1, 0, + error = xfs_bmapi(NULL, ip, offset_fsb, 1, 0, NULL, 0, &imap, &nimap, NULL, NULL); if (error || nimap < 1) break; @@ -4141,7 +4085,7 @@ xfs_free_file_space( vp = XFS_ITOV(ip); mp = ip->i_mount; - vn_trace_entry(ip, __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(ip); if ((error = XFS_QM_DQATTACH(mp, ip, 0))) return error; @@ -4149,7 +4093,7 @@ xfs_free_file_space( error = 0; if (len <= 0) /* if nothing being freed */ return error; - rt = (ip->i_d.di_flags & XFS_DIFLAG_REALTIME); + rt = XFS_IS_REALTIME_INODE(ip); startoffset_fsb = XFS_B_TO_FSB(mp, offset); end_dmi_offset = offset + len; endoffset_fsb = XFS_B_TO_FSBT(mp, end_dmi_offset); @@ -4172,15 +4116,12 @@ xfs_free_file_space( vn_iowait(ip); /* wait for the completion of any pending DIOs */ } - rounding = max_t(uint, 1 << mp->m_sb.sb_blocklog, NBPP); + rounding = max_t(uint, 1 << mp->m_sb.sb_blocklog, PAGE_CACHE_SIZE); ioffset = offset & ~(rounding - 1); if (VN_CACHED(vp) != 0) { - xfs_inval_cached_trace(&ip->i_iocore, ioffset, -1, - ctooff(offtoct(ioffset)), -1); - error = xfs_flushinval_pages(ip, - ctooff(offtoct(ioffset)), - -1, FI_REMAPF_LOCKED); + xfs_inval_cached_trace(ip, ioffset, -1, ioffset, -1); + error = xfs_flushinval_pages(ip, ioffset, -1, FI_REMAPF_LOCKED); if (error) goto out_unlock_iolock; } @@ -4191,9 +4132,9 @@ xfs_free_file_space( * actually need to zero the extent edges. Otherwise xfs_bunmapi * will take care of it for us. */ - if (rt && !XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb)) { + if (rt && !xfs_sb_version_hasextflgbit(&mp->m_sb)) { nimap = 1; - error = XFS_BMAPI(mp, NULL, &ip->i_iocore, startoffset_fsb, + error = xfs_bmapi(NULL, ip, startoffset_fsb, 1, 0, NULL, 0, &imap, &nimap, NULL, NULL); if (error) goto out_unlock_iolock; @@ -4208,7 +4149,7 @@ xfs_free_file_space( startoffset_fsb += mp->m_sb.sb_rextsize - mod; } nimap = 1; - error = XFS_BMAPI(mp, NULL, &ip->i_iocore, endoffset_fsb - 1, + error = xfs_bmapi(NULL, ip, endoffset_fsb - 1, 1, 0, NULL, 0, &imap, &nimap, NULL, NULL); if (error) goto out_unlock_iolock; @@ -4284,7 +4225,7 @@ xfs_free_file_space( * issue the bunmapi() call to free the blocks */ XFS_BMAP_INIT(&free_list, &firstfsb); - error = XFS_BUNMAPI(mp, tp, &ip->i_iocore, startoffset_fsb, + error = xfs_bunmapi(tp, ip, startoffset_fsb, endoffset_fsb - startoffset_fsb, 0, 2, &firstfsb, &free_list, NULL, &done); if (error) { @@ -4347,23 +4288,11 @@ xfs_change_file_space( xfs_trans_t *tp; bhv_vattr_t va; - vn_trace_entry(ip, __FUNCTION__, (inst_t *)__return_address); + xfs_itrace_entry(ip); - /* - * must be a regular file and have write permission - */ if (!S_ISREG(ip->i_d.di_mode)) return XFS_ERROR(EINVAL); - xfs_ilock(ip, XFS_ILOCK_SHARED); - - if ((error = xfs_iaccess(ip, S_IWUSR, credp))) { - xfs_iunlock(ip, XFS_ILOCK_SHARED); - return error; - } - - xfs_iunlock(ip, XFS_ILOCK_SHARED); - switch (bf->l_whence) { case 0: /*SEEK_SET*/ break; diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h index b7e461c40cf..4e3970f0e5e 100644 --- a/fs/xfs/xfs_vnodeops.h +++ b/fs/xfs/xfs_vnodeops.h @@ -18,7 +18,6 @@ int xfs_open(struct xfs_inode *ip); int xfs_getattr(struct xfs_inode *ip, struct bhv_vattr *vap, int flags); int xfs_setattr(struct xfs_inode *ip, struct bhv_vattr *vap, int flags, struct cred *credp); -int xfs_access(struct xfs_inode *ip, int mode, struct cred *credp); int xfs_readlink(struct xfs_inode *ip, char *link); int xfs_fsync(struct xfs_inode *ip, int flag, xfs_off_t start, xfs_off_t stop); @@ -39,7 +38,6 @@ int xfs_readdir(struct xfs_inode *dp, void *dirent, size_t bufsize, int xfs_symlink(struct xfs_inode *dp, bhv_vname_t *dentry, char *target_path, mode_t mode, bhv_vnode_t **vpp, struct cred *credp); -int xfs_fid2(struct xfs_inode *ip, struct xfs_fid *xfid); int xfs_rwlock(struct xfs_inode *ip, bhv_vrwlock_t locktype); void xfs_rwunlock(struct xfs_inode *ip, bhv_vrwlock_t locktype); int xfs_inode_flush(struct xfs_inode *ip, int flags); |