From 83451879ab213e152c6fe5c743f257ba58d7acd1 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 1 Dec 2005 17:12:59 -0800 Subject: [CIFS] Use fsuid (fsgid) more consistently instead of uid/gid in assembling smb requests when setuids and Linux protocol extensions enabled and in checking more matching sessions in multiuser mount mode. Pointed out by Shaggy. Signed-off-by: Steve French --- fs/cifs/CHANGES | 6 ++++++ fs/cifs/cifsfs.h | 2 +- fs/cifs/dir.c | 8 ++++---- fs/cifs/inode.c | 4 ++-- fs/cifs/misc.c | 8 ++++---- 5 files changed, 17 insertions(+), 11 deletions(-) diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index 943ef9b8224..8bef2f3a413 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -1,3 +1,7 @@ +Version 1.40 +------------ +Use fsuid (fsgid) more consistently instead of uid (gid). + Version 1.39 ------------ Defer close of a file handle slightly if pending writes depend on that handle @@ -7,6 +11,8 @@ Fix SFU style symlinks and mknod needed for servers which do not support the CIFS Unix Extensions. Fix setfacl/getfacl on bigendian. Timeout negative dentries so files that the client sees as deleted but that later get created on the server will be recognized. Add client side permission check on setattr. +Timeout stuck requests better (where server has never responded or sent corrupt +responses) Version 1.38 ------------ diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 9ec40e0e54f..821a8eb2255 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -99,5 +99,5 @@ extern ssize_t cifs_getxattr(struct dentry *, const char *, void *, size_t); extern ssize_t cifs_listxattr(struct dentry *, char *, size_t); extern int cifs_ioctl (struct inode * inode, struct file * filep, unsigned int command, unsigned long arg); -#define CIFS_VERSION "1.39" +#define CIFS_VERSION "1.40" #endif /* _CIFSFS_H */ diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index 32cc96cafa3..fed55e3c53d 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., 2002,2003 + * Copyright (C) International Business Machines Corp., 2002,2005 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify @@ -200,8 +200,8 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, (oplock & CIFS_CREATE_ACTION)) if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID) { CIFSSMBUnixSetPerms(xid, pTcon, full_path, mode, - (__u64)current->euid, - (__u64)current->egid, + (__u64)current->fsuid, + (__u64)current->fsgid, 0 /* dev */, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & @@ -325,7 +325,7 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode, else if (pTcon->ses->capabilities & CAP_UNIX) { if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID) { rc = CIFSSMBUnixSetPerms(xid, pTcon, full_path, - mode,(__u64)current->euid,(__u64)current->egid, + mode,(__u64)current->fsuid,(__u64)current->fsgid, device_number, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 411c1f7f84d..d1e99575743 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -750,8 +750,8 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID) { CIFSSMBUnixSetPerms(xid, pTcon, full_path, mode, - (__u64)current->euid, - (__u64)current->egid, + (__u64)current->fsuid, + (__u64)current->fsgid, 0 /* dev_t */, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 94baf6c8ecb..9af37f86468 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., 2002,2004 + * Copyright (C) International Business Machines Corp., 2002,2005 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify @@ -348,12 +348,12 @@ header_assemble(struct smb_hdr *buffer, char smb_command /* command */ , /* BB Add support for establishing new tCon and SMB Session */ /* with userid/password pairs found on the smb session */ /* for other target tcp/ip addresses BB */ - if(current->uid != treeCon->ses->linux_uid) { - cFYI(1,("Multiuser mode and UID did not match tcon uid ")); + if(current->fsuid != treeCon->ses->linux_uid) { + cFYI(1,("Multiuser mode and UID did not match tcon uid")); read_lock(&GlobalSMBSeslock); list_for_each(temp_item, &GlobalSMBSessionList) { ses = list_entry(temp_item, struct cifsSesInfo, cifsSessionList); - if(ses->linux_uid == current->uid) { + if(ses->linux_uid == current->fsuid) { if(ses->server == treeCon->ses->server) { cFYI(1,("found matching uid substitute right smb_uid")); buffer->Uid = ses->Suid; -- cgit v1.2.3 From bf8206791750854bc6668266b694e8fe2cacb924 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 1 Dec 2005 22:32:42 -0800 Subject: [CIFS] Kerberos and CIFS ACL support part 1 Signed-off-by: Steve French --- fs/cifs/README | 12 +++++++++++- fs/cifs/cifsacl.h | 36 ++++++++++++++++++++++++++++++++++++ fs/cifs/cifspdu.h | 2 +- fs/cifs/connect.c | 53 ++++++++++++++++++++++++++++++++++++++++++++++------- 4 files changed, 94 insertions(+), 9 deletions(-) create mode 100644 fs/cifs/cifsacl.h diff --git a/fs/cifs/README b/fs/cifs/README index e5d09a2fc7a..b0070d1b149 100644 --- a/fs/cifs/README +++ b/fs/cifs/README @@ -436,7 +436,17 @@ A partial list of the supported mount options follows: SFU does). In the future the bottom 9 bits of the mode mode also will be emulated using queries of the security descriptor (ACL). - +sec Security mode. Allowed values are: + none attempt to connection as a null user (no name) + krb5 Use Kerberos version 5 authentication + krb5i Use Kerberos authentication and packet signing + ntlm Use NTLM password hashing (default) + ntlmi Use NTLM password hashing with signing (if + /proc/fs/cifs/PacketSigningEnabled on or if + server requires signing also can be the default) + ntlmv2 Use NTLMv2 password hashing + ntlmv2i Use NTLMv2 password hashing with packet signing + The mount.cifs mount helper also accepts a few mount options before -o including: diff --git a/fs/cifs/cifsacl.h b/fs/cifs/cifsacl.h new file mode 100644 index 00000000000..4cfcdf2e630 --- /dev/null +++ b/fs/cifs/cifsacl.h @@ -0,0 +1,36 @@ +/* + * fs/cifs/cifsacl.h + * + * Copyright (c) International Business Machines Corp., 2005 + * 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 _CIFSACL_H +#define _CIFSACL_H + +struct cifs_sid { + __u8 revision; /* revision level */ + __u8 num_subauths; + __u8 authority[6]; + __u8 sub_auth[4]; + /* next sub_auth if any ... */ +} __attribute__((packed)); + +/* everyone */ +const cifs_sid sid_everyone = {1, 1, {0, 0, 0, 0, 0, 0}, {0, 0, 0, 0}}; +/* group users */ +const cifs_sid sid_user = {1, 2 , {0, 0, 0, 0, 0, 5}, {32, 545, 0, 0}}; diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index 33e1859fd2f..5253e779b3a 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -1,7 +1,7 @@ /* * fs/cifs/cifspdu.h * - * Copyright (c) International Business Machines Corp., 2002 + * Copyright (c) International Business Machines Corp., 2002,2005 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index c467de85761..651f3b6cebe 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -82,6 +82,12 @@ struct smb_vol { unsigned remap:1; /* set to remap seven reserved chars in filenames */ unsigned posix_paths:1; /* unset to not ask for posix pathnames. */ unsigned sfu_emul:1; + unsigned krb5:1; + unsigned ntlm:1; + unsigned ntlmv2:1; + unsigned nullauth:1; /* attempt to authenticate with null user */ + unsigned sign:1; + unsigned seal:1; /* encrypt */ unsigned nocase; /* request case insensitive filenames */ unsigned nobrl; /* disable sending byte range locks to srv */ unsigned int rsize; @@ -777,7 +783,7 @@ cifs_parse_mount_options(char *options, const char *devname,struct smb_vol *vol) /* vol->retry default is 0 (i.e. "soft" limited retry not hard retry) */ vol->rw = TRUE; - + vol->ntlm = TRUE; /* default is always to request posix paths. */ vol->posix_paths = 1; @@ -903,6 +909,39 @@ cifs_parse_mount_options(char *options, const char *devname,struct smb_vol *vol) printk(KERN_WARNING "CIFS: ip address too long\n"); return 1; } + } else if (strnicmp(data, "sec", 3) == 0) { + if (!value || !*value) { + cERROR(1,("no security value specified")); + continue; + } else if (strnicmp(value, "krb5i", 5) == 0) { + vol->sign = 1; + vol->krb5 = 1; + } else if (strnicmp(value, "krb5p", 5) == 0) { + /* vol->seal = 1; + vol->krb5 = 1; */ + cERROR(1,("Krb5 cifs privacy not supported")); + return 1; + } else if (strnicmp(value, "krb5", 4) == 0) { + vol->krb5 = 1; + } else if (strnicmp(value, "ntlmv2i", 7) == 0) { + vol->ntlmv2 = 1; + vol->sign = 1; + } else if (strnicmp(value, "ntlmv2", 6) == 0) { + vol->ntlmv2 = 1; + } else if (strnicmp(value, "ntlmi", 5) == 0) { + vol->ntlm = 1; + vol->sign = 1; + } else if (strnicmp(value, "ntlm", 4) == 0) { + /* ntlm is default so can be turned off too */ + vol->ntlm = 1; + } else if (strnicmp(value, "nontlm", 6) == 0) { + vol->ntlm = 0; + } else if (strnicmp(value, "none", 4) == 0) { + vol->nullauth = 1; + } else { + cERROR(1,("bad security option: %s", value)); + return 1; + } } else if ((strnicmp(data, "unc", 3) == 0) || (strnicmp(data, "target", 6) == 0) || (strnicmp(data, "path", 4) == 0)) { @@ -1546,7 +1585,7 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, cFYI(1, ("Username: %s ", volume_info.username)); } else { - cifserror("No username specified "); + cifserror("No username specified"); /* In userspace mount helper we can get user name from alternate locations such as env variables and files on disk */ kfree(volume_info.UNC); @@ -1587,7 +1626,7 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, return -EINVAL; } else /* which servers DFS root would we conect to */ { cERROR(1, - ("CIFS mount error: No UNC path (e.g. -o unc=//192.168.1.100/public) specified ")); + ("CIFS mount error: No UNC path (e.g. -o unc=//192.168.1.100/public) specified")); kfree(volume_info.UNC); kfree(volume_info.password); FreeXid(xid); @@ -1626,7 +1665,7 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, if (srvTcp) { - cFYI(1, ("Existing tcp session with server found ")); + cFYI(1, ("Existing tcp session with server found")); } else { /* create socket */ if(volume_info.port) sin_server.sin_port = htons(volume_info.port); @@ -1689,11 +1728,11 @@ 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")); kfree(volume_info.password); /* volume_info.UNC freed at end of function */ } else if (!rc) { - cFYI(1, ("Existing smb sess not found ")); + cFYI(1, ("Existing smb sess not found")); pSesInfo = sesInfoAlloc(); if (pSesInfo == NULL) rc = -ENOMEM; @@ -1777,7 +1816,7 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, find_unc(sin_server.sin_addr.s_addr, volume_info.UNC, volume_info.username); if (tcon) { - cFYI(1, ("Found match on UNC path ")); + cFYI(1, ("Found match on UNC path")); /* we can have only one retry value for a connection to a share so for resources mounted more than once to the same server share the last value passed in -- cgit v1.2.3 From 84afc29b185334f489975a003b128e1b15e24a54 Mon Sep 17 00:00:00 2001 From: Steve French Date: Fri, 2 Dec 2005 13:32:45 -0800 Subject: [CIFS] Readpages and readir performance improvements - eliminate extra memcpy. Part 1 Signed-off-by: Steve French --- fs/cifs/CHANGES | 3 ++- fs/cifs/cifsencrypt.c | 55 ++++++++++++++++++++++++++++++++++++++++++++++++++- fs/cifs/cifsproto.h | 6 +++--- fs/cifs/cifssmb.c | 5 +---- fs/cifs/file.c | 6 ------ fs/cifs/transport.c | 19 ++++++++++-------- 6 files changed, 71 insertions(+), 23 deletions(-) diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index 8bef2f3a413..1d2137561c5 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -1,6 +1,7 @@ Version 1.40 ------------ -Use fsuid (fsgid) more consistently instead of uid (gid). +Use fsuid (fsgid) more consistently instead of uid (gid). Improve performance +of readpages by eliminating one extra memcpy. Version 1.39 ------------ diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c index fe2bb7c4c91..a2c24858d40 100644 --- a/fs/cifs/cifsencrypt.c +++ b/fs/cifs/cifsencrypt.c @@ -1,7 +1,7 @@ /* * fs/cifs/cifsencrypt.c * - * Copyright (C) International Business Machines Corp., 2003 + * Copyright (C) International Business Machines Corp., 2005 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify @@ -82,6 +82,59 @@ int cifs_sign_smb(struct smb_hdr * cifs_pdu, struct TCP_Server_Info * server, return rc; } +static int cifs_calc_signature2(const struct kvec * iov, int n_vec, + const char * key, char * signature) +{ + struct MD5Context context; + + if((iov == NULL) || (signature == NULL)) + return -EINVAL; + + MD5Init(&context); + MD5Update(&context,key,CIFS_SESSION_KEY_SIZE+16); + +/* MD5Update(&context,cifs_pdu->Protocol,cifs_pdu->smb_buf_length); */ /* BB FIXME BB */ + + MD5Final(signature,&context); + + return -EOPNOTSUPP; +/* return 0; */ +} + + +int cifs_sign_smb2(struct kvec * iov, int n_vec, struct TCP_Server_Info *server, + __u32 * pexpected_response_sequence_number) +{ + int rc = 0; + char smb_signature[20]; + struct smb_hdr * cifs_pdu = iov[0].iov_base; + + if((cifs_pdu == NULL) || (server == NULL)) + return -EINVAL; + + if((cifs_pdu->Flags2 & SMBFLG2_SECURITY_SIGNATURE) == 0) + return rc; + + spin_lock(&GlobalMid_Lock); + cifs_pdu->Signature.Sequence.SequenceNumber = + cpu_to_le32(server->sequence_number); + cifs_pdu->Signature.Sequence.Reserved = 0; + + *pexpected_response_sequence_number = server->sequence_number++; + server->sequence_number++; + spin_unlock(&GlobalMid_Lock); + + rc = cifs_calc_signature2(iov, n_vec, server->mac_signing_key, + smb_signature); + if(rc) + memset(cifs_pdu->Signature.SecuritySignature, 0, 8); + else + memcpy(cifs_pdu->Signature.SecuritySignature, smb_signature, 8); + + return rc; + +} + int cifs_verify_signature(struct smb_hdr * cifs_pdu, const char * mac_key, __u32 expected_sequence_number) { diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 1b73f4f4c5c..c058f8a45b2 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -48,7 +48,7 @@ extern int SendReceive(const unsigned int /* xid */ , struct cifsSesInfo *, struct smb_hdr * /* out */ , int * /* bytes returned */ , const int long_op); extern int SendReceive2(const unsigned int /* xid */ , struct cifsSesInfo *, - struct kvec *, int /* nvec */, + struct kvec *, int /* nvec to send */, int * /* bytes returned */ , const int long_op); extern int checkSMBhdr(struct smb_hdr *smb, __u16 mid); extern int checkSMB(struct smb_hdr *smb, __u16 mid, int length); @@ -237,12 +237,10 @@ extern int CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, const __u64 lseek, unsigned int *nbytes, const char *buf, const char __user *ubuf, const int long_op); -#ifdef CONFIG_CIFS_EXPERIMENTAL extern int CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, const int netfid, const unsigned int count, const __u64 offset, unsigned int *nbytes, struct kvec *iov, const int nvec, const int long_op); -#endif /* CONFIG_CIFS_EXPERIMENTAL */ extern int CIFSGetSrvInodeNumber(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, __u64 * inode_number, const struct nls_table *nls_codepage, @@ -269,6 +267,8 @@ extern void tconInfoFree(struct cifsTconInfo *); extern int cifs_reconnect(struct TCP_Server_Info *server); extern int cifs_sign_smb(struct smb_hdr *, struct TCP_Server_Info *,__u32 *); +extern int cifs_sign_smb2(struct kvec *iov, int n_vec, struct TCP_Server_Info *, + __u32 *); extern int cifs_verify_signature(struct smb_hdr *, const char * mac_key, __u32 expected_sequence_number); extern int cifs_calculate_mac_key(char * key,const char * rn,const char * pass); diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 6867e556d37..3565d3bf2e3 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -1155,7 +1155,6 @@ CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, return rc; } -#ifdef CONFIG_CIFS_EXPERIMENTAL int CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, const int netfid, const unsigned int count, @@ -1223,7 +1222,7 @@ CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, *nbytes = le16_to_cpu(pSMBr->CountHigh); *nbytes = (*nbytes) << 16; *nbytes += le16_to_cpu(pSMBr->Count); - } + } cifs_small_buf_release(pSMB); @@ -1234,8 +1233,6 @@ CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, } -#endif /* CIFS_EXPERIMENTAL */ - int CIFSSMBLock(const int xid, struct cifsTconInfo *tcon, const __u16 smb_file_id, const __u64 len, diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 14a1c72ced9..b67be3d8c01 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -870,7 +870,6 @@ static ssize_t cifs_write(struct file *file, const char *write_data, if (rc != 0) break; } -#ifdef CONFIG_CIFS_EXPERIMENTAL /* BB FIXME We can not sign across two buffers yet */ if((experimEnabled) && ((pTcon->ses->server->secMode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) == 0)) { @@ -889,7 +888,6 @@ static ssize_t cifs_write(struct file *file, const char *write_data, iov, 1, long_op); } else /* BB FIXME fixup indentation of line below */ -#endif rc = CIFSSMBWrite(xid, pTcon, open_file->netfid, min_t(const int, cifs_sb->wsize, @@ -1026,7 +1024,6 @@ static int cifs_partialpagewrite(struct page *page, unsigned from, unsigned to) return rc; } -#ifdef CONFIG_CIFS_EXPERIMENTAL static int cifs_writepages(struct address_space *mapping, struct writeback_control *wbc) { @@ -1229,7 +1226,6 @@ retry: return rc; } -#endif static int cifs_writepage(struct page* page, struct writeback_control *wbc) { @@ -1875,9 +1871,7 @@ struct address_space_operations cifs_addr_ops = { .readpage = cifs_readpage, .readpages = cifs_readpages, .writepage = cifs_writepage, -#ifdef CONFIG_CIFS_EXPERIMENTAL .writepages = cifs_writepages, -#endif .prepare_write = cifs_prepare_write, .commit_write = cifs_commit_write, .set_page_dirty = __set_page_dirty_nobuffers, diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index f8871196098..0abfbf4e4a4 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -206,7 +206,6 @@ smb_send(struct socket *ssocket, struct smb_hdr *smb_buffer, return rc; } -#ifdef CONFIG_CIFS_EXPERIMENTAL static int smb_send2(struct socket *ssocket, struct kvec *iov, int n_vec, struct sockaddr *sin) @@ -392,8 +391,7 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, return -ENOMEM; } -/* BB FIXME */ -/* rc = cifs_sign_smb2(iov, n_vec, ses->server, &midQ->sequence_number); */ + rc = cifs_sign_smb2(iov, n_vec, ses->server, &midQ->sequence_number); midQ->midState = MID_REQUEST_SUBMITTED; #ifdef CONFIG_CIFS_STATS2 @@ -492,11 +490,17 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, if (midQ->resp_buf && (midQ->midState == MID_RESPONSE_RECEIVED)) { + in_buf->smb_buf_length = receive_len; - /* BB verify that length would not overrun small buf */ - memcpy((char *)in_buf + 4, - (char *)midQ->resp_buf + 4, - receive_len); + if(receive_len > 500) { + /* use multiple buffers on way out */ + } else { + memcpy((char *)in_buf + 4, + (char *)midQ->resp_buf + 4, + receive_len); + iov[0].iov_len = receive_len + 4; + iov[1].iov_len = 0; + } dump_smb(in_buf, 80); /* convert the length into a more usable form */ @@ -549,7 +553,6 @@ out_unlock2: return rc; } -#endif /* CIFS_EXPERIMENTAL */ int SendReceive(const unsigned int xid, struct cifsSesInfo *ses, -- cgit v1.2.3 From 4498eed50a114565debd38f173acd62cce6e7cb8 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sat, 3 Dec 2005 13:58:57 -0800 Subject: [CIFS] Add extended stats (STATS2) for total buffer allocations for better performance debugging. Signed-off-by: Steve French --- fs/cifs/cifs_debug.c | 4 ++++ fs/cifs/cifsfs.c | 6 ++++++ fs/cifs/cifsglob.h | 8 ++++++-- fs/cifs/misc.c | 7 +++++++ 4 files changed, 23 insertions(+), 2 deletions(-) diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index 22a444a3fe4..7f19f2547d0 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -219,6 +219,10 @@ cifs_stats_write(struct file *file, const char __user *buffer, if (c == '1' || c == 'y' || c == 'Y' || c == '0') { read_lock(&GlobalSMBSeslock); +#ifdef CONFIG_CIFS_STATS2 + atomic_set(&totBufAllocCount, 0); + atomic_set(&totSmBufAllocCount, 0); +#endif /* CONFIG_CIFS_STATS2 */ list_for_each(tmp, &GlobalTreeConnectionList) { tcon = list_entry(tmp, struct cifsTconInfo, cifsConnectionList); diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 2a13a2bac8f..ba90903909a 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -954,6 +954,12 @@ init_cifs(void) atomic_set(&tconInfoReconnectCount, 0); atomic_set(&bufAllocCount, 0); + atomic_set(&smBufAllocCount, 0); +#ifdef CONFIG_CIFS_STATS2 + atomic_set(&totBufAllocCount, 0); + atomic_set(&totSmBufAllocCount, 0); +#endif /* CONFIG_CIFS_STATS2 */ + atomic_set(&midCount, 0); GlobalCurrentXid = 0; GlobalTotalActiveXid = 0; diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 1ba08f8c5bc..c011c278af4 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -505,8 +505,12 @@ GLOBAL_EXTERN atomic_t tcpSesReconnectCount; GLOBAL_EXTERN atomic_t tconInfoReconnectCount; /* Various Debug counters to remove someday (BB) */ -GLOBAL_EXTERN atomic_t bufAllocCount; -GLOBAL_EXTERN atomic_t smBufAllocCount; +GLOBAL_EXTERN atomic_t bufAllocCount; /* current number allocated */ +#ifdef CONFIG_CIFS_STATS2 +GLOBAL_EXTERN atomic_t totBufAllocCount; /* total allocated over all time */ +GLOBAL_EXTERN atomic_t totSmBufAllocCount; +#endif +GLOBAL_EXTERN atomic_t smBufAllocCount; GLOBAL_EXTERN atomic_t midCount; /* Misc globals */ diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 9af37f86468..ac5a72a299a 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -161,6 +161,9 @@ cifs_buf_get(void) if (ret_buf) { memset(ret_buf, 0, sizeof(struct smb_hdr) + 3); atomic_inc(&bufAllocCount); +#ifdef CONFIG_CIFS_STATS2 + atomic_inc(&totBufAllocCount); +#endif /* CONFIG_CIFS_STATS2 */ } return ret_buf; @@ -195,6 +198,10 @@ cifs_small_buf_get(void) /* No need to clear memory here, cleared in header assemble */ /* memset(ret_buf, 0, sizeof(struct smb_hdr) + 27);*/ atomic_inc(&smBufAllocCount); +#ifdef CONFIG_CIFS_STATS2 + atomic_inc(&totSmBufAllocCount); +#endif /* CONFIG_CIFS_STATS2 */ + } return ret_buf; } -- cgit v1.2.3 From 07475ffba5800c53573180dd521273642adcd0e9 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sat, 3 Dec 2005 14:11:37 -0800 Subject: [CIFS] Display large/small total buffer allocations in /proc/fs/cifs/Stats when CONFIG_CIFS_STATS2 is on (helps in debugging performance) Signed-off-by: Steve French --- fs/cifs/cifs_debug.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index 7f19f2547d0..6f5d81f5eac 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -280,6 +280,14 @@ cifs_stats_read(char *buf, char **beginBuffer, off_t offset, smBufAllocCount.counter,cifs_min_small); length += item_length; buf += item_length; +#ifdef CONFIG_CIFS_STATS2 + item_length = sprintf(buf, "Total Large %d Small %d Allocations\n", + atomic_read(&totBufAllocCount), + atomic_read(&totSmBufAllocCount)); + length += item_length; + buf += item_length; +#endif /* CONFIG_CIFS_STATS2 */ + item_length = sprintf(buf,"Operations (MIDs): %d\n", midCount.counter); -- cgit v1.2.3 From ec637e3ffb6b978143652477c7c5f96c9519b691 Mon Sep 17 00:00:00 2001 From: Steve French Date: Mon, 12 Dec 2005 20:53:18 -0800 Subject: [CIFS] Avoid extra large buffer allocation (and memcpy) in cifs_readpages Signed-off-by: Steve French --- fs/cifs/cifs_debug.c | 39 +++++++++++----------- fs/cifs/cifsfs.c | 5 +-- fs/cifs/cifsglob.h | 8 ++++- fs/cifs/cifspdu.h | 6 +++- fs/cifs/cifsproto.h | 14 ++++---- fs/cifs/cifssmb.c | 93 ++++++++++++++++++++++++++++++++-------------------- fs/cifs/connect.c | 2 +- fs/cifs/file.c | 52 ++++++++++++++++++----------- fs/cifs/inode.c | 5 +-- fs/cifs/misc.c | 2 +- fs/cifs/transport.c | 38 ++++++++++----------- 11 files changed, 156 insertions(+), 108 deletions(-) diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index 6f5d81f5eac..f4124a32bef 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -401,8 +401,8 @@ static read_proc_t ntlmv2_enabled_read; static write_proc_t ntlmv2_enabled_write; static read_proc_t packet_signing_enabled_read; static write_proc_t packet_signing_enabled_write; -static read_proc_t quotaEnabled_read; -static write_proc_t quotaEnabled_write; +static read_proc_t experimEnabled_read; +static write_proc_t experimEnabled_write; static read_proc_t linuxExtensionsEnabled_read; static write_proc_t linuxExtensionsEnabled_write; @@ -442,9 +442,9 @@ cifs_proc_init(void) pde->write_proc = oplockEnabled_write; pde = create_proc_read_entry("Experimental", 0, proc_fs_cifs, - quotaEnabled_read, NULL); + experimEnabled_read, NULL); if (pde) - pde->write_proc = quotaEnabled_write; + pde->write_proc = experimEnabled_write; pde = create_proc_read_entry("LinuxExtensionsEnabled", 0, proc_fs_cifs, linuxExtensionsEnabled_read, NULL); @@ -586,14 +586,13 @@ oplockEnabled_write(struct file *file, const char __user *buffer, } static int -quotaEnabled_read(char *page, char **start, off_t off, +experimEnabled_read(char *page, char **start, off_t off, int count, int *eof, void *data) { int len; len = sprintf(page, "%d\n", experimEnabled); -/* could also check if quotas are enabled in kernel - as a whole first */ + len -= off; *start = page + off; @@ -608,21 +607,23 @@ quotaEnabled_read(char *page, char **start, off_t off, return len; } static int -quotaEnabled_write(struct file *file, const char __user *buffer, +experimEnabled_write(struct file *file, const char __user *buffer, unsigned long count, void *data) { - char c; - int rc; + char c; + int rc; - rc = get_user(c, buffer); - if (rc) - return rc; - if (c == '0' || c == 'n' || c == 'N') - experimEnabled = 0; - else if (c == '1' || c == 'y' || c == 'Y') - experimEnabled = 1; + rc = get_user(c, buffer); + if (rc) + return rc; + if (c == '0' || c == 'n' || c == 'N') + experimEnabled = 0; + else if (c == '1' || c == 'y' || c == 'Y') + experimEnabled = 1; + else if (c == '2') + experimEnabled = 2; - return count; + return count; } static int @@ -632,8 +633,6 @@ linuxExtensionsEnabled_read(char *page, char **start, off_t off, int len; len = sprintf(page, "%d\n", linuxExtEnabled); -/* could also check if quotas are enabled in kernel - as a whole first */ len -= off; *start = page + off; diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index ba90903909a..582d66ca6da 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -733,7 +733,7 @@ cifs_init_request_bufs(void) kmem_cache_destroy(cifs_req_cachep); return -ENOMEM; } - /* 256 (MAX_CIFS_HDR_SIZE bytes is enough for most SMB responses and + /* MAX_CIFS_SMALL_BUFFER_SIZE bytes is enough for most SMB responses and almost all handle based requests (but not write response, nor is it sufficient for path based requests). A smaller size would have been more efficient (compacting multiple slab items on one 4k page) @@ -742,7 +742,8 @@ cifs_init_request_bufs(void) efficient to alloc 1 per page off the slab compared to 17K (5page) alloc of large cifs buffers even when page debugging is on */ cifs_sm_req_cachep = kmem_cache_create("cifs_small_rq", - MAX_CIFS_HDR_SIZE, 0, SLAB_HWCACHE_ALIGN, NULL, NULL); + MAX_CIFS_SMALL_BUFFER_SIZE, 0, SLAB_HWCACHE_ALIGN, + NULL, NULL); if (cifs_sm_req_cachep == NULL) { mempool_destroy(cifs_req_poolp); kmem_cache_destroy(cifs_req_cachep); diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index c011c278af4..862e403ff21 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -285,6 +285,7 @@ struct cifs_search_info { unsigned endOfSearch:1; unsigned emptyDir:1; unsigned unicode:1; + unsigned smallBuf:1; /* so we know which buf_release function to call */ }; struct cifsFileInfo { @@ -420,7 +421,12 @@ struct dir_notify_req { #define MID_RESPONSE_RECEIVED 4 #define MID_RETRY_NEEDED 8 /* session closed while this request out */ #define MID_NO_RESP_NEEDED 0x10 -#define MID_SMALL_BUFFER 0x20 /* 112 byte response buffer instead of 4K */ + +/* Types of response buffer returned from SendReceive2 */ +#define CIFS_NO_BUFFER 0 /* Response buffer not returned */ +#define CIFS_SMALL_BUFFER 1 +#define CIFS_LARGE_BUFFER 2 +#define CIFS_IOVEC 4 /* array of response buffers */ /* ***************************************************************** diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index 5253e779b3a..3c09fb9cc56 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -80,7 +80,11 @@ #define NT_TRANSACT_GET_USER_QUOTA 0x07 #define NT_TRANSACT_SET_USER_QUOTA 0x08 -#define MAX_CIFS_HDR_SIZE 256 /* is future chained NTCreateXReadX bigger? */ +#define MAX_CIFS_SMALL_BUFFER_SIZE 448 /* big enough for most */ +/* future chained NTCreateXReadX bigger, but for time being NTCreateX biggest */ +/* among the requests (NTCreateX response is bigger with wct of 34) */ +#define MAX_CIFS_HDR_SIZE 0x58 /* 4 len + 32 hdr + (2*24 wct) + 2 bct + 2 pad */ +#define CIFS_SMALL_PATH 120 /* allows for (448-88)/3 */ /* internal cifs vfs structures */ /***************************************************************** diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index c058f8a45b2..8ef0ce47f5b 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -49,7 +49,7 @@ extern int SendReceive(const unsigned int /* xid */ , struct cifsSesInfo *, int * /* bytes returned */ , const int long_op); extern int SendReceive2(const unsigned int /* xid */ , struct cifsSesInfo *, struct kvec *, int /* nvec to send */, - int * /* bytes returned */ , const int long_op); + int * /* type of buf returned */ , const int long_op); extern int checkSMBhdr(struct smb_hdr *smb, __u16 mid); extern int checkSMB(struct smb_hdr *smb, __u16 mid, int length); extern int is_valid_oplock_break(struct smb_hdr *smb); @@ -93,11 +93,12 @@ extern int CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, const struct nls_table *); extern int CIFSFindFirst(const int xid, struct cifsTconInfo *tcon, - const char *searchName, const struct nls_table *nls_codepage, - __u16 *searchHandle, struct cifs_search_info * psrch_inf, int map, const char dirsep); + const char *searchName, const struct nls_table *nls_codepage, + __u16 *searchHandle, struct cifs_search_info * psrch_inf, + int map, const char dirsep); extern int CIFSFindNext(const int xid, struct cifsTconInfo *tcon, - __u16 searchHandle, struct cifs_search_info * psrch_inf); + __u16 searchHandle, struct cifs_search_info * psrch_inf); extern int CIFSFindClose(const int, struct cifsTconInfo *tcon, const __u16 search_handle); @@ -230,8 +231,9 @@ extern int CIFSSMBClose(const int xid, struct cifsTconInfo *tcon, const int smb_file_id); extern int CIFSSMBRead(const int xid, struct cifsTconInfo *tcon, - const int netfid, unsigned int count, - const __u64 lseek, unsigned int *nbytes, char **buf); + const int netfid, unsigned int count, + const __u64 lseek, unsigned int *nbytes, char **buf, + int * return_buf_type); extern int CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, const int netfid, const unsigned int count, const __u64 lseek, unsigned int *nbytes, diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 3565d3bf2e3..1620991cd4d 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -958,21 +958,19 @@ openRetry: return rc; } -/* If no buffer passed in, then caller wants to do the copy - as in the case of readpages so the SMB buffer must be - freed by the caller */ - int CIFSSMBRead(const int xid, struct cifsTconInfo *tcon, - const int netfid, const unsigned int count, - const __u64 lseek, unsigned int *nbytes, char **buf) + const int netfid, const unsigned int count, + const __u64 lseek, unsigned int *nbytes, char **buf, + int * pbuf_type) { int rc = -EACCES; READ_REQ *pSMB = NULL; READ_RSP *pSMBr = NULL; char *pReadData = NULL; - int bytes_returned; int wct; + int resp_buf_type = 0; + struct kvec iov[1]; cFYI(1,("Reading %d bytes on fid %d",count,netfid)); if(tcon->ses->capabilities & CAP_LARGE_FILES) @@ -981,8 +979,7 @@ CIFSSMBRead(const int xid, struct cifsTconInfo *tcon, wct = 10; /* old style read */ *nbytes = 0; - rc = smb_init(SMB_COM_READ_ANDX, wct, tcon, (void **) &pSMB, - (void **) &pSMBr); + rc = small_smb_init(SMB_COM_READ_ANDX, wct, tcon, (void **) &pSMB); if (rc) return rc; @@ -990,13 +987,13 @@ CIFSSMBRead(const int xid, struct cifsTconInfo *tcon, if (tcon->ses->server == NULL) return -ECONNABORTED; - pSMB->AndXCommand = 0xFF; /* none */ + pSMB->AndXCommand = 0xFF; /* none */ pSMB->Fid = netfid; pSMB->OffsetLow = cpu_to_le32(lseek & 0xFFFFFFFF); if(wct == 12) pSMB->OffsetHigh = cpu_to_le32(lseek >> 32); - else if((lseek >> 32) > 0) /* can not handle this big offset for old */ - return -EIO; + else if((lseek >> 32) > 0) /* can not handle this big offset for old */ + return -EIO; pSMB->Remaining = 0; pSMB->MaxCount = cpu_to_le16(count & 0xFFFF); @@ -1005,14 +1002,18 @@ CIFSSMBRead(const int xid, struct cifsTconInfo *tcon, pSMB->ByteCount = 0; /* no need to do le conversion since 0 */ else { /* old style read */ - struct smb_com_readx_req * pSMBW = + struct smb_com_readx_req * pSMBW = (struct smb_com_readx_req *)pSMB; - pSMBW->ByteCount = 0; + pSMBW->ByteCount = 0; } - - rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, - (struct smb_hdr *) pSMBr, &bytes_returned, 0); + + iov[0].iov_base = (char *)pSMB; + iov[0].iov_len = pSMB->hdr.smb_buf_length + 4; + rc = SendReceive2(xid, tcon->ses, iov, + 1 /* num iovecs */, + &resp_buf_type, 0); cifs_stats_inc(&tcon->num_reads); + pSMBr = (READ_RSP *)iov[0].iov_base; if (rc) { cERROR(1, ("Send error in read = %d", rc)); } else { @@ -1022,33 +1023,43 @@ CIFSSMBRead(const int xid, struct cifsTconInfo *tcon, *nbytes = data_length; /*check that DataLength would not go beyond end of SMB */ - if ((data_length > CIFSMaxBufSize) + if ((data_length > CIFSMaxBufSize) || (data_length > count)) { cFYI(1,("bad length %d for count %d",data_length,count)); rc = -EIO; *nbytes = 0; } else { - pReadData = - (char *) (&pSMBr->hdr.Protocol) + + pReadData = (char *) (&pSMBr->hdr.Protocol) + le16_to_cpu(pSMBr->DataOffset); -/* if(rc = copy_to_user(buf, pReadData, data_length)) { - cERROR(1,("Faulting on read rc = %d",rc)); - rc = -EFAULT; - }*/ /* can not use copy_to_user when using page cache*/ +/* if(rc = copy_to_user(buf, pReadData, data_length)) { + cERROR(1,("Faulting on read rc = %d",rc)); + rc = -EFAULT; + }*/ /* can not use copy_to_user when using page cache*/ if(*buf) - memcpy(*buf,pReadData,data_length); + memcpy(*buf,pReadData,data_length); } } - if(*buf) - cifs_buf_release(pSMB); - else - *buf = (char *)pSMB; - /* Note: On -EAGAIN error only caller can retry on handle based calls + cifs_small_buf_release(pSMB); + if(*buf) { + if(resp_buf_type == CIFS_SMALL_BUFFER) + cifs_small_buf_release(iov[0].iov_base); + else if(resp_buf_type == CIFS_LARGE_BUFFER) + cifs_buf_release(iov[0].iov_base); + } else /* return buffer to caller to free */ /* BB FIXME how do we tell caller if it is not a large buffer */ { + *buf = iov[0].iov_base; + if(resp_buf_type == CIFS_SMALL_BUFFER) + *pbuf_type = CIFS_SMALL_BUFFER; + else if(resp_buf_type == CIFS_LARGE_BUFFER) + *pbuf_type = CIFS_LARGE_BUFFER; + } + + /* Note: On -EAGAIN error only caller can retry on handle based calls since file handle passed in no longer valid */ return rc; } + int CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, const int netfid, const unsigned int count, @@ -1163,10 +1174,10 @@ CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, { int rc = -EACCES; WRITE_REQ *pSMB = NULL; - int bytes_returned, wct; + int wct; int smb_hdr_len; + int resp_buf_type = 0; - /* BB removeme BB */ cFYI(1,("write2 at %lld %d bytes", (long long)offset, count)); if(tcon->ses->capabilities & CAP_LARGE_FILES) @@ -1209,22 +1220,34 @@ CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, pSMBW->ByteCount = cpu_to_le16(count + 5); } iov[0].iov_base = pSMB; - iov[0].iov_len = smb_hdr_len + 4; + if(wct == 14) + iov[0].iov_len = smb_hdr_len + 4; + else /* wct == 12 pad bigger by four bytes */ + iov[0].iov_len = smb_hdr_len + 8; + - rc = SendReceive2(xid, tcon->ses, iov, n_vec + 1, &bytes_returned, + rc = SendReceive2(xid, tcon->ses, iov, n_vec + 1, &resp_buf_type, long_op); cifs_stats_inc(&tcon->num_writes); if (rc) { cFYI(1, ("Send error Write2 = %d", rc)); *nbytes = 0; + } else if(resp_buf_type == 0) { + /* presumably this can not happen, but best to be safe */ + rc = -EIO; + *nbytes = 0; } else { - WRITE_RSP * pSMBr = (WRITE_RSP *)pSMB; + WRITE_RSP * pSMBr = (WRITE_RSP *)iov[0].iov_base; *nbytes = le16_to_cpu(pSMBr->CountHigh); *nbytes = (*nbytes) << 16; *nbytes += le16_to_cpu(pSMBr->Count); } cifs_small_buf_release(pSMB); + if(resp_buf_type == CIFS_SMALL_BUFFER) + cifs_small_buf_release(iov[0].iov_base); + else if(resp_buf_type == CIFS_LARGE_BUFFER) + cifs_buf_release(iov[0].iov_base); /* Note: On -EAGAIN error only caller can retry on handle based calls since file handle passed in no longer valid */ diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 651f3b6cebe..45c9d726c00 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -514,7 +514,7 @@ cifs_demultiplex_thread(struct TCP_Server_Info *server) /* else length ok */ reconnect = 0; - if(pdu_length > MAX_CIFS_HDR_SIZE - 4) { + if(pdu_length > MAX_CIFS_SMALL_BUFFER_SIZE - 4) { isLargeBuf = TRUE; memcpy(bigbuf, smallbuf, 4); smb_buffer = bigbuf; diff --git a/fs/cifs/file.c b/fs/cifs/file.c index b67be3d8c01..c249b628fd1 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -555,13 +555,13 @@ int cifs_closedir(struct inode *inode, struct file *file) } ptmp = pCFileStruct->srch_inf.ntwrk_buf_start; if (ptmp) { - /* BB removeme BB */ cFYI(1, ("freeing smb buf in srch struct in closedir")); + cFYI(1, ("closedir free smb buf in srch struct")); pCFileStruct->srch_inf.ntwrk_buf_start = NULL; cifs_buf_release(ptmp); } ptmp = pCFileStruct->search_resume_name; if (ptmp) { - /* BB removeme BB */ cFYI(1, ("freeing resume name in closedir")); + cFYI(1, ("closedir free resume name")); pCFileStruct->search_resume_name = NULL; kfree(ptmp); } @@ -871,8 +871,8 @@ static ssize_t cifs_write(struct file *file, const char *write_data, break; } /* BB FIXME We can not sign across two buffers yet */ - if((experimEnabled) && ((pTcon->ses->server->secMode & - (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) == 0)) { + if((pTcon->ses->server->secMode & + (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) == 0) { struct kvec iov[2]; unsigned int len; @@ -1424,6 +1424,7 @@ ssize_t cifs_user_read(struct file *file, char __user *read_data, rc = -EAGAIN; smb_read_data = NULL; while (rc == -EAGAIN) { + int buf_type = CIFS_NO_BUFFER; if ((open_file->invalidHandle) && (!open_file->closePend)) { rc = cifs_reopen_file(file->f_dentry->d_inode, @@ -1432,20 +1433,22 @@ ssize_t cifs_user_read(struct file *file, char __user *read_data, break; } rc = CIFSSMBRead(xid, pTcon, - open_file->netfid, - current_read_size, *poffset, - &bytes_read, &smb_read_data); + open_file->netfid, + current_read_size, *poffset, + &bytes_read, &smb_read_data, + &buf_type); pSMBr = (struct smb_com_read_rsp *)smb_read_data; if (copy_to_user(current_offset, smb_read_data + 4 /* RFC1001 hdr */ + le16_to_cpu(pSMBr->DataOffset), bytes_read)) { rc = -EFAULT; - FreeXid(xid); - return rc; - } + } if (smb_read_data) { - cifs_buf_release(smb_read_data); + if(buf_type == CIFS_SMALL_BUFFER) + cifs_small_buf_release(smb_read_data); + else if(buf_type == CIFS_LARGE_BUFFER) + cifs_buf_release(smb_read_data); smb_read_data = NULL; } } @@ -1478,6 +1481,7 @@ static ssize_t cifs_read(struct file *file, char *read_data, size_t read_size, int xid; char *current_offset; struct cifsFileInfo *open_file; + int buf_type = CIFS_NO_BUFFER; xid = GetXid(); cifs_sb = CIFS_SB(file->f_dentry->d_sb); @@ -1514,9 +1518,10 @@ static ssize_t cifs_read(struct file *file, char *read_data, size_t read_size, break; } rc = CIFSSMBRead(xid, pTcon, - open_file->netfid, - current_read_size, *poffset, - &bytes_read, ¤t_offset); + open_file->netfid, + current_read_size, *poffset, + &bytes_read, ¤t_offset, + &buf_type); } if (rc || (bytes_read == 0)) { if (total_read) { @@ -1614,6 +1619,7 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, struct smb_com_read_rsp *pSMBr; struct pagevec lru_pvec; struct cifsFileInfo *open_file; + int buf_type = CIFS_NO_BUFFER; xid = GetXid(); if (file->private_data == NULL) { @@ -1670,14 +1676,17 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, } rc = CIFSSMBRead(xid, pTcon, - open_file->netfid, - read_size, offset, - &bytes_read, &smb_read_data); - + open_file->netfid, + read_size, offset, + &bytes_read, &smb_read_data, + &buf_type); /* BB more RC checks ? */ if (rc== -EAGAIN) { if (smb_read_data) { - cifs_buf_release(smb_read_data); + if(buf_type == CIFS_SMALL_BUFFER) + cifs_small_buf_release(smb_read_data); + else if(buf_type == CIFS_LARGE_BUFFER) + cifs_buf_release(smb_read_data); smb_read_data = NULL; } } @@ -1734,7 +1743,10 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, break; } if (smb_read_data) { - cifs_buf_release(smb_read_data); + if(buf_type == CIFS_SMALL_BUFFER) + cifs_small_buf_release(smb_read_data); + else if(buf_type == CIFS_LARGE_BUFFER) + cifs_buf_release(smb_read_data); smb_read_data = NULL; } bytes_read = 0; diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index d1e99575743..f65310cc60a 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -229,11 +229,12 @@ static int decode_sfu_inode(struct inode * inode, __u64 size, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); if (rc==0) { + int buf_type = CIFS_NO_BUFFER; /* Read header */ rc = CIFSSMBRead(xid, pTcon, netfid, 24 /* length */, 0 /* offset */, - &bytes_read, &pbuf); + &bytes_read, &pbuf, &buf_type); if((rc == 0) && (bytes_read >= 8)) { if(memcmp("IntxBLK", pbuf, 8) == 0) { cFYI(1,("Block device")); @@ -267,7 +268,7 @@ static int decode_sfu_inode(struct inode * inode, __u64 size, } else { inode->i_mode |= S_IFREG; /* then it is a file */ rc = -EOPNOTSUPP; /* or some unknown SFU type */ - } + } CIFSSMBClose(xid, pTcon, netfid); } return rc; diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index ac5a72a299a..812c6bb0fe3 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -299,7 +299,7 @@ header_assemble(struct smb_hdr *buffer, char smb_command /* command */ , struct cifsSesInfo * ses; char *temp = (char *) buffer; - memset(temp,0,MAX_CIFS_HDR_SIZE); + memset(temp,0,256); /* bigger than MAX_CIFS_HDR_SIZE */ buffer->smb_buf_length = (2 * word_count) + sizeof (struct smb_hdr) - diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 0abfbf4e4a4..c96a148c39b 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -298,7 +298,7 @@ smb_send2(struct socket *ssocket, struct kvec *iov, int n_vec, int SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, - struct kvec *iov, int n_vec, int *pbytes_returned, + struct kvec *iov, int n_vec, int * pRespBufType /* ret */, const int long_op) { int rc = 0; @@ -306,6 +306,8 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, unsigned long timeout; struct mid_q_entry *midQ; struct smb_hdr *in_buf = iov[0].iov_base; + + *pRespBufType = CIFS_NO_BUFFER; /* no response buf yet */ if (ses == NULL) { cERROR(1,("Null smb session")); @@ -491,23 +493,20 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, if (midQ->resp_buf && (midQ->midState == MID_RESPONSE_RECEIVED)) { - in_buf->smb_buf_length = receive_len; - if(receive_len > 500) { - /* use multiple buffers on way out */ - } else { - memcpy((char *)in_buf + 4, - (char *)midQ->resp_buf + 4, - receive_len); - iov[0].iov_len = receive_len + 4; - iov[1].iov_len = 0; - } + iov[0].iov_base = (char *)midQ->resp_buf; + if(midQ->largeBuf) + *pRespBufType = CIFS_LARGE_BUFFER; + else + *pRespBufType = CIFS_SMALL_BUFFER; + iov[0].iov_len = receive_len + 4; + iov[1].iov_len = 0; - dump_smb(in_buf, 80); + dump_smb(midQ->resp_buf, 80); /* convert the length into a more usable form */ if((receive_len > 24) && (ses->server->secMode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED))) { - rc = cifs_verify_signature(in_buf, + rc = cifs_verify_signature(midQ->resp_buf, ses->server->mac_signing_key, midQ->sequence_number+1); if(rc) { @@ -516,18 +515,19 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, } } - *pbytes_returned = in_buf->smb_buf_length; - /* BB special case reconnect tid and uid here? */ /* BB special case Errbadpassword and pwdexpired here */ - rc = map_smb_to_linux_error(in_buf); + rc = map_smb_to_linux_error(midQ->resp_buf); /* convert ByteCount if necessary */ if (receive_len >= sizeof (struct smb_hdr) - 4 /* do not count RFC1001 header */ + - (2 * in_buf->WordCount) + 2 /* bcc */ ) - BCC(in_buf) = le16_to_cpu(BCC_LE(in_buf)); + (2 * midQ->resp_buf->WordCount) + 2 /* bcc */ ) + BCC(midQ->resp_buf) = + le16_to_cpu(BCC_LE(midQ->resp_buf)); + midQ->resp_buf = NULL; /* mark it so will not be freed + by DeleteMidQEntry */ } else { rc = -EIO; cFYI(1,("Bad MID state?")); @@ -793,7 +793,7 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, BCC(out_buf) = le16_to_cpu(BCC_LE(out_buf)); } else { rc = -EIO; - cERROR(1,("Bad MID state? ")); + cERROR(1,("Bad MID state?")); } } cifs_no_response_exit: -- cgit v1.2.3 From fdf7f2e91981938702ab2f58f44dc5685dd1bdc1 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sat, 31 Dec 2005 10:29:42 -0800 Subject: [CIFS] Fix typos in rfc1002pdu.h Pointed out by Leo Comitale Signed-off-by: Steve French --- fs/cifs/rfc1002pdu.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/cifs/rfc1002pdu.h b/fs/cifs/rfc1002pdu.h index 9222033cad8..aede606132a 100644 --- a/fs/cifs/rfc1002pdu.h +++ b/fs/cifs/rfc1002pdu.h @@ -24,11 +24,11 @@ /* NB: unlike smb/cifs packets, the RFC1002 structures are big endian */ /* RFC 1002 session packet types */ -#define RFC1002_SESSION_MESASAGE 0x00 +#define RFC1002_SESSION_MESSAGE 0x00 #define RFC1002_SESSION_REQUEST 0x81 #define RFC1002_POSITIVE_SESSION_RESPONSE 0x82 #define RFC1002_NEGATIVE_SESSION_RESPONSE 0x83 -#define RFC1002_RETARGET_SESSION_RESPONSE 0x83 +#define RFC1002_RETARGET_SESSION_RESPONSE 0x84 #define RFC1002_SESSION_KEEP_ALIVE 0x85 /* RFC 1002 flags (only one defined */ -- cgit v1.2.3 From a6230af7bdffcd3837cb9fbefc17aa6aaada486c Mon Sep 17 00:00:00 2001 From: Steve French Date: Sun, 8 Jan 2006 20:04:55 -0800 Subject: [CIFS] Minor cleanup to new cifs acl header. Signed-off-by: Steve French --- fs/cifs/cifsacl.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/cifs/cifsacl.h b/fs/cifs/cifsacl.h index 4cfcdf2e630..d152ff50ea8 100644 --- a/fs/cifs/cifsacl.h +++ b/fs/cifs/cifsacl.h @@ -34,3 +34,5 @@ struct cifs_sid { const cifs_sid sid_everyone = {1, 1, {0, 0, 0, 0, 0, 0}, {0, 0, 0, 0}}; /* group users */ const cifs_sid sid_user = {1, 2 , {0, 0, 0, 0, 0, 5}, {32, 545, 0, 0}}; + +#endif /* _CIFSACL_H */ -- cgit v1.2.3 From f3f6ec4b77f627a6427460d6f8884e1042eef134 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sun, 8 Jan 2006 20:12:58 -0800 Subject: [CIFS] Fix cifs trying to write to f_ops patch 2ea55c01e0c5dfead8699484b0bae2a375b1f61c fixed CIFS clobbering the global fops structure for some per mount setting, by duplicating and having 2 fops structs. However the write to the fops was left behind, which is a NOP in practice (due to the fact that we KNOW the fops has that field set to NULL already due to the duplication). So remove it... In addition, another instance of the same bug was forgotten in november. Signed-off-by: Arjan van de Ven Signed-off-by: Steve French --- fs/cifs/readdir.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 9bdaaecae36..288cc048d37 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -214,8 +214,7 @@ static void fill_in_inode(struct inode *tmp_inode, tmp_inode->i_fop = &cifs_file_nobrl_ops; else tmp_inode->i_fop = &cifs_file_ops; - if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) - tmp_inode->i_fop->lock = NULL; + tmp_inode->i_data.a_ops = &cifs_addr_ops; if((cifs_sb->tcon) && (cifs_sb->tcon->ses) && (cifs_sb->tcon->ses->server->maxBuf < @@ -327,12 +326,18 @@ static void unix_fill_in_inode(struct inode *tmp_inode, 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) - tmp_inode->i_fop = &cifs_file_direct_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; - if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) - tmp_inode->i_fop->lock = NULL; + tmp_inode->i_data.a_ops = &cifs_addr_ops; if((cifs_sb->tcon) && (cifs_sb->tcon->ses) && (cifs_sb->tcon->ses->server->maxBuf < -- cgit v1.2.3 From c32a0b689cb9cc160cfcd19735bbf50bb70c6ef4 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 12 Jan 2006 14:41:28 -0800 Subject: [CIFS] Allow local filesize for file that is open for write to be updated from server when mount forcedirectio. Allowing update of file size with non forcedirectio mounts should be allowed in the fiture but requires carefully writing out the last page in the local file if it is a partial page in order to avoid corruption and careful serialization Thanks to Maximiliano Curia who suggested similar changes and provided a testcase. Signed-off-by: Steve French --- fs/cifs/cifsfs.c | 17 +++++++++++++++-- fs/cifs/file.c | 10 ++++++++++ 2 files changed, 25 insertions(+), 2 deletions(-) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 582d66ca6da..136af8a08f4 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -513,6 +513,17 @@ static ssize_t cifs_file_aio_write(struct kiocb *iocb, const char __user *buf, return written; } +static loff_t cifs_llseek(struct file *file, loff_t offset, int origin) +{ + /* origin == SEEK_END => we must revalidate the cached file length */ + if (origin == 2) { + int retval = cifs_revalidate(file->f_dentry); + if (retval < 0) + return (loff_t)retval; + } + return remote_llseek(file, offset, origin); +} + static struct file_system_type cifs_fs_type = { .owner = THIS_MODULE, .name = "cifs", @@ -586,6 +597,7 @@ struct file_operations cifs_file_ops = { .flush = cifs_flush, .mmap = cifs_file_mmap, .sendfile = generic_file_sendfile, + .llseek = cifs_llseek, #ifdef CONFIG_CIFS_POSIX .ioctl = cifs_ioctl, #endif /* CONFIG_CIFS_POSIX */ @@ -609,7 +621,7 @@ struct file_operations cifs_file_direct_ops = { #ifdef CONFIG_CIFS_POSIX .ioctl = cifs_ioctl, #endif /* CONFIG_CIFS_POSIX */ - + .llseek = cifs_llseek, #ifdef CONFIG_CIFS_EXPERIMENTAL .dir_notify = cifs_dir_notify, #endif /* CONFIG_CIFS_EXPERIMENTAL */ @@ -627,6 +639,7 @@ struct file_operations cifs_file_nobrl_ops = { .flush = cifs_flush, .mmap = cifs_file_mmap, .sendfile = generic_file_sendfile, + .llseek = cifs_llseek, #ifdef CONFIG_CIFS_POSIX .ioctl = cifs_ioctl, #endif /* CONFIG_CIFS_POSIX */ @@ -649,7 +662,7 @@ struct file_operations cifs_file_direct_nobrl_ops = { #ifdef CONFIG_CIFS_POSIX .ioctl = cifs_ioctl, #endif /* CONFIG_CIFS_POSIX */ - + .llseek = cifs_llseek, #ifdef CONFIG_CIFS_EXPERIMENTAL .dir_notify = cifs_dir_notify, #endif /* CONFIG_CIFS_EXPERIMENTAL */ diff --git a/fs/cifs/file.c b/fs/cifs/file.c index c249b628fd1..670ec1e84da 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -1835,10 +1835,20 @@ int is_size_safe_to_change(struct cifsInodeInfo *cifsInode) open_file = find_writable_file(cifsInode); if(open_file) { + struct cifs_sb_info *cifs_sb; + /* there is not actually a write pending so let this handle go free and allow it to be closable if needed */ atomic_dec(&open_file->wrtPending); + + cifs_sb = CIFS_SB(cifsInode->vfs_inode.i_sb); + 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; + } + return 0; } else return 1; -- cgit v1.2.3 From 0a4b92c05ed02ad7abdd165823eaf4bbcb33ae5c Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 12 Jan 2006 15:44:21 -0800 Subject: [CIFS] Add worker function for Get ACL cifs style Signed-off-by: Steve French --- fs/cifs/CHANGES | 4 +- fs/cifs/cifs_fs_sb.h | 5 +- fs/cifs/cifsglob.h | 2 + fs/cifs/cifspdu.h | 89 +++++++++++++++++++++---- fs/cifs/cifsproto.h | 3 + fs/cifs/cifssmb.c | 179 ++++++++++++++++++++++++++++++++++++++++++++++++++- fs/cifs/connect.c | 7 ++ fs/cifs/transport.c | 1 - fs/cifs/xattr.c | 22 ++++++- 9 files changed, 292 insertions(+), 20 deletions(-) diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index 1d2137561c5..c7995c96f06 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -1,7 +1,9 @@ Version 1.40 ------------ Use fsuid (fsgid) more consistently instead of uid (gid). Improve performance -of readpages by eliminating one extra memcpy. +of readpages by eliminating one extra memcpy. Allow update of file size +from remote server even if file is open for write as long as mount is +directio. Version 1.39 ------------ diff --git a/fs/cifs/cifs_fs_sb.h b/fs/cifs/cifs_fs_sb.h index f799f6f0e72..ad58eb0c4d6 100644 --- a/fs/cifs/cifs_fs_sb.h +++ b/fs/cifs/cifs_fs_sb.h @@ -24,9 +24,10 @@ #define CIFS_MOUNT_DIRECT_IO 8 /* do not write nor read through page cache */ #define CIFS_MOUNT_NO_XATTR 0x10 /* if set - disable xattr support */ #define CIFS_MOUNT_MAP_SPECIAL_CHR 0x20 /* remap illegal chars in filenames */ -#define CIFS_MOUNT_POSIX_PATHS 0x40 /* Negotiate posix pathnames if possible. */ -#define CIFS_MOUNT_UNX_EMUL 0x80 /* Network compat with SFUnix emulation */ +#define CIFS_MOUNT_POSIX_PATHS 0x40 /* Negotiate posix pathnames if possible. */ +#define CIFS_MOUNT_UNX_EMUL 0x80 /* Network compat with SFUnix emulation */ #define CIFS_MOUNT_NO_BRL 0x100 /* No sending byte range locks to srv */ +#define CIFS_MOUNT_CIFS_ACL 0x200 /* send ACL requests to non-POSIX srv */ struct cifs_sb_info { struct cifsTconInfo *tcon; /* primary mount */ diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 862e403ff21..7bed27601ce 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -233,6 +233,8 @@ struct cifsTconInfo { atomic_t num_hardlinks; atomic_t num_symlinks; atomic_t num_locks; + atomic_t num_acl_get; + atomic_t num_acl_set; #ifdef CONFIG_CIFS_STATS2 unsigned long long time_writes; unsigned long long time_reads; diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index 3c09fb9cc56..cc2471094ca 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -528,7 +528,7 @@ typedef union smb_com_session_setup_andx { /* STRING PrimaryDomain */ /* STRING NativeOS */ /* STRING NativeLanMan */ - } __attribute__((packed)) old_req; /* pre-NTLM (LANMAN2.1) request format */ + } __attribute__((packed)) old_req; /* pre-NTLM (LANMAN2.1) req format */ struct { /* default (NTLM) response format */ struct smb_hdr hdr; /* wct = 3 */ @@ -540,7 +540,7 @@ typedef union smb_com_session_setup_andx { unsigned char NativeOS[1]; /* followed by */ /* unsigned char * NativeLanMan; */ /* unsigned char * PrimaryDomain; */ - } __attribute__((packed)) old_resp; /* pre-NTLM (LANMAN2.1) response format */ + } __attribute__((packed)) old_resp; /* pre-NTLM (LANMAN2.1) response */ } __attribute__((packed)) SESSION_SETUP_ANDX; #define CIFS_NETWORK_OPSYS "CIFS VFS Client for Linux" @@ -1007,10 +1007,49 @@ typedef struct smb_com_setattr_rsp { /* empty wct response to setattr */ -/***************************************************/ -/* NT Transact structure defintions follow */ -/* Currently only ioctl and notify are implemented */ -/***************************************************/ +/*******************************************************/ +/* NT Transact structure defintions follow */ +/* Currently only ioctl, acl (get security descriptor) */ +/* and notify are implemented */ +/*******************************************************/ +typedef struct smb_com_ntransact_req { + struct smb_hdr hdr; /* wct >= 19 */ + __u8 MaxSetupCount; + __u16 Reserved; + __le32 TotalParameterCount; + __le32 TotalDataCount; + __le32 MaxParameterCount; + __le32 MaxDataCount; + __le32 ParameterCount; + __le32 ParameterOffset; + __le32 DataCount; + __le32 DataOffset; + __u8 SetupCount; /* four setup words follow subcommand */ + /* SNIA spec incorrectly included spurious pad here */ + __le16 SubCommand; /* 2 = IOCTL/FSCTL */ + /* SetupCount words follow then */ + __le16 ByteCount; + __u8 Pad[3]; + __u8 Parms[0]; +} __attribute__((packed)) NTRANSACT_REQ; + +typedef struct smb_com_ntransact_rsp { + struct smb_hdr hdr; /* wct = 18 */ + __u8 Reserved[3]; + __le32 TotalParameterCount; + __le32 TotalDataCount; + __le32 ParameterCount; + __le32 ParameterOffset; + __le32 ParameterDisplacement; + __le32 DataCount; + __le32 DataOffset; + __le32 DataDisplacement; + __u8 SetupCount; /* 0 */ + __u16 ByteCount; + /* __u8 Pad[3]; */ + /* parms and data follow */ +} __attribute__((packed)) NTRANSACT_RSP; + typedef struct smb_com_transaction_ioctl_req { struct smb_hdr hdr; /* wct = 23 */ __u8 MaxSetupCount; @@ -1025,11 +1064,11 @@ typedef struct smb_com_transaction_ioctl_req { __le32 DataOffset; __u8 SetupCount; /* four setup words follow subcommand */ /* SNIA spec incorrectly included spurious pad here */ - __le16 SubCommand;/* 2 = IOCTL/FSCTL */ + __le16 SubCommand; /* 2 = IOCTL/FSCTL */ __le32 FunctionCode; __u16 Fid; - __u8 IsFsctl; /* 1 = File System Control, 0 = device control (IOCTL)*/ - __u8 IsRootFlag; /* 1 = apply command to root of share (must be DFS share)*/ + __u8 IsFsctl; /* 1 = File System Control 0 = device control (IOCTL) */ + __u8 IsRootFlag; /* 1 = apply command to root of share (must be DFS) */ __le16 ByteCount; __u8 Pad[3]; __u8 Data[1]; @@ -1049,9 +1088,35 @@ typedef struct smb_com_transaction_ioctl_rsp { __u8 SetupCount; /* 1 */ __le16 ReturnedDataLen; __u16 ByteCount; - __u8 Pad[3]; } __attribute__((packed)) TRANSACT_IOCTL_RSP; +#define CIFS_ACL_OWNER 1 +#define CIFS_ACL_GROUP 2 +#define CIFS_ACL_DACL 4 +#define CIFS_ACL_SACL 8 + +typedef struct smb_com_transaction_qsec_req { + struct smb_hdr hdr; /* wct = 19 */ + __u8 MaxSetupCount; + __u16 Reserved; + __le32 TotalParameterCount; + __le32 TotalDataCount; + __le32 MaxParameterCount; + __le32 MaxDataCount; + __le32 ParameterCount; + __le32 ParameterOffset; + __le32 DataCount; + __le32 DataOffset; + __u8 SetupCount; /* no setup words follow subcommand */ + /* SNIA spec incorrectly included spurious pad here */ + __le16 SubCommand; /* 6 = QUERY_SECURITY_DESC */ + __le16 ByteCount; /* bcc = 3 + 8 */ + __u8 Pad[3]; + __u16 Fid; + __u16 Reserved2; + __le32 AclFlags; +} __attribute__((packed)) QUERY_SEC_DESC_REQ; + typedef struct smb_com_transaction_change_notify_req { struct smb_hdr hdr; /* wct = 23 */ __u8 MaxSetupCount; @@ -1072,10 +1137,12 @@ typedef struct smb_com_transaction_change_notify_req { __u8 WatchTree; /* 1 = Monitor subdirectories */ __u8 Reserved2; __le16 ByteCount; -/* __u8 Pad[3];*/ +/* __u8 Pad[3];*/ /* __u8 Data[1];*/ } __attribute__((packed)) TRANSACT_CHANGE_NOTIFY_REQ; +/* BB eventually change to use generic ntransact rsp struct + and validation routine */ typedef struct smb_com_transaction_change_notify_rsp { struct smb_hdr hdr; /* wct = 18 */ __u8 Reserved[3]; diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 8ef0ce47f5b..3c03aadaff0 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -299,6 +299,9 @@ extern int CIFSSMBSetEA(const int xid, struct cifsTconInfo *tcon, const char *fileName, const char * ea_name, const void * ea_value, const __u16 ea_value_len, const struct nls_table *nls_codepage, int remap_special_chars); +extern int CIFSSMBGetCIFSACL(const int xid, struct cifsTconInfo *tcon, + __u16 fid, char *acl_inf, const int buflen, + const int acl_type /* ACCESS vs. DEFAULT */); 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 1620991cd4d..20300bc9ae7 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -1926,6 +1926,90 @@ querySymLinkRetry: return rc; } +/* Initialize NT TRANSACT SMB into small smb request buffer. + This assumes that all NT TRANSACTS that we init here have + total parm and data under about 400 bytes (to fit in small cifs + buffer size), which is the case so far, it easily fits. NB: + Setup words themselves and ByteCount + MaxSetupCount (size of returned setup area) and + MaxParameterCount (returned parms size) must be set by caller */ +static int +smb_init_ntransact(const __u16 sub_command, const int setup_count, + const int parm_len, struct cifsTconInfo *tcon, + void ** ret_buf) +{ + int rc; + __u32 temp_offset; + struct smb_com_ntransact_req * pSMB; + + rc = small_smb_init(SMB_COM_NT_TRANSACT, 19 + setup_count, tcon, + (void **)&pSMB); + if (rc) + return rc; + *ret_buf = (void *)pSMB; + pSMB->Reserved = 0; + pSMB->TotalParameterCount = cpu_to_le32(parm_len); + pSMB->TotalDataCount = 0; + pSMB->MaxDataCount = cpu_to_le32((tcon->ses->server->maxBuf - + MAX_CIFS_HDR_SIZE) & 0xFFFFFF00); + pSMB->ParameterCount = pSMB->TotalParameterCount; + pSMB->DataCount = pSMB->TotalDataCount; + temp_offset = offsetof(struct smb_com_ntransact_req, Parms) + + (setup_count * 2) - 4 /* for rfc1001 length itself */; + pSMB->ParameterOffset = cpu_to_le32(temp_offset); + pSMB->DataOffset = cpu_to_le32(temp_offset + parm_len); + pSMB->SetupCount = setup_count; /* no need to le convert byte fields */ + pSMB->SubCommand = cpu_to_le16(sub_command); + return 0; +} + +static int +validate_ntransact(char * buf, char ** ppparm, char ** ppdata, + int * pdatalen, int * pparmlen) +{ + char * end_of_smb; + __u32 data_count, data_offset, parm_count, parm_offset; + struct smb_com_ntransact_rsp * pSMBr; + + if(buf == NULL) + return -EINVAL; + + pSMBr = (struct smb_com_ntransact_rsp *)buf; + + /* ByteCount was converted from little endian in SendReceive */ + end_of_smb = 2 /* sizeof byte count */ + pSMBr->ByteCount + + (char *)&pSMBr->ByteCount; + + + data_offset = le32_to_cpu(pSMBr->DataOffset); + data_count = le32_to_cpu(pSMBr->DataCount); + parm_offset = le32_to_cpu(pSMBr->ParameterOffset); + parm_count = le32_to_cpu(pSMBr->ParameterCount); + + *ppparm = (char *)&pSMBr->hdr.Protocol + parm_offset; + *ppdata = (char *)&pSMBr->hdr.Protocol + data_offset; + + /* should we also check that parm and data areas do not overlap? */ + if(*ppparm > end_of_smb) { + cFYI(1,("parms start after end of smb")); + return -EINVAL; + } else if(parm_count + *ppparm > end_of_smb) { + cFYI(1,("parm end after end of smb")); + return -EINVAL; + } else if(*ppdata > end_of_smb) { + cFYI(1,("data starts after end of smb")); + return -EINVAL; + } else if(data_count + *ppdata > end_of_smb) { + cFYI(1,("data %p + count %d (%p) ends after end of smb %p start %p", + *ppdata, data_count, (data_count + *ppdata), end_of_smb, pSMBr)); /* BB FIXME */ + return -EINVAL; + } else if(parm_count + data_count > pSMBr->ByteCount) { + cFYI(1,("parm count and data count larger than SMB")); + return -EINVAL; + } + return 0; +} + int CIFSSMBQueryReparseLinkInfo(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, @@ -1948,7 +2032,8 @@ CIFSSMBQueryReparseLinkInfo(const int xid, struct cifsTconInfo *tcon, pSMB->TotalDataCount = 0; pSMB->MaxParameterCount = cpu_to_le32(2); /* BB find exact data count max from sess structure BB */ - pSMB->MaxDataCount = cpu_to_le32(4000); + pSMB->MaxDataCount = cpu_to_le32((tcon->ses->server->maxBuf - + MAX_CIFS_HDR_SIZE) & 0xFFFFFF00); pSMB->MaxSetupCount = 4; pSMB->Reserved = 0; pSMB->ParameterOffset = 0; @@ -1975,7 +2060,9 @@ CIFSSMBQueryReparseLinkInfo(const int xid, struct cifsTconInfo *tcon, rc = -EIO; /* bad smb */ else { if(data_count && (data_count < 2048)) { - char * end_of_smb = pSMBr->ByteCount + (char *)&pSMBr->ByteCount; + char * end_of_smb = 2 /* sizeof byte count */ + + pSMBr->ByteCount + + (char *)&pSMBr->ByteCount; struct reparse_data * reparse_buf = (struct reparse_data *) ((char *)&pSMBr->hdr.Protocol + data_offset); @@ -2219,6 +2306,7 @@ queryAclRetry: rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); + cifs_stats_inc(&tcon->num_acl_get); if (rc) { cFYI(1, ("Send error in Query POSIX ACL = %d", rc)); } else { @@ -2406,6 +2494,87 @@ GetExtAttrOut: #endif /* CONFIG_POSIX */ +/* Convert CIFS ACL to POSIX form */ +static int parse_sec_desc(struct sec_desc * psec_desc, int acl_len) +{ + CHECK ON OTHER COMPUTER TO SEE IF FORMAT ENCODED IN CIFSPDU.H ALREADY + return 0; +} + +/* Get Security Descriptor (by handle) from remote server for a file or dir */ +int +CIFSSMBGetCIFSACL(const int xid, struct cifsTconInfo *tcon, __u16 fid, + /* BB fix up return info */ char *acl_inf, const int buflen, + const int acl_type /* ACCESS/DEFAULT not sure implication */) +{ + int rc = 0; + int buf_type = 0; + QUERY_SEC_DESC_REQ * pSMB; + struct kvec iov[1]; + + cFYI(1, ("GetCifsACL")); + + rc = smb_init_ntransact(NT_TRANSACT_QUERY_SECURITY_DESC, 0, + 8 /* parm len */, tcon, (void **) &pSMB); + if (rc) + return rc; + + pSMB->MaxParameterCount = cpu_to_le32(4); + /* BB TEST with big acls that might need to be e.g. larger than 16K */ + pSMB->MaxSetupCount = 0; + pSMB->Fid = fid; /* file handle always le */ + pSMB->AclFlags = cpu_to_le32(CIFS_ACL_OWNER | CIFS_ACL_GROUP | + CIFS_ACL_DACL); + pSMB->ByteCount = cpu_to_le16(11); /* 3 bytes pad + 8 bytes parm */ + pSMB->hdr.smb_buf_length += 11; + iov[0].iov_base = (char *)pSMB; + iov[0].iov_len = pSMB->hdr.smb_buf_length + 4; + + rc = SendReceive2(xid, tcon->ses, iov, 1 /* num iovec */, &buf_type, 0); + cifs_stats_inc(&tcon->num_acl_get); + if (rc) { + cFYI(1, ("Send error in QuerySecDesc = %d", rc)); + } else { /* decode response */ + struct sec_desc * psec_desc; + __le32 * parm; + int parm_len; + int data_len; + int acl_len; + struct smb_com_ntransact_rsp * pSMBr; + +/* validate_nttransact */ + rc = validate_ntransact(iov[0].iov_base, (char **)&parm, + (char **)&psec_desc, + &parm_len, &data_len); + + if(rc) + goto qsec_out; + pSMBr = (struct smb_com_ntransact_rsp *)iov[0].iov_base; + + cERROR(1,("smb %p parm %p data %p",pSMBr,parm,psec_desc)); /* BB removeme BB */ + + if (le32_to_cpu(pSMBr->ParameterCount) != 4) { + rc = -EIO; /* bad smb */ + goto qsec_out; + } + +/* BB check that data area is minimum length and as big as acl_len */ + + acl_len = le32_to_cpu(*(__le32 *)parm); + /* BB check if(acl_len > bufsize) */ + + parse_sec_desc(psec_desc, acl_len); + } +qsec_out: + if(buf_type == CIFS_SMALL_BUFFER) + cifs_small_buf_release(iov[0].iov_base); + else if(buf_type == CIFS_LARGE_BUFFER) + cifs_buf_release(iov[0].iov_base); + cifs_small_buf_release(pSMB); + return rc; +} + + /* Legacy Query Path Information call for lookup to old servers such as Win9x/WinME */ int SMBQueryInformation(const int xid, struct cifsTconInfo *tcon, @@ -4304,7 +4473,7 @@ int CIFSSMBNotify(const int xid, struct cifsTconInfo *tcon, { int rc = 0; struct smb_com_transaction_change_notify_req * pSMB = NULL; - struct smb_com_transaction_change_notify_rsp * pSMBr = NULL; + struct smb_com_ntransaction_change_notify_rsp * pSMBr = NULL; struct dir_notify_req *dnotify_req; int bytes_returned; @@ -4319,6 +4488,10 @@ int CIFSSMBNotify(const int xid, struct cifsTconInfo *tcon, pSMB->MaxParameterCount = cpu_to_le32(2); /* BB find exact data count max from sess structure BB */ pSMB->MaxDataCount = 0; /* same in little endian or be */ +/* BB VERIFY verify which is correct for above BB */ + pSMB->MaxDataCount = cpu_to_le32((tcon->ses->server->maxBuf - + MAX_CIFS_HDR_SIZE) & 0xFFFFFF00); + pSMB->MaxSetupCount = 4; pSMB->Reserved = 0; pSMB->ParameterOffset = 0; diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 45c9d726c00..1817d5313a8 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -76,6 +76,7 @@ struct smb_vol { unsigned setuids:1; unsigned noperm:1; unsigned no_psx_acl:1; /* set if posix acl support should be disabled */ + unsigned cifs_acl:1; unsigned no_xattr:1; /* set if xattr (EA) support should be disabled*/ unsigned server_ino:1; /* use inode numbers from server ie UniqueId */ unsigned direct_io:1; @@ -1159,6 +1160,10 @@ cifs_parse_mount_options(char *options, const char *devname,struct smb_vol *vol) vol->server_ino = 1; } else if (strnicmp(data, "noserverino",9) == 0) { vol->server_ino = 0; + } else if (strnicmp(data, "cifsacl",7) == 0) { + vol->cifs_acl = 1; + } else if (strnicmp(data, "nocifsacl", 9) == 0) { + vol->cifs_acl = 0; } else if (strnicmp(data, "acl",3) == 0) { vol->no_psx_acl = 0; } else if (strnicmp(data, "noacl",5) == 0) { @@ -1806,6 +1811,8 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_UNX_EMUL; if(volume_info.nobrl) cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NO_BRL; + if(volume_info.cifs_acl) + cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_CIFS_ACL; if(volume_info.direct_io) { cFYI(1,("mounting share using direct i/o")); diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index c96a148c39b..7b98792150e 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -489,7 +489,6 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, receive_len, xid)); rc = -EIO; } else { /* rcvd frame is ok */ - if (midQ->resp_buf && (midQ->midState == MID_RESPONSE_RECEIVED)) { diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c index f375f87c7db..777e3363c2a 100644 --- a/fs/cifs/xattr.c +++ b/fs/cifs/xattr.c @@ -254,7 +254,8 @@ ssize_t cifs_getxattr(struct dentry * direntry, const char * ea_name, rc = CIFSSMBQueryEA(xid,pTcon,full_path,ea_name,ea_value, buf_size, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); - } else if(strncmp(ea_name,POSIX_ACL_XATTR_ACCESS,strlen(POSIX_ACL_XATTR_ACCESS)) == 0) { + } else if(strncmp(ea_name,POSIX_ACL_XATTR_ACCESS, + strlen(POSIX_ACL_XATTR_ACCESS)) == 0) { #ifdef CONFIG_CIFS_POSIX if(sb->s_flags & MS_POSIXACL) rc = CIFSSMBGetPosixACL(xid, pTcon, full_path, @@ -262,10 +263,27 @@ ssize_t cifs_getxattr(struct dentry * direntry, const char * ea_name, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); +/* else if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) { + __u16 fid; + int oplock = FALSE; + rc = CIFSSMBOpen(xid, pTcon, full_path, + FILE_OPEN, GENERIC_READ, 0, &fid, + &oplock, NULL, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + if(rc == 0) { + rc = CIFSSMBGetCIFSACL(xid, pTcon, fid, + ea_value, buf_size, + ACL_TYPE_ACCESS); + CIFSSMBClose(xid, pTcon, fid) + } + } */ /* BB enable after fixing up return data */ + #else cFYI(1,("query POSIX ACL not supported yet")); #endif /* CONFIG_CIFS_POSIX */ - } else if(strncmp(ea_name,POSIX_ACL_XATTR_DEFAULT,strlen(POSIX_ACL_XATTR_DEFAULT)) == 0) { + } else if(strncmp(ea_name,POSIX_ACL_XATTR_DEFAULT, + strlen(POSIX_ACL_XATTR_DEFAULT)) == 0) { #ifdef CONFIG_CIFS_POSIX if(sb->s_flags & MS_POSIXACL) rc = CIFSSMBGetPosixACL(xid, pTcon, full_path, -- cgit v1.2.3 From 84153973a29dfb3f3d9fe2fe75c2cd613a3cdf27 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 12 Jan 2006 20:57:56 -0800 Subject: [CIFS] Fix typo Signed-off-by: Steve French --- fs/cifs/cifssmb.c | 1 - 1 file changed, 1 deletion(-) diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 20300bc9ae7..ca0f573e979 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -2497,7 +2497,6 @@ GetExtAttrOut: /* Convert CIFS ACL to POSIX form */ static int parse_sec_desc(struct sec_desc * psec_desc, int acl_len) { - CHECK ON OTHER COMPUTER TO SEE IF FORMAT ENCODED IN CIFSPDU.H ALREADY return 0; } -- cgit v1.2.3 From eeac8047fcf4c659eb15f2e27a0ef4aeba64157f Mon Sep 17 00:00:00 2001 From: Steve French Date: Fri, 13 Jan 2006 21:34:58 -0800 Subject: [CIFS] Fix CIFS to recognize share mode security Fix Samba bugzilla bug 3301 In share mode encrypted password must be sent on tree connection (in our case only the NTLM password is sent, not the older LANMAN one). Signed-off-by: Steve French --- fs/cifs/CHANGES | 3 ++- fs/cifs/cifsacl.h | 6 +++--- fs/cifs/cifssmb.c | 20 ++++++++++++++------ fs/cifs/connect.c | 27 ++++++++++++++++++++++----- 4 files changed, 41 insertions(+), 15 deletions(-) diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index c7995c96f06..d335015473a 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -3,7 +3,8 @@ Version 1.40 Use fsuid (fsgid) more consistently instead of uid (gid). Improve performance of readpages by eliminating one extra memcpy. Allow update of file size from remote server even if file is open for write as long as mount is -directio. +directio. Recognize share mode security and send NTLM encrypted password +on tree connect if share mode negotiated. Version 1.39 ------------ diff --git a/fs/cifs/cifsacl.h b/fs/cifs/cifsacl.h index d152ff50ea8..d0776ac2b80 100644 --- a/fs/cifs/cifsacl.h +++ b/fs/cifs/cifsacl.h @@ -26,13 +26,13 @@ struct cifs_sid { __u8 revision; /* revision level */ __u8 num_subauths; __u8 authority[6]; - __u8 sub_auth[4]; + __u32 sub_auth[4]; /* next sub_auth if any ... */ } __attribute__((packed)); /* everyone */ -const cifs_sid sid_everyone = {1, 1, {0, 0, 0, 0, 0, 0}, {0, 0, 0, 0}}; +extern const struct cifs_sid sid_everyone; /* group users */ -const cifs_sid sid_user = {1, 2 , {0, 0, 0, 0, 0, 5}, {32, 545, 0, 0}}; +extern const struct cifs_sid sid_user; #endif /* _CIFSACL_H */ diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index ca0f573e979..7a4e936d726 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -37,6 +37,7 @@ #include "cifsproto.h" #include "cifs_unicode.h" #include "cifs_debug.h" +#include "cifsacl.h" #ifdef CONFIG_CIFS_POSIX static struct { @@ -372,8 +373,10 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) rc = SendReceive(xid, ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); if (rc == 0) { - server->secMode = pSMBr->SecurityMode; - server->secType = NTLM; /* BB override default for + server->secMode = pSMBr->SecurityMode; + if((server->secMode & SECMODE_USER) == 0) + cFYI(1,("share mode security")); + server->secType = NTLM; /* BB override default for NTLMv2 or kerberos v5 */ /* one byte - no need to convert this or EncryptionKeyLen from little endian */ @@ -383,7 +386,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) 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(0, ("Max buf = %d", ses->server->maxBuf)); GETU32(ses->server->sessid) = le32_to_cpu(pSMBr->SessionKey); server->capabilities = le32_to_cpu(pSMBr->Capabilities); server->timeZone = le16_to_cpu(pSMBr->ServerTimeZone); @@ -411,8 +414,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) (server->server_GUID, pSMBr->u.extended_response. GUID, 16) != 0) { - cFYI(1, - ("UID of server does not match previous connection to same ip address")); + cFYI(1, ("server UID changed")); memcpy(server-> server_GUID, pSMBr->u. @@ -2494,8 +2496,14 @@ GetExtAttrOut: #endif /* CONFIG_POSIX */ + +/* security id for everyone */ +const struct cifs_sid sid_everyone = {1, 1, {0, 0, 0, 0, 0, 0}, {0, 0, 0, 0}}; +/* group users */ +const struct cifs_sid sid_user = {1, 2 , {0, 0, 0, 0, 0, 5}, {32, 545, 0, 0}}; + /* Convert CIFS ACL to POSIX form */ -static int parse_sec_desc(struct sec_desc * psec_desc, int acl_len) +static int parse_sec_desc(struct cifs_sid * psec_desc, int acl_len) { return 0; } diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 1817d5313a8..88f60aa5205 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -1795,7 +1795,8 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, cifs_sb->mnt_gid = volume_info.linux_gid; cifs_sb->mnt_file_mode = volume_info.file_mode; cifs_sb->mnt_dir_mode = volume_info.dir_mode; - cFYI(1,("file mode: 0x%x dir mode: 0x%x",cifs_sb->mnt_file_mode,cifs_sb->mnt_dir_mode)); + cFYI(1,("file mode: 0x%x dir mode: 0x%x", + cifs_sb->mnt_file_mode,cifs_sb->mnt_dir_mode)); if(volume_info.noperm) cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NO_PERM; @@ -1972,7 +1973,7 @@ CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, __u32 capabilities; __u16 count; - cFYI(1, ("In sesssetup ")); + cFYI(1, ("In sesssetup")); if(ses == NULL) return -EINVAL; user = ses->userName; @@ -3248,9 +3249,26 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, pSMB->AndXCommand = 0xFF; pSMB->Flags = cpu_to_le16(TCON_EXTENDED_SECINFO); - pSMB->PasswordLength = cpu_to_le16(1); /* minimum */ bcc_ptr = &pSMB->Password[0]; - bcc_ptr++; /* skip password */ + if((ses->server->secMode) & SECMODE_USER) { + pSMB->PasswordLength = cpu_to_le16(1); /* minimum */ + bcc_ptr++; /* skip password */ + } else { + pSMB->PasswordLength = cpu_to_le16(CIFS_SESSION_KEY_SIZE); + /* BB FIXME add code to fail this if NTLMv2 or Kerberos + specified as required (when that support is added to + the vfs in the future) as only NTLM or the much + weaker LANMAN (which we do not send) is accepted + by Samba (not sure whether other servers allow + NTLMv2 password here) */ + SMBNTencrypt(ses->password, + ses->server->cryptKey, + bcc_ptr); + + bcc_ptr += CIFS_SESSION_KEY_SIZE; + *bcc_ptr = 0; + bcc_ptr++; /* align */ + } if(ses->server->secMode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) smb_buffer->Flags2 |= SMBFLG2_SECURITY_SIGNATURE; @@ -3268,7 +3286,6 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, bcc_ptr += 2 * length; /* convert num of 16 bit words to bytes */ bcc_ptr += 2; /* skip trailing null */ } else { /* ASCII */ - strcpy(bcc_ptr, tree); bcc_ptr += strlen(tree) + 1; } -- cgit v1.2.3 From 5f7f5b0c99c9c913b649d065a9773e6ed4ed8168 Mon Sep 17 00:00:00 2001 From: Prarit Bhargava Date: Mon, 16 Jan 2006 15:31:52 -0500 Subject: [IA64-SGI] Older PROM WAR for device flush code Work-around to temporarily support older PROMs with new flush device code. This code allows systems running older PROMs to continue to run on the new kernel base until a new official PROM is released. Signed-off-by: Prarit Bhargava Acked-by: Jes Sorensen Signed-off-by: Tony Luck --- arch/ia64/sn/include/xtalk/hubdev.h | 9 +++++++ arch/ia64/sn/kernel/io_init.c | 54 ++++++++++++++++++++++++++++++++++--- 2 files changed, 60 insertions(+), 3 deletions(-) diff --git a/arch/ia64/sn/include/xtalk/hubdev.h b/arch/ia64/sn/include/xtalk/hubdev.h index 7c88e9a5851..8182583c762 100644 --- a/arch/ia64/sn/include/xtalk/hubdev.h +++ b/arch/ia64/sn/include/xtalk/hubdev.h @@ -51,6 +51,15 @@ struct sn_flush_device_kernel { struct sn_flush_device_common *common; }; +/* 01/16/06 This struct is the old PROM/kernel struct and needs to be included + * for older official PROMs to function on the new kernel base. This struct + * will be removed when the next official PROM release occurs. */ + +struct sn_flush_device_war { + struct sn_flush_device_common common; + u32 filler; /* older PROMs expect the default size of a spinlock_t */ +}; + /* * **widget_p - Used as an array[wid_num][device] of sn_flush_device_kernel. */ diff --git a/arch/ia64/sn/kernel/io_init.c b/arch/ia64/sn/kernel/io_init.c index 233d55115d3..00700f7e683 100644 --- a/arch/ia64/sn/kernel/io_init.c +++ b/arch/ia64/sn/kernel/io_init.c @@ -165,8 +165,45 @@ sn_pcidev_info_get(struct pci_dev *dev) return NULL; } +/* Older PROM flush WAR + * + * 01/16/06 -- This war will be in place until a new official PROM is released. + * Additionally note that the struct sn_flush_device_war also has to be + * removed from arch/ia64/sn/include/xtalk/hubdev.h + */ +static u8 war_implemented = 0; + +static void sn_device_fixup_war(u64 nasid, u64 widget, int device, + struct sn_flush_device_common *common) +{ + struct sn_flush_device_war *war_list; + struct sn_flush_device_war *dev_entry; + struct ia64_sal_retval isrv = {0,0,0,0}; + + if (!war_implemented) { + printk(KERN_WARNING "PROM version < 4.50 -- implementing old " + "PROM flush WAR\n"); + war_implemented = 1; + } + + war_list = kzalloc(DEV_PER_WIDGET * sizeof(*war_list), GFP_KERNEL); + if (!war_list) + BUG(); + + SAL_CALL_NOLOCK(isrv, SN_SAL_IOIF_GET_WIDGET_DMAFLUSH_LIST, + nasid, widget, __pa(war_list), 0, 0, 0 ,0); + if (isrv.status) + panic("sn_device_fixup_war failed: %s\n", + ia64_sal_strerror(isrv.status)); + + dev_entry = war_list + device; + memcpy(common,dev_entry, sizeof(*common)); + + kfree(war_list); +} + /* - * sn_fixup_ionodes() - This routine initializes the HUB data strcuture for + * sn_fixup_ionodes() - This routine initializes the HUB data strcuture for * each node in the system. */ static void sn_fixup_ionodes(void) @@ -246,8 +283,19 @@ static void sn_fixup_ionodes(void) widget, device, (u64)(dev_entry->common)); - if (status) - BUG(); + if (status) { + if (sn_sal_rev() < 0x0450) { + /* shortlived WAR for older + * PROM images + */ + sn_device_fixup_war(nasid, + widget, + device, + dev_entry->common); + } + else + BUG(); + } spin_lock_init(&dev_entry->sfdl_flush_lock); } -- cgit v1.2.3 From ac354a899b91239aac4d5893fc4288bc400e82b4 Mon Sep 17 00:00:00 2001 From: Mike Habeck Date: Wed, 11 Jan 2006 16:40:24 -0600 Subject: [IA64-SGI] pass segment# on SN_SAL_IOIF_SLOT_{DIS,EN}ABLE calls Bugfix... the altix SN_SAL_IOIF_SLOT_ENABLE & SN_SAL_IOIF_SLOT_DISABLE SAL calls need to pass the segment# down Signed-off-by: Mike Habeck Signed-off-by: Tony Luck --- arch/ia64/sn/pci/pcibr/pcibr_provider.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/arch/ia64/sn/pci/pcibr/pcibr_provider.c b/arch/ia64/sn/pci/pcibr/pcibr_provider.c index 77a1262751d..2fac27049bf 100644 --- a/arch/ia64/sn/pci/pcibr/pcibr_provider.c +++ b/arch/ia64/sn/pci/pcibr/pcibr_provider.c @@ -24,13 +24,15 @@ sal_pcibr_slot_enable(struct pcibus_info *soft, int device, void *resp) { struct ia64_sal_retval ret_stuff; u64 busnum; + u64 segment; ret_stuff.status = 0; ret_stuff.v0 = 0; + segment = soft->pbi_buscommon.bs_persist_segment; busnum = soft->pbi_buscommon.bs_persist_busnum; - SAL_CALL_NOLOCK(ret_stuff, (u64) SN_SAL_IOIF_SLOT_ENABLE, (u64) busnum, - (u64) device, (u64) resp, 0, 0, 0, 0); + SAL_CALL_NOLOCK(ret_stuff, (u64) SN_SAL_IOIF_SLOT_ENABLE, segment, + busnum, (u64) device, (u64) resp, 0, 0, 0); return (int)ret_stuff.v0; } @@ -41,14 +43,16 @@ sal_pcibr_slot_disable(struct pcibus_info *soft, int device, int action, { struct ia64_sal_retval ret_stuff; u64 busnum; + u64 segment; ret_stuff.status = 0; ret_stuff.v0 = 0; + segment = soft->pbi_buscommon.bs_persist_segment; busnum = soft->pbi_buscommon.bs_persist_busnum; SAL_CALL_NOLOCK(ret_stuff, (u64) SN_SAL_IOIF_SLOT_DISABLE, - (u64) busnum, (u64) device, (u64) action, - (u64) resp, 0, 0, 0); + segment, busnum, (u64) device, (u64) action, + (u64) resp, 0, 0); return (int)ret_stuff.v0; } -- cgit v1.2.3 From 7a48f923b8b27bfaa5f7b2a449a6fe268724ddd5 Mon Sep 17 00:00:00 2001 From: Sridhar Samudrala Date: Tue, 17 Jan 2006 11:51:28 -0800 Subject: [SCTP]: Fix potential race condition between sctp_close() and sctp_rcv(). Do not release the reference to association/endpoint if an incoming skb is added to backlog. Instead release it after the chunk is processed in sctp_backlog_rcv(). Signed-off-by: Sridhar Samudrala Signed-off-by: Vlad Yasevich --- net/sctp/input.c | 29 ++++++++++++++++++++--------- net/sctp/inqueue.c | 4 +++- 2 files changed, 23 insertions(+), 10 deletions(-) diff --git a/net/sctp/input.c b/net/sctp/input.c index 4aa6fc60357..c463e4049c5 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -262,15 +262,12 @@ int sctp_rcv(struct sk_buff *skb) else sctp_backlog_rcv(sk, skb); - /* Release the sock and any reference counts we took in the - * lookup calls. + /* Release the sock and the sock ref we took in the lookup calls. + * The asoc/ep ref will be released in sctp_backlog_rcv. */ sctp_bh_unlock_sock(sk); - if (asoc) - sctp_association_put(asoc); - else - sctp_endpoint_put(ep); sock_put(sk); + return ret; discard_it: @@ -296,9 +293,23 @@ discard_release: int sctp_backlog_rcv(struct sock *sk, struct sk_buff *skb) { struct sctp_chunk *chunk = SCTP_INPUT_CB(skb)->chunk; - struct sctp_inq *inqueue = &chunk->rcvr->inqueue; - - sctp_inq_push(inqueue, chunk); + struct sctp_inq *inqueue = NULL; + struct sctp_ep_common *rcvr = NULL; + + rcvr = chunk->rcvr; + if (rcvr->dead) { + sctp_chunk_free(chunk); + } else { + inqueue = &chunk->rcvr->inqueue; + sctp_inq_push(inqueue, chunk); + } + + /* Release the asoc/ep ref we took in the lookup calls in sctp_rcv. */ + if (SCTP_EP_TYPE_ASSOCIATION == rcvr->type) + sctp_association_put(sctp_assoc(rcvr)); + else + sctp_endpoint_put(sctp_ep(rcvr)); + return 0; } diff --git a/net/sctp/inqueue.c b/net/sctp/inqueue.c index 2d33922c044..297b8951463 100644 --- a/net/sctp/inqueue.c +++ b/net/sctp/inqueue.c @@ -73,8 +73,10 @@ void sctp_inq_free(struct sctp_inq *queue) /* If there is a packet which is currently being worked on, * free it as well. */ - if (queue->in_progress) + if (queue->in_progress) { sctp_chunk_free(queue->in_progress); + queue->in_progress = NULL; + } if (queue->malloced) { /* Dump the master memory segment. */ -- cgit v1.2.3 From 9834a2bb4970547540222fcba04e0a37d04cb0a0 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Tue, 17 Jan 2006 11:52:12 -0800 Subject: [SCTP]: Fix sctp_cookie alignment in the packet. On 64 bit architectures, sctp_cookie sent as part of INIT-ACK is not aligned on a 64 bit boundry and thus causes unaligned access exceptions. The layout of the cookie prameter is this: |<----- Parameter Header --------------------|<--- Cookie DATA -------- ----------------------------------------------------------------------- | param type (16 bits) | param len (16 bits) | sig [32 bytes] | cookie.. ----------------------------------------------------------------------- The cookie data portion contains 64 bit values on 64 bit architechtures (timeval) that fall on a 32 bit alignment boundry when used as part of the on-wire format, but align correctly when used in internal structures. This patch explicitely pads the on-wire format so that it is properly aligned. Signed-off-by: Vlad Yasevich Signed-off-by: Sridhar Samudrala --- include/net/sctp/structs.h | 3 ++- net/sctp/sm_make_chunk.c | 14 ++++++++++++-- 2 files changed, 14 insertions(+), 3 deletions(-) diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index f5c22d77fea..72aeae4a006 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -405,8 +405,9 @@ struct sctp_cookie { /* The format of our cookie that we send to our peer. */ struct sctp_signed_cookie { __u8 signature[SCTP_SECRET_SIZE]; + __u32 __pad; /* force sctp_cookie alignment to 64 bits */ struct sctp_cookie c; -}; +} __attribute__((packed)); /* This is another convenience type to allocate memory for address * params for the maximum size and pass such structures around diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 556c495c692..4fe1d6c863b 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -1275,7 +1275,12 @@ static sctp_cookie_param_t *sctp_pack_cookie(const struct sctp_endpoint *ep, unsigned int keylen; char *key; - headersize = sizeof(sctp_paramhdr_t) + SCTP_SECRET_SIZE; + /* Header size is static data prior to the actual cookie, including + * any padding. + */ + headersize = sizeof(sctp_paramhdr_t) + + (sizeof(struct sctp_signed_cookie) - + sizeof(struct sctp_cookie)); bodysize = sizeof(struct sctp_cookie) + ntohs(init_chunk->chunk_hdr->length) + addrs_len; @@ -1362,7 +1367,12 @@ struct sctp_association *sctp_unpack_cookie( struct sk_buff *skb = chunk->skb; struct timeval tv; - headersize = sizeof(sctp_chunkhdr_t) + SCTP_SECRET_SIZE; + /* Header size is static data prior to the actual cookie, including + * any padding. + */ + headersize = sizeof(sctp_chunkhdr_t) + + (sizeof(struct sctp_signed_cookie) - + sizeof(struct sctp_cookie)); bodysize = ntohs(chunk->chunk_hdr->length) - headersize; fixed_size = headersize + sizeof(struct sctp_cookie); -- cgit v1.2.3 From 49392e5ecf608da6770fd8723b534a0fc851edc4 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Tue, 17 Jan 2006 11:53:06 -0800 Subject: [SCTP]: sctp doesn't show all associations/endpoints in /proc When creating a very large number of associations (and endpoints), /proc/assocs and /proc/eps will not show all of them. As a result netstat will not show all of the either. This is particularly evident when creating 1000+ associations (or endpoints). As an example with 1500 tcp style associations over loopback, netstat showed 1420 on my system instead of 3000. The reason for this is that the seq_operations start method is invoked multiple times bacause of the amount of data that is provided. The start method always increments the position parameter and since we use the position as the hash bucket id, we end up skipping hash buckets. This patch corrects this situation and get's rid of the silly hash-1 decrement. Signed-off-by: Vlad Yasevich Signed-off-by: Sridhar Samudrala --- net/sctp/proc.c | 28 ++++++++++------------------ 1 file changed, 10 insertions(+), 18 deletions(-) diff --git a/net/sctp/proc.c b/net/sctp/proc.c index 6e4dc28874d..1b5e5b119f7 100644 --- a/net/sctp/proc.c +++ b/net/sctp/proc.c @@ -176,7 +176,7 @@ static void sctp_seq_dump_remote_addrs(struct seq_file *seq, struct sctp_associa static void * sctp_eps_seq_start(struct seq_file *seq, loff_t *pos) { - if (*pos > sctp_ep_hashsize) + if (*pos >= sctp_ep_hashsize) return NULL; if (*pos < 0) @@ -185,8 +185,6 @@ static void * sctp_eps_seq_start(struct seq_file *seq, loff_t *pos) if (*pos == 0) seq_printf(seq, " ENDPT SOCK STY SST HBKT LPORT UID INODE LADDRS\n"); - ++*pos; - return (void *)pos; } @@ -198,11 +196,9 @@ static void sctp_eps_seq_stop(struct seq_file *seq, void *v) static void * sctp_eps_seq_next(struct seq_file *seq, void *v, loff_t *pos) { - if (*pos > sctp_ep_hashsize) + if (++*pos >= sctp_ep_hashsize) return NULL; - ++*pos; - return pos; } @@ -216,17 +212,17 @@ static int sctp_eps_seq_show(struct seq_file *seq, void *v) struct sock *sk; int hash = *(int *)v; - if (hash > sctp_ep_hashsize) + if (hash >= sctp_ep_hashsize) return -ENOMEM; - head = &sctp_ep_hashtable[hash-1]; + head = &sctp_ep_hashtable[hash]; sctp_local_bh_disable(); read_lock(&head->lock); for (epb = head->chain; epb; epb = epb->next) { ep = sctp_ep(epb); sk = epb->sk; seq_printf(seq, "%8p %8p %-3d %-3d %-4d %-5d %5d %5lu ", ep, sk, - sctp_sk(sk)->type, sk->sk_state, hash-1, + sctp_sk(sk)->type, sk->sk_state, hash, epb->bind_addr.port, sock_i_uid(sk), sock_i_ino(sk)); @@ -283,7 +279,7 @@ void sctp_eps_proc_exit(void) static void * sctp_assocs_seq_start(struct seq_file *seq, loff_t *pos) { - if (*pos > sctp_assoc_hashsize) + if (*pos >= sctp_assoc_hashsize) return NULL; if (*pos < 0) @@ -293,8 +289,6 @@ static void * sctp_assocs_seq_start(struct seq_file *seq, loff_t *pos) seq_printf(seq, " ASSOC SOCK STY SST ST HBKT ASSOC-ID TX_QUEUE RX_QUEUE UID INODE LPORT " "RPORT LADDRS <-> RADDRS\n"); - ++*pos; - return (void *)pos; } @@ -306,11 +300,9 @@ static void sctp_assocs_seq_stop(struct seq_file *seq, void *v) static void * sctp_assocs_seq_next(struct seq_file *seq, void *v, loff_t *pos) { - if (*pos > sctp_assoc_hashsize) + if (++*pos >= sctp_assoc_hashsize) return NULL; - ++*pos; - return pos; } @@ -323,10 +315,10 @@ static int sctp_assocs_seq_show(struct seq_file *seq, void *v) struct sock *sk; int hash = *(int *)v; - if (hash > sctp_assoc_hashsize) + if (hash >= sctp_assoc_hashsize) return -ENOMEM; - head = &sctp_assoc_hashtable[hash-1]; + head = &sctp_assoc_hashtable[hash]; sctp_local_bh_disable(); read_lock(&head->lock); for (epb = head->chain; epb; epb = epb->next) { @@ -335,7 +327,7 @@ static int sctp_assocs_seq_show(struct seq_file *seq, void *v) seq_printf(seq, "%8p %8p %-3d %-3d %-2d %-4d %4d %8d %8d %7d %5lu %-5d %5d ", assoc, sk, sctp_sk(sk)->type, sk->sk_state, - assoc->state, hash-1, assoc->assoc_id, + assoc->state, hash, assoc->assoc_id, (sk->sk_rcvbuf - assoc->rwnd), assoc->sndbuf_used, sock_i_uid(sk), sock_i_ino(sk), -- cgit v1.2.3 From 38b0e42aba928d9929a26ec23b850c36a31fca5f Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Tue, 17 Jan 2006 11:54:06 -0800 Subject: [SCTP]: Fix sctp_assoc_seq_show() panics on big-endian systems. This patch corrects the panic by casting the argument to the pointer of correct size. On big-endian systems we ended up loading only 32 bits of data because we are treating the pointer as an int*. By treating this pointer as loff_t*, we'll load the full 64 bits and then let regular integer demotion take place which will give us the correct value. Signed-off-by: Vlad Yaseivch Signed-off-by: Sridhar Samudrala --- net/sctp/proc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/sctp/proc.c b/net/sctp/proc.c index 1b5e5b119f7..d47a52c303a 100644 --- a/net/sctp/proc.c +++ b/net/sctp/proc.c @@ -210,7 +210,7 @@ static int sctp_eps_seq_show(struct seq_file *seq, void *v) struct sctp_ep_common *epb; struct sctp_endpoint *ep; struct sock *sk; - int hash = *(int *)v; + int hash = *(loff_t *)v; if (hash >= sctp_ep_hashsize) return -ENOMEM; @@ -313,7 +313,7 @@ static int sctp_assocs_seq_show(struct seq_file *seq, void *v) struct sctp_ep_common *epb; struct sctp_association *assoc; struct sock *sk; - int hash = *(int *)v; + int hash = *(loff_t *)v; if (hash >= sctp_assoc_hashsize) return -ENOMEM; -- cgit v1.2.3 From 8116ffad4180b39d7a755345c1fde09da83930c0 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Tue, 17 Jan 2006 11:55:17 -0800 Subject: [SCTP]: Fix bad sysctl formatting of SCTP timeout values on 64-bit m/cs. Change all the structure members that hold jiffies to be of type unsigned long. This also corrects bad sysctl formating on 64 bit architectures. Signed-off-by: Vlad Yasevich Signed-off-by: Sridhar Samudrala --- include/net/sctp/structs.h | 78 ++++++++++++++++++++++++---------------------- net/sctp/sm_sideeffect.c | 4 +-- net/sctp/socket.c | 2 +- net/sctp/sysctl.c | 7 ++--- net/sctp/transport.c | 2 +- 5 files changed, 47 insertions(+), 46 deletions(-) diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 72aeae4a006..ad3d15cb0a0 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -127,9 +127,9 @@ extern struct sctp_globals { * RTO.Alpha - 1/8 (3 when converted to right shifts.) * RTO.Beta - 1/4 (2 when converted to right shifts.) */ - __u32 rto_initial; - __u32 rto_min; - __u32 rto_max; + unsigned long rto_initial; + unsigned long rto_min; + unsigned long rto_max; /* Note: rto_alpha and rto_beta are really defined as inverse * powers of two to facilitate integer operations. @@ -140,12 +140,18 @@ extern struct sctp_globals { /* Max.Burst - 4 */ int max_burst; - /* Valid.Cookie.Life - 60 seconds */ - int valid_cookie_life; - /* Whether Cookie Preservative is enabled(1) or not(0) */ int cookie_preserve_enable; + /* Valid.Cookie.Life - 60 seconds */ + unsigned long valid_cookie_life; + + /* Delayed SACK timeout 200ms default*/ + unsigned long sack_timeout; + + /* HB.interval - 30 seconds */ + unsigned long hb_interval; + /* Association.Max.Retrans - 10 attempts * Path.Max.Retrans - 5 attempts (per destination address) * Max.Init.Retransmits - 8 attempts @@ -168,12 +174,6 @@ extern struct sctp_globals { */ int rcvbuf_policy; - /* Delayed SACK timeout 200ms default*/ - int sack_timeout; - - /* HB.interval - 30 seconds */ - int hb_interval; - /* The following variables are implementation specific. */ /* Default initialization values to be applied to new associations. */ @@ -828,7 +828,7 @@ struct sctp_transport { __u32 rtt; /* This is the most recent RTT. */ /* RTO : The current retransmission timeout value. */ - __u32 rto; + unsigned long rto; /* RTTVAR : The current RTT variation. */ __u32 rttvar; @@ -878,22 +878,10 @@ struct sctp_transport { /* Heartbeat interval: The endpoint sends out a Heartbeat chunk to * the destination address every heartbeat interval. */ - __u32 hbinterval; - - /* This is the max_retrans value for the transport and will - * be initialized from the assocs value. This can be changed - * using SCTP_SET_PEER_ADDR_PARAMS socket option. - */ - __u16 pathmaxrxt; - - /* PMTU : The current known path MTU. */ - __u32 pathmtu; + unsigned long hbinterval; /* SACK delay timeout */ - __u32 sackdelay; - - /* Flags controling Heartbeat, SACK delay, and Path MTU Discovery. */ - __u32 param_flags; + unsigned long sackdelay; /* When was the last time (in jiffies) that we heard from this * transport? We use this to pick new active and retran paths. @@ -905,6 +893,18 @@ struct sctp_transport { */ unsigned long last_time_ecne_reduced; + /* This is the max_retrans value for the transport and will + * be initialized from the assocs value. This can be changed + * using SCTP_SET_PEER_ADDR_PARAMS socket option. + */ + __u16 pathmaxrxt; + + /* PMTU : The current known path MTU. */ + __u32 pathmtu; + + /* Flags controling Heartbeat, SACK delay, and Path MTU Discovery. */ + __u32 param_flags; + /* The number of times INIT has been sent on this transport. */ int init_sent_count; @@ -1500,9 +1500,9 @@ struct sctp_association { * These values will be initialized by system defaults, but can * be modified via the SCTP_RTOINFO socket option. */ - __u32 rto_initial; - __u32 rto_max; - __u32 rto_min; + unsigned long rto_initial; + unsigned long rto_max; + unsigned long rto_min; /* Maximum number of new data packets that can be sent in a burst. */ int max_burst; @@ -1520,13 +1520,13 @@ struct sctp_association { __u16 init_retries; /* The largest timeout or RTO value to use in attempting an INIT */ - __u16 max_init_timeo; + unsigned long max_init_timeo; /* Heartbeat interval: The endpoint sends out a Heartbeat chunk to * the destination address every heartbeat interval. This value * will be inherited by all new transports. */ - __u32 hbinterval; + unsigned long hbinterval; /* This is the max_retrans value for new transports in the * association. @@ -1538,13 +1538,14 @@ struct sctp_association { */ __u32 pathmtu; - /* SACK delay timeout */ - __u32 sackdelay; - /* Flags controling Heartbeat, SACK delay, and Path MTU Discovery. */ __u32 param_flags; - int timeouts[SCTP_NUM_TIMEOUT_TYPES]; + /* SACK delay timeout */ + unsigned long sackdelay; + + + unsigned long timeouts[SCTP_NUM_TIMEOUT_TYPES]; struct timer_list timers[SCTP_NUM_TIMEOUT_TYPES]; /* Transport to which SHUTDOWN chunk was last sent. */ @@ -1649,7 +1650,10 @@ struct sctp_association { /* How many duplicated TSNs have we seen? */ int numduptsns; - /* Number of seconds of idle time before an association is closed. */ + /* Number of seconds of idle time before an association is closed. + * In the association context, this is really used as a boolean + * since the real timeout is stored in the timeouts array + */ __u32 autoclose; /* These are to support diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index b8b38aba92b..8d1dc24bab4 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -1300,7 +1300,7 @@ static int sctp_cmd_interpreter(sctp_event_t event_type, "T1 INIT Timeout adjustment" " init_err_counter: %d" " cycle: %d" - " timeout: %d\n", + " timeout: %ld\n", asoc->init_err_counter, asoc->init_cycle, asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT]); @@ -1328,7 +1328,7 @@ static int sctp_cmd_interpreter(sctp_event_t event_type, SCTP_DEBUG_PRINTK( "T1 COOKIE Timeout adjustment" " init_err_counter: %d" - " timeout: %d\n", + " timeout: %ld\n", asoc->init_err_counter, asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE]); diff --git a/net/sctp/socket.c b/net/sctp/socket.c index c98ee375ba5..6a0b1af8993 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -2995,7 +2995,7 @@ SCTP_STATIC int sctp_init_sock(struct sock *sk) sp->hbinterval = jiffies_to_msecs(sctp_hb_interval); sp->pathmaxrxt = sctp_max_retrans_path; sp->pathmtu = 0; // allow default discovery - sp->sackdelay = sctp_sack_timeout; + sp->sackdelay = jiffies_to_msecs(sctp_sack_timeout); sp->param_flags = SPP_HB_ENABLE | SPP_PMTUD_ENABLE | SPP_SACKDELAY_ENABLE; diff --git a/net/sctp/sysctl.c b/net/sctp/sysctl.c index fcd7096c953..dc6f3ff3235 100644 --- a/net/sctp/sysctl.c +++ b/net/sctp/sysctl.c @@ -159,12 +159,9 @@ static ctl_table sctp_table[] = { .ctl_name = NET_SCTP_PRESERVE_ENABLE, .procname = "cookie_preserve_enable", .data = &sctp_cookie_preserve_enable, - .maxlen = sizeof(long), + .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_doulongvec_ms_jiffies_minmax, - .strategy = &sctp_sysctl_jiffies_ms, - .extra1 = &rto_timer_min, - .extra2 = &rto_timer_max + .proc_handler = &proc_dointvec }, { .ctl_name = NET_SCTP_RTO_ALPHA, diff --git a/net/sctp/transport.c b/net/sctp/transport.c index 68d73e2dd15..160f62ad1cc 100644 --- a/net/sctp/transport.c +++ b/net/sctp/transport.c @@ -350,7 +350,7 @@ void sctp_transport_update_rto(struct sctp_transport *tp, __u32 rtt) tp->rto_pending = 0; SCTP_DEBUG_PRINTK("%s: transport: %p, rtt: %d, srtt: %d " - "rttvar: %d, rto: %d\n", __FUNCTION__, + "rttvar: %d, rto: %ld\n", __FUNCTION__, tp, rtt, tp->srtt, tp->rttvar, tp->rto); } -- cgit v1.2.3 From 313e7b4d2588539e388d31c1febd50503a0083fc Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Tue, 17 Jan 2006 11:55:57 -0800 Subject: [SCTP]: Fix machine check/connection hang on IA64. sctp_unpack_cookie used an on-stack array called digest as a result/out parameter in the call to crypto_hmac. However, hmac code (crypto_hmac_final) assumes that the 'out' argument is in virtual memory (identity mapped region) and can use virt_to_page call on it. This does not work with the on-stack declared digest. The problems observed so far have been: a) incorrect hmac digest b) machine check and hardware reset. Solution is to define the digest in an identity mapped region by kmalloc'ing it. We can do this once as part of the endpoint structure and re-use it when verifying the SCTP cookie. Signed-off-by: Vlad Yasevich Signed-off-by: Sridhar Samudrala --- include/net/sctp/structs.h | 8 ++++++++ net/sctp/sm_make_chunk.c | 2 +- 2 files changed, 9 insertions(+), 1 deletion(-) diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index ad3d15cb0a0..8c522ae031b 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1250,6 +1250,14 @@ struct sctp_endpoint { int last_key; int key_changed_at; + /* digest: This is a digest of the sctp cookie. This field is + * only used on the receive path when we try to validate + * that the cookie has not been tampered with. We put + * this here so we pre-allocate this once and can re-use + * on every receive. + */ + __u8 digest[SCTP_SIGNATURE_SIZE]; + /* sendbuf acct. policy. */ __u32 sndbuf_policy; diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 4fe1d6c863b..5e0de3c0eea 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -1359,7 +1359,7 @@ struct sctp_association *sctp_unpack_cookie( struct sctp_signed_cookie *cookie; struct sctp_cookie *bear_cookie; int headersize, bodysize, fixed_size; - __u8 digest[SCTP_SIGNATURE_SIZE]; + __u8 *digest = ep->digest; struct scatterlist sg; unsigned int keylen, len; char *key; -- cgit v1.2.3 From c4d2444e992c4eda1d7fc3287e93ba58295bf6b9 Mon Sep 17 00:00:00 2001 From: Sridhar Samudrala Date: Tue, 17 Jan 2006 11:56:26 -0800 Subject: [SCTP]: Fix couple of races between sctp_peeloff() and sctp_rcv(). Validate and update the sk in sctp_rcv() to avoid the race where an assoc/ep could move to a different socket after we get the sk, but before the skb is added to the backlog. Also migrate the skb's in backlog queue to new sk when doing a peeloff. Signed-off-by: Sridhar Samudrala --- include/net/sctp/sctp.h | 2 ++ net/sctp/input.c | 35 ++++++++++++++++++++++++++++++++++- net/sctp/socket.c | 4 ++++ 3 files changed, 40 insertions(+), 1 deletion(-) diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index a553f39f6ae..e673b2c984e 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -175,6 +175,8 @@ void sctp_icmp_frag_needed(struct sock *, struct sctp_association *, void sctp_icmp_proto_unreachable(struct sock *sk, struct sctp_association *asoc, struct sctp_transport *t); +void sctp_backlog_migrate(struct sctp_association *assoc, + struct sock *oldsk, struct sock *newsk); /* * Section: Macros, externs, and inlines diff --git a/net/sctp/input.c b/net/sctp/input.c index c463e4049c5..71fd5637564 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -257,12 +257,21 @@ int sctp_rcv(struct sk_buff *skb) */ sctp_bh_lock_sock(sk); + /* It is possible that the association could have moved to a different + * socket if it is peeled off. If so, update the sk. + */ + if (sk != rcvr->sk) { + sctp_bh_lock_sock(rcvr->sk); + sctp_bh_unlock_sock(sk); + sk = rcvr->sk; + } + if (sock_owned_by_user(sk)) sk_add_backlog(sk, skb); else sctp_backlog_rcv(sk, skb); - /* Release the sock and the sock ref we took in the lookup calls. + /* Release the sock and the sock ref we took in the lookup calls. * The asoc/ep ref will be released in sctp_backlog_rcv. */ sctp_bh_unlock_sock(sk); @@ -297,6 +306,9 @@ int sctp_backlog_rcv(struct sock *sk, struct sk_buff *skb) struct sctp_ep_common *rcvr = NULL; rcvr = chunk->rcvr; + + BUG_TRAP(rcvr->sk == sk); + if (rcvr->dead) { sctp_chunk_free(chunk); } else { @@ -313,6 +325,27 @@ int sctp_backlog_rcv(struct sock *sk, struct sk_buff *skb) return 0; } +void sctp_backlog_migrate(struct sctp_association *assoc, + struct sock *oldsk, struct sock *newsk) +{ + struct sk_buff *skb; + struct sctp_chunk *chunk; + + skb = oldsk->sk_backlog.head; + oldsk->sk_backlog.head = oldsk->sk_backlog.tail = NULL; + while (skb != NULL) { + struct sk_buff *next = skb->next; + + chunk = SCTP_INPUT_CB(skb)->chunk; + skb->next = NULL; + if (&assoc->base == chunk->rcvr) + sk_add_backlog(newsk, skb); + else + sk_add_backlog(oldsk, skb); + skb = next; + } +} + /* Handle icmp frag needed error. */ void sctp_icmp_frag_needed(struct sock *sk, struct sctp_association *asoc, struct sctp_transport *t, __u32 pmtu) diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 6a0b1af8993..fb1821d9f33 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -5602,8 +5602,12 @@ static void sctp_sock_migrate(struct sock *oldsk, struct sock *newsk, */ newsp->type = type; + spin_lock_bh(&oldsk->sk_lock.slock); + /* Migrate the backlog from oldsk to newsk. */ + sctp_backlog_migrate(assoc, oldsk, newsk); /* Migrate the association to the new socket. */ sctp_assoc_migrate(assoc, newsk); + spin_unlock_bh(&oldsk->sk_lock.slock); /* If the association on the newsk is already closed before accept() * is called, set RCV_SHUTDOWN flag. -- cgit v1.2.3 From a7d1f1b66c05ef4ebb58a34be7caad9af15546a4 Mon Sep 17 00:00:00 2001 From: Tsutomu Fujii Date: Tue, 17 Jan 2006 11:57:09 -0800 Subject: [SCTP]: Fix sctp_rcv_ootb() to handle the last chunk of a packet correctly. Signed-off-by: Tsutomu Fujii Signed-off-by: Sridhar Samudrala --- net/sctp/input.c | 13 +++++++++---- net/sctp/sm_statefuns.c | 2 ++ 2 files changed, 11 insertions(+), 4 deletions(-) diff --git a/net/sctp/input.c b/net/sctp/input.c index 71fd5637564..cb78b50868e 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -588,10 +588,16 @@ int sctp_rcv_ootb(struct sk_buff *skb) sctp_errhdr_t *err; ch = (sctp_chunkhdr_t *) skb->data; - ch_end = ((__u8 *) ch) + WORD_ROUND(ntohs(ch->length)); /* Scan through all the chunks in the packet. */ - while (ch_end > (__u8 *)ch && ch_end < skb->tail) { + do { + /* Break out if chunk length is less then minimal. */ + if (ntohs(ch->length) < sizeof(sctp_chunkhdr_t)) + break; + + ch_end = ((__u8 *)ch) + WORD_ROUND(ntohs(ch->length)); + if (ch_end > skb->tail) + break; /* RFC 8.4, 2) If the OOTB packet contains an ABORT chunk, the * receiver MUST silently discard the OOTB packet and take no @@ -622,8 +628,7 @@ int sctp_rcv_ootb(struct sk_buff *skb) } ch = (sctp_chunkhdr_t *) ch_end; - ch_end = ((__u8 *) ch) + WORD_ROUND(ntohs(ch->length)); - } + } while (ch_end < skb->tail); return 0; diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 477d7f80dba..71c9a961c32 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -3090,6 +3090,8 @@ sctp_disposition_t sctp_sf_ootb(const struct sctp_endpoint *ep, break; ch_end = ((__u8 *)ch) + WORD_ROUND(ntohs(ch->length)); + if (ch_end > skb->tail) + break; if (SCTP_CID_SHUTDOWN_ACK == ch->type) ootb_shut_ack = 1; -- cgit v1.2.3 From f9e505a9a03df5acace6e758c8d12982635a1c64 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Tue, 17 Jan 2006 12:52:21 -0500 Subject: [IA64-SGI] sn2 mutex conversion Migrate sn2 code to use mutex and completion events rather than semaphores. Signed-off-by: Jes Sorensen Acked-by: Dean Nelson Signed-off-by: Tony Luck --- arch/ia64/sn/kernel/mca.c | 7 ++++--- arch/ia64/sn/kernel/xp_main.c | 17 +++++++++-------- arch/ia64/sn/kernel/xpc_channel.c | 34 ++++++++++++++-------------------- arch/ia64/sn/kernel/xpc_main.c | 17 +++++++++-------- include/asm-ia64/sn/xp.h | 3 ++- include/asm-ia64/sn/xpc.h | 9 +++++---- 6 files changed, 43 insertions(+), 44 deletions(-) diff --git a/arch/ia64/sn/kernel/mca.c b/arch/ia64/sn/kernel/mca.c index 6546db6abdb..9ab684d1bb5 100644 --- a/arch/ia64/sn/kernel/mca.c +++ b/arch/ia64/sn/kernel/mca.c @@ -10,6 +10,7 @@ #include #include #include +#include #include #include #include @@ -27,7 +28,7 @@ void sn_init_cpei_timer(void); /* Printing oemdata from mca uses data that is not passed through SAL, it is * global. Only one user at a time. */ -static DECLARE_MUTEX(sn_oemdata_mutex); +static DEFINE_MUTEX(sn_oemdata_mutex); static u8 **sn_oemdata; static u64 *sn_oemdata_size, sn_oemdata_bufsize; @@ -89,7 +90,7 @@ static int sn_platform_plat_specific_err_print(const u8 * sect_header, u8 ** oemdata, u64 * oemdata_size) { - down(&sn_oemdata_mutex); + mutex_lock(&sn_oemdata_mutex); sn_oemdata = oemdata; sn_oemdata_size = oemdata_size; sn_oemdata_bufsize = 0; @@ -107,7 +108,7 @@ sn_platform_plat_specific_err_print(const u8 * sect_header, u8 ** oemdata, *sn_oemdata_size = 0; ia64_sn_plat_specific_err_print(print_hook, (char *)sect_header); } - up(&sn_oemdata_mutex); + mutex_unlock(&sn_oemdata_mutex); return 0; } diff --git a/arch/ia64/sn/kernel/xp_main.c b/arch/ia64/sn/kernel/xp_main.c index 3be52a34c80..b7ea46645e1 100644 --- a/arch/ia64/sn/kernel/xp_main.c +++ b/arch/ia64/sn/kernel/xp_main.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include #include @@ -136,13 +137,13 @@ xpc_connect(int ch_number, xpc_channel_func func, void *key, u16 payload_size, registration = &xpc_registrations[ch_number]; - if (down_interruptible(®istration->sema) != 0) { + if (mutex_lock_interruptible(®istration->mutex) != 0) { return xpcInterrupted; } /* if XPC_CHANNEL_REGISTERED(ch_number) */ if (registration->func != NULL) { - up(®istration->sema); + mutex_unlock(®istration->mutex); return xpcAlreadyRegistered; } @@ -154,7 +155,7 @@ xpc_connect(int ch_number, xpc_channel_func func, void *key, u16 payload_size, registration->key = key; registration->func = func; - up(®istration->sema); + mutex_unlock(®istration->mutex); xpc_interface.connect(ch_number); @@ -190,11 +191,11 @@ xpc_disconnect(int ch_number) * figured XPC's users will just turn around and call xpc_disconnect() * again anyways, so we might as well wait, if need be. */ - down(®istration->sema); + mutex_lock(®istration->mutex); /* if !XPC_CHANNEL_REGISTERED(ch_number) */ if (registration->func == NULL) { - up(®istration->sema); + mutex_unlock(®istration->mutex); return; } @@ -208,7 +209,7 @@ xpc_disconnect(int ch_number) xpc_interface.disconnect(ch_number); - up(®istration->sema); + mutex_unlock(®istration->mutex); return; } @@ -250,9 +251,9 @@ xp_init(void) xp_nofault_PIOR_target = SH1_IPI_ACCESS; } - /* initialize the connection registration semaphores */ + /* initialize the connection registration mutex */ for (ch_number = 0; ch_number < XPC_NCHANNELS; ch_number++) { - sema_init(&xpc_registrations[ch_number].sema, 1); /* mutex */ + mutex_init(&xpc_registrations[ch_number].mutex); } return 0; diff --git a/arch/ia64/sn/kernel/xpc_channel.c b/arch/ia64/sn/kernel/xpc_channel.c index 0c0a6890240..8d950c778bb 100644 --- a/arch/ia64/sn/kernel/xpc_channel.c +++ b/arch/ia64/sn/kernel/xpc_channel.c @@ -22,6 +22,8 @@ #include #include #include +#include +#include #include #include #include @@ -56,8 +58,8 @@ xpc_initialize_channels(struct xpc_partition *part, partid_t partid) atomic_set(&ch->n_to_notify, 0); spin_lock_init(&ch->lock); - sema_init(&ch->msg_to_pull_sema, 1); /* mutex */ - sema_init(&ch->wdisconnect_sema, 0); /* event wait */ + mutex_init(&ch->msg_to_pull_mutex); + init_completion(&ch->wdisconnect_wait); atomic_set(&ch->n_on_msg_allocate_wq, 0); init_waitqueue_head(&ch->msg_allocate_wq); @@ -534,7 +536,6 @@ static enum xpc_retval xpc_allocate_msgqueues(struct xpc_channel *ch) { unsigned long irq_flags; - int i; enum xpc_retval ret; @@ -552,11 +553,6 @@ xpc_allocate_msgqueues(struct xpc_channel *ch) return ret; } - for (i = 0; i < ch->local_nentries; i++) { - /* use a semaphore as an event wait queue */ - sema_init(&ch->notify_queue[i].sema, 0); - } - spin_lock_irqsave(&ch->lock, irq_flags); ch->flags |= XPC_C_SETUP; spin_unlock_irqrestore(&ch->lock, irq_flags); @@ -799,10 +795,8 @@ xpc_process_disconnect(struct xpc_channel *ch, unsigned long *irq_flags) } if (ch->flags & XPC_C_WDISCONNECT) { - spin_unlock_irqrestore(&ch->lock, *irq_flags); - up(&ch->wdisconnect_sema); - spin_lock_irqsave(&ch->lock, *irq_flags); - + /* we won't lose the CPU since we're holding ch->lock */ + complete(&ch->wdisconnect_wait); } else if (ch->delayed_IPI_flags) { if (part->act_state != XPC_P_DEACTIVATING) { /* time to take action on any delayed IPI flags */ @@ -1092,12 +1086,12 @@ xpc_connect_channel(struct xpc_channel *ch) struct xpc_registration *registration = &xpc_registrations[ch->number]; - if (down_trylock(®istration->sema) != 0) { + if (mutex_trylock(®istration->mutex) == 0) { return xpcRetry; } if (!XPC_CHANNEL_REGISTERED(ch->number)) { - up(®istration->sema); + mutex_unlock(®istration->mutex); return xpcUnregistered; } @@ -1108,7 +1102,7 @@ xpc_connect_channel(struct xpc_channel *ch) if (ch->flags & XPC_C_DISCONNECTING) { spin_unlock_irqrestore(&ch->lock, irq_flags); - up(®istration->sema); + mutex_unlock(®istration->mutex); return ch->reason; } @@ -1140,7 +1134,7 @@ xpc_connect_channel(struct xpc_channel *ch) * channel lock be locked and will unlock and relock * the channel lock as needed. */ - up(®istration->sema); + mutex_unlock(®istration->mutex); XPC_DISCONNECT_CHANNEL(ch, xpcUnequalMsgSizes, &irq_flags); spin_unlock_irqrestore(&ch->lock, irq_flags); @@ -1155,7 +1149,7 @@ xpc_connect_channel(struct xpc_channel *ch) atomic_inc(&xpc_partitions[ch->partid].nchannels_active); } - up(®istration->sema); + mutex_unlock(®istration->mutex); /* initiate the connection */ @@ -2089,7 +2083,7 @@ xpc_pull_remote_msg(struct xpc_channel *ch, s64 get) enum xpc_retval ret; - if (down_interruptible(&ch->msg_to_pull_sema) != 0) { + if (mutex_lock_interruptible(&ch->msg_to_pull_mutex) != 0) { /* we were interrupted by a signal */ return NULL; } @@ -2125,7 +2119,7 @@ xpc_pull_remote_msg(struct xpc_channel *ch, s64 get) XPC_DEACTIVATE_PARTITION(part, ret); - up(&ch->msg_to_pull_sema); + mutex_unlock(&ch->msg_to_pull_mutex); return NULL; } @@ -2134,7 +2128,7 @@ xpc_pull_remote_msg(struct xpc_channel *ch, s64 get) ch->next_msg_to_pull += nmsgs; } - up(&ch->msg_to_pull_sema); + mutex_unlock(&ch->msg_to_pull_mutex); /* return the message we were looking for */ msg_offset = (get % ch->remote_nentries) * ch->msg_size; diff --git a/arch/ia64/sn/kernel/xpc_main.c b/arch/ia64/sn/kernel/xpc_main.c index 8930586e0eb..c75f8aeefc2 100644 --- a/arch/ia64/sn/kernel/xpc_main.c +++ b/arch/ia64/sn/kernel/xpc_main.c @@ -55,6 +55,7 @@ #include #include #include +#include #include #include #include @@ -177,10 +178,10 @@ static DECLARE_WAIT_QUEUE_HEAD(xpc_act_IRQ_wq); static unsigned long xpc_hb_check_timeout; /* notification that the xpc_hb_checker thread has exited */ -static DECLARE_MUTEX_LOCKED(xpc_hb_checker_exited); +static DECLARE_COMPLETION(xpc_hb_checker_exited); /* notification that the xpc_discovery thread has exited */ -static DECLARE_MUTEX_LOCKED(xpc_discovery_exited); +static DECLARE_COMPLETION(xpc_discovery_exited); static struct timer_list xpc_hb_timer; @@ -321,7 +322,7 @@ xpc_hb_checker(void *ignore) /* mark this thread as having exited */ - up(&xpc_hb_checker_exited); + complete(&xpc_hb_checker_exited); return 0; } @@ -341,7 +342,7 @@ xpc_initiate_discovery(void *ignore) dev_dbg(xpc_part, "discovery thread is exiting\n"); /* mark this thread as having exited */ - up(&xpc_discovery_exited); + complete(&xpc_discovery_exited); return 0; } @@ -893,7 +894,7 @@ xpc_disconnect_wait(int ch_number) continue; } - (void) down(&ch->wdisconnect_sema); + wait_for_completion(&ch->wdisconnect_wait); spin_lock_irqsave(&ch->lock, irq_flags); DBUG_ON(!(ch->flags & XPC_C_DISCONNECTED)); @@ -946,10 +947,10 @@ xpc_do_exit(enum xpc_retval reason) free_irq(SGI_XPC_ACTIVATE, NULL); /* wait for the discovery thread to exit */ - down(&xpc_discovery_exited); + wait_for_completion(&xpc_discovery_exited); /* wait for the heartbeat checker thread to exit */ - down(&xpc_hb_checker_exited); + wait_for_completion(&xpc_hb_checker_exited); /* sleep for a 1/3 of a second or so */ @@ -1367,7 +1368,7 @@ xpc_init(void) dev_err(xpc_part, "failed while forking discovery thread\n"); /* mark this new thread as a non-starter */ - up(&xpc_discovery_exited); + complete(&xpc_discovery_exited); xpc_do_exit(xpcUnloading); return -EBUSY; diff --git a/include/asm-ia64/sn/xp.h b/include/asm-ia64/sn/xp.h index 203945ae034..9bd2f9bf329 100644 --- a/include/asm-ia64/sn/xp.h +++ b/include/asm-ia64/sn/xp.h @@ -18,6 +18,7 @@ #include #include +#include #include #include @@ -359,7 +360,7 @@ typedef void (*xpc_notify_func)(enum xpc_retval reason, partid_t partid, * the channel. */ struct xpc_registration { - struct semaphore sema; + struct mutex mutex; xpc_channel_func func; /* function to call */ void *key; /* pointer to user's key */ u16 nentries; /* #of msg entries in local msg queue */ diff --git a/include/asm-ia64/sn/xpc.h b/include/asm-ia64/sn/xpc.h index 87e9cd58851..0c36928ffd8 100644 --- a/include/asm-ia64/sn/xpc.h +++ b/include/asm-ia64/sn/xpc.h @@ -19,6 +19,8 @@ #include #include #include +#include +#include #include #include #include @@ -335,8 +337,7 @@ struct xpc_openclose_args { * and consumed by the intended recipient. */ struct xpc_notify { - struct semaphore sema; /* notify semaphore */ - volatile u8 type; /* type of notification */ + volatile u8 type; /* type of notification */ /* the following two fields are only used if type == XPC_N_CALL */ xpc_notify_func func; /* user's notify function */ @@ -465,8 +466,8 @@ struct xpc_channel { xpc_channel_func func; /* user's channel function */ void *key; /* pointer to user's key */ - struct semaphore msg_to_pull_sema; /* next msg to pull serialization */ - struct semaphore wdisconnect_sema; /* wait for channel disconnect */ + struct mutex msg_to_pull_mutex; /* next msg to pull serialization */ + struct completion wdisconnect_wait; /* wait for channel disconnect */ struct xpc_openclose_args *local_openclose_args; /* args passed on */ /* opening or closing of channel */ -- cgit v1.2.3 From d171e519da635a82ab759cbfd46617ac160c9ec0 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Tue, 17 Jan 2006 13:05:12 -0500 Subject: [IA64-SGI] sn_console.c minor cleanup Fix printk level and remove unnecessary CONFIG_SMP|CONFIG_PREEMPT tests as this is taken care through the spinlock macros anyway. Signed-off-by: Jes Sorensen Signed-off-by: Pat Gefre Signed-off-by: Tony Luck --- drivers/serial/sn_console.c | 129 ++++++++++++++++++++------------------------ 1 file changed, 58 insertions(+), 71 deletions(-) diff --git a/drivers/serial/sn_console.c b/drivers/serial/sn_console.c index 5468e5a767e..43e67d6c29d 100644 --- a/drivers/serial/sn_console.c +++ b/drivers/serial/sn_console.c @@ -6,7 +6,7 @@ * driver for that. * * - * Copyright (c) 2004-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2004-2006 Silicon Graphics, Inc. All Rights Reserved. * * This program is free software; you can redistribute it and/or modify it * under the terms of version 2 of the GNU General Public License @@ -829,8 +829,8 @@ static int __init sn_sal_module_init(void) misc.name = DEVICE_NAME_DYNAMIC; retval = misc_register(&misc); if (retval != 0) { - printk - ("Failed to register console device using misc_register.\n"); + printk(KERN_WARNING "Failed to register console " + "device using misc_register.\n"); return -ENODEV; } sal_console_uart.major = MISC_MAJOR; @@ -942,88 +942,75 @@ sn_sal_console_write(struct console *co, const char *s, unsigned count) { unsigned long flags = 0; struct sn_cons_port *port = &sal_console_port; -#if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT) static int stole_lock = 0; -#endif BUG_ON(!port->sc_is_asynch); /* We can't look at the xmit buffer if we're not registered with serial core * yet. So only do the fancy recovery after registering */ - if (port->sc_port.info) { - - /* somebody really wants this output, might be an - * oops, kdb, panic, etc. make sure they get it. */ -#if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT) - if (spin_is_locked(&port->sc_port.lock)) { - int lhead = port->sc_port.info->xmit.head; - int ltail = port->sc_port.info->xmit.tail; - int counter, got_lock = 0; + if (!port->sc_port.info) { + /* Not yet registered with serial core - simple case */ + puts_raw_fixed(port->sc_ops->sal_puts_raw, s, count); + return; + } - /* - * We attempt to determine if someone has died with the - * lock. We wait ~20 secs after the head and tail ptrs - * stop moving and assume the lock holder is not functional - * and plow ahead. If the lock is freed within the time out - * period we re-get the lock and go ahead normally. We also - * remember if we have plowed ahead so that we don't have - * to wait out the time out period again - the asumption - * is that we will time out again. - */ + /* somebody really wants this output, might be an + * oops, kdb, panic, etc. make sure they get it. */ + if (spin_is_locked(&port->sc_port.lock)) { + int lhead = port->sc_port.info->xmit.head; + int ltail = port->sc_port.info->xmit.tail; + int counter, got_lock = 0; + + /* + * We attempt to determine if someone has died with the + * lock. We wait ~20 secs after the head and tail ptrs + * stop moving and assume the lock holder is not functional + * and plow ahead. If the lock is freed within the time out + * period we re-get the lock and go ahead normally. We also + * remember if we have plowed ahead so that we don't have + * to wait out the time out period again - the asumption + * is that we will time out again. + */ - for (counter = 0; counter < 150; mdelay(125), counter++) { - if (!spin_is_locked(&port->sc_port.lock) - || stole_lock) { - if (!stole_lock) { - spin_lock_irqsave(&port-> - sc_port.lock, - flags); - got_lock = 1; - } - break; - } else { - /* still locked */ - if ((lhead != - port->sc_port.info->xmit.head) - || (ltail != - port->sc_port.info->xmit. - tail)) { - lhead = - port->sc_port.info->xmit. - head; - ltail = - port->sc_port.info->xmit. - tail; - counter = 0; - } + for (counter = 0; counter < 150; mdelay(125), counter++) { + if (!spin_is_locked(&port->sc_port.lock) + || stole_lock) { + if (!stole_lock) { + spin_lock_irqsave(&port->sc_port.lock, + flags); + got_lock = 1; } - } - /* flush anything in the serial core xmit buffer, raw */ - sn_transmit_chars(port, 1); - if (got_lock) { - spin_unlock_irqrestore(&port->sc_port.lock, - flags); - stole_lock = 0; + break; } else { - /* fell thru */ - stole_lock = 1; + /* still locked */ + if ((lhead != port->sc_port.info->xmit.head) + || (ltail != + port->sc_port.info->xmit.tail)) { + lhead = + port->sc_port.info->xmit.head; + ltail = + port->sc_port.info->xmit.tail; + counter = 0; + } } - puts_raw_fixed(port->sc_ops->sal_puts_raw, s, count); - } else { - stole_lock = 0; -#endif - spin_lock_irqsave(&port->sc_port.lock, flags); - sn_transmit_chars(port, 1); + } + /* flush anything in the serial core xmit buffer, raw */ + sn_transmit_chars(port, 1); + if (got_lock) { spin_unlock_irqrestore(&port->sc_port.lock, flags); - - puts_raw_fixed(port->sc_ops->sal_puts_raw, s, count); -#if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT) + stole_lock = 0; + } else { + /* fell thru */ + stole_lock = 1; } -#endif - } - else { - /* Not yet registered with serial core - simple case */ + puts_raw_fixed(port->sc_ops->sal_puts_raw, s, count); + } else { + stole_lock = 0; + spin_lock_irqsave(&port->sc_port.lock, flags); + sn_transmit_chars(port, 1); + spin_unlock_irqrestore(&port->sc_port.lock, flags); + puts_raw_fixed(port->sc_ops->sal_puts_raw, s, count); } } -- cgit v1.2.3 From 8d08aed8d7714683b33666cc066e20f957dda01d Mon Sep 17 00:00:00 2001 From: Jack Steiner Date: Tue, 17 Jan 2006 15:42:46 -0600 Subject: [IA64] Zonelists for nodes without cpus If a node runs out of memory, ensure that memory on nodes w/o cpus is used before using memory on nodes with cpus. Signed-off-by: Jack Steiner Signed-off-by: Tony Luck --- include/asm-ia64/topology.h | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/include/asm-ia64/topology.h b/include/asm-ia64/topology.h index d8aae4da397..412ef8e493a 100644 --- a/include/asm-ia64/topology.h +++ b/include/asm-ia64/topology.h @@ -18,6 +18,10 @@ #include #ifdef CONFIG_NUMA + +/* Nodes w/o CPUs are preferred for memory allocations, see build_zonelists */ +#define PENALTY_FOR_NODE_WITH_CPUS 255 + /* * Returns the number of the node containing CPU 'cpu' */ -- cgit v1.2.3 From 4b16bfbf8f8013fefb49592d030ff87651ab48cb Mon Sep 17 00:00:00 2001 From: Zoltan Menyhart Date: Fri, 13 Jan 2006 17:25:23 +0100 Subject: [IA64] Fix bug in ia64 specific down() function Chen, Kenneth W wrote: > The memory order semantics for include/asm-ia64/semaphore.h:down() > doesn't look right. It is using atomic_dec_return, which eventually > translate into ia64_fetch_and_add() that uses release semantics. > Shouldn't it use acquire semantics? Use ia64_fetchadd() instead of atomic_dec_return() Acked-by: Ken Chen Signed-off-by: Tony Luck --- include/asm-ia64/semaphore.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/include/asm-ia64/semaphore.h b/include/asm-ia64/semaphore.h index bb8906285fa..f483eeb95dd 100644 --- a/include/asm-ia64/semaphore.h +++ b/include/asm-ia64/semaphore.h @@ -61,7 +61,7 @@ static inline void down (struct semaphore *sem) { might_sleep(); - if (atomic_dec_return(&sem->count) < 0) + if (ia64_fetchadd(-1, &sem->count.counter, acq) < 1) __down(sem); } @@ -75,7 +75,7 @@ down_interruptible (struct semaphore * sem) int ret = 0; might_sleep(); - if (atomic_dec_return(&sem->count) < 0) + if (ia64_fetchadd(-1, &sem->count.counter, acq) < 1) ret = __down_interruptible(sem); return ret; } @@ -85,7 +85,7 @@ down_trylock (struct semaphore *sem) { int ret = 0; - if (atomic_dec_return(&sem->count) < 0) + if (ia64_fetchadd(-1, &sem->count.counter, acq) < 1) ret = __down_trylock(sem); return ret; } @@ -93,7 +93,7 @@ down_trylock (struct semaphore *sem) static inline void up (struct semaphore * sem) { - if (atomic_inc_return(&sem->count) <= 0) + if (ia64_fetchadd(1, &sem->count.counter, rel) <= -1) __up(sem); } -- cgit v1.2.3 From f07adc591e6ff100773b93b643f58d9773df6e21 Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Tue, 17 Jan 2006 15:27:09 -0800 Subject: ARM: OMAP: 1/4 Fix clock framework to use clk_enable/disable This patch fixes OMAP clock framework to use clk_enable/disable instead of clk_use/unuse as specified in include/linux/clk.h. Instances of clk_use/unuse are renamed to clk_enable/disable, and references clk_use/unuse are removed. Signed-off-by: Tony Lindgren --- arch/arm/plat-omap/clock.c | 41 ++++----------------------------------- include/asm-arm/arch-omap/clock.h | 2 -- 2 files changed, 4 insertions(+), 39 deletions(-) diff --git a/arch/arm/plat-omap/clock.c b/arch/arm/plat-omap/clock.c index 7ebc5a29db8..3c2bfc0efda 100644 --- a/arch/arm/plat-omap/clock.c +++ b/arch/arm/plat-omap/clock.c @@ -34,7 +34,7 @@ DEFINE_SPINLOCK(clockfw_lock); static struct clk_functions *arch_clock; /*------------------------------------------------------------------------- - * Standard clock functions defined in asm/hardware/clock.h + * Standard clock functions defined in include/linux/clk.h *-------------------------------------------------------------------------*/ struct clk * clk_get(struct device *dev, const char *id) @@ -60,12 +60,8 @@ int clk_enable(struct clk *clk) int ret = 0; spin_lock_irqsave(&clockfw_lock, flags); - if (clk->enable) - ret = clk->enable(clk); - else if (arch_clock->clk_enable) + if (arch_clock->clk_enable) ret = arch_clock->clk_enable(clk); - else - printk(KERN_ERR "Could not enable clock %s\n", clk->name); spin_unlock_irqrestore(&clockfw_lock, flags); return ret; @@ -77,41 +73,12 @@ void clk_disable(struct clk *clk) unsigned long flags; spin_lock_irqsave(&clockfw_lock, flags); - if (clk->disable) - clk->disable(clk); - else if (arch_clock->clk_disable) + if (arch_clock->clk_disable) arch_clock->clk_disable(clk); - else - printk(KERN_ERR "Could not disable clock %s\n", clk->name); spin_unlock_irqrestore(&clockfw_lock, flags); } EXPORT_SYMBOL(clk_disable); -int clk_use(struct clk *clk) -{ - unsigned long flags; - int ret = 0; - - spin_lock_irqsave(&clockfw_lock, flags); - if (arch_clock->clk_use) - ret = arch_clock->clk_use(clk); - spin_unlock_irqrestore(&clockfw_lock, flags); - - return ret; -} -EXPORT_SYMBOL(clk_use); - -void clk_unuse(struct clk *clk) -{ - unsigned long flags; - - spin_lock_irqsave(&clockfw_lock, flags); - if (arch_clock->clk_unuse) - arch_clock->clk_unuse(clk); - spin_unlock_irqrestore(&clockfw_lock, flags); -} -EXPORT_SYMBOL(clk_unuse); - int clk_get_usecount(struct clk *clk) { unsigned long flags; @@ -146,7 +113,7 @@ void clk_put(struct clk *clk) EXPORT_SYMBOL(clk_put); /*------------------------------------------------------------------------- - * Optional clock functions defined in asm/hardware/clock.h + * Optional clock functions defined in include/linux/clk.h *-------------------------------------------------------------------------*/ long clk_round_rate(struct clk *clk, unsigned long rate) diff --git a/include/asm-arm/arch-omap/clock.h b/include/asm-arm/arch-omap/clock.h index 740c297eb11..46a0402696d 100644 --- a/include/asm-arm/arch-omap/clock.h +++ b/include/asm-arm/arch-omap/clock.h @@ -38,8 +38,6 @@ struct clk { struct clk_functions { int (*clk_enable)(struct clk *clk); void (*clk_disable)(struct clk *clk); - int (*clk_use)(struct clk *clk); - void (*clk_unuse)(struct clk *clk); long (*clk_round_rate)(struct clk *clk, unsigned long rate); int (*clk_set_rate)(struct clk *clk, unsigned long rate); int (*clk_set_parent)(struct clk *clk, struct clk *parent); -- cgit v1.2.3 From 10b55794134b279e2ce37713972e324c0dd507ab Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Tue, 17 Jan 2006 15:30:42 -0800 Subject: ARM: OMAP: 2/4 Fix clock framework to use clk_enable/disable for omap1 This patch fixes OMAP clock framework to use clk_enable/disable instead of clk_use/unuse as specified in include/linux/clk.h. Signed-off-by: Tony Lindgren --- arch/arm/mach-omap1/clock.c | 44 ++++++------ arch/arm/mach-omap1/clock.h | 168 ++++++++++++++++++++++---------------------- 2 files changed, 106 insertions(+), 106 deletions(-) diff --git a/arch/arm/mach-omap1/clock.c b/arch/arm/mach-omap1/clock.c index 9d862f86bba..75110ba1042 100644 --- a/arch/arm/mach-omap1/clock.c +++ b/arch/arm/mach-omap1/clock.c @@ -50,10 +50,10 @@ static int omap1_clk_enable_dsp_domain(struct clk *clk) { int retval; - retval = omap1_clk_use(&api_ck.clk); + retval = omap1_clk_enable(&api_ck.clk); if (!retval) { - retval = omap1_clk_enable(clk); - omap1_clk_unuse(&api_ck.clk); + retval = omap1_clk_enable_generic(clk); + omap1_clk_disable(&api_ck.clk); } return retval; @@ -61,9 +61,9 @@ static int omap1_clk_enable_dsp_domain(struct clk *clk) static void omap1_clk_disable_dsp_domain(struct clk *clk) { - if (omap1_clk_use(&api_ck.clk) == 0) { - omap1_clk_disable(clk); - omap1_clk_unuse(&api_ck.clk); + if (omap1_clk_enable(&api_ck.clk) == 0) { + omap1_clk_disable_generic(clk); + omap1_clk_disable(&api_ck.clk); } } @@ -72,7 +72,7 @@ static int omap1_clk_enable_uart_functional(struct clk *clk) int ret; struct uart_clk *uclk; - ret = omap1_clk_enable(clk); + ret = omap1_clk_enable_generic(clk); if (ret == 0) { /* Set smart idle acknowledgement mode */ uclk = (struct uart_clk *)clk; @@ -91,7 +91,7 @@ static void omap1_clk_disable_uart_functional(struct clk *clk) uclk = (struct uart_clk *)clk; omap_writeb((omap_readb(uclk->sysc_addr) & ~0x18), uclk->sysc_addr); - omap1_clk_disable(clk); + omap1_clk_disable_generic(clk); } static void omap1_clk_allow_idle(struct clk *clk) @@ -230,9 +230,9 @@ static void omap1_ckctl_recalc_dsp_domain(struct clk * clk) * Note that DSP_CKCTL virt addr = phys addr, so * we must use __raw_readw() instead of omap_readw(). */ - omap1_clk_use(&api_ck.clk); + omap1_clk_enable(&api_ck.clk); dsor = 1 << (3 & (__raw_readw(DSP_CKCTL) >> clk->rate_offset)); - omap1_clk_unuse(&api_ck.clk); + omap1_clk_disable(&api_ck.clk); if (unlikely(clk->rate == clk->parent->rate / dsor)) return; /* No change, quick exit */ @@ -412,12 +412,12 @@ static void omap1_init_ext_clk(struct clk * clk) clk-> rate = 96000000 / dsor; } -static int omap1_clk_use(struct clk *clk) +static int omap1_clk_enable(struct clk *clk) { int ret = 0; if (clk->usecount++ == 0) { if (likely(clk->parent)) { - ret = omap1_clk_use(clk->parent); + ret = omap1_clk_enable(clk->parent); if (unlikely(ret != 0)) { clk->usecount--; @@ -432,7 +432,7 @@ static int omap1_clk_use(struct clk *clk) ret = clk->enable(clk); if (unlikely(ret != 0) && clk->parent) { - omap1_clk_unuse(clk->parent); + omap1_clk_disable(clk->parent); clk->usecount--; } } @@ -440,12 +440,12 @@ static int omap1_clk_use(struct clk *clk) return ret; } -static void omap1_clk_unuse(struct clk *clk) +static void omap1_clk_disable(struct clk *clk) { if (clk->usecount > 0 && !(--clk->usecount)) { clk->disable(clk); if (likely(clk->parent)) { - omap1_clk_unuse(clk->parent); + omap1_clk_disable(clk->parent); if (clk->flags & CLOCK_NO_IDLE_PARENT) if (!cpu_is_omap24xx()) omap1_clk_allow_idle(clk->parent); @@ -453,7 +453,7 @@ static void omap1_clk_unuse(struct clk *clk) } } -static int omap1_clk_enable(struct clk *clk) +static int omap1_clk_enable_generic(struct clk *clk) { __u16 regval16; __u32 regval32; @@ -492,7 +492,7 @@ static int omap1_clk_enable(struct clk *clk) return 0; } -static void omap1_clk_disable(struct clk *clk) +static void omap1_clk_disable_generic(struct clk *clk) { __u16 regval16; __u32 regval32; @@ -654,8 +654,8 @@ late_initcall(omap1_late_clk_reset); #endif static struct clk_functions omap1_clk_functions = { - .clk_use = omap1_clk_use, - .clk_unuse = omap1_clk_unuse, + .clk_enable = omap1_clk_enable, + .clk_disable = omap1_clk_disable, .clk_round_rate = omap1_clk_round_rate, .clk_set_rate = omap1_clk_set_rate, }; @@ -780,9 +780,9 @@ int __init omap1_clk_init(void) * Only enable those clocks we will need, let the drivers * enable other clocks as necessary */ - clk_use(&armper_ck.clk); - clk_use(&armxor_ck.clk); - clk_use(&armtim_ck.clk); /* This should be done by timer code */ + clk_enable(&armper_ck.clk); + clk_enable(&armxor_ck.clk); + clk_enable(&armtim_ck.clk); /* This should be done by timer code */ if (cpu_is_omap1510()) clk_enable(&arm_gpio_ck); diff --git a/arch/arm/mach-omap1/clock.h b/arch/arm/mach-omap1/clock.h index f3bdfb50e01..4f18d1b9444 100644 --- a/arch/arm/mach-omap1/clock.h +++ b/arch/arm/mach-omap1/clock.h @@ -13,8 +13,8 @@ #ifndef __ARCH_ARM_MACH_OMAP1_CLOCK_H #define __ARCH_ARM_MACH_OMAP1_CLOCK_H -static int omap1_clk_enable(struct clk * clk); -static void omap1_clk_disable(struct clk * clk); +static int omap1_clk_enable_generic(struct clk * clk); +static void omap1_clk_disable_generic(struct clk * clk); static void omap1_ckctl_recalc(struct clk * clk); static void omap1_watchdog_recalc(struct clk * clk); static void omap1_ckctl_recalc_dsp_domain(struct clk * clk); @@ -30,8 +30,8 @@ static long omap1_round_ext_clk_rate(struct clk * clk, unsigned long rate); static void omap1_init_ext_clk(struct clk * clk); static int omap1_select_table_rate(struct clk * clk, unsigned long rate); static long omap1_round_to_table_rate(struct clk * clk, unsigned long rate); -static int omap1_clk_use(struct clk *clk); -static void omap1_clk_unuse(struct clk *clk); +static int omap1_clk_enable(struct clk *clk); +static void omap1_clk_disable(struct clk *clk); struct mpu_rate { unsigned long rate; @@ -152,8 +152,8 @@ static struct clk ck_ref = { .rate = 12000000, .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | ALWAYS_ENABLED, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk ck_dpll1 = { @@ -161,8 +161,8 @@ static struct clk ck_dpll1 = { .parent = &ck_ref, .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | RATE_PROPAGATES | ALWAYS_ENABLED, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct arm_idlect1_clk ck_dpll1out = { @@ -173,8 +173,8 @@ static struct arm_idlect1_clk ck_dpll1out = { .enable_reg = (void __iomem *)ARM_IDLECT2, .enable_bit = EN_CKOUT_ARM, .recalc = &followparent_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }, .idlect_shift = 12, }; @@ -186,8 +186,8 @@ static struct clk arm_ck = { RATE_CKCTL | RATE_PROPAGATES | ALWAYS_ENABLED, .rate_offset = CKCTL_ARMDIV_OFFSET, .recalc = &omap1_ckctl_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct arm_idlect1_clk armper_ck = { @@ -200,8 +200,8 @@ static struct arm_idlect1_clk armper_ck = { .enable_bit = EN_PERCK, .rate_offset = CKCTL_PERDIV_OFFSET, .recalc = &omap1_ckctl_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }, .idlect_shift = 2, }; @@ -213,8 +213,8 @@ static struct clk arm_gpio_ck = { .enable_reg = (void __iomem *)ARM_IDLECT2, .enable_bit = EN_GPIOCK, .recalc = &followparent_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct arm_idlect1_clk armxor_ck = { @@ -226,8 +226,8 @@ static struct arm_idlect1_clk armxor_ck = { .enable_reg = (void __iomem *)ARM_IDLECT2, .enable_bit = EN_XORPCK, .recalc = &followparent_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }, .idlect_shift = 1, }; @@ -241,8 +241,8 @@ static struct arm_idlect1_clk armtim_ck = { .enable_reg = (void __iomem *)ARM_IDLECT2, .enable_bit = EN_TIMCK, .recalc = &followparent_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }, .idlect_shift = 9, }; @@ -256,8 +256,8 @@ static struct arm_idlect1_clk armwdt_ck = { .enable_reg = (void __iomem *)ARM_IDLECT2, .enable_bit = EN_WDTCK, .recalc = &omap1_watchdog_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }, .idlect_shift = 0, }; @@ -272,8 +272,8 @@ static struct clk arminth_ck16xx = { * * 1510 version is in TC clocks. */ - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk dsp_ck = { @@ -285,8 +285,8 @@ static struct clk dsp_ck = { .enable_bit = EN_DSPCK, .rate_offset = CKCTL_DSPDIV_OFFSET, .recalc = &omap1_ckctl_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk dspmmu_ck = { @@ -296,8 +296,8 @@ static struct clk dspmmu_ck = { RATE_CKCTL | ALWAYS_ENABLED, .rate_offset = CKCTL_DSPMMUDIV_OFFSET, .recalc = &omap1_ckctl_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk dspper_ck = { @@ -349,8 +349,8 @@ static struct arm_idlect1_clk tc_ck = { CLOCK_IDLE_CONTROL, .rate_offset = CKCTL_TCDIV_OFFSET, .recalc = &omap1_ckctl_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }, .idlect_shift = 6, }; @@ -364,8 +364,8 @@ static struct clk arminth_ck1510 = { * * 16xx version is in MPU clocks. */ - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk tipb_ck = { @@ -374,8 +374,8 @@ static struct clk tipb_ck = { .parent = &tc_ck.clk, .flags = CLOCK_IN_OMAP1510 | ALWAYS_ENABLED, .recalc = &followparent_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk l3_ocpi_ck = { @@ -386,8 +386,8 @@ static struct clk l3_ocpi_ck = { .enable_reg = (void __iomem *)ARM_IDLECT3, .enable_bit = EN_OCPI_CK, .recalc = &followparent_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk tc1_ck = { @@ -397,8 +397,8 @@ static struct clk tc1_ck = { .enable_reg = (void __iomem *)ARM_IDLECT3, .enable_bit = EN_TC1_CK, .recalc = &followparent_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk tc2_ck = { @@ -408,8 +408,8 @@ static struct clk tc2_ck = { .enable_reg = (void __iomem *)ARM_IDLECT3, .enable_bit = EN_TC2_CK, .recalc = &followparent_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk dma_ck = { @@ -419,8 +419,8 @@ static struct clk dma_ck = { .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | ALWAYS_ENABLED, .recalc = &followparent_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk dma_lcdfree_ck = { @@ -428,8 +428,8 @@ static struct clk dma_lcdfree_ck = { .parent = &tc_ck.clk, .flags = CLOCK_IN_OMAP16XX | ALWAYS_ENABLED, .recalc = &followparent_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct arm_idlect1_clk api_ck = { @@ -441,8 +441,8 @@ static struct arm_idlect1_clk api_ck = { .enable_reg = (void __iomem *)ARM_IDLECT2, .enable_bit = EN_APICK, .recalc = &followparent_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }, .idlect_shift = 8, }; @@ -455,8 +455,8 @@ static struct arm_idlect1_clk lb_ck = { .enable_reg = (void __iomem *)ARM_IDLECT2, .enable_bit = EN_LBCK, .recalc = &followparent_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }, .idlect_shift = 4, }; @@ -466,8 +466,8 @@ static struct clk rhea1_ck = { .parent = &tc_ck.clk, .flags = CLOCK_IN_OMAP16XX | ALWAYS_ENABLED, .recalc = &followparent_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk rhea2_ck = { @@ -475,8 +475,8 @@ static struct clk rhea2_ck = { .parent = &tc_ck.clk, .flags = CLOCK_IN_OMAP16XX | ALWAYS_ENABLED, .recalc = &followparent_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk lcd_ck_16xx = { @@ -487,8 +487,8 @@ static struct clk lcd_ck_16xx = { .enable_bit = EN_LCDCK, .rate_offset = CKCTL_LCDDIV_OFFSET, .recalc = &omap1_ckctl_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct arm_idlect1_clk lcd_ck_1510 = { @@ -501,8 +501,8 @@ static struct arm_idlect1_clk lcd_ck_1510 = { .enable_bit = EN_LCDCK, .rate_offset = CKCTL_LCDDIV_OFFSET, .recalc = &omap1_ckctl_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }, .idlect_shift = 3, }; @@ -518,8 +518,8 @@ static struct clk uart1_1510 = { .enable_bit = 29, /* Chooses between 12MHz and 48MHz */ .set_rate = &omap1_set_uart_rate, .recalc = &omap1_uart_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct uart_clk uart1_16xx = { @@ -550,8 +550,8 @@ static struct clk uart2_ck = { .enable_bit = 30, /* Chooses between 12MHz and 48MHz */ .set_rate = &omap1_set_uart_rate, .recalc = &omap1_uart_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk uart3_1510 = { @@ -565,8 +565,8 @@ static struct clk uart3_1510 = { .enable_bit = 31, /* Chooses between 12MHz and 48MHz */ .set_rate = &omap1_set_uart_rate, .recalc = &omap1_uart_recalc, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct uart_clk uart3_16xx = { @@ -593,8 +593,8 @@ static struct clk usb_clko = { /* 6 MHz output on W4_USB_CLKO */ RATE_FIXED | ENABLE_REG_32BIT, .enable_reg = (void __iomem *)ULPD_CLOCK_CTRL, .enable_bit = USB_MCLK_EN_BIT, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk usb_hhc_ck1510 = { @@ -605,8 +605,8 @@ static struct clk usb_hhc_ck1510 = { RATE_FIXED | ENABLE_REG_32BIT, .enable_reg = (void __iomem *)MOD_CONF_CTRL_0, .enable_bit = USB_HOST_HHC_UHOST_EN, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk usb_hhc_ck16xx = { @@ -618,8 +618,8 @@ static struct clk usb_hhc_ck16xx = { RATE_FIXED | ENABLE_REG_32BIT, .enable_reg = (void __iomem *)OTG_BASE + 0x08 /* OTG_SYSCON_2 */, .enable_bit = 8 /* UHOST_EN */, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk usb_dc_ck = { @@ -629,8 +629,8 @@ static struct clk usb_dc_ck = { .flags = CLOCK_IN_OMAP16XX | RATE_FIXED, .enable_reg = (void __iomem *)SOFT_REQ_REG, .enable_bit = 4, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk mclk_1510 = { @@ -638,8 +638,8 @@ static struct clk mclk_1510 = { /* Direct from ULPD, no parent. May be enabled by ext hardware. */ .rate = 12000000, .flags = CLOCK_IN_OMAP1510 | RATE_FIXED, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk mclk_16xx = { @@ -651,8 +651,8 @@ static struct clk mclk_16xx = { .set_rate = &omap1_set_ext_clk_rate, .round_rate = &omap1_round_ext_clk_rate, .init = &omap1_init_ext_clk, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk bclk_1510 = { @@ -660,8 +660,8 @@ static struct clk bclk_1510 = { /* Direct from ULPD, no parent. May be enabled by ext hardware. */ .rate = 12000000, .flags = CLOCK_IN_OMAP1510 | RATE_FIXED, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk bclk_16xx = { @@ -673,8 +673,8 @@ static struct clk bclk_16xx = { .set_rate = &omap1_set_ext_clk_rate, .round_rate = &omap1_round_ext_clk_rate, .init = &omap1_init_ext_clk, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk mmc1_ck = { @@ -686,8 +686,8 @@ static struct clk mmc1_ck = { RATE_FIXED | ENABLE_REG_32BIT | CLOCK_NO_IDLE_PARENT, .enable_reg = (void __iomem *)MOD_CONF_CTRL_0, .enable_bit = 23, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk mmc2_ck = { @@ -699,8 +699,8 @@ static struct clk mmc2_ck = { RATE_FIXED | ENABLE_REG_32BIT | CLOCK_NO_IDLE_PARENT, .enable_reg = (void __iomem *)MOD_CONF_CTRL_0, .enable_bit = 20, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk virtual_ck_mpu = { @@ -711,8 +711,8 @@ static struct clk virtual_ck_mpu = { .recalc = &followparent_recalc, .set_rate = &omap1_select_table_rate, .round_rate = &omap1_round_to_table_rate, - .enable = &omap1_clk_enable, - .disable = &omap1_clk_disable, + .enable = &omap1_clk_enable_generic, + .disable = &omap1_clk_disable_generic, }; static struct clk * onchip_clks[] = { -- cgit v1.2.3 From fde0fd49419177ddd69254b8d532edde9ce6a543 Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Tue, 17 Jan 2006 15:31:18 -0800 Subject: ARM: OMAP: 3/4 Fix clock framework to use clk_enable/disable for omap2 This patch fixes OMAP clock framework to use clk_enable/disable instead of clk_use/unuse as specified in include/linux/clk.h. Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/clock.c | 32 +++++++++++++++----------------- arch/arm/mach-omap2/clock.h | 12 ++++++------ 2 files changed, 21 insertions(+), 23 deletions(-) diff --git a/arch/arm/mach-omap2/clock.c b/arch/arm/mach-omap2/clock.c index 5407b954915..180f675c906 100644 --- a/arch/arm/mach-omap2/clock.c +++ b/arch/arm/mach-omap2/clock.c @@ -111,7 +111,7 @@ static void omap2_clk_fixed_enable(struct clk *clk) /* Enables clock without considering parent dependencies or use count * REVISIT: Maybe change this to use clk->enable like on omap1? */ -static int omap2_clk_enable(struct clk * clk) +static int _omap2_clk_enable(struct clk * clk) { u32 regval32; @@ -150,7 +150,7 @@ static void omap2_clk_fixed_disable(struct clk *clk) } /* Disables clock without considering parent dependencies or use count */ -static void omap2_clk_disable(struct clk *clk) +static void _omap2_clk_disable(struct clk *clk) { u32 regval32; @@ -167,23 +167,23 @@ static void omap2_clk_disable(struct clk *clk) __raw_writel(regval32, clk->enable_reg); } -static int omap2_clk_use(struct clk *clk) +static int omap2_clk_enable(struct clk *clk) { int ret = 0; if (clk->usecount++ == 0) { if (likely((u32)clk->parent)) - ret = omap2_clk_use(clk->parent); + ret = omap2_clk_enable(clk->parent); if (unlikely(ret != 0)) { clk->usecount--; return ret; } - ret = omap2_clk_enable(clk); + ret = _omap2_clk_enable(clk); if (unlikely(ret != 0) && clk->parent) { - omap2_clk_unuse(clk->parent); + omap2_clk_disable(clk->parent); clk->usecount--; } } @@ -191,12 +191,12 @@ static int omap2_clk_use(struct clk *clk) return ret; } -static void omap2_clk_unuse(struct clk *clk) +static void omap2_clk_disable(struct clk *clk) { if (clk->usecount > 0 && !(--clk->usecount)) { - omap2_clk_disable(clk); + _omap2_clk_disable(clk); if (likely((u32)clk->parent)) - omap2_clk_unuse(clk->parent); + omap2_clk_disable(clk->parent); } } @@ -873,7 +873,7 @@ static int omap2_clk_set_parent(struct clk *clk, struct clk *new_parent) reg = (void __iomem *)src_sel; if (clk->usecount > 0) - omap2_clk_disable(clk); + _omap2_clk_disable(clk); /* Set new source value (previous dividers if any in effect) */ reg_val = __raw_readl(reg) & ~(field_mask << src_off); @@ -884,7 +884,7 @@ static int omap2_clk_set_parent(struct clk *clk, struct clk *new_parent) __raw_writel(0x1, (void __iomem *)&PRCM_CLKCFG_CTRL); if (clk->usecount > 0) - omap2_clk_enable(clk); + _omap2_clk_enable(clk); clk->parent = new_parent; @@ -999,8 +999,6 @@ static int omap2_select_table_rate(struct clk * clk, unsigned long rate) static struct clk_functions omap2_clk_functions = { .clk_enable = omap2_clk_enable, .clk_disable = omap2_clk_disable, - .clk_use = omap2_clk_use, - .clk_unuse = omap2_clk_unuse, .clk_round_rate = omap2_clk_round_rate, .clk_set_rate = omap2_clk_set_rate, .clk_set_parent = omap2_clk_set_parent, @@ -1045,7 +1043,7 @@ static void __init omap2_disable_unused_clocks(void) continue; printk(KERN_INFO "Disabling unused clock \"%s\"\n", ck->name); - omap2_clk_disable(ck); + _omap2_clk_disable(ck); } } late_initcall(omap2_disable_unused_clocks); @@ -1120,10 +1118,10 @@ int __init omap2_clk_init(void) * Only enable those clocks we will need, let the drivers * enable other clocks as necessary */ - clk_use(&sync_32k_ick); - clk_use(&omapctrl_ick); + clk_enable(&sync_32k_ick); + clk_enable(&omapctrl_ick); if (cpu_is_omap2430()) - clk_use(&sdrc_ick); + clk_enable(&sdrc_ick); return 0; } diff --git a/arch/arm/mach-omap2/clock.h b/arch/arm/mach-omap2/clock.h index 4aeab5591bd..6cab20b1d3c 100644 --- a/arch/arm/mach-omap2/clock.h +++ b/arch/arm/mach-omap2/clock.h @@ -24,7 +24,7 @@ static void omap2_propagate_rate(struct clk * clk); static void omap2_mpu_recalc(struct clk * clk); static int omap2_select_table_rate(struct clk * clk, unsigned long rate); static long omap2_round_to_table_rate(struct clk * clk, unsigned long rate); -static void omap2_clk_unuse(struct clk *clk); +static void omap2_clk_disable(struct clk *clk); static void omap2_sys_clk_recalc(struct clk * clk); static u32 omap2_clksel_to_divisor(u32 div_sel, u32 field_val); static u32 omap2_clksel_get_divisor(struct clk *clk); @@ -859,7 +859,7 @@ static struct clk core_l3_ck = { /* Used for ick and fck, interconnect */ static struct clk usb_l4_ick = { /* FS-USB interface clock */ .name = "usb_l4_ick", - .parent = &core_ck, + .parent = &core_l3_ck, .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | RATE_CKCTL | CM_CORE_SEL1 | DELAYED_APP | CONFIG_PARTICIPANT, @@ -1045,7 +1045,7 @@ static struct clk gpt1_ick = { .name = "gpt1_ick", .parent = &l4_ck, .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, - .enable_reg = (void __iomem *)&CM_ICLKEN_WKUP, /* Bit4 */ + .enable_reg = (void __iomem *)&CM_ICLKEN_WKUP, /* Bit0 */ .enable_bit = 0, .recalc = &omap2_followparent_recalc, }; @@ -1055,7 +1055,7 @@ static struct clk gpt1_fck = { .parent = &func_32k_ck, .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | CM_WKUP_SEL1, - .enable_reg = (void __iomem *)&CM_FCLKEN_WKUP, + .enable_reg = (void __iomem *)&CM_FCLKEN_WKUP, /* Bit0 */ .enable_bit = 0, .src_offset = 0, .recalc = &omap2_followparent_recalc, @@ -1065,7 +1065,7 @@ static struct clk gpt2_ick = { .name = "gpt2_ick", .parent = &l4_ck, .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, - .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, /* bit4 */ + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, /* Bit4 */ .enable_bit = 0, .recalc = &omap2_followparent_recalc, }; @@ -1839,7 +1839,7 @@ static struct clk usb_fck = { static struct clk usbhs_ick = { .name = "usbhs_ick", - .parent = &l4_ck, + .parent = &core_l3_ck, .flags = CLOCK_IN_OMAP243X, .enable_reg = (void __iomem *)&CM_ICLKEN2_CORE, .enable_bit = 6, -- cgit v1.2.3 From 30ff720b40ba64b0e9c8974673b95970e68503ac Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Tue, 17 Jan 2006 15:33:51 -0800 Subject: ARM: OMAP: 4/4 Fix clock framework to use clk_enable/disable misc This patch fixes OMAP clock framework to use clk_enable/disable instead of clk_use/unuse as specified in include/linux/clk.h. Signed-off-by: Tony Lindgren --- arch/arm/mach-omap1/serial.c | 6 +++--- arch/arm/mach-omap2/serial.c | 12 ++++++------ arch/arm/mach-omap2/timer-gp.c | 2 +- arch/arm/plat-omap/gpio.c | 6 +++--- arch/arm/plat-omap/mcbsp.c | 12 ++++++------ arch/arm/plat-omap/ocpi.c | 4 ++-- 6 files changed, 21 insertions(+), 21 deletions(-) diff --git a/arch/arm/mach-omap1/serial.c b/arch/arm/mach-omap1/serial.c index 7a68f098a02..e924e0c6a4c 100644 --- a/arch/arm/mach-omap1/serial.c +++ b/arch/arm/mach-omap1/serial.c @@ -146,7 +146,7 @@ void __init omap_serial_init(void) if (IS_ERR(uart1_ck)) printk("Could not get uart1_ck\n"); else { - clk_use(uart1_ck); + clk_enable(uart1_ck); if (cpu_is_omap1510()) clk_set_rate(uart1_ck, 12000000); } @@ -166,7 +166,7 @@ void __init omap_serial_init(void) if (IS_ERR(uart2_ck)) printk("Could not get uart2_ck\n"); else { - clk_use(uart2_ck); + clk_enable(uart2_ck); if (cpu_is_omap1510()) clk_set_rate(uart2_ck, 12000000); else @@ -188,7 +188,7 @@ void __init omap_serial_init(void) if (IS_ERR(uart3_ck)) printk("Could not get uart3_ck\n"); else { - clk_use(uart3_ck); + clk_enable(uart3_ck); if (cpu_is_omap1510()) clk_set_rate(uart3_ck, 12000000); } diff --git a/arch/arm/mach-omap2/serial.c b/arch/arm/mach-omap2/serial.c index e1bd46a96e1..24dd374224a 100644 --- a/arch/arm/mach-omap2/serial.c +++ b/arch/arm/mach-omap2/serial.c @@ -119,14 +119,14 @@ void __init omap_serial_init() if (IS_ERR(uart1_ick)) printk("Could not get uart1_ick\n"); else { - clk_use(uart1_ick); + clk_enable(uart1_ick); } uart1_fck = clk_get(NULL, "uart1_fck"); if (IS_ERR(uart1_fck)) printk("Could not get uart1_fck\n"); else { - clk_use(uart1_fck); + clk_enable(uart1_fck); } break; case 1: @@ -134,14 +134,14 @@ void __init omap_serial_init() if (IS_ERR(uart2_ick)) printk("Could not get uart2_ick\n"); else { - clk_use(uart2_ick); + clk_enable(uart2_ick); } uart2_fck = clk_get(NULL, "uart2_fck"); if (IS_ERR(uart2_fck)) printk("Could not get uart2_fck\n"); else { - clk_use(uart2_fck); + clk_enable(uart2_fck); } break; case 2: @@ -149,14 +149,14 @@ void __init omap_serial_init() if (IS_ERR(uart3_ick)) printk("Could not get uart3_ick\n"); else { - clk_use(uart3_ick); + clk_enable(uart3_ick); } uart3_fck = clk_get(NULL, "uart3_fck"); if (IS_ERR(uart3_fck)) printk("Could not get uart3_fck\n"); else { - clk_use(uart3_fck); + clk_enable(uart3_fck); } break; } diff --git a/arch/arm/mach-omap2/timer-gp.c b/arch/arm/mach-omap2/timer-gp.c index 23d36b1c40f..1d2f5ac2f69 100644 --- a/arch/arm/mach-omap2/timer-gp.c +++ b/arch/arm/mach-omap2/timer-gp.c @@ -104,7 +104,7 @@ static void __init omap2_gp_timer_init(void) if (IS_ERR(sys_ck)) printk(KERN_ERR "Could not get sys_ck\n"); else { - clk_use(sys_ck); + clk_enable(sys_ck); tick_period = clk_get_rate(sys_ck) / 100; clk_put(sys_ck); } diff --git a/arch/arm/plat-omap/gpio.c b/arch/arm/plat-omap/gpio.c index ca3681a824a..b4d5b9e4bfc 100644 --- a/arch/arm/plat-omap/gpio.c +++ b/arch/arm/plat-omap/gpio.c @@ -853,19 +853,19 @@ static int __init _omap_gpio_init(void) if (IS_ERR(gpio_ick)) printk("Could not get arm_gpio_ck\n"); else - clk_use(gpio_ick); + clk_enable(gpio_ick); } if (cpu_is_omap24xx()) { gpio_ick = clk_get(NULL, "gpios_ick"); if (IS_ERR(gpio_ick)) printk("Could not get gpios_ick\n"); else - clk_use(gpio_ick); + clk_enable(gpio_ick); gpio_fck = clk_get(NULL, "gpios_fck"); if (IS_ERR(gpio_ick)) printk("Could not get gpios_fck\n"); else - clk_use(gpio_fck); + clk_enable(gpio_fck); } #ifdef CONFIG_ARCH_OMAP15XX diff --git a/arch/arm/plat-omap/mcbsp.c b/arch/arm/plat-omap/mcbsp.c index be0e0f32a59..1cd2cace7e1 100644 --- a/arch/arm/plat-omap/mcbsp.c +++ b/arch/arm/plat-omap/mcbsp.c @@ -190,11 +190,11 @@ static int omap_mcbsp_check(unsigned int id) static void omap_mcbsp_dsp_request(void) { if (cpu_is_omap1510() || cpu_is_omap16xx()) { - clk_use(mcbsp_dsp_ck); - clk_use(mcbsp_api_ck); + clk_enable(mcbsp_dsp_ck); + clk_enable(mcbsp_api_ck); /* enable 12MHz clock to mcbsp 1 & 3 */ - clk_use(mcbsp_dspxor_ck); + clk_enable(mcbsp_dspxor_ck); /* * DSP external peripheral reset @@ -208,9 +208,9 @@ static void omap_mcbsp_dsp_request(void) static void omap_mcbsp_dsp_free(void) { if (cpu_is_omap1510() || cpu_is_omap16xx()) { - clk_unuse(mcbsp_dspxor_ck); - clk_unuse(mcbsp_dsp_ck); - clk_unuse(mcbsp_api_ck); + clk_disable(mcbsp_dspxor_ck); + clk_disable(mcbsp_dsp_ck); + clk_disable(mcbsp_api_ck); } } diff --git a/arch/arm/plat-omap/ocpi.c b/arch/arm/plat-omap/ocpi.c index e40fcc8b43d..5cc6775c789 100644 --- a/arch/arm/plat-omap/ocpi.c +++ b/arch/arm/plat-omap/ocpi.c @@ -88,7 +88,7 @@ static int __init omap_ocpi_init(void) if (IS_ERR(ocpi_ck)) return PTR_ERR(ocpi_ck); - clk_use(ocpi_ck); + clk_enable(ocpi_ck); ocpi_enable(); printk("OMAP OCPI interconnect driver loaded\n"); @@ -102,7 +102,7 @@ static void __exit omap_ocpi_exit(void) if (!cpu_is_omap16xx()) return; - clk_unuse(ocpi_ck); + clk_disable(ocpi_ck); clk_put(ocpi_ck); } -- cgit v1.2.3 From 5f7c3ff6a2e227418d363069ff89cf9d7f01fbc1 Mon Sep 17 00:00:00 2001 From: Jon Maloy Date: Fri, 13 Jan 2006 10:45:44 +0000 Subject: [TIPC] Minor changes to #includes Signed-off-by: Jon Maloy --- net/tipc/bearer.h | 2 +- net/tipc/config.h | 3 +-- net/tipc/core.h | 5 +++++ net/tipc/discover.h | 2 +- net/tipc/msg.h | 2 +- net/tipc/port.h | 2 +- 6 files changed, 10 insertions(+), 6 deletions(-) diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h index 21e63d3f018..0c40cc2b43c 100644 --- a/net/tipc/bearer.h +++ b/net/tipc/bearer.h @@ -37,7 +37,7 @@ #ifndef _TIPC_BEARER_H #define _TIPC_BEARER_H -#include +#include "core.h" #include "bcast.h" #define MAX_BEARERS 8 diff --git a/net/tipc/config.h b/net/tipc/config.h index 646377d4045..e74d94f753c 100644 --- a/net/tipc/config.h +++ b/net/tipc/config.h @@ -39,8 +39,7 @@ /* ---------------------------------------------------------------------- */ -#include -#include +#include "core.h" #include "link.h" struct sk_buff *cfg_reply_alloc(int payload_size); diff --git a/net/tipc/core.h b/net/tipc/core.h index b69b60b2cc8..1a34cc95b9e 100644 --- a/net/tipc/core.h +++ b/net/tipc/core.h @@ -37,6 +37,11 @@ #ifndef _TIPC_CORE_H #define _TIPC_CORE_H +#include +#include +#include +#include +#include #include #include #include diff --git a/net/tipc/discover.h b/net/tipc/discover.h index 2a6114d9162..f4acb360d6c 100644 --- a/net/tipc/discover.h +++ b/net/tipc/discover.h @@ -37,7 +37,7 @@ #ifndef _TIPC_DISCOVER_H #define _TIPC_DISCOVER_H -#include +#include "core.h" struct link_req; diff --git a/net/tipc/msg.h b/net/tipc/msg.h index 662c81862a0..6574aab11fa 100644 --- a/net/tipc/msg.h +++ b/net/tipc/msg.h @@ -37,7 +37,7 @@ #ifndef _TIPC_MSG_H #define _TIPC_MSG_H -#include +#include "core.h" #define TIPC_VERSION 2 #define DATA_LOW TIPC_LOW_IMPORTANCE diff --git a/net/tipc/port.h b/net/tipc/port.h index e829a99d3b7..f4a8c2be3fa 100644 --- a/net/tipc/port.h +++ b/net/tipc/port.h @@ -37,7 +37,7 @@ #ifndef _TIPC_PORT_H #define _TIPC_PORT_H -#include +#include "core.h" #include "ref.h" #include "net.h" #include "msg.h" -- cgit v1.2.3 From 16cb4b333c9e7a00ce3b1d74ec0c9b4c2e956910 Mon Sep 17 00:00:00 2001 From: Per Liden Date: Fri, 13 Jan 2006 22:22:22 +0100 Subject: [TIPC] Updated link priority macros Added macros for min/default/max link priority in tipc_config.h. Also renamed TIPC_NUM_LINK_PRI to TIPC_MEDIA_LINK_PRI since that is a more accurate description of what it is used for. Signed-off-by: Per Liden --- include/linux/tipc_config.h | 7 +++++-- net/tipc/bcast.c | 4 ++-- net/tipc/bearer.c | 19 +++++++++++++------ net/tipc/eth_media.c | 6 +++--- net/tipc/link.c | 3 ++- 5 files changed, 25 insertions(+), 14 deletions(-) diff --git a/include/linux/tipc_config.h b/include/linux/tipc_config.h index a52c8c64a5a..33a653913d9 100644 --- a/include/linux/tipc_config.h +++ b/include/linux/tipc_config.h @@ -168,10 +168,13 @@ #define TIPC_MAX_LINK_NAME 60 /* format = Z.C.N:interface-Z.C.N:interface */ /* - * Link priority limits (range from 0 to # priorities - 1) + * Link priority limits (min, default, max, media default) */ -#define TIPC_NUM_LINK_PRI 32 +#define TIPC_MIN_LINK_PRI 0 +#define TIPC_DEF_LINK_PRI 10 +#define TIPC_MAX_LINK_PRI 31 +#define TIPC_MEDIA_LINK_PRI (TIPC_MAX_LINK_PRI + 1) /* * Link tolerance limits (min, default, max), in ms diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c index 9713d622efb..af9743a52d6 100644 --- a/net/tipc/bcast.c +++ b/net/tipc/bcast.c @@ -82,7 +82,7 @@ struct bcbearer { struct bearer bearer; struct media media; struct bcbearer_pair bpairs[MAX_BEARERS]; - struct bcbearer_pair bpairs_temp[TIPC_NUM_LINK_PRI]; + struct bcbearer_pair bpairs_temp[TIPC_MAX_LINK_PRI + 1]; }; /** @@ -630,7 +630,7 @@ void bcbearer_sort(void) bp_curr = bcbearer->bpairs; memset(bcbearer->bpairs, 0, sizeof(bcbearer->bpairs)); - for (pri = (TIPC_NUM_LINK_PRI - 1); pri >= 0; pri--) { + for (pri = TIPC_MAX_LINK_PRI; pri >= 0; pri--) { if (!bp_temp[pri].primary) continue; diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 3dd19fdc5a2..02b6cf6ab7a 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -119,7 +119,8 @@ int tipc_register_media(u32 media_type, warn("Media registration error: no broadcast address supplied\n"); goto exit; } - if (bearer_priority >= TIPC_NUM_LINK_PRI) { + if ((bearer_priority < TIPC_MIN_LINK_PRI) && + (bearer_priority > TIPC_MAX_LINK_PRI)) { warn("Media registration error: priority %u\n", bearer_priority); goto exit; } @@ -476,10 +477,15 @@ int tipc_enable_bearer(const char *name, u32 bcast_scope, u32 priority) if (tipc_mode != TIPC_NET_MODE) return -ENOPROTOOPT; + if (!bearer_name_validate(name, &b_name) || !addr_domain_valid(bcast_scope) || - !in_scope(bcast_scope, tipc_own_addr) || - (priority > TIPC_NUM_LINK_PRI)) + !in_scope(bcast_scope, tipc_own_addr)) + return -EINVAL; + + if ((priority < TIPC_MIN_LINK_PRI || + priority > TIPC_MAX_LINK_PRI) && + (priority != TIPC_MEDIA_LINK_PRI)) return -EINVAL; write_lock_bh(&net_lock); @@ -491,7 +497,8 @@ int tipc_enable_bearer(const char *name, u32 bcast_scope, u32 priority) warn("No media <%s>\n", b_name.media_name); goto failed; } - if (priority == TIPC_NUM_LINK_PRI) + + if (priority == TIPC_MEDIA_LINK_PRI) priority = m_ptr->priority; restart: @@ -547,8 +554,8 @@ restart: } b_ptr->publ.lock = SPIN_LOCK_UNLOCKED; write_unlock_bh(&net_lock); - info("Enabled bearer <%s>, discovery domain %s\n", - name, addr_string_fill(addr_string, bcast_scope)); + info("Enabled bearer <%s>, discovery domain %s, priority %u\n", + name, addr_string_fill(addr_string, bcast_scope), priority); return 0; failed: write_unlock_bh(&net_lock); diff --git a/net/tipc/eth_media.c b/net/tipc/eth_media.c index 34d0462db3a..2ab6c16280e 100644 --- a/net/tipc/eth_media.c +++ b/net/tipc/eth_media.c @@ -42,9 +42,9 @@ #define MAX_ETH_BEARERS 2 #define TIPC_PROTOCOL 0x88ca -#define ETH_LINK_PRIORITY 10 +#define ETH_LINK_PRIORITY TIPC_DEF_LINK_PRI #define ETH_LINK_TOLERANCE TIPC_DEF_LINK_TOL - +#define ETH_LINK_WINDOW TIPC_DEF_LINK_WIN /** * struct eth_bearer - Ethernet bearer data structure @@ -260,7 +260,7 @@ int eth_media_start(void) res = tipc_register_media(TIPC_MEDIA_TYPE_ETH, "eth", enable_bearer, disable_bearer, send_msg, eth_addr2str, &bcast_addr, ETH_LINK_PRIORITY, - ETH_LINK_TOLERANCE, TIPC_DEF_LINK_WIN); + ETH_LINK_TOLERANCE, ETH_LINK_WINDOW); if (res) return res; diff --git a/net/tipc/link.c b/net/tipc/link.c index 7265f4be476..d1e1ae66464 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -2812,7 +2812,8 @@ struct sk_buff *link_cmd_config(const void *req_tlv_area, int req_tlv_space, } break; case TIPC_CMD_SET_LINK_PRI: - if (new_value < TIPC_NUM_LINK_PRI) { + if ((new_value >= TIPC_MIN_LINK_PRI) && + (new_value <= TIPC_MAX_LINK_PRI)) { l_ptr->priority = new_value; link_send_proto_msg(l_ptr, STATE_MSG, 0, 0, 0, new_value, 0); -- cgit v1.2.3 From 7c2b2aaee45c365596a83bcf8758b1f222e0eca4 Mon Sep 17 00:00:00 2001 From: Per Liden Date: Sat, 14 Jan 2006 12:42:21 +0100 Subject: [TIPC] Provide real email addresses in MAINTAINERS Signed-off-by: Per Liden --- MAINTAINERS | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 6d1b048c62a..a731fbfefb0 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2534,11 +2534,11 @@ S: Maintained TIPC NETWORK LAYER P: Per Liden -M: per.liden@nospam.ericsson.com +M: per.liden@ericsson.com P: Jon Maloy -M: jon.maloy@nospam.ericsson.com +M: jon.maloy@ericsson.com P: Allan Stephens -M: allan.stephens@nospam.windriver.com +M: allan.stephens@windriver.com L: tipc-discussion@lists.sourceforge.net W: http://tipc.sourceforge.net/ W: http://tipc.cslab.ericsson.net/ -- cgit v1.2.3 From 33a9c4da5ab16192ef1e961d4c4e45c18031cd67 Mon Sep 17 00:00:00 2001 From: Per Liden Date: Mon, 16 Jan 2006 11:42:12 +0100 Subject: [TIPC] Move ethernet protocol id to linux/if_ether.h Signed-off-by: Per Liden --- include/linux/if_ether.h | 1 + net/tipc/eth_media.c | 5 ++--- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/include/linux/if_ether.h b/include/linux/if_ether.h index fe26d431de8..7a92c1ce145 100644 --- a/include/linux/if_ether.h +++ b/include/linux/if_ether.h @@ -72,6 +72,7 @@ * over Ethernet */ #define ETH_P_AOE 0x88A2 /* ATA over Ethernet */ +#define ETH_P_TIPC 0x88CA /* TIPC */ /* * Non DIX types. Won't clash for 1500 types. diff --git a/net/tipc/eth_media.c b/net/tipc/eth_media.c index 2ab6c16280e..64a880b5935 100644 --- a/net/tipc/eth_media.c +++ b/net/tipc/eth_media.c @@ -41,7 +41,6 @@ #include #define MAX_ETH_BEARERS 2 -#define TIPC_PROTOCOL 0x88ca #define ETH_LINK_PRIORITY TIPC_DEF_LINK_PRI #define ETH_LINK_TOLERANCE TIPC_DEF_LINK_TOL #define ETH_LINK_WINDOW TIPC_DEF_LINK_WIN @@ -78,7 +77,7 @@ static int send_msg(struct sk_buff *buf, struct tipc_bearer *tb_ptr, clone->nh.raw = clone->data; dev = ((struct eth_bearer *)(tb_ptr->usr_handle))->dev; clone->dev = dev; - dev->hard_header(clone, dev, TIPC_PROTOCOL, + dev->hard_header(clone, dev, ETH_P_TIPC, &dest->dev_addr.eth_addr, dev->dev_addr, clone->len); dev_queue_xmit(clone); @@ -141,7 +140,7 @@ static int enable_bearer(struct tipc_bearer *tb_ptr) return -EDQUOT; if (!eb_ptr->dev) { eb_ptr->dev = dev; - eb_ptr->tipc_packet_type.type = __constant_htons(TIPC_PROTOCOL); + eb_ptr->tipc_packet_type.type = __constant_htons(ETH_P_TIPC); eb_ptr->tipc_packet_type.dev = dev; eb_ptr->tipc_packet_type.func = recv_msg; eb_ptr->tipc_packet_type.af_packet_priv = eb_ptr; -- cgit v1.2.3 From 50f9bcddf8be147678c2d4ef8ac5279222d0ae3a Mon Sep 17 00:00:00 2001 From: Per Liden Date: Mon, 16 Jan 2006 12:42:35 +0100 Subject: [TIPC] Remove unused #includes Signed-off-by: Per Liden --- net/tipc/core.c | 1 - net/tipc/eth_media.c | 1 - net/tipc/socket.c | 2 -- 3 files changed, 4 deletions(-) diff --git a/net/tipc/core.c b/net/tipc/core.c index e83ac06e31b..9a1ab178b44 100644 --- a/net/tipc/core.c +++ b/net/tipc/core.c @@ -37,7 +37,6 @@ #include #include #include -#include #include #include "core.h" diff --git a/net/tipc/eth_media.c b/net/tipc/eth_media.c index 64a880b5935..cb507c4b7e6 100644 --- a/net/tipc/eth_media.c +++ b/net/tipc/eth_media.c @@ -38,7 +38,6 @@ #include #include #include -#include #define MAX_ETH_BEARERS 2 #define ETH_LINK_PRIORITY TIPC_DEF_LINK_PRI diff --git a/net/tipc/socket.c b/net/tipc/socket.c index d21f8c0cd25..107f654db0b 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -42,9 +42,7 @@ #include #include #include -#include #include -#include #include #include #include -- cgit v1.2.3 From c11ac3f236d73dd00c21ab4f06166606dea71ef3 Mon Sep 17 00:00:00 2001 From: Per Liden Date: Mon, 16 Jan 2006 16:32:18 +0100 Subject: [TIPC] Add help text for TIPC configuration option Signed-off-by: Per Liden --- net/tipc/Kconfig | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/net/tipc/Kconfig b/net/tipc/Kconfig index 05ab18e62de..3891cc00087 100644 --- a/net/tipc/Kconfig +++ b/net/tipc/Kconfig @@ -8,7 +8,12 @@ menu "TIPC Configuration (EXPERIMENTAL)" config TIPC tristate "The TIPC Protocol (EXPERIMENTAL)" ---help--- - TBD. + The Transparent Inter Process Communication (TIPC) protocol is + specially designed for intra cluster communication. This protocol + originates from Ericsson where it has been used in carrier grade + cluster applications for many years. + + For more information about TIPC, see http://tipc.sourceforge.net. This protocol support is also available as a module ( = code which can be inserted in and removed from the running kernel whenever you -- cgit v1.2.3 From 1e63e681e06d438fdc542d40924a4f155d461bbd Mon Sep 17 00:00:00 2001 From: Per Liden Date: Mon, 16 Jan 2006 16:39:13 +0100 Subject: [TIPC] Group protocols with sub-options in Kconfig This is just a cosmetic change that moves the TIPC configuration entry next to the other protocols that also have sub-options. Makes the the networking options menu look a bit better. Signed-off-by: Per Liden --- net/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/Kconfig b/net/Kconfig index 9296b269d67..bc603d9aea5 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -150,6 +150,7 @@ endif source "net/dccp/Kconfig" source "net/sctp/Kconfig" +source "net/tipc/Kconfig" source "net/atm/Kconfig" source "net/bridge/Kconfig" source "net/8021q/Kconfig" @@ -159,7 +160,6 @@ source "net/ipx/Kconfig" source "drivers/net/appletalk/Kconfig" source "net/x25/Kconfig" source "net/lapb/Kconfig" -source "net/tipc/Kconfig" config NET_DIVERT bool "Frame Diverter (EXPERIMENTAL)" -- cgit v1.2.3 From 4323add67792ced172d0d93b8b2e6187023115f1 Mon Sep 17 00:00:00 2001 From: Per Liden Date: Wed, 18 Jan 2006 00:38:21 +0100 Subject: [TIPC] Avoid polluting the global namespace This patch adds a tipc_ prefix to all externally visible symbols. Signed-off-by: Per Liden --- net/tipc/addr.c | 10 +- net/tipc/addr.h | 4 +- net/tipc/bcast.c | 162 ++++++------- net/tipc/bcast.h | 44 ++-- net/tipc/bearer.c | 170 +++++++------- net/tipc/bearer.h | 38 ++-- net/tipc/cluster.c | 154 ++++++------- net/tipc/cluster.h | 40 ++-- net/tipc/config.c | 218 +++++++++--------- net/tipc/config.h | 38 ++-- net/tipc/core.c | 86 +++---- net/tipc/core.h | 30 +-- net/tipc/dbg.c | 112 ++++----- net/tipc/dbg.h | 18 +- net/tipc/discover.c | 52 ++--- net/tipc/discover.h | 16 +- net/tipc/eth_media.c | 8 +- net/tipc/handler.c | 6 +- net/tipc/link.c | 602 ++++++++++++++++++++++++------------------------- net/tipc/link.h | 73 +++--- net/tipc/msg.c | 19 +- net/tipc/msg.h | 16 +- net/tipc/name_distr.c | 92 ++++---- net/tipc/name_distr.h | 10 +- net/tipc/name_table.c | 206 ++++++++--------- net/tipc/name_table.h | 26 +-- net/tipc/net.c | 126 +++++------ net/tipc/net.h | 20 +- net/tipc/netlink.c | 16 +- net/tipc/node.c | 215 +++++++++--------- net/tipc/node.h | 50 ++-- net/tipc/node_subscr.c | 20 +- net/tipc/node_subscr.h | 6 +- net/tipc/port.c | 274 +++++++++++----------- net/tipc/port.h | 58 ++--- net/tipc/ref.c | 72 +++--- net/tipc/ref.h | 38 ++-- net/tipc/socket.c | 10 +- net/tipc/subscr.c | 58 ++--- net/tipc/subscr.h | 24 +- net/tipc/user_reg.c | 22 +- net/tipc/user_reg.h | 8 +- net/tipc/zone.c | 40 ++-- net/tipc/zone.h | 20 +- 44 files changed, 1654 insertions(+), 1673 deletions(-) diff --git a/net/tipc/addr.c b/net/tipc/addr.c index eca22260c98..0be25e175b9 100644 --- a/net/tipc/addr.c +++ b/net/tipc/addr.c @@ -47,7 +47,7 @@ u32 tipc_get_addr(void) } /** - * addr_domain_valid - validates a network domain address + * tipc_addr_domain_valid - validates a network domain address * * Accepts , , , and <0.0.0>, * where Z, C, and N are non-zero and do not exceed the configured limits. @@ -55,7 +55,7 @@ u32 tipc_get_addr(void) * Returns 1 if domain address is valid, otherwise 0 */ -int addr_domain_valid(u32 addr) +int tipc_addr_domain_valid(u32 addr) { u32 n = tipc_node(addr); u32 c = tipc_cluster(addr); @@ -79,7 +79,7 @@ int addr_domain_valid(u32 addr) } /** - * addr_node_valid - validates a proposed network address for this node + * tipc_addr_node_valid - validates a proposed network address for this node * * Accepts , where Z, C, and N are non-zero and do not exceed * the configured limits. @@ -87,8 +87,8 @@ int addr_domain_valid(u32 addr) * Returns 1 if address can be used, otherwise 0 */ -int addr_node_valid(u32 addr) +int tipc_addr_node_valid(u32 addr) { - return (addr_domain_valid(addr) && tipc_node(addr)); + return (tipc_addr_domain_valid(addr) && tipc_node(addr)); } diff --git a/net/tipc/addr.h b/net/tipc/addr.h index 02ca71783e2..bcfebb3cbbf 100644 --- a/net/tipc/addr.h +++ b/net/tipc/addr.h @@ -122,7 +122,7 @@ static inline char *addr_string_fill(char *string, u32 addr) return string; } -int addr_domain_valid(u32); -int addr_node_valid(u32 addr); +int tipc_addr_domain_valid(u32); +int tipc_addr_node_valid(u32 addr); #endif diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c index af9743a52d6..a7b04f397c1 100644 --- a/net/tipc/bcast.c +++ b/net/tipc/bcast.c @@ -104,7 +104,7 @@ static struct bclink *bclink = NULL; static struct link *bcl = NULL; static spinlock_t bc_lock = SPIN_LOCK_UNLOCKED; -char bc_link_name[] = "multicast-link"; +char tipc_bclink_name[] = "multicast-link"; static inline u32 buf_seqno(struct sk_buff *buf) @@ -178,19 +178,19 @@ static void bclink_retransmit_pkt(u32 after, u32 to) buf = buf->next; } if (buf != NULL) - link_retransmit(bcl, buf, mod(to - after)); + tipc_link_retransmit(bcl, buf, mod(to - after)); spin_unlock_bh(&bc_lock); } /** - * bclink_acknowledge - handle acknowledgement of broadcast packets + * tipc_bclink_acknowledge - handle acknowledgement of broadcast packets * @n_ptr: node that sent acknowledgement info * @acked: broadcast sequence # that has been acknowledged * * Node is locked, bc_lock unlocked. */ -void bclink_acknowledge(struct node *n_ptr, u32 acked) +void tipc_bclink_acknowledge(struct node *n_ptr, u32 acked) { struct sk_buff *crs; struct sk_buff *next; @@ -226,16 +226,16 @@ void bclink_acknowledge(struct node *n_ptr, u32 acked) /* Try resolving broadcast link congestion, if necessary */ if (unlikely(bcl->next_out)) - link_push_queue(bcl); + tipc_link_push_queue(bcl); if (unlikely(released && !list_empty(&bcl->waiting_ports))) - link_wakeup_ports(bcl, 0); + tipc_link_wakeup_ports(bcl, 0); spin_unlock_bh(&bc_lock); } /** * bclink_send_ack - unicast an ACK msg * - * net_lock and node lock set + * tipc_net_lock and node lock set */ static void bclink_send_ack(struct node *n_ptr) @@ -243,13 +243,13 @@ static void bclink_send_ack(struct node *n_ptr) struct link *l_ptr = n_ptr->active_links[n_ptr->addr & 1]; if (l_ptr != NULL) - link_send_proto_msg(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); + tipc_link_send_proto_msg(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); } /** * bclink_send_nack- broadcast a NACK msg * - * net_lock and node lock set + * tipc_net_lock and node lock set */ static void bclink_send_nack(struct node *n_ptr) @@ -271,11 +271,11 @@ static void bclink_send_nack(struct node *n_ptr) msg_set_bcgap_to(msg, n_ptr->bclink.gap_to); msg_set_bcast_tag(msg, tipc_own_tag); - if (bearer_send(&bcbearer->bearer, buf, 0)) { + if (tipc_bearer_send(&bcbearer->bearer, buf, 0)) { bcl->stats.sent_nacks++; buf_discard(buf); } else { - bearer_schedule(bcl->b_ptr, bcl); + tipc_bearer_schedule(bcl->b_ptr, bcl); bcl->proto_msg_queue = buf; bcl->stats.bearer_congs++; } @@ -291,12 +291,12 @@ static void bclink_send_nack(struct node *n_ptr) } /** - * bclink_check_gap - send a NACK if a sequence gap exists + * tipc_bclink_check_gap - send a NACK if a sequence gap exists * - * net_lock and node lock set + * tipc_net_lock and node lock set */ -void bclink_check_gap(struct node *n_ptr, u32 last_sent) +void tipc_bclink_check_gap(struct node *n_ptr, u32 last_sent) { if (!n_ptr->bclink.supported || less_eq(last_sent, mod(n_ptr->bclink.last_in))) @@ -309,19 +309,19 @@ void bclink_check_gap(struct node *n_ptr, u32 last_sent) } /** - * bclink_peek_nack - process a NACK msg meant for another node + * tipc_bclink_peek_nack - process a NACK msg meant for another node * - * Only net_lock set. + * Only tipc_net_lock set. */ -void bclink_peek_nack(u32 dest, u32 sender_tag, u32 gap_after, u32 gap_to) +void tipc_bclink_peek_nack(u32 dest, u32 sender_tag, u32 gap_after, u32 gap_to) { - struct node *n_ptr = node_find(dest); + struct node *n_ptr = tipc_node_find(dest); u32 my_after, my_to; - if (unlikely(!n_ptr || !node_is_up(n_ptr))) + if (unlikely(!n_ptr || !tipc_node_is_up(n_ptr))) return; - node_lock(n_ptr); + tipc_node_lock(n_ptr); /* * Modify gap to suppress unnecessary NACKs from this node */ @@ -364,20 +364,20 @@ void bclink_peek_nack(u32 dest, u32 sender_tag, u32 gap_after, u32 gap_to) bclink_set_gap(n_ptr); } } - node_unlock(n_ptr); + tipc_node_unlock(n_ptr); } /** - * bclink_send_msg - broadcast a packet to all nodes in cluster + * tipc_bclink_send_msg - broadcast a packet to all nodes in cluster */ -int bclink_send_msg(struct sk_buff *buf) +int tipc_bclink_send_msg(struct sk_buff *buf) { int res; spin_lock_bh(&bc_lock); - res = link_send_buf(bcl, buf); + res = tipc_link_send_buf(bcl, buf); if (unlikely(res == -ELINKCONG)) buf_discard(buf); else @@ -393,22 +393,22 @@ int bclink_send_msg(struct sk_buff *buf) } /** - * bclink_recv_pkt - receive a broadcast packet, and deliver upwards + * tipc_bclink_recv_pkt - receive a broadcast packet, and deliver upwards * - * net_lock is read_locked, no other locks set + * tipc_net_lock is read_locked, no other locks set */ -void bclink_recv_pkt(struct sk_buff *buf) +void tipc_bclink_recv_pkt(struct sk_buff *buf) { struct tipc_msg *msg = buf_msg(buf); - struct node* node = node_find(msg_prevnode(msg)); + struct node* node = tipc_node_find(msg_prevnode(msg)); u32 next_in; u32 seqno; struct sk_buff *deferred; msg_dbg(msg, "bclink.supported || + if (unlikely(!node || !tipc_node_is_up(node) || !node->bclink.supported || (msg_mc_netid(msg) != tipc_net_id))) { buf_discard(buf); return; @@ -417,14 +417,14 @@ void bclink_recv_pkt(struct sk_buff *buf) if (unlikely(msg_user(msg) == BCAST_PROTOCOL)) { msg_dbg(msg, "stats.recv_nacks++; bclink_retransmit_pkt(msg_bcgap_after(msg), msg_bcgap_to(msg)); } else { - bclink_peek_nack(msg_destnode(msg), + tipc_bclink_peek_nack(msg_destnode(msg), msg_bcast_tag(msg), msg_bcgap_after(msg), msg_bcgap_to(msg)); @@ -433,7 +433,7 @@ void bclink_recv_pkt(struct sk_buff *buf) return; } - node_lock(node); + tipc_node_lock(node); receive: deferred = node->bclink.deferred_head; next_in = mod(node->bclink.last_in + 1); @@ -448,26 +448,26 @@ receive: bcl->stats.sent_acks++; } if (likely(msg_isdata(msg))) { - node_unlock(node); - port_recv_mcast(buf, NULL); + tipc_node_unlock(node); + tipc_port_recv_mcast(buf, NULL); } else if (msg_user(msg) == MSG_BUNDLER) { bcl->stats.recv_bundles++; bcl->stats.recv_bundled += msg_msgcnt(msg); - node_unlock(node); - link_recv_bundle(buf); + tipc_node_unlock(node); + tipc_link_recv_bundle(buf); } else if (msg_user(msg) == MSG_FRAGMENTER) { bcl->stats.recv_fragments++; - if (link_recv_fragment(&node->bclink.defragm, - &buf, &msg)) + if (tipc_link_recv_fragment(&node->bclink.defragm, + &buf, &msg)) bcl->stats.recv_fragmented++; - node_unlock(node); - net_route_msg(buf); + tipc_node_unlock(node); + tipc_net_route_msg(buf); } else { - node_unlock(node); - net_route_msg(buf); + tipc_node_unlock(node); + tipc_net_route_msg(buf); } if (deferred && (buf_seqno(deferred) == mod(next_in + 1))) { - node_lock(node); + tipc_node_lock(node); buf = deferred; msg = buf_msg(buf); node->bclink.deferred_head = deferred->next; @@ -478,9 +478,9 @@ receive: u32 gap_after = node->bclink.gap_after; u32 gap_to = node->bclink.gap_to; - if (link_defer_pkt(&node->bclink.deferred_head, - &node->bclink.deferred_tail, - buf)) { + if (tipc_link_defer_pkt(&node->bclink.deferred_head, + &node->bclink.deferred_tail, + buf)) { node->bclink.nack_sync++; bcl->stats.deferred_recv++; if (seqno == mod(gap_after + 1)) @@ -497,10 +497,10 @@ receive: bcl->stats.duplicates++; buf_discard(buf); } - node_unlock(node); + tipc_node_unlock(node); } -u32 bclink_get_last_sent(void) +u32 tipc_bclink_get_last_sent(void) { u32 last_sent = mod(bcl->next_out_no - 1); @@ -509,15 +509,15 @@ u32 bclink_get_last_sent(void) return last_sent; } -u32 bclink_acks_missing(struct node *n_ptr) +u32 tipc_bclink_acks_missing(struct node *n_ptr) { return (n_ptr->bclink.supported && - (bclink_get_last_sent() != n_ptr->bclink.acked)); + (tipc_bclink_get_last_sent() != n_ptr->bclink.acked)); } /** - * bcbearer_send - send a packet through the broadcast pseudo-bearer + * tipc_bcbearer_send - send a packet through the broadcast pseudo-bearer * * Send through as many bearers as necessary to reach all nodes * that support TIPC multicasting. @@ -525,9 +525,9 @@ u32 bclink_acks_missing(struct node *n_ptr) * Returns 0 if packet sent successfully, non-zero if not */ -int bcbearer_send(struct sk_buff *buf, - struct tipc_bearer *unused1, - struct tipc_media_addr *unused2) +int tipc_bcbearer_send(struct sk_buff *buf, + struct tipc_bearer *unused1, + struct tipc_media_addr *unused2) { static int send_count = 0; @@ -541,8 +541,8 @@ int bcbearer_send(struct sk_buff *buf, if (likely(!msg_non_seq(buf_msg(buf)))) { struct tipc_msg *msg; - assert(cluster_bcast_nodes.count != 0); - bcbuf_set_acks(buf, cluster_bcast_nodes.count); + assert(tipc_cltr_bcast_nodes.count != 0); + bcbuf_set_acks(buf, tipc_cltr_bcast_nodes.count); msg = buf_msg(buf); msg_set_non_seq(msg); msg_set_mc_netid(msg, tipc_net_id); @@ -555,7 +555,7 @@ int bcbearer_send(struct sk_buff *buf, /* Send buffer over bearers until all targets reached */ - remains = cluster_bcast_nodes; + remains = tipc_cltr_bcast_nodes; for (bp_index = 0; bp_index < MAX_BEARERS; bp_index++) { struct bearer *p = bcbearer->bpairs[bp_index].primary; @@ -564,7 +564,7 @@ int bcbearer_send(struct sk_buff *buf, if (!p) break; /* no more bearers to try */ - nmap_diff(&remains, &p->nodes, &remains_new); + tipc_nmap_diff(&remains, &p->nodes, &remains_new); if (remains_new.count == remains.count) continue; /* bearer pair doesn't add anything */ @@ -597,10 +597,10 @@ update: } /** - * bcbearer_sort - create sets of bearer pairs used by broadcast bearer + * tipc_bcbearer_sort - create sets of bearer pairs used by broadcast bearer */ -void bcbearer_sort(void) +void tipc_bcbearer_sort(void) { struct bcbearer_pair *bp_temp = bcbearer->bpairs_temp; struct bcbearer_pair *bp_curr; @@ -614,7 +614,7 @@ void bcbearer_sort(void) memset(bp_temp, 0, sizeof(bcbearer->bpairs_temp)); for (b_index = 0; b_index < MAX_BEARERS; b_index++) { - struct bearer *b = &bearers[b_index]; + struct bearer *b = &tipc_bearers[b_index]; if (!b->active || !b->nodes.count) continue; @@ -638,8 +638,8 @@ void bcbearer_sort(void) bp_curr->primary = bp_temp[pri].primary; if (bp_temp[pri].secondary) { - if (nmap_equal(&bp_temp[pri].primary->nodes, - &bp_temp[pri].secondary->nodes)) { + if (tipc_nmap_equal(&bp_temp[pri].primary->nodes, + &bp_temp[pri].secondary->nodes)) { bp_curr->secondary = bp_temp[pri].secondary; } else { bp_curr++; @@ -654,14 +654,14 @@ void bcbearer_sort(void) } /** - * bcbearer_push - resolve bearer congestion + * tipc_bcbearer_push - resolve bearer congestion * * Forces bclink to push out any unsent packets, until all packets are gone * or congestion reoccurs. * No locks set when function called */ -void bcbearer_push(void) +void tipc_bcbearer_push(void) { struct bearer *b_ptr; @@ -669,20 +669,20 @@ void bcbearer_push(void) b_ptr = &bcbearer->bearer; if (b_ptr->publ.blocked) { b_ptr->publ.blocked = 0; - bearer_lock_push(b_ptr); + tipc_bearer_lock_push(b_ptr); } spin_unlock_bh(&bc_lock); } -int bclink_stats(char *buf, const u32 buf_size) +int tipc_bclink_stats(char *buf, const u32 buf_size) { struct print_buf pb; if (!bcl) return 0; - printbuf_init(&pb, buf, buf_size); + tipc_printbuf_init(&pb, buf, buf_size); spin_lock_bh(&bc_lock); @@ -718,10 +718,10 @@ int bclink_stats(char *buf, const u32 buf_size) : 0); spin_unlock_bh(&bc_lock); - return printbuf_validate(&pb); + return tipc_printbuf_validate(&pb); } -int bclink_reset_stats(void) +int tipc_bclink_reset_stats(void) { if (!bcl) return -ENOPROTOOPT; @@ -732,7 +732,7 @@ int bclink_reset_stats(void) return TIPC_OK; } -int bclink_set_queue_limits(u32 limit) +int tipc_bclink_set_queue_limits(u32 limit) { if (!bcl) return -ENOPROTOOPT; @@ -740,12 +740,12 @@ int bclink_set_queue_limits(u32 limit) return -EINVAL; spin_lock_bh(&bc_lock); - link_set_queue_limits(bcl, limit); + tipc_link_set_queue_limits(bcl, limit); spin_unlock_bh(&bc_lock); return TIPC_OK; } -int bclink_init(void) +int tipc_bclink_init(void) { bcbearer = kmalloc(sizeof(*bcbearer), GFP_ATOMIC); bclink = kmalloc(sizeof(*bclink), GFP_ATOMIC); @@ -762,7 +762,7 @@ int bclink_init(void) memset(bcbearer, 0, sizeof(struct bcbearer)); INIT_LIST_HEAD(&bcbearer->bearer.cong_links); bcbearer->bearer.media = &bcbearer->media; - bcbearer->media.send_msg = bcbearer_send; + bcbearer->media.send_msg = tipc_bcbearer_send; sprintf(bcbearer->media.name, "tipc-multicast"); bcl = &bclink->link; @@ -772,27 +772,27 @@ int bclink_init(void) bclink->node.lock = SPIN_LOCK_UNLOCKED; bcl->owner = &bclink->node; bcl->max_pkt = MAX_PKT_DEFAULT_MCAST; - link_set_queue_limits(bcl, BCLINK_WIN_DEFAULT); + tipc_link_set_queue_limits(bcl, BCLINK_WIN_DEFAULT); bcl->b_ptr = &bcbearer->bearer; bcl->state = WORKING_WORKING; - sprintf(bcl->name, bc_link_name); + sprintf(bcl->name, tipc_bclink_name); if (BCLINK_LOG_BUF_SIZE) { char *pb = kmalloc(BCLINK_LOG_BUF_SIZE, GFP_ATOMIC); if (!pb) goto nomem; - printbuf_init(&bcl->print_buf, pb, BCLINK_LOG_BUF_SIZE); + tipc_printbuf_init(&bcl->print_buf, pb, BCLINK_LOG_BUF_SIZE); } return TIPC_OK; } -void bclink_stop(void) +void tipc_bclink_stop(void) { spin_lock_bh(&bc_lock); if (bcbearer) { - link_stop(bcl); + tipc_link_stop(bcl); if (BCLINK_LOG_BUF_SIZE) kfree(bcl->print_buf.buf); bcl = NULL; diff --git a/net/tipc/bcast.h b/net/tipc/bcast.h index 5430e524b4f..0e3be2ab330 100644 --- a/net/tipc/bcast.h +++ b/net/tipc/bcast.h @@ -70,14 +70,14 @@ struct port_list { struct node; -extern char bc_link_name[]; +extern char tipc_bclink_name[]; /** * nmap_get - determine if node exists in a node map */ -static inline int nmap_get(struct node_map *nm_ptr, u32 node) +static inline int tipc_nmap_get(struct node_map *nm_ptr, u32 node) { int n = tipc_node(node); int w = n / WSIZE; @@ -90,7 +90,7 @@ static inline int nmap_get(struct node_map *nm_ptr, u32 node) * nmap_add - add a node to a node map */ -static inline void nmap_add(struct node_map *nm_ptr, u32 node) +static inline void tipc_nmap_add(struct node_map *nm_ptr, u32 node) { int n = tipc_node(node); int w = n / WSIZE; @@ -106,7 +106,7 @@ static inline void nmap_add(struct node_map *nm_ptr, u32 node) * nmap_remove - remove a node from a node map */ -static inline void nmap_remove(struct node_map *nm_ptr, u32 node) +static inline void tipc_nmap_remove(struct node_map *nm_ptr, u32 node) { int n = tipc_node(node); int w = n / WSIZE; @@ -122,7 +122,7 @@ static inline void nmap_remove(struct node_map *nm_ptr, u32 node) * nmap_equal - test for equality of node maps */ -static inline int nmap_equal(struct node_map *nm_a, struct node_map *nm_b) +static inline int tipc_nmap_equal(struct node_map *nm_a, struct node_map *nm_b) { return !memcmp(nm_a, nm_b, sizeof(*nm_a)); } @@ -134,8 +134,8 @@ static inline int nmap_equal(struct node_map *nm_a, struct node_map *nm_b) * @nm_diff: output node map A-B (i.e. nodes of A that are not in B) */ -static inline void nmap_diff(struct node_map *nm_a, struct node_map *nm_b, - struct node_map *nm_diff) +static inline void tipc_nmap_diff(struct node_map *nm_a, struct node_map *nm_b, + struct node_map *nm_diff) { int stop = sizeof(nm_a->map) / sizeof(u32); int w; @@ -159,7 +159,7 @@ static inline void nmap_diff(struct node_map *nm_a, struct node_map *nm_b, * port_list_add - add a port to a port list, ensuring no duplicates */ -static inline void port_list_add(struct port_list *pl_ptr, u32 port) +static inline void tipc_port_list_add(struct port_list *pl_ptr, u32 port) { struct port_list *item = pl_ptr; int i; @@ -194,7 +194,7 @@ static inline void port_list_add(struct port_list *pl_ptr, u32 port) * Note: First item is on stack, so it doesn't need to be released */ -static inline void port_list_free(struct port_list *pl_ptr) +static inline void tipc_port_list_free(struct port_list *pl_ptr) { struct port_list *item; struct port_list *next; @@ -206,18 +206,18 @@ static inline void port_list_free(struct port_list *pl_ptr) } -int bclink_init(void); -void bclink_stop(void); -void bclink_acknowledge(struct node *n_ptr, u32 acked); -int bclink_send_msg(struct sk_buff *buf); -void bclink_recv_pkt(struct sk_buff *buf); -u32 bclink_get_last_sent(void); -u32 bclink_acks_missing(struct node *n_ptr); -void bclink_check_gap(struct node *n_ptr, u32 seqno); -int bclink_stats(char *stats_buf, const u32 buf_size); -int bclink_reset_stats(void); -int bclink_set_queue_limits(u32 limit); -void bcbearer_sort(void); -void bcbearer_push(void); +int tipc_bclink_init(void); +void tipc_bclink_stop(void); +void tipc_bclink_acknowledge(struct node *n_ptr, u32 acked); +int tipc_bclink_send_msg(struct sk_buff *buf); +void tipc_bclink_recv_pkt(struct sk_buff *buf); +u32 tipc_bclink_get_last_sent(void); +u32 tipc_bclink_acks_missing(struct node *n_ptr); +void tipc_bclink_check_gap(struct node *n_ptr, u32 seqno); +int tipc_bclink_stats(char *stats_buf, const u32 buf_size); +int tipc_bclink_reset_stats(void); +int tipc_bclink_set_queue_limits(u32 limit); +void tipc_bcbearer_sort(void); +void tipc_bcbearer_push(void); #endif diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 02b6cf6ab7a..64dcb0f3a8b 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -48,7 +48,7 @@ static struct media *media_list = 0; static u32 media_count = 0; -struct bearer *bearers = 0; +struct bearer *tipc_bearers = 0; /** * media_name_valid - validate media name @@ -107,7 +107,7 @@ int tipc_register_media(u32 media_type, u32 i; int res = -EINVAL; - write_lock_bh(&net_lock); + write_lock_bh(&tipc_net_lock); if (!media_list) goto exit; @@ -165,15 +165,15 @@ int tipc_register_media(u32 media_type, dbg("Media <%s> registered\n", name); res = 0; exit: - write_unlock_bh(&net_lock); + write_unlock_bh(&tipc_net_lock); return res; } /** - * media_addr_printf - record media address in print buffer + * tipc_media_addr_printf - record media address in print buffer */ -void media_addr_printf(struct print_buf *pb, struct tipc_media_addr *a) +void tipc_media_addr_printf(struct print_buf *pb, struct tipc_media_addr *a) { struct media *m_ptr; u32 media_type; @@ -201,25 +201,25 @@ void media_addr_printf(struct print_buf *pb, struct tipc_media_addr *a) } /** - * media_get_names - record names of registered media in buffer + * tipc_media_get_names - record names of registered media in buffer */ -struct sk_buff *media_get_names(void) +struct sk_buff *tipc_media_get_names(void) { struct sk_buff *buf; struct media *m_ptr; int i; - buf = cfg_reply_alloc(MAX_MEDIA * TLV_SPACE(TIPC_MAX_MEDIA_NAME)); + buf = tipc_cfg_reply_alloc(MAX_MEDIA * TLV_SPACE(TIPC_MAX_MEDIA_NAME)); if (!buf) return NULL; - read_lock_bh(&net_lock); + read_lock_bh(&tipc_net_lock); for (i = 0, m_ptr = media_list; i < media_count; i++, m_ptr++) { - cfg_append_tlv(buf, TIPC_TLV_MEDIA_NAME, m_ptr->name, - strlen(m_ptr->name) + 1); + tipc_cfg_append_tlv(buf, TIPC_TLV_MEDIA_NAME, m_ptr->name, + strlen(m_ptr->name) + 1); } - read_unlock_bh(&net_lock); + read_unlock_bh(&tipc_net_lock); return buf; } @@ -283,7 +283,7 @@ static struct bearer *bearer_find(const char *name) struct bearer *b_ptr; u32 i; - for (i = 0, b_ptr = bearers; i < MAX_BEARERS; i++, b_ptr++) { + for (i = 0, b_ptr = tipc_bearers; i < MAX_BEARERS; i++, b_ptr++) { if (b_ptr->active && (!strcmp(b_ptr->publ.name, name))) return b_ptr; } @@ -291,16 +291,16 @@ static struct bearer *bearer_find(const char *name) } /** - * bearer_find - locates bearer object with matching interface name + * tipc_bearer_find_interface - locates bearer object with matching interface name */ -struct bearer *bearer_find_interface(const char *if_name) +struct bearer *tipc_bearer_find_interface(const char *if_name) { struct bearer *b_ptr; char *b_if_name; u32 i; - for (i = 0, b_ptr = bearers; i < MAX_BEARERS; i++, b_ptr++) { + for (i = 0, b_ptr = tipc_bearers; i < MAX_BEARERS; i++, b_ptr++) { if (!b_ptr->active) continue; b_if_name = strchr(b_ptr->publ.name, ':') + 1; @@ -311,54 +311,54 @@ struct bearer *bearer_find_interface(const char *if_name) } /** - * bearer_get_names - record names of bearers in buffer + * tipc_bearer_get_names - record names of bearers in buffer */ -struct sk_buff *bearer_get_names(void) +struct sk_buff *tipc_bearer_get_names(void) { struct sk_buff *buf; struct media *m_ptr; struct bearer *b_ptr; int i, j; - buf = cfg_reply_alloc(MAX_BEARERS * TLV_SPACE(TIPC_MAX_BEARER_NAME)); + buf = tipc_cfg_reply_alloc(MAX_BEARERS * TLV_SPACE(TIPC_MAX_BEARER_NAME)); if (!buf) return NULL; - read_lock_bh(&net_lock); + read_lock_bh(&tipc_net_lock); for (i = 0, m_ptr = media_list; i < media_count; i++, m_ptr++) { for (j = 0; j < MAX_BEARERS; j++) { - b_ptr = &bearers[j]; + b_ptr = &tipc_bearers[j]; if (b_ptr->active && (b_ptr->media == m_ptr)) { - cfg_append_tlv(buf, TIPC_TLV_BEARER_NAME, - b_ptr->publ.name, - strlen(b_ptr->publ.name) + 1); + tipc_cfg_append_tlv(buf, TIPC_TLV_BEARER_NAME, + b_ptr->publ.name, + strlen(b_ptr->publ.name) + 1); } } } - read_unlock_bh(&net_lock); + read_unlock_bh(&tipc_net_lock); return buf; } -void bearer_add_dest(struct bearer *b_ptr, u32 dest) +void tipc_bearer_add_dest(struct bearer *b_ptr, u32 dest) { - nmap_add(&b_ptr->nodes, dest); - disc_update_link_req(b_ptr->link_req); - bcbearer_sort(); + tipc_nmap_add(&b_ptr->nodes, dest); + tipc_disc_update_link_req(b_ptr->link_req); + tipc_bcbearer_sort(); } -void bearer_remove_dest(struct bearer *b_ptr, u32 dest) +void tipc_bearer_remove_dest(struct bearer *b_ptr, u32 dest) { - nmap_remove(&b_ptr->nodes, dest); - disc_update_link_req(b_ptr->link_req); - bcbearer_sort(); + tipc_nmap_remove(&b_ptr->nodes, dest); + tipc_disc_update_link_req(b_ptr->link_req); + tipc_bcbearer_sort(); } /* * bearer_push(): Resolve bearer congestion. Force the waiting * links to push out their unsent packets, one packet per link * per iteration, until all packets are gone or congestion reoccurs. - * 'net_lock' is read_locked when this function is called + * 'tipc_net_lock' is read_locked when this function is called * bearer.lock must be taken before calling * Returns binary true(1) ore false(0) */ @@ -372,7 +372,7 @@ static int bearer_push(struct bearer *b_ptr) while (!list_empty(&b_ptr->cong_links) && (res != PUSH_FAILED)) { list_for_each_entry_safe(ln, tln, &b_ptr->cong_links, link_list) { - res = link_push_packet(ln); + res = tipc_link_push_packet(ln); if (res == PUSH_FAILED) break; if (res == PUSH_FINISHED) @@ -382,7 +382,7 @@ static int bearer_push(struct bearer *b_ptr) return list_empty(&b_ptr->cong_links); } -void bearer_lock_push(struct bearer *b_ptr) +void tipc_bearer_lock_push(struct bearer *b_ptr) { int res; @@ -390,7 +390,7 @@ void bearer_lock_push(struct bearer *b_ptr) res = bearer_push(b_ptr); spin_unlock_bh(&b_ptr->publ.lock); if (res) - bcbearer_push(); + tipc_bcbearer_push(); } @@ -405,7 +405,7 @@ void tipc_continue(struct tipc_bearer *tb_ptr) spin_lock_bh(&b_ptr->publ.lock); b_ptr->continue_count++; if (!list_empty(&b_ptr->cong_links)) - k_signal((Handler)bearer_lock_push, (unsigned long)b_ptr); + tipc_k_signal((Handler)tipc_bearer_lock_push, (unsigned long)b_ptr); b_ptr->publ.blocked = 0; spin_unlock_bh(&b_ptr->publ.lock); } @@ -414,11 +414,11 @@ void tipc_continue(struct tipc_bearer *tb_ptr) * Schedule link for sending of messages after the bearer * has been deblocked by 'continue()'. This method is called * when somebody tries to send a message via this link while - * the bearer is congested. 'net_lock' is in read_lock here + * the bearer is congested. 'tipc_net_lock' is in read_lock here * bearer.lock is busy */ -static void bearer_schedule_unlocked(struct bearer *b_ptr, struct link *l_ptr) +static void tipc_bearer_schedule_unlocked(struct bearer *b_ptr, struct link *l_ptr) { list_move_tail(&l_ptr->link_list, &b_ptr->cong_links); } @@ -427,24 +427,24 @@ static void bearer_schedule_unlocked(struct bearer *b_ptr, struct link *l_ptr) * Schedule link for sending of messages after the bearer * has been deblocked by 'continue()'. This method is called * when somebody tries to send a message via this link while - * the bearer is congested. 'net_lock' is in read_lock here, + * the bearer is congested. 'tipc_net_lock' is in read_lock here, * bearer.lock is free */ -void bearer_schedule(struct bearer *b_ptr, struct link *l_ptr) +void tipc_bearer_schedule(struct bearer *b_ptr, struct link *l_ptr) { spin_lock_bh(&b_ptr->publ.lock); - bearer_schedule_unlocked(b_ptr, l_ptr); + tipc_bearer_schedule_unlocked(b_ptr, l_ptr); spin_unlock_bh(&b_ptr->publ.lock); } /* - * bearer_resolve_congestion(): Check if there is bearer congestion, + * tipc_bearer_resolve_congestion(): Check if there is bearer congestion, * and if there is, try to resolve it before returning. - * 'net_lock' is read_locked when this function is called + * 'tipc_net_lock' is read_locked when this function is called */ -int bearer_resolve_congestion(struct bearer *b_ptr, struct link *l_ptr) +int tipc_bearer_resolve_congestion(struct bearer *b_ptr, struct link *l_ptr) { int res = 1; @@ -452,7 +452,7 @@ int bearer_resolve_congestion(struct bearer *b_ptr, struct link *l_ptr) return 1; spin_lock_bh(&b_ptr->publ.lock); if (!bearer_push(b_ptr)) { - bearer_schedule_unlocked(b_ptr, l_ptr); + tipc_bearer_schedule_unlocked(b_ptr, l_ptr); res = 0; } spin_unlock_bh(&b_ptr->publ.lock); @@ -479,7 +479,7 @@ int tipc_enable_bearer(const char *name, u32 bcast_scope, u32 priority) return -ENOPROTOOPT; if (!bearer_name_validate(name, &b_name) || - !addr_domain_valid(bcast_scope) || + !tipc_addr_domain_valid(bcast_scope) || !in_scope(bcast_scope, tipc_own_addr)) return -EINVAL; @@ -488,8 +488,8 @@ int tipc_enable_bearer(const char *name, u32 bcast_scope, u32 priority) (priority != TIPC_MEDIA_LINK_PRI)) return -EINVAL; - write_lock_bh(&net_lock); - if (!bearers) + write_lock_bh(&tipc_net_lock); + if (!tipc_bearers) goto failed; m_ptr = media_find(b_name.media_name); @@ -505,15 +505,15 @@ restart: bearer_id = MAX_BEARERS; with_this_prio = 1; for (i = MAX_BEARERS; i-- != 0; ) { - if (!bearers[i].active) { + if (!tipc_bearers[i].active) { bearer_id = i; continue; } - if (!strcmp(name, bearers[i].publ.name)) { + if (!strcmp(name, tipc_bearers[i].publ.name)) { warn("Bearer <%s> already enabled\n", name); goto failed; } - if ((bearers[i].priority == priority) && + if ((tipc_bearers[i].priority == priority) && (++with_this_prio > 2)) { if (priority-- == 0) { warn("Third bearer <%s> with priority %u, unable to lower to %u\n", @@ -530,7 +530,7 @@ restart: goto failed; } - b_ptr = &bearers[bearer_id]; + b_ptr = &tipc_bearers[bearer_id]; memset(b_ptr, 0, sizeof(struct bearer)); strcpy(b_ptr->publ.name, name); @@ -549,16 +549,16 @@ restart: INIT_LIST_HEAD(&b_ptr->cong_links); INIT_LIST_HEAD(&b_ptr->links); if (m_ptr->bcast) { - b_ptr->link_req = disc_init_link_req(b_ptr, &m_ptr->bcast_addr, - bcast_scope, 2); + b_ptr->link_req = tipc_disc_init_link_req(b_ptr, &m_ptr->bcast_addr, + bcast_scope, 2); } b_ptr->publ.lock = SPIN_LOCK_UNLOCKED; - write_unlock_bh(&net_lock); + write_unlock_bh(&tipc_net_lock); info("Enabled bearer <%s>, discovery domain %s, priority %u\n", name, addr_string_fill(addr_string, bcast_scope), priority); return 0; failed: - write_unlock_bh(&net_lock); + write_unlock_bh(&tipc_net_lock); return res; } @@ -576,11 +576,11 @@ int tipc_block_bearer(const char *name) if (tipc_mode != TIPC_NET_MODE) return -ENOPROTOOPT; - read_lock_bh(&net_lock); + read_lock_bh(&tipc_net_lock); b_ptr = bearer_find(name); if (!b_ptr) { warn("Attempt to block unknown bearer <%s>\n", name); - read_unlock_bh(&net_lock); + read_unlock_bh(&tipc_net_lock); return -EINVAL; } @@ -590,11 +590,11 @@ int tipc_block_bearer(const char *name) struct node *n_ptr = l_ptr->owner; spin_lock_bh(&n_ptr->lock); - link_reset(l_ptr); + tipc_link_reset(l_ptr); spin_unlock_bh(&n_ptr->lock); } spin_unlock_bh(&b_ptr->publ.lock); - read_unlock_bh(&net_lock); + read_unlock_bh(&tipc_net_lock); info("Blocked bearer <%s>\n", name); return TIPC_OK; } @@ -602,7 +602,7 @@ int tipc_block_bearer(const char *name) /** * bearer_disable - * - * Note: This routine assumes caller holds net_lock. + * Note: This routine assumes caller holds tipc_net_lock. */ static int bearer_disable(const char *name) @@ -620,19 +620,19 @@ static int bearer_disable(const char *name) return -EINVAL; } - disc_stop_link_req(b_ptr->link_req); + tipc_disc_stop_link_req(b_ptr->link_req); spin_lock_bh(&b_ptr->publ.lock); b_ptr->link_req = NULL; b_ptr->publ.blocked = 1; if (b_ptr->media->disable_bearer) { spin_unlock_bh(&b_ptr->publ.lock); - write_unlock_bh(&net_lock); + write_unlock_bh(&tipc_net_lock); b_ptr->media->disable_bearer(&b_ptr->publ); - write_lock_bh(&net_lock); + write_lock_bh(&tipc_net_lock); spin_lock_bh(&b_ptr->publ.lock); } list_for_each_entry_safe(l_ptr, temp_l_ptr, &b_ptr->links, link_list) { - link_delete(l_ptr); + tipc_link_delete(l_ptr); } spin_unlock_bh(&b_ptr->publ.lock); info("Disabled bearer <%s>\n", name); @@ -644,54 +644,54 @@ int tipc_disable_bearer(const char *name) { int res; - write_lock_bh(&net_lock); + write_lock_bh(&tipc_net_lock); res = bearer_disable(name); - write_unlock_bh(&net_lock); + write_unlock_bh(&tipc_net_lock); return res; } -int bearer_init(void) +int tipc_bearer_init(void) { int res; - write_lock_bh(&net_lock); - bearers = kmalloc(MAX_BEARERS * sizeof(struct bearer), GFP_ATOMIC); + write_lock_bh(&tipc_net_lock); + tipc_bearers = kmalloc(MAX_BEARERS * sizeof(struct bearer), GFP_ATOMIC); media_list = kmalloc(MAX_MEDIA * sizeof(struct media), GFP_ATOMIC); - if (bearers && media_list) { - memset(bearers, 0, MAX_BEARERS * sizeof(struct bearer)); + if (tipc_bearers && media_list) { + memset(tipc_bearers, 0, MAX_BEARERS * sizeof(struct bearer)); memset(media_list, 0, MAX_MEDIA * sizeof(struct media)); res = TIPC_OK; } else { - kfree(bearers); + kfree(tipc_bearers); kfree(media_list); - bearers = 0; + tipc_bearers = 0; media_list = 0; res = -ENOMEM; } - write_unlock_bh(&net_lock); + write_unlock_bh(&tipc_net_lock); return res; } -void bearer_stop(void) +void tipc_bearer_stop(void) { u32 i; - if (!bearers) + if (!tipc_bearers) return; for (i = 0; i < MAX_BEARERS; i++) { - if (bearers[i].active) - bearers[i].publ.blocked = 1; + if (tipc_bearers[i].active) + tipc_bearers[i].publ.blocked = 1; } for (i = 0; i < MAX_BEARERS; i++) { - if (bearers[i].active) - bearer_disable(bearers[i].publ.name); + if (tipc_bearers[i].active) + bearer_disable(tipc_bearers[i].publ.name); } - kfree(bearers); + kfree(tipc_bearers); kfree(media_list); - bearers = 0; + tipc_bearers = 0; media_list = 0; media_count = 0; } diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h index 0c40cc2b43c..c4e7c1c3655 100644 --- a/net/tipc/bearer.h +++ b/net/tipc/bearer.h @@ -114,26 +114,24 @@ struct bearer_name { struct link; -extern struct bearer *bearers; +extern struct bearer *tipc_bearers; -void media_addr_printf(struct print_buf *pb, struct tipc_media_addr *a); -struct sk_buff *media_get_names(void); +void tipc_media_addr_printf(struct print_buf *pb, struct tipc_media_addr *a); +struct sk_buff *tipc_media_get_names(void); -struct sk_buff *bearer_get_names(void); -void bearer_add_dest(struct bearer *b_ptr, u32 dest); -void bearer_remove_dest(struct bearer *b_ptr, u32 dest); -void bearer_schedule(struct bearer *b_ptr, struct link *l_ptr); -struct bearer *bearer_find_interface(const char *if_name); -int bearer_resolve_congestion(struct bearer *b_ptr, struct link *l_ptr); -int bearer_init(void); -void bearer_stop(void); -int bearer_broadcast(struct sk_buff *buf, struct tipc_bearer *b_ptr, - struct tipc_media_addr *dest); -void bearer_lock_push(struct bearer *b_ptr); +struct sk_buff *tipc_bearer_get_names(void); +void tipc_bearer_add_dest(struct bearer *b_ptr, u32 dest); +void tipc_bearer_remove_dest(struct bearer *b_ptr, u32 dest); +void tipc_bearer_schedule(struct bearer *b_ptr, struct link *l_ptr); +struct bearer *tipc_bearer_find_interface(const char *if_name); +int tipc_bearer_resolve_congestion(struct bearer *b_ptr, struct link *l_ptr); +int tipc_bearer_init(void); +void tipc_bearer_stop(void); +void tipc_bearer_lock_push(struct bearer *b_ptr); /** - * bearer_send- sends buffer to destination over bearer + * tipc_bearer_send- sends buffer to destination over bearer * * Returns true (1) if successful, or false (0) if unable to send * @@ -150,23 +148,23 @@ void bearer_lock_push(struct bearer *b_ptr); * and let TIPC's link code deal with the undelivered message. */ -static inline int bearer_send(struct bearer *b_ptr, struct sk_buff *buf, - struct tipc_media_addr *dest) +static inline int tipc_bearer_send(struct bearer *b_ptr, struct sk_buff *buf, + struct tipc_media_addr *dest) { return !b_ptr->media->send_msg(buf, &b_ptr->publ, dest); } /** - * bearer_congested - determines if bearer is currently congested + * tipc_bearer_congested - determines if bearer is currently congested */ -static inline int bearer_congested(struct bearer *b_ptr, struct link *l_ptr) +static inline int tipc_bearer_congested(struct bearer *b_ptr, struct link *l_ptr) { if (unlikely(b_ptr->publ.blocked)) return 1; if (likely(list_empty(&b_ptr->cong_links))) return 0; - return !bearer_resolve_congestion(b_ptr, l_ptr); + return !tipc_bearer_resolve_congestion(b_ptr, l_ptr); } #endif diff --git a/net/tipc/cluster.c b/net/tipc/cluster.c index f0f7bac51d4..ab974ca1937 100644 --- a/net/tipc/cluster.c +++ b/net/tipc/cluster.c @@ -44,15 +44,15 @@ #include "msg.h" #include "bearer.h" -void cluster_multicast(struct cluster *c_ptr, struct sk_buff *buf, - u32 lower, u32 upper); -struct sk_buff *cluster_prepare_routing_msg(u32 data_size, u32 dest); +void tipc_cltr_multicast(struct cluster *c_ptr, struct sk_buff *buf, + u32 lower, u32 upper); +struct sk_buff *tipc_cltr_prepare_routing_msg(u32 data_size, u32 dest); -struct node **local_nodes = 0; -struct node_map cluster_bcast_nodes = {0,{0,}}; -u32 highest_allowed_slave = 0; +struct node **tipc_local_nodes = 0; +struct node_map tipc_cltr_bcast_nodes = {0,{0,}}; +u32 tipc_highest_allowed_slave = 0; -struct cluster *cluster_create(u32 addr) +struct cluster *tipc_cltr_create(u32 addr) { struct _zone *z_ptr; struct cluster *c_ptr; @@ -77,16 +77,16 @@ struct cluster *cluster_create(u32 addr) } memset(c_ptr->nodes, 0, alloc); if (in_own_cluster(addr)) - local_nodes = c_ptr->nodes; + tipc_local_nodes = c_ptr->nodes; c_ptr->highest_slave = LOWEST_SLAVE - 1; c_ptr->highest_node = 0; - z_ptr = zone_find(tipc_zone(addr)); + z_ptr = tipc_zone_find(tipc_zone(addr)); if (z_ptr == NULL) { - z_ptr = zone_create(addr); + z_ptr = tipc_zone_create(addr); } if (z_ptr != NULL) { - zone_attach_cluster(z_ptr, c_ptr); + tipc_zone_attach_cluster(z_ptr, c_ptr); c_ptr->owner = z_ptr; } else { @@ -97,23 +97,23 @@ struct cluster *cluster_create(u32 addr) return c_ptr; } -void cluster_delete(struct cluster *c_ptr) +void tipc_cltr_delete(struct cluster *c_ptr) { u32 n_num; if (!c_ptr) return; for (n_num = 1; n_num <= c_ptr->highest_node; n_num++) { - node_delete(c_ptr->nodes[n_num]); + tipc_node_delete(c_ptr->nodes[n_num]); } for (n_num = LOWEST_SLAVE; n_num <= c_ptr->highest_slave; n_num++) { - node_delete(c_ptr->nodes[n_num]); + tipc_node_delete(c_ptr->nodes[n_num]); } kfree(c_ptr->nodes); kfree(c_ptr); } -u32 cluster_next_node(struct cluster *c_ptr, u32 addr) +u32 tipc_cltr_next_node(struct cluster *c_ptr, u32 addr) { struct node *n_ptr; u32 n_num = tipc_node(addr) + 1; @@ -122,24 +122,24 @@ u32 cluster_next_node(struct cluster *c_ptr, u32 addr) return addr; for (; n_num <= c_ptr->highest_node; n_num++) { n_ptr = c_ptr->nodes[n_num]; - if (n_ptr && node_has_active_links(n_ptr)) + if (n_ptr && tipc_node_has_active_links(n_ptr)) return n_ptr->addr; } for (n_num = 1; n_num < tipc_node(addr); n_num++) { n_ptr = c_ptr->nodes[n_num]; - if (n_ptr && node_has_active_links(n_ptr)) + if (n_ptr && tipc_node_has_active_links(n_ptr)) return n_ptr->addr; } return 0; } -void cluster_attach_node(struct cluster *c_ptr, struct node *n_ptr) +void tipc_cltr_attach_node(struct cluster *c_ptr, struct node *n_ptr) { u32 n_num = tipc_node(n_ptr->addr); u32 max_n_num = tipc_max_nodes; if (in_own_cluster(n_ptr->addr)) - max_n_num = highest_allowed_slave; + max_n_num = tipc_highest_allowed_slave; assert(n_num > 0); assert(n_num <= max_n_num); assert(c_ptr->nodes[n_num] == 0); @@ -149,12 +149,12 @@ void cluster_attach_node(struct cluster *c_ptr, struct node *n_ptr) } /** - * cluster_select_router - select router to a cluster + * tipc_cltr_select_router - select router to a cluster * * Uses deterministic and fair algorithm. */ -u32 cluster_select_router(struct cluster *c_ptr, u32 ref) +u32 tipc_cltr_select_router(struct cluster *c_ptr, u32 ref) { u32 n_num; u32 ulim = c_ptr->highest_node; @@ -174,29 +174,29 @@ u32 cluster_select_router(struct cluster *c_ptr, u32 ref) /* Lookup upwards with wrap-around */ do { - if (node_is_up(c_ptr->nodes[n_num])) + if (tipc_node_is_up(c_ptr->nodes[n_num])) break; } while (++n_num <= ulim); if (n_num > ulim) { n_num = 1; do { - if (node_is_up(c_ptr->nodes[n_num])) + if (tipc_node_is_up(c_ptr->nodes[n_num])) break; } while (++n_num < tstart); if (n_num == tstart) return 0; } assert(n_num <= ulim); - return node_select_router(c_ptr->nodes[n_num], ref); + return tipc_node_select_router(c_ptr->nodes[n_num], ref); } /** - * cluster_select_node - select destination node within a remote cluster + * tipc_cltr_select_node - select destination node within a remote cluster * * Uses deterministic and fair algorithm. */ -struct node *cluster_select_node(struct cluster *c_ptr, u32 selector) +struct node *tipc_cltr_select_node(struct cluster *c_ptr, u32 selector) { u32 n_num; u32 mask = tipc_max_nodes; @@ -215,11 +215,11 @@ struct node *cluster_select_node(struct cluster *c_ptr, u32 selector) /* Lookup upwards with wrap-around */ for (n_num = start_entry; n_num <= c_ptr->highest_node; n_num++) { - if (node_has_active_links(c_ptr->nodes[n_num])) + if (tipc_node_has_active_links(c_ptr->nodes[n_num])) return c_ptr->nodes[n_num]; } for (n_num = 1; n_num < start_entry; n_num++) { - if (node_has_active_links(c_ptr->nodes[n_num])) + if (tipc_node_has_active_links(c_ptr->nodes[n_num])) return c_ptr->nodes[n_num]; } return 0; @@ -229,7 +229,7 @@ struct node *cluster_select_node(struct cluster *c_ptr, u32 selector) * Routing table management: See description in node.c */ -struct sk_buff *cluster_prepare_routing_msg(u32 data_size, u32 dest) +struct sk_buff *tipc_cltr_prepare_routing_msg(u32 data_size, u32 dest) { u32 size = INT_H_SIZE + data_size; struct sk_buff *buf = buf_acquire(size); @@ -243,39 +243,39 @@ struct sk_buff *cluster_prepare_routing_msg(u32 data_size, u32 dest) return buf; } -void cluster_bcast_new_route(struct cluster *c_ptr, u32 dest, +void tipc_cltr_bcast_new_route(struct cluster *c_ptr, u32 dest, u32 lower, u32 upper) { - struct sk_buff *buf = cluster_prepare_routing_msg(0, c_ptr->addr); + struct sk_buff *buf = tipc_cltr_prepare_routing_msg(0, c_ptr->addr); struct tipc_msg *msg; if (buf) { msg = buf_msg(buf); msg_set_remote_node(msg, dest); msg_set_type(msg, ROUTE_ADDITION); - cluster_multicast(c_ptr, buf, lower, upper); + tipc_cltr_multicast(c_ptr, buf, lower, upper); } else { warn("Memory squeeze: broadcast of new route failed\n"); } } -void cluster_bcast_lost_route(struct cluster *c_ptr, u32 dest, - u32 lower, u32 upper) +void tipc_cltr_bcast_lost_route(struct cluster *c_ptr, u32 dest, + u32 lower, u32 upper) { - struct sk_buff *buf = cluster_prepare_routing_msg(0, c_ptr->addr); + struct sk_buff *buf = tipc_cltr_prepare_routing_msg(0, c_ptr->addr); struct tipc_msg *msg; if (buf) { msg = buf_msg(buf); msg_set_remote_node(msg, dest); msg_set_type(msg, ROUTE_REMOVAL); - cluster_multicast(c_ptr, buf, lower, upper); + tipc_cltr_multicast(c_ptr, buf, lower, upper); } else { warn("Memory squeeze: broadcast of lost route failed\n"); } } -void cluster_send_slave_routes(struct cluster *c_ptr, u32 dest) +void tipc_cltr_send_slave_routes(struct cluster *c_ptr, u32 dest) { struct sk_buff *buf; struct tipc_msg *msg; @@ -288,21 +288,21 @@ void cluster_send_slave_routes(struct cluster *c_ptr, u32 dest) assert(in_own_cluster(c_ptr->addr)); if (highest <= LOWEST_SLAVE) return; - buf = cluster_prepare_routing_msg(highest - LOWEST_SLAVE + 1, - c_ptr->addr); + buf = tipc_cltr_prepare_routing_msg(highest - LOWEST_SLAVE + 1, + c_ptr->addr); if (buf) { msg = buf_msg(buf); msg_set_remote_node(msg, c_ptr->addr); msg_set_type(msg, SLAVE_ROUTING_TABLE); for (n_num = LOWEST_SLAVE; n_num <= highest; n_num++) { if (c_ptr->nodes[n_num] && - node_has_active_links(c_ptr->nodes[n_num])) { + tipc_node_has_active_links(c_ptr->nodes[n_num])) { send = 1; msg_set_dataoctet(msg, n_num); } } if (send) - link_send(buf, dest, dest); + tipc_link_send(buf, dest, dest); else buf_discard(buf); } else { @@ -310,7 +310,7 @@ void cluster_send_slave_routes(struct cluster *c_ptr, u32 dest) } } -void cluster_send_ext_routes(struct cluster *c_ptr, u32 dest) +void tipc_cltr_send_ext_routes(struct cluster *c_ptr, u32 dest) { struct sk_buff *buf; struct tipc_msg *msg; @@ -323,20 +323,20 @@ void cluster_send_ext_routes(struct cluster *c_ptr, u32 dest) assert(!is_slave(dest)); assert(in_own_cluster(dest)); highest = c_ptr->highest_node; - buf = cluster_prepare_routing_msg(highest + 1, c_ptr->addr); + buf = tipc_cltr_prepare_routing_msg(highest + 1, c_ptr->addr); if (buf) { msg = buf_msg(buf); msg_set_remote_node(msg, c_ptr->addr); msg_set_type(msg, EXT_ROUTING_TABLE); for (n_num = 1; n_num <= highest; n_num++) { if (c_ptr->nodes[n_num] && - node_has_active_links(c_ptr->nodes[n_num])) { + tipc_node_has_active_links(c_ptr->nodes[n_num])) { send = 1; msg_set_dataoctet(msg, n_num); } } if (send) - link_send(buf, dest, dest); + tipc_link_send(buf, dest, dest); else buf_discard(buf); } else { @@ -344,7 +344,7 @@ void cluster_send_ext_routes(struct cluster *c_ptr, u32 dest) } } -void cluster_send_local_routes(struct cluster *c_ptr, u32 dest) +void tipc_cltr_send_local_routes(struct cluster *c_ptr, u32 dest) { struct sk_buff *buf; struct tipc_msg *msg; @@ -354,20 +354,20 @@ void cluster_send_local_routes(struct cluster *c_ptr, u32 dest) assert(is_slave(dest)); assert(in_own_cluster(c_ptr->addr)); - buf = cluster_prepare_routing_msg(highest, c_ptr->addr); + buf = tipc_cltr_prepare_routing_msg(highest, c_ptr->addr); if (buf) { msg = buf_msg(buf); msg_set_remote_node(msg, c_ptr->addr); msg_set_type(msg, LOCAL_ROUTING_TABLE); for (n_num = 1; n_num <= highest; n_num++) { if (c_ptr->nodes[n_num] && - node_has_active_links(c_ptr->nodes[n_num])) { + tipc_node_has_active_links(c_ptr->nodes[n_num])) { send = 1; msg_set_dataoctet(msg, n_num); } } if (send) - link_send(buf, dest, dest); + tipc_link_send(buf, dest, dest); else buf_discard(buf); } else { @@ -375,7 +375,7 @@ void cluster_send_local_routes(struct cluster *c_ptr, u32 dest) } } -void cluster_recv_routing_table(struct sk_buff *buf) +void tipc_cltr_recv_routing_table(struct sk_buff *buf) { struct tipc_msg *msg = buf_msg(buf); struct cluster *c_ptr; @@ -388,9 +388,9 @@ void cluster_recv_routing_table(struct sk_buff *buf) u32 c_num; u32 n_num; - c_ptr = cluster_find(rem_node); + c_ptr = tipc_cltr_find(rem_node); if (!c_ptr) { - c_ptr = cluster_create(rem_node); + c_ptr = tipc_cltr_create(rem_node); if (!c_ptr) { buf_discard(buf); return; @@ -412,10 +412,10 @@ void cluster_recv_routing_table(struct sk_buff *buf) u32 addr = tipc_addr(z_num, c_num, n_num); n_ptr = c_ptr->nodes[n_num]; if (!n_ptr) { - n_ptr = node_create(addr); + n_ptr = tipc_node_create(addr); } if (n_ptr) - node_add_router(n_ptr, router); + tipc_node_add_router(n_ptr, router); } } break; @@ -428,10 +428,10 @@ void cluster_recv_routing_table(struct sk_buff *buf) u32 addr = tipc_addr(z_num, c_num, slave_num); n_ptr = c_ptr->nodes[slave_num]; if (!n_ptr) { - n_ptr = node_create(addr); + n_ptr = tipc_node_create(addr); } if (n_ptr) - node_add_router(n_ptr, router); + tipc_node_add_router(n_ptr, router); } } break; @@ -445,9 +445,9 @@ void cluster_recv_routing_table(struct sk_buff *buf) } n_ptr = c_ptr->nodes[tipc_node(rem_node)]; if (!n_ptr) - n_ptr = node_create(rem_node); + n_ptr = tipc_node_create(rem_node); if (n_ptr) - node_add_router(n_ptr, router); + tipc_node_add_router(n_ptr, router); break; case ROUTE_REMOVAL: if (!is_slave(tipc_own_addr)) { @@ -459,7 +459,7 @@ void cluster_recv_routing_table(struct sk_buff *buf) } n_ptr = c_ptr->nodes[tipc_node(rem_node)]; if (n_ptr) - node_remove_router(n_ptr, router); + tipc_node_remove_router(n_ptr, router); break; default: assert(!"Illegal routing manager message received\n"); @@ -467,7 +467,7 @@ void cluster_recv_routing_table(struct sk_buff *buf) buf_discard(buf); } -void cluster_remove_as_router(struct cluster *c_ptr, u32 router) +void tipc_cltr_remove_as_router(struct cluster *c_ptr, u32 router) { u32 start_entry; u32 tstop; @@ -486,17 +486,17 @@ void cluster_remove_as_router(struct cluster *c_ptr, u32 router) for (n_num = start_entry; n_num <= tstop; n_num++) { if (c_ptr->nodes[n_num]) { - node_remove_router(c_ptr->nodes[n_num], router); + tipc_node_remove_router(c_ptr->nodes[n_num], router); } } } /** - * cluster_multicast - multicast message to local nodes + * tipc_cltr_multicast - multicast message to local nodes */ -void cluster_multicast(struct cluster *c_ptr, struct sk_buff *buf, - u32 lower, u32 upper) +void tipc_cltr_multicast(struct cluster *c_ptr, struct sk_buff *buf, + u32 lower, u32 upper) { struct sk_buff *buf_copy; struct node *n_ptr; @@ -505,9 +505,9 @@ void cluster_multicast(struct cluster *c_ptr, struct sk_buff *buf, assert(lower <= upper); assert(((lower >= 1) && (lower <= tipc_max_nodes)) || - ((lower >= LOWEST_SLAVE) && (lower <= highest_allowed_slave))); + ((lower >= LOWEST_SLAVE) && (lower <= tipc_highest_allowed_slave))); assert(((upper >= 1) && (upper <= tipc_max_nodes)) || - ((upper >= LOWEST_SLAVE) && (upper <= highest_allowed_slave))); + ((upper >= LOWEST_SLAVE) && (upper <= tipc_highest_allowed_slave))); assert(in_own_cluster(c_ptr->addr)); tstop = is_slave(upper) ? c_ptr->highest_slave : c_ptr->highest_node; @@ -515,22 +515,22 @@ void cluster_multicast(struct cluster *c_ptr, struct sk_buff *buf, tstop = upper; for (n_num = lower; n_num <= tstop; n_num++) { n_ptr = c_ptr->nodes[n_num]; - if (n_ptr && node_has_active_links(n_ptr)) { + if (n_ptr && tipc_node_has_active_links(n_ptr)) { buf_copy = skb_copy(buf, GFP_ATOMIC); if (buf_copy == NULL) break; msg_set_destnode(buf_msg(buf_copy), n_ptr->addr); - link_send(buf_copy, n_ptr->addr, n_ptr->addr); + tipc_link_send(buf_copy, n_ptr->addr, n_ptr->addr); } } buf_discard(buf); } /** - * cluster_broadcast - broadcast message to all nodes within cluster + * tipc_cltr_broadcast - broadcast message to all nodes within cluster */ -void cluster_broadcast(struct sk_buff *buf) +void tipc_cltr_broadcast(struct sk_buff *buf) { struct sk_buff *buf_copy; struct cluster *c_ptr; @@ -541,7 +541,7 @@ void cluster_broadcast(struct sk_buff *buf) u32 node_type; if (tipc_mode == TIPC_NET_MODE) { - c_ptr = cluster_find(tipc_own_addr); + c_ptr = tipc_cltr_find(tipc_own_addr); assert(in_own_cluster(c_ptr->addr)); /* For now */ /* Send to standard nodes, then repeat loop sending to slaves */ @@ -550,14 +550,14 @@ void cluster_broadcast(struct sk_buff *buf) for (node_type = 1; node_type <= 2; node_type++) { for (n_num = tstart; n_num <= tstop; n_num++) { n_ptr = c_ptr->nodes[n_num]; - if (n_ptr && node_has_active_links(n_ptr)) { + if (n_ptr && tipc_node_has_active_links(n_ptr)) { buf_copy = skb_copy(buf, GFP_ATOMIC); if (buf_copy == NULL) goto exit; msg_set_destnode(buf_msg(buf_copy), n_ptr->addr); - link_send(buf_copy, n_ptr->addr, - n_ptr->addr); + tipc_link_send(buf_copy, n_ptr->addr, + n_ptr->addr); } } tstart = LOWEST_SLAVE; @@ -568,9 +568,9 @@ exit: buf_discard(buf); } -int cluster_init(void) +int tipc_cltr_init(void) { - highest_allowed_slave = LOWEST_SLAVE + tipc_max_slaves; - return cluster_create(tipc_own_addr) ? TIPC_OK : -ENOMEM; + tipc_highest_allowed_slave = LOWEST_SLAVE + tipc_max_slaves; + return tipc_cltr_create(tipc_own_addr) ? TIPC_OK : -ENOMEM; } diff --git a/net/tipc/cluster.h b/net/tipc/cluster.h index 1ffb095991d..9963642e105 100644 --- a/net/tipc/cluster.h +++ b/net/tipc/cluster.h @@ -60,29 +60,29 @@ struct cluster { }; -extern struct node **local_nodes; -extern u32 highest_allowed_slave; -extern struct node_map cluster_bcast_nodes; +extern struct node **tipc_local_nodes; +extern u32 tipc_highest_allowed_slave; +extern struct node_map tipc_cltr_bcast_nodes; -void cluster_remove_as_router(struct cluster *c_ptr, u32 router); -void cluster_send_ext_routes(struct cluster *c_ptr, u32 dest); -struct node *cluster_select_node(struct cluster *c_ptr, u32 selector); -u32 cluster_select_router(struct cluster *c_ptr, u32 ref); -void cluster_recv_routing_table(struct sk_buff *buf); -struct cluster *cluster_create(u32 addr); -void cluster_delete(struct cluster *c_ptr); -void cluster_attach_node(struct cluster *c_ptr, struct node *n_ptr); -void cluster_send_slave_routes(struct cluster *c_ptr, u32 dest); -void cluster_broadcast(struct sk_buff *buf); -int cluster_init(void); -u32 cluster_next_node(struct cluster *c_ptr, u32 addr); -void cluster_bcast_new_route(struct cluster *c_ptr, u32 dest, u32 lo, u32 hi); -void cluster_send_local_routes(struct cluster *c_ptr, u32 dest); -void cluster_bcast_lost_route(struct cluster *c_ptr, u32 dest, u32 lo, u32 hi); +void tipc_cltr_remove_as_router(struct cluster *c_ptr, u32 router); +void tipc_cltr_send_ext_routes(struct cluster *c_ptr, u32 dest); +struct node *tipc_cltr_select_node(struct cluster *c_ptr, u32 selector); +u32 tipc_cltr_select_router(struct cluster *c_ptr, u32 ref); +void tipc_cltr_recv_routing_table(struct sk_buff *buf); +struct cluster *tipc_cltr_create(u32 addr); +void tipc_cltr_delete(struct cluster *c_ptr); +void tipc_cltr_attach_node(struct cluster *c_ptr, struct node *n_ptr); +void tipc_cltr_send_slave_routes(struct cluster *c_ptr, u32 dest); +void tipc_cltr_broadcast(struct sk_buff *buf); +int tipc_cltr_init(void); +u32 tipc_cltr_next_node(struct cluster *c_ptr, u32 addr); +void tipc_cltr_bcast_new_route(struct cluster *c_ptr, u32 dest, u32 lo, u32 hi); +void tipc_cltr_send_local_routes(struct cluster *c_ptr, u32 dest); +void tipc_cltr_bcast_lost_route(struct cluster *c_ptr, u32 dest, u32 lo, u32 hi); -static inline struct cluster *cluster_find(u32 addr) +static inline struct cluster *tipc_cltr_find(u32 addr) { - struct _zone *z_ptr = zone_find(addr); + struct _zone *z_ptr = tipc_zone_find(addr); if (z_ptr) return z_ptr->clusters[1]; diff --git a/net/tipc/config.c b/net/tipc/config.c index 8ddef4fce2c..3c8e6740e5a 100644 --- a/net/tipc/config.c +++ b/net/tipc/config.c @@ -70,13 +70,13 @@ static int req_tlv_space; /* request message TLV area size */ static int rep_headroom; /* reply message headroom to use */ -void cfg_link_event(u32 addr, char *name, int up) +void tipc_cfg_link_event(u32 addr, char *name, int up) { /* TIPC DOESN'T HANDLE LINK EVENT SUBSCRIPTIONS AT THE MOMENT */ } -struct sk_buff *cfg_reply_alloc(int payload_size) +struct sk_buff *tipc_cfg_reply_alloc(int payload_size) { struct sk_buff *buf; @@ -86,14 +86,14 @@ struct sk_buff *cfg_reply_alloc(int payload_size) return buf; } -int cfg_append_tlv(struct sk_buff *buf, int tlv_type, - void *tlv_data, int tlv_data_size) +int tipc_cfg_append_tlv(struct sk_buff *buf, int tlv_type, + void *tlv_data, int tlv_data_size) { struct tlv_desc *tlv = (struct tlv_desc *)buf->tail; int new_tlv_space = TLV_SPACE(tlv_data_size); if (skb_tailroom(buf) < new_tlv_space) { - dbg("cfg_append_tlv unable to append TLV\n"); + dbg("tipc_cfg_append_tlv unable to append TLV\n"); return 0; } skb_put(buf, new_tlv_space); @@ -104,28 +104,28 @@ int cfg_append_tlv(struct sk_buff *buf, int tlv_type, return 1; } -struct sk_buff *cfg_reply_unsigned_type(u16 tlv_type, u32 value) +struct sk_buff *tipc_cfg_reply_unsigned_type(u16 tlv_type, u32 value) { struct sk_buff *buf; u32 value_net; - buf = cfg_reply_alloc(TLV_SPACE(sizeof(value))); + buf = tipc_cfg_reply_alloc(TLV_SPACE(sizeof(value))); if (buf) { value_net = htonl(value); - cfg_append_tlv(buf, tlv_type, &value_net, - sizeof(value_net)); + tipc_cfg_append_tlv(buf, tlv_type, &value_net, + sizeof(value_net)); } return buf; } -struct sk_buff *cfg_reply_string_type(u16 tlv_type, char *string) +struct sk_buff *tipc_cfg_reply_string_type(u16 tlv_type, char *string) { struct sk_buff *buf; int string_len = strlen(string) + 1; - buf = cfg_reply_alloc(TLV_SPACE(string_len)); + buf = tipc_cfg_reply_alloc(TLV_SPACE(string_len)); if (buf) - cfg_append_tlv(buf, tlv_type, string, string_len); + tipc_cfg_append_tlv(buf, tlv_type, string, string_len); return buf; } @@ -246,7 +246,7 @@ static void cfg_cmd_event(struct tipc_cmd_msg *msg, exit: rmsg.result_len = htonl(msg_sect[1].iov_len); rmsg.retval = htonl(rv); - cfg_respond(msg_sect, 2u, orig); + tipc_cfg_respond(msg_sect, 2u, orig); } #endif @@ -255,26 +255,26 @@ static struct sk_buff *cfg_enable_bearer(void) struct tipc_bearer_config *args; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_BEARER_CONFIG)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); args = (struct tipc_bearer_config *)TLV_DATA(req_tlv_area); if (tipc_enable_bearer(args->name, ntohl(args->detect_scope), ntohl(args->priority))) - return cfg_reply_error_string("unable to enable bearer"); + return tipc_cfg_reply_error_string("unable to enable bearer"); - return cfg_reply_none(); + return tipc_cfg_reply_none(); } static struct sk_buff *cfg_disable_bearer(void) { if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_BEARER_NAME)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); if (tipc_disable_bearer((char *)TLV_DATA(req_tlv_area))) - return cfg_reply_error_string("unable to disable bearer"); + return tipc_cfg_reply_error_string("unable to disable bearer"); - return cfg_reply_none(); + return tipc_cfg_reply_none(); } static struct sk_buff *cfg_set_own_addr(void) @@ -282,25 +282,25 @@ static struct sk_buff *cfg_set_own_addr(void) u32 addr; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); addr = *(u32 *)TLV_DATA(req_tlv_area); addr = ntohl(addr); if (addr == tipc_own_addr) - return cfg_reply_none(); - if (!addr_node_valid(addr)) - return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (node address)"); + return tipc_cfg_reply_none(); + if (!tipc_addr_node_valid(addr)) + return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (node address)"); if (tipc_own_addr) - return cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (cannot change node address once assigned)"); + return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (cannot change node address once assigned)"); spin_unlock_bh(&config_lock); - stop_net(); + tipc_core_stop_net(); tipc_own_addr = addr; - start_net(); + tipc_core_start_net(); spin_lock_bh(&config_lock); - return cfg_reply_none(); + return tipc_cfg_reply_none(); } static struct sk_buff *cfg_set_remote_mng(void) @@ -308,12 +308,12 @@ static struct sk_buff *cfg_set_remote_mng(void) u32 value; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); value = *(u32 *)TLV_DATA(req_tlv_area); value = ntohl(value); tipc_remote_management = (value != 0); - return cfg_reply_none(); + return tipc_cfg_reply_none(); } static struct sk_buff *cfg_set_max_publications(void) @@ -321,15 +321,15 @@ static struct sk_buff *cfg_set_max_publications(void) u32 value; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); value = *(u32 *)TLV_DATA(req_tlv_area); value = ntohl(value); if (value != delimit(value, 1, 65535)) - return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (max publications must be 1-65535)"); + return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (max publications must be 1-65535)"); tipc_max_publications = value; - return cfg_reply_none(); + return tipc_cfg_reply_none(); } static struct sk_buff *cfg_set_max_subscriptions(void) @@ -337,15 +337,15 @@ static struct sk_buff *cfg_set_max_subscriptions(void) u32 value; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); value = *(u32 *)TLV_DATA(req_tlv_area); value = ntohl(value); if (value != delimit(value, 1, 65535)) - return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (max subscriptions must be 1-65535"); + return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (max subscriptions must be 1-65535"); tipc_max_subscriptions = value; - return cfg_reply_none(); + return tipc_cfg_reply_none(); } static struct sk_buff *cfg_set_max_ports(void) @@ -354,31 +354,31 @@ static struct sk_buff *cfg_set_max_ports(void) u32 value; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); value = *(u32 *)TLV_DATA(req_tlv_area); value = ntohl(value); if (value != delimit(value, 127, 65535)) - return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (max ports must be 127-65535)"); + return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (max ports must be 127-65535)"); if (value == tipc_max_ports) - return cfg_reply_none(); + return tipc_cfg_reply_none(); if (atomic_read(&tipc_user_count) > 2) - return cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (cannot change max ports while TIPC users exist)"); + return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (cannot change max ports while TIPC users exist)"); spin_unlock_bh(&config_lock); orig_mode = tipc_get_mode(); if (orig_mode == TIPC_NET_MODE) - stop_net(); - stop_core(); + tipc_core_stop_net(); + tipc_core_stop(); tipc_max_ports = value; - start_core(); + tipc_core_start(); if (orig_mode == TIPC_NET_MODE) - start_net(); + tipc_core_start_net(); spin_lock_bh(&config_lock); - return cfg_reply_none(); + return tipc_cfg_reply_none(); } static struct sk_buff *set_net_max(int value, int *parameter) @@ -388,13 +388,13 @@ static struct sk_buff *set_net_max(int value, int *parameter) if (value != *parameter) { orig_mode = tipc_get_mode(); if (orig_mode == TIPC_NET_MODE) - stop_net(); + tipc_core_stop_net(); *parameter = value; if (orig_mode == TIPC_NET_MODE) - start_net(); + tipc_core_start_net(); } - return cfg_reply_none(); + return tipc_cfg_reply_none(); } static struct sk_buff *cfg_set_max_zones(void) @@ -402,12 +402,12 @@ static struct sk_buff *cfg_set_max_zones(void) u32 value; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); value = *(u32 *)TLV_DATA(req_tlv_area); value = ntohl(value); if (value != delimit(value, 1, 255)) - return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (max zones must be 1-255)"); + return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (max zones must be 1-255)"); return set_net_max(value, &tipc_max_zones); } @@ -416,13 +416,13 @@ static struct sk_buff *cfg_set_max_clusters(void) u32 value; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); value = *(u32 *)TLV_DATA(req_tlv_area); value = ntohl(value); if (value != 1) - return cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (max clusters fixed at 1)"); - return cfg_reply_none(); + return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (max clusters fixed at 1)"); + return tipc_cfg_reply_none(); } static struct sk_buff *cfg_set_max_nodes(void) @@ -430,12 +430,12 @@ static struct sk_buff *cfg_set_max_nodes(void) u32 value; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); value = *(u32 *)TLV_DATA(req_tlv_area); value = ntohl(value); if (value != delimit(value, 8, 2047)) - return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (max nodes must be 8-2047)"); + return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (max nodes must be 8-2047)"); return set_net_max(value, &tipc_max_nodes); } @@ -444,13 +444,13 @@ static struct sk_buff *cfg_set_max_slaves(void) u32 value; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); value = *(u32 *)TLV_DATA(req_tlv_area); value = ntohl(value); if (value != 0) - return cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (max secondary nodes fixed at 0)"); - return cfg_reply_none(); + return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (max secondary nodes fixed at 0)"); + return tipc_cfg_reply_none(); } static struct sk_buff *cfg_set_netid(void) @@ -458,22 +458,22 @@ static struct sk_buff *cfg_set_netid(void) u32 value; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); value = *(u32 *)TLV_DATA(req_tlv_area); value = ntohl(value); if (value != delimit(value, 1, 9999)) - return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (network id must be 1-9999)"); + return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (network id must be 1-9999)"); if (tipc_own_addr) - return cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (cannot change network id once part of network)"); + return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (cannot change network id once part of network)"); return set_net_max(value, &tipc_net_id); } -struct sk_buff *cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area, - int request_space, int reply_headroom) +struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area, + int request_space, int reply_headroom) { struct sk_buff *rep_tlv_buf; @@ -490,19 +490,19 @@ struct sk_buff *cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area, if (likely(orig_node == tipc_own_addr)) { /* command is permitted */ } else if (cmd >= 0x8000) { - rep_tlv_buf = cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (cannot be done remotely)"); + rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (cannot be done remotely)"); goto exit; } else if (!tipc_remote_management) { - rep_tlv_buf = cfg_reply_error_string(TIPC_CFG_NO_REMOTE); + rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NO_REMOTE); goto exit; } else if (cmd >= 0x4000) { u32 domain = 0; - if ((nametbl_translate(TIPC_ZM_SRV, 0, &domain) == 0) || + if ((tipc_nametbl_translate(TIPC_ZM_SRV, 0, &domain) == 0) || (domain != orig_node)) { - rep_tlv_buf = cfg_reply_error_string(TIPC_CFG_NOT_ZONE_MSTR); + rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_ZONE_MSTR); goto exit; } } @@ -511,50 +511,50 @@ struct sk_buff *cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area, switch (cmd) { case TIPC_CMD_NOOP: - rep_tlv_buf = cfg_reply_none(); + rep_tlv_buf = tipc_cfg_reply_none(); break; case TIPC_CMD_GET_NODES: - rep_tlv_buf = node_get_nodes(req_tlv_area, req_tlv_space); + rep_tlv_buf = tipc_node_get_nodes(req_tlv_area, req_tlv_space); break; case TIPC_CMD_GET_LINKS: - rep_tlv_buf = node_get_links(req_tlv_area, req_tlv_space); + rep_tlv_buf = tipc_node_get_links(req_tlv_area, req_tlv_space); break; case TIPC_CMD_SHOW_LINK_STATS: - rep_tlv_buf = link_cmd_show_stats(req_tlv_area, req_tlv_space); + rep_tlv_buf = tipc_link_cmd_show_stats(req_tlv_area, req_tlv_space); break; case TIPC_CMD_RESET_LINK_STATS: - rep_tlv_buf = link_cmd_reset_stats(req_tlv_area, req_tlv_space); + rep_tlv_buf = tipc_link_cmd_reset_stats(req_tlv_area, req_tlv_space); break; case TIPC_CMD_SHOW_NAME_TABLE: - rep_tlv_buf = nametbl_get(req_tlv_area, req_tlv_space); + rep_tlv_buf = tipc_nametbl_get(req_tlv_area, req_tlv_space); break; case TIPC_CMD_GET_BEARER_NAMES: - rep_tlv_buf = bearer_get_names(); + rep_tlv_buf = tipc_bearer_get_names(); break; case TIPC_CMD_GET_MEDIA_NAMES: - rep_tlv_buf = media_get_names(); + rep_tlv_buf = tipc_media_get_names(); break; case TIPC_CMD_SHOW_PORTS: - rep_tlv_buf = port_get_ports(); + rep_tlv_buf = tipc_port_get_ports(); break; #if 0 case TIPC_CMD_SHOW_PORT_STATS: rep_tlv_buf = port_show_stats(req_tlv_area, req_tlv_space); break; case TIPC_CMD_RESET_PORT_STATS: - rep_tlv_buf = cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED); + rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED); break; #endif case TIPC_CMD_SET_LOG_SIZE: - rep_tlv_buf = log_resize(req_tlv_area, req_tlv_space); + rep_tlv_buf = tipc_log_resize(req_tlv_area, req_tlv_space); break; case TIPC_CMD_DUMP_LOG: - rep_tlv_buf = log_dump(); + rep_tlv_buf = tipc_log_dump(); break; case TIPC_CMD_SET_LINK_TOL: case TIPC_CMD_SET_LINK_PRI: case TIPC_CMD_SET_LINK_WINDOW: - rep_tlv_buf = link_cmd_config(req_tlv_area, req_tlv_space, cmd); + rep_tlv_buf = tipc_link_cmd_config(req_tlv_area, req_tlv_space, cmd); break; case TIPC_CMD_ENABLE_BEARER: rep_tlv_buf = cfg_enable_bearer(); @@ -593,31 +593,31 @@ struct sk_buff *cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area, rep_tlv_buf = cfg_set_netid(); break; case TIPC_CMD_GET_REMOTE_MNG: - rep_tlv_buf = cfg_reply_unsigned(tipc_remote_management); + rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_remote_management); break; case TIPC_CMD_GET_MAX_PORTS: - rep_tlv_buf = cfg_reply_unsigned(tipc_max_ports); + rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_ports); break; case TIPC_CMD_GET_MAX_PUBL: - rep_tlv_buf = cfg_reply_unsigned(tipc_max_publications); + rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_publications); break; case TIPC_CMD_GET_MAX_SUBSCR: - rep_tlv_buf = cfg_reply_unsigned(tipc_max_subscriptions); + rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_subscriptions); break; case TIPC_CMD_GET_MAX_ZONES: - rep_tlv_buf = cfg_reply_unsigned(tipc_max_zones); + rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_zones); break; case TIPC_CMD_GET_MAX_CLUSTERS: - rep_tlv_buf = cfg_reply_unsigned(tipc_max_clusters); + rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_clusters); break; case TIPC_CMD_GET_MAX_NODES: - rep_tlv_buf = cfg_reply_unsigned(tipc_max_nodes); + rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_nodes); break; case TIPC_CMD_GET_MAX_SLAVES: - rep_tlv_buf = cfg_reply_unsigned(tipc_max_slaves); + rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_slaves); break; case TIPC_CMD_GET_NETID: - rep_tlv_buf = cfg_reply_unsigned(tipc_net_id); + rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_net_id); break; default: rep_tlv_buf = NULL; @@ -655,11 +655,11 @@ static void cfg_named_msg_event(void *userdata, /* Generate reply for request (if can't, return request) */ - rep_buf = cfg_do_cmd(orig->node, - ntohs(req_hdr->tcm_type), - msg + sizeof(*req_hdr), - size - sizeof(*req_hdr), - BUF_HEADROOM + MAX_H_SIZE + sizeof(*rep_hdr)); + rep_buf = tipc_cfg_do_cmd(orig->node, + ntohs(req_hdr->tcm_type), + msg + sizeof(*req_hdr), + size - sizeof(*req_hdr), + BUF_HEADROOM + MAX_H_SIZE + sizeof(*rep_hdr)); if (rep_buf) { skb_push(rep_buf, sizeof(*rep_hdr)); rep_hdr = (struct tipc_cfg_msg_hdr *)rep_buf->data; @@ -675,7 +675,7 @@ static void cfg_named_msg_event(void *userdata, tipc_send_buf2port(port_ref, orig, rep_buf, rep_buf->len); } -int cfg_init(void) +int tipc_cfg_init(void) { struct tipc_name_seq seq; int res; @@ -696,7 +696,7 @@ int cfg_init(void) seq.type = TIPC_CFG_SRV; seq.lower = seq.upper = tipc_own_addr; - res = nametbl_publish_rsv(mng.port_ref, TIPC_ZONE_SCOPE, &seq); + res = tipc_nametbl_publish_rsv(mng.port_ref, TIPC_ZONE_SCOPE, &seq); if (res) goto failed; @@ -709,7 +709,7 @@ failed: return res; } -void cfg_stop(void) +void tipc_cfg_stop(void) { if (mng.user_ref) { tipc_detach(mng.user_ref); diff --git a/net/tipc/config.h b/net/tipc/config.h index e74d94f753c..7a728f954d8 100644 --- a/net/tipc/config.h +++ b/net/tipc/config.h @@ -42,38 +42,38 @@ #include "core.h" #include "link.h" -struct sk_buff *cfg_reply_alloc(int payload_size); -int cfg_append_tlv(struct sk_buff *buf, int tlv_type, - void *tlv_data, int tlv_data_size); -struct sk_buff *cfg_reply_unsigned_type(u16 tlv_type, u32 value); -struct sk_buff *cfg_reply_string_type(u16 tlv_type, char *string); +struct sk_buff *tipc_cfg_reply_alloc(int payload_size); +int tipc_cfg_append_tlv(struct sk_buff *buf, int tlv_type, + void *tlv_data, int tlv_data_size); +struct sk_buff *tipc_cfg_reply_unsigned_type(u16 tlv_type, u32 value); +struct sk_buff *tipc_cfg_reply_string_type(u16 tlv_type, char *string); -static inline struct sk_buff *cfg_reply_none(void) +static inline struct sk_buff *tipc_cfg_reply_none(void) { - return cfg_reply_alloc(0); + return tipc_cfg_reply_alloc(0); } -static inline struct sk_buff *cfg_reply_unsigned(u32 value) +static inline struct sk_buff *tipc_cfg_reply_unsigned(u32 value) { - return cfg_reply_unsigned_type(TIPC_TLV_UNSIGNED, value); + return tipc_cfg_reply_unsigned_type(TIPC_TLV_UNSIGNED, value); } -static inline struct sk_buff *cfg_reply_error_string(char *string) +static inline struct sk_buff *tipc_cfg_reply_error_string(char *string) { - return cfg_reply_string_type(TIPC_TLV_ERROR_STRING, string); + return tipc_cfg_reply_string_type(TIPC_TLV_ERROR_STRING, string); } -static inline struct sk_buff *cfg_reply_ultra_string(char *string) +static inline struct sk_buff *tipc_cfg_reply_ultra_string(char *string) { - return cfg_reply_string_type(TIPC_TLV_ULTRA_STRING, string); + return tipc_cfg_reply_string_type(TIPC_TLV_ULTRA_STRING, string); } -struct sk_buff *cfg_do_cmd(u32 orig_node, u16 cmd, - const void *req_tlv_area, int req_tlv_space, - int headroom); +struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, + const void *req_tlv_area, int req_tlv_space, + int headroom); -void cfg_link_event(u32 addr, char *name, int up); -int cfg_init(void); -void cfg_stop(void); +void tipc_cfg_link_event(u32 addr, char *name, int up); +int tipc_cfg_init(void); +void tipc_cfg_stop(void); #endif diff --git a/net/tipc/core.c b/net/tipc/core.c index 9a1ab178b44..3d0a8ee4e1d 100644 --- a/net/tipc/core.c +++ b/net/tipc/core.c @@ -48,14 +48,14 @@ #include "subscr.h" #include "config.h" -int eth_media_start(void); -void eth_media_stop(void); -int handler_start(void); -void handler_stop(void); -int socket_init(void); -void socket_stop(void); -int netlink_start(void); -void netlink_stop(void); +int tipc_eth_media_start(void); +void tipc_eth_media_stop(void); +int tipc_handler_start(void); +void tipc_handler_stop(void); +int tipc_socket_init(void); +void tipc_socket_stop(void); +int tipc_netlink_start(void); +void tipc_netlink_stop(void); #define MOD_NAME "tipc_start: " @@ -112,56 +112,56 @@ int tipc_get_mode(void) } /** - * stop_net - shut down TIPC networking sub-systems + * tipc_core_stop_net - shut down TIPC networking sub-systems */ -void stop_net(void) +void tipc_core_stop_net(void) { - eth_media_stop(); - tipc_stop_net(); + tipc_eth_media_stop(); + tipc_net_stop(); } /** * start_net - start TIPC networking sub-systems */ -int start_net(void) +int tipc_core_start_net(void) { int res; - if ((res = tipc_start_net()) || - (res = eth_media_start())) { - stop_net(); + if ((res = tipc_net_start()) || + (res = tipc_eth_media_start())) { + tipc_core_stop_net(); } return res; } /** - * stop_core - switch TIPC from SINGLE NODE to NOT RUNNING mode + * tipc_core_stop - switch TIPC from SINGLE NODE to NOT RUNNING mode */ -void stop_core(void) +void tipc_core_stop(void) { if (tipc_mode != TIPC_NODE_MODE) return; tipc_mode = TIPC_NOT_RUNNING; - netlink_stop(); - handler_stop(); - cfg_stop(); - subscr_stop(); - reg_stop(); - nametbl_stop(); - ref_table_stop(); - socket_stop(); + tipc_netlink_stop(); + tipc_handler_stop(); + tipc_cfg_stop(); + tipc_subscr_stop(); + tipc_reg_stop(); + tipc_nametbl_stop(); + tipc_ref_table_stop(); + tipc_socket_stop(); } /** - * start_core - switch TIPC from NOT RUNNING to SINGLE NODE mode + * tipc_core_start - switch TIPC from NOT RUNNING to SINGLE NODE mode */ -int start_core(void) +int tipc_core_start(void) { int res; @@ -171,16 +171,16 @@ int start_core(void) get_random_bytes(&tipc_random, sizeof(tipc_random)); tipc_mode = TIPC_NODE_MODE; - if ((res = handler_start()) || - (res = ref_table_init(tipc_max_ports + tipc_max_subscriptions, - tipc_random)) || - (res = reg_start()) || - (res = nametbl_init()) || - (res = k_signal((Handler)subscr_start, 0)) || - (res = k_signal((Handler)cfg_init, 0)) || - (res = netlink_start()) || - (res = socket_init())) { - stop_core(); + if ((res = tipc_handler_start()) || + (res = tipc_ref_table_init(tipc_max_ports + tipc_max_subscriptions, + tipc_random)) || + (res = tipc_reg_start()) || + (res = tipc_nametbl_init()) || + (res = tipc_k_signal((Handler)tipc_subscr_start, 0)) || + (res = tipc_k_signal((Handler)tipc_cfg_init, 0)) || + (res = tipc_netlink_start()) || + (res = tipc_socket_init())) { + tipc_core_stop(); } return res; } @@ -190,7 +190,7 @@ static int __init tipc_init(void) { int res; - log_reinit(CONFIG_TIPC_LOG); + tipc_log_reinit(CONFIG_TIPC_LOG); info("Activated (compiled " __DATE__ " " __TIME__ ")\n"); tipc_own_addr = 0; @@ -204,7 +204,7 @@ static int __init tipc_init(void) tipc_max_slaves = delimit(CONFIG_TIPC_SLAVE_NODES, 0, 2047); tipc_net_id = 4711; - if ((res = start_core())) + if ((res = tipc_core_start())) err("Unable to start in single node mode\n"); else info("Started in single node mode\n"); @@ -213,10 +213,10 @@ static int __init tipc_init(void) static void __exit tipc_exit(void) { - stop_net(); - stop_core(); + tipc_core_stop_net(); + tipc_core_stop(); info("Deactivated\n"); - log_stop(); + tipc_log_stop(); } module_init(tipc_init); diff --git a/net/tipc/core.h b/net/tipc/core.h index 1a34cc95b9e..1f2e8b27a13 100644 --- a/net/tipc/core.h +++ b/net/tipc/core.h @@ -65,9 +65,9 @@ #define assert(i) BUG_ON(!(i)) struct tipc_msg; -extern struct print_buf *CONS, *LOG; -extern struct print_buf *TEE(struct print_buf *, struct print_buf *); -void msg_print(struct print_buf*,struct tipc_msg *,const char*); +extern struct print_buf *TIPC_CONS, *TIPC_LOG; +extern struct print_buf *TIPC_TEE(struct print_buf *, struct print_buf *); +void tipc_msg_print(struct print_buf*,struct tipc_msg *,const char*); void tipc_printf(struct print_buf *, const char *fmt, ...); void tipc_dump(struct print_buf*,const char *fmt, ...); @@ -84,7 +84,7 @@ void tipc_dump(struct print_buf*,const char *fmt, ...); #define info(fmt, arg...) tipc_printf(TIPC_OUTPUT, KERN_NOTICE "TIPC: " fmt, ## arg) #define dbg(fmt, arg...) do {if (DBG_OUTPUT) tipc_printf(DBG_OUTPUT, fmt, ## arg);} while(0) -#define msg_dbg(msg, txt) do {if (DBG_OUTPUT) msg_print(DBG_OUTPUT, msg, txt);} while(0) +#define msg_dbg(msg, txt) do {if (DBG_OUTPUT) tipc_msg_print(DBG_OUTPUT, msg, txt);} while(0) #define dump(fmt, arg...) do {if (DBG_OUTPUT) tipc_dump(DBG_OUTPUT, fmt, ##arg);} while(0) @@ -94,15 +94,15 @@ void tipc_dump(struct print_buf*,const char *fmt, ...); * here, or on a per .c file basis, by redefining these symbols. The following * print buffer options are available: * - * NULL : Output to null print buffer (i.e. print nowhere) - * CONS : Output to system console - * LOG : Output to TIPC log buffer - * &buf : Output to user-defined buffer (struct print_buf *) - * TEE(&buf_a,&buf_b) : Output to two print buffers (eg. TEE(CONS,LOG) ) + * NULL : Output to null print buffer (i.e. print nowhere) + * TIPC_CONS : Output to system console + * TIPC_LOG : Output to TIPC log buffer + * &buf : Output to user-defined buffer (struct print_buf *) + * TIPC_TEE(&buf_a,&buf_b) : Output to two print buffers (eg. TIPC_TEE(TIPC_CONS,TIPC_LOG) ) */ #ifndef TIPC_OUTPUT -#define TIPC_OUTPUT TEE(CONS,LOG) +#define TIPC_OUTPUT TIPC_TEE(TIPC_CONS,TIPC_LOG) #endif #ifndef DBG_OUTPUT @@ -167,10 +167,10 @@ extern atomic_t tipc_user_count; * Routines available to privileged subsystems */ -extern int start_core(void); -extern void stop_core(void); -extern int start_net(void); -extern void stop_net(void); +extern int tipc_core_start(void); +extern void tipc_core_stop(void); +extern int tipc_core_start_net(void); +extern void tipc_core_stop_net(void); static inline int delimit(int val, int min, int max) { @@ -188,7 +188,7 @@ static inline int delimit(int val, int min, int max) typedef void (*Handler) (unsigned long); -u32 k_signal(Handler routine, unsigned long argument); +u32 tipc_k_signal(Handler routine, unsigned long argument); /** * k_init_timer - initialize a timer diff --git a/net/tipc/dbg.c b/net/tipc/dbg.c index 7ed60a1cfbb..4f4beefa783 100644 --- a/net/tipc/dbg.c +++ b/net/tipc/dbg.c @@ -44,10 +44,10 @@ static char print_string[MAX_STRING]; static spinlock_t print_lock = SPIN_LOCK_UNLOCKED; static struct print_buf cons_buf = { NULL, 0, NULL, NULL }; -struct print_buf *CONS = &cons_buf; +struct print_buf *TIPC_CONS = &cons_buf; static struct print_buf log_buf = { NULL, 0, NULL, NULL }; -struct print_buf *LOG = &log_buf; +struct print_buf *TIPC_LOG = &log_buf; #define FORMAT(PTR,LEN,FMT) \ @@ -66,15 +66,15 @@ struct print_buf *LOG = &log_buf; * simultaneous use of the print buffer(s) being manipulated. * 2) tipc_printf() uses 'print_lock' to prevent simultaneous use of * 'print_string' and to protect its print buffer(s). - * 3) TEE() uses 'print_lock' to protect its print buffer(s). - * 4) Routines of the form log_XXX() uses 'print_lock' to protect LOG. + * 3) TIPC_TEE() uses 'print_lock' to protect its print buffer(s). + * 4) Routines of the form log_XXX() uses 'print_lock' to protect TIPC_LOG. */ /** - * printbuf_init - initialize print buffer to empty + * tipc_printbuf_init - initialize print buffer to empty */ -void printbuf_init(struct print_buf *pb, char *raw, u32 sz) +void tipc_printbuf_init(struct print_buf *pb, char *raw, u32 sz) { if (!pb || !raw || (sz < (MAX_STRING + 1))) return; @@ -87,26 +87,26 @@ void printbuf_init(struct print_buf *pb, char *raw, u32 sz) } /** - * printbuf_reset - reinitialize print buffer to empty state + * tipc_printbuf_reset - reinitialize print buffer to empty state */ -void printbuf_reset(struct print_buf *pb) +void tipc_printbuf_reset(struct print_buf *pb) { if (pb && pb->buf) - printbuf_init(pb, pb->buf, pb->size); + tipc_printbuf_init(pb, pb->buf, pb->size); } /** - * printbuf_empty - test if print buffer is in empty state + * tipc_printbuf_empty - test if print buffer is in empty state */ -int printbuf_empty(struct print_buf *pb) +int tipc_printbuf_empty(struct print_buf *pb) { return (!pb || !pb->buf || (pb->crs == pb->buf)); } /** - * printbuf_validate - check for print buffer overflow + * tipc_printbuf_validate - check for print buffer overflow * * Verifies that a print buffer has captured all data written to it. * If data has been lost, linearize buffer and prepend an error message @@ -114,7 +114,7 @@ int printbuf_empty(struct print_buf *pb) * Returns length of print buffer data string (including trailing NULL) */ -int printbuf_validate(struct print_buf *pb) +int tipc_printbuf_validate(struct print_buf *pb) { char *err = " *** PRINT BUFFER WRAPPED AROUND ***\n"; char *cp_buf; @@ -126,13 +126,13 @@ int printbuf_validate(struct print_buf *pb) if (pb->buf[pb->size - 1] == '\0') { cp_buf = kmalloc(pb->size, GFP_ATOMIC); if (cp_buf != NULL){ - printbuf_init(&cb, cp_buf, pb->size); - printbuf_move(&cb, pb); - printbuf_move(pb, &cb); + tipc_printbuf_init(&cb, cp_buf, pb->size); + tipc_printbuf_move(&cb, pb); + tipc_printbuf_move(pb, &cb); kfree(cp_buf); memcpy(pb->buf, err, strlen(err)); } else { - printbuf_reset(pb); + tipc_printbuf_reset(pb); tipc_printf(pb, err); } } @@ -140,13 +140,13 @@ int printbuf_validate(struct print_buf *pb) } /** - * printbuf_move - move print buffer contents to another print buffer + * tipc_printbuf_move - move print buffer contents to another print buffer * * Current contents of destination print buffer (if any) are discarded. * Source print buffer becomes empty if a successful move occurs. */ -void printbuf_move(struct print_buf *pb_to, struct print_buf *pb_from) +void tipc_printbuf_move(struct print_buf *pb_to, struct print_buf *pb_from) { int len; @@ -156,12 +156,12 @@ void printbuf_move(struct print_buf *pb_to, struct print_buf *pb_from) return; if (!pb_from || !pb_from->buf) { - printbuf_reset(pb_to); + tipc_printbuf_reset(pb_to); return; } if (pb_to->size < pb_from->size) { - printbuf_reset(pb_to); + tipc_printbuf_reset(pb_to); tipc_printf(pb_to, "*** PRINT BUFFER OVERFLOW ***"); return; } @@ -179,7 +179,7 @@ void printbuf_move(struct print_buf *pb_to, struct print_buf *pb_from) strcpy(pb_to->crs, pb_from->buf); pb_to->crs += len; - printbuf_reset(pb_from); + tipc_printbuf_reset(pb_from); } /** @@ -199,7 +199,7 @@ void tipc_printf(struct print_buf *pb, const char *fmt, ...) strcpy(print_string, "*** STRING TOO LONG ***"); while (pb) { - if (pb == CONS) + if (pb == TIPC_CONS) printk(print_string); else if (pb->buf) { chars_left = pb->buf + pb->size - pb->crs - 1; @@ -223,10 +223,10 @@ void tipc_printf(struct print_buf *pb, const char *fmt, ...) } /** - * TEE - perform next output operation on both print buffers + * TIPC_TEE - perform next output operation on both print buffers */ -struct print_buf *TEE(struct print_buf *b0, struct print_buf *b1) +struct print_buf *TIPC_TEE(struct print_buf *b0, struct print_buf *b1) { struct print_buf *pb = b0; @@ -294,96 +294,96 @@ void tipc_dump(struct print_buf *pb, const char *fmt, ...) int len; spin_lock_bh(&print_lock); - FORMAT(CONS->buf, len, fmt); - printk(CONS->buf); + FORMAT(TIPC_CONS->buf, len, fmt); + printk(TIPC_CONS->buf); for (; pb; pb = pb->next) { - if (pb == CONS) + if (pb == TIPC_CONS) continue; printk("\n---- Start of dump,%s log ----\n\n", - (pb == LOG) ? "global" : "local"); + (pb == TIPC_LOG) ? "global" : "local"); printbuf_dump(pb); - printbuf_reset(pb); + tipc_printbuf_reset(pb); printk("\n-------- End of dump --------\n"); } spin_unlock_bh(&print_lock); } /** - * log_stop - free up TIPC log print buffer + * tipc_log_stop - free up TIPC log print buffer */ -void log_stop(void) +void tipc_log_stop(void) { spin_lock_bh(&print_lock); - if (LOG->buf) { - kfree(LOG->buf); - LOG->buf = NULL; + if (TIPC_LOG->buf) { + kfree(TIPC_LOG->buf); + TIPC_LOG->buf = NULL; } spin_unlock_bh(&print_lock); } /** - * log_reinit - set TIPC log print buffer to specified size + * tipc_log_reinit - set TIPC log print buffer to specified size */ -void log_reinit(int log_size) +void tipc_log_reinit(int log_size) { - log_stop(); + tipc_log_stop(); if (log_size) { if (log_size <= MAX_STRING) log_size = MAX_STRING + 1; spin_lock_bh(&print_lock); - printbuf_init(LOG, kmalloc(log_size, GFP_ATOMIC), log_size); + tipc_printbuf_init(TIPC_LOG, kmalloc(log_size, GFP_ATOMIC), log_size); spin_unlock_bh(&print_lock); } } /** - * log_resize - reconfigure size of TIPC log buffer + * tipc_log_resize - reconfigure size of TIPC log buffer */ -struct sk_buff *log_resize(const void *req_tlv_area, int req_tlv_space) +struct sk_buff *tipc_log_resize(const void *req_tlv_area, int req_tlv_space) { u32 value; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); value = *(u32 *)TLV_DATA(req_tlv_area); value = ntohl(value); if (value != delimit(value, 0, 32768)) - return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (log size must be 0-32768)"); - log_reinit(value); - return cfg_reply_none(); + return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (log size must be 0-32768)"); + tipc_log_reinit(value); + return tipc_cfg_reply_none(); } /** - * log_dump - capture TIPC log buffer contents in configuration message + * tipc_log_dump - capture TIPC log buffer contents in configuration message */ -struct sk_buff *log_dump(void) +struct sk_buff *tipc_log_dump(void) { struct sk_buff *reply; spin_lock_bh(&print_lock); - if (!LOG->buf) - reply = cfg_reply_ultra_string("log not activated\n"); - else if (printbuf_empty(LOG)) - reply = cfg_reply_ultra_string("log is empty\n"); + if (!TIPC_LOG->buf) + reply = tipc_cfg_reply_ultra_string("log not activated\n"); + else if (tipc_printbuf_empty(TIPC_LOG)) + reply = tipc_cfg_reply_ultra_string("log is empty\n"); else { struct tlv_desc *rep_tlv; struct print_buf pb; int str_len; - str_len = min(LOG->size, 32768u); - reply = cfg_reply_alloc(TLV_SPACE(str_len)); + str_len = min(TIPC_LOG->size, 32768u); + reply = tipc_cfg_reply_alloc(TLV_SPACE(str_len)); if (reply) { rep_tlv = (struct tlv_desc *)reply->data; - printbuf_init(&pb, TLV_DATA(rep_tlv), str_len); - printbuf_move(&pb, LOG); + tipc_printbuf_init(&pb, TLV_DATA(rep_tlv), str_len); + tipc_printbuf_move(&pb, TIPC_LOG); str_len = strlen(TLV_DATA(rep_tlv)) + 1; skb_put(reply, TLV_SPACE(str_len)); TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); diff --git a/net/tipc/dbg.h b/net/tipc/dbg.h index c6b2a64c224..227f050d2a5 100644 --- a/net/tipc/dbg.h +++ b/net/tipc/dbg.h @@ -44,16 +44,16 @@ struct print_buf { struct print_buf *next; }; -void printbuf_init(struct print_buf *pb, char *buf, u32 sz); -void printbuf_reset(struct print_buf *pb); -int printbuf_empty(struct print_buf *pb); -int printbuf_validate(struct print_buf *pb); -void printbuf_move(struct print_buf *pb_to, struct print_buf *pb_from); +void tipc_printbuf_init(struct print_buf *pb, char *buf, u32 sz); +void tipc_printbuf_reset(struct print_buf *pb); +int tipc_printbuf_empty(struct print_buf *pb); +int tipc_printbuf_validate(struct print_buf *pb); +void tipc_printbuf_move(struct print_buf *pb_to, struct print_buf *pb_from); -void log_reinit(int log_size); -void log_stop(void); +void tipc_log_reinit(int log_size); +void tipc_log_stop(void); -struct sk_buff *log_resize(const void *req_tlv_area, int req_tlv_space); -struct sk_buff *log_dump(void); +struct sk_buff *tipc_log_resize(const void *req_tlv_area, int req_tlv_space); +struct sk_buff *tipc_log_dump(void); #endif diff --git a/net/tipc/discover.c b/net/tipc/discover.c index b106ef1621c..53ba4630c10 100644 --- a/net/tipc/discover.c +++ b/net/tipc/discover.c @@ -93,7 +93,7 @@ int disc_create_link(const struct tipc_link_create *argv) * disc_lost_link(): A link has lost contact */ -void disc_link_event(u32 addr, char *name, int up) +void tipc_disc_link_event(u32 addr, char *name, int up) { if (in_own_cluster(addr)) return; @@ -103,17 +103,17 @@ void disc_link_event(u32 addr, char *name, int up) } /** - * disc_init_msg - initialize a link setup message + * tipc_disc_init_msg - initialize a link setup message * @type: message type (request or response) * @req_links: number of links associated with message * @dest_domain: network domain of node(s) which should respond to message * @b_ptr: ptr to bearer issuing message */ -struct sk_buff *disc_init_msg(u32 type, - u32 req_links, - u32 dest_domain, - struct bearer *b_ptr) +struct sk_buff *tipc_disc_init_msg(u32 type, + u32 req_links, + u32 dest_domain, + struct bearer *b_ptr) { struct sk_buff *buf = buf_acquire(DSC_H_SIZE); struct tipc_msg *msg; @@ -132,11 +132,11 @@ struct sk_buff *disc_init_msg(u32 type, } /** - * disc_recv_msg - handle incoming link setup message (request or response) + * tipc_disc_recv_msg - handle incoming link setup message (request or response) * @buf: buffer containing message */ -void disc_recv_msg(struct sk_buff *buf) +void tipc_disc_recv_msg(struct sk_buff *buf) { struct bearer *b_ptr = (struct bearer *)TIPC_SKB_CB(buf)->handle; struct link *link; @@ -153,9 +153,9 @@ void disc_recv_msg(struct sk_buff *buf) if (net_id != tipc_net_id) return; - if (!addr_domain_valid(dest)) + if (!tipc_addr_domain_valid(dest)) return; - if (!addr_node_valid(orig)) + if (!tipc_addr_node_valid(orig)) return; if (orig == tipc_own_addr) return; @@ -169,11 +169,11 @@ void disc_recv_msg(struct sk_buff *buf) /* Always accept link here */ struct sk_buff *rbuf; struct tipc_media_addr *addr; - struct node *n_ptr = node_find(orig); + struct node *n_ptr = tipc_node_find(orig); int link_up; dbg(" in own cluster\n"); if (n_ptr == NULL) { - n_ptr = node_create(orig); + n_ptr = tipc_node_create(orig); } if (n_ptr == NULL) { warn("Memory squeeze; Failed to create node\n"); @@ -183,7 +183,7 @@ void disc_recv_msg(struct sk_buff *buf) link = n_ptr->links[b_ptr->identity]; if (!link) { dbg("creating link\n"); - link = link_create(b_ptr, orig, &media_addr); + link = tipc_link_create(b_ptr, orig, &media_addr); if (!link) { spin_unlock_bh(&n_ptr->lock); return; @@ -196,13 +196,13 @@ void disc_recv_msg(struct sk_buff *buf) warn("New bearer address for %s\n", addr_string_fill(addr_string, orig)); memcpy(addr, &media_addr, sizeof(*addr)); - link_reset(link); + tipc_link_reset(link); } - link_up = link_is_up(link); + link_up = tipc_link_is_up(link); spin_unlock_bh(&n_ptr->lock); if ((type == DSC_RESP_MSG) || link_up) return; - rbuf = disc_init_msg(DSC_RESP_MSG, 1, orig, b_ptr); + rbuf = tipc_disc_init_msg(DSC_RESP_MSG, 1, orig, b_ptr); if (rbuf != NULL) { msg_dbg(buf_msg(rbuf),"SEND:"); b_ptr->media->send_msg(rbuf, &b_ptr->publ, &media_addr); @@ -212,11 +212,11 @@ void disc_recv_msg(struct sk_buff *buf) } /** - * disc_stop_link_req - stop sending periodic link setup requests + * tipc_disc_stop_link_req - stop sending periodic link setup requests * @req: ptr to link request structure */ -void disc_stop_link_req(struct link_req *req) +void tipc_disc_stop_link_req(struct link_req *req) { if (!req) return; @@ -228,11 +228,11 @@ void disc_stop_link_req(struct link_req *req) } /** - * disc_update_link_req - update frequency of periodic link setup requests + * tipc_disc_update_link_req - update frequency of periodic link setup requests * @req: ptr to link request structure */ -void disc_update_link_req(struct link_req *req) +void tipc_disc_update_link_req(struct link_req *req) { if (!req) return; @@ -282,7 +282,7 @@ static void disc_timeout(struct link_req *req) } /** - * disc_init_link_req - start sending periodic link setup requests + * tipc_disc_init_link_req - start sending periodic link setup requests * @b_ptr: ptr to bearer issuing requests * @dest: destination address for request messages * @dest_domain: network domain of node(s) which should respond to message @@ -291,10 +291,10 @@ static void disc_timeout(struct link_req *req) * Returns pointer to link request structure, or NULL if unable to create. */ -struct link_req *disc_init_link_req(struct bearer *b_ptr, - const struct tipc_media_addr *dest, - u32 dest_domain, - u32 req_links) +struct link_req *tipc_disc_init_link_req(struct bearer *b_ptr, + const struct tipc_media_addr *dest, + u32 dest_domain, + u32 req_links) { struct link_req *req; @@ -302,7 +302,7 @@ struct link_req *disc_init_link_req(struct bearer *b_ptr, if (!req) return NULL; - req->buf = disc_init_msg(DSC_REQ_MSG, req_links, dest_domain, b_ptr); + req->buf = tipc_disc_init_msg(DSC_REQ_MSG, req_links, dest_domain, b_ptr); if (!req->buf) { kfree(req); return NULL; diff --git a/net/tipc/discover.h b/net/tipc/discover.h index f4acb360d6c..0454fd1ae7f 100644 --- a/net/tipc/discover.h +++ b/net/tipc/discover.h @@ -41,16 +41,16 @@ struct link_req; -struct link_req *disc_init_link_req(struct bearer *b_ptr, - const struct tipc_media_addr *dest, - u32 dest_domain, - u32 req_links); -void disc_update_link_req(struct link_req *req); -void disc_stop_link_req(struct link_req *req); +struct link_req *tipc_disc_init_link_req(struct bearer *b_ptr, + const struct tipc_media_addr *dest, + u32 dest_domain, + u32 req_links); +void tipc_disc_update_link_req(struct link_req *req); +void tipc_disc_stop_link_req(struct link_req *req); -void disc_recv_msg(struct sk_buff *buf); +void tipc_disc_recv_msg(struct sk_buff *buf); -void disc_link_event(u32 addr, char *name, int up); +void tipc_disc_link_event(u32 addr, char *name, int up); #if 0 int disc_create_link(const struct tipc_link_create *argv); #endif diff --git a/net/tipc/eth_media.c b/net/tipc/eth_media.c index cb507c4b7e6..1f8d83b9c8b 100644 --- a/net/tipc/eth_media.c +++ b/net/tipc/eth_media.c @@ -238,13 +238,13 @@ static char *eth_addr2str(struct tipc_media_addr *a, char *str_buf, int str_size } /** - * eth_media_start - activate Ethernet bearer support + * tipc_eth_media_start - activate Ethernet bearer support * * Register Ethernet media type with TIPC bearer code. Also register * with OS for notifications about device state changes. */ -int eth_media_start(void) +int tipc_eth_media_start(void) { struct tipc_media_addr bcast_addr; int res; @@ -271,10 +271,10 @@ int eth_media_start(void) } /** - * eth_media_stop - deactivate Ethernet bearer support + * tipc_eth_media_stop - deactivate Ethernet bearer support */ -void eth_media_stop(void) +void tipc_eth_media_stop(void) { int i; diff --git a/net/tipc/handler.c b/net/tipc/handler.c index f320010f8a6..966f70a1b60 100644 --- a/net/tipc/handler.c +++ b/net/tipc/handler.c @@ -52,7 +52,7 @@ static void process_signal_queue(unsigned long dummy); static DECLARE_TASKLET_DISABLED(tipc_tasklet, process_signal_queue, 0); -unsigned int k_signal(Handler routine, unsigned long argument) +unsigned int tipc_k_signal(Handler routine, unsigned long argument) { struct queue_item *item; @@ -93,7 +93,7 @@ static void process_signal_queue(unsigned long dummy) spin_unlock_bh(&qitem_lock); } -int handler_start(void) +int tipc_handler_start(void) { tipc_queue_item_cache = kmem_cache_create("tipc_queue_items", sizeof(struct queue_item), @@ -107,7 +107,7 @@ int handler_start(void) return 0; } -void handler_stop(void) +void tipc_handler_stop(void) { struct list_head *l, *n; struct queue_item *item; diff --git a/net/tipc/link.c b/net/tipc/link.c index d1e1ae66464..511872afa45 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -148,12 +148,12 @@ static void link_print(struct link *l_ptr, struct print_buf *buf, #define LINK_LOG_BUF_SIZE 0 #define dbg_link(fmt, arg...) do {if (LINK_LOG_BUF_SIZE) tipc_printf(&l_ptr->print_buf, fmt, ## arg); } while(0) -#define dbg_link_msg(msg, txt) do {if (LINK_LOG_BUF_SIZE) msg_print(&l_ptr->print_buf, msg, txt); } while(0) +#define dbg_link_msg(msg, txt) do {if (LINK_LOG_BUF_SIZE) tipc_msg_print(&l_ptr->print_buf, msg, txt); } while(0) #define dbg_link_state(txt) do {if (LINK_LOG_BUF_SIZE) link_print(l_ptr, &l_ptr->print_buf, txt); } while(0) #define dbg_link_dump() do { \ if (LINK_LOG_BUF_SIZE) { \ tipc_printf(LOG, "\n\nDumping link <%s>:\n", l_ptr->name); \ - printbuf_move(LOG, &l_ptr->print_buf); \ + tipc_printbuf_move(LOG, &l_ptr->print_buf); \ } \ } while (0) @@ -252,14 +252,14 @@ static inline u32 link_last_sent(struct link *l_ptr) * Simple non-inlined link routines (i.e. referenced outside this file) */ -int link_is_up(struct link *l_ptr) +int tipc_link_is_up(struct link *l_ptr) { if (!l_ptr) return 0; return (link_working_working(l_ptr) || link_working_unknown(l_ptr)); } -int link_is_active(struct link *l_ptr) +int tipc_link_is_active(struct link *l_ptr) { return ((l_ptr->owner->active_links[0] == l_ptr) || (l_ptr->owner->active_links[1] == l_ptr)); @@ -338,15 +338,15 @@ static int link_name_validate(const char *name, struct link_name *name_parts) * link_timeout - handle expiration of link timer * @l_ptr: pointer to link * - * This routine must not grab "net_lock" to avoid a potential deadlock conflict - * with link_delete(). (There is no risk that the node will be deleted by - * another thread because link_delete() always cancels the link timer before - * node_delete() is called.) + * This routine must not grab "tipc_net_lock" to avoid a potential deadlock conflict + * with tipc_link_delete(). (There is no risk that the node will be deleted by + * another thread because tipc_link_delete() always cancels the link timer before + * tipc_node_delete() is called.) */ static void link_timeout(struct link *l_ptr) { - node_lock(l_ptr->owner); + tipc_node_lock(l_ptr->owner); /* update counters used in statistical profiling of send traffic */ @@ -391,9 +391,9 @@ static void link_timeout(struct link *l_ptr) link_state_event(l_ptr, TIMEOUT_EVT); if (l_ptr->next_out) - link_push_queue(l_ptr); + tipc_link_push_queue(l_ptr); - node_unlock(l_ptr->owner); + tipc_node_unlock(l_ptr->owner); } static inline void link_set_timer(struct link *l_ptr, u32 time) @@ -402,7 +402,7 @@ static inline void link_set_timer(struct link *l_ptr, u32 time) } /** - * link_create - create a new link + * tipc_link_create - create a new link * @b_ptr: pointer to associated bearer * @peer: network address of node at other end of link * @media_addr: media address to use when sending messages over link @@ -410,8 +410,8 @@ static inline void link_set_timer(struct link *l_ptr, u32 time) * Returns pointer to link. */ -struct link *link_create(struct bearer *b_ptr, const u32 peer, - const struct tipc_media_addr *media_addr) +struct link *tipc_link_create(struct bearer *b_ptr, const u32 peer, + const struct tipc_media_addr *media_addr) { struct link *l_ptr; struct tipc_msg *msg; @@ -449,7 +449,7 @@ struct link *link_create(struct bearer *b_ptr, const u32 peer, strcpy((char *)msg_data(msg), if_name); l_ptr->priority = b_ptr->priority; - link_set_queue_limits(l_ptr, b_ptr->media->window); + tipc_link_set_queue_limits(l_ptr, b_ptr->media->window); link_init_max_pkt(l_ptr); @@ -458,7 +458,7 @@ struct link *link_create(struct bearer *b_ptr, const u32 peer, link_reset_statistics(l_ptr); - l_ptr->owner = node_attach_link(l_ptr); + l_ptr->owner = tipc_node_attach_link(l_ptr); if (!l_ptr->owner) { kfree(l_ptr); return NULL; @@ -472,52 +472,52 @@ struct link *link_create(struct bearer *b_ptr, const u32 peer, warn("Memory squeeze; Failed to create link\n"); return NULL; } - printbuf_init(&l_ptr->print_buf, pb, LINK_LOG_BUF_SIZE); + tipc_printbuf_init(&l_ptr->print_buf, pb, LINK_LOG_BUF_SIZE); } - k_signal((Handler)link_start, (unsigned long)l_ptr); + tipc_k_signal((Handler)tipc_link_start, (unsigned long)l_ptr); - dbg("link_create(): tolerance = %u,cont intv = %u, abort_limit = %u\n", + dbg("tipc_link_create(): tolerance = %u,cont intv = %u, abort_limit = %u\n", l_ptr->tolerance, l_ptr->continuity_interval, l_ptr->abort_limit); return l_ptr; } /** - * link_delete - delete a link + * tipc_link_delete - delete a link * @l_ptr: pointer to link * - * Note: 'net_lock' is write_locked, bearer is locked. + * Note: 'tipc_net_lock' is write_locked, bearer is locked. * This routine must not grab the node lock until after link timer cancellation * to avoid a potential deadlock situation. */ -void link_delete(struct link *l_ptr) +void tipc_link_delete(struct link *l_ptr) { if (!l_ptr) { err("Attempt to delete non-existent link\n"); return; } - dbg("link_delete()\n"); + dbg("tipc_link_delete()\n"); k_cancel_timer(&l_ptr->timer); - node_lock(l_ptr->owner); - link_reset(l_ptr); - node_detach_link(l_ptr->owner, l_ptr); - link_stop(l_ptr); + tipc_node_lock(l_ptr->owner); + tipc_link_reset(l_ptr); + tipc_node_detach_link(l_ptr->owner, l_ptr); + tipc_link_stop(l_ptr); list_del_init(&l_ptr->link_list); if (LINK_LOG_BUF_SIZE) kfree(l_ptr->print_buf.buf); - node_unlock(l_ptr->owner); + tipc_node_unlock(l_ptr->owner); k_term_timer(&l_ptr->timer); kfree(l_ptr); } -void link_start(struct link *l_ptr) +void tipc_link_start(struct link *l_ptr) { - dbg("link_start %x\n", l_ptr); + dbg("tipc_link_start %x\n", l_ptr); link_state_event(l_ptr, STARTING_EVT); } @@ -535,8 +535,8 @@ static int link_schedule_port(struct link *l_ptr, u32 origport, u32 sz) { struct port *p_ptr; - spin_lock_bh(&port_list_lock); - p_ptr = port_lock(origport); + spin_lock_bh(&tipc_port_list_lock); + p_ptr = tipc_port_lock(origport); if (p_ptr) { if (!p_ptr->wakeup) goto exit; @@ -548,13 +548,13 @@ static int link_schedule_port(struct link *l_ptr, u32 origport, u32 sz) list_add_tail(&p_ptr->wait_list, &l_ptr->waiting_ports); l_ptr->stats.link_congs++; exit: - port_unlock(p_ptr); + tipc_port_unlock(p_ptr); } - spin_unlock_bh(&port_list_lock); + spin_unlock_bh(&tipc_port_list_lock); return -ELINKCONG; } -void link_wakeup_ports(struct link *l_ptr, int all) +void tipc_link_wakeup_ports(struct link *l_ptr, int all) { struct port *p_ptr; struct port *temp_p_ptr; @@ -564,7 +564,7 @@ void link_wakeup_ports(struct link *l_ptr, int all) win = 100000; if (win <= 0) return; - if (!spin_trylock_bh(&port_list_lock)) + if (!spin_trylock_bh(&tipc_port_list_lock)) return; if (link_congested(l_ptr)) goto exit; @@ -583,7 +583,7 @@ void link_wakeup_ports(struct link *l_ptr, int all) } exit: - spin_unlock_bh(&port_list_lock); + spin_unlock_bh(&tipc_port_list_lock); } /** @@ -606,11 +606,11 @@ static void link_release_outqueue(struct link *l_ptr) } /** - * link_reset_fragments - purge link's inbound message fragments queue + * tipc_link_reset_fragments - purge link's inbound message fragments queue * @l_ptr: pointer to link */ -void link_reset_fragments(struct link *l_ptr) +void tipc_link_reset_fragments(struct link *l_ptr) { struct sk_buff *buf = l_ptr->defragm_buf; struct sk_buff *next; @@ -624,11 +624,11 @@ void link_reset_fragments(struct link *l_ptr) } /** - * link_stop - purge all inbound and outbound messages associated with link + * tipc_link_stop - purge all inbound and outbound messages associated with link * @l_ptr: pointer to link */ -void link_stop(struct link *l_ptr) +void tipc_link_stop(struct link *l_ptr) { struct sk_buff *buf; struct sk_buff *next; @@ -647,7 +647,7 @@ void link_stop(struct link *l_ptr) buf = next; } - link_reset_fragments(l_ptr); + tipc_link_reset_fragments(l_ptr); buf_discard(l_ptr->proto_msg_queue); l_ptr->proto_msg_queue = NULL; @@ -677,7 +677,7 @@ static void link_send_event(void (*fcn)(u32 a, char *n, int up), ev->up = up; ev->fcn = fcn; memcpy(ev->name, l_ptr->name, TIPC_MAX_LINK_NAME); - k_signal((Handler)link_recv_event, (unsigned long)ev); + tipc_k_signal((Handler)link_recv_event, (unsigned long)ev); } #else @@ -686,7 +686,7 @@ static void link_send_event(void (*fcn)(u32 a, char *n, int up), #endif -void link_reset(struct link *l_ptr) +void tipc_link_reset(struct link *l_ptr) { struct sk_buff *buf; u32 prev_state = l_ptr->state; @@ -706,13 +706,13 @@ void link_reset(struct link *l_ptr) if ((prev_state == RESET_UNKNOWN) || (prev_state == RESET_RESET)) return; - node_link_down(l_ptr->owner, l_ptr); - bearer_remove_dest(l_ptr->b_ptr, l_ptr->addr); + tipc_node_link_down(l_ptr->owner, l_ptr); + tipc_bearer_remove_dest(l_ptr->b_ptr, l_ptr->addr); #if 0 - tipc_printf(CONS, "\nReset link <%s>\n", l_ptr->name); + tipc_printf(TIPC_CONS, "\nReset link <%s>\n", l_ptr->name); dbg_link_dump(); #endif - if (node_has_active_links(l_ptr->owner) && + if (tipc_node_has_active_links(l_ptr->owner) && l_ptr->owner->permit_changeover) { l_ptr->reset_checkpoint = checkpoint; l_ptr->exp_msg_count = START_CHANGEOVER; @@ -730,7 +730,7 @@ void link_reset(struct link *l_ptr) buf = next; } if (!list_empty(&l_ptr->waiting_ports)) - link_wakeup_ports(l_ptr, 1); + tipc_link_wakeup_ports(l_ptr, 1); l_ptr->retransm_queue_head = 0; l_ptr->retransm_queue_size = 0; @@ -747,20 +747,20 @@ void link_reset(struct link *l_ptr) l_ptr->stale_count = 0; link_reset_statistics(l_ptr); - link_send_event(cfg_link_event, l_ptr, 0); + link_send_event(tipc_cfg_link_event, l_ptr, 0); if (!in_own_cluster(l_ptr->addr)) - link_send_event(disc_link_event, l_ptr, 0); + link_send_event(tipc_disc_link_event, l_ptr, 0); } static void link_activate(struct link *l_ptr) { l_ptr->next_in_no = 1; - node_link_up(l_ptr->owner, l_ptr); - bearer_add_dest(l_ptr->b_ptr, l_ptr->addr); - link_send_event(cfg_link_event, l_ptr, 1); + tipc_node_link_up(l_ptr->owner, l_ptr); + tipc_bearer_add_dest(l_ptr->b_ptr, l_ptr->addr); + link_send_event(tipc_cfg_link_event, l_ptr, 1); if (!in_own_cluster(l_ptr->addr)) - link_send_event(disc_link_event, l_ptr, 1); + link_send_event(tipc_disc_link_event, l_ptr, 1); } /** @@ -799,13 +799,13 @@ static void link_state_event(struct link *l_ptr, unsigned event) dbg_link("TIM "); if (l_ptr->next_in_no != l_ptr->checkpoint) { l_ptr->checkpoint = l_ptr->next_in_no; - if (bclink_acks_missing(l_ptr->owner)) { - link_send_proto_msg(l_ptr, STATE_MSG, - 0, 0, 0, 0, 0); + if (tipc_bclink_acks_missing(l_ptr->owner)) { + tipc_link_send_proto_msg(l_ptr, STATE_MSG, + 0, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; } else if (l_ptr->max_pkt < l_ptr->max_pkt_target) { - link_send_proto_msg(l_ptr, STATE_MSG, - 1, 0, 0, 0, 0); + tipc_link_send_proto_msg(l_ptr, STATE_MSG, + 1, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; } link_set_timer(l_ptr, cont_intv); @@ -814,16 +814,16 @@ static void link_state_event(struct link *l_ptr, unsigned event) dbg_link(" -> WU\n"); l_ptr->state = WORKING_UNKNOWN; l_ptr->fsm_msg_cnt = 0; - link_send_proto_msg(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + tipc_link_send_proto_msg(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv / 4); break; case RESET_MSG: dbg_link("RES -> RR\n"); - link_reset(l_ptr); + tipc_link_reset(l_ptr); l_ptr->state = RESET_RESET; l_ptr->fsm_msg_cnt = 0; - link_send_proto_msg(l_ptr, ACTIVATE_MSG, 0, 0, 0, 0, 0); + tipc_link_send_proto_msg(l_ptr, ACTIVATE_MSG, 0, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); break; @@ -844,10 +844,10 @@ static void link_state_event(struct link *l_ptr, unsigned event) break; case RESET_MSG: dbg_link("RES -> RR\n"); - link_reset(l_ptr); + tipc_link_reset(l_ptr); l_ptr->state = RESET_RESET; l_ptr->fsm_msg_cnt = 0; - link_send_proto_msg(l_ptr, ACTIVATE_MSG, 0, 0, 0, 0, 0); + tipc_link_send_proto_msg(l_ptr, ACTIVATE_MSG, 0, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); break; @@ -858,9 +858,9 @@ static void link_state_event(struct link *l_ptr, unsigned event) l_ptr->state = WORKING_WORKING; l_ptr->fsm_msg_cnt = 0; l_ptr->checkpoint = l_ptr->next_in_no; - if (bclink_acks_missing(l_ptr->owner)) { - link_send_proto_msg(l_ptr, STATE_MSG, - 0, 0, 0, 0, 0); + if (tipc_bclink_acks_missing(l_ptr->owner)) { + tipc_link_send_proto_msg(l_ptr, STATE_MSG, + 0, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; } link_set_timer(l_ptr, cont_intv); @@ -868,18 +868,18 @@ static void link_state_event(struct link *l_ptr, unsigned event) dbg_link("Probing %u/%u,timer = %u ms)\n", l_ptr->fsm_msg_cnt, l_ptr->abort_limit, cont_intv / 4); - link_send_proto_msg(l_ptr, STATE_MSG, - 1, 0, 0, 0, 0); + tipc_link_send_proto_msg(l_ptr, STATE_MSG, + 1, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv / 4); } else { /* Link has failed */ dbg_link("-> RU (%u probes unanswered)\n", l_ptr->fsm_msg_cnt); - link_reset(l_ptr); + tipc_link_reset(l_ptr); l_ptr->state = RESET_UNKNOWN; l_ptr->fsm_msg_cnt = 0; - link_send_proto_msg(l_ptr, RESET_MSG, - 0, 0, 0, 0, 0); + tipc_link_send_proto_msg(l_ptr, RESET_MSG, + 0, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); } @@ -904,7 +904,7 @@ static void link_state_event(struct link *l_ptr, unsigned event) l_ptr->state = WORKING_WORKING; l_ptr->fsm_msg_cnt = 0; link_activate(l_ptr); - link_send_proto_msg(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + tipc_link_send_proto_msg(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); break; @@ -913,7 +913,7 @@ static void link_state_event(struct link *l_ptr, unsigned event) dbg_link(" -> RR\n"); l_ptr->state = RESET_RESET; l_ptr->fsm_msg_cnt = 0; - link_send_proto_msg(l_ptr, ACTIVATE_MSG, 1, 0, 0, 0, 0); + tipc_link_send_proto_msg(l_ptr, ACTIVATE_MSG, 1, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); break; @@ -923,7 +923,7 @@ static void link_state_event(struct link *l_ptr, unsigned event) /* fall through */ case TIMEOUT_EVT: dbg_link("TIM \n"); - link_send_proto_msg(l_ptr, RESET_MSG, 0, 0, 0, 0, 0); + tipc_link_send_proto_msg(l_ptr, RESET_MSG, 0, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); break; @@ -947,7 +947,7 @@ static void link_state_event(struct link *l_ptr, unsigned event) l_ptr->state = WORKING_WORKING; l_ptr->fsm_msg_cnt = 0; link_activate(l_ptr); - link_send_proto_msg(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + tipc_link_send_proto_msg(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); break; @@ -956,7 +956,7 @@ static void link_state_event(struct link *l_ptr, unsigned event) break; case TIMEOUT_EVT: dbg_link("TIM\n"); - link_send_proto_msg(l_ptr, ACTIVATE_MSG, 0, 0, 0, 0, 0); + tipc_link_send_proto_msg(l_ptr, ACTIVATE_MSG, 0, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); dbg_link("fsm_msg_cnt %u\n", l_ptr->fsm_msg_cnt); @@ -1023,12 +1023,12 @@ static inline void link_add_to_outqueue(struct link *l_ptr, } /* - * link_send_buf() is the 'full path' for messages, called from + * tipc_link_send_buf() is the 'full path' for messages, called from * inside TIPC when the 'fast path' in tipc_send_buf * has failed, and from link_send() */ -int link_send_buf(struct link *l_ptr, struct sk_buff *buf) +int tipc_link_send_buf(struct link *l_ptr, struct sk_buff *buf) { struct tipc_msg *msg = buf_msg(buf); u32 size = msg_size(msg); @@ -1051,7 +1051,7 @@ int link_send_buf(struct link *l_ptr, struct sk_buff *buf) buf_discard(buf); if (imp > CONN_MANAGER) { warn("Resetting <%s>, send queue full", l_ptr->name); - link_reset(l_ptr); + tipc_link_reset(l_ptr); } return dsz; } @@ -1059,21 +1059,21 @@ int link_send_buf(struct link *l_ptr, struct sk_buff *buf) /* Fragmentation needed ? */ if (size > max_packet) - return link_send_long_buf(l_ptr, buf); + return tipc_link_send_long_buf(l_ptr, buf); /* Packet can be queued or sent: */ if (queue_size > l_ptr->stats.max_queue_sz) l_ptr->stats.max_queue_sz = queue_size; - if (likely(!bearer_congested(l_ptr->b_ptr, l_ptr) && + if (likely(!tipc_bearer_congested(l_ptr->b_ptr, l_ptr) && !link_congested(l_ptr))) { link_add_to_outqueue(l_ptr, buf, msg); - if (likely(bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr))) { + if (likely(tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr))) { l_ptr->unacked_window = 0; } else { - bearer_schedule(l_ptr->b_ptr, l_ptr); + tipc_bearer_schedule(l_ptr->b_ptr, l_ptr); l_ptr->stats.bearer_congs++; l_ptr->next_out = buf; } @@ -1088,7 +1088,7 @@ int link_send_buf(struct link *l_ptr, struct sk_buff *buf) if (l_ptr->next_out && link_bundle_buf(l_ptr, l_ptr->last_out, buf)) { - bearer_resolve_congestion(l_ptr->b_ptr, l_ptr); + tipc_bearer_resolve_congestion(l_ptr->b_ptr, l_ptr); return dsz; } @@ -1114,38 +1114,38 @@ int link_send_buf(struct link *l_ptr, struct sk_buff *buf) if (!l_ptr->next_out) l_ptr->next_out = buf; link_add_to_outqueue(l_ptr, buf, msg); - bearer_resolve_congestion(l_ptr->b_ptr, l_ptr); + tipc_bearer_resolve_congestion(l_ptr->b_ptr, l_ptr); return dsz; } /* - * link_send(): same as link_send_buf(), but the link to use has + * tipc_link_send(): same as tipc_link_send_buf(), but the link to use has * not been selected yet, and the the owner node is not locked * Called by TIPC internal users, e.g. the name distributor */ -int link_send(struct sk_buff *buf, u32 dest, u32 selector) +int tipc_link_send(struct sk_buff *buf, u32 dest, u32 selector) { struct link *l_ptr; struct node *n_ptr; int res = -ELINKCONG; - read_lock_bh(&net_lock); - n_ptr = node_select(dest, selector); + read_lock_bh(&tipc_net_lock); + n_ptr = tipc_node_select(dest, selector); if (n_ptr) { - node_lock(n_ptr); + tipc_node_lock(n_ptr); l_ptr = n_ptr->active_links[selector & 1]; - dbg("link_send: found link %x for dest %x\n", l_ptr, dest); + dbg("tipc_link_send: found link %x for dest %x\n", l_ptr, dest); if (l_ptr) { - res = link_send_buf(l_ptr, buf); + res = tipc_link_send_buf(l_ptr, buf); } - node_unlock(n_ptr); + tipc_node_unlock(n_ptr); } else { dbg("Attempt to send msg to unknown node:\n"); msg_dbg(buf_msg(buf),">>>"); buf_discard(buf); } - read_unlock_bh(&net_lock); + read_unlock_bh(&tipc_net_lock); return res; } @@ -1166,14 +1166,14 @@ static inline int link_send_buf_fast(struct link *l_ptr, struct sk_buff *buf, if (likely(msg_size(msg) <= link_max_pkt(l_ptr))) { if (likely(list_empty(&l_ptr->b_ptr->cong_links))) { link_add_to_outqueue(l_ptr, buf, msg); - if (likely(bearer_send(l_ptr->b_ptr, buf, - &l_ptr->media_addr))) { + if (likely(tipc_bearer_send(l_ptr->b_ptr, buf, + &l_ptr->media_addr))) { l_ptr->unacked_window = 0; msg_dbg(msg,"SENT_FAST:"); return res; } dbg("failed sent fast...\n"); - bearer_schedule(l_ptr->b_ptr, l_ptr); + tipc_bearer_schedule(l_ptr->b_ptr, l_ptr); l_ptr->stats.bearer_congs++; l_ptr->next_out = buf; return res; @@ -1182,7 +1182,7 @@ static inline int link_send_buf_fast(struct link *l_ptr, struct sk_buff *buf, else *used_max_pkt = link_max_pkt(l_ptr); } - return link_send_buf(l_ptr, buf); /* All other cases */ + return tipc_link_send_buf(l_ptr, buf); /* All other cases */ } /* @@ -1200,24 +1200,24 @@ int tipc_send_buf_fast(struct sk_buff *buf, u32 destnode) u32 dummy; if (destnode == tipc_own_addr) - return port_recv_msg(buf); + return tipc_port_recv_msg(buf); - read_lock_bh(&net_lock); - n_ptr = node_select(destnode, selector); + read_lock_bh(&tipc_net_lock); + n_ptr = tipc_node_select(destnode, selector); if (likely(n_ptr)) { - node_lock(n_ptr); + tipc_node_lock(n_ptr); l_ptr = n_ptr->active_links[selector]; dbg("send_fast: buf %x selected %x, destnode = %x\n", buf, l_ptr, destnode); if (likely(l_ptr)) { res = link_send_buf_fast(l_ptr, buf, &dummy); - node_unlock(n_ptr); - read_unlock_bh(&net_lock); + tipc_node_unlock(n_ptr); + read_unlock_bh(&tipc_net_lock); return res; } - node_unlock(n_ptr); + tipc_node_unlock(n_ptr); } - read_unlock_bh(&net_lock); + read_unlock_bh(&tipc_net_lock); res = msg_data_sz(buf_msg(buf)); tipc_reject_msg(buf, TIPC_ERR_NO_NODE); return res; @@ -1225,15 +1225,15 @@ int tipc_send_buf_fast(struct sk_buff *buf, u32 destnode) /* - * link_send_sections_fast: Entry for messages where the + * tipc_link_send_sections_fast: Entry for messages where the * destination processor is known and the header is complete, * except for total message length. * Returns user data length or errno. */ -int link_send_sections_fast(struct port *sender, - struct iovec const *msg_sect, - const u32 num_sect, - u32 destaddr) +int tipc_link_send_sections_fast(struct port *sender, + struct iovec const *msg_sect, + const u32 num_sect, + u32 destaddr) { struct tipc_msg *hdr = &sender->publ.phdr; struct link *l_ptr; @@ -1253,10 +1253,10 @@ again: res = msg_build(hdr, msg_sect, num_sect, sender->max_pkt, !sender->user_port, &buf); - read_lock_bh(&net_lock); - node = node_select(destaddr, selector); + read_lock_bh(&tipc_net_lock); + node = tipc_node_select(destaddr, selector); if (likely(node)) { - node_lock(node); + tipc_node_lock(node); l_ptr = node->active_links[selector]; if (likely(l_ptr)) { if (likely(buf)) { @@ -1265,8 +1265,8 @@ again: if (unlikely(res < 0)) buf_discard(buf); exit: - node_unlock(node); - read_unlock_bh(&net_lock); + tipc_node_unlock(node); + read_unlock_bh(&tipc_net_lock); return res; } @@ -1290,8 +1290,8 @@ exit: */ sender->max_pkt = link_max_pkt(l_ptr); - node_unlock(node); - read_unlock_bh(&net_lock); + tipc_node_unlock(node); + read_unlock_bh(&tipc_net_lock); if ((msg_hdr_sz(hdr) + res) <= sender->max_pkt) @@ -1300,17 +1300,17 @@ exit: return link_send_sections_long(sender, msg_sect, num_sect, destaddr); } - node_unlock(node); + tipc_node_unlock(node); } - read_unlock_bh(&net_lock); + read_unlock_bh(&tipc_net_lock); /* Couldn't find a link to the destination node */ if (buf) return tipc_reject_msg(buf, TIPC_ERR_NO_NODE); if (res >= 0) - return port_reject_sections(sender, hdr, msg_sect, num_sect, - TIPC_ERR_NO_NODE); + return tipc_port_reject_sections(sender, hdr, msg_sect, num_sect, + TIPC_ERR_NO_NODE); return res; } @@ -1444,17 +1444,17 @@ error: * Now we have a buffer chain. Select a link and check * that packet size is still OK */ - node = node_select(destaddr, sender->publ.ref & 1); + node = tipc_node_select(destaddr, sender->publ.ref & 1); if (likely(node)) { - node_lock(node); + tipc_node_lock(node); l_ptr = node->active_links[sender->publ.ref & 1]; if (!l_ptr) { - node_unlock(node); + tipc_node_unlock(node); goto reject; } if (link_max_pkt(l_ptr) < max_pkt) { sender->max_pkt = link_max_pkt(l_ptr); - node_unlock(node); + tipc_node_unlock(node); for (; buf_chain; buf_chain = buf) { buf = buf_chain->next; buf_discard(buf_chain); @@ -1467,8 +1467,8 @@ reject: buf = buf_chain->next; buf_discard(buf_chain); } - return port_reject_sections(sender, hdr, msg_sect, num_sect, - TIPC_ERR_NO_NODE); + return tipc_port_reject_sections(sender, hdr, msg_sect, num_sect, + TIPC_ERR_NO_NODE); } /* Append whole chain to send queue: */ @@ -1491,15 +1491,15 @@ reject: /* Send it, if possible: */ - link_push_queue(l_ptr); - node_unlock(node); + tipc_link_push_queue(l_ptr); + tipc_node_unlock(node); return dsz; } /* - * link_push_packet: Push one unsent packet to the media + * tipc_link_push_packet: Push one unsent packet to the media */ -u32 link_push_packet(struct link *l_ptr) +u32 tipc_link_push_packet(struct link *l_ptr) { struct sk_buff *buf = l_ptr->first_out; u32 r_q_size = l_ptr->retransm_queue_size; @@ -1526,7 +1526,7 @@ u32 link_push_packet(struct link *l_ptr) if (r_q_size && buf && !skb_cloned(buf)) { msg_set_ack(buf_msg(buf), mod(l_ptr->next_in_no - 1)); msg_set_bcast_ack(buf_msg(buf), l_ptr->owner->bclink.last_in); - if (bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { + if (tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { msg_dbg(buf_msg(buf), ">DEF-RETR>"); l_ptr->retransm_queue_head = mod(++r_q_head); l_ptr->retransm_queue_size = --r_q_size; @@ -1545,7 +1545,7 @@ u32 link_push_packet(struct link *l_ptr) if (buf) { msg_set_ack(buf_msg(buf), mod(l_ptr->next_in_no - 1)); msg_set_bcast_ack(buf_msg(buf),l_ptr->owner->bclink.last_in); - if (bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { + if (tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { msg_dbg(buf_msg(buf), ">DEF-PROT>"); l_ptr->unacked_window = 0; buf_discard(buf); @@ -1569,7 +1569,7 @@ u32 link_push_packet(struct link *l_ptr) if (mod(next - first) < l_ptr->queue_limit[0]) { msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); - if (bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { + if (tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { if (msg_user(msg) == MSG_BUNDLER) msg_set_type(msg, CLOSED_MSG); msg_dbg(msg, ">PUSH-DATA>"); @@ -1589,29 +1589,29 @@ u32 link_push_packet(struct link *l_ptr) * push_queue(): push out the unsent messages of a link where * congestion has abated. Node is locked */ -void link_push_queue(struct link *l_ptr) +void tipc_link_push_queue(struct link *l_ptr) { u32 res; - if (bearer_congested(l_ptr->b_ptr, l_ptr)) + if (tipc_bearer_congested(l_ptr->b_ptr, l_ptr)) return; do { - res = link_push_packet(l_ptr); + res = tipc_link_push_packet(l_ptr); } while (res == TIPC_OK); if (res == PUSH_FAILED) - bearer_schedule(l_ptr->b_ptr, l_ptr); + tipc_bearer_schedule(l_ptr->b_ptr, l_ptr); } -void link_retransmit(struct link *l_ptr, struct sk_buff *buf, - u32 retransmits) +void tipc_link_retransmit(struct link *l_ptr, struct sk_buff *buf, + u32 retransmits) { struct tipc_msg *msg; dbg("Retransmitting %u in link %x\n", retransmits, l_ptr); - if (bearer_congested(l_ptr->b_ptr, l_ptr) && buf && !skb_cloned(buf)) { + if (tipc_bearer_congested(l_ptr->b_ptr, l_ptr) && buf && !skb_cloned(buf)) { msg_dbg(buf_msg(buf), ">NO_RETR->BCONG>"); dbg_print_link(l_ptr, " "); l_ptr->retransm_queue_head = msg_seqno(buf_msg(buf)); @@ -1622,15 +1622,15 @@ void link_retransmit(struct link *l_ptr, struct sk_buff *buf, msg = buf_msg(buf); msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); - if (bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { + if (tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { /* Catch if retransmissions fail repeatedly: */ if (l_ptr->last_retransmitted == msg_seqno(msg)) { if (++l_ptr->stale_count > 100) { - msg_print(CONS, buf_msg(buf), ">RETR>"); + tipc_msg_print(TIPC_CONS, buf_msg(buf), ">RETR>"); info("...Retransmitted %u times\n", l_ptr->stale_count); - link_print(l_ptr, CONS, "Resetting Link\n");; - link_reset(l_ptr); + link_print(l_ptr, TIPC_CONS, "Resetting Link\n");; + tipc_link_reset(l_ptr); break; } } else { @@ -1643,7 +1643,7 @@ void link_retransmit(struct link *l_ptr, struct sk_buff *buf, retransmits--; l_ptr->stats.retransmitted++; } else { - bearer_schedule(l_ptr->b_ptr, l_ptr); + tipc_bearer_schedule(l_ptr->b_ptr, l_ptr); l_ptr->stats.bearer_congs++; l_ptr->retransm_queue_head = msg_seqno(buf_msg(buf)); l_ptr->retransm_queue_size = retransmits; @@ -1663,9 +1663,9 @@ static void link_recv_non_seq(struct sk_buff *buf) struct tipc_msg *msg = buf_msg(buf); if (msg_user(msg) == LINK_CONFIG) - disc_recv_msg(buf); + tipc_disc_recv_msg(buf); else - bclink_recv_pkt(buf); + tipc_bclink_recv_pkt(buf); } /** @@ -1692,7 +1692,7 @@ static struct sk_buff *link_insert_deferred_queue(struct link *l_ptr, void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *tb_ptr) { - read_lock_bh(&net_lock); + read_lock_bh(&tipc_net_lock); while (head) { struct bearer *b_ptr; struct node *n_ptr; @@ -1720,22 +1720,22 @@ void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *tb_ptr) link_recv_non_seq(buf); continue; } - n_ptr = node_find(msg_prevnode(msg)); + n_ptr = tipc_node_find(msg_prevnode(msg)); if (unlikely(!n_ptr)) goto cont; - node_lock(n_ptr); + tipc_node_lock(n_ptr); l_ptr = n_ptr->links[b_ptr->identity]; if (unlikely(!l_ptr)) { - node_unlock(n_ptr); + tipc_node_unlock(n_ptr); goto cont; } /* * Release acked messages */ if (less(n_ptr->bclink.acked, msg_bcast_ack(msg))) { - if (node_is_up(n_ptr) && n_ptr->bclink.supported) - bclink_acknowledge(n_ptr, msg_bcast_ack(msg)); + if (tipc_node_is_up(n_ptr) && n_ptr->bclink.supported) + tipc_bclink_acknowledge(n_ptr, msg_bcast_ack(msg)); } crs = l_ptr->first_out; @@ -1752,12 +1752,12 @@ void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *tb_ptr) l_ptr->out_queue_size -= released; } if (unlikely(l_ptr->next_out)) - link_push_queue(l_ptr); + tipc_link_push_queue(l_ptr); if (unlikely(!list_empty(&l_ptr->waiting_ports))) - link_wakeup_ports(l_ptr, 0); + tipc_link_wakeup_ports(l_ptr, 0); if (unlikely(++l_ptr->unacked_window >= TIPC_MIN_LINK_WIN)) { l_ptr->stats.sent_acks++; - link_send_proto_msg(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); + tipc_link_send_proto_msg(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); } protocol_check: @@ -1770,8 +1770,8 @@ protocol_check: if (likely(msg_is_dest(msg, tipc_own_addr))) { deliver: if (likely(msg_isdata(msg))) { - node_unlock(n_ptr); - port_recv_msg(buf); + tipc_node_unlock(n_ptr); + tipc_port_recv_msg(buf); continue; } switch (msg_user(msg)) { @@ -1779,34 +1779,32 @@ deliver: l_ptr->stats.recv_bundles++; l_ptr->stats.recv_bundled += msg_msgcnt(msg); - node_unlock(n_ptr); - link_recv_bundle(buf); + tipc_node_unlock(n_ptr); + tipc_link_recv_bundle(buf); continue; case ROUTE_DISTRIBUTOR: - node_unlock(n_ptr); - cluster_recv_routing_table(buf); + tipc_node_unlock(n_ptr); + tipc_cltr_recv_routing_table(buf); continue; case NAME_DISTRIBUTOR: - node_unlock(n_ptr); - named_recv(buf); + tipc_node_unlock(n_ptr); + tipc_named_recv(buf); continue; case CONN_MANAGER: - node_unlock(n_ptr); - port_recv_proto_msg(buf); + tipc_node_unlock(n_ptr); + tipc_port_recv_proto_msg(buf); continue; case MSG_FRAGMENTER: l_ptr->stats.recv_fragments++; - if (link_recv_fragment( - &l_ptr->defragm_buf, - &buf, &msg)) { + if (tipc_link_recv_fragment(&l_ptr->defragm_buf, + &buf, &msg)) { l_ptr->stats.recv_fragmented++; goto deliver; } break; case CHANGEOVER_PROTOCOL: type = msg_type(msg); - if (link_recv_changeover_msg( - &l_ptr, &buf)) { + if (link_recv_changeover_msg(&l_ptr, &buf)) { msg = buf_msg(buf); seq_no = msg_seqno(msg); TIPC_SKB_CB(buf)->handle @@ -1818,20 +1816,20 @@ deliver: break; } } - node_unlock(n_ptr); - net_route_msg(buf); + tipc_node_unlock(n_ptr); + tipc_net_route_msg(buf); continue; } link_handle_out_of_seq_msg(l_ptr, buf); head = link_insert_deferred_queue(l_ptr, head); - node_unlock(n_ptr); + tipc_node_unlock(n_ptr); continue; } if (msg_user(msg) == LINK_PROTOCOL) { link_recv_proto_msg(l_ptr, buf); head = link_insert_deferred_queue(l_ptr, head); - node_unlock(n_ptr); + tipc_node_unlock(n_ptr); continue; } msg_dbg(msg,"NSEQnext = head; head = buf; - node_unlock(n_ptr); + tipc_node_unlock(n_ptr); continue; } - node_unlock(n_ptr); + tipc_node_unlock(n_ptr); cont: buf_discard(buf); } - read_unlock_bh(&net_lock); + read_unlock_bh(&tipc_net_lock); } /* @@ -1858,9 +1856,9 @@ cont: * Returns the increase of the queue length,i.e. 0 or 1 */ -u32 link_defer_pkt(struct sk_buff **head, - struct sk_buff **tail, - struct sk_buff *buf) +u32 tipc_link_defer_pkt(struct sk_buff **head, + struct sk_buff **tail, + struct sk_buff *buf) { struct sk_buff *prev = 0; struct sk_buff *crs = *head; @@ -1939,12 +1937,12 @@ static void link_handle_out_of_seq_msg(struct link *l_ptr, return; } - if (link_defer_pkt(&l_ptr->oldest_deferred_in, - &l_ptr->newest_deferred_in, buf)) { + if (tipc_link_defer_pkt(&l_ptr->oldest_deferred_in, + &l_ptr->newest_deferred_in, buf)) { l_ptr->deferred_inqueue_sz++; l_ptr->stats.deferred_recv++; if ((l_ptr->deferred_inqueue_sz % 16) == 1) - link_send_proto_msg(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); + tipc_link_send_proto_msg(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); } else l_ptr->stats.duplicates++; } @@ -1952,8 +1950,8 @@ static void link_handle_out_of_seq_msg(struct link *l_ptr, /* * Send protocol message to the other endpoint. */ -void link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int probe_msg, - u32 gap, u32 tolerance, u32 priority, u32 ack_mtu) +void tipc_link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int probe_msg, + u32 gap, u32 tolerance, u32 priority, u32 ack_mtu) { struct sk_buff *buf = 0; struct tipc_msg *msg = l_ptr->pmsg; @@ -1964,12 +1962,12 @@ void link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int probe_msg, msg_set_type(msg, msg_typ); msg_set_net_plane(msg, l_ptr->b_ptr->net_plane); msg_set_bcast_ack(msg, mod(l_ptr->owner->bclink.last_in)); - msg_set_last_bcast(msg, bclink_get_last_sent()); + msg_set_last_bcast(msg, tipc_bclink_get_last_sent()); if (msg_typ == STATE_MSG) { u32 next_sent = mod(l_ptr->next_out_no); - if (!link_is_up(l_ptr)) + if (!tipc_link_is_up(l_ptr)) return; if (l_ptr->next_out) next_sent = msg_seqno(buf_msg(l_ptr->next_out)); @@ -2013,7 +2011,7 @@ void link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int probe_msg, msg_set_max_pkt(msg, l_ptr->max_pkt_target); } - if (node_has_redundant_links(l_ptr->owner)) { + if (tipc_node_has_redundant_links(l_ptr->owner)) { msg_set_redundant_link(msg); } else { msg_clear_redundant_link(msg); @@ -2026,7 +2024,7 @@ void link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int probe_msg, /* Congestion? */ - if (bearer_congested(l_ptr->b_ptr, l_ptr)) { + if (tipc_bearer_congested(l_ptr->b_ptr, l_ptr)) { if (!l_ptr->proto_msg_queue) { l_ptr->proto_msg_queue = buf_acquire(sizeof(l_ptr->proto_msg)); @@ -2050,14 +2048,14 @@ void link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int probe_msg, memcpy(buf->data, (unchar *)msg, sizeof(l_ptr->proto_msg)); msg_set_size(buf_msg(buf), msg_size); - if (bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { + if (tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { l_ptr->unacked_window = 0; buf_discard(buf); return; } /* New congestion */ - bearer_schedule(l_ptr->b_ptr, l_ptr); + tipc_bearer_schedule(l_ptr->b_ptr, l_ptr); l_ptr->proto_msg_queue = buf; l_ptr->stats.bearer_congs++; } @@ -2131,7 +2129,7 @@ static void link_recv_proto_msg(struct link *l_ptr, struct sk_buff *buf) l_ptr->peer_bearer_id = msg_bearer_id(msg); /* Synchronize broadcast sequence numbers */ - if (!node_has_redundant_links(l_ptr->owner)) { + if (!tipc_node_has_redundant_links(l_ptr->owner)) { l_ptr->owner->bclink.last_in = mod(msg_last_bcast(msg)); } break; @@ -2145,7 +2143,7 @@ static void link_recv_proto_msg(struct link *l_ptr, struct sk_buff *buf) warn("Changing prio <%s>: %u->%u\n", l_ptr->name, l_ptr->priority, msg_linkprio(msg)); l_ptr->priority = msg_linkprio(msg); - link_reset(l_ptr); /* Enforce change to take effect */ + tipc_link_reset(l_ptr); /* Enforce change to take effect */ break; } link_state_event(l_ptr, TRAFFIC_MSG_EVT); @@ -2176,17 +2174,17 @@ static void link_recv_proto_msg(struct link *l_ptr, struct sk_buff *buf) /* Protocol message before retransmits, reduce loss risk */ - bclink_check_gap(l_ptr->owner, msg_last_bcast(msg)); + tipc_bclink_check_gap(l_ptr->owner, msg_last_bcast(msg)); if (rec_gap || (msg_probe(msg))) { - link_send_proto_msg(l_ptr, STATE_MSG, - 0, rec_gap, 0, 0, max_pkt_ack); + tipc_link_send_proto_msg(l_ptr, STATE_MSG, + 0, rec_gap, 0, 0, max_pkt_ack); } if (msg_seq_gap(msg)) { msg_dbg(msg, "With Gap:"); l_ptr->stats.recv_nacks++; - link_retransmit(l_ptr, l_ptr->first_out, - msg_seq_gap(msg)); + tipc_link_retransmit(l_ptr, l_ptr->first_out, + msg_seq_gap(msg)); } break; default: @@ -2198,20 +2196,20 @@ exit: /* - * link_tunnel(): Send one message via a link belonging to + * tipc_link_tunnel(): Send one message via a link belonging to * another bearer. Owner node is locked. */ -void link_tunnel(struct link *l_ptr, - struct tipc_msg *tunnel_hdr, - struct tipc_msg *msg, - u32 selector) +void tipc_link_tunnel(struct link *l_ptr, + struct tipc_msg *tunnel_hdr, + struct tipc_msg *msg, + u32 selector) { struct link *tunnel; struct sk_buff *buf; u32 length = msg_size(msg); tunnel = l_ptr->owner->active_links[selector & 1]; - if (!link_is_up(tunnel)) + if (!tipc_link_is_up(tunnel)) return; msg_set_size(tunnel_hdr, length + INT_H_SIZE); buf = buf_acquire(length + INT_H_SIZE); @@ -2222,7 +2220,7 @@ void link_tunnel(struct link *l_ptr, dbg("%c->%c:", l_ptr->b_ptr->net_plane, tunnel->b_ptr->net_plane); msg_dbg(buf_msg(buf), ">SEND>"); assert(tunnel); - link_send_buf(tunnel, buf); + tipc_link_send_buf(tunnel, buf); } @@ -2232,12 +2230,12 @@ void link_tunnel(struct link *l_ptr, * Owner node is locked. */ -void link_changeover(struct link *l_ptr) +void tipc_link_changeover(struct link *l_ptr) { u32 msgcount = l_ptr->out_queue_size; struct sk_buff *crs = l_ptr->first_out; struct link *tunnel = l_ptr->owner->active_links[0]; - int split_bundles = node_has_redundant_links(l_ptr->owner); + int split_bundles = tipc_node_has_redundant_links(l_ptr->owner); struct tipc_msg tunnel_hdr; if (!tunnel) @@ -2261,7 +2259,7 @@ void link_changeover(struct link *l_ptr) dbg("%c->%c:", l_ptr->b_ptr->net_plane, tunnel->b_ptr->net_plane); msg_dbg(&tunnel_hdr, "EMPTY>SEND>"); - link_send_buf(tunnel, buf); + tipc_link_send_buf(tunnel, buf); } else { warn("Memory squeeze; link changeover failed\n"); } @@ -2277,20 +2275,20 @@ void link_changeover(struct link *l_ptr) while (msgcount--) { msg_set_seqno(m,msg_seqno(msg)); - link_tunnel(l_ptr, &tunnel_hdr, m, - msg_link_selector(m)); + tipc_link_tunnel(l_ptr, &tunnel_hdr, m, + msg_link_selector(m)); pos += align(msg_size(m)); m = (struct tipc_msg *)pos; } } else { - link_tunnel(l_ptr, &tunnel_hdr, msg, - msg_link_selector(msg)); + tipc_link_tunnel(l_ptr, &tunnel_hdr, msg, + msg_link_selector(msg)); } crs = crs->next; } } -void link_send_duplicate(struct link *l_ptr, struct link *tunnel) +void tipc_link_send_duplicate(struct link *l_ptr, struct link *tunnel) { struct sk_buff *iter; struct tipc_msg tunnel_hdr; @@ -2320,8 +2318,8 @@ void link_send_duplicate(struct link *l_ptr, struct link *tunnel) dbg("%c->%c:", l_ptr->b_ptr->net_plane, tunnel->b_ptr->net_plane); msg_dbg(buf_msg(outbuf), ">SEND>"); - link_send_buf(tunnel, outbuf); - if (!link_is_up(l_ptr)) + tipc_link_send_buf(tunnel, outbuf); + if (!tipc_link_is_up(l_ptr)) return; iter = iter->next; } @@ -2393,9 +2391,9 @@ static int link_recv_changeover_msg(struct link **l_ptr, /* First original message ?: */ - if (link_is_up(dest_link)) { + if (tipc_link_is_up(dest_link)) { msg_dbg(tunnel_msg, "UP/FIRST/exp_msg_count = msg_count; if (!msg_count) goto exit; @@ -2436,7 +2434,7 @@ exit: /* * Bundler functionality: */ -void link_recv_bundle(struct sk_buff *buf) +void tipc_link_recv_bundle(struct sk_buff *buf) { u32 msgcount = msg_msgcnt(buf_msg(buf)); u32 pos = INT_H_SIZE; @@ -2456,7 +2454,7 @@ void link_recv_bundle(struct sk_buff *buf) }; pos += align(msg_size(buf_msg(obuf))); msg_dbg(buf_msg(obuf), " /"); - net_route_msg(obuf); + tipc_net_route_msg(obuf); } buf_discard(buf); } @@ -2467,11 +2465,11 @@ void link_recv_bundle(struct sk_buff *buf) /* - * link_send_long_buf: Entry for buffers needing fragmentation. + * tipc_link_send_long_buf: Entry for buffers needing fragmentation. * The buffer is complete, inclusive total message length. * Returns user data length. */ -int link_send_long_buf(struct link *l_ptr, struct sk_buff *buf) +int tipc_link_send_long_buf(struct link *l_ptr, struct sk_buff *buf) { struct tipc_msg *inmsg = buf_msg(buf); struct tipc_msg fragm_hdr; @@ -2521,8 +2519,8 @@ int link_send_long_buf(struct link *l_ptr, struct sk_buff *buf) /* Send queued messages first, if any: */ l_ptr->stats.sent_fragments++; - link_send_buf(l_ptr, fragm); - if (!link_is_up(l_ptr)) + tipc_link_send_buf(l_ptr, fragm); + if (!tipc_link_is_up(l_ptr)) return dsz; msg_set_fragm_no(&fragm_hdr, ++fragm_no); rest -= fragm_sz; @@ -2582,11 +2580,11 @@ static inline void incr_timer_cnt(struct sk_buff *buf) } /* - * link_recv_fragment(): Called with node lock on. Returns + * tipc_link_recv_fragment(): Called with node lock on. Returns * the reassembled buffer if message is complete. */ -int link_recv_fragment(struct sk_buff **pending, struct sk_buff **fb, - struct tipc_msg **m) +int tipc_link_recv_fragment(struct sk_buff **pending, struct sk_buff **fb, + struct tipc_msg **m) { struct sk_buff *prev = 0; struct sk_buff *fbuf = *fb; @@ -2714,7 +2712,7 @@ static void link_set_supervision_props(struct link *l_ptr, u32 tolerance) } -void link_set_queue_limits(struct link *l_ptr, u32 window) +void tipc_link_set_queue_limits(struct link *l_ptr, u32 window) { /* Data messages from this node, inclusive FIRST_FRAGM */ l_ptr->queue_limit[DATA_LOW] = window; @@ -2739,7 +2737,7 @@ void link_set_queue_limits(struct link *l_ptr, u32 window) * @name - ptr to link name string * @node - ptr to area to be filled with ptr to associated node * - * Caller must hold 'net_lock' to ensure node and bearer are not deleted; + * Caller must hold 'tipc_net_lock' to ensure node and bearer are not deleted; * this also prevents link deletion. * * Returns pointer to link (or 0 if invalid link name). @@ -2754,11 +2752,11 @@ static struct link *link_find_link(const char *name, struct node **node) if (!link_name_validate(name, &link_name_parts)) return 0; - b_ptr = bearer_find_interface(link_name_parts.if_local); + b_ptr = tipc_bearer_find_interface(link_name_parts.if_local); if (!b_ptr) return 0; - *node = node_find(link_name_parts.addr_peer); + *node = tipc_node_find(link_name_parts.addr_peer); if (!*node) return 0; @@ -2769,8 +2767,8 @@ static struct link *link_find_link(const char *name, struct node **node) return l_ptr; } -struct sk_buff *link_cmd_config(const void *req_tlv_area, int req_tlv_space, - u16 cmd) +struct sk_buff *tipc_link_cmd_config(const void *req_tlv_area, int req_tlv_space, + u16 cmd) { struct tipc_link_config *args; u32 new_value; @@ -2779,35 +2777,35 @@ struct sk_buff *link_cmd_config(const void *req_tlv_area, int req_tlv_space, int res; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_CONFIG)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); args = (struct tipc_link_config *)TLV_DATA(req_tlv_area); new_value = ntohl(args->value); - if (!strcmp(args->name, bc_link_name)) { + if (!strcmp(args->name, tipc_bclink_name)) { if ((cmd == TIPC_CMD_SET_LINK_WINDOW) && - (bclink_set_queue_limits(new_value) == 0)) - return cfg_reply_none(); - return cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (cannot change setting on broadcast link)"); + (tipc_bclink_set_queue_limits(new_value) == 0)) + return tipc_cfg_reply_none(); + return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (cannot change setting on broadcast link)"); } - read_lock_bh(&net_lock); + read_lock_bh(&tipc_net_lock); l_ptr = link_find_link(args->name, &node); if (!l_ptr) { - read_unlock_bh(&net_lock); - return cfg_reply_error_string("link not found"); + read_unlock_bh(&tipc_net_lock); + return tipc_cfg_reply_error_string("link not found"); } - node_lock(node); + tipc_node_lock(node); res = -EINVAL; switch (cmd) { case TIPC_CMD_SET_LINK_TOL: if ((new_value >= TIPC_MIN_LINK_TOL) && (new_value <= TIPC_MAX_LINK_TOL)) { link_set_supervision_props(l_ptr, new_value); - link_send_proto_msg(l_ptr, STATE_MSG, - 0, 0, new_value, 0, 0); + tipc_link_send_proto_msg(l_ptr, STATE_MSG, + 0, 0, new_value, 0, 0); res = TIPC_OK; } break; @@ -2815,26 +2813,26 @@ struct sk_buff *link_cmd_config(const void *req_tlv_area, int req_tlv_space, if ((new_value >= TIPC_MIN_LINK_PRI) && (new_value <= TIPC_MAX_LINK_PRI)) { l_ptr->priority = new_value; - link_send_proto_msg(l_ptr, STATE_MSG, - 0, 0, 0, new_value, 0); + tipc_link_send_proto_msg(l_ptr, STATE_MSG, + 0, 0, 0, new_value, 0); res = TIPC_OK; } break; case TIPC_CMD_SET_LINK_WINDOW: if ((new_value >= TIPC_MIN_LINK_WIN) && (new_value <= TIPC_MAX_LINK_WIN)) { - link_set_queue_limits(l_ptr, new_value); + tipc_link_set_queue_limits(l_ptr, new_value); res = TIPC_OK; } break; } - node_unlock(node); + tipc_node_unlock(node); - read_unlock_bh(&net_lock); + read_unlock_bh(&tipc_net_lock); if (res) - return cfg_reply_error_string("cannot change link setting"); + return tipc_cfg_reply_error_string("cannot change link setting"); - return cfg_reply_none(); + return tipc_cfg_reply_none(); } /** @@ -2849,34 +2847,34 @@ static void link_reset_statistics(struct link *l_ptr) l_ptr->stats.recv_info = l_ptr->next_in_no; } -struct sk_buff *link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_space) +struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_space) { char *link_name; struct link *l_ptr; struct node *node; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_NAME)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); link_name = (char *)TLV_DATA(req_tlv_area); - if (!strcmp(link_name, bc_link_name)) { - if (bclink_reset_stats()) - return cfg_reply_error_string("link not found"); - return cfg_reply_none(); + if (!strcmp(link_name, tipc_bclink_name)) { + if (tipc_bclink_reset_stats()) + return tipc_cfg_reply_error_string("link not found"); + return tipc_cfg_reply_none(); } - read_lock_bh(&net_lock); + read_lock_bh(&tipc_net_lock); l_ptr = link_find_link(link_name, &node); if (!l_ptr) { - read_unlock_bh(&net_lock); - return cfg_reply_error_string("link not found"); + read_unlock_bh(&tipc_net_lock); + return tipc_cfg_reply_error_string("link not found"); } - node_lock(node); + tipc_node_lock(node); link_reset_statistics(l_ptr); - node_unlock(node); - read_unlock_bh(&net_lock); - return cfg_reply_none(); + tipc_node_unlock(node); + read_unlock_bh(&tipc_net_lock); + return tipc_cfg_reply_none(); } /** @@ -2889,7 +2887,7 @@ static u32 percent(u32 count, u32 total) } /** - * link_stats - print link statistics + * tipc_link_stats - print link statistics * @name: link name * @buf: print buffer area * @buf_size: size of print buffer area @@ -2897,7 +2895,7 @@ static u32 percent(u32 count, u32 total) * Returns length of print buffer data string (or 0 if error) */ -static int link_stats(const char *name, char *buf, const u32 buf_size) +static int tipc_link_stats(const char *name, char *buf, const u32 buf_size) { struct print_buf pb; struct link *l_ptr; @@ -2905,22 +2903,22 @@ static int link_stats(const char *name, char *buf, const u32 buf_size) char *status; u32 profile_total = 0; - if (!strcmp(name, bc_link_name)) - return bclink_stats(buf, buf_size); + if (!strcmp(name, tipc_bclink_name)) + return tipc_bclink_stats(buf, buf_size); - printbuf_init(&pb, buf, buf_size); + tipc_printbuf_init(&pb, buf, buf_size); - read_lock_bh(&net_lock); + read_lock_bh(&tipc_net_lock); l_ptr = link_find_link(name, &node); if (!l_ptr) { - read_unlock_bh(&net_lock); + read_unlock_bh(&tipc_net_lock); return 0; } - node_lock(node); + tipc_node_lock(node); - if (link_is_active(l_ptr)) + if (tipc_link_is_active(l_ptr)) status = "ACTIVE"; - else if (link_is_up(l_ptr)) + else if (tipc_link_is_up(l_ptr)) status = "STANDBY"; else status = "DEFUNCT"; @@ -2976,33 +2974,33 @@ static int link_stats(const char *name, char *buf, const u32 buf_size) ? (l_ptr->stats.accu_queue_sz / l_ptr->stats.queue_sz_counts) : 0); - node_unlock(node); - read_unlock_bh(&net_lock); - return printbuf_validate(&pb); + tipc_node_unlock(node); + read_unlock_bh(&tipc_net_lock); + return tipc_printbuf_validate(&pb); } #define MAX_LINK_STATS_INFO 2000 -struct sk_buff *link_cmd_show_stats(const void *req_tlv_area, int req_tlv_space) +struct sk_buff *tipc_link_cmd_show_stats(const void *req_tlv_area, int req_tlv_space) { struct sk_buff *buf; struct tlv_desc *rep_tlv; int str_len; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_NAME)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - buf = cfg_reply_alloc(TLV_SPACE(MAX_LINK_STATS_INFO)); + buf = tipc_cfg_reply_alloc(TLV_SPACE(MAX_LINK_STATS_INFO)); if (!buf) return NULL; rep_tlv = (struct tlv_desc *)buf->data; - str_len = link_stats((char *)TLV_DATA(req_tlv_area), - (char *)TLV_DATA(rep_tlv), MAX_LINK_STATS_INFO); + str_len = tipc_link_stats((char *)TLV_DATA(req_tlv_area), + (char *)TLV_DATA(rep_tlv), MAX_LINK_STATS_INFO); if (!str_len) { buf_discard(buf); - return cfg_reply_error_string("link not found"); + return tipc_cfg_reply_error_string("link not found"); } skb_put(buf, TLV_SPACE(str_len)); @@ -3021,20 +3019,20 @@ int link_control(const char *name, u32 op, u32 val) u32 a; a = link_name2addr(name, &bearer_id); - read_lock_bh(&net_lock); - node = node_find(a); + read_lock_bh(&tipc_net_lock); + node = tipc_node_find(a); if (node) { - node_lock(node); + tipc_node_lock(node); l_ptr = node->links[bearer_id]; if (l_ptr) { if (op == TIPC_REMOVE_LINK) { struct bearer *b_ptr = l_ptr->b_ptr; spin_lock_bh(&b_ptr->publ.lock); - link_delete(l_ptr); + tipc_link_delete(l_ptr); spin_unlock_bh(&b_ptr->publ.lock); } if (op == TIPC_CMD_BLOCK_LINK) { - link_reset(l_ptr); + tipc_link_reset(l_ptr); l_ptr->blocked = 1; } if (op == TIPC_CMD_UNBLOCK_LINK) { @@ -3042,22 +3040,22 @@ int link_control(const char *name, u32 op, u32 val) } res = TIPC_OK; } - node_unlock(node); + tipc_node_unlock(node); } - read_unlock_bh(&net_lock); + read_unlock_bh(&tipc_net_lock); return res; } #endif /** - * link_get_max_pkt - get maximum packet size to use when sending to destination + * tipc_link_get_max_pkt - get maximum packet size to use when sending to destination * @dest: network address of destination node * @selector: used to select from set of active links * * If no active link can be found, uses default maximum packet size. */ -u32 link_get_max_pkt(u32 dest, u32 selector) +u32 tipc_link_get_max_pkt(u32 dest, u32 selector) { struct node *n_ptr; struct link *l_ptr; @@ -3066,16 +3064,16 @@ u32 link_get_max_pkt(u32 dest, u32 selector) if (dest == tipc_own_addr) return MAX_MSG_SIZE; - read_lock_bh(&net_lock); - n_ptr = node_select(dest, selector); + read_lock_bh(&tipc_net_lock); + n_ptr = tipc_node_select(dest, selector); if (n_ptr) { - node_lock(n_ptr); + tipc_node_lock(n_ptr); l_ptr = n_ptr->active_links[selector & 1]; if (l_ptr) res = link_max_pkt(l_ptr); - node_unlock(n_ptr); + tipc_node_unlock(n_ptr); } - read_unlock_bh(&net_lock); + read_unlock_bh(&tipc_net_lock); return res; } diff --git a/net/tipc/link.h b/net/tipc/link.h index c2553f07375..2d3c157f707 100644 --- a/net/tipc/link.h +++ b/net/tipc/link.h @@ -221,44 +221,43 @@ struct link { struct port; -struct link *link_create(struct bearer *b_ptr, const u32 peer, - const struct tipc_media_addr *media_addr); -void link_delete(struct link *l_ptr); -void link_changeover(struct link *l_ptr); -void link_send_duplicate(struct link *l_ptr, struct link *dest); -void link_reset_fragments(struct link *l_ptr); -int link_is_up(struct link *l_ptr); -int link_is_active(struct link *l_ptr); -void link_start(struct link *l_ptr); -u32 link_push_packet(struct link *l_ptr); -void link_stop(struct link *l_ptr); -struct sk_buff *link_cmd_config(const void *req_tlv_area, int req_tlv_space, u16 cmd); -struct sk_buff *link_cmd_show_stats(const void *req_tlv_area, int req_tlv_space); -struct sk_buff *link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_space); -void link_reset(struct link *l_ptr); -int link_send(struct sk_buff *buf, u32 dest, u32 selector); -int link_send_buf(struct link *l_ptr, struct sk_buff *buf); -u32 link_get_max_pkt(u32 dest,u32 selector); -int link_send_sections_fast(struct port* sender, - struct iovec const *msg_sect, - const u32 num_sect, - u32 destnode); - -int link_send_long_buf(struct link *l_ptr, struct sk_buff *buf); -void link_tunnel(struct link *l_ptr, struct tipc_msg *tnl_hdr, - struct tipc_msg *msg, u32 selector); -void link_recv_bundle(struct sk_buff *buf); -int link_recv_fragment(struct sk_buff **pending, - struct sk_buff **fb, - struct tipc_msg **msg); -void link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int prob, u32 gap, - u32 tolerance, u32 priority, u32 acked_mtu); -void link_push_queue(struct link *l_ptr); -u32 link_defer_pkt(struct sk_buff **head, struct sk_buff **tail, +struct link *tipc_link_create(struct bearer *b_ptr, const u32 peer, + const struct tipc_media_addr *media_addr); +void tipc_link_delete(struct link *l_ptr); +void tipc_link_changeover(struct link *l_ptr); +void tipc_link_send_duplicate(struct link *l_ptr, struct link *dest); +void tipc_link_reset_fragments(struct link *l_ptr); +int tipc_link_is_up(struct link *l_ptr); +int tipc_link_is_active(struct link *l_ptr); +void tipc_link_start(struct link *l_ptr); +u32 tipc_link_push_packet(struct link *l_ptr); +void tipc_link_stop(struct link *l_ptr); +struct sk_buff *tipc_link_cmd_config(const void *req_tlv_area, int req_tlv_space, u16 cmd); +struct sk_buff *tipc_link_cmd_show_stats(const void *req_tlv_area, int req_tlv_space); +struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_space); +void tipc_link_reset(struct link *l_ptr); +int tipc_link_send(struct sk_buff *buf, u32 dest, u32 selector); +int tipc_link_send_buf(struct link *l_ptr, struct sk_buff *buf); +u32 tipc_link_get_max_pkt(u32 dest,u32 selector); +int tipc_link_send_sections_fast(struct port* sender, + struct iovec const *msg_sect, + const u32 num_sect, + u32 destnode); +int tipc_link_send_long_buf(struct link *l_ptr, struct sk_buff *buf); +void tipc_link_tunnel(struct link *l_ptr, struct tipc_msg *tnl_hdr, + struct tipc_msg *msg, u32 selector); +void tipc_link_recv_bundle(struct sk_buff *buf); +int tipc_link_recv_fragment(struct sk_buff **pending, + struct sk_buff **fb, + struct tipc_msg **msg); +void tipc_link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int prob, u32 gap, + u32 tolerance, u32 priority, u32 acked_mtu); +void tipc_link_push_queue(struct link *l_ptr); +u32 tipc_link_defer_pkt(struct sk_buff **head, struct sk_buff **tail, struct sk_buff *buf); -void link_wakeup_ports(struct link *l_ptr, int all); -void link_set_queue_limits(struct link *l_ptr, u32 window); -void link_retransmit(struct link *l_ptr, struct sk_buff *start, u32 retransmits); +void tipc_link_wakeup_ports(struct link *l_ptr, int all); +void tipc_link_set_queue_limits(struct link *l_ptr, u32 window); +void tipc_link_retransmit(struct link *l_ptr, struct sk_buff *start, u32 retransmits); /* * Link sequence number manipulation routines (uses modulo 2**16 arithmetic) diff --git a/net/tipc/msg.c b/net/tipc/msg.c index 03dbc55cb04..3bd345a344e 100644 --- a/net/tipc/msg.c +++ b/net/tipc/msg.c @@ -41,18 +41,7 @@ #include "bearer.h" -void msg_set_media_addr(struct tipc_msg *m, struct tipc_media_addr *a) -{ - memcpy(&((int *)m)[5], a, sizeof(*a)); -} - -void msg_get_media_addr(struct tipc_msg *m, struct tipc_media_addr *a) -{ - memcpy(a, &((int*)m)[5], sizeof(*a)); -} - - -void msg_print(struct print_buf *buf, struct tipc_msg *msg, const char *str) +void tipc_msg_print(struct print_buf *buf, struct tipc_msg *msg, const char *str) { u32 usr = msg_user(msg); tipc_printf(buf, str); @@ -318,7 +307,7 @@ void msg_print(struct print_buf *buf, struct tipc_msg *msg, const char *str) tipc_printf(buf, ":REQL(%u):", msg_req_links(msg)); tipc_printf(buf, ":DDOM(%x):", msg_dest_domain(msg)); tipc_printf(buf, ":NETID(%u):", msg_bc_netid(msg)); - media_addr_printf(buf, orig); + tipc_media_addr_printf(buf, orig); } if (msg_user(msg) == BCAST_PROTOCOL) { tipc_printf(buf, "BCNACK:AFTER(%u):", msg_bcgap_after(msg)); @@ -326,9 +315,9 @@ void msg_print(struct print_buf *buf, struct tipc_msg *msg, const char *str) } tipc_printf(buf, "\n"); if ((usr == CHANGEOVER_PROTOCOL) && (msg_msgcnt(msg))) { - msg_print(buf,msg_get_wrapped(msg)," /"); + tipc_msg_print(buf,msg_get_wrapped(msg)," /"); } if ((usr == MSG_FRAGMENTER) && (msg_type(msg) == FIRST_FRAGMENT)) { - msg_print(buf,msg_get_wrapped(msg)," /"); + tipc_msg_print(buf,msg_get_wrapped(msg)," /"); } } diff --git a/net/tipc/msg.h b/net/tipc/msg.h index 6574aab11fa..6699aaf7bd4 100644 --- a/net/tipc/msg.h +++ b/net/tipc/msg.h @@ -805,14 +805,14 @@ static inline int msg_build(struct tipc_msg *hdr, return -EFAULT; } +static inline void msg_set_media_addr(struct tipc_msg *m, struct tipc_media_addr *a) +{ + memcpy(&((int *)m)[5], a, sizeof(*a)); +} -struct tipc_media_addr; - -extern void msg_set_media_addr(struct tipc_msg *m, - struct tipc_media_addr *a); - -extern void msg_get_media_addr(struct tipc_msg *m, - struct tipc_media_addr *a); - +static inline void msg_get_media_addr(struct tipc_msg *m, struct tipc_media_addr *a) +{ + memcpy(a, &((int*)m)[5], sizeof(*a)); +} #endif diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c index 41cbaf1a4a7..830f9099904 100644 --- a/net/tipc/name_distr.c +++ b/net/tipc/name_distr.c @@ -114,10 +114,10 @@ static struct sk_buff *named_prepare_buf(u32 type, u32 size, u32 dest) } /** - * named_publish - tell other nodes about a new publication by this node + * tipc_named_publish - tell other nodes about a new publication by this node */ -void named_publish(struct publication *publ) +void tipc_named_publish(struct publication *publ) { struct sk_buff *buf; struct distr_item *item; @@ -133,15 +133,15 @@ void named_publish(struct publication *publ) item = (struct distr_item *)msg_data(buf_msg(buf)); publ_to_item(item, publ); - dbg("named_withdraw: broadcasting publish msg\n"); - cluster_broadcast(buf); + dbg("tipc_named_withdraw: broadcasting publish msg\n"); + tipc_cltr_broadcast(buf); } /** - * named_withdraw - tell other nodes about a withdrawn publication by this node + * tipc_named_withdraw - tell other nodes about a withdrawn publication by this node */ -void named_withdraw(struct publication *publ) +void tipc_named_withdraw(struct publication *publ) { struct sk_buff *buf; struct distr_item *item; @@ -157,15 +157,15 @@ void named_withdraw(struct publication *publ) item = (struct distr_item *)msg_data(buf_msg(buf)); publ_to_item(item, publ); - dbg("named_withdraw: broadcasting withdraw msg\n"); - cluster_broadcast(buf); + dbg("tipc_named_withdraw: broadcasting withdraw msg\n"); + tipc_cltr_broadcast(buf); } /** - * named_node_up - tell specified node about all publications by this node + * tipc_named_node_up - tell specified node about all publications by this node */ -void named_node_up(unsigned long node) +void tipc_named_node_up(unsigned long node) { struct publication *publ; struct distr_item *item = 0; @@ -175,7 +175,7 @@ void named_node_up(unsigned long node) u32 max_item_buf; assert(in_own_cluster(node)); - read_lock_bh(&nametbl_lock); + read_lock_bh(&tipc_nametbl_lock); max_item_buf = TIPC_MAX_USER_MSG_SIZE / ITEM_SIZE; max_item_buf *= ITEM_SIZE; rest = publ_cnt * ITEM_SIZE; @@ -196,15 +196,15 @@ void named_node_up(unsigned long node) left -= ITEM_SIZE; if (!left) { msg_set_link_selector(buf_msg(buf), node); - dbg("named_node_up: sending publish msg to " + dbg("tipc_named_node_up: sending publish msg to " "<%u.%u.%u>\n", tipc_zone(node), tipc_cluster(node), tipc_node(node)); - link_send(buf, node, node); + tipc_link_send(buf, node, node); buf = 0; } } exit: - read_unlock_bh(&nametbl_lock); + read_unlock_bh(&tipc_nametbl_lock); } /** @@ -221,73 +221,73 @@ exit: static void node_is_down(struct publication *publ) { struct publication *p; - write_lock_bh(&nametbl_lock); + write_lock_bh(&tipc_nametbl_lock); dbg("node_is_down: withdrawing %u, %u, %u\n", publ->type, publ->lower, publ->upper); publ->key += 1222345; - p = nametbl_remove_publ(publ->type, publ->lower, - publ->node, publ->ref, publ->key); + p = tipc_nametbl_remove_publ(publ->type, publ->lower, + publ->node, publ->ref, publ->key); assert(p == publ); - write_unlock_bh(&nametbl_lock); + write_unlock_bh(&tipc_nametbl_lock); if (publ) kfree(publ); } /** - * named_recv - process name table update message sent by another node + * tipc_named_recv - process name table update message sent by another node */ -void named_recv(struct sk_buff *buf) +void tipc_named_recv(struct sk_buff *buf) { struct publication *publ; struct tipc_msg *msg = buf_msg(buf); struct distr_item *item = (struct distr_item *)msg_data(msg); u32 count = msg_data_sz(msg) / ITEM_SIZE; - write_lock_bh(&nametbl_lock); + write_lock_bh(&tipc_nametbl_lock); while (count--) { if (msg_type(msg) == PUBLICATION) { - dbg("named_recv: got publication for %u, %u, %u\n", + dbg("tipc_named_recv: got publication for %u, %u, %u\n", ntohl(item->type), ntohl(item->lower), ntohl(item->upper)); - publ = nametbl_insert_publ(ntohl(item->type), - ntohl(item->lower), - ntohl(item->upper), - TIPC_CLUSTER_SCOPE, - msg_orignode(msg), - ntohl(item->ref), - ntohl(item->key)); + publ = tipc_nametbl_insert_publ(ntohl(item->type), + ntohl(item->lower), + ntohl(item->upper), + TIPC_CLUSTER_SCOPE, + msg_orignode(msg), + ntohl(item->ref), + ntohl(item->key)); if (publ) { - nodesub_subscribe(&publ->subscr, - msg_orignode(msg), - publ, - (net_ev_handler)node_is_down); + tipc_nodesub_subscribe(&publ->subscr, + msg_orignode(msg), + publ, + (net_ev_handler)node_is_down); } } else if (msg_type(msg) == WITHDRAWAL) { - dbg("named_recv: got withdrawl for %u, %u, %u\n", + dbg("tipc_named_recv: got withdrawl for %u, %u, %u\n", ntohl(item->type), ntohl(item->lower), ntohl(item->upper)); - publ = nametbl_remove_publ(ntohl(item->type), - ntohl(item->lower), - msg_orignode(msg), - ntohl(item->ref), - ntohl(item->key)); + publ = tipc_nametbl_remove_publ(ntohl(item->type), + ntohl(item->lower), + msg_orignode(msg), + ntohl(item->ref), + ntohl(item->key)); if (publ) { - nodesub_unsubscribe(&publ->subscr); + tipc_nodesub_unsubscribe(&publ->subscr); kfree(publ); } } else { - warn("named_recv: unknown msg\n"); + warn("tipc_named_recv: unknown msg\n"); } item++; } - write_unlock_bh(&nametbl_lock); + write_unlock_bh(&tipc_nametbl_lock); buf_discard(buf); } /** - * named_reinit - re-initialize local publication list + * tipc_named_reinit - re-initialize local publication list * * This routine is called whenever TIPC networking is (re)enabled. * All existing publications by this node that have "cluster" or "zone" scope @@ -295,15 +295,15 @@ void named_recv(struct sk_buff *buf) * (If the node's address is unchanged, the update loop terminates immediately.) */ -void named_reinit(void) +void tipc_named_reinit(void) { struct publication *publ; - write_lock_bh(&nametbl_lock); + write_lock_bh(&tipc_nametbl_lock); list_for_each_entry(publ, &publ_root, local_list) { if (publ->node == tipc_own_addr) break; publ->node = tipc_own_addr; } - write_unlock_bh(&nametbl_lock); + write_unlock_bh(&tipc_nametbl_lock); } diff --git a/net/tipc/name_distr.h b/net/tipc/name_distr.h index a04bdeac84e..843da0172f4 100644 --- a/net/tipc/name_distr.h +++ b/net/tipc/name_distr.h @@ -39,10 +39,10 @@ #include "name_table.h" -void named_publish(struct publication *publ); -void named_withdraw(struct publication *publ); -void named_node_up(unsigned long node); -void named_recv(struct sk_buff *buf); -void named_reinit(void); +void tipc_named_publish(struct publication *publ); +void tipc_named_withdraw(struct publication *publ); +void tipc_named_node_up(unsigned long node); +void tipc_named_recv(struct sk_buff *buf); +void tipc_named_reinit(void); #endif diff --git a/net/tipc/name_table.c b/net/tipc/name_table.c index 972c83eb83b..3f4b23bd08f 100644 --- a/net/tipc/name_table.c +++ b/net/tipc/name_table.c @@ -99,9 +99,9 @@ struct name_table { u32 local_publ_count; }; -struct name_table table = { NULL } ; +static struct name_table table = { NULL } ; static atomic_t rsv_publ_ok = ATOMIC_INIT(0); -rwlock_t nametbl_lock = RW_LOCK_UNLOCKED; +rwlock_t tipc_nametbl_lock = RW_LOCK_UNLOCKED; static inline int hash(int x) @@ -139,10 +139,10 @@ static struct publication *publ_create(u32 type, u32 lower, u32 upper, } /** - * subseq_alloc - allocate a specified number of sub-sequence structures + * tipc_subseq_alloc - allocate a specified number of sub-sequence structures */ -struct sub_seq *subseq_alloc(u32 cnt) +struct sub_seq *tipc_subseq_alloc(u32 cnt) { u32 sz = cnt * sizeof(struct sub_seq); struct sub_seq *sseq = (struct sub_seq *)kmalloc(sz, GFP_ATOMIC); @@ -153,16 +153,16 @@ struct sub_seq *subseq_alloc(u32 cnt) } /** - * nameseq_create - create a name sequence structure for the specified 'type' + * tipc_nameseq_create - create a name sequence structure for the specified 'type' * * Allocates a single sub-sequence structure and sets it to all 0's. */ -struct name_seq *nameseq_create(u32 type, struct hlist_head *seq_head) +struct name_seq *tipc_nameseq_create(u32 type, struct hlist_head *seq_head) { struct name_seq *nseq = (struct name_seq *)kmalloc(sizeof(*nseq), GFP_ATOMIC); - struct sub_seq *sseq = subseq_alloc(1); + struct sub_seq *sseq = tipc_subseq_alloc(1); if (!nseq || !sseq) { warn("Memory squeeze; failed to create name sequence\n"); @@ -175,7 +175,7 @@ struct name_seq *nameseq_create(u32 type, struct hlist_head *seq_head) nseq->lock = SPIN_LOCK_UNLOCKED; nseq->type = type; nseq->sseqs = sseq; - dbg("nameseq_create() nseq = %x type %u, ssseqs %x, ff: %u\n", + dbg("tipc_nameseq_create() nseq = %x type %u, ssseqs %x, ff: %u\n", nseq, type, nseq->sseqs, nseq->first_free); nseq->alloc = 1; INIT_HLIST_NODE(&nseq->ns_list); @@ -240,10 +240,10 @@ static u32 nameseq_locate_subseq(struct name_seq *nseq, u32 instance) } /** - * nameseq_insert_publ - + * tipc_nameseq_insert_publ - */ -struct publication *nameseq_insert_publ(struct name_seq *nseq, +struct publication *tipc_nameseq_insert_publ(struct name_seq *nseq, u32 type, u32 lower, u32 upper, u32 scope, u32 node, u32 port, u32 key) { @@ -285,7 +285,7 @@ struct publication *nameseq_insert_publ(struct name_seq *nseq, if (nseq->first_free == nseq->alloc) { struct sub_seq *sseqs = nseq->sseqs; - nseq->sseqs = subseq_alloc(nseq->alloc * 2); + nseq->sseqs = tipc_subseq_alloc(nseq->alloc * 2); if (nseq->sseqs != NULL) { memcpy(nseq->sseqs, sseqs, nseq->alloc * sizeof (struct sub_seq)); @@ -354,23 +354,23 @@ struct publication *nameseq_insert_publ(struct name_seq *nseq, */ list_for_each_entry_safe(s, st, &nseq->subscriptions, nameseq_list) { dbg("calling report_overlap()\n"); - subscr_report_overlap(s, - publ->lower, - publ->upper, - TIPC_PUBLISHED, - publ->ref, - publ->node, - created_subseq); + tipc_subscr_report_overlap(s, + publ->lower, + publ->upper, + TIPC_PUBLISHED, + publ->ref, + publ->node, + created_subseq); } return publ; } /** - * nameseq_remove_publ - + * tipc_nameseq_remove_publ - */ -struct publication *nameseq_remove_publ(struct name_seq *nseq, u32 inst, - u32 node, u32 ref, u32 key) +struct publication *tipc_nameseq_remove_publ(struct name_seq *nseq, u32 inst, + u32 node, u32 ref, u32 key) { struct publication *publ; struct publication *prev; @@ -470,24 +470,24 @@ struct publication *nameseq_remove_publ(struct name_seq *nseq, u32 inst, * Any subscriptions waiting ? */ list_for_each_entry_safe(s, st, &nseq->subscriptions, nameseq_list) { - subscr_report_overlap(s, - publ->lower, - publ->upper, - TIPC_WITHDRAWN, - publ->ref, - publ->node, - removed_subseq); + tipc_subscr_report_overlap(s, + publ->lower, + publ->upper, + TIPC_WITHDRAWN, + publ->ref, + publ->node, + removed_subseq); } return publ; } /** - * nameseq_subscribe: attach a subscription, and issue + * tipc_nameseq_subscribe: attach a subscription, and issue * the prescribed number of events if there is any sub- * sequence overlapping with the requested sequence */ -void nameseq_subscribe(struct name_seq *nseq, struct subscription *s) +void tipc_nameseq_subscribe(struct name_seq *nseq, struct subscription *s) { struct sub_seq *sseq = nseq->sseqs; @@ -498,18 +498,18 @@ void nameseq_subscribe(struct name_seq *nseq, struct subscription *s) while (sseq != &nseq->sseqs[nseq->first_free]) { struct publication *zl = sseq->zone_list; - if (zl && subscr_overlap(s,sseq->lower,sseq->upper)) { + if (zl && tipc_subscr_overlap(s,sseq->lower,sseq->upper)) { struct publication *crs = zl; int must_report = 1; do { - subscr_report_overlap(s, - sseq->lower, - sseq->upper, - TIPC_PUBLISHED, - crs->ref, - crs->node, - must_report); + tipc_subscr_report_overlap(s, + sseq->lower, + sseq->upper, + TIPC_PUBLISHED, + crs->ref, + crs->node, + must_report); must_report = 0; crs = crs->zone_list_next; } while (crs != zl); @@ -538,8 +538,8 @@ static struct name_seq *nametbl_find_seq(u32 type) return 0; }; -struct publication *nametbl_insert_publ(u32 type, u32 lower, u32 upper, - u32 scope, u32 node, u32 port, u32 key) +struct publication *tipc_nametbl_insert_publ(u32 type, u32 lower, u32 upper, + u32 scope, u32 node, u32 port, u32 key) { struct name_seq *seq = nametbl_find_seq(type); @@ -552,19 +552,19 @@ struct publication *nametbl_insert_publ(u32 type, u32 lower, u32 upper, dbg("Publishing <%u,%u,%u> from %x\n", type, lower, upper, node); if (!seq) { - seq = nameseq_create(type, &table.types[hash(type)]); - dbg("nametbl_insert_publ: created %x\n", seq); + seq = tipc_nameseq_create(type, &table.types[hash(type)]); + dbg("tipc_nametbl_insert_publ: created %x\n", seq); } if (!seq) return 0; assert(seq->type == type); - return nameseq_insert_publ(seq, type, lower, upper, - scope, node, port, key); + return tipc_nameseq_insert_publ(seq, type, lower, upper, + scope, node, port, key); } -struct publication *nametbl_remove_publ(u32 type, u32 lower, - u32 node, u32 ref, u32 key) +struct publication *tipc_nametbl_remove_publ(u32 type, u32 lower, + u32 node, u32 ref, u32 key) { struct publication *publ; struct name_seq *seq = nametbl_find_seq(type); @@ -573,7 +573,7 @@ struct publication *nametbl_remove_publ(u32 type, u32 lower, return 0; dbg("Withdrawing <%u,%u> from %x\n", type, lower, node); - publ = nameseq_remove_publ(seq, lower, node, ref, key); + publ = tipc_nameseq_remove_publ(seq, lower, node, ref, key); if (!seq->first_free && list_empty(&seq->subscriptions)) { hlist_del_init(&seq->ns_list); @@ -584,14 +584,14 @@ struct publication *nametbl_remove_publ(u32 type, u32 lower, } /* - * nametbl_translate(): Translate tipc_name -> tipc_portid. + * tipc_nametbl_translate(): Translate tipc_name -> tipc_portid. * Very time-critical. * * Note: on entry 'destnode' is the search domain used during translation; * on exit it passes back the node address of the matching port (if any) */ -u32 nametbl_translate(u32 type, u32 instance, u32 *destnode) +u32 tipc_nametbl_translate(u32 type, u32 instance, u32 *destnode) { struct sub_seq *sseq; struct publication *publ = 0; @@ -601,7 +601,7 @@ u32 nametbl_translate(u32 type, u32 instance, u32 *destnode) if (!in_scope(*destnode, tipc_own_addr)) return 0; - read_lock_bh(&nametbl_lock); + read_lock_bh(&tipc_nametbl_lock); seq = nametbl_find_seq(type); if (unlikely(!seq)) goto not_found; @@ -619,7 +619,7 @@ found: ref = publ->ref; *destnode = publ->node; spin_unlock_bh(&seq->lock); - read_unlock_bh(&nametbl_lock); + read_unlock_bh(&tipc_nametbl_lock); return ref; } publ = sseq->cluster_list; @@ -657,12 +657,12 @@ found: spin_unlock_bh(&seq->lock); not_found: *destnode = 0; - read_unlock_bh(&nametbl_lock); + read_unlock_bh(&tipc_nametbl_lock); return 0; } /** - * nametbl_mc_translate - find multicast destinations + * tipc_nametbl_mc_translate - find multicast destinations * * Creates list of all local ports that overlap the given multicast address; * also determines if any off-node ports overlap. @@ -674,15 +674,15 @@ not_found: * Returns non-zero if any off-node ports overlap */ -int nametbl_mc_translate(u32 type, u32 lower, u32 upper, u32 limit, - struct port_list *dports) +int tipc_nametbl_mc_translate(u32 type, u32 lower, u32 upper, u32 limit, + struct port_list *dports) { struct name_seq *seq; struct sub_seq *sseq; struct sub_seq *sseq_stop; int res = 0; - read_lock_bh(&nametbl_lock); + read_lock_bh(&tipc_nametbl_lock); seq = nametbl_find_seq(type); if (!seq) goto exit; @@ -700,7 +700,7 @@ int nametbl_mc_translate(u32 type, u32 lower, u32 upper, u32 limit, if (publ && (publ->scope <= limit)) do { if (publ->node == tipc_own_addr) - port_list_add(dports, publ->ref); + tipc_port_list_add(dports, publ->ref); else res = 1; publ = publ->cluster_list_next; @@ -709,15 +709,15 @@ int nametbl_mc_translate(u32 type, u32 lower, u32 upper, u32 limit, spin_unlock_bh(&seq->lock); exit: - read_unlock_bh(&nametbl_lock); + read_unlock_bh(&tipc_nametbl_lock); return res; } /** - * nametbl_publish_rsv - publish port name using a reserved name type + * tipc_nametbl_publish_rsv - publish port name using a reserved name type */ -int nametbl_publish_rsv(u32 ref, unsigned int scope, +int tipc_nametbl_publish_rsv(u32 ref, unsigned int scope, struct tipc_name_seq const *seq) { int res; @@ -729,10 +729,10 @@ int nametbl_publish_rsv(u32 ref, unsigned int scope, } /** - * nametbl_publish - add name publication to network name tables + * tipc_nametbl_publish - add name publication to network name tables */ -struct publication *nametbl_publish(u32 type, u32 lower, u32 upper, +struct publication *tipc_nametbl_publish(u32 type, u32 lower, u32 upper, u32 scope, u32 port_ref, u32 key) { struct publication *publ; @@ -748,77 +748,77 @@ struct publication *nametbl_publish(u32 type, u32 lower, u32 upper, return 0; } - write_lock_bh(&nametbl_lock); + write_lock_bh(&tipc_nametbl_lock); table.local_publ_count++; - publ = nametbl_insert_publ(type, lower, upper, scope, + publ = tipc_nametbl_insert_publ(type, lower, upper, scope, tipc_own_addr, port_ref, key); if (publ && (scope != TIPC_NODE_SCOPE)) { - named_publish(publ); + tipc_named_publish(publ); } - write_unlock_bh(&nametbl_lock); + write_unlock_bh(&tipc_nametbl_lock); return publ; } /** - * nametbl_withdraw - withdraw name publication from network name tables + * tipc_nametbl_withdraw - withdraw name publication from network name tables */ -int nametbl_withdraw(u32 type, u32 lower, u32 ref, u32 key) +int tipc_nametbl_withdraw(u32 type, u32 lower, u32 ref, u32 key) { struct publication *publ; - dbg("nametbl_withdraw:<%d,%d,%d>\n", type, lower, key); - write_lock_bh(&nametbl_lock); - publ = nametbl_remove_publ(type, lower, tipc_own_addr, ref, key); + dbg("tipc_nametbl_withdraw:<%d,%d,%d>\n", type, lower, key); + write_lock_bh(&tipc_nametbl_lock); + publ = tipc_nametbl_remove_publ(type, lower, tipc_own_addr, ref, key); if (publ) { table.local_publ_count--; if (publ->scope != TIPC_NODE_SCOPE) - named_withdraw(publ); - write_unlock_bh(&nametbl_lock); + tipc_named_withdraw(publ); + write_unlock_bh(&tipc_nametbl_lock); list_del_init(&publ->pport_list); kfree(publ); return 1; } - write_unlock_bh(&nametbl_lock); + write_unlock_bh(&tipc_nametbl_lock); return 0; } /** - * nametbl_subscribe - add a subscription object to the name table + * tipc_nametbl_subscribe - add a subscription object to the name table */ void -nametbl_subscribe(struct subscription *s) +tipc_nametbl_subscribe(struct subscription *s) { u32 type = s->seq.type; struct name_seq *seq; - write_lock_bh(&nametbl_lock); + write_lock_bh(&tipc_nametbl_lock); seq = nametbl_find_seq(type); if (!seq) { - seq = nameseq_create(type, &table.types[hash(type)]); + seq = tipc_nameseq_create(type, &table.types[hash(type)]); } if (seq){ spin_lock_bh(&seq->lock); - dbg("nametbl_subscribe:found %x for <%u,%u,%u>\n", + dbg("tipc_nametbl_subscribe:found %x for <%u,%u,%u>\n", seq, type, s->seq.lower, s->seq.upper); assert(seq->type == type); - nameseq_subscribe(seq, s); + tipc_nameseq_subscribe(seq, s); spin_unlock_bh(&seq->lock); } - write_unlock_bh(&nametbl_lock); + write_unlock_bh(&tipc_nametbl_lock); } /** - * nametbl_unsubscribe - remove a subscription object from name table + * tipc_nametbl_unsubscribe - remove a subscription object from name table */ void -nametbl_unsubscribe(struct subscription *s) +tipc_nametbl_unsubscribe(struct subscription *s) { struct name_seq *seq; - write_lock_bh(&nametbl_lock); + write_lock_bh(&tipc_nametbl_lock); seq = nametbl_find_seq(s->seq.type); if (seq != NULL){ spin_lock_bh(&seq->lock); @@ -830,7 +830,7 @@ nametbl_unsubscribe(struct subscription *s) kfree(seq); } } - write_unlock_bh(&nametbl_lock); + write_unlock_bh(&tipc_nametbl_lock); } @@ -983,17 +983,17 @@ static void nametbl_list(struct print_buf *buf, u32 depth_info, } } -void nametbl_print(struct print_buf *buf, const char *str) +void tipc_nametbl_print(struct print_buf *buf, const char *str) { tipc_printf(buf, str); - read_lock_bh(&nametbl_lock); + read_lock_bh(&tipc_nametbl_lock); nametbl_list(buf, 0, 0, 0, 0); - read_unlock_bh(&nametbl_lock); + read_unlock_bh(&tipc_nametbl_lock); } #define MAX_NAME_TBL_QUERY 32768 -struct sk_buff *nametbl_get(const void *req_tlv_area, int req_tlv_space) +struct sk_buff *tipc_nametbl_get(const void *req_tlv_area, int req_tlv_space) { struct sk_buff *buf; struct tipc_name_table_query *argv; @@ -1002,20 +1002,20 @@ struct sk_buff *nametbl_get(const void *req_tlv_area, int req_tlv_space) int str_len; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NAME_TBL_QUERY)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - buf = cfg_reply_alloc(TLV_SPACE(MAX_NAME_TBL_QUERY)); + buf = tipc_cfg_reply_alloc(TLV_SPACE(MAX_NAME_TBL_QUERY)); if (!buf) return NULL; rep_tlv = (struct tlv_desc *)buf->data; - printbuf_init(&b, TLV_DATA(rep_tlv), MAX_NAME_TBL_QUERY); + tipc_printbuf_init(&b, TLV_DATA(rep_tlv), MAX_NAME_TBL_QUERY); argv = (struct tipc_name_table_query *)TLV_DATA(req_tlv_area); - read_lock_bh(&nametbl_lock); + read_lock_bh(&tipc_nametbl_lock); nametbl_list(&b, ntohl(argv->depth), ntohl(argv->type), ntohl(argv->lowbound), ntohl(argv->upbound)); - read_unlock_bh(&nametbl_lock); - str_len = printbuf_validate(&b); + read_unlock_bh(&tipc_nametbl_lock); + str_len = tipc_printbuf_validate(&b); skb_put(buf, TLV_SPACE(str_len)); TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); @@ -1023,12 +1023,12 @@ struct sk_buff *nametbl_get(const void *req_tlv_area, int req_tlv_space) return buf; } -void nametbl_dump(void) +void tipc_nametbl_dump(void) { - nametbl_list(CONS, 0, 0, 0, 0); + nametbl_list(TIPC_CONS, 0, 0, 0, 0); } -int nametbl_init(void) +int tipc_nametbl_init(void) { int array_size = sizeof(struct hlist_head) * tipc_nametbl_size; @@ -1036,14 +1036,14 @@ int nametbl_init(void) if (!table.types) return -ENOMEM; - write_lock_bh(&nametbl_lock); + write_lock_bh(&tipc_nametbl_lock); memset(table.types, 0, array_size); table.local_publ_count = 0; - write_unlock_bh(&nametbl_lock); + write_unlock_bh(&tipc_nametbl_lock); return 0; } -void nametbl_stop(void) +void tipc_nametbl_stop(void) { struct hlist_head *seq_head; struct hlist_node *seq_node; @@ -1054,7 +1054,7 @@ void nametbl_stop(void) if (!table.types) return; - write_lock_bh(&nametbl_lock); + write_lock_bh(&tipc_nametbl_lock); for (i = 0; i < tipc_nametbl_size; i++) { seq_head = &table.types[i]; hlist_for_each_entry_safe(seq, seq_node, tmp, seq_head, ns_list) { @@ -1075,5 +1075,5 @@ void nametbl_stop(void) } kfree(table.types); table.types = NULL; - write_unlock_bh(&nametbl_lock); + write_unlock_bh(&tipc_nametbl_lock); } diff --git a/net/tipc/name_table.h b/net/tipc/name_table.h index f82693384f6..e8a3d71763c 100644 --- a/net/tipc/name_table.h +++ b/net/tipc/name_table.h @@ -85,24 +85,24 @@ struct publication { }; -extern rwlock_t nametbl_lock; +extern rwlock_t tipc_nametbl_lock; -struct sk_buff *nametbl_get(const void *req_tlv_area, int req_tlv_space); -u32 nametbl_translate(u32 type, u32 instance, u32 *node); -int nametbl_mc_translate(u32 type, u32 lower, u32 upper, u32 limit, +struct sk_buff *tipc_nametbl_get(const void *req_tlv_area, int req_tlv_space); +u32 tipc_nametbl_translate(u32 type, u32 instance, u32 *node); +int tipc_nametbl_mc_translate(u32 type, u32 lower, u32 upper, u32 limit, struct port_list *dports); -int nametbl_publish_rsv(u32 ref, unsigned int scope, +int tipc_nametbl_publish_rsv(u32 ref, unsigned int scope, struct tipc_name_seq const *seq); -struct publication *nametbl_publish(u32 type, u32 lower, u32 upper, +struct publication *tipc_nametbl_publish(u32 type, u32 lower, u32 upper, u32 scope, u32 port_ref, u32 key); -int nametbl_withdraw(u32 type, u32 lower, u32 ref, u32 key); -struct publication *nametbl_insert_publ(u32 type, u32 lower, u32 upper, +int tipc_nametbl_withdraw(u32 type, u32 lower, u32 ref, u32 key); +struct publication *tipc_nametbl_insert_publ(u32 type, u32 lower, u32 upper, u32 scope, u32 node, u32 ref, u32 key); -struct publication *nametbl_remove_publ(u32 type, u32 lower, +struct publication *tipc_nametbl_remove_publ(u32 type, u32 lower, u32 node, u32 ref, u32 key); -void nametbl_subscribe(struct subscription *s); -void nametbl_unsubscribe(struct subscription *s); -int nametbl_init(void); -void nametbl_stop(void); +void tipc_nametbl_subscribe(struct subscription *s); +void tipc_nametbl_unsubscribe(struct subscription *s); +int tipc_nametbl_init(void); +void tipc_nametbl_stop(void); #endif diff --git a/net/tipc/net.c b/net/tipc/net.c index 6826b493c1d..074891ad4f0 100644 --- a/net/tipc/net.c +++ b/net/tipc/net.c @@ -58,25 +58,25 @@ * 1: The routing hierarchy. * Comprises the structures 'zone', 'cluster', 'node', 'link' * and 'bearer'. The whole hierarchy is protected by a big - * read/write lock, net_lock, to enssure that nothing is added + * read/write lock, tipc_net_lock, to enssure that nothing is added * or removed while code is accessing any of these structures. * This layer must not be called from the two others while they * hold any of their own locks. * Neither must it itself do any upcalls to the other two before - * it has released net_lock and other protective locks. + * it has released tipc_net_lock and other protective locks. * - * Within the net_lock domain there are two sub-domains;'node' and + * Within the tipc_net_lock domain there are two sub-domains;'node' and * 'bearer', where local write operations are permitted, * provided that those are protected by individual spin_locks - * per instance. Code holding net_lock(read) and a node spin_lock + * per instance. Code holding tipc_net_lock(read) and a node spin_lock * is permitted to poke around in both the node itself and its * subordinate links. I.e, it can update link counters and queues, * change link state, send protocol messages, and alter the * "active_links" array in the node; but it can _not_ remove a link * or a node from the overall structure. * Correspondingly, individual bearers may change status within a - * net_lock(read), protected by an individual spin_lock ber bearer - * instance, but it needs net_lock(write) to remove/add any bearers. + * tipc_net_lock(read), protected by an individual spin_lock ber bearer + * instance, but it needs tipc_net_lock(write) to remove/add any bearers. * * * 2: The transport level of the protocol. @@ -97,91 +97,91 @@ * (Nobody is using read-only access to this, so it can just as * well be changed to a spin_lock) * - A spin lock to protect the registry of kernel/driver users (reg.c) - * - A global spin_lock (port_lock), which only task is to ensure + * - A global spin_lock (tipc_port_lock), which only task is to ensure * consistency where more than one port is involved in an operation, * i.e., whe a port is part of a linked list of ports. * There are two such lists; 'port_list', which is used for management, * and 'wait_list', which is used to queue ports during congestion. * * 3: The name table (name_table.c, name_distr.c, subscription.c) - * - There is one big read/write-lock (nametbl_lock) protecting the + * - There is one big read/write-lock (tipc_nametbl_lock) protecting the * overall name table structure. Nothing must be added/removed to * this structure without holding write access to it. * - There is one local spin_lock per sub_sequence, which can be seen - * as a sub-domain to the nametbl_lock domain. It is used only + * as a sub-domain to the tipc_nametbl_lock domain. It is used only * for translation operations, and is needed because a translation * steps the root of the 'publication' linked list between each lookup. - * This is always used within the scope of a nametbl_lock(read). + * This is always used within the scope of a tipc_nametbl_lock(read). * - A local spin_lock protecting the queue of subscriber events. */ -rwlock_t net_lock = RW_LOCK_UNLOCKED; -struct network net = { 0 }; +rwlock_t tipc_net_lock = RW_LOCK_UNLOCKED; +struct network tipc_net = { 0 }; -struct node *net_select_remote_node(u32 addr, u32 ref) +struct node *tipc_net_select_remote_node(u32 addr, u32 ref) { - return zone_select_remote_node(net.zones[tipc_zone(addr)], addr, ref); + return tipc_zone_select_remote_node(tipc_net.zones[tipc_zone(addr)], addr, ref); } -u32 net_select_router(u32 addr, u32 ref) +u32 tipc_net_select_router(u32 addr, u32 ref) { - return zone_select_router(net.zones[tipc_zone(addr)], addr, ref); + return tipc_zone_select_router(tipc_net.zones[tipc_zone(addr)], addr, ref); } -u32 net_next_node(u32 a) +u32 tipc_net_next_node(u32 a) { - if (net.zones[tipc_zone(a)]) - return zone_next_node(a); + if (tipc_net.zones[tipc_zone(a)]) + return tipc_zone_next_node(a); return 0; } -void net_remove_as_router(u32 router) +void tipc_net_remove_as_router(u32 router) { u32 z_num; for (z_num = 1; z_num <= tipc_max_zones; z_num++) { - if (!net.zones[z_num]) + if (!tipc_net.zones[z_num]) continue; - zone_remove_as_router(net.zones[z_num], router); + tipc_zone_remove_as_router(tipc_net.zones[z_num], router); } } -void net_send_external_routes(u32 dest) +void tipc_net_send_external_routes(u32 dest) { u32 z_num; for (z_num = 1; z_num <= tipc_max_zones; z_num++) { - if (net.zones[z_num]) - zone_send_external_routes(net.zones[z_num], dest); + if (tipc_net.zones[z_num]) + tipc_zone_send_external_routes(tipc_net.zones[z_num], dest); } } -int net_init(void) +static int net_init(void) { u32 sz = sizeof(struct _zone *) * (tipc_max_zones + 1); - memset(&net, 0, sizeof(net)); - net.zones = (struct _zone **)kmalloc(sz, GFP_ATOMIC); - if (!net.zones) { + memset(&tipc_net, 0, sizeof(tipc_net)); + tipc_net.zones = (struct _zone **)kmalloc(sz, GFP_ATOMIC); + if (!tipc_net.zones) { return -ENOMEM; } - memset(net.zones, 0, sz); + memset(tipc_net.zones, 0, sz); return TIPC_OK; } -void net_stop(void) +static void net_stop(void) { u32 z_num; - if (!net.zones) + if (!tipc_net.zones) return; for (z_num = 1; z_num <= tipc_max_zones; z_num++) { - zone_delete(net.zones[z_num]); + tipc_zone_delete(tipc_net.zones[z_num]); } - kfree(net.zones); - net.zones = 0; + kfree(tipc_net.zones); + tipc_net.zones = 0; } static void net_route_named_msg(struct sk_buff *buf) @@ -191,26 +191,26 @@ static void net_route_named_msg(struct sk_buff *buf) u32 dport; if (!msg_named(msg)) { - msg_dbg(msg, "net->drop_nam:"); + msg_dbg(msg, "tipc_net->drop_nam:"); buf_discard(buf); return; } dnode = addr_domain(msg_lookup_scope(msg)); - dport = nametbl_translate(msg_nametype(msg), msg_nameinst(msg), &dnode); - dbg("net->lookup<%u,%u>-><%u,%x>\n", + dport = tipc_nametbl_translate(msg_nametype(msg), msg_nameinst(msg), &dnode); + dbg("tipc_net->lookup<%u,%u>-><%u,%x>\n", msg_nametype(msg), msg_nameinst(msg), dport, dnode); if (dport) { msg_set_destnode(msg, dnode); msg_set_destport(msg, dport); - net_route_msg(buf); + tipc_net_route_msg(buf); return; } - msg_dbg(msg, "net->rej:NO NAME: "); + msg_dbg(msg, "tipc_net->rej:NO NAME: "); tipc_reject_msg(buf, TIPC_ERR_NO_NAME); } -void net_route_msg(struct sk_buff *buf) +void tipc_net_route_msg(struct sk_buff *buf) { struct tipc_msg *msg; u32 dnode; @@ -232,29 +232,29 @@ void net_route_msg(struct sk_buff *buf) return; } - msg_dbg(msg, "net->rout: "); + msg_dbg(msg, "tipc_net->rout: "); /* Handle message for this node */ dnode = msg_short(msg) ? tipc_own_addr : msg_destnode(msg); if (in_scope(dnode, tipc_own_addr)) { if (msg_isdata(msg)) { if (msg_mcast(msg)) - port_recv_mcast(buf, NULL); + tipc_port_recv_mcast(buf, NULL); else if (msg_destport(msg)) - port_recv_msg(buf); + tipc_port_recv_msg(buf); else net_route_named_msg(buf); return; } switch (msg_user(msg)) { case ROUTE_DISTRIBUTOR: - cluster_recv_routing_table(buf); + tipc_cltr_recv_routing_table(buf); break; case NAME_DISTRIBUTOR: - named_recv(buf); + tipc_named_recv(buf); break; case CONN_MANAGER: - port_recv_proto_msg(buf); + tipc_port_recv_proto_msg(buf); break; default: msg_dbg(msg,"DROP/NET/SEND>: "); - link_send(buf, dnode, msg_link_selector(msg)); + tipc_link_send(buf, dnode, msg_link_selector(msg)); } -int tipc_start_net(void) +int tipc_net_start(void) { char addr_string[16]; int res; @@ -277,35 +277,35 @@ int tipc_start_net(void) return -ENOPROTOOPT; tipc_mode = TIPC_NET_MODE; - named_reinit(); - port_reinit(); + tipc_named_reinit(); + tipc_port_reinit(); - if ((res = bearer_init()) || + if ((res = tipc_bearer_init()) || (res = net_init()) || - (res = cluster_init()) || - (res = bclink_init())) { + (res = tipc_cltr_init()) || + (res = tipc_bclink_init())) { return res; } - subscr_stop(); - cfg_stop(); - k_signal((Handler)subscr_start, 0); - k_signal((Handler)cfg_init, 0); + tipc_subscr_stop(); + tipc_cfg_stop(); + tipc_k_signal((Handler)tipc_subscr_start, 0); + tipc_k_signal((Handler)tipc_cfg_init, 0); info("Started in network mode\n"); info("Own node address %s, network identity %u\n", addr_string_fill(addr_string, tipc_own_addr), tipc_net_id); return TIPC_OK; } -void tipc_stop_net(void) +void tipc_net_stop(void) { if (tipc_mode != TIPC_NET_MODE) return; - write_lock_bh(&net_lock); - bearer_stop(); + write_lock_bh(&tipc_net_lock); + tipc_bearer_stop(); tipc_mode = TIPC_NODE_MODE; - bclink_stop(); + tipc_bclink_stop(); net_stop(); - write_unlock_bh(&net_lock); + write_unlock_bh(&tipc_net_lock); info("Left network mode \n"); } diff --git a/net/tipc/net.h b/net/tipc/net.h index 948c6d42102..f3e0b85e647 100644 --- a/net/tipc/net.h +++ b/net/tipc/net.h @@ -49,18 +49,16 @@ struct network { }; -extern struct network net; -extern rwlock_t net_lock; +extern struct network tipc_net; +extern rwlock_t tipc_net_lock; -int net_init(void); -void net_stop(void); -void net_remove_as_router(u32 router); -void net_send_external_routes(u32 dest); -void net_route_msg(struct sk_buff *buf); -struct node *net_select_remote_node(u32 addr, u32 ref); -u32 net_select_router(u32 addr, u32 ref); +void tipc_net_remove_as_router(u32 router); +void tipc_net_send_external_routes(u32 dest); +void tipc_net_route_msg(struct sk_buff *buf); +struct node *tipc_net_select_remote_node(u32 addr, u32 ref); +u32 tipc_net_select_router(u32 addr, u32 ref); -int tipc_start_net(void); -void tipc_stop_net(void); +int tipc_net_start(void); +void tipc_net_stop(void); #endif diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c index 19b3f402253..eb1bb4dce7a 100644 --- a/net/tipc/netlink.c +++ b/net/tipc/netlink.c @@ -47,13 +47,13 @@ static int handle_cmd(struct sk_buff *skb, struct genl_info *info) int hdr_space = NLMSG_SPACE(GENL_HDRLEN + TIPC_GENL_HDRLEN); if ((req_userhdr->cmd & 0xC000) && (!capable(CAP_NET_ADMIN))) - rep_buf = cfg_reply_error_string(TIPC_CFG_NOT_NET_ADMIN); + rep_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_NET_ADMIN); else - rep_buf = cfg_do_cmd(req_userhdr->dest, - req_userhdr->cmd, - NLMSG_DATA(req_nlh) + GENL_HDRLEN + TIPC_GENL_HDRLEN, - NLMSG_PAYLOAD(req_nlh, GENL_HDRLEN + TIPC_GENL_HDRLEN), - hdr_space); + rep_buf = tipc_cfg_do_cmd(req_userhdr->dest, + req_userhdr->cmd, + NLMSG_DATA(req_nlh) + GENL_HDRLEN + TIPC_GENL_HDRLEN, + NLMSG_PAYLOAD(req_nlh, GENL_HDRLEN + TIPC_GENL_HDRLEN), + hdr_space); if (rep_buf) { skb_push(rep_buf, hdr_space); @@ -81,7 +81,7 @@ static struct genl_ops ops = { static int family_registered = 0; -int netlink_start(void) +int tipc_netlink_start(void) { @@ -103,7 +103,7 @@ int netlink_start(void) return -EFAULT; } -void netlink_stop(void) +void tipc_netlink_stop(void) { if (family_registered) { genl_unregister_family(&family); diff --git a/net/tipc/node.c b/net/tipc/node.c index 05688d01138..6d65010e5fa 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -45,17 +45,16 @@ #include "port.h" #include "bearer.h" #include "name_distr.h" -#include "net.h" void node_print(struct print_buf *buf, struct node *n_ptr, char *str); static void node_lost_contact(struct node *n_ptr); static void node_established_contact(struct node *n_ptr); -struct node *nodes = NULL; /* sorted list of nodes within cluster */ +struct node *tipc_nodes = NULL; /* sorted list of nodes within cluster */ u32 tipc_own_tag = 0; -struct node *node_create(u32 addr) +struct node *tipc_node_create(u32 addr) { struct cluster *c_ptr; struct node *n_ptr; @@ -68,16 +67,16 @@ struct node *node_create(u32 addr) n_ptr->lock = SPIN_LOCK_UNLOCKED; INIT_LIST_HEAD(&n_ptr->nsub); - c_ptr = cluster_find(addr); + c_ptr = tipc_cltr_find(addr); if (c_ptr == NULL) - c_ptr = cluster_create(addr); + c_ptr = tipc_cltr_create(addr); if (c_ptr != NULL) { n_ptr->owner = c_ptr; - cluster_attach_node(c_ptr, n_ptr); + tipc_cltr_attach_node(c_ptr, n_ptr); n_ptr->last_router = -1; /* Insert node into ordered list */ - for (curr_node = &nodes; *curr_node; + for (curr_node = &tipc_nodes; *curr_node; curr_node = &(*curr_node)->next) { if (addr < (*curr_node)->addr) { n_ptr->next = *curr_node; @@ -93,13 +92,13 @@ struct node *node_create(u32 addr) return n_ptr; } -void node_delete(struct node *n_ptr) +void tipc_node_delete(struct node *n_ptr) { if (!n_ptr) return; #if 0 - /* Not needed because links are already deleted via bearer_stop() */ + /* Not needed because links are already deleted via tipc_bearer_stop() */ u32 l_num; @@ -114,12 +113,12 @@ void node_delete(struct node *n_ptr) /** - * node_link_up - handle addition of link + * tipc_node_link_up - handle addition of link * * Link becomes active (alone or shared) or standby, depending on its priority. */ -void node_link_up(struct node *n_ptr, struct link *l_ptr) +void tipc_node_link_up(struct node *n_ptr, struct link *l_ptr) { struct link **active = &n_ptr->active_links[0]; @@ -136,7 +135,7 @@ void node_link_up(struct node *n_ptr, struct link *l_ptr) info("Link is standby\n"); return; } - link_send_duplicate(active[0], l_ptr); + tipc_link_send_duplicate(active[0], l_ptr); if (l_ptr->priority == active[0]->priority) { active[0] = l_ptr; return; @@ -161,7 +160,7 @@ static void node_select_active_links(struct node *n_ptr) for (i = 0; i < MAX_BEARERS; i++) { struct link *l_ptr = n_ptr->links[i]; - if (!l_ptr || !link_is_up(l_ptr) || + if (!l_ptr || !tipc_link_is_up(l_ptr) || (l_ptr->priority < highest_prio)) continue; @@ -175,14 +174,14 @@ static void node_select_active_links(struct node *n_ptr) } /** - * node_link_down - handle loss of link + * tipc_node_link_down - handle loss of link */ -void node_link_down(struct node *n_ptr, struct link *l_ptr) +void tipc_node_link_down(struct node *n_ptr, struct link *l_ptr) { struct link **active; - if (!link_is_active(l_ptr)) { + if (!tipc_link_is_active(l_ptr)) { info("Lost standby link <%s> on network plane %c\n", l_ptr->name, l_ptr->b_ptr->net_plane); return; @@ -197,40 +196,40 @@ void node_link_down(struct node *n_ptr, struct link *l_ptr) active[1] = active[0]; if (active[0] == l_ptr) node_select_active_links(n_ptr); - if (node_is_up(n_ptr)) - link_changeover(l_ptr); + if (tipc_node_is_up(n_ptr)) + tipc_link_changeover(l_ptr); else node_lost_contact(n_ptr); } -int node_has_active_links(struct node *n_ptr) +int tipc_node_has_active_links(struct node *n_ptr) { return (n_ptr && ((n_ptr->active_links[0]) || (n_ptr->active_links[1]))); } -int node_has_redundant_links(struct node *n_ptr) +int tipc_node_has_redundant_links(struct node *n_ptr) { - return (node_has_active_links(n_ptr) && + return (tipc_node_has_active_links(n_ptr) && (n_ptr->active_links[0] != n_ptr->active_links[1])); } -int node_has_active_routes(struct node *n_ptr) +int tipc_node_has_active_routes(struct node *n_ptr) { return (n_ptr && (n_ptr->last_router >= 0)); } -int node_is_up(struct node *n_ptr) +int tipc_node_is_up(struct node *n_ptr) { - return (node_has_active_links(n_ptr) || node_has_active_routes(n_ptr)); + return (tipc_node_has_active_links(n_ptr) || tipc_node_has_active_routes(n_ptr)); } -struct node *node_attach_link(struct link *l_ptr) +struct node *tipc_node_attach_link(struct link *l_ptr) { - struct node *n_ptr = node_find(l_ptr->addr); + struct node *n_ptr = tipc_node_find(l_ptr->addr); if (!n_ptr) - n_ptr = node_create(l_ptr->addr); + n_ptr = tipc_node_create(l_ptr->addr); if (n_ptr) { u32 bearer_id = l_ptr->b_ptr->identity; char addr_string[16]; @@ -246,7 +245,7 @@ struct node *node_attach_link(struct link *l_ptr) if (!n_ptr->links[bearer_id]) { n_ptr->links[bearer_id] = l_ptr; - net.zones[tipc_zone(l_ptr->addr)]->links++; + tipc_net.zones[tipc_zone(l_ptr->addr)]->links++; n_ptr->link_cnt++; return n_ptr; } @@ -257,10 +256,10 @@ struct node *node_attach_link(struct link *l_ptr) return 0; } -void node_detach_link(struct node *n_ptr, struct link *l_ptr) +void tipc_node_detach_link(struct node *n_ptr, struct link *l_ptr) { n_ptr->links[l_ptr->b_ptr->identity] = 0; - net.zones[tipc_zone(l_ptr->addr)]->links--; + tipc_net.zones[tipc_zone(l_ptr->addr)]->links--; n_ptr->link_cnt--; } @@ -315,45 +314,45 @@ static void node_established_contact(struct node *n_ptr) struct cluster *c_ptr; dbg("node_established_contact:-> %x\n", n_ptr->addr); - if (!node_has_active_routes(n_ptr)) { - k_signal((Handler)named_node_up, n_ptr->addr); + if (!tipc_node_has_active_routes(n_ptr)) { + tipc_k_signal((Handler)tipc_named_node_up, n_ptr->addr); } /* Syncronize broadcast acks */ - n_ptr->bclink.acked = bclink_get_last_sent(); + n_ptr->bclink.acked = tipc_bclink_get_last_sent(); if (is_slave(tipc_own_addr)) return; if (!in_own_cluster(n_ptr->addr)) { /* Usage case 1 (see above) */ - c_ptr = cluster_find(tipc_own_addr); + c_ptr = tipc_cltr_find(tipc_own_addr); if (!c_ptr) - c_ptr = cluster_create(tipc_own_addr); + c_ptr = tipc_cltr_create(tipc_own_addr); if (c_ptr) - cluster_bcast_new_route(c_ptr, n_ptr->addr, 1, - tipc_max_nodes); + tipc_cltr_bcast_new_route(c_ptr, n_ptr->addr, 1, + tipc_max_nodes); return; } c_ptr = n_ptr->owner; if (is_slave(n_ptr->addr)) { /* Usage case 2 (see above) */ - cluster_bcast_new_route(c_ptr, n_ptr->addr, 1, tipc_max_nodes); - cluster_send_local_routes(c_ptr, n_ptr->addr); + tipc_cltr_bcast_new_route(c_ptr, n_ptr->addr, 1, tipc_max_nodes); + tipc_cltr_send_local_routes(c_ptr, n_ptr->addr); return; } if (n_ptr->bclink.supported) { - nmap_add(&cluster_bcast_nodes, n_ptr->addr); + tipc_nmap_add(&tipc_cltr_bcast_nodes, n_ptr->addr); if (n_ptr->addr < tipc_own_addr) tipc_own_tag++; } /* Case 3 (see above) */ - net_send_external_routes(n_ptr->addr); - cluster_send_slave_routes(c_ptr, n_ptr->addr); - cluster_bcast_new_route(c_ptr, n_ptr->addr, LOWEST_SLAVE, - highest_allowed_slave); + tipc_net_send_external_routes(n_ptr->addr); + tipc_cltr_send_slave_routes(c_ptr, n_ptr->addr); + tipc_cltr_bcast_new_route(c_ptr, n_ptr->addr, LOWEST_SLAVE, + tipc_highest_allowed_slave); } static void node_lost_contact(struct node *n_ptr) @@ -375,39 +374,39 @@ static void node_lost_contact(struct node *n_ptr) n_ptr->bclink.defragm = NULL; } if (in_own_cluster(n_ptr->addr) && n_ptr->bclink.supported) { - bclink_acknowledge(n_ptr, mod(n_ptr->bclink.acked + 10000)); + tipc_bclink_acknowledge(n_ptr, mod(n_ptr->bclink.acked + 10000)); } /* Update routing tables */ if (is_slave(tipc_own_addr)) { - net_remove_as_router(n_ptr->addr); + tipc_net_remove_as_router(n_ptr->addr); } else { if (!in_own_cluster(n_ptr->addr)) { /* Case 4 (see above) */ - c_ptr = cluster_find(tipc_own_addr); - cluster_bcast_lost_route(c_ptr, n_ptr->addr, 1, - tipc_max_nodes); + c_ptr = tipc_cltr_find(tipc_own_addr); + tipc_cltr_bcast_lost_route(c_ptr, n_ptr->addr, 1, + tipc_max_nodes); } else { /* Case 5 (see above) */ - c_ptr = cluster_find(n_ptr->addr); + c_ptr = tipc_cltr_find(n_ptr->addr); if (is_slave(n_ptr->addr)) { - cluster_bcast_lost_route(c_ptr, n_ptr->addr, 1, - tipc_max_nodes); + tipc_cltr_bcast_lost_route(c_ptr, n_ptr->addr, 1, + tipc_max_nodes); } else { if (n_ptr->bclink.supported) { - nmap_remove(&cluster_bcast_nodes, - n_ptr->addr); + tipc_nmap_remove(&tipc_cltr_bcast_nodes, + n_ptr->addr); if (n_ptr->addr < tipc_own_addr) tipc_own_tag--; } - net_remove_as_router(n_ptr->addr); - cluster_bcast_lost_route(c_ptr, n_ptr->addr, - LOWEST_SLAVE, - highest_allowed_slave); + tipc_net_remove_as_router(n_ptr->addr); + tipc_cltr_bcast_lost_route(c_ptr, n_ptr->addr, + LOWEST_SLAVE, + tipc_highest_allowed_slave); } } } - if (node_has_active_routes(n_ptr)) + if (tipc_node_has_active_routes(n_ptr)) return; info("Lost contact with %s\n", @@ -420,35 +419,35 @@ static void node_lost_contact(struct node *n_ptr) continue; l_ptr->reset_checkpoint = l_ptr->next_in_no; l_ptr->exp_msg_count = 0; - link_reset_fragments(l_ptr); + tipc_link_reset_fragments(l_ptr); } /* Notify subscribers */ list_for_each_entry_safe(ns, tns, &n_ptr->nsub, nodesub_list) { ns->node = 0; list_del_init(&ns->nodesub_list); - k_signal((Handler)ns->handle_node_down, - (unsigned long)ns->usr_handle); + tipc_k_signal((Handler)ns->handle_node_down, + (unsigned long)ns->usr_handle); } } /** - * node_select_next_hop - find the next-hop node for a message + * tipc_node_select_next_hop - find the next-hop node for a message * * Called by when cluster local lookup has failed. */ -struct node *node_select_next_hop(u32 addr, u32 selector) +struct node *tipc_node_select_next_hop(u32 addr, u32 selector) { struct node *n_ptr; u32 router_addr; - if (!addr_domain_valid(addr)) + if (!tipc_addr_domain_valid(addr)) return 0; /* Look for direct link to destination processsor */ - n_ptr = node_find(addr); - if (n_ptr && node_has_active_links(n_ptr)) + n_ptr = tipc_node_find(addr); + if (n_ptr && tipc_node_has_active_links(n_ptr)) return n_ptr; /* Cluster local system nodes *must* have direct links */ @@ -456,9 +455,9 @@ struct node *node_select_next_hop(u32 addr, u32 selector) return 0; /* Look for cluster local router with direct link to node */ - router_addr = node_select_router(n_ptr, selector); + router_addr = tipc_node_select_router(n_ptr, selector); if (router_addr) - return node_select(router_addr, selector); + return tipc_node_select(router_addr, selector); /* Slave nodes can only be accessed within own cluster via a known router with direct link -- if no router was found,give up */ @@ -467,25 +466,25 @@ struct node *node_select_next_hop(u32 addr, u32 selector) /* Inter zone/cluster -- find any direct link to remote cluster */ addr = tipc_addr(tipc_zone(addr), tipc_cluster(addr), 0); - n_ptr = net_select_remote_node(addr, selector); - if (n_ptr && node_has_active_links(n_ptr)) + n_ptr = tipc_net_select_remote_node(addr, selector); + if (n_ptr && tipc_node_has_active_links(n_ptr)) return n_ptr; /* Last resort -- look for any router to anywhere in remote zone */ - router_addr = net_select_router(addr, selector); + router_addr = tipc_net_select_router(addr, selector); if (router_addr) - return node_select(router_addr, selector); + return tipc_node_select(router_addr, selector); return 0; } /** - * node_select_router - select router to reach specified node + * tipc_node_select_router - select router to reach specified node * * Uses a deterministic and fair algorithm for selecting router node. */ -u32 node_select_router(struct node *n_ptr, u32 ref) +u32 tipc_node_select_router(struct node *n_ptr, u32 ref) { u32 ulim; u32 mask; @@ -523,7 +522,7 @@ u32 node_select_router(struct node *n_ptr, u32 ref) return tipc_addr(own_zone(), own_cluster(), r); } -void node_add_router(struct node *n_ptr, u32 router) +void tipc_node_add_router(struct node *n_ptr, u32 router) { u32 r_num = tipc_node(router); @@ -534,7 +533,7 @@ void node_add_router(struct node *n_ptr, u32 router) !n_ptr->routers[n_ptr->last_router]); } -void node_remove_router(struct node *n_ptr, u32 router) +void tipc_node_remove_router(struct node *n_ptr, u32 router) { u32 r_num = tipc_node(router); @@ -547,7 +546,7 @@ void node_remove_router(struct node *n_ptr, u32 router) while ((--n_ptr->last_router >= 0) && !n_ptr->routers[n_ptr->last_router]); - if (!node_is_up(n_ptr)) + if (!tipc_node_is_up(n_ptr)) node_lost_contact(n_ptr); } @@ -572,16 +571,16 @@ u32 tipc_available_nodes(const u32 domain) struct node *n_ptr; u32 cnt = 0; - for (n_ptr = nodes; n_ptr; n_ptr = n_ptr->next) { + for (n_ptr = tipc_nodes; n_ptr; n_ptr = n_ptr->next) { if (!in_scope(domain, n_ptr->addr)) continue; - if (node_is_up(n_ptr)) + if (tipc_node_is_up(n_ptr)) cnt++; } return cnt; } -struct sk_buff *node_get_nodes(const void *req_tlv_area, int req_tlv_space) +struct sk_buff *tipc_node_get_nodes(const void *req_tlv_area, int req_tlv_space) { u32 domain; struct sk_buff *buf; @@ -589,40 +588,40 @@ struct sk_buff *node_get_nodes(const void *req_tlv_area, int req_tlv_space) struct tipc_node_info node_info; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); domain = *(u32 *)TLV_DATA(req_tlv_area); domain = ntohl(domain); - if (!addr_domain_valid(domain)) - return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (network address)"); + if (!tipc_addr_domain_valid(domain)) + return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (network address)"); - if (!nodes) - return cfg_reply_none(); + if (!tipc_nodes) + return tipc_cfg_reply_none(); /* For now, get space for all other nodes (will need to modify this when slave nodes are supported */ - buf = cfg_reply_alloc(TLV_SPACE(sizeof(node_info)) * - (tipc_max_nodes - 1)); + buf = tipc_cfg_reply_alloc(TLV_SPACE(sizeof(node_info)) * + (tipc_max_nodes - 1)); if (!buf) return NULL; /* Add TLVs for all nodes in scope */ - for (n_ptr = nodes; n_ptr; n_ptr = n_ptr->next) { + for (n_ptr = tipc_nodes; n_ptr; n_ptr = n_ptr->next) { if (!in_scope(domain, n_ptr->addr)) continue; node_info.addr = htonl(n_ptr->addr); - node_info.up = htonl(node_is_up(n_ptr)); - cfg_append_tlv(buf, TIPC_TLV_NODE_INFO, - &node_info, sizeof(node_info)); + node_info.up = htonl(tipc_node_is_up(n_ptr)); + tipc_cfg_append_tlv(buf, TIPC_TLV_NODE_INFO, + &node_info, sizeof(node_info)); } return buf; } -struct sk_buff *node_get_links(const void *req_tlv_area, int req_tlv_space) +struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) { u32 domain; struct sk_buff *buf; @@ -630,22 +629,22 @@ struct sk_buff *node_get_links(const void *req_tlv_area, int req_tlv_space) struct tipc_link_info link_info; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); domain = *(u32 *)TLV_DATA(req_tlv_area); domain = ntohl(domain); - if (!addr_domain_valid(domain)) - return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (network address)"); + if (!tipc_addr_domain_valid(domain)) + return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (network address)"); - if (!nodes) - return cfg_reply_none(); + if (!tipc_nodes) + return tipc_cfg_reply_none(); /* For now, get space for 2 links to all other nodes + bcast link (will need to modify this when slave nodes are supported */ - buf = cfg_reply_alloc(TLV_SPACE(sizeof(link_info)) * - (2 * (tipc_max_nodes - 1) + 1)); + buf = tipc_cfg_reply_alloc(TLV_SPACE(sizeof(link_info)) * + (2 * (tipc_max_nodes - 1) + 1)); if (!buf) return NULL; @@ -654,12 +653,12 @@ struct sk_buff *node_get_links(const void *req_tlv_area, int req_tlv_space) link_info.dest = tipc_own_addr & 0xfffff00; link_info.dest = htonl(link_info.dest); link_info.up = htonl(1); - sprintf(link_info.str, bc_link_name); - cfg_append_tlv(buf, TIPC_TLV_LINK_INFO, &link_info, sizeof(link_info)); + sprintf(link_info.str, tipc_bclink_name); + tipc_cfg_append_tlv(buf, TIPC_TLV_LINK_INFO, &link_info, sizeof(link_info)); /* Add TLVs for any other links in scope */ - for (n_ptr = nodes; n_ptr; n_ptr = n_ptr->next) { + for (n_ptr = tipc_nodes; n_ptr; n_ptr = n_ptr->next) { u32 i; if (!in_scope(domain, n_ptr->addr)) @@ -668,10 +667,10 @@ struct sk_buff *node_get_links(const void *req_tlv_area, int req_tlv_space) if (!n_ptr->links[i]) continue; link_info.dest = htonl(n_ptr->addr); - link_info.up = htonl(link_is_up(n_ptr->links[i])); + link_info.up = htonl(tipc_link_is_up(n_ptr->links[i])); strcpy(link_info.str, n_ptr->links[i]->name); - cfg_append_tlv(buf, TIPC_TLV_LINK_INFO, - &link_info, sizeof(link_info)); + tipc_cfg_append_tlv(buf, TIPC_TLV_LINK_INFO, + &link_info, sizeof(link_info)); } } diff --git a/net/tipc/node.h b/net/tipc/node.h index b39442badcc..29f7ae6992d 100644 --- a/net/tipc/node.h +++ b/net/tipc/node.h @@ -92,31 +92,31 @@ struct node { } bclink; }; -extern struct node *nodes; +extern struct node *tipc_nodes; extern u32 tipc_own_tag; -struct node *node_create(u32 addr); -void node_delete(struct node *n_ptr); -struct node *node_attach_link(struct link *l_ptr); -void node_detach_link(struct node *n_ptr, struct link *l_ptr); -void node_link_down(struct node *n_ptr, struct link *l_ptr); -void node_link_up(struct node *n_ptr, struct link *l_ptr); -int node_has_active_links(struct node *n_ptr); -int node_has_redundant_links(struct node *n_ptr); -u32 node_select_router(struct node *n_ptr, u32 ref); -struct node *node_select_next_hop(u32 addr, u32 selector); -int node_is_up(struct node *n_ptr); -void node_add_router(struct node *n_ptr, u32 router); -void node_remove_router(struct node *n_ptr, u32 router); -struct sk_buff *node_get_links(const void *req_tlv_area, int req_tlv_space); -struct sk_buff *node_get_nodes(const void *req_tlv_area, int req_tlv_space); +struct node *tipc_node_create(u32 addr); +void tipc_node_delete(struct node *n_ptr); +struct node *tipc_node_attach_link(struct link *l_ptr); +void tipc_node_detach_link(struct node *n_ptr, struct link *l_ptr); +void tipc_node_link_down(struct node *n_ptr, struct link *l_ptr); +void tipc_node_link_up(struct node *n_ptr, struct link *l_ptr); +int tipc_node_has_active_links(struct node *n_ptr); +int tipc_node_has_redundant_links(struct node *n_ptr); +u32 tipc_node_select_router(struct node *n_ptr, u32 ref); +struct node *tipc_node_select_next_hop(u32 addr, u32 selector); +int tipc_node_is_up(struct node *n_ptr); +void tipc_node_add_router(struct node *n_ptr, u32 router); +void tipc_node_remove_router(struct node *n_ptr, u32 router); +struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space); +struct sk_buff *tipc_node_get_nodes(const void *req_tlv_area, int req_tlv_space); -static inline struct node *node_find(u32 addr) +static inline struct node *tipc_node_find(u32 addr) { if (likely(in_own_cluster(addr))) - return local_nodes[tipc_node(addr)]; - else if (addr_domain_valid(addr)) { - struct cluster *c_ptr = cluster_find(addr); + return tipc_local_nodes[tipc_node(addr)]; + else if (tipc_addr_domain_valid(addr)) { + struct cluster *c_ptr = tipc_cltr_find(addr); if (c_ptr) return c_ptr->nodes[tipc_node(addr)]; @@ -124,19 +124,19 @@ static inline struct node *node_find(u32 addr) return 0; } -static inline struct node *node_select(u32 addr, u32 selector) +static inline struct node *tipc_node_select(u32 addr, u32 selector) { if (likely(in_own_cluster(addr))) - return local_nodes[tipc_node(addr)]; - return node_select_next_hop(addr, selector); + return tipc_local_nodes[tipc_node(addr)]; + return tipc_node_select_next_hop(addr, selector); } -static inline void node_lock(struct node *n_ptr) +static inline void tipc_node_lock(struct node *n_ptr) { spin_lock_bh(&n_ptr->lock); } -static inline void node_unlock(struct node *n_ptr) +static inline void tipc_node_unlock(struct node *n_ptr) { spin_unlock_bh(&n_ptr->lock); } diff --git a/net/tipc/node_subscr.c b/net/tipc/node_subscr.c index 79375927916..afeea121d8b 100644 --- a/net/tipc/node_subscr.c +++ b/net/tipc/node_subscr.c @@ -41,39 +41,39 @@ #include "addr.h" /** - * nodesub_subscribe - create "node down" subscription for specified node + * tipc_nodesub_subscribe - create "node down" subscription for specified node */ -void nodesub_subscribe(struct node_subscr *node_sub, u32 addr, +void tipc_nodesub_subscribe(struct node_subscr *node_sub, u32 addr, void *usr_handle, net_ev_handler handle_down) { node_sub->node = 0; if (addr == tipc_own_addr) return; - if (!addr_node_valid(addr)) { + if (!tipc_addr_node_valid(addr)) { warn("node_subscr with illegal %x\n", addr); return; } node_sub->handle_node_down = handle_down; node_sub->usr_handle = usr_handle; - node_sub->node = node_find(addr); + node_sub->node = tipc_node_find(addr); assert(node_sub->node); - node_lock(node_sub->node); + tipc_node_lock(node_sub->node); list_add_tail(&node_sub->nodesub_list, &node_sub->node->nsub); - node_unlock(node_sub->node); + tipc_node_unlock(node_sub->node); } /** - * nodesub_unsubscribe - cancel "node down" subscription (if any) + * tipc_nodesub_unsubscribe - cancel "node down" subscription (if any) */ -void nodesub_unsubscribe(struct node_subscr *node_sub) +void tipc_nodesub_unsubscribe(struct node_subscr *node_sub) { if (!node_sub->node) return; - node_lock(node_sub->node); + tipc_node_lock(node_sub->node); list_del_init(&node_sub->nodesub_list); - node_unlock(node_sub->node); + tipc_node_unlock(node_sub->node); } diff --git a/net/tipc/node_subscr.h b/net/tipc/node_subscr.h index a3b87ac4859..01751c4fbb4 100644 --- a/net/tipc/node_subscr.h +++ b/net/tipc/node_subscr.h @@ -56,8 +56,8 @@ struct node_subscr { struct list_head nodesub_list; }; -void nodesub_subscribe(struct node_subscr *node_sub, u32 addr, - void *usr_handle, net_ev_handler handle_down); -void nodesub_unsubscribe(struct node_subscr *node_sub); +void tipc_nodesub_subscribe(struct node_subscr *node_sub, u32 addr, + void *usr_handle, net_ev_handler handle_down); +void tipc_nodesub_unsubscribe(struct node_subscr *node_sub); #endif diff --git a/net/tipc/port.c b/net/tipc/port.c index 66caca7abe9..72aae52bfec 100644 --- a/net/tipc/port.c +++ b/net/tipc/port.c @@ -57,10 +57,10 @@ static struct sk_buff *msg_queue_head = 0; static struct sk_buff *msg_queue_tail = 0; -spinlock_t port_list_lock = SPIN_LOCK_UNLOCKED; +spinlock_t tipc_port_list_lock = SPIN_LOCK_UNLOCKED; static spinlock_t queue_lock = SPIN_LOCK_UNLOCKED; -LIST_HEAD(ports); +static LIST_HEAD(ports); static void port_handle_node_down(unsigned long ref); static struct sk_buff* port_build_self_abort_msg(struct port *,u32 err); static struct sk_buff* port_build_peer_abort_msg(struct port *,u32 err); @@ -107,7 +107,7 @@ int tipc_multicast(u32 ref, struct tipc_name_seq const *seq, u32 domain, struct sk_buff *buf; struct sk_buff *ibuf = NULL; struct port_list dports = {0, NULL, }; - struct port *oport = port_deref(ref); + struct port *oport = tipc_port_deref(ref); int ext_targets; int res; @@ -129,8 +129,8 @@ int tipc_multicast(u32 ref, struct tipc_name_seq const *seq, u32 domain, /* Figure out where to send multicast message */ - ext_targets = nametbl_mc_translate(seq->type, seq->lower, seq->upper, - TIPC_NODE_SCOPE, &dports); + ext_targets = tipc_nametbl_mc_translate(seq->type, seq->lower, seq->upper, + TIPC_NODE_SCOPE, &dports); /* Send message to destinations (duplicate it only if necessary) */ @@ -138,12 +138,12 @@ int tipc_multicast(u32 ref, struct tipc_name_seq const *seq, u32 domain, if (dports.count != 0) { ibuf = skb_copy(buf, GFP_ATOMIC); if (ibuf == NULL) { - port_list_free(&dports); + tipc_port_list_free(&dports); buf_discard(buf); return -ENOMEM; } } - res = bclink_send_msg(buf); + res = tipc_bclink_send_msg(buf); if ((res < 0) && (dports.count != 0)) { buf_discard(ibuf); } @@ -153,20 +153,20 @@ int tipc_multicast(u32 ref, struct tipc_name_seq const *seq, u32 domain, if (res >= 0) { if (ibuf) - port_recv_mcast(ibuf, &dports); + tipc_port_recv_mcast(ibuf, &dports); } else { - port_list_free(&dports); + tipc_port_list_free(&dports); } return res; } /** - * port_recv_mcast - deliver multicast message to all destination ports + * tipc_port_recv_mcast - deliver multicast message to all destination ports * * If there is no port list, perform a lookup to create one */ -void port_recv_mcast(struct sk_buff *buf, struct port_list *dp) +void tipc_port_recv_mcast(struct sk_buff *buf, struct port_list *dp) { struct tipc_msg* msg; struct port_list dports = {0, NULL, }; @@ -179,7 +179,7 @@ void port_recv_mcast(struct sk_buff *buf, struct port_list *dp) /* Create destination port list, if one wasn't supplied */ if (dp == NULL) { - nametbl_mc_translate(msg_nametype(msg), + tipc_nametbl_mc_translate(msg_nametype(msg), msg_namelower(msg), msg_nameupper(msg), TIPC_CLUSTER_SCOPE, @@ -192,8 +192,8 @@ void port_recv_mcast(struct sk_buff *buf, struct port_list *dp) if (dp->count != 0) { if (dp->count == 1) { msg_set_destport(msg, dp->ports[0]); - port_recv_msg(buf); - port_list_free(dp); + tipc_port_recv_msg(buf); + tipc_port_list_free(dp); return; } for (; cnt < dp->count; cnt++) { @@ -209,12 +209,12 @@ void port_recv_mcast(struct sk_buff *buf, struct port_list *dp) item = item->next; } msg_set_destport(buf_msg(b),item->ports[index]); - port_recv_msg(b); + tipc_port_recv_msg(b); } } exit: buf_discard(buf); - port_list_free(dp); + tipc_port_list_free(dp); } /** @@ -238,14 +238,14 @@ u32 tipc_createport_raw(void *usr_handle, return 0; } memset(p_ptr, 0, sizeof(*p_ptr)); - ref = ref_acquire(p_ptr, &p_ptr->publ.lock); + ref = tipc_ref_acquire(p_ptr, &p_ptr->publ.lock); if (!ref) { warn("Reference Table Exhausted\n"); kfree(p_ptr); return 0; } - port_lock(ref); + tipc_port_lock(ref); p_ptr->publ.ref = ref; msg = &p_ptr->publ.phdr; msg_init(msg, DATA_LOW, TIPC_NAMED_MSG, TIPC_OK, LONG_H_SIZE, 0); @@ -264,12 +264,12 @@ u32 tipc_createport_raw(void *usr_handle, p_ptr->wakeup = wakeup; p_ptr->user_port = 0; k_init_timer(&p_ptr->timer, (Handler)port_timeout, ref); - spin_lock_bh(&port_list_lock); + spin_lock_bh(&tipc_port_list_lock); INIT_LIST_HEAD(&p_ptr->publications); INIT_LIST_HEAD(&p_ptr->port_list); list_add_tail(&p_ptr->port_list, &ports); - spin_unlock_bh(&port_list_lock); - port_unlock(p_ptr); + spin_unlock_bh(&tipc_port_list_lock); + tipc_port_unlock(p_ptr); return ref; } @@ -279,31 +279,31 @@ int tipc_deleteport(u32 ref) struct sk_buff *buf = 0; tipc_withdraw(ref, 0, 0); - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; - ref_discard(ref); - port_unlock(p_ptr); + tipc_ref_discard(ref); + tipc_port_unlock(p_ptr); k_cancel_timer(&p_ptr->timer); if (p_ptr->publ.connected) { buf = port_build_peer_abort_msg(p_ptr, TIPC_ERR_NO_PORT); - nodesub_unsubscribe(&p_ptr->subscription); + tipc_nodesub_unsubscribe(&p_ptr->subscription); } if (p_ptr->user_port) { - reg_remove_port(p_ptr->user_port); + tipc_reg_remove_port(p_ptr->user_port); kfree(p_ptr->user_port); } - spin_lock_bh(&port_list_lock); + spin_lock_bh(&tipc_port_list_lock); list_del(&p_ptr->port_list); list_del(&p_ptr->wait_list); - spin_unlock_bh(&port_list_lock); + spin_unlock_bh(&tipc_port_list_lock); k_term_timer(&p_ptr->timer); kfree(p_ptr); dbg("Deleted port %u\n", ref); - net_route_msg(buf); + tipc_net_route_msg(buf); return TIPC_OK; } @@ -315,7 +315,7 @@ int tipc_deleteport(u32 ref) struct tipc_port *tipc_get_port(const u32 ref) { - return (struct tipc_port *)ref_deref(ref); + return (struct tipc_port *)tipc_ref_deref(ref); } /** @@ -327,11 +327,11 @@ void *tipc_get_handle(const u32 ref) struct port *p_ptr; void * handle; - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) return 0; handle = p_ptr->publ.usr_handle; - port_unlock(p_ptr); + tipc_port_unlock(p_ptr); return handle; } @@ -344,7 +344,7 @@ int tipc_portunreliable(u32 ref, unsigned int *isunreliable) { struct port *p_ptr; - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; *isunreliable = port_unreliable(p_ptr); @@ -356,11 +356,11 @@ int tipc_set_portunreliable(u32 ref, unsigned int isunreliable) { struct port *p_ptr; - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; msg_set_src_droppable(&p_ptr->publ.phdr, (isunreliable != 0)); - port_unlock(p_ptr); + tipc_port_unlock(p_ptr); return TIPC_OK; } @@ -373,7 +373,7 @@ int tipc_portunreturnable(u32 ref, unsigned int *isunrejectable) { struct port *p_ptr; - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; *isunrejectable = port_unreturnable(p_ptr); @@ -385,11 +385,11 @@ int tipc_set_portunreturnable(u32 ref, unsigned int isunrejectable) { struct port *p_ptr; - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; msg_set_dest_droppable(&p_ptr->publ.phdr, (isunrejectable != 0)); - port_unlock(p_ptr); + tipc_port_unlock(p_ptr); return TIPC_OK; } @@ -476,25 +476,25 @@ int tipc_reject_msg(struct sk_buff *buf, u32 err) /* send self-abort message when rejecting on a connected port */ if (msg_connected(msg)) { struct sk_buff *abuf = 0; - struct port *p_ptr = port_lock(msg_destport(msg)); + struct port *p_ptr = tipc_port_lock(msg_destport(msg)); if (p_ptr) { if (p_ptr->publ.connected) abuf = port_build_self_abort_msg(p_ptr, err); - port_unlock(p_ptr); + tipc_port_unlock(p_ptr); } - net_route_msg(abuf); + tipc_net_route_msg(abuf); } /* send rejected message */ buf_discard(buf); - net_route_msg(rbuf); + tipc_net_route_msg(rbuf); return data_sz; } -int port_reject_sections(struct port *p_ptr, struct tipc_msg *hdr, - struct iovec const *msg_sect, u32 num_sect, - int err) +int tipc_port_reject_sections(struct port *p_ptr, struct tipc_msg *hdr, + struct iovec const *msg_sect, u32 num_sect, + int err) { struct sk_buff *buf; int res; @@ -509,7 +509,7 @@ int port_reject_sections(struct port *p_ptr, struct tipc_msg *hdr, static void port_timeout(unsigned long ref) { - struct port *p_ptr = port_lock(ref); + struct port *p_ptr = tipc_port_lock(ref); struct sk_buff *buf = 0; if (!p_ptr || !p_ptr->publ.connected) @@ -532,21 +532,21 @@ static void port_timeout(unsigned long ref) p_ptr->probing_state = PROBING; k_start_timer(&p_ptr->timer, p_ptr->probing_interval); } - port_unlock(p_ptr); - net_route_msg(buf); + tipc_port_unlock(p_ptr); + tipc_net_route_msg(buf); } static void port_handle_node_down(unsigned long ref) { - struct port *p_ptr = port_lock(ref); + struct port *p_ptr = tipc_port_lock(ref); struct sk_buff* buf = 0; if (!p_ptr) return; buf = port_build_self_abort_msg(p_ptr, TIPC_ERR_NO_NODE); - port_unlock(p_ptr); - net_route_msg(buf); + tipc_port_unlock(p_ptr); + tipc_net_route_msg(buf); } @@ -589,10 +589,10 @@ static struct sk_buff *port_build_peer_abort_msg(struct port *p_ptr, u32 err) 0); } -void port_recv_proto_msg(struct sk_buff *buf) +void tipc_port_recv_proto_msg(struct sk_buff *buf) { struct tipc_msg *msg = buf_msg(buf); - struct port *p_ptr = port_lock(msg_destport(msg)); + struct port *p_ptr = tipc_port_lock(msg_destport(msg)); u32 err = TIPC_OK; struct sk_buff *r_buf = 0; struct sk_buff *abort_buf = 0; @@ -615,11 +615,11 @@ void port_recv_proto_msg(struct sk_buff *buf) } } if (msg_type(msg) == CONN_ACK) { - int wakeup = port_congested(p_ptr) && + int wakeup = tipc_port_congested(p_ptr) && p_ptr->publ.congested && p_ptr->wakeup; p_ptr->acked += msg_msgcnt(msg); - if (port_congested(p_ptr)) + if (tipc_port_congested(p_ptr)) goto exit; p_ptr->publ.congested = 0; if (!wakeup) @@ -659,9 +659,9 @@ void port_recv_proto_msg(struct sk_buff *buf) port_incr_out_seqno(p_ptr); exit: if (p_ptr) - port_unlock(p_ptr); - net_route_msg(r_buf); - net_route_msg(abort_buf); + tipc_port_unlock(p_ptr); + tipc_net_route_msg(r_buf); + tipc_net_route_msg(abort_buf); buf_discard(buf); } @@ -704,7 +704,7 @@ static void port_print(struct port *p_ptr, struct print_buf *buf, int full_id) #define MAX_PORT_QUERY 32768 -struct sk_buff *port_get_ports(void) +struct sk_buff *tipc_port_get_ports(void) { struct sk_buff *buf; struct tlv_desc *rep_tlv; @@ -712,20 +712,20 @@ struct sk_buff *port_get_ports(void) struct port *p_ptr; int str_len; - buf = cfg_reply_alloc(TLV_SPACE(MAX_PORT_QUERY)); + buf = tipc_cfg_reply_alloc(TLV_SPACE(MAX_PORT_QUERY)); if (!buf) return NULL; rep_tlv = (struct tlv_desc *)buf->data; - printbuf_init(&pb, TLV_DATA(rep_tlv), MAX_PORT_QUERY); - spin_lock_bh(&port_list_lock); + tipc_printbuf_init(&pb, TLV_DATA(rep_tlv), MAX_PORT_QUERY); + spin_lock_bh(&tipc_port_list_lock); list_for_each_entry(p_ptr, &ports, port_list) { spin_lock_bh(p_ptr->publ.lock); port_print(p_ptr, &pb, 0); spin_unlock_bh(p_ptr->publ.lock); } - spin_unlock_bh(&port_list_lock); - str_len = printbuf_validate(&pb); + spin_unlock_bh(&tipc_port_list_lock); + str_len = tipc_printbuf_validate(&pb); skb_put(buf, TLV_SPACE(str_len)); TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); @@ -752,22 +752,22 @@ struct sk_buff *port_show_stats(const void *req_tlv_area, int req_tlv_space) ref = *(u32 *)TLV_DATA(req_tlv_area); ref = ntohl(ref); - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) return cfg_reply_error_string("port not found"); - buf = cfg_reply_alloc(TLV_SPACE(MAX_PORT_STATS)); + buf = tipc_cfg_reply_alloc(TLV_SPACE(MAX_PORT_STATS)); if (!buf) { - port_unlock(p_ptr); + tipc_port_unlock(p_ptr); return NULL; } rep_tlv = (struct tlv_desc *)buf->data; - printbuf_init(&pb, TLV_DATA(rep_tlv), MAX_PORT_STATS); + tipc_printbuf_init(&pb, TLV_DATA(rep_tlv), MAX_PORT_STATS); port_print(p_ptr, &pb, 1); /* NEED TO FILL IN ADDITIONAL PORT STATISTICS HERE */ - port_unlock(p_ptr); - str_len = printbuf_validate(&pb); + tipc_port_unlock(p_ptr); + str_len = tipc_printbuf_validate(&pb); skb_put(buf, TLV_SPACE(str_len)); TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); @@ -777,19 +777,19 @@ struct sk_buff *port_show_stats(const void *req_tlv_area, int req_tlv_space) #endif -void port_reinit(void) +void tipc_port_reinit(void) { struct port *p_ptr; struct tipc_msg *msg; - spin_lock_bh(&port_list_lock); + spin_lock_bh(&tipc_port_list_lock); list_for_each_entry(p_ptr, &ports, port_list) { msg = &p_ptr->publ.phdr; if (msg_orignode(msg) == tipc_own_addr) break; msg_set_orignode(msg, tipc_own_addr); } - spin_unlock_bh(&port_list_lock); + spin_unlock_bh(&tipc_port_list_lock); } @@ -820,7 +820,7 @@ static void port_dispatcher_sigh(void *dummy) struct tipc_msg *msg = buf_msg(buf); u32 dref = msg_destport(msg); - p_ptr = port_lock(dref); + p_ptr = tipc_port_lock(dref); if (!p_ptr) { /* Port deleted while msg in queue */ tipc_reject_msg(buf, TIPC_ERR_NO_PORT); @@ -976,7 +976,7 @@ static u32 port_dispatcher(struct tipc_port *dummy, struct sk_buff *buf) msg_queue_tail = buf; } else { msg_queue_tail = msg_queue_head = buf; - k_signal((Handler)port_dispatcher_sigh, 0); + tipc_k_signal((Handler)port_dispatcher_sigh, 0); } spin_unlock_bh(&queue_lock); return TIPC_OK; @@ -994,14 +994,14 @@ static void port_wakeup_sh(unsigned long ref) tipc_continue_event cb = 0; void *uh = 0; - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (p_ptr) { up_ptr = p_ptr->user_port; if (up_ptr) { cb = up_ptr->continue_event_cb; uh = up_ptr->usr_handle; } - port_unlock(p_ptr); + tipc_port_unlock(p_ptr); } if (cb) cb(uh, ref); @@ -1010,7 +1010,7 @@ static void port_wakeup_sh(unsigned long ref) static void port_wakeup(struct tipc_port *p_ptr) { - k_signal((Handler)port_wakeup_sh, p_ptr->ref); + tipc_k_signal((Handler)port_wakeup_sh, p_ptr->ref); } void tipc_acknowledge(u32 ref, u32 ack) @@ -1018,7 +1018,7 @@ void tipc_acknowledge(u32 ref, u32 ack) struct port *p_ptr; struct sk_buff *buf = 0; - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) return; if (p_ptr->publ.connected) { @@ -1033,8 +1033,8 @@ void tipc_acknowledge(u32 ref, u32 ack) port_out_seqno(p_ptr), ack); } - port_unlock(p_ptr); - net_route_msg(buf); + tipc_port_unlock(p_ptr); + tipc_net_route_msg(buf); } /* @@ -1063,7 +1063,7 @@ int tipc_createport(u32 user_ref, return -ENOMEM; } ref = tipc_createport_raw(0, port_dispatcher, port_wakeup, importance); - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) { kfree(up_ptr); return -ENOMEM; @@ -1081,10 +1081,10 @@ int tipc_createport(u32 user_ref, up_ptr->conn_msg_cb = conn_msg_cb; up_ptr->continue_event_cb = continue_event_cb; INIT_LIST_HEAD(&up_ptr->uport_list); - reg_add_port(up_ptr); + tipc_reg_add_port(up_ptr); *portref = p_ptr->publ.ref; dbg(" tipc_createport: %x with ref %u\n", p_ptr, p_ptr->publ.ref); - port_unlock(p_ptr); + tipc_port_unlock(p_ptr); return TIPC_OK; } @@ -1099,7 +1099,7 @@ int tipc_portimportance(u32 ref, unsigned int *importance) { struct port *p_ptr; - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; *importance = (unsigned int)msg_importance(&p_ptr->publ.phdr); @@ -1114,7 +1114,7 @@ int tipc_set_portimportance(u32 ref, unsigned int imp) if (imp > TIPC_CRITICAL_IMPORTANCE) return -EINVAL; - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; msg_set_importance(&p_ptr->publ.phdr, (u32)imp); @@ -1130,7 +1130,7 @@ int tipc_publish(u32 ref, unsigned int scope, struct tipc_name_seq const *seq) u32 key; int res = -EINVAL; - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); dbg("tipc_publ %u, p_ptr = %x, conn = %x, scope = %x, " "lower = %u, upper = %u\n", ref, p_ptr, p_ptr->publ.connected, scope, seq->lower, seq->upper); @@ -1147,8 +1147,8 @@ int tipc_publish(u32 ref, unsigned int scope, struct tipc_name_seq const *seq) res = -EADDRINUSE; goto exit; } - publ = nametbl_publish(seq->type, seq->lower, seq->upper, - scope, p_ptr->publ.ref, key); + publ = tipc_nametbl_publish(seq->type, seq->lower, seq->upper, + scope, p_ptr->publ.ref, key); if (publ) { list_add(&publ->pport_list, &p_ptr->publications); p_ptr->pub_count++; @@ -1156,7 +1156,7 @@ int tipc_publish(u32 ref, unsigned int scope, struct tipc_name_seq const *seq) res = TIPC_OK; } exit: - port_unlock(p_ptr); + tipc_port_unlock(p_ptr); return res; } @@ -1167,7 +1167,7 @@ int tipc_withdraw(u32 ref, unsigned int scope, struct tipc_name_seq const *seq) struct publication *tpubl; int res = -EINVAL; - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; if (!p_ptr->publ.published) @@ -1175,8 +1175,8 @@ int tipc_withdraw(u32 ref, unsigned int scope, struct tipc_name_seq const *seq) if (!seq) { list_for_each_entry_safe(publ, tpubl, &p_ptr->publications, pport_list) { - nametbl_withdraw(publ->type, publ->lower, - publ->ref, publ->key); + tipc_nametbl_withdraw(publ->type, publ->lower, + publ->ref, publ->key); } res = TIPC_OK; } else { @@ -1190,8 +1190,8 @@ int tipc_withdraw(u32 ref, unsigned int scope, struct tipc_name_seq const *seq) continue; if (publ->upper != seq->upper) break; - nametbl_withdraw(publ->type, publ->lower, - publ->ref, publ->key); + tipc_nametbl_withdraw(publ->type, publ->lower, + publ->ref, publ->key); res = TIPC_OK; break; } @@ -1199,7 +1199,7 @@ int tipc_withdraw(u32 ref, unsigned int scope, struct tipc_name_seq const *seq) if (list_empty(&p_ptr->publications)) p_ptr->publ.published = 0; exit: - port_unlock(p_ptr); + tipc_port_unlock(p_ptr); return res; } @@ -1209,7 +1209,7 @@ int tipc_connect2port(u32 ref, struct tipc_portid const *peer) struct tipc_msg *msg; int res = -EINVAL; - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; if (p_ptr->publ.published || p_ptr->publ.connected) @@ -1234,13 +1234,13 @@ int tipc_connect2port(u32 ref, struct tipc_portid const *peer) p_ptr->publ.connected = 1; k_start_timer(&p_ptr->timer, p_ptr->probing_interval); - nodesub_subscribe(&p_ptr->subscription,peer->node, + tipc_nodesub_subscribe(&p_ptr->subscription,peer->node, (void *)(unsigned long)ref, (net_ev_handler)port_handle_node_down); res = TIPC_OK; exit: - port_unlock(p_ptr); - p_ptr->max_pkt = link_get_max_pkt(peer->node, ref); + tipc_port_unlock(p_ptr); + p_ptr->max_pkt = tipc_link_get_max_pkt(peer->node, ref); return res; } @@ -1254,16 +1254,16 @@ int tipc_disconnect(u32 ref) struct port *p_ptr; int res = -ENOTCONN; - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; if (p_ptr->publ.connected) { p_ptr->publ.connected = 0; /* let timer expire on it's own to avoid deadlock! */ - nodesub_unsubscribe(&p_ptr->subscription); + tipc_nodesub_unsubscribe(&p_ptr->subscription); res = TIPC_OK; } - port_unlock(p_ptr); + tipc_port_unlock(p_ptr); return res; } @@ -1275,7 +1275,7 @@ int tipc_shutdown(u32 ref) struct port *p_ptr; struct sk_buff *buf = 0; - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; @@ -1293,8 +1293,8 @@ int tipc_shutdown(u32 ref) port_out_seqno(p_ptr), 0); } - port_unlock(p_ptr); - net_route_msg(buf); + tipc_port_unlock(p_ptr); + tipc_net_route_msg(buf); return tipc_disconnect(ref); } @@ -1302,11 +1302,11 @@ int tipc_isconnected(u32 ref, int *isconnected) { struct port *p_ptr; - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; *isconnected = p_ptr->publ.connected; - port_unlock(p_ptr); + tipc_port_unlock(p_ptr); return TIPC_OK; } @@ -1315,7 +1315,7 @@ int tipc_peer(u32 ref, struct tipc_portid *peer) struct port *p_ptr; int res; - p_ptr = port_lock(ref); + p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; if (p_ptr->publ.connected) { @@ -1324,23 +1324,23 @@ int tipc_peer(u32 ref, struct tipc_portid *peer) res = TIPC_OK; } else res = -ENOTCONN; - port_unlock(p_ptr); + tipc_port_unlock(p_ptr); return res; } int tipc_ref_valid(u32 ref) { /* Works irrespective of type */ - return !!ref_deref(ref); + return !!tipc_ref_deref(ref); } /* - * port_recv_sections(): Concatenate and deliver sectioned + * tipc_port_recv_sections(): Concatenate and deliver sectioned * message for this node. */ -int port_recv_sections(struct port *sender, unsigned int num_sect, +int tipc_port_recv_sections(struct port *sender, unsigned int num_sect, struct iovec const *msg_sect) { struct sk_buff *buf; @@ -1349,7 +1349,7 @@ int port_recv_sections(struct port *sender, unsigned int num_sect, res = msg_build(&sender->publ.phdr, msg_sect, num_sect, MAX_MSG_SIZE, !sender->user_port, &buf); if (likely(buf)) - port_recv_msg(buf); + tipc_port_recv_msg(buf); return res; } @@ -1363,18 +1363,18 @@ int tipc_send(u32 ref, unsigned int num_sect, struct iovec const *msg_sect) u32 destnode; int res; - p_ptr = port_deref(ref); + p_ptr = tipc_port_deref(ref); if (!p_ptr || !p_ptr->publ.connected) return -EINVAL; p_ptr->publ.congested = 1; - if (!port_congested(p_ptr)) { + if (!tipc_port_congested(p_ptr)) { destnode = port_peernode(p_ptr); if (likely(destnode != tipc_own_addr)) - res = link_send_sections_fast(p_ptr, msg_sect, num_sect, - destnode); + res = tipc_link_send_sections_fast(p_ptr, msg_sect, num_sect, + destnode); else - res = port_recv_sections(p_ptr, num_sect, msg_sect); + res = tipc_port_recv_sections(p_ptr, num_sect, msg_sect); if (likely(res != -ELINKCONG)) { port_incr_out_seqno(p_ptr); @@ -1404,7 +1404,7 @@ int tipc_send_buf(u32 ref, struct sk_buff *buf, unsigned int dsz) u32 sz; u32 res; - p_ptr = port_deref(ref); + p_ptr = tipc_port_deref(ref); if (!p_ptr || !p_ptr->publ.connected) return -EINVAL; @@ -1419,11 +1419,11 @@ int tipc_send_buf(u32 ref, struct sk_buff *buf, unsigned int dsz) memcpy(buf->data, (unchar *)msg, hsz); destnode = msg_destnode(msg); p_ptr->publ.congested = 1; - if (!port_congested(p_ptr)) { + if (!tipc_port_congested(p_ptr)) { if (likely(destnode != tipc_own_addr)) res = tipc_send_buf_fast(buf, destnode); else { - port_recv_msg(buf); + tipc_port_recv_msg(buf); res = sz; } if (likely(res != -ELINKCONG)) { @@ -1458,7 +1458,7 @@ int tipc_forward2name(u32 ref, u32 destport = 0; int res; - p_ptr = port_deref(ref); + p_ptr = tipc_port_deref(ref); if (!p_ptr || p_ptr->publ.connected) return -EINVAL; @@ -1472,16 +1472,16 @@ int tipc_forward2name(u32 ref, msg_set_lookup_scope(msg, addr_scope(domain)); if (importance <= TIPC_CRITICAL_IMPORTANCE) msg_set_importance(msg,importance); - destport = nametbl_translate(name->type, name->instance, &destnode); + destport = tipc_nametbl_translate(name->type, name->instance, &destnode); msg_set_destnode(msg, destnode); msg_set_destport(msg, destport); if (likely(destport || destnode)) { p_ptr->sent++; if (likely(destnode == tipc_own_addr)) - return port_recv_sections(p_ptr, num_sect, msg_sect); - res = link_send_sections_fast(p_ptr, msg_sect, num_sect, - destnode); + return tipc_port_recv_sections(p_ptr, num_sect, msg_sect); + res = tipc_link_send_sections_fast(p_ptr, msg_sect, num_sect, + destnode); if (likely(res != -ELINKCONG)) return res; if (port_unreliable(p_ptr)) { @@ -1490,8 +1490,8 @@ int tipc_forward2name(u32 ref, } return -ELINKCONG; } - return port_reject_sections(p_ptr, msg, msg_sect, num_sect, - TIPC_ERR_NO_NAME); + return tipc_port_reject_sections(p_ptr, msg, msg_sect, num_sect, + TIPC_ERR_NO_NAME); } /** @@ -1530,7 +1530,7 @@ int tipc_forward_buf2name(u32 ref, u32 destport = 0; int res; - p_ptr = (struct port *)ref_deref(ref); + p_ptr = (struct port *)tipc_ref_deref(ref); if (!p_ptr || p_ptr->publ.connected) return -EINVAL; @@ -1545,7 +1545,7 @@ int tipc_forward_buf2name(u32 ref, msg_set_lookup_scope(msg, addr_scope(domain)); msg_set_hdr_sz(msg, LONG_H_SIZE); msg_set_size(msg, LONG_H_SIZE + dsz); - destport = nametbl_translate(name->type, name->instance, &destnode); + destport = tipc_nametbl_translate(name->type, name->instance, &destnode); msg_set_destnode(msg, destnode); msg_set_destport(msg, destport); msg_dbg(msg, "forw2name ==> "); @@ -1557,7 +1557,7 @@ int tipc_forward_buf2name(u32 ref, if (likely(destport || destnode)) { p_ptr->sent++; if (destnode == tipc_own_addr) - return port_recv_msg(buf); + return tipc_port_recv_msg(buf); res = tipc_send_buf_fast(buf, destnode); if (likely(res != -ELINKCONG)) return res; @@ -1601,7 +1601,7 @@ int tipc_forward2port(u32 ref, struct tipc_msg *msg; int res; - p_ptr = port_deref(ref); + p_ptr = tipc_port_deref(ref); if (!p_ptr || p_ptr->publ.connected) return -EINVAL; @@ -1616,8 +1616,8 @@ int tipc_forward2port(u32 ref, msg_set_importance(msg, importance); p_ptr->sent++; if (dest->node == tipc_own_addr) - return port_recv_sections(p_ptr, num_sect, msg_sect); - res = link_send_sections_fast(p_ptr, msg_sect, num_sect, dest->node); + return tipc_port_recv_sections(p_ptr, num_sect, msg_sect); + res = tipc_link_send_sections_fast(p_ptr, msg_sect, num_sect, dest->node); if (likely(res != -ELINKCONG)) return res; if (port_unreliable(p_ptr)) { @@ -1658,7 +1658,7 @@ int tipc_forward_buf2port(u32 ref, struct tipc_msg *msg; int res; - p_ptr = (struct port *)ref_deref(ref); + p_ptr = (struct port *)tipc_ref_deref(ref); if (!p_ptr || p_ptr->publ.connected) return -EINVAL; @@ -1680,7 +1680,7 @@ int tipc_forward_buf2port(u32 ref, msg_dbg(msg, "buf2port: "); p_ptr->sent++; if (dest->node == tipc_own_addr) - return port_recv_msg(buf); + return tipc_port_recv_msg(buf); res = tipc_send_buf_fast(buf, dest->node); if (likely(res != -ELINKCONG)) return res; diff --git a/net/tipc/port.h b/net/tipc/port.h index f4a8c2be3fa..839f100da64 100644 --- a/net/tipc/port.h +++ b/net/tipc/port.h @@ -110,65 +110,65 @@ struct port { struct node_subscr subscription; }; -extern spinlock_t port_list_lock; +extern spinlock_t tipc_port_list_lock; struct port_list; -int port_recv_sections(struct port *p_ptr, u32 num_sect, - struct iovec const *msg_sect); -int port_reject_sections(struct port *p_ptr, struct tipc_msg *hdr, - struct iovec const *msg_sect, u32 num_sect, - int err); -struct sk_buff *port_get_ports(void); +int tipc_port_recv_sections(struct port *p_ptr, u32 num_sect, + struct iovec const *msg_sect); +int tipc_port_reject_sections(struct port *p_ptr, struct tipc_msg *hdr, + struct iovec const *msg_sect, u32 num_sect, + int err); +struct sk_buff *tipc_port_get_ports(void); struct sk_buff *port_show_stats(const void *req_tlv_area, int req_tlv_space); -void port_recv_proto_msg(struct sk_buff *buf); -void port_recv_mcast(struct sk_buff *buf, struct port_list *dp); -void port_reinit(void); +void tipc_port_recv_proto_msg(struct sk_buff *buf); +void tipc_port_recv_mcast(struct sk_buff *buf, struct port_list *dp); +void tipc_port_reinit(void); /** - * port_lock - lock port instance referred to and return its pointer + * tipc_port_lock - lock port instance referred to and return its pointer */ -static inline struct port *port_lock(u32 ref) +static inline struct port *tipc_port_lock(u32 ref) { - return (struct port *)ref_lock(ref); + return (struct port *)tipc_ref_lock(ref); } /** - * port_unlock - unlock a port instance + * tipc_port_unlock - unlock a port instance * - * Can use pointer instead of ref_unlock() since port is already locked. + * Can use pointer instead of tipc_ref_unlock() since port is already locked. */ -static inline void port_unlock(struct port *p_ptr) +static inline void tipc_port_unlock(struct port *p_ptr) { spin_unlock_bh(p_ptr->publ.lock); } -static inline struct port* port_deref(u32 ref) +static inline struct port* tipc_port_deref(u32 ref) { - return (struct port *)ref_deref(ref); + return (struct port *)tipc_ref_deref(ref); } -static inline u32 peer_port(struct port *p_ptr) +static inline u32 tipc_peer_port(struct port *p_ptr) { return msg_destport(&p_ptr->publ.phdr); } -static inline u32 peer_node(struct port *p_ptr) +static inline u32 tipc_peer_node(struct port *p_ptr) { return msg_destnode(&p_ptr->publ.phdr); } -static inline int port_congested(struct port *p_ptr) +static inline int tipc_port_congested(struct port *p_ptr) { return((p_ptr->sent - p_ptr->acked) >= (TIPC_FLOW_CONTROL_WIN * 2)); } /** - * port_recv_msg - receive message from lower layer and deliver to port user + * tipc_port_recv_msg - receive message from lower layer and deliver to port user */ -static inline int port_recv_msg(struct sk_buff *buf) +static inline int tipc_port_recv_msg(struct sk_buff *buf) { struct port *p_ptr; struct tipc_msg *msg = buf_msg(buf); @@ -178,24 +178,24 @@ static inline int port_recv_msg(struct sk_buff *buf) /* forward unresolved named message */ if (unlikely(!destport)) { - net_route_msg(buf); + tipc_net_route_msg(buf); return dsz; } /* validate destination & pass to port, otherwise reject message */ - p_ptr = port_lock(destport); + p_ptr = tipc_port_lock(destport); if (likely(p_ptr)) { if (likely(p_ptr->publ.connected)) { - if ((unlikely(msg_origport(msg) != peer_port(p_ptr))) || - (unlikely(msg_orignode(msg) != peer_node(p_ptr))) || + if ((unlikely(msg_origport(msg) != tipc_peer_port(p_ptr))) || + (unlikely(msg_orignode(msg) != tipc_peer_node(p_ptr))) || (unlikely(!msg_connected(msg)))) { err = TIPC_ERR_NO_PORT; - port_unlock(p_ptr); + tipc_port_unlock(p_ptr); goto reject; } } err = p_ptr->dispatcher(&p_ptr->publ, buf); - port_unlock(p_ptr); + tipc_port_unlock(p_ptr); if (likely(!err)) return dsz; } else { diff --git a/net/tipc/ref.c b/net/tipc/ref.c index 944093fe246..5a13c2defe4 100644 --- a/net/tipc/ref.c +++ b/net/tipc/ref.c @@ -61,15 +61,15 @@ * because entry 0's reference field has the form XXXX|1--1. */ -struct ref_table ref_table = { 0 }; +struct ref_table tipc_ref_table = { 0 }; -rwlock_t reftbl_lock = RW_LOCK_UNLOCKED; +static rwlock_t ref_table_lock = RW_LOCK_UNLOCKED; /** - * ref_table_init - create reference table for objects + * tipc_ref_table_init - create reference table for objects */ -int ref_table_init(u32 requested_size, u32 start) +int tipc_ref_table_init(u32 requested_size, u32 start) { struct reference *table; u32 sz = 1 << 4; @@ -83,43 +83,43 @@ int ref_table_init(u32 requested_size, u32 start) if (table == NULL) return -ENOMEM; - write_lock_bh(&reftbl_lock); + write_lock_bh(&ref_table_lock); index_mask = sz - 1; for (i = sz - 1; i >= 0; i--) { table[i].object = 0; table[i].lock = SPIN_LOCK_UNLOCKED; table[i].data.next_plus_upper = (start & ~index_mask) + i - 1; } - ref_table.entries = table; - ref_table.index_mask = index_mask; - ref_table.first_free = sz - 1; - ref_table.last_free = 1; - write_unlock_bh(&reftbl_lock); + tipc_ref_table.entries = table; + tipc_ref_table.index_mask = index_mask; + tipc_ref_table.first_free = sz - 1; + tipc_ref_table.last_free = 1; + write_unlock_bh(&ref_table_lock); return TIPC_OK; } /** - * ref_table_stop - destroy reference table for objects + * tipc_ref_table_stop - destroy reference table for objects */ -void ref_table_stop(void) +void tipc_ref_table_stop(void) { - if (!ref_table.entries) + if (!tipc_ref_table.entries) return; - vfree(ref_table.entries); - ref_table.entries = 0; + vfree(tipc_ref_table.entries); + tipc_ref_table.entries = 0; } /** - * ref_acquire - create reference to an object + * tipc_ref_acquire - create reference to an object * * Return a unique reference value which can be translated back to the pointer * 'object' at a later time. Also, pass back a pointer to the lock protecting * the object, but without locking it. */ -u32 ref_acquire(void *object, spinlock_t **lock) +u32 tipc_ref_acquire(void *object, spinlock_t **lock) { struct reference *entry; u32 index; @@ -127,17 +127,17 @@ u32 ref_acquire(void *object, spinlock_t **lock) u32 next_plus_upper; u32 reference = 0; - assert(ref_table.entries && object); + assert(tipc_ref_table.entries && object); - write_lock_bh(&reftbl_lock); - if (ref_table.first_free) { - index = ref_table.first_free; - entry = &(ref_table.entries[index]); - index_mask = ref_table.index_mask; + write_lock_bh(&ref_table_lock); + if (tipc_ref_table.first_free) { + index = tipc_ref_table.first_free; + entry = &(tipc_ref_table.entries[index]); + index_mask = tipc_ref_table.index_mask; /* take lock in case a previous user of entry still holds it */ spin_lock_bh(&entry->lock); next_plus_upper = entry->data.next_plus_upper; - ref_table.first_free = next_plus_upper & index_mask; + tipc_ref_table.first_free = next_plus_upper & index_mask; reference = (next_plus_upper & ~index_mask) + index; entry->data.reference = reference; entry->object = object; @@ -145,45 +145,45 @@ u32 ref_acquire(void *object, spinlock_t **lock) *lock = &entry->lock; spin_unlock_bh(&entry->lock); } - write_unlock_bh(&reftbl_lock); + write_unlock_bh(&ref_table_lock); return reference; } /** - * ref_discard - invalidate references to an object + * tipc_ref_discard - invalidate references to an object * * Disallow future references to an object and free up the entry for re-use. * Note: The entry's spin_lock may still be busy after discard */ -void ref_discard(u32 ref) +void tipc_ref_discard(u32 ref) { struct reference *entry; u32 index; u32 index_mask; - assert(ref_table.entries); + assert(tipc_ref_table.entries); assert(ref != 0); - write_lock_bh(&reftbl_lock); - index_mask = ref_table.index_mask; + write_lock_bh(&ref_table_lock); + index_mask = tipc_ref_table.index_mask; index = ref & index_mask; - entry = &(ref_table.entries[index]); + entry = &(tipc_ref_table.entries[index]); assert(entry->object != 0); assert(entry->data.reference == ref); /* mark entry as unused */ entry->object = 0; - if (ref_table.first_free == 0) - ref_table.first_free = index; + if (tipc_ref_table.first_free == 0) + tipc_ref_table.first_free = index; else /* next_plus_upper is always XXXX|0--0 for last free entry */ - ref_table.entries[ref_table.last_free].data.next_plus_upper + tipc_ref_table.entries[tipc_ref_table.last_free].data.next_plus_upper |= index; - ref_table.last_free = index; + tipc_ref_table.last_free = index; /* increment upper bits of entry to invalidate subsequent references */ entry->data.next_plus_upper = (ref & ~index_mask) + (index_mask + 1); - write_unlock_bh(&reftbl_lock); + write_unlock_bh(&ref_table_lock); } diff --git a/net/tipc/ref.h b/net/tipc/ref.h index 429cde57228..4f8f9f40dca 100644 --- a/net/tipc/ref.h +++ b/net/tipc/ref.h @@ -54,7 +54,7 @@ struct reference { }; /** - * struct ref_table - table of TIPC object reference entries + * struct tipc_ref_table - table of TIPC object reference entries * @entries: pointer to array of reference entries * @index_mask: bitmask for array index portion of reference values * @first_free: array index of first unused object reference entry @@ -68,24 +68,24 @@ struct ref_table { u32 last_free; }; -extern struct ref_table ref_table; +extern struct ref_table tipc_ref_table; -int ref_table_init(u32 requested_size, u32 start); -void ref_table_stop(void); +int tipc_ref_table_init(u32 requested_size, u32 start); +void tipc_ref_table_stop(void); -u32 ref_acquire(void *object, spinlock_t **lock); -void ref_discard(u32 ref); +u32 tipc_ref_acquire(void *object, spinlock_t **lock); +void tipc_ref_discard(u32 ref); /** - * ref_lock - lock referenced object and return pointer to it + * tipc_ref_lock - lock referenced object and return pointer to it */ -static inline void *ref_lock(u32 ref) +static inline void *tipc_ref_lock(u32 ref) { - if (likely(ref_table.entries)) { + if (likely(tipc_ref_table.entries)) { struct reference *r = - &ref_table.entries[ref & ref_table.index_mask]; + &tipc_ref_table.entries[ref & tipc_ref_table.index_mask]; spin_lock_bh(&r->lock); if (likely(r->data.reference == ref)) @@ -96,31 +96,31 @@ static inline void *ref_lock(u32 ref) } /** - * ref_unlock - unlock referenced object + * tipc_ref_unlock - unlock referenced object */ -static inline void ref_unlock(u32 ref) +static inline void tipc_ref_unlock(u32 ref) { - if (likely(ref_table.entries)) { + if (likely(tipc_ref_table.entries)) { struct reference *r = - &ref_table.entries[ref & ref_table.index_mask]; + &tipc_ref_table.entries[ref & tipc_ref_table.index_mask]; if (likely(r->data.reference == ref)) spin_unlock_bh(&r->lock); else - err("ref_unlock() invoked using obsolete reference\n"); + err("tipc_ref_unlock() invoked using obsolete reference\n"); } } /** - * ref_deref - return pointer referenced object (without locking it) + * tipc_ref_deref - return pointer referenced object (without locking it) */ -static inline void *ref_deref(u32 ref) +static inline void *tipc_ref_deref(u32 ref) { - if (likely(ref_table.entries)) { + if (likely(tipc_ref_table.entries)) { struct reference *r = - &ref_table.entries[ref & ref_table.index_mask]; + &tipc_ref_table.entries[ref & tipc_ref_table.index_mask]; if (likely(r->data.reference == ref)) return r->object; diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 107f654db0b..67253bfcd70 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -1183,7 +1183,7 @@ static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf) if (unlikely(msg_errcode(msg) && (sock->state == SS_CONNECTED))) { sock->state = SS_DISCONNECTING; /* Note: Use signal since port lock is already taken! */ - k_signal((Handler)async_disconnect, tport->ref); + tipc_k_signal((Handler)async_disconnect, tport->ref); } /* Enqueue message (finally!) */ @@ -1683,11 +1683,11 @@ static struct proto tipc_proto = { }; /** - * socket_init - initialize TIPC socket interface + * tipc_socket_init - initialize TIPC socket interface * * Returns 0 on success, errno otherwise */ -int socket_init(void) +int tipc_socket_init(void) { int res; @@ -1710,9 +1710,9 @@ int socket_init(void) } /** - * sock_stop - stop TIPC socket interface + * tipc_socket_stop - stop TIPC socket interface */ -void socket_stop(void) +void tipc_socket_stop(void) { if (!sockets_enabled) return; diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c index 80e219ba527..5ff38b9f319 100644 --- a/net/tipc/subscr.c +++ b/net/tipc/subscr.c @@ -118,14 +118,14 @@ static void subscr_send_event(struct subscription *sub, } /** - * subscr_overlap - test for subscription overlap with the given values + * tipc_subscr_overlap - test for subscription overlap with the given values * * Returns 1 if there is overlap, otherwise 0. */ -int subscr_overlap(struct subscription *sub, - u32 found_lower, - u32 found_upper) +int tipc_subscr_overlap(struct subscription *sub, + u32 found_lower, + u32 found_upper) { if (found_lower < sub->seq.lower) @@ -138,22 +138,22 @@ int subscr_overlap(struct subscription *sub, } /** - * subscr_report_overlap - issue event if there is subscription overlap + * tipc_subscr_report_overlap - issue event if there is subscription overlap * * Protected by nameseq.lock in name_table.c */ -void subscr_report_overlap(struct subscription *sub, - u32 found_lower, - u32 found_upper, - u32 event, - u32 port_ref, - u32 node, - int must) +void tipc_subscr_report_overlap(struct subscription *sub, + u32 found_lower, + u32 found_upper, + u32 event, + u32 port_ref, + u32 node, + int must) { dbg("Rep overlap %u:%u,%u<->%u,%u\n", sub->seq.type, sub->seq.lower, sub->seq.upper, found_lower, found_upper); - if (!subscr_overlap(sub, found_lower, found_upper)) + if (!tipc_subscr_overlap(sub, found_lower, found_upper)) return; if (!must && (sub->filter != TIPC_SUB_PORTS)) return; @@ -172,13 +172,13 @@ static void subscr_timeout(struct subscription *sub) /* Validate subscriber reference (in case subscriber is terminating) */ subscriber_ref = sub->owner->ref; - subscriber = (struct subscriber *)ref_lock(subscriber_ref); + subscriber = (struct subscriber *)tipc_ref_lock(subscriber_ref); if (subscriber == NULL) return; /* Unlink subscription from name table */ - nametbl_unsubscribe(sub); + tipc_nametbl_unsubscribe(sub); /* Notify subscriber of timeout, then unlink subscription */ @@ -192,7 +192,7 @@ static void subscr_timeout(struct subscription *sub) /* Now destroy subscription */ - ref_unlock(subscriber_ref); + tipc_ref_unlock(subscriber_ref); k_term_timer(&sub->timer); kfree(sub); atomic_dec(&topsrv.subscription_count); @@ -216,7 +216,7 @@ static void subscr_terminate(struct subscriber *subscriber) /* Invalidate subscriber reference */ - ref_discard(subscriber->ref); + tipc_ref_discard(subscriber->ref); spin_unlock_bh(subscriber->lock); /* Destroy any existing subscriptions for subscriber */ @@ -227,7 +227,7 @@ static void subscr_terminate(struct subscriber *subscriber) k_cancel_timer(&sub->timer); k_term_timer(&sub->timer); } - nametbl_unsubscribe(sub); + tipc_nametbl_unsubscribe(sub); list_del(&sub->subscription_list); dbg("Term: Removed sub %u,%u,%u from subscriber %x list\n", sub->seq.type, sub->seq.lower, sub->seq.upper, subscriber); @@ -315,7 +315,7 @@ static void subscr_subscribe(struct tipc_subscr *s, k_start_timer(&sub->timer, sub->timeout); } sub->owner = subscriber; - nametbl_subscribe(sub); + tipc_nametbl_subscribe(sub); } /** @@ -332,7 +332,7 @@ static void subscr_conn_shutdown_event(void *usr_handle, struct subscriber *subscriber; spinlock_t *subscriber_lock; - subscriber = ref_lock((u32)(unsigned long)usr_handle); + subscriber = tipc_ref_lock((u32)(unsigned long)usr_handle); if (subscriber == NULL) return; @@ -354,7 +354,7 @@ static void subscr_conn_msg_event(void *usr_handle, struct subscriber *subscriber; spinlock_t *subscriber_lock; - subscriber = ref_lock((u32)(unsigned long)usr_handle); + subscriber = tipc_ref_lock((u32)(unsigned long)usr_handle); if (subscriber == NULL) return; @@ -401,7 +401,7 @@ static void subscr_named_msg_event(void *usr_handle, memset(subscriber, 0, sizeof(struct subscriber)); INIT_LIST_HEAD(&subscriber->subscription_list); INIT_LIST_HEAD(&subscriber->subscriber_list); - subscriber->ref = ref_acquire(subscriber, &subscriber->lock); + subscriber->ref = tipc_ref_acquire(subscriber, &subscriber->lock); if (subscriber->ref == 0) { warn("Failed to acquire subscriber reference\n"); kfree(subscriber); @@ -423,7 +423,7 @@ static void subscr_named_msg_event(void *usr_handle, &subscriber->port_ref); if (subscriber->port_ref == 0) { warn("Memory squeeze; failed to create subscription port\n"); - ref_discard(subscriber->ref); + tipc_ref_discard(subscriber->ref); kfree(subscriber); return; } @@ -432,7 +432,7 @@ static void subscr_named_msg_event(void *usr_handle, /* Add subscriber to topology server's subscriber list */ - ref_lock(subscriber->ref); + tipc_ref_lock(subscriber->ref); spin_lock_bh(&topsrv.lock); list_add(&subscriber->subscriber_list, &topsrv.subscriber_list); spin_unlock_bh(&topsrv.lock); @@ -451,7 +451,7 @@ static void subscr_named_msg_event(void *usr_handle, spin_unlock_bh(subscriber_lock); } -int subscr_start(void) +int tipc_subscr_start(void) { struct tipc_name_seq seq = {TIPC_TOP_SRV, TIPC_TOP_SRV, TIPC_TOP_SRV}; int res = -1; @@ -481,7 +481,7 @@ int subscr_start(void) if (res) goto failed; - res = nametbl_publish_rsv(topsrv.setup_port, TIPC_NODE_SCOPE, &seq); + res = tipc_nametbl_publish_rsv(topsrv.setup_port, TIPC_NODE_SCOPE, &seq); if (res) goto failed; @@ -496,7 +496,7 @@ failed: return res; } -void subscr_stop(void) +void tipc_subscr_stop(void) { struct subscriber *subscriber; struct subscriber *subscriber_temp; @@ -507,7 +507,7 @@ void subscr_stop(void) list_for_each_entry_safe(subscriber, subscriber_temp, &topsrv.subscriber_list, subscriber_list) { - ref_lock(subscriber->ref); + tipc_ref_lock(subscriber->ref); subscriber_lock = subscriber->lock; subscr_terminate(subscriber); spin_unlock_bh(subscriber_lock); @@ -522,6 +522,6 @@ int tipc_ispublished(struct tipc_name const *name) { u32 domain = 0; - return(nametbl_translate(name->type, name->instance,&domain) != 0); + return(tipc_nametbl_translate(name->type, name->instance,&domain) != 0); } diff --git a/net/tipc/subscr.h b/net/tipc/subscr.h index ccff4efcb75..1e5090465d2 100644 --- a/net/tipc/subscr.h +++ b/net/tipc/subscr.h @@ -60,21 +60,21 @@ struct subscription { struct subscriber *owner; }; -int subscr_overlap(struct subscription * sub, - u32 found_lower, - u32 found_upper); +int tipc_subscr_overlap(struct subscription * sub, + u32 found_lower, + u32 found_upper); -void subscr_report_overlap(struct subscription * sub, - u32 found_lower, - u32 found_upper, - u32 event, - u32 port_ref, - u32 node, - int must_report); +void tipc_subscr_report_overlap(struct subscription * sub, + u32 found_lower, + u32 found_upper, + u32 event, + u32 port_ref, + u32 node, + int must_report); -int subscr_start(void); +int tipc_subscr_start(void); -void subscr_stop(void); +void tipc_subscr_stop(void); #endif diff --git a/net/tipc/user_reg.c b/net/tipc/user_reg.c index 35ec7dc8211..106200d7658 100644 --- a/net/tipc/user_reg.c +++ b/net/tipc/user_reg.c @@ -114,10 +114,10 @@ static void reg_callback(struct tipc_user *user_ptr) } /** - * reg_start - activate TIPC user registry + * tipc_reg_start - activate TIPC user registry */ -int reg_start(void) +int tipc_reg_start(void) { u32 u; int res; @@ -127,17 +127,17 @@ int reg_start(void) for (u = 1; u <= MAX_USERID; u++) { if (users[u].callback) - k_signal((Handler)reg_callback, - (unsigned long)&users[u]); + tipc_k_signal((Handler)reg_callback, + (unsigned long)&users[u]); } return TIPC_OK; } /** - * reg_stop - shut down & delete TIPC user registry + * tipc_reg_stop - shut down & delete TIPC user registry */ -void reg_stop(void) +void tipc_reg_stop(void) { int id; @@ -184,7 +184,7 @@ int tipc_attach(u32 *userid, tipc_mode_event cb, void *usr_handle) atomic_inc(&tipc_user_count); if (cb && (tipc_mode != TIPC_NOT_RUNNING)) - k_signal((Handler)reg_callback, (unsigned long)user_ptr); + tipc_k_signal((Handler)reg_callback, (unsigned long)user_ptr); return TIPC_OK; } @@ -223,10 +223,10 @@ void tipc_detach(u32 userid) } /** - * reg_add_port - register a user's driver port + * tipc_reg_add_port - register a user's driver port */ -int reg_add_port(struct user_port *up_ptr) +int tipc_reg_add_port(struct user_port *up_ptr) { struct tipc_user *user_ptr; @@ -245,10 +245,10 @@ int reg_add_port(struct user_port *up_ptr) } /** - * reg_remove_port - deregister a user's driver port + * tipc_reg_remove_port - deregister a user's driver port */ -int reg_remove_port(struct user_port *up_ptr) +int tipc_reg_remove_port(struct user_port *up_ptr) { if (up_ptr->user_ref == 0) return TIPC_OK; diff --git a/net/tipc/user_reg.h b/net/tipc/user_reg.h index 122ca9be367..d0e88794ed1 100644 --- a/net/tipc/user_reg.h +++ b/net/tipc/user_reg.h @@ -39,10 +39,10 @@ #include "port.h" -int reg_start(void); -void reg_stop(void); +int tipc_reg_start(void); +void tipc_reg_stop(void); -int reg_add_port(struct user_port *up_ptr); -int reg_remove_port(struct user_port *up_ptr); +int tipc_reg_add_port(struct user_port *up_ptr); +int tipc_reg_remove_port(struct user_port *up_ptr); #endif diff --git a/net/tipc/zone.c b/net/tipc/zone.c index 4eaef662d56..7c11f7f83a2 100644 --- a/net/tipc/zone.c +++ b/net/tipc/zone.c @@ -42,12 +42,12 @@ #include "cluster.h" #include "node.h" -struct _zone *zone_create(u32 addr) +struct _zone *tipc_zone_create(u32 addr) { struct _zone *z_ptr = 0; u32 z_num; - if (!addr_domain_valid(addr)) + if (!tipc_addr_domain_valid(addr)) return 0; z_ptr = (struct _zone *)kmalloc(sizeof(*z_ptr), GFP_ATOMIC); @@ -55,24 +55,24 @@ struct _zone *zone_create(u32 addr) memset(z_ptr, 0, sizeof(*z_ptr)); z_num = tipc_zone(addr); z_ptr->addr = tipc_addr(z_num, 0, 0); - net.zones[z_num] = z_ptr; + tipc_net.zones[z_num] = z_ptr; } return z_ptr; } -void zone_delete(struct _zone *z_ptr) +void tipc_zone_delete(struct _zone *z_ptr) { u32 c_num; if (!z_ptr) return; for (c_num = 1; c_num <= tipc_max_clusters; c_num++) { - cluster_delete(z_ptr->clusters[c_num]); + tipc_cltr_delete(z_ptr->clusters[c_num]); } kfree(z_ptr); } -void zone_attach_cluster(struct _zone *z_ptr, struct cluster *c_ptr) +void tipc_zone_attach_cluster(struct _zone *z_ptr, struct cluster *c_ptr) { u32 c_num = tipc_cluster(c_ptr->addr); @@ -82,19 +82,19 @@ void zone_attach_cluster(struct _zone *z_ptr, struct cluster *c_ptr) z_ptr->clusters[c_num] = c_ptr; } -void zone_remove_as_router(struct _zone *z_ptr, u32 router) +void tipc_zone_remove_as_router(struct _zone *z_ptr, u32 router) { u32 c_num; for (c_num = 1; c_num <= tipc_max_clusters; c_num++) { if (z_ptr->clusters[c_num]) { - cluster_remove_as_router(z_ptr->clusters[c_num], - router); + tipc_cltr_remove_as_router(z_ptr->clusters[c_num], + router); } } } -void zone_send_external_routes(struct _zone *z_ptr, u32 dest) +void tipc_zone_send_external_routes(struct _zone *z_ptr, u32 dest) { u32 c_num; @@ -102,12 +102,12 @@ void zone_send_external_routes(struct _zone *z_ptr, u32 dest) if (z_ptr->clusters[c_num]) { if (in_own_cluster(z_ptr->addr)) continue; - cluster_send_ext_routes(z_ptr->clusters[c_num], dest); + tipc_cltr_send_ext_routes(z_ptr->clusters[c_num], dest); } } } -struct node *zone_select_remote_node(struct _zone *z_ptr, u32 addr, u32 ref) +struct node *tipc_zone_select_remote_node(struct _zone *z_ptr, u32 addr, u32 ref) { struct cluster *c_ptr; struct node *n_ptr; @@ -118,7 +118,7 @@ struct node *zone_select_remote_node(struct _zone *z_ptr, u32 addr, u32 ref) c_ptr = z_ptr->clusters[tipc_cluster(addr)]; if (!c_ptr) return 0; - n_ptr = cluster_select_node(c_ptr, ref); + n_ptr = tipc_cltr_select_node(c_ptr, ref); if (n_ptr) return n_ptr; @@ -127,14 +127,14 @@ struct node *zone_select_remote_node(struct _zone *z_ptr, u32 addr, u32 ref) c_ptr = z_ptr->clusters[c_num]; if (!c_ptr) return 0; - n_ptr = cluster_select_node(c_ptr, ref); + n_ptr = tipc_cltr_select_node(c_ptr, ref); if (n_ptr) return n_ptr; } return 0; } -u32 zone_select_router(struct _zone *z_ptr, u32 addr, u32 ref) +u32 tipc_zone_select_router(struct _zone *z_ptr, u32 addr, u32 ref) { struct cluster *c_ptr; u32 c_num; @@ -143,14 +143,14 @@ u32 zone_select_router(struct _zone *z_ptr, u32 addr, u32 ref) if (!z_ptr) return 0; c_ptr = z_ptr->clusters[tipc_cluster(addr)]; - router = c_ptr ? cluster_select_router(c_ptr, ref) : 0; + router = c_ptr ? tipc_cltr_select_router(c_ptr, ref) : 0; if (router) return router; /* Links to any other clusters within the zone? */ for (c_num = 1; c_num <= tipc_max_clusters; c_num++) { c_ptr = z_ptr->clusters[c_num]; - router = c_ptr ? cluster_select_router(c_ptr, ref) : 0; + router = c_ptr ? tipc_cltr_select_router(c_ptr, ref) : 0; if (router) return router; } @@ -158,12 +158,12 @@ u32 zone_select_router(struct _zone *z_ptr, u32 addr, u32 ref) } -u32 zone_next_node(u32 addr) +u32 tipc_zone_next_node(u32 addr) { - struct cluster *c_ptr = cluster_find(addr); + struct cluster *c_ptr = tipc_cltr_find(addr); if (c_ptr) - return cluster_next_node(c_ptr, addr); + return tipc_cltr_next_node(c_ptr, addr); return 0; } diff --git a/net/tipc/zone.h b/net/tipc/zone.h index 4326f78d829..267999c5a24 100644 --- a/net/tipc/zone.h +++ b/net/tipc/zone.h @@ -54,18 +54,18 @@ struct _zone { u32 links; }; -struct node *zone_select_remote_node(struct _zone *z_ptr, u32 addr, u32 ref); -u32 zone_select_router(struct _zone *z_ptr, u32 addr, u32 ref); -void zone_remove_as_router(struct _zone *z_ptr, u32 router); -void zone_send_external_routes(struct _zone *z_ptr, u32 dest); -struct _zone *zone_create(u32 addr); -void zone_delete(struct _zone *z_ptr); -void zone_attach_cluster(struct _zone *z_ptr, struct cluster *c_ptr); -u32 zone_next_node(u32 addr); +struct node *tipc_zone_select_remote_node(struct _zone *z_ptr, u32 addr, u32 ref); +u32 tipc_zone_select_router(struct _zone *z_ptr, u32 addr, u32 ref); +void tipc_zone_remove_as_router(struct _zone *z_ptr, u32 router); +void tipc_zone_send_external_routes(struct _zone *z_ptr, u32 dest); +struct _zone *tipc_zone_create(u32 addr); +void tipc_zone_delete(struct _zone *z_ptr); +void tipc_zone_attach_cluster(struct _zone *z_ptr, struct cluster *c_ptr); +u32 tipc_zone_next_node(u32 addr); -static inline struct _zone *zone_find(u32 addr) +static inline struct _zone *tipc_zone_find(u32 addr) { - return net.zones[tipc_zone(addr)]; + return tipc_net.zones[tipc_zone(addr)]; } #endif -- cgit v1.2.3 From d41f084a74de860fe879403fbbad13abdf7aea8e Mon Sep 17 00:00:00 2001 From: Steve French Date: Tue, 17 Jan 2006 19:16:53 -0800 Subject: [CIFS] Remove compiler warning Signed-off-by: Benoit Boissinot Signed-off-by: Steve French --- fs/cifs/cifssmb.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 7a4e936d726..217323b0c89 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -2542,7 +2542,7 @@ CIFSSMBGetCIFSACL(const int xid, struct cifsTconInfo *tcon, __u16 fid, if (rc) { cFYI(1, ("Send error in QuerySecDesc = %d", rc)); } else { /* decode response */ - struct sec_desc * psec_desc; + struct cifs_sid * psec_desc; __le32 * parm; int parm_len; int data_len; -- cgit v1.2.3 From dbd2915ce87e811165da0717f8e159276ebb803e Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 17 Jan 2006 21:58:01 -0800 Subject: [IPV4]: RT_CACHE_STAT_INC() warning fix BUG: using smp_processor_id() in preemptible [00000001] code: rpc.statd/2408 And it _is_ a bug, but I guess we don't care enough to add preempt_disable(). Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- net/ipv4/route.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index f2e82afc15b..d82c242ea70 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -241,7 +241,8 @@ static int rt_hash_log; static unsigned int rt_hash_rnd; static DEFINE_PER_CPU(struct rt_cache_stat, rt_cache_stat); -#define RT_CACHE_STAT_INC(field) (__get_cpu_var(rt_cache_stat).field++) +#define RT_CACHE_STAT_INC(field) \ + (per_cpu(rt_cache_stat, raw_smp_processor_id()).field++) static int rt_intern_hash(unsigned hash, struct rtable *rth, struct rtable **res); -- cgit v1.2.3 From bc965a7f43d3fc37584f9c16c6d7a84a911a9d9c Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 18 Jan 2006 09:54:29 +0000 Subject: [SERIAL] Fix serial8250 driver initialisation ordering Commit 7493a314cb83797ce612a577475aacaedc553fed changed the ordering of the registration of the platform device driver vs the 8250 drivers internal initialisation. This led to the probe function being called before the driver had finished its internal initialisation, causing mayhem. Revert the ordering change. Signed-off-by: Russell King --- drivers/serial/8250.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c index d9ce8c54941..bc36edff205 100644 --- a/drivers/serial/8250.c +++ b/drivers/serial/8250.c @@ -2595,15 +2595,11 @@ static int __init serial8250_init(void) if (ret) goto out; - ret = platform_driver_register(&serial8250_isa_driver); - if (ret) - goto unreg_uart_drv; - serial8250_isa_devs = platform_device_alloc("serial8250", PLAT8250_DEV_LEGACY); if (!serial8250_isa_devs) { ret = -ENOMEM; - goto unreg_plat_drv; + goto unreg_uart_drv; } ret = platform_device_add(serial8250_isa_devs); @@ -2612,12 +2608,13 @@ static int __init serial8250_init(void) serial8250_register_ports(&serial8250_reg, &serial8250_isa_devs->dev); - goto out; + ret = platform_driver_register(&serial8250_isa_driver); + if (ret == 0) + goto out; + platform_device_del(serial8250_isa_devs); put_dev: platform_device_put(serial8250_isa_devs); - unreg_plat_drv: - platform_driver_unregister(&serial8250_isa_driver); unreg_uart_drv: uart_unregister_driver(&serial8250_reg); out: -- cgit v1.2.3 From d9004eb466d03b7900ed432fecec6819012b4ed3 Mon Sep 17 00:00:00 2001 From: Alon Bar-Lev Date: Wed, 18 Jan 2006 11:47:33 +0000 Subject: [SERIAL] Add 8250 support for Decision Computer International Co. PCCOM2 There is a new device which is look like: Serial controller: Decision Computer International Co. PCCOM2 (rev 02) (prog-if 02 [16550]) 0700: 6666:0004 (rev 02) (prog-if 02) Flags: medium devsel, IRQ 177 Memory at fe000000 (32-bit, non-prefetchable) [size=128] I/O ports at e880 [size=128] I/O ports at e400 [size=256] It has two 16550A, and is not listed in kernel, although the manufacturer clams that it is supported... I've created the following patch, it only add the new PCI id and the card to the repository, it seems to work. Signed-off-by: Alon Bar-Lev Signed-off-by: Russell King --- drivers/serial/8250_pci.c | 10 ++++++++++ include/linux/pci_ids.h | 1 + 2 files changed, 11 insertions(+) diff --git a/drivers/serial/8250_pci.c b/drivers/serial/8250_pci.c index 589fb076654..2a912153321 100644 --- a/drivers/serial/8250_pci.c +++ b/drivers/serial/8250_pci.c @@ -940,6 +940,7 @@ enum pci_board_num_t { pbn_b2_bt_2_921600, pbn_b2_bt_4_921600, + pbn_b3_2_115200, pbn_b3_4_115200, pbn_b3_8_115200, @@ -1311,6 +1312,12 @@ static struct pciserial_board pci_boards[] __devinitdata = { .uart_offset = 8, }, + [pbn_b3_2_115200] = { + .flags = FL_BASE3, + .num_ports = 2, + .base_baud = 115200, + .uart_offset = 8, + }, [pbn_b3_4_115200] = { .flags = FL_BASE3, .num_ports = 4, @@ -2272,6 +2279,9 @@ static struct pci_device_id serial_pci_tbl[] = { PCI_ANY_ID, PCI_ANY_ID, 0, 0, pbn_nec_nile4 }, + { PCI_VENDOR_ID_DCI, PCI_DEVICE_ID_DCI_PCCOM2, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, + pbn_b3_2_115200 }, { PCI_VENDOR_ID_DCI, PCI_DEVICE_ID_DCI_PCCOM4, PCI_ANY_ID, PCI_ANY_ID, 0, 0, pbn_b3_4_115200 }, diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 5403257ae3e..ecc1fc1f0f0 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1992,6 +1992,7 @@ #define PCI_VENDOR_ID_DCI 0x6666 #define PCI_DEVICE_ID_DCI_PCCOM4 0x0001 #define PCI_DEVICE_ID_DCI_PCCOM8 0x0002 +#define PCI_DEVICE_ID_DCI_PCCOM2 0x0004 #define PCI_VENDOR_ID_INTEL 0x8086 #define PCI_DEVICE_ID_INTEL_EESSC 0x0008 -- cgit v1.2.3 From c07a8475ddcadb55b11379c35c9586971a9cedbf Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 18 Jan 2006 13:41:36 -0800 Subject: [SPARC64]: Update defconfig. Signed-off-by: David S. Miller --- arch/sparc64/defconfig | 24 ++++++++++++++++++------ 1 file changed, 18 insertions(+), 6 deletions(-) diff --git a/arch/sparc64/defconfig b/arch/sparc64/defconfig index a3fb3376ffa..4770fbf4360 100644 --- a/arch/sparc64/defconfig +++ b/arch/sparc64/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.15 -# Mon Jan 9 14:36:29 2006 +# Linux kernel version: 2.6.16-rc1 +# Wed Jan 18 13:41:02 2006 # CONFIG_SPARC=y CONFIG_SPARC64=y @@ -233,6 +233,11 @@ CONFIG_VLAN_8021Q=m # CONFIG_ATALK is not set # CONFIG_X25 is not set # CONFIG_LAPB is not set + +# +# TIPC Configuration (EXPERIMENTAL) +# +# CONFIG_TIPC is not set # CONFIG_NET_DIVERT is not set # CONFIG_ECONET is not set # CONFIG_WAN_ROUTER is not set @@ -420,8 +425,7 @@ CONFIG_ISCSI_TCP=m # CONFIG_SCSI_QLOGIC_FC is not set # CONFIG_SCSI_QLOGIC_1280 is not set # CONFIG_SCSI_QLOGICPTI is not set -CONFIG_SCSI_QLA2XXX=y -# CONFIG_SCSI_QLA2XXX_EMBEDDED_FIRMWARE is not set +# CONFIG_SCSI_QLA_FC is not set # CONFIG_SCSI_LPFC is not set # CONFIG_SCSI_DC395x is not set # CONFIG_SCSI_DC390T is not set @@ -652,7 +656,6 @@ CONFIG_SERIAL_SUNSU_CONSOLE=y CONFIG_SERIAL_SUNSAB=m CONFIG_SERIAL_CORE=y CONFIG_SERIAL_CORE_CONSOLE=y -# CONFIG_SERIAL_JSM is not set CONFIG_UNIX98_PTYS=y # CONFIG_LEGACY_PTYS is not set @@ -737,6 +740,12 @@ CONFIG_I2C_ALGOBIT=y # CONFIG_I2C_DEBUG_BUS is not set # CONFIG_I2C_DEBUG_CHIP is not set +# +# SPI support +# +# CONFIG_SPI is not set +# CONFIG_SPI_MASTER is not set + # # Dallas's 1-wire bus # @@ -1014,6 +1023,7 @@ CONFIG_USB_UHCI_HCD=m # CONFIG_USB_HID=y CONFIG_USB_HIDINPUT=y +# CONFIG_USB_HIDINPUT_POWERBOOK is not set # CONFIG_HID_FF is not set CONFIG_USB_HIDDEV=y # CONFIG_USB_AIPTEK is not set @@ -1268,12 +1278,13 @@ CONFIG_KPROBES=y # Kernel hacking # CONFIG_PRINTK_TIME=y -CONFIG_DEBUG_KERNEL=y CONFIG_MAGIC_SYSRQ=y +CONFIG_DEBUG_KERNEL=y CONFIG_LOG_BUF_SHIFT=18 CONFIG_DETECT_SOFTLOCKUP=y CONFIG_SCHEDSTATS=y # CONFIG_DEBUG_SLAB is not set +# CONFIG_DEBUG_MUTEXES is not set # CONFIG_DEBUG_SPINLOCK is not set # CONFIG_DEBUG_SPINLOCK_SLEEP is not set # CONFIG_DEBUG_KOBJECT is not set @@ -1281,6 +1292,7 @@ CONFIG_DEBUG_BUGVERBOSE=y # CONFIG_DEBUG_INFO is not set CONFIG_DEBUG_FS=y # CONFIG_DEBUG_VM is not set +CONFIG_FORCED_INLINING=y # CONFIG_RCU_TORTURE_TEST is not set # CONFIG_DEBUG_STACK_USAGE is not set # CONFIG_DEBUG_DCFLUSH is not set -- cgit v1.2.3 From fa4f0774d7c6cccb4d1fda76b91dd8eddcb2dd6a Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Wed, 18 Jan 2006 14:05:16 -0800 Subject: [CASSINI]: dont touch page_count Remove page refcount manipulations from cassini driver by using another field in struct page. Needed for lockless pagecache. Signed-off-by: Nick Piggin Signed-off-by: David S. Miller --- drivers/net/cassini.c | 36 ++++++++++++++++++++++++++++++++---- 1 file changed, 32 insertions(+), 4 deletions(-) diff --git a/drivers/net/cassini.c b/drivers/net/cassini.c index dde631f8f68..6e295fce5c6 100644 --- a/drivers/net/cassini.c +++ b/drivers/net/cassini.c @@ -335,6 +335,30 @@ static inline void cas_mask_intr(struct cas *cp) cas_disable_irq(cp, i); } +static inline void cas_buffer_init(cas_page_t *cp) +{ + struct page *page = cp->buffer; + atomic_set((atomic_t *)&page->lru.next, 1); +} + +static inline int cas_buffer_count(cas_page_t *cp) +{ + struct page *page = cp->buffer; + return atomic_read((atomic_t *)&page->lru.next); +} + +static inline void cas_buffer_inc(cas_page_t *cp) +{ + struct page *page = cp->buffer; + atomic_inc((atomic_t *)&page->lru.next); +} + +static inline void cas_buffer_dec(cas_page_t *cp) +{ + struct page *page = cp->buffer; + atomic_dec((atomic_t *)&page->lru.next); +} + static void cas_enable_irq(struct cas *cp, const int ring) { if (ring == 0) { /* all but TX_DONE */ @@ -472,6 +496,7 @@ static int cas_page_free(struct cas *cp, cas_page_t *page) { pci_unmap_page(cp->pdev, page->dma_addr, cp->page_size, PCI_DMA_FROMDEVICE); + cas_buffer_dec(page); __free_pages(page->buffer, cp->page_order); kfree(page); return 0; @@ -501,6 +526,7 @@ static cas_page_t *cas_page_alloc(struct cas *cp, const gfp_t flags) page->buffer = alloc_pages(flags, cp->page_order); if (!page->buffer) goto page_err; + cas_buffer_init(page); page->dma_addr = pci_map_page(cp->pdev, page->buffer, 0, cp->page_size, PCI_DMA_FROMDEVICE); return page; @@ -579,7 +605,7 @@ static void cas_spare_recover(struct cas *cp, const gfp_t flags) list_for_each_safe(elem, tmp, &list) { cas_page_t *page = list_entry(elem, cas_page_t, list); - if (page_count(page->buffer) > 1) + if (cas_buffer_count(page) > 1) continue; list_del(elem); @@ -1347,7 +1373,7 @@ static inline cas_page_t *cas_page_spare(struct cas *cp, const int index) cas_page_t *page = cp->rx_pages[1][index]; cas_page_t *new; - if (page_count(page->buffer) == 1) + if (cas_buffer_count(page) == 1) return page; new = cas_page_dequeue(cp); @@ -1367,7 +1393,7 @@ static cas_page_t *cas_page_swap(struct cas *cp, const int ring, cas_page_t **page1 = cp->rx_pages[1]; /* swap if buffer is in use */ - if (page_count(page0[index]->buffer) > 1) { + if (cas_buffer_count(page0[index]) > 1) { cas_page_t *new = cas_page_spare(cp, index); if (new) { page1[index] = page0[index]; @@ -2039,6 +2065,7 @@ static int cas_rx_process_pkt(struct cas *cp, struct cas_rx_comp *rxc, skb->len += hlen - swivel; get_page(page->buffer); + cas_buffer_inc(page); frag->page = page->buffer; frag->page_offset = off; frag->size = hlen - swivel; @@ -2063,6 +2090,7 @@ static int cas_rx_process_pkt(struct cas *cp, struct cas_rx_comp *rxc, frag++; get_page(page->buffer); + cas_buffer_inc(page); frag->page = page->buffer; frag->page_offset = 0; frag->size = hlen; @@ -2225,7 +2253,7 @@ static int cas_post_rxds_ringN(struct cas *cp, int ring, int num) released = 0; while (entry != last) { /* make a new buffer if it's still in use */ - if (page_count(page[entry]->buffer) > 1) { + if (cas_buffer_count(page[entry]) > 1) { cas_page_t *new = cas_page_dequeue(cp); if (!new) { /* let the timer know that we need to -- cgit v1.2.3 From e048a374d0d7beb9cf3529eac71ddd79699c8669 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 18 Jan 2006 14:06:59 -0800 Subject: [IRDA]: maintainer status Jean says he really doesn't have time to much IRDA any more. The following would help motivate someone who has more time. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index ff16eac8cf5..f98955ed503 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1398,7 +1398,7 @@ IRDA SUBSYSTEM P: Jean Tourrilhes L: irda-users@lists.sourceforge.net (subscribers-only) W: http://irda.sourceforge.net/ -S: Maintained +S: Odd Fixes ISAPNP P: Jaroslav Kysela -- cgit v1.2.3 From 7ac5459ec0f074022818af35c589b9e2b406d7c3 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 18 Jan 2006 14:19:10 -0800 Subject: [PKTGEN]: Respect hard_header_len of device. Don't assume 16. Found by Ben Greear. Signed-off-by: David S. Miller --- net/core/pktgen.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/net/core/pktgen.c b/net/core/pktgen.c index 3827f881f42..da16f8fd149 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -1860,13 +1860,14 @@ static struct sk_buff *fill_packet_ipv4(struct net_device *odev, */ mod_cur_headers(pkt_dev); - skb = alloc_skb(pkt_dev->cur_pkt_size + 64 + 16, GFP_ATOMIC); + datalen = (odev->hard_header_len + 16) & ~0xf; + skb = alloc_skb(pkt_dev->cur_pkt_size + 64 + datalen, GFP_ATOMIC); if (!skb) { sprintf(pkt_dev->result, "No memory"); return NULL; } - skb_reserve(skb, 16); + skb_reserve(skb, datalen); /* Reserve for ethernet and IP header */ eth = (__u8 *) skb_push(skb, 14); -- cgit v1.2.3 From 47c886b3123a335f0622136e021e7691d60d742c Mon Sep 17 00:00:00 2001 From: Steve French Date: Wed, 18 Jan 2006 14:20:39 -0800 Subject: [CIFS] Fix oops in cifs_readpages caused by not checking buf_type in an error path of new cifs_readpages code. Signed-off-by: Steve French --- fs/cifs/file.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 378095a442d..77c990f0cb9 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -1754,7 +1754,10 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, /* need to free smb_read_data buf before exit */ if (smb_read_data) { - cifs_buf_release(smb_read_data); + if(buf_type == CIFS_SMALL_BUFFER) + cifs_small_buf_release(smb_read_data); + else if(buf_type == CIFS_LARGE_BUFFER) + cifs_buf_release(smb_read_data); smb_read_data = NULL; } -- cgit v1.2.3 From ad12583f46bcb6ce93ccd99fa063c0d701146b2e Mon Sep 17 00:00:00 2001 From: David L Stevens Date: Wed, 18 Jan 2006 14:20:56 -0800 Subject: [IPV4]: Fix multiple bugs in IGMPv3 1) fix "mld_marksources()" to a) send nothing when all queried sources are excluded b) send full exclude report when source queried sources are not excluded c) don't schedule a timer when there's nothing to report 2) fix "add_grec()" to send empty-source records when it should The original check doesn't account for a non-empty source list with all sources inactive; the new code keeps that short-circuit case, and also generates the group header with an empty list if needed. 3) fix mca_crcount decrement to be after add_grec(), which needs its original value 4) add/remove delete records and prevent current advertisements when an exclude-mode filter moves from "active" to "inactive" or vice versa based on new filter additions. Items 1-3 are just IPv4 versions of the IPv6 bugs found by Yan Zheng and fixed earlier. Item #4 is a related bug that affects exclude-mode change records only (but not queries) and also occurs in IPv6 (IPv6 version coming soon). Signed-off-by: David L Stevens Signed-off-by: David S. Miller --- net/ipv4/igmp.c | 152 +++++++++++++++++++++++++++++++++++++++++++++----------- 1 file changed, 122 insertions(+), 30 deletions(-) diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 192092b89e5..d8ce7133cd8 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -233,7 +233,18 @@ static int is_in(struct ip_mc_list *pmc, struct ip_sf_list *psf, int type, case IGMPV3_MODE_IS_EXCLUDE: if (gdeleted || sdeleted) return 0; - return !(pmc->gsquery && !psf->sf_gsresp); + if (!(pmc->gsquery && !psf->sf_gsresp)) { + if (pmc->sfmode == MCAST_INCLUDE) + return 1; + /* don't include if this source is excluded + * in all filters + */ + if (psf->sf_count[MCAST_INCLUDE]) + return type == IGMPV3_MODE_IS_INCLUDE; + return pmc->sfcount[MCAST_EXCLUDE] == + psf->sf_count[MCAST_EXCLUDE]; + } + return 0; case IGMPV3_CHANGE_TO_INCLUDE: if (gdeleted || sdeleted) return 0; @@ -385,7 +396,7 @@ static struct sk_buff *add_grec(struct sk_buff *skb, struct ip_mc_list *pmc, struct igmpv3_report *pih; struct igmpv3_grec *pgr = NULL; struct ip_sf_list *psf, *psf_next, *psf_prev, **psf_list; - int scount, first, isquery, truncate; + int scount, stotal, first, isquery, truncate; if (pmc->multiaddr == IGMP_ALL_HOSTS) return skb; @@ -395,25 +406,13 @@ static struct sk_buff *add_grec(struct sk_buff *skb, struct ip_mc_list *pmc, truncate = type == IGMPV3_MODE_IS_EXCLUDE || type == IGMPV3_CHANGE_TO_EXCLUDE; + stotal = scount = 0; + psf_list = sdeleted ? &pmc->tomb : &pmc->sources; - if (!*psf_list) { - if (type == IGMPV3_ALLOW_NEW_SOURCES || - type == IGMPV3_BLOCK_OLD_SOURCES) - return skb; - if (pmc->crcount || isquery) { - /* make sure we have room for group header and at - * least one source. - */ - if (skb && AVAILABLE(skb) < sizeof(struct igmpv3_grec)+ - sizeof(__u32)) { - igmpv3_sendpack(skb); - skb = NULL; /* add_grhead will get a new one */ - } - skb = add_grhead(skb, pmc, type, &pgr); - } - return skb; - } + if (!*psf_list) + goto empty_source; + pih = skb ? (struct igmpv3_report *)skb->h.igmph : NULL; /* EX and TO_EX get a fresh packet, if needed */ @@ -426,7 +425,6 @@ static struct sk_buff *add_grec(struct sk_buff *skb, struct ip_mc_list *pmc, } } first = 1; - scount = 0; psf_prev = NULL; for (psf=*psf_list; psf; psf=psf_next) { u32 *psrc; @@ -460,7 +458,7 @@ static struct sk_buff *add_grec(struct sk_buff *skb, struct ip_mc_list *pmc, } psrc = (u32 *)skb_put(skb, sizeof(u32)); *psrc = psf->sf_inaddr; - scount++; + scount++; stotal++; if ((type == IGMPV3_ALLOW_NEW_SOURCES || type == IGMPV3_BLOCK_OLD_SOURCES) && psf->sf_crcount) { psf->sf_crcount--; @@ -475,6 +473,21 @@ static struct sk_buff *add_grec(struct sk_buff *skb, struct ip_mc_list *pmc, } psf_prev = psf; } + +empty_source: + if (!stotal) { + if (type == IGMPV3_ALLOW_NEW_SOURCES || + type == IGMPV3_BLOCK_OLD_SOURCES) + return skb; + if (pmc->crcount || isquery) { + /* make sure we have room for group header */ + if (skb && AVAILABLE(skb)grec_nsrcs = htons(scount); @@ -557,11 +570,11 @@ static void igmpv3_send_cr(struct in_device *in_dev) skb = add_grec(skb, pmc, dtype, 1, 1); } if (pmc->crcount) { - pmc->crcount--; if (pmc->sfmode == MCAST_EXCLUDE) { type = IGMPV3_CHANGE_TO_INCLUDE; skb = add_grec(skb, pmc, type, 1, 0); } + pmc->crcount--; if (pmc->crcount == 0) { igmpv3_clear_zeros(&pmc->tomb); igmpv3_clear_zeros(&pmc->sources); @@ -594,12 +607,12 @@ static void igmpv3_send_cr(struct in_device *in_dev) /* filter mode changes */ if (pmc->crcount) { - pmc->crcount--; if (pmc->sfmode == MCAST_EXCLUDE) type = IGMPV3_CHANGE_TO_EXCLUDE; else type = IGMPV3_CHANGE_TO_INCLUDE; skb = add_grec(skb, pmc, type, 0, 0); + pmc->crcount--; } spin_unlock_bh(&pmc->lock); } @@ -735,11 +748,43 @@ static void igmp_timer_expire(unsigned long data) ip_ma_put(im); } -static void igmp_marksources(struct ip_mc_list *pmc, int nsrcs, __u32 *srcs) +/* mark EXCLUDE-mode sources */ +static int igmp_xmarksources(struct ip_mc_list *pmc, int nsrcs, __u32 *srcs) +{ + struct ip_sf_list *psf; + int i, scount; + + scount = 0; + for (psf=pmc->sources; psf; psf=psf->sf_next) { + if (scount == nsrcs) + break; + for (i=0; isfcount[MCAST_INCLUDE] || + pmc->sfcount[MCAST_EXCLUDE] != + psf->sf_count[MCAST_EXCLUDE]) + continue; + if (srcs[i] == psf->sf_inaddr) { + scount++; + break; + } + } + } + pmc->gsquery = 0; + if (scount == nsrcs) /* all sources excluded */ + return 0; + return 1; +} + +static int igmp_marksources(struct ip_mc_list *pmc, int nsrcs, __u32 *srcs) { struct ip_sf_list *psf; int i, scount; + if (pmc->sfmode == MCAST_EXCLUDE) + return igmp_xmarksources(pmc, nsrcs, srcs); + + /* mark INCLUDE-mode sources */ scount = 0; for (psf=pmc->sources; psf; psf=psf->sf_next) { if (scount == nsrcs) @@ -751,6 +796,12 @@ static void igmp_marksources(struct ip_mc_list *pmc, int nsrcs, __u32 *srcs) break; } } + if (!scount) { + pmc->gsquery = 0; + return 0; + } + pmc->gsquery = 1; + return 1; } static void igmp_heard_report(struct in_device *in_dev, u32 group) @@ -845,6 +896,8 @@ static void igmp_heard_query(struct in_device *in_dev, struct sk_buff *skb, */ read_lock(&in_dev->mc_list_lock); for (im=in_dev->mc_list; im!=NULL; im=im->next) { + int changed; + if (group && group != im->multiaddr) continue; if (im->multiaddr == IGMP_ALL_HOSTS) @@ -854,10 +907,11 @@ static void igmp_heard_query(struct in_device *in_dev, struct sk_buff *skb, im->gsquery = im->gsquery && mark; else im->gsquery = mark; - if (im->gsquery) - igmp_marksources(im, ntohs(ih3->nsrcs), ih3->srcs); + changed = !im->gsquery || + igmp_marksources(im, ntohs(ih3->nsrcs), ih3->srcs); spin_unlock_bh(&im->lock); - igmp_mod_timer(im, max_delay); + if (changed) + igmp_mod_timer(im, max_delay); } read_unlock(&in_dev->mc_list_lock); } @@ -1510,7 +1564,7 @@ static void sf_markstate(struct ip_mc_list *pmc) static int sf_setstate(struct ip_mc_list *pmc) { - struct ip_sf_list *psf; + struct ip_sf_list *psf, *dpsf; int mca_xcount = pmc->sfcount[MCAST_EXCLUDE]; int qrv = pmc->interface->mr_qrv; int new_in, rv; @@ -1522,8 +1576,46 @@ static int sf_setstate(struct ip_mc_list *pmc) !psf->sf_count[MCAST_INCLUDE]; } else new_in = psf->sf_count[MCAST_INCLUDE] != 0; - if (new_in != psf->sf_oldin) { - psf->sf_crcount = qrv; + if (new_in) { + if (!psf->sf_oldin) { + struct ip_sf_list *prev = 0; + + for (dpsf=pmc->tomb; dpsf; dpsf=dpsf->sf_next) { + if (dpsf->sf_inaddr == psf->sf_inaddr) + break; + prev = dpsf; + } + if (dpsf) { + if (prev) + prev->sf_next = dpsf->sf_next; + else + pmc->tomb = dpsf->sf_next; + kfree(dpsf); + } + psf->sf_crcount = qrv; + rv++; + } + } else if (psf->sf_oldin) { + + psf->sf_crcount = 0; + /* + * add or update "delete" records if an active filter + * is now inactive + */ + for (dpsf=pmc->tomb; dpsf; dpsf=dpsf->sf_next) + if (dpsf->sf_inaddr == psf->sf_inaddr) + break; + if (!dpsf) { + dpsf = (struct ip_sf_list *) + kmalloc(sizeof(*dpsf), GFP_ATOMIC); + if (!dpsf) + continue; + *dpsf = *psf; + /* pmc->lock held by callers */ + dpsf->sf_next = pmc->tomb; + pmc->tomb = dpsf; + } + dpsf->sf_crcount = qrv; rv++; } } -- cgit v1.2.3 From 68477d11769ce8c6830523f08637894c43885c7e Mon Sep 17 00:00:00 2001 From: David Vrabel Date: Wed, 18 Jan 2006 22:38:44 +0000 Subject: [ARM] 3267/1: PXA27x SSP controller register defines Patch from David Vrabel PXA27x SSP controller has a few different registers, including SCR (serial clock rate) in SSCR0. Signed-off-by: David Vrabel Signed-off-by: Russell King --- include/asm-arm/arch-pxa/pxa-regs.h | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/include/asm-arm/arch-pxa/pxa-regs.h b/include/asm-arm/arch-pxa/pxa-regs.h index dae138b9cac..1409c5bd703 100644 --- a/include/asm-arm/arch-pxa/pxa-regs.h +++ b/include/asm-arm/arch-pxa/pxa-regs.h @@ -108,6 +108,7 @@ #define DCSR_STARTINTR (1 << 1) /* Start Interrupt (read / write) */ #define DCSR_BUSERR (1 << 0) /* Bus Error Interrupt (read / write) */ +#define DALGN __REG(0x400000a0) /* DMA Alignment Register */ #define DINT __REG(0x400000f0) /* DMA Interrupt Register */ #define DRCMR(n) __REG2(0x40000100, (n)<<2) @@ -1614,8 +1615,21 @@ #define SSCR0_National (0x2 << 4) /* National Microwire */ #define SSCR0_ECS (1 << 6) /* External clock select */ #define SSCR0_SSE (1 << 7) /* Synchronous Serial Port Enable */ +#if defined(CONFIG_PXA25x) #define SSCR0_SCR (0x0000ff00) /* Serial Clock Rate (mask) */ #define SSCR0_SerClkDiv(x) ((((x) - 2)/2) << 8) /* Divisor [2..512] */ +#elif defined(CONFIG_PXA27x) +#define SSCR0_SCR (0x000fff00) /* Serial Clock Rate (mask) */ +#define SSCR0_SerClkDiv(x) (((x) - 1) << 8) /* Divisor [1..4096] */ +#define SSCR0_EDSS (1 << 20) /* Extended data size select */ +#define SSCR0_NCS (1 << 21) /* Network clock select */ +#define SSCR0_RIM (1 << 22) /* Receive FIFO overrrun interrupt mask */ +#define SSCR0_TUM (1 << 23) /* Transmit FIFO underrun interrupt mask */ +#define SSCR0_FRDC (0x07000000) /* Frame rate divider control (mask) */ +#define SSCR0_SlotsPerFrm(c) ((x) - 1) /* Time slots per frame [1..8] */ +#define SSCR0_ADC (1 << 30) /* Audio clock select */ +#define SSCR0_MOD (1 << 31) /* Mode (normal or network) */ +#endif #define SSCR1_RIE (1 << 0) /* Receive FIFO Interrupt Enable */ #define SSCR1_TIE (1 << 1) /* Transmit FIFO Interrupt Enable */ -- cgit v1.2.3 From 1230b4046b9da77ccb83ef2eaffc54dc4a96093a Mon Sep 17 00:00:00 2001 From: Andrew Victor Date: Wed, 18 Jan 2006 22:38:46 +0000 Subject: [ARM] 3268/1: AT91RM9200 serial update for 2.6.15-git12 Patch from Andrew Victor This patch fixes two small issues with 2.6.15-git12. 1) Corrected major/minor numbers for ttyAT devices in the KConfig help. (Patch from Karl Olsen) 2) tty->flip.count has been removed. Signed-off-by: Andrew Victor Signed-off-by: Russell King --- drivers/serial/Kconfig | 2 +- drivers/serial/at91_serial.c | 2 -- 2 files changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig index 5e7199f7b59..9fd1925de36 100644 --- a/drivers/serial/Kconfig +++ b/drivers/serial/Kconfig @@ -301,7 +301,7 @@ config SERIAL_AT91_TTYAT depends on SERIAL_AT91=y help Say Y here if you wish to have the five internal AT91RM9200 UARTs - appear as /dev/ttyAT0-4 (major 240, minor 0-4) instead of the + appear as /dev/ttyAT0-4 (major 204, minor 154-158) instead of the normal /dev/ttyS0-4 (major 4, minor 64-68). This is necessary if you also want other UARTs, such as external 8250/16C550 compatible UARTs. diff --git a/drivers/serial/at91_serial.c b/drivers/serial/at91_serial.c index 0e206063d68..2113feb75c3 100644 --- a/drivers/serial/at91_serial.c +++ b/drivers/serial/at91_serial.c @@ -222,8 +222,6 @@ static void at91_rx_chars(struct uart_port *port, struct pt_regs *regs) while (status & (AT91_US_RXRDY)) { ch = UART_GET_CHAR(port); - if (tty->flip.count >= TTY_FLIPBUF_SIZE) - goto ignore_char; port->icount.rx++; flg = TTY_NORMAL; -- cgit v1.2.3 From fcca538b83f2984095f15f0f90f6c7686e3a32d4 Mon Sep 17 00:00:00 2001 From: Nicolas Pitre Date: Wed, 18 Jan 2006 22:38:47 +0000 Subject: [ARM] 3270/1: ARM EABI: fix sigreturn and rt_sigreturn Patch from Nicolas Pitre The signal return path consists of user code provided by the kernel. Since a syscall is used, it has to be updated to work with EABI. Noticed by Daniel Jacobowitz. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King --- arch/arm/kernel/signal.c | 29 ++++++++++++++++++----------- arch/arm/kernel/signal.h | 2 +- 2 files changed, 19 insertions(+), 12 deletions(-) diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c index 765922bcf9e..a0cd0a90a10 100644 --- a/arch/arm/kernel/signal.c +++ b/arch/arm/kernel/signal.c @@ -29,6 +29,12 @@ #define SWI_SYS_SIGRETURN (0xef000000|(__NR_sigreturn)) #define SWI_SYS_RT_SIGRETURN (0xef000000|(__NR_rt_sigreturn)) +/* + * With EABI, the syscall number has to be loaded into r7. + */ +#define MOV_R7_NR_SIGRETURN (0xe3a07000 | (__NR_sigreturn - __NR_SYSCALL_BASE)) +#define MOV_R7_NR_RT_SIGRETURN (0xe3a07000 | (__NR_rt_sigreturn - __NR_SYSCALL_BASE)) + /* * For Thumb syscalls, we pass the syscall number via r7. We therefore * need two 16-bit instructions. @@ -36,9 +42,9 @@ #define SWI_THUMB_SIGRETURN (0xdf00 << 16 | 0x2700 | (__NR_sigreturn - __NR_SYSCALL_BASE)) #define SWI_THUMB_RT_SIGRETURN (0xdf00 << 16 | 0x2700 | (__NR_rt_sigreturn - __NR_SYSCALL_BASE)) -const unsigned long sigreturn_codes[4] = { - SWI_SYS_SIGRETURN, SWI_THUMB_SIGRETURN, - SWI_SYS_RT_SIGRETURN, SWI_THUMB_RT_SIGRETURN +const unsigned long sigreturn_codes[7] = { + MOV_R7_NR_SIGRETURN, SWI_SYS_SIGRETURN, SWI_THUMB_SIGRETURN, + MOV_R7_NR_RT_SIGRETURN, SWI_SYS_RT_SIGRETURN, SWI_THUMB_RT_SIGRETURN, }; static int do_signal(sigset_t *oldset, struct pt_regs * regs, int syscall); @@ -189,7 +195,7 @@ struct aux_sigframe { struct sigframe { struct sigcontext sc; unsigned long extramask[_NSIG_WORDS-1]; - unsigned long retcode; + unsigned long retcode[2]; struct aux_sigframe aux __attribute__((aligned(8))); }; @@ -198,7 +204,7 @@ struct rt_sigframe { void __user *puc; struct siginfo info; struct ucontext uc; - unsigned long retcode; + unsigned long retcode[2]; struct aux_sigframe aux __attribute__((aligned(8))); }; @@ -436,12 +442,13 @@ setup_return(struct pt_regs *regs, struct k_sigaction *ka, if (ka->sa.sa_flags & SA_RESTORER) { retcode = (unsigned long)ka->sa.sa_restorer; } else { - unsigned int idx = thumb; + unsigned int idx = thumb << 1; if (ka->sa.sa_flags & SA_SIGINFO) - idx += 2; + idx += 3; - if (__put_user(sigreturn_codes[idx], rc)) + if (__put_user(sigreturn_codes[idx], rc) || + __put_user(sigreturn_codes[idx+1], rc+1)) return 1; if (cpsr & MODE32_BIT) { @@ -456,7 +463,7 @@ setup_return(struct pt_regs *regs, struct k_sigaction *ka, * the return code written onto the stack. */ flush_icache_range((unsigned long)rc, - (unsigned long)(rc + 1)); + (unsigned long)(rc + 2)); retcode = ((unsigned long)rc) + thumb; } @@ -488,7 +495,7 @@ setup_frame(int usig, struct k_sigaction *ka, sigset_t *set, struct pt_regs *reg } if (err == 0) - err = setup_return(regs, ka, &frame->retcode, frame, usig); + err = setup_return(regs, ka, frame->retcode, frame, usig); return err; } @@ -522,7 +529,7 @@ setup_rt_frame(int usig, struct k_sigaction *ka, siginfo_t *info, err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); if (err == 0) - err = setup_return(regs, ka, &frame->retcode, frame, usig); + err = setup_return(regs, ka, frame->retcode, frame, usig); if (err == 0) { /* diff --git a/arch/arm/kernel/signal.h b/arch/arm/kernel/signal.h index 91d26faca62..9991049c522 100644 --- a/arch/arm/kernel/signal.h +++ b/arch/arm/kernel/signal.h @@ -9,4 +9,4 @@ */ #define KERN_SIGRETURN_CODE 0xffff0500 -extern const unsigned long sigreturn_codes[4]; +extern const unsigned long sigreturn_codes[7]; -- cgit v1.2.3 From 5e0974459df9acd6cd0952c98a58816ad820ae66 Mon Sep 17 00:00:00 2001 From: Nicolas Pitre Date: Wed, 18 Jan 2006 22:38:49 +0000 Subject: [ARM] 3271/1: ARM EABI: fix calling of cmpxchg syscall emulation Patch from Nicolas Pitre This is kernel provided user space code. Since a syscall is used, it has to be updated to work with EABI. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King --- arch/arm/kernel/entry-armv.S | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S index 874e6bb7940..d401d908c46 100644 --- a/arch/arm/kernel/entry-armv.S +++ b/arch/arm/kernel/entry-armv.S @@ -735,8 +735,11 @@ __kuser_cmpxchg: @ 0xffff0fc0 * The kernel itself must perform the operation. * A special ghost syscall is used for that (see traps.c). */ + stmfd sp!, {r7, lr} + mov r7, #0xff00 @ 0xfff0 into r7 for EABI + orr r7, r7, #0xf0 swi #0x9ffff0 - mov pc, lr + ldmfd sp!, {r7, pc} #elif __LINUX_ARM_ARCH__ < 6 -- cgit v1.2.3 From 265d5e48dd1fd4c53f2b454e1e82b7713ddca37e Mon Sep 17 00:00:00 2001 From: Nicolas Pitre Date: Wed, 18 Jan 2006 22:38:51 +0000 Subject: [ARM] 3272/1: fix kernel decompressor crash Patch from Nicolas Pitre Commit f4619025a51747a3788fd1bb6bdc46e368a889a7 broke the kernel decompressor (at least on PXA). Here's the fix. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King --- arch/arm/boot/compressed/head.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/boot/compressed/head.S b/arch/arm/boot/compressed/head.S index aaa47400eb9..db3389d8e02 100644 --- a/arch/arm/boot/compressed/head.S +++ b/arch/arm/boot/compressed/head.S @@ -334,7 +334,7 @@ __setup_mmu: sub r3, r4, #16384 @ Page directory size mov r1, #0x12 orr r1, r1, #3 << 10 add r2, r3, #16384 -1: cmp r1, r8 @ if virt > start of RAM +1: cmp r1, r9 @ if virt > start of RAM orrhs r1, r1, #0x0c @ set cacheable, bufferable cmp r1, r10 @ if virt > end of RAM bichs r1, r1, #0x0c @ clear cacheable, bufferable -- cgit v1.2.3 From 1e74c891252941ac195bb47978f079c592271a3f Mon Sep 17 00:00:00 2001 From: David Vrabel Date: Wed, 18 Jan 2006 22:46:43 +0000 Subject: [ARM] 3281/1: ixp4xx: export ixp4xx_exp_bus_size for modules Patch from David Vrabel Export ixp4xx_exp_bus_size so modules can use the IXP4XX_EXP_BUS_BASE(n) macro. Also, fix a printk format warning. Signed-off-by: David Vrabel Signed-off-by: Russell King --- arch/arm/mach-ixp4xx/common.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/arm/mach-ixp4xx/common.c b/arch/arm/mach-ixp4xx/common.c index 6b393691d0e..4bdc9d4526c 100644 --- a/arch/arm/mach-ixp4xx/common.c +++ b/arch/arm/mach-ixp4xx/common.c @@ -333,6 +333,7 @@ static struct platform_device *ixp46x_devices[] __initdata = { }; unsigned long ixp4xx_exp_bus_size; +EXPORT_SYMBOL(ixp4xx_exp_bus_size); void __init ixp4xx_sys_init(void) { @@ -352,7 +353,7 @@ void __init ixp4xx_sys_init(void) } } - printk("IXP4xx: Using %uMiB expansion bus window size\n", + printk("IXP4xx: Using %luMiB expansion bus window size\n", ixp4xx_exp_bus_size >> 20); } -- cgit v1.2.3 From d5ca3117b07ebb0e7dcd5e7a1219277cc4478266 Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Wed, 18 Jan 2006 14:52:18 -0800 Subject: [MAINTAINERS]: correct location for net-2.6.git Correct location info for net-2.6 git tree. Signed-off-by: John W. Linville Signed-off-by: David S. Miller --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index c4d0e5913b4..105ad85f5d6 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1843,7 +1843,7 @@ M: yoshfuji@linux-ipv6.org P: Patrick McHardy M: kaber@coreworks.de L: netdev@vger.kernel.org -T: git kernel.org:/pub/scm/linux/kernel/davem/net-2.6.git +T: git kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6.git S: Maintained IPVS -- cgit v1.2.3 From 29f8f63272bd286549d1c7f75ad22686191d35d8 Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Wed, 18 Jan 2006 14:52:48 -0800 Subject: [MAINTAINERS]: add entry for wireless networking Add an entry to MAINTAINERS for wireless networking, just so people know whom to bless with patches. Signed-off-by: John W. Linville Signed-off-by: David S. Miller --- MAINTAINERS | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 105ad85f5d6..e6dbb21a8e5 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1846,6 +1846,13 @@ L: netdev@vger.kernel.org T: git kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6.git S: Maintained +NETWORKING [WIRELESS] +P: John W. Linville +M: linville@tuxdriver.com +L: netdev@vger.kernel.org +T: git kernel.org:/pub/scm/linux/kernel/git/linville/wireless-2.6.git +S: Maintained + IPVS P: Wensong Zhang M: wensong@linux-vs.org -- cgit v1.2.3 From c126cf80d450a4d0aac3de7162d4c14b5c971b24 Mon Sep 17 00:00:00 2001 From: "Eddie C. Dost" Date: Wed, 18 Jan 2006 14:54:31 -0800 Subject: [SPARC64]: Serial Console for E250 Patch From: Eddie C. Dost I have the following patch for serial console over the RSC (remote system controller) on my E250 machine. It basically adds support for input-device=rsc and output-device=rsc from OBP, and allows 115200,8,n,1,- serial mode setting. Signed-off-by: David S. Miller --- arch/sparc64/kernel/setup.c | 2 ++ arch/sparc64/prom/console.c | 8 ++++++++ drivers/serial/suncore.c | 34 ++++++++++++++++++++++++++++++++-- drivers/serial/sunsab.c | 7 ++++--- include/asm-sparc64/oplib.h | 2 ++ 5 files changed, 48 insertions(+), 5 deletions(-) diff --git a/arch/sparc64/kernel/setup.c b/arch/sparc64/kernel/setup.c index 250745896ae..054461e6946 100644 --- a/arch/sparc64/kernel/setup.c +++ b/arch/sparc64/kernel/setup.c @@ -561,6 +561,8 @@ static int __init set_preferred_console(void) serial_console = 1; } else if (idev == PROMDEV_ITTYB && odev == PROMDEV_OTTYB) { serial_console = 2; + } else if (idev == PROMDEV_IRSC && odev == PROMDEV_ORSC) { + serial_console = 3; } else { prom_printf("Inconsistent console: " "input %d, output %d\n", diff --git a/arch/sparc64/prom/console.c b/arch/sparc64/prom/console.c index eae5db8dda5..ac6d035dd15 100644 --- a/arch/sparc64/prom/console.c +++ b/arch/sparc64/prom/console.c @@ -99,8 +99,12 @@ prom_query_input_device(void) if (!strncmp(propb, "keyboard", 8)) return PROMDEV_ITTYA; + if (!strncmp (propb, "rsc", 3)) + return PROMDEV_IRSC; + if (strncmp (propb, "tty", 3) || !propb[3]) return PROMDEV_I_UNK; + switch (propb[3]) { case 'a': return PROMDEV_ITTYA; case 'b': return PROMDEV_ITTYB; @@ -136,8 +140,12 @@ prom_query_output_device(void) if (!strncmp(propb, "screen", 6)) return PROMDEV_OTTYA; + if (!strncmp (propb, "rsc", 3)) + return PROMDEV_ORSC; + if (strncmp (propb, "tty", 3) || !propb[3]) return PROMDEV_O_UNK; + switch (propb[3]) { case 'a': return PROMDEV_OTTYA; case 'b': return PROMDEV_OTTYB; diff --git a/drivers/serial/suncore.c b/drivers/serial/suncore.c index 5fc4a62173d..fa4ae94243c 100644 --- a/drivers/serial/suncore.c +++ b/drivers/serial/suncore.c @@ -34,6 +34,7 @@ sunserial_console_termios(struct console *con) char *mode_prop = "ttyX-mode"; char *cd_prop = "ttyX-ignore-cd"; char *dtr_prop = "ttyX-rts-dtr-off"; + char *ssp_console_modes_prop = "ssp-console-modes"; int baud, bits, stop, cflag; char parity; int carrier = 0; @@ -43,14 +44,39 @@ sunserial_console_termios(struct console *con) if (!serial_console) return; - if (serial_console == 1) { + switch (serial_console) { + case PROMDEV_OTTYA: mode_prop[3] = 'a'; cd_prop[3] = 'a'; dtr_prop[3] = 'a'; - } else { + break; + + case PROMDEV_OTTYB: mode_prop[3] = 'b'; cd_prop[3] = 'b'; dtr_prop[3] = 'b'; + break; + + case PROMDEV_ORSC: + + nd = prom_pathtoinode("rsc"); + if (!nd) { + strcpy(mode, "115200,8,n,1,-"); + goto no_options; + } + + if (!prom_node_has_property(nd, ssp_console_modes_prop)) { + strcpy(mode, "115200,8,n,1,-"); + goto no_options; + } + + memset(mode, 0, sizeof(mode)); + prom_getstring(nd, ssp_console_modes_prop, mode, sizeof(mode)); + goto no_options; + + default: + strcpy(mode, "9600,8,n,1,-"); + goto no_options; } topnd = prom_getchild(prom_root_node); @@ -110,6 +136,10 @@ no_options: case 9600: cflag |= B9600; break; case 19200: cflag |= B19200; break; case 38400: cflag |= B38400; break; + case 57600: cflag |= B57600; break; + case 115200: cflag |= B115200; break; + case 230400: cflag |= B230400; break; + case 460800: cflag |= B460800; break; default: baud = 9600; cflag |= B9600; break; } diff --git a/drivers/serial/sunsab.c b/drivers/serial/sunsab.c index 7e773ff76c6..8bcaebcc0ad 100644 --- a/drivers/serial/sunsab.c +++ b/drivers/serial/sunsab.c @@ -897,9 +897,6 @@ static int sunsab_console_setup(struct console *con, char *options) sunserial_console_termios(con); - /* Firmware console speed is limited to 150-->38400 baud so - * this hackish cflag thing is OK. - */ switch (con->cflag & CBAUD) { case B150: baud = 150; break; case B300: baud = 300; break; @@ -910,6 +907,10 @@ static int sunsab_console_setup(struct console *con, char *options) default: case B9600: baud = 9600; break; case B19200: baud = 19200; break; case B38400: baud = 38400; break; + case B57600: baud = 57600; break; + case B115200: baud = 115200; break; + case B230400: baud = 230400; break; + case B460800: baud = 460800; break; }; /* diff --git a/include/asm-sparc64/oplib.h b/include/asm-sparc64/oplib.h index d02f1e8ae1a..3c59b2693fb 100644 --- a/include/asm-sparc64/oplib.h +++ b/include/asm-sparc64/oplib.h @@ -163,6 +163,7 @@ enum prom_input_device { PROMDEV_IKBD, /* input from keyboard */ PROMDEV_ITTYA, /* input from ttya */ PROMDEV_ITTYB, /* input from ttyb */ + PROMDEV_IRSC, /* input from rsc */ PROMDEV_I_UNK, }; @@ -174,6 +175,7 @@ enum prom_output_device { PROMDEV_OSCREEN, /* to screen */ PROMDEV_OTTYA, /* to ttya */ PROMDEV_OTTYB, /* to ttyb */ + PROMDEV_ORSC, /* to rsc */ PROMDEV_O_UNK, }; -- cgit v1.2.3 From 959a85ada599fd5bab7fab8251795c3a58af28d4 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 18 Jan 2006 14:58:05 -0800 Subject: [SPARC64]: Fix build with CONFIG_COMPAT disabled. Based upon a report and preliminary patch from Jim Gifford. Signed-off-by: David S. Miller --- arch/sparc64/kernel/process.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/sparc64/kernel/process.c b/arch/sparc64/kernel/process.c index 1dc3650c5ca..059b0d02522 100644 --- a/arch/sparc64/kernel/process.c +++ b/arch/sparc64/kernel/process.c @@ -164,6 +164,7 @@ void machine_restart(char * cmd) panic("Reboot failed!"); } +#ifdef CONFIG_COMPAT static void show_regwindow32(struct pt_regs *regs) { struct reg_window32 __user *rw; @@ -189,6 +190,9 @@ static void show_regwindow32(struct pt_regs *regs) r_w.ins[0], r_w.ins[1], r_w.ins[2], r_w.ins[3], r_w.ins[4], r_w.ins[5], r_w.ins[6], r_w.ins[7]); } +#else +#define show_regwindow32(regs) do { } while (0) +#endif static void show_regwindow(struct pt_regs *regs) { -- cgit v1.2.3 From 8d3c7fce2d20ecc3264c8d8c91ae3beacdeaed1b Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Wed, 18 Jan 2006 17:42:19 -0800 Subject: [PATCH] jbd: log_do_checkpoint fix While checkpointing we have to check that our transaction still is in the checkpoint list *and* (not or) that it's not just a different transaction with the same address. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/jbd/checkpoint.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c index cb3cef525c3..e6265a0b56b 100644 --- a/fs/jbd/checkpoint.c +++ b/fs/jbd/checkpoint.c @@ -338,7 +338,7 @@ restart: * done (maybe it's a new transaction, but it fell at the same * address). */ - if (journal->j_checkpoint_transactions == transaction || + if (journal->j_checkpoint_transactions == transaction && transaction->t_tid == this_tid) { int batch_count = 0; struct buffer_head *bhs[NR_BATCH]; -- cgit v1.2.3 From 43c3e6f5abdf6acac9b90c86bf03f995bf7d3d92 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Wed, 18 Jan 2006 17:42:20 -0800 Subject: [PATCH] jbd: remove_transaction fix We have to check that also the second checkpoint list is non-empty before dropping the transaction. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/jbd/commit.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c index 002ad2bbc76..29e62d98bae 100644 --- a/fs/jbd/commit.c +++ b/fs/jbd/commit.c @@ -829,7 +829,8 @@ restart_loop: journal->j_committing_transaction = NULL; spin_unlock(&journal->j_state_lock); - if (commit_transaction->t_checkpoint_list == NULL) { + if (commit_transaction->t_checkpoint_list == NULL && + commit_transaction->t_checkpoint_io_list == NULL) { __journal_drop_transaction(journal, commit_transaction); } else { if (journal->j_checkpoint_transactions == NULL) { -- cgit v1.2.3 From ab26a20bb0ca0cafb3190972c982f9bfb819040f Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Wed, 18 Jan 2006 17:42:21 -0800 Subject: [PATCH] x86_64: Fix MCE exception stack for boot CPU Fix a typo/mis-merge in one of the previous patches. Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/setup64.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c index 8ac4db09610..70f1bb808a2 100644 --- a/arch/x86_64/kernel/setup64.c +++ b/arch/x86_64/kernel/setup64.c @@ -146,7 +146,7 @@ void pda_init(int cpu) pda->irqstackptr += IRQSTACKSIZE-64; } -char boot_exception_stacks[(N_EXCEPTION_STACKS - 2) * EXCEPTION_STKSZ + DEBUG_STKSZ] +char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ + DEBUG_STKSZ] __attribute__((section(".bss.page_aligned"))); /* May not be marked __init: used by software suspend */ -- cgit v1.2.3 From c8d338c8dbc4461a6de1171c2332b8ed547f8f3b Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 18 Jan 2006 17:42:22 -0800 Subject: [PATCH] scsi_transport_spi build fix On alpha: In file included from drivers/scsi/sym53c8xx_2/sym_glue.h:59, from drivers/scsi/sym53c8xx_2/sym_fw.c:40: include/scsi/scsi_transport_spi.h:57: error: field `dv_mutex' has incomplete type Cc: James Bottomley Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/scsi/scsi_transport_spi.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/scsi/scsi_transport_spi.h b/include/scsi/scsi_transport_spi.h index 2b5930ba69e..fb5a2ffae93 100644 --- a/include/scsi/scsi_transport_spi.h +++ b/include/scsi/scsi_transport_spi.h @@ -22,6 +22,7 @@ #include #include +#include struct scsi_transport_template; struct scsi_target; -- cgit v1.2.3 From 35fbd397f0ce2addd5ab889bf0bcd95f45837e3e Mon Sep 17 00:00:00 2001 From: Paul Fulghum Date: Wed, 18 Jan 2006 17:42:24 -0800 Subject: [PATCH] synclink_gt fix size of register value storage Fix incorrect variable size used to hold register value. This bug might wipe out a portion of the TCR value when setting the interface options. Signed-off-by: Paul Fulghum Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/synclink_gt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c index 07c9be6a6bb..a85a60a93de 100644 --- a/drivers/char/synclink_gt.c +++ b/drivers/char/synclink_gt.c @@ -2630,7 +2630,7 @@ static int get_interface(struct slgt_info *info, int __user *if_mode) static int set_interface(struct slgt_info *info, int if_mode) { unsigned long flags; - unsigned char val; + unsigned short val; DBGINFO(("%s set_interface=%x)\n", info->device_name, if_mode)); spin_lock_irqsave(&info->lock,flags); -- cgit v1.2.3 From c2aeacd48ca2418fbac985a58d2200c263444d65 Mon Sep 17 00:00:00 2001 From: Jon Mason Date: Wed, 18 Jan 2006 17:42:25 -0800 Subject: [PATCH] Prevent trident driver from grabbing pcnet32 hardware Some pcnet32 hardware erroneously has the Vendor ID for Trident. The pcnet32 driver looks for the PCI ethernet class before grabbing the hardware, but the current trident driver does not check against the PCI audio class. This allows the trident driver to claim the pcnet32 hardware. This patch prevents that. This revised version of the OSS Trident patch includes PCI_DEVICE Macro usage. Signed-off-by: Jon Mason Signed-off-by: Muli Ben-Yehuda Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- sound/oss/trident.c | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-) diff --git a/sound/oss/trident.c b/sound/oss/trident.c index 5f0ad6bb43b..a21c663e7e1 100644 --- a/sound/oss/trident.c +++ b/sound/oss/trident.c @@ -278,16 +278,14 @@ static char *card_names[] = { }; static struct pci_device_id trident_pci_tbl[] = { - {PCI_VENDOR_ID_TRIDENT, PCI_DEVICE_ID_TRIDENT_4DWAVE_DX, - PCI_ANY_ID, PCI_ANY_ID, 0, 0, TRIDENT_4D_DX}, - {PCI_VENDOR_ID_TRIDENT, PCI_DEVICE_ID_TRIDENT_4DWAVE_NX, - PCI_ANY_ID, PCI_ANY_ID, 0, 0, TRIDENT_4D_NX}, - {PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_7018, - PCI_ANY_ID, PCI_ANY_ID, 0, 0, SIS_7018}, - {PCI_VENDOR_ID_ALI, PCI_DEVICE_ID_ALI_5451, - PCI_ANY_ID, PCI_ANY_ID, 0, 0, ALI_5451}, - {PCI_VENDOR_ID_INTERG, PCI_DEVICE_ID_INTERG_5050, - PCI_ANY_ID, PCI_ANY_ID, 0, 0, CYBER5050}, + {PCI_DEVICE(PCI_VENDOR_ID_TRIDENT, PCI_DEVICE_ID_TRIDENT_4DWAVE_DX), + PCI_CLASS_MULTIMEDIA_AUDIO << 8, 0xffff00, TRIDENT_4D_DX}, + {PCI_DEVICE(PCI_VENDOR_ID_TRIDENT, PCI_DEVICE_ID_TRIDENT_4DWAVE_NX), + 0, 0, TRIDENT_4D_NX}, + {PCI_DEVICE(PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_7018), 0, 0, SIS_7018}, + {PCI_DEVICE(PCI_VENDOR_ID_ALI, PCI_DEVICE_ID_ALI_5451), 0, 0, ALI_5451}, + {PCI_DEVICE(PCI_VENDOR_ID_INTERG, PCI_DEVICE_ID_INTERG_5050), + 0, 0, CYBER5050}, {0,} }; -- cgit v1.2.3 From e236a166b2bc437769a9b8b5d19186a3761bde48 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 18 Jan 2006 17:42:26 -0800 Subject: [PATCH] mm: dirty_exceeded speedup Ravikiran reports that this variable is bouncing all around nodes on NUMA machines, causing measurable performance problems. Fix that up by only writing to it when it actually changed. And put it in a new cacheline to prevent it sharing with other things (this happened). Signed-off-by: Ravikiran Thirumalai Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page-writeback.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 5240e426c1f..945559fb63d 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -46,7 +46,7 @@ static long ratelimit_pages = 32; static long total_pages; /* The total number of pages in the machine. */ -static int dirty_exceeded; /* Dirty mem may be over limit */ +static int dirty_exceeded __cacheline_aligned_in_smp; /* Dirty mem may be over limit */ /* * When balance_dirty_pages decides that the caller needs to perform some @@ -212,7 +212,8 @@ static void balance_dirty_pages(struct address_space *mapping) if (nr_reclaimable + wbs.nr_writeback <= dirty_thresh) break; - dirty_exceeded = 1; + if (!dirty_exceeded) + dirty_exceeded = 1; /* Note: nr_reclaimable denotes nr_dirty + nr_unstable. * Unstable writes are a feature of certain networked @@ -234,7 +235,7 @@ static void balance_dirty_pages(struct address_space *mapping) blk_congestion_wait(WRITE, HZ/10); } - if (nr_reclaimable + wbs.nr_writeback <= dirty_thresh) + if (nr_reclaimable + wbs.nr_writeback <= dirty_thresh && dirty_exceeded) dirty_exceeded = 0; if (writeback_in_progress(bdi)) -- cgit v1.2.3 From 053837fce7aa79025ed57656855df09f80175527 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Wed, 18 Jan 2006 17:42:27 -0800 Subject: [PATCH] mm: migration page refcounting fix Migration code currently does not take a reference to target page properly, so between unlocking the pte and trying to take a new reference to the page with isolate_lru_page, anything could happen to it. Fix this by holding the pte lock until we get a chance to elevate the refcount. Other small cleanups while we're here. Signed-off-by: Nick Piggin Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm_inline.h | 21 -------------- include/linux/swap.h | 1 + mm/filemap.c | 1 + mm/mempolicy.c | 29 +++++++++++-------- mm/rmap.c | 2 +- mm/swap.c | 26 +++++++++++++++++ mm/vmscan.c | 71 ++++++++++++++++++++--------------------------- 7 files changed, 76 insertions(+), 75 deletions(-) diff --git a/include/linux/mm_inline.h b/include/linux/mm_inline.h index 49cc68af01f..8ac854f7f19 100644 --- a/include/linux/mm_inline.h +++ b/include/linux/mm_inline.h @@ -39,24 +39,3 @@ del_page_from_lru(struct zone *zone, struct page *page) } } -/* - * Isolate one page from the LRU lists. - * - * - zone->lru_lock must be held - */ -static inline int __isolate_lru_page(struct page *page) -{ - if (unlikely(!TestClearPageLRU(page))) - return 0; - - if (get_page_testone(page)) { - /* - * It is being freed elsewhere - */ - __put_page(page); - SetPageLRU(page); - return -ENOENT; - } - - return 1; -} diff --git a/include/linux/swap.h b/include/linux/swap.h index e92054d6530..d01f7efb0f2 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -167,6 +167,7 @@ extern void FASTCALL(lru_cache_add_active(struct page *)); extern void FASTCALL(activate_page(struct page *)); extern void FASTCALL(mark_page_accessed(struct page *)); extern void lru_add_drain(void); +extern int lru_add_drain_all(void); extern int rotate_reclaimable_page(struct page *page); extern void swap_setup(void); diff --git a/mm/filemap.c b/mm/filemap.c index a965b6b35f2..44da3d47699 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -94,6 +94,7 @@ generic_file_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, * ->private_lock (try_to_unmap_one) * ->tree_lock (try_to_unmap_one) * ->zone.lru_lock (follow_page->mark_page_accessed) + * ->zone.lru_lock (check_pte_range->isolate_lru_page) * ->private_lock (page_remove_rmap->set_page_dirty) * ->tree_lock (page_remove_rmap->set_page_dirty) * ->inode_lock (page_remove_rmap->set_page_dirty) diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 3171f884d24..551cde40520 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -208,6 +208,17 @@ static int check_pte_range(struct vm_area_struct *vma, pmd_t *pmd, page = vm_normal_page(vma, addr, *pte); if (!page) continue; + /* + * The check for PageReserved here is important to avoid + * handling zero pages and other pages that may have been + * marked special by the system. + * + * If the PageReserved would not be checked here then f.e. + * the location of the zero page could have an influence + * on MPOL_MF_STRICT, zero pages would be counted for + * the per node stats, and there would be useless attempts + * to put zero pages on the migration list. + */ if (PageReserved(page)) continue; nid = page_to_nid(page); @@ -216,11 +227,8 @@ static int check_pte_range(struct vm_area_struct *vma, pmd_t *pmd, if (flags & MPOL_MF_STATS) gather_stats(page, private); - else if (flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) { - spin_unlock(ptl); + else if (flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) migrate_page_add(vma, page, private, flags); - spin_lock(ptl); - } else break; } while (pte++, addr += PAGE_SIZE, addr != end); @@ -309,6 +317,10 @@ check_range(struct mm_struct *mm, unsigned long start, unsigned long end, int err; struct vm_area_struct *first, *vma, *prev; + /* Clear the LRU lists so pages can be isolated */ + if (flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) + lru_add_drain_all(); + first = find_vma(mm, start); if (!first) return ERR_PTR(-EFAULT); @@ -555,15 +567,8 @@ static void migrate_page_add(struct vm_area_struct *vma, if ((flags & MPOL_MF_MOVE_ALL) || !page->mapping || PageAnon(page) || mapping_writably_mapped(page->mapping) || single_mm_mapping(vma->vm_mm, page->mapping)) { - int rc = isolate_lru_page(page); - - if (rc == 1) + if (isolate_lru_page(page)) list_add(&page->lru, pagelist); - /* - * If the isolate attempt was not successful then we just - * encountered an unswappable page. Something must be wrong. - */ - WARN_ON(rc == 0); } } diff --git a/mm/rmap.c b/mm/rmap.c index dfbb89f99a1..d85a99d28c0 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -33,7 +33,7 @@ * mapping->i_mmap_lock * anon_vma->lock * mm->page_table_lock or pte_lock - * zone->lru_lock (in mark_page_accessed) + * zone->lru_lock (in mark_page_accessed, isolate_lru_page) * swap_lock (in swap_duplicate, swap_info_get) * mmlist_lock (in mmput, drain_mmlist and others) * mapping->private_lock (in __set_page_dirty_buffers) diff --git a/mm/swap.c b/mm/swap.c index cbb48e721ab..bc2442a7b0e 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -174,6 +174,32 @@ void lru_add_drain(void) put_cpu(); } +#ifdef CONFIG_NUMA +static void lru_add_drain_per_cpu(void *dummy) +{ + lru_add_drain(); +} + +/* + * Returns 0 for success + */ +int lru_add_drain_all(void) +{ + return schedule_on_each_cpu(lru_add_drain_per_cpu, NULL); +} + +#else + +/* + * Returns 0 for success + */ +int lru_add_drain_all(void) +{ + lru_add_drain(); + return 0; +} +#endif + /* * This path almost never happens for VM activity - pages are normally * freed via pagevecs. But it gets used by networking. diff --git a/mm/vmscan.c b/mm/vmscan.c index bf903b2d198..827bf674577 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -586,7 +586,7 @@ static inline void move_to_lru(struct page *page) } /* - * Add isolated pages on the list back to the LRU + * Add isolated pages on the list back to the LRU. * * returns the number of pages put back. */ @@ -760,46 +760,33 @@ next: return nr_failed + retry; } -static void lru_add_drain_per_cpu(void *dummy) -{ - lru_add_drain(); -} - /* * Isolate one page from the LRU lists and put it on the - * indicated list. Do necessary cache draining if the - * page is not on the LRU lists yet. + * indicated list with elevated refcount. * * Result: * 0 = page not on LRU list * 1 = page removed from LRU list and added to the specified list. - * -ENOENT = page is being freed elsewhere. */ int isolate_lru_page(struct page *page) { - int rc = 0; - struct zone *zone = page_zone(page); + int ret = 0; -redo: - spin_lock_irq(&zone->lru_lock); - rc = __isolate_lru_page(page); - if (rc == 1) { - if (PageActive(page)) - del_page_from_active_list(zone, page); - else - del_page_from_inactive_list(zone, page); - } - spin_unlock_irq(&zone->lru_lock); - if (rc == 0) { - /* - * Maybe this page is still waiting for a cpu to drain it - * from one of the lru lists? - */ - rc = schedule_on_each_cpu(lru_add_drain_per_cpu, NULL); - if (rc == 0 && PageLRU(page)) - goto redo; + if (PageLRU(page)) { + struct zone *zone = page_zone(page); + spin_lock_irq(&zone->lru_lock); + if (TestClearPageLRU(page)) { + ret = 1; + get_page(page); + if (PageActive(page)) + del_page_from_active_list(zone, page); + else + del_page_from_inactive_list(zone, page); + } + spin_unlock_irq(&zone->lru_lock); } - return rc; + + return ret; } #endif @@ -831,18 +818,20 @@ static int isolate_lru_pages(int nr_to_scan, struct list_head *src, page = lru_to_page(src); prefetchw_prev_lru_page(page, src, flags); - switch (__isolate_lru_page(page)) { - case 1: - /* Succeeded to isolate page */ - list_move(&page->lru, dst); - nr_taken++; - break; - case -ENOENT: - /* Not possible to isolate */ - list_move(&page->lru, src); - break; - default: + if (!TestClearPageLRU(page)) BUG(); + list_del(&page->lru); + if (get_page_testone(page)) { + /* + * It is being freed elsewhere + */ + __put_page(page); + SetPageLRU(page); + list_add(&page->lru, src); + continue; + } else { + list_add(&page->lru, dst); + nr_taken++; } } -- cgit v1.2.3 From fc3012896337c83a056c496d7cfb0072e1591181 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 18 Jan 2006 17:42:29 -0800 Subject: [PATCH] Simplify migrate_page_add Simplify migrate_page_add after feedback from Hugh. This also allows us to drop one parameter from migrate_page_add. Signed-off-by: Christoph Lameter Cc: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/mempolicy.c | 43 +++++++------------------------------------ 1 file changed, 7 insertions(+), 36 deletions(-) diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 551cde40520..a683a66599b 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -185,8 +185,8 @@ static struct mempolicy *mpol_new(int mode, nodemask_t *nodes) } static void gather_stats(struct page *, void *); -static void migrate_page_add(struct vm_area_struct *vma, - struct page *page, struct list_head *pagelist, unsigned long flags); +static void migrate_page_add(struct page *page, struct list_head *pagelist, + unsigned long flags); /* Scan through pages checking if pages follow certain conditions. */ static int check_pte_range(struct vm_area_struct *vma, pmd_t *pmd, @@ -228,7 +228,7 @@ static int check_pte_range(struct vm_area_struct *vma, pmd_t *pmd, if (flags & MPOL_MF_STATS) gather_stats(page, private); else if (flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) - migrate_page_add(vma, page, private, flags); + migrate_page_add(page, private, flags); else break; } while (pte++, addr += PAGE_SIZE, addr != end); @@ -531,42 +531,13 @@ long do_get_mempolicy(int *policy, nodemask_t *nmask, * page migration */ -/* Check if we are the only process mapping the page in question */ -static inline int single_mm_mapping(struct mm_struct *mm, - struct address_space *mapping) -{ - struct vm_area_struct *vma; - struct prio_tree_iter iter; - int rc = 1; - - spin_lock(&mapping->i_mmap_lock); - vma_prio_tree_foreach(vma, &iter, &mapping->i_mmap, 0, ULONG_MAX) - if (mm != vma->vm_mm) { - rc = 0; - goto out; - } - list_for_each_entry(vma, &mapping->i_mmap_nonlinear, shared.vm_set.list) - if (mm != vma->vm_mm) { - rc = 0; - goto out; - } -out: - spin_unlock(&mapping->i_mmap_lock); - return rc; -} - -/* - * Add a page to be migrated to the pagelist - */ -static void migrate_page_add(struct vm_area_struct *vma, - struct page *page, struct list_head *pagelist, unsigned long flags) +static void migrate_page_add(struct page *page, struct list_head *pagelist, + unsigned long flags) { /* - * Avoid migrating a page that is shared by others and not writable. + * Avoid migrating a page that is shared with others. */ - if ((flags & MPOL_MF_MOVE_ALL) || !page->mapping || PageAnon(page) || - mapping_writably_mapped(page->mapping) || - single_mm_mapping(vma->vm_mm, page->mapping)) { + if ((flags & MPOL_MF_MOVE_ALL) || page_mapcount(page) == 1) { if (isolate_lru_page(page)) list_add(&page->lru, pagelist); } -- cgit v1.2.3 From f1fd1067ece574ab56e4a70878b9a5a1ed4c3c42 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 18 Jan 2006 17:42:30 -0800 Subject: [PATCH] Zone reclaim: resurrect may_swap Zone reclaim has a huge impact on NUMA performance (f.e. our maximum throughput with XFS is raised from 4GB to 6GB/sec / page cache contamination of numa nodes destroys locality if one just does a large copy operation which results in performance dropping for good until reboot). This patch: Resurrect may_swap in struct scan_control Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmscan.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/mm/vmscan.c b/mm/vmscan.c index 827bf674577..e5117b6897a 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -71,6 +71,9 @@ struct scan_control { int may_writepage; + /* Can pages be swapped as part of reclaim? */ + int may_swap; + /* This context's SWAP_CLUSTER_MAX. If freeing memory for * suspend, we effectively ignore SWAP_CLUSTER_MAX. * In this context, it doesn't matter that we scan the @@ -458,6 +461,8 @@ static int shrink_list(struct list_head *page_list, struct scan_control *sc) * Try to allocate it some swap space here. */ if (PageAnon(page) && !PageSwapCache(page)) { + if (!sc->may_swap) + goto keep_locked; if (!add_to_swap(page, GFP_ATOMIC)) goto activate_locked; } @@ -1166,6 +1171,7 @@ int try_to_free_pages(struct zone **zones, gfp_t gfp_mask) sc.gfp_mask = gfp_mask; sc.may_writepage = 0; + sc.may_swap = 1; inc_page_state(allocstall); @@ -1268,6 +1274,7 @@ loop_again: total_reclaimed = 0; sc.gfp_mask = GFP_KERNEL; sc.may_writepage = 0; + sc.may_swap = 1; sc.nr_mapped = read_page_state(nr_mapped); inc_page_state(pageoutrun); -- cgit v1.2.3 From 9eeff2395e3cfd05c9b2e6074ff943a34b0c5c21 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 18 Jan 2006 17:42:31 -0800 Subject: [PATCH] Zone reclaim: Reclaim logic Some bits for zone reclaim exists in 2.6.15 but they are not usable. This patch fixes them up, removes unused code and makes zone reclaim usable. Zone reclaim allows the reclaiming of pages from a zone if the number of free pages falls below the watermarks even if other zones still have enough pages available. Zone reclaim is of particular importance for NUMA machines. It can be more beneficial to reclaim a page than taking the performance penalties that come with allocating a page on a remote zone. Zone reclaim is enabled if the maximum distance to another node is higher than RECLAIM_DISTANCE, which may be defined by an arch. By default RECLAIM_DISTANCE is 20. 20 is the distance to another node in the same component (enclosure or motherboard) on IA64. The meaning of the NUMA distance information seems to vary by arch. If zone reclaim is not successful then no further reclaim attempts will occur for a certain time period (ZONE_RECLAIM_INTERVAL). This patch was discussed before. See http://marc.theaimsgroup.com/?l=linux-kernel&m=113519961504207&w=2 http://marc.theaimsgroup.com/?l=linux-kernel&m=113408418232531&w=2 http://marc.theaimsgroup.com/?l=linux-kernel&m=113389027420032&w=2 http://marc.theaimsgroup.com/?l=linux-kernel&m=113380938612205&w=2 Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 12 +++++---- include/linux/swap.h | 11 ++++++++ include/linux/topology.h | 8 ++++++ mm/page_alloc.c | 17 +++++++++--- mm/vmscan.c | 68 ++++++++++++++++++++++++++++++++++++++++++++++++ 5 files changed, 108 insertions(+), 8 deletions(-) diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 34cbefd2ebd..93a849f742d 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -149,14 +149,16 @@ struct zone { unsigned long pages_scanned; /* since last reclaim */ int all_unreclaimable; /* All pages pinned */ - /* - * Does the allocator try to reclaim pages from the zone as soon - * as it fails a watermark_ok() in __alloc_pages? - */ - int reclaim_pages; /* A count of how many reclaimers are scanning this zone */ atomic_t reclaim_in_progress; + /* + * timestamp (in jiffies) of the last zone reclaim that did not + * result in freeing of pages. This is used to avoid repeated scans + * if all memory in the zone is in use. + */ + unsigned long last_unsuccessful_zone_reclaim; + /* * prev_priority holds the scanning priority for this zone. It is * defined as the scanning priority at which we achieved our reclaim diff --git a/include/linux/swap.h b/include/linux/swap.h index d01f7efb0f2..4a99e4a7fbf 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -176,6 +176,17 @@ extern int try_to_free_pages(struct zone **, gfp_t); extern int shrink_all_memory(int); extern int vm_swappiness; +#ifdef CONFIG_NUMA +extern int zone_reclaim_mode; +extern int zone_reclaim(struct zone *, gfp_t, unsigned int); +#else +#define zone_reclaim_mode 0 +static inline int zone_reclaim(struct zone *z, gfp_t mask, unsigned int order) +{ + return 0; +} +#endif + #ifdef CONFIG_MIGRATION extern int isolate_lru_page(struct page *p); extern int putback_lru_pages(struct list_head *l); diff --git a/include/linux/topology.h b/include/linux/topology.h index 315a5163d6a..e8eb0040ce3 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -56,6 +56,14 @@ #define REMOTE_DISTANCE 20 #define node_distance(from,to) ((from) == (to) ? LOCAL_DISTANCE : REMOTE_DISTANCE) #endif +#ifndef RECLAIM_DISTANCE +/* + * If the distance between nodes in a system is larger than RECLAIM_DISTANCE + * (in whatever arch specific measurement units returned by node_distance()) + * then switch on zone reclaim on boot. + */ +#define RECLAIM_DISTANCE 20 +#endif #ifndef PENALTY_FOR_NODE_WITH_CPUS #define PENALTY_FOR_NODE_WITH_CPUS (1) #endif diff --git a/mm/page_alloc.c b/mm/page_alloc.c index c2e29743a8d..df54e2fc8ee 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -878,7 +878,9 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, mark = (*z)->pages_high; if (!zone_watermark_ok(*z, order, mark, classzone_idx, alloc_flags)) - continue; + if (!zone_reclaim_mode || + !zone_reclaim(*z, gfp_mask, order)) + continue; } page = buffered_rmqueue(zonelist, *z, order, gfp_mask); @@ -1595,13 +1597,22 @@ static void __init build_zonelists(pg_data_t *pgdat) prev_node = local_node; nodes_clear(used_mask); while ((node = find_next_best_node(local_node, &used_mask)) >= 0) { + int distance = node_distance(local_node, node); + + /* + * If another node is sufficiently far away then it is better + * to reclaim pages in a zone before going off node. + */ + if (distance > RECLAIM_DISTANCE) + zone_reclaim_mode = 1; + /* * We don't want to pressure a particular node. * So adding penalty to the first node in same * distance group to make it round-robin. */ - if (node_distance(local_node, node) != - node_distance(local_node, prev_node)) + + if (distance != node_distance(local_node, prev_node)) node_load[node] += load; prev_node = node; load--; diff --git a/mm/vmscan.c b/mm/vmscan.c index e5117b6897a..2e34b61a70c 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1572,3 +1572,71 @@ static int __init kswapd_init(void) } module_init(kswapd_init) + +#ifdef CONFIG_NUMA +/* + * Zone reclaim mode + * + * If non-zero call zone_reclaim when the number of free pages falls below + * the watermarks. + * + * In the future we may add flags to the mode. However, the page allocator + * should only have to check that zone_reclaim_mode != 0 before calling + * zone_reclaim(). + */ +int zone_reclaim_mode __read_mostly; + +/* + * Mininum time between zone reclaim scans + */ +#define ZONE_RECLAIM_INTERVAL HZ/2 +/* + * Try to free up some pages from this zone through reclaim. + */ +int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) +{ + int nr_pages = 1 << order; + struct task_struct *p = current; + struct reclaim_state reclaim_state; + struct scan_control sc = { + .gfp_mask = gfp_mask, + .may_writepage = 0, + .may_swap = 0, + .nr_mapped = read_page_state(nr_mapped), + .nr_scanned = 0, + .nr_reclaimed = 0, + .priority = 0 + }; + + if (!(gfp_mask & __GFP_WAIT) || + zone->zone_pgdat->node_id != numa_node_id() || + zone->all_unreclaimable || + atomic_read(&zone->reclaim_in_progress) > 0) + return 0; + + if (time_before(jiffies, + zone->last_unsuccessful_zone_reclaim + ZONE_RECLAIM_INTERVAL)) + return 0; + + disable_swap_token(); + + if (nr_pages > SWAP_CLUSTER_MAX) + sc.swap_cluster_max = nr_pages; + else + sc.swap_cluster_max = SWAP_CLUSTER_MAX; + + cond_resched(); + p->flags |= PF_MEMALLOC; + reclaim_state.reclaimed_slab = 0; + p->reclaim_state = &reclaim_state; + shrink_zone(zone, &sc); + p->reclaim_state = NULL; + current->flags &= ~PF_MEMALLOC; + + if (sc.nr_reclaimed == 0) + zone->last_unsuccessful_zone_reclaim = jiffies; + + return sc.nr_reclaimed > nr_pages; +} +#endif + -- cgit v1.2.3 From 1743660b911bfb849b1fb33830522254561b9f9b Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 18 Jan 2006 17:42:32 -0800 Subject: [PATCH] Zone reclaim: proc override proc support for zone reclaim This patch creates a proc entry /proc/sys/vm/zone_reclaim_mode that may be used to override the automatic determination of the zone reclaim made on bootup. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/sysctl/vm.txt | 18 ++++++++++++++++++ include/linux/sysctl.h | 1 + kernel/sysctl.c | 11 +++++++++++ 3 files changed, 30 insertions(+) diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt index 6910c0136f8..391dd64363e 100644 --- a/Documentation/sysctl/vm.txt +++ b/Documentation/sysctl/vm.txt @@ -27,6 +27,7 @@ Currently, these files are in /proc/sys/vm: - laptop_mode - block_dump - drop-caches +- zone_reclaim_mode ============================================================== @@ -120,3 +121,20 @@ set to pcp->high/4. The upper limit of batch is (PAGE_SHIFT * 8) The initial value is zero. Kernel does not use this value at boot time to set the high water marks for each per cpu page list. + +=============================================================== + +zone_reclaim_mode: + +This is set during bootup to 1 if it is determined that pages from +remote zones will cause a significant performance reduction. The +page allocator will then reclaim easily reusable pages (those page +cache pages that are currently not used) before going off node. + +The user can override this setting. It may be beneficial to switch +off zone reclaim if the system is used for a file server and all +of memory should be used for caching files from disk. + +It may be beneficial to switch this on if one wants to do zone +reclaim regardless of the numa distances in the system. + diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 7f472127b7b..8352a7ce589 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -182,6 +182,7 @@ enum VM_SWAP_TOKEN_TIMEOUT=28, /* default time for token time out */ VM_DROP_PAGECACHE=29, /* int: nuke lots of pagecache */ VM_PERCPU_PAGELIST_FRACTION=30,/* int: fraction of pages in each percpu_pagelist */ + VM_ZONE_RECLAIM_MODE=31,/* reclaim local zone memory before going off node */ }; diff --git a/kernel/sysctl.c b/kernel/sysctl.c index f5d69b6e29f..cb99a42f8b3 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -869,6 +869,17 @@ static ctl_table vm_table[] = { .proc_handler = &proc_dointvec_jiffies, .strategy = &sysctl_jiffies, }, +#endif +#ifdef CONFIG_NUMA + { + .ctl_name = VM_ZONE_RECLAIM_MODE, + .procname = "zone_reclaim_mode", + .data = &zone_reclaim_mode, + .maxlen = sizeof(zone_reclaim_mode), + .mode = 0644, + .proc_handler = &proc_dointvec, + .strategy = &zero, + }, #endif { .ctl_name = 0 } }; -- cgit v1.2.3 From fc0abb1451c64c79ac80665d5ba74450ce274e4d Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 18 Jan 2006 17:42:33 -0800 Subject: [PATCH] sem2mutex: mm/slab.c Convert mm/swapfile.c's swapon_sem to swapon_mutex. Signed-off-by: Ingo Molnar Cc: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/slab.c | 46 +++++++++++++++++++++++----------------------- mm/swapfile.c | 17 +++++++++-------- 2 files changed, 32 insertions(+), 31 deletions(-) diff --git a/mm/slab.c b/mm/slab.c index 9374293a301..bd0317f1e06 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -68,7 +68,7 @@ * Further notes from the original documentation: * * 11 April '97. Started multi-threading - markhe - * The global cache-chain is protected by the semaphore 'cache_chain_sem'. + * The global cache-chain is protected by the mutex 'cache_chain_mutex'. * The sem is only needed when accessing/extending the cache-chain, which * can never happen inside an interrupt (kmem_cache_create(), * kmem_cache_shrink() and kmem_cache_reap()). @@ -103,6 +103,7 @@ #include #include #include +#include #include #include @@ -631,7 +632,7 @@ static kmem_cache_t cache_cache = { }; /* Guard access to the cache-chain. */ -static struct semaphore cache_chain_sem; +static DEFINE_MUTEX(cache_chain_mutex); static struct list_head cache_chain; /* @@ -857,7 +858,7 @@ static int __devinit cpuup_callback(struct notifier_block *nfb, switch (action) { case CPU_UP_PREPARE: - down(&cache_chain_sem); + mutex_lock(&cache_chain_mutex); /* we need to do this right in the beginning since * alloc_arraycache's are going to use this list. * kmalloc_node allows us to add the slab to the right @@ -912,7 +913,7 @@ static int __devinit cpuup_callback(struct notifier_block *nfb, l3->shared = nc; } } - up(&cache_chain_sem); + mutex_unlock(&cache_chain_mutex); break; case CPU_ONLINE: start_cpu_timer(cpu); @@ -921,7 +922,7 @@ static int __devinit cpuup_callback(struct notifier_block *nfb, case CPU_DEAD: /* fall thru */ case CPU_UP_CANCELED: - down(&cache_chain_sem); + mutex_lock(&cache_chain_mutex); list_for_each_entry(cachep, &cache_chain, next) { struct array_cache *nc; @@ -973,13 +974,13 @@ static int __devinit cpuup_callback(struct notifier_block *nfb, spin_unlock_irq(&cachep->spinlock); kfree(nc); } - up(&cache_chain_sem); + mutex_unlock(&cache_chain_mutex); break; #endif } return NOTIFY_OK; bad: - up(&cache_chain_sem); + mutex_unlock(&cache_chain_mutex); return NOTIFY_BAD; } @@ -1047,7 +1048,6 @@ void __init kmem_cache_init(void) */ /* 1) create the cache_cache */ - init_MUTEX(&cache_chain_sem); INIT_LIST_HEAD(&cache_chain); list_add(&cache_cache.next, &cache_chain); cache_cache.colour_off = cache_line_size(); @@ -1168,10 +1168,10 @@ void __init kmem_cache_init(void) /* 6) resize the head arrays to their final sizes */ { kmem_cache_t *cachep; - down(&cache_chain_sem); + mutex_lock(&cache_chain_mutex); list_for_each_entry(cachep, &cache_chain, next) enable_cpucache(cachep); - up(&cache_chain_sem); + mutex_unlock(&cache_chain_mutex); } /* Done! */ @@ -1590,7 +1590,7 @@ kmem_cache_create (const char *name, size_t size, size_t align, BUG(); } - down(&cache_chain_sem); + mutex_lock(&cache_chain_mutex); list_for_each(p, &cache_chain) { kmem_cache_t *pc = list_entry(p, kmem_cache_t, next); @@ -1856,7 +1856,7 @@ kmem_cache_create (const char *name, size_t size, size_t align, if (!cachep && (flags & SLAB_PANIC)) panic("kmem_cache_create(): failed to create slab `%s'\n", name); - up(&cache_chain_sem); + mutex_unlock(&cache_chain_mutex); return cachep; } EXPORT_SYMBOL(kmem_cache_create); @@ -2044,18 +2044,18 @@ int kmem_cache_destroy(kmem_cache_t *cachep) lock_cpu_hotplug(); /* Find the cache in the chain of caches. */ - down(&cache_chain_sem); + mutex_lock(&cache_chain_mutex); /* * the chain is never empty, cache_cache is never destroyed */ list_del(&cachep->next); - up(&cache_chain_sem); + mutex_unlock(&cache_chain_mutex); if (__cache_shrink(cachep)) { slab_error(cachep, "Can't free all objects"); - down(&cache_chain_sem); + mutex_lock(&cache_chain_mutex); list_add(&cachep->next, &cache_chain); - up(&cache_chain_sem); + mutex_unlock(&cache_chain_mutex); unlock_cpu_hotplug(); return 1; } @@ -3314,7 +3314,7 @@ static void drain_array_locked(kmem_cache_t *cachep, struct array_cache *ac, * - clear the per-cpu caches for this CPU. * - return freeable pages to the main free memory pool. * - * If we cannot acquire the cache chain semaphore then just give up - we'll + * If we cannot acquire the cache chain mutex then just give up - we'll * try again on the next iteration. */ static void cache_reap(void *unused) @@ -3322,7 +3322,7 @@ static void cache_reap(void *unused) struct list_head *walk; struct kmem_list3 *l3; - if (down_trylock(&cache_chain_sem)) { + if (!mutex_trylock(&cache_chain_mutex)) { /* Give up. Setup the next iteration. */ schedule_delayed_work(&__get_cpu_var(reap_work), REAPTIMEOUT_CPUC); @@ -3393,7 +3393,7 @@ static void cache_reap(void *unused) cond_resched(); } check_irq_on(); - up(&cache_chain_sem); + mutex_unlock(&cache_chain_mutex); drain_remote_pages(); /* Setup the next iteration */ schedule_delayed_work(&__get_cpu_var(reap_work), REAPTIMEOUT_CPUC); @@ -3429,7 +3429,7 @@ static void *s_start(struct seq_file *m, loff_t *pos) loff_t n = *pos; struct list_head *p; - down(&cache_chain_sem); + mutex_lock(&cache_chain_mutex); if (!n) print_slabinfo_header(m); p = cache_chain.next; @@ -3451,7 +3451,7 @@ static void *s_next(struct seq_file *m, void *p, loff_t *pos) static void s_stop(struct seq_file *m, void *p) { - up(&cache_chain_sem); + mutex_unlock(&cache_chain_mutex); } static int s_show(struct seq_file *m, void *p) @@ -3603,7 +3603,7 @@ ssize_t slabinfo_write(struct file *file, const char __user * buffer, return -EINVAL; /* Find the cache in the chain of caches. */ - down(&cache_chain_sem); + mutex_lock(&cache_chain_mutex); res = -EINVAL; list_for_each(p, &cache_chain) { kmem_cache_t *cachep = list_entry(p, kmem_cache_t, next); @@ -3620,7 +3620,7 @@ ssize_t slabinfo_write(struct file *file, const char __user * buffer, break; } } - up(&cache_chain_sem); + mutex_unlock(&cache_chain_mutex); if (res >= 0) res = count; return res; diff --git a/mm/swapfile.c b/mm/swapfile.c index 957fef43fa6..f1e69c30d20 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -25,6 +25,7 @@ #include #include #include +#include #include #include @@ -46,12 +47,12 @@ struct swap_list_t swap_list = {-1, -1}; struct swap_info_struct swap_info[MAX_SWAPFILES]; -static DECLARE_MUTEX(swapon_sem); +static DEFINE_MUTEX(swapon_mutex); /* * We need this because the bdev->unplug_fn can sleep and we cannot * hold swap_lock while calling the unplug_fn. And swap_lock - * cannot be turned into a semaphore. + * cannot be turned into a mutex. */ static DECLARE_RWSEM(swap_unplug_sem); @@ -1161,7 +1162,7 @@ asmlinkage long sys_swapoff(const char __user * specialfile) up_write(&swap_unplug_sem); destroy_swap_extents(p); - down(&swapon_sem); + mutex_lock(&swapon_mutex); spin_lock(&swap_lock); drain_mmlist(); @@ -1180,7 +1181,7 @@ asmlinkage long sys_swapoff(const char __user * specialfile) p->swap_map = NULL; p->flags = 0; spin_unlock(&swap_lock); - up(&swapon_sem); + mutex_unlock(&swapon_mutex); vfree(swap_map); inode = mapping->host; if (S_ISBLK(inode->i_mode)) { @@ -1209,7 +1210,7 @@ static void *swap_start(struct seq_file *swap, loff_t *pos) int i; loff_t l = *pos; - down(&swapon_sem); + mutex_lock(&swapon_mutex); for (i = 0; i < nr_swapfiles; i++, ptr++) { if (!(ptr->flags & SWP_USED) || !ptr->swap_map) @@ -1238,7 +1239,7 @@ static void *swap_next(struct seq_file *swap, void *v, loff_t *pos) static void swap_stop(struct seq_file *swap, void *v) { - up(&swapon_sem); + mutex_unlock(&swapon_mutex); } static int swap_show(struct seq_file *swap, void *v) @@ -1540,7 +1541,7 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) goto bad_swap; } - down(&swapon_sem); + mutex_lock(&swapon_mutex); spin_lock(&swap_lock); p->flags = SWP_ACTIVE; nr_swap_pages += nr_good_pages; @@ -1566,7 +1567,7 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) swap_info[prev].next = p - swap_info; } spin_unlock(&swap_lock); - up(&swapon_sem); + mutex_unlock(&swapon_mutex); error = 0; goto out; bad_swap: -- cgit v1.2.3 From dc85da15d42b0efc792b0f5eab774dc5dbc1ceec Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 18 Jan 2006 17:42:36 -0800 Subject: [PATCH] NUMA policies in the slab allocator V2 This patch fixes a regression in 2.6.14 against 2.6.13 that causes an imbalance in memory allocation during bootup. The slab allocator in 2.6.13 is not numa aware and simply calls alloc_pages(). This means that memory policies may control the behavior of alloc_pages(). During bootup the memory policy is set to MPOL_INTERLEAVE resulting in the spreading out of allocations during bootup over all available nodes. The slab allocator in 2.6.13 has only a single list of slab pages. As a result the per cpu slab cache and the spinlock controlled page lists may contain slab entries from off node memory. The slab allocator in 2.6.13 makes no effort to discern the locality of an entry on its lists. The NUMA aware slab allocator in 2.6.14 controls locality of the slab pages explicitly by calling alloc_pages_node(). The NUMA slab allocator manages slab entries by having lists of available slab pages for each node. The per cpu slab cache can only contain slab entries associated with the node local to the processor. This guarantees that the default allocation mode of the slab allocator always assigns local memory if available. Setting MPOL_INTERLEAVE as a default policy during bootup has no effect anymore. In 2.6.14 all node unspecific slab allocations are performed on the boot processor. This means that most of key data structures are allocated on one node. Most processors will have to refer to these structures making the boot node a potential bottleneck. This may reduce performance and cause unnecessary memory pressure on the boot node. This patch implements NUMA policies in the slab layer. There is the need of explicit application of NUMA memory policies by the slab allcator itself since the NUMA slab allocator does no longer let the page_allocator control locality. The check for policies is made directly at the beginning of __cache_alloc using current->mempolicy. The memory policy is already frequently checked by the page allocator (alloc_page_vma() and alloc_page_current()). So it is highly likely that the cacheline is present. For MPOL_INTERLEAVE kmalloc() will spread out each request to one node after another so that an equal distribution of allocations can be obtained during bootup. It is not possible to push the policy check to lower layers of the NUMA slab allocator since the per cpu caches are now only containing slab entries from the current node. If the policy says that the local node is not to be preferred or forbidden then there is no point in checking the slab cache or local list of slab pages. The allocation better be directed immediately to the lists containing slab entries for the allowed set of nodes. This way of applying policy also fixes another strange behavior in 2.6.13. alloc_pages() is controlled by the memory allocation policy of the current process. It could therefore be that one process is running with MPOL_INTERLEAVE and would f.e. obtain a new page following that policy since no slab entries are in the lists anymore. A page can typically be used for multiple slab entries but lets say that the current process is only using one. The other entries are then added to the slab lists. These are now non local entries in the slab lists despite of the possible availability of local pages that would provide faster access and increase the performance of the application. Another process without MPOL_INTERLEAVE may now run and expect a local slab entry from kmalloc(). However, there are still these free slab entries from the off node page obtained from the other process via MPOL_INTERLEAVE in the cache. The process will then get an off node slab entry although other slab entries may be available that are local to that process. This means that the policy if one process may contaminate the locality of the slab caches for other processes. This patch in effect insures that a per process policy is followed for the allocation of slab entries and that there cannot be a memory policy influence from one process to another. A process with default policy will always get a local slab entry if one is available. And the process using memory policies will get its memory arranged as requested. Off-node slab allocation will require the use of spinlocks and will make the use of per cpu caches not possible. A process using memory policies to redirect allocations offnode will have to cope with additional lock overhead in addition to the latency added by the need to access a remote slab entry. Changes V1->V2 - Remove #ifdef CONFIG_NUMA by moving forward declaration into prior #ifdef CONFIG_NUMA section. - Give the function determining the node number to use a saner name. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mempolicy.h | 1 + mm/mempolicy.c | 30 ++++++++++++++++++++++++++++++ mm/slab.c | 12 ++++++++++++ 3 files changed, 43 insertions(+) diff --git a/include/linux/mempolicy.h b/include/linux/mempolicy.h index d6a53ed6ab6..bbd2221923c 100644 --- a/include/linux/mempolicy.h +++ b/include/linux/mempolicy.h @@ -159,6 +159,7 @@ extern void mpol_rebind_mm(struct mm_struct *mm, nodemask_t *new); extern struct mempolicy default_policy; extern struct zonelist *huge_zonelist(struct vm_area_struct *vma, unsigned long addr); +extern unsigned slab_node(struct mempolicy *policy); extern int policy_zone; diff --git a/mm/mempolicy.c b/mm/mempolicy.c index a683a66599b..71430d44082 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -976,6 +976,36 @@ static unsigned interleave_nodes(struct mempolicy *policy) return nid; } +/* + * Depending on the memory policy provide a node from which to allocate the + * next slab entry. + */ +unsigned slab_node(struct mempolicy *policy) +{ + if (in_interrupt()) + return numa_node_id(); + + switch (policy->policy) { + case MPOL_INTERLEAVE: + return interleave_nodes(policy); + + case MPOL_BIND: + /* + * Follow bind policy behavior and start allocation at the + * first node. + */ + return policy->v.zonelist->zones[0]->zone_pgdat->node_id; + + case MPOL_PREFERRED: + if (policy->v.preferred_node >= 0) + return policy->v.preferred_node; + /* Fall through */ + + default: + return numa_node_id(); + } +} + /* Do static interleaving for a VMA with known offset. */ static unsigned offset_il_node(struct mempolicy *pol, struct vm_area_struct *vma, unsigned long off) diff --git a/mm/slab.c b/mm/slab.c index bd0317f1e06..9025608696e 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -103,6 +103,7 @@ #include #include #include +#include #include #include @@ -773,6 +774,8 @@ static struct array_cache *alloc_arraycache(int node, int entries, } #ifdef CONFIG_NUMA +static void *__cache_alloc_node(kmem_cache_t *, gfp_t, int); + static inline struct array_cache **alloc_alien_cache(int node, int limit) { struct array_cache **ac_ptr; @@ -2570,6 +2573,15 @@ static inline void *____cache_alloc(kmem_cache_t *cachep, gfp_t flags) void *objp; struct array_cache *ac; +#ifdef CONFIG_NUMA + if (current->mempolicy) { + int nid = slab_node(current->mempolicy); + + if (nid != numa_node_id()) + return __cache_alloc_node(cachep, flags, nid); + } +#endif + check_irq_off(); ac = ac_data(cachep); if (likely(ac->avail)) { -- cgit v1.2.3 From 86c562a9d6683063e071692fe14e0a18e64ee1be Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 18 Jan 2006 17:42:37 -0800 Subject: [PATCH] mm: optimize numa policy handling in slab allocator Move the interrupt check from slab_node into ___cache_alloc and adds an "unlikely()" to avoid pipeline stalls on some architectures. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/mempolicy.c | 3 --- mm/slab.c | 2 +- 2 files changed, 1 insertion(+), 4 deletions(-) diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 71430d44082..73790188b0e 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -982,9 +982,6 @@ static unsigned interleave_nodes(struct mempolicy *policy) */ unsigned slab_node(struct mempolicy *policy) { - if (in_interrupt()) - return numa_node_id(); - switch (policy->policy) { case MPOL_INTERLEAVE: return interleave_nodes(policy); diff --git a/mm/slab.c b/mm/slab.c index 9025608696e..6f8495e2185 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -2574,7 +2574,7 @@ static inline void *____cache_alloc(kmem_cache_t *cachep, gfp_t flags) struct array_cache *ac; #ifdef CONFIG_NUMA - if (current->mempolicy) { + if (unlikely(current->mempolicy && !in_interrupt())) { int nid = slab_node(current->mempolicy); if (nid != numa_node_id()) -- cgit v1.2.3 From ea1eae75eb596e0628dc5e01d32c78b1f6b257fb Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Wed, 18 Jan 2006 17:42:38 -0800 Subject: [PATCH] uml: add __raw_writel definition Add implementations of the write* and __raw_write* functions. __raw_writel is needed by lib/iocopy.c, which shouldn't be used in UML, but which is unconditionally linked in anyway. Signed-off-by: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-um/io.h | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/include/asm-um/io.h b/include/asm-um/io.h index 90674056dce..1934d9340e2 100644 --- a/include/asm-um/io.h +++ b/include/asm-um/io.h @@ -33,4 +33,20 @@ static inline void * phys_to_virt(unsigned long address) */ #define xlate_dev_kmem_ptr(p) p +static inline void writeb(unsigned char b, volatile void __iomem *addr) +{ + *(volatile unsigned char __force *) addr = b; +} +static inline void writew(unsigned short b, volatile void __iomem *addr) +{ + *(volatile unsigned short __force *) addr = b; +} +static inline void writel(unsigned int b, volatile void __iomem *addr) +{ + *(volatile unsigned int __force *) addr = b; +} +#define __raw_writeb writeb +#define __raw_writew writew +#define __raw_writel writel + #endif -- cgit v1.2.3 From 12919aa6e015dd85170fc3b1a3e10a5dfd116c72 Mon Sep 17 00:00:00 2001 From: Bodo Stroesser Date: Wed, 18 Jan 2006 17:42:39 -0800 Subject: [PATCH] uml: move LDT creation s390 doesn't have a LDT. So MM_COPY_SEGMENTS will not be supported on s390. The only user of MM_COPY_SEGMENTS is new_mm(), but that's no longer useful, as arch/sys-i386/ldt.c defines init_new_ldt(), which is called immediately after new_mm(). So we should copy host's LDT in init_new_ldt(), if /proc/mm is available, to have this subarch specific call in subarch code. Signed-off-by: Bodo Stroesser Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/kernel/skas/include/skas.h | 2 +- arch/um/kernel/skas/mmu.c | 8 ++------ arch/um/kernel/skas/process_kern.c | 16 ++-------------- arch/um/sys-i386/ldt.c | 26 ++++++++++++++++++++++++-- 4 files changed, 29 insertions(+), 23 deletions(-) diff --git a/arch/um/kernel/skas/include/skas.h b/arch/um/kernel/skas/include/skas.h index 01d489de398..a7f7cd556f1 100644 --- a/arch/um/kernel/skas/include/skas.h +++ b/arch/um/kernel/skas/include/skas.h @@ -33,7 +33,7 @@ extern int protect(struct mm_id * mm_idp, unsigned long addr, unsigned long len, int r, int w, int x, int done, void **data); extern void user_signal(int sig, union uml_pt_regs *regs, int pid); -extern int new_mm(int from, unsigned long stack); +extern int new_mm(unsigned long stack); extern int start_userspace(unsigned long stub_stack); extern int copy_context_skas0(unsigned long stack, int pid); extern void get_skas_faultinfo(int pid, struct faultinfo * fi); diff --git a/arch/um/kernel/skas/mmu.c b/arch/um/kernel/skas/mmu.c index 677871f1b37..c5c9885a829 100644 --- a/arch/um/kernel/skas/mmu.c +++ b/arch/um/kernel/skas/mmu.c @@ -78,7 +78,7 @@ int init_new_context_skas(struct task_struct *task, struct mm_struct *mm) struct mmu_context_skas *from_mm = NULL; struct mmu_context_skas *to_mm = &mm->context.skas; unsigned long stack = 0; - int from_fd, ret = -ENOMEM; + int ret = -ENOMEM; if(skas_needs_stub){ stack = get_zeroed_page(GFP_KERNEL); @@ -108,11 +108,7 @@ int init_new_context_skas(struct task_struct *task, struct mm_struct *mm) from_mm = ¤t->mm->context.skas; if(proc_mm){ - if(from_mm) - from_fd = from_mm->id.u.mm_fd; - else from_fd = -1; - - ret = new_mm(from_fd, stack); + ret = new_mm(stack); if(ret < 0){ printk("init_new_context_skas - new_mm failed, " "errno = %d\n", ret); diff --git a/arch/um/kernel/skas/process_kern.c b/arch/um/kernel/skas/process_kern.c index dc41c6dc2f3..fcddee76c7c 100644 --- a/arch/um/kernel/skas/process_kern.c +++ b/arch/um/kernel/skas/process_kern.c @@ -20,7 +20,6 @@ #include "tlb.h" #include "kern.h" #include "mode.h" -#include "proc_mm.h" #include "registers.h" void switch_to_skas(void *prev, void *next) @@ -125,25 +124,14 @@ int copy_thread_skas(int nr, unsigned long clone_flags, unsigned long sp, extern void map_stub_pages(int fd, unsigned long code, unsigned long data, unsigned long stack); -int new_mm(int from, unsigned long stack) +int new_mm(unsigned long stack) { - struct proc_mm_op copy; - int n, fd; + int fd; fd = os_open_file("/proc/mm", of_cloexec(of_write(OPENFLAGS())), 0); if(fd < 0) return(fd); - if(from != -1){ - copy = ((struct proc_mm_op) { .op = MM_COPY_SEGMENTS, - .u = - { .copy_segments = from } } ); - n = os_write_file(fd, ©, sizeof(copy)); - if(n != sizeof(copy)) - printk("new_mm : /proc/mm copy_segments failed, " - "err = %d\n", -n); - } - if(skas_needs_stub) map_stub_pages(fd, CONFIG_STUB_CODE, CONFIG_STUB_DATA, stack); diff --git a/arch/um/sys-i386/ldt.c b/arch/um/sys-i386/ldt.c index 17746b4c08f..0cdfd4481d5 100644 --- a/arch/um/sys-i386/ldt.c +++ b/arch/um/sys-i386/ldt.c @@ -16,6 +16,8 @@ #include "choose-mode.h" #include "kern.h" #include "mode_kern.h" +#include "proc_mm.h" +#include "os.h" extern int modify_ldt(int func, void *ptr, unsigned long bytecount); @@ -456,13 +458,14 @@ long init_new_ldt(struct mmu_context_skas * new_mm, int i; long page, err=0; void *addr = NULL; + struct proc_mm_op copy; - memset(&desc, 0, sizeof(desc)); if(!ptrace_ldt) init_MUTEX(&new_mm->ldt.semaphore); if(!from_mm){ + memset(&desc, 0, sizeof(desc)); /* * We have to initialize a clean ldt. */ @@ -494,8 +497,26 @@ long init_new_ldt(struct mmu_context_skas * new_mm, } } new_mm->ldt.entry_count = 0; + + goto out; } - else if (!ptrace_ldt) { + + if(proc_mm){ + /* We have a valid from_mm, so we now have to copy the LDT of + * from_mm to new_mm, because using proc_mm an new mm with + * an empty/default LDT was created in new_mm() + */ + copy = ((struct proc_mm_op) { .op = MM_COPY_SEGMENTS, + .u = + { .copy_segments = + from_mm->id.u.mm_fd } } ); + i = os_write_file(new_mm->id.u.mm_fd, ©, sizeof(copy)); + if(i != sizeof(copy)) + printk("new_mm : /proc/mm copy_segments failed, " + "err = %d\n", -i); + } + + if(!ptrace_ldt) { /* Our local LDT is used to supply the data for * modify_ldt(READLDT), if PTRACE_LDT isn't available, * i.e., we have to use the stub for modify_ldt, which @@ -524,6 +545,7 @@ long init_new_ldt(struct mmu_context_skas * new_mm, up(&from_mm->ldt.semaphore); } + out: return err; } -- cgit v1.2.3 From 4fef0c10fa174b57a10854b8b4b2b90d155706e0 Mon Sep 17 00:00:00 2001 From: Gennady Sharapov Date: Wed, 18 Jan 2006 17:42:41 -0800 Subject: [PATCH] uml: move libc-dependent utility procedures The serial UML OS-abstraction layer patch (um/kernel dir). This moves all systemcalls from user_util.c file under os-Linux dir Signed-off-by: Gennady Sharapov Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/fd.c | 1 + arch/um/include/os.h | 10 +++ arch/um/include/user_util.h | 18 ---- arch/um/kernel/Makefile | 5 +- arch/um/kernel/skas/uaccess.c | 2 +- arch/um/kernel/tt/gdb.c | 1 + arch/um/kernel/tt/ptproxy/ptrace.c | 1 + arch/um/kernel/tt/ptproxy/sysdep.c | 1 + arch/um/kernel/user_util.c | 174 ------------------------------------- arch/um/os-Linux/Makefile | 4 +- arch/um/os-Linux/tt.c | 48 ++++++++++ arch/um/os-Linux/util.c | 116 +++++++++++++++++++++++++ 12 files changed, 183 insertions(+), 198 deletions(-) delete mode 100644 arch/um/kernel/user_util.c create mode 100644 arch/um/os-Linux/util.c diff --git a/arch/um/drivers/fd.c b/arch/um/drivers/fd.c index 3296e86a03a..c41f75e4acb 100644 --- a/arch/um/drivers/fd.c +++ b/arch/um/drivers/fd.c @@ -11,6 +11,7 @@ #include "user.h" #include "user_util.h" #include "chan_user.h" +#include "os.h" struct fd_chan { int fd; diff --git a/arch/um/include/os.h b/arch/um/include/os.h index dd72d66cf0e..7f27dfe0eca 100644 --- a/arch/um/include/os.h +++ b/arch/um/include/os.h @@ -195,6 +195,8 @@ extern unsigned long long os_usecs(void); /* tt.c * for tt mode only (will be deleted in future...) */ +extern void stop(void); +extern int wait_for_stop(int pid, int sig, int cont_type, void *relay); extern int protect_memory(unsigned long addr, unsigned long len, int r, int w, int x, int must_succeed); extern void forward_pending_sigio(int target); @@ -235,4 +237,12 @@ extern int set_signals(int enable); extern void os_fill_handlinfo(struct kern_handlers h); extern void do_longjmp(void *p, int val); +/* util.c */ +extern void stack_protections(unsigned long address); +extern void task_protections(unsigned long address); +extern int raw(int fd); +extern void setup_machinename(char *machine_out); +extern void setup_hostinfo(void); +extern int setjmp_wrapper(void (*proc)(void *, void *), ...); + #endif diff --git a/arch/um/include/user_util.h b/arch/um/include/user_util.h index c1dbd77b073..a6f1f176cf8 100644 --- a/arch/um/include/user_util.h +++ b/arch/um/include/user_util.h @@ -44,10 +44,6 @@ extern unsigned long brk_start; extern int pty_output_sigio; extern int pty_close_sigio; -extern void stop(void); -extern void stack_protections(unsigned long address); -extern void task_protections(unsigned long address); -extern int wait_for_stop(int pid, int sig, int cont_type, void *relay); extern void *add_signal_handler(int sig, void (*handler)(int)); extern int linux_main(int argc, char **argv); extern void set_cmdline(char *cmd); @@ -55,8 +51,6 @@ extern void input_cb(void (*proc)(void *), void *arg, int arg_len); extern int get_pty(void); extern void *um_kmalloc(int size); extern int switcheroo(int fd, int prot, void *from, void *to, int size); -extern void setup_machinename(char *machine_out); -extern void setup_hostinfo(void); extern void do_exec(int old_pid, int new_pid); extern void tracer_panic(char *msg, ...); extern int detach(int pid, int sig); @@ -70,18 +64,6 @@ extern int cpu_feature(char *what, char *buf, int len); extern int arch_handle_signal(int sig, union uml_pt_regs *regs); extern int arch_fixup(unsigned long address, void *sc_ptr); extern void arch_init_thread(void); -extern int setjmp_wrapper(void (*proc)(void *, void *), ...); extern int raw(int fd); #endif - -/* - * 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/arch/um/kernel/Makefile b/arch/um/kernel/Makefile index 193cc2b7448..165c8639c9f 100644 --- a/arch/um/kernel/Makefile +++ b/arch/um/kernel/Makefile @@ -10,8 +10,7 @@ obj-y = config.o exec_kern.o exitcode.o \ init_task.o irq.o irq_user.o ksyms.o mem.o physmem.o \ process_kern.o ptrace.o reboot.o resource.o sigio_user.o sigio_kern.o \ signal_kern.o smp.o syscall_kern.o sysrq.o time.o \ - time_kern.o tlb.o trap_kern.o uaccess.o um_arch.o umid.o \ - user_util.o + time_kern.o tlb.o trap_kern.o uaccess.o um_arch.o umid.o obj-$(CONFIG_BLK_DEV_INITRD) += initrd.o obj-$(CONFIG_GPROF) += gprof_syms.o @@ -24,7 +23,7 @@ obj-$(CONFIG_MODE_SKAS) += skas/ user-objs-$(CONFIG_TTY_LOG) += tty_log.o -USER_OBJS := $(user-objs-y) config.o time.o tty_log.o user_util.o +USER_OBJS := $(user-objs-y) config.o time.o tty_log.o include arch/um/scripts/Makefile.rules diff --git a/arch/um/kernel/skas/uaccess.c b/arch/um/kernel/skas/uaccess.c index a5a47528dec..5992c325716 100644 --- a/arch/um/kernel/skas/uaccess.c +++ b/arch/um/kernel/skas/uaccess.c @@ -13,7 +13,7 @@ #include "asm/pgtable.h" #include "asm/uaccess.h" #include "kern_util.h" -#include "user_util.h" +#include "os.h" extern void *um_virt_to_phys(struct task_struct *task, unsigned long addr, pte_t *pte_out); diff --git a/arch/um/kernel/tt/gdb.c b/arch/um/kernel/tt/gdb.c index 37e22d71a0d..786e4edd86c 100644 --- a/arch/um/kernel/tt/gdb.c +++ b/arch/um/kernel/tt/gdb.c @@ -20,6 +20,7 @@ #include "user_util.h" #include "tt.h" #include "sysdep/thread.h" +#include "os.h" extern int debugger_pid; extern int debugger_fd; diff --git a/arch/um/kernel/tt/ptproxy/ptrace.c b/arch/um/kernel/tt/ptproxy/ptrace.c index 528a5fc8d88..03774427d46 100644 --- a/arch/um/kernel/tt/ptproxy/ptrace.c +++ b/arch/um/kernel/tt/ptproxy/ptrace.c @@ -20,6 +20,7 @@ Jeff Dike (jdike@karaya.com) : Modified for integration into uml #include "kern_util.h" #include "ptrace_user.h" #include "tt.h" +#include "os.h" long proxy_ptrace(struct debugger *debugger, int arg1, pid_t arg2, long arg3, long arg4, pid_t child, int *ret) diff --git a/arch/um/kernel/tt/ptproxy/sysdep.c b/arch/um/kernel/tt/ptproxy/sysdep.c index a5f0e01e214..99f178319d0 100644 --- a/arch/um/kernel/tt/ptproxy/sysdep.c +++ b/arch/um/kernel/tt/ptproxy/sysdep.c @@ -15,6 +15,7 @@ terms and conditions. #include "ptrace_user.h" #include "user_util.h" #include "user.h" +#include "os.h" int get_syscall(pid_t pid, long *arg1, long *arg2, long *arg3, long *arg4, long *arg5) diff --git a/arch/um/kernel/user_util.c b/arch/um/kernel/user_util.c deleted file mode 100644 index 4c231161f25..00000000000 --- a/arch/um/kernel/user_util.c +++ /dev/null @@ -1,174 +0,0 @@ -/* - * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "asm/types.h" -#include -#include -#include -#include -#include -#include -#include -#include -#include "user_util.h" -#include "kern_util.h" -#include "user.h" -#include "mem_user.h" -#include "init.h" -#include "ptrace_user.h" -#include "uml-config.h" - -void stop(void) -{ - while(1) sleep(1000000); -} - -void stack_protections(unsigned long address) -{ - int prot = PROT_READ | PROT_WRITE | PROT_EXEC; - - if(mprotect((void *) address, page_size(), prot) < 0) - panic("protecting stack failed, errno = %d", errno); -} - -void task_protections(unsigned long address) -{ - unsigned long guard = address + page_size(); - unsigned long stack = guard + page_size(); - int prot = 0, pages; - -#ifdef notdef - if(mprotect((void *) stack, page_size(), prot) < 0) - panic("protecting guard page failed, errno = %d", errno); -#endif - pages = (1 << UML_CONFIG_KERNEL_STACK_ORDER) - 2; - prot = PROT_READ | PROT_WRITE | PROT_EXEC; - if(mprotect((void *) stack, pages * page_size(), prot) < 0) - panic("protecting stack failed, errno = %d", errno); -} - -int wait_for_stop(int pid, int sig, int cont_type, void *relay) -{ - sigset_t *relay_signals = relay; - int status, ret; - - while(1){ - CATCH_EINTR(ret = waitpid(pid, &status, WUNTRACED)); - if((ret < 0) || - !WIFSTOPPED(status) || (WSTOPSIG(status) != sig)){ - if(ret < 0){ - printk("wait failed, errno = %d\n", - errno); - } - else if(WIFEXITED(status)) - printk("process %d exited with status %d\n", - pid, WEXITSTATUS(status)); - else if(WIFSIGNALED(status)) - printk("process %d exited with signal %d\n", - pid, WTERMSIG(status)); - else if((WSTOPSIG(status) == SIGVTALRM) || - (WSTOPSIG(status) == SIGALRM) || - (WSTOPSIG(status) == SIGIO) || - (WSTOPSIG(status) == SIGPROF) || - (WSTOPSIG(status) == SIGCHLD) || - (WSTOPSIG(status) == SIGWINCH) || - (WSTOPSIG(status) == SIGINT)){ - ptrace(cont_type, pid, 0, WSTOPSIG(status)); - continue; - } - else if((relay_signals != NULL) && - sigismember(relay_signals, WSTOPSIG(status))){ - ptrace(cont_type, pid, 0, WSTOPSIG(status)); - continue; - } - else printk("process %d stopped with signal %d\n", - pid, WSTOPSIG(status)); - panic("wait_for_stop failed to wait for %d to stop " - "with %d\n", pid, sig); - } - return(status); - } -} - -int raw(int fd) -{ - struct termios tt; - int err; - - CATCH_EINTR(err = tcgetattr(fd, &tt)); - if(err < 0) - return -errno; - - cfmakeraw(&tt); - - CATCH_EINTR(err = tcsetattr(fd, TCSADRAIN, &tt)); - if(err < 0) - return -errno; - - /* XXX tcsetattr could have applied only some changes - * (and cfmakeraw() is a set of changes) */ - return(0); -} - -void setup_machinename(char *machine_out) -{ - struct utsname host; - - uname(&host); -#if defined(UML_CONFIG_UML_X86) && !defined(UML_CONFIG_64BIT) - if (!strcmp(host.machine, "x86_64")) { - strcpy(machine_out, "i686"); - return; - } -#endif - strcpy(machine_out, host.machine); -} - -char host_info[(_UTSNAME_LENGTH + 1) * 4 + _UTSNAME_NODENAME_LENGTH + 1]; - -void setup_hostinfo(void) -{ - struct utsname host; - - uname(&host); - sprintf(host_info, "%s %s %s %s %s", host.sysname, host.nodename, - host.release, host.version, host.machine); -} - -int setjmp_wrapper(void (*proc)(void *, void *), ...) -{ - va_list args; - sigjmp_buf buf; - int n; - - n = sigsetjmp(buf, 1); - if(n == 0){ - va_start(args, proc); - (*proc)(&buf, &args); - } - va_end(args); - return(n); -} - -/* - * 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/arch/um/os-Linux/Makefile b/arch/um/os-Linux/Makefile index 40c7d6b1df6..08a4e628b24 100644 --- a/arch/um/os-Linux/Makefile +++ b/arch/um/os-Linux/Makefile @@ -5,12 +5,12 @@ obj-y = aio.o elf_aux.o file.o helper.o main.o mem.o process.o signal.o \ start_up.o time.o trap.o tt.o tty.o uaccess.o umid.o user_syms.o \ - drivers/ sys-$(SUBARCH)/ + util.o drivers/ sys-$(SUBARCH)/ obj-$(CONFIG_MODE_SKAS) += skas/ USER_OBJS := aio.o elf_aux.o file.o helper.o main.o mem.o process.o signal.o \ - start_up.o time.o trap.o tt.o tty.o uaccess.o umid.o + start_up.o time.o trap.o tt.o tty.o uaccess.o umid.o util.o elf_aux.o: $(ARCH_DIR)/kernel-offsets.h CFLAGS_elf_aux.o += -I$(objtree)/arch/um diff --git a/arch/um/os-Linux/tt.c b/arch/um/os-Linux/tt.c index cb2648b79d0..404bb63a74a 100644 --- a/arch/um/os-Linux/tt.c +++ b/arch/um/os-Linux/tt.c @@ -63,6 +63,54 @@ void kill_child_dead(int pid) } while(1); } +void stop(void) +{ + while(1) sleep(1000000); +} + +int wait_for_stop(int pid, int sig, int cont_type, void *relay) +{ + sigset_t *relay_signals = relay; + int status, ret; + + while(1){ + CATCH_EINTR(ret = waitpid(pid, &status, WUNTRACED)); + if((ret < 0) || + !WIFSTOPPED(status) || (WSTOPSIG(status) != sig)){ + if(ret < 0){ + printk("wait failed, errno = %d\n", + errno); + } + else if(WIFEXITED(status)) + printk("process %d exited with status %d\n", + pid, WEXITSTATUS(status)); + else if(WIFSIGNALED(status)) + printk("process %d exited with signal %d\n", + pid, WTERMSIG(status)); + else if((WSTOPSIG(status) == SIGVTALRM) || + (WSTOPSIG(status) == SIGALRM) || + (WSTOPSIG(status) == SIGIO) || + (WSTOPSIG(status) == SIGPROF) || + (WSTOPSIG(status) == SIGCHLD) || + (WSTOPSIG(status) == SIGWINCH) || + (WSTOPSIG(status) == SIGINT)){ + ptrace(cont_type, pid, 0, WSTOPSIG(status)); + continue; + } + else if((relay_signals != NULL) && + sigismember(relay_signals, WSTOPSIG(status))){ + ptrace(cont_type, pid, 0, WSTOPSIG(status)); + continue; + } + else printk("process %d stopped with signal %d\n", + pid, WSTOPSIG(status)); + panic("wait_for_stop failed to wait for %d to stop " + "with %d\n", pid, sig); + } + return(status); + } +} + /* *------------------------- * only for tt mode (will be deleted in future...) diff --git a/arch/um/os-Linux/util.c b/arch/um/os-Linux/util.c new file mode 100644 index 00000000000..d224434d561 --- /dev/null +++ b/arch/um/os-Linux/util.c @@ -0,0 +1,116 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "asm/types.h" +#include +#include +#include +#include +#include +#include +#include +#include +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "mem_user.h" +#include "init.h" +#include "ptrace_user.h" +#include "uml-config.h" +#include "os.h" + +void stack_protections(unsigned long address) +{ + int prot = PROT_READ | PROT_WRITE | PROT_EXEC; + + if(mprotect((void *) address, page_size(), prot) < 0) + panic("protecting stack failed, errno = %d", errno); +} + +void task_protections(unsigned long address) +{ + unsigned long guard = address + page_size(); + unsigned long stack = guard + page_size(); + int prot = 0, pages; + +#ifdef notdef + if(mprotect((void *) stack, page_size(), prot) < 0) + panic("protecting guard page failed, errno = %d", errno); +#endif + pages = (1 << UML_CONFIG_KERNEL_STACK_ORDER) - 2; + prot = PROT_READ | PROT_WRITE | PROT_EXEC; + if(mprotect((void *) stack, pages * page_size(), prot) < 0) + panic("protecting stack failed, errno = %d", errno); +} + +int raw(int fd) +{ + struct termios tt; + int err; + + CATCH_EINTR(err = tcgetattr(fd, &tt)); + if(err < 0) + return -errno; + + cfmakeraw(&tt); + + CATCH_EINTR(err = tcsetattr(fd, TCSADRAIN, &tt)); + if(err < 0) + return -errno; + + /* XXX tcsetattr could have applied only some changes + * (and cfmakeraw() is a set of changes) */ + return(0); +} + +void setup_machinename(char *machine_out) +{ + struct utsname host; + + uname(&host); +#if defined(UML_CONFIG_UML_X86) && !defined(UML_CONFIG_64BIT) + if (!strcmp(host.machine, "x86_64")) { + strcpy(machine_out, "i686"); + return; + } +#endif + strcpy(machine_out, host.machine); +} + +char host_info[(_UTSNAME_LENGTH + 1) * 4 + _UTSNAME_NODENAME_LENGTH + 1]; + +void setup_hostinfo(void) +{ + struct utsname host; + + uname(&host); + sprintf(host_info, "%s %s %s %s %s", host.sysname, host.nodename, + host.release, host.version, host.machine); +} + +int setjmp_wrapper(void (*proc)(void *, void *), ...) +{ + va_list args; + sigjmp_buf buf; + int n; + + n = sigsetjmp(buf, 1); + if(n == 0){ + va_start(args, proc); + (*proc)(&buf, &args); + } + va_end(args); + return(n); +} -- cgit v1.2.3 From cff65c4f0ea6662124bbb7bf3806e5df1c6d735d Mon Sep 17 00:00:00 2001 From: Gennady Sharapov Date: Wed, 18 Jan 2006 17:42:42 -0800 Subject: [PATCH] uml: move libc-dependent time code The serial UML OS-abstraction layer patch (um/kernel dir). This moves all systemcalls from time.c file under os-Linux dir and joins time.c and tine_kernel.c files Signed-off-by: Gennady Sharapov Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/include/os.h | 12 +++- arch/um/include/time_user.h | 19 ------ arch/um/kernel/Makefile | 4 +- arch/um/kernel/exec_kern.c | 1 - arch/um/kernel/process_kern.c | 1 - arch/um/kernel/skas/process.c | 1 - arch/um/kernel/skas/process_kern.c | 1 - arch/um/kernel/syscall.c | 4 +- arch/um/kernel/time_kern.c | 124 +++++++++++++++++++++-------------- arch/um/kernel/tt/exec_kern.c | 1 - arch/um/kernel/tt/process_kern.c | 1 - arch/um/os-Linux/main.c | 1 - arch/um/os-Linux/signal.c | 2 +- arch/um/os-Linux/start_up.c | 1 - arch/um/os-Linux/time.c | 131 +++++++++++++++++++++++++++++++++---- arch/um/os-Linux/tt.c | 1 - 16 files changed, 211 insertions(+), 94 deletions(-) delete mode 100644 arch/um/include/time_user.h diff --git a/arch/um/include/os.h b/arch/um/include/os.h index 7f27dfe0eca..624938ad9e1 100644 --- a/arch/um/include/os.h +++ b/arch/um/include/os.h @@ -190,7 +190,6 @@ extern int os_protect_memory(void *addr, unsigned long len, int r, int w, int x); extern int os_unmap_memory(void *addr, int len); extern void os_flush_stdout(void); -extern unsigned long long os_usecs(void); /* tt.c * for tt mode only (will be deleted in future...) @@ -245,4 +244,15 @@ extern void setup_machinename(char *machine_out); extern void setup_hostinfo(void); extern int setjmp_wrapper(void (*proc)(void *, void *), ...); +/* time.c */ +#define BILLION (1000 * 1000 * 1000) + +extern void switch_timers(int to_real); +extern void idle_sleep(int secs); +extern void enable_timer(void); +extern void disable_timer(void); +extern void user_time_init(void); +extern void uml_idle_timer(void); +extern unsigned long long os_nsecs(void); + #endif diff --git a/arch/um/include/time_user.h b/arch/um/include/time_user.h deleted file mode 100644 index 17d7ef2141f..00000000000 --- a/arch/um/include/time_user.h +++ /dev/null @@ -1,19 +0,0 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#ifndef __TIME_USER_H__ -#define __TIME_USER_H__ - -extern void timer(void); -extern void switch_timers(int to_real); -extern void idle_sleep(int secs); -extern void enable_timer(void); -extern void prepare_timer(void * ptr); -extern void disable_timer(void); -extern unsigned long time_lock(void); -extern void time_unlock(unsigned long); -extern void user_time_init(void); - -#endif diff --git a/arch/um/kernel/Makefile b/arch/um/kernel/Makefile index 165c8639c9f..693018ba80f 100644 --- a/arch/um/kernel/Makefile +++ b/arch/um/kernel/Makefile @@ -9,7 +9,7 @@ clean-files := obj-y = config.o exec_kern.o exitcode.o \ init_task.o irq.o irq_user.o ksyms.o mem.o physmem.o \ process_kern.o ptrace.o reboot.o resource.o sigio_user.o sigio_kern.o \ - signal_kern.o smp.o syscall_kern.o sysrq.o time.o \ + signal_kern.o smp.o syscall_kern.o sysrq.o \ time_kern.o tlb.o trap_kern.o uaccess.o um_arch.o umid.o obj-$(CONFIG_BLK_DEV_INITRD) += initrd.o @@ -23,7 +23,7 @@ obj-$(CONFIG_MODE_SKAS) += skas/ user-objs-$(CONFIG_TTY_LOG) += tty_log.o -USER_OBJS := $(user-objs-y) config.o time.o tty_log.o +USER_OBJS := $(user-objs-y) config.o tty_log.o include arch/um/scripts/Makefile.rules diff --git a/arch/um/kernel/exec_kern.c b/arch/um/kernel/exec_kern.c index efd222ffe20..569fe8b9b05 100644 --- a/arch/um/kernel/exec_kern.c +++ b/arch/um/kernel/exec_kern.c @@ -17,7 +17,6 @@ #include "irq_user.h" #include "tlb.h" #include "os.h" -#include "time_user.h" #include "choose-mode.h" #include "mode_kern.h" diff --git a/arch/um/kernel/process_kern.c b/arch/um/kernel/process_kern.c index 7f13b85d265..e167cf0a71f 100644 --- a/arch/um/kernel/process_kern.c +++ b/arch/um/kernel/process_kern.c @@ -39,7 +39,6 @@ #include "init.h" #include "irq_user.h" #include "mem_user.h" -#include "time_user.h" #include "tlb.h" #include "frame_kern.h" #include "sigcontext.h" diff --git a/arch/um/kernel/skas/process.c b/arch/um/kernel/skas/process.c index 3b3955d8440..eea1c9c4bb0 100644 --- a/arch/um/kernel/skas/process.c +++ b/arch/um/kernel/skas/process.c @@ -18,7 +18,6 @@ #include #include "user.h" #include "ptrace_user.h" -#include "time_user.h" #include "sysdep/ptrace.h" #include "user_util.h" #include "kern_util.h" diff --git a/arch/um/kernel/skas/process_kern.c b/arch/um/kernel/skas/process_kern.c index fcddee76c7c..a340006dec4 100644 --- a/arch/um/kernel/skas/process_kern.c +++ b/arch/um/kernel/skas/process_kern.c @@ -13,7 +13,6 @@ #include "asm/uaccess.h" #include "asm/atomic.h" #include "kern_util.h" -#include "time_user.h" #include "skas.h" #include "os.h" #include "user_util.h" diff --git a/arch/um/kernel/syscall.c b/arch/um/kernel/syscall.c index 1429c131879..1731d90e685 100644 --- a/arch/um/kernel/syscall.c +++ b/arch/um/kernel/syscall.c @@ -25,12 +25,12 @@ int record_syscall_start(int syscall) syscall_record[index].syscall = syscall; syscall_record[index].pid = current_pid(); syscall_record[index].result = 0xdeadbeef; - syscall_record[index].start = os_usecs(); + syscall_record[index].start = os_nsecs(); return(index); } void record_syscall_end(int index, long result) { syscall_record[index].result = result; - syscall_record[index].end = os_usecs(); + syscall_record[index].end = os_nsecs(); } diff --git a/arch/um/kernel/time_kern.c b/arch/um/kernel/time_kern.c index 020ca79b8d3..6712ffad024 100644 --- a/arch/um/kernel/time_kern.c +++ b/arch/um/kernel/time_kern.c @@ -1,4 +1,4 @@ -/* +/* * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) * Licensed under the GPL */ @@ -13,12 +13,12 @@ #include "linux/interrupt.h" #include "linux/init.h" #include "linux/delay.h" +#include "linux/hrtimer.h" #include "asm/irq.h" #include "asm/param.h" #include "asm/current.h" #include "kern_util.h" #include "user_util.h" -#include "time_user.h" #include "mode.h" #include "os.h" @@ -39,7 +39,7 @@ unsigned long long sched_clock(void) int timer_irq_inited = 0; static int first_tick; -static unsigned long long prev_usecs; +static unsigned long long prev_nsecs; #ifdef CONFIG_UML_REAL_TIME_CLOCK static long long delta; /* Deviation per interval */ #endif @@ -58,23 +58,23 @@ void timer_irq(union uml_pt_regs *regs) if(first_tick){ #ifdef CONFIG_UML_REAL_TIME_CLOCK /* We've had 1 tick */ - unsigned long long usecs = os_usecs(); + unsigned long long nsecs = os_nsecs(); - delta += usecs - prev_usecs; - prev_usecs = usecs; + delta += nsecs - prev_nsecs; + prev_nsecs = nsecs; /* Protect against the host clock being set backwards */ if(delta < 0) delta = 0; - ticks += (delta * HZ) / MILLION; - delta -= (ticks * MILLION) / HZ; + ticks += (delta * HZ) / BILLION; + delta -= (ticks * BILLION) / HZ; #else ticks = 1; #endif } else { - prev_usecs = os_usecs(); + prev_nsecs = os_nsecs(); first_tick = 1; } @@ -88,45 +88,99 @@ void boot_timer_handler(int sig) { struct pt_regs regs; - CHOOSE_MODE((void) + CHOOSE_MODE((void) (UPT_SC(®s.regs) = (struct sigcontext *) (&sig + 1)), (void) (regs.regs.skas.is_user = 0)); do_timer(®s); } +static DEFINE_SPINLOCK(timer_spinlock); + +static unsigned long long local_offset = 0; + +static inline unsigned long long get_time(void) +{ + unsigned long long nsecs; + unsigned long flags; + + spin_lock_irqsave(&timer_spinlock, flags); + nsecs = os_nsecs(); + nsecs += local_offset; + spin_unlock_irqrestore(&timer_spinlock, flags); + + return nsecs; +} + irqreturn_t um_timer(int irq, void *dev, struct pt_regs *regs) { + unsigned long long nsecs; unsigned long flags; do_timer(regs); + write_seqlock_irqsave(&xtime_lock, flags); - timer(); + nsecs = get_time() + local_offset; + xtime.tv_sec = nsecs / NSEC_PER_SEC; + xtime.tv_nsec = nsecs - xtime.tv_sec * NSEC_PER_SEC; write_sequnlock_irqrestore(&xtime_lock, flags); + return(IRQ_HANDLED); } long um_time(int __user *tloc) { - struct timeval now; + long ret = get_time() / NSEC_PER_SEC; - do_gettimeofday(&now); - if (tloc) { - if (put_user(now.tv_sec, tloc)) - now.tv_sec = -EFAULT; - } - return now.tv_sec; + if((tloc != NULL) && put_user(ret, tloc)) + return -EFAULT; + + return ret; +} + +void do_gettimeofday(struct timeval *tv) +{ + unsigned long long nsecs = get_time(); + + tv->tv_sec = nsecs / NSEC_PER_SEC; + /* Careful about calculations here - this was originally done as + * (nsecs - tv->tv_sec * NSEC_PER_SEC) / NSEC_PER_USEC + * which gave bogus (> 1000000) values. Dunno why, suspect gcc + * (4.0.0) miscompiled it, or there's a subtle 64/32-bit conversion + * problem that I missed. + */ + nsecs -= tv->tv_sec * NSEC_PER_SEC; + tv->tv_usec = (unsigned long) nsecs / NSEC_PER_USEC; +} + +static inline void set_time(unsigned long long nsecs) +{ + unsigned long long now; + unsigned long flags; + + spin_lock_irqsave(&timer_spinlock, flags); + now = os_nsecs(); + local_offset = nsecs - now; + spin_unlock_irqrestore(&timer_spinlock, flags); + + clock_was_set(); } long um_stime(int __user *tptr) { int value; - struct timespec new; if (get_user(value, tptr)) return -EFAULT; - new.tv_sec = value; - new.tv_nsec = 0; - do_settimeofday(&new); + + set_time((unsigned long long) value * NSEC_PER_SEC); + + return 0; +} + +int do_settimeofday(struct timespec *tv) +{ + set_time((unsigned long long) tv->tv_sec * NSEC_PER_SEC + tv->tv_nsec); + return 0; } @@ -142,21 +196,6 @@ void timer_handler(int sig, union uml_pt_regs *regs) timer_irq(regs); } -static DEFINE_SPINLOCK(timer_spinlock); - -unsigned long time_lock(void) -{ - unsigned long flags; - - spin_lock_irqsave(&timer_spinlock, flags); - return(flags); -} - -void time_unlock(unsigned long flags) -{ - spin_unlock_irqrestore(&timer_spinlock, flags); -} - int __init timer_init(void) { int err; @@ -171,14 +210,3 @@ int __init timer_init(void) } __initcall(timer_init); - -/* - * 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/arch/um/kernel/tt/exec_kern.c b/arch/um/kernel/tt/exec_kern.c index 8f40e483873..5c1e4cc1c04 100644 --- a/arch/um/kernel/tt/exec_kern.c +++ b/arch/um/kernel/tt/exec_kern.c @@ -13,7 +13,6 @@ #include "user_util.h" #include "kern_util.h" #include "irq_user.h" -#include "time_user.h" #include "mem_user.h" #include "os.h" #include "tlb.h" diff --git a/arch/um/kernel/tt/process_kern.c b/arch/um/kernel/tt/process_kern.c index 62535303aa2..295c1ac817b 100644 --- a/arch/um/kernel/tt/process_kern.c +++ b/arch/um/kernel/tt/process_kern.c @@ -18,7 +18,6 @@ #include "os.h" #include "kern.h" #include "sigcontext.h" -#include "time_user.h" #include "mem_user.h" #include "tlb.h" #include "mode.h" diff --git a/arch/um/os-Linux/main.c b/arch/um/os-Linux/main.c index 172c8474453..98becd18f21 100644 --- a/arch/um/os-Linux/main.c +++ b/arch/um/os-Linux/main.c @@ -16,7 +16,6 @@ #include "user_util.h" #include "kern_util.h" #include "mem_user.h" -#include "time_user.h" #include "irq_user.h" #include "user.h" #include "init.h" diff --git a/arch/um/os-Linux/signal.c b/arch/um/os-Linux/signal.c index c1f46a0fef1..da474a797fb 100644 --- a/arch/um/os-Linux/signal.c +++ b/arch/um/os-Linux/signal.c @@ -18,8 +18,8 @@ #include "sysdep/sigcontext.h" #include "sysdep/signal.h" #include "sigcontext.h" -#include "time_user.h" #include "mode.h" +#include "os.h" void sig_handler(ARCH_SIGHDLR_PARAM) { diff --git a/arch/um/os-Linux/start_up.c b/arch/um/os-Linux/start_up.c index b47e5e71d1a..6c5b17ed59e 100644 --- a/arch/um/os-Linux/start_up.c +++ b/arch/um/os-Linux/start_up.c @@ -29,7 +29,6 @@ #include "irq_user.h" #include "ptrace_user.h" #include "mem_user.h" -#include "time_user.h" #include "init.h" #include "os.h" #include "uml-config.h" diff --git a/arch/um/os-Linux/time.c b/arch/um/os-Linux/time.c index cf30a39bc48..6f7626775ac 100644 --- a/arch/um/os-Linux/time.c +++ b/arch/um/os-Linux/time.c @@ -1,21 +1,128 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include #include +#include +#include #include +#include +#include +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "process.h" +#include "kern_constants.h" +#include "os.h" + +/* XXX This really needs to be declared and initialized in a kernel file since + * it's in + */ +extern struct timespec wall_to_monotonic; + +static void set_interval(int timer_type) +{ + int usec = 1000000/hz(); + struct itimerval interval = ((struct itimerval) { { 0, usec }, + { 0, usec } }); + + if(setitimer(timer_type, &interval, NULL) == -1) + panic("setitimer failed - errno = %d\n", errno); +} + +void enable_timer(void) +{ + set_interval(ITIMER_VIRTUAL); +} + +void disable_timer(void) +{ + struct itimerval disable = ((struct itimerval) { { 0, 0 }, { 0, 0 }}); + if((setitimer(ITIMER_VIRTUAL, &disable, NULL) < 0) || + (setitimer(ITIMER_REAL, &disable, NULL) < 0)) + printk("disnable_timer - setitimer failed, errno = %d\n", + errno); + /* If there are signals already queued, after unblocking ignore them */ + set_handler(SIGALRM, SIG_IGN, 0, -1); + set_handler(SIGVTALRM, SIG_IGN, 0, -1); +} + +void switch_timers(int to_real) +{ + struct itimerval disable = ((struct itimerval) { { 0, 0 }, { 0, 0 }}); + struct itimerval enable = ((struct itimerval) { { 0, 1000000/hz() }, + { 0, 1000000/hz() }}); + int old, new; + + if(to_real){ + old = ITIMER_VIRTUAL; + new = ITIMER_REAL; + } + else { + old = ITIMER_REAL; + new = ITIMER_VIRTUAL; + } + + if((setitimer(old, &disable, NULL) < 0) || + (setitimer(new, &enable, NULL))) + printk("switch_timers - setitimer failed, errno = %d\n", + errno); +} -unsigned long long os_usecs(void) +void uml_idle_timer(void) +{ + if(signal(SIGVTALRM, SIG_IGN) == SIG_ERR) + panic("Couldn't unset SIGVTALRM handler"); + + set_handler(SIGALRM, (__sighandler_t) alarm_handler, + SA_RESTART, SIGUSR1, SIGIO, SIGWINCH, SIGVTALRM, -1); + set_interval(ITIMER_REAL); +} + +extern void ktime_get_ts(struct timespec *ts); +#define do_posix_clock_monotonic_gettime(ts) ktime_get_ts(ts) + +void time_init(void) +{ + struct timespec now; + + if(signal(SIGVTALRM, boot_timer_handler) == SIG_ERR) + panic("Couldn't set SIGVTALRM handler"); + set_interval(ITIMER_VIRTUAL); + + do_posix_clock_monotonic_gettime(&now); + wall_to_monotonic.tv_sec = -now.tv_sec; + wall_to_monotonic.tv_nsec = -now.tv_nsec; +} + +unsigned long long os_nsecs(void) { struct timeval tv; gettimeofday(&tv, NULL); - return((unsigned long long) tv.tv_sec * 1000000 + tv.tv_usec); + return((unsigned long long) tv.tv_sec * BILLION + tv.tv_usec * 1000); } -/* - * 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: - */ +void idle_sleep(int secs) +{ + struct timespec ts; + + ts.tv_sec = secs; + ts.tv_nsec = 0; + nanosleep(&ts, NULL); +} + +/* XXX This partly duplicates init_irq_signals */ + +void user_time_init(void) +{ + set_handler(SIGVTALRM, (__sighandler_t) alarm_handler, + SA_ONSTACK | SA_RESTART, SIGUSR1, SIGIO, SIGWINCH, + SIGALRM, SIGUSR2, -1); + set_handler(SIGALRM, (__sighandler_t) alarm_handler, + SA_ONSTACK | SA_RESTART, SIGUSR1, SIGIO, SIGWINCH, + SIGVTALRM, SIGUSR2, -1); + set_interval(ITIMER_VIRTUAL); +} diff --git a/arch/um/os-Linux/tt.c b/arch/um/os-Linux/tt.c index 404bb63a74a..919d19f1153 100644 --- a/arch/um/os-Linux/tt.c +++ b/arch/um/os-Linux/tt.c @@ -27,7 +27,6 @@ #include "sysdep/sigcontext.h" #include "irq_user.h" #include "ptrace_user.h" -#include "time_user.h" #include "init.h" #include "os.h" #include "uml-config.h" -- cgit v1.2.3 From 2c332a251302873cf8301c2aad27737b6df70255 Mon Sep 17 00:00:00 2001 From: Bodo Stroesser Date: Wed, 18 Jan 2006 17:42:43 -0800 Subject: [PATCH] uml: change interface to boot_timer_handler Current implementation of boot_timer_handler isn't usable for s390. So I changed its name to do_boot_timer_handler, taking (struct sigcontext *)sc as argument. do_boot_timer_handler is called from new boot_timer_handler() in arch/um/os-Linux/signal.c, which uses the same mechanisms as other signal handler to find out sigcontext pointer. Signed-off-by: Bodo Stroesser Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/kernel/time_kern.c | 5 ++--- arch/um/os-Linux/signal.c | 12 +++++++++++- 2 files changed, 13 insertions(+), 4 deletions(-) diff --git a/arch/um/kernel/time_kern.c b/arch/um/kernel/time_kern.c index 6712ffad024..6fa4e1b892b 100644 --- a/arch/um/kernel/time_kern.c +++ b/arch/um/kernel/time_kern.c @@ -84,12 +84,11 @@ void timer_irq(union uml_pt_regs *regs) } } -void boot_timer_handler(int sig) +void do_boot_timer_handler(struct sigcontext * sc) { struct pt_regs regs; - CHOOSE_MODE((void) - (UPT_SC(®s.regs) = (struct sigcontext *) (&sig + 1)), + CHOOSE_MODE((void) (UPT_SC(®s.regs) = sc), (void) (regs.regs.skas.is_user = 0)); do_timer(®s); } diff --git a/arch/um/os-Linux/signal.c b/arch/um/os-Linux/signal.c index da474a797fb..884e4575194 100644 --- a/arch/um/os-Linux/signal.c +++ b/arch/um/os-Linux/signal.c @@ -12,7 +12,6 @@ #include #include #include "user_util.h" -#include "kern_util.h" #include "user.h" #include "signal_kern.h" #include "sysdep/sigcontext.h" @@ -49,6 +48,17 @@ void alarm_handler(ARCH_SIGHDLR_PARAM) switch_timers(1); } +extern void do_boot_timer_handler(struct sigcontext * sc); + +void boot_timer_handler(ARCH_SIGHDLR_PARAM) +{ + struct sigcontext *sc; + + ARCH_GET_SIGCONTEXT(sc, sig); + + do_boot_timer_handler(sc); +} + void set_sigstack(void *sig_stack, int size) { stack_t stack = ((stack_t) { .ss_flags = 0, -- cgit v1.2.3 From 4abfbf4034b419736de5797a3860ab0bcf5c5c8d Mon Sep 17 00:00:00 2001 From: Gennady Sharapov Date: Wed, 18 Jan 2006 17:42:44 -0800 Subject: [PATCH] uml: move headers to arch/um/include The serial UML OS-abstraction layer patch (um/kernel dir). This moves skas headers to arch/um/include. Signed-off-by: Gennady Sharapov Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/Makefile | 2 +- arch/um/include/mode_kern.h | 15 +------ arch/um/include/skas/mm_id.h | 17 ++++++++ arch/um/include/skas/mmu-skas.h | 24 +++++++++++ arch/um/include/skas/mode-skas.h | 22 +++++++++++ arch/um/include/skas/mode_kern_skas.h | 41 +++++++++++++++++++ arch/um/include/skas/proc_mm.h | 44 +++++++++++++++++++++ arch/um/include/skas/skas.h | 49 +++++++++++++++++++++++ arch/um/include/skas/stub-data.h | 18 +++++++++ arch/um/include/skas/uaccess-skas.h | 21 ++++++++++ arch/um/include/tt/debug.h | 18 +++++++++ arch/um/include/tt/mmu-tt.h | 12 ++++++ arch/um/include/tt/mode-tt.h | 23 +++++++++++ arch/um/include/tt/mode_kern_tt.h | 41 +++++++++++++++++++ arch/um/include/tt/tt.h | 36 +++++++++++++++++ arch/um/include/tt/uaccess-tt.h | 48 ++++++++++++++++++++++ arch/um/kernel/skas/include/mm_id.h | 17 -------- arch/um/kernel/skas/include/mmu-skas.h | 35 ----------------- arch/um/kernel/skas/include/mode-skas.h | 33 ---------------- arch/um/kernel/skas/include/mode_kern-skas.h | 52 ------------------------ arch/um/kernel/skas/include/proc_mm.h | 55 -------------------------- arch/um/kernel/skas/include/skas.h | 49 ----------------------- arch/um/kernel/skas/include/stub-data.h | 18 --------- arch/um/kernel/skas/include/uaccess-skas.h | 32 --------------- arch/um/kernel/tt/include/debug.h | 18 --------- arch/um/kernel/tt/include/mmu-tt.h | 23 ----------- arch/um/kernel/tt/include/tt.h | 46 ---------------------- arch/um/kernel/tt/include/uaccess-tt.h | 59 ---------------------------- 28 files changed, 417 insertions(+), 451 deletions(-) create mode 100644 arch/um/include/skas/mm_id.h create mode 100644 arch/um/include/skas/mmu-skas.h create mode 100644 arch/um/include/skas/mode-skas.h create mode 100644 arch/um/include/skas/mode_kern_skas.h create mode 100644 arch/um/include/skas/proc_mm.h create mode 100644 arch/um/include/skas/skas.h create mode 100644 arch/um/include/skas/stub-data.h create mode 100644 arch/um/include/skas/uaccess-skas.h create mode 100644 arch/um/include/tt/debug.h create mode 100644 arch/um/include/tt/mmu-tt.h create mode 100644 arch/um/include/tt/mode-tt.h create mode 100644 arch/um/include/tt/mode_kern_tt.h create mode 100644 arch/um/include/tt/tt.h create mode 100644 arch/um/include/tt/uaccess-tt.h delete mode 100644 arch/um/kernel/skas/include/mm_id.h delete mode 100644 arch/um/kernel/skas/include/mmu-skas.h delete mode 100644 arch/um/kernel/skas/include/mode-skas.h delete mode 100644 arch/um/kernel/skas/include/mode_kern-skas.h delete mode 100644 arch/um/kernel/skas/include/proc_mm.h delete mode 100644 arch/um/kernel/skas/include/skas.h delete mode 100644 arch/um/kernel/skas/include/stub-data.h delete mode 100644 arch/um/kernel/skas/include/uaccess-skas.h delete mode 100644 arch/um/kernel/tt/include/debug.h delete mode 100644 arch/um/kernel/tt/include/mmu-tt.h delete mode 100644 arch/um/kernel/tt/include/tt.h delete mode 100644 arch/um/kernel/tt/include/uaccess-tt.h diff --git a/arch/um/Makefile b/arch/um/Makefile index 45435ff589c..6430a638385 100644 --- a/arch/um/Makefile +++ b/arch/um/Makefile @@ -32,7 +32,7 @@ um-modes-$(CONFIG_MODE_TT) += tt um-modes-$(CONFIG_MODE_SKAS) += skas MODE_INCLUDE += $(foreach mode,$(um-modes-y),\ - -I$(srctree)/$(ARCH_DIR)/kernel/$(mode)/include) + -I$(srctree)/$(ARCH_DIR)/include/$(mode)) MAKEFILES-INCL += $(foreach mode,$(um-modes-y),\ $(srctree)/$(ARCH_DIR)/Makefile-$(mode)) diff --git a/arch/um/include/mode_kern.h b/arch/um/include/mode_kern.h index 2d88afd0cf1..e7539a8451e 100644 --- a/arch/um/include/mode_kern.h +++ b/arch/um/include/mode_kern.h @@ -9,22 +9,11 @@ #include "linux/config.h" #ifdef CONFIG_MODE_TT -#include "mode_kern-tt.h" +#include "mode_kern_tt.h" #endif #ifdef CONFIG_MODE_SKAS -#include "mode_kern-skas.h" +#include "mode_kern_skas.h" #endif #endif - -/* - * 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/arch/um/include/skas/mm_id.h b/arch/um/include/skas/mm_id.h new file mode 100644 index 00000000000..48dd0989dda --- /dev/null +++ b/arch/um/include/skas/mm_id.h @@ -0,0 +1,17 @@ +/* + * Copyright (C) 2005 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __MM_ID_H +#define __MM_ID_H + +struct mm_id { + union { + int mm_fd; + int pid; + } u; + unsigned long stack; +}; + +#endif diff --git a/arch/um/include/skas/mmu-skas.h b/arch/um/include/skas/mmu-skas.h new file mode 100644 index 00000000000..d8869a6ef1b --- /dev/null +++ b/arch/um/include/skas/mmu-skas.h @@ -0,0 +1,24 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SKAS_MMU_H +#define __SKAS_MMU_H + +#include "linux/config.h" +#include "mm_id.h" +#include "asm/ldt.h" + +struct mmu_context_skas { + struct mm_id id; + unsigned long last_page_table; +#ifdef CONFIG_3_LEVEL_PGTABLES + unsigned long last_pmd; +#endif + uml_ldt_t ldt; +}; + +extern void switch_mm_skas(struct mm_id * mm_idp); + +#endif diff --git a/arch/um/include/skas/mode-skas.h b/arch/um/include/skas/mode-skas.h new file mode 100644 index 00000000000..64b960006f1 --- /dev/null +++ b/arch/um/include/skas/mode-skas.h @@ -0,0 +1,22 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __MODE_SKAS_H__ +#define __MODE_SKAS_H__ + +#include + +extern unsigned long exec_regs[]; +extern unsigned long exec_fp_regs[]; +extern unsigned long exec_fpx_regs[]; +extern int have_fpx_regs; + +extern void sig_handler_common_skas(int sig, void *sc_ptr); +extern void halt_skas(void); +extern void reboot_skas(void); +extern void kill_off_processes_skas(void); +extern int is_skas_winch(int pid, int fd, void *data); + +#endif diff --git a/arch/um/include/skas/mode_kern_skas.h b/arch/um/include/skas/mode_kern_skas.h new file mode 100644 index 00000000000..dd9f2d722fb --- /dev/null +++ b/arch/um/include/skas/mode_kern_skas.h @@ -0,0 +1,41 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SKAS_MODE_KERN_H__ +#define __SKAS_MODE_KERN_H__ + +#include "linux/sched.h" +#include "asm/page.h" +#include "asm/ptrace.h" + +extern void flush_thread_skas(void); +extern void switch_to_skas(void *prev, void *next); +extern void start_thread_skas(struct pt_regs *regs, unsigned long eip, + unsigned long esp); +extern int copy_thread_skas(int nr, unsigned long clone_flags, + unsigned long sp, unsigned long stack_top, + struct task_struct *p, struct pt_regs *regs); +extern void release_thread_skas(struct task_struct *task); +extern void initial_thread_cb_skas(void (*proc)(void *), void *arg); +extern void init_idle_skas(void); +extern void flush_tlb_kernel_range_skas(unsigned long start, + unsigned long end); +extern void flush_tlb_kernel_vm_skas(void); +extern void __flush_tlb_one_skas(unsigned long addr); +extern void flush_tlb_range_skas(struct vm_area_struct *vma, + unsigned long start, unsigned long end); +extern void flush_tlb_mm_skas(struct mm_struct *mm); +extern void force_flush_all_skas(void); +extern long execute_syscall_skas(void *r); +extern void before_mem_skas(unsigned long unused); +extern unsigned long set_task_sizes_skas(int arg, unsigned long *host_size_out, + unsigned long *task_size_out); +extern int start_uml_skas(void); +extern int external_pid_skas(struct task_struct *task); +extern int thread_pid_skas(struct task_struct *task); + +#define kmem_end_skas (host_task_size - 1024 * 1024) + +#endif diff --git a/arch/um/include/skas/proc_mm.h b/arch/um/include/skas/proc_mm.h new file mode 100644 index 00000000000..90280920960 --- /dev/null +++ b/arch/um/include/skas/proc_mm.h @@ -0,0 +1,44 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SKAS_PROC_MM_H +#define __SKAS_PROC_MM_H + +#define MM_MMAP 54 +#define MM_MUNMAP 55 +#define MM_MPROTECT 56 +#define MM_COPY_SEGMENTS 57 + +struct mm_mmap { + unsigned long addr; + unsigned long len; + unsigned long prot; + unsigned long flags; + unsigned long fd; + unsigned long offset; +}; + +struct mm_munmap { + unsigned long addr; + unsigned long len; +}; + +struct mm_mprotect { + unsigned long addr; + unsigned long len; + unsigned int prot; +}; + +struct proc_mm_op { + int op; + union { + struct mm_mmap mmap; + struct mm_munmap munmap; + struct mm_mprotect mprotect; + int copy_segments; + } u; +}; + +#endif diff --git a/arch/um/include/skas/skas.h b/arch/um/include/skas/skas.h new file mode 100644 index 00000000000..158f322248e --- /dev/null +++ b/arch/um/include/skas/skas.h @@ -0,0 +1,49 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SKAS_H +#define __SKAS_H + +#include "mm_id.h" +#include "sysdep/ptrace.h" + +extern int userspace_pid[]; +extern int proc_mm, ptrace_faultinfo, ptrace_ldt; +extern int skas_needs_stub; + +extern void switch_threads(void *me, void *next); +extern void thread_wait(void *sw, void *fb); +extern void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr, + void (*handler)(int)); +extern int start_idle_thread(void *stack, void *switch_buf_ptr, + void **fork_buf_ptr); +extern int user_thread(unsigned long stack, int flags); +extern void userspace(union uml_pt_regs *regs); +extern void new_thread_proc(void *stack, void (*handler)(int sig)); +extern void new_thread_handler(int sig); +extern void handle_syscall(union uml_pt_regs *regs); +extern int map(struct mm_id * mm_idp, unsigned long virt, + unsigned long len, int r, int w, int x, int phys_fd, + unsigned long long offset, int done, void **data); +extern int unmap(struct mm_id * mm_idp, void *addr, unsigned long len, + int done, void **data); +extern int protect(struct mm_id * mm_idp, unsigned long addr, + unsigned long len, int r, int w, int x, int done, + void **data); +extern void user_signal(int sig, union uml_pt_regs *regs, int pid); +extern int new_mm(unsigned long stack); +extern int start_userspace(unsigned long stub_stack); +extern int copy_context_skas0(unsigned long stack, int pid); +extern void get_skas_faultinfo(int pid, struct faultinfo * fi); +extern long execute_syscall_skas(void *r); +extern unsigned long current_stub_stack(void); +extern long run_syscall_stub(struct mm_id * mm_idp, + int syscall, unsigned long *args, long expected, + void **addr, int done); +extern long syscall_stub_data(struct mm_id * mm_idp, + unsigned long *data, int data_count, + void **addr, void **stub_addr); + +#endif diff --git a/arch/um/include/skas/stub-data.h b/arch/um/include/skas/stub-data.h new file mode 100644 index 00000000000..f6ed92c3727 --- /dev/null +++ b/arch/um/include/skas/stub-data.h @@ -0,0 +1,18 @@ +/* + * Copyright (C) 2005 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __STUB_DATA_H +#define __STUB_DATA_H + +#include + +struct stub_data { + long offset; + int fd; + struct itimerval timer; + long err; +}; + +#endif diff --git a/arch/um/include/skas/uaccess-skas.h b/arch/um/include/skas/uaccess-skas.h new file mode 100644 index 00000000000..224a75f4c02 --- /dev/null +++ b/arch/um/include/skas/uaccess-skas.h @@ -0,0 +1,21 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SKAS_UACCESS_H +#define __SKAS_UACCESS_H + +#include "asm/errno.h" + +/* No SKAS-specific checking. */ +#define access_ok_skas(type, addr, size) 0 + +extern int copy_from_user_skas(void *to, const void __user *from, int n); +extern int copy_to_user_skas(void __user *to, const void *from, int n); +extern int strncpy_from_user_skas(char *dst, const char __user *src, int count); +extern int __clear_user_skas(void __user *mem, int len); +extern int clear_user_skas(void __user *mem, int len); +extern int strnlen_user_skas(const void __user *str, int len); + +#endif diff --git a/arch/um/include/tt/debug.h b/arch/um/include/tt/debug.h new file mode 100644 index 00000000000..9778fa83829 --- /dev/null +++ b/arch/um/include/tt/debug.h @@ -0,0 +1,18 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) and + * Lars Brinkhoff. + * Licensed under the GPL + */ + +#ifndef __UML_TT_DEBUG_H +#define __UML_TT_DEBUG_H + +extern int debugger_proxy(int status, pid_t pid); +extern void child_proxy(pid_t pid, int status); +extern void init_proxy (pid_t pid, int waiting, int status); +extern int start_debugger(char *prog, int startup, int stop, int *debugger_fd); +extern void fake_child_exit(void); +extern int gdb_config(char *str); +extern int gdb_remove(int unused); + +#endif diff --git a/arch/um/include/tt/mmu-tt.h b/arch/um/include/tt/mmu-tt.h new file mode 100644 index 00000000000..572a78b2258 --- /dev/null +++ b/arch/um/include/tt/mmu-tt.h @@ -0,0 +1,12 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __TT_MMU_H +#define __TT_MMU_H + +struct mmu_context_tt { +}; + +#endif diff --git a/arch/um/include/tt/mode-tt.h b/arch/um/include/tt/mode-tt.h new file mode 100644 index 00000000000..2823cd56eea --- /dev/null +++ b/arch/um/include/tt/mode-tt.h @@ -0,0 +1,23 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __MODE_TT_H__ +#define __MODE_TT_H__ + +#include "sysdep/ptrace.h" + +enum { OP_NONE, OP_EXEC, OP_FORK, OP_TRACE_ON, OP_REBOOT, OP_HALT, OP_CB }; + +extern int tracing_pid; + +extern int tracer(int (*init_proc)(void *), void *sp); +extern void sig_handler_common_tt(int sig, void *sc); +extern void syscall_handler_tt(int sig, union uml_pt_regs *regs); +extern void reboot_tt(void); +extern void halt_tt(void); +extern int is_tracer_winch(int pid, int fd, void *data); +extern void kill_off_processes_tt(void); + +#endif diff --git a/arch/um/include/tt/mode_kern_tt.h b/arch/um/include/tt/mode_kern_tt.h new file mode 100644 index 00000000000..efa0012550d --- /dev/null +++ b/arch/um/include/tt/mode_kern_tt.h @@ -0,0 +1,41 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __TT_MODE_KERN_H__ +#define __TT_MODE_KERN_H__ + +#include "linux/sched.h" +#include "asm/page.h" +#include "asm/ptrace.h" +#include "asm/uaccess.h" + +extern void switch_to_tt(void *prev, void *next); +extern void flush_thread_tt(void); +extern void start_thread_tt(struct pt_regs *regs, unsigned long eip, + unsigned long esp); +extern int copy_thread_tt(int nr, unsigned long clone_flags, unsigned long sp, + unsigned long stack_top, struct task_struct *p, + struct pt_regs *regs); +extern void release_thread_tt(struct task_struct *task); +extern void initial_thread_cb_tt(void (*proc)(void *), void *arg); +extern void init_idle_tt(void); +extern void flush_tlb_kernel_range_tt(unsigned long start, unsigned long end); +extern void flush_tlb_kernel_vm_tt(void); +extern void __flush_tlb_one_tt(unsigned long addr); +extern void flush_tlb_range_tt(struct vm_area_struct *vma, + unsigned long start, unsigned long end); +extern void flush_tlb_mm_tt(struct mm_struct *mm); +extern void force_flush_all_tt(void); +extern long execute_syscall_tt(void *r); +extern void before_mem_tt(unsigned long brk_start); +extern unsigned long set_task_sizes_tt(int arg, unsigned long *host_size_out, + unsigned long *task_size_out); +extern int start_uml_tt(void); +extern int external_pid_tt(struct task_struct *task); +extern int thread_pid_tt(struct task_struct *task); + +#define kmem_end_tt (host_task_size - ABOVE_KMEM) + +#endif diff --git a/arch/um/include/tt/tt.h b/arch/um/include/tt/tt.h new file mode 100644 index 00000000000..80852198018 --- /dev/null +++ b/arch/um/include/tt/tt.h @@ -0,0 +1,36 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __TT_H__ +#define __TT_H__ + +#include "sysdep/ptrace.h" + +extern int gdb_pid; +extern int debug; +extern int debug_stop; +extern int debug_trace; + +extern int honeypot; + +extern int fork_tramp(void *sig_stack); +extern int do_proc_op(void *t, int proc_id); +extern int tracer(int (*init_proc)(void *), void *sp); +extern void attach_process(int pid); +extern void tracer_panic(char *format, ...); +extern void set_init_pid(int pid); +extern int set_user_mode(void *task); +extern void set_tracing(void *t, int tracing); +extern int is_tracing(void *task); +extern void syscall_handler(int sig, union uml_pt_regs *regs); +extern void exit_kernel(int pid, void *task); +extern void do_syscall(void *task, int pid, int local_using_sysemu); +extern void do_sigtrap(void *task); +extern int is_valid_pid(int pid); +extern void remap_data(void *segment_start, void *segment_end, int w); +extern long execute_syscall_tt(void *r); + +#endif + diff --git a/arch/um/include/tt/uaccess-tt.h b/arch/um/include/tt/uaccess-tt.h new file mode 100644 index 00000000000..b19645f32f2 --- /dev/null +++ b/arch/um/include/tt/uaccess-tt.h @@ -0,0 +1,48 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#ifndef __TT_UACCESS_H +#define __TT_UACCESS_H + +#include "linux/string.h" +#include "linux/sched.h" +#include "asm/processor.h" +#include "asm/errno.h" +#include "asm/current.h" +#include "asm/a.out.h" +#include "uml_uaccess.h" + +#define ABOVE_KMEM (16 * 1024 * 1024) + +extern unsigned long end_vm; +extern unsigned long uml_physmem; + +#define is_stack(addr, size) \ + (((unsigned long) (addr) < STACK_TOP) && \ + ((unsigned long) (addr) >= STACK_TOP - ABOVE_KMEM) && \ + (((unsigned long) (addr) + (size)) <= STACK_TOP)) + +#define access_ok_tt(type, addr, size) \ + (is_stack(addr, size)) + +extern unsigned long get_fault_addr(void); + +extern int __do_copy_from_user(void *to, const void *from, int n, + void **fault_addr, void **fault_catcher); +extern int __do_strncpy_from_user(char *dst, const char *src, size_t n, + void **fault_addr, void **fault_catcher); +extern int __do_clear_user(void *mem, size_t len, void **fault_addr, + void **fault_catcher); +extern int __do_strnlen_user(const char *str, unsigned long n, + void **fault_addr, void **fault_catcher); + +extern int copy_from_user_tt(void *to, const void __user *from, int n); +extern int copy_to_user_tt(void __user *to, const void *from, int n); +extern int strncpy_from_user_tt(char *dst, const char __user *src, int count); +extern int __clear_user_tt(void __user *mem, int len); +extern int clear_user_tt(void __user *mem, int len); +extern int strnlen_user_tt(const void __user *str, int len); + +#endif diff --git a/arch/um/kernel/skas/include/mm_id.h b/arch/um/kernel/skas/include/mm_id.h deleted file mode 100644 index 48dd0989dda..00000000000 --- a/arch/um/kernel/skas/include/mm_id.h +++ /dev/null @@ -1,17 +0,0 @@ -/* - * Copyright (C) 2005 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#ifndef __MM_ID_H -#define __MM_ID_H - -struct mm_id { - union { - int mm_fd; - int pid; - } u; - unsigned long stack; -}; - -#endif diff --git a/arch/um/kernel/skas/include/mmu-skas.h b/arch/um/kernel/skas/include/mmu-skas.h deleted file mode 100644 index 44110c521e4..00000000000 --- a/arch/um/kernel/skas/include/mmu-skas.h +++ /dev/null @@ -1,35 +0,0 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#ifndef __SKAS_MMU_H -#define __SKAS_MMU_H - -#include "linux/config.h" -#include "mm_id.h" -#include "asm/ldt.h" - -struct mmu_context_skas { - struct mm_id id; - unsigned long last_page_table; -#ifdef CONFIG_3_LEVEL_PGTABLES - unsigned long last_pmd; -#endif - uml_ldt_t ldt; -}; - -extern void switch_mm_skas(struct mm_id * mm_idp); - -#endif - -/* - * 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/arch/um/kernel/skas/include/mode-skas.h b/arch/um/kernel/skas/include/mode-skas.h deleted file mode 100644 index bcd26a6a388..00000000000 --- a/arch/um/kernel/skas/include/mode-skas.h +++ /dev/null @@ -1,33 +0,0 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#ifndef __MODE_SKAS_H__ -#define __MODE_SKAS_H__ - -#include - -extern unsigned long exec_regs[]; -extern unsigned long exec_fp_regs[]; -extern unsigned long exec_fpx_regs[]; -extern int have_fpx_regs; - -extern void sig_handler_common_skas(int sig, void *sc_ptr); -extern void halt_skas(void); -extern void reboot_skas(void); -extern void kill_off_processes_skas(void); -extern int is_skas_winch(int pid, int fd, void *data); - -#endif - -/* - * 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/arch/um/kernel/skas/include/mode_kern-skas.h b/arch/um/kernel/skas/include/mode_kern-skas.h deleted file mode 100644 index c97a80dfe37..00000000000 --- a/arch/um/kernel/skas/include/mode_kern-skas.h +++ /dev/null @@ -1,52 +0,0 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#ifndef __SKAS_MODE_KERN_H__ -#define __SKAS_MODE_KERN_H__ - -#include "linux/sched.h" -#include "asm/page.h" -#include "asm/ptrace.h" - -extern void flush_thread_skas(void); -extern void switch_to_skas(void *prev, void *next); -extern void start_thread_skas(struct pt_regs *regs, unsigned long eip, - unsigned long esp); -extern int copy_thread_skas(int nr, unsigned long clone_flags, - unsigned long sp, unsigned long stack_top, - struct task_struct *p, struct pt_regs *regs); -extern void release_thread_skas(struct task_struct *task); -extern void initial_thread_cb_skas(void (*proc)(void *), void *arg); -extern void init_idle_skas(void); -extern void flush_tlb_kernel_range_skas(unsigned long start, - unsigned long end); -extern void flush_tlb_kernel_vm_skas(void); -extern void __flush_tlb_one_skas(unsigned long addr); -extern void flush_tlb_range_skas(struct vm_area_struct *vma, - unsigned long start, unsigned long end); -extern void flush_tlb_mm_skas(struct mm_struct *mm); -extern void force_flush_all_skas(void); -extern long execute_syscall_skas(void *r); -extern void before_mem_skas(unsigned long unused); -extern unsigned long set_task_sizes_skas(int arg, unsigned long *host_size_out, - unsigned long *task_size_out); -extern int start_uml_skas(void); -extern int external_pid_skas(struct task_struct *task); -extern int thread_pid_skas(struct task_struct *task); - -#define kmem_end_skas (host_task_size - 1024 * 1024) - -#endif - -/* - * 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/arch/um/kernel/skas/include/proc_mm.h b/arch/um/kernel/skas/include/proc_mm.h deleted file mode 100644 index cce61a67905..00000000000 --- a/arch/um/kernel/skas/include/proc_mm.h +++ /dev/null @@ -1,55 +0,0 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#ifndef __SKAS_PROC_MM_H -#define __SKAS_PROC_MM_H - -#define MM_MMAP 54 -#define MM_MUNMAP 55 -#define MM_MPROTECT 56 -#define MM_COPY_SEGMENTS 57 - -struct mm_mmap { - unsigned long addr; - unsigned long len; - unsigned long prot; - unsigned long flags; - unsigned long fd; - unsigned long offset; -}; - -struct mm_munmap { - unsigned long addr; - unsigned long len; -}; - -struct mm_mprotect { - unsigned long addr; - unsigned long len; - unsigned int prot; -}; - -struct proc_mm_op { - int op; - union { - struct mm_mmap mmap; - struct mm_munmap munmap; - struct mm_mprotect mprotect; - int copy_segments; - } u; -}; - -#endif - -/* - * 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/arch/um/kernel/skas/include/skas.h b/arch/um/kernel/skas/include/skas.h deleted file mode 100644 index a7f7cd556f1..00000000000 --- a/arch/um/kernel/skas/include/skas.h +++ /dev/null @@ -1,49 +0,0 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#ifndef __SKAS_H -#define __SKAS_H - -#include "mm_id.h" -#include "sysdep/ptrace.h" - -extern int userspace_pid[]; -extern int proc_mm, ptrace_faultinfo, ptrace_ldt; -extern int skas_needs_stub; - -extern void switch_threads(void *me, void *next); -extern void thread_wait(void *sw, void *fb); -extern void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr, - void (*handler)(int)); -extern int start_idle_thread(void *stack, void *switch_buf_ptr, - void **fork_buf_ptr); -extern int user_thread(unsigned long stack, int flags); -extern void userspace(union uml_pt_regs *regs); -extern void new_thread_proc(void *stack, void (*handler)(int sig)); -extern void new_thread_handler(int sig); -extern void handle_syscall(union uml_pt_regs *regs); -extern int map(struct mm_id * mm_idp, unsigned long virt, - unsigned long len, int r, int w, int x, int phys_fd, - unsigned long long offset, int done, void **data); -extern int unmap(struct mm_id * mm_idp, void *addr, unsigned long len, - int done, void **data); -extern int protect(struct mm_id * mm_idp, unsigned long addr, - unsigned long len, int r, int w, int x, int done, - void **data); -extern void user_signal(int sig, union uml_pt_regs *regs, int pid); -extern int new_mm(unsigned long stack); -extern int start_userspace(unsigned long stub_stack); -extern int copy_context_skas0(unsigned long stack, int pid); -extern void get_skas_faultinfo(int pid, struct faultinfo * fi); -extern long execute_syscall_skas(void *r); -extern unsigned long current_stub_stack(void); -extern long run_syscall_stub(struct mm_id * mm_idp, - int syscall, unsigned long *args, long expected, - void **addr, int done); -extern long syscall_stub_data(struct mm_id * mm_idp, - unsigned long *data, int data_count, - void **addr, void **stub_addr); - -#endif diff --git a/arch/um/kernel/skas/include/stub-data.h b/arch/um/kernel/skas/include/stub-data.h deleted file mode 100644 index f6ed92c3727..00000000000 --- a/arch/um/kernel/skas/include/stub-data.h +++ /dev/null @@ -1,18 +0,0 @@ -/* - * Copyright (C) 2005 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#ifndef __STUB_DATA_H -#define __STUB_DATA_H - -#include - -struct stub_data { - long offset; - int fd; - struct itimerval timer; - long err; -}; - -#endif diff --git a/arch/um/kernel/skas/include/uaccess-skas.h b/arch/um/kernel/skas/include/uaccess-skas.h deleted file mode 100644 index 64516c556cd..00000000000 --- a/arch/um/kernel/skas/include/uaccess-skas.h +++ /dev/null @@ -1,32 +0,0 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#ifndef __SKAS_UACCESS_H -#define __SKAS_UACCESS_H - -#include "asm/errno.h" - -/* No SKAS-specific checking. */ -#define access_ok_skas(type, addr, size) 0 - -extern int copy_from_user_skas(void *to, const void __user *from, int n); -extern int copy_to_user_skas(void __user *to, const void *from, int n); -extern int strncpy_from_user_skas(char *dst, const char __user *src, int count); -extern int __clear_user_skas(void __user *mem, int len); -extern int clear_user_skas(void __user *mem, int len); -extern int strnlen_user_skas(const void __user *str, int len); - -#endif - -/* - * 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/arch/um/kernel/tt/include/debug.h b/arch/um/kernel/tt/include/debug.h deleted file mode 100644 index 738435461e1..00000000000 --- a/arch/um/kernel/tt/include/debug.h +++ /dev/null @@ -1,18 +0,0 @@ -/* - * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) and - * Lars Brinkhoff. - * Licensed under the GPL - */ - -#ifndef __UML_TT_DEBUG_H -#define __UML_TT_DEBUG_H - -extern int debugger_proxy(int status, pid_t pid); -extern void child_proxy(pid_t pid, int status); -extern void init_proxy (pid_t pid, int waiting, int status); -extern int start_debugger(char *prog, int startup, int stop, int *debugger_fd); -extern void fake_child_exit(void); -extern int gdb_config(char *str); -extern int gdb_remove(int unused); - -#endif diff --git a/arch/um/kernel/tt/include/mmu-tt.h b/arch/um/kernel/tt/include/mmu-tt.h deleted file mode 100644 index 0440510ab3f..00000000000 --- a/arch/um/kernel/tt/include/mmu-tt.h +++ /dev/null @@ -1,23 +0,0 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#ifndef __TT_MMU_H -#define __TT_MMU_H - -struct mmu_context_tt { -}; - -#endif - -/* - * 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/arch/um/kernel/tt/include/tt.h b/arch/um/kernel/tt/include/tt.h deleted file mode 100644 index c667b67af40..00000000000 --- a/arch/um/kernel/tt/include/tt.h +++ /dev/null @@ -1,46 +0,0 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#ifndef __TT_H__ -#define __TT_H__ - -#include "sysdep/ptrace.h" - -extern int gdb_pid; -extern int debug; -extern int debug_stop; -extern int debug_trace; - -extern int honeypot; - -extern int fork_tramp(void *sig_stack); -extern int do_proc_op(void *t, int proc_id); -extern int tracer(int (*init_proc)(void *), void *sp); -extern void attach_process(int pid); -extern void tracer_panic(char *format, ...); -extern void set_init_pid(int pid); -extern int set_user_mode(void *task); -extern void set_tracing(void *t, int tracing); -extern int is_tracing(void *task); -extern void syscall_handler(int sig, union uml_pt_regs *regs); -extern void exit_kernel(int pid, void *task); -extern void do_syscall(void *task, int pid, int local_using_sysemu); -extern void do_sigtrap(void *task); -extern int is_valid_pid(int pid); -extern void remap_data(void *segment_start, void *segment_end, int w); -extern long execute_syscall_tt(void *r); - -#endif - -/* - * 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/arch/um/kernel/tt/include/uaccess-tt.h b/arch/um/kernel/tt/include/uaccess-tt.h deleted file mode 100644 index b9bfe9c481c..00000000000 --- a/arch/um/kernel/tt/include/uaccess-tt.h +++ /dev/null @@ -1,59 +0,0 @@ -/* - * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) - * Licensed under the GPL - */ - -#ifndef __TT_UACCESS_H -#define __TT_UACCESS_H - -#include "linux/string.h" -#include "linux/sched.h" -#include "asm/processor.h" -#include "asm/errno.h" -#include "asm/current.h" -#include "asm/a.out.h" -#include "uml_uaccess.h" - -#define ABOVE_KMEM (16 * 1024 * 1024) - -extern unsigned long end_vm; -extern unsigned long uml_physmem; - -#define is_stack(addr, size) \ - (((unsigned long) (addr) < STACK_TOP) && \ - ((unsigned long) (addr) >= STACK_TOP - ABOVE_KMEM) && \ - (((unsigned long) (addr) + (size)) <= STACK_TOP)) - -#define access_ok_tt(type, addr, size) \ - (is_stack(addr, size)) - -extern unsigned long get_fault_addr(void); - -extern int __do_copy_from_user(void *to, const void *from, int n, - void **fault_addr, void **fault_catcher); -extern int __do_strncpy_from_user(char *dst, const char *src, size_t n, - void **fault_addr, void **fault_catcher); -extern int __do_clear_user(void *mem, size_t len, void **fault_addr, - void **fault_catcher); -extern int __do_strnlen_user(const char *str, unsigned long n, - void **fault_addr, void **fault_catcher); - -extern int copy_from_user_tt(void *to, const void __user *from, int n); -extern int copy_to_user_tt(void __user *to, const void *from, int n); -extern int strncpy_from_user_tt(char *dst, const char __user *src, int count); -extern int __clear_user_tt(void __user *mem, int len); -extern int clear_user_tt(void __user *mem, int len); -extern int strnlen_user_tt(const void __user *str, int len); - -#endif - -/* - * 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: - */ -- cgit v1.2.3 From f45d9fc9d80678c2ee22c578e503055207c46fd0 Mon Sep 17 00:00:00 2001 From: Gennady Sharapov Date: Wed, 18 Jan 2006 17:42:45 -0800 Subject: [PATCH] uml: move libc-dependent skas memory mapping code The serial UML OS-abstraction layer patch (um/kernel/skas dir). This moves all systemcalls from skas/mem_user.c file under os-Linux dir and join skas/mem_user.c and skas/mem.c files. Signed-off-by: Gennady Sharapov Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/include/os.h | 17 +++ arch/um/include/skas/skas.h | 14 -- arch/um/kernel/skas/Makefile | 4 +- arch/um/kernel/skas/mem_user.c | 281 ---------------------------------------- arch/um/os-Linux/skas/Makefile | 4 +- arch/um/os-Linux/skas/mem.c | 283 +++++++++++++++++++++++++++++++++++++++++ 6 files changed, 304 insertions(+), 299 deletions(-) delete mode 100644 arch/um/kernel/skas/mem_user.c create mode 100644 arch/um/os-Linux/skas/mem.c diff --git a/arch/um/include/os.h b/arch/um/include/os.h index 624938ad9e1..8006e085b67 100644 --- a/arch/um/include/os.h +++ b/arch/um/include/os.h @@ -11,6 +11,7 @@ #include "../os/include/file.h" #include "sysdep/ptrace.h" #include "kern_util.h" +#include "skas/mm_id.h" #define OS_TYPE_FILE 1 #define OS_TYPE_DIR 2 @@ -255,4 +256,20 @@ extern void user_time_init(void); extern void uml_idle_timer(void); extern unsigned long long os_nsecs(void); +/* skas/mem.c */ +extern long run_syscall_stub(struct mm_id * mm_idp, + int syscall, unsigned long *args, long expected, + void **addr, int done); +extern long syscall_stub_data(struct mm_id * mm_idp, + unsigned long *data, int data_count, + void **addr, void **stub_addr); +extern int map(struct mm_id * mm_idp, unsigned long virt, + unsigned long len, int r, int w, int x, int phys_fd, + unsigned long long offset, int done, void **data); +extern int unmap(struct mm_id * mm_idp, void *addr, unsigned long len, + int done, void **data); +extern int protect(struct mm_id * mm_idp, unsigned long addr, + unsigned long len, int r, int w, int x, int done, + void **data); + #endif diff --git a/arch/um/include/skas/skas.h b/arch/um/include/skas/skas.h index 158f322248e..7418f2addda 100644 --- a/arch/um/include/skas/skas.h +++ b/arch/um/include/skas/skas.h @@ -24,14 +24,6 @@ extern void userspace(union uml_pt_regs *regs); extern void new_thread_proc(void *stack, void (*handler)(int sig)); extern void new_thread_handler(int sig); extern void handle_syscall(union uml_pt_regs *regs); -extern int map(struct mm_id * mm_idp, unsigned long virt, - unsigned long len, int r, int w, int x, int phys_fd, - unsigned long long offset, int done, void **data); -extern int unmap(struct mm_id * mm_idp, void *addr, unsigned long len, - int done, void **data); -extern int protect(struct mm_id * mm_idp, unsigned long addr, - unsigned long len, int r, int w, int x, int done, - void **data); extern void user_signal(int sig, union uml_pt_regs *regs, int pid); extern int new_mm(unsigned long stack); extern int start_userspace(unsigned long stub_stack); @@ -39,11 +31,5 @@ extern int copy_context_skas0(unsigned long stack, int pid); extern void get_skas_faultinfo(int pid, struct faultinfo * fi); extern long execute_syscall_skas(void *r); extern unsigned long current_stub_stack(void); -extern long run_syscall_stub(struct mm_id * mm_idp, - int syscall, unsigned long *args, long expected, - void **addr, int done); -extern long syscall_stub_data(struct mm_id * mm_idp, - unsigned long *data, int data_count, - void **addr, void **stub_addr); #endif diff --git a/arch/um/kernel/skas/Makefile b/arch/um/kernel/skas/Makefile index 7a9fc16d71d..eb664c67c3f 100644 --- a/arch/um/kernel/skas/Makefile +++ b/arch/um/kernel/skas/Makefile @@ -1,9 +1,9 @@ -# +# # Copyright (C) 2002 - 2004 Jeff Dike (jdike@addtoit.com) # Licensed under the GPL # -obj-y := clone.o exec_kern.o mem.o mem_user.o mmu.o process.o process_kern.o \ +obj-y := clone.o exec_kern.o mem.o mmu.o process.o process_kern.o \ syscall.o tlb.o uaccess.o USER_OBJS := process.o clone.o diff --git a/arch/um/kernel/skas/mem_user.c b/arch/um/kernel/skas/mem_user.c deleted file mode 100644 index 1d89640bd50..00000000000 --- a/arch/um/kernel/skas/mem_user.c +++ /dev/null @@ -1,281 +0,0 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#include -#include -#include -#include -#include -#include -#include -#include "mem_user.h" -#include "mem.h" -#include "skas.h" -#include "user.h" -#include "os.h" -#include "proc_mm.h" -#include "ptrace_user.h" -#include "user_util.h" -#include "kern_util.h" -#include "task.h" -#include "registers.h" -#include "uml-config.h" -#include "sysdep/ptrace.h" -#include "sysdep/stub.h" - -extern unsigned long batch_syscall_stub, __syscall_stub_start; - -extern void wait_stub_done(int pid, int sig, char * fname); - -static inline unsigned long *check_init_stack(struct mm_id * mm_idp, - unsigned long *stack) -{ - if(stack == NULL){ - stack = (unsigned long *) mm_idp->stack + 2; - *stack = 0; - } - return stack; -} - -extern int proc_mm; - -int single_count = 0; -int multi_count = 0; -int multi_op_count = 0; - -static long do_syscall_stub(struct mm_id *mm_idp, void **addr) -{ - unsigned long regs[MAX_REG_NR]; - unsigned long *data; - unsigned long *syscall; - long ret, offset; - int n, pid = mm_idp->u.pid; - - if(proc_mm) -#warning Need to look up userspace_pid by cpu - pid = userspace_pid[0]; - - multi_count++; - - get_safe_registers(regs); - regs[REGS_IP_INDEX] = UML_CONFIG_STUB_CODE + - ((unsigned long) &batch_syscall_stub - - (unsigned long) &__syscall_stub_start); - n = ptrace_setregs(pid, regs); - if(n < 0) - panic("do_syscall_stub : PTRACE_SETREGS failed, errno = %d\n", - n); - - wait_stub_done(pid, 0, "do_syscall_stub"); - - /* When the stub stops, we find the following values on the - * beginning of the stack: - * (long )return_value - * (long )offset to failed sycall-data (0, if no error) - */ - ret = *((unsigned long *) mm_idp->stack); - offset = *((unsigned long *) mm_idp->stack + 1); - if (offset) { - data = (unsigned long *)(mm_idp->stack + - offset - UML_CONFIG_STUB_DATA); - syscall = (unsigned long *)((unsigned long)data + data[0]); - printk("do_syscall_stub: syscall %ld failed, return value = " - "0x%lx, expected return value = 0x%lx\n", - syscall[0], ret, syscall[7]); - printk(" syscall parameters: " - "0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx\n", - syscall[1], syscall[2], syscall[3], - syscall[4], syscall[5], syscall[6]); - for(n = 1; n < data[0]/sizeof(long); n++) { - if(n == 1) - printk(" additional syscall data:"); - if(n % 4 == 1) - printk("\n "); - printk(" 0x%lx", data[n]); - } - if(n > 1) - printk("\n"); - } - else ret = 0; - - *addr = check_init_stack(mm_idp, NULL); - - return ret; -} - -long run_syscall_stub(struct mm_id * mm_idp, int syscall, - unsigned long *args, long expected, void **addr, - int done) -{ - unsigned long *stack = check_init_stack(mm_idp, *addr); - - if(done && *addr == NULL) - single_count++; - - *stack += sizeof(long); - stack += *stack / sizeof(long); - - *stack++ = syscall; - *stack++ = args[0]; - *stack++ = args[1]; - *stack++ = args[2]; - *stack++ = args[3]; - *stack++ = args[4]; - *stack++ = args[5]; - *stack++ = expected; - *stack = 0; - multi_op_count++; - - if(!done && ((((unsigned long) stack) & ~PAGE_MASK) < - PAGE_SIZE - 10 * sizeof(long))){ - *addr = stack; - return 0; - } - - return do_syscall_stub(mm_idp, addr); -} - -long syscall_stub_data(struct mm_id * mm_idp, - unsigned long *data, int data_count, - void **addr, void **stub_addr) -{ - unsigned long *stack; - int ret = 0; - - /* If *addr still is uninitialized, it *must* contain NULL. - * Thus in this case do_syscall_stub correctly won't be called. - */ - if((((unsigned long) *addr) & ~PAGE_MASK) >= - PAGE_SIZE - (10 + data_count) * sizeof(long)) { - ret = do_syscall_stub(mm_idp, addr); - /* in case of error, don't overwrite data on stack */ - if(ret) - return ret; - } - - stack = check_init_stack(mm_idp, *addr); - *addr = stack; - - *stack = data_count * sizeof(long); - - memcpy(stack + 1, data, data_count * sizeof(long)); - - *stub_addr = (void *)(((unsigned long)(stack + 1) & ~PAGE_MASK) + - UML_CONFIG_STUB_DATA); - - return 0; -} - -int map(struct mm_id * mm_idp, unsigned long virt, unsigned long len, - int r, int w, int x, int phys_fd, unsigned long long offset, - int done, void **data) -{ - int prot, ret; - - prot = (r ? PROT_READ : 0) | (w ? PROT_WRITE : 0) | - (x ? PROT_EXEC : 0); - - if(proc_mm){ - struct proc_mm_op map; - int fd = mm_idp->u.mm_fd; - - map = ((struct proc_mm_op) { .op = MM_MMAP, - .u = - { .mmap = - { .addr = virt, - .len = len, - .prot = prot, - .flags = MAP_SHARED | - MAP_FIXED, - .fd = phys_fd, - .offset= offset - } } } ); - ret = os_write_file(fd, &map, sizeof(map)); - if(ret != sizeof(map)) - printk("map : /proc/mm map failed, err = %d\n", -ret); - else ret = 0; - } - else { - unsigned long args[] = { virt, len, prot, - MAP_SHARED | MAP_FIXED, phys_fd, - MMAP_OFFSET(offset) }; - - ret = run_syscall_stub(mm_idp, STUB_MMAP_NR, args, virt, - data, done); - } - - return ret; -} - -int unmap(struct mm_id * mm_idp, void *addr, unsigned long len, int done, - void **data) -{ - int ret; - - if(proc_mm){ - struct proc_mm_op unmap; - int fd = mm_idp->u.mm_fd; - - unmap = ((struct proc_mm_op) { .op = MM_MUNMAP, - .u = - { .munmap = - { .addr = - (unsigned long) addr, - .len = len } } } ); - ret = os_write_file(fd, &unmap, sizeof(unmap)); - if(ret != sizeof(unmap)) - printk("unmap - proc_mm write returned %d\n", ret); - else ret = 0; - } - else { - unsigned long args[] = { (unsigned long) addr, len, 0, 0, 0, - 0 }; - - ret = run_syscall_stub(mm_idp, __NR_munmap, args, 0, - data, done); - if(ret < 0) - printk("munmap stub failed, errno = %d\n", ret); - } - - return ret; -} - -int protect(struct mm_id * mm_idp, unsigned long addr, unsigned long len, - int r, int w, int x, int done, void **data) -{ - struct proc_mm_op protect; - int prot, ret; - - prot = (r ? PROT_READ : 0) | (w ? PROT_WRITE : 0) | - (x ? PROT_EXEC : 0); - - if(proc_mm){ - int fd = mm_idp->u.mm_fd; - protect = ((struct proc_mm_op) { .op = MM_MPROTECT, - .u = - { .mprotect = - { .addr = - (unsigned long) addr, - .len = len, - .prot = prot } } } ); - - ret = os_write_file(fd, &protect, sizeof(protect)); - if(ret != sizeof(protect)) - printk("protect failed, err = %d", -ret); - else ret = 0; - } - else { - unsigned long args[] = { addr, len, prot, 0, 0, 0 }; - - ret = run_syscall_stub(mm_idp, __NR_mprotect, args, 0, - data, done); - } - - return ret; -} - -void before_mem_skas(unsigned long unused) -{ -} diff --git a/arch/um/os-Linux/skas/Makefile b/arch/um/os-Linux/skas/Makefile index eab5386d60a..79cc6c72aa9 100644 --- a/arch/um/os-Linux/skas/Makefile +++ b/arch/um/os-Linux/skas/Makefile @@ -3,8 +3,8 @@ # Licensed under the GPL # -obj-y := trap.o +obj-y := mem.o trap.o -USER_OBJS := trap.o +USER_OBJS := mem.o trap.o include arch/um/scripts/Makefile.rules diff --git a/arch/um/os-Linux/skas/mem.c b/arch/um/os-Linux/skas/mem.c new file mode 100644 index 00000000000..9890e9090f5 --- /dev/null +++ b/arch/um/os-Linux/skas/mem.c @@ -0,0 +1,283 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include +#include +#include +#include +#include +#include +#include +#include "mem_user.h" +#include "mem.h" +#include "skas.h" +#include "user.h" +#include "os.h" +#include "proc_mm.h" +#include "ptrace_user.h" +#include "user_util.h" +#include "kern_util.h" +#include "task.h" +#include "registers.h" +#include "uml-config.h" +#include "sysdep/ptrace.h" +#include "sysdep/stub.h" + +extern unsigned long batch_syscall_stub, __syscall_stub_start; + +extern void wait_stub_done(int pid, int sig, char * fname); + +static inline unsigned long *check_init_stack(struct mm_id * mm_idp, + unsigned long *stack) +{ + if(stack == NULL) { + stack = (unsigned long *) mm_idp->stack + 2; + *stack = 0; + } + return stack; +} + +extern int proc_mm; + +int single_count = 0; +int multi_count = 0; +int multi_op_count = 0; + +static inline long do_syscall_stub(struct mm_id * mm_idp, void **addr) +{ + unsigned long regs[MAX_REG_NR]; + int n; + long ret, offset; + unsigned long * data; + unsigned long * syscall; + int pid = mm_idp->u.pid; + + if(proc_mm) +#warning Need to look up userspace_pid by cpu + pid = userspace_pid[0]; + + multi_count++; + + get_safe_registers(regs); + regs[REGS_IP_INDEX] = UML_CONFIG_STUB_CODE + + ((unsigned long) &batch_syscall_stub - + (unsigned long) &__syscall_stub_start); + + n = ptrace_setregs(pid, regs); + if(n < 0) + panic("do_syscall_stub : PTRACE_SETREGS failed, errno = %d\n", + n); + + wait_stub_done(pid, 0, "do_syscall_stub"); + + /* When the stub stops, we find the following values on the + * beginning of the stack: + * (long )return_value + * (long )offset to failed sycall-data (0, if no error) + */ + ret = *((unsigned long *) mm_idp->stack); + offset = *((unsigned long *) mm_idp->stack + 1); + if (offset) { + data = (unsigned long *)(mm_idp->stack + + offset - UML_CONFIG_STUB_DATA); + printk("do_syscall_stub : ret = %d, offset = %d, " + "data = 0x%x\n", ret, offset, data); + syscall = (unsigned long *)((unsigned long)data + data[0]); + printk("do_syscall_stub: syscall %ld failed, return value = " + "0x%lx, expected return value = 0x%lx\n", + syscall[0], ret, syscall[7]); + printk(" syscall parameters: " + "0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx\n", + syscall[1], syscall[2], syscall[3], + syscall[4], syscall[5], syscall[6]); + for(n = 1; n < data[0]/sizeof(long); n++) { + if(n == 1) + printk(" additional syscall data:"); + if(n % 4 == 1) + printk("\n "); + printk(" 0x%lx", data[n]); + } + if(n > 1) + printk("\n"); + } + else ret = 0; + + *addr = check_init_stack(mm_idp, NULL); + + return ret; +} + +long run_syscall_stub(struct mm_id * mm_idp, int syscall, + unsigned long *args, long expected, void **addr, + int done) +{ + unsigned long *stack = check_init_stack(mm_idp, *addr); + + if(done && *addr == NULL) + single_count++; + + *stack += sizeof(long); + stack += *stack / sizeof(long); + + *stack++ = syscall; + *stack++ = args[0]; + *stack++ = args[1]; + *stack++ = args[2]; + *stack++ = args[3]; + *stack++ = args[4]; + *stack++ = args[5]; + *stack++ = expected; + *stack = 0; + multi_op_count++; + + if(!done && ((((unsigned long) stack) & ~PAGE_MASK) < + PAGE_SIZE - 10 * sizeof(long))){ + *addr = stack; + return 0; + } + + return do_syscall_stub(mm_idp, addr); +} + +long syscall_stub_data(struct mm_id * mm_idp, + unsigned long *data, int data_count, + void **addr, void **stub_addr) +{ + unsigned long *stack; + int ret = 0; + + /* If *addr still is uninitialized, it *must* contain NULL. + * Thus in this case do_syscall_stub correctly won't be called. + */ + if((((unsigned long) *addr) & ~PAGE_MASK) >= + PAGE_SIZE - (10 + data_count) * sizeof(long)) { + ret = do_syscall_stub(mm_idp, addr); + /* in case of error, don't overwrite data on stack */ + if(ret) + return ret; + } + + stack = check_init_stack(mm_idp, *addr); + *addr = stack; + + *stack = data_count * sizeof(long); + + memcpy(stack + 1, data, data_count * sizeof(long)); + + *stub_addr = (void *)(((unsigned long)(stack + 1) & ~PAGE_MASK) + + UML_CONFIG_STUB_DATA); + + return 0; +} + +int map(struct mm_id * mm_idp, unsigned long virt, unsigned long len, + int r, int w, int x, int phys_fd, unsigned long long offset, + int done, void **data) +{ + int prot, ret; + + prot = (r ? PROT_READ : 0) | (w ? PROT_WRITE : 0) | + (x ? PROT_EXEC : 0); + + if(proc_mm){ + struct proc_mm_op map; + int fd = mm_idp->u.mm_fd; + + map = ((struct proc_mm_op) { .op = MM_MMAP, + .u = + { .mmap = + { .addr = virt, + .len = len, + .prot = prot, + .flags = MAP_SHARED | + MAP_FIXED, + .fd = phys_fd, + .offset= offset + } } } ); + ret = os_write_file(fd, &map, sizeof(map)); + if(ret != sizeof(map)) + printk("map : /proc/mm map failed, err = %d\n", -ret); + else ret = 0; + } + else { + unsigned long args[] = { virt, len, prot, + MAP_SHARED | MAP_FIXED, phys_fd, + MMAP_OFFSET(offset) }; + + ret = run_syscall_stub(mm_idp, STUB_MMAP_NR, args, virt, + data, done); + } + + return ret; +} + +int unmap(struct mm_id * mm_idp, void *addr, unsigned long len, int done, + void **data) +{ + int ret; + + if(proc_mm){ + struct proc_mm_op unmap; + int fd = mm_idp->u.mm_fd; + + unmap = ((struct proc_mm_op) { .op = MM_MUNMAP, + .u = + { .munmap = + { .addr = + (unsigned long) addr, + .len = len } } } ); + ret = os_write_file(fd, &unmap, sizeof(unmap)); + if(ret != sizeof(unmap)) + printk("unmap - proc_mm write returned %d\n", ret); + else ret = 0; + } + else { + unsigned long args[] = { (unsigned long) addr, len, 0, 0, 0, + 0 }; + + ret = run_syscall_stub(mm_idp, __NR_munmap, args, 0, + data, done); + } + + return ret; +} + +int protect(struct mm_id * mm_idp, unsigned long addr, unsigned long len, + int r, int w, int x, int done, void **data) +{ + struct proc_mm_op protect; + int prot, ret; + + prot = (r ? PROT_READ : 0) | (w ? PROT_WRITE : 0) | + (x ? PROT_EXEC : 0); + if(proc_mm){ + int fd = mm_idp->u.mm_fd; + + protect = ((struct proc_mm_op) { .op = MM_MPROTECT, + .u = + { .mprotect = + { .addr = + (unsigned long) addr, + .len = len, + .prot = prot } } } ); + + ret = os_write_file(fd, &protect, sizeof(protect)); + if(ret != sizeof(protect)) + printk("protect failed, err = %d", -ret); + else ret = 0; + } + else { + unsigned long args[] = { addr, len, prot, 0, 0, 0 }; + + ret = run_syscall_stub(mm_idp, __NR_mprotect, args, 0, + data, done); + } + + return ret; +} + +void before_mem_skas(unsigned long unused) +{ +} -- cgit v1.2.3 From abaf69773d8dda98b917d94c07757f6520da7bec Mon Sep 17 00:00:00 2001 From: Gennady Sharapov Date: Wed, 18 Jan 2006 17:42:46 -0800 Subject: [PATCH] uml: move libc-dependent skas process handling The serial UML OS-abstraction layer patch (um/kernel/skas dir). This moves all systemcalls from skas/process.c file under os-Linux dir and join skas/process.c and skas/process_kern.c files. Signed-off-by: Gennady Sharapov Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/include/os.h | 19 ++ arch/um/include/skas/mode-skas.h | 3 - arch/um/include/skas/mode_kern_skas.h | 1 - arch/um/include/skas/skas.h | 9 - arch/um/kernel/skas/Makefile | 4 +- arch/um/kernel/skas/process_kern.c | 18 +- arch/um/os-Linux/signal.c | 5 + arch/um/os-Linux/skas/Makefile | 4 +- arch/um/os-Linux/skas/process.c | 565 ++++++++++++++++++++++++++++++++++ 9 files changed, 601 insertions(+), 27 deletions(-) create mode 100644 arch/um/os-Linux/skas/process.c diff --git a/arch/um/include/os.h b/arch/um/include/os.h index 8006e085b67..eb1710b8125 100644 --- a/arch/um/include/os.h +++ b/arch/um/include/os.h @@ -232,6 +232,7 @@ extern void block_signals(void); extern void unblock_signals(void); extern int get_signals(void); extern int set_signals(int enable); +extern void os_usr1_signal(int on); /* trap.c */ extern void os_fill_handlinfo(struct kern_handlers h); @@ -272,4 +273,22 @@ extern int protect(struct mm_id * mm_idp, unsigned long addr, unsigned long len, int r, int w, int x, int done, void **data); +/* skas/process.c */ +extern int is_skas_winch(int pid, int fd, void *data); +extern int start_userspace(unsigned long stub_stack); +extern int copy_context_skas0(unsigned long stack, int pid); +extern void userspace(union uml_pt_regs *regs); +extern void map_stub_pages(int fd, unsigned long code, + unsigned long data, unsigned long stack); +extern void new_thread(void *stack, void **switch_buf_ptr, + void **fork_buf_ptr, void (*handler)(int)); +extern void thread_wait(void *sw, void *fb); +extern void switch_threads(void *me, void *next); +extern int start_idle_thread(void *stack, void *switch_buf_ptr, + void **fork_buf_ptr); +extern void initial_thread_cb_skas(void (*proc)(void *), + void *arg); +extern void halt_skas(void); +extern void reboot_skas(void); + #endif diff --git a/arch/um/include/skas/mode-skas.h b/arch/um/include/skas/mode-skas.h index 64b960006f1..260065cfeef 100644 --- a/arch/um/include/skas/mode-skas.h +++ b/arch/um/include/skas/mode-skas.h @@ -14,9 +14,6 @@ extern unsigned long exec_fpx_regs[]; extern int have_fpx_regs; extern void sig_handler_common_skas(int sig, void *sc_ptr); -extern void halt_skas(void); -extern void reboot_skas(void); extern void kill_off_processes_skas(void); -extern int is_skas_winch(int pid, int fd, void *data); #endif diff --git a/arch/um/include/skas/mode_kern_skas.h b/arch/um/include/skas/mode_kern_skas.h index dd9f2d722fb..63c58739bde 100644 --- a/arch/um/include/skas/mode_kern_skas.h +++ b/arch/um/include/skas/mode_kern_skas.h @@ -18,7 +18,6 @@ extern int copy_thread_skas(int nr, unsigned long clone_flags, unsigned long sp, unsigned long stack_top, struct task_struct *p, struct pt_regs *regs); extern void release_thread_skas(struct task_struct *task); -extern void initial_thread_cb_skas(void (*proc)(void *), void *arg); extern void init_idle_skas(void); extern void flush_tlb_kernel_range_skas(unsigned long start, unsigned long end); diff --git a/arch/um/include/skas/skas.h b/arch/um/include/skas/skas.h index 7418f2addda..86357282d68 100644 --- a/arch/um/include/skas/skas.h +++ b/arch/um/include/skas/skas.h @@ -13,21 +13,12 @@ extern int userspace_pid[]; extern int proc_mm, ptrace_faultinfo, ptrace_ldt; extern int skas_needs_stub; -extern void switch_threads(void *me, void *next); -extern void thread_wait(void *sw, void *fb); -extern void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr, - void (*handler)(int)); -extern int start_idle_thread(void *stack, void *switch_buf_ptr, - void **fork_buf_ptr); extern int user_thread(unsigned long stack, int flags); -extern void userspace(union uml_pt_regs *regs); extern void new_thread_proc(void *stack, void (*handler)(int sig)); extern void new_thread_handler(int sig); extern void handle_syscall(union uml_pt_regs *regs); extern void user_signal(int sig, union uml_pt_regs *regs, int pid); extern int new_mm(unsigned long stack); -extern int start_userspace(unsigned long stub_stack); -extern int copy_context_skas0(unsigned long stack, int pid); extern void get_skas_faultinfo(int pid, struct faultinfo * fi); extern long execute_syscall_skas(void *r); extern unsigned long current_stub_stack(void); diff --git a/arch/um/kernel/skas/Makefile b/arch/um/kernel/skas/Makefile index eb664c67c3f..57181a920d4 100644 --- a/arch/um/kernel/skas/Makefile +++ b/arch/um/kernel/skas/Makefile @@ -3,10 +3,10 @@ # Licensed under the GPL # -obj-y := clone.o exec_kern.o mem.o mmu.o process.o process_kern.o \ +obj-y := clone.o exec_kern.o mem.o mmu.o process_kern.o \ syscall.o tlb.o uaccess.o -USER_OBJS := process.o clone.o +USER_OBJS := clone.o include arch/um/scripts/Makefile.rules diff --git a/arch/um/kernel/skas/process_kern.c b/arch/um/kernel/skas/process_kern.c index a340006dec4..3f70a2e12f0 100644 --- a/arch/um/kernel/skas/process_kern.c +++ b/arch/um/kernel/skas/process_kern.c @@ -1,4 +1,4 @@ -/* +/* * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) * Licensed under the GPL */ @@ -32,7 +32,7 @@ void switch_to_skas(void *prev, void *next) if(current->pid == 0) switch_timers(0); - switch_threads(&from->thread.mode.skas.switch_buf, + switch_threads(&from->thread.mode.skas.switch_buf, to->thread.mode.skas.switch_buf); if(current->pid == 0) @@ -48,8 +48,8 @@ void new_thread_handler(int sig) fn = current->thread.request.u.thread.proc; arg = current->thread.request.u.thread.arg; - change_sig(SIGUSR1, 1); - thread_wait(¤t->thread.mode.skas.switch_buf, + os_usr1_signal(1); + thread_wait(¤t->thread.mode.skas.switch_buf, current->thread.mode.skas.fork_buf); if(current->thread.prev_sched != NULL) @@ -80,8 +80,8 @@ void release_thread_skas(struct task_struct *task) void fork_handler(int sig) { - change_sig(SIGUSR1, 1); - thread_wait(¤t->thread.mode.skas.switch_buf, + os_usr1_signal(1); + thread_wait(¤t->thread.mode.skas.switch_buf, current->thread.mode.skas.fork_buf); force_flush_all(); @@ -91,13 +91,13 @@ void fork_handler(int sig) schedule_tail(current->thread.prev_sched); current->thread.prev_sched = NULL; - /* Handle any immediate reschedules or signals */ +/* Handle any immediate reschedules or signals */ interrupt_end(); userspace(¤t->thread.regs.regs); } int copy_thread_skas(int nr, unsigned long clone_flags, unsigned long sp, - unsigned long stack_top, struct task_struct * p, + unsigned long stack_top, struct task_struct * p, struct pt_regs *regs) { void (*handler)(int); @@ -121,8 +121,6 @@ int copy_thread_skas(int nr, unsigned long clone_flags, unsigned long sp, return(0); } -extern void map_stub_pages(int fd, unsigned long code, - unsigned long data, unsigned long stack); int new_mm(unsigned long stack) { int fd; diff --git a/arch/um/os-Linux/signal.c b/arch/um/os-Linux/signal.c index 884e4575194..56ca95931b4 100644 --- a/arch/um/os-Linux/signal.c +++ b/arch/um/os-Linux/signal.c @@ -194,3 +194,8 @@ int set_signals(int enable) return(ret); } + +void os_usr1_signal(int on) +{ + change_sig(SIGUSR1, on); +} diff --git a/arch/um/os-Linux/skas/Makefile b/arch/um/os-Linux/skas/Makefile index 79cc6c72aa9..5fd8d4dad66 100644 --- a/arch/um/os-Linux/skas/Makefile +++ b/arch/um/os-Linux/skas/Makefile @@ -3,8 +3,8 @@ # Licensed under the GPL # -obj-y := mem.o trap.o +obj-y := mem.o process.o trap.o -USER_OBJS := mem.o trap.o +USER_OBJS := mem.o process.o trap.o include arch/um/scripts/Makefile.rules diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c new file mode 100644 index 00000000000..f787854322c --- /dev/null +++ b/arch/um/os-Linux/skas/process.c @@ -0,0 +1,565 @@ +/* + * Copyright (C) 2002- 2004 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include +#include +#include +#include +#include +#include +#include +#include "ptrace_user.h" +#include +#include +#include +#include +#include +#include +#include "user.h" +#include "sysdep/ptrace.h" +#include "user_util.h" +#include "kern_util.h" +#include "skas.h" +#include "stub-data.h" +#include "mm_id.h" +#include "sysdep/sigcontext.h" +#include "sysdep/stub.h" +#include "os.h" +#include "proc_mm.h" +#include "skas_ptrace.h" +#include "chan_user.h" +#include "registers.h" +#include "mem.h" +#include "uml-config.h" +#include "process.h" + +int is_skas_winch(int pid, int fd, void *data) +{ + if(pid != os_getpgrp()) + return(0); + + register_winch_irq(-1, fd, -1, data); + return(1); +} + +void wait_stub_done(int pid, int sig, char * fname) +{ + int n, status, err; + + do { + if ( sig != -1 ) { + err = ptrace(PTRACE_CONT, pid, 0, sig); + if(err) + panic("%s : continue failed, errno = %d\n", + fname, errno); + } + sig = 0; + + CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED)); + } while((n >= 0) && WIFSTOPPED(status) && + ((WSTOPSIG(status) == SIGVTALRM) || + /* running UML inside a detached screen can cause + * SIGWINCHes + */ + (WSTOPSIG(status) == SIGWINCH))); + + if((n < 0) || !WIFSTOPPED(status) || + (WSTOPSIG(status) != SIGUSR1 && WSTOPSIG(status) != SIGTRAP)){ + unsigned long regs[HOST_FRAME_SIZE]; + + if(ptrace(PTRACE_GETREGS, pid, 0, regs) < 0) + printk("Failed to get registers from stub, " + "errno = %d\n", errno); + else { + int i; + + printk("Stub registers -\n"); + for(i = 0; i < HOST_FRAME_SIZE; i++) + printk("\t%d - %lx\n", i, regs[i]); + } + panic("%s : failed to wait for SIGUSR1/SIGTRAP, " + "pid = %d, n = %d, errno = %d, status = 0x%x\n", + fname, pid, n, errno, status); + } +} + +extern unsigned long current_stub_stack(void); + +void get_skas_faultinfo(int pid, struct faultinfo * fi) +{ + int err; + + if(ptrace_faultinfo){ + err = ptrace(PTRACE_FAULTINFO, pid, 0, fi); + if(err) + panic("get_skas_faultinfo - PTRACE_FAULTINFO failed, " + "errno = %d\n", errno); + + /* Special handling for i386, which has different structs */ + if (sizeof(struct ptrace_faultinfo) < sizeof(struct faultinfo)) + memset((char *)fi + sizeof(struct ptrace_faultinfo), 0, + sizeof(struct faultinfo) - + sizeof(struct ptrace_faultinfo)); + } + else { + wait_stub_done(pid, SIGSEGV, "get_skas_faultinfo"); + + /* faultinfo is prepared by the stub-segv-handler at start of + * the stub stack page. We just have to copy it. + */ + memcpy(fi, (void *)current_stub_stack(), sizeof(*fi)); + } +} + +static void handle_segv(int pid, union uml_pt_regs * regs) +{ + get_skas_faultinfo(pid, ®s->skas.faultinfo); + segv(regs->skas.faultinfo, 0, 1, NULL); +} + +/*To use the same value of using_sysemu as the caller, ask it that value (in local_using_sysemu)*/ +static void handle_trap(int pid, union uml_pt_regs *regs, int local_using_sysemu) +{ + int err, status; + + /* Mark this as a syscall */ + UPT_SYSCALL_NR(regs) = PT_SYSCALL_NR(regs->skas.regs); + + if (!local_using_sysemu) + { + err = ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_NR_OFFSET, + __NR_getpid); + if(err < 0) + panic("handle_trap - nullifying syscall failed errno = %d\n", + errno); + + err = ptrace(PTRACE_SYSCALL, pid, 0, 0); + if(err < 0) + panic("handle_trap - continuing to end of syscall failed, " + "errno = %d\n", errno); + + CATCH_EINTR(err = waitpid(pid, &status, WUNTRACED)); + if((err < 0) || !WIFSTOPPED(status) || + (WSTOPSIG(status) != SIGTRAP + 0x80)) + panic("handle_trap - failed to wait at end of syscall, " + "errno = %d, status = %d\n", errno, status); + } + + handle_syscall(regs); +} + +extern int __syscall_stub_start; +int stub_code_fd = -1; +__u64 stub_code_offset; + +static int userspace_tramp(void *stack) +{ + void *addr; + + ptrace(PTRACE_TRACEME, 0, 0, 0); + + init_new_thread_signals(1); + enable_timer(); + + if(!proc_mm){ + /* This has a pte, but it can't be mapped in with the usual + * tlb_flush mechanism because this is part of that mechanism + */ + addr = mmap64((void *) UML_CONFIG_STUB_CODE, page_size(), + PROT_EXEC, MAP_FIXED | MAP_PRIVATE, + stub_code_fd, stub_code_offset); + if(addr == MAP_FAILED){ + printk("mapping stub code failed, errno = %d\n", + errno); + exit(1); + } + + if(stack != NULL){ + int fd; + __u64 offset; + fd = phys_mapping(to_phys(stack), &offset); + addr = mmap((void *) UML_CONFIG_STUB_DATA, page_size(), + PROT_READ | PROT_WRITE, + MAP_FIXED | MAP_SHARED, fd, offset); + if(addr == MAP_FAILED){ + printk("mapping stub stack failed, " + "errno = %d\n", errno); + exit(1); + } + } + } + if(!ptrace_faultinfo){ + unsigned long v = UML_CONFIG_STUB_CODE + + (unsigned long) stub_segv_handler - + (unsigned long) &__syscall_stub_start; + + set_sigstack((void *) UML_CONFIG_STUB_DATA, page_size()); + set_handler(SIGSEGV, (void *) v, SA_ONSTACK, + SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, + SIGUSR1, -1); + } + + os_stop_process(os_getpid()); + return(0); +} + +/* Each element set once, and only accessed by a single processor anyway */ +#undef NR_CPUS +#define NR_CPUS 1 +int userspace_pid[NR_CPUS]; + +int start_userspace(unsigned long stub_stack) +{ + void *stack; + unsigned long sp; + int pid, status, n, flags; + + if ( stub_code_fd == -1 ) + stub_code_fd = phys_mapping(to_phys(&__syscall_stub_start), + &stub_code_offset); + + stack = mmap(NULL, PAGE_SIZE, PROT_READ | PROT_WRITE | PROT_EXEC, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + if(stack == MAP_FAILED) + panic("start_userspace : mmap failed, errno = %d", errno); + sp = (unsigned long) stack + PAGE_SIZE - sizeof(void *); + + flags = CLONE_FILES | SIGCHLD; + if(proc_mm) flags |= CLONE_VM; + pid = clone(userspace_tramp, (void *) sp, flags, (void *) stub_stack); + if(pid < 0) + panic("start_userspace : clone failed, errno = %d", errno); + + do { + CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED)); + if(n < 0) + panic("start_userspace : wait failed, errno = %d", + errno); + } while(WIFSTOPPED(status) && (WSTOPSIG(status) == SIGVTALRM)); + + if(!WIFSTOPPED(status) || (WSTOPSIG(status) != SIGSTOP)) + panic("start_userspace : expected SIGSTOP, got status = %d", + status); + + if (ptrace(PTRACE_OLDSETOPTIONS, pid, NULL, (void *)PTRACE_O_TRACESYSGOOD) < 0) + panic("start_userspace : PTRACE_OLDSETOPTIONS failed, errno=%d\n", + errno); + + if(munmap(stack, PAGE_SIZE) < 0) + panic("start_userspace : munmap failed, errno = %d\n", errno); + + return(pid); +} + +void userspace(union uml_pt_regs *regs) +{ + int err, status, op, pid = userspace_pid[0]; + int local_using_sysemu; /*To prevent races if using_sysemu changes under us.*/ + + while(1){ + restore_registers(pid, regs); + + /* Now we set local_using_sysemu to be used for one loop */ + local_using_sysemu = get_using_sysemu(); + + op = SELECT_PTRACE_OPERATION(local_using_sysemu, singlestepping(NULL)); + + err = ptrace(op, pid, 0, 0); + if(err) + panic("userspace - could not resume userspace process, " + "pid=%d, ptrace operation = %d, errno = %d\n", + op, errno); + + CATCH_EINTR(err = waitpid(pid, &status, WUNTRACED)); + if(err < 0) + panic("userspace - waitpid failed, errno = %d\n", + errno); + + regs->skas.is_user = 1; + save_registers(pid, regs); + UPT_SYSCALL_NR(regs) = -1; /* Assume: It's not a syscall */ + + if(WIFSTOPPED(status)){ + switch(WSTOPSIG(status)){ + case SIGSEGV: + if(PTRACE_FULL_FAULTINFO || !ptrace_faultinfo) + user_signal(SIGSEGV, regs, pid); + else handle_segv(pid, regs); + break; + case SIGTRAP + 0x80: + handle_trap(pid, regs, local_using_sysemu); + break; + case SIGTRAP: + relay_signal(SIGTRAP, regs); + break; + case SIGIO: + case SIGVTALRM: + case SIGILL: + case SIGBUS: + case SIGFPE: + case SIGWINCH: + user_signal(WSTOPSIG(status), regs, pid); + break; + default: + printk("userspace - child stopped with signal " + "%d\n", WSTOPSIG(status)); + } + again: + pid = userspace_pid[0]; + interrupt_end(); + + /* Avoid -ERESTARTSYS handling in host */ + if(PT_SYSCALL_NR_OFFSET != PT_SYSCALL_RET_OFFSET) + PT_SYSCALL_NR(regs->skas.regs) = -1; + } + } +} +#define INIT_JMP_NEW_THREAD 0 +#define INIT_JMP_REMOVE_SIGSTACK 1 +#define INIT_JMP_CALLBACK 2 +#define INIT_JMP_HALT 3 +#define INIT_JMP_REBOOT 4 + +int copy_context_skas0(unsigned long new_stack, int pid) +{ + int err; + unsigned long regs[MAX_REG_NR]; + unsigned long current_stack = current_stub_stack(); + struct stub_data *data = (struct stub_data *) current_stack; + struct stub_data *child_data = (struct stub_data *) new_stack; + __u64 new_offset; + int new_fd = phys_mapping(to_phys((void *)new_stack), &new_offset); + + /* prepare offset and fd of child's stack as argument for parent's + * and child's mmap2 calls + */ + *data = ((struct stub_data) { .offset = MMAP_OFFSET(new_offset), + .fd = new_fd, + .timer = ((struct itimerval) + { { 0, 1000000 / hz() }, + { 0, 1000000 / hz() }})}); + get_safe_registers(regs); + + /* Set parent's instruction pointer to start of clone-stub */ + regs[REGS_IP_INDEX] = UML_CONFIG_STUB_CODE + + (unsigned long) stub_clone_handler - + (unsigned long) &__syscall_stub_start; + regs[REGS_SP_INDEX] = UML_CONFIG_STUB_DATA + PAGE_SIZE - + sizeof(void *); +#ifdef __SIGNAL_FRAMESIZE + regs[REGS_SP_INDEX] -= __SIGNAL_FRAMESIZE; +#endif + err = ptrace_setregs(pid, regs); + if(err < 0) + panic("copy_context_skas0 : PTRACE_SETREGS failed, " + "pid = %d, errno = %d\n", pid, errno); + + /* set a well known return code for detection of child write failure */ + child_data->err = 12345678; + + /* Wait, until parent has finished its work: read child's pid from + * parent's stack, and check, if bad result. + */ + wait_stub_done(pid, 0, "copy_context_skas0"); + + pid = data->err; + if(pid < 0) + panic("copy_context_skas0 - stub-parent reports error %d\n", + pid); + + /* Wait, until child has finished too: read child's result from + * child's stack and check it. + */ + wait_stub_done(pid, -1, "copy_context_skas0"); + if (child_data->err != UML_CONFIG_STUB_DATA) + panic("copy_context_skas0 - stub-child reports error %d\n", + child_data->err); + + if (ptrace(PTRACE_OLDSETOPTIONS, pid, NULL, + (void *)PTRACE_O_TRACESYSGOOD) < 0) + panic("copy_context_skas0 : PTRACE_OLDSETOPTIONS failed, " + "errno = %d\n", errno); + + return pid; +} + +/* + * This is used only, if stub pages are needed, while proc_mm is + * availabl. Opening /proc/mm creates a new mm_context, which lacks + * the stub-pages. Thus, we map them using /proc/mm-fd + */ +void map_stub_pages(int fd, unsigned long code, + unsigned long data, unsigned long stack) +{ + struct proc_mm_op mmop; + int n; + + mmop = ((struct proc_mm_op) { .op = MM_MMAP, + .u = + { .mmap = + { .addr = code, + .len = PAGE_SIZE, + .prot = PROT_EXEC, + .flags = MAP_FIXED | MAP_PRIVATE, + .fd = stub_code_fd, + .offset = stub_code_offset + } } }); + n = os_write_file(fd, &mmop, sizeof(mmop)); + if(n != sizeof(mmop)) + panic("map_stub_pages : /proc/mm map for code failed, " + "err = %d\n", -n); + + if ( stack ) { + __u64 map_offset; + int map_fd = phys_mapping(to_phys((void *)stack), &map_offset); + mmop = ((struct proc_mm_op) + { .op = MM_MMAP, + .u = + { .mmap = + { .addr = data, + .len = PAGE_SIZE, + .prot = PROT_READ | PROT_WRITE, + .flags = MAP_FIXED | MAP_SHARED, + .fd = map_fd, + .offset = map_offset + } } }); + n = os_write_file(fd, &mmop, sizeof(mmop)); + if(n != sizeof(mmop)) + panic("map_stub_pages : /proc/mm map for data failed, " + "err = %d\n", -n); + } +} + +void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr, + void (*handler)(int)) +{ + unsigned long flags; + sigjmp_buf switch_buf, fork_buf; + + *switch_buf_ptr = &switch_buf; + *fork_buf_ptr = &fork_buf; + + /* Somewhat subtle - siglongjmp restores the signal mask before doing + * the longjmp. This means that when jumping from one stack to another + * when the target stack has interrupts enabled, an interrupt may occur + * on the source stack. This is bad when starting up a process because + * it's not supposed to get timer ticks until it has been scheduled. + * So, we disable interrupts around the sigsetjmp to ensure that + * they can't happen until we get back here where they are safe. + */ + flags = get_signals(); + block_signals(); + if(sigsetjmp(fork_buf, 1) == 0) + new_thread_proc(stack, handler); + + remove_sigstack(); + + set_signals(flags); +} + +void thread_wait(void *sw, void *fb) +{ + sigjmp_buf buf, **switch_buf = sw, *fork_buf; + + *switch_buf = &buf; + fork_buf = fb; + if(sigsetjmp(buf, 1) == 0) + siglongjmp(*fork_buf, INIT_JMP_REMOVE_SIGSTACK); +} + +void switch_threads(void *me, void *next) +{ + sigjmp_buf my_buf, **me_ptr = me, *next_buf = next; + + *me_ptr = &my_buf; + if(sigsetjmp(my_buf, 1) == 0) + siglongjmp(*next_buf, 1); +} + +static sigjmp_buf initial_jmpbuf; + +/* XXX Make these percpu */ +static void (*cb_proc)(void *arg); +static void *cb_arg; +static sigjmp_buf *cb_back; + +int start_idle_thread(void *stack, void *switch_buf_ptr, void **fork_buf_ptr) +{ + sigjmp_buf **switch_buf = switch_buf_ptr; + int n; + + set_handler(SIGWINCH, (__sighandler_t) sig_handler, + SA_ONSTACK | SA_RESTART, SIGUSR1, SIGIO, SIGALRM, + SIGVTALRM, -1); + + *fork_buf_ptr = &initial_jmpbuf; + n = sigsetjmp(initial_jmpbuf, 1); + switch(n){ + case INIT_JMP_NEW_THREAD: + new_thread_proc((void *) stack, new_thread_handler); + break; + case INIT_JMP_REMOVE_SIGSTACK: + remove_sigstack(); + break; + case INIT_JMP_CALLBACK: + (*cb_proc)(cb_arg); + siglongjmp(*cb_back, 1); + break; + case INIT_JMP_HALT: + kmalloc_ok = 0; + return(0); + case INIT_JMP_REBOOT: + kmalloc_ok = 0; + return(1); + default: + panic("Bad sigsetjmp return in start_idle_thread - %d\n", n); + } + siglongjmp(**switch_buf, 1); +} + +void initial_thread_cb_skas(void (*proc)(void *), void *arg) +{ + sigjmp_buf here; + + cb_proc = proc; + cb_arg = arg; + cb_back = &here; + + block_signals(); + if(sigsetjmp(here, 1) == 0) + siglongjmp(initial_jmpbuf, INIT_JMP_CALLBACK); + unblock_signals(); + + cb_proc = NULL; + cb_arg = NULL; + cb_back = NULL; +} + +void halt_skas(void) +{ + block_signals(); + siglongjmp(initial_jmpbuf, INIT_JMP_HALT); +} + +void reboot_skas(void) +{ + block_signals(); + siglongjmp(initial_jmpbuf, INIT_JMP_REBOOT); +} + +void switch_mm_skas(struct mm_id *mm_idp) +{ + int err; + +#warning need cpu pid in switch_mm_skas + if(proc_mm){ + err = ptrace(PTRACE_SWITCH_MM, userspace_pid[0], 0, + mm_idp->u.mm_fd); + if(err) + panic("switch_mm_skas - PTRACE_SWITCH_MM failed, " + "errno = %d\n", errno); + } + else userspace_pid[0] = mm_idp->u.pid; +} -- cgit v1.2.3 From 09ee011eb322c2072ec184a88763c250a5485d8b Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Wed, 18 Jan 2006 17:42:48 -0800 Subject: [PATCH] uml: eliminate some globals Stop using global variables to hold the file descriptor and offset used to map the skas0 stubs. Instead, calculate them using the page physical addresses. Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/os-Linux/skas/process.c | 28 ++++++++++++---------------- 1 file changed, 12 insertions(+), 16 deletions(-) diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c index f787854322c..b959b2618b7 100644 --- a/arch/um/os-Linux/skas/process.c +++ b/arch/um/os-Linux/skas/process.c @@ -151,8 +151,6 @@ static void handle_trap(int pid, union uml_pt_regs *regs, int local_using_sysemu } extern int __syscall_stub_start; -int stub_code_fd = -1; -__u64 stub_code_offset; static int userspace_tramp(void *stack) { @@ -167,30 +165,30 @@ static int userspace_tramp(void *stack) /* This has a pte, but it can't be mapped in with the usual * tlb_flush mechanism because this is part of that mechanism */ + int fd; + __u64 offset; + fd = phys_mapping(to_phys(&__syscall_stub_start), &offset); addr = mmap64((void *) UML_CONFIG_STUB_CODE, page_size(), - PROT_EXEC, MAP_FIXED | MAP_PRIVATE, - stub_code_fd, stub_code_offset); + PROT_EXEC, MAP_FIXED | MAP_PRIVATE, fd, offset); if(addr == MAP_FAILED){ - printk("mapping stub code failed, errno = %d\n", + printk("mapping mmap stub failed, errno = %d\n", errno); exit(1); } if(stack != NULL){ - int fd; - __u64 offset; fd = phys_mapping(to_phys(stack), &offset); addr = mmap((void *) UML_CONFIG_STUB_DATA, page_size(), PROT_READ | PROT_WRITE, MAP_FIXED | MAP_SHARED, fd, offset); if(addr == MAP_FAILED){ - printk("mapping stub stack failed, " + printk("mapping segfault stack failed, " "errno = %d\n", errno); exit(1); } } } - if(!ptrace_faultinfo){ + if(!ptrace_faultinfo && (stack != NULL)){ unsigned long v = UML_CONFIG_STUB_CODE + (unsigned long) stub_segv_handler - (unsigned long) &__syscall_stub_start; @@ -216,10 +214,6 @@ int start_userspace(unsigned long stub_stack) unsigned long sp; int pid, status, n, flags; - if ( stub_code_fd == -1 ) - stub_code_fd = phys_mapping(to_phys(&__syscall_stub_start), - &stub_code_offset); - stack = mmap(NULL, PAGE_SIZE, PROT_READ | PROT_WRITE | PROT_EXEC, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); if(stack == MAP_FAILED) @@ -306,7 +300,6 @@ void userspace(union uml_pt_regs *regs) printk("userspace - child stopped with signal " "%d\n", WSTOPSIG(status)); } - again: pid = userspace_pid[0]; interrupt_end(); @@ -395,6 +388,9 @@ void map_stub_pages(int fd, unsigned long code, { struct proc_mm_op mmop; int n; + __u64 code_offset; + int code_fd = phys_mapping(to_phys((void *) &__syscall_stub_start), + &code_offset); mmop = ((struct proc_mm_op) { .op = MM_MMAP, .u = @@ -403,8 +399,8 @@ void map_stub_pages(int fd, unsigned long code, .len = PAGE_SIZE, .prot = PROT_EXEC, .flags = MAP_FIXED | MAP_PRIVATE, - .fd = stub_code_fd, - .offset = stub_code_offset + .fd = code_fd, + .offset = code_offset } } }); n = os_write_file(fd, &mmop, sizeof(mmop)); if(n != sizeof(mmop)) -- cgit v1.2.3 From 1d7173baf286c8b720f97f119ec92be43076ebde Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Wed, 18 Jan 2006 17:42:49 -0800 Subject: [PATCH] uml: implement soft interrupts This patch implements soft interrupts. Interrupt enabling and disabling no longer map to sigprocmask. Rather, a flag is set indicating whether interrupts may be handled. If a signal comes in and interrupts are marked as OK, then it is handled normally. If interrupts are marked as off, then the signal handler simply returns after noting that a signal needs handling. When interrupts are enabled later on, this pending signals flag is checked, and the IRQ handlers are called at that point. The point of this is to reduce the cost of local_irq_save et al, since they are very much more common than the signals that they are enabling and disabling. Soft interrupts produce a speed-up of ~25% on a kernel build. Subtleties - UML uses sigsetjmp/siglongjmp to switch contexts. sigsetjmp has been wrapped in a save_flags-like macro which remembers the interrupt state at setjmp time, and restores it when it is longjmp-ed back to. The enable_signals function has to loop because the IRQ handler disables interrupts before returning. enable_signals has to return with signals enabled, and signals may come in between the disabling and the return to enable_signals. So, it loops for as long as there are pending signals, ensuring that signals are enabled when it finally returns, and that there are no pending signals that need to be dealt with. Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/include/longjmp.h | 19 ++++ arch/um/os-Linux/main.c | 12 --- arch/um/os-Linux/process.c | 30 ++---- arch/um/os-Linux/signal.c | 203 ++++++++++++++++++++++++++-------------- arch/um/os-Linux/skas/process.c | 29 +++--- arch/um/os-Linux/trap.c | 3 +- arch/um/os-Linux/uaccess.c | 4 +- arch/um/os-Linux/util.c | 1 + 8 files changed, 183 insertions(+), 118 deletions(-) create mode 100644 arch/um/include/longjmp.h diff --git a/arch/um/include/longjmp.h b/arch/um/include/longjmp.h new file mode 100644 index 00000000000..50f5df18e46 --- /dev/null +++ b/arch/um/include/longjmp.h @@ -0,0 +1,19 @@ +#ifndef __UML_LONGJMP_H +#define __UML_LONGJMP_H + +#include +#include "os.h" + +#define UML_SIGLONGJMP(buf, val) do { \ + siglongjmp(*buf, val); \ +} while(0) + +#define UML_SIGSETJMP(buf, enable) ({ \ + int n; \ + enable = get_signals(); \ + n = sigsetjmp(*buf, 1); \ + if(n != 0) \ + set_signals(enable); \ + n; }) + +#endif diff --git a/arch/um/os-Linux/main.c b/arch/um/os-Linux/main.c index 98becd18f21..2878e89a674 100644 --- a/arch/um/os-Linux/main.c +++ b/arch/um/os-Linux/main.c @@ -81,20 +81,8 @@ extern void scan_elf_aux( char **envp); int main(int argc, char **argv, char **envp) { char **new_argv; - sigset_t mask; int ret, i, err; - /* Enable all signals except SIGIO - in some environments, we can - * enter with some signals blocked - */ - - sigemptyset(&mask); - sigaddset(&mask, SIGIO); - if(sigprocmask(SIG_SETMASK, &mask, NULL) < 0){ - perror("sigprocmask"); - exit(1); - } - #ifdef UML_CONFIG_CMDLINE_ON_HOST /* Allocate memory for thread command lines */ if(argc < 2 || strlen(argv[1]) < THREAD_NAME_LEN - 1){ diff --git a/arch/um/os-Linux/process.c b/arch/um/os-Linux/process.c index 39815c6b5e4..7f5e2dac2a3 100644 --- a/arch/um/os-Linux/process.c +++ b/arch/um/os-Linux/process.c @@ -18,6 +18,7 @@ #include "process.h" #include "irq_user.h" #include "kern_util.h" +#include "longjmp.h" #define ARBITRARY_ADDR -1 #define FAILURE_PID -1 @@ -205,24 +206,13 @@ void init_new_thread_signals(int altstack) int run_kernel_thread(int (*fn)(void *), void *arg, void **jmp_ptr) { - sigjmp_buf buf; - int n; - - *jmp_ptr = &buf; - n = sigsetjmp(buf, 1); - if(n != 0) - return(n); - (*fn)(arg); - return(0); + sigjmp_buf buf; + int n, enable; + + *jmp_ptr = &buf; + n = UML_SIGSETJMP(&buf, enable); + if(n != 0) + return(n); + (*fn)(arg); + return(0); } - -/* - * 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/arch/um/os-Linux/signal.c b/arch/um/os-Linux/signal.c index 56ca95931b4..f11b3124a0c 100644 --- a/arch/um/os-Linux/signal.c +++ b/arch/um/os-Linux/signal.c @@ -20,23 +20,58 @@ #include "mode.h" #include "os.h" +/* These are the asynchronous signals. SIGVTALRM and SIGARLM are handled + * together under SIGVTALRM_BIT. SIGPROF is excluded because we want to + * be able to profile all of UML, not just the non-critical sections. If + * profiling is not thread-safe, then that is not my problem. We can disable + * profiling when SMP is enabled in that case. + */ +#define SIGIO_BIT 0 +#define SIGIO_MASK (1 << SIGIO_BIT) + +#define SIGVTALRM_BIT 1 +#define SIGVTALRM_MASK (1 << SIGVTALRM_BIT) + +#define SIGALRM_BIT 2 +#define SIGALRM_MASK (1 << SIGALRM_BIT) + +static int signals_enabled = 1; +static int pending = 0; + void sig_handler(ARCH_SIGHDLR_PARAM) { struct sigcontext *sc; + int enabled; + + /* Must be the first thing that this handler does - x86_64 stores + * the sigcontext in %rdx, and we need to save it before it has a + * chance to get trashed. + */ ARCH_GET_SIGCONTEXT(sc, sig); + + enabled = signals_enabled; + if(!enabled && (sig == SIGIO)){ + pending |= SIGIO_MASK; + return; + } + + block_signals(); + CHOOSE_MODE_PROC(sig_handler_common_tt, sig_handler_common_skas, sig, sc); + + set_signals(enabled); } extern int timer_irq_inited; -void alarm_handler(ARCH_SIGHDLR_PARAM) +static void real_alarm_handler(int sig, struct sigcontext *sc) { - struct sigcontext *sc; - - ARCH_GET_SIGCONTEXT(sc, sig); - if(!timer_irq_inited) return; + if(!timer_irq_inited){ + signals_enabled = 1; + return; + } if(sig == SIGALRM) switch_timers(0); @@ -46,6 +81,29 @@ void alarm_handler(ARCH_SIGHDLR_PARAM) if(sig == SIGALRM) switch_timers(1); + +} + +void alarm_handler(ARCH_SIGHDLR_PARAM) +{ + struct sigcontext *sc; + int enabled; + + ARCH_GET_SIGCONTEXT(sc, sig); + + enabled = signals_enabled; + if(!signals_enabled){ + if(sig == SIGVTALRM) + pending |= SIGVTALRM_MASK; + else pending |= SIGALRM_MASK; + + return; + } + + block_signals(); + + real_alarm_handler(sig, sc); + set_signals(enabled); } extern void do_boot_timer_handler(struct sigcontext * sc); @@ -53,10 +111,22 @@ extern void do_boot_timer_handler(struct sigcontext * sc); void boot_timer_handler(ARCH_SIGHDLR_PARAM) { struct sigcontext *sc; + int enabled; ARCH_GET_SIGCONTEXT(sc, sig); + enabled = signals_enabled; + if(!enabled){ + if(sig == SIGVTALRM) + pending |= SIGVTALRM_MASK; + else pending |= SIGALRM_MASK; + return; + } + + block_signals(); + do_boot_timer_handler(sc); + set_signals(enabled); } void set_sigstack(void *sig_stack, int size) @@ -83,6 +153,7 @@ void set_handler(int sig, void (*handler)(int), int flags, ...) { struct sigaction action; va_list ap; + sigset_t sig_mask; int mask; va_start(ap, flags); @@ -95,7 +166,12 @@ void set_handler(int sig, void (*handler)(int), int flags, ...) action.sa_flags = flags; action.sa_restorer = NULL; if(sigaction(sig, &action, NULL) < 0) - panic("sigaction failed"); + panic("sigaction failed - errno = %d\n", errno); + + sigemptyset(&sig_mask); + sigaddset(&sig_mask, sig); + if(sigprocmask(SIG_UNBLOCK, &sig_mask, NULL) < 0) + panic("sigprocmask failed - errno = %d\n", errno); } int change_sig(int signal, int on) @@ -108,91 +184,74 @@ int change_sig(int signal, int on) return(!sigismember(&old, signal)); } -/* Both here and in set/get_signal we don't touch SIGPROF, because we must not - * disable profiling; it's safe because the profiling code does not interact - * with the kernel code at all.*/ - -static void change_signals(int type) -{ - sigset_t mask; - - sigemptyset(&mask); - sigaddset(&mask, SIGVTALRM); - sigaddset(&mask, SIGALRM); - sigaddset(&mask, SIGIO); - if(sigprocmask(type, &mask, NULL) < 0) - panic("Failed to change signal mask - errno = %d", errno); -} - void block_signals(void) { - change_signals(SIG_BLOCK); + signals_enabled = 0; } void unblock_signals(void) { - change_signals(SIG_UNBLOCK); -} + int save_pending; -/* These are the asynchronous signals. SIGVTALRM and SIGARLM are handled - * together under SIGVTALRM_BIT. SIGPROF is excluded because we want to - * be able to profile all of UML, not just the non-critical sections. If - * profiling is not thread-safe, then that is not my problem. We can disable - * profiling when SMP is enabled in that case. - */ -#define SIGIO_BIT 0 -#define SIGVTALRM_BIT 1 - -static int enable_mask(sigset_t *mask) -{ - int sigs; + if(signals_enabled == 1) + return; - sigs = sigismember(mask, SIGIO) ? 0 : 1 << SIGIO_BIT; - sigs |= sigismember(mask, SIGVTALRM) ? 0 : 1 << SIGVTALRM_BIT; - sigs |= sigismember(mask, SIGALRM) ? 0 : 1 << SIGVTALRM_BIT; - return(sigs); + /* We loop because the IRQ handler returns with interrupts off. So, + * interrupts may have arrived and we need to re-enable them and + * recheck pending. + */ + while(1){ + /* Save and reset save_pending after enabling signals. This + * way, pending won't be changed while we're reading it. + */ + signals_enabled = 1; + + save_pending = pending; + if(save_pending == 0) + return; + + pending = 0; + + /* We have pending interrupts, so disable signals, as the + * handlers expect them off when they are called. They will + * be enabled again above. + */ + + signals_enabled = 0; + + /* Deal with SIGIO first because the alarm handler might + * schedule, leaving the pending SIGIO stranded until we come + * back here. + */ + if(save_pending & SIGIO_MASK) + CHOOSE_MODE_PROC(sig_handler_common_tt, + sig_handler_common_skas, SIGIO, NULL); + + if(save_pending & SIGALRM_MASK) + real_alarm_handler(SIGALRM, NULL); + + if(save_pending & SIGVTALRM_MASK) + real_alarm_handler(SIGVTALRM, NULL); + } } int get_signals(void) { - sigset_t mask; - - if(sigprocmask(SIG_SETMASK, NULL, &mask) < 0) - panic("Failed to get signal mask"); - return(enable_mask(&mask)); + return signals_enabled; } int set_signals(int enable) { - sigset_t mask; int ret; + if(signals_enabled == enable) + return enable; - sigemptyset(&mask); - if(enable & (1 << SIGIO_BIT)) - sigaddset(&mask, SIGIO); - if(enable & (1 << SIGVTALRM_BIT)){ - sigaddset(&mask, SIGVTALRM); - sigaddset(&mask, SIGALRM); - } - - /* This is safe - sigprocmask is guaranteed to copy locally the - * value of new_set, do his work and then, at the end, write to - * old_set. - */ - if(sigprocmask(SIG_UNBLOCK, &mask, &mask) < 0) - panic("Failed to enable signals"); - ret = enable_mask(&mask); - sigemptyset(&mask); - if((enable & (1 << SIGIO_BIT)) == 0) - sigaddset(&mask, SIGIO); - if((enable & (1 << SIGVTALRM_BIT)) == 0){ - sigaddset(&mask, SIGVTALRM); - sigaddset(&mask, SIGALRM); - } - if(sigprocmask(SIG_BLOCK, &mask, NULL) < 0) - panic("Failed to block signals"); + ret = signals_enabled; + if(enable) + unblock_signals(); + else block_signals(); - return(ret); + return ret; } void os_usr1_signal(int on) diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c index b959b2618b7..120a21c5883 100644 --- a/arch/um/os-Linux/skas/process.c +++ b/arch/um/os-Linux/skas/process.c @@ -34,6 +34,7 @@ #include "mem.h" #include "uml-config.h" #include "process.h" +#include "longjmp.h" int is_skas_winch(int pid, int fd, void *data) { @@ -433,6 +434,7 @@ void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr, { unsigned long flags; sigjmp_buf switch_buf, fork_buf; + int enable; *switch_buf_ptr = &switch_buf; *fork_buf_ptr = &fork_buf; @@ -447,7 +449,7 @@ void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr, */ flags = get_signals(); block_signals(); - if(sigsetjmp(fork_buf, 1) == 0) + if(UML_SIGSETJMP(&fork_buf, enable) == 0) new_thread_proc(stack, handler); remove_sigstack(); @@ -458,20 +460,22 @@ void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr, void thread_wait(void *sw, void *fb) { sigjmp_buf buf, **switch_buf = sw, *fork_buf; + int enable; *switch_buf = &buf; fork_buf = fb; - if(sigsetjmp(buf, 1) == 0) + if(UML_SIGSETJMP(&buf, enable) == 0) siglongjmp(*fork_buf, INIT_JMP_REMOVE_SIGSTACK); } void switch_threads(void *me, void *next) { sigjmp_buf my_buf, **me_ptr = me, *next_buf = next; + int enable; *me_ptr = &my_buf; - if(sigsetjmp(my_buf, 1) == 0) - siglongjmp(*next_buf, 1); + if(UML_SIGSETJMP(&my_buf, enable) == 0) + UML_SIGLONGJMP(next_buf, 1); } static sigjmp_buf initial_jmpbuf; @@ -484,14 +488,14 @@ static sigjmp_buf *cb_back; int start_idle_thread(void *stack, void *switch_buf_ptr, void **fork_buf_ptr) { sigjmp_buf **switch_buf = switch_buf_ptr; - int n; + int n, enable; set_handler(SIGWINCH, (__sighandler_t) sig_handler, SA_ONSTACK | SA_RESTART, SIGUSR1, SIGIO, SIGALRM, SIGVTALRM, -1); *fork_buf_ptr = &initial_jmpbuf; - n = sigsetjmp(initial_jmpbuf, 1); + n = UML_SIGSETJMP(&initial_jmpbuf, enable); switch(n){ case INIT_JMP_NEW_THREAD: new_thread_proc((void *) stack, new_thread_handler); @@ -501,7 +505,7 @@ int start_idle_thread(void *stack, void *switch_buf_ptr, void **fork_buf_ptr) break; case INIT_JMP_CALLBACK: (*cb_proc)(cb_arg); - siglongjmp(*cb_back, 1); + UML_SIGLONGJMP(cb_back, 1); break; case INIT_JMP_HALT: kmalloc_ok = 0; @@ -512,20 +516,21 @@ int start_idle_thread(void *stack, void *switch_buf_ptr, void **fork_buf_ptr) default: panic("Bad sigsetjmp return in start_idle_thread - %d\n", n); } - siglongjmp(**switch_buf, 1); + UML_SIGLONGJMP(*switch_buf, 1); } void initial_thread_cb_skas(void (*proc)(void *), void *arg) { sigjmp_buf here; + int enable; cb_proc = proc; cb_arg = arg; cb_back = &here; block_signals(); - if(sigsetjmp(here, 1) == 0) - siglongjmp(initial_jmpbuf, INIT_JMP_CALLBACK); + if(UML_SIGSETJMP(&here, enable) == 0) + UML_SIGLONGJMP(&initial_jmpbuf, INIT_JMP_CALLBACK); unblock_signals(); cb_proc = NULL; @@ -536,13 +541,13 @@ void initial_thread_cb_skas(void (*proc)(void *), void *arg) void halt_skas(void) { block_signals(); - siglongjmp(initial_jmpbuf, INIT_JMP_HALT); + UML_SIGLONGJMP(&initial_jmpbuf, INIT_JMP_HALT); } void reboot_skas(void) { block_signals(); - siglongjmp(initial_jmpbuf, INIT_JMP_REBOOT); + UML_SIGLONGJMP(&initial_jmpbuf, INIT_JMP_REBOOT); } void switch_mm_skas(struct mm_id *mm_idp) diff --git a/arch/um/os-Linux/trap.c b/arch/um/os-Linux/trap.c index 321e1c8e227..a9f6b26f982 100644 --- a/arch/um/os-Linux/trap.c +++ b/arch/um/os-Linux/trap.c @@ -10,6 +10,7 @@ #include "user_util.h" #include "os.h" #include "mode.h" +#include "longjmp.h" void usr2_handler(int sig, union uml_pt_regs *regs) { @@ -36,5 +37,5 @@ void do_longjmp(void *b, int val) { sigjmp_buf *buf = b; - siglongjmp(*buf, val); + UML_SIGLONGJMP(buf, val); } diff --git a/arch/um/os-Linux/uaccess.c b/arch/um/os-Linux/uaccess.c index 38d710158c3..166fb66995d 100644 --- a/arch/um/os-Linux/uaccess.c +++ b/arch/um/os-Linux/uaccess.c @@ -6,6 +6,7 @@ #include #include +#include "longjmp.h" unsigned long __do_user_copy(void *to, const void *from, int n, void **fault_addr, void **fault_catcher, @@ -13,10 +14,11 @@ unsigned long __do_user_copy(void *to, const void *from, int n, int n), int *faulted_out) { unsigned long *faddrp = (unsigned long *) fault_addr, ret; + int enable; sigjmp_buf jbuf; *fault_catcher = &jbuf; - if(sigsetjmp(jbuf, 1) == 0){ + if(UML_SIGSETJMP(&jbuf, enable) == 0){ (*op)(to, from, n); ret = 0; *faulted_out = 0; diff --git a/arch/um/os-Linux/util.c b/arch/um/os-Linux/util.c index d224434d561..e32065e2fdc 100644 --- a/arch/um/os-Linux/util.c +++ b/arch/um/os-Linux/util.c @@ -30,6 +30,7 @@ #include "ptrace_user.h" #include "uml-config.h" #include "os.h" +#include "longjmp.h" void stack_protections(unsigned long address) { -- cgit v1.2.3 From c83d4635ee8c8fe16046ff6cabcff708be16df75 Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Wed, 18 Jan 2006 17:42:50 -0800 Subject: [PATCH] uml: use setjmp/longjmp instead of sigsetjmp/siglongjmp Now that we are doing soft interrupts, there's no point in using sigsetjmp and siglongjmp. Using setjmp and longjmp saves a sigprocmask on every jump. Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/include/longjmp.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/um/include/longjmp.h b/arch/um/include/longjmp.h index 50f5df18e46..018b3819ab0 100644 --- a/arch/um/include/longjmp.h +++ b/arch/um/include/longjmp.h @@ -5,13 +5,13 @@ #include "os.h" #define UML_SIGLONGJMP(buf, val) do { \ - siglongjmp(*buf, val); \ + longjmp(*buf, val); \ } while(0) #define UML_SIGSETJMP(buf, enable) ({ \ int n; \ enable = get_signals(); \ - n = sigsetjmp(*buf, 1); \ + n = setjmp(*buf); \ if(n != 0) \ set_signals(enable); \ n; }) -- cgit v1.2.3 From 097fdf06c63741e6ac1a4e01c2255861dd0a1c49 Mon Sep 17 00:00:00 2001 From: Bodo Stroesser Date: Wed, 18 Jan 2006 17:42:51 -0800 Subject: [PATCH] uml: TT mode softint fixes Some fixes to make softints work in tt mode. Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/kernel/time_kern.c | 5 +++-- arch/um/kernel/tt/trap_user.c | 5 +++-- 2 files changed, 6 insertions(+), 4 deletions(-) diff --git a/arch/um/kernel/time_kern.c b/arch/um/kernel/time_kern.c index 6fa4e1b892b..3c7626cdba4 100644 --- a/arch/um/kernel/time_kern.c +++ b/arch/um/kernel/time_kern.c @@ -187,8 +187,9 @@ void timer_handler(int sig, union uml_pt_regs *regs) { local_irq_disable(); irq_enter(); - update_process_times(CHOOSE_MODE(user_context(UPT_SP(regs)), - (regs)->skas.is_user)); + update_process_times(CHOOSE_MODE( + (UPT_SC(regs) && user_context(UPT_SP(regs))), + (regs)->skas.is_user)); irq_exit(); local_irq_enable(); if(current_thread->cpu == 0) diff --git a/arch/um/kernel/tt/trap_user.c b/arch/um/kernel/tt/trap_user.c index a414c529fbc..b5d9d64d91e 100644 --- a/arch/um/kernel/tt/trap_user.c +++ b/arch/um/kernel/tt/trap_user.c @@ -18,7 +18,7 @@ void sig_handler_common_tt(int sig, void *sc_ptr) { struct sigcontext *sc = sc_ptr; struct tt_regs save_regs, *r; - int save_errno = errno, is_user; + int save_errno = errno, is_user = 0; void (*handler)(int, union uml_pt_regs *); /* This is done because to allow SIGSEGV to be delivered inside a SEGV @@ -35,7 +35,8 @@ void sig_handler_common_tt(int sig, void *sc_ptr) GET_FAULTINFO_FROM_SC(r->faultinfo, sc); } save_regs = *r; - is_user = user_context(SC_SP(sc)); + if (sc) + is_user = user_context(SC_SP(sc)); r->sc = sc; if(sig != SIGUSR2) r->syscall = -1; -- cgit v1.2.3 From 3b948068b84b9759cdf0965abf3074dcb9230e98 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 18 Jan 2006 17:42:52 -0800 Subject: [PATCH] uml: remove leftover from patch revertal I added this line to share this file with UML, but now it's no longer shared so remove this useless leftover. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Acked-by: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-i386/futex.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/asm-i386/futex.h b/include/asm-i386/futex.h index e7a271d3930..44b9db80647 100644 --- a/include/asm-i386/futex.h +++ b/include/asm-i386/futex.h @@ -61,7 +61,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) if (op == FUTEX_OP_SET) __futex_atomic_op1("xchgl %0, %2", ret, oldval, uaddr, oparg); else { -#if !defined(CONFIG_X86_BSWAP) && !defined(CONFIG_UML) +#ifndef CONFIG_X86_BSWAP if (boot_cpu_data.x86 == 3) ret = -ENOSYS; else -- cgit v1.2.3 From c42791b6ec5453cd7910eac7bfdd88f27173f81c Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 18 Jan 2006 17:42:53 -0800 Subject: [PATCH] uml: make daemon transport behave properly Avoid uninitialized data in the daemon_data structure. I used this transport before doing proper setup before-hand, and I got some very nice SLAB corruption due to freeing crap pointers. So just make sure to clear everything when appropriate. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/daemon_kern.c | 4 ++++ arch/um/drivers/daemon_user.c | 6 ++++++ 2 files changed, 10 insertions(+) diff --git a/arch/um/drivers/daemon_kern.c b/arch/um/drivers/daemon_kern.c index 30d285b266a..507e3cbac9d 100644 --- a/arch/um/drivers/daemon_kern.c +++ b/arch/um/drivers/daemon_kern.c @@ -31,6 +31,10 @@ void daemon_init(struct net_device *dev, void *data) dpri->fd = -1; dpri->control = -1; dpri->dev = dev; + /* We will free this pointer. If it contains crap we're burned. */ + dpri->ctl_addr = NULL; + dpri->data_addr = NULL; + dpri->local_addr = NULL; printk("daemon backend (uml_switch version %d) - %s:%s", SWITCH_VERSION, dpri->sock_type, dpri->ctl_sock); diff --git a/arch/um/drivers/daemon_user.c b/arch/um/drivers/daemon_user.c index 1bb085b2824..c944265955e 100644 --- a/arch/um/drivers/daemon_user.c +++ b/arch/um/drivers/daemon_user.c @@ -158,10 +158,16 @@ static void daemon_remove(void *data) struct daemon_data *pri = data; os_close_file(pri->fd); + pri->fd = -1; os_close_file(pri->control); + pri->control = -1; + kfree(pri->data_addr); + pri->data_addr = NULL; kfree(pri->ctl_addr); + pri->ctl_addr = NULL; kfree(pri->local_addr); + pri->local_addr = NULL; } int daemon_user_write(int fd, void *buf, int len, struct daemon_data *pri) -- cgit v1.2.3 From e56a78855a4f72fc658bfd21d08939dd6e09fa4c Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 18 Jan 2006 17:42:55 -0800 Subject: [PATCH] uml: networking - clear transport-specific structure Pre-clear transport-specific private structure before passing it down. In fact, I just got a slab corruption and kernel panic on exit because kfree() was called on a pointer which probably was never allocated, BUT hadn't been set to NULL by the driver. As the code is full of such errors, I've decided for now to go the safe way (we're talking about drivers), and to do the simple thing. I'm also starting to fix drivers, and already sent a patch for the daemon transport. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/net_kern.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/um/drivers/net_kern.c b/arch/um/drivers/net_kern.c index fb1f9fb9b87..f3442ce29a2 100644 --- a/arch/um/drivers/net_kern.c +++ b/arch/um/drivers/net_kern.c @@ -317,6 +317,11 @@ static int eth_configure(int n, void *init, char *mac, return 1; } + lp = dev->priv; + /* This points to the transport private data. It's still clear, but we + * must memset it to 0 *now*. Let's help the drivers. */ + memset(lp, 0, size); + /* sysfs register */ if (!driver_registered) { platform_driver_register(¨_net_driver); @@ -358,7 +363,6 @@ static int eth_configure(int n, void *init, char *mac, free_netdev(dev); return 1; } - lp = dev->priv; /* lp.user is the first four bytes of the transport data, which * has already been initialized. This structure assignment will -- cgit v1.2.3 From 71c8d4c3aad3132765d30b05dce98bb8a9508f02 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 18 Jan 2006 17:42:56 -0800 Subject: [PATCH] uml: fix spinlock recursion and sleep-inside-spinlock in error path In this error path, when the interface has had a problem, we call dev_close(), which is disallowed for two reasons: *) takes again the UML internal spinlock, inside the ->stop method of this device *) can be called in process context only, while we're in interrupt context. I've also thought that calling dev_close() may be a wrong policy to follow, but it's not up to me to decide that. However, we may end up with multiple dev_close() queued on the same device. But the initial test for (dev->flags & IFF_UP) makes this harmless, though - and dev_close() is supposed to care about races with itself. So there's no harm in delaying the shutdown, IMHO. Something to mark the interface as "going to shutdown" would be appreciated, but dev_deactivate has the same problems as dev_close(), so we can't use it either. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/net_kern.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/arch/um/drivers/net_kern.c b/arch/um/drivers/net_kern.c index f3442ce29a2..8ebb2241ad4 100644 --- a/arch/um/drivers/net_kern.c +++ b/arch/um/drivers/net_kern.c @@ -68,6 +68,11 @@ static int uml_net_rx(struct net_device *dev) return pkt_len; } +static void uml_dev_close(void* dev) +{ + dev_close( (struct net_device *) dev); +} + irqreturn_t uml_net_interrupt(int irq, void *dev_id, struct pt_regs *regs) { struct net_device *dev = dev_id; @@ -80,15 +85,21 @@ irqreturn_t uml_net_interrupt(int irq, void *dev_id, struct pt_regs *regs) spin_lock(&lp->lock); while((err = uml_net_rx(dev)) > 0) ; if(err < 0) { + DECLARE_WORK(close_work, uml_dev_close, dev); printk(KERN_ERR "Device '%s' read returned %d, shutting it down\n", dev->name, err); - dev_close(dev); + /* dev_close can't be called in interrupt context, and takes + * again lp->lock. + * And dev_close() can be safely called multiple times on the + * same device, since it tests for (dev->flags & IFF_UP). So + * there's no harm in delaying the device shutdown. */ + schedule_work(&close_work); goto out; } reactivate_fd(lp->fd, UM_ETH_IRQ); - out: +out: spin_unlock(&lp->lock); return(IRQ_HANDLED); } -- cgit v1.2.3 From b6a2b13778873bd9edd0b4a7d24a7bd730369021 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 18 Jan 2006 17:42:57 -0800 Subject: [PATCH] uml: sigio code - reduce spinlock hold time In a previous patch I shifted an allocation to being atomic. In this patch, a better but more intrusive solution is implemented, i.e. hold the lock only when really needing it, especially not over pipe operations, nor over the culprit allocation. Additionally, while at it, add a missing kfree in the failure path, and make sure that if we fail in forking, write_sigio_pid is -1 and not, say, -ENOMEM. And fix whitespace, at least for things I was touching anyway. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/kernel/sigio_user.c | 83 +++++++++++++++++++++++++++++++-------------- 1 file changed, 58 insertions(+), 25 deletions(-) diff --git a/arch/um/kernel/sigio_user.c b/arch/um/kernel/sigio_user.c index 62e5cfdf218..f7b18e157d3 100644 --- a/arch/um/kernel/sigio_user.c +++ b/arch/um/kernel/sigio_user.c @@ -337,70 +337,103 @@ int ignore_sigio_fd(int fd) return(err); } -static int setup_initial_poll(int fd) +static struct pollfd* setup_initial_poll(int fd) { struct pollfd *p; - p = um_kmalloc_atomic(sizeof(struct pollfd)); - if(p == NULL){ + p = um_kmalloc(sizeof(struct pollfd)); + if (p == NULL) { printk("setup_initial_poll : failed to allocate poll\n"); - return(-1); + return NULL; } *p = ((struct pollfd) { .fd = fd, .events = POLLIN, .revents = 0 }); - current_poll = ((struct pollfds) { .poll = p, - .used = 1, - .size = 1 }); - return(0); + return p; } void write_sigio_workaround(void) { unsigned long stack; + struct pollfd *p; int err; + int l_write_sigio_fds[2]; + int l_sigio_private[2]; + int l_write_sigio_pid; + /* We call this *tons* of times - and most ones we must just fail. */ sigio_lock(); - if(write_sigio_pid != -1) - goto out; + l_write_sigio_pid = write_sigio_pid; + sigio_unlock(); - err = os_pipe(write_sigio_fds, 1, 1); + if (l_write_sigio_pid != -1) + return; + + err = os_pipe(l_write_sigio_fds, 1, 1); if(err < 0){ printk("write_sigio_workaround - os_pipe 1 failed, " "err = %d\n", -err); - goto out; + return; } - err = os_pipe(sigio_private, 1, 1); + err = os_pipe(l_sigio_private, 1, 1); if(err < 0){ - printk("write_sigio_workaround - os_pipe 2 failed, " + printk("write_sigio_workaround - os_pipe 1 failed, " "err = %d\n", -err); goto out_close1; } - if(setup_initial_poll(sigio_private[1])) + + p = setup_initial_poll(l_sigio_private[1]); + if(!p) goto out_close2; - write_sigio_pid = run_helper_thread(write_sigio_thread, NULL, + sigio_lock(); + + /* Did we race? Don't try to optimize this, please, it's not so likely + * to happen, and no more than once at the boot. */ + if(write_sigio_pid != -1) + goto out_unlock; + + write_sigio_pid = run_helper_thread(write_sigio_thread, NULL, CLONE_FILES | CLONE_VM, &stack, 0); - if(write_sigio_pid < 0) goto out_close2; + if (write_sigio_pid < 0) + goto out_clear; - if(write_sigio_irq(write_sigio_fds[0])) + if (write_sigio_irq(l_write_sigio_fds[0])) goto out_kill; - out: + /* Success, finally. */ + memcpy(write_sigio_fds, l_write_sigio_fds, sizeof(l_write_sigio_fds)); + memcpy(sigio_private, l_sigio_private, sizeof(l_sigio_private)); + + current_poll = ((struct pollfds) { .poll = p, + .used = 1, + .size = 1 }); + sigio_unlock(); return; out_kill: - os_kill_process(write_sigio_pid, 1); + l_write_sigio_pid = write_sigio_pid; write_sigio_pid = -1; + sigio_unlock(); + /* Going to call waitpid, avoid holding the lock. */ + os_kill_process(l_write_sigio_pid, 1); + goto out_free; + + out_clear: + write_sigio_pid = -1; + out_unlock: + sigio_unlock(); + out_free: + kfree(p); out_close2: - os_close_file(sigio_private[0]); - os_close_file(sigio_private[1]); + os_close_file(l_sigio_private[0]); + os_close_file(l_sigio_private[1]); out_close1: - os_close_file(write_sigio_fds[0]); - os_close_file(write_sigio_fds[1]); - sigio_unlock(); + os_close_file(l_write_sigio_fds[0]); + os_close_file(l_write_sigio_fds[1]); + return; } int read_sigio_fd(int fd) -- cgit v1.2.3 From b63162939cd797c8269964ce856ed1f2fec5f70e Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 18 Jan 2006 17:42:58 -0800 Subject: [PATCH] uml: avoid malloc to sleep in atomic sections Ugly trick to help make malloc not sleeping - we can't do anything else. But this is not yet optimal, since spinlock don't trigger in_atomic() when preemption is disabled. Also, even if ugly, this was already used in one place, and was even more bogus. Fix it. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/include/kern_util.h | 4 +++- arch/um/include/user.h | 1 + arch/um/kernel/process_kern.c | 21 +++++++++++++-------- arch/um/os-Linux/helper.c | 4 ++-- 4 files changed, 19 insertions(+), 11 deletions(-) diff --git a/arch/um/include/kern_util.h b/arch/um/include/kern_util.h index 8f4e46d677a..c649108a9e9 100644 --- a/arch/um/include/kern_util.h +++ b/arch/um/include/kern_util.h @@ -120,8 +120,10 @@ extern void machine_halt(void); extern int is_syscall(unsigned long addr); extern void arch_switch(void); extern void free_irq(unsigned int, void *); -extern int um_in_interrupt(void); extern int cpu(void); + +/* Are we disallowed to sleep? Used to choose between GFP_KERNEL and GFP_ATOMIC. */ +extern int __cant_sleep(void); extern void segv_handler(int sig, union uml_pt_regs *regs); extern void sigio_handler(int sig, union uml_pt_regs *regs); diff --git a/arch/um/include/user.h b/arch/um/include/user.h index 0f865ef4691..91b0ac4ad88 100644 --- a/arch/um/include/user.h +++ b/arch/um/include/user.h @@ -18,6 +18,7 @@ extern int open_gdb_chan(void); extern unsigned long strlcpy(char *, const char *, unsigned long); extern unsigned long strlcat(char *, const char *, unsigned long); extern void *um_vmalloc(int size); +extern void *um_vmalloc_atomic(int size); extern void vfree(void *ptr); #endif diff --git a/arch/um/kernel/process_kern.c b/arch/um/kernel/process_kern.c index e167cf0a71f..3113cab8675 100644 --- a/arch/um/kernel/process_kern.c +++ b/arch/um/kernel/process_kern.c @@ -287,17 +287,27 @@ EXPORT_SYMBOL(disable_hlt); void *um_kmalloc(int size) { - return(kmalloc(size, GFP_KERNEL)); + return kmalloc(size, GFP_KERNEL); } void *um_kmalloc_atomic(int size) { - return(kmalloc(size, GFP_ATOMIC)); + return kmalloc(size, GFP_ATOMIC); } void *um_vmalloc(int size) { - return(vmalloc(size)); + return vmalloc(size); +} + +void *um_vmalloc_atomic(int size) +{ + return __vmalloc(size, GFP_ATOMIC | __GFP_HIGHMEM, PAGE_KERNEL); +} + +int __cant_sleep(void) { + return in_atomic() || irqs_disabled() || in_interrupt(); + /* Is in_interrupt() really needed? */ } unsigned long get_fault_addr(void) @@ -369,11 +379,6 @@ int smp_sigio_handler(void) return(0); } -int um_in_interrupt(void) -{ - return(in_interrupt()); -} - int cpu(void) { return(current_thread->cpu); diff --git a/arch/um/os-Linux/helper.c b/arch/um/os-Linux/helper.c index 36cc8475bcd..6490a4ff40a 100644 --- a/arch/um/os-Linux/helper.c +++ b/arch/um/os-Linux/helper.c @@ -60,7 +60,7 @@ int run_helper(void (*pre_exec)(void *), void *pre_data, char **argv, if((stack_out != NULL) && (*stack_out != 0)) stack = *stack_out; - else stack = alloc_stack(0, um_in_interrupt()); + else stack = alloc_stack(0, __cant_sleep()); if(stack == 0) return(-ENOMEM); @@ -124,7 +124,7 @@ int run_helper_thread(int (*proc)(void *), void *arg, unsigned int flags, unsigned long stack, sp; int pid, status, err; - stack = alloc_stack(stack_order, um_in_interrupt()); + stack = alloc_stack(stack_order, __cant_sleep()); if(stack == 0) return(-ENOMEM); sp = stack + (page_size() << stack_order) - sizeof(void *); -- cgit v1.2.3 From ce2d2aedcc3ca582fed90f44970e8b3e4f006a7d Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 18 Jan 2006 17:42:59 -0800 Subject: [PATCH] uml: arch Kconfig menu cleanups *) mark as "EXPERIMENTAL" various items that either aren't very stable or that are actively crashing the setup of users which don't really need them (i.e. HIGHMEM and 3-level pagetables on x86 - nobody needs either, everybody reports "I'm using it and getting trouble"). *) move net/Kconfig near to the rest of network configurations, and drivers/block/Kconfig near "Block layer" submenu. *) it's useless and doesn't work well to force NETDEVICES on and to disable the prompt like it's done. Better remove the attempt, and change that to a simple "default y if UML". *) drop the warning about "report problems about HPPFS" - it's redundant anyway, as that's the usual procedure, and HPPFS users are especially technical (i.e. they know reporting bugs is _good_). Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/Kconfig | 27 ++++++++++++++------------- arch/um/Kconfig.i386 | 6 +++++- drivers/net/Kconfig | 1 + 3 files changed, 20 insertions(+), 14 deletions(-) diff --git a/arch/um/Kconfig b/arch/um/Kconfig index 8ff3bcbce5f..5982fe2753e 100644 --- a/arch/um/Kconfig +++ b/arch/um/Kconfig @@ -143,6 +143,7 @@ config HOSTFS config HPPFS tristate "HoneyPot ProcFS (EXPERIMENTAL)" + depends on EXPERIMENTAL help hppfs (HoneyPot ProcFS) is a filesystem which allows UML /proc entries to be overridden, removed, or fabricated from the host. @@ -155,10 +156,6 @@ config HPPFS You only need this if you are setting up a UML honeypot. Otherwise, it is safe to say 'N' here. - If you are actively using it, please report any problems, since it's - getting fixed. In this moment, it is experimental on 2.6 (it works on - 2.4). - config MCONSOLE bool "Management console" default y @@ -243,8 +240,16 @@ config NEST_LEVEL Only change this if you are running nested UMLs. config HIGHMEM - bool "Highmem support" - depends on !64BIT + bool "Highmem support (EXPERIMENTAL)" + depends on !64BIT && EXPERIMENTAL + default n + help + This was used to allow UML to run with big amounts of memory. + Currently it is unstable, so if unsure say N. + + To use big amounts of memory, it is recommended to disable TT mode (i.e. + CONFIG_MODE_TT) and enable static linking (i.e. CONFIG_STATIC_LINK) - + this should allow the guest to use up to 2.75G of memory. config KERNEL_STACK_ORDER int "Kernel stack size order" @@ -269,17 +274,13 @@ endmenu source "init/Kconfig" -source "net/Kconfig" - -source "drivers/base/Kconfig" +source "drivers/block/Kconfig" source "arch/um/Kconfig.char" -source "drivers/block/Kconfig" +source "drivers/base/Kconfig" -config NETDEVICES - bool - default NET +source "net/Kconfig" source "arch/um/Kconfig.net" diff --git a/arch/um/Kconfig.i386 b/arch/um/Kconfig.i386 index c71b39a677a..ef79ed25aec 100644 --- a/arch/um/Kconfig.i386 +++ b/arch/um/Kconfig.i386 @@ -22,13 +22,17 @@ config TOP_ADDR default 0x80000000 if HOST_2G_2G config 3_LEVEL_PGTABLES - bool "Three-level pagetables" + bool "Three-level pagetables (EXPERIMENTAL)" default n + depends on EXPERIMENTAL help Three-level pagetables will let UML have more than 4G of physical memory. All the memory that can't be mapped directly will be treated as high memory. + However, this it experimental on 32-bit architectures, so if unsure say + N (on x86-64 it's automatically enabled, instead, as it's safe there). + config STUB_CODE hex default 0xbfffe000 diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 0c69918671c..626508afe1b 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -7,6 +7,7 @@ menu "Network device support" config NETDEVICES depends on NET + default y if UML bool "Network device support" ---help--- You can say N here if you don't intend to connect your Linux box to -- cgit v1.2.3 From 4833aff757b747b51b878a13f3a6e2b1a2abb619 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 18 Jan 2006 17:43:00 -0800 Subject: [PATCH] uml: allow again to move backing file and to override saved location When the user specifies both a COW file and its backing file, if the previous backing file is not found, currently UML tries again to use it and fails. This can be corrected by changing same_backing_files() return value in that case, so that the caller will try to change the COW file to point to the new location, as already done in other cases. Additionally, given the change in the meaning of the func, change its name, invert its return value, so all values are inverted except when stat(from_cow,&buf2) fails. And add some comments and two minor bugfixes - remove a fd leak (return err rather than goto out) and a repeated check. Tested well. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/ubd_kern.c | 28 +++++++++++++++------------- 1 file changed, 15 insertions(+), 13 deletions(-) diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index 7696f8d2d89..c171d0aa5e9 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -1103,31 +1103,33 @@ static int ubd_ioctl(struct inode * inode, struct file * file, return(-EINVAL); } -static int same_backing_files(char *from_cmdline, char *from_cow, char *cow) +static int path_requires_switch(char *from_cmdline, char *from_cow, char *cow) { struct uml_stat buf1, buf2; int err; - if(from_cmdline == NULL) return(1); - if(!strcmp(from_cmdline, from_cow)) return(1); + if(from_cmdline == NULL) + return 0; + if(!strcmp(from_cmdline, from_cow)) + return 0; err = os_stat_file(from_cmdline, &buf1); if(err < 0){ printk("Couldn't stat '%s', err = %d\n", from_cmdline, -err); - return(1); + return 0; } err = os_stat_file(from_cow, &buf2); if(err < 0){ printk("Couldn't stat '%s', err = %d\n", from_cow, -err); - return(1); + return 1; } if((buf1.ust_dev == buf2.ust_dev) && (buf1.ust_ino == buf2.ust_ino)) - return(1); + return 0; printk("Backing file mismatch - \"%s\" requested,\n" "\"%s\" specified in COW header of \"%s\"\n", from_cmdline, from_cow, cow); - return(0); + return 1; } static int backing_file_mismatch(char *file, __u64 size, time_t mtime) @@ -1189,7 +1191,7 @@ int open_ubd_file(char *file, struct openflags *openflags, unsigned long long size; __u32 version, align; char *backing_file; - int fd, err, sectorsize, same, mode = 0644; + int fd, err, sectorsize, asked_switch, mode = 0644; fd = os_open_file(file, *openflags, mode); if(fd < 0){ @@ -1209,6 +1211,7 @@ int open_ubd_file(char *file, struct openflags *openflags, goto out_close; } + /* Succesful return case! */ if(backing_file_out == NULL) return(fd); err = read_cow_header(file_reader, &fd, &version, &backing_file, &mtime, @@ -1220,17 +1223,16 @@ int open_ubd_file(char *file, struct openflags *openflags, } if(err) return(fd); - if(backing_file_out == NULL) return(fd); - - same = same_backing_files(*backing_file_out, backing_file, file); + asked_switch = path_requires_switch(*backing_file_out, backing_file, file); - if(!same && !backing_file_mismatch(*backing_file_out, size, mtime)){ + /* Allow switching only if no mismatch. */ + if (asked_switch && !backing_file_mismatch(*backing_file_out, size, mtime)) { printk("Switching backing file to '%s'\n", *backing_file_out); err = write_cow_header(file, fd, *backing_file_out, sectorsize, align, &size); if(err){ printk("Switch failed, errno = %d\n", -err); - return(err); + goto out_close; } } else { -- cgit v1.2.3 From a374a48ffbd7d1ffd40d1d30d2751df159a1aca7 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 18 Jan 2006 17:43:01 -0800 Subject: [PATCH] uml ubd code: fix a bit of whitespace Correct a bit of whitespace problems while working here. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/ubd_kern.c | 31 +++++++++++++++++-------------- 1 file changed, 17 insertions(+), 14 deletions(-) diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index c171d0aa5e9..101efd26d46 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -1194,15 +1194,16 @@ int open_ubd_file(char *file, struct openflags *openflags, int fd, err, sectorsize, asked_switch, mode = 0644; fd = os_open_file(file, *openflags, mode); - if(fd < 0){ - if((fd == -ENOENT) && (create_cow_out != NULL)) + if (fd < 0) { + if ((fd == -ENOENT) && (create_cow_out != NULL)) *create_cow_out = 1; - if(!openflags->w || - ((fd != -EROFS) && (fd != -EACCES))) return(fd); + if (!openflags->w || + ((fd != -EROFS) && (fd != -EACCES))) + return fd; openflags->w = 0; fd = os_open_file(file, *openflags, mode); - if(fd < 0) - return(fd); + if (fd < 0) + return fd; } err = os_lock_file(fd, openflags->w); @@ -1212,7 +1213,8 @@ int open_ubd_file(char *file, struct openflags *openflags, } /* Succesful return case! */ - if(backing_file_out == NULL) return(fd); + if(backing_file_out == NULL) + return(fd); err = read_cow_header(file_reader, &fd, &version, &backing_file, &mtime, &size, §orsize, &align, bitmap_offset_out); @@ -1221,7 +1223,8 @@ int open_ubd_file(char *file, struct openflags *openflags, "errno = %d\n", file, -err); goto out_close; } - if(err) return(fd); + if(err) + return(fd); asked_switch = path_requires_switch(*backing_file_out, backing_file, file); @@ -1230,24 +1233,24 @@ int open_ubd_file(char *file, struct openflags *openflags, printk("Switching backing file to '%s'\n", *backing_file_out); err = write_cow_header(file, fd, *backing_file_out, sectorsize, align, &size); - if(err){ + if (err) { printk("Switch failed, errno = %d\n", -err); goto out_close; } - } - else { + } else { *backing_file_out = backing_file; err = backing_file_mismatch(*backing_file_out, size, mtime); - if(err) goto out_close; + if (err) + goto out_close; } cow_sizes(version, size, sectorsize, align, *bitmap_offset_out, bitmap_len_out, data_offset_out); - return(fd); + return fd; out_close: os_close_file(fd); - return(err); + return err; } int create_cow_file(char *cow_file, char *backing_file, struct openflags flags, -- cgit v1.2.3 From 147b31cf09ee493aa71c87c0dd2eef74b6b2aeba Mon Sep 17 00:00:00 2001 From: Eric Van Hensbergen Date: Wed, 18 Jan 2006 17:43:02 -0800 Subject: [PATCH] v9fs: add readpage support v9fs mmap support was originally removed from v9fs at Al Viro's request, but recently there have been requests from folks who want readpage functionality (primarily to enable execution of files mounted via 9P). This patch adds readpage support (but not writepage which contained most of the objectionable code). It passes fsx-linux (and other regressions) so it should be relatively safe. Signed-off-by: Eric Van Hensbergen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/9p/Makefile | 1 + fs/9p/v9fs_vfs.h | 1 + fs/9p/vfs_addr.c | 109 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ fs/9p/vfs_file.c | 4 ++ fs/9p/vfs_inode.c | 1 + 5 files changed, 116 insertions(+) create mode 100644 fs/9p/vfs_addr.c diff --git a/fs/9p/Makefile b/fs/9p/Makefile index 3d023089707..2f4ce43f7b6 100644 --- a/fs/9p/Makefile +++ b/fs/9p/Makefile @@ -8,6 +8,7 @@ obj-$(CONFIG_9P_FS) := 9p2000.o conv.o \ vfs_super.o \ vfs_inode.o \ + vfs_addr.o \ vfs_file.o \ vfs_dir.o \ vfs_dentry.o \ diff --git a/fs/9p/v9fs_vfs.h b/fs/9p/v9fs_vfs.h index c78502ad00e..69cf2905dc9 100644 --- a/fs/9p/v9fs_vfs.h +++ b/fs/9p/v9fs_vfs.h @@ -39,6 +39,7 @@ */ extern struct file_system_type v9fs_fs_type; +extern struct address_space_operations v9fs_addr_operations; extern struct file_operations v9fs_file_operations; extern struct file_operations v9fs_dir_operations; extern struct dentry_operations v9fs_dentry_operations; diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c new file mode 100644 index 00000000000..8100fb5171b --- /dev/null +++ b/fs/9p/vfs_addr.c @@ -0,0 +1,109 @@ +/* + * linux/fs/9p/vfs_addr.c + * + * This file contians vfs address (mmap) ops for 9P2000. + * + * Copyright (C) 2005 by Eric Van Hensbergen + * Copyright (C) 2002 by Ron Minnich + * + * 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: + * Free Software Foundation + * 51 Franklin Street, Fifth Floor + * Boston, MA 02111-1301 USA + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "debug.h" +#include "v9fs.h" +#include "9p.h" +#include "v9fs_vfs.h" +#include "fid.h" + +/** + * v9fs_vfs_readpage - read an entire page in from 9P + * + * @file: file being read + * @page: structure to page + * + */ + +static int v9fs_vfs_readpage(struct file *filp, struct page *page) +{ + char *buffer = NULL; + int retval = -EIO; + loff_t offset = page_offset(page); + int count = PAGE_CACHE_SIZE; + struct inode *inode = filp->f_dentry->d_inode; + struct v9fs_session_info *v9ses = v9fs_inode2v9ses(inode); + int rsize = v9ses->maxdata - V9FS_IOHDRSZ; + struct v9fs_fid *v9f = filp->private_data; + struct v9fs_fcall *fcall = NULL; + int fid = v9f->fid; + int total = 0; + int result = 0; + + buffer = kmap(page); + do { + if (count < rsize) + rsize = count; + + result = v9fs_t_read(v9ses, fid, offset, rsize, &fcall); + + if (result < 0) { + printk(KERN_ERR "v9fs_t_read returned %d\n", + result); + + kfree(fcall); + goto UnmapAndUnlock; + } else + offset += result; + + memcpy(buffer, fcall->params.rread.data, result); + + count -= result; + buffer += result; + total += result; + + kfree(fcall); + + if (result < rsize) + break; + } while (count); + + memset(buffer, 0, count); + flush_dcache_page(page); + SetPageUptodate(page); + retval = 0; + +UnmapAndUnlock: + kunmap(page); + unlock_page(page); + return retval; +} + +struct address_space_operations v9fs_addr_operations = { + .readpage = v9fs_vfs_readpage, +}; diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c index 6852f0eb96e..c7e14d91721 100644 --- a/fs/9p/vfs_file.c +++ b/fs/9p/vfs_file.c @@ -289,6 +289,9 @@ v9fs_file_write(struct file *filp, const char __user * data, total += result; } while (count); + if(inode->i_mapping->nrpages) + invalidate_inode_pages2(inode->i_mapping); + return total; } @@ -299,4 +302,5 @@ struct file_operations v9fs_file_operations = { .open = v9fs_file_open, .release = v9fs_dir_release, .lock = v9fs_file_lock, + .mmap = generic_file_mmap, }; diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index a17b2885428..91f552454c7 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -177,6 +177,7 @@ struct inode *v9fs_get_inode(struct super_block *sb, int mode) inode->i_blocks = 0; inode->i_rdev = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; + inode->i_mapping->a_ops = &v9fs_addr_operations; switch (mode & S_IFMT) { case S_IFIFO: -- cgit v1.2.3 From c21761f168894b356626c847fe13be39605d76b4 Mon Sep 17 00:00:00 2001 From: Jason Baron Date: Wed, 18 Jan 2006 17:43:03 -0800 Subject: [PATCH] fix sched_setscheduler semantics Currently, a negative policy argument passed into the 'sys_sched_setscheduler()' system call, will return with success. However, the manpage for 'sys_sched_setscheduler' says: EINVAL The scheduling policy is not one of the recognized policies, or the parameter p does not make sense for the policy. Signed-off-by: Jason Baron Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sched.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/kernel/sched.c b/kernel/sched.c index 788ecce1e0e..3ee2ae45125 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -3850,6 +3850,10 @@ do_sched_setscheduler(pid_t pid, int policy, struct sched_param __user *param) asmlinkage long sys_sched_setscheduler(pid_t pid, int policy, struct sched_param __user *param) { + /* negative values for policy are not valid */ + if (policy < 0) + return -EINVAL; + return do_sched_setscheduler(pid, policy, param); } -- cgit v1.2.3 From 5131cf154ad1c6e584efa58d17a469d0b80f49bd Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 18 Jan 2006 17:43:04 -0800 Subject: [PATCH] add missing syscall declarations All standard system calls should be declared in include/linux/syscalls.h. Add some of the new additions that were previously missed. Signed-off-by: Arnd Bergmann Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/inotify.c | 1 + include/linux/syscalls.h | 19 +++++++++++++++++-- 2 files changed, 18 insertions(+), 2 deletions(-) diff --git a/fs/inotify.c b/fs/inotify.c index 2fecb7af4a7..878ccca6121 100644 --- a/fs/inotify.c +++ b/fs/inotify.c @@ -33,6 +33,7 @@ #include #include #include +#include #include diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 3eed4734701..e666d607056 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -510,9 +510,24 @@ asmlinkage long sys_keyctl(int cmd, unsigned long arg2, unsigned long arg3, asmlinkage long sys_ioprio_set(int which, int who, int ioprio); asmlinkage long sys_ioprio_get(int which, int who); asmlinkage long sys_set_mempolicy(int mode, unsigned long __user *nmask, - unsigned long maxnode); + unsigned long maxnode); asmlinkage long sys_migrate_pages(pid_t pid, unsigned long maxnode, - const unsigned long __user *from, const unsigned long __user *to); + const unsigned long __user *from, + const unsigned long __user *to); +asmlinkage long sys_mbind(unsigned long start, unsigned long len, + unsigned long mode, + unsigned long __user *nmask, + unsigned long maxnode, + unsigned flags); +asmlinkage long sys_get_mempolicy(int __user *policy, + unsigned long __user *nmask, + unsigned long maxnode, + unsigned long addr, unsigned long flags); + +asmlinkage long sys_inotify_init(void); +asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, + u32 mask); +asmlinkage long sys_inotify_rm_watch(int fd, u32 wd); asmlinkage long sys_spu_run(int fd, __u32 __user *unpc, __u32 __user *ustatus); -- cgit v1.2.3 From 634725a92938b0f282b17cec0b007dca77adebd2 Mon Sep 17 00:00:00 2001 From: Roman Zippel Date: Wed, 18 Jan 2006 17:43:05 -0800 Subject: [PATCH] hfs: cleanup HFS+ prints Add the log level and a "hfs: " prefix to all kernel prints. (HFS and HFS+ will use the same prefix, as they share some code and could be merged at some point.) Signed-off-by: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hfsplus/bfind.c | 3 +-- fs/hfsplus/bnode.c | 11 ++++++----- fs/hfsplus/brec.c | 2 +- fs/hfsplus/btree.c | 16 ++++++---------- fs/hfsplus/catalog.c | 2 +- fs/hfsplus/dir.c | 14 +++++++------- fs/hfsplus/extents.c | 3 +-- fs/hfsplus/inode.c | 5 +---- fs/hfsplus/options.c | 18 +++++++++--------- fs/hfsplus/super.c | 49 ++++++++++++++++++++----------------------------- fs/hfsplus/wrapper.c | 4 ++-- 11 files changed, 55 insertions(+), 72 deletions(-) diff --git a/fs/hfsplus/bfind.c b/fs/hfsplus/bfind.c index 257cdde0514..5007a41f1be 100644 --- a/fs/hfsplus/bfind.c +++ b/fs/hfsplus/bfind.c @@ -64,7 +64,6 @@ int __hfs_brec_find(struct hfs_bnode *bnode, struct hfs_find_data *fd) else e = rec - 1; } while (b <= e); - //printk("%d: %d,%d,%d\n", bnode->this, b, e, rec); if (rec != e && e >= 0) { len = hfs_brec_lenoff(bnode, e, &off); keylen = hfs_brec_keylen(bnode, e); @@ -127,7 +126,7 @@ int hfs_brec_find(struct hfs_find_data *fd) return res; invalid: - printk("HFS+-fs: inconsistency in B*Tree (%d,%d,%d,%u,%u)\n", + printk(KERN_ERR "hfs: inconsistency in B*Tree (%d,%d,%d,%u,%u)\n", height, bnode->height, bnode->type, nidx, parent); res = -EIO; release: diff --git a/fs/hfsplus/bnode.c b/fs/hfsplus/bnode.c index 930cd9212de..8f07e8fbd03 100644 --- a/fs/hfsplus/bnode.c +++ b/fs/hfsplus/bnode.c @@ -358,7 +358,7 @@ void hfs_bnode_unlink(struct hfs_bnode *node) // move down? if (!node->prev && !node->next) { - printk("hfs_btree_del_level\n"); + printk(KERN_DEBUG "hfs_btree_del_level\n"); } if (!node->parent) { tree->root = 0; @@ -379,7 +379,7 @@ struct hfs_bnode *hfs_bnode_findhash(struct hfs_btree *tree, u32 cnid) struct hfs_bnode *node; if (cnid >= tree->node_count) { - printk("HFS+-fs: request for non-existent node %d in B*Tree\n", cnid); + printk(KERN_ERR "hfs: request for non-existent node %d in B*Tree\n", cnid); return NULL; } @@ -402,7 +402,7 @@ static struct hfs_bnode *__hfs_bnode_create(struct hfs_btree *tree, u32 cnid) loff_t off; if (cnid >= tree->node_count) { - printk("HFS+-fs: request for non-existent node %d in B*Tree\n", cnid); + printk(KERN_ERR "hfs: request for non-existent node %d in B*Tree\n", cnid); return NULL; } @@ -576,8 +576,9 @@ struct hfs_bnode *hfs_bnode_create(struct hfs_btree *tree, u32 num) node = hfs_bnode_findhash(tree, num); spin_unlock(&tree->hash_lock); if (node) { - printk("new node %u already hashed?\n", num); - BUG(); + printk(KERN_CRIT "new node %u already hashed?\n", num); + WARN_ON(1); + return node; } node = __hfs_bnode_create(tree, num); if (!node) diff --git a/fs/hfsplus/brec.c b/fs/hfsplus/brec.c index 0ccef2ab790..c88e5d72a40 100644 --- a/fs/hfsplus/brec.c +++ b/fs/hfsplus/brec.c @@ -360,7 +360,7 @@ again: end_off = hfs_bnode_read_u16(parent, end_rec_off); if (end_rec_off - end_off < diff) { - printk("splitting index node...\n"); + printk(KERN_DEBUG "hfs: splitting index node...\n"); fd->bnode = parent; new_node = hfs_bnode_split(fd); if (IS_ERR(new_node)) diff --git a/fs/hfsplus/btree.c b/fs/hfsplus/btree.c index 44326aa2bd3..67129066383 100644 --- a/fs/hfsplus/btree.c +++ b/fs/hfsplus/btree.c @@ -39,7 +39,7 @@ struct hfs_btree *hfs_btree_open(struct super_block *sb, u32 id) } else if (id == HFSPLUS_CAT_CNID) { tree->keycmp = hfsplus_cat_cmp_key; } else { - printk("HFS+-fs: unknown B*Tree requested\n"); + printk(KERN_ERR "hfs: unknown B*Tree requested\n"); goto free_tree; } tree->inode = iget(sb, id); @@ -99,7 +99,7 @@ void hfs_btree_close(struct hfs_btree *tree) while ((node = tree->node_hash[i])) { tree->node_hash[i] = node->next_hash; if (atomic_read(&node->refcnt)) - printk("HFS+: node %d:%d still has %d user(s)!\n", + printk(KERN_CRIT "hfs: node %d:%d still has %d user(s)!\n", node->tree->cnid, node->this, atomic_read(&node->refcnt)); hfs_bnode_free(node); tree->node_hash_cnt--; @@ -223,10 +223,6 @@ struct hfs_bnode *hfs_bmap_alloc(struct hfs_btree *tree) tree->free_nodes--; mark_inode_dirty(tree->inode); hfs_bnode_put(node); - if (!idx) { - printk("unexpected idx %u (%u)\n", idx, node->this); - BUG(); - } return hfs_bnode_create(tree, idx); } } @@ -242,7 +238,7 @@ struct hfs_bnode *hfs_bmap_alloc(struct hfs_btree *tree) kunmap(*pagep); nidx = node->next; if (!nidx) { - printk("create new bmap node...\n"); + printk(KERN_DEBUG "hfs: create new bmap node...\n"); next_node = hfs_bmap_new_bmap(node, idx); } else next_node = hfs_bnode_find(tree, nidx); @@ -284,7 +280,7 @@ void hfs_bmap_free(struct hfs_bnode *node) hfs_bnode_put(node); if (!i) { /* panic */; - printk("HFS: unable to free bnode %u. bmap not found!\n", node->this); + printk(KERN_CRIT "hfs: unable to free bnode %u. bmap not found!\n", node->this); return; } node = hfs_bnode_find(tree, i); @@ -292,7 +288,7 @@ void hfs_bmap_free(struct hfs_bnode *node) return; if (node->type != HFS_NODE_MAP) { /* panic */; - printk("HFS: invalid bmap found! (%u,%d)\n", node->this, node->type); + printk(KERN_CRIT "hfs: invalid bmap found! (%u,%d)\n", node->this, node->type); hfs_bnode_put(node); return; } @@ -305,7 +301,7 @@ void hfs_bmap_free(struct hfs_bnode *node) m = 1 << (~nidx & 7); byte = data[off]; if (!(byte & m)) { - printk("HFS: trying to free free bnode %u(%d)\n", node->this, node->type); + printk(KERN_CRIT "hfs: trying to free free bnode %u(%d)\n", node->this, node->type); kunmap(page); hfs_bnode_put(node); return; diff --git a/fs/hfsplus/catalog.c b/fs/hfsplus/catalog.c index 94712790c8b..074451f1d95 100644 --- a/fs/hfsplus/catalog.c +++ b/fs/hfsplus/catalog.c @@ -139,7 +139,7 @@ int hfsplus_find_cat(struct super_block *sb, u32 cnid, type = be16_to_cpu(tmp.type); if (type != HFSPLUS_FOLDER_THREAD && type != HFSPLUS_FILE_THREAD) { - printk("HFS+-fs: Found bad thread record in catalog\n"); + printk(KERN_ERR "hfs: found bad thread record in catalog\n"); return -EIO; } diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c index 50c8f44b6c6..82c22376547 100644 --- a/fs/hfsplus/dir.c +++ b/fs/hfsplus/dir.c @@ -84,7 +84,7 @@ again: } else if (!dentry->d_fsdata) dentry->d_fsdata = (void *)(unsigned long)cnid; } else { - printk("HFS+-fs: Illegal catalog entry type in lookup\n"); + printk(KERN_ERR "hfs: invalid catalog entry type in lookup\n"); err = -EIO; goto fail; } @@ -132,12 +132,12 @@ static int hfsplus_readdir(struct file *filp, void *dirent, filldir_t filldir) case 1: hfs_bnode_read(fd.bnode, &entry, fd.entryoffset, fd.entrylength); if (be16_to_cpu(entry.type) != HFSPLUS_FOLDER_THREAD) { - printk("HFS+-fs: bad catalog folder thread\n"); + printk(KERN_ERR "hfs: bad catalog folder thread\n"); err = -EIO; goto out; } if (fd.entrylength < HFSPLUS_MIN_THREAD_SZ) { - printk("HFS+-fs: truncated catalog thread\n"); + printk(KERN_ERR "hfs: truncated catalog thread\n"); err = -EIO; goto out; } @@ -156,7 +156,7 @@ static int hfsplus_readdir(struct file *filp, void *dirent, filldir_t filldir) for (;;) { if (be32_to_cpu(fd.key->cat.parent) != inode->i_ino) { - printk("HFS+-fs: walked past end of dir\n"); + printk(KERN_ERR "hfs: walked past end of dir\n"); err = -EIO; goto out; } @@ -168,7 +168,7 @@ static int hfsplus_readdir(struct file *filp, void *dirent, filldir_t filldir) goto out; if (type == HFSPLUS_FOLDER) { if (fd.entrylength < sizeof(struct hfsplus_cat_folder)) { - printk("HFS+-fs: small dir entry\n"); + printk(KERN_ERR "hfs: small dir entry\n"); err = -EIO; goto out; } @@ -180,7 +180,7 @@ static int hfsplus_readdir(struct file *filp, void *dirent, filldir_t filldir) break; } else if (type == HFSPLUS_FILE) { if (fd.entrylength < sizeof(struct hfsplus_cat_file)) { - printk("HFS+-fs: small file entry\n"); + printk(KERN_ERR "hfs: small file entry\n"); err = -EIO; goto out; } @@ -188,7 +188,7 @@ static int hfsplus_readdir(struct file *filp, void *dirent, filldir_t filldir) be32_to_cpu(entry.file.id), DT_REG)) break; } else { - printk("HFS+-fs: bad catalog entry type\n"); + printk(KERN_ERR "hfs: bad catalog entry type\n"); err = -EIO; goto out; } diff --git a/fs/hfsplus/extents.c b/fs/hfsplus/extents.c index e3ff56a0301..c95559f69bc 100644 --- a/fs/hfsplus/extents.c +++ b/fs/hfsplus/extents.c @@ -349,10 +349,9 @@ int hfsplus_file_extend(struct inode *inode) if (HFSPLUS_SB(sb).alloc_file->i_size * 8 < HFSPLUS_SB(sb).total_blocks - HFSPLUS_SB(sb).free_blocks + 8) { // extend alloc file - printk("extend alloc file! (%Lu,%u,%u)\n", HFSPLUS_SB(sb).alloc_file->i_size * 8, + printk(KERN_ERR "hfs: extend alloc file! (%Lu,%u,%u)\n", HFSPLUS_SB(sb).alloc_file->i_size * 8, HFSPLUS_SB(sb).total_blocks, HFSPLUS_SB(sb).free_blocks); return -ENOSPC; - //BUG(); } down(&HFSPLUS_I(inode).extents_lock); diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c index 7acff6c5464..182eb317797 100644 --- a/fs/hfsplus/inode.c +++ b/fs/hfsplus/inode.c @@ -18,13 +18,11 @@ static int hfsplus_readpage(struct file *file, struct page *page) { - //printk("readpage: %lu\n", page->index); return block_read_full_page(page, hfsplus_get_block); } static int hfsplus_writepage(struct page *page, struct writeback_control *wbc) { - //printk("writepage: %lu\n", page->index); return block_write_full_page(page, hfsplus_get_block, wbc); } @@ -92,7 +90,6 @@ static int hfsplus_releasepage(struct page *page, gfp_t mask) } while (--i && nidx < tree->node_count); spin_unlock(&tree->hash_lock); } - //printk("releasepage: %lu,%x = %d\n", page->index, mask, res); return res ? try_to_free_buffers(page) : 0; } @@ -467,7 +464,7 @@ int hfsplus_cat_read_inode(struct inode *inode, struct hfs_find_data *fd) inode->i_mtime = hfsp_mt2ut(file->content_mod_date); inode->i_ctime = inode->i_mtime; } else { - printk("HFS+-fs: bad catalog entry used to create inode\n"); + printk(KERN_ERR "hfs: bad catalog entry used to create inode\n"); res = -EIO; } return res; diff --git a/fs/hfsplus/options.c b/fs/hfsplus/options.c index 935dafba007..dc64fac0083 100644 --- a/fs/hfsplus/options.c +++ b/fs/hfsplus/options.c @@ -83,58 +83,58 @@ int hfsplus_parse_options(char *input, struct hfsplus_sb_info *sbi) switch (token) { case opt_creator: if (match_fourchar(&args[0], &sbi->creator)) { - printk("HFS+-fs: creator requires a 4 character value\n"); + printk(KERN_ERR "hfs: creator requires a 4 character value\n"); return 0; } break; case opt_type: if (match_fourchar(&args[0], &sbi->type)) { - printk("HFS+-fs: type requires a 4 character value\n"); + printk(KERN_ERR "hfs: type requires a 4 character value\n"); return 0; } break; case opt_umask: if (match_octal(&args[0], &tmp)) { - printk("HFS+-fs: umask requires a value\n"); + printk(KERN_ERR "hfs: umask requires a value\n"); return 0; } sbi->umask = (umode_t)tmp; break; case opt_uid: if (match_int(&args[0], &tmp)) { - printk("HFS+-fs: uid requires an argument\n"); + printk(KERN_ERR "hfs: uid requires an argument\n"); return 0; } sbi->uid = (uid_t)tmp; break; case opt_gid: if (match_int(&args[0], &tmp)) { - printk("HFS+-fs: gid requires an argument\n"); + printk(KERN_ERR "hfs: gid requires an argument\n"); return 0; } sbi->gid = (gid_t)tmp; break; case opt_part: if (match_int(&args[0], &sbi->part)) { - printk("HFS+-fs: part requires an argument\n"); + printk(KERN_ERR "hfs: part requires an argument\n"); return 0; } break; case opt_session: if (match_int(&args[0], &sbi->session)) { - printk("HFS+-fs: session requires an argument\n"); + printk(KERN_ERR "hfs: session requires an argument\n"); return 0; } break; case opt_nls: if (sbi->nls) { - printk("HFS+-fs: unable to change nls mapping\n"); + printk(KERN_ERR "hfs: unable to change nls mapping\n"); return 0; } p = match_strdup(&args[0]); sbi->nls = load_nls(p); if (!sbi->nls) { - printk("HFS+-fs: unable to load nls mapping \"%s\"\n", p); + printk(KERN_ERR "hfs: unable to load nls mapping \"%s\"\n", p); kfree(p); return 0; } diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index d791780def5..b712d34d458 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -169,7 +169,7 @@ static void hfsplus_write_super(struct super_block *sb) block = HFSPLUS_SB(sb).blockoffset; block += (HFSPLUS_SB(sb).sect_count - 2) >> (sb->s_blocksize_bits - 9); offset = ((HFSPLUS_SB(sb).sect_count - 2) << 9) & (sb->s_blocksize - 1); - printk("backup: %u,%u,%u,%u\n", HFSPLUS_SB(sb).blockoffset, + printk(KERN_DEBUG "hfs: backup: %u,%u,%u,%u\n", HFSPLUS_SB(sb).blockoffset, HFSPLUS_SB(sb).sect_count, block, offset); bh = sb_bread(sb, block); if (bh) { @@ -179,7 +179,7 @@ static void hfsplus_write_super(struct super_block *sb) mark_buffer_dirty(bh); brelse(bh); } else - printk("backup not found!\n"); + printk(KERN_WARNING "hfs: backup not found!\n"); } } HFSPLUS_SB(sb).flags &= ~HFSPLUS_SB_WRITEBACKUP; @@ -240,18 +240,18 @@ static int hfsplus_remount(struct super_block *sb, int *flags, char *data) return -EINVAL; if (!(vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_UNMNT))) { - printk("HFS+-fs warning: Filesystem was not cleanly unmounted, " + printk(KERN_WARNING "hfs: filesystem was not cleanly unmounted, " "running fsck.hfsplus is recommended. leaving read-only.\n"); sb->s_flags |= MS_RDONLY; *flags |= MS_RDONLY; } else if (sbi.flags & HFSPLUS_SB_FORCE) { /* nothing */ } else if (vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_SOFTLOCK)) { - printk("HFS+-fs: Filesystem is marked locked, leaving read-only.\n"); + printk(KERN_WARNING "hfs: filesystem is marked locked, leaving read-only.\n"); sb->s_flags |= MS_RDONLY; *flags |= MS_RDONLY; } else if (vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_JOURNALED)) { - printk("HFS+-fs: Filesystem is marked journaled, leaving read-only.\n"); + printk(KERN_WARNING "hfs: filesystem is marked journaled, leaving read-only.\n"); sb->s_flags |= MS_RDONLY; *flags |= MS_RDONLY; } @@ -292,8 +292,7 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) INIT_HLIST_HEAD(&sbi->rsrc_inodes); hfsplus_fill_defaults(sbi); if (!hfsplus_parse_options(data, sbi)) { - if (!silent) - printk("HFS+-fs: unable to parse mount options\n"); + printk(KERN_ERR "hfs: unable to parse mount options\n"); err = -EINVAL; goto cleanup; } @@ -302,7 +301,7 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) nls = sbi->nls; sbi->nls = load_nls("utf8"); if (!sbi->nls) { - printk("HFS+: unable to load nls for utf8\n"); + printk(KERN_ERR "hfs: unable to load nls for utf8\n"); err = -EINVAL; goto cleanup; } @@ -310,7 +309,7 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) /* Grab the volume header */ if (hfsplus_read_wrapper(sb)) { if (!silent) - printk("HFS+-fs: unable to find HFS+ superblock\n"); + printk(KERN_WARNING "hfs: unable to find HFS+ superblock\n"); err = -EINVAL; goto cleanup; } @@ -319,8 +318,7 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) /* Copy parts of the volume header into the superblock */ sb->s_magic = be16_to_cpu(vhdr->signature); if (be16_to_cpu(vhdr->version) != HFSPLUS_CURRENT_VERSION) { - if (!silent) - printk("HFS+-fs: wrong filesystem version\n"); + printk(KERN_ERR "hfs: wrong filesystem version\n"); goto cleanup; } HFSPLUS_SB(sb).total_blocks = be32_to_cpu(vhdr->total_blocks); @@ -341,20 +339,17 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) sb->s_maxbytes = MAX_LFS_FILESIZE; if (!(vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_UNMNT))) { - if (!silent) - printk("HFS+-fs warning: Filesystem was not cleanly unmounted, " - "running fsck.hfsplus is recommended. mounting read-only.\n"); + printk(KERN_WARNING "hfs: Filesystem was not cleanly unmounted, " + "running fsck.hfsplus is recommended. mounting read-only.\n"); sb->s_flags |= MS_RDONLY; } else if (sbi->flags & HFSPLUS_SB_FORCE) { /* nothing */ } else if (vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_SOFTLOCK)) { - if (!silent) - printk("HFS+-fs: Filesystem is marked locked, mounting read-only.\n"); + printk(KERN_WARNING "hfs: Filesystem is marked locked, mounting read-only.\n"); sb->s_flags |= MS_RDONLY; } else if (vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_JOURNALED)) { - if (!silent) - printk("HFS+-fs: write access to a jounaled filesystem is not supported, " - "use the force option at your own risk, mounting read-only.\n"); + printk(KERN_WARNING "hfs: write access to a jounaled filesystem is not supported, " + "use the force option at your own risk, mounting read-only.\n"); sb->s_flags |= MS_RDONLY; } sbi->flags &= ~HFSPLUS_SB_FORCE; @@ -362,21 +357,18 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) /* Load metadata objects (B*Trees) */ HFSPLUS_SB(sb).ext_tree = hfs_btree_open(sb, HFSPLUS_EXT_CNID); if (!HFSPLUS_SB(sb).ext_tree) { - if (!silent) - printk("HFS+-fs: failed to load extents file\n"); + printk(KERN_ERR "hfs: failed to load extents file\n"); goto cleanup; } HFSPLUS_SB(sb).cat_tree = hfs_btree_open(sb, HFSPLUS_CAT_CNID); if (!HFSPLUS_SB(sb).cat_tree) { - if (!silent) - printk("HFS+-fs: failed to load catalog file\n"); + printk(KERN_ERR "hfs: failed to load catalog file\n"); goto cleanup; } HFSPLUS_SB(sb).alloc_file = iget(sb, HFSPLUS_ALLOC_CNID); if (!HFSPLUS_SB(sb).alloc_file) { - if (!silent) - printk("HFS+-fs: failed to load allocation file\n"); + printk(KERN_ERR "hfs: failed to load allocation file\n"); goto cleanup; } @@ -384,8 +376,7 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) root = iget(sb, HFSPLUS_ROOT_CNID); sb->s_root = d_alloc_root(root); if (!sb->s_root) { - if (!silent) - printk("HFS+-fs: failed to load root directory\n"); + printk(KERN_ERR "hfs: failed to load root directory\n"); iput(root); goto cleanup; } @@ -419,7 +410,7 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) sync_dirty_buffer(HFSPLUS_SB(sb).s_vhbh); if (!HFSPLUS_SB(sb).hidden_dir) { - printk("HFS+: create hidden dir...\n"); + printk(KERN_DEBUG "hfs: create hidden dir...\n"); HFSPLUS_SB(sb).hidden_dir = hfsplus_new_inode(sb, S_IFDIR); hfsplus_create_cat(HFSPLUS_SB(sb).hidden_dir->i_ino, sb->s_root->d_inode, &str, HFSPLUS_SB(sb).hidden_dir); @@ -499,7 +490,7 @@ static void __exit exit_hfsplus_fs(void) { unregister_filesystem(&hfsplus_fs_type); if (kmem_cache_destroy(hfsplus_inode_cachep)) - printk(KERN_INFO "hfsplus_inode_cache: not all structures were freed\n"); + printk(KERN_ERR "hfsplus_inode_cache: not all structures were freed\n"); } module_init(init_hfsplus_fs) diff --git a/fs/hfsplus/wrapper.c b/fs/hfsplus/wrapper.c index 95455e83923..6b2dc3a061a 100644 --- a/fs/hfsplus/wrapper.c +++ b/fs/hfsplus/wrapper.c @@ -70,7 +70,7 @@ static int hfsplus_get_last_session(struct super_block *sb, *start = (sector_t)te.cdte_addr.lba << 2; return 0; } - printk(KERN_ERR "HFS: Invalid session number or type of track\n"); + printk(KERN_ERR "hfs: invalid session number or type of track\n"); return -EINVAL; } ms_info.addr_format = CDROM_LBA; @@ -143,7 +143,7 @@ int hfsplus_read_wrapper(struct super_block *sb) blocksize >>= 1; if (sb_set_blocksize(sb, blocksize) != blocksize) { - printk("HFS+: unable to blocksize to %u!\n", blocksize); + printk(KERN_ERR "hfs: unable to set blocksize to %u!\n", blocksize); return -EINVAL; } -- cgit v1.2.3 From 7cf3cc3036cb7b1147350bf7c3f1ab98c160eb7b Mon Sep 17 00:00:00 2001 From: Roman Zippel Date: Wed, 18 Jan 2006 17:43:07 -0800 Subject: [PATCH] hfs: cleanup HFS prints Add the log level and a "hfs: " prefix to all kernel prints. Signed-off-by: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hfs/bfind.c | 3 +-- fs/hfs/bnode.c | 6 +++--- fs/hfs/brec.c | 2 +- fs/hfs/btree.c | 10 +++++----- fs/hfs/catalog.c | 2 +- fs/hfs/dir.c | 12 ++++++------ fs/hfs/hfs_fs.h | 3 --- fs/hfs/inode.c | 1 - fs/hfs/mdb.c | 22 +++++++++++----------- fs/hfs/super.c | 40 ++++++++++++++++++++-------------------- 10 files changed, 48 insertions(+), 53 deletions(-) diff --git a/fs/hfs/bfind.c b/fs/hfs/bfind.c index 89450ae3222..f13f1494d4f 100644 --- a/fs/hfs/bfind.c +++ b/fs/hfs/bfind.c @@ -64,7 +64,6 @@ int __hfs_brec_find(struct hfs_bnode *bnode, struct hfs_find_data *fd) else e = rec - 1; } while (b <= e); - //printk("%d: %d,%d,%d\n", bnode->this, b, e, rec); if (rec != e && e >= 0) { len = hfs_brec_lenoff(bnode, e, &off); keylen = hfs_brec_keylen(bnode, e); @@ -127,7 +126,7 @@ int hfs_brec_find(struct hfs_find_data *fd) return res; invalid: - printk("HFS: inconsistency in B*Tree (%d,%d,%d,%u,%u)\n", + printk(KERN_ERR "hfs: inconsistency in B*Tree (%d,%d,%d,%u,%u)\n", height, bnode->height, bnode->type, nidx, parent); res = -EIO; release: diff --git a/fs/hfs/bnode.c b/fs/hfs/bnode.c index 3d5cdc6847c..a7a7d77f3fd 100644 --- a/fs/hfs/bnode.c +++ b/fs/hfs/bnode.c @@ -198,7 +198,7 @@ void hfs_bnode_unlink(struct hfs_bnode *node) // move down? if (!node->prev && !node->next) { - printk("hfs_btree_del_level\n"); + printk(KERN_DEBUG "hfs_btree_del_level\n"); } if (!node->parent) { tree->root = 0; @@ -219,7 +219,7 @@ struct hfs_bnode *hfs_bnode_findhash(struct hfs_btree *tree, u32 cnid) struct hfs_bnode *node; if (cnid >= tree->node_count) { - printk("HFS: request for non-existent node %d in B*Tree\n", cnid); + printk(KERN_ERR "hfs: request for non-existent node %d in B*Tree\n", cnid); return NULL; } @@ -242,7 +242,7 @@ static struct hfs_bnode *__hfs_bnode_create(struct hfs_btree *tree, u32 cnid) loff_t off; if (cnid >= tree->node_count) { - printk("HFS: request for non-existent node %d in B*Tree\n", cnid); + printk(KERN_ERR "hfs: request for non-existent node %d in B*Tree\n", cnid); return NULL; } diff --git a/fs/hfs/brec.c b/fs/hfs/brec.c index 7d8fff2c25f..5c87cf4801f 100644 --- a/fs/hfs/brec.c +++ b/fs/hfs/brec.c @@ -362,7 +362,7 @@ again: end_off = hfs_bnode_read_u16(parent, end_rec_off); if (end_rec_off - end_off < diff) { - printk("splitting index node...\n"); + printk(KERN_DEBUG "hfs: splitting index node...\n"); fd->bnode = parent; new_node = hfs_bnode_split(fd); if (IS_ERR(new_node)) diff --git a/fs/hfs/btree.c b/fs/hfs/btree.c index 394725efa1c..7bb11edd148 100644 --- a/fs/hfs/btree.c +++ b/fs/hfs/btree.c @@ -111,7 +111,7 @@ void hfs_btree_close(struct hfs_btree *tree) while ((node = tree->node_hash[i])) { tree->node_hash[i] = node->next_hash; if (atomic_read(&node->refcnt)) - printk("HFS: node %d:%d still has %d user(s)!\n", + printk(KERN_ERR "hfs: node %d:%d still has %d user(s)!\n", node->tree->cnid, node->this, atomic_read(&node->refcnt)); hfs_bnode_free(node); tree->node_hash_cnt--; @@ -252,7 +252,7 @@ struct hfs_bnode *hfs_bmap_alloc(struct hfs_btree *tree) kunmap(*pagep); nidx = node->next; if (!nidx) { - printk("create new bmap node...\n"); + printk(KERN_DEBUG "hfs: create new bmap node...\n"); next_node = hfs_bmap_new_bmap(node, idx); } else next_node = hfs_bnode_find(tree, nidx); @@ -292,7 +292,7 @@ void hfs_bmap_free(struct hfs_bnode *node) hfs_bnode_put(node); if (!i) { /* panic */; - printk("HFS: unable to free bnode %u. bmap not found!\n", node->this); + printk(KERN_CRIT "hfs: unable to free bnode %u. bmap not found!\n", node->this); return; } node = hfs_bnode_find(tree, i); @@ -300,7 +300,7 @@ void hfs_bmap_free(struct hfs_bnode *node) return; if (node->type != HFS_NODE_MAP) { /* panic */; - printk("HFS: invalid bmap found! (%u,%d)\n", node->this, node->type); + printk(KERN_CRIT "hfs: invalid bmap found! (%u,%d)\n", node->this, node->type); hfs_bnode_put(node); return; } @@ -313,7 +313,7 @@ void hfs_bmap_free(struct hfs_bnode *node) m = 1 << (~nidx & 7); byte = data[off]; if (!(byte & m)) { - printk("HFS: trying to free free bnode %u(%d)\n", node->this, node->type); + printk(KERN_CRIT "hfs: trying to free free bnode %u(%d)\n", node->this, node->type); kunmap(page); hfs_bnode_put(node); return; diff --git a/fs/hfs/catalog.c b/fs/hfs/catalog.c index 2fcd679f023..ba851576ebb 100644 --- a/fs/hfs/catalog.c +++ b/fs/hfs/catalog.c @@ -184,7 +184,7 @@ int hfs_cat_find_brec(struct super_block *sb, u32 cnid, type = rec.type; if (type != HFS_CDR_THD && type != HFS_CDR_FTH) { - printk("HFS-fs: Found bad thread record in catalog\n"); + printk(KERN_ERR "hfs: found bad thread record in catalog\n"); return -EIO; } diff --git a/fs/hfs/dir.c b/fs/hfs/dir.c index e1f24befba5..534e5a7480e 100644 --- a/fs/hfs/dir.c +++ b/fs/hfs/dir.c @@ -81,12 +81,12 @@ static int hfs_readdir(struct file *filp, void *dirent, filldir_t filldir) case 1: hfs_bnode_read(fd.bnode, &entry, fd.entryoffset, fd.entrylength); if (entry.type != HFS_CDR_THD) { - printk("HFS: bad catalog folder thread\n"); + printk(KERN_ERR "hfs: bad catalog folder thread\n"); err = -EIO; goto out; } //if (fd.entrylength < HFS_MIN_THREAD_SZ) { - // printk("HFS: truncated catalog thread\n"); + // printk(KERN_ERR "hfs: truncated catalog thread\n"); // err = -EIO; // goto out; //} @@ -105,7 +105,7 @@ static int hfs_readdir(struct file *filp, void *dirent, filldir_t filldir) for (;;) { if (be32_to_cpu(fd.key->cat.ParID) != inode->i_ino) { - printk("HFS: walked past end of dir\n"); + printk(KERN_ERR "hfs: walked past end of dir\n"); err = -EIO; goto out; } @@ -114,7 +114,7 @@ static int hfs_readdir(struct file *filp, void *dirent, filldir_t filldir) len = hfs_mac2asc(sb, strbuf, &fd.key->cat.CName); if (type == HFS_CDR_DIR) { if (fd.entrylength < sizeof(struct hfs_cat_dir)) { - printk("HFS: small dir entry\n"); + printk(KERN_ERR "hfs: small dir entry\n"); err = -EIO; goto out; } @@ -123,7 +123,7 @@ static int hfs_readdir(struct file *filp, void *dirent, filldir_t filldir) break; } else if (type == HFS_CDR_FIL) { if (fd.entrylength < sizeof(struct hfs_cat_file)) { - printk("HFS: small file entry\n"); + printk(KERN_ERR "hfs: small file entry\n"); err = -EIO; goto out; } @@ -131,7 +131,7 @@ static int hfs_readdir(struct file *filp, void *dirent, filldir_t filldir) be32_to_cpu(entry.file.FlNum), DT_REG)) break; } else { - printk("HFS: bad catalog entry type %d\n", type); + printk(KERN_ERR "hfs: bad catalog entry type %d\n", type); err = -EIO; goto out; } diff --git a/fs/hfs/hfs_fs.h b/fs/hfs/hfs_fs.h index cc5dcd52e23..18ce47ab1b7 100644 --- a/fs/hfs/hfs_fs.h +++ b/fs/hfs/hfs_fs.h @@ -35,9 +35,6 @@ #define dprint(flg, fmt, args...) \ if (flg & DBG_MASK) printk(fmt , ## args) -#define hfs_warn(format, args...) printk(KERN_WARNING format , ## args) -#define hfs_error(format, args...) printk(KERN_ERR format , ## args) - /* * struct hfs_inode_info * diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c index 050a4927649..39fd85b9b91 100644 --- a/fs/hfs/inode.c +++ b/fs/hfs/inode.c @@ -95,7 +95,6 @@ static int hfs_releasepage(struct page *page, gfp_t mask) } while (--i && nidx < tree->node_count); spin_unlock(&tree->hash_lock); } - //printk("releasepage: %lu,%x = %d\n", page->index, mask, res); return res ? try_to_free_buffers(page) : 0; } diff --git a/fs/hfs/mdb.c b/fs/hfs/mdb.c index 0a473f79c89..b4651e128d7 100644 --- a/fs/hfs/mdb.c +++ b/fs/hfs/mdb.c @@ -47,7 +47,7 @@ static int hfs_get_last_session(struct super_block *sb, *start = (sector_t)te.cdte_addr.lba << 2; return 0; } - printk(KERN_ERR "HFS: Invalid session number or type of track\n"); + printk(KERN_ERR "hfs: invalid session number or type of track\n"); return -EINVAL; } ms_info.addr_format = CDROM_LBA; @@ -100,7 +100,7 @@ int hfs_mdb_get(struct super_block *sb) HFS_SB(sb)->alloc_blksz = size = be32_to_cpu(mdb->drAlBlkSiz); if (!size || (size & (HFS_SECTOR_SIZE - 1))) { - hfs_warn("hfs_fs: bad allocation block size %d\n", size); + printk(KERN_ERR "hfs: bad allocation block size %d\n", size); goto out_bh; } @@ -117,7 +117,7 @@ int hfs_mdb_get(struct super_block *sb) size >>= 1; brelse(bh); if (!sb_set_blocksize(sb, size)) { - printk("hfs_fs: unable to set blocksize to %u\n", size); + printk(KERN_ERR "hfs: unable to set blocksize to %u\n", size); goto out; } @@ -161,8 +161,8 @@ int hfs_mdb_get(struct super_block *sb) } if (!HFS_SB(sb)->alt_mdb) { - hfs_warn("hfs_fs: unable to locate alternate MDB\n"); - hfs_warn("hfs_fs: continuing without an alternate MDB\n"); + printk(KERN_WARNING "hfs: unable to locate alternate MDB\n"); + printk(KERN_WARNING "hfs: continuing without an alternate MDB\n"); } HFS_SB(sb)->bitmap = (__be32 *)__get_free_pages(GFP_KERNEL, PAGE_SIZE < 8192 ? 1 : 0); @@ -177,7 +177,7 @@ int hfs_mdb_get(struct super_block *sb) while (size) { bh = sb_bread(sb, off >> sb->s_blocksize_bits); if (!bh) { - hfs_warn("hfs_fs: unable to read volume bitmap\n"); + printk(KERN_ERR "hfs: unable to read volume bitmap\n"); goto out; } off2 = off & (sb->s_blocksize - 1); @@ -191,23 +191,23 @@ int hfs_mdb_get(struct super_block *sb) HFS_SB(sb)->ext_tree = hfs_btree_open(sb, HFS_EXT_CNID, hfs_ext_keycmp); if (!HFS_SB(sb)->ext_tree) { - hfs_warn("hfs_fs: unable to open extent tree\n"); + printk(KERN_ERR "hfs: unable to open extent tree\n"); goto out; } HFS_SB(sb)->cat_tree = hfs_btree_open(sb, HFS_CAT_CNID, hfs_cat_keycmp); if (!HFS_SB(sb)->cat_tree) { - hfs_warn("hfs_fs: unable to open catalog tree\n"); + printk(KERN_ERR "hfs: unable to open catalog tree\n"); goto out; } attrib = mdb->drAtrb; if (!(attrib & cpu_to_be16(HFS_SB_ATTRIB_UNMNT))) { - hfs_warn("HFS-fs warning: Filesystem was not cleanly unmounted, " + printk(KERN_WARNING "hfs: filesystem was not cleanly unmounted, " "running fsck.hfs is recommended. mounting read-only.\n"); sb->s_flags |= MS_RDONLY; } if ((attrib & cpu_to_be16(HFS_SB_ATTRIB_SLOCK))) { - hfs_warn("HFS-fs: Filesystem is marked locked, mounting read-only.\n"); + printk(KERN_WARNING "hfs: filesystem is marked locked, mounting read-only.\n"); sb->s_flags |= MS_RDONLY; } if (!(sb->s_flags & MS_RDONLY)) { @@ -303,7 +303,7 @@ void hfs_mdb_commit(struct super_block *sb) while (size) { bh = sb_bread(sb, block); if (!bh) { - hfs_warn("hfs_fs: unable to read volume bitmap\n"); + printk(KERN_ERR "hfs: unable to read volume bitmap\n"); break; } len = min((int)sb->s_blocksize - off, size); diff --git a/fs/hfs/super.c b/fs/hfs/super.c index c5074aeafca..1181d116117 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -101,12 +101,12 @@ static int hfs_remount(struct super_block *sb, int *flags, char *data) return 0; if (!(*flags & MS_RDONLY)) { if (!(HFS_SB(sb)->mdb->drAtrb & cpu_to_be16(HFS_SB_ATTRIB_UNMNT))) { - printk("HFS-fs warning: Filesystem was not cleanly unmounted, " + printk(KERN_WARNING "hfs: filesystem was not cleanly unmounted, " "running fsck.hfs is recommended. leaving read-only.\n"); sb->s_flags |= MS_RDONLY; *flags |= MS_RDONLY; } else if (HFS_SB(sb)->mdb->drAtrb & cpu_to_be16(HFS_SB_ATTRIB_SLOCK)) { - printk("HFS-fs: Filesystem is marked locked, leaving read-only.\n"); + printk(KERN_WARNING "hfs: filesystem is marked locked, leaving read-only.\n"); sb->s_flags |= MS_RDONLY; *flags |= MS_RDONLY; } @@ -229,21 +229,21 @@ static int parse_options(char *options, struct hfs_sb_info *hsb) switch (token) { case opt_uid: if (match_int(&args[0], &tmp)) { - printk("HFS: uid requires an argument\n"); + printk(KERN_ERR "hfs: uid requires an argument\n"); return 0; } hsb->s_uid = (uid_t)tmp; break; case opt_gid: if (match_int(&args[0], &tmp)) { - printk("HFS: gid requires an argument\n"); + printk(KERN_ERR "hfs: gid requires an argument\n"); return 0; } hsb->s_gid = (gid_t)tmp; break; case opt_umask: if (match_octal(&args[0], &tmp)) { - printk("HFS: umask requires a value\n"); + printk(KERN_ERR "hfs: umask requires a value\n"); return 0; } hsb->s_file_umask = (umode_t)tmp; @@ -251,39 +251,39 @@ static int parse_options(char *options, struct hfs_sb_info *hsb) break; case opt_file_umask: if (match_octal(&args[0], &tmp)) { - printk("HFS: file_umask requires a value\n"); + printk(KERN_ERR "hfs: file_umask requires a value\n"); return 0; } hsb->s_file_umask = (umode_t)tmp; break; case opt_dir_umask: if (match_octal(&args[0], &tmp)) { - printk("HFS: dir_umask requires a value\n"); + printk(KERN_ERR "hfs: dir_umask requires a value\n"); return 0; } hsb->s_dir_umask = (umode_t)tmp; break; case opt_part: if (match_int(&args[0], &hsb->part)) { - printk("HFS: part requires an argument\n"); + printk(KERN_ERR "hfs: part requires an argument\n"); return 0; } break; case opt_session: if (match_int(&args[0], &hsb->session)) { - printk("HFS: session requires an argument\n"); + printk(KERN_ERR "hfs: session requires an argument\n"); return 0; } break; case opt_type: if (match_fourchar(&args[0], &hsb->s_type)) { - printk("HFS+-fs: type requires a 4 character value\n"); + printk(KERN_ERR "hfs: type requires a 4 character value\n"); return 0; } break; case opt_creator: if (match_fourchar(&args[0], &hsb->s_creator)) { - printk("HFS+-fs: creator requires a 4 character value\n"); + printk(KERN_ERR "hfs: creator requires a 4 character value\n"); return 0; } break; @@ -292,13 +292,13 @@ static int parse_options(char *options, struct hfs_sb_info *hsb) break; case opt_codepage: if (hsb->nls_disk) { - printk("HFS+-fs: unable to change codepage\n"); + printk(KERN_ERR "hfs: unable to change codepage\n"); return 0; } p = match_strdup(&args[0]); hsb->nls_disk = load_nls(p); if (!hsb->nls_disk) { - printk("HFS+-fs: unable to load codepage \"%s\"\n", p); + printk(KERN_ERR "hfs: unable to load codepage \"%s\"\n", p); kfree(p); return 0; } @@ -306,13 +306,13 @@ static int parse_options(char *options, struct hfs_sb_info *hsb) break; case opt_iocharset: if (hsb->nls_io) { - printk("HFS: unable to change iocharset\n"); + printk(KERN_ERR "hfs: unable to change iocharset\n"); return 0; } p = match_strdup(&args[0]); hsb->nls_io = load_nls(p); if (!hsb->nls_io) { - printk("HFS: unable to load iocharset \"%s\"\n", p); + printk(KERN_ERR "hfs: unable to load iocharset \"%s\"\n", p); kfree(p); return 0; } @@ -326,7 +326,7 @@ static int parse_options(char *options, struct hfs_sb_info *hsb) if (hsb->nls_disk && !hsb->nls_io) { hsb->nls_io = load_nls_default(); if (!hsb->nls_io) { - printk("HFS: unable to load default iocharset\n"); + printk(KERN_ERR "hfs: unable to load default iocharset\n"); return 0; } } @@ -364,7 +364,7 @@ static int hfs_fill_super(struct super_block *sb, void *data, int silent) res = -EINVAL; if (!parse_options((char *)data, sbi)) { - hfs_warn("hfs_fs: unable to parse mount options.\n"); + printk(KERN_ERR "hfs: unable to parse mount options.\n"); goto bail; } @@ -375,7 +375,7 @@ static int hfs_fill_super(struct super_block *sb, void *data, int silent) res = hfs_mdb_get(sb); if (res) { if (!silent) - hfs_warn("VFS: Can't find a HFS filesystem on dev %s.\n", + printk(KERN_WARNING "hfs: can't find a HFS filesystem on dev %s.\n", hfs_mdb_name(sb)); res = -EINVAL; goto bail; @@ -407,7 +407,7 @@ static int hfs_fill_super(struct super_block *sb, void *data, int silent) bail_iput: iput(root_inode); bail_no_root: - hfs_warn("hfs_fs: get root inode failed.\n"); + printk(KERN_ERR "hfs: get root inode failed.\n"); bail: hfs_mdb_put(sb); return res; @@ -454,7 +454,7 @@ static void __exit exit_hfs_fs(void) { unregister_filesystem(&hfs_fs_type); if (kmem_cache_destroy(hfs_inode_cachep)) - printk(KERN_INFO "hfs_inode_cache: not all structures were freed\n"); + printk(KERN_ERR "hfs_inode_cache: not all structures were freed\n"); } module_init(init_hfs_fs) -- cgit v1.2.3 From 2179d372d9f8b5fc5c189c89bc6a565a42151b23 Mon Sep 17 00:00:00 2001 From: David Elliott Date: Wed, 18 Jan 2006 17:43:08 -0800 Subject: [PATCH] hfs: add HFSX support Add support for HFSX, which allows for case-sensitive filenames. Signed-off-by: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hfsplus/btree.c | 23 ++++++++++++++--------- fs/hfsplus/catalog.c | 18 ++++++++++++++++-- fs/hfsplus/extents.c | 3 ++- fs/hfsplus/hfsplus_fs.h | 11 +++++++---- fs/hfsplus/hfsplus_raw.h | 10 ++++++++-- fs/hfsplus/super.c | 5 +++-- fs/hfsplus/unicode.c | 30 +++++++++++++++++++++++++++++- fs/hfsplus/wrapper.c | 13 +++++++++++-- 8 files changed, 90 insertions(+), 23 deletions(-) diff --git a/fs/hfsplus/btree.c b/fs/hfsplus/btree.c index 67129066383..a67edfa34e9 100644 --- a/fs/hfsplus/btree.c +++ b/fs/hfsplus/btree.c @@ -31,17 +31,8 @@ struct hfs_btree *hfs_btree_open(struct super_block *sb, u32 id) init_MUTEX(&tree->tree_lock); spin_lock_init(&tree->hash_lock); - /* Set the correct compare function */ tree->sb = sb; tree->cnid = id; - if (id == HFSPLUS_EXT_CNID) { - tree->keycmp = hfsplus_ext_cmp_key; - } else if (id == HFSPLUS_CAT_CNID) { - tree->keycmp = hfsplus_cat_cmp_key; - } else { - printk(KERN_ERR "hfs: unknown B*Tree requested\n"); - goto free_tree; - } tree->inode = iget(sb, id); if (!tree->inode) goto free_tree; @@ -64,6 +55,20 @@ struct hfs_btree *hfs_btree_open(struct super_block *sb, u32 id) tree->max_key_len = be16_to_cpu(head->max_key_len); tree->depth = be16_to_cpu(head->depth); + /* Set the correct compare function */ + if (id == HFSPLUS_EXT_CNID) { + tree->keycmp = hfsplus_ext_cmp_key; + } else if (id == HFSPLUS_CAT_CNID) { + if ((HFSPLUS_SB(sb).flags & HFSPLUS_SB_HFSX) && + (head->key_type == HFSPLUS_KEY_BINARY)) + tree->keycmp = hfsplus_cat_bin_cmp_key; + else + tree->keycmp = hfsplus_cat_case_cmp_key; + } else { + printk(KERN_ERR "hfs: unknown B*Tree requested\n"); + goto fail_page; + } + size = tree->node_size; if (!size || size & (size - 1)) goto fail_page; diff --git a/fs/hfsplus/catalog.c b/fs/hfsplus/catalog.c index 074451f1d95..662d176856d 100644 --- a/fs/hfsplus/catalog.c +++ b/fs/hfsplus/catalog.c @@ -13,7 +13,8 @@ #include "hfsplus_fs.h" #include "hfsplus_raw.h" -int hfsplus_cat_cmp_key(hfsplus_btree_key *k1, hfsplus_btree_key *k2) +int hfsplus_cat_case_cmp_key(const hfsplus_btree_key *k1, + const hfsplus_btree_key *k2) { __be32 k1p, k2p; @@ -22,7 +23,20 @@ int hfsplus_cat_cmp_key(hfsplus_btree_key *k1, hfsplus_btree_key *k2) if (k1p != k2p) return be32_to_cpu(k1p) < be32_to_cpu(k2p) ? -1 : 1; - return hfsplus_unistrcmp(&k1->cat.name, &k2->cat.name); + return hfsplus_strcasecmp(&k1->cat.name, &k2->cat.name); +} + +int hfsplus_cat_bin_cmp_key(const hfsplus_btree_key *k1, + const hfsplus_btree_key *k2) +{ + __be32 k1p, k2p; + + k1p = k1->cat.parent; + k2p = k2->cat.parent; + if (k1p != k2p) + return be32_to_cpu(k1p) < be32_to_cpu(k2p) ? -1 : 1; + + return hfsplus_strcmp(&k1->cat.name, &k2->cat.name); } void hfsplus_cat_build_key(struct super_block *sb, hfsplus_btree_key *key, diff --git a/fs/hfsplus/extents.c b/fs/hfsplus/extents.c index c95559f69bc..1a7480089e8 100644 --- a/fs/hfsplus/extents.c +++ b/fs/hfsplus/extents.c @@ -16,7 +16,8 @@ #include "hfsplus_raw.h" /* Compare two extents keys, returns 0 on same, pos/neg for difference */ -int hfsplus_ext_cmp_key(hfsplus_btree_key *k1, hfsplus_btree_key *k2) +int hfsplus_ext_cmp_key(const hfsplus_btree_key *k1, + const hfsplus_btree_key *k2) { __be32 k1id, k2id; __be32 k1s, k2s; diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h index 0fa1ab6250b..4608171f45d 100644 --- a/fs/hfsplus/hfsplus_fs.h +++ b/fs/hfsplus/hfsplus_fs.h @@ -36,7 +36,7 @@ #define HFSPLUS_TYPE_DATA 0x00 #define HFSPLUS_TYPE_RSRC 0xFF -typedef int (*btree_keycmp)(hfsplus_btree_key *, hfsplus_btree_key *); +typedef int (*btree_keycmp)(const hfsplus_btree_key *, const hfsplus_btree_key *); #define NODE_HASH_SIZE 256 @@ -149,6 +149,7 @@ struct hfsplus_sb_info { #define HFSPLUS_SB_WRITEBACKUP 0x0001 #define HFSPLUS_SB_NODECOMPOSE 0x0002 #define HFSPLUS_SB_FORCE 0x0004 +#define HFSPLUS_SB_HFSX 0x0008 struct hfsplus_inode_info { @@ -303,7 +304,8 @@ int hfs_brec_read(struct hfs_find_data *, void *, int); int hfs_brec_goto(struct hfs_find_data *, int); /* catalog.c */ -int hfsplus_cat_cmp_key(hfsplus_btree_key *, hfsplus_btree_key *); +int hfsplus_cat_case_cmp_key(const hfsplus_btree_key *, const hfsplus_btree_key *); +int hfsplus_cat_bin_cmp_key(const hfsplus_btree_key *, const hfsplus_btree_key *); void hfsplus_cat_build_key(struct super_block *sb, hfsplus_btree_key *, u32, struct qstr *); int hfsplus_find_cat(struct super_block *, u32, struct hfs_find_data *); int hfsplus_create_cat(u32, struct inode *, struct qstr *, struct inode *); @@ -312,7 +314,7 @@ int hfsplus_rename_cat(u32, struct inode *, struct qstr *, struct inode *, struct qstr *); /* extents.c */ -int hfsplus_ext_cmp_key(hfsplus_btree_key *, hfsplus_btree_key *); +int hfsplus_ext_cmp_key(const hfsplus_btree_key *, const hfsplus_btree_key *); void hfsplus_ext_write_extent(struct inode *); int hfsplus_get_block(struct inode *, sector_t, struct buffer_head *, int); int hfsplus_free_fork(struct super_block *, u32, struct hfsplus_fork_raw *, int); @@ -350,7 +352,8 @@ extern u16 hfsplus_decompose_table[]; extern u16 hfsplus_compose_table[]; /* unicode.c */ -int hfsplus_unistrcmp(const struct hfsplus_unistr *, const struct hfsplus_unistr *); +int hfsplus_strcasecmp(const struct hfsplus_unistr *, const struct hfsplus_unistr *); +int hfsplus_strcmp(const struct hfsplus_unistr *, const struct hfsplus_unistr *); int hfsplus_uni2asc(struct super_block *, const struct hfsplus_unistr *, char *, int *); int hfsplus_asc2uni(struct super_block *, struct hfsplus_unistr *, const char *, int); diff --git a/fs/hfsplus/hfsplus_raw.h b/fs/hfsplus/hfsplus_raw.h index b4fbed63321..ccc47966dc5 100644 --- a/fs/hfsplus/hfsplus_raw.h +++ b/fs/hfsplus/hfsplus_raw.h @@ -22,8 +22,10 @@ #define HFSPLUS_SECTOR_SHIFT 9 #define HFSPLUS_VOLHEAD_SECTOR 2 #define HFSPLUS_VOLHEAD_SIG 0x482b +#define HFSPLUS_VOLHEAD_SIGX 0x4858 #define HFSPLUS_SUPER_MAGIC 0x482b -#define HFSPLUS_CURRENT_VERSION 4 +#define HFSPLUS_MIN_VERSION 4 +#define HFSPLUS_CURRENT_VERSION 5 #define HFSP_WRAP_MAGIC 0x4244 #define HFSP_WRAP_ATTRIB_SLOCK 0x8000 @@ -161,7 +163,7 @@ struct hfs_btree_header_rec { u16 reserved1; __be32 clump_size; u8 btree_type; - u8 reserved2; + u8 key_type; __be32 attributes; u32 reserved3[16]; } __packed; @@ -186,6 +188,10 @@ struct hfs_btree_header_rec { #define HFSPLUS_EXCH_CNID 15 /* ExchangeFiles temp id */ #define HFSPLUS_FIRSTUSER_CNID 16 /* first available user id */ +/* btree key type */ +#define HFSPLUS_KEY_CASEFOLDING 0xCF /* case-insensitive */ +#define HFSPLUS_KEY_BINARY 0xBC /* case-sensitive */ + /* HFS+ catalog entry key */ struct hfsplus_cat_key { __be16 key_len; diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index b712d34d458..7843f792a4b 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -316,8 +316,9 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) vhdr = HFSPLUS_SB(sb).s_vhdr; /* Copy parts of the volume header into the superblock */ - sb->s_magic = be16_to_cpu(vhdr->signature); - if (be16_to_cpu(vhdr->version) != HFSPLUS_CURRENT_VERSION) { + sb->s_magic = HFSPLUS_VOLHEAD_SIG; + if (be16_to_cpu(vhdr->version) < HFSPLUS_MIN_VERSION || + be16_to_cpu(vhdr->version) > HFSPLUS_CURRENT_VERSION) { printk(KERN_ERR "hfs: wrong filesystem version\n"); goto cleanup; } diff --git a/fs/hfsplus/unicode.c b/fs/hfsplus/unicode.c index 060c69048c3..689c8bd721f 100644 --- a/fs/hfsplus/unicode.c +++ b/fs/hfsplus/unicode.c @@ -28,7 +28,8 @@ static inline u16 case_fold(u16 c) } /* Compare unicode strings, return values like normal strcmp */ -int hfsplus_unistrcmp(const struct hfsplus_unistr *s1, const struct hfsplus_unistr *s2) +int hfsplus_strcasecmp(const struct hfsplus_unistr *s1, + const struct hfsplus_unistr *s2) { u16 len1, len2, c1, c2; const hfsplus_unichr *p1, *p2; @@ -59,6 +60,33 @@ int hfsplus_unistrcmp(const struct hfsplus_unistr *s1, const struct hfsplus_unis } } +/* Compare names as a sequence of 16-bit unsigned integers */ +int hfsplus_strcmp(const struct hfsplus_unistr *s1, + const struct hfsplus_unistr *s2) +{ + u16 len1, len2, c1, c2; + const hfsplus_unichr *p1, *p2; + int len; + + len1 = be16_to_cpu(s1->length); + len2 = be16_to_cpu(s2->length); + p1 = s1->unicode; + p2 = s2->unicode; + + for (len = min(len1, len2); len > 0; len--) { + c1 = be16_to_cpu(*p1); + c2 = be16_to_cpu(*p2); + if (c1 != c2) + return c1 < c2 ? -1 : 1; + p1++; + p2++; + } + + return len1 < len2 ? -1 : + len1 > len2 ? 1 : 0; +} + + #define Hangul_SBase 0xac00 #define Hangul_LBase 0x1100 #define Hangul_VBase 0x1161 diff --git a/fs/hfsplus/wrapper.c b/fs/hfsplus/wrapper.c index 6b2dc3a061a..72cab78f050 100644 --- a/fs/hfsplus/wrapper.c +++ b/fs/hfsplus/wrapper.c @@ -28,8 +28,11 @@ static int hfsplus_read_mdb(void *bufptr, struct hfsplus_wd *wd) { u32 extent; u16 attrib; + __be16 sig; - if (be16_to_cpu(*(__be16 *)(bufptr + HFSP_WRAPOFF_EMBEDSIG)) != HFSPLUS_VOLHEAD_SIG) + sig = *(__be16 *)(bufptr + HFSP_WRAPOFF_EMBEDSIG); + if (sig != cpu_to_be16(HFSPLUS_VOLHEAD_SIG) && + sig != cpu_to_be16(HFSPLUS_VOLHEAD_SIGX)) return 0; attrib = be16_to_cpu(*(__be16 *)(bufptr + HFSP_WRAPOFF_ATTRIB)); @@ -114,6 +117,10 @@ int hfsplus_read_wrapper(struct super_block *sb) } if (vhdr->signature == cpu_to_be16(HFSPLUS_VOLHEAD_SIG)) break; + if (vhdr->signature == cpu_to_be16(HFSPLUS_VOLHEAD_SIGX)) { + HFSPLUS_SB(sb).flags |= HFSPLUS_SB_HFSX; + break; + } brelse(bh); /* check for a partition block @@ -158,7 +165,9 @@ int hfsplus_read_wrapper(struct super_block *sb) return -EIO; /* should still be the same... */ - if (be16_to_cpu(vhdr->signature) != HFSPLUS_VOLHEAD_SIG) + if (vhdr->signature != (HFSPLUS_SB(sb).flags & HFSPLUS_SB_HFSX ? + cpu_to_be16(HFSPLUS_VOLHEAD_SIGX) : + cpu_to_be16(HFSPLUS_VOLHEAD_SIG))) goto error; HFSPLUS_SB(sb).s_vhbh = bh; HFSPLUS_SB(sb).s_vhdr = vhdr; -- cgit v1.2.3 From 9a4cad95c9338077487226e22d4e01bc9edebf21 Mon Sep 17 00:00:00 2001 From: Roman Zippel Date: Wed, 18 Jan 2006 17:43:09 -0800 Subject: [PATCH] hfs: set correct ctime Read the correct ctime from disk (it was written but never read for some reason). Read also creation date, which is used in the next patch. (Problem found by Olivier Castan ) Signed-off-by: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hfsplus/catalog.c | 14 ++++++++++---- fs/hfsplus/hfsplus_fs.h | 1 + fs/hfsplus/inode.c | 6 ++++-- 3 files changed, 15 insertions(+), 6 deletions(-) diff --git a/fs/hfsplus/catalog.c b/fs/hfsplus/catalog.c index 662d176856d..04255af3470 100644 --- a/fs/hfsplus/catalog.c +++ b/fs/hfsplus/catalog.c @@ -94,8 +94,11 @@ static int hfsplus_cat_build_record(hfsplus_cat_entry *entry, u32 cnid, struct i memset(folder, 0, sizeof(*folder)); folder->type = cpu_to_be16(HFSPLUS_FOLDER); folder->id = cpu_to_be32(inode->i_ino); - folder->create_date = folder->content_mod_date = - folder->attribute_mod_date = folder->access_date = hfsp_now2mt(); + HFSPLUS_I(inode).create_date = + folder->create_date = + folder->content_mod_date = + folder->attribute_mod_date = + folder->access_date = hfsp_now2mt(); hfsplus_set_perms(inode, &folder->permissions); if (inode == HFSPLUS_SB(inode->i_sb).hidden_dir) /* invisible and namelocked */ @@ -109,8 +112,11 @@ static int hfsplus_cat_build_record(hfsplus_cat_entry *entry, u32 cnid, struct i file->type = cpu_to_be16(HFSPLUS_FILE); file->flags = cpu_to_be16(HFSPLUS_FILE_THREAD_EXISTS); file->id = cpu_to_be32(cnid); - file->create_date = file->content_mod_date = - file->attribute_mod_date = file->access_date = hfsp_now2mt(); + HFSPLUS_I(inode).create_date = + file->create_date = + file->content_mod_date = + file->attribute_mod_date = + file->access_date = hfsp_now2mt(); if (cnid == inode->i_ino) { hfsplus_set_perms(inode, &file->permissions); file->user_info.fdType = cpu_to_be32(HFSPLUS_SB(inode->i_sb).type); diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h index 4608171f45d..7ae393637a0 100644 --- a/fs/hfsplus/hfsplus_fs.h +++ b/fs/hfsplus/hfsplus_fs.h @@ -166,6 +166,7 @@ struct hfsplus_inode_info { struct inode *rsrc_inode; unsigned long flags; + __be32 create_date; /* Device number in hfsplus_permissions in catalog */ u32 dev; /* BSD system and user file flags */ diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c index 182eb317797..12ed2b7d046 100644 --- a/fs/hfsplus/inode.c +++ b/fs/hfsplus/inode.c @@ -431,7 +431,8 @@ int hfsplus_cat_read_inode(struct inode *inode, struct hfs_find_data *fd) inode->i_size = 2 + be32_to_cpu(folder->valence); inode->i_atime = hfsp_mt2ut(folder->access_date); inode->i_mtime = hfsp_mt2ut(folder->content_mod_date); - inode->i_ctime = inode->i_mtime; + inode->i_ctime = hfsp_mt2ut(folder->attribute_mod_date); + HFSPLUS_I(inode).create_date = folder->create_date; HFSPLUS_I(inode).fs_blocks = 0; inode->i_op = &hfsplus_dir_inode_operations; inode->i_fop = &hfsplus_dir_operations; @@ -462,7 +463,8 @@ int hfsplus_cat_read_inode(struct inode *inode, struct hfs_find_data *fd) } inode->i_atime = hfsp_mt2ut(file->access_date); inode->i_mtime = hfsp_mt2ut(file->content_mod_date); - inode->i_ctime = inode->i_mtime; + inode->i_ctime = hfsp_mt2ut(file->attribute_mod_date); + HFSPLUS_I(inode).create_date = file->create_date; } else { printk(KERN_ERR "hfs: bad catalog entry used to create inode\n"); res = -EIO; -- cgit v1.2.3 From af8c85bb6d4e5352551277edd8448c4dfb2328ab Mon Sep 17 00:00:00 2001 From: Roman Zippel Date: Wed, 18 Jan 2006 17:43:10 -0800 Subject: [PATCH] hfs: set correct create date for links HFS+ also requires the correct creation date so recent version of OS X recognize it as link. Improve link handling: - if something is wrong with the link, ignore the link attribute and treat it as regular file (this also fixes a missing unlock during lookup). - check for incorrect link counts during unlink. Signed-off-by: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hfsplus/catalog.c | 1 + fs/hfsplus/dir.c | 33 +++++++++++++++++++++------------ 2 files changed, 22 insertions(+), 12 deletions(-) diff --git a/fs/hfsplus/catalog.c b/fs/hfsplus/catalog.c index 04255af3470..04058c8096d 100644 --- a/fs/hfsplus/catalog.c +++ b/fs/hfsplus/catalog.c @@ -127,6 +127,7 @@ static int hfsplus_cat_build_record(hfsplus_cat_entry *entry, u32 cnid, struct i file->user_info.fdType = cpu_to_be32(HFSP_HARDLINK_TYPE); file->user_info.fdCreator = cpu_to_be32(HFSP_HFSPLUS_CREATOR); file->user_info.fdFlags = cpu_to_be16(0x100); + file->create_date = HFSPLUS_I(HFSPLUS_SB(inode->i_sb).hidden_dir).create_date; file->permissions.dev = cpu_to_be32(HFSPLUS_I(inode).dev); } return sizeof(*file); diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c index 82c22376547..01a6fe3a395 100644 --- a/fs/hfsplus/dir.c +++ b/fs/hfsplus/dir.c @@ -66,21 +66,28 @@ again: } cnid = be32_to_cpu(entry.file.id); if (entry.file.user_info.fdType == cpu_to_be32(HFSP_HARDLINK_TYPE) && - entry.file.user_info.fdCreator == cpu_to_be32(HFSP_HFSPLUS_CREATOR)) { + entry.file.user_info.fdCreator == cpu_to_be32(HFSP_HFSPLUS_CREATOR) && + (entry.file.create_date == HFSPLUS_I(HFSPLUS_SB(sb).hidden_dir).create_date || + entry.file.create_date == HFSPLUS_I(sb->s_root->d_inode).create_date) && + HFSPLUS_SB(sb).hidden_dir) { struct qstr str; char name[32]; if (dentry->d_fsdata) { - err = -ENOENT; - inode = NULL; - goto out; + /* + * We found a link pointing to another link, + * so ignore it and treat it as regular file. + */ + cnid = (unsigned long)dentry->d_fsdata; + linkid = 0; + } else { + dentry->d_fsdata = (void *)(unsigned long)cnid; + linkid = be32_to_cpu(entry.file.permissions.dev); + str.len = sprintf(name, "iNode%d", linkid); + str.name = name; + hfsplus_cat_build_key(sb, fd.search_key, HFSPLUS_SB(sb).hidden_dir->i_ino, &str); + goto again; } - dentry->d_fsdata = (void *)(unsigned long)cnid; - linkid = be32_to_cpu(entry.file.permissions.dev); - str.len = sprintf(name, "iNode%d", linkid); - str.name = name; - hfsplus_cat_build_key(sb, fd.search_key, HFSPLUS_SB(sb).hidden_dir->i_ino, &str); - goto again; } else if (!dentry->d_fsdata) dentry->d_fsdata = (void *)(unsigned long)cnid; } else { @@ -330,7 +337,8 @@ static int hfsplus_unlink(struct inode *dir, struct dentry *dentry) if (res) return res; - inode->i_nlink--; + if (inode->i_nlink > 0) + inode->i_nlink--; hfsplus_delete_inode(inode); if (inode->i_ino != cnid && !inode->i_nlink) { if (!atomic_read(&HFSPLUS_I(inode).opencnt)) { @@ -339,7 +347,8 @@ static int hfsplus_unlink(struct inode *dir, struct dentry *dentry) hfsplus_delete_inode(inode); } else inode->i_flags |= S_DEAD; - } + } else + inode->i_nlink = 0; inode->i_ctime = CURRENT_TIME_SEC; mark_inode_dirty(inode); -- cgit v1.2.3 From 6b192832daae6d141063c49ae1ded6f7dddee50e Mon Sep 17 00:00:00 2001 From: Roman Zippel Date: Wed, 18 Jan 2006 17:43:12 -0800 Subject: [PATCH] hfs: set type/creator for symlinks Set the correct type and creator for symlinks. Signed-off-by: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hfsplus/catalog.c | 9 +++++++-- fs/hfsplus/hfsplus_raw.h | 3 +++ 2 files changed, 10 insertions(+), 2 deletions(-) diff --git a/fs/hfsplus/catalog.c b/fs/hfsplus/catalog.c index 04058c8096d..f2d7c49ce75 100644 --- a/fs/hfsplus/catalog.c +++ b/fs/hfsplus/catalog.c @@ -119,8 +119,13 @@ static int hfsplus_cat_build_record(hfsplus_cat_entry *entry, u32 cnid, struct i file->access_date = hfsp_now2mt(); if (cnid == inode->i_ino) { hfsplus_set_perms(inode, &file->permissions); - file->user_info.fdType = cpu_to_be32(HFSPLUS_SB(inode->i_sb).type); - file->user_info.fdCreator = cpu_to_be32(HFSPLUS_SB(inode->i_sb).creator); + if (S_ISLNK(inode->i_mode)) { + file->user_info.fdType = cpu_to_be32(HFSP_SYMLINK_TYPE); + file->user_info.fdCreator = cpu_to_be32(HFSP_SYMLINK_CREATOR); + } else { + file->user_info.fdType = cpu_to_be32(HFSPLUS_SB(inode->i_sb).type); + file->user_info.fdCreator = cpu_to_be32(HFSPLUS_SB(inode->i_sb).creator); + } if ((file->permissions.rootflags | file->permissions.userflags) & HFSPLUS_FLG_IMMUTABLE) file->flags |= cpu_to_be16(HFSPLUS_FILE_LOCKED); } else { diff --git a/fs/hfsplus/hfsplus_raw.h b/fs/hfsplus/hfsplus_raw.h index ccc47966dc5..49205531a50 100644 --- a/fs/hfsplus/hfsplus_raw.h +++ b/fs/hfsplus/hfsplus_raw.h @@ -43,6 +43,9 @@ #define HFSP_HARDLINK_TYPE 0x686c6e6b /* 'hlnk' */ #define HFSP_HFSPLUS_CREATOR 0x6866732b /* 'hfs+' */ +#define HFSP_SYMLINK_TYPE 0x736c6e6b /* 'slnk' */ +#define HFSP_SYMLINK_CREATOR 0x72686170 /* 'rhap' */ + #define HFSP_MOUNT_VERSION 0x482b4c78 /* 'H+Lx' */ /* Structures used on disk */ -- cgit v1.2.3 From f193fbab2e4710f629e7c1859d4908646b47b126 Mon Sep 17 00:00:00 2001 From: YAMAMOTO Takashi Date: Wed, 18 Jan 2006 17:43:13 -0800 Subject: [PATCH] nfsd: check error status from nfsd_sync_dir Change nfsd_sync_dir to return an error if ->sync fails, and pass that error up through the stack. This involves a number of rearrangements of error paths, and care to distinguish between Linux -errno numbers and NFSERR numbers. In the 'create' routines, we continue with the 'setattr' even if a previous sync_dir failed. This patch is quite different from Takashi's in a few ways, but there is still a strong lineage. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/vfs.c | 80 ++++++++++++++++++++++++++--------------------- include/linux/nfsd/nfsd.h | 2 +- 2 files changed, 45 insertions(+), 37 deletions(-) diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index eef0576a778..acf637869cc 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -710,14 +710,15 @@ static inline int nfsd_dosync(struct file *filp, struct dentry *dp, { struct inode *inode = dp->d_inode; int (*fsync) (struct file *, struct dentry *, int); - int err = nfs_ok; + int err; - filemap_fdatawrite(inode->i_mapping); - if (fop && (fsync = fop->fsync)) - err=fsync(filp, dp, 0); - filemap_fdatawait(inode->i_mapping); + err = filemap_fdatawrite(inode->i_mapping); + if (err == 0 && fop && (fsync = fop->fsync)) + err = fsync(filp, dp, 0); + if (err == 0) + err = filemap_fdatawait(inode->i_mapping); - return nfserrno(err); + return err; } @@ -734,10 +735,10 @@ nfsd_sync(struct file *filp) return err; } -void +int nfsd_sync_dir(struct dentry *dp) { - nfsd_dosync(NULL, dp, dp->d_inode->i_fop); + return nfsd_dosync(NULL, dp, dp->d_inode->i_fop); } /* @@ -1065,6 +1066,7 @@ nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, if (EX_ISSYNC(fhp->fh_export)) { if (file->f_op && file->f_op->fsync) { err = nfsd_sync(file); + err = nfserrno(err); } else { err = nfserr_notsupp; } @@ -1175,7 +1177,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, goto out_nfserr; if (EX_ISSYNC(fhp->fh_export)) { - nfsd_sync_dir(dentry); + err = nfsd_sync_dir(dentry); write_inode_now(dchild->d_inode, 1); } @@ -1185,9 +1187,11 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, * send along the gid when it tries to implement setgid * directories via NFS. */ - err = 0; - if ((iap->ia_valid &= ~(ATTR_UID|ATTR_GID|ATTR_MODE)) != 0) - err = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); + if ((iap->ia_valid &= ~(ATTR_UID|ATTR_GID|ATTR_MODE)) != 0) { + int err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); + if (err2) + err = err2; + } /* * Update the file handle to get the new inode info. */ @@ -1306,17 +1310,12 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, goto out_nfserr; if (EX_ISSYNC(fhp->fh_export)) { - nfsd_sync_dir(dentry); + err = nfsd_sync_dir(dentry); + if (err) + err = nfserrno(err); /* setattr will sync the child (or not) */ } - /* - * Update the filehandle to get the new inode info. - */ - err = fh_update(resfhp); - if (err) - goto out; - if (createmode == NFS3_CREATE_EXCLUSIVE) { /* Cram the verifier into atime/mtime/mode */ iap->ia_valid = ATTR_MTIME|ATTR_ATIME @@ -1337,8 +1336,17 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, * implement setgid directories via NFS. Clear out all that cruft. */ set_attr: - if ((iap->ia_valid &= ~(ATTR_UID|ATTR_GID)) != 0) - err = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); + if ((iap->ia_valid &= ~(ATTR_UID|ATTR_GID)) != 0) { + int err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); + if (err2) + err = nfserrno(err2); + } + + /* + * Update the filehandle to get the new inode info. + */ + if (!err) + err = fh_update(resfhp); out: fh_unlock(fhp); @@ -1447,10 +1455,10 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, } else err = vfs_symlink(dentry->d_inode, dnew, path, mode); - if (!err) { + if (!err) if (EX_ISSYNC(fhp->fh_export)) - nfsd_sync_dir(dentry); - } else + err = nfsd_sync_dir(dentry); + if (err) err = nfserrno(err); fh_unlock(fhp); @@ -1506,8 +1514,10 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp, err = vfs_link(dold, dirp, dnew); if (!err) { if (EX_ISSYNC(ffhp->fh_export)) { - nfsd_sync_dir(ddir); + err = nfsd_sync_dir(ddir); write_inode_now(dest, 1); + if (err) + err = nfserrno(err); } } else { if (err == -EXDEV && rqstp->rq_vers == 2) @@ -1595,8 +1605,9 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, #endif err = vfs_rename(fdir, odentry, tdir, ndentry); if (!err && EX_ISSYNC(tfhp->fh_export)) { - nfsd_sync_dir(tdentry); - nfsd_sync_dir(fdentry); + err = nfsd_sync_dir(tdentry); + if (!err) + err = nfsd_sync_dir(fdentry); } out_dput_new: @@ -1671,17 +1682,14 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, dput(rdentry); - if (err) - goto out_nfserr; - if (EX_ISSYNC(fhp->fh_export)) - nfsd_sync_dir(dentry); - -out: - return err; + if (err == 0 && + EX_ISSYNC(fhp->fh_export)) + err = nfsd_sync_dir(dentry); out_nfserr: err = nfserrno(err); - goto out; +out: + return err; } /* diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 51c231a1e5a..ec7c2e872d7 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -124,7 +124,7 @@ int nfsd_statfs(struct svc_rqst *, struct svc_fh *, int nfsd_notify_change(struct inode *, struct iattr *); int nfsd_permission(struct svc_export *, struct dentry *, int); -void nfsd_sync_dir(struct dentry *dp); +int nfsd_sync_dir(struct dentry *dp); #if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) #ifdef CONFIG_NFSD_V2_ACL -- cgit v1.2.3 From 7e8f05934d540aeb45c1cc3cc3a0e811bd61866b Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Wed, 18 Jan 2006 17:43:14 -0800 Subject: [PATCH] nfsd: remove inline from a couple of large NFS functions These are both called from two places close together. I could rearrange that code so there is only one call site, but just removing the 'inline' is probably best. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/vfs.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index acf637869cc..6acec341676 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -815,7 +815,7 @@ nfsd_read_actor(read_descriptor_t *desc, struct page *page, unsigned long offset return size; } -static inline int +static int nfsd_vfs_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, unsigned long *count) { @@ -879,7 +879,7 @@ static void kill_suid(struct dentry *dentry) mutex_unlock(&dentry->d_inode->i_mutex); } -static inline int +static int nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, unsigned long cnt, int *stablep) -- cgit v1.2.3 From 1918e341383ab787d6c5b17200f4ed901b10c777 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 18 Jan 2006 17:43:16 -0800 Subject: [PATCH] svcrpc: save and restore the daddr field when request deferred The server code currently keeps track of the destination address on every request so that it can reply using the same address. However we forget to do that in the case of a deferred request. Remedy this oversight. >From folks at PolyServe. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sunrpc/svc.h | 1 + net/sunrpc/svcsock.c | 2 ++ 2 files changed, 3 insertions(+) diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index e4086ec8b95..50cab2a09f2 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -246,6 +246,7 @@ struct svc_deferred_req { u32 prot; /* protocol (UDP or TCP) */ struct sockaddr_in addr; struct svc_sock *svsk; /* where reply must go */ + u32 daddr; /* where reply must come from */ struct cache_deferred_req handle; int argslen; u32 args[0]; diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index e67613e4eb1..50580620e89 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -1527,6 +1527,7 @@ svc_defer(struct cache_req *req) dr->handle.owner = rqstp->rq_server; dr->prot = rqstp->rq_prot; dr->addr = rqstp->rq_addr; + dr->daddr = rqstp->rq_daddr; dr->argslen = rqstp->rq_arg.len >> 2; memcpy(dr->args, rqstp->rq_arg.head[0].iov_base-skip, dr->argslen<<2); } @@ -1552,6 +1553,7 @@ static int svc_deferred_recv(struct svc_rqst *rqstp) rqstp->rq_arg.len = dr->argslen<<2; rqstp->rq_prot = dr->prot; rqstp->rq_addr = dr->addr; + rqstp->rq_daddr = dr->daddr; return dr->argslen<<2; } -- cgit v1.2.3 From a6f6ef2f1d7329111fcad7db48fb7adba5062d0a Mon Sep 17 00:00:00 2001 From: Andy Adamson Date: Wed, 18 Jan 2006 17:43:17 -0800 Subject: [PATCH] nfsd4: misc lock fixes Logic fixes for LOCK and UNLOCK. - Move the permission check on the current file handle outside of nfs4_lock_state() - remove the file manager fl_release_private calls; fl_ops is not set. Signed-off-by: Andy Adamson Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 6bbefd06f10..3d4a2ec97ca 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -2700,6 +2700,11 @@ nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock if (check_lock_length(lock->lk_offset, lock->lk_length)) return nfserr_inval; + if ((status = fh_verify(rqstp, current_fh, S_IFREG, MAY_LOCK))) { + dprintk("NFSD: nfsd4_lock: permission denied!\n"); + return status; + } + nfs4_lock_state(); if (lock->lk_is_new) { @@ -2757,11 +2762,6 @@ nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock /* lock->lk_stateowner and lock_stp have been created or found */ filp = lock_stp->st_vfs_file; - if ((status = fh_verify(rqstp, current_fh, S_IFREG, MAY_LOCK))) { - dprintk("NFSD: nfsd4_lock: permission denied!\n"); - goto out; - } - status = nfserr_grace; if (nfs4_in_grace() && !lock->lk_reclaim) goto out; @@ -2802,8 +2802,6 @@ nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock */ status = posix_lock_file(filp, &file_lock); - if (file_lock.fl_ops && file_lock.fl_ops->fl_release_private) - file_lock.fl_ops->fl_release_private(&file_lock); dprintk("NFSD: nfsd4_lock: posix_lock_file status %d\n",status); switch (-status) { case 0: /* success! */ @@ -2977,8 +2975,6 @@ nfsd4_locku(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock * Try to unlock the file in the VFS. */ status = posix_lock_file(filp, &file_lock); - if (file_lock.fl_ops && file_lock.fl_ops->fl_release_private) - file_lock.fl_ops->fl_release_private(&file_lock); if (status) { dprintk("NFSD: nfs4_locku: posix_lock_file failed!\n"); goto out_nfserr; -- cgit v1.2.3 From 8a280510852959c0d51b1d625e90c0491c238368 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 18 Jan 2006 17:43:18 -0800 Subject: [PATCH] nfsd4: fix nfsd4_lock cleanup on failure release_state_owner also puts the lock owner on the close_lru. There's no need for that, though; replays of the failed lock would be handled by the openowner not the lockowner. Also consolidate the cleanup a bit, fixing leaks that can happen if errors occur between the time a new lock owner is allocated and the lock is done. Remove a comment and dprintk that look a little redundant. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 18 ++++-------------- 1 file changed, 4 insertions(+), 14 deletions(-) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 3d4a2ec97ca..5bf7fd3947c 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -2744,10 +2744,8 @@ nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock if (lock_sop == NULL) goto out; lock_stp = alloc_init_lock_stateid(lock_sop, fp, open_stp); - if (lock_stp == NULL) { - release_stateowner(lock_sop); + if (lock_stp == NULL) goto out; - } } else { /* lock (lock owner + lock stateid) already exists */ status = nfs4_preprocess_seqid_op(current_fh, @@ -2815,7 +2813,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock status = nfserr_deadlock; default: dprintk("NFSD: nfsd4_lock: posix_lock_file() failed! status %d\n",status); - goto out_destroy_new_stateid; + goto out; } conflicting_lock: @@ -2829,17 +2827,9 @@ conflicting_lock: goto out; } nfs4_set_lock_denied(conflock, &lock->lk_denied); - -out_destroy_new_stateid: - if (lock->lk_is_new) { - dprintk("NFSD: nfsd4_lock: destroy new stateid!\n"); - /* - * An error encountered after instantiation of the new - * stateid has forced us to destroy it. - */ - release_state_owner(lock_stp, LOCK_STATE); - } out: + if (status && lock->lk_is_new && lock_sop) + release_stateowner(lock_sop); if (lock->lk_stateowner) { nfs4_get_stateowner(lock->lk_stateowner); *replay_owner = lock->lk_stateowner; -- cgit v1.2.3 From 3a65588adc4401622b204caa897123e16a4a0318 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 18 Jan 2006 17:43:19 -0800 Subject: [PATCH] nfsd4: rename lk_stateowner One of the things that's confusing about nfsd4_lock is that the lk_stateowner field could be set to either of two different lockowners: the open owner or the lock owner. Rename to lk_replay_owner and add a comment to make it clear that it's used for whichever stateowner has its sequence id bumped for replay detection. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 16 ++++++++-------- fs/nfsd/nfs4xdr.c | 5 ++--- include/linux/nfsd/xdr4.h | 5 +++-- 3 files changed, 13 insertions(+), 13 deletions(-) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 5bf7fd3947c..578ea521c82 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -2725,11 +2725,11 @@ nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock lock->lk_new_open_seqid, &lock->lk_new_open_stateid, CHECK_FH | OPEN_STATE, - &lock->lk_stateowner, &open_stp, + &lock->lk_replay_owner, &open_stp, lock); if (status) goto out; - open_sop = lock->lk_stateowner; + open_sop = lock->lk_replay_owner; /* create lockowner and lock stateid */ fp = open_stp->st_file; strhashval = lock_ownerstr_hashval(fp->fi_inode, @@ -2752,12 +2752,12 @@ nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock lock->lk_old_lock_seqid, &lock->lk_old_lock_stateid, CHECK_FH | LOCK_STATE, - &lock->lk_stateowner, &lock_stp, lock); + &lock->lk_replay_owner, &lock_stp, lock); if (status) goto out; - lock_sop = lock->lk_stateowner; + lock_sop = lock->lk_replay_owner; } - /* lock->lk_stateowner and lock_stp have been created or found */ + /* lock->lk_replay_owner and lock_stp have been created or found */ filp = lock_stp->st_vfs_file; status = nfserr_grace; @@ -2830,9 +2830,9 @@ conflicting_lock: out: if (status && lock->lk_is_new && lock_sop) release_stateowner(lock_sop); - if (lock->lk_stateowner) { - nfs4_get_stateowner(lock->lk_stateowner); - *replay_owner = lock->lk_stateowner; + if (lock->lk_replay_owner) { + nfs4_get_stateowner(lock->lk_replay_owner); + *replay_owner = lock->lk_replay_owner; } nfs4_unlock_state(); return status; diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index dcd67318694..6b743327686 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -528,7 +528,7 @@ nfsd4_decode_lock(struct nfsd4_compoundargs *argp, struct nfsd4_lock *lock) { DECODE_HEAD; - lock->lk_stateowner = NULL; + lock->lk_replay_owner = NULL; /* * type, reclaim(boolean), offset, length, new_lock_owner(boolean) */ @@ -1895,7 +1895,6 @@ nfsd4_encode_lock_denied(struct nfsd4_compoundres *resp, struct nfsd4_lock_denie static void nfsd4_encode_lock(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_lock *lock) { - ENCODE_SEQID_OP_HEAD; if (!nfserr) { @@ -1906,7 +1905,7 @@ nfsd4_encode_lock(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_lock } else if (nfserr == nfserr_denied) nfsd4_encode_lock_denied(resp, &lock->lk_denied); - ENCODE_SEQID_OP_TAIL(lock->lk_stateowner); + ENCODE_SEQID_OP_TAIL(lock->lk_replay_owner); } static void diff --git a/include/linux/nfsd/xdr4.h b/include/linux/nfsd/xdr4.h index 8903688890c..77adba7d228 100644 --- a/include/linux/nfsd/xdr4.h +++ b/include/linux/nfsd/xdr4.h @@ -145,8 +145,9 @@ struct nfsd4_lock { } ok; struct nfsd4_lock_denied denied; } u; - - struct nfs4_stateowner *lk_stateowner; + /* The lk_replay_owner is the open owner in the open_to_lock_owner + * case and the lock owner otherwise: */ + struct nfs4_stateowner *lk_replay_owner; }; #define lk_new_open_seqid v.new.open_seqid #define lk_new_open_stateid v.new.open_stateid -- cgit v1.2.3 From 04ef59548470b81829e8593c1b39776ce0534d68 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 18 Jan 2006 17:43:21 -0800 Subject: [PATCH] nfsd4: remove release_state_owner() It's confusing having both release_stateowner() and release_state_owner(). And as it turns out, release_state_owner() is short and only called from one place; so just remove it. Also note the confirmed check is superfluous there--preprocess_seqid_op already check this. And remove a redundant comment and a superfluous line assignment while we're at it. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 31 +++++++++---------------------- 1 file changed, 9 insertions(+), 22 deletions(-) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 578ea521c82..3510e2ca40d 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1178,7 +1178,6 @@ release_stateid(struct nfs4_stateid *stp, int flags) locks_remove_posix(filp, (fl_owner_t) stp->st_stateowner); put_nfs4_file(stp->st_file); kmem_cache_free(stateid_slab, stp); - stp = NULL; } static void @@ -1191,22 +1190,6 @@ move_to_close_lru(struct nfs4_stateowner *sop) sop->so_time = get_seconds(); } -static void -release_state_owner(struct nfs4_stateid *stp, int flag) -{ - struct nfs4_stateowner *sop = stp->st_stateowner; - - dprintk("NFSD: release_state_owner\n"); - release_stateid(stp, flag); - - /* place unused nfs4_stateowners on so_close_lru list to be - * released by the laundromat service after the lease period - * to enable us to handle CLOSE replay - */ - if (sop->so_confirmed && list_empty(&sop->so_stateids)) - move_to_close_lru(sop); -} - static int cmp_owner_str(struct nfs4_stateowner *sop, struct xdr_netobj *owner, clientid_t *clid) { return ((sop->so_owner.len == owner->len) && @@ -2423,15 +2406,19 @@ nfsd4_close(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_clos CHECK_FH | OPEN_STATE | CLOSE_STATE, &close->cl_stateowner, &stp, NULL))) goto out; - /* - * Return success, but first update the stateid. - */ status = nfs_ok; update_stateid(&stp->st_stateid); memcpy(&close->cl_stateid, &stp->st_stateid, sizeof(stateid_t)); - /* release_state_owner() calls nfsd_close() if needed */ - release_state_owner(stp, OPEN_STATE); + /* release_stateid() calls nfsd_close() if needed */ + release_stateid(stp, OPEN_STATE); + + /* place unused nfs4_stateowners on so_close_lru list to be + * released by the laundromat service after the lease period + * to enable us to handle CLOSE replay + */ + if (list_empty(&close->cl_stateowner->so_stateids)) + move_to_close_lru(close->cl_stateowner); out: if (close->cl_stateowner) { nfs4_get_stateowner(close->cl_stateowner); -- cgit v1.2.3 From 796dadfd02eda1e3e6e42ecc8379d8b1c1523ddf Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 18 Jan 2006 17:43:22 -0800 Subject: [PATCH] nfsd4: fix check_for_locks Fix some bad logic. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 3510e2ca40d..e13d2233ff8 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -2989,9 +2989,10 @@ check_for_locks(struct file *filp, struct nfs4_stateowner *lowner) lock_kernel(); for (flpp = &inode->i_flock; *flpp != NULL; flpp = &(*flpp)->fl_next) { - if ((*flpp)->fl_owner == (fl_owner_t)lowner) + if ((*flpp)->fl_owner == (fl_owner_t)lowner) { status = 1; goto out; + } } out: unlock_kernel(); -- cgit v1.2.3 From fd44527707f2697fd2959e8bdb321ae588d150e2 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 18 Jan 2006 17:43:23 -0800 Subject: [PATCH] nfsd4: operation debugging Simple, useful debugging printk: print the number of each op as we process it. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4proc.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 361b4007d4a..2a1766ce216 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -768,6 +768,8 @@ nfsd4_proc_compound(struct svc_rqst *rqstp, while (!status && resp->opcnt < args->opcnt) { op = &args->ops[resp->opcnt++]; + dprintk("nfsv4 compound op #%d: %d\n", resp->opcnt, op->opnum); + /* * The XDR decode routines may have pre-set op->status; * for example, if there is a miscellaneous XDR error -- cgit v1.2.3 From 822f1005ae1f3a4a8b136f38a6933d3f719f4c4a Mon Sep 17 00:00:00 2001 From: Andy Adamson Date: Wed, 18 Jan 2006 17:43:24 -0800 Subject: [PATCH] svcrpc: gss: handle the GSS_S_CONTINUE Kerberos context initiation is handled in a single round trip, but other mechanisms (including spkm3) may require more, so we need to handle the GSS_S_CONTINUE case in svcauth_gss_accept. Send a null verifier. Signed-off-by: Andy Adamson Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- net/sunrpc/auth_gss/svcauth_gss.c | 30 +++++++++++++++++++++++++----- 1 file changed, 25 insertions(+), 5 deletions(-) diff --git a/net/sunrpc/auth_gss/svcauth_gss.c b/net/sunrpc/auth_gss/svcauth_gss.c index e4ada15ed85..d2ccc7e8faa 100644 --- a/net/sunrpc/auth_gss/svcauth_gss.c +++ b/net/sunrpc/auth_gss/svcauth_gss.c @@ -585,6 +585,20 @@ gss_verify_header(struct svc_rqst *rqstp, struct rsc *rsci, return SVC_OK; } +static int +gss_write_null_verf(struct svc_rqst *rqstp) +{ + u32 *p; + + svc_putu32(rqstp->rq_res.head, htonl(RPC_AUTH_NULL)); + p = rqstp->rq_res.head->iov_base + rqstp->rq_res.head->iov_len; + /* don't really need to check if head->iov_len > PAGE_SIZE ... */ + *p++ = 0; + if (!xdr_ressize_check(rqstp, p)) + return -1; + return 0; +} + static int gss_write_verf(struct svc_rqst *rqstp, struct gss_ctx *ctx_id, u32 seq) { @@ -876,12 +890,18 @@ svcauth_gss_accept(struct svc_rqst *rqstp, u32 *authp) case -ENOENT: goto drop; case 0: - rsci = gss_svc_searchbyctx(&rsip->out_handle); - if (!rsci) { - goto drop; + if (rsip->major_status == GSS_S_COMPLETE) { + rsci = gss_svc_searchbyctx(&rsip->out_handle); + if (!rsci) { + goto drop; + } + if (gss_write_verf(rqstp, rsci->mechctx, + GSS_SEQ_WIN)) + goto drop; + } else { + if (gss_write_null_verf(rqstp)) + goto drop; } - if (gss_write_verf(rqstp, rsci->mechctx, GSS_SEQ_WIN)) - goto drop; if (resv->iov_len + 4 > PAGE_SIZE) goto drop; svc_putu32(resv, rpc_success); -- cgit v1.2.3 From 91a4762e0ab0880fa00e8f0b7a052e4929d867a6 Mon Sep 17 00:00:00 2001 From: Kevin Coffman Date: Wed, 18 Jan 2006 17:43:25 -0800 Subject: [PATCH] svcrpc: gss: server context init failure handling We require the server's gssd to create a completed context before asking the kernel to send a final context init reply. However, gssd could be buggy, or under some bizarre circumstances we might purge the context from our cache before we get the chance to use it here. Handle this case by returning GSS_S_NO_CONTEXT to the client. Also move the relevant code here to a separate function rather than nesting excessively. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- net/sunrpc/auth_gss/svcauth_gss.c | 29 +++++++++++++++++------------ 1 file changed, 17 insertions(+), 12 deletions(-) diff --git a/net/sunrpc/auth_gss/svcauth_gss.c b/net/sunrpc/auth_gss/svcauth_gss.c index d2ccc7e8faa..fdad66dc9a9 100644 --- a/net/sunrpc/auth_gss/svcauth_gss.c +++ b/net/sunrpc/auth_gss/svcauth_gss.c @@ -755,6 +755,21 @@ svcauth_gss_set_client(struct svc_rqst *rqstp) return SVC_OK; } +static inline int +gss_write_init_verf(struct svc_rqst *rqstp, struct rsi *rsip) +{ + struct rsc *rsci; + + if (rsip->major_status != GSS_S_COMPLETE) + return gss_write_null_verf(rqstp); + rsci = gss_svc_searchbyctx(&rsip->out_handle); + if (rsci == NULL) { + rsip->major_status = GSS_S_NO_CONTEXT; + return gss_write_null_verf(rqstp); + } + return gss_write_verf(rqstp, rsci->mechctx, GSS_SEQ_WIN); +} + /* * Accept an rpcsec packet. * If context establishment, punt to user space @@ -890,18 +905,8 @@ svcauth_gss_accept(struct svc_rqst *rqstp, u32 *authp) case -ENOENT: goto drop; case 0: - if (rsip->major_status == GSS_S_COMPLETE) { - rsci = gss_svc_searchbyctx(&rsip->out_handle); - if (!rsci) { - goto drop; - } - if (gss_write_verf(rqstp, rsci->mechctx, - GSS_SEQ_WIN)) - goto drop; - } else { - if (gss_write_null_verf(rqstp)) - goto drop; - } + if (gss_write_init_verf(rqstp, rsip)) + goto drop; if (resv->iov_len + 4 > PAGE_SIZE) goto drop; svc_putu32(resv, rpc_success); -- cgit v1.2.3 From 5fb8b49e2955cc473929c5994b8389111daed59d Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 18 Jan 2006 17:43:26 -0800 Subject: [PATCH] svcrpc: gss: svc context creation error handling Allow mechanisms to return more varied errors on the context creation downcall. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- net/sunrpc/auth_gss/svcauth_gss.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/sunrpc/auth_gss/svcauth_gss.c b/net/sunrpc/auth_gss/svcauth_gss.c index fdad66dc9a9..23632d84d8d 100644 --- a/net/sunrpc/auth_gss/svcauth_gss.c +++ b/net/sunrpc/auth_gss/svcauth_gss.c @@ -420,7 +420,8 @@ static int rsc_parse(struct cache_detail *cd, gss_mech_put(gm); goto out; } - if (gss_import_sec_context(buf, len, gm, &rsci.mechctx)) { + status = gss_import_sec_context(buf, len, gm, &rsci.mechctx); + if (status) { gss_mech_put(gm); goto out; } -- cgit v1.2.3 From d22749b62f3e12de26b86a200f9a5bf7afe3590e Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 18 Jan 2006 17:43:27 -0800 Subject: [PATCH] nfsd4: fix open of recovery directory We should be opening this directory RDONLY, not RDWR. Thanks to Christoph Hellwig for the bug report. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4recover.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index be963a133aa..64f4af3651a 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -222,8 +222,7 @@ 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_RDWR); + filp = dentry_open(dget(dir), mntget(rec_dir.mnt), O_RDONLY); status = PTR_ERR(filp); if (IS_ERR(filp)) goto out; -- cgit v1.2.3 From c2642ab05b855d2d3b850ddf90dbb02b1b9358ac Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 18 Jan 2006 17:43:29 -0800 Subject: [PATCH] nfsd4: recovery lookup dir check Make sure we get a directory when we look up the recovery directory. Thanks to Christoph Hellwig for the bug report. Based on feedback from Christoph and others, we may remove the need for this lookup and just pass in a file descriptor from userspace instead, and/or completely move the directory handling to userspace. For now we're just fixing the obvious bugs. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4recover.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 64f4af3651a..06da7506363 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -399,9 +399,10 @@ nfsd4_init_recdir(char *rec_dirname) nfs4_save_user(&uid, &gid); - status = path_lookup(rec_dirname, LOOKUP_FOLLOW, &rec_dir); - if (status == -ENOENT) - printk("NFSD: recovery directory %s doesn't exist\n", + status = path_lookup(rec_dirname, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, + &rec_dir); + if (status) + printk("NFSD: unable to find recovery directory %s\n", rec_dirname); if (!status) -- cgit v1.2.3 From a525825df15221a95d4c1f5a291d9fde77ef10bc Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 18 Jan 2006 17:43:30 -0800 Subject: [PATCH] nfsd4: handle replays of failed open reclaims We need to make sure open reclaims are marked confirmed immediately so that we can handle replays even if they fail (e.g. with a seqid-incrementing error). (See 8.1.8.) Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4proc.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 2a1766ce216..69ee182575a 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -210,6 +210,7 @@ nfsd4_open(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open goto out; break; case NFS4_OPEN_CLAIM_PREVIOUS: + open->op_stateowner->so_confirmed = 1; /* * The CURRENT_FH is already set to the file being * opened. (1) set open->op_cinfo, (2) set @@ -221,6 +222,7 @@ nfsd4_open(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open goto out; break; case NFS4_OPEN_CLAIM_DELEGATE_PREV: + open->op_stateowner->so_confirmed = 1; printk("NFSD: unsupported OPEN claim type %d\n", open->op_claim_type); status = nfserr_notsupp; -- cgit v1.2.3 From ae8b625313db4dd4b060962c2a02f3a2837ca61b Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 18 Jan 2006 17:43:32 -0800 Subject: [PATCH] nfsd4: no replays on unconfirmed owners We shouldn't check for replays until after checking whether the open owner is confirmed. Clients are allowed to reuse openowners without bumping the seqid. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 24 +++++++++++------------- 1 file changed, 11 insertions(+), 13 deletions(-) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index e13d2233ff8..dc792b6b451 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1465,8 +1465,16 @@ nfsd4_process_open1(struct nfsd4_open *open) sop = find_openstateowner_str(strhashval, open); if (sop) { open->op_stateowner = sop; - /* check for replay */ - if (open->op_seqid == sop->so_seqid - 1){ + if (!sop->so_confirmed) { + /* Replace any unconfirmed stateowner without + * even checking for replays */ + clp = sop->so_client; + release_stateowner(sop); + } else if (open->op_seqid == sop->so_seqid) { + /* normal case */ + goto renew; + } else if (open->op_seqid == sop->so_seqid - 1) { + /* replay */ if (sop->so_replay.rp_buflen) return NFSERR_REPLAY_ME; else { @@ -1480,19 +1488,9 @@ nfsd4_process_open1(struct nfsd4_open *open) " replay with no replay cache\n"); goto renew; } - } else if (sop->so_confirmed) { - if (open->op_seqid == sop->so_seqid) - goto renew; + } else { status = nfserr_bad_seqid; goto out; - } else { - /* If we get here, we received an OPEN for an - * unconfirmed nfs4_stateowner. Since the seqid's are - * different, purge the existing nfs4_stateowner, and - * instantiate a new one. - */ - clp = sop->so_client; - release_stateowner(sop); } } else { /* nfs4_stateowner not found. -- cgit v1.2.3 From 375c5547cbf39423078535affac66c8afdc8fafb Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 18 Jan 2006 17:43:33 -0800 Subject: [PATCH] nfsd4: nfs4state.c miscellaneous goto removals Remove some goto's that made the logic here a little more tortuous than necessary. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4proc.c | 33 ++++++++++++++------------------- 1 file changed, 14 insertions(+), 19 deletions(-) diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 69ee182575a..89028f2d8d2 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -592,25 +592,21 @@ nfsd4_setattr(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_se status = nfs_ok; if (setattr->sa_iattr.ia_valid & ATTR_SIZE) { nfs4_lock_state(); - if ((status = nfs4_preprocess_stateid_op(current_fh, - &setattr->sa_stateid, - CHECK_FH | WR_STATE, NULL))) { - dprintk("NFSD: nfsd4_setattr: couldn't process stateid!\n"); - goto out_unlock; - } + status = nfs4_preprocess_stateid_op(current_fh, + &setattr->sa_stateid, CHECK_FH | WR_STATE, NULL); nfs4_unlock_state(); + if (status) { + dprintk("NFSD: nfsd4_setattr: couldn't process stateid!"); + return status; + } } status = nfs_ok; if (setattr->sa_acl != NULL) status = nfsd4_set_nfs4_acl(rqstp, current_fh, setattr->sa_acl); if (status) - goto out; + return status; status = nfsd_setattr(rqstp, current_fh, &setattr->sa_iattr, 0, (time_t)0); -out: - return status; -out_unlock: - nfs4_unlock_state(); return status; } @@ -628,15 +624,17 @@ nfsd4_write(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_writ return nfserr_inval; nfs4_lock_state(); - if ((status = nfs4_preprocess_stateid_op(current_fh, stateid, - CHECK_FH | WR_STATE, &filp))) { - dprintk("NFSD: nfsd4_write: couldn't process stateid!\n"); - goto out; - } + status = nfs4_preprocess_stateid_op(current_fh, stateid, + CHECK_FH | WR_STATE, &filp); if (filp) get_file(filp); nfs4_unlock_state(); + if (status) { + dprintk("NFSD: nfsd4_write: couldn't process stateid!\n"); + return status; + } + write->wr_bytes_written = write->wr_buflen; write->wr_how_written = write->wr_stable_how; p = (u32 *)write->wr_verifier.data; @@ -652,9 +650,6 @@ nfsd4_write(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_writ if (status == nfserr_symlink) status = nfserr_inval; return status; -out: - nfs4_unlock_state(); - return status; } /* This routine never returns NFS_OK! If there are no other errors, it -- cgit v1.2.3 From 0f442aa299bb47046db4077d9c122a1b539311ce Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 18 Jan 2006 17:43:34 -0800 Subject: [PATCH] nfsd4: simplify process-open1 logic nfsd4_process_open1 is very highly nested; flatten it out a bit. Also, the preceding comment, which just outlines the logic, seems redundant. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 99 +++++++++++++++++++---------------------------------- 1 file changed, 35 insertions(+), 64 deletions(-) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index dc792b6b451..b92dc9e0897 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1429,90 +1429,61 @@ static struct lock_manager_operations nfsd_lease_mng_ops = { }; -/* - * nfsd4_process_open1() - * lookup stateowner. - * found: - * check confirmed - * confirmed: - * check seqid - * not confirmed: - * delete owner - * create new owner - * notfound: - * verify clientid - * create new owner - * - * called with nfs4_lock_state() held. - */ int nfsd4_process_open1(struct nfsd4_open *open) { - int status; clientid_t *clientid = &open->op_clientid; struct nfs4_client *clp = NULL; unsigned int strhashval; struct nfs4_stateowner *sop = NULL; - status = nfserr_inval; if (!check_name(open->op_owner)) - goto out; + return nfserr_inval; if (STALE_CLIENTID(&open->op_clientid)) return nfserr_stale_clientid; strhashval = ownerstr_hashval(clientid->cl_id, open->op_owner); sop = find_openstateowner_str(strhashval, open); - if (sop) { - open->op_stateowner = sop; - if (!sop->so_confirmed) { - /* Replace any unconfirmed stateowner without - * even checking for replays */ - clp = sop->so_client; - release_stateowner(sop); - } else if (open->op_seqid == sop->so_seqid) { - /* normal case */ - goto renew; - } else if (open->op_seqid == sop->so_seqid - 1) { - /* replay */ - if (sop->so_replay.rp_buflen) - return NFSERR_REPLAY_ME; - else { - /* The original OPEN failed so spectacularly - * that we don't even have replay data saved! - * Therefore, we have no choice but to continue - * processing this OPEN; presumably, we'll - * fail again for the same reason. - */ - dprintk("nfsd4_process_open1:" - " replay with no replay cache\n"); - goto renew; - } - } else { - status = nfserr_bad_seqid; - goto out; - } - } else { - /* nfs4_stateowner not found. - * Verify clientid and instantiate new nfs4_stateowner. - * If verify fails this is presumably the result of the - * client's lease expiring. - */ - status = nfserr_expired; + open->op_stateowner = sop; + if (!sop) { + /* Make sure the client's lease hasn't expired. */ clp = find_confirmed_client(clientid); if (clp == NULL) - goto out; + return nfserr_expired; + goto renew; } - status = nfserr_resource; - sop = alloc_init_open_stateowner(strhashval, clp, open); - if (sop == NULL) - goto out; - open->op_stateowner = sop; + if (!sop->so_confirmed) { + /* Replace unconfirmed owners without checking for replay. */ + clp = sop->so_client; + release_stateowner(sop); + open->op_stateowner = NULL; + goto renew; + } + if (open->op_seqid == sop->so_seqid - 1) { + if (sop->so_replay.rp_buflen) + return NFSERR_REPLAY_ME; + /* The original OPEN failed so spectacularly + * that we don't even have replay data saved! + * Therefore, we have no choice but to continue + * processing this OPEN; presumably, we'll + * fail again for the same reason. + */ + dprintk("nfsd4_process_open1: replay with no replay cache\n"); + goto renew; + } + if (open->op_seqid != sop->so_seqid) + return nfserr_bad_seqid; renew: - status = nfs_ok; + if (open->op_stateowner == NULL) { + sop = alloc_init_open_stateowner(strhashval, clp, open); + if (sop == NULL) + return nfserr_resource; + open->op_stateowner = sop; + } + list_del_init(&sop->so_close_lru); renew_client(sop->so_client); -out: - return status; + return nfs_ok; } static inline int -- cgit v1.2.3 From fb553c0f17444e090db951b96df4d2d71b4f4b6b Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 18 Jan 2006 17:43:36 -0800 Subject: [PATCH] nfsd4: don't create on open that fails due to ERR_GRACE In an earlier patch (commit b648330a1d741d5df8a5076b2a0a2519c69c8f41) I noted that a too-early grace-period check was preventing us from bumping the sequence id on open. Unfortunately in that patch I stupidly moved the grace-period check back too far, so now an open for create can succesfully create the file while still returning ERR_GRACE. The correct place for that check is after we've set the open_owner and handled any replays, but before we actually start mucking with the filesystem. Thanks to Avishay Traeger for reporting the bug. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4proc.c | 8 ++++++++ fs/nfsd/nfs4state.c | 6 ------ 2 files changed, 8 insertions(+), 6 deletions(-) diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 89028f2d8d2..b3f169f400d 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -192,6 +192,14 @@ nfsd4_open(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open } if (status) goto out; + + /* Openowner is now set, so sequence id will get bumped. Now we need + * these checks before we do any creates: */ + if (nfs4_in_grace() && open->op_claim_type != NFS4_OPEN_CLAIM_PREVIOUS) + return nfserr_grace; + if (!nfs4_in_grace() && open->op_claim_type == NFS4_OPEN_CLAIM_PREVIOUS) + return nfserr_no_grace; + switch (open->op_claim_type) { case NFS4_OPEN_CLAIM_DELEGATE_CUR: status = nfserr_inval; diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index b92dc9e0897..7167dcf8e1f 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1732,12 +1732,6 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf struct nfs4_delegation *dp = NULL; int status; - if (nfs4_in_grace() && open->op_claim_type != NFS4_OPEN_CLAIM_PREVIOUS) - return nfserr_grace; - - if (!nfs4_in_grace() && open->op_claim_type == NFS4_OPEN_CLAIM_PREVIOUS) - return nfserr_no_grace; - status = nfserr_inval; if (!TEST_ACCESS(open->op_share_access) || !TEST_DENY(open->op_share_deny)) goto out; -- cgit v1.2.3 From 6c26d08f02f829a833d393c3f1b196538a9ec2c4 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 18 Jan 2006 17:43:38 -0800 Subject: [PATCH] nfsd4: fix open_downgrade Bad bookkeeping of the share reservations when handling open upgrades was causing open downgrade to fail. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 7167dcf8e1f..82c36ccd8b5 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1609,26 +1609,26 @@ nfs4_upgrade_open(struct svc_rqst *rqstp, struct svc_fh *cur_fh, struct nfs4_sta { struct file *filp = stp->st_vfs_file; struct inode *inode = filp->f_dentry->d_inode; - unsigned int share_access; + unsigned int share_access, new_writer; int status; set_access(&share_access, stp->st_access_bmap); - share_access = ~share_access; - share_access &= open->op_share_access; - - if (!(share_access & NFS4_SHARE_ACCESS_WRITE)) - return nfsd4_truncate(rqstp, cur_fh, open); + new_writer = (~share_access) & open->op_share_access + & NFS4_SHARE_ACCESS_WRITE; - status = get_write_access(inode); - if (status) - return nfserrno(status); + if (new_writer) { + status = get_write_access(inode); + if (status) + return nfserrno(status); + } status = nfsd4_truncate(rqstp, cur_fh, open); if (status) { - put_write_access(inode); + if (new_writer) + put_write_access(inode); return status; } /* remember the open */ - filp->f_mode = (filp->f_mode | FMODE_WRITE) & ~FMODE_READ; + filp->f_mode |= open->op_share_access; set_bit(open->op_share_access, &stp->st_access_bmap); set_bit(open->op_share_deny, &stp->st_deny_bmap); -- cgit v1.2.3 From 34081efc12aaaa12f20e5b59f3cb98ba6e27fb34 Mon Sep 17 00:00:00 2001 From: Fred Isaman Date: Wed, 18 Jan 2006 17:43:40 -0800 Subject: [PATCH] nfsd4: Fix bug in rdattr_error return Fix bug in rdattr_error return which causes correct error code to be overwritten by nfserr_toosmall. Signed-off-by: Fred Isaman Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4xdr.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 6b743327686..69d3501173a 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -1764,10 +1764,11 @@ nfsd4_encode_dirent(struct readdir_cd *ccd, const char *name, int namlen, */ if (!(cd->rd_bmval[0] & FATTR4_WORD0_RDATTR_ERROR)) goto fail; - nfserr = nfserr_toosmall; p = nfsd4_encode_rdattr_error(p, buflen, nfserr); - if (p == NULL) + if (p == NULL) { + nfserr = nfserr_toosmall; goto fail; + } } cd->buflen -= (p - cd->buffer); cd->buffer = p; -- cgit v1.2.3 From 5274881992b8a632620f69346401da66e480a23b Mon Sep 17 00:00:00 2001 From: Fred Isaman Date: Wed, 18 Jan 2006 17:43:43 -0800 Subject: [PATCH] nfsd4: clean up settattr code Clean up some unnecessary special-casing in the setattr code.. Signed-off-by: Fred Isaman Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4proc.c | 10 +--------- 1 file changed, 1 insertion(+), 9 deletions(-) diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index b3f169f400d..a00fe868629 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -594,10 +594,6 @@ nfsd4_setattr(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_se { int status = nfs_ok; - if (!current_fh->fh_dentry) - return nfserr_nofilehandle; - - status = nfs_ok; if (setattr->sa_iattr.ia_valid & ATTR_SIZE) { nfs4_lock_state(); status = nfs4_preprocess_stateid_op(current_fh, @@ -799,17 +795,13 @@ nfsd4_proc_compound(struct svc_rqst *rqstp, /* All operations except RENEW, SETCLIENTID, RESTOREFH * SETCLIENTID_CONFIRM, PUTFH and PUTROOTFH * require a valid current filehandle - * - * SETATTR NOFILEHANDLE error handled in nfsd4_setattr - * due to required returned bitmap argument */ if ((!current_fh->fh_dentry) && !((op->opnum == OP_PUTFH) || (op->opnum == OP_PUTROOTFH) || (op->opnum == OP_SETCLIENTID) || (op->opnum == OP_SETCLIENTID_CONFIRM) || (op->opnum == OP_RENEW) || (op->opnum == OP_RESTOREFH) || - (op->opnum == OP_RELEASE_LOCKOWNER) || - (op->opnum == OP_SETATTR))) { + (op->opnum == OP_RELEASE_LOCKOWNER))) { op->status = nfserr_nofilehandle; goto encode_op; } -- cgit v1.2.3 From d75f2b9f5da71667aba7e97a962c49f1e17aa4ca Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 18 Jan 2006 17:43:44 -0800 Subject: [PATCH] nfsd/vfs.c: endianness fixes Several failure exits return -E instead of nfserr_ and vice versa. Signed-off-by: Al Viro Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/vfs.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 6acec341676..e6dfa149a30 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1134,7 +1134,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, "nfsd_create: parent %s/%s not locked!\n", dentry->d_parent->d_name.name, dentry->d_name.name); - err = -EIO; + err = nfserr_io; goto out; } } @@ -1600,7 +1600,7 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, if ((ffhp->fh_export->ex_flags & NFSEXP_MSNFS) && ((atomic_read(&odentry->d_count) > 1) || (atomic_read(&ndentry->d_count) > 1))) { - err = nfserr_perm; + err = -EPERM; } else #endif err = vfs_rename(fdir, odentry, tdir, ndentry); @@ -1672,7 +1672,7 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, #ifdef MSNFS if ((fhp->fh_export->ex_flags & NFSEXP_MSNFS) && (atomic_read(&rdentry->d_count) > 1)) { - err = nfserr_perm; + err = -EPERM; } else #endif err = vfs_unlink(dirp, rdentry); -- cgit v1.2.3 From 9246585a117f182d26a5a5c15872c3e8fcf44dd6 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 18 Jan 2006 17:43:46 -0800 Subject: [PATCH] nfsd4_truncate() bogus return value -EINVAL (in host order, no less) is not a good thing to return to client. nfsd4_truncate() returns it in one case and its callers expect nfs_.... from it. AFAICS, it should be nfserr_inval Signed-off-by: Al Viro Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 82c36ccd8b5..ddee0e0371a 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1600,7 +1600,7 @@ nfsd4_truncate(struct svc_rqst *rqstp, struct svc_fh *fh, if (!open->op_truncate) return 0; if (!(open->op_share_access & NFS4_SHARE_ACCESS_WRITE)) - return -EINVAL; + return nfserr_inval; return nfsd_setattr(rqstp, fh, &iattr, 0, (time_t)0); } -- cgit v1.2.3 From de1ae286f863c46b7c8f9bed97df17d7f5ea510c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 18 Jan 2006 17:43:47 -0800 Subject: [PATCH] NFSERR_SERVERFAULT returned host-endian ->rp_status is network-endian and nobody byteswaps it before sending to client; putting NFSERR_SERVERFAULT instead of nfserr_serverfault in there is not nice... Signed-off-by: Al Viro Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index ddee0e0371a..856ed4fd662 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1088,7 +1088,7 @@ alloc_init_open_stateowner(unsigned int strhashval, struct nfs4_client *clp, str sop->so_seqid = open->op_seqid; sop->so_confirmed = 0; rp = &sop->so_replay; - rp->rp_status = NFSERR_SERVERFAULT; + rp->rp_status = nfserr_serverfault; rp->rp_buflen = 0; rp->rp_buf = rp->rp_ibuf; return sop; @@ -2583,7 +2583,7 @@ alloc_init_lock_stateowner(unsigned int strhashval, struct nfs4_client *clp, str sop->so_seqid = lock->lk_new_lock_seqid + 1; sop->so_confirmed = 1; rp = &sop->so_replay; - rp->rp_status = NFSERR_SERVERFAULT; + rp->rp_status = nfserr_serverfault; rp->rp_buflen = 0; rp->rp_buf = rp->rp_ibuf; return sop; -- cgit v1.2.3 From 7fcd53303de8dbbed863f6471ca92eb96a1faa28 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 18 Jan 2006 17:43:48 -0800 Subject: [PATCH] nfsd4_lock() returns bogus values to clients missing nfserrno() in default case of a switch by return value of posix_lock_file(); as the result we send negative host-endian to clients that expect positive network-endian, preferably mentioned in RFC... BTW, that case is not impossible - posix_lock_file() can return -ENOLCK and we do not handle that one explicitly. Signed-off-by: Al Viro Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4state.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 856ed4fd662..1143cfb6454 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -2761,7 +2761,10 @@ nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock goto conflicting_lock; case (EDEADLK): status = nfserr_deadlock; + dprintk("NFSD: nfsd4_lock: posix_lock_file() failed! status %d\n",status); + goto out; default: + status = nfserrno(status); dprintk("NFSD: nfsd4_lock: posix_lock_file() failed! status %d\n",status); goto out; } -- cgit v1.2.3 From 45bd3b3dffaa9fb1706fe001bf66276d0997a850 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Wed, 18 Jan 2006 17:43:50 -0800 Subject: [PATCH] knfsd: Fix some more errno/nfserr confusion in vfs.c nfsd_sync* return an errno, which usually needs to be converted to an errno, sometimes immediately, sometimes a little later. Also, nfsd_setattr returns an nfserr which SHOULDN'T be converted from an errno (because it isn't one). Also some tidyups of the form: err = XX err = nfserrno(err) and err = XX if (err) err = nfserrno(err) become err = nfserrno(XX) Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/vfs.c | 17 ++++++----------- 1 file changed, 6 insertions(+), 11 deletions(-) diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index e6dfa149a30..5320e5afadd 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -891,9 +891,9 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, int err = 0; int stable = *stablep; +#ifdef MSNFS err = nfserr_perm; -#ifdef MSNFS if ((fhp->fh_export->ex_flags & NFSEXP_MSNFS) && (!lock_may_write(file->f_dentry->d_inode, offset, cnt))) goto out; @@ -1065,8 +1065,7 @@ nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, return err; if (EX_ISSYNC(fhp->fh_export)) { if (file->f_op && file->f_op->fsync) { - err = nfsd_sync(file); - err = nfserrno(err); + err = nfserrno(nfsd_sync(file)); } else { err = nfserr_notsupp; } @@ -1177,7 +1176,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, goto out_nfserr; if (EX_ISSYNC(fhp->fh_export)) { - err = nfsd_sync_dir(dentry); + err = nfserrno(nfsd_sync_dir(dentry)); write_inode_now(dchild->d_inode, 1); } @@ -1310,9 +1309,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, goto out_nfserr; if (EX_ISSYNC(fhp->fh_export)) { - err = nfsd_sync_dir(dentry); - if (err) - err = nfserrno(err); + err = nfserrno(nfsd_sync_dir(dentry)); /* setattr will sync the child (or not) */ } @@ -1339,7 +1336,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, if ((iap->ia_valid &= ~(ATTR_UID|ATTR_GID)) != 0) { int err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); if (err2) - err = nfserrno(err2); + err = err2; } /* @@ -1514,10 +1511,8 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp, err = vfs_link(dold, dirp, dnew); if (!err) { if (EX_ISSYNC(ffhp->fh_export)) { - err = nfsd_sync_dir(ddir); + err = nfserrno(nfsd_sync_dir(ddir)); write_inode_now(dest, 1); - if (err) - err = nfserrno(err); } } else { if (err == -EXDEV && rqstp->rq_vers == 2) -- cgit v1.2.3 From 846f2fcd77850ef8f0aab46df9cadd5c35a5fef0 Mon Sep 17 00:00:00 2001 From: David Shaw Date: Wed, 18 Jan 2006 17:43:51 -0800 Subject: [PATCH] knfsd: Provide missing NFSv2 part of patch for checking vfs_getattr. A recent patch which checked the return status of vfs_getattr in nfsd, completely missed the nfsproc.c (NFSv2) part. Here is it. This patch moved the call to vfs_getattr from the xdr encoding (at which point it is too late to return an error) to the call handling. This means several calls to vfs_getattr are needed in nfsproc.c. Many are encapsulated in nfsd_return_attrs and nfsd_return_dirop. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfsproc.c | 37 ++++++++++++++++++++++++++++++------- 1 file changed, 30 insertions(+), 7 deletions(-) diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c index 0aa1b9603d7..3e6b75cd90f 100644 --- a/fs/nfsd/nfsproc.c +++ b/fs/nfsd/nfsproc.c @@ -36,6 +36,22 @@ nfsd_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) return nfs_ok; } +static int +nfsd_return_attrs(int err, struct nfsd_attrstat *resp) +{ + if (err) return err; + return nfserrno(vfs_getattr(resp->fh.fh_export->ex_mnt, + resp->fh.fh_dentry, + &resp->stat)); +} +static int +nfsd_return_dirop(int err, struct nfsd_diropres *resp) +{ + if (err) return err; + return nfserrno(vfs_getattr(resp->fh.fh_export->ex_mnt, + resp->fh.fh_dentry, + &resp->stat)); +} /* * Get a file's attributes * N.B. After this call resp->fh needs an fh_put @@ -44,10 +60,12 @@ static int nfsd_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle *argp, struct nfsd_attrstat *resp) { + int nfserr; dprintk("nfsd: GETATTR %s\n", SVCFH_fmt(&argp->fh)); fh_copy(&resp->fh, &argp->fh); - return fh_verify(rqstp, &resp->fh, 0, MAY_NOP); + nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP); + return nfsd_return_attrs(nfserr, resp); } /* @@ -58,12 +76,14 @@ static int nfsd_proc_setattr(struct svc_rqst *rqstp, struct nfsd_sattrargs *argp, struct nfsd_attrstat *resp) { + int nfserr; dprintk("nfsd: SETATTR %s, valid=%x, size=%ld\n", SVCFH_fmt(&argp->fh), argp->attrs.ia_valid, (long) argp->attrs.ia_size); fh_copy(&resp->fh, &argp->fh); - return nfsd_setattr(rqstp, &resp->fh, &argp->attrs,0, (time_t)0); + nfserr = nfsd_setattr(rqstp, &resp->fh, &argp->attrs,0, (time_t)0); + return nfsd_return_attrs(nfserr, resp); } /* @@ -86,7 +106,7 @@ nfsd_proc_lookup(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, &resp->fh); fh_put(&argp->fh); - return nfserr; + return nfsd_return_dirop(nfserr, resp); } /* @@ -142,7 +162,10 @@ nfsd_proc_read(struct svc_rqst *rqstp, struct nfsd_readargs *argp, argp->vec, argp->vlen, &resp->count); - return nfserr; + if (nfserr) return nfserr; + return nfserrno(vfs_getattr(resp->fh.fh_export->ex_mnt, + resp->fh.fh_dentry, + &resp->stat)); } /* @@ -165,7 +188,7 @@ nfsd_proc_write(struct svc_rqst *rqstp, struct nfsd_writeargs *argp, argp->vec, argp->vlen, argp->len, &stable); - return nfserr; + return nfsd_return_attrs(nfserr, resp); } /* @@ -322,7 +345,7 @@ out_unlock: done: fh_put(dirfhp); - return nfserr; + return nfsd_return_dirop(nfserr, resp); } static int @@ -425,7 +448,7 @@ nfsd_proc_mkdir(struct svc_rqst *rqstp, struct nfsd_createargs *argp, nfserr = nfsd_create(rqstp, &argp->fh, argp->name, argp->len, &argp->attrs, S_IFDIR, 0, &resp->fh); fh_put(&argp->fh); - return nfserr; + return nfsd_return_dirop(nfserr, resp); } /* -- cgit v1.2.3 From e2f99018eb7b29954747a2dd78e9fc0c36a60f0f Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 18 Jan 2006 17:43:52 -0800 Subject: [PATCH] exportfs: add find_acceptable_alias helper find_exported_dentry contains two duplicate loops to find an alias that the acceptable callback likes. Split this out to a new helper and switch from list_for_each to list_for_each_entry to make it more readable. Signed-off-by: Christoph Hellwig Acked-by: NeilBrown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exportfs/expfs.c | 79 ++++++++++++++++++++++++----------------------------- 1 file changed, 36 insertions(+), 43 deletions(-) diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c index 5bfe40085fb..b06b54f1bbb 100644 --- a/fs/exportfs/expfs.c +++ b/fs/exportfs/expfs.c @@ -11,6 +11,33 @@ struct export_operations export_op_default; #define dprintk(fmt, args...) do{}while(0) +static struct dentry * +find_acceptable_alias(struct dentry *result, + int (*acceptable)(void *context, struct dentry *dentry), + void *context) +{ + struct dentry *dentry, *toput = NULL; + + spin_lock(&dcache_lock); + list_for_each_entry(dentry, &result->d_inode->i_dentry, d_alias) { + dget_locked(dentry); + spin_unlock(&dcache_lock); + if (toput) + dput(toput); + if (dentry != result && acceptable(context, dentry)) { + dput(result); + return dentry; + } + spin_lock(&dcache_lock); + toput = dentry; + } + spin_unlock(&dcache_lock); + + if (toput) + dput(toput); + return NULL; +} + /** * find_exported_dentry - helper routine to implement export_operations->decode_fh * @sb: The &super_block identifying the filesystem @@ -52,8 +79,7 @@ find_exported_dentry(struct super_block *sb, void *obj, void *parent, struct dentry *target_dir; int err; struct export_operations *nops = sb->s_export_op; - struct list_head *le, *head; - struct dentry *toput = NULL; + struct dentry *alias; int noprogress; char nbuf[NAME_MAX+1]; @@ -79,27 +105,10 @@ find_exported_dentry(struct super_block *sb, void *obj, void *parent, /* there is no other dentry, so fail */ goto err_result; } - /* try any other aliases */ - spin_lock(&dcache_lock); - head = &result->d_inode->i_dentry; - list_for_each(le, head) { - struct dentry *dentry = list_entry(le, struct dentry, d_alias); - dget_locked(dentry); - spin_unlock(&dcache_lock); - if (toput) - dput(toput); - toput = NULL; - if (dentry != result && - acceptable(context, dentry)) { - dput(result); - return dentry; - } - spin_lock(&dcache_lock); - toput = dentry; - } - spin_unlock(&dcache_lock); - if (toput) - dput(toput); + + alias = find_acceptable_alias(result, acceptable, context); + if (alias) + return alias; } /* It's a directory, or we are required to confirm the file's @@ -258,26 +267,10 @@ find_exported_dentry(struct super_block *sb, void *obj, void *parent, /* now result is properly connected, it is our best bet */ if (acceptable(context, result)) return result; - /* one last try of the aliases.. */ - spin_lock(&dcache_lock); - toput = NULL; - head = &result->d_inode->i_dentry; - list_for_each(le, head) { - struct dentry *dentry = list_entry(le, struct dentry, d_alias); - dget_locked(dentry); - spin_unlock(&dcache_lock); - if (toput) dput(toput); - if (dentry != result && - acceptable(context, dentry)) { - dput(result); - return dentry; - } - spin_lock(&dcache_lock); - toput = dentry; - } - spin_unlock(&dcache_lock); - if (toput) - dput(toput); + + alias = find_acceptable_alias(result, acceptable, context); + if (alias) + return alias; /* drat - I just cannot find anything acceptable */ dput(result); -- cgit v1.2.3 From 5590ff0d5528b60153c0b4e7b771472b5a95e297 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 18 Jan 2006 17:43:53 -0800 Subject: [PATCH] vfs: *at functions: core Here is a series of patches which introduce in total 13 new system calls which take a file descriptor/filename pair instead of a single file name. These functions, openat etc, have been discussed on numerous occasions. They are needed to implement race-free filesystem traversal, they are necessary to implement a virtual per-thread current working directory (think multi-threaded backup software), etc. We have in glibc today implementations of the interfaces which use the /proc/self/fd magic. But this code is rather expensive. Here are some results (similar to what Jim Meyering posted before). The test creates a deep directory hierarchy on a tmpfs filesystem. Then rm -fr is used to remove all directories. Without syscall support I get this: real 0m31.921s user 0m0.688s sys 0m31.234s With syscall support the results are much better: real 0m20.699s user 0m0.536s sys 0m20.149s The interfaces are for obvious reasons currently not much used. But they'll be used. coreutils (and Jeff's posixutils) are already using them. Furthermore, code like ftw/fts in libc (maybe even glob) will also start using them. I expect a patch to make follow soon. Every program which is walking the filesystem tree will benefit. Signed-off-by: Ulrich Drepper Signed-off-by: Alexey Dobriyan Cc: Christoph Hellwig Cc: Al Viro Acked-by: Ingo Molnar Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/kernel/osf_sys.c | 2 +- arch/sparc64/kernel/sys_sparc32.c | 4 +- fs/compat.c | 48 +++++++++-- fs/exec.c | 2 +- fs/namei.c | 172 ++++++++++++++++++++++++++++++-------- fs/open.c | 83 ++++++++++++++---- fs/stat.c | 66 ++++++++++++--- include/linux/fcntl.h | 7 ++ include/linux/fs.h | 7 +- include/linux/namei.h | 7 +- include/linux/time.h | 2 +- 11 files changed, 320 insertions(+), 80 deletions(-) diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index 01fe990d3e5..7fb14f42a12 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -960,7 +960,7 @@ osf_utimes(char __user *filename, struct timeval32 __user *tvs) return -EFAULT; } - return do_utimes(filename, tvs ? ktvs : NULL); + return do_utimes(AT_FDCWD, filename, tvs ? ktvs : NULL); } #define MAX_SELECT_SECONDS \ diff --git a/arch/sparc64/kernel/sys_sparc32.c b/arch/sparc64/kernel/sys_sparc32.c index d4b7a100cb8..9264ccbaaaf 100644 --- a/arch/sparc64/kernel/sys_sparc32.c +++ b/arch/sparc64/kernel/sys_sparc32.c @@ -821,7 +821,7 @@ asmlinkage long sys32_utimes(char __user *filename, return -EFAULT; } - return do_utimes(filename, (tvs ? &ktvs[0] : NULL)); + return do_utimes(AT_FDCWD, filename, (tvs ? &ktvs[0] : NULL)); } /* These are here just in case some old sparc32 binary calls it. */ @@ -1003,7 +1003,7 @@ asmlinkage long sys32_adjtimex(struct timex32 __user *utp) asmlinkage long sparc32_open(const char __user *filename, int flags, int mode) { - return do_sys_open(filename, flags, mode); + return do_sys_open(AT_FDCWD, filename, flags, mode); } extern unsigned long do_mremap(unsigned long addr, diff --git a/fs/compat.c b/fs/compat.c index 2468ac1df2f..c6ba9deabad 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -68,10 +68,10 @@ asmlinkage long compat_sys_utime(char __user *filename, struct compat_utimbuf __ tv[0].tv_usec = 0; tv[1].tv_usec = 0; } - return do_utimes(filename, t ? tv : NULL); + return do_utimes(AT_FDCWD, filename, t ? tv : NULL); } -asmlinkage long compat_sys_utimes(char __user *filename, struct compat_timeval __user *t) +asmlinkage long compat_sys_futimesat(int dfd, char __user *filename, struct compat_timeval __user *t) { struct timeval tv[2]; @@ -82,14 +82,19 @@ asmlinkage long compat_sys_utimes(char __user *filename, struct compat_timeval _ get_user(tv[1].tv_usec, &t[1].tv_usec)) return -EFAULT; } - return do_utimes(filename, t ? tv : NULL); + return do_utimes(dfd, filename, t ? tv : NULL); +} + +asmlinkage long compat_sys_utimes(char __user *filename, struct compat_timeval __user *t) +{ + return compat_sys_futimesat(AT_FDCWD, filename, t); } asmlinkage long compat_sys_newstat(char __user * filename, struct compat_stat __user *statbuf) { struct kstat stat; - int error = vfs_stat(filename, &stat); + int error = vfs_stat_fd(AT_FDCWD, filename, &stat); if (!error) error = cp_compat_stat(&stat, statbuf); @@ -100,13 +105,34 @@ asmlinkage long compat_sys_newlstat(char __user * filename, struct compat_stat __user *statbuf) { struct kstat stat; - int error = vfs_lstat(filename, &stat); + int error = vfs_lstat_fd(AT_FDCWD, filename, &stat); if (!error) error = cp_compat_stat(&stat, statbuf); return error; } +asmlinkage long compat_sys_newfstatat(int dfd, char __user *filename, + struct compat_stat __user *statbuf, int flag) +{ + struct kstat stat; + int error = -EINVAL; + + if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) + goto out; + + if (flag & AT_SYMLINK_NOFOLLOW) + error = vfs_lstat_fd(dfd, filename, &stat); + else + error = vfs_stat_fd(dfd, filename, &stat); + + if (!error) + error = cp_compat_stat(&stat, statbuf); + +out: + return error; +} + asmlinkage long compat_sys_newfstat(unsigned int fd, struct compat_stat __user * statbuf) { @@ -1290,7 +1316,17 @@ out: asmlinkage long compat_sys_open(const char __user *filename, int flags, int mode) { - return do_sys_open(filename, flags, mode); + return do_sys_open(AT_FDCWD, filename, flags, mode); +} + +/* + * Exactly like fs/open.c:sys_openat(), except that it doesn't set the + * O_LARGEFILE flag. + */ +asmlinkage long +compat_sys_openat(int dfd, const char __user *filename, int flags, int mode) +{ + return do_sys_open(dfd, filename, flags, mode); } /* diff --git a/fs/exec.c b/fs/exec.c index 62b40af68cc..055378d2513 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -477,7 +477,7 @@ struct file *open_exec(const char *name) int err; struct file *file; - err = path_lookup_open(name, LOOKUP_FOLLOW, &nd, FMODE_READ); + err = path_lookup_open(AT_FDCWD, name, LOOKUP_FOLLOW, &nd, FMODE_READ); file = ERR_PTR(err); if (!err) { diff --git a/fs/namei.c b/fs/namei.c index 33fb5bd34a8..4acdac043b6 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -30,6 +30,8 @@ #include #include #include +#include +#include #include #include @@ -1063,7 +1065,8 @@ set_it: } /* Returns 0 and nd will be valid on success; Retuns error, otherwise. */ -int fastcall path_lookup(const char *name, unsigned int flags, struct nameidata *nd) +static int fastcall do_path_lookup(int dfd, const char *name, + unsigned int flags, struct nameidata *nd) { int retval = 0; @@ -1083,9 +1086,38 @@ int fastcall path_lookup(const char *name, unsigned int flags, struct nameidata } nd->mnt = mntget(current->fs->rootmnt); nd->dentry = dget(current->fs->root); - } else { + } else if (dfd == AT_FDCWD) { nd->mnt = mntget(current->fs->pwdmnt); nd->dentry = dget(current->fs->pwd); + } else { + struct file *file; + int fput_needed; + struct dentry *dentry; + + file = fget_light(dfd, &fput_needed); + if (!file) { + retval = -EBADF; + goto out_fail; + } + + dentry = file->f_dentry; + + if (!S_ISDIR(dentry->d_inode->i_mode)) { + retval = -ENOTDIR; + fput_light(file, fput_needed); + goto out_fail; + } + + retval = file_permission(file, MAY_EXEC); + if (retval) { + fput_light(file, fput_needed); + goto out_fail; + } + + nd->mnt = mntget(file->f_vfsmnt); + nd->dentry = dget(dentry); + + fput_light(file, fput_needed); } read_unlock(¤t->fs->lock); current->total_link_count = 0; @@ -1094,11 +1126,19 @@ out: if (unlikely(current->audit_context && nd && nd->dentry && nd->dentry->d_inode)) audit_inode(name, nd->dentry->d_inode, flags); +out_fail: return retval; } -static int __path_lookup_intent_open(const char *name, unsigned int lookup_flags, - struct nameidata *nd, int open_flags, int create_mode) +int fastcall path_lookup(const char *name, unsigned int flags, + struct nameidata *nd) +{ + return do_path_lookup(AT_FDCWD, name, flags, nd); +} + +static int __path_lookup_intent_open(int dfd, const char *name, + unsigned int lookup_flags, struct nameidata *nd, + int open_flags, int create_mode) { struct file *filp = get_empty_filp(); int err; @@ -1108,7 +1148,7 @@ static int __path_lookup_intent_open(const char *name, unsigned int lookup_flags nd->intent.open.file = filp; nd->intent.open.flags = open_flags; nd->intent.open.create_mode = create_mode; - err = path_lookup(name, lookup_flags|LOOKUP_OPEN, nd); + err = do_path_lookup(dfd, name, lookup_flags|LOOKUP_OPEN, nd); if (IS_ERR(nd->intent.open.file)) { if (err == 0) { err = PTR_ERR(nd->intent.open.file); @@ -1126,10 +1166,10 @@ static int __path_lookup_intent_open(const char *name, unsigned int lookup_flags * @nd: pointer to nameidata * @open_flags: open intent flags */ -int path_lookup_open(const char *name, unsigned int lookup_flags, +int path_lookup_open(int dfd, const char *name, unsigned int lookup_flags, struct nameidata *nd, int open_flags) { - return __path_lookup_intent_open(name, lookup_flags, nd, + return __path_lookup_intent_open(dfd, name, lookup_flags, nd, open_flags, 0); } @@ -1141,12 +1181,12 @@ int path_lookup_open(const char *name, unsigned int lookup_flags, * @open_flags: open intent flags * @create_mode: create intent flags */ -static int path_lookup_create(const char *name, unsigned int lookup_flags, - struct nameidata *nd, int open_flags, - int create_mode) +static int path_lookup_create(int dfd, const char *name, + unsigned int lookup_flags, struct nameidata *nd, + int open_flags, int create_mode) { - return __path_lookup_intent_open(name, lookup_flags|LOOKUP_CREATE, nd, - open_flags, create_mode); + return __path_lookup_intent_open(dfd, name, lookup_flags|LOOKUP_CREATE, + nd, open_flags, create_mode); } int __user_path_lookup_open(const char __user *name, unsigned int lookup_flags, @@ -1156,7 +1196,7 @@ int __user_path_lookup_open(const char __user *name, unsigned int lookup_flags, int err = PTR_ERR(tmp); if (!IS_ERR(tmp)) { - err = __path_lookup_intent_open(tmp, lookup_flags, nd, open_flags, 0); + err = __path_lookup_intent_open(AT_FDCWD, tmp, lookup_flags, nd, open_flags, 0); putname(tmp); } return err; @@ -1248,18 +1288,24 @@ access: * that namei follows links, while lnamei does not. * SMP-safe */ -int fastcall __user_walk(const char __user *name, unsigned flags, struct nameidata *nd) +int fastcall __user_walk_fd(int dfd, const char __user *name, unsigned flags, + struct nameidata *nd) { char *tmp = getname(name); int err = PTR_ERR(tmp); if (!IS_ERR(tmp)) { - err = path_lookup(tmp, flags, nd); + err = do_path_lookup(dfd, tmp, flags, nd); putname(tmp); } return err; } +int fastcall __user_walk(const char __user *name, unsigned flags, struct nameidata *nd) +{ + return __user_walk_fd(AT_FDCWD, name, flags, nd); +} + /* * It's inline, so penalty for filesystems that don't use sticky bit is * minimal. @@ -1518,7 +1564,8 @@ int may_open(struct nameidata *nd, int acc_mode, int flag) * for symlinks (where the permissions are checked later). * SMP-safe */ -int open_namei(const char * pathname, int flag, int mode, struct nameidata *nd) +int open_namei(int dfd, const char *pathname, int flag, + int mode, struct nameidata *nd) { int acc_mode, error; struct path path; @@ -1540,7 +1587,8 @@ int open_namei(const char * pathname, int flag, int mode, struct nameidata *nd) * The simplest case - just a plain lookup. */ if (!(flag & O_CREAT)) { - error = path_lookup_open(pathname, lookup_flags(flag), nd, flag); + error = path_lookup_open(dfd, pathname, lookup_flags(flag), + nd, flag); if (error) return error; goto ok; @@ -1549,7 +1597,7 @@ int open_namei(const char * pathname, int flag, int mode, struct nameidata *nd) /* * Create - we need to know the parent. */ - error = path_lookup_create(pathname, LOOKUP_PARENT, nd, flag, mode); + error = path_lookup_create(dfd,pathname,LOOKUP_PARENT,nd,flag,mode); if (error) return error; @@ -1744,7 +1792,8 @@ int vfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) return error; } -asmlinkage long sys_mknod(const char __user * filename, int mode, unsigned dev) +asmlinkage long sys_mknodat(int dfd, const char __user *filename, int mode, + unsigned dev) { int error = 0; char * tmp; @@ -1757,7 +1806,7 @@ asmlinkage long sys_mknod(const char __user * filename, int mode, unsigned dev) if (IS_ERR(tmp)) return PTR_ERR(tmp); - error = path_lookup(tmp, LOOKUP_PARENT, &nd); + error = do_path_lookup(dfd, tmp, LOOKUP_PARENT, &nd); if (error) goto out; dentry = lookup_create(&nd, 0); @@ -1793,6 +1842,11 @@ out: return error; } +asmlinkage long sys_mknod(const char __user *filename, int mode, unsigned dev) +{ + return sys_mknodat(AT_FDCWD, filename, mode, dev); +} + int vfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) { int error = may_create(dir, dentry, NULL); @@ -1815,7 +1869,7 @@ int vfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) return error; } -asmlinkage long sys_mkdir(const char __user * pathname, int mode) +asmlinkage long sys_mkdirat(int dfd, const char __user *pathname, int mode) { int error = 0; char * tmp; @@ -1826,7 +1880,7 @@ asmlinkage long sys_mkdir(const char __user * pathname, int mode) struct dentry *dentry; struct nameidata nd; - error = path_lookup(tmp, LOOKUP_PARENT, &nd); + error = do_path_lookup(dfd, tmp, LOOKUP_PARENT, &nd); if (error) goto out; dentry = lookup_create(&nd, 1); @@ -1846,6 +1900,11 @@ out: return error; } +asmlinkage long sys_mkdir(const char __user *pathname, int mode) +{ + return sys_mkdirat(AT_FDCWD, pathname, mode); +} + /* * We try to drop the dentry early: we should have * a usage count of 2 if we're the only user of this @@ -1907,7 +1966,7 @@ int vfs_rmdir(struct inode *dir, struct dentry *dentry) return error; } -asmlinkage long sys_rmdir(const char __user * pathname) +static long do_rmdir(int dfd, const char __user *pathname) { int error = 0; char * name; @@ -1918,7 +1977,7 @@ asmlinkage long sys_rmdir(const char __user * pathname) if(IS_ERR(name)) return PTR_ERR(name); - error = path_lookup(name, LOOKUP_PARENT, &nd); + error = do_path_lookup(dfd, name, LOOKUP_PARENT, &nd); if (error) goto exit; @@ -1948,6 +2007,11 @@ exit: return error; } +asmlinkage long sys_rmdir(const char __user *pathname) +{ + return do_rmdir(AT_FDCWD, pathname); +} + int vfs_unlink(struct inode *dir, struct dentry *dentry) { int error = may_delete(dir, dentry, 0); @@ -1984,7 +2048,7 @@ int vfs_unlink(struct inode *dir, struct dentry *dentry) * writeout happening, and we don't want to prevent access to the directory * while waiting on the I/O. */ -asmlinkage long sys_unlink(const char __user * pathname) +static long do_unlinkat(int dfd, const char __user *pathname) { int error = 0; char * name; @@ -1996,7 +2060,7 @@ asmlinkage long sys_unlink(const char __user * pathname) if(IS_ERR(name)) return PTR_ERR(name); - error = path_lookup(name, LOOKUP_PARENT, &nd); + error = do_path_lookup(dfd, name, LOOKUP_PARENT, &nd); if (error) goto exit; error = -EISDIR; @@ -2031,6 +2095,22 @@ slashes: goto exit2; } +asmlinkage long sys_unlinkat(int dfd, const char __user *pathname, int flag) +{ + if ((flag & ~AT_REMOVEDIR) != 0) + return -EINVAL; + + if (flag & AT_REMOVEDIR) + return do_rmdir(dfd, pathname); + + return do_unlinkat(dfd, pathname); +} + +asmlinkage long sys_unlink(const char __user *pathname) +{ + return do_unlinkat(AT_FDCWD, pathname); +} + int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname, int mode) { int error = may_create(dir, dentry, NULL); @@ -2052,7 +2132,8 @@ int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname, i return error; } -asmlinkage long sys_symlink(const char __user * oldname, const char __user * newname) +asmlinkage long sys_symlinkat(const char __user *oldname, + int newdfd, const char __user *newname) { int error = 0; char * from; @@ -2067,7 +2148,7 @@ asmlinkage long sys_symlink(const char __user * oldname, const char __user * new struct dentry *dentry; struct nameidata nd; - error = path_lookup(to, LOOKUP_PARENT, &nd); + error = do_path_lookup(newdfd, to, LOOKUP_PARENT, &nd); if (error) goto out; dentry = lookup_create(&nd, 0); @@ -2085,6 +2166,11 @@ out: return error; } +asmlinkage long sys_symlink(const char __user *oldname, const char __user *newname) +{ + return sys_symlinkat(oldname, AT_FDCWD, newname); +} + int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_dentry) { struct inode *inode = old_dentry->d_inode; @@ -2132,7 +2218,8 @@ int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_de * with linux 2.0, and to avoid hard-linking to directories * and other special files. --ADM */ -asmlinkage long sys_link(const char __user * oldname, const char __user * newname) +asmlinkage long sys_linkat(int olddfd, const char __user *oldname, + int newdfd, const char __user *newname) { struct dentry *new_dentry; struct nameidata nd, old_nd; @@ -2143,10 +2230,10 @@ asmlinkage long sys_link(const char __user * oldname, const char __user * newnam if (IS_ERR(to)) return PTR_ERR(to); - error = __user_walk(oldname, 0, &old_nd); + error = __user_walk_fd(olddfd, oldname, 0, &old_nd); if (error) goto exit; - error = path_lookup(to, LOOKUP_PARENT, &nd); + error = do_path_lookup(newdfd, to, LOOKUP_PARENT, &nd); if (error) goto out; error = -EXDEV; @@ -2169,6 +2256,11 @@ exit: return error; } +asmlinkage long sys_link(const char __user *oldname, const char __user *newname) +{ + return sys_linkat(AT_FDCWD, oldname, AT_FDCWD, newname); +} + /* * The worst of all namespace operations - renaming directory. "Perverted" * doesn't even start to describe it. Somebody in UCB had a heck of a trip... @@ -2315,7 +2407,8 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry, return error; } -static int do_rename(const char * oldname, const char * newname) +static int do_rename(int olddfd, const char *oldname, + int newdfd, const char *newname) { int error = 0; struct dentry * old_dir, * new_dir; @@ -2323,11 +2416,11 @@ static int do_rename(const char * oldname, const char * newname) struct dentry * trap; struct nameidata oldnd, newnd; - error = path_lookup(oldname, LOOKUP_PARENT, &oldnd); + error = do_path_lookup(olddfd, oldname, LOOKUP_PARENT, &oldnd); if (error) goto exit; - error = path_lookup(newname, LOOKUP_PARENT, &newnd); + error = do_path_lookup(newdfd, newname, LOOKUP_PARENT, &newnd); if (error) goto exit1; @@ -2391,7 +2484,8 @@ exit: return error; } -asmlinkage long sys_rename(const char __user * oldname, const char __user * newname) +asmlinkage long sys_renameat(int olddfd, const char __user *oldname, + int newdfd, const char __user *newname) { int error; char * from; @@ -2403,13 +2497,18 @@ asmlinkage long sys_rename(const char __user * oldname, const char __user * newn to = getname(newname); error = PTR_ERR(to); if (!IS_ERR(to)) { - error = do_rename(from,to); + error = do_rename(olddfd, from, newdfd, to); putname(to); } putname(from); return error; } +asmlinkage long sys_rename(const char __user *oldname, const char __user *newname) +{ + return sys_renameat(AT_FDCWD, oldname, AT_FDCWD, newname); +} + int vfs_readlink(struct dentry *dentry, char __user *buffer, int buflen, const char *link) { int len; @@ -2553,6 +2652,7 @@ struct inode_operations page_symlink_inode_operations = { }; EXPORT_SYMBOL(__user_walk); +EXPORT_SYMBOL(__user_walk_fd); EXPORT_SYMBOL(follow_down); EXPORT_SYMBOL(follow_up); EXPORT_SYMBOL(get_write_access); /* binfmt_aout */ diff --git a/fs/open.c b/fs/open.c index 8e20c1f3256..70e0230d8e7 100644 --- a/fs/open.c +++ b/fs/open.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include #include @@ -383,7 +384,7 @@ asmlinkage long sys_utime(char __user * filename, struct utimbuf __user * times) error = get_user(newattrs.ia_atime.tv_sec, ×->actime); newattrs.ia_atime.tv_nsec = 0; - if (!error) + if (!error) error = get_user(newattrs.ia_mtime.tv_sec, ×->modtime); newattrs.ia_mtime.tv_nsec = 0; if (error) @@ -414,14 +415,14 @@ out: * must be owner or have write permission. * Else, update from *times, must be owner or super user. */ -long do_utimes(char __user * filename, struct timeval * times) +long do_utimes(int dfd, char __user *filename, struct timeval *times) { int error; struct nameidata nd; struct inode * inode; struct iattr newattrs; - error = user_path_walk(filename, &nd); + error = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW, &nd); if (error) goto out; @@ -461,13 +462,18 @@ out: return error; } -asmlinkage long sys_utimes(char __user * filename, struct timeval __user * utimes) +asmlinkage long sys_futimesat(int dfd, char __user *filename, struct timeval __user *utimes) { struct timeval times[2]; if (utimes && copy_from_user(×, utimes, sizeof(times))) return -EFAULT; - return do_utimes(filename, utimes ? times : NULL); + return do_utimes(dfd, filename, utimes ? times : NULL); +} + +asmlinkage long sys_utimes(char __user *filename, struct timeval __user *utimes) +{ + return sys_futimesat(AT_FDCWD, filename, utimes); } @@ -476,7 +482,7 @@ asmlinkage long sys_utimes(char __user * filename, struct timeval __user * utime * We do this by temporarily clearing all FS-related capabilities and * switching the fsuid/fsgid around to the real ones. */ -asmlinkage long sys_access(const char __user * filename, int mode) +asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) { struct nameidata nd; int old_fsuid, old_fsgid; @@ -506,7 +512,7 @@ asmlinkage long sys_access(const char __user * filename, int mode) else current->cap_effective = current->cap_permitted; - res = __user_walk(filename, LOOKUP_FOLLOW|LOOKUP_ACCESS, &nd); + res = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW|LOOKUP_ACCESS, &nd); if (!res) { res = vfs_permission(&nd, mode); /* SuS v2 requires we report a read only fs too */ @@ -523,6 +529,11 @@ asmlinkage long sys_access(const char __user * filename, int mode) return res; } +asmlinkage long sys_access(const char __user *filename, int mode) +{ + return sys_faccessat(AT_FDCWD, filename, mode); +} + asmlinkage long sys_chdir(const char __user * filename) { struct nameidata nd; @@ -635,14 +646,15 @@ out: return err; } -asmlinkage long sys_chmod(const char __user * filename, mode_t mode) +asmlinkage long sys_fchmodat(int dfd, const char __user *filename, + mode_t mode) { struct nameidata nd; struct inode * inode; int error; struct iattr newattrs; - error = user_path_walk(filename, &nd); + error = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW, &nd); if (error) goto out; inode = nd.dentry->d_inode; @@ -669,6 +681,11 @@ out: return error; } +asmlinkage long sys_chmod(const char __user *filename, mode_t mode) +{ + return sys_fchmodat(AT_FDCWD, filename, mode); +} + static int chown_common(struct dentry * dentry, uid_t user, gid_t group) { struct inode * inode; @@ -717,6 +734,26 @@ asmlinkage long sys_chown(const char __user * filename, uid_t user, gid_t group) return error; } +asmlinkage long sys_fchownat(int dfd, const char __user *filename, uid_t user, + gid_t group, int flag) +{ + struct nameidata nd; + int error = -EINVAL; + int follow; + + if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) + goto out; + + follow = (flag & AT_SYMLINK_NOFOLLOW) ? 0 : LOOKUP_FOLLOW; + error = __user_walk_fd(dfd, filename, follow, &nd); + if (!error) { + error = chown_common(nd.dentry, user, group); + path_release(&nd); + } +out: + return error; +} + asmlinkage long sys_lchown(const char __user * filename, uid_t user, gid_t group) { struct nameidata nd; @@ -820,7 +857,8 @@ cleanup_file: * for the internal routines (ie open_namei()/follow_link() etc). 00 is * used by symlinks. */ -struct file *filp_open(const char * filename, int flags, int mode) +static struct file *do_filp_open(int dfd, const char *filename, int flags, + int mode) { int namei_flags, error; struct nameidata nd; @@ -829,12 +867,17 @@ struct file *filp_open(const char * filename, int flags, int mode) if ((namei_flags+1) & O_ACCMODE) namei_flags++; - error = open_namei(filename, namei_flags, mode, &nd); + error = open_namei(dfd, filename, namei_flags, mode, &nd); if (!error) return nameidata_to_filp(&nd, flags); return ERR_PTR(error); } + +struct file *filp_open(const char *filename, int flags, int mode) +{ + return do_filp_open(AT_FDCWD, filename, flags, mode); +} EXPORT_SYMBOL(filp_open); /** @@ -991,7 +1034,7 @@ void fastcall put_unused_fd(unsigned int fd) EXPORT_SYMBOL(put_unused_fd); /* - * Install a file pointer in the fd array. + * Install a file pointer in the fd array. * * The VFS is full of places where we drop the files lock between * setting the open_fds bitmap and installing the file in the file @@ -1016,7 +1059,7 @@ void fastcall fd_install(unsigned int fd, struct file * file) EXPORT_SYMBOL(fd_install); -long do_sys_open(const char __user *filename, int flags, int mode) +long do_sys_open(int dfd, const char __user *filename, int flags, int mode) { char *tmp = getname(filename); int fd = PTR_ERR(tmp); @@ -1024,7 +1067,7 @@ long do_sys_open(const char __user *filename, int flags, int mode) if (!IS_ERR(tmp)) { fd = get_unused_fd(); if (fd >= 0) { - struct file *f = filp_open(tmp, flags, mode); + struct file *f = do_filp_open(dfd, tmp, flags, mode); if (IS_ERR(f)) { put_unused_fd(fd); fd = PTR_ERR(f); @@ -1043,10 +1086,20 @@ asmlinkage long sys_open(const char __user *filename, int flags, int mode) if (force_o_largefile()) flags |= O_LARGEFILE; - return do_sys_open(filename, flags, mode); + return do_sys_open(AT_FDCWD, filename, flags, mode); } EXPORT_SYMBOL_GPL(sys_open); +asmlinkage long sys_openat(int dfd, const char __user *filename, int flags, + int mode) +{ + if (force_o_largefile()) + flags |= O_LARGEFILE; + + return do_sys_open(dfd, filename, flags, mode); +} +EXPORT_SYMBOL_GPL(sys_openat); + #ifndef __alpha__ /* diff --git a/fs/stat.c b/fs/stat.c index b8a0e5110ab..24211b030f3 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -63,12 +63,12 @@ int vfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) EXPORT_SYMBOL(vfs_getattr); -int vfs_stat(char __user *name, struct kstat *stat) +int vfs_stat_fd(int dfd, char __user *name, struct kstat *stat) { struct nameidata nd; int error; - error = user_path_walk(name, &nd); + error = __user_walk_fd(dfd, name, LOOKUP_FOLLOW, &nd); if (!error) { error = vfs_getattr(nd.mnt, nd.dentry, stat); path_release(&nd); @@ -76,14 +76,19 @@ int vfs_stat(char __user *name, struct kstat *stat) return error; } +int vfs_stat(char __user *name, struct kstat *stat) +{ + return vfs_stat_fd(AT_FDCWD, name, stat); +} + EXPORT_SYMBOL(vfs_stat); -int vfs_lstat(char __user *name, struct kstat *stat) +int vfs_lstat_fd(int dfd, char __user *name, struct kstat *stat) { struct nameidata nd; int error; - error = user_path_walk_link(name, &nd); + error = __user_walk_fd(dfd, name, 0, &nd); if (!error) { error = vfs_getattr(nd.mnt, nd.dentry, stat); path_release(&nd); @@ -91,6 +96,11 @@ int vfs_lstat(char __user *name, struct kstat *stat) return error; } +int vfs_lstat(char __user *name, struct kstat *stat) +{ + return vfs_lstat_fd(AT_FDCWD, name, stat); +} + EXPORT_SYMBOL(vfs_lstat); int vfs_fstat(unsigned int fd, struct kstat *stat) @@ -151,7 +161,7 @@ static int cp_old_stat(struct kstat *stat, struct __old_kernel_stat __user * sta asmlinkage long sys_stat(char __user * filename, struct __old_kernel_stat __user * statbuf) { struct kstat stat; - int error = vfs_stat(filename, &stat); + int error = vfs_stat_fd(AT_FDCWD, filename, &stat); if (!error) error = cp_old_stat(&stat, statbuf); @@ -161,7 +171,7 @@ asmlinkage long sys_stat(char __user * filename, struct __old_kernel_stat __user asmlinkage long sys_lstat(char __user * filename, struct __old_kernel_stat __user * statbuf) { struct kstat stat; - int error = vfs_lstat(filename, &stat); + int error = vfs_lstat_fd(AT_FDCWD, filename, &stat); if (!error) error = cp_old_stat(&stat, statbuf); @@ -229,27 +239,50 @@ static int cp_new_stat(struct kstat *stat, struct stat __user *statbuf) return copy_to_user(statbuf,&tmp,sizeof(tmp)) ? -EFAULT : 0; } -asmlinkage long sys_newstat(char __user * filename, struct stat __user * statbuf) +asmlinkage long sys_newstat(char __user *filename, struct stat __user *statbuf) { struct kstat stat; - int error = vfs_stat(filename, &stat); + int error = vfs_stat_fd(AT_FDCWD, filename, &stat); if (!error) error = cp_new_stat(&stat, statbuf); return error; } -asmlinkage long sys_newlstat(char __user * filename, struct stat __user * statbuf) + +asmlinkage long sys_newlstat(char __user *filename, struct stat __user *statbuf) { struct kstat stat; - int error = vfs_lstat(filename, &stat); + int error = vfs_lstat_fd(AT_FDCWD, filename, &stat); if (!error) error = cp_new_stat(&stat, statbuf); return error; } -asmlinkage long sys_newfstat(unsigned int fd, struct stat __user * statbuf) + +asmlinkage long sys_newfstatat(int dfd, char __user *filename, + struct stat __user *statbuf, int flag) +{ + struct kstat stat; + int error = -EINVAL; + + if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) + goto out; + + if (flag & AT_SYMLINK_NOFOLLOW) + error = vfs_lstat_fd(dfd, filename, &stat); + else + error = vfs_stat_fd(dfd, filename, &stat); + + if (!error) + error = cp_new_stat(&stat, statbuf); + +out: + return error; +} + +asmlinkage long sys_newfstat(unsigned int fd, struct stat __user *statbuf) { struct kstat stat; int error = vfs_fstat(fd, &stat); @@ -260,7 +293,8 @@ asmlinkage long sys_newfstat(unsigned int fd, struct stat __user * statbuf) return error; } -asmlinkage long sys_readlink(const char __user * path, char __user * buf, int bufsiz) +asmlinkage long sys_readlinkat(int dfd, const char __user *path, + char __user *buf, int bufsiz) { struct nameidata nd; int error; @@ -268,7 +302,7 @@ asmlinkage long sys_readlink(const char __user * path, char __user * buf, int bu if (bufsiz <= 0) return -EINVAL; - error = user_path_walk_link(path, &nd); + error = __user_walk_fd(dfd, path, 0, &nd); if (!error) { struct inode * inode = nd.dentry->d_inode; @@ -285,6 +319,12 @@ asmlinkage long sys_readlink(const char __user * path, char __user * buf, int bu return error; } +asmlinkage long sys_readlink(const char __user *path, char __user *buf, + int bufsiz) +{ + return sys_readlinkat(AT_FDCWD, path, buf, bufsiz); +} + /* ---------- LFS-64 ----------- */ #ifdef __ARCH_WANT_STAT64 diff --git a/include/linux/fcntl.h b/include/linux/fcntl.h index 8a7c82151de..c52a63755fd 100644 --- a/include/linux/fcntl.h +++ b/include/linux/fcntl.h @@ -23,6 +23,13 @@ #define DN_ATTRIB 0x00000020 /* File changed attibutes */ #define DN_MULTISHOT 0x80000000 /* Don't remove notifier */ +#define AT_FDCWD -100 /* Special value used to indicate + openat should use the current + working directory. */ +#define AT_SYMLINK_NOFOLLOW 0x100 /* Do not follow symbolic links. */ +#define AT_REMOVEDIR 0x200 /* Remove directory instead of + unlinking file. */ + #ifdef __KERNEL__ #ifndef force_o_largefile diff --git a/include/linux/fs.h b/include/linux/fs.h index b77f2608eef..84bb449b9b0 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1340,7 +1340,8 @@ static inline int break_lease(struct inode *inode, unsigned int mode) extern int do_truncate(struct dentry *, loff_t start, unsigned int time_attrs, struct file *filp); -extern long do_sys_open(const char __user *filename, int flags, int mode); +extern long do_sys_open(int fdf, const char __user *filename, int flags, + int mode); extern struct file *filp_open(const char *, int, int); extern struct file * dentry_open(struct dentry *, struct vfsmount *, int); extern int filp_close(struct file *, fl_owner_t id); @@ -1479,7 +1480,7 @@ static inline void allow_write_access(struct file *file) } extern int do_pipe(int *); -extern int open_namei(const char *, int, int, struct nameidata *); +extern int open_namei(int dfd, const char *, int, int, struct nameidata *); extern int may_open(struct nameidata *, int, int); extern int kernel_read(struct file *, unsigned long, char *, unsigned long); @@ -1677,6 +1678,8 @@ extern int vfs_readdir(struct file *, filldir_t, void *); extern int vfs_stat(char __user *, struct kstat *); extern int vfs_lstat(char __user *, struct kstat *); +extern int vfs_stat_fd(int dfd, char __user *, struct kstat *); +extern int vfs_lstat_fd(int dfd, char __user *, struct kstat *); extern int vfs_fstat(unsigned int, struct kstat *); extern int vfs_ioctl(struct file *, unsigned int, unsigned int, unsigned long); diff --git a/include/linux/namei.h b/include/linux/namei.h index b699e427c00..e6698013e4d 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -56,10 +56,11 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND}; #define LOOKUP_ACCESS (0x0400) extern int FASTCALL(__user_walk(const char __user *, unsigned, struct nameidata *)); +extern int FASTCALL(__user_walk_fd(int dfd, const char __user *, unsigned, struct nameidata *)); #define user_path_walk(name,nd) \ - __user_walk(name, LOOKUP_FOLLOW, nd) + __user_walk_fd(AT_FDCWD, name, LOOKUP_FOLLOW, nd) #define user_path_walk_link(name,nd) \ - __user_walk(name, 0, nd) + __user_walk_fd(AT_FDCWD, name, 0, nd) extern int FASTCALL(path_lookup(const char *, unsigned, struct nameidata *)); extern int FASTCALL(path_walk(const char *, struct nameidata *)); extern int FASTCALL(link_path_walk(const char *, struct nameidata *)); @@ -67,7 +68,7 @@ extern void path_release(struct nameidata *); extern void path_release_on_umount(struct nameidata *); extern int __user_path_lookup_open(const char __user *, unsigned lookup_flags, struct nameidata *nd, int open_flags); -extern int path_lookup_open(const char *, unsigned lookup_flags, struct nameidata *, int open_flags); +extern int path_lookup_open(int dfd, const char *name, unsigned lookup_flags, struct nameidata *, int open_flags); extern struct file *lookup_instantiate_filp(struct nameidata *nd, struct dentry *dentry, int (*open)(struct inode *, struct file *)); extern struct file *nameidata_to_filp(struct nameidata *nd, int flags); diff --git a/include/linux/time.h b/include/linux/time.h index f2aca7ec632..614dd846583 100644 --- a/include/linux/time.h +++ b/include/linux/time.h @@ -74,7 +74,7 @@ extern void do_gettimeofday(struct timeval *tv); extern int do_settimeofday(struct timespec *tv); extern int do_sys_settimeofday(struct timespec *tv, struct timezone *tz); #define do_posix_clock_monotonic_gettime(ts) ktime_get_ts(ts) -extern long do_utimes(char __user *filename, struct timeval *times); +extern long do_utimes(int dfd, char __user *filename, struct timeval *times); struct itimerval; extern int do_setitimer(int which, struct itimerval *value, struct itimerval *ovalue); -- cgit v1.2.3 From 4f085507231e8003c66ed12e38c73b76e938ee95 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 18 Jan 2006 17:43:55 -0800 Subject: [PATCH] vfs: *at functions: i386 Wire up the x86 syscalls Signed-off-by: Ulrich Drepper Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/syscall_table.S | 13 +++++++++++++ include/asm-i386/unistd.h | 15 ++++++++++++++- 2 files changed, 27 insertions(+), 1 deletion(-) diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S index 6ff3e524322..a988b40e594 100644 --- a/arch/i386/kernel/syscall_table.S +++ b/arch/i386/kernel/syscall_table.S @@ -294,3 +294,16 @@ ENTRY(sys_call_table) .long sys_inotify_add_watch .long sys_inotify_rm_watch .long sys_migrate_pages + .long sys_openat /* 295 */ + .long sys_mkdirat + .long sys_mknodat + .long sys_fchownat + .long sys_futimesat + .long sys_newfstatat /* 300 */ + .long sys_unlinkat + .long sys_renameat + .long sys_linkat + .long sys_symlinkat + .long sys_readlinkat /* 305 */ + .long sys_fchmodat + .long sys_faccessat diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h index 481c3c0ea72..e3028aaf6d8 100644 --- a/include/asm-i386/unistd.h +++ b/include/asm-i386/unistd.h @@ -300,8 +300,21 @@ #define __NR_inotify_add_watch 292 #define __NR_inotify_rm_watch 293 #define __NR_migrate_pages 294 +#define __NR_openat 295 +#define __NR_mkdirat 296 +#define __NR_mknodat 297 +#define __NR_fchownat 298 +#define __NR_futimesat 299 +#define __NR_newfstatat 300 +#define __NR_unlinkat 301 +#define __NR_renameat 302 +#define __NR_linkat 303 +#define __NR_symlinkat 304 +#define __NR_readlinkat 305 +#define __NR_fchmodat 306 +#define __NR_faccessat 307 -#define NR_syscalls 295 +#define NR_syscalls 308 /* * user-visible error numbers are in the range -1 - -128: see -- cgit v1.2.3 From a60fc5190a31d98508ea6a76f74217f4104e74b7 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 18 Jan 2006 17:43:56 -0800 Subject: [PATCH] vfs: *at functions: x86_64 Wire up the x86_64 syscalls. Signed-off-by: Ulrich Drepper Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/ia32/ia32entry.S | 13 +++++++++++++ include/asm-x86_64/ia32_unistd.h | 15 ++++++++++++++- include/asm-x86_64/unistd.h | 29 ++++++++++++++++++++++++++++- 3 files changed, 55 insertions(+), 2 deletions(-) diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S index 58f5bfb52c6..f05c2a80248 100644 --- a/arch/x86_64/ia32/ia32entry.S +++ b/arch/x86_64/ia32/ia32entry.S @@ -672,6 +672,19 @@ ia32_sys_call_table: .quad sys_inotify_add_watch .quad sys_inotify_rm_watch .quad sys_migrate_pages + .quad compat_sys_openat /* 295 */ + .quad sys_mkdirat + .quad sys_mknodat + .quad sys_fchownat + .quad sys_futimesat + .quad compat_sys_newfstatat /* 300 */ + .quad sys_unlinkat + .quad sys_renameat + .quad sys_linkat + .quad sys_symlinkat + .quad sys_readlinkat /* 305 */ + .quad sys_fchmodat + .quad sys_faccessat ia32_syscall_end: .rept IA32_NR_syscalls-(ia32_syscall_end-ia32_sys_call_table)/8 .quad ni_syscall diff --git a/include/asm-x86_64/ia32_unistd.h b/include/asm-x86_64/ia32_unistd.h index e8843362a6c..e87cd83a0e8 100644 --- a/include/asm-x86_64/ia32_unistd.h +++ b/include/asm-x86_64/ia32_unistd.h @@ -300,7 +300,20 @@ #define __NR_ia32_inotify_add_watch 292 #define __NR_ia32_inotify_rm_watch 293 #define __NR_ia32_migrate_pages 294 +#define __NR_ia32_opanat 295 +#define __NR_ia32_mkdirat 296 +#define __NR_ia32_mknodat 297 +#define __NR_ia32_fchownat 298 +#define __NR_ia32_futimesat 299 +#define __NR_ia32_newfstatat 300 +#define __NR_ia32_unlinkat 301 +#define __NR_ia32_renameat 302 +#define __NR_ia32_linkat 303 +#define __NR_ia32_symlinkat 304 +#define __NR_ia32_readlinkat 305 +#define __NR_ia32_fchmodat 306 +#define __NR_ia32_faccessat 307 -#define IA32_NR_syscalls 295 /* must be > than biggest syscall! */ +#define IA32_NR_syscalls 308 /* must be > than biggest syscall! */ #endif /* _ASM_X86_64_IA32_UNISTD_H_ */ diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h index e6f896161c1..436d099b5b6 100644 --- a/include/asm-x86_64/unistd.h +++ b/include/asm-x86_64/unistd.h @@ -573,8 +573,35 @@ __SYSCALL(__NR_inotify_add_watch, sys_inotify_add_watch) __SYSCALL(__NR_inotify_rm_watch, sys_inotify_rm_watch) #define __NR_migrate_pages 256 __SYSCALL(__NR_migrate_pages, sys_migrate_pages) +#define __NR_openat 257 +__SYSCALL(__NR_openat, sys_openat) +#define __NR_mkdirat 258 +__SYSCALL(__NR_mkdirat, sys_mkdirat) +#define __NR_mknodat 259 +__SYSCALL(__NR_mknodat, sys_mknodat) +#define __NR_fchownat 260 +__SYSCALL(__NR_fchownat, sys_fchownat) +#define __NR_futimesat 261 +__SYSCALL(__NR_futimesat, sys_futimesat) +#define __NR_newfstatat 262 +__SYSCALL(__NR_newfstatat, sys_newfstatat) +#define __NR_unlinkat 263 +__SYSCALL(__NR_unlinkat, sys_unlinkat) +#define __NR_renameat 264 +__SYSCALL(__NR_renameat, sys_renameat) +#define __NR_linkat 265 +__SYSCALL(__NR_linkat, sys_linkat) +#define __NR_symlinkat 266 +__SYSCALL(__NR_symlinkat, sys_symlinkat) +#define __NR_readlinkat 267 +__SYSCALL(__NR_readlinkat, sys_readlinkat) +#define __NR_fchmodat 268 +__SYSCALL(__NR_fchmodat, sys_fchmodat) +#define __NR_faccessat 269 +__SYSCALL(__NR_faccessat, sys_faccessat) + +#define __NR_syscall_max __NR_faccessat -#define __NR_syscall_max __NR_migrate_pages #ifndef __NO_STUBS /* user-visible error numbers are in the range -1 - -4095 */ -- cgit v1.2.3 From 150256d8aadb3a337c31efa9e175cbd25bf06b06 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 18 Jan 2006 17:43:57 -0800 Subject: [PATCH] Generic sys_rt_sigsuspend() The TIF_RESTORE_SIGMASK flag allows us to have a generic implementation of sys_rt_sigsuspend() instead of duplicating it for each architecture. This provides such an implementation and makes arch/powerpc use it. It also tidies up the ppc32 sys_sigsuspend() to use TIF_RESTORE_SIGMASK. Signed-off-by: David Woodhouse Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/kernel/signal_32.c | 56 ++++------------------------------------- arch/powerpc/kernel/signal_64.c | 36 -------------------------- include/asm-powerpc/unistd.h | 2 ++ include/linux/sched.h | 1 + kernel/compat.c | 28 +++++++++++++++++++++ kernel/signal.c | 26 +++++++++++++++++++ 6 files changed, 62 insertions(+), 87 deletions(-) diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c index 177bba78fb0..7f0d5ce2567 100644 --- a/arch/powerpc/kernel/signal_32.c +++ b/arch/powerpc/kernel/signal_32.c @@ -252,8 +252,7 @@ int do_signal(sigset_t *oldset, struct pt_regs *regs); /* * Atomically swap in the new signal mask, and wait for a signal. */ -long sys_sigsuspend(old_sigset_t mask, int p2, int p3, int p4, int p6, int p7, - struct pt_regs *regs) +long sys_sigsuspend(old_sigset_t mask) { sigset_t saveset; @@ -264,55 +263,10 @@ long sys_sigsuspend(old_sigset_t mask, int p2, int p3, int p4, int p6, int p7, recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); - regs->result = -EINTR; - regs->gpr[3] = EINTR; - regs->ccr |= 0x10000000; - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - if (do_signal(&saveset, regs)) { - set_thread_flag(TIF_RESTOREALL); - return 0; - } - } -} - -long sys_rt_sigsuspend( -#ifdef CONFIG_PPC64 - compat_sigset_t __user *unewset, -#else - sigset_t __user *unewset, -#endif - size_t sigsetsize, int p3, int p4, - int p6, int p7, struct pt_regs *regs) -{ - sigset_t saveset, newset; - - /* XXX: Don't preclude handling different sized sigset_t's. */ - if (sigsetsize != sizeof(sigset_t)) - return -EINVAL; - - if (get_sigset_t(&newset, unewset)) - return -EFAULT; - sigdelsetmask(&newset, ~_BLOCKABLE); - - spin_lock_irq(¤t->sighand->siglock); - saveset = current->blocked; - current->blocked = newset; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - regs->result = -EINTR; - regs->gpr[3] = EINTR; - regs->ccr |= 0x10000000; - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - if (do_signal(&saveset, regs)) { - set_thread_flag(TIF_RESTOREALL); - return 0; - } - } + current->state = TASK_INTERRUPTIBLE; + schedule(); + set_thread_flag(TIF_RESTORE_SIGMASK); + return -ERESTARTNOHAND; } #ifdef CONFIG_PPC32 diff --git a/arch/powerpc/kernel/signal_64.c b/arch/powerpc/kernel/signal_64.c index 7b9d999e211..a4a6812e815 100644 --- a/arch/powerpc/kernel/signal_64.c +++ b/arch/powerpc/kernel/signal_64.c @@ -67,42 +67,6 @@ struct rt_sigframe { char abigap[288]; } __attribute__ ((aligned (16))); - -/* - * Atomically swap in the new signal mask, and wait for a signal. - */ -long sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize, int p3, int p4, - int p6, int p7, struct pt_regs *regs) -{ - sigset_t saveset, newset; - - /* XXX: Don't preclude handling different sized sigset_t's. */ - if (sigsetsize != sizeof(sigset_t)) - return -EINVAL; - - if (copy_from_user(&newset, unewset, sizeof(newset))) - return -EFAULT; - sigdelsetmask(&newset, ~_BLOCKABLE); - - spin_lock_irq(¤t->sighand->siglock); - saveset = current->blocked; - current->blocked = newset; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - regs->result = -EINTR; - regs->gpr[3] = EINTR; - regs->ccr |= 0x10000000; - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - if (do_signal(&saveset, regs)) { - set_thread_flag(TIF_RESTOREALL); - return 0; - } - } -} - long sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss, unsigned long r5, unsigned long r6, unsigned long r7, unsigned long r8, struct pt_regs *regs) diff --git a/include/asm-powerpc/unistd.h b/include/asm-powerpc/unistd.h index 19eaac3fbbf..76daec49606 100644 --- a/include/asm-powerpc/unistd.h +++ b/include/asm-powerpc/unistd.h @@ -444,11 +444,13 @@ type name(type1 arg1, type2 arg2, type3 arg3, type4 arg4, type5 arg5, type6 arg6 #define __ARCH_WANT_SYS_SIGPENDING #define __ARCH_WANT_SYS_SIGPROCMASK #define __ARCH_WANT_SYS_RT_SIGACTION +#define __ARCH_WANT_SYS_RT_SIGSUSPEND #ifdef CONFIG_PPC32 #define __ARCH_WANT_OLD_STAT #endif #ifdef CONFIG_PPC64 #define __ARCH_WANT_COMPAT_SYS_TIME +#define __ARCH_WANT_COMPAT_SYS_RT_SIGSUSPEND #endif /* diff --git a/include/linux/sched.h b/include/linux/sched.h index 2df1a1a2fee..0cfcd1c7865 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -809,6 +809,7 @@ struct task_struct { struct sighand_struct *sighand; sigset_t blocked, real_blocked; + sigset_t saved_sigmask; /* To be restored with TIF_RESTORE_SIGMASK */ struct sigpending pending; unsigned long sas_ss_sp; diff --git a/kernel/compat.c b/kernel/compat.c index 256e5d9f064..1867290c37e 100644 --- a/kernel/compat.c +++ b/kernel/compat.c @@ -871,3 +871,31 @@ asmlinkage long compat_sys_stime(compat_time_t __user *tptr) } #endif /* __ARCH_WANT_COMPAT_SYS_TIME */ + +#ifdef __ARCH_WANT_COMPAT_SYS_RT_SIGSUSPEND +asmlinkage long compat_sys_rt_sigsuspend(compat_sigset_t __user *unewset, compat_size_t sigsetsize) +{ + sigset_t newset; + compat_sigset_t newset32; + + /* XXX: Don't preclude handling different sized sigset_t's. */ + if (sigsetsize != sizeof(sigset_t)) + return -EINVAL; + + if (copy_from_user(&newset32, unewset, sizeof(compat_sigset_t))) + return -EFAULT; + sigset_from_compat(&newset, &newset32); + sigdelsetmask(&newset, sigmask(SIGKILL)|sigmask(SIGSTOP)); + + spin_lock_irq(¤t->sighand->siglock); + current->saved_sigmask = current->blocked; + current->blocked = newset; + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + + current->state = TASK_INTERRUPTIBLE; + schedule(); + set_thread_flag(TIF_RESTORE_SIGMASK); + return -ERESTARTNOHAND; +} +#endif /* __ARCH_WANT_COMPAT_SYS_RT_SIGSUSPEND */ diff --git a/kernel/signal.c b/kernel/signal.c index 5dafbd36d62..d3efafd8109 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -2721,6 +2721,32 @@ sys_pause(void) #endif +#ifdef __ARCH_WANT_SYS_RT_SIGSUSPEND +asmlinkage long sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize) +{ + sigset_t newset; + + /* XXX: Don't preclude handling different sized sigset_t's. */ + if (sigsetsize != sizeof(sigset_t)) + return -EINVAL; + + if (copy_from_user(&newset, unewset, sizeof(newset))) + return -EFAULT; + sigdelsetmask(&newset, sigmask(SIGKILL)|sigmask(SIGSTOP)); + + spin_lock_irq(¤t->sighand->siglock); + current->saved_sigmask = current->blocked; + current->blocked = newset; + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + + current->state = TASK_INTERRUPTIBLE; + schedule(); + set_thread_flag(TIF_RESTORE_SIGMASK); + return -ERESTARTNOHAND; +} +#endif /* __ARCH_WANT_SYS_RT_SIGSUSPEND */ + void __init signals_init(void) { sigqueue_cachep = -- cgit v1.2.3 From a411aee96ea7fe6fe065df65bf29ea755bcdb554 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 18 Jan 2006 17:43:59 -0800 Subject: [PATCH] Handle TIF_RESTORE_SIGMASK for FRV Handle TIF_RESTORE_SIGMASK as added by David Woodhouse's patch entitled: [PATCH] 2/3 Add TIF_RESTORE_SIGMASK support for arch/powerpc [PATCH] 3/3 Generic sys_rt_sigsuspend It does the following: (1) Declares TIF_RESTORE_SIGMASK for FRV. (2) Invokes it over to do_signal() when TIF_RESTORE_SIGMASK is set. (3) Makes do_signal() support TIF_RESTORE_SIGMASK, using the signal mask saved in current->saved_sigmask. (4) Discards sys_rt_sigsuspend() from the arch, using the generic one instead. (5) Makes sys_sigsuspend() save the signal mask and set TIF_RESTORE_SIGMASK rather than attempting to fudge the return registers. (6) Makes sys_sigsuspend() return -ERESTARTNOHAND rather than looping intrinsically. (7) Makes setup_frame(), setup_rt_frame() and handle_signal() return 0 or -EFAULT rather than true/false to be consistent with the rest of the kernel. Due to the fact do_signal() is then only called from one place: (8) Make do_signal() no longer have a return value is it was just being ignored; force_sig() takes care of this. (9) Discards the old sigmask argument to do_signal() as it's no longer necessary. This patch depends on the FRV signalling patches as well as the sys_rt_sigsuspend patch. Signed-off-by: David Howells Signed-off-by: David Woodhouse Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/frv/kernel/signal.c | 120 ++++++++++++++++-------------------------- include/asm-frv/thread_info.h | 2 + include/asm-frv/unistd.h | 1 + 3 files changed, 47 insertions(+), 76 deletions(-) diff --git a/arch/frv/kernel/signal.c b/arch/frv/kernel/signal.c index 5b7146f54fd..679c1d5cc95 100644 --- a/arch/frv/kernel/signal.c +++ b/arch/frv/kernel/signal.c @@ -35,74 +35,22 @@ struct fdpic_func_descriptor { unsigned long GOT; }; -static int do_signal(sigset_t *oldset); - /* * Atomically swap in the new signal mask, and wait for a signal. */ asmlinkage int sys_sigsuspend(int history0, int history1, old_sigset_t mask) { - sigset_t saveset; - mask &= _BLOCKABLE; spin_lock_irq(¤t->sighand->siglock); - saveset = current->blocked; + current->saved_sigmask = current->blocked; siginitset(¤t->blocked, mask); recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); - __frame->gr8 = -EINTR; - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - if (do_signal(&saveset)) - /* return the signal number as the return value of this function - * - this is an utterly evil hack. syscalls should not invoke do_signal() - * as entry.S sets regs->gr8 to the return value of the system call - * - we can't just use sigpending() as we'd have to discard SIG_IGN signals - * and call waitpid() if SIGCHLD needed discarding - * - this only works on the i386 because it passes arguments to the signal - * handler on the stack, and the return value in EAX is effectively - * discarded - */ - return __frame->gr8; - } -} - -asmlinkage int sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize) -{ - sigset_t saveset, newset; - - /* XXX: Don't preclude handling different sized sigset_t's. */ - if (sigsetsize != sizeof(sigset_t)) - return -EINVAL; - - if (copy_from_user(&newset, unewset, sizeof(newset))) - return -EFAULT; - sigdelsetmask(&newset, ~_BLOCKABLE); - - spin_lock_irq(¤t->sighand->siglock); - saveset = current->blocked; - current->blocked = newset; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - __frame->gr8 = -EINTR; - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - if (do_signal(&saveset)) - /* return the signal number as the return value of this function - * - this is an utterly evil hack. syscalls should not invoke do_signal() - * as entry.S sets regs->gr8 to the return value of the system call - * - we can't just use sigpending() as we'd have to discard SIG_IGN signals - * and call waitpid() if SIGCHLD needed discarding - * - this only works on the i386 because it passes arguments to the signal - * handler on the stack, and the return value in EAX is effectively - * discarded - */ - return __frame->gr8; - } + current->state = TASK_INTERRUPTIBLE; + schedule(); + set_thread_flag(TIF_RESTORE_SIGMASK); + return -ERESTARTNOHAND; } asmlinkage int sys_sigaction(int sig, @@ -372,11 +320,11 @@ static int setup_frame(int sig, struct k_sigaction *ka, sigset_t *set) frame->pretcode); #endif - return 1; + return 0; give_sigsegv: force_sig(SIGSEGV, current); - return 0; + return -EFAULT; } /* end setup_frame() */ @@ -471,11 +419,11 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, frame->pretcode); #endif - return 1; + return 0; give_sigsegv: force_sig(SIGSEGV, current); - return 0; + return -EFAULT; } /* end setup_rt_frame() */ @@ -516,7 +464,7 @@ static int handle_signal(unsigned long sig, siginfo_t *info, else ret = setup_frame(sig, ka, oldset); - if (ret) { + if (ret == 0) { spin_lock_irq(¤t->sighand->siglock); sigorsets(¤t->blocked, ¤t->blocked, &ka->sa.sa_mask); @@ -536,10 +484,11 @@ static int handle_signal(unsigned long sig, siginfo_t *info, * want to handle. Thus you cannot kill init even with a SIGKILL even by * mistake. */ -static int do_signal(sigset_t *oldset) +static void do_signal(void) { struct k_sigaction ka; siginfo_t info; + sigset_t *oldset; int signr; /* @@ -549,43 +498,62 @@ static int do_signal(sigset_t *oldset) * if so. */ if (!user_mode(__frame)) - return 1; + return; if (try_to_freeze()) goto no_signal; - if (!oldset) + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + oldset = ¤t->saved_sigmask; + else oldset = ¤t->blocked; signr = get_signal_to_deliver(&info, &ka, __frame, NULL); - if (signr > 0) - return handle_signal(signr, &info, &ka, oldset); + if (signr > 0) { + if (handle_signal(signr, &info, &ka, oldset) == 0) { + /* a signal was successfully delivered; the saved + * sigmask will have been stored in the signal frame, + * and will be restored by sigreturn, so we can simply + * clear the TIF_RESTORE_SIGMASK flag */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + clear_thread_flag(TIF_RESTORE_SIGMASK); + } + + return; + } no_signal: /* Did we come from a system call? */ if (__frame->syscallno >= 0) { /* Restart the system call - no handlers present */ - if (__frame->gr8 == -ERESTARTNOHAND || - __frame->gr8 == -ERESTARTSYS || - __frame->gr8 == -ERESTARTNOINTR) { + switch (__frame->gr8) { + case -ERESTARTNOHAND: + case -ERESTARTSYS: + case -ERESTARTNOINTR: __frame->gr8 = __frame->orig_gr8; __frame->pc -= 4; - } + break; - if (__frame->gr8 == -ERESTART_RESTARTBLOCK){ + case -ERESTART_RESTARTBLOCK: __frame->gr8 = __NR_restart_syscall; __frame->pc -= 4; + break; } } - return 0; + /* if there's no signal to deliver, we just put the saved sigmask + * back */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) { + clear_thread_flag(TIF_RESTORE_SIGMASK); + sigprocmask(SIG_SETMASK, ¤t->saved_sigmask, NULL); + } } /* end do_signal() */ /*****************************************************************************/ /* * notification of userspace execution resumption - * - triggered by current->work.notify_resume + * - triggered by the TIF_WORK_MASK flags */ asmlinkage void do_notify_resume(__u32 thread_info_flags) { @@ -594,7 +562,7 @@ asmlinkage void do_notify_resume(__u32 thread_info_flags) clear_thread_flag(TIF_SINGLESTEP); /* deal with pending signal delivery */ - if (thread_info_flags & _TIF_SIGPENDING) - do_signal(NULL); + if (thread_info_flags & (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK)) + do_signal(); } /* end do_notify_resume() */ diff --git a/include/asm-frv/thread_info.h b/include/asm-frv/thread_info.h index a5576e02dd1..ea426abf01d 100644 --- a/include/asm-frv/thread_info.h +++ b/include/asm-frv/thread_info.h @@ -129,6 +129,7 @@ register struct thread_info *__current_thread_info asm("gr15"); #define TIF_NEED_RESCHED 3 /* rescheduling necessary */ #define TIF_SINGLESTEP 4 /* restore singlestep on return to user mode */ #define TIF_IRET 5 /* return with iret */ +#define TIF_RESTORE_SIGMASK 6 /* restore signal mask in do_signal() */ #define TIF_POLLING_NRFLAG 16 /* true if poll_idle() is polling TIF_NEED_RESCHED */ #define TIF_MEMDIE 17 /* OOM killer killed process */ @@ -138,6 +139,7 @@ register struct thread_info *__current_thread_info asm("gr15"); #define _TIF_NEED_RESCHED (1 << TIF_NEED_RESCHED) #define _TIF_SINGLESTEP (1 << TIF_SINGLESTEP) #define _TIF_IRET (1 << TIF_IRET) +#define _TIF_RESTORE_SIGMASK (1 << TIF_RESTORE_SIGMASK) #define _TIF_POLLING_NRFLAG (1 << TIF_POLLING_NRFLAG) #define _TIF_WORK_MASK 0x0000FFFE /* work to do on interrupt/exception return */ diff --git a/include/asm-frv/unistd.h b/include/asm-frv/unistd.h index cde376a7a85..4d994d2e99e 100644 --- a/include/asm-frv/unistd.h +++ b/include/asm-frv/unistd.h @@ -486,6 +486,7 @@ static inline pid_t wait(int * wait_stat) /* #define __ARCH_WANT_SYS_SIGPENDING */ #define __ARCH_WANT_SYS_SIGPROCMASK #define __ARCH_WANT_SYS_RT_SIGACTION +#define __ARCH_WANT_SYS_RT_SIGSUSPEND #endif /* -- cgit v1.2.3 From 283828f3c19ceb3a64a8544d42cc189003e8b0fe Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 18 Jan 2006 17:44:00 -0800 Subject: [PATCH] Handle TIF_RESTORE_SIGMASK for i386 Handle TIF_RESTORE_SIGMASK as added by David Woodhouse's patch entitled: [PATCH] 2/3 Add TIF_RESTORE_SIGMASK support for arch/powerpc [PATCH] 3/3 Generic sys_rt_sigsuspend It does the following: (1) Declares TIF_RESTORE_SIGMASK for i386. (2) Invokes it over to do_signal() when TIF_RESTORE_SIGMASK is set. (3) Makes do_signal() support TIF_RESTORE_SIGMASK, using the signal mask saved in current->saved_sigmask. (4) Discards sys_rt_sigsuspend() from the arch, using the generic one instead. (5) Makes sys_sigsuspend() save the signal mask and set TIF_RESTORE_SIGMASK rather than attempting to fudge the return registers. (6) Makes sys_sigsuspend() return -ERESTARTNOHAND rather than looping intrinsically. (7) Makes setup_frame(), setup_rt_frame() and handle_signal() return 0 or -EFAULT rather than true/false to be consistent with the rest of the kernel. Due to the fact do_signal() is then only called from one place: (8) Makes do_signal() no longer have a return value is it was just being ignored; force_sig() takes care of this. (9) Discards the old sigmask argument to do_signal() as it's no longer necessary. (10) Makes do_signal() static. (11) Marks the second argument to do_notify_resume() as unused. The unused argument should remain in the middle as the arguments are passed in as registers, and the ordering is specific in entry.S Given the way do_signal() is now no longer called from sys_{,rt_}sigsuspend(), they no longer need access to the exception frame, and so can just take arguments normally. This patch depends on sys_rt_sigsuspend patch. Signed-off-by: David Howells Signed-off-by: David Woodhouse Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/signal.c | 109 ++++++++++++++++++----------------------- include/asm-i386/signal.h | 1 - include/asm-i386/thread_info.h | 2 + include/asm-i386/unistd.h | 1 + 4 files changed, 51 insertions(+), 62 deletions(-) diff --git a/arch/i386/kernel/signal.c b/arch/i386/kernel/signal.c index adcd069db91..963616d364e 100644 --- a/arch/i386/kernel/signal.c +++ b/arch/i386/kernel/signal.c @@ -37,51 +37,17 @@ asmlinkage int sys_sigsuspend(int history0, int history1, old_sigset_t mask) { - struct pt_regs * regs = (struct pt_regs *) &history0; - sigset_t saveset; - mask &= _BLOCKABLE; spin_lock_irq(¤t->sighand->siglock); - saveset = current->blocked; + current->saved_sigmask = current->blocked; siginitset(¤t->blocked, mask); recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); - regs->eax = -EINTR; - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - if (do_signal(regs, &saveset)) - return -EINTR; - } -} - -asmlinkage int -sys_rt_sigsuspend(struct pt_regs regs) -{ - sigset_t saveset, newset; - - /* XXX: Don't preclude handling different sized sigset_t's. */ - if (regs.ecx != sizeof(sigset_t)) - return -EINVAL; - - if (copy_from_user(&newset, (sigset_t __user *)regs.ebx, sizeof(newset))) - return -EFAULT; - sigdelsetmask(&newset, ~_BLOCKABLE); - - spin_lock_irq(¤t->sighand->siglock); - saveset = current->blocked; - current->blocked = newset; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - regs.eax = -EINTR; - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - if (do_signal(®s, &saveset)) - return -EINTR; - } + current->state = TASK_INTERRUPTIBLE; + schedule(); + set_thread_flag(TIF_RESTORE_SIGMASK); + return -ERESTARTNOHAND; } asmlinkage int @@ -433,11 +399,11 @@ static int setup_frame(int sig, struct k_sigaction *ka, current->comm, current->pid, frame, regs->eip, frame->pretcode); #endif - return 1; + return 0; give_sigsegv: force_sigsegv(sig, current); - return 0; + return -EFAULT; } static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, @@ -527,11 +493,11 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, current->comm, current->pid, frame, regs->eip, frame->pretcode); #endif - return 1; + return 0; give_sigsegv: force_sigsegv(sig, current); - return 0; + return -EFAULT; } /* @@ -581,7 +547,7 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, else ret = setup_frame(sig, ka, oldset, regs); - if (ret) { + if (ret == 0) { spin_lock_irq(¤t->sighand->siglock); sigorsets(¤t->blocked,¤t->blocked,&ka->sa.sa_mask); if (!(ka->sa.sa_flags & SA_NODEFER)) @@ -598,11 +564,12 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, * want to handle. Thus you cannot kill init even with a SIGKILL even by * mistake. */ -int fastcall do_signal(struct pt_regs *regs, sigset_t *oldset) +static void fastcall do_signal(struct pt_regs *regs) { siginfo_t info; int signr; struct k_sigaction ka; + sigset_t *oldset; /* * We want the common case to go fast, which @@ -613,12 +580,14 @@ int fastcall do_signal(struct pt_regs *regs, sigset_t *oldset) * CS suffices. */ if (!user_mode(regs)) - return 1; + return; if (try_to_freeze()) goto no_signal; - if (!oldset) + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + oldset = ¤t->saved_sigmask; + else oldset = ¤t->blocked; signr = get_signal_to_deliver(&info, &ka, regs, NULL); @@ -628,38 +597,55 @@ int fastcall do_signal(struct pt_regs *regs, sigset_t *oldset) * have been cleared if the watchpoint triggered * inside the kernel. */ - if (unlikely(current->thread.debugreg[7])) { + if (unlikely(current->thread.debugreg[7])) set_debugreg(current->thread.debugreg[7], 7); - } /* Whee! Actually deliver the signal. */ - return handle_signal(signr, &info, &ka, oldset, regs); + if (handle_signal(signr, &info, &ka, oldset, regs) == 0) { + /* a signal was successfully delivered; the saved + * sigmask will have been stored in the signal frame, + * and will be restored by sigreturn, so we can simply + * clear the TIF_RESTORE_SIGMASK flag */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + clear_thread_flag(TIF_RESTORE_SIGMASK); + } + + return; } - no_signal: +no_signal: /* Did we come from a system call? */ if (regs->orig_eax >= 0) { /* Restart the system call - no handlers present */ - if (regs->eax == -ERESTARTNOHAND || - regs->eax == -ERESTARTSYS || - regs->eax == -ERESTARTNOINTR) { + switch (regs->eax) { + case -ERESTARTNOHAND: + case -ERESTARTSYS: + case -ERESTARTNOINTR: regs->eax = regs->orig_eax; regs->eip -= 2; - } - if (regs->eax == -ERESTART_RESTARTBLOCK){ + break; + + case -ERESTART_RESTARTBLOCK: regs->eax = __NR_restart_syscall; regs->eip -= 2; + break; } } - return 0; + + /* if there's no signal to deliver, we just put the saved sigmask + * back */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) { + clear_thread_flag(TIF_RESTORE_SIGMASK); + sigprocmask(SIG_SETMASK, ¤t->saved_sigmask, NULL); + } } /* * notification of userspace execution resumption - * - triggered by current->work.notify_resume + * - triggered by the TIF_WORK_MASK flags */ __attribute__((regparm(3))) -void do_notify_resume(struct pt_regs *regs, sigset_t *oldset, +void do_notify_resume(struct pt_regs *regs, void *_unused, __u32 thread_info_flags) { /* Pending single-step? */ @@ -667,9 +653,10 @@ void do_notify_resume(struct pt_regs *regs, sigset_t *oldset, regs->eflags |= TF_MASK; clear_thread_flag(TIF_SINGLESTEP); } + /* deal with pending signal delivery */ - if (thread_info_flags & _TIF_SIGPENDING) - do_signal(regs,oldset); + if (thread_info_flags & (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK)) + do_signal(regs); clear_thread_flag(TIF_IRET); } diff --git a/include/asm-i386/signal.h b/include/asm-i386/signal.h index 76524b4052a..026fd231488 100644 --- a/include/asm-i386/signal.h +++ b/include/asm-i386/signal.h @@ -218,7 +218,6 @@ static __inline__ int sigfindinword(unsigned long word) } struct pt_regs; -extern int FASTCALL(do_signal(struct pt_regs *regs, sigset_t *oldset)); #define ptrace_signal_deliver(regs, cookie) \ do { \ diff --git a/include/asm-i386/thread_info.h b/include/asm-i386/thread_info.h index 2493e77e8c3..e20e99551d7 100644 --- a/include/asm-i386/thread_info.h +++ b/include/asm-i386/thread_info.h @@ -140,6 +140,7 @@ register unsigned long current_stack_pointer asm("esp") __attribute_used__; #define TIF_SYSCALL_EMU 6 /* syscall emulation active */ #define TIF_SYSCALL_AUDIT 7 /* syscall auditing active */ #define TIF_SECCOMP 8 /* secure computing */ +#define TIF_RESTORE_SIGMASK 9 /* restore signal mask in do_signal() */ #define TIF_POLLING_NRFLAG 16 /* true if poll_idle() is polling TIF_NEED_RESCHED */ #define TIF_MEMDIE 17 @@ -152,6 +153,7 @@ register unsigned long current_stack_pointer asm("esp") __attribute_used__; #define _TIF_SYSCALL_EMU (1< Date: Wed, 18 Jan 2006 17:44:01 -0800 Subject: [PATCH] TIF_RESTORE_SIGMASK support for arch/powerpc Implement the TIF_RESTORE_SIGMASK flag in the new arch/powerpc kernel, for both 32-bit and 64-bit system call paths. Signed-off-by: David Woodhouse Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/kernel/entry_32.S | 8 ++++---- arch/powerpc/kernel/entry_64.S | 2 +- arch/powerpc/kernel/signal_32.c | 20 +++++++++++++++----- arch/powerpc/kernel/signal_64.c | 20 ++++++++++++++++++-- arch/powerpc/kernel/systbl.S | 2 ++ include/asm-powerpc/thread_info.h | 5 ++++- include/asm-powerpc/unistd.h | 4 +++- 7 files changed, 47 insertions(+), 14 deletions(-) diff --git a/arch/powerpc/kernel/entry_32.S b/arch/powerpc/kernel/entry_32.S index d8da2a35c0a..f20a67261ec 100644 --- a/arch/powerpc/kernel/entry_32.S +++ b/arch/powerpc/kernel/entry_32.S @@ -227,7 +227,7 @@ ret_from_syscall: MTMSRD(r10) lwz r9,TI_FLAGS(r12) li r8,-_LAST_ERRNO - andi. r0,r9,(_TIF_SYSCALL_T_OR_A|_TIF_SIGPENDING|_TIF_NEED_RESCHED|_TIF_RESTOREALL) + andi. r0,r9,(_TIF_SYSCALL_T_OR_A|_TIF_SIGPENDING|_TIF_NEED_RESCHED|_TIF_RESTOREALL|_TIF_RESTORE_SIGMASK) bne- syscall_exit_work cmplw 0,r3,r8 blt+ syscall_exit_cont @@ -357,7 +357,7 @@ save_user_nvgprs_cont: lwz r5,_MSR(r1) andi. r5,r5,MSR_PR beq ret_from_except - andi. r0,r9,_TIF_SIGPENDING + andi. r0,r9,_TIF_SIGPENDING|_TIF_RESTORE_SIGMASK beq ret_from_except b do_user_signal 8: @@ -683,7 +683,7 @@ user_exc_return: /* r10 contains MSR_KERNEL here */ /* Check current_thread_info()->flags */ rlwinm r9,r1,0,0,(31-THREAD_SHIFT) lwz r9,TI_FLAGS(r9) - andi. r0,r9,(_TIF_SIGPENDING|_TIF_NEED_RESCHED|_TIF_RESTOREALL) + andi. r0,r9,(_TIF_SIGPENDING|_TIF_NEED_RESCHED|_TIF_RESTOREALL|_TIF_RESTORE_SIGMASK) bne do_work restore_user: @@ -917,7 +917,7 @@ recheck: lwz r9,TI_FLAGS(r9) andi. r0,r9,_TIF_NEED_RESCHED bne- do_resched - andi. r0,r9,_TIF_SIGPENDING + andi. r0,r9,_TIF_SIGPENDING|_TIF_RESTORE_SIGMASK beq restore_user do_user_signal: /* r10 contains MSR_KERNEL here */ ori r10,r10,MSR_EE diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S index 54203631886..388f861b8ed 100644 --- a/arch/powerpc/kernel/entry_64.S +++ b/arch/powerpc/kernel/entry_64.S @@ -160,7 +160,7 @@ syscall_exit: mtmsrd r10,1 ld r9,TI_FLAGS(r12) li r11,-_LAST_ERRNO - andi. r0,r9,(_TIF_SYSCALL_T_OR_A|_TIF_SINGLESTEP|_TIF_SIGPENDING|_TIF_NEED_RESCHED|_TIF_RESTOREALL|_TIF_SAVE_NVGPRS|_TIF_NOERROR) + andi. r0,r9,(_TIF_SYSCALL_T_OR_A|_TIF_SINGLESTEP|_TIF_SIGPENDING|_TIF_NEED_RESCHED|_TIF_RESTOREALL|_TIF_SAVE_NVGPRS|_TIF_NOERROR|_TIF_RESTORE_SIGMASK) bne- syscall_exit_work cmpld r3,r11 ld r5,_CCR(r1) diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c index 7f0d5ce2567..3747ab0dac3 100644 --- a/arch/powerpc/kernel/signal_32.c +++ b/arch/powerpc/kernel/signal_32.c @@ -1128,7 +1128,7 @@ int do_signal(sigset_t *oldset, struct pt_regs *regs) { siginfo_t info; struct k_sigaction ka; - unsigned int frame, newsp; + unsigned int newsp; int signr, ret; #ifdef CONFIG_PPC32 @@ -1139,11 +1139,11 @@ int do_signal(sigset_t *oldset, struct pt_regs *regs) } #endif - if (!oldset) + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + oldset = ¤t->saved_sigmask; + else if (!oldset) oldset = ¤t->blocked; - newsp = frame = 0; - signr = get_signal_to_deliver(&info, &ka, regs, NULL); #ifdef CONFIG_PPC32 no_signal: @@ -1173,8 +1173,14 @@ no_signal: } } - if (signr == 0) + if (signr == 0) { + /* No signal to deliver -- put the saved sigmask back */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) { + clear_thread_flag(TIF_RESTORE_SIGMASK); + sigprocmask(SIG_SETMASK, ¤t->saved_sigmask, NULL); + } return 0; /* no signals delivered */ + } if ((ka.sa.sa_flags & SA_ONSTACK) && current->sas_ss_size && !on_sig_stack(regs->gpr[1])) @@ -1207,6 +1213,10 @@ no_signal: sigaddset(¤t->blocked, signr); recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); + /* A signal was successfully delivered; the saved sigmask is in + its frame, and we can clear the TIF_RESTORE_SIGMASK flag */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + clear_thread_flag(TIF_RESTORE_SIGMASK); } return ret; diff --git a/arch/powerpc/kernel/signal_64.c b/arch/powerpc/kernel/signal_64.c index a4a6812e815..b3193116e68 100644 --- a/arch/powerpc/kernel/signal_64.c +++ b/arch/powerpc/kernel/signal_64.c @@ -520,11 +520,15 @@ int do_signal(sigset_t *oldset, struct pt_regs *regs) if (test_thread_flag(TIF_32BIT)) return do_signal32(oldset, regs); - if (!oldset) + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + oldset = ¤t->saved_sigmask; + else if (!oldset) oldset = ¤t->blocked; signr = get_signal_to_deliver(&info, &ka, regs, NULL); if (signr > 0) { + int ret; + /* Whee! Actually deliver the signal. */ if (TRAP(regs) == 0x0C00) syscall_restart(regs, &ka); @@ -537,7 +541,14 @@ int do_signal(sigset_t *oldset, struct pt_regs *regs) if (current->thread.dabr) set_dabr(current->thread.dabr); - return handle_signal(signr, &ka, &info, oldset, regs); + ret = handle_signal(signr, &ka, &info, oldset, regs); + + /* If a signal was successfully delivered, the saved sigmask is in + its frame, and we can clear the TIF_RESTORE_SIGMASK flag */ + if (ret && test_thread_flag(TIF_RESTORE_SIGMASK)) + clear_thread_flag(TIF_RESTORE_SIGMASK); + + return ret; } if (TRAP(regs) == 0x0C00) { /* System Call! */ @@ -553,6 +564,11 @@ int do_signal(sigset_t *oldset, struct pt_regs *regs) regs->result = 0; } } + /* No signal to deliver -- put the saved sigmask back */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) { + clear_thread_flag(TIF_RESTORE_SIGMASK); + sigprocmask(SIG_SETMASK, ¤t->saved_sigmask, NULL); + } return 0; } diff --git a/arch/powerpc/kernel/systbl.S b/arch/powerpc/kernel/systbl.S index 68013179a50..007b15ee36d 100644 --- a/arch/powerpc/kernel/systbl.S +++ b/arch/powerpc/kernel/systbl.S @@ -321,3 +321,5 @@ SYSCALL(inotify_add_watch) SYSCALL(inotify_rm_watch) SYSCALL(spu_run) SYSCALL(spu_create) +COMPAT_SYS(pselect6) +COMPAT_SYS(ppoll) diff --git a/include/asm-powerpc/thread_info.h b/include/asm-powerpc/thread_info.h index 7e09d7cda93..67cdaf3ae9f 100644 --- a/include/asm-powerpc/thread_info.h +++ b/include/asm-powerpc/thread_info.h @@ -122,6 +122,7 @@ static inline struct thread_info *current_thread_info(void) #define TIF_RESTOREALL 12 /* Restore all regs (implies NOERROR) */ #define TIF_SAVE_NVGPRS 13 /* Save r14-r31 in signal frame */ #define TIF_NOERROR 14 /* Force successful syscall return */ +#define TIF_RESTORE_SIGMASK 15 /* Restore signal mask in do_signal */ /* as above, but as bit values */ #define _TIF_SYSCALL_TRACE (1< Date: Wed, 18 Jan 2006 17:44:02 -0800 Subject: [PATCH] uml: add TIF_RESTORE_SIGMASK support Add support for TIF_RESTORE_SIGMASK. I copy the i386 handling of the flag. sys_sigsuspend is also changed to follow i386. Also a bit of cleanup - turn an if into a switch get rid of a couple more emacs formatting comments Signed-off-by: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/kernel/signal_kern.c | 79 +++++++++++++++++++++++--------------------- include/asm-um/thread_info.h | 13 ++------ 2 files changed, 43 insertions(+), 49 deletions(-) diff --git a/arch/um/kernel/signal_kern.c b/arch/um/kernel/signal_kern.c index 7b0e0e81c16..7a54708db50 100644 --- a/arch/um/kernel/signal_kern.c +++ b/arch/um/kernel/signal_kern.c @@ -99,31 +99,46 @@ static int handle_signal(struct pt_regs *regs, unsigned long signr, return err; } -static int kern_do_signal(struct pt_regs *regs, sigset_t *oldset) +static int kern_do_signal(struct pt_regs *regs) { struct k_sigaction ka_copy; siginfo_t info; + sigset_t *oldset; int sig, handled_sig = 0; + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + oldset = ¤t->saved_sigmask; + else + oldset = ¤t->blocked; + while((sig = get_signal_to_deliver(&info, &ka_copy, regs, NULL)) > 0){ handled_sig = 1; /* Whee! Actually deliver the signal. */ - if(!handle_signal(regs, sig, &ka_copy, &info, oldset)) + if(!handle_signal(regs, sig, &ka_copy, &info, oldset)){ + /* a signal was successfully delivered; the saved + * sigmask will have been stored in the signal frame, + * and will be restored by sigreturn, so we can simply + * clear the TIF_RESTORE_SIGMASK flag */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + clear_thread_flag(TIF_RESTORE_SIGMASK); break; + } } /* Did we come from a system call? */ if(!handled_sig && (PT_REGS_SYSCALL_NR(regs) >= 0)){ /* Restart the system call - no handlers present */ - if(PT_REGS_SYSCALL_RET(regs) == -ERESTARTNOHAND || - PT_REGS_SYSCALL_RET(regs) == -ERESTARTSYS || - PT_REGS_SYSCALL_RET(regs) == -ERESTARTNOINTR){ + switch(PT_REGS_SYSCALL_RET(regs)){ + case -ERESTARTNOHAND: + case -ERESTARTSYS: + case -ERESTARTNOINTR: PT_REGS_ORIG_SYSCALL(regs) = PT_REGS_SYSCALL_NR(regs); PT_REGS_RESTART_SYSCALL(regs); - } - else if(PT_REGS_SYSCALL_RET(regs) == -ERESTART_RESTARTBLOCK){ + break; + case -ERESTART_RESTARTBLOCK: PT_REGS_SYSCALL_RET(regs) = __NR_restart_syscall; PT_REGS_RESTART_SYSCALL(regs); + break; } } @@ -137,12 +152,19 @@ static int kern_do_signal(struct pt_regs *regs, sigset_t *oldset) if(current->ptrace & PT_DTRACE) current->thread.singlestep_syscall = is_syscall(PT_REGS_IP(¤t->thread.regs)); + + /* if there's no signal to deliver, we just put the saved sigmask + * back */ + if (!handled_sig && test_thread_flag(TIF_RESTORE_SIGMASK)) { + clear_thread_flag(TIF_RESTORE_SIGMASK); + sigprocmask(SIG_SETMASK, ¤t->saved_sigmask, NULL); + } return(handled_sig); } int do_signal(void) { - return(kern_do_signal(¤t->thread.regs, ¤t->blocked)); + return(kern_do_signal(¤t->thread.regs)); } /* @@ -150,27 +172,22 @@ int do_signal(void) */ long sys_sigsuspend(int history0, int history1, old_sigset_t mask) { - sigset_t saveset; - mask &= _BLOCKABLE; spin_lock_irq(¤t->sighand->siglock); - saveset = current->blocked; + current->saved_sigmask = current->blocked; siginitset(¤t->blocked, mask); recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); - PT_REGS_SYSCALL_RET(¤t->thread.regs) = -EINTR; - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - if(kern_do_signal(¤t->thread.regs, &saveset)) - return(-EINTR); - } + current->state = TASK_INTERRUPTIBLE; + schedule(); + set_thread_flag(TIF_RESTORE_SIGMASK); + return -ERESTARTNOHAND; } long sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize) { - sigset_t saveset, newset; + sigset_t newset; /* XXX: Don't preclude handling different sized sigset_t's. */ if (sigsetsize != sizeof(sigset_t)) @@ -181,32 +198,18 @@ long sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize) sigdelsetmask(&newset, ~_BLOCKABLE); spin_lock_irq(¤t->sighand->siglock); - saveset = current->blocked; + current->saved_sigmask = current->blocked; current->blocked = newset; recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); - PT_REGS_SYSCALL_RET(¤t->thread.regs) = -EINTR; - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - if (kern_do_signal(¤t->thread.regs, &saveset)) - return(-EINTR); - } + current->state = TASK_INTERRUPTIBLE; + schedule(); + set_thread_flag(TIF_RESTORE_SIGMASK); + return -ERESTARTNOHAND; } long sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss) { return(do_sigaltstack(uss, uoss, PT_REGS_SP(¤t->thread.regs))); } - -/* - * 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/include/asm-um/thread_info.h b/include/asm-um/thread_info.h index 705c71972c3..17b6b07c433 100644 --- a/include/asm-um/thread_info.h +++ b/include/asm-um/thread_info.h @@ -69,6 +69,7 @@ static inline struct thread_info *current_thread_info(void) #define TIF_RESTART_BLOCK 4 #define TIF_MEMDIE 5 #define TIF_SYSCALL_AUDIT 6 +#define TIF_RESTORE_SIGMASK 7 #define _TIF_SYSCALL_TRACE (1 << TIF_SYSCALL_TRACE) #define _TIF_SIGPENDING (1 << TIF_SIGPENDING) @@ -76,16 +77,6 @@ static inline struct thread_info *current_thread_info(void) #define _TIF_POLLING_NRFLAG (1 << TIF_POLLING_NRFLAG) #define _TIF_MEMDIE (1 << TIF_MEMDIE) #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) +#define _TIF_RESTORE_SIGMASK (1 << TIF_RESTORE_SIGMASK) #endif - -/* - * 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: - */ -- cgit v1.2.3 From 36a7878a224c18aa4a5e098dc93d19cf5601462b Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Wed, 18 Jan 2006 17:44:03 -0800 Subject: [PATCH] uml: use generic sys_rt_sigsuspend Use the generic sys_rt_sigsuspend. Signed-off-by: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/kernel/signal_kern.c | 24 ------------------------ include/asm-um/unistd.h | 1 + 2 files changed, 1 insertion(+), 24 deletions(-) diff --git a/arch/um/kernel/signal_kern.c b/arch/um/kernel/signal_kern.c index 7a54708db50..da17b7541e0 100644 --- a/arch/um/kernel/signal_kern.c +++ b/arch/um/kernel/signal_kern.c @@ -185,30 +185,6 @@ long sys_sigsuspend(int history0, int history1, old_sigset_t mask) return -ERESTARTNOHAND; } -long sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize) -{ - sigset_t newset; - - /* XXX: Don't preclude handling different sized sigset_t's. */ - if (sigsetsize != sizeof(sigset_t)) - return -EINVAL; - - if (copy_from_user(&newset, unewset, sizeof(newset))) - return -EFAULT; - sigdelsetmask(&newset, ~_BLOCKABLE); - - spin_lock_irq(¤t->sighand->siglock); - current->saved_sigmask = current->blocked; - current->blocked = newset; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - current->state = TASK_INTERRUPTIBLE; - schedule(); - set_thread_flag(TIF_RESTORE_SIGMASK); - return -ERESTARTNOHAND; -} - long sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss) { return(do_sigaltstack(uss, uoss, PT_REGS_SP(¤t->thread.regs))); diff --git a/include/asm-um/unistd.h b/include/asm-um/unistd.h index 6fdde45cc05..afccfcaa9ea 100644 --- a/include/asm-um/unistd.h +++ b/include/asm-um/unistd.h @@ -34,6 +34,7 @@ extern int um_execve(const char *file, char *const argv[], char *const env[]); #define __ARCH_WANT_SYS_SIGPENDING #define __ARCH_WANT_SYS_SIGPROCMASK #define __ARCH_WANT_SYS_RT_SIGACTION +#define __ARCH_WANT_SYS_RT_SIGSUSPEND #endif #ifdef __KERNEL_SYSCALLS__ -- cgit v1.2.3 From 9f72949f679df06021c9e43886c9191494fdb007 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 18 Jan 2006 17:44:05 -0800 Subject: [PATCH] Add pselect/ppoll system call implementation The following implementation of ppoll() and pselect() system calls depends on the architecture providing a TIF_RESTORE_SIGMASK flag in the thread_info. These system calls have to change the signal mask during their operation, and signal handlers must be invoked using the new, temporary signal mask. The old signal mask must be restored either upon successful exit from the system call, or upon returning from the invoked signal handler if the system call is interrupted. We can't simply restore the original signal mask and return to userspace, since the restored signal mask may actually block the signal which interrupted the system call. The TIF_RESTORE_SIGMASK flag deals with this by causing the syscall exit path to trap into do_signal() just as TIF_SIGPENDING does, and by causing do_signal() to use the saved signal mask instead of the current signal mask when setting up the stack frame for the signal handler -- or by causing do_signal() to simply restore the saved signal mask in the case where there is no handler to be invoked. The first patch implements the sys_pselect() and sys_ppoll() system calls, which are present only if TIF_RESTORE_SIGMASK is defined. That #ifdef should go away in time when all architectures have implemented it. The second patch implements TIF_RESTORE_SIGMASK for the PowerPC kernel (in the -mm tree), and the third patch then removes the arch-specific implementations of sys_rt_sigsuspend() and replaces them with generic versions using the same trick. The fourth and fifth patches, provided by David Howells, implement TIF_RESTORE_SIGMASK for FR-V and i386 respectively, and the sixth patch adds the syscalls to the i386 syscall table. This patch: Add the pselect() and ppoll() system calls, providing core routines usable by the original select() and poll() system calls and also the new calls (with their semantics w.r.t timeouts). Signed-off-by: David Woodhouse Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/compat.c | 260 ++++++++++++++++++++++++++++++++------ fs/select.c | 348 ++++++++++++++++++++++++++++++++++++++++++--------- include/linux/poll.h | 6 +- 3 files changed, 519 insertions(+), 95 deletions(-) diff --git a/fs/compat.c b/fs/compat.c index c6ba9deabad..18b21b4c9e3 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -53,6 +53,8 @@ #include #include +extern void sigset_from_compat(sigset_t *set, compat_sigset_t *compat); + /* * Not all architectures have sys_utime, so implement this in terms * of sys_utimes. @@ -1657,36 +1659,14 @@ static void select_bits_free(void *bits, int size) #define MAX_SELECT_SECONDS \ ((unsigned long) (MAX_SCHEDULE_TIMEOUT / HZ)-1) -asmlinkage long -compat_sys_select(int n, compat_ulong_t __user *inp, compat_ulong_t __user *outp, - compat_ulong_t __user *exp, struct compat_timeval __user *tvp) +int compat_core_sys_select(int n, compat_ulong_t __user *inp, + compat_ulong_t __user *outp, compat_ulong_t __user *exp, s64 *timeout) { fd_set_bits fds; char *bits; - long timeout; int size, max_fdset, ret = -EINVAL; struct fdtable *fdt; - timeout = MAX_SCHEDULE_TIMEOUT; - if (tvp) { - time_t sec, usec; - - if (!access_ok(VERIFY_READ, tvp, sizeof(*tvp)) - || __get_user(sec, &tvp->tv_sec) - || __get_user(usec, &tvp->tv_usec)) { - ret = -EFAULT; - goto out_nofds; - } - - if (sec < 0 || usec < 0) - goto out_nofds; - - if ((unsigned long) sec < MAX_SELECT_SECONDS) { - timeout = ROUND_UP(usec, 1000000/HZ); - timeout += sec * (unsigned long) HZ; - } - } - if (n < 0) goto out_nofds; @@ -1723,19 +1703,7 @@ compat_sys_select(int n, compat_ulong_t __user *inp, compat_ulong_t __user *outp zero_fd_set(n, fds.res_out); zero_fd_set(n, fds.res_ex); - ret = do_select(n, &fds, &timeout); - - if (tvp && !(current->personality & STICKY_TIMEOUTS)) { - time_t sec = 0, usec = 0; - if (timeout) { - sec = timeout / HZ; - usec = timeout % HZ; - usec *= (1000000/HZ); - } - if (put_user(sec, &tvp->tv_sec) || - put_user(usec, &tvp->tv_usec)) - ret = -EFAULT; - } + ret = do_select(n, &fds, timeout); if (ret < 0) goto out; @@ -1756,6 +1724,224 @@ out_nofds: return ret; } +asmlinkage long compat_sys_select(int n, compat_ulong_t __user *inp, + compat_ulong_t __user *outp, compat_ulong_t __user *exp, + struct compat_timeval __user *tvp) +{ + s64 timeout = -1; + struct compat_timeval tv; + int ret; + + if (tvp) { + if (copy_from_user(&tv, tvp, sizeof(tv))) + return -EFAULT; + + if (tv.tv_sec < 0 || tv.tv_usec < 0) + return -EINVAL; + + /* Cast to u64 to make GCC stop complaining */ + if ((u64)tv.tv_sec >= (u64)MAX_INT64_SECONDS) + timeout = -1; /* infinite */ + else { + timeout = ROUND_UP(tv.tv_sec, 1000000/HZ); + timeout += tv.tv_sec * HZ; + } + } + + ret = compat_core_sys_select(n, inp, outp, exp, &timeout); + + if (tvp) { + if (current->personality & STICKY_TIMEOUTS) + goto sticky; + tv.tv_usec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)); + tv.tv_sec = timeout; + if (copy_to_user(tvp, &tv, sizeof(tv))) { +sticky: + /* + * If an application puts its timeval in read-only + * memory, we don't want the Linux-specific update to + * the timeval to cause a fault after the select has + * completed successfully. However, because we're not + * updating the timeval, we can't restart the system + * call. + */ + if (ret == -ERESTARTNOHAND) + ret = -EINTR; + } + } + + return ret; +} + +#ifdef TIF_RESTORE_SIGMASK +asmlinkage long compat_sys_pselect7(int n, compat_ulong_t __user *inp, + compat_ulong_t __user *outp, compat_ulong_t __user *exp, + struct compat_timespec __user *tsp, compat_sigset_t __user *sigmask, + compat_size_t sigsetsize) +{ + compat_sigset_t ss32; + sigset_t ksigmask, sigsaved; + long timeout = MAX_SCHEDULE_TIMEOUT; + struct compat_timespec ts; + int ret; + + if (tsp) { + if (copy_from_user(&ts, tsp, sizeof(ts))) + return -EFAULT; + + if (ts.tv_sec < 0 || ts.tv_nsec < 0) + return -EINVAL; + } + + if (sigmask) { + if (sigsetsize != sizeof(compat_sigset_t)) + return -EINVAL; + if (copy_from_user(&ss32, sigmask, sizeof(ss32))) + return -EFAULT; + sigset_from_compat(&ksigmask, &ss32); + + sigdelsetmask(&ksigmask, sigmask(SIGKILL)|sigmask(SIGSTOP)); + sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); + } + + do { + if (tsp) { + if ((unsigned long)ts.tv_sec < MAX_SELECT_SECONDS) { + timeout = ROUND_UP(ts.tv_nsec, 1000000000/HZ); + timeout += ts.tv_sec * (unsigned long)HZ; + ts.tv_sec = 0; + ts.tv_nsec = 0; + } else { + ts.tv_sec -= MAX_SELECT_SECONDS; + timeout = MAX_SELECT_SECONDS * HZ; + } + } + + ret = compat_core_sys_select(n, inp, outp, exp, &timeout); + + } while (!ret && !timeout && tsp && (ts.tv_sec || ts.tv_nsec)); + + if (tsp && !(current->personality & STICKY_TIMEOUTS)) { + ts.tv_sec += timeout / HZ; + ts.tv_nsec += (timeout % HZ) * (1000000000/HZ); + if (ts.tv_nsec >= 1000000000) { + ts.tv_sec++; + ts.tv_nsec -= 1000000000; + } + (void)copy_to_user(tsp, &ts, sizeof(ts)); + } + + if (ret == -ERESTARTNOHAND) { + /* + * Don't restore the signal mask yet. Let do_signal() deliver + * the signal on the way back to userspace, before the signal + * mask is restored. + */ + if (sigmask) { + memcpy(¤t->saved_sigmask, &sigsaved, + sizeof(sigsaved)); + set_thread_flag(TIF_RESTORE_SIGMASK); + } + } else if (sigmask) + sigprocmask(SIG_SETMASK, &sigsaved, NULL); + + return ret; +} + +asmlinkage long compat_sys_pselect6(int n, compat_ulong_t __user *inp, + compat_ulong_t __user *outp, compat_ulong_t __user *exp, + struct compat_timespec __user *tsp, void __user *sig) +{ + compat_size_t sigsetsize = 0; + compat_uptr_t up = 0; + + if (sig) { + if (!access_ok(VERIFY_READ, sig, + sizeof(compat_uptr_t)+sizeof(compat_size_t)) || + __get_user(up, (compat_uptr_t __user *)sig) || + __get_user(sigsetsize, + (compat_size_t __user *)(sig+sizeof(up)))) + return -EFAULT; + } + return compat_sys_pselect7(n, inp, outp, exp, tsp, compat_ptr(up), + sigsetsize); +} + +asmlinkage long compat_sys_ppoll(struct pollfd __user *ufds, + unsigned int nfds, struct compat_timespec __user *tsp, + const compat_sigset_t __user *sigmask, compat_size_t sigsetsize) +{ + compat_sigset_t ss32; + sigset_t ksigmask, sigsaved; + struct compat_timespec ts; + s64 timeout = -1; + int ret; + + if (tsp) { + if (copy_from_user(&ts, tsp, sizeof(ts))) + return -EFAULT; + + /* We assume that ts.tv_sec is always lower than + the number of seconds that can be expressed in + an s64. Otherwise the compiler bitches at us */ + timeout = ROUND_UP(ts.tv_sec, 1000000000/HZ); + timeout += ts.tv_sec * HZ; + } + + if (sigmask) { + if (sigsetsize |= sizeof(compat_sigset_t)) + return -EINVAL; + if (copy_from_user(&ss32, sigmask, sizeof(ss32))) + return -EFAULT; + sigset_from_compat(&ksigmask, &ss32); + + sigdelsetmask(&ksigmask, sigmask(SIGKILL)|sigmask(SIGSTOP)); + sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); + } + + ret = do_sys_poll(ufds, nfds, &timeout); + + /* We can restart this syscall, usually */ + if (ret == -EINTR) { + /* + * Don't restore the signal mask yet. Let do_signal() deliver + * the signal on the way back to userspace, before the signal + * mask is restored. + */ + if (sigmask) { + memcpy(¤t->saved_sigmask, &sigsaved, + sizeof(sigsaved)); + set_thread_flag(TIF_RESTORE_SIGMASK); + } + ret = -ERESTARTNOHAND; + } else if (sigmask) + sigprocmask(SIG_SETMASK, &sigsaved, NULL); + + if (tsp && timeout >= 0) { + if (current->personality & STICKY_TIMEOUTS) + goto sticky; + /* Yes, we know it's actually an s64, but it's also positive. */ + ts.tv_nsec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)) * 1000; + ts.tv_sec = timeout; + if (copy_to_user(tsp, &ts, sizeof(ts))) { +sticky: + /* + * If an application puts its timeval in read-only + * memory, we don't want the Linux-specific update to + * the timeval to cause a fault after the select has + * completed successfully. However, because we're not + * updating the timeval, we can't restart the system + * call. + */ + if (ret == -ERESTARTNOHAND && timeout >= 0) + ret = -EINTR; + } + } + + return ret; +} +#endif /* TIF_RESTORE_SIGMASK */ + #if defined(CONFIG_NFSD) || defined(CONFIG_NFSD_MODULE) /* Stuff for NFS server syscalls... */ struct compat_nfsctl_svc { diff --git a/fs/select.c b/fs/select.c index f10a10317d5..c0f02d36c60 100644 --- a/fs/select.c +++ b/fs/select.c @@ -179,12 +179,11 @@ get_max: #define POLLOUT_SET (POLLWRBAND | POLLWRNORM | POLLOUT | POLLERR) #define POLLEX_SET (POLLPRI) -int do_select(int n, fd_set_bits *fds, long *timeout) +int do_select(int n, fd_set_bits *fds, s64 *timeout) { struct poll_wqueues table; poll_table *wait; int retval, i; - long __timeout = *timeout; rcu_read_lock(); retval = max_select_fd(n, fds); @@ -196,11 +195,12 @@ int do_select(int n, fd_set_bits *fds, long *timeout) poll_initwait(&table); wait = &table.pt; - if (!__timeout) + if (!*timeout) wait = NULL; retval = 0; for (;;) { unsigned long *rinp, *routp, *rexp, *inp, *outp, *exp; + long __timeout; set_current_state(TASK_INTERRUPTIBLE); @@ -255,22 +255,32 @@ int do_select(int n, fd_set_bits *fds, long *timeout) *rexp = res_ex; } wait = NULL; - if (retval || !__timeout || signal_pending(current)) + if (retval || !*timeout || signal_pending(current)) break; if(table.error) { retval = table.error; break; } + + if (*timeout < 0) { + /* Wait indefinitely */ + __timeout = MAX_SCHEDULE_TIMEOUT; + } else if (unlikely(*timeout >= (s64)MAX_SCHEDULE_TIMEOUT - 1)) { + /* Wait for longer than MAX_SCHEDULE_TIMEOUT. Do it in a loop */ + __timeout = MAX_SCHEDULE_TIMEOUT - 1; + *timeout -= __timeout; + } else { + __timeout = *timeout; + *timeout = 0; + } __timeout = schedule_timeout(__timeout); + if (*timeout >= 0) + *timeout += __timeout; } __set_current_state(TASK_RUNNING); poll_freewait(&table); - /* - * Up-to-date the caller timeout. - */ - *timeout = __timeout; return retval; } @@ -295,36 +305,14 @@ static void select_bits_free(void *bits, int size) #define MAX_SELECT_SECONDS \ ((unsigned long) (MAX_SCHEDULE_TIMEOUT / HZ)-1) -asmlinkage long -sys_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, struct timeval __user *tvp) +static int core_sys_select(int n, fd_set __user *inp, fd_set __user *outp, + fd_set __user *exp, s64 *timeout) { fd_set_bits fds; char *bits; - long timeout; int ret, size, max_fdset; struct fdtable *fdt; - timeout = MAX_SCHEDULE_TIMEOUT; - if (tvp) { - time_t sec, usec; - - if (!access_ok(VERIFY_READ, tvp, sizeof(*tvp)) - || __get_user(sec, &tvp->tv_sec) - || __get_user(usec, &tvp->tv_usec)) { - ret = -EFAULT; - goto out_nofds; - } - - ret = -EINVAL; - if (sec < 0 || usec < 0) - goto out_nofds; - - if ((unsigned long) sec < MAX_SELECT_SECONDS) { - timeout = ROUND_UP(usec, 1000000/HZ); - timeout += sec * (unsigned long) HZ; - } - } - ret = -EINVAL; if (n < 0) goto out_nofds; @@ -362,18 +350,7 @@ sys_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, s zero_fd_set(n, fds.res_out); zero_fd_set(n, fds.res_ex); - ret = do_select(n, &fds, &timeout); - - if (tvp && !(current->personality & STICKY_TIMEOUTS)) { - time_t sec = 0, usec = 0; - if (timeout) { - sec = timeout / HZ; - usec = timeout % HZ; - usec *= (1000000/HZ); - } - put_user(sec, &tvp->tv_sec); - put_user(usec, &tvp->tv_usec); - } + ret = do_select(n, &fds, timeout); if (ret < 0) goto out; @@ -395,6 +372,154 @@ out_nofds: return ret; } +asmlinkage long sys_select(int n, fd_set __user *inp, fd_set __user *outp, + fd_set __user *exp, struct timeval __user *tvp) +{ + s64 timeout = -1; + struct timeval tv; + int ret; + + if (tvp) { + if (copy_from_user(&tv, tvp, sizeof(tv))) + return -EFAULT; + + if (tv.tv_sec < 0 || tv.tv_usec < 0) + return -EINVAL; + + /* Cast to u64 to make GCC stop complaining */ + if ((u64)tv.tv_sec >= (u64)MAX_INT64_SECONDS) + timeout = -1; /* infinite */ + else { + timeout = ROUND_UP(tv.tv_usec, USEC_PER_SEC/HZ); + timeout += tv.tv_sec * HZ; + } + } + + ret = core_sys_select(n, inp, outp, exp, &timeout); + + if (tvp) { + if (current->personality & STICKY_TIMEOUTS) + goto sticky; + tv.tv_usec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)); + tv.tv_sec = timeout; + if (copy_to_user(tvp, &tv, sizeof(tv))) { +sticky: + /* + * If an application puts its timeval in read-only + * memory, we don't want the Linux-specific update to + * the timeval to cause a fault after the select has + * completed successfully. However, because we're not + * updating the timeval, we can't restart the system + * call. + */ + if (ret == -ERESTARTNOHAND) + ret = -EINTR; + } + } + + return ret; +} + +#ifdef TIF_RESTORE_SIGMASK +asmlinkage long sys_pselect7(int n, fd_set __user *inp, fd_set __user *outp, + fd_set __user *exp, struct timespec __user *tsp, + const sigset_t __user *sigmask, size_t sigsetsize) +{ + s64 timeout = MAX_SCHEDULE_TIMEOUT; + sigset_t ksigmask, sigsaved; + struct timespec ts; + int ret; + + if (tsp) { + if (copy_from_user(&ts, tsp, sizeof(ts))) + return -EFAULT; + + if (ts.tv_sec < 0 || ts.tv_nsec < 0) + return -EINVAL; + + /* Cast to u64 to make GCC stop complaining */ + if ((u64)ts.tv_sec >= (u64)MAX_INT64_SECONDS) + timeout = -1; /* infinite */ + else { + timeout = ROUND_UP(ts.tv_nsec, NSEC_PER_SEC/HZ); + timeout += ts.tv_sec * HZ; + } + } + + if (sigmask) { + /* XXX: Don't preclude handling different sized sigset_t's. */ + if (sigsetsize != sizeof(sigset_t)) + return -EINVAL; + if (copy_from_user(&ksigmask, sigmask, sizeof(ksigmask))) + return -EFAULT; + + sigdelsetmask(&ksigmask, sigmask(SIGKILL)|sigmask(SIGSTOP)); + sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); + } + + ret = core_sys_select(n, inp, outp, exp, &timeout); + + if (tsp) { + if (current->personality & STICKY_TIMEOUTS) + goto sticky; + ts.tv_nsec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)) * 1000; + ts.tv_sec = timeout; + if (copy_to_user(tsp, &ts, sizeof(ts))) { +sticky: + /* + * If an application puts its timeval in read-only + * memory, we don't want the Linux-specific update to + * the timeval to cause a fault after the select has + * completed successfully. However, because we're not + * updating the timeval, we can't restart the system + * call. + */ + if (ret == -ERESTARTNOHAND) + ret = -EINTR; + } + } + + if (ret == -ERESTARTNOHAND) { + /* + * Don't restore the signal mask yet. Let do_signal() deliver + * the signal on the way back to userspace, before the signal + * mask is restored. + */ + if (sigmask) { + memcpy(¤t->saved_sigmask, &sigsaved, + sizeof(sigsaved)); + set_thread_flag(TIF_RESTORE_SIGMASK); + } + } else if (sigmask) + sigprocmask(SIG_SETMASK, &sigsaved, NULL); + + return ret; +} + +/* + * Most architectures can't handle 7-argument syscalls. So we provide a + * 6-argument version where the sixth argument is a pointer to a structure + * which has a pointer to the sigset_t itself followed by a size_t containing + * the sigset size. + */ +asmlinkage long sys_pselect6(int n, fd_set __user *inp, fd_set __user *outp, + fd_set __user *exp, struct timespec __user *tsp, void __user *sig) +{ + size_t sigsetsize = 0; + sigset_t __user *up = NULL; + + if (sig) { + if (!access_ok(VERIFY_READ, sig, sizeof(void *)+sizeof(size_t)) + || __get_user(up, (sigset_t * __user *)sig) + || __get_user(sigsetsize, + (size_t * __user)(sig+sizeof(void *)))) + return -EFAULT; + } + + return sys_pselect7(n, inp, outp, exp, tsp, up, sigsetsize); +} +#endif /* TIF_RESTORE_SIGMASK */ + struct poll_list { struct poll_list *next; int len; @@ -436,16 +561,19 @@ static void do_pollfd(unsigned int num, struct pollfd * fdpage, } static int do_poll(unsigned int nfds, struct poll_list *list, - struct poll_wqueues *wait, long timeout) + struct poll_wqueues *wait, s64 *timeout) { int count = 0; poll_table* pt = &wait->pt; - if (!timeout) + /* Optimise the no-wait case */ + if (!(*timeout)) pt = NULL; for (;;) { struct poll_list *walk; + long __timeout; + set_current_state(TASK_INTERRUPTIBLE); walk = list; while(walk != NULL) { @@ -453,18 +581,36 @@ static int do_poll(unsigned int nfds, struct poll_list *list, walk = walk->next; } pt = NULL; - if (count || !timeout || signal_pending(current)) + if (count || !*timeout || signal_pending(current)) break; count = wait->error; if (count) break; - timeout = schedule_timeout(timeout); + + if (*timeout < 0) { + /* Wait indefinitely */ + __timeout = MAX_SCHEDULE_TIMEOUT; + } else if (unlikely(*timeout >= (s64)MAX_SCHEDULE_TIMEOUT-1)) { + /* + * Wait for longer than MAX_SCHEDULE_TIMEOUT. Do it in + * a loop + */ + __timeout = MAX_SCHEDULE_TIMEOUT - 1; + *timeout -= __timeout; + } else { + __timeout = *timeout; + *timeout = 0; + } + + __timeout = schedule_timeout(__timeout); + if (*timeout >= 0) + *timeout += __timeout; } __set_current_state(TASK_RUNNING); return count; } -asmlinkage long sys_poll(struct pollfd __user * ufds, unsigned int nfds, long timeout) +int do_sys_poll(struct pollfd __user *ufds, unsigned int nfds, s64 *timeout) { struct poll_wqueues table; int fdcount, err; @@ -482,14 +628,6 @@ asmlinkage long sys_poll(struct pollfd __user * ufds, unsigned int nfds, long ti if (nfds > max_fdset && nfds > OPEN_MAX) return -EINVAL; - if (timeout) { - /* Careful about overflow in the intermediate values */ - if ((unsigned long) timeout < MAX_SCHEDULE_TIMEOUT / HZ) - timeout = (unsigned long)(timeout*HZ+999)/1000+1; - else /* Negative or overflow */ - timeout = MAX_SCHEDULE_TIMEOUT; - } - poll_initwait(&table); head = NULL; @@ -519,6 +657,7 @@ asmlinkage long sys_poll(struct pollfd __user * ufds, unsigned int nfds, long ti } i -= pp->len; } + fdcount = do_poll(nfds, head, &table, timeout); /* OK, now copy the revents fields back to user space. */ @@ -547,3 +686,98 @@ out_fds: poll_freewait(&table); return err; } + +asmlinkage long sys_poll(struct pollfd __user *ufds, unsigned int nfds, + long timeout_msecs) +{ + s64 timeout_jiffies = 0; + + if (timeout_msecs) { +#if HZ > 1000 + /* We can only overflow if HZ > 1000 */ + if (timeout_msecs / 1000 > (s64)0x7fffffffffffffffULL / (s64)HZ) + timeout_jiffies = -1; + else +#endif + timeout_jiffies = msecs_to_jiffies(timeout_msecs); + } + + return do_sys_poll(ufds, nfds, &timeout_jiffies); +} + +#ifdef TIF_RESTORE_SIGMASK +asmlinkage long sys_ppoll(struct pollfd __user *ufds, unsigned int nfds, + struct timespec __user *tsp, const sigset_t __user *sigmask, + size_t sigsetsize) +{ + sigset_t ksigmask, sigsaved; + struct timespec ts; + s64 timeout = -1; + int ret; + + if (tsp) { + if (copy_from_user(&ts, tsp, sizeof(ts))) + return -EFAULT; + + /* Cast to u64 to make GCC stop complaining */ + if ((u64)ts.tv_sec >= (u64)MAX_INT64_SECONDS) + timeout = -1; /* infinite */ + else { + timeout = ROUND_UP(ts.tv_nsec, NSEC_PER_SEC/HZ); + timeout += ts.tv_sec * HZ; + } + } + + if (sigmask) { + /* XXX: Don't preclude handling different sized sigset_t's. */ + if (sigsetsize != sizeof(sigset_t)) + return -EINVAL; + if (copy_from_user(&ksigmask, sigmask, sizeof(ksigmask))) + return -EFAULT; + + sigdelsetmask(&ksigmask, sigmask(SIGKILL)|sigmask(SIGSTOP)); + sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); + } + + ret = do_sys_poll(ufds, nfds, &timeout); + + /* We can restart this syscall, usually */ + if (ret == -EINTR) { + /* + * Don't restore the signal mask yet. Let do_signal() deliver + * the signal on the way back to userspace, before the signal + * mask is restored. + */ + if (sigmask) { + memcpy(¤t->saved_sigmask, &sigsaved, + sizeof(sigsaved)); + set_thread_flag(TIF_RESTORE_SIGMASK); + } + ret = -ERESTARTNOHAND; + } else if (sigmask) + sigprocmask(SIG_SETMASK, &sigsaved, NULL); + + if (tsp && timeout >= 0) { + if (current->personality & STICKY_TIMEOUTS) + goto sticky; + /* Yes, we know it's actually an s64, but it's also positive. */ + ts.tv_nsec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)) * 1000; + ts.tv_sec = timeout; + if (copy_to_user(tsp, &ts, sizeof(ts))) { + sticky: + /* + * If an application puts its timeval in read-only + * memory, we don't want the Linux-specific update to + * the timeval to cause a fault after the select has + * completed successfully. However, because we're not + * updating the timeval, we can't restart the system + * call. + */ + if (ret == -ERESTARTNOHAND && timeout >= 0) + ret = -EINTR; + } + } + + return ret; +} +#endif /* TIF_RESTORE_SIGMASK */ diff --git a/include/linux/poll.h b/include/linux/poll.h index f6da702088f..8e8f6098508 100644 --- a/include/linux/poll.h +++ b/include/linux/poll.h @@ -92,7 +92,11 @@ void zero_fd_set(unsigned long nr, unsigned long *fdset) memset(fdset, 0, FDS_BYTES(nr)); } -extern int do_select(int n, fd_set_bits *fds, long *timeout); +#define MAX_INT64_SECONDS (((s64)(~((u64)0)>>1)/HZ)-1) + +extern int do_select(int n, fd_set_bits *fds, s64 *timeout); +extern int do_sys_poll(struct pollfd __user * ufds, unsigned int nfds, + s64 *timeout); #endif /* KERNEL */ -- cgit v1.2.3 From 3213e913b0d6baeb28aa1affbdd4bfa7efedc35f Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 18 Jan 2006 17:44:06 -0800 Subject: [PATCH] Add pselect/ppoll system calls on i386 Add the sys_pselect6() and sys_poll() calls to the i386 syscall table. Signed-off-by: David Woodhouse Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/syscall_table.S | 2 ++ include/asm-i386/unistd.h | 4 +++- 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S index a988b40e594..1b665928336 100644 --- a/arch/i386/kernel/syscall_table.S +++ b/arch/i386/kernel/syscall_table.S @@ -307,3 +307,5 @@ ENTRY(sys_call_table) .long sys_readlinkat /* 305 */ .long sys_fchmodat .long sys_faccessat + .long sys_pselect6 + .long sys_ppoll diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h index 3400441b3f9..597496ed2ae 100644 --- a/include/asm-i386/unistd.h +++ b/include/asm-i386/unistd.h @@ -313,8 +313,10 @@ #define __NR_readlinkat 305 #define __NR_fchmodat 306 #define __NR_faccessat 307 +#define __NR_pselect6 308 +#define __NR_ppoll 309 -#define NR_syscalls 308 +#define NR_syscalls 310 /* * user-visible error numbers are in the range -1 - -128: see -- cgit v1.2.3 From 715b49ef2de6fcead0776d9349071670282faf65 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 18 Jan 2006 17:44:07 -0800 Subject: [PATCH] EDAC: atomic scrub operations EDAC requires a way to scrub memory if an ECC error is found and the chipset does not do the work automatically. That means rewriting memory locations atomically with respect to all CPUs _and_ bus masters. That means we can't use atomic_add(foo, 0) as it gets optimised for non-SMP This adds a function to include/asm-foo/atomic.h for the platforms currently supported which implements a scrub of a mapped block. It also adjusts a few other files include order where atomic.h is included before types.h as this now causes an error as atomic_scrub uses u32. Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/kcopyd.c | 1 + fs/nfsctl.c | 1 + include/asm-i386/atomic.h | 12 ++++++++++++ include/asm-x86_64/atomic.h | 12 ++++++++++++ kernel/audit.c | 2 +- kernel/auditsc.c | 2 +- net/ipv4/raw.c | 2 +- 7 files changed, 29 insertions(+), 3 deletions(-) diff --git a/drivers/md/kcopyd.c b/drivers/md/kcopyd.c index ca99979c868..8b3515f394a 100644 --- a/drivers/md/kcopyd.c +++ b/drivers/md/kcopyd.c @@ -8,6 +8,7 @@ * completion notification. */ +#include #include #include diff --git a/fs/nfsctl.c b/fs/nfsctl.c index 0b14938b5b6..0d4cf948606 100644 --- a/fs/nfsctl.c +++ b/fs/nfsctl.c @@ -5,6 +5,7 @@ * */ #include +#include #include #include #include diff --git a/include/asm-i386/atomic.h b/include/asm-i386/atomic.h index de649d3aa2d..e2c00c95a5e 100644 --- a/include/asm-i386/atomic.h +++ b/include/asm-i386/atomic.h @@ -255,5 +255,17 @@ __asm__ __volatile__(LOCK "orl %0,%1" \ #define smp_mb__before_atomic_inc() barrier() #define smp_mb__after_atomic_inc() barrier() +/* ECC atomic, DMA, SMP and interrupt safe scrub function */ + +static __inline__ void atomic_scrub(unsigned long *virt_addr, u32 size) +{ + u32 i; + for (i = 0; i < size / 4; i++, virt_addr++) + /* Very carefully read and write to memory atomically + * so we are interrupt, DMA and SMP safe. + */ + __asm__ __volatile__("lock; addl $0, %0"::"m"(*virt_addr)); +} + #include #endif diff --git a/include/asm-x86_64/atomic.h b/include/asm-x86_64/atomic.h index 4b5cd553e77..4048508c4f4 100644 --- a/include/asm-x86_64/atomic.h +++ b/include/asm-x86_64/atomic.h @@ -426,5 +426,17 @@ __asm__ __volatile__(LOCK "orl %0,%1" \ #define smp_mb__before_atomic_inc() barrier() #define smp_mb__after_atomic_inc() barrier() +/* ECC atomic, DMA, SMP and interrupt safe scrub function */ + +static __inline__ void atomic_scrub(u32 *virt_addr, u32 size) +{ + u32 i; + for (i = 0; i < size / 4; i++, virt_addr++) + /* Very carefully read and write to memory atomically + * so we are interrupt, DMA and SMP safe. + */ + __asm__ __volatile__("lock; addl $0, %0"::"m"(*virt_addr)); +} + #include #endif diff --git a/kernel/audit.c b/kernel/audit.c index d13ab7d2d89..0a813d2883e 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -42,8 +42,8 @@ */ #include -#include #include +#include #include #include #include diff --git a/kernel/auditsc.c b/kernel/auditsc.c index d8a68509e72..685c25175d9 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -30,8 +30,8 @@ */ #include -#include #include +#include #include #include #include diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 165a4d81efa..f29a12da510 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -40,12 +40,12 @@ */ #include +#include #include #include #include #include #include -#include #include #include #include -- cgit v1.2.3 From 806c35f5057a64d3061ee4e2b1023bf6f6d328e2 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 18 Jan 2006 17:44:08 -0800 Subject: [PATCH] EDAC: drivers for AMD 76x and Intel E750x, E752x Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/edac/amd76x_edac.c | 356 +++++++++++++++ drivers/edac/e752x_edac.c | 1069 ++++++++++++++++++++++++++++++++++++++++++++ drivers/edac/e7xxx_edac.c | 558 +++++++++++++++++++++++ 3 files changed, 1983 insertions(+) create mode 100644 drivers/edac/amd76x_edac.c create mode 100644 drivers/edac/e752x_edac.c create mode 100644 drivers/edac/e7xxx_edac.c diff --git a/drivers/edac/amd76x_edac.c b/drivers/edac/amd76x_edac.c new file mode 100644 index 00000000000..8e2b1295e70 --- /dev/null +++ b/drivers/edac/amd76x_edac.c @@ -0,0 +1,356 @@ +/* + * AMD 76x Memory Controller kernel module + * (C) 2003 Linux Networx (http://lnxi.com) + * This file may be distributed under the terms of the + * GNU General Public License. + * + * Written by Thayne Harbaugh + * Based on work by Dan Hollis and others. + * http://www.anime.net/~goemon/linux-ecc/ + * + * $Id: edac_amd76x.c,v 1.4.2.5 2005/10/05 00:43:44 dsp_llnl Exp $ + * + */ + + +#include +#include +#include + +#include +#include + +#include + +#include "edac_mc.h" + + +#define AMD76X_NR_CSROWS 8 +#define AMD76X_NR_CHANS 1 +#define AMD76X_NR_DIMMS 4 + + +/* AMD 76x register addresses - device 0 function 0 - PCI bridge */ +#define AMD76X_ECC_MODE_STATUS 0x48 /* Mode and status of ECC (32b) + * + * 31:16 reserved + * 15:14 SERR enabled: x1=ue 1x=ce + * 13 reserved + * 12 diag: disabled, enabled + * 11:10 mode: dis, EC, ECC, ECC+scrub + * 9:8 status: x1=ue 1x=ce + * 7:4 UE cs row + * 3:0 CE cs row + */ +#define AMD76X_DRAM_MODE_STATUS 0x58 /* DRAM Mode and status (32b) + * + * 31:26 clock disable 5 - 0 + * 25 SDRAM init + * 24 reserved + * 23 mode register service + * 22:21 suspend to RAM + * 20 burst refresh enable + * 19 refresh disable + * 18 reserved + * 17:16 cycles-per-refresh + * 15:8 reserved + * 7:0 x4 mode enable 7 - 0 + */ +#define AMD76X_MEM_BASE_ADDR 0xC0 /* Memory base address (8 x 32b) + * + * 31:23 chip-select base + * 22:16 reserved + * 15:7 chip-select mask + * 6:3 reserved + * 2:1 address mode + * 0 chip-select enable + */ + + +struct amd76x_error_info { + u32 ecc_mode_status; +}; + + +enum amd76x_chips { + AMD761 = 0, + AMD762 +}; + + +struct amd76x_dev_info { + const char *ctl_name; +}; + + +static const struct amd76x_dev_info amd76x_devs[] = { + [AMD761] = {.ctl_name = "AMD761"}, + [AMD762] = {.ctl_name = "AMD762"}, +}; + + +/** + * amd76x_get_error_info - fetch error information + * @mci: Memory controller + * @info: Info to fill in + * + * Fetch and store the AMD76x ECC status. Clear pending status + * on the chip so that further errors will be reported + */ + +static void amd76x_get_error_info (struct mem_ctl_info *mci, + struct amd76x_error_info *info) +{ + pci_read_config_dword(mci->pdev, AMD76X_ECC_MODE_STATUS, + &info->ecc_mode_status); + + if (info->ecc_mode_status & BIT(8)) + pci_write_bits32(mci->pdev, AMD76X_ECC_MODE_STATUS, + (u32) BIT(8), (u32) BIT(8)); + + if (info->ecc_mode_status & BIT(9)) + pci_write_bits32(mci->pdev, AMD76X_ECC_MODE_STATUS, + (u32) BIT(9), (u32) BIT(9)); +} + + +/** + * amd76x_process_error_info - Error check + * @mci: Memory controller + * @info: Previously fetched information from chip + * @handle_errors: 1 if we should do recovery + * + * Process the chip state and decide if an error has occurred. + * A return of 1 indicates an error. Also if handle_errors is true + * then attempt to handle and clean up after the error + */ + +static int amd76x_process_error_info (struct mem_ctl_info *mci, + struct amd76x_error_info *info, int handle_errors) +{ + int error_found; + u32 row; + + error_found = 0; + + /* + * Check for an uncorrectable error + */ + if (info->ecc_mode_status & BIT(8)) { + error_found = 1; + + if (handle_errors) { + row = (info->ecc_mode_status >> 4) & 0xf; + edac_mc_handle_ue(mci, + mci->csrows[row].first_page, 0, row, + mci->ctl_name); + } + } + + /* + * Check for a correctable error + */ + if (info->ecc_mode_status & BIT(9)) { + error_found = 1; + + if (handle_errors) { + row = info->ecc_mode_status & 0xf; + edac_mc_handle_ce(mci, + mci->csrows[row].first_page, 0, 0, row, 0, + mci->ctl_name); + } + } + return error_found; +} + +/** + * amd76x_check - Poll the controller + * @mci: Memory controller + * + * Called by the poll handlers this function reads the status + * from the controller and checks for errors. + */ + +static void amd76x_check(struct mem_ctl_info *mci) +{ + struct amd76x_error_info info; + debugf3("MC: " __FILE__ ": %s()\n", __func__); + amd76x_get_error_info(mci, &info); + amd76x_process_error_info(mci, &info, 1); +} + + +/** + * amd76x_probe1 - Perform set up for detected device + * @pdev; PCI device detected + * @dev_idx: Device type index + * + * We have found an AMD76x and now need to set up the memory + * controller status reporting. We configure and set up the + * memory controller reporting and claim the device. + */ + +static int amd76x_probe1(struct pci_dev *pdev, int dev_idx) +{ + int rc = -ENODEV; + int index; + struct mem_ctl_info *mci = NULL; + enum edac_type ems_modes[] = { + EDAC_NONE, + EDAC_EC, + EDAC_SECDED, + EDAC_SECDED + }; + u32 ems; + u32 ems_mode; + + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + pci_read_config_dword(pdev, AMD76X_ECC_MODE_STATUS, &ems); + ems_mode = (ems >> 10) & 0x3; + + mci = edac_mc_alloc(0, AMD76X_NR_CSROWS, AMD76X_NR_CHANS); + + if (mci == NULL) { + rc = -ENOMEM; + goto fail; + } + + debugf0("MC: " __FILE__ ": %s(): mci = %p\n", __func__, mci); + + mci->pdev = pci_dev_get(pdev); + mci->mtype_cap = MEM_FLAG_RDDR; + + mci->edac_ctl_cap = EDAC_FLAG_NONE | EDAC_FLAG_EC | EDAC_FLAG_SECDED; + mci->edac_cap = ems_mode ? + (EDAC_FLAG_EC | EDAC_FLAG_SECDED) : EDAC_FLAG_NONE; + + mci->mod_name = BS_MOD_STR; + mci->mod_ver = "$Revision: 1.4.2.5 $"; + mci->ctl_name = amd76x_devs[dev_idx].ctl_name; + mci->edac_check = amd76x_check; + mci->ctl_page_to_phys = NULL; + + for (index = 0; index < mci->nr_csrows; index++) { + struct csrow_info *csrow = &mci->csrows[index]; + u32 mba; + u32 mba_base; + u32 mba_mask; + u32 dms; + + /* find the DRAM Chip Select Base address and mask */ + pci_read_config_dword(mci->pdev, + AMD76X_MEM_BASE_ADDR + (index * 4), + &mba); + + if (!(mba & BIT(0))) + continue; + + mba_base = mba & 0xff800000UL; + mba_mask = ((mba & 0xff80) << 16) | 0x7fffffUL; + + pci_read_config_dword(mci->pdev, AMD76X_DRAM_MODE_STATUS, + &dms); + + csrow->first_page = mba_base >> PAGE_SHIFT; + csrow->nr_pages = (mba_mask + 1) >> PAGE_SHIFT; + csrow->last_page = csrow->first_page + csrow->nr_pages - 1; + csrow->page_mask = mba_mask >> PAGE_SHIFT; + csrow->grain = csrow->nr_pages << PAGE_SHIFT; + csrow->mtype = MEM_RDDR; + csrow->dtype = ((dms >> index) & 0x1) ? DEV_X4 : DEV_UNKNOWN; + csrow->edac_mode = ems_modes[ems_mode]; + } + + /* clear counters */ + pci_write_bits32(mci->pdev, AMD76X_ECC_MODE_STATUS, (u32) (0x3 << 8), + (u32) (0x3 << 8)); + + if (edac_mc_add_mc(mci)) { + debugf3("MC: " __FILE__ + ": %s(): failed edac_mc_add_mc()\n", __func__); + goto fail; + } + + /* get this far and it's successful */ + debugf3("MC: " __FILE__ ": %s(): success\n", __func__); + return 0; + +fail: + if (mci) { + if(mci->pdev) + pci_dev_put(mci->pdev); + edac_mc_free(mci); + } + return rc; +} + +/* returns count (>= 0), or negative on error */ +static int __devinit amd76x_init_one(struct pci_dev *pdev, + const struct pci_device_id *ent) +{ + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + /* don't need to call pci_device_enable() */ + return amd76x_probe1(pdev, ent->driver_data); +} + + +/** + * amd76x_remove_one - driver shutdown + * @pdev: PCI device being handed back + * + * Called when the driver is unloaded. Find the matching mci + * structure for the device then delete the mci and free the + * resources. + */ + +static void __devexit amd76x_remove_one(struct pci_dev *pdev) +{ + struct mem_ctl_info *mci; + + debugf0(__FILE__ ": %s()\n", __func__); + + if ((mci = edac_mc_find_mci_by_pdev(pdev)) == NULL) + return; + if (edac_mc_del_mc(mci)) + return; + pci_dev_put(mci->pdev); + edac_mc_free(mci); +} + + +static const struct pci_device_id amd76x_pci_tbl[] __devinitdata = { + {PCI_VEND_DEV(AMD, FE_GATE_700C), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + AMD762}, + {PCI_VEND_DEV(AMD, FE_GATE_700E), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + AMD761}, + {0,} /* 0 terminated list. */ +}; + +MODULE_DEVICE_TABLE(pci, amd76x_pci_tbl); + + +static struct pci_driver amd76x_driver = { + .name = BS_MOD_STR, + .probe = amd76x_init_one, + .remove = __devexit_p(amd76x_remove_one), + .id_table = amd76x_pci_tbl, +}; + +int __init amd76x_init(void) +{ + return pci_register_driver(&amd76x_driver); +} + +static void __exit amd76x_exit(void) +{ + pci_unregister_driver(&amd76x_driver); +} + +module_init(amd76x_init); +module_exit(amd76x_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Linux Networx (http://lnxi.com) Thayne Harbaugh"); +MODULE_DESCRIPTION("MC support for AMD 76x memory controllers"); diff --git a/drivers/edac/e752x_edac.c b/drivers/edac/e752x_edac.c new file mode 100644 index 00000000000..959f584f568 --- /dev/null +++ b/drivers/edac/e752x_edac.c @@ -0,0 +1,1069 @@ +/* + * Intel e752x Memory Controller kernel module + * (C) 2004 Linux Networx (http://lnxi.com) + * This file may be distributed under the terms of the + * GNU General Public License. + * + * See "enum e752x_chips" below for supported chipsets + * + * Written by Tom Zimmerman + * + * Contributors: + * Thayne Harbaugh at realmsys.com (?) + * Wang Zhenyu at intel.com + * Dave Jiang at mvista.com + * + * $Id: bluesmoke_e752x.c,v 1.5.2.11 2005/10/05 00:43:44 dsp_llnl Exp $ + * + */ + + +#include +#include +#include + +#include +#include + +#include + +#include "edac_mc.h" + + +#ifndef PCI_DEVICE_ID_INTEL_7520_0 +#define PCI_DEVICE_ID_INTEL_7520_0 0x3590 +#endif /* PCI_DEVICE_ID_INTEL_7520_0 */ + +#ifndef PCI_DEVICE_ID_INTEL_7520_1_ERR +#define PCI_DEVICE_ID_INTEL_7520_1_ERR 0x3591 +#endif /* PCI_DEVICE_ID_INTEL_7520_1_ERR */ + +#ifndef PCI_DEVICE_ID_INTEL_7525_0 +#define PCI_DEVICE_ID_INTEL_7525_0 0x359E +#endif /* PCI_DEVICE_ID_INTEL_7525_0 */ + +#ifndef PCI_DEVICE_ID_INTEL_7525_1_ERR +#define PCI_DEVICE_ID_INTEL_7525_1_ERR 0x3593 +#endif /* PCI_DEVICE_ID_INTEL_7525_1_ERR */ + +#ifndef PCI_DEVICE_ID_INTEL_7320_0 +#define PCI_DEVICE_ID_INTEL_7320_0 0x3592 +#endif /* PCI_DEVICE_ID_INTEL_7320_0 */ + +#ifndef PCI_DEVICE_ID_INTEL_7320_1_ERR +#define PCI_DEVICE_ID_INTEL_7320_1_ERR 0x3593 +#endif /* PCI_DEVICE_ID_INTEL_7320_1_ERR */ + +#define E752X_NR_CSROWS 8 /* number of csrows */ + + +/* E752X register addresses - device 0 function 0 */ +#define E752X_DRB 0x60 /* DRAM row boundary register (8b) */ +#define E752X_DRA 0x70 /* DRAM row attribute register (8b) */ + /* + * 31:30 Device width row 7 + * 01=x8 10=x4 11=x8 DDR2 + * 27:26 Device width row 6 + * 23:22 Device width row 5 + * 19:20 Device width row 4 + * 15:14 Device width row 3 + * 11:10 Device width row 2 + * 7:6 Device width row 1 + * 3:2 Device width row 0 + */ +#define E752X_DRC 0x7C /* DRAM controller mode reg (32b) */ + /* FIXME:IS THIS RIGHT? */ + /* + * 22 Number channels 0=1,1=2 + * 19:18 DRB Granularity 32/64MB + */ +#define E752X_DRM 0x80 /* Dimm mapping register */ +#define E752X_DDRCSR 0x9A /* DDR control and status reg (16b) */ + /* + * 14:12 1 single A, 2 single B, 3 dual + */ +#define E752X_TOLM 0xC4 /* DRAM top of low memory reg (16b) */ +#define E752X_REMAPBASE 0xC6 /* DRAM remap base address reg (16b) */ +#define E752X_REMAPLIMIT 0xC8 /* DRAM remap limit address reg (16b) */ +#define E752X_REMAPOFFSET 0xCA /* DRAM remap limit offset reg (16b) */ + +/* E752X register addresses - device 0 function 1 */ +#define E752X_FERR_GLOBAL 0x40 /* Global first error register (32b) */ +#define E752X_NERR_GLOBAL 0x44 /* Global next error register (32b) */ +#define E752X_HI_FERR 0x50 /* Hub interface first error reg (8b) */ +#define E752X_HI_NERR 0x52 /* Hub interface next error reg (8b) */ +#define E752X_HI_ERRMASK 0x54 /* Hub interface error mask reg (8b) */ +#define E752X_HI_SMICMD 0x5A /* Hub interface SMI command reg (8b) */ +#define E752X_SYSBUS_FERR 0x60 /* System buss first error reg (16b) */ +#define E752X_SYSBUS_NERR 0x62 /* System buss next error reg (16b) */ +#define E752X_SYSBUS_ERRMASK 0x64 /* System buss error mask reg (16b) */ +#define E752X_SYSBUS_SMICMD 0x6A /* System buss SMI command reg (16b) */ +#define E752X_BUF_FERR 0x70 /* Memory buffer first error reg (8b) */ +#define E752X_BUF_NERR 0x72 /* Memory buffer next error reg (8b) */ +#define E752X_BUF_ERRMASK 0x74 /* Memory buffer error mask reg (8b) */ +#define E752X_BUF_SMICMD 0x7A /* Memory buffer SMI command reg (8b) */ +#define E752X_DRAM_FERR 0x80 /* DRAM first error register (16b) */ +#define E752X_DRAM_NERR 0x82 /* DRAM next error register (16b) */ +#define E752X_DRAM_ERRMASK 0x84 /* DRAM error mask register (8b) */ +#define E752X_DRAM_SMICMD 0x8A /* DRAM SMI command register (8b) */ +#define E752X_DRAM_RETR_ADD 0xAC /* DRAM Retry address register (32b) */ +#define E752X_DRAM_SEC1_ADD 0xA0 /* DRAM first correctable memory */ + /* error address register (32b) */ + /* + * 31 Reserved + * 30:2 CE address (64 byte block 34:6) + * 1 Reserved + * 0 HiLoCS + */ +#define E752X_DRAM_SEC2_ADD 0xC8 /* DRAM first correctable memory */ + /* error address register (32b) */ + /* + * 31 Reserved + * 30:2 CE address (64 byte block 34:6) + * 1 Reserved + * 0 HiLoCS + */ +#define E752X_DRAM_DED_ADD 0xA4 /* DRAM first uncorrectable memory */ + /* error address register (32b) */ + /* + * 31 Reserved + * 30:2 CE address (64 byte block 34:6) + * 1 Reserved + * 0 HiLoCS + */ +#define E752X_DRAM_SCRB_ADD 0xA8 /* DRAM first uncorrectable scrub memory */ + /* error address register (32b) */ + /* + * 31 Reserved + * 30:2 CE address (64 byte block 34:6) + * 1 Reserved + * 0 HiLoCS + */ +#define E752X_DRAM_SEC1_SYNDROME 0xC4 /* DRAM first correctable memory */ + /* error syndrome register (16b) */ +#define E752X_DRAM_SEC2_SYNDROME 0xC6 /* DRAM second correctable memory */ + /* error syndrome register (16b) */ +#define E752X_DEVPRES1 0xF4 /* Device Present 1 register (8b) */ + +/* ICH5R register addresses - device 30 function 0 */ +#define ICH5R_PCI_STAT 0x06 /* PCI status register (16b) */ +#define ICH5R_PCI_2ND_STAT 0x1E /* PCI status secondary reg (16b) */ +#define ICH5R_PCI_BRIDGE_CTL 0x3E /* PCI bridge control register (16b) */ + +enum e752x_chips { + E7520 = 0, + E7525 = 1, + E7320 = 2 +}; + + +struct e752x_pvt { + struct pci_dev *bridge_ck; + struct pci_dev *dev_d0f0; + struct pci_dev *dev_d0f1; + u32 tolm; + u32 remapbase; + u32 remaplimit; + int mc_symmetric; + u8 map[8]; + int map_type; + const struct e752x_dev_info *dev_info; +}; + + +struct e752x_dev_info { + u16 err_dev; + const char *ctl_name; +}; + +struct e752x_error_info { + u32 ferr_global; + u32 nerr_global; + u8 hi_ferr; + u8 hi_nerr; + u16 sysbus_ferr; + u16 sysbus_nerr; + u8 buf_ferr; + u8 buf_nerr; + u16 dram_ferr; + u16 dram_nerr; + u32 dram_sec1_add; + u32 dram_sec2_add; + u16 dram_sec1_syndrome; + u16 dram_sec2_syndrome; + u32 dram_ded_add; + u32 dram_scrb_add; + u32 dram_retr_add; +}; + +static const struct e752x_dev_info e752x_devs[] = { + [E7520] = { + .err_dev = PCI_DEVICE_ID_INTEL_7520_1_ERR, + .ctl_name = "E7520"}, + [E7525] = { + .err_dev = PCI_DEVICE_ID_INTEL_7525_1_ERR, + .ctl_name = "E7525"}, + [E7320] = { + .err_dev = PCI_DEVICE_ID_INTEL_7320_1_ERR, + .ctl_name = "E7320"}, +}; + + +static unsigned long ctl_page_to_phys(struct mem_ctl_info *mci, + unsigned long page) +{ + u32 remap; + struct e752x_pvt *pvt = (struct e752x_pvt *) mci->pvt_info; + + debugf3("MC: " __FILE__ ": %s()\n", __func__); + + if (page < pvt->tolm) + return page; + if ((page >= 0x100000) && (page < pvt->remapbase)) + return page; + remap = (page - pvt->tolm) + pvt->remapbase; + if (remap < pvt->remaplimit) + return remap; + printk(KERN_ERR "Invalid page %lx - out of range\n", page); + return pvt->tolm - 1; +} + +static void do_process_ce(struct mem_ctl_info *mci, u16 error_one, + u32 sec1_add, u16 sec1_syndrome) +{ + u32 page; + int row; + int channel; + int i; + struct e752x_pvt *pvt = (struct e752x_pvt *) mci->pvt_info; + + debugf3("MC: " __FILE__ ": %s()\n", __func__); + + /* convert the addr to 4k page */ + page = sec1_add >> (PAGE_SHIFT - 4); + + /* FIXME - check for -1 */ + if (pvt->mc_symmetric) { + /* chip select are bits 14 & 13 */ + row = ((page >> 1) & 3); + printk(KERN_WARNING + "Test row %d Table %d %d %d %d %d %d %d %d\n", + row, pvt->map[0], pvt->map[1], pvt->map[2], + pvt->map[3], pvt->map[4], pvt->map[5], + pvt->map[6], pvt->map[7]); + + /* test for channel remapping */ + for (i = 0; i < 8; i++) { + if (pvt->map[i] == row) + break; + } + printk(KERN_WARNING "Test computed row %d\n", i); + if (i < 8) + row = i; + else + printk(KERN_WARNING + "MC%d: row %d not found in remap table\n", + mci->mc_idx, row); + } else + row = edac_mc_find_csrow_by_page(mci, page); + /* 0 = channel A, 1 = channel B */ + channel = !(error_one & 1); + + if (!pvt->map_type) + row = 7 - row; + edac_mc_handle_ce(mci, page, 0, sec1_syndrome, row, channel, + "e752x CE"); +} + + +static inline void process_ce(struct mem_ctl_info *mci, u16 error_one, + u32 sec1_add, u16 sec1_syndrome, int *error_found, + int handle_error) +{ + *error_found = 1; + + if (handle_error) + do_process_ce(mci, error_one, sec1_add, sec1_syndrome); +} + +static void do_process_ue(struct mem_ctl_info *mci, u16 error_one, u32 ded_add, + u32 scrb_add) +{ + u32 error_2b, block_page; + int row; + struct e752x_pvt *pvt = (struct e752x_pvt *) mci->pvt_info; + + debugf3("MC: " __FILE__ ": %s()\n", __func__); + + if (error_one & 0x0202) { + error_2b = ded_add; + /* convert to 4k address */ + block_page = error_2b >> (PAGE_SHIFT - 4); + row = pvt->mc_symmetric ? + /* chip select are bits 14 & 13 */ + ((block_page >> 1) & 3) : + edac_mc_find_csrow_by_page(mci, block_page); + edac_mc_handle_ue(mci, block_page, 0, row, + "e752x UE from Read"); + } + if (error_one & 0x0404) { + error_2b = scrb_add; + /* convert to 4k address */ + block_page = error_2b >> (PAGE_SHIFT - 4); + row = pvt->mc_symmetric ? + /* chip select are bits 14 & 13 */ + ((block_page >> 1) & 3) : + edac_mc_find_csrow_by_page(mci, block_page); + edac_mc_handle_ue(mci, block_page, 0, row, + "e752x UE from Scruber"); + } +} + +static inline void process_ue(struct mem_ctl_info *mci, u16 error_one, + u32 ded_add, u32 scrb_add, int *error_found, int handle_error) +{ + *error_found = 1; + + if (handle_error) + do_process_ue(mci, error_one, ded_add, scrb_add); +} + +static inline void process_ue_no_info_wr(struct mem_ctl_info *mci, + int *error_found, int handle_error) +{ + *error_found = 1; + + if (!handle_error) + return; + + debugf3("MC: " __FILE__ ": %s()\n", __func__); + edac_mc_handle_ue_no_info(mci, "e752x UE log memory write"); +} + +static void do_process_ded_retry(struct mem_ctl_info *mci, u16 error, + u32 retry_add) +{ + u32 error_1b, page; + int row; + struct e752x_pvt *pvt = (struct e752x_pvt *) mci->pvt_info; + + error_1b = retry_add; + page = error_1b >> (PAGE_SHIFT - 4); /* convert the addr to 4k page */ + row = pvt->mc_symmetric ? + ((page >> 1) & 3) : /* chip select are bits 14 & 13 */ + edac_mc_find_csrow_by_page(mci, page); + printk(KERN_WARNING + "MC%d: CE page 0x%lx, row %d : Memory read retry\n", + mci->mc_idx, (long unsigned int) page, row); +} + +static inline void process_ded_retry(struct mem_ctl_info *mci, u16 error, + u32 retry_add, int *error_found, int handle_error) +{ + *error_found = 1; + + if (handle_error) + do_process_ded_retry(mci, error, retry_add); +} + +static inline void process_threshold_ce(struct mem_ctl_info *mci, u16 error, + int *error_found, int handle_error) +{ + *error_found = 1; + + if (handle_error) + printk(KERN_WARNING "MC%d: Memory threshold CE\n", + mci->mc_idx); +} + +char *global_message[11] = { + "PCI Express C1", "PCI Express C", "PCI Express B1", + "PCI Express B", "PCI Express A1", "PCI Express A", + "DMA Controler", "HUB Interface", "System Bus", + "DRAM Controler", "Internal Buffer" +}; + +char *fatal_message[2] = { "Non-Fatal ", "Fatal " }; + +static void do_global_error(int fatal, u32 errors) +{ + int i; + + for (i = 0; i < 11; i++) { + if (errors & (1 << i)) + printk(KERN_WARNING "%sError %s\n", + fatal_message[fatal], global_message[i]); + } +} + +static inline void global_error(int fatal, u32 errors, int *error_found, + int handle_error) +{ + *error_found = 1; + + if (handle_error) + do_global_error(fatal, errors); +} + +char *hub_message[7] = { + "HI Address or Command Parity", "HI Illegal Access", + "HI Internal Parity", "Out of Range Access", + "HI Data Parity", "Enhanced Config Access", + "Hub Interface Target Abort" +}; + +static void do_hub_error(int fatal, u8 errors) +{ + int i; + + for (i = 0; i < 7; i++) { + if (errors & (1 << i)) + printk(KERN_WARNING "%sError %s\n", + fatal_message[fatal], hub_message[i]); + } +} + +static inline void hub_error(int fatal, u8 errors, int *error_found, + int handle_error) +{ + *error_found = 1; + + if (handle_error) + do_hub_error(fatal, errors); +} + +char *membuf_message[4] = { + "Internal PMWB to DRAM parity", + "Internal PMWB to System Bus Parity", + "Internal System Bus or IO to PMWB Parity", + "Internal DRAM to PMWB Parity" +}; + +static void do_membuf_error(u8 errors) +{ + int i; + + for (i = 0; i < 4; i++) { + if (errors & (1 << i)) + printk(KERN_WARNING "Non-Fatal Error %s\n", + membuf_message[i]); + } +} + +static inline void membuf_error(u8 errors, int *error_found, int handle_error) +{ + *error_found = 1; + + if (handle_error) + do_membuf_error(errors); +} + +char *sysbus_message[10] = { + "Addr or Request Parity", + "Data Strobe Glitch", + "Addr Strobe Glitch", + "Data Parity", + "Addr Above TOM", + "Non DRAM Lock Error", + "MCERR", "BINIT", + "Memory Parity", + "IO Subsystem Parity" +}; + +static void do_sysbus_error(int fatal, u32 errors) +{ + int i; + + for (i = 0; i < 10; i++) { + if (errors & (1 << i)) + printk(KERN_WARNING "%sError System Bus %s\n", + fatal_message[fatal], global_message[i]); + } +} + +static inline void sysbus_error(int fatal, u32 errors, int *error_found, + int handle_error) +{ + *error_found = 1; + + if (handle_error) + do_sysbus_error(fatal, errors); +} + +static void e752x_check_hub_interface (struct e752x_error_info *info, + int *error_found, int handle_error) +{ + u8 stat8; + + //pci_read_config_byte(dev,E752X_HI_FERR,&stat8); + stat8 = info->hi_ferr; + if(stat8 & 0x7f) { /* Error, so process */ + stat8 &= 0x7f; + if(stat8 & 0x2b) + hub_error(1, stat8 & 0x2b, error_found, handle_error); + if(stat8 & 0x54) + hub_error(0, stat8 & 0x54, error_found, handle_error); + } + //pci_read_config_byte(dev,E752X_HI_NERR,&stat8); + stat8 = info->hi_nerr; + if(stat8 & 0x7f) { /* Error, so process */ + stat8 &= 0x7f; + if (stat8 & 0x2b) + hub_error(1, stat8 & 0x2b, error_found, handle_error); + if(stat8 & 0x54) + hub_error(0, stat8 & 0x54, error_found, handle_error); + } +} + +static void e752x_check_sysbus (struct e752x_error_info *info, int *error_found, + int handle_error) +{ + u32 stat32, error32; + + //pci_read_config_dword(dev,E752X_SYSBUS_FERR,&stat32); + stat32 = info->sysbus_ferr + (info->sysbus_nerr << 16); + + if (stat32 == 0) + return; /* no errors */ + + error32 = (stat32 >> 16) & 0x3ff; + stat32 = stat32 & 0x3ff; + if(stat32 & 0x083) + sysbus_error(1, stat32 & 0x083, error_found, handle_error); + if(stat32 & 0x37c) + sysbus_error(0, stat32 & 0x37c, error_found, handle_error); + if(error32 & 0x083) + sysbus_error(1, error32 & 0x083, error_found, handle_error); + if(error32 & 0x37c) + sysbus_error(0, error32 & 0x37c, error_found, handle_error); +} + +static void e752x_check_membuf (struct e752x_error_info *info, int *error_found, + int handle_error) +{ + u8 stat8; + + stat8 = info->buf_ferr; + if (stat8 & 0x0f) { /* Error, so process */ + stat8 &= 0x0f; + membuf_error(stat8, error_found, handle_error); + } + stat8 = info->buf_nerr; + if (stat8 & 0x0f) { /* Error, so process */ + stat8 &= 0x0f; + membuf_error(stat8, error_found, handle_error); + } +} + +static void e752x_check_dram (struct mem_ctl_info *mci, + struct e752x_error_info *info, int *error_found, int handle_error) +{ + u16 error_one, error_next; + + error_one = info->dram_ferr; + error_next = info->dram_nerr; + + /* decode and report errors */ + if(error_one & 0x0101) /* check first error correctable */ + process_ce(mci, error_one, info->dram_sec1_add, + info->dram_sec1_syndrome, error_found, + handle_error); + + if(error_next & 0x0101) /* check next error correctable */ + process_ce(mci, error_next, info->dram_sec2_add, + info->dram_sec2_syndrome, error_found, + handle_error); + + if(error_one & 0x4040) + process_ue_no_info_wr(mci, error_found, handle_error); + + if(error_next & 0x4040) + process_ue_no_info_wr(mci, error_found, handle_error); + + if(error_one & 0x2020) + process_ded_retry(mci, error_one, info->dram_retr_add, + error_found, handle_error); + + if(error_next & 0x2020) + process_ded_retry(mci, error_next, info->dram_retr_add, + error_found, handle_error); + + if(error_one & 0x0808) + process_threshold_ce(mci, error_one, error_found, + handle_error); + + if(error_next & 0x0808) + process_threshold_ce(mci, error_next, error_found, + handle_error); + + if(error_one & 0x0606) + process_ue(mci, error_one, info->dram_ded_add, + info->dram_scrb_add, error_found, handle_error); + + if(error_next & 0x0606) + process_ue(mci, error_next, info->dram_ded_add, + info->dram_scrb_add, error_found, handle_error); +} + +static void e752x_get_error_info (struct mem_ctl_info *mci, + struct e752x_error_info *info) +{ + struct pci_dev *dev; + struct e752x_pvt *pvt; + + memset(info, 0, sizeof(*info)); + pvt = (struct e752x_pvt *) mci->pvt_info; + dev = pvt->dev_d0f1; + + pci_read_config_dword(dev, E752X_FERR_GLOBAL, &info->ferr_global); + + if (info->ferr_global) { + pci_read_config_byte(dev, E752X_HI_FERR, &info->hi_ferr); + pci_read_config_word(dev, E752X_SYSBUS_FERR, + &info->sysbus_ferr); + pci_read_config_byte(dev, E752X_BUF_FERR, &info->buf_ferr); + pci_read_config_word(dev, E752X_DRAM_FERR, + &info->dram_ferr); + pci_read_config_dword(dev, E752X_DRAM_SEC1_ADD, + &info->dram_sec1_add); + pci_read_config_word(dev, E752X_DRAM_SEC1_SYNDROME, + &info->dram_sec1_syndrome); + pci_read_config_dword(dev, E752X_DRAM_DED_ADD, + &info->dram_ded_add); + pci_read_config_dword(dev, E752X_DRAM_SCRB_ADD, + &info->dram_scrb_add); + pci_read_config_dword(dev, E752X_DRAM_RETR_ADD, + &info->dram_retr_add); + + if (info->hi_ferr & 0x7f) + pci_write_config_byte(dev, E752X_HI_FERR, + info->hi_ferr); + + if (info->sysbus_ferr) + pci_write_config_word(dev, E752X_SYSBUS_FERR, + info->sysbus_ferr); + + if (info->buf_ferr & 0x0f) + pci_write_config_byte(dev, E752X_BUF_FERR, + info->buf_ferr); + + if (info->dram_ferr) + pci_write_bits16(pvt->bridge_ck, E752X_DRAM_FERR, + info->dram_ferr, info->dram_ferr); + + pci_write_config_dword(dev, E752X_FERR_GLOBAL, + info->ferr_global); + } + + pci_read_config_dword(dev, E752X_NERR_GLOBAL, &info->nerr_global); + + if (info->nerr_global) { + pci_read_config_byte(dev, E752X_HI_NERR, &info->hi_nerr); + pci_read_config_word(dev, E752X_SYSBUS_NERR, + &info->sysbus_nerr); + pci_read_config_byte(dev, E752X_BUF_NERR, &info->buf_nerr); + pci_read_config_word(dev, E752X_DRAM_NERR, + &info->dram_nerr); + pci_read_config_dword(dev, E752X_DRAM_SEC2_ADD, + &info->dram_sec2_add); + pci_read_config_word(dev, E752X_DRAM_SEC2_SYNDROME, + &info->dram_sec2_syndrome); + + if (info->hi_nerr & 0x7f) + pci_write_config_byte(dev, E752X_HI_NERR, + info->hi_nerr); + + if (info->sysbus_nerr) + pci_write_config_word(dev, E752X_SYSBUS_NERR, + info->sysbus_nerr); + + if (info->buf_nerr & 0x0f) + pci_write_config_byte(dev, E752X_BUF_NERR, + info->buf_nerr); + + if (info->dram_nerr) + pci_write_bits16(pvt->bridge_ck, E752X_DRAM_NERR, + info->dram_nerr, info->dram_nerr); + + pci_write_config_dword(dev, E752X_NERR_GLOBAL, + info->nerr_global); + } +} + +static int e752x_process_error_info (struct mem_ctl_info *mci, + struct e752x_error_info *info, int handle_errors) +{ + u32 error32, stat32; + int error_found; + + error_found = 0; + error32 = (info->ferr_global >> 18) & 0x3ff; + stat32 = (info->ferr_global >> 4) & 0x7ff; + + if (error32) + global_error(1, error32, &error_found, handle_errors); + + if (stat32) + global_error(0, stat32, &error_found, handle_errors); + + error32 = (info->nerr_global >> 18) & 0x3ff; + stat32 = (info->nerr_global >> 4) & 0x7ff; + + if (error32) + global_error(1, error32, &error_found, handle_errors); + + if (stat32) + global_error(0, stat32, &error_found, handle_errors); + + e752x_check_hub_interface(info, &error_found, handle_errors); + e752x_check_sysbus(info, &error_found, handle_errors); + e752x_check_membuf(info, &error_found, handle_errors); + e752x_check_dram(mci, info, &error_found, handle_errors); + return error_found; +} + +static void e752x_check(struct mem_ctl_info *mci) +{ + struct e752x_error_info info; + debugf3("MC: " __FILE__ ": %s()\n", __func__); + e752x_get_error_info(mci, &info); + e752x_process_error_info(mci, &info, 1); +} + +static int e752x_probe1(struct pci_dev *pdev, int dev_idx) +{ + int rc = -ENODEV; + int index; + u16 pci_data, stat; + u32 stat32; + u16 stat16; + u8 stat8; + struct mem_ctl_info *mci = NULL; + struct e752x_pvt *pvt = NULL; + u16 ddrcsr; + u32 drc; + int drc_chan; /* Number of channels 0=1chan,1=2chan */ + int drc_drbg; /* DRB granularity 0=64mb,1=128mb */ + int drc_ddim; /* DRAM Data Integrity Mode 0=none,2=edac */ + u32 dra; + unsigned long last_cumul_size; + struct pci_dev *pres_dev; + struct pci_dev *dev = NULL; + + debugf0("MC: " __FILE__ ": %s(): mci\n", __func__); + debugf0("Starting Probe1\n"); + + /* enable device 0 function 1 */ + pci_read_config_byte(pdev, E752X_DEVPRES1, &stat8); + stat8 |= (1 << 5); + pci_write_config_byte(pdev, E752X_DEVPRES1, stat8); + + /* need to find out the number of channels */ + pci_read_config_dword(pdev, E752X_DRC, &drc); + pci_read_config_word(pdev, E752X_DDRCSR, &ddrcsr); + /* FIXME: should check >>12 or 0xf, true for all? */ + /* Dual channel = 1, Single channel = 0 */ + drc_chan = (((ddrcsr >> 12) & 3) == 3); + drc_drbg = drc_chan + 1; /* 128 in dual mode, 64 in single */ + drc_ddim = (drc >> 20) & 0x3; + + mci = edac_mc_alloc(sizeof(*pvt), E752X_NR_CSROWS, drc_chan + 1); + + if (mci == NULL) { + rc = -ENOMEM; + goto fail; + } + + debugf3("MC: " __FILE__ ": %s(): init mci\n", __func__); + + mci->mtype_cap = MEM_FLAG_RDDR; + mci->edac_ctl_cap = EDAC_FLAG_NONE | EDAC_FLAG_SECDED | + EDAC_FLAG_S4ECD4ED; + /* FIXME - what if different memory types are in different csrows? */ + mci->mod_name = BS_MOD_STR; + mci->mod_ver = "$Revision: 1.5.2.11 $"; + mci->pdev = pdev; + + debugf3("MC: " __FILE__ ": %s(): init pvt\n", __func__); + pvt = (struct e752x_pvt *) mci->pvt_info; + pvt->dev_info = &e752x_devs[dev_idx]; + pvt->bridge_ck = pci_get_device(PCI_VENDOR_ID_INTEL, + pvt->dev_info->err_dev, + pvt->bridge_ck); + if (pvt->bridge_ck == NULL) + pvt->bridge_ck = pci_scan_single_device(pdev->bus, + PCI_DEVFN(0, 1)); + if (pvt->bridge_ck == NULL) { + printk(KERN_ERR "MC: error reporting device not found:" + "vendor %x device 0x%x (broken BIOS?)\n", + PCI_VENDOR_ID_INTEL, e752x_devs[dev_idx].err_dev); + goto fail; + } + pvt->mc_symmetric = ((ddrcsr & 0x10) != 0); + + debugf3("MC: " __FILE__ ": %s(): more mci init\n", __func__); + mci->ctl_name = pvt->dev_info->ctl_name; + mci->edac_check = e752x_check; + mci->ctl_page_to_phys = ctl_page_to_phys; + + /* find out the device types */ + pci_read_config_dword(pdev, E752X_DRA, &dra); + + /* + * The dram row boundary (DRB) reg values are boundary address for + * each DRAM row with a granularity of 64 or 128MB (single/dual + * channel operation). DRB regs are cumulative; therefore DRB7 will + * contain the total memory contained in all eight rows. + */ + for (last_cumul_size = index = 0; index < mci->nr_csrows; index++) { + u8 value; + u32 cumul_size; + /* mem_dev 0=x8, 1=x4 */ + int mem_dev = (dra >> (index * 4 + 2)) & 0x3; + struct csrow_info *csrow = &mci->csrows[index]; + + mem_dev = (mem_dev == 2); + pci_read_config_byte(mci->pdev, E752X_DRB + index, &value); + /* convert a 128 or 64 MiB DRB to a page size. */ + cumul_size = value << (25 + drc_drbg - PAGE_SHIFT); + debugf3("MC: " __FILE__ ": %s(): (%d) cumul_size 0x%x\n", + __func__, index, cumul_size); + if (cumul_size == last_cumul_size) + continue; /* not populated */ + + csrow->first_page = last_cumul_size; + csrow->last_page = cumul_size - 1; + csrow->nr_pages = cumul_size - last_cumul_size; + last_cumul_size = cumul_size; + csrow->grain = 1 << 12; /* 4KiB - resolution of CELOG */ + csrow->mtype = MEM_RDDR; /* only one type supported */ + csrow->dtype = mem_dev ? DEV_X4 : DEV_X8; + + /* + * if single channel or x8 devices then SECDED + * if dual channel and x4 then S4ECD4ED + */ + if (drc_ddim) { + if (drc_chan && mem_dev) { + csrow->edac_mode = EDAC_S4ECD4ED; + mci->edac_cap |= EDAC_FLAG_S4ECD4ED; + } else { + csrow->edac_mode = EDAC_SECDED; + mci->edac_cap |= EDAC_FLAG_SECDED; + } + } else + csrow->edac_mode = EDAC_NONE; + } + + /* Fill in the memory map table */ + { + u8 value; + u8 last = 0; + u8 row = 0; + for (index = 0; index < 8; index += 2) { + + pci_read_config_byte(mci->pdev, E752X_DRB + index, + &value); + /* test if there is a dimm in this slot */ + if (value == last) { + /* no dimm in the slot, so flag it as empty */ + pvt->map[index] = 0xff; + pvt->map[index + 1] = 0xff; + } else { /* there is a dimm in the slot */ + pvt->map[index] = row; + row++; + last = value; + /* test the next value to see if the dimm is + double sided */ + pci_read_config_byte(mci->pdev, + E752X_DRB + index + 1, + &value); + pvt->map[index + 1] = (value == last) ? + 0xff : /* the dimm is single sided, + so flag as empty */ + row; /* this is a double sided dimm + to save the next row # */ + row++; + last = value; + } + } + } + + /* set the map type. 1 = normal, 0 = reversed */ + pci_read_config_byte(mci->pdev, E752X_DRM, &stat8); + pvt->map_type = ((stat8 & 0x0f) > ((stat8 >> 4) & 0x0f)); + + mci->edac_cap |= EDAC_FLAG_NONE; + + debugf3("MC: " __FILE__ ": %s(): tolm, remapbase, remaplimit\n", + __func__); + /* load the top of low memory, remap base, and remap limit vars */ + pci_read_config_word(mci->pdev, E752X_TOLM, &pci_data); + pvt->tolm = ((u32) pci_data) << 4; + pci_read_config_word(mci->pdev, E752X_REMAPBASE, &pci_data); + pvt->remapbase = ((u32) pci_data) << 14; + pci_read_config_word(mci->pdev, E752X_REMAPLIMIT, &pci_data); + pvt->remaplimit = ((u32) pci_data) << 14; + printk("tolm = %x, remapbase = %x, remaplimit = %x\n", pvt->tolm, + pvt->remapbase, pvt->remaplimit); + + if (edac_mc_add_mc(mci)) { + debugf3("MC: " __FILE__ + ": %s(): failed edac_mc_add_mc()\n", + __func__); + goto fail; + } + + /* Walk through the PCI table and clear errors */ + switch (dev_idx) { + case E7520: + dev = pci_get_device(PCI_VENDOR_ID_INTEL, + PCI_DEVICE_ID_INTEL_7520_0, NULL); + break; + case E7525: + dev = pci_get_device(PCI_VENDOR_ID_INTEL, + PCI_DEVICE_ID_INTEL_7525_0, NULL); + break; + case E7320: + dev = pci_get_device(PCI_VENDOR_ID_INTEL, + PCI_DEVICE_ID_INTEL_7320_0, NULL); + break; + } + + + pvt->dev_d0f0 = dev; + for (pres_dev = dev; + ((struct pci_dev *) pres_dev->global_list.next != dev); + pres_dev = (struct pci_dev *) pres_dev->global_list.next) { + pci_read_config_dword(pres_dev, PCI_COMMAND, &stat32); + stat = (u16) (stat32 >> 16); + /* clear any error bits */ + if (stat32 & ((1 << 6) + (1 << 8))) + pci_write_config_word(pres_dev, PCI_STATUS, stat); + } + /* find the error reporting device and clear errors */ + dev = pvt->dev_d0f1 = pci_dev_get(pvt->bridge_ck); + /* Turn off error disable & SMI in case the BIOS turned it on */ + pci_write_config_byte(dev, E752X_HI_ERRMASK, 0x00); + pci_write_config_byte(dev, E752X_HI_SMICMD, 0x00); + pci_write_config_word(dev, E752X_SYSBUS_ERRMASK, 0x00); + pci_write_config_word(dev, E752X_SYSBUS_SMICMD, 0x00); + pci_write_config_byte(dev, E752X_BUF_ERRMASK, 0x00); + pci_write_config_byte(dev, E752X_BUF_SMICMD, 0x00); + pci_write_config_byte(dev, E752X_DRAM_ERRMASK, 0x00); + pci_write_config_byte(dev, E752X_DRAM_SMICMD, 0x00); + /* clear other MCH errors */ + pci_read_config_dword(dev, E752X_FERR_GLOBAL, &stat32); + pci_write_config_dword(dev, E752X_FERR_GLOBAL, stat32); + pci_read_config_dword(dev, E752X_NERR_GLOBAL, &stat32); + pci_write_config_dword(dev, E752X_NERR_GLOBAL, stat32); + pci_read_config_byte(dev, E752X_HI_FERR, &stat8); + pci_write_config_byte(dev, E752X_HI_FERR, stat8); + pci_read_config_byte(dev, E752X_HI_NERR, &stat8); + pci_write_config_byte(dev, E752X_HI_NERR, stat8); + pci_read_config_dword(dev, E752X_SYSBUS_FERR, &stat32); + pci_write_config_dword(dev, E752X_SYSBUS_FERR, stat32); + pci_read_config_byte(dev, E752X_BUF_FERR, &stat8); + pci_write_config_byte(dev, E752X_BUF_FERR, stat8); + pci_read_config_byte(dev, E752X_BUF_NERR, &stat8); + pci_write_config_byte(dev, E752X_BUF_NERR, stat8); + pci_read_config_word(dev, E752X_DRAM_FERR, &stat16); + pci_write_config_word(dev, E752X_DRAM_FERR, stat16); + pci_read_config_word(dev, E752X_DRAM_NERR, &stat16); + pci_write_config_word(dev, E752X_DRAM_NERR, stat16); + + /* get this far and it's successful */ + debugf3("MC: " __FILE__ ": %s(): success\n", __func__); + return 0; + +fail: + if (mci) { + if (pvt->dev_d0f0) + pci_dev_put(pvt->dev_d0f0); + if (pvt->dev_d0f1) + pci_dev_put(pvt->dev_d0f1); + if (pvt->bridge_ck) + pci_dev_put(pvt->bridge_ck); + edac_mc_free(mci); + } + return rc; +} + +/* returns count (>= 0), or negative on error */ +static int __devinit e752x_init_one(struct pci_dev *pdev, + const struct pci_device_id *ent) +{ + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + /* wake up and enable device */ + if(pci_enable_device(pdev) < 0) + return -EIO; + return e752x_probe1(pdev, ent->driver_data); +} + + +static void __devexit e752x_remove_one(struct pci_dev *pdev) +{ + struct mem_ctl_info *mci; + struct e752x_pvt *pvt; + + debugf0(__FILE__ ": %s()\n", __func__); + + if ((mci = edac_mc_find_mci_by_pdev(pdev)) == NULL) + return; + + if (edac_mc_del_mc(mci)) + return; + + pvt = (struct e752x_pvt *) mci->pvt_info; + pci_dev_put(pvt->dev_d0f0); + pci_dev_put(pvt->dev_d0f1); + pci_dev_put(pvt->bridge_ck); + edac_mc_free(mci); +} + + +static const struct pci_device_id e752x_pci_tbl[] __devinitdata = { + {PCI_VEND_DEV(INTEL, 7520_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + E7520}, + {PCI_VEND_DEV(INTEL, 7525_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + E7525}, + {PCI_VEND_DEV(INTEL, 7320_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + E7320}, + {0,} /* 0 terminated list. */ +}; + +MODULE_DEVICE_TABLE(pci, e752x_pci_tbl); + + +static struct pci_driver e752x_driver = { + name: BS_MOD_STR, + probe: e752x_init_one, + remove: __devexit_p(e752x_remove_one), + id_table: e752x_pci_tbl, +}; + + +int __init e752x_init(void) +{ + int pci_rc; + + debugf3("MC: " __FILE__ ": %s()\n", __func__); + pci_rc = pci_register_driver(&e752x_driver); + return (pci_rc < 0) ? pci_rc : 0; +} + + +static void __exit e752x_exit(void) +{ + debugf3("MC: " __FILE__ ": %s()\n", __func__); + pci_unregister_driver(&e752x_driver); +} + + +module_init(e752x_init); +module_exit(e752x_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Linux Networx (http://lnxi.com) Tom Zimmerman\n"); +MODULE_DESCRIPTION("MC support for Intel e752x memory controllers"); diff --git a/drivers/edac/e7xxx_edac.c b/drivers/edac/e7xxx_edac.c new file mode 100644 index 00000000000..066be43f5ec --- /dev/null +++ b/drivers/edac/e7xxx_edac.c @@ -0,0 +1,558 @@ +/* + * Intel e7xxx Memory Controller kernel module + * (C) 2003 Linux Networx (http://lnxi.com) + * This file may be distributed under the terms of the + * GNU General Public License. + * + * See "enum e7xxx_chips" below for supported chipsets + * + * Written by Thayne Harbaugh + * Based on work by Dan Hollis and others. + * http://www.anime.net/~goemon/linux-ecc/ + * + * Contributors: + * Eric Biederman (Linux Networx) + * Tom Zimmerman (Linux Networx) + * Jim Garlick (Lawrence Livermore National Labs) + * Dave Peterson (Lawrence Livermore National Labs) + * That One Guy (Some other place) + * Wang Zhenyu (intel.com) + * + * $Id: edac_e7xxx.c,v 1.5.2.9 2005/10/05 00:43:44 dsp_llnl Exp $ + * + */ + + +#include +#include +#include +#include +#include +#include +#include "edac_mc.h" + + +#ifndef PCI_DEVICE_ID_INTEL_7205_0 +#define PCI_DEVICE_ID_INTEL_7205_0 0x255d +#endif /* PCI_DEVICE_ID_INTEL_7205_0 */ + +#ifndef PCI_DEVICE_ID_INTEL_7205_1_ERR +#define PCI_DEVICE_ID_INTEL_7205_1_ERR 0x2551 +#endif /* PCI_DEVICE_ID_INTEL_7205_1_ERR */ + +#ifndef PCI_DEVICE_ID_INTEL_7500_0 +#define PCI_DEVICE_ID_INTEL_7500_0 0x2540 +#endif /* PCI_DEVICE_ID_INTEL_7500_0 */ + +#ifndef PCI_DEVICE_ID_INTEL_7500_1_ERR +#define PCI_DEVICE_ID_INTEL_7500_1_ERR 0x2541 +#endif /* PCI_DEVICE_ID_INTEL_7500_1_ERR */ + +#ifndef PCI_DEVICE_ID_INTEL_7501_0 +#define PCI_DEVICE_ID_INTEL_7501_0 0x254c +#endif /* PCI_DEVICE_ID_INTEL_7501_0 */ + +#ifndef PCI_DEVICE_ID_INTEL_7501_1_ERR +#define PCI_DEVICE_ID_INTEL_7501_1_ERR 0x2541 +#endif /* PCI_DEVICE_ID_INTEL_7501_1_ERR */ + +#ifndef PCI_DEVICE_ID_INTEL_7505_0 +#define PCI_DEVICE_ID_INTEL_7505_0 0x2550 +#endif /* PCI_DEVICE_ID_INTEL_7505_0 */ + +#ifndef PCI_DEVICE_ID_INTEL_7505_1_ERR +#define PCI_DEVICE_ID_INTEL_7505_1_ERR 0x2551 +#endif /* PCI_DEVICE_ID_INTEL_7505_1_ERR */ + + +#define E7XXX_NR_CSROWS 8 /* number of csrows */ +#define E7XXX_NR_DIMMS 8 /* FIXME - is this correct? */ + + +/* E7XXX register addresses - device 0 function 0 */ +#define E7XXX_DRB 0x60 /* DRAM row boundary register (8b) */ +#define E7XXX_DRA 0x70 /* DRAM row attribute register (8b) */ + /* + * 31 Device width row 7 0=x8 1=x4 + * 27 Device width row 6 + * 23 Device width row 5 + * 19 Device width row 4 + * 15 Device width row 3 + * 11 Device width row 2 + * 7 Device width row 1 + * 3 Device width row 0 + */ +#define E7XXX_DRC 0x7C /* DRAM controller mode reg (32b) */ + /* + * 22 Number channels 0=1,1=2 + * 19:18 DRB Granularity 32/64MB + */ +#define E7XXX_TOLM 0xC4 /* DRAM top of low memory reg (16b) */ +#define E7XXX_REMAPBASE 0xC6 /* DRAM remap base address reg (16b) */ +#define E7XXX_REMAPLIMIT 0xC8 /* DRAM remap limit address reg (16b) */ + +/* E7XXX register addresses - device 0 function 1 */ +#define E7XXX_DRAM_FERR 0x80 /* DRAM first error register (8b) */ +#define E7XXX_DRAM_NERR 0x82 /* DRAM next error register (8b) */ +#define E7XXX_DRAM_CELOG_ADD 0xA0 /* DRAM first correctable memory */ + /* error address register (32b) */ + /* + * 31:28 Reserved + * 27:6 CE address (4k block 33:12) + * 5:0 Reserved + */ +#define E7XXX_DRAM_UELOG_ADD 0xB0 /* DRAM first uncorrectable memory */ + /* error address register (32b) */ + /* + * 31:28 Reserved + * 27:6 CE address (4k block 33:12) + * 5:0 Reserved + */ +#define E7XXX_DRAM_CELOG_SYNDROME 0xD0 /* DRAM first correctable memory */ + /* error syndrome register (16b) */ + +enum e7xxx_chips { + E7500 = 0, + E7501, + E7505, + E7205, +}; + + +struct e7xxx_pvt { + struct pci_dev *bridge_ck; + u32 tolm; + u32 remapbase; + u32 remaplimit; + const struct e7xxx_dev_info *dev_info; +}; + + +struct e7xxx_dev_info { + u16 err_dev; + const char *ctl_name; +}; + + +struct e7xxx_error_info { + u8 dram_ferr; + u8 dram_nerr; + u32 dram_celog_add; + u16 dram_celog_syndrome; + u32 dram_uelog_add; +}; + +static const struct e7xxx_dev_info e7xxx_devs[] = { + [E7500] = { + .err_dev = PCI_DEVICE_ID_INTEL_7500_1_ERR, + .ctl_name = "E7500"}, + [E7501] = { + .err_dev = PCI_DEVICE_ID_INTEL_7501_1_ERR, + .ctl_name = "E7501"}, + [E7505] = { + .err_dev = PCI_DEVICE_ID_INTEL_7505_1_ERR, + .ctl_name = "E7505"}, + [E7205] = { + .err_dev = PCI_DEVICE_ID_INTEL_7205_1_ERR, + .ctl_name = "E7205"}, +}; + + +/* FIXME - is this valid for both SECDED and S4ECD4ED? */ +static inline int e7xxx_find_channel(u16 syndrome) +{ + debugf3("MC: " __FILE__ ": %s()\n", __func__); + + if ((syndrome & 0xff00) == 0) + return 0; + if ((syndrome & 0x00ff) == 0) + return 1; + if ((syndrome & 0xf000) == 0 || (syndrome & 0x0f00) == 0) + return 0; + return 1; +} + + +static unsigned long +ctl_page_to_phys(struct mem_ctl_info *mci, unsigned long page) +{ + u32 remap; + struct e7xxx_pvt *pvt = (struct e7xxx_pvt *) mci->pvt_info; + + debugf3("MC: " __FILE__ ": %s()\n", __func__); + + if ((page < pvt->tolm) || + ((page >= 0x100000) && (page < pvt->remapbase))) + return page; + remap = (page - pvt->tolm) + pvt->remapbase; + if (remap < pvt->remaplimit) + return remap; + printk(KERN_ERR "Invalid page %lx - out of range\n", page); + return pvt->tolm - 1; +} + + +static void process_ce(struct mem_ctl_info *mci, struct e7xxx_error_info *info) +{ + u32 error_1b, page; + u16 syndrome; + int row; + int channel; + + debugf3("MC: " __FILE__ ": %s()\n", __func__); + + /* read the error address */ + error_1b = info->dram_celog_add; + /* FIXME - should use PAGE_SHIFT */ + page = error_1b >> 6; /* convert the address to 4k page */ + /* read the syndrome */ + syndrome = info->dram_celog_syndrome; + /* FIXME - check for -1 */ + row = edac_mc_find_csrow_by_page(mci, page); + /* convert syndrome to channel */ + channel = e7xxx_find_channel(syndrome); + edac_mc_handle_ce(mci, page, 0, syndrome, row, channel, + "e7xxx CE"); +} + + +static void process_ce_no_info(struct mem_ctl_info *mci) +{ + debugf3("MC: " __FILE__ ": %s()\n", __func__); + edac_mc_handle_ce_no_info(mci, "e7xxx CE log register overflow"); +} + + +static void process_ue(struct mem_ctl_info *mci, struct e7xxx_error_info *info) +{ + u32 error_2b, block_page; + int row; + + debugf3("MC: " __FILE__ ": %s()\n", __func__); + + /* read the error address */ + error_2b = info->dram_uelog_add; + /* FIXME - should use PAGE_SHIFT */ + block_page = error_2b >> 6; /* convert to 4k address */ + row = edac_mc_find_csrow_by_page(mci, block_page); + edac_mc_handle_ue(mci, block_page, 0, row, "e7xxx UE"); +} + + +static void process_ue_no_info(struct mem_ctl_info *mci) +{ + debugf3("MC: " __FILE__ ": %s()\n", __func__); + edac_mc_handle_ue_no_info(mci, "e7xxx UE log register overflow"); +} + + +static void e7xxx_get_error_info (struct mem_ctl_info *mci, + struct e7xxx_error_info *info) +{ + struct e7xxx_pvt *pvt; + + pvt = (struct e7xxx_pvt *) mci->pvt_info; + pci_read_config_byte(pvt->bridge_ck, E7XXX_DRAM_FERR, + &info->dram_ferr); + pci_read_config_byte(pvt->bridge_ck, E7XXX_DRAM_NERR, + &info->dram_nerr); + + if ((info->dram_ferr & 1) || (info->dram_nerr & 1)) { + pci_read_config_dword(pvt->bridge_ck, E7XXX_DRAM_CELOG_ADD, + &info->dram_celog_add); + pci_read_config_word(pvt->bridge_ck, + E7XXX_DRAM_CELOG_SYNDROME, &info->dram_celog_syndrome); + } + + if ((info->dram_ferr & 2) || (info->dram_nerr & 2)) + pci_read_config_dword(pvt->bridge_ck, E7XXX_DRAM_UELOG_ADD, + &info->dram_uelog_add); + + if (info->dram_ferr & 3) + pci_write_bits8(pvt->bridge_ck, E7XXX_DRAM_FERR, 0x03, + 0x03); + + if (info->dram_nerr & 3) + pci_write_bits8(pvt->bridge_ck, E7XXX_DRAM_NERR, 0x03, + 0x03); +} + + +static int e7xxx_process_error_info (struct mem_ctl_info *mci, + struct e7xxx_error_info *info, int handle_errors) +{ + int error_found; + + error_found = 0; + + /* decode and report errors */ + if (info->dram_ferr & 1) { /* check first error correctable */ + error_found = 1; + + if (handle_errors) + process_ce(mci, info); + } + + if (info->dram_ferr & 2) { /* check first error uncorrectable */ + error_found = 1; + + if (handle_errors) + process_ue(mci, info); + } + + if (info->dram_nerr & 1) { /* check next error correctable */ + error_found = 1; + + if (handle_errors) { + if (info->dram_ferr & 1) + process_ce_no_info(mci); + else + process_ce(mci, info); + } + } + + if (info->dram_nerr & 2) { /* check next error uncorrectable */ + error_found = 1; + + if (handle_errors) { + if (info->dram_ferr & 2) + process_ue_no_info(mci); + else + process_ue(mci, info); + } + } + + return error_found; +} + + +static void e7xxx_check(struct mem_ctl_info *mci) +{ + struct e7xxx_error_info info; + + debugf3("MC: " __FILE__ ": %s()\n", __func__); + e7xxx_get_error_info(mci, &info); + e7xxx_process_error_info(mci, &info, 1); +} + + +static int e7xxx_probe1(struct pci_dev *pdev, int dev_idx) +{ + int rc = -ENODEV; + int index; + u16 pci_data; + struct mem_ctl_info *mci = NULL; + struct e7xxx_pvt *pvt = NULL; + u32 drc; + int drc_chan = 1; /* Number of channels 0=1chan,1=2chan */ + int drc_drbg = 1; /* DRB granularity 0=32mb,1=64mb */ + int drc_ddim; /* DRAM Data Integrity Mode 0=none,2=edac */ + u32 dra; + unsigned long last_cumul_size; + + + debugf0("MC: " __FILE__ ": %s(): mci\n", __func__); + + /* need to find out the number of channels */ + pci_read_config_dword(pdev, E7XXX_DRC, &drc); + /* only e7501 can be single channel */ + if (dev_idx == E7501) { + drc_chan = ((drc >> 22) & 0x1); + drc_drbg = (drc >> 18) & 0x3; + } + drc_ddim = (drc >> 20) & 0x3; + + mci = edac_mc_alloc(sizeof(*pvt), E7XXX_NR_CSROWS, drc_chan + 1); + + if (mci == NULL) { + rc = -ENOMEM; + goto fail; + } + + debugf3("MC: " __FILE__ ": %s(): init mci\n", __func__); + + mci->mtype_cap = MEM_FLAG_RDDR; + mci->edac_ctl_cap = + EDAC_FLAG_NONE | EDAC_FLAG_SECDED | EDAC_FLAG_S4ECD4ED; + /* FIXME - what if different memory types are in different csrows? */ + mci->mod_name = BS_MOD_STR; + mci->mod_ver = "$Revision: 1.5.2.9 $"; + mci->pdev = pdev; + + debugf3("MC: " __FILE__ ": %s(): init pvt\n", __func__); + pvt = (struct e7xxx_pvt *) mci->pvt_info; + pvt->dev_info = &e7xxx_devs[dev_idx]; + pvt->bridge_ck = pci_get_device(PCI_VENDOR_ID_INTEL, + pvt->dev_info->err_dev, + pvt->bridge_ck); + if (!pvt->bridge_ck) { + printk(KERN_ERR + "MC: error reporting device not found:" + "vendor %x device 0x%x (broken BIOS?)\n", + PCI_VENDOR_ID_INTEL, e7xxx_devs[dev_idx].err_dev); + goto fail; + } + + debugf3("MC: " __FILE__ ": %s(): more mci init\n", __func__); + mci->ctl_name = pvt->dev_info->ctl_name; + + mci->edac_check = e7xxx_check; + mci->ctl_page_to_phys = ctl_page_to_phys; + + /* find out the device types */ + pci_read_config_dword(pdev, E7XXX_DRA, &dra); + + /* + * The dram row boundary (DRB) reg values are boundary address + * for each DRAM row with a granularity of 32 or 64MB (single/dual + * channel operation). DRB regs are cumulative; therefore DRB7 will + * contain the total memory contained in all eight rows. + */ + for (last_cumul_size = index = 0; index < mci->nr_csrows; index++) { + u8 value; + u32 cumul_size; + /* mem_dev 0=x8, 1=x4 */ + int mem_dev = (dra >> (index * 4 + 3)) & 0x1; + struct csrow_info *csrow = &mci->csrows[index]; + + pci_read_config_byte(mci->pdev, E7XXX_DRB + index, &value); + /* convert a 64 or 32 MiB DRB to a page size. */ + cumul_size = value << (25 + drc_drbg - PAGE_SHIFT); + debugf3("MC: " __FILE__ ": %s(): (%d) cumul_size 0x%x\n", + __func__, index, cumul_size); + if (cumul_size == last_cumul_size) + continue; /* not populated */ + + csrow->first_page = last_cumul_size; + csrow->last_page = cumul_size - 1; + csrow->nr_pages = cumul_size - last_cumul_size; + last_cumul_size = cumul_size; + csrow->grain = 1 << 12; /* 4KiB - resolution of CELOG */ + csrow->mtype = MEM_RDDR; /* only one type supported */ + csrow->dtype = mem_dev ? DEV_X4 : DEV_X8; + + /* + * if single channel or x8 devices then SECDED + * if dual channel and x4 then S4ECD4ED + */ + if (drc_ddim) { + if (drc_chan && mem_dev) { + csrow->edac_mode = EDAC_S4ECD4ED; + mci->edac_cap |= EDAC_FLAG_S4ECD4ED; + } else { + csrow->edac_mode = EDAC_SECDED; + mci->edac_cap |= EDAC_FLAG_SECDED; + } + } else + csrow->edac_mode = EDAC_NONE; + } + + mci->edac_cap |= EDAC_FLAG_NONE; + + debugf3("MC: " __FILE__ ": %s(): tolm, remapbase, remaplimit\n", + __func__); + /* load the top of low memory, remap base, and remap limit vars */ + pci_read_config_word(mci->pdev, E7XXX_TOLM, &pci_data); + pvt->tolm = ((u32) pci_data) << 4; + pci_read_config_word(mci->pdev, E7XXX_REMAPBASE, &pci_data); + pvt->remapbase = ((u32) pci_data) << 14; + pci_read_config_word(mci->pdev, E7XXX_REMAPLIMIT, &pci_data); + pvt->remaplimit = ((u32) pci_data) << 14; + printk("tolm = %x, remapbase = %x, remaplimit = %x\n", pvt->tolm, + pvt->remapbase, pvt->remaplimit); + + /* clear any pending errors, or initial state bits */ + pci_write_bits8(pvt->bridge_ck, E7XXX_DRAM_FERR, 0x03, 0x03); + pci_write_bits8(pvt->bridge_ck, E7XXX_DRAM_NERR, 0x03, 0x03); + + if (edac_mc_add_mc(mci) != 0) { + debugf3("MC: " __FILE__ + ": %s(): failed edac_mc_add_mc()\n", + __func__); + goto fail; + } + + /* get this far and it's successful */ + debugf3("MC: " __FILE__ ": %s(): success\n", __func__); + return 0; + +fail: + if (mci != NULL) { + if(pvt != NULL && pvt->bridge_ck) + pci_dev_put(pvt->bridge_ck); + edac_mc_free(mci); + } + + return rc; +} + +/* returns count (>= 0), or negative on error */ +static int __devinit +e7xxx_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) +{ + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + /* wake up and enable device */ + return pci_enable_device(pdev) ? + -EIO : e7xxx_probe1(pdev, ent->driver_data); +} + + +static void __devexit e7xxx_remove_one(struct pci_dev *pdev) +{ + struct mem_ctl_info *mci; + struct e7xxx_pvt *pvt; + + debugf0(__FILE__ ": %s()\n", __func__); + + if (((mci = edac_mc_find_mci_by_pdev(pdev)) != 0) && + edac_mc_del_mc(mci)) { + pvt = (struct e7xxx_pvt *) mci->pvt_info; + pci_dev_put(pvt->bridge_ck); + edac_mc_free(mci); + } +} + + +static const struct pci_device_id e7xxx_pci_tbl[] __devinitdata = { + {PCI_VEND_DEV(INTEL, 7205_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + E7205}, + {PCI_VEND_DEV(INTEL, 7500_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + E7500}, + {PCI_VEND_DEV(INTEL, 7501_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + E7501}, + {PCI_VEND_DEV(INTEL, 7505_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + E7505}, + {0,} /* 0 terminated list. */ +}; + +MODULE_DEVICE_TABLE(pci, e7xxx_pci_tbl); + + +static struct pci_driver e7xxx_driver = { + .name = BS_MOD_STR, + .probe = e7xxx_init_one, + .remove = __devexit_p(e7xxx_remove_one), + .id_table = e7xxx_pci_tbl, +}; + + +int __init e7xxx_init(void) +{ + return pci_register_driver(&e7xxx_driver); +} + + +static void __exit e7xxx_exit(void) +{ + pci_unregister_driver(&e7xxx_driver); +} + +module_init(e7xxx_init); +module_exit(e7xxx_exit); + + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Linux Networx (http://lnxi.com) Thayne Harbaugh et al\n" + "Based on.work by Dan Hollis et al"); +MODULE_DESCRIPTION("MC support for Intel e7xxx memory controllers"); -- cgit v1.2.3 From 0d88a10e566d46bffc214c974e5cf5abe38d8da8 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 18 Jan 2006 17:44:10 -0800 Subject: [PATCH] EDAC: drivers for Intel i82860, i82875 Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/edac/i82860_edac.c | 299 +++++++++++++++++++++++++ drivers/edac/i82875p_edac.c | 532 ++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 831 insertions(+) create mode 100644 drivers/edac/i82860_edac.c create mode 100644 drivers/edac/i82875p_edac.c diff --git a/drivers/edac/i82860_edac.c b/drivers/edac/i82860_edac.c new file mode 100644 index 00000000000..bfb7ae02e37 --- /dev/null +++ b/drivers/edac/i82860_edac.c @@ -0,0 +1,299 @@ +/* + * Intel 82860 Memory Controller kernel module + * (C) 2005 Red Hat (http://www.redhat.com) + * This file may be distributed under the terms of the + * GNU General Public License. + * + * Written by Ben Woodard + * shamelessly copied from and based upon the edac_i82875 driver + * by Thayne Harbaugh of Linux Networx. (http://lnxi.com) + */ + + +#include +#include +#include +#include +#include +#include +#include "edac_mc.h" + + +#ifndef PCI_DEVICE_ID_INTEL_82860_0 +#define PCI_DEVICE_ID_INTEL_82860_0 0x2531 +#endif /* PCI_DEVICE_ID_INTEL_82860_0 */ + +#define I82860_MCHCFG 0x50 +#define I82860_GBA 0x60 +#define I82860_GBA_MASK 0x7FF +#define I82860_GBA_SHIFT 24 +#define I82860_ERRSTS 0xC8 +#define I82860_EAP 0xE4 +#define I82860_DERRCTL_STS 0xE2 + +enum i82860_chips { + I82860 = 0, +}; + +struct i82860_dev_info { + const char *ctl_name; +}; + +struct i82860_error_info { + u16 errsts; + u32 eap; + u16 derrsyn; + u16 errsts2; +}; + +static const struct i82860_dev_info i82860_devs[] = { + [I82860] = { + .ctl_name = "i82860"}, +}; + +static struct pci_dev *mci_pdev = NULL; /* init dev: in case that AGP code + has already registered driver */ + +static int i82860_registered = 1; + +static void i82860_get_error_info (struct mem_ctl_info *mci, + struct i82860_error_info *info) +{ + /* + * This is a mess because there is no atomic way to read all the + * registers at once and the registers can transition from CE being + * overwritten by UE. + */ + pci_read_config_word(mci->pdev, I82860_ERRSTS, &info->errsts); + pci_read_config_dword(mci->pdev, I82860_EAP, &info->eap); + pci_read_config_word(mci->pdev, I82860_DERRCTL_STS, &info->derrsyn); + pci_read_config_word(mci->pdev, I82860_ERRSTS, &info->errsts2); + + pci_write_bits16(mci->pdev, I82860_ERRSTS, 0x0003, 0x0003); + + /* + * If the error is the same for both reads then the first set of reads + * is valid. If there is a change then there is a CE no info and the + * second set of reads is valid and should be UE info. + */ + if (!(info->errsts2 & 0x0003)) + return; + if ((info->errsts ^ info->errsts2) & 0x0003) { + pci_read_config_dword(mci->pdev, I82860_EAP, &info->eap); + pci_read_config_word(mci->pdev, I82860_DERRCTL_STS, + &info->derrsyn); + } +} + +static int i82860_process_error_info (struct mem_ctl_info *mci, + struct i82860_error_info *info, int handle_errors) +{ + int row; + + if (!(info->errsts2 & 0x0003)) + return 0; + + if (!handle_errors) + return 1; + + if ((info->errsts ^ info->errsts2) & 0x0003) { + edac_mc_handle_ce_no_info(mci, "UE overwrote CE"); + info->errsts = info->errsts2; + } + + info->eap >>= PAGE_SHIFT; + row = edac_mc_find_csrow_by_page(mci, info->eap); + + if (info->errsts & 0x0002) + edac_mc_handle_ue(mci, info->eap, 0, row, "i82860 UE"); + else + edac_mc_handle_ce(mci, info->eap, 0, info->derrsyn, row, + 0, "i82860 UE"); + + return 1; +} + +static void i82860_check(struct mem_ctl_info *mci) +{ + struct i82860_error_info info; + + debugf1("MC%d: " __FILE__ ": %s()\n", mci->mc_idx, __func__); + i82860_get_error_info(mci, &info); + i82860_process_error_info(mci, &info, 1); +} + +static int i82860_probe1(struct pci_dev *pdev, int dev_idx) +{ + int rc = -ENODEV; + int index; + struct mem_ctl_info *mci = NULL; + unsigned long last_cumul_size; + + u16 mchcfg_ddim; /* DRAM Data Integrity Mode 0=none,2=edac */ + + /* RDRAM has channels but these don't map onto the abstractions that + edac uses. + The device groups from the GRA registers seem to map reasonably + well onto the notion of a chip select row. + There are 16 GRA registers and since the name is associated with + the channel and the GRA registers map to physical devices so we are + going to make 1 channel for group. + */ + mci = edac_mc_alloc(0, 16, 1); + if (!mci) + return -ENOMEM; + + debugf3("MC: " __FILE__ ": %s(): init mci\n", __func__); + + mci->pdev = pdev; + mci->mtype_cap = MEM_FLAG_DDR; + + + mci->edac_ctl_cap = EDAC_FLAG_NONE | EDAC_FLAG_SECDED; + /* I"m not sure about this but I think that all RDRAM is SECDED */ + mci->edac_cap = EDAC_FLAG_SECDED; + /* adjust FLAGS */ + + mci->mod_name = BS_MOD_STR; + mci->mod_ver = "$Revision: 1.1.2.6 $"; + mci->ctl_name = i82860_devs[dev_idx].ctl_name; + mci->edac_check = i82860_check; + mci->ctl_page_to_phys = NULL; + + pci_read_config_word(mci->pdev, I82860_MCHCFG, &mchcfg_ddim); + mchcfg_ddim = mchcfg_ddim & 0x180; + + /* + * The group row boundary (GRA) reg values are boundary address + * for each DRAM row with a granularity of 16MB. GRA regs are + * cumulative; therefore GRA15 will contain the total memory contained + * in all eight rows. + */ + for (last_cumul_size = index = 0; index < mci->nr_csrows; index++) { + u16 value; + u32 cumul_size; + struct csrow_info *csrow = &mci->csrows[index]; + + pci_read_config_word(mci->pdev, I82860_GBA + index * 2, + &value); + + cumul_size = (value & I82860_GBA_MASK) << + (I82860_GBA_SHIFT - PAGE_SHIFT); + debugf3("MC: " __FILE__ ": %s(): (%d) cumul_size 0x%x\n", + __func__, index, cumul_size); + if (cumul_size == last_cumul_size) + continue; /* not populated */ + + csrow->first_page = last_cumul_size; + csrow->last_page = cumul_size - 1; + csrow->nr_pages = cumul_size - last_cumul_size; + last_cumul_size = cumul_size; + csrow->grain = 1 << 12; /* I82860_EAP has 4KiB reolution */ + csrow->mtype = MEM_RMBS; + csrow->dtype = DEV_UNKNOWN; + csrow->edac_mode = mchcfg_ddim ? EDAC_SECDED : EDAC_NONE; + } + + /* clear counters */ + pci_write_bits16(mci->pdev, I82860_ERRSTS, 0x0003, 0x0003); + + if (edac_mc_add_mc(mci)) { + debugf3("MC: " __FILE__ + ": %s(): failed edac_mc_add_mc()\n", + __func__); + edac_mc_free(mci); + } else { + /* get this far and it's successful */ + debugf3("MC: " __FILE__ ": %s(): success\n", __func__); + rc = 0; + } + return rc; +} + +/* returns count (>= 0), or negative on error */ +static int __devinit i82860_init_one(struct pci_dev *pdev, + const struct pci_device_id *ent) +{ + int rc; + + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + printk(KERN_INFO "i82860 init one\n"); + if(pci_enable_device(pdev) < 0) + return -EIO; + rc = i82860_probe1(pdev, ent->driver_data); + if(rc == 0) + mci_pdev = pci_dev_get(pdev); + return rc; +} + +static void __devexit i82860_remove_one(struct pci_dev *pdev) +{ + struct mem_ctl_info *mci; + + debugf0(__FILE__ ": %s()\n", __func__); + + mci = edac_mc_find_mci_by_pdev(pdev); + if ((mci != NULL) && (edac_mc_del_mc(mci) == 0)) + edac_mc_free(mci); +} + +static const struct pci_device_id i82860_pci_tbl[] __devinitdata = { + {PCI_VEND_DEV(INTEL, 82860_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + I82860}, + {0,} /* 0 terminated list. */ +}; + +MODULE_DEVICE_TABLE(pci, i82860_pci_tbl); + +static struct pci_driver i82860_driver = { + .name = BS_MOD_STR, + .probe = i82860_init_one, + .remove = __devexit_p(i82860_remove_one), + .id_table = i82860_pci_tbl, +}; + +int __init i82860_init(void) +{ + int pci_rc; + + debugf3("MC: " __FILE__ ": %s()\n", __func__); + if ((pci_rc = pci_register_driver(&i82860_driver)) < 0) + return pci_rc; + + if (!mci_pdev) { + i82860_registered = 0; + mci_pdev = pci_get_device(PCI_VENDOR_ID_INTEL, + PCI_DEVICE_ID_INTEL_82860_0, NULL); + if (mci_pdev == NULL) { + debugf0("860 pci_get_device fail\n"); + return -ENODEV; + } + pci_rc = i82860_init_one(mci_pdev, i82860_pci_tbl); + if (pci_rc < 0) { + debugf0("860 init fail\n"); + pci_dev_put(mci_pdev); + return -ENODEV; + } + } + return 0; +} + +static void __exit i82860_exit(void) +{ + debugf3("MC: " __FILE__ ": %s()\n", __func__); + + pci_unregister_driver(&i82860_driver); + if (!i82860_registered) { + i82860_remove_one(mci_pdev); + pci_dev_put(mci_pdev); + } +} + +module_init(i82860_init); +module_exit(i82860_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR + ("Red Hat Inc. (http://www.redhat.com.com) Ben Woodard "); +MODULE_DESCRIPTION("ECC support for Intel 82860 memory hub controllers"); diff --git a/drivers/edac/i82875p_edac.c b/drivers/edac/i82875p_edac.c new file mode 100644 index 00000000000..79d14dfbcbd --- /dev/null +++ b/drivers/edac/i82875p_edac.c @@ -0,0 +1,532 @@ +/* + * Intel D82875P Memory Controller kernel module + * (C) 2003 Linux Networx (http://lnxi.com) + * This file may be distributed under the terms of the + * GNU General Public License. + * + * Written by Thayne Harbaugh + * Contributors: + * Wang Zhenyu at intel.com + * + * $Id: edac_i82875p.c,v 1.5.2.11 2005/10/05 00:43:44 dsp_llnl Exp $ + * + * Note: E7210 appears same as D82875P - zhenyu.z.wang at intel.com + */ + + +#include +#include +#include + +#include +#include + +#include + +#include "edac_mc.h" + + +#ifndef PCI_DEVICE_ID_INTEL_82875_0 +#define PCI_DEVICE_ID_INTEL_82875_0 0x2578 +#endif /* PCI_DEVICE_ID_INTEL_82875_0 */ + +#ifndef PCI_DEVICE_ID_INTEL_82875_6 +#define PCI_DEVICE_ID_INTEL_82875_6 0x257e +#endif /* PCI_DEVICE_ID_INTEL_82875_6 */ + + +/* four csrows in dual channel, eight in single channel */ +#define I82875P_NR_CSROWS(nr_chans) (8/(nr_chans)) + + +/* Intel 82875p register addresses - device 0 function 0 - DRAM Controller */ +#define I82875P_EAP 0x58 /* Error Address Pointer (32b) + * + * 31:12 block address + * 11:0 reserved + */ + +#define I82875P_DERRSYN 0x5c /* DRAM Error Syndrome (8b) + * + * 7:0 DRAM ECC Syndrome + */ + +#define I82875P_DES 0x5d /* DRAM Error Status (8b) + * + * 7:1 reserved + * 0 Error channel 0/1 + */ + +#define I82875P_ERRSTS 0xc8 /* Error Status Register (16b) + * + * 15:10 reserved + * 9 non-DRAM lock error (ndlock) + * 8 Sftwr Generated SMI + * 7 ECC UE + * 6 reserved + * 5 MCH detects unimplemented cycle + * 4 AGP access outside GA + * 3 Invalid AGP access + * 2 Invalid GA translation table + * 1 Unsupported AGP command + * 0 ECC CE + */ + +#define I82875P_ERRCMD 0xca /* Error Command (16b) + * + * 15:10 reserved + * 9 SERR on non-DRAM lock + * 8 SERR on ECC UE + * 7 SERR on ECC CE + * 6 target abort on high exception + * 5 detect unimplemented cyc + * 4 AGP access outside of GA + * 3 SERR on invalid AGP access + * 2 invalid translation table + * 1 SERR on unsupported AGP command + * 0 reserved + */ + + +/* Intel 82875p register addresses - device 6 function 0 - DRAM Controller */ +#define I82875P_PCICMD6 0x04 /* PCI Command Register (16b) + * + * 15:10 reserved + * 9 fast back-to-back - ro 0 + * 8 SERR enable - ro 0 + * 7 addr/data stepping - ro 0 + * 6 parity err enable - ro 0 + * 5 VGA palette snoop - ro 0 + * 4 mem wr & invalidate - ro 0 + * 3 special cycle - ro 0 + * 2 bus master - ro 0 + * 1 mem access dev6 - 0(dis),1(en) + * 0 IO access dev3 - 0(dis),1(en) + */ + +#define I82875P_BAR6 0x10 /* Mem Delays Base ADDR Reg (32b) + * + * 31:12 mem base addr [31:12] + * 11:4 address mask - ro 0 + * 3 prefetchable - ro 0(non),1(pre) + * 2:1 mem type - ro 0 + * 0 mem space - ro 0 + */ + +/* Intel 82875p MMIO register space - device 0 function 0 - MMR space */ + +#define I82875P_DRB_SHIFT 26 /* 64MiB grain */ +#define I82875P_DRB 0x00 /* DRAM Row Boundary (8b x 8) + * + * 7 reserved + * 6:0 64MiB row boundary addr + */ + +#define I82875P_DRA 0x10 /* DRAM Row Attribute (4b x 8) + * + * 7 reserved + * 6:4 row attr row 1 + * 3 reserved + * 2:0 row attr row 0 + * + * 000 = 4KiB + * 001 = 8KiB + * 010 = 16KiB + * 011 = 32KiB + */ + +#define I82875P_DRC 0x68 /* DRAM Controller Mode (32b) + * + * 31:30 reserved + * 29 init complete + * 28:23 reserved + * 22:21 nr chan 00=1,01=2 + * 20 reserved + * 19:18 Data Integ Mode 00=none,01=ecc + * 17:11 reserved + * 10:8 refresh mode + * 7 reserved + * 6:4 mode select + * 3:2 reserved + * 1:0 DRAM type 01=DDR + */ + + +enum i82875p_chips { + I82875P = 0, +}; + + +struct i82875p_pvt { + struct pci_dev *ovrfl_pdev; + void *ovrfl_window; +}; + + +struct i82875p_dev_info { + const char *ctl_name; +}; + + +struct i82875p_error_info { + u16 errsts; + u32 eap; + u8 des; + u8 derrsyn; + u16 errsts2; +}; + + +static const struct i82875p_dev_info i82875p_devs[] = { + [I82875P] = { + .ctl_name = "i82875p"}, +}; + +static struct pci_dev *mci_pdev = NULL; /* init dev: in case that AGP code + has already registered driver */ +static int i82875p_registered = 1; + +static void i82875p_get_error_info (struct mem_ctl_info *mci, + struct i82875p_error_info *info) +{ + /* + * This is a mess because there is no atomic way to read all the + * registers at once and the registers can transition from CE being + * overwritten by UE. + */ + pci_read_config_word(mci->pdev, I82875P_ERRSTS, &info->errsts); + pci_read_config_dword(mci->pdev, I82875P_EAP, &info->eap); + pci_read_config_byte(mci->pdev, I82875P_DES, &info->des); + pci_read_config_byte(mci->pdev, I82875P_DERRSYN, &info->derrsyn); + pci_read_config_word(mci->pdev, I82875P_ERRSTS, &info->errsts2); + + pci_write_bits16(mci->pdev, I82875P_ERRSTS, 0x0081, 0x0081); + + /* + * If the error is the same then we can for both reads then + * the first set of reads is valid. If there is a change then + * there is a CE no info and the second set of reads is valid + * and should be UE info. + */ + if (!(info->errsts2 & 0x0081)) + return; + if ((info->errsts ^ info->errsts2) & 0x0081) { + pci_read_config_dword(mci->pdev, I82875P_EAP, &info->eap); + pci_read_config_byte(mci->pdev, I82875P_DES, &info->des); + pci_read_config_byte(mci->pdev, I82875P_DERRSYN, + &info->derrsyn); + } +} + +static int i82875p_process_error_info (struct mem_ctl_info *mci, + struct i82875p_error_info *info, int handle_errors) +{ + int row, multi_chan; + + multi_chan = mci->csrows[0].nr_channels - 1; + + if (!(info->errsts2 & 0x0081)) + return 0; + + if (!handle_errors) + return 1; + + if ((info->errsts ^ info->errsts2) & 0x0081) { + edac_mc_handle_ce_no_info(mci, "UE overwrote CE"); + info->errsts = info->errsts2; + } + + info->eap >>= PAGE_SHIFT; + row = edac_mc_find_csrow_by_page(mci, info->eap); + + if (info->errsts & 0x0080) + edac_mc_handle_ue(mci, info->eap, 0, row, "i82875p UE"); + else + edac_mc_handle_ce(mci, info->eap, 0, info->derrsyn, row, + multi_chan ? (info->des & 0x1) : 0, + "i82875p CE"); + + return 1; +} + + +static void i82875p_check(struct mem_ctl_info *mci) +{ + struct i82875p_error_info info; + + debugf1("MC%d: " __FILE__ ": %s()\n", mci->mc_idx, __func__); + i82875p_get_error_info(mci, &info); + i82875p_process_error_info(mci, &info, 1); +} + + +#ifdef CONFIG_PROC_FS +extern int pci_proc_attach_device(struct pci_dev *); +#endif + +static int i82875p_probe1(struct pci_dev *pdev, int dev_idx) +{ + int rc = -ENODEV; + int index; + struct mem_ctl_info *mci = NULL; + struct i82875p_pvt *pvt = NULL; + unsigned long last_cumul_size; + struct pci_dev *ovrfl_pdev; + void __iomem *ovrfl_window = NULL; + + u32 drc; + u32 drc_chan; /* Number of channels 0=1chan,1=2chan */ + u32 nr_chans; + u32 drc_ddim; /* DRAM Data Integrity Mode 0=none,2=edac */ + + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + ovrfl_pdev = pci_find_device(PCI_VEND_DEV(INTEL, 82875_6), NULL); + + if (!ovrfl_pdev) { + /* + * Intel tells BIOS developers to hide device 6 which + * configures the overflow device access containing + * the DRBs - this is where we expose device 6. + * http://www.x86-secret.com/articles/tweak/pat/patsecrets-2.htm + */ + pci_write_bits8(pdev, 0xf4, 0x2, 0x2); + ovrfl_pdev = + pci_scan_single_device(pdev->bus, PCI_DEVFN(6, 0)); + if (!ovrfl_pdev) + goto fail; + } +#ifdef CONFIG_PROC_FS + if (!ovrfl_pdev->procent && pci_proc_attach_device(ovrfl_pdev)) { + printk(KERN_ERR "MC: " __FILE__ + ": %s(): Failed to attach overflow device\n", + __func__); + goto fail; + } +#endif /* CONFIG_PROC_FS */ + if (pci_enable_device(ovrfl_pdev)) { + printk(KERN_ERR "MC: " __FILE__ + ": %s(): Failed to enable overflow device\n", + __func__); + goto fail; + } + + if (pci_request_regions(ovrfl_pdev, pci_name(ovrfl_pdev))) { +#ifdef CORRECT_BIOS + goto fail; +#endif + } + /* cache is irrelevant for PCI bus reads/writes */ + ovrfl_window = ioremap_nocache(pci_resource_start(ovrfl_pdev, 0), + pci_resource_len(ovrfl_pdev, 0)); + + if (!ovrfl_window) { + printk(KERN_ERR "MC: " __FILE__ + ": %s(): Failed to ioremap bar6\n", __func__); + goto fail; + } + + /* need to find out the number of channels */ + drc = readl(ovrfl_window + I82875P_DRC); + drc_chan = ((drc >> 21) & 0x1); + nr_chans = drc_chan + 1; + drc_ddim = (drc >> 18) & 0x1; + + mci = edac_mc_alloc(sizeof(*pvt), I82875P_NR_CSROWS(nr_chans), + nr_chans); + + if (!mci) { + rc = -ENOMEM; + goto fail; + } + + debugf3("MC: " __FILE__ ": %s(): init mci\n", __func__); + + mci->pdev = pdev; + mci->mtype_cap = MEM_FLAG_DDR; + + mci->edac_ctl_cap = EDAC_FLAG_NONE | EDAC_FLAG_SECDED; + mci->edac_cap = EDAC_FLAG_UNKNOWN; + /* adjust FLAGS */ + + mci->mod_name = BS_MOD_STR; + mci->mod_ver = "$Revision: 1.5.2.11 $"; + mci->ctl_name = i82875p_devs[dev_idx].ctl_name; + mci->edac_check = i82875p_check; + mci->ctl_page_to_phys = NULL; + + debugf3("MC: " __FILE__ ": %s(): init pvt\n", __func__); + + pvt = (struct i82875p_pvt *) mci->pvt_info; + pvt->ovrfl_pdev = ovrfl_pdev; + pvt->ovrfl_window = ovrfl_window; + + /* + * The dram row boundary (DRB) reg values are boundary address + * for each DRAM row with a granularity of 32 or 64MB (single/dual + * channel operation). DRB regs are cumulative; therefore DRB7 will + * contain the total memory contained in all eight rows. + */ + for (last_cumul_size = index = 0; index < mci->nr_csrows; index++) { + u8 value; + u32 cumul_size; + struct csrow_info *csrow = &mci->csrows[index]; + + value = readb(ovrfl_window + I82875P_DRB + index); + cumul_size = value << (I82875P_DRB_SHIFT - PAGE_SHIFT); + debugf3("MC: " __FILE__ ": %s(): (%d) cumul_size 0x%x\n", + __func__, index, cumul_size); + if (cumul_size == last_cumul_size) + continue; /* not populated */ + + csrow->first_page = last_cumul_size; + csrow->last_page = cumul_size - 1; + csrow->nr_pages = cumul_size - last_cumul_size; + last_cumul_size = cumul_size; + csrow->grain = 1 << 12; /* I82875P_EAP has 4KiB reolution */ + csrow->mtype = MEM_DDR; + csrow->dtype = DEV_UNKNOWN; + csrow->edac_mode = drc_ddim ? EDAC_SECDED : EDAC_NONE; + } + + /* clear counters */ + pci_write_bits16(mci->pdev, I82875P_ERRSTS, 0x0081, 0x0081); + + if (edac_mc_add_mc(mci)) { + debugf3("MC: " __FILE__ + ": %s(): failed edac_mc_add_mc()\n", __func__); + goto fail; + } + + /* get this far and it's successful */ + debugf3("MC: " __FILE__ ": %s(): success\n", __func__); + return 0; + + fail: + if (mci) + edac_mc_free(mci); + + if (ovrfl_window) + iounmap(ovrfl_window); + + if (ovrfl_pdev) { + pci_release_regions(ovrfl_pdev); + pci_disable_device(ovrfl_pdev); + } + + /* NOTE: the ovrfl proc entry and pci_dev are intentionally left */ + return rc; +} + + +/* returns count (>= 0), or negative on error */ +static int __devinit i82875p_init_one(struct pci_dev *pdev, + const struct pci_device_id *ent) +{ + int rc; + + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + printk(KERN_INFO "i82875p init one\n"); + if(pci_enable_device(pdev) < 0) + return -EIO; + rc = i82875p_probe1(pdev, ent->driver_data); + if (mci_pdev == NULL) + mci_pdev = pci_dev_get(pdev); + return rc; +} + + +static void __devexit i82875p_remove_one(struct pci_dev *pdev) +{ + struct mem_ctl_info *mci; + struct i82875p_pvt *pvt = NULL; + + debugf0(__FILE__ ": %s()\n", __func__); + + if ((mci = edac_mc_find_mci_by_pdev(pdev)) == NULL) + return; + + pvt = (struct i82875p_pvt *) mci->pvt_info; + if (pvt->ovrfl_window) + iounmap(pvt->ovrfl_window); + + if (pvt->ovrfl_pdev) { +#ifdef CORRECT_BIOS + pci_release_regions(pvt->ovrfl_pdev); +#endif /*CORRECT_BIOS */ + pci_disable_device(pvt->ovrfl_pdev); + pci_dev_put(pvt->ovrfl_pdev); + } + + if (edac_mc_del_mc(mci)) + return; + + edac_mc_free(mci); +} + + +static const struct pci_device_id i82875p_pci_tbl[] __devinitdata = { + {PCI_VEND_DEV(INTEL, 82875_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + I82875P}, + {0,} /* 0 terminated list. */ +}; + +MODULE_DEVICE_TABLE(pci, i82875p_pci_tbl); + + +static struct pci_driver i82875p_driver = { + .name = BS_MOD_STR, + .probe = i82875p_init_one, + .remove = __devexit_p(i82875p_remove_one), + .id_table = i82875p_pci_tbl, +}; + + +int __init i82875p_init(void) +{ + int pci_rc; + + debugf3("MC: " __FILE__ ": %s()\n", __func__); + pci_rc = pci_register_driver(&i82875p_driver); + if (pci_rc < 0) + return pci_rc; + if (mci_pdev == NULL) { + i82875p_registered = 0; + mci_pdev = + pci_get_device(PCI_VENDOR_ID_INTEL, + PCI_DEVICE_ID_INTEL_82875_0, NULL); + if (!mci_pdev) { + debugf0("875p pci_get_device fail\n"); + return -ENODEV; + } + pci_rc = i82875p_init_one(mci_pdev, i82875p_pci_tbl); + if (pci_rc < 0) { + debugf0("875p init fail\n"); + pci_dev_put(mci_pdev); + return -ENODEV; + } + } + return 0; +} + + +static void __exit i82875p_exit(void) +{ + debugf3("MC: " __FILE__ ": %s()\n", __func__); + + pci_unregister_driver(&i82875p_driver); + if (!i82875p_registered) { + i82875p_remove_one(mci_pdev); + pci_dev_put(mci_pdev); + } +} + + +module_init(i82875p_init); +module_exit(i82875p_exit); + + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Linux Networx (http://lnxi.com) Thayne Harbaugh"); +MODULE_DESCRIPTION("MC support for Intel 82875 memory hub controllers"); -- cgit v1.2.3 From 2f768af73fea4c70f9046388a7ff648ad11f028e Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 18 Jan 2006 17:44:12 -0800 Subject: [PATCH] EDAC: drivers for Radisys 82600 Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/edac/r82600_edac.c | 407 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 407 insertions(+) create mode 100644 drivers/edac/r82600_edac.c diff --git a/drivers/edac/r82600_edac.c b/drivers/edac/r82600_edac.c new file mode 100644 index 00000000000..b6399a54256 --- /dev/null +++ b/drivers/edac/r82600_edac.c @@ -0,0 +1,407 @@ +/* + * Radisys 82600 Embedded chipset Memory Controller kernel module + * (C) 2005 EADS Astrium + * This file may be distributed under the terms of the + * GNU General Public License. + * + * Written by Tim Small , based on work by Thayne + * Harbaugh, Dan Hollis and others. + * + * $Id: edac_r82600.c,v 1.1.2.6 2005/10/05 00:43:44 dsp_llnl Exp $ + * + * Written with reference to 82600 High Integration Dual PCI System + * Controller Data Book: + * http://www.radisys.com/files/support_downloads/007-01277-0002.82600DataBook.pdf + * references to this document given in [] + */ + +#include +#include +#include + +#include +#include + +#include + +#include "edac_mc.h" + +/* Radisys say "The 82600 integrates a main memory SDRAM controller that + * supports up to four banks of memory. The four banks can support a mix of + * sizes of 64 bit wide (72 bits with ECC) Synchronous DRAM (SDRAM) DIMMs, + * each of which can be any size from 16MB to 512MB. Both registered (control + * signals buffered) and unbuffered DIMM types are supported. Mixing of + * registered and unbuffered DIMMs as well as mixing of ECC and non-ECC DIMMs + * is not allowed. The 82600 SDRAM interface operates at the same frequency as + * the CPU bus, 66MHz, 100MHz or 133MHz." + */ + +#define R82600_NR_CSROWS 4 +#define R82600_NR_CHANS 1 +#define R82600_NR_DIMMS 4 + +#define R82600_BRIDGE_ID 0x8200 + +/* Radisys 82600 register addresses - device 0 function 0 - PCI bridge */ +#define R82600_DRAMC 0x57 /* Various SDRAM related control bits + * all bits are R/W + * + * 7 SDRAM ISA Hole Enable + * 6 Flash Page Mode Enable + * 5 ECC Enable: 1=ECC 0=noECC + * 4 DRAM DIMM Type: 1= + * 3 BIOS Alias Disable + * 2 SDRAM BIOS Flash Write Enable + * 1:0 SDRAM Refresh Rate: 00=Disabled + * 01=7.8usec (256Mbit SDRAMs) + * 10=15.6us 11=125usec + */ + +#define R82600_SDRAMC 0x76 /* "SDRAM Control Register" + * More SDRAM related control bits + * all bits are R/W + * + * 15:8 Reserved. + * + * 7:5 Special SDRAM Mode Select + * + * 4 Force ECC + * + * 1=Drive ECC bits to 0 during + * write cycles (i.e. ECC test mode) + * + * 0=Normal ECC functioning + * + * 3 Enhanced Paging Enable + * + * 2 CAS# Latency 0=3clks 1=2clks + * + * 1 RAS# to CAS# Delay 0=3 1=2 + * + * 0 RAS# Precharge 0=3 1=2 + */ + +#define R82600_EAP 0x80 /* ECC Error Address Pointer Register + * + * 31 Disable Hardware Scrubbing (RW) + * 0=Scrub on corrected read + * 1=Don't scrub on corrected read + * + * 30:12 Error Address Pointer (RO) + * Upper 19 bits of error address + * + * 11:4 Syndrome Bits (RO) + * + * 3 BSERR# on multibit error (RW) + * 1=enable 0=disable + * + * 2 NMI on Single Bit Eror (RW) + * 1=NMI triggered by SBE n.b. other + * prerequeists + * 0=NMI not triggered + * + * 1 MBE (R/WC) + * read 1=MBE at EAP (see above) + * read 0=no MBE, or SBE occurred first + * write 1=Clear MBE status (must also + * clear SBE) + * write 0=NOP + * + * 1 SBE (R/WC) + * read 1=SBE at EAP (see above) + * read 0=no SBE, or MBE occurred first + * write 1=Clear SBE status (must also + * clear MBE) + * write 0=NOP + */ + +#define R82600_DRBA 0x60 /* + 0x60..0x63 SDRAM Row Boundry Address + * Registers + * + * 7:0 Address lines 30:24 - upper limit of + * each row [p57] + */ + +struct r82600_error_info { + u32 eapr; +}; + + +static unsigned int disable_hardware_scrub = 0; + + +static void r82600_get_error_info (struct mem_ctl_info *mci, + struct r82600_error_info *info) +{ + pci_read_config_dword(mci->pdev, R82600_EAP, &info->eapr); + + if (info->eapr & BIT(0)) + /* Clear error to allow next error to be reported [p.62] */ + pci_write_bits32(mci->pdev, R82600_EAP, + ((u32) BIT(0) & (u32) BIT(1)), + ((u32) BIT(0) & (u32) BIT(1))); + + if (info->eapr & BIT(1)) + /* Clear error to allow next error to be reported [p.62] */ + pci_write_bits32(mci->pdev, R82600_EAP, + ((u32) BIT(0) & (u32) BIT(1)), + ((u32) BIT(0) & (u32) BIT(1))); +} + + +static int r82600_process_error_info (struct mem_ctl_info *mci, + struct r82600_error_info *info, int handle_errors) +{ + int error_found; + u32 eapaddr, page; + u32 syndrome; + + error_found = 0; + + /* bits 30:12 store the upper 19 bits of the 32 bit error address */ + eapaddr = ((info->eapr >> 12) & 0x7FFF) << 13; + /* Syndrome in bits 11:4 [p.62] */ + syndrome = (info->eapr >> 4) & 0xFF; + + /* the R82600 reports at less than page * + * granularity (upper 19 bits only) */ + page = eapaddr >> PAGE_SHIFT; + + if (info->eapr & BIT(0)) { /* CE? */ + error_found = 1; + + if (handle_errors) + edac_mc_handle_ce( + mci, page, 0, /* not avail */ + syndrome, + edac_mc_find_csrow_by_page(mci, page), + 0, /* channel */ + mci->ctl_name); + } + + if (info->eapr & BIT(1)) { /* UE? */ + error_found = 1; + + if (handle_errors) + /* 82600 doesn't give enough info */ + edac_mc_handle_ue(mci, page, 0, + edac_mc_find_csrow_by_page(mci, page), + mci->ctl_name); + } + + return error_found; +} + +static void r82600_check(struct mem_ctl_info *mci) +{ + struct r82600_error_info info; + + debugf1("MC%d: " __FILE__ ": %s()\n", mci->mc_idx, __func__); + r82600_get_error_info(mci, &info); + r82600_process_error_info(mci, &info, 1); +} + +static int r82600_probe1(struct pci_dev *pdev, int dev_idx) +{ + int rc = -ENODEV; + int index; + struct mem_ctl_info *mci = NULL; + u8 dramcr; + u32 ecc_on; + u32 reg_sdram; + u32 eapr; + u32 scrub_disabled; + u32 sdram_refresh_rate; + u32 row_high_limit_last = 0; + u32 eap_init_bits; + + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + + pci_read_config_byte(pdev, R82600_DRAMC, &dramcr); + pci_read_config_dword(pdev, R82600_EAP, &eapr); + + ecc_on = dramcr & BIT(5); + reg_sdram = dramcr & BIT(4); + scrub_disabled = eapr & BIT(31); + sdram_refresh_rate = dramcr & (BIT(0) | BIT(1)); + + debugf2("MC: " __FILE__ ": %s(): sdram refresh rate = %#0x\n", + __func__, sdram_refresh_rate); + + debugf2("MC: " __FILE__ ": %s(): DRAMC register = %#0x\n", __func__, + dramcr); + + mci = edac_mc_alloc(0, R82600_NR_CSROWS, R82600_NR_CHANS); + + if (mci == NULL) { + rc = -ENOMEM; + goto fail; + } + + debugf0("MC: " __FILE__ ": %s(): mci = %p\n", __func__, mci); + + mci->pdev = pdev; + mci->mtype_cap = MEM_FLAG_RDDR | MEM_FLAG_DDR; + + mci->edac_ctl_cap = EDAC_FLAG_NONE | EDAC_FLAG_EC | EDAC_FLAG_SECDED; + /* FIXME try to work out if the chip leads have been * + * used for COM2 instead on this board? [MA6?] MAYBE: */ + + /* On the R82600, the pins for memory bits 72:65 - i.e. the * + * EC bits are shared with the pins for COM2 (!), so if COM2 * + * is enabled, we assume COM2 is wired up, and thus no EDAC * + * is possible. */ + mci->edac_cap = EDAC_FLAG_NONE | EDAC_FLAG_EC | EDAC_FLAG_SECDED; + if (ecc_on) { + if (scrub_disabled) + debugf3("MC: " __FILE__ ": %s(): mci = %p - " + "Scrubbing disabled! EAP: %#0x\n", __func__, + mci, eapr); + } else + mci->edac_cap = EDAC_FLAG_NONE; + + mci->mod_name = BS_MOD_STR; + mci->mod_ver = "$Revision: 1.1.2.6 $"; + mci->ctl_name = "R82600"; + mci->edac_check = r82600_check; + mci->ctl_page_to_phys = NULL; + + for (index = 0; index < mci->nr_csrows; index++) { + struct csrow_info *csrow = &mci->csrows[index]; + u8 drbar; /* sDram Row Boundry Address Register */ + u32 row_high_limit; + u32 row_base; + + /* find the DRAM Chip Select Base address and mask */ + pci_read_config_byte(mci->pdev, R82600_DRBA + index, &drbar); + + debugf1("MC%d: " __FILE__ ": %s() Row=%d DRBA = %#0x\n", + mci->mc_idx, __func__, index, drbar); + + row_high_limit = ((u32) drbar << 24); +/* row_high_limit = ((u32)drbar << 24) | 0xffffffUL; */ + + debugf1("MC%d: " __FILE__ ": %s() Row=%d, " + "Boundry Address=%#0x, Last = %#0x \n", + mci->mc_idx, __func__, index, row_high_limit, + row_high_limit_last); + + /* Empty row [p.57] */ + if (row_high_limit == row_high_limit_last) + continue; + + row_base = row_high_limit_last; + + csrow->first_page = row_base >> PAGE_SHIFT; + csrow->last_page = (row_high_limit >> PAGE_SHIFT) - 1; + csrow->nr_pages = csrow->last_page - csrow->first_page + 1; + /* Error address is top 19 bits - so granularity is * + * 14 bits */ + csrow->grain = 1 << 14; + csrow->mtype = reg_sdram ? MEM_RDDR : MEM_DDR; + /* FIXME - check that this is unknowable with this chipset */ + csrow->dtype = DEV_UNKNOWN; + + /* Mode is global on 82600 */ + csrow->edac_mode = ecc_on ? EDAC_SECDED : EDAC_NONE; + row_high_limit_last = row_high_limit; + } + + /* clear counters */ + /* FIXME should we? */ + + if (edac_mc_add_mc(mci)) { + debugf3("MC: " __FILE__ + ": %s(): failed edac_mc_add_mc()\n", __func__); + goto fail; + } + + /* get this far and it's successful */ + + /* Clear error flags to allow next error to be reported [p.62] */ + /* Test systems seem to always have the UE flag raised on boot */ + + eap_init_bits = BIT(0) & BIT(1); + if (disable_hardware_scrub) { + eap_init_bits |= BIT(31); + debugf3("MC: " __FILE__ ": %s(): Disabling Hardware Scrub " + "(scrub on error)\n", __func__); + } + + pci_write_bits32(mci->pdev, R82600_EAP, eap_init_bits, + eap_init_bits); + + debugf3("MC: " __FILE__ ": %s(): success\n", __func__); + return 0; + +fail: + if (mci) + edac_mc_free(mci); + + return rc; +} + +/* returns count (>= 0), or negative on error */ +static int __devinit r82600_init_one(struct pci_dev *pdev, + const struct pci_device_id *ent) +{ + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + /* don't need to call pci_device_enable() */ + return r82600_probe1(pdev, ent->driver_data); +} + + +static void __devexit r82600_remove_one(struct pci_dev *pdev) +{ + struct mem_ctl_info *mci; + + debugf0(__FILE__ ": %s()\n", __func__); + + if (((mci = edac_mc_find_mci_by_pdev(pdev)) != NULL) && + !edac_mc_del_mc(mci)) + edac_mc_free(mci); +} + + +static const struct pci_device_id r82600_pci_tbl[] __devinitdata = { + {PCI_DEVICE(PCI_VENDOR_ID_RADISYS, R82600_BRIDGE_ID)}, + {0,} /* 0 terminated list. */ +}; + +MODULE_DEVICE_TABLE(pci, r82600_pci_tbl); + + +static struct pci_driver r82600_driver = { + .name = BS_MOD_STR, + .probe = r82600_init_one, + .remove = __devexit_p(r82600_remove_one), + .id_table = r82600_pci_tbl, +}; + + +int __init r82600_init(void) +{ + return pci_register_driver(&r82600_driver); +} + + +static void __exit r82600_exit(void) +{ + pci_unregister_driver(&r82600_driver); +} + + +module_init(r82600_init); +module_exit(r82600_exit); + + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Tim Small - WPAD Ltd. " + "on behalf of EADS Astrium"); +MODULE_DESCRIPTION("MC support for Radisys 82600 memory controllers"); + +module_param(disable_hardware_scrub, bool, 0644); +MODULE_PARM_DESC(disable_hardware_scrub, + "If set, disable the chipset's automatic scrub for CEs"); -- cgit v1.2.3 From da9bb1d27b21cb24cbb6a2efb5d3c464d357a01e Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 18 Jan 2006 17:44:13 -0800 Subject: [PATCH] EDAC: core EDAC support code This is a subset of the bluesmoke project core code, stripped of the NMI work which isn't ready to merge and some of the "interesting" proc functionality that needs reworking or just has no place in kernel. It requires no core kernel changes except the added scrub functions already posted. The goal is to merge further functionality only after the core code is accepted and proven in the base kernel, and only at the point the upstream extras are really ready to merge. From: doug thompson This converts EDAC to sysfs and is the final chunk neccessary before EDAC has a stable user space API and can be considered for submission into the base kernel. Signed-off-by: Alan Cox Signed-off-by: Adrian Bunk Signed-off-by: Jesper Juhl Signed-off-by: doug thompson Signed-off-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/drivers/edac/edac.txt | 673 +++++++++++ MAINTAINERS | 9 + arch/i386/kernel/quirks.c | 9 +- drivers/Kconfig | 2 + drivers/Makefile | 1 + drivers/edac/Kconfig | 102 ++ drivers/edac/Makefile | 18 + drivers/edac/amd76x_edac.c | 2 +- drivers/edac/e752x_edac.c | 14 +- drivers/edac/e7xxx_edac.c | 2 +- drivers/edac/edac_mc.c | 2209 +++++++++++++++++++++++++++++++++++ drivers/edac/edac_mc.h | 448 +++++++ drivers/edac/i82860_edac.c | 2 +- drivers/edac/i82875p_edac.c | 2 +- drivers/edac/r82600_edac.c | 2 +- include/asm-i386/atomic.h | 12 - include/asm-i386/edac.h | 18 + include/asm-x86_64/atomic.h | 12 - include/asm-x86_64/edac.h | 18 + 19 files changed, 3515 insertions(+), 40 deletions(-) create mode 100644 Documentation/drivers/edac/edac.txt create mode 100644 drivers/edac/Kconfig create mode 100644 drivers/edac/Makefile create mode 100644 drivers/edac/edac_mc.c create mode 100644 drivers/edac/edac_mc.h create mode 100644 include/asm-i386/edac.h create mode 100644 include/asm-x86_64/edac.h diff --git a/Documentation/drivers/edac/edac.txt b/Documentation/drivers/edac/edac.txt new file mode 100644 index 00000000000..d37191fe568 --- /dev/null +++ b/Documentation/drivers/edac/edac.txt @@ -0,0 +1,673 @@ + + +EDAC - Error Detection And Correction + +Written by Doug Thompson +7 Dec 2005 + + +EDAC was written by: + Thayne Harbaugh, + modified by Dave Peterson, Doug Thompson, et al, + from the bluesmoke.sourceforge.net project. + + +============================================================================ +EDAC PURPOSE + +The 'edac' kernel module goal is to detect and report errors that occur +within the computer system. In the initial release, memory Correctable Errors +(CE) and Uncorrectable Errors (UE) are the primary errors being harvested. + +Detecting CE events, then harvesting those events and reporting them, +CAN be a predictor of future UE events. With CE events, the system can +continue to operate, but with less safety. Preventive maintainence and +proactive part replacement of memory DIMMs exhibiting CEs can reduce +the likelihood of the dreaded UE events and system 'panics'. + + +In addition, PCI Bus Parity and SERR Errors are scanned for on PCI devices +in order to determine if errors are occurring on data transfers. +The presence of PCI Parity errors must be examined with a grain of salt. +There are several addin adapters that do NOT follow the PCI specification +with regards to Parity generation and reporting. The specification says +the vendor should tie the parity status bits to 0 if they do not intend +to generate parity. Some vendors do not do this, and thus the parity bit +can "float" giving false positives. + +The PCI Parity EDAC device has the ability to "skip" known flakey +cards during the parity scan. These are set by the parity "blacklist" +interface in the sysfs for PCI Parity. (See the PCI section in the sysfs +section below.) There is also a parity "whitelist" which is used as +an explicit list of devices to scan, while the blacklist is a list +of devices to skip. + +EDAC will have future error detectors that will be added or integrated +into EDAC in the following list: + + MCE Machine Check Exception + MCA Machine Check Architecture + NMI NMI notification of ECC errors + MSRs Machine Specific Register error cases + and other mechanisms. + +These errors are usually bus errors, ECC errors, thermal throttling +and the like. + + +============================================================================ +EDAC VERSIONING + +EDAC is composed of a "core" module (edac_mc.ko) and several Memory +Controller (MC) driver modules. On a given system, the CORE +is loaded and one MC driver will be loaded. Both the CORE and +the MC driver have individual versions that reflect current release +level of their respective modules. Thus, to "report" on what version +a system is running, one must report both the CORE's and the +MC driver's versions. + + +LOADING + +If 'edac' was statically linked with the kernel then no loading is +necessary. If 'edac' was built as modules then simply modprobe the +'edac' pieces that you need. You should be able to modprobe +hardware-specific modules and have the dependencies load the necessary core +modules. + +Example: + +$> modprobe amd76x_edac + +loads both the amd76x_edac.ko memory controller module and the edac_mc.ko +core module. + + +============================================================================ +EDAC sysfs INTERFACE + +EDAC presents a 'sysfs' interface for control, reporting and attribute +reporting purposes. + +EDAC lives in the /sys/devices/system/edac directory. Within this directory +there currently reside 2 'edac' components: + + mc memory controller(s) system + pci PCI status system + + +============================================================================ +Memory Controller (mc) Model + +First a background on the memory controller's model abstracted in EDAC. +Each mc device controls a set of DIMM memory modules. These modules are +layed out in a Chip-Select Row (csrowX) and Channel table (chX). There can +be multiple csrows and two channels. + +Memory controllers allow for several csrows, with 8 csrows being a typical value. +Yet, the actual number of csrows depends on the electrical "loading" +of a given motherboard, memory controller and DIMM characteristics. + +Dual channels allows for 128 bit data transfers to the CPU from memory. + + + Channel 0 Channel 1 + =================================== + csrow0 | DIMM_A0 | DIMM_B0 | + csrow1 | DIMM_A0 | DIMM_B0 | + =================================== + + =================================== + csrow2 | DIMM_A1 | DIMM_B1 | + csrow3 | DIMM_A1 | DIMM_B1 | + =================================== + +In the above example table there are 4 physical slots on the motherboard +for memory DIMMs: + + DIMM_A0 + DIMM_B0 + DIMM_A1 + DIMM_B1 + +Labels for these slots are usually silk screened on the motherboard. Slots +labeled 'A' are channel 0 in this example. Slots labled 'B' +are channel 1. Notice that there are two csrows possible on a +physical DIMM. These csrows are allocated their csrow assignment +based on the slot into which the memory DIMM is placed. Thus, when 1 DIMM +is placed in each Channel, the csrows cross both DIMMs. + +Memory DIMMs come single or dual "ranked". A rank is a populated csrow. +Thus, 2 single ranked DIMMs, placed in slots DIMM_A0 and DIMM_B0 above +will have 1 csrow, csrow0. csrow1 will be empty. On the other hand, +when 2 dual ranked DIMMs are similiaryly placed, then both csrow0 and +csrow1 will be populated. The pattern repeats itself for csrow2 and +csrow3. + +The representation of the above is reflected in the directory tree +in EDAC's sysfs interface. Starting in directory +/sys/devices/system/edac/mc each memory controller will be represented +by its own 'mcX' directory, where 'X" is the index of the MC. + + + ..../edac/mc/ + | + |->mc0 + |->mc1 + |->mc2 + .... + +Under each 'mcX' directory each 'csrowX' is again represented by a +'csrowX', where 'X" is the csrow index: + + + .../mc/mc0/ + | + |->csrow0 + |->csrow2 + |->csrow3 + .... + +Notice that there is no csrow1, which indicates that csrow0 is +composed of a single ranked DIMMs. This should also apply in both +Channels, in order to have dual-channel mode be operational. Since +both csrow2 and csrow3 are populated, this indicates a dual ranked +set of DIMMs for channels 0 and 1. + + +Within each of the 'mc','mcX' and 'csrowX' directories are several +EDAC control and attribute files. + + +============================================================================ +DIRECTORY 'mc' + +In directory 'mc' are EDAC system overall control and attribute files: + + +Panic on UE control file: + + 'panic_on_ue' + + An uncorrectable error will cause a machine panic. This is usually + desirable. It is a bad idea to continue when an uncorrectable error + occurs - it is indeterminate what was uncorrected and the operating + system context might be so mangled that continuing will lead to further + corruption. If the kernel has MCE configured, then EDAC will never + notice the UE. + + LOAD TIME: module/kernel parameter: panic_on_ue=[0|1] + + RUN TIME: echo "1" >/sys/devices/system/edac/mc/panic_on_ue + + +Log UE control file: + + 'log_ue' + + Generate kernel messages describing uncorrectable errors. These errors + are reported through the system message log system. UE statistics + will be accumulated even when UE logging is disabled. + + LOAD TIME: module/kernel parameter: log_ue=[0|1] + + RUN TIME: echo "1" >/sys/devices/system/edac/mc/log_ue + + +Log CE control file: + + 'log_ce' + + Generate kernel messages describing correctable errors. These + errors are reported through the system message log system. + CE statistics will be accumulated even when CE logging is disabled. + + LOAD TIME: module/kernel parameter: log_ce=[0|1] + + RUN TIME: echo "1" >/sys/devices/system/edac/mc/log_ce + + +Polling period control file: + + 'poll_msec' + + The time period, in milliseconds, for polling for error information. + Too small a value wastes resources. Too large a value might delay + necessary handling of errors and might loose valuable information for + locating the error. 1000 milliseconds (once each second) is about + right for most uses. + + LOAD TIME: module/kernel parameter: poll_msec=[0|1] + + RUN TIME: echo "1000" >/sys/devices/system/edac/mc/poll_msec + + +Module Version read-only attribute file: + + 'mc_version' + + The EDAC CORE modules's version and compile date are shown here to + indicate what EDAC is running. + + + +============================================================================ +'mcX' DIRECTORIES + + +In 'mcX' directories are EDAC control and attribute files for +this 'X" instance of the memory controllers: + + +Counter reset control file: + + 'reset_counters' + + This write-only control file will zero all the statistical counters + for UE and CE errors. Zeroing the counters will also reset the timer + indicating how long since the last counter zero. This is useful + for computing errors/time. Since the counters are always reset at + driver initialization time, no module/kernel parameter is available. + + RUN TIME: echo "anything" >/sys/devices/system/edac/mc/mc0/counter_reset + + This resets the counters on memory controller 0 + + +Seconds since last counter reset control file: + + 'seconds_since_reset' + + This attribute file displays how many seconds have elapsed since the + last counter reset. This can be used with the error counters to + measure error rates. + + + +DIMM capability attribute file: + + 'edac_capability' + + The EDAC (Error Detection and Correction) capabilities/modes of + the memory controller hardware. + + +DIMM Current Capability attribute file: + + 'edac_current_capability' + + The EDAC capabilities available with the hardware + configuration. This may not be the same as "EDAC capability" + if the correct memory is not used. If a memory controller is + capable of EDAC, but DIMMs without check bits are in use, then + Parity, SECDED, S4ECD4ED capabilities will not be available + even though the memory controller might be capable of those + modes with the proper memory loaded. + + +Memory Type supported on this controller attribute file: + + 'supported_mem_type' + + This attribute file displays the memory type, usually + buffered and unbuffered DIMMs. + + +Memory Controller name attribute file: + + 'mc_name' + + This attribute file displays the type of memory controller + that is being utilized. + + +Memory Controller Module name attribute file: + + 'module_name' + + This attribute file displays the memory controller module name, + version and date built. The name of the memory controller + hardware - some drivers work with multiple controllers and + this field shows which hardware is present. + + +Total memory managed by this memory controller attribute file: + + 'size_mb' + + This attribute file displays, in count of megabytes, of memory + that this instance of memory controller manages. + + +Total Uncorrectable Errors count attribute file: + + 'ue_count' + + This attribute file displays the total count of uncorrectable + errors that have occurred on this memory controller. If panic_on_ue + is set this counter will not have a chance to increment, + since EDAC will panic the system. + + +Total UE count that had no information attribute fileY: + + 'ue_noinfo_count' + + This attribute file displays the number of UEs that + have occurred have occurred with no informations as to which DIMM + slot is having errors. + + +Total Correctable Errors count attribute file: + + 'ce_count' + + This attribute file displays the total count of correctable + errors that have occurred on this memory controller. This + count is very important to examine. CEs provide early + indications that a DIMM is beginning to fail. This count + field should be monitored for non-zero values and report + such information to the system administrator. + + +Total Correctable Errors count attribute file: + + 'ce_noinfo_count' + + This attribute file displays the number of CEs that + have occurred wherewith no informations as to which DIMM slot + is having errors. Memory is handicapped, but operational, + yet no information is available to indicate which slot + the failing memory is in. This count field should be also + be monitored for non-zero values. + +Device Symlink: + + 'device' + + Symlink to the memory controller device + + + +============================================================================ +'csrowX' DIRECTORIES + +In the 'csrowX' directories are EDAC control and attribute files for +this 'X" instance of csrow: + + +Total Uncorrectable Errors count attribute file: + + 'ue_count' + + This attribute file displays the total count of uncorrectable + errors that have occurred on this csrow. If panic_on_ue is set + this counter will not have a chance to increment, since EDAC + will panic the system. + + +Total Correctable Errors count attribute file: + + 'ce_count' + + This attribute file displays the total count of correctable + errors that have occurred on this csrow. This + count is very important to examine. CEs provide early + indications that a DIMM is beginning to fail. This count + field should be monitored for non-zero values and report + such information to the system administrator. + + +Total memory managed by this csrow attribute file: + + 'size_mb' + + This attribute file displays, in count of megabytes, of memory + that this csrow contatins. + + +Memory Type attribute file: + + 'mem_type' + + This attribute file will display what type of memory is currently + on this csrow. Normally, either buffered or unbuffered memory. + + +EDAC Mode of operation attribute file: + + 'edac_mode' + + This attribute file will display what type of Error detection + and correction is being utilized. + + +Device type attribute file: + + 'dev_type' + + This attribute file will display what type of DIMM device is + being utilized. Example: x4 + + +Channel 0 CE Count attribute file: + + 'ch0_ce_count' + + This attribute file will display the count of CEs on this + DIMM located in channel 0. + + +Channel 0 UE Count attribute file: + + 'ch0_ue_count' + + This attribute file will display the count of UEs on this + DIMM located in channel 0. + + +Channel 0 DIMM Label control file: + + 'ch0_dimm_label' + + This control file allows this DIMM to have a label assigned + to it. With this label in the module, when errors occur + the output can provide the DIMM label in the system log. + This becomes vital for panic events to isolate the + cause of the UE event. + + DIMM Labels must be assigned after booting, with information + that correctly identifies the physical slot with its + silk screen label. This information is currently very + motherboard specific and determination of this information + must occur in userland at this time. + + +Channel 1 CE Count attribute file: + + 'ch1_ce_count' + + This attribute file will display the count of CEs on this + DIMM located in channel 1. + + +Channel 1 UE Count attribute file: + + 'ch1_ue_count' + + This attribute file will display the count of UEs on this + DIMM located in channel 0. + + +Channel 1 DIMM Label control file: + + 'ch1_dimm_label' + + This control file allows this DIMM to have a label assigned + to it. With this label in the module, when errors occur + the output can provide the DIMM label in the system log. + This becomes vital for panic events to isolate the + cause of the UE event. + + DIMM Labels must be assigned after booting, with information + that correctly identifies the physical slot with its + silk screen label. This information is currently very + motherboard specific and determination of this information + must occur in userland at this time. + + +============================================================================ +SYSTEM LOGGING + +If logging for UEs and CEs are enabled then system logs will have +error notices indicating errors that have been detected: + +MC0: CE page 0x283, offset 0xce0, grain 8, syndrome 0x6ec3, row 0, +channel 1 "DIMM_B1": amd76x_edac + +MC0: CE page 0x1e5, offset 0xfb0, grain 8, syndrome 0xb741, row 0, +channel 1 "DIMM_B1": amd76x_edac + + +The structure of the message is: + the memory controller (MC0) + Error type (CE) + memory page (0x283) + offset in the page (0xce0) + the byte granularity (grain 8) + or resolution of the error + the error syndrome (0xb741) + memory row (row 0) + memory channel (channel 1) + DIMM label, if set prior (DIMM B1 + and then an optional, driver-specific message that may + have additional information. + +Both UEs and CEs with no info will lack all but memory controller, +error type, a notice of "no info" and then an optional, +driver-specific error message. + + + +============================================================================ +PCI Bus Parity Detection + + +On Header Type 00 devices the primary status is looked at +for any parity error regardless of whether Parity is enabled on the +device. (The spec indicates parity is generated in some cases). +On Header Type 01 bridges, the secondary status register is also +looked at to see if parity ocurred on the bus on the other side of +the bridge. + + +SYSFS CONFIGURATION + +Under /sys/devices/system/edac/pci are control and attribute files as follows: + + +Enable/Disable PCI Parity checking control file: + + 'check_pci_parity' + + + This control file enables or disables the PCI Bus Parity scanning + operation. Writing a 1 to this file enables the scanning. Writing + a 0 to this file disables the scanning. + + Enable: + echo "1" >/sys/devices/system/edac/pci/check_pci_parity + + Disable: + echo "0" >/sys/devices/system/edac/pci/check_pci_parity + + + +Panic on PCI PARITY Error: + + 'panic_on_pci_parity' + + + This control files enables or disables panic'ing when a parity + error has been detected. + + + module/kernel parameter: panic_on_pci_parity=[0|1] + + Enable: + echo "1" >/sys/devices/system/edac/pci/panic_on_pci_parity + + Disable: + echo "0" >/sys/devices/system/edac/pci/panic_on_pci_parity + + +Parity Count: + + 'pci_parity_count' + + This attribute file will display the number of parity errors that + have been detected. + + + +PCI Device Whitelist: + + 'pci_parity_whitelist' + + This control file allows for an explicit list of PCI devices to be + scanned for parity errors. Only devices found on this list will + be examined. The list is a line of hexadecimel VENDOR and DEVICE + ID tuples: + + 1022:7450,1434:16a6 + + One or more can be inserted, seperated by a comma. + + To write the above list doing the following as one command line: + + echo "1022:7450,1434:16a6" + > /sys/devices/system/edac/pci/pci_parity_whitelist + + + + To display what the whitelist is, simply 'cat' the same file. + + +PCI Device Blacklist: + + 'pci_parity_blacklist' + + This control file allows for a list of PCI devices to be + skipped for scanning. + The list is a line of hexadecimel VENDOR and DEVICE ID tuples: + + 1022:7450,1434:16a6 + + One or more can be inserted, seperated by a comma. + + To write the above list doing the following as one command line: + + echo "1022:7450,1434:16a6" + > /sys/devices/system/edac/pci/pci_parity_blacklist + + + To display what the whitelist current contatins, + simply 'cat' the same file. + +======================================================================= + +PCI Vendor and Devices IDs can be obtained with the lspci command. Using +the -n option lspci will display the vendor and device IDs. The system +adminstrator will have to determine which devices should be scanned or +skipped. + + + +The two lists (white and black) are prioritized. blacklist is the lower +priority and will NOT be utilized when a whitelist has been set. +Turn OFF a whitelist by an empty echo command: + + echo > /sys/devices/system/edac/pci/pci_parity_whitelist + +and any previous blacklist will be utililzed. + diff --git a/MAINTAINERS b/MAINTAINERS index e6dbb21a8e5..3f8a90ac47d 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -867,6 +867,15 @@ L: ebtables-devel@lists.sourceforge.net W: http://ebtables.sourceforge.net/ S: Maintained +EDAC-CORE +P: Doug Thompson +M: norsk5@xmission.com, dthompson@linuxnetworx.com +P: Dave Peterson +M: dsp@llnl.gov, dave_peterson@pobox.com +L: bluesmoke-devel@lists.sourceforge.net +W: bluesmoke.sourceforge.net +S: Maintained + EEPRO100 NETWORK DRIVER P: Andrey V. Savochkin M: saw@saw.sw.com.sg diff --git a/arch/i386/kernel/quirks.c b/arch/i386/kernel/quirks.c index aaf89cb2bc5..87ccdac8492 100644 --- a/arch/i386/kernel/quirks.c +++ b/arch/i386/kernel/quirks.c @@ -25,8 +25,7 @@ static void __devinit quirk_intel_irqbalance(struct pci_dev *dev) /* enable access to config space*/ pci_read_config_byte(dev, 0xf4, &config); - config |= 0x2; - pci_write_config_byte(dev, 0xf4, config); + pci_write_config_byte(dev, 0xf4, config|0x2); /* read xTPR register */ raw_pci_ops->read(0, 0, 0x40, 0x4c, 2, &word); @@ -42,9 +41,9 @@ static void __devinit quirk_intel_irqbalance(struct pci_dev *dev) #endif } - config &= ~0x2; - /* disable access to config space*/ - pci_write_config_byte(dev, 0xf4, config); + /* put back the original value for config space*/ + if (!(config & 0x2)) + pci_write_config_byte(dev, 0xf4, config); } DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_E7320_MCH, quirk_intel_irqbalance); DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_E7525_MCH, quirk_intel_irqbalance); diff --git a/drivers/Kconfig b/drivers/Kconfig index 283c089537b..bddf431bbb7 100644 --- a/drivers/Kconfig +++ b/drivers/Kconfig @@ -68,4 +68,6 @@ source "drivers/infiniband/Kconfig" source "drivers/sn/Kconfig" +source "drivers/edac/Kconfig" + endmenu diff --git a/drivers/Makefile b/drivers/Makefile index 7c45050ecd0..619dd964c51 100644 --- a/drivers/Makefile +++ b/drivers/Makefile @@ -63,6 +63,7 @@ obj-$(CONFIG_PHONE) += telephony/ obj-$(CONFIG_MD) += md/ obj-$(CONFIG_BT) += bluetooth/ obj-$(CONFIG_ISDN) += isdn/ +obj-$(CONFIG_EDAC) += edac/ obj-$(CONFIG_MCA) += mca/ obj-$(CONFIG_EISA) += eisa/ obj-$(CONFIG_CPU_FREQ) += cpufreq/ diff --git a/drivers/edac/Kconfig b/drivers/edac/Kconfig new file mode 100644 index 00000000000..4819e7fc00d --- /dev/null +++ b/drivers/edac/Kconfig @@ -0,0 +1,102 @@ +# +# EDAC Kconfig +# Copyright (c) 2003 Linux Networx +# Licensed and distributed under the GPL +# +# $Id: Kconfig,v 1.4.2.7 2005/07/08 22:05:38 dsp_llnl Exp $ +# + +menu 'EDAC - error detection and reporting (RAS)' + +config EDAC + tristate "EDAC core system error reporting" + depends on X86 + default y + help + EDAC is designed to report errors in the core system. + These are low-level errors that are reported in the CPU or + supporting chipset: memory errors, cache errors, PCI errors, + thermal throttling, etc.. If unsure, select 'Y'. + + +comment "Reporting subsystems" + depends on EDAC + +config EDAC_DEBUG + bool "Debugging" + depends on EDAC + help + This turns on debugging information for the entire EDAC + sub-system. You can insert module with "debug_level=x", current + there're four debug levels (x=0,1,2,3 from low to high). + Usually you should select 'N'. + +config EDAC_MM_EDAC + tristate "Main Memory EDAC (Error Detection And Correction) reporting" + depends on EDAC + default y + help + Some systems are able to detect and correct errors in main + memory. EDAC can report statistics on memory error + detection and correction (EDAC - or commonly referred to ECC + errors). EDAC will also try to decode where these errors + occurred so that a particular failing memory module can be + replaced. If unsure, select 'Y'. + + +config EDAC_AMD76X + tristate "AMD 76x (760, 762, 768)" + depends on EDAC_MM_EDAC && PCI + help + Support for error detection and correction on the AMD 76x + series of chipsets used with the Athlon processor. + +config EDAC_E7XXX + tristate "Intel e7xxx (e7205, e7500, e7501, e7505)" + depends on EDAC_MM_EDAC && PCI + help + Support for error detection and correction on the Intel + E7205, E7500, E7501 and E7505 server chipsets. + +config EDAC_E752X + tristate "Intel e752x (e7520, e7525, e7320)" + depends on EDAC_MM_EDAC && PCI + help + Support for error detection and correction on the Intel + E7520, E7525, E7320 server chipsets. + +config EDAC_I82875P + tristate "Intel 82875p (D82875P, E7210)" + depends on EDAC_MM_EDAC && PCI + help + Support for error detection and correction on the Intel + DP82785P and E7210 server chipsets. + +config EDAC_I82860 + tristate "Intel 82860" + depends on EDAC_MM_EDAC && PCI + help + Support for error detection and correction on the Intel + 82860 chipset. + +config EDAC_R82600 + tristate "Radisys 82600 embedded chipset" + depends on EDAC_MM_EDAC + help + Support for error detection and correction on the Radisys + 82600 embedded chipset. + +choice + prompt "Error detecting method" + depends on EDAC + default EDAC_POLL + +config EDAC_POLL + bool "Poll for errors" + depends on EDAC + help + Poll the chipset periodically to detect errors. + +endchoice + +endmenu diff --git a/drivers/edac/Makefile b/drivers/edac/Makefile new file mode 100644 index 00000000000..93137fdab4b --- /dev/null +++ b/drivers/edac/Makefile @@ -0,0 +1,18 @@ +# +# Makefile for the Linux kernel EDAC drivers. +# +# Copyright 02 Jul 2003, Linux Networx (http://lnxi.com) +# This file may be distributed under the terms of the +# GNU General Public License. +# +# $Id: Makefile,v 1.4.2.3 2005/07/08 22:05:38 dsp_llnl Exp $ + + +obj-$(CONFIG_EDAC_MM_EDAC) += edac_mc.o +obj-$(CONFIG_EDAC_AMD76X) += amd76x_edac.o +obj-$(CONFIG_EDAC_E7XXX) += e7xxx_edac.o +obj-$(CONFIG_EDAC_E752X) += e752x_edac.o +obj-$(CONFIG_EDAC_I82875P) += i82875p_edac.o +obj-$(CONFIG_EDAC_I82860) += i82860_edac.o +obj-$(CONFIG_EDAC_R82600) += r82600_edac.o + diff --git a/drivers/edac/amd76x_edac.c b/drivers/edac/amd76x_edac.c index 8e2b1295e70..2fcc8120b53 100644 --- a/drivers/edac/amd76x_edac.c +++ b/drivers/edac/amd76x_edac.c @@ -338,7 +338,7 @@ static struct pci_driver amd76x_driver = { .id_table = amd76x_pci_tbl, }; -int __init amd76x_init(void) +static int __init amd76x_init(void) { return pci_register_driver(&amd76x_driver); } diff --git a/drivers/edac/e752x_edac.c b/drivers/edac/e752x_edac.c index 959f584f568..770a5a63307 100644 --- a/drivers/edac/e752x_edac.c +++ b/drivers/edac/e752x_edac.c @@ -13,7 +13,7 @@ * Wang Zhenyu at intel.com * Dave Jiang at mvista.com * - * $Id: bluesmoke_e752x.c,v 1.5.2.11 2005/10/05 00:43:44 dsp_llnl Exp $ + * $Id: edac_e752x.c,v 1.5.2.11 2005/10/05 00:43:44 dsp_llnl Exp $ * */ @@ -376,14 +376,14 @@ static inline void process_threshold_ce(struct mem_ctl_info *mci, u16 error, mci->mc_idx); } -char *global_message[11] = { +static char *global_message[11] = { "PCI Express C1", "PCI Express C", "PCI Express B1", "PCI Express B", "PCI Express A1", "PCI Express A", "DMA Controler", "HUB Interface", "System Bus", "DRAM Controler", "Internal Buffer" }; -char *fatal_message[2] = { "Non-Fatal ", "Fatal " }; +static char *fatal_message[2] = { "Non-Fatal ", "Fatal " }; static void do_global_error(int fatal, u32 errors) { @@ -405,7 +405,7 @@ static inline void global_error(int fatal, u32 errors, int *error_found, do_global_error(fatal, errors); } -char *hub_message[7] = { +static char *hub_message[7] = { "HI Address or Command Parity", "HI Illegal Access", "HI Internal Parity", "Out of Range Access", "HI Data Parity", "Enhanced Config Access", @@ -432,7 +432,7 @@ static inline void hub_error(int fatal, u8 errors, int *error_found, do_hub_error(fatal, errors); } -char *membuf_message[4] = { +static char *membuf_message[4] = { "Internal PMWB to DRAM parity", "Internal PMWB to System Bus Parity", "Internal System Bus or IO to PMWB Parity", @@ -458,6 +458,7 @@ static inline void membuf_error(u8 errors, int *error_found, int handle_error) do_membuf_error(errors); } +#if 0 char *sysbus_message[10] = { "Addr or Request Parity", "Data Strobe Glitch", @@ -469,6 +470,7 @@ char *sysbus_message[10] = { "Memory Parity", "IO Subsystem Parity" }; +#endif /* 0 */ static void do_sysbus_error(int fatal, u32 errors) { @@ -1044,7 +1046,7 @@ static struct pci_driver e752x_driver = { }; -int __init e752x_init(void) +static int __init e752x_init(void) { int pci_rc; diff --git a/drivers/edac/e7xxx_edac.c b/drivers/edac/e7xxx_edac.c index 066be43f5ec..d5e320dfc66 100644 --- a/drivers/edac/e7xxx_edac.c +++ b/drivers/edac/e7xxx_edac.c @@ -537,7 +537,7 @@ static struct pci_driver e7xxx_driver = { }; -int __init e7xxx_init(void) +static int __init e7xxx_init(void) { return pci_register_driver(&e7xxx_driver); } diff --git a/drivers/edac/edac_mc.c b/drivers/edac/edac_mc.c new file mode 100644 index 00000000000..4be9bd0a126 --- /dev/null +++ b/drivers/edac/edac_mc.c @@ -0,0 +1,2209 @@ +/* + * edac_mc kernel module + * (C) 2005 Linux Networx (http://lnxi.com) + * This file may be distributed under the terms of the + * GNU General Public License. + * + * Written by Thayne Harbaugh + * Based on work by Dan Hollis and others. + * http://www.anime.net/~goemon/linux-ecc/ + * + * Modified by Dave Peterson and Doug Thompson + * + */ + + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#include "edac_mc.h" + +#define EDAC_MC_VERSION "edac_mc Ver: 2.0.0 " __DATE__ + +#ifdef CONFIG_EDAC_DEBUG +/* Values of 0 to 4 will generate output */ +int edac_debug_level = 1; +EXPORT_SYMBOL(edac_debug_level); +#endif + +/* EDAC Controls, setable by module parameter, and sysfs */ +static int log_ue = 1; +static int log_ce = 1; +static int panic_on_ue = 1; +static int poll_msec = 1000; + +static int check_pci_parity = 0; /* default YES check PCI parity */ +static int panic_on_pci_parity; /* default no panic on PCI Parity */ +static atomic_t pci_parity_count = ATOMIC_INIT(0); + +/* lock to memory controller's control array */ +static DECLARE_MUTEX(mem_ctls_mutex); +static struct list_head mc_devices = LIST_HEAD_INIT(mc_devices); + +/* Structure of the whitelist and blacklist arrays */ +struct edac_pci_device_list { + unsigned int vendor; /* Vendor ID */ + unsigned int device; /* Deviice ID */ +}; + + +#define MAX_LISTED_PCI_DEVICES 32 + +/* List of PCI devices (vendor-id:device-id) that should be skipped */ +static struct edac_pci_device_list pci_blacklist[MAX_LISTED_PCI_DEVICES]; +static int pci_blacklist_count; + +/* List of PCI devices (vendor-id:device-id) that should be scanned */ +static struct edac_pci_device_list pci_whitelist[MAX_LISTED_PCI_DEVICES]; +static int pci_whitelist_count ; + +/* START sysfs data and methods */ + +static const char *mem_types[] = { + [MEM_EMPTY] = "Empty", + [MEM_RESERVED] = "Reserved", + [MEM_UNKNOWN] = "Unknown", + [MEM_FPM] = "FPM", + [MEM_EDO] = "EDO", + [MEM_BEDO] = "BEDO", + [MEM_SDR] = "Unbuffered-SDR", + [MEM_RDR] = "Registered-SDR", + [MEM_DDR] = "Unbuffered-DDR", + [MEM_RDDR] = "Registered-DDR", + [MEM_RMBS] = "RMBS" +}; + +static const char *dev_types[] = { + [DEV_UNKNOWN] = "Unknown", + [DEV_X1] = "x1", + [DEV_X2] = "x2", + [DEV_X4] = "x4", + [DEV_X8] = "x8", + [DEV_X16] = "x16", + [DEV_X32] = "x32", + [DEV_X64] = "x64" +}; + +static const char *edac_caps[] = { + [EDAC_UNKNOWN] = "Unknown", + [EDAC_NONE] = "None", + [EDAC_RESERVED] = "Reserved", + [EDAC_PARITY] = "PARITY", + [EDAC_EC] = "EC", + [EDAC_SECDED] = "SECDED", + [EDAC_S2ECD2ED] = "S2ECD2ED", + [EDAC_S4ECD4ED] = "S4ECD4ED", + [EDAC_S8ECD8ED] = "S8ECD8ED", + [EDAC_S16ECD16ED] = "S16ECD16ED" +}; + + +/* sysfs object: /sys/devices/system/edac */ +static struct sysdev_class edac_class = { + set_kset_name("edac"), +}; + +/* sysfs objects: + * /sys/devices/system/edac/mc + * /sys/devices/system/edac/pci + */ +static struct kobject edac_memctrl_kobj; +static struct kobject edac_pci_kobj; + +/* + * /sys/devices/system/edac/mc; + * data structures and methods + */ +static ssize_t memctrl_string_show(void *ptr, char *buffer) +{ + char *value = (char*) ptr; + return sprintf(buffer, "%s\n", value); +} + +static ssize_t memctrl_int_show(void *ptr, char *buffer) +{ + int *value = (int*) ptr; + return sprintf(buffer, "%d\n", *value); +} + +static ssize_t memctrl_int_store(void *ptr, const char *buffer, size_t count) +{ + int *value = (int*) ptr; + + if (isdigit(*buffer)) + *value = simple_strtoul(buffer, NULL, 0); + + return count; +} + +struct memctrl_dev_attribute { + struct attribute attr; + void *value; + ssize_t (*show)(void *,char *); + ssize_t (*store)(void *, const char *, size_t); +}; + +/* Set of show/store abstract level functions for memory control object */ +static ssize_t +memctrl_dev_show(struct kobject *kobj, struct attribute *attr, char *buffer) +{ + struct memctrl_dev_attribute *memctrl_dev; + memctrl_dev = (struct memctrl_dev_attribute*)attr; + + if (memctrl_dev->show) + return memctrl_dev->show(memctrl_dev->value, buffer); + return -EIO; +} + +static ssize_t +memctrl_dev_store(struct kobject *kobj, struct attribute *attr, + const char *buffer, size_t count) +{ + struct memctrl_dev_attribute *memctrl_dev; + memctrl_dev = (struct memctrl_dev_attribute*)attr; + + if (memctrl_dev->store) + return memctrl_dev->store(memctrl_dev->value, buffer, count); + return -EIO; +} + +static struct sysfs_ops memctrlfs_ops = { + .show = memctrl_dev_show, + .store = memctrl_dev_store +}; + +#define MEMCTRL_ATTR(_name,_mode,_show,_store) \ +struct memctrl_dev_attribute attr_##_name = { \ + .attr = {.name = __stringify(_name), .mode = _mode }, \ + .value = &_name, \ + .show = _show, \ + .store = _store, \ +}; + +#define MEMCTRL_STRING_ATTR(_name,_data,_mode,_show,_store) \ +struct memctrl_dev_attribute attr_##_name = { \ + .attr = {.name = __stringify(_name), .mode = _mode }, \ + .value = _data, \ + .show = _show, \ + .store = _store, \ +}; + +/* cwrow attribute f*/ +MEMCTRL_STRING_ATTR(mc_version,EDAC_MC_VERSION,S_IRUGO,memctrl_string_show,NULL); + +/* csrow control files */ +MEMCTRL_ATTR(panic_on_ue,S_IRUGO|S_IWUSR,memctrl_int_show,memctrl_int_store); +MEMCTRL_ATTR(log_ue,S_IRUGO|S_IWUSR,memctrl_int_show,memctrl_int_store); +MEMCTRL_ATTR(log_ce,S_IRUGO|S_IWUSR,memctrl_int_show,memctrl_int_store); +MEMCTRL_ATTR(poll_msec,S_IRUGO|S_IWUSR,memctrl_int_show,memctrl_int_store); + + +/* Base Attributes of the memory ECC object */ +static struct memctrl_dev_attribute *memctrl_attr[] = { + &attr_panic_on_ue, + &attr_log_ue, + &attr_log_ce, + &attr_poll_msec, + &attr_mc_version, + NULL, +}; + +/* Main MC kobject release() function */ +static void edac_memctrl_master_release(struct kobject *kobj) +{ + debugf1("EDAC MC: " __FILE__ ": %s()\n", __func__); +} + +static struct kobj_type ktype_memctrl = { + .release = edac_memctrl_master_release, + .sysfs_ops = &memctrlfs_ops, + .default_attrs = (struct attribute **) memctrl_attr, +}; + + +/* Initialize the main sysfs entries for edac: + * /sys/devices/system/edac + * + * and children + * + * Return: 0 SUCCESS + * !0 FAILURE + */ +static int edac_sysfs_memctrl_setup(void) +{ + int err=0; + + debugf1("MC: " __FILE__ ": %s()\n", __func__); + + /* create the /sys/devices/system/edac directory */ + err = sysdev_class_register(&edac_class); + if (!err) { + /* Init the MC's kobject */ + memset(&edac_memctrl_kobj, 0, sizeof (edac_memctrl_kobj)); + kobject_init(&edac_memctrl_kobj); + + edac_memctrl_kobj.parent = &edac_class.kset.kobj; + edac_memctrl_kobj.ktype = &ktype_memctrl; + + /* generate sysfs "..../edac/mc" */ + err = kobject_set_name(&edac_memctrl_kobj,"mc"); + if (!err) { + /* FIXME: maybe new sysdev_create_subdir() */ + err = kobject_register(&edac_memctrl_kobj); + if (err) { + debugf1("Failed to register '.../edac/mc'\n"); + } else { + debugf1("Registered '.../edac/mc' kobject\n"); + } + } + } else { + debugf1(KERN_WARNING "__FILE__ %s() error=%d\n", __func__,err); + } + + return err; +} + +/* + * MC teardown: + * the '..../edac/mc' kobject followed by '..../edac' itself + */ +static void edac_sysfs_memctrl_teardown(void) +{ + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + /* Unregister the MC's kobject */ + kobject_unregister(&edac_memctrl_kobj); + + /* release the master edac mc kobject */ + kobject_put(&edac_memctrl_kobj); + + /* Unregister the 'edac' object */ + sysdev_class_unregister(&edac_class); +} + +/* + * /sys/devices/system/edac/pci; + * data structures and methods + */ + +struct list_control { + struct edac_pci_device_list *list; + int *count; +}; + +/* Output the list as: vendor_id:device:id<,vendor_id:device_id> */ +static ssize_t edac_pci_list_string_show(void *ptr, char *buffer) +{ + struct list_control *listctl; + struct edac_pci_device_list *list; + char *p = buffer; + int len=0; + int i; + + listctl = ptr; + list = listctl->list; + + for (i = 0; i < *(listctl->count); i++, list++ ) { + if (len > 0) + len += snprintf(p + len, (PAGE_SIZE-len), ","); + + len += snprintf(p + len, + (PAGE_SIZE-len), + "%x:%x", + list->vendor,list->device); + } + + len += snprintf(p + len,(PAGE_SIZE-len), "\n"); + + return (ssize_t) len; +} + +/** + * + * Scan string from **s to **e looking for one 'vendor:device' tuple + * where each field is a hex value + * + * return 0 if an entry is NOT found + * return 1 if an entry is found + * fill in *vendor_id and *device_id with values found + * + * In both cases, make sure *s has been moved forward toward *e + */ +static int parse_one_device(const char **s,const char **e, + unsigned int *vendor_id, unsigned int *device_id) +{ + const char *runner, *p; + + /* if null byte, we are done */ + if (!**s) { + (*s)++; /* keep *s moving */ + return 0; + } + + /* skip over newlines & whitespace */ + if ((**s == '\n') || isspace(**s)) { + (*s)++; + return 0; + } + + if (!isxdigit(**s)) { + (*s)++; + return 0; + } + + /* parse vendor_id */ + runner = *s; + while (runner < *e) { + /* scan for vendor:device delimiter */ + if (*runner == ':') { + *vendor_id = simple_strtol((char*) *s, (char**) &p, 16); + runner = p + 1; + break; + } + runner++; + } + + if (!isxdigit(*runner)) { + *s = ++runner; + return 0; + } + + /* parse device_id */ + if (runner < *e) { + *device_id = simple_strtol((char*)runner, (char**)&p, 16); + runner = p; + } + + *s = runner; + + return 1; +} + +static ssize_t edac_pci_list_string_store(void *ptr, const char *buffer, + size_t count) +{ + struct list_control *listctl; + struct edac_pci_device_list *list; + unsigned int vendor_id, device_id; + const char *s, *e; + int *index; + + s = (char*)buffer; + e = s + count; + + listctl = ptr; + list = listctl->list; + index = listctl->count; + + *index = 0; + while (*index < MAX_LISTED_PCI_DEVICES) { + + if (parse_one_device(&s,&e,&vendor_id,&device_id)) { + list[ *index ].vendor = vendor_id; + list[ *index ].device = device_id; + (*index)++; + } + + /* check for all data consume */ + if (s >= e) + break; + } + + return count; +} + +static ssize_t edac_pci_int_show(void *ptr, char *buffer) +{ + int *value = ptr; + return sprintf(buffer,"%d\n",*value); +} + +static ssize_t edac_pci_int_store(void *ptr, const char *buffer, size_t count) +{ + int *value = ptr; + + if (isdigit(*buffer)) + *value = simple_strtoul(buffer,NULL,0); + + return count; +} + +struct edac_pci_dev_attribute { + struct attribute attr; + void *value; + ssize_t (*show)(void *,char *); + ssize_t (*store)(void *, const char *,size_t); +}; + +/* Set of show/store abstract level functions for PCI Parity object */ +static ssize_t edac_pci_dev_show(struct kobject *kobj, struct attribute *attr, + char *buffer) +{ + struct edac_pci_dev_attribute *edac_pci_dev; + edac_pci_dev= (struct edac_pci_dev_attribute*)attr; + + if (edac_pci_dev->show) + return edac_pci_dev->show(edac_pci_dev->value, buffer); + return -EIO; +} + +static ssize_t edac_pci_dev_store(struct kobject *kobj, struct attribute *attr, + const char *buffer, size_t count) +{ + struct edac_pci_dev_attribute *edac_pci_dev; + edac_pci_dev= (struct edac_pci_dev_attribute*)attr; + + if (edac_pci_dev->show) + return edac_pci_dev->store(edac_pci_dev->value, buffer, count); + return -EIO; +} + +static struct sysfs_ops edac_pci_sysfs_ops = { + .show = edac_pci_dev_show, + .store = edac_pci_dev_store +}; + + +#define EDAC_PCI_ATTR(_name,_mode,_show,_store) \ +struct edac_pci_dev_attribute edac_pci_attr_##_name = { \ + .attr = {.name = __stringify(_name), .mode = _mode }, \ + .value = &_name, \ + .show = _show, \ + .store = _store, \ +}; + +#define EDAC_PCI_STRING_ATTR(_name,_data,_mode,_show,_store) \ +struct edac_pci_dev_attribute edac_pci_attr_##_name = { \ + .attr = {.name = __stringify(_name), .mode = _mode }, \ + .value = _data, \ + .show = _show, \ + .store = _store, \ +}; + +static struct list_control pci_whitelist_control = { + .list = pci_whitelist, + .count = &pci_whitelist_count +}; + +static struct list_control pci_blacklist_control = { + .list = pci_blacklist, + .count = &pci_blacklist_count +}; + +/* whitelist attribute */ +EDAC_PCI_STRING_ATTR(pci_parity_whitelist, + &pci_whitelist_control, + S_IRUGO|S_IWUSR, + edac_pci_list_string_show, + edac_pci_list_string_store); + +EDAC_PCI_STRING_ATTR(pci_parity_blacklist, + &pci_blacklist_control, + S_IRUGO|S_IWUSR, + edac_pci_list_string_show, + edac_pci_list_string_store); + +/* PCI Parity control files */ +EDAC_PCI_ATTR(check_pci_parity,S_IRUGO|S_IWUSR,edac_pci_int_show,edac_pci_int_store); +EDAC_PCI_ATTR(panic_on_pci_parity,S_IRUGO|S_IWUSR,edac_pci_int_show,edac_pci_int_store); +EDAC_PCI_ATTR(pci_parity_count,S_IRUGO,edac_pci_int_show,NULL); + +/* Base Attributes of the memory ECC object */ +static struct edac_pci_dev_attribute *edac_pci_attr[] = { + &edac_pci_attr_check_pci_parity, + &edac_pci_attr_panic_on_pci_parity, + &edac_pci_attr_pci_parity_count, + &edac_pci_attr_pci_parity_whitelist, + &edac_pci_attr_pci_parity_blacklist, + NULL, +}; + +/* No memory to release */ +static void edac_pci_release(struct kobject *kobj) +{ + debugf1("EDAC PCI: " __FILE__ ": %s()\n", __func__); +} + +static struct kobj_type ktype_edac_pci = { + .release = edac_pci_release, + .sysfs_ops = &edac_pci_sysfs_ops, + .default_attrs = (struct attribute **) edac_pci_attr, +}; + +/** + * edac_sysfs_pci_setup() + * + */ +static int edac_sysfs_pci_setup(void) +{ + int err; + + debugf1("MC: " __FILE__ ": %s()\n", __func__); + + memset(&edac_pci_kobj, 0, sizeof(edac_pci_kobj)); + + kobject_init(&edac_pci_kobj); + edac_pci_kobj.parent = &edac_class.kset.kobj; + edac_pci_kobj.ktype = &ktype_edac_pci; + + err = kobject_set_name(&edac_pci_kobj, "pci"); + if (!err) { + /* Instanstiate the csrow object */ + /* FIXME: maybe new sysdev_create_subdir() */ + err = kobject_register(&edac_pci_kobj); + if (err) + debugf1("Failed to register '.../edac/pci'\n"); + else + debugf1("Registered '.../edac/pci' kobject\n"); + } + return err; +} + + +static void edac_sysfs_pci_teardown(void) +{ + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + kobject_unregister(&edac_pci_kobj); + kobject_put(&edac_pci_kobj); +} + +/* EDAC sysfs CSROW data structures and methods */ + +/* Set of more detailed csrow attribute show/store functions */ +static ssize_t csrow_ch0_dimm_label_show(struct csrow_info *csrow, char *data) +{ + ssize_t size = 0; + + if (csrow->nr_channels > 0) { + size = snprintf(data, EDAC_MC_LABEL_LEN,"%s\n", + csrow->channels[0].label); + } + return size; +} + +static ssize_t csrow_ch1_dimm_label_show(struct csrow_info *csrow, char *data) +{ + ssize_t size = 0; + + if (csrow->nr_channels > 0) { + size = snprintf(data, EDAC_MC_LABEL_LEN, "%s\n", + csrow->channels[1].label); + } + return size; +} + +static ssize_t csrow_ch0_dimm_label_store(struct csrow_info *csrow, + const char *data, size_t size) +{ + ssize_t max_size = 0; + + if (csrow->nr_channels > 0) { + max_size = min((ssize_t)size,(ssize_t)EDAC_MC_LABEL_LEN-1); + strncpy(csrow->channels[0].label, data, max_size); + csrow->channels[0].label[max_size] = '\0'; + } + return size; +} + +static ssize_t csrow_ch1_dimm_label_store(struct csrow_info *csrow, + const char *data, size_t size) +{ + ssize_t max_size = 0; + + if (csrow->nr_channels > 1) { + max_size = min((ssize_t)size,(ssize_t)EDAC_MC_LABEL_LEN-1); + strncpy(csrow->channels[1].label, data, max_size); + csrow->channels[1].label[max_size] = '\0'; + } + return max_size; +} + +static ssize_t csrow_ue_count_show(struct csrow_info *csrow, char *data) +{ + return sprintf(data,"%u\n", csrow->ue_count); +} + +static ssize_t csrow_ce_count_show(struct csrow_info *csrow, char *data) +{ + return sprintf(data,"%u\n", csrow->ce_count); +} + +static ssize_t csrow_ch0_ce_count_show(struct csrow_info *csrow, char *data) +{ + ssize_t size = 0; + + if (csrow->nr_channels > 0) { + size = sprintf(data,"%u\n", csrow->channels[0].ce_count); + } + return size; +} + +static ssize_t csrow_ch1_ce_count_show(struct csrow_info *csrow, char *data) +{ + ssize_t size = 0; + + if (csrow->nr_channels > 1) { + size = sprintf(data,"%u\n", csrow->channels[1].ce_count); + } + return size; +} + +static ssize_t csrow_size_show(struct csrow_info *csrow, char *data) +{ + return sprintf(data,"%u\n", PAGES_TO_MiB(csrow->nr_pages)); +} + +static ssize_t csrow_mem_type_show(struct csrow_info *csrow, char *data) +{ + return sprintf(data,"%s\n", mem_types[csrow->mtype]); +} + +static ssize_t csrow_dev_type_show(struct csrow_info *csrow, char *data) +{ + return sprintf(data,"%s\n", dev_types[csrow->dtype]); +} + +static ssize_t csrow_edac_mode_show(struct csrow_info *csrow, char *data) +{ + return sprintf(data,"%s\n", edac_caps[csrow->edac_mode]); +} + +struct csrowdev_attribute { + struct attribute attr; + ssize_t (*show)(struct csrow_info *,char *); + ssize_t (*store)(struct csrow_info *, const char *,size_t); +}; + +#define to_csrow(k) container_of(k, struct csrow_info, kobj) +#define to_csrowdev_attr(a) container_of(a, struct csrowdev_attribute, attr) + +/* Set of show/store higher level functions for csrow objects */ +static ssize_t csrowdev_show(struct kobject *kobj, struct attribute *attr, + char *buffer) +{ + struct csrow_info *csrow = to_csrow(kobj); + struct csrowdev_attribute *csrowdev_attr = to_csrowdev_attr(attr); + + if (csrowdev_attr->show) + return csrowdev_attr->show(csrow, buffer); + return -EIO; +} + +static ssize_t csrowdev_store(struct kobject *kobj, struct attribute *attr, + const char *buffer, size_t count) +{ + struct csrow_info *csrow = to_csrow(kobj); + struct csrowdev_attribute * csrowdev_attr = to_csrowdev_attr(attr); + + if (csrowdev_attr->store) + return csrowdev_attr->store(csrow, buffer, count); + return -EIO; +} + +static struct sysfs_ops csrowfs_ops = { + .show = csrowdev_show, + .store = csrowdev_store +}; + +#define CSROWDEV_ATTR(_name,_mode,_show,_store) \ +struct csrowdev_attribute attr_##_name = { \ + .attr = {.name = __stringify(_name), .mode = _mode }, \ + .show = _show, \ + .store = _store, \ +}; + +/* cwrow/attribute files */ +CSROWDEV_ATTR(size_mb,S_IRUGO,csrow_size_show,NULL); +CSROWDEV_ATTR(dev_type,S_IRUGO,csrow_dev_type_show,NULL); +CSROWDEV_ATTR(mem_type,S_IRUGO,csrow_mem_type_show,NULL); +CSROWDEV_ATTR(edac_mode,S_IRUGO,csrow_edac_mode_show,NULL); +CSROWDEV_ATTR(ue_count,S_IRUGO,csrow_ue_count_show,NULL); +CSROWDEV_ATTR(ce_count,S_IRUGO,csrow_ce_count_show,NULL); +CSROWDEV_ATTR(ch0_ce_count,S_IRUGO,csrow_ch0_ce_count_show,NULL); +CSROWDEV_ATTR(ch1_ce_count,S_IRUGO,csrow_ch1_ce_count_show,NULL); + +/* control/attribute files */ +CSROWDEV_ATTR(ch0_dimm_label,S_IRUGO|S_IWUSR, + csrow_ch0_dimm_label_show, + csrow_ch0_dimm_label_store); +CSROWDEV_ATTR(ch1_dimm_label,S_IRUGO|S_IWUSR, + csrow_ch1_dimm_label_show, + csrow_ch1_dimm_label_store); + + +/* Attributes of the CSROW object */ +static struct csrowdev_attribute *csrow_attr[] = { + &attr_dev_type, + &attr_mem_type, + &attr_edac_mode, + &attr_size_mb, + &attr_ue_count, + &attr_ce_count, + &attr_ch0_ce_count, + &attr_ch1_ce_count, + &attr_ch0_dimm_label, + &attr_ch1_dimm_label, + NULL, +}; + + +/* No memory to release */ +static void edac_csrow_instance_release(struct kobject *kobj) +{ + debugf1("EDAC MC: " __FILE__ ": %s()\n", __func__); +} + +static struct kobj_type ktype_csrow = { + .release = edac_csrow_instance_release, + .sysfs_ops = &csrowfs_ops, + .default_attrs = (struct attribute **) csrow_attr, +}; + +/* Create a CSROW object under specifed edac_mc_device */ +static int edac_create_csrow_object(struct kobject *edac_mci_kobj, + struct csrow_info *csrow, int index ) +{ + int err = 0; + + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + memset(&csrow->kobj, 0, sizeof(csrow->kobj)); + + /* generate ..../edac/mc/mc/csrow */ + + kobject_init(&csrow->kobj); + csrow->kobj.parent = edac_mci_kobj; + csrow->kobj.ktype = &ktype_csrow; + + /* name this instance of csrow */ + err = kobject_set_name(&csrow->kobj,"csrow%d",index); + if (!err) { + /* Instanstiate the csrow object */ + err = kobject_register(&csrow->kobj); + if (err) + debugf0("Failed to register CSROW%d\n",index); + else + debugf0("Registered CSROW%d\n",index); + } + + return err; +} + +/* sysfs data structures and methods for the MCI kobjects */ + +static ssize_t mci_reset_counters_store(struct mem_ctl_info *mci, + const char *data, size_t count ) +{ + int row, chan; + + mci->ue_noinfo_count = 0; + mci->ce_noinfo_count = 0; + mci->ue_count = 0; + mci->ce_count = 0; + for (row = 0; row < mci->nr_csrows; row++) { + struct csrow_info *ri = &mci->csrows[row]; + + ri->ue_count = 0; + ri->ce_count = 0; + for (chan = 0; chan < ri->nr_channels; chan++) + ri->channels[chan].ce_count = 0; + } + mci->start_time = jiffies; + + return count; +} + +static ssize_t mci_ue_count_show(struct mem_ctl_info *mci, char *data) +{ + return sprintf(data,"%d\n", mci->ue_count); +} + +static ssize_t mci_ce_count_show(struct mem_ctl_info *mci, char *data) +{ + return sprintf(data,"%d\n", mci->ce_count); +} + +static ssize_t mci_ce_noinfo_show(struct mem_ctl_info *mci, char *data) +{ + return sprintf(data,"%d\n", mci->ce_noinfo_count); +} + +static ssize_t mci_ue_noinfo_show(struct mem_ctl_info *mci, char *data) +{ + return sprintf(data,"%d\n", mci->ue_noinfo_count); +} + +static ssize_t mci_seconds_show(struct mem_ctl_info *mci, char *data) +{ + return sprintf(data,"%ld\n", (jiffies - mci->start_time) / HZ); +} + +static ssize_t mci_mod_name_show(struct mem_ctl_info *mci, char *data) +{ + return sprintf(data,"%s %s\n", mci->mod_name, mci->mod_ver); +} + +static ssize_t mci_ctl_name_show(struct mem_ctl_info *mci, char *data) +{ + return sprintf(data,"%s\n", mci->ctl_name); +} + +static int mci_output_edac_cap(char *buf, unsigned long edac_cap) +{ + char *p = buf; + int bit_idx; + + for (bit_idx = 0; bit_idx < 8 * sizeof(edac_cap); bit_idx++) { + if ((edac_cap >> bit_idx) & 0x1) + p += sprintf(p, "%s ", edac_caps[bit_idx]); + } + + return p - buf; +} + +static ssize_t mci_edac_capability_show(struct mem_ctl_info *mci, char *data) +{ + char *p = data; + + p += mci_output_edac_cap(p,mci->edac_ctl_cap); + p += sprintf(p, "\n"); + + return p - data; +} + +static ssize_t mci_edac_current_capability_show(struct mem_ctl_info *mci, + char *data) +{ + char *p = data; + + p += mci_output_edac_cap(p,mci->edac_cap); + p += sprintf(p, "\n"); + + return p - data; +} + +static int mci_output_mtype_cap(char *buf, unsigned long mtype_cap) +{ + char *p = buf; + int bit_idx; + + for (bit_idx = 0; bit_idx < 8 * sizeof(mtype_cap); bit_idx++) { + if ((mtype_cap >> bit_idx) & 0x1) + p += sprintf(p, "%s ", mem_types[bit_idx]); + } + + return p - buf; +} + +static ssize_t mci_supported_mem_type_show(struct mem_ctl_info *mci, char *data) +{ + char *p = data; + + p += mci_output_mtype_cap(p,mci->mtype_cap); + p += sprintf(p, "\n"); + + return p - data; +} + +static ssize_t mci_size_mb_show(struct mem_ctl_info *mci, char *data) +{ + int total_pages, csrow_idx; + + for (total_pages = csrow_idx = 0; csrow_idx < mci->nr_csrows; + csrow_idx++) { + struct csrow_info *csrow = &mci->csrows[csrow_idx]; + + if (!csrow->nr_pages) + continue; + total_pages += csrow->nr_pages; + } + + return sprintf(data,"%u\n", PAGES_TO_MiB(total_pages)); +} + +struct mcidev_attribute { + struct attribute attr; + ssize_t (*show)(struct mem_ctl_info *,char *); + ssize_t (*store)(struct mem_ctl_info *, const char *,size_t); +}; + +#define to_mci(k) container_of(k, struct mem_ctl_info, edac_mci_kobj) +#define to_mcidev_attr(a) container_of(a, struct mcidev_attribute, attr) + +static ssize_t mcidev_show(struct kobject *kobj, struct attribute *attr, + char *buffer) +{ + struct mem_ctl_info *mem_ctl_info = to_mci(kobj); + struct mcidev_attribute * mcidev_attr = to_mcidev_attr(attr); + + if (mcidev_attr->show) + return mcidev_attr->show(mem_ctl_info, buffer); + return -EIO; +} + +static ssize_t mcidev_store(struct kobject *kobj, struct attribute *attr, + const char *buffer, size_t count) +{ + struct mem_ctl_info *mem_ctl_info = to_mci(kobj); + struct mcidev_attribute * mcidev_attr = to_mcidev_attr(attr); + + if (mcidev_attr->store) + return mcidev_attr->store(mem_ctl_info, buffer, count); + return -EIO; +} + +static struct sysfs_ops mci_ops = { + .show = mcidev_show, + .store = mcidev_store +}; + +#define MCIDEV_ATTR(_name,_mode,_show,_store) \ +struct mcidev_attribute mci_attr_##_name = { \ + .attr = {.name = __stringify(_name), .mode = _mode }, \ + .show = _show, \ + .store = _store, \ +}; + +/* Control file */ +MCIDEV_ATTR(reset_counters,S_IWUSR,NULL,mci_reset_counters_store); + +/* Attribute files */ +MCIDEV_ATTR(mc_name,S_IRUGO,mci_ctl_name_show,NULL); +MCIDEV_ATTR(module_name,S_IRUGO,mci_mod_name_show,NULL); +MCIDEV_ATTR(edac_capability,S_IRUGO,mci_edac_capability_show,NULL); +MCIDEV_ATTR(size_mb,S_IRUGO,mci_size_mb_show,NULL); +MCIDEV_ATTR(seconds_since_reset,S_IRUGO,mci_seconds_show,NULL); +MCIDEV_ATTR(ue_noinfo_count,S_IRUGO,mci_ue_noinfo_show,NULL); +MCIDEV_ATTR(ce_noinfo_count,S_IRUGO,mci_ce_noinfo_show,NULL); +MCIDEV_ATTR(ue_count,S_IRUGO,mci_ue_count_show,NULL); +MCIDEV_ATTR(ce_count,S_IRUGO,mci_ce_count_show,NULL); +MCIDEV_ATTR(edac_current_capability,S_IRUGO, + mci_edac_current_capability_show,NULL); +MCIDEV_ATTR(supported_mem_type,S_IRUGO, + mci_supported_mem_type_show,NULL); + + +static struct mcidev_attribute *mci_attr[] = { + &mci_attr_reset_counters, + &mci_attr_module_name, + &mci_attr_mc_name, + &mci_attr_edac_capability, + &mci_attr_edac_current_capability, + &mci_attr_supported_mem_type, + &mci_attr_size_mb, + &mci_attr_seconds_since_reset, + &mci_attr_ue_noinfo_count, + &mci_attr_ce_noinfo_count, + &mci_attr_ue_count, + &mci_attr_ce_count, + NULL +}; + + +/* + * Release of a MC controlling instance + */ +static void edac_mci_instance_release(struct kobject *kobj) +{ + struct mem_ctl_info *mci; + mci = container_of(kobj,struct mem_ctl_info,edac_mci_kobj); + + debugf0("MC: " __FILE__ ": %s() idx=%d calling kfree\n", + __func__, mci->mc_idx); + + kfree(mci); +} + +static struct kobj_type ktype_mci = { + .release = edac_mci_instance_release, + .sysfs_ops = &mci_ops, + .default_attrs = (struct attribute **) mci_attr, +}; + +#define EDAC_DEVICE_SYMLINK "device" + +/* + * Create a new Memory Controller kobject instance, + * mc under the 'mc' directory + * + * Return: + * 0 Success + * !0 Failure + */ +static int edac_create_sysfs_mci_device(struct mem_ctl_info *mci) +{ + int i; + int err; + struct csrow_info *csrow; + struct kobject *edac_mci_kobj=&mci->edac_mci_kobj; + + debugf0("MC: " __FILE__ ": %s() idx=%d\n", __func__, mci->mc_idx); + + memset(edac_mci_kobj, 0, sizeof(*edac_mci_kobj)); + kobject_init(edac_mci_kobj); + + /* set the name of the mc object */ + err = kobject_set_name(edac_mci_kobj,"mc%d",mci->mc_idx); + if (err) + return err; + + /* link to our parent the '..../edac/mc' object */ + edac_mci_kobj->parent = &edac_memctrl_kobj; + edac_mci_kobj->ktype = &ktype_mci; + + /* register the mc kobject */ + err = kobject_register(edac_mci_kobj); + if (err) + return err; + + /* create a symlink for the device */ + err = sysfs_create_link(edac_mci_kobj, &mci->pdev->dev.kobj, + EDAC_DEVICE_SYMLINK); + if (err) { + kobject_unregister(edac_mci_kobj); + return err; + } + + /* Make directories for each CSROW object + * under the mc kobject + */ + for (i = 0; i < mci->nr_csrows; i++) { + + csrow = &mci->csrows[i]; + + /* Only expose populated CSROWs */ + if (csrow->nr_pages > 0) { + err = edac_create_csrow_object(edac_mci_kobj,csrow,i); + if (err) + goto fail; + } + } + + /* Mark this MCI instance as having sysfs entries */ + mci->sysfs_active = MCI_SYSFS_ACTIVE; + + return 0; + + + /* CSROW error: backout what has already been registered, */ +fail: + for ( i--; i >= 0; i--) { + if (csrow->nr_pages > 0) { + kobject_unregister(&mci->csrows[i].kobj); + kobject_put(&mci->csrows[i].kobj); + } + } + + kobject_unregister(edac_mci_kobj); + kobject_put(edac_mci_kobj); + + return err; +} + +/* + * remove a Memory Controller instance + */ +static void edac_remove_sysfs_mci_device(struct mem_ctl_info *mci) +{ + int i; + + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + /* remove all csrow kobjects */ + for (i = 0; i < mci->nr_csrows; i++) { + if (mci->csrows[i].nr_pages > 0) { + kobject_unregister(&mci->csrows[i].kobj); + kobject_put(&mci->csrows[i].kobj); + } + } + + sysfs_remove_link(&mci->edac_mci_kobj, EDAC_DEVICE_SYMLINK); + + kobject_unregister(&mci->edac_mci_kobj); + kobject_put(&mci->edac_mci_kobj); +} + +/* END OF sysfs data and methods */ + +#ifdef CONFIG_EDAC_DEBUG + +EXPORT_SYMBOL(edac_mc_dump_channel); + +void edac_mc_dump_channel(struct channel_info *chan) +{ + debugf4("\tchannel = %p\n", chan); + debugf4("\tchannel->chan_idx = %d\n", chan->chan_idx); + debugf4("\tchannel->ce_count = %d\n", chan->ce_count); + debugf4("\tchannel->label = '%s'\n", chan->label); + debugf4("\tchannel->csrow = %p\n\n", chan->csrow); +} + + +EXPORT_SYMBOL(edac_mc_dump_csrow); + +void edac_mc_dump_csrow(struct csrow_info *csrow) +{ + debugf4("\tcsrow = %p\n", csrow); + debugf4("\tcsrow->csrow_idx = %d\n", csrow->csrow_idx); + debugf4("\tcsrow->first_page = 0x%lx\n", + csrow->first_page); + debugf4("\tcsrow->last_page = 0x%lx\n", csrow->last_page); + debugf4("\tcsrow->page_mask = 0x%lx\n", csrow->page_mask); + debugf4("\tcsrow->nr_pages = 0x%x\n", csrow->nr_pages); + debugf4("\tcsrow->nr_channels = %d\n", + csrow->nr_channels); + debugf4("\tcsrow->channels = %p\n", csrow->channels); + debugf4("\tcsrow->mci = %p\n\n", csrow->mci); +} + + +EXPORT_SYMBOL(edac_mc_dump_mci); + +void edac_mc_dump_mci(struct mem_ctl_info *mci) +{ + debugf3("\tmci = %p\n", mci); + debugf3("\tmci->mtype_cap = %lx\n", mci->mtype_cap); + debugf3("\tmci->edac_ctl_cap = %lx\n", mci->edac_ctl_cap); + debugf3("\tmci->edac_cap = %lx\n", mci->edac_cap); + debugf4("\tmci->edac_check = %p\n", mci->edac_check); + debugf3("\tmci->nr_csrows = %d, csrows = %p\n", + mci->nr_csrows, mci->csrows); + debugf3("\tpdev = %p\n", mci->pdev); + debugf3("\tmod_name:ctl_name = %s:%s\n", + mci->mod_name, mci->ctl_name); + debugf3("\tpvt_info = %p\n\n", mci->pvt_info); +} + + +#endif /* CONFIG_EDAC_DEBUG */ + +/* 'ptr' points to a possibly unaligned item X such that sizeof(X) is 'size'. + * Adjust 'ptr' so that its alignment is at least as stringent as what the + * compiler would provide for X and return the aligned result. + * + * If 'size' is a constant, the compiler will optimize this whole function + * down to either a no-op or the addition of a constant to the value of 'ptr'. + */ +static inline char * align_ptr (void *ptr, unsigned size) +{ + unsigned align, r; + + /* Here we assume that the alignment of a "long long" is the most + * stringent alignment that the compiler will ever provide by default. + * As far as I know, this is a reasonable assumption. + */ + if (size > sizeof(long)) + align = sizeof(long long); + else if (size > sizeof(int)) + align = sizeof(long); + else if (size > sizeof(short)) + align = sizeof(int); + else if (size > sizeof(char)) + align = sizeof(short); + else + return (char *) ptr; + + r = size % align; + + if (r == 0) + return (char *) ptr; + + return (char *) (((unsigned long) ptr) + align - r); +} + + +EXPORT_SYMBOL(edac_mc_alloc); + +/** + * edac_mc_alloc: Allocate a struct mem_ctl_info structure + * @size_pvt: size of private storage needed + * @nr_csrows: Number of CWROWS needed for this MC + * @nr_chans: Number of channels for the MC + * + * Everything is kmalloc'ed as one big chunk - more efficient. + * Only can be used if all structures have the same lifetime - otherwise + * you have to allocate and initialize your own structures. + * + * Use edac_mc_free() to free mc structures allocated by this function. + * + * Returns: + * NULL allocation failed + * struct mem_ctl_info pointer + */ +struct mem_ctl_info *edac_mc_alloc(unsigned sz_pvt, unsigned nr_csrows, + unsigned nr_chans) +{ + struct mem_ctl_info *mci; + struct csrow_info *csi, *csrow; + struct channel_info *chi, *chp, *chan; + void *pvt; + unsigned size; + int row, chn; + + /* Figure out the offsets of the various items from the start of an mc + * structure. We want the alignment of each item to be at least as + * stringent as what the compiler would provide if we could simply + * hardcode everything into a single struct. + */ + mci = (struct mem_ctl_info *) 0; + csi = (struct csrow_info *)align_ptr(&mci[1], sizeof(*csi)); + chi = (struct channel_info *) + align_ptr(&csi[nr_csrows], sizeof(*chi)); + pvt = align_ptr(&chi[nr_chans * nr_csrows], sz_pvt); + size = ((unsigned long) pvt) + sz_pvt; + + if ((mci = kmalloc(size, GFP_KERNEL)) == NULL) + return NULL; + + /* Adjust pointers so they point within the memory we just allocated + * rather than an imaginary chunk of memory located at address 0. + */ + csi = (struct csrow_info *) (((char *) mci) + ((unsigned long) csi)); + chi = (struct channel_info *) (((char *) mci) + ((unsigned long) chi)); + pvt = sz_pvt ? (((char *) mci) + ((unsigned long) pvt)) : NULL; + + memset(mci, 0, size); /* clear all fields */ + + mci->csrows = csi; + mci->pvt_info = pvt; + mci->nr_csrows = nr_csrows; + + for (row = 0; row < nr_csrows; row++) { + csrow = &csi[row]; + csrow->csrow_idx = row; + csrow->mci = mci; + csrow->nr_channels = nr_chans; + chp = &chi[row * nr_chans]; + csrow->channels = chp; + + for (chn = 0; chn < nr_chans; chn++) { + chan = &chp[chn]; + chan->chan_idx = chn; + chan->csrow = csrow; + } + } + + return mci; +} + + +EXPORT_SYMBOL(edac_mc_free); + +/** + * edac_mc_free: Free a previously allocated 'mci' structure + * @mci: pointer to a struct mem_ctl_info structure + * + * Free up a previously allocated mci structure + * A MCI structure can be in 2 states after being allocated + * by edac_mc_alloc(). + * 1) Allocated in a MC driver's probe, but not yet committed + * 2) Allocated and committed, by a call to edac_mc_add_mc() + * edac_mc_add_mc() is the function that adds the sysfs entries + * thus, this free function must determine which state the 'mci' + * structure is in, then either free it directly or + * perform kobject cleanup by calling edac_remove_sysfs_mci_device(). + * + * VOID Return + */ +void edac_mc_free(struct mem_ctl_info *mci) +{ + /* only if sysfs entries for this mci instance exist + * do we remove them and defer the actual kfree via + * the kobject 'release()' callback. + * + * Otherwise, do a straight kfree now. + */ + if (mci->sysfs_active == MCI_SYSFS_ACTIVE) + edac_remove_sysfs_mci_device(mci); + else + kfree(mci); +} + + + +EXPORT_SYMBOL(edac_mc_find_mci_by_pdev); + +struct mem_ctl_info *edac_mc_find_mci_by_pdev(struct pci_dev *pdev) +{ + struct mem_ctl_info *mci; + struct list_head *item; + + debugf3("MC: " __FILE__ ": %s()\n", __func__); + + list_for_each(item, &mc_devices) { + mci = list_entry(item, struct mem_ctl_info, link); + + if (mci->pdev == pdev) + return mci; + } + + return NULL; +} + +static int add_mc_to_global_list (struct mem_ctl_info *mci) +{ + struct list_head *item, *insert_before; + struct mem_ctl_info *p; + int i; + + if (list_empty(&mc_devices)) { + mci->mc_idx = 0; + insert_before = &mc_devices; + } else { + if (edac_mc_find_mci_by_pdev(mci->pdev)) { + printk(KERN_WARNING + "EDAC MC: %s (%s) %s %s already assigned %d\n", + mci->pdev->dev.bus_id, pci_name(mci->pdev), + mci->mod_name, mci->ctl_name, mci->mc_idx); + return 1; + } + + insert_before = NULL; + i = 0; + + list_for_each(item, &mc_devices) { + p = list_entry(item, struct mem_ctl_info, link); + + if (p->mc_idx != i) { + insert_before = item; + break; + } + + i++; + } + + mci->mc_idx = i; + + if (insert_before == NULL) + insert_before = &mc_devices; + } + + list_add_tail_rcu(&mci->link, insert_before); + return 0; +} + + + +EXPORT_SYMBOL(edac_mc_add_mc); + +/** + * edac_mc_add_mc: Insert the 'mci' structure into the mci global list + * @mci: pointer to the mci structure to be added to the list + * + * Return: + * 0 Success + * !0 Failure + */ + +/* FIXME - should a warning be printed if no error detection? correction? */ +int edac_mc_add_mc(struct mem_ctl_info *mci) +{ + int rc = 1; + + debugf0("MC: " __FILE__ ": %s()\n", __func__); +#ifdef CONFIG_EDAC_DEBUG + if (edac_debug_level >= 3) + edac_mc_dump_mci(mci); + if (edac_debug_level >= 4) { + int i; + + for (i = 0; i < mci->nr_csrows; i++) { + int j; + edac_mc_dump_csrow(&mci->csrows[i]); + for (j = 0; j < mci->csrows[i].nr_channels; j++) + edac_mc_dump_channel(&mci->csrows[i]. + channels[j]); + } + } +#endif + down(&mem_ctls_mutex); + + if (add_mc_to_global_list(mci)) + goto finish; + + /* set load time so that error rate can be tracked */ + mci->start_time = jiffies; + + if (edac_create_sysfs_mci_device(mci)) { + printk(KERN_WARNING + "EDAC MC%d: failed to create sysfs device\n", + mci->mc_idx); + /* FIXME - should there be an error code and unwind? */ + goto finish; + } + + /* Report action taken */ + printk(KERN_INFO + "EDAC MC%d: Giving out device to %s %s: PCI %s\n", + mci->mc_idx, mci->mod_name, mci->ctl_name, + pci_name(mci->pdev)); + + + rc = 0; + +finish: + up(&mem_ctls_mutex); + return rc; +} + + + +static void complete_mc_list_del (struct rcu_head *head) +{ + struct mem_ctl_info *mci; + + mci = container_of(head, struct mem_ctl_info, rcu); + INIT_LIST_HEAD(&mci->link); + complete(&mci->complete); +} + +static void del_mc_from_global_list (struct mem_ctl_info *mci) +{ + list_del_rcu(&mci->link); + init_completion(&mci->complete); + call_rcu(&mci->rcu, complete_mc_list_del); + wait_for_completion(&mci->complete); +} + +EXPORT_SYMBOL(edac_mc_del_mc); + +/** + * edac_mc_del_mc: Remove the specified mci structure from global list + * @mci: Pointer to struct mem_ctl_info structure + * + * Returns: + * 0 Success + * 1 Failure + */ +int edac_mc_del_mc(struct mem_ctl_info *mci) +{ + int rc = 1; + + debugf0("MC%d: " __FILE__ ": %s()\n", mci->mc_idx, __func__); + down(&mem_ctls_mutex); + del_mc_from_global_list(mci); + printk(KERN_INFO + "EDAC MC%d: Removed device %d for %s %s: PCI %s\n", + mci->mc_idx, mci->mc_idx, mci->mod_name, mci->ctl_name, + pci_name(mci->pdev)); + rc = 0; + up(&mem_ctls_mutex); + + return rc; +} + + +EXPORT_SYMBOL(edac_mc_scrub_block); + +void edac_mc_scrub_block(unsigned long page, unsigned long offset, + u32 size) +{ + struct page *pg; + void *virt_addr; + unsigned long flags = 0; + + debugf3("MC: " __FILE__ ": %s()\n", __func__); + + /* ECC error page was not in our memory. Ignore it. */ + if(!pfn_valid(page)) + return; + + /* Find the actual page structure then map it and fix */ + pg = pfn_to_page(page); + + if (PageHighMem(pg)) + local_irq_save(flags); + + virt_addr = kmap_atomic(pg, KM_BOUNCE_READ); + + /* Perform architecture specific atomic scrub operation */ + atomic_scrub(virt_addr + offset, size); + + /* Unmap and complete */ + kunmap_atomic(virt_addr, KM_BOUNCE_READ); + + if (PageHighMem(pg)) + local_irq_restore(flags); +} + + +/* FIXME - should return -1 */ +EXPORT_SYMBOL(edac_mc_find_csrow_by_page); + +int edac_mc_find_csrow_by_page(struct mem_ctl_info *mci, + unsigned long page) +{ + struct csrow_info *csrows = mci->csrows; + int row, i; + + debugf1("MC%d: " __FILE__ ": %s(): 0x%lx\n", mci->mc_idx, __func__, + page); + row = -1; + + for (i = 0; i < mci->nr_csrows; i++) { + struct csrow_info *csrow = &csrows[i]; + + if (csrow->nr_pages == 0) + continue; + + debugf3("MC%d: " __FILE__ + ": %s(): first(0x%lx) page(0x%lx)" + " last(0x%lx) mask(0x%lx)\n", mci->mc_idx, + __func__, csrow->first_page, page, + csrow->last_page, csrow->page_mask); + + if ((page >= csrow->first_page) && + (page <= csrow->last_page) && + ((page & csrow->page_mask) == + (csrow->first_page & csrow->page_mask))) { + row = i; + break; + } + } + + if (row == -1) + printk(KERN_ERR + "EDAC MC%d: could not look up page error address %lx\n", + mci->mc_idx, (unsigned long) page); + + return row; +} + + +EXPORT_SYMBOL(edac_mc_handle_ce); + +/* FIXME - setable log (warning/emerg) levels */ +/* FIXME - integrate with evlog: http://evlog.sourceforge.net/ */ +void edac_mc_handle_ce(struct mem_ctl_info *mci, + unsigned long page_frame_number, + unsigned long offset_in_page, + unsigned long syndrome, int row, int channel, + const char *msg) +{ + unsigned long remapped_page; + + debugf3("MC%d: " __FILE__ ": %s()\n", mci->mc_idx, __func__); + + /* FIXME - maybe make panic on INTERNAL ERROR an option */ + if (row >= mci->nr_csrows || row < 0) { + /* something is wrong */ + printk(KERN_ERR + "EDAC MC%d: INTERNAL ERROR: row out of range (%d >= %d)\n", + mci->mc_idx, row, mci->nr_csrows); + edac_mc_handle_ce_no_info(mci, "INTERNAL ERROR"); + return; + } + if (channel >= mci->csrows[row].nr_channels || channel < 0) { + /* something is wrong */ + printk(KERN_ERR + "EDAC MC%d: INTERNAL ERROR: channel out of range " + "(%d >= %d)\n", + mci->mc_idx, channel, mci->csrows[row].nr_channels); + edac_mc_handle_ce_no_info(mci, "INTERNAL ERROR"); + return; + } + + if (log_ce) + /* FIXME - put in DIMM location */ + printk(KERN_WARNING + "EDAC MC%d: CE page 0x%lx, offset 0x%lx," + " grain %d, syndrome 0x%lx, row %d, channel %d," + " label \"%s\": %s\n", mci->mc_idx, + page_frame_number, offset_in_page, + mci->csrows[row].grain, syndrome, row, channel, + mci->csrows[row].channels[channel].label, msg); + + mci->ce_count++; + mci->csrows[row].ce_count++; + mci->csrows[row].channels[channel].ce_count++; + + if (mci->scrub_mode & SCRUB_SW_SRC) { + /* + * Some MC's can remap memory so that it is still available + * at a different address when PCI devices map into memory. + * MC's that can't do this lose the memory where PCI devices + * are mapped. This mapping is MC dependant and so we call + * back into the MC driver for it to map the MC page to + * a physical (CPU) page which can then be mapped to a virtual + * page - which can then be scrubbed. + */ + remapped_page = mci->ctl_page_to_phys ? + mci->ctl_page_to_phys(mci, page_frame_number) : + page_frame_number; + + edac_mc_scrub_block(remapped_page, offset_in_page, + mci->csrows[row].grain); + } +} + + +EXPORT_SYMBOL(edac_mc_handle_ce_no_info); + +void edac_mc_handle_ce_no_info(struct mem_ctl_info *mci, + const char *msg) +{ + if (log_ce) + printk(KERN_WARNING + "EDAC MC%d: CE - no information available: %s\n", + mci->mc_idx, msg); + mci->ce_noinfo_count++; + mci->ce_count++; +} + + +EXPORT_SYMBOL(edac_mc_handle_ue); + +void edac_mc_handle_ue(struct mem_ctl_info *mci, + unsigned long page_frame_number, + unsigned long offset_in_page, int row, + const char *msg) +{ + int len = EDAC_MC_LABEL_LEN * 4; + char labels[len + 1]; + char *pos = labels; + int chan; + int chars; + + debugf3("MC%d: " __FILE__ ": %s()\n", mci->mc_idx, __func__); + + /* FIXME - maybe make panic on INTERNAL ERROR an option */ + if (row >= mci->nr_csrows || row < 0) { + /* something is wrong */ + printk(KERN_ERR + "EDAC MC%d: INTERNAL ERROR: row out of range (%d >= %d)\n", + mci->mc_idx, row, mci->nr_csrows); + edac_mc_handle_ue_no_info(mci, "INTERNAL ERROR"); + return; + } + + chars = snprintf(pos, len + 1, "%s", + mci->csrows[row].channels[0].label); + len -= chars; + pos += chars; + for (chan = 1; (chan < mci->csrows[row].nr_channels) && (len > 0); + chan++) { + chars = snprintf(pos, len + 1, ":%s", + mci->csrows[row].channels[chan].label); + len -= chars; + pos += chars; + } + + if (log_ue) + printk(KERN_EMERG + "EDAC MC%d: UE page 0x%lx, offset 0x%lx, grain %d, row %d," + " labels \"%s\": %s\n", mci->mc_idx, + page_frame_number, offset_in_page, + mci->csrows[row].grain, row, labels, msg); + + if (panic_on_ue) + panic + ("EDAC MC%d: UE page 0x%lx, offset 0x%lx, grain %d, row %d," + " labels \"%s\": %s\n", mci->mc_idx, + page_frame_number, offset_in_page, + mci->csrows[row].grain, row, labels, msg); + + mci->ue_count++; + mci->csrows[row].ue_count++; +} + + +EXPORT_SYMBOL(edac_mc_handle_ue_no_info); + +void edac_mc_handle_ue_no_info(struct mem_ctl_info *mci, + const char *msg) +{ + if (panic_on_ue) + panic("EDAC MC%d: Uncorrected Error", mci->mc_idx); + + if (log_ue) + printk(KERN_WARNING + "EDAC MC%d: UE - no information available: %s\n", + mci->mc_idx, msg); + mci->ue_noinfo_count++; + mci->ue_count++; +} + + +#ifdef CONFIG_PCI + +static u16 get_pci_parity_status(struct pci_dev *dev, int secondary) +{ + int where; + u16 status; + + where = secondary ? PCI_SEC_STATUS : PCI_STATUS; + pci_read_config_word(dev, where, &status); + + /* If we get back 0xFFFF then we must suspect that the card has been pulled but + the Linux PCI layer has not yet finished cleaning up. We don't want to report + on such devices */ + + if (status == 0xFFFF) { + u32 sanity; + pci_read_config_dword(dev, 0, &sanity); + if (sanity == 0xFFFFFFFF) + return 0; + } + status &= PCI_STATUS_DETECTED_PARITY | PCI_STATUS_SIG_SYSTEM_ERROR | + PCI_STATUS_PARITY; + + if (status) + /* reset only the bits we are interested in */ + pci_write_config_word(dev, where, status); + + return status; +} + +typedef void (*pci_parity_check_fn_t) (struct pci_dev *dev); + +/* Clear any PCI parity errors logged by this device. */ +static void edac_pci_dev_parity_clear( struct pci_dev *dev ) +{ + u8 header_type; + + get_pci_parity_status(dev, 0); + + /* read the device TYPE, looking for bridges */ + pci_read_config_byte(dev, PCI_HEADER_TYPE, &header_type); + + if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) + get_pci_parity_status(dev, 1); +} + +/* + * PCI Parity polling + * + */ +static void edac_pci_dev_parity_test(struct pci_dev *dev) +{ + u16 status; + u8 header_type; + + /* read the STATUS register on this device + */ + status = get_pci_parity_status(dev, 0); + + debugf2("PCI STATUS= 0x%04x %s\n", status, dev->dev.bus_id ); + + /* check the status reg for errors */ + if (status) { + if (status & (PCI_STATUS_SIG_SYSTEM_ERROR)) + printk(KERN_CRIT + "EDAC PCI- " + "Signaled System Error on %s\n", + pci_name (dev)); + + if (status & (PCI_STATUS_PARITY)) { + printk(KERN_CRIT + "EDAC PCI- " + "Master Data Parity Error on %s\n", + pci_name (dev)); + + atomic_inc(&pci_parity_count); + } + + if (status & (PCI_STATUS_DETECTED_PARITY)) { + printk(KERN_CRIT + "EDAC PCI- " + "Detected Parity Error on %s\n", + pci_name (dev)); + + atomic_inc(&pci_parity_count); + } + } + + /* read the device TYPE, looking for bridges */ + pci_read_config_byte(dev, PCI_HEADER_TYPE, &header_type); + + debugf2("PCI HEADER TYPE= 0x%02x %s\n", header_type, dev->dev.bus_id ); + + if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { + /* On bridges, need to examine secondary status register */ + status = get_pci_parity_status(dev, 1); + + debugf2("PCI SEC_STATUS= 0x%04x %s\n", + status, dev->dev.bus_id ); + + /* check the secondary status reg for errors */ + if (status) { + if (status & (PCI_STATUS_SIG_SYSTEM_ERROR)) + printk(KERN_CRIT + "EDAC PCI-Bridge- " + "Signaled System Error on %s\n", + pci_name (dev)); + + if (status & (PCI_STATUS_PARITY)) { + printk(KERN_CRIT + "EDAC PCI-Bridge- " + "Master Data Parity Error on %s\n", + pci_name (dev)); + + atomic_inc(&pci_parity_count); + } + + if (status & (PCI_STATUS_DETECTED_PARITY)) { + printk(KERN_CRIT + "EDAC PCI-Bridge- " + "Detected Parity Error on %s\n", + pci_name (dev)); + + atomic_inc(&pci_parity_count); + } + } + } +} + +/* + * check_dev_on_list: Scan for a PCI device on a white/black list + * @list: an EDAC &edac_pci_device_list white/black list pointer + * @free_index: index of next free entry on the list + * @pci_dev: PCI Device pointer + * + * see if list contains the device. + * + * Returns: 0 not found + * 1 found on list + */ +static int check_dev_on_list(struct edac_pci_device_list *list, int free_index, + struct pci_dev *dev) +{ + int i; + int rc = 0; /* Assume not found */ + unsigned short vendor=dev->vendor; + unsigned short device=dev->device; + + /* Scan the list, looking for a vendor/device match + */ + for (i = 0; i < free_index; i++, list++ ) { + if ( (list->vendor == vendor ) && + (list->device == device )) { + rc = 1; + break; + } + } + + return rc; +} + +/* + * pci_dev parity list iterator + * Scan the PCI device list for one iteration, looking for SERRORs + * Master Parity ERRORS or Parity ERRORs on primary or secondary devices + */ +static inline void edac_pci_dev_parity_iterator(pci_parity_check_fn_t fn) +{ + struct pci_dev *dev=NULL; + + /* request for kernel access to the next PCI device, if any, + * and while we are looking at it have its reference count + * bumped until we are done with it + */ + while((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) { + + /* if whitelist exists then it has priority, so only scan those + * devices on the whitelist + */ + if (pci_whitelist_count > 0 ) { + if (check_dev_on_list(pci_whitelist, + pci_whitelist_count, dev)) + fn(dev); + } else { + /* + * if no whitelist, then check if this devices is + * blacklisted + */ + if (!check_dev_on_list(pci_blacklist, + pci_blacklist_count, dev)) + fn(dev); + } + } +} + +static void do_pci_parity_check(void) +{ + unsigned long flags; + int before_count; + + debugf3("MC: " __FILE__ ": %s()\n", __func__); + + if (!check_pci_parity) + return; + + before_count = atomic_read(&pci_parity_count); + + /* scan all PCI devices looking for a Parity Error on devices and + * bridges + */ + local_irq_save(flags); + edac_pci_dev_parity_iterator(edac_pci_dev_parity_test); + local_irq_restore(flags); + + /* Only if operator has selected panic on PCI Error */ + if (panic_on_pci_parity) { + /* If the count is different 'after' from 'before' */ + if (before_count != atomic_read(&pci_parity_count)) + panic("EDAC: PCI Parity Error"); + } +} + + +static inline void clear_pci_parity_errors(void) +{ + /* Clear any PCI bus parity errors that devices initially have logged + * in their registers. + */ + edac_pci_dev_parity_iterator(edac_pci_dev_parity_clear); +} + + +#else /* CONFIG_PCI */ + + +static inline void do_pci_parity_check(void) +{ + /* no-op */ +} + + +static inline void clear_pci_parity_errors(void) +{ + /* no-op */ +} + + +#endif /* CONFIG_PCI */ + +/* + * Iterate over all MC instances and check for ECC, et al, errors + */ +static inline void check_mc_devices (void) +{ + unsigned long flags; + struct list_head *item; + struct mem_ctl_info *mci; + + debugf3("MC: " __FILE__ ": %s()\n", __func__); + + /* during poll, have interrupts off */ + local_irq_save(flags); + + list_for_each(item, &mc_devices) { + mci = list_entry(item, struct mem_ctl_info, link); + + if (mci->edac_check != NULL) + mci->edac_check(mci); + } + + local_irq_restore(flags); +} + + +/* + * Check MC status every poll_msec. + * Check PCI status every poll_msec as well. + * + * This where the work gets done for edac. + * + * SMP safe, doesn't use NMI, and auto-rate-limits. + */ +static void do_edac_check(void) +{ + + debugf3("MC: " __FILE__ ": %s()\n", __func__); + + check_mc_devices(); + + do_pci_parity_check(); +} + + +/* + * EDAC thread state information + */ +struct bs_thread_info +{ + struct task_struct *task; + struct completion *event; + char *name; + void (*run)(void); +}; + +static struct bs_thread_info bs_thread; + +/* + * edac_kernel_thread + * This the kernel thread that processes edac operations + * in a normal thread environment + */ +static int edac_kernel_thread(void *arg) +{ + struct bs_thread_info *thread = (struct bs_thread_info *) arg; + + /* detach thread */ + daemonize(thread->name); + + current->exit_signal = SIGCHLD; + allow_signal(SIGKILL); + thread->task = current; + + /* indicate to starting task we have started */ + complete(thread->event); + + /* loop forever, until we are told to stop */ + while(thread->run != NULL) { + void (*run)(void); + + /* call the function to check the memory controllers */ + run = thread->run; + if (run) + run(); + + if (signal_pending(current)) + flush_signals(current); + + /* ensure we are interruptable */ + set_current_state(TASK_INTERRUPTIBLE); + + /* goto sleep for the interval */ + schedule_timeout((HZ * poll_msec) / 1000); + try_to_freeze(); + } + + /* notify waiter that we are exiting */ + complete(thread->event); + + return 0; +} + +/* + * edac_mc_init + * module initialization entry point + */ +static int __init edac_mc_init(void) +{ + int ret; + struct completion event; + + printk(KERN_INFO "MC: " __FILE__ " version " EDAC_MC_VERSION "\n"); + + /* + * Harvest and clear any boot/initialization PCI parity errors + * + * FIXME: This only clears errors logged by devices present at time of + * module initialization. We should also do an initial clear + * of each newly hotplugged device. + */ + clear_pci_parity_errors(); + + /* perform check for first time to harvest boot leftovers */ + do_edac_check(); + + /* Create the MC sysfs entires */ + if (edac_sysfs_memctrl_setup()) { + printk(KERN_ERR "EDAC MC: Error initializing sysfs code\n"); + return -ENODEV; + } + + /* Create the PCI parity sysfs entries */ + if (edac_sysfs_pci_setup()) { + edac_sysfs_memctrl_teardown(); + printk(KERN_ERR "EDAC PCI: Error initializing sysfs code\n"); + return -ENODEV; + } + + /* Create our kernel thread */ + init_completion(&event); + bs_thread.event = &event; + bs_thread.name = "kedac"; + bs_thread.run = do_edac_check; + + /* create our kernel thread */ + ret = kernel_thread(edac_kernel_thread, &bs_thread, CLONE_KERNEL); + if (ret < 0) { + /* remove the sysfs entries */ + edac_sysfs_memctrl_teardown(); + edac_sysfs_pci_teardown(); + return -ENOMEM; + } + + /* wait for our kernel theard ack that it is up and running */ + wait_for_completion(&event); + + return 0; +} + + +/* + * edac_mc_exit() + * module exit/termination functioni + */ +static void __exit edac_mc_exit(void) +{ + struct completion event; + + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + init_completion(&event); + bs_thread.event = &event; + + /* As soon as ->run is set to NULL, the task could disappear, + * so we need to hold tasklist_lock until we have sent the signal + */ + read_lock(&tasklist_lock); + bs_thread.run = NULL; + send_sig(SIGKILL, bs_thread.task, 1); + read_unlock(&tasklist_lock); + wait_for_completion(&event); + + /* tear down the sysfs device */ + edac_sysfs_memctrl_teardown(); + edac_sysfs_pci_teardown(); +} + + + + +module_init(edac_mc_init); +module_exit(edac_mc_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Linux Networx (http://lnxi.com) Thayne Harbaugh et al\n" + "Based on.work by Dan Hollis et al"); +MODULE_DESCRIPTION("Core library routines for MC reporting"); + +module_param(panic_on_ue, int, 0644); +MODULE_PARM_DESC(panic_on_ue, "Panic on uncorrected error: 0=off 1=on"); +module_param(check_pci_parity, int, 0644); +MODULE_PARM_DESC(check_pci_parity, "Check for PCI bus parity errors: 0=off 1=on"); +module_param(panic_on_pci_parity, int, 0644); +MODULE_PARM_DESC(panic_on_pci_parity, "Panic on PCI Bus Parity error: 0=off 1=on"); +module_param(log_ue, int, 0644); +MODULE_PARM_DESC(log_ue, "Log uncorrectable error to console: 0=off 1=on"); +module_param(log_ce, int, 0644); +MODULE_PARM_DESC(log_ce, "Log correctable error to console: 0=off 1=on"); +module_param(poll_msec, int, 0644); +MODULE_PARM_DESC(poll_msec, "Polling period in milliseconds"); +#ifdef CONFIG_EDAC_DEBUG +module_param(edac_debug_level, int, 0644); +MODULE_PARM_DESC(edac_debug_level, "Debug level"); +#endif diff --git a/drivers/edac/edac_mc.h b/drivers/edac/edac_mc.h new file mode 100644 index 00000000000..75ecf484a43 --- /dev/null +++ b/drivers/edac/edac_mc.h @@ -0,0 +1,448 @@ +/* + * MC kernel module + * (C) 2003 Linux Networx (http://lnxi.com) + * This file may be distributed under the terms of the + * GNU General Public License. + * + * Written by Thayne Harbaugh + * Based on work by Dan Hollis and others. + * http://www.anime.net/~goemon/linux-ecc/ + * + * NMI handling support added by + * Dave Peterson + * + * $Id: edac_mc.h,v 1.4.2.10 2005/10/05 00:43:44 dsp_llnl Exp $ + * + */ + + +#ifndef _EDAC_MC_H_ +#define _EDAC_MC_H_ + + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +#define EDAC_MC_LABEL_LEN 31 +#define MC_PROC_NAME_MAX_LEN 7 + +#if PAGE_SHIFT < 20 +#define PAGES_TO_MiB( pages ) ( ( pages ) >> ( 20 - PAGE_SHIFT ) ) +#else /* PAGE_SHIFT > 20 */ +#define PAGES_TO_MiB( pages ) ( ( pages ) << ( PAGE_SHIFT - 20 ) ) +#endif + +#ifdef CONFIG_EDAC_DEBUG +extern int edac_debug_level; +#define edac_debug_printk(level, fmt, args...) \ +do { if (level <= edac_debug_level) printk(KERN_DEBUG fmt, ##args); } while(0) +#define debugf0( ... ) edac_debug_printk(0, __VA_ARGS__ ) +#define debugf1( ... ) edac_debug_printk(1, __VA_ARGS__ ) +#define debugf2( ... ) edac_debug_printk(2, __VA_ARGS__ ) +#define debugf3( ... ) edac_debug_printk(3, __VA_ARGS__ ) +#define debugf4( ... ) edac_debug_printk(4, __VA_ARGS__ ) +#else /* !CONFIG_EDAC_DEBUG */ +#define debugf0( ... ) +#define debugf1( ... ) +#define debugf2( ... ) +#define debugf3( ... ) +#define debugf4( ... ) +#endif /* !CONFIG_EDAC_DEBUG */ + + +#define bs_xstr(s) bs_str(s) +#define bs_str(s) #s +#define BS_MOD_STR bs_xstr(KBUILD_BASENAME) + +#define BIT(x) (1 << (x)) + +#define PCI_VEND_DEV(vend, dev) PCI_VENDOR_ID_ ## vend, PCI_DEVICE_ID_ ## vend ## _ ## dev + +/* memory devices */ +enum dev_type { + DEV_UNKNOWN = 0, + DEV_X1, + DEV_X2, + DEV_X4, + DEV_X8, + DEV_X16, + DEV_X32, /* Do these parts exist? */ + DEV_X64 /* Do these parts exist? */ +}; + +#define DEV_FLAG_UNKNOWN BIT(DEV_UNKNOWN) +#define DEV_FLAG_X1 BIT(DEV_X1) +#define DEV_FLAG_X2 BIT(DEV_X2) +#define DEV_FLAG_X4 BIT(DEV_X4) +#define DEV_FLAG_X8 BIT(DEV_X8) +#define DEV_FLAG_X16 BIT(DEV_X16) +#define DEV_FLAG_X32 BIT(DEV_X32) +#define DEV_FLAG_X64 BIT(DEV_X64) + +/* memory types */ +enum mem_type { + MEM_EMPTY = 0, /* Empty csrow */ + MEM_RESERVED, /* Reserved csrow type */ + MEM_UNKNOWN, /* Unknown csrow type */ + MEM_FPM, /* Fast page mode */ + MEM_EDO, /* Extended data out */ + MEM_BEDO, /* Burst Extended data out */ + MEM_SDR, /* Single data rate SDRAM */ + MEM_RDR, /* Registered single data rate SDRAM */ + MEM_DDR, /* Double data rate SDRAM */ + MEM_RDDR, /* Registered Double data rate SDRAM */ + MEM_RMBS /* Rambus DRAM */ +}; + +#define MEM_FLAG_EMPTY BIT(MEM_EMPTY) +#define MEM_FLAG_RESERVED BIT(MEM_RESERVED) +#define MEM_FLAG_UNKNOWN BIT(MEM_UNKNOWN) +#define MEM_FLAG_FPM BIT(MEM_FPM) +#define MEM_FLAG_EDO BIT(MEM_EDO) +#define MEM_FLAG_BEDO BIT(MEM_BEDO) +#define MEM_FLAG_SDR BIT(MEM_SDR) +#define MEM_FLAG_RDR BIT(MEM_RDR) +#define MEM_FLAG_DDR BIT(MEM_DDR) +#define MEM_FLAG_RDDR BIT(MEM_RDDR) +#define MEM_FLAG_RMBS BIT(MEM_RMBS) + + +/* chipset Error Detection and Correction capabilities and mode */ +enum edac_type { + EDAC_UNKNOWN = 0, /* Unknown if ECC is available */ + EDAC_NONE, /* Doesnt support ECC */ + EDAC_RESERVED, /* Reserved ECC type */ + EDAC_PARITY, /* Detects parity errors */ + EDAC_EC, /* Error Checking - no correction */ + EDAC_SECDED, /* Single bit error correction, Double detection */ + EDAC_S2ECD2ED, /* Chipkill x2 devices - do these exist? */ + EDAC_S4ECD4ED, /* Chipkill x4 devices */ + EDAC_S8ECD8ED, /* Chipkill x8 devices */ + EDAC_S16ECD16ED, /* Chipkill x16 devices */ +}; + +#define EDAC_FLAG_UNKNOWN BIT(EDAC_UNKNOWN) +#define EDAC_FLAG_NONE BIT(EDAC_NONE) +#define EDAC_FLAG_PARITY BIT(EDAC_PARITY) +#define EDAC_FLAG_EC BIT(EDAC_EC) +#define EDAC_FLAG_SECDED BIT(EDAC_SECDED) +#define EDAC_FLAG_S2ECD2ED BIT(EDAC_S2ECD2ED) +#define EDAC_FLAG_S4ECD4ED BIT(EDAC_S4ECD4ED) +#define EDAC_FLAG_S8ECD8ED BIT(EDAC_S8ECD8ED) +#define EDAC_FLAG_S16ECD16ED BIT(EDAC_S16ECD16ED) + + +/* scrubbing capabilities */ +enum scrub_type { + SCRUB_UNKNOWN = 0, /* Unknown if scrubber is available */ + SCRUB_NONE, /* No scrubber */ + SCRUB_SW_PROG, /* SW progressive (sequential) scrubbing */ + SCRUB_SW_SRC, /* Software scrub only errors */ + SCRUB_SW_PROG_SRC, /* Progressive software scrub from an error */ + SCRUB_SW_TUNABLE, /* Software scrub frequency is tunable */ + SCRUB_HW_PROG, /* HW progressive (sequential) scrubbing */ + SCRUB_HW_SRC, /* Hardware scrub only errors */ + SCRUB_HW_PROG_SRC, /* Progressive hardware scrub from an error */ + SCRUB_HW_TUNABLE /* Hardware scrub frequency is tunable */ +}; + +#define SCRUB_FLAG_SW_PROG BIT(SCRUB_SW_PROG) +#define SCRUB_FLAG_SW_SRC BIT(SCRUB_SW_SRC_CORR) +#define SCRUB_FLAG_SW_PROG_SRC BIT(SCRUB_SW_PROG_SRC_CORR) +#define SCRUB_FLAG_SW_TUN BIT(SCRUB_SW_SCRUB_TUNABLE) +#define SCRUB_FLAG_HW_PROG BIT(SCRUB_HW_PROG) +#define SCRUB_FLAG_HW_SRC BIT(SCRUB_HW_SRC_CORR) +#define SCRUB_FLAG_HW_PROG_SRC BIT(SCRUB_HW_PROG_SRC_CORR) +#define SCRUB_FLAG_HW_TUN BIT(SCRUB_HW_TUNABLE) + +enum mci_sysfs_status { + MCI_SYSFS_INACTIVE = 0, /* sysfs entries NOT registered */ + MCI_SYSFS_ACTIVE /* sysfs entries ARE registered */ +}; + +/* FIXME - should have notify capabilities: NMI, LOG, PROC, etc */ + +/* + * There are several things to be aware of that aren't at all obvious: + * + * + * SOCKETS, SOCKET SETS, BANKS, ROWS, CHIP-SELECT ROWS, CHANNELS, etc.. + * + * These are some of the many terms that are thrown about that don't always + * mean what people think they mean (Inconceivable!). In the interest of + * creating a common ground for discussion, terms and their definitions + * will be established. + * + * Memory devices: The individual chip on a memory stick. These devices + * commonly output 4 and 8 bits each. Grouping several + * of these in parallel provides 64 bits which is common + * for a memory stick. + * + * Memory Stick: A printed circuit board that agregates multiple + * memory devices in parallel. This is the atomic + * memory component that is purchaseable by Joe consumer + * and loaded into a memory socket. + * + * Socket: A physical connector on the motherboard that accepts + * a single memory stick. + * + * Channel: Set of memory devices on a memory stick that must be + * grouped in parallel with one or more additional + * channels from other memory sticks. This parallel + * grouping of the output from multiple channels are + * necessary for the smallest granularity of memory access. + * Some memory controllers are capable of single channel - + * which means that memory sticks can be loaded + * individually. Other memory controllers are only + * capable of dual channel - which means that memory + * sticks must be loaded as pairs (see "socket set"). + * + * Chip-select row: All of the memory devices that are selected together. + * for a single, minimum grain of memory access. + * This selects all of the parallel memory devices across + * all of the parallel channels. Common chip-select rows + * for single channel are 64 bits, for dual channel 128 + * bits. + * + * Single-Ranked stick: A Single-ranked stick has 1 chip-select row of memmory. + * Motherboards commonly drive two chip-select pins to + * a memory stick. A single-ranked stick, will occupy + * only one of those rows. The other will be unused. + * + * Double-Ranked stick: A double-ranked stick has two chip-select rows which + * access different sets of memory devices. The two + * rows cannot be accessed concurrently. + * + * Double-sided stick: DEPRECATED TERM, see Double-Ranked stick. + * A double-sided stick has two chip-select rows which + * access different sets of memory devices. The two + * rows cannot be accessed concurrently. "Double-sided" + * is irrespective of the memory devices being mounted + * on both sides of the memory stick. + * + * Socket set: All of the memory sticks that are required for for + * a single memory access or all of the memory sticks + * spanned by a chip-select row. A single socket set + * has two chip-select rows and if double-sided sticks + * are used these will occupy those chip-select rows. + * + * Bank: This term is avoided because it is unclear when + * needing to distinguish between chip-select rows and + * socket sets. + * + * Controller pages: + * + * Physical pages: + * + * Virtual pages: + * + * + * STRUCTURE ORGANIZATION AND CHOICES + * + * + * + * PS - I enjoyed writing all that about as much as you enjoyed reading it. + */ + + +struct channel_info { + int chan_idx; /* channel index */ + u32 ce_count; /* Correctable Errors for this CHANNEL */ + char label[EDAC_MC_LABEL_LEN + 1]; /* DIMM label on motherboard */ + struct csrow_info *csrow; /* the parent */ +}; + + +struct csrow_info { + unsigned long first_page; /* first page number in dimm */ + unsigned long last_page; /* last page number in dimm */ + unsigned long page_mask; /* used for interleaving - + 0UL for non intlv */ + u32 nr_pages; /* number of pages in csrow */ + u32 grain; /* granularity of reported error in bytes */ + int csrow_idx; /* the chip-select row */ + enum dev_type dtype; /* memory device type */ + u32 ue_count; /* Uncorrectable Errors for this csrow */ + u32 ce_count; /* Correctable Errors for this csrow */ + enum mem_type mtype; /* memory csrow type */ + enum edac_type edac_mode; /* EDAC mode for this csrow */ + struct mem_ctl_info *mci; /* the parent */ + + struct kobject kobj; /* sysfs kobject for this csrow */ + + /* FIXME the number of CHANNELs might need to become dynamic */ + u32 nr_channels; + struct channel_info *channels; +}; + + +struct mem_ctl_info { + struct list_head link; /* for global list of mem_ctl_info structs */ + unsigned long mtype_cap; /* memory types supported by mc */ + unsigned long edac_ctl_cap; /* Mem controller EDAC capabilities */ + unsigned long edac_cap; /* configuration capabilities - this is + closely related to edac_ctl_cap. The + difference is that the controller + may be capable of s4ecd4ed which would + be listed in edac_ctl_cap, but if + channels aren't capable of s4ecd4ed then the + edac_cap would not have that capability. */ + unsigned long scrub_cap; /* chipset scrub capabilities */ + enum scrub_type scrub_mode; /* current scrub mode */ + + enum mci_sysfs_status sysfs_active; /* status of sysfs */ + + /* pointer to edac checking routine */ + void (*edac_check) (struct mem_ctl_info * mci); + /* + * Remaps memory pages: controller pages to physical pages. + * For most MC's, this will be NULL. + */ + /* FIXME - why not send the phys page to begin with? */ + unsigned long (*ctl_page_to_phys) (struct mem_ctl_info * mci, + unsigned long page); + int mc_idx; + int nr_csrows; + struct csrow_info *csrows; + /* + * FIXME - what about controllers on other busses? - IDs must be + * unique. pdev pointer should be sufficiently unique, but + * BUS:SLOT.FUNC numbers may not be unique. + */ + struct pci_dev *pdev; + const char *mod_name; + const char *mod_ver; + const char *ctl_name; + char proc_name[MC_PROC_NAME_MAX_LEN + 1]; + void *pvt_info; + u32 ue_noinfo_count; /* Uncorrectable Errors w/o info */ + u32 ce_noinfo_count; /* Correctable Errors w/o info */ + u32 ue_count; /* Total Uncorrectable Errors for this MC */ + u32 ce_count; /* Total Correctable Errors for this MC */ + unsigned long start_time; /* mci load start time (in jiffies) */ + + /* this stuff is for safe removal of mc devices from global list while + * NMI handlers may be traversing list + */ + struct rcu_head rcu; + struct completion complete; + + /* edac sysfs device control */ + struct kobject edac_mci_kobj; +}; + + + +/* write all or some bits in a byte-register*/ +static inline void pci_write_bits8(struct pci_dev *pdev, int offset, + u8 value, u8 mask) +{ + if (mask != 0xff) { + u8 buf; + pci_read_config_byte(pdev, offset, &buf); + value &= mask; + buf &= ~mask; + value |= buf; + } + pci_write_config_byte(pdev, offset, value); +} + + +/* write all or some bits in a word-register*/ +static inline void pci_write_bits16(struct pci_dev *pdev, int offset, + u16 value, u16 mask) +{ + if (mask != 0xffff) { + u16 buf; + pci_read_config_word(pdev, offset, &buf); + value &= mask; + buf &= ~mask; + value |= buf; + } + pci_write_config_word(pdev, offset, value); +} + + +/* write all or some bits in a dword-register*/ +static inline void pci_write_bits32(struct pci_dev *pdev, int offset, + u32 value, u32 mask) +{ + if (mask != 0xffff) { + u32 buf; + pci_read_config_dword(pdev, offset, &buf); + value &= mask; + buf &= ~mask; + value |= buf; + } + pci_write_config_dword(pdev, offset, value); +} + + +#ifdef CONFIG_EDAC_DEBUG +void edac_mc_dump_channel(struct channel_info *chan); +void edac_mc_dump_mci(struct mem_ctl_info *mci); +void edac_mc_dump_csrow(struct csrow_info *csrow); +#endif /* CONFIG_EDAC_DEBUG */ + +extern int edac_mc_add_mc(struct mem_ctl_info *mci); +extern int edac_mc_del_mc(struct mem_ctl_info *mci); + +extern int edac_mc_find_csrow_by_page(struct mem_ctl_info *mci, + unsigned long page); + +extern struct mem_ctl_info *edac_mc_find_mci_by_pdev(struct pci_dev + *pdev); + +extern void edac_mc_scrub_block(unsigned long page, + unsigned long offset, u32 size); + +/* + * The no info errors are used when error overflows are reported. + * There are a limited number of error logging registers that can + * be exausted. When all registers are exhausted and an additional + * error occurs then an error overflow register records that an + * error occured and the type of error, but doesn't have any + * further information. The ce/ue versions make for cleaner + * reporting logic and function interface - reduces conditional + * statement clutter and extra function arguments. + */ +extern void edac_mc_handle_ce(struct mem_ctl_info *mci, + unsigned long page_frame_number, + unsigned long offset_in_page, + unsigned long syndrome, + int row, int channel, const char *msg); + +extern void edac_mc_handle_ce_no_info(struct mem_ctl_info *mci, + const char *msg); + +extern void edac_mc_handle_ue(struct mem_ctl_info *mci, + unsigned long page_frame_number, + unsigned long offset_in_page, + int row, const char *msg); + +extern void edac_mc_handle_ue_no_info(struct mem_ctl_info *mci, + const char *msg); + +/* + * This kmalloc's and initializes all the structures. + * Can't be used if all structures don't have the same lifetime. + */ +extern struct mem_ctl_info *edac_mc_alloc(unsigned sz_pvt, + unsigned nr_csrows, unsigned nr_chans); + +/* Free an mc previously allocated by edac_mc_alloc() */ +extern void edac_mc_free(struct mem_ctl_info *mci); + + +#endif /* _EDAC_MC_H_ */ diff --git a/drivers/edac/i82860_edac.c b/drivers/edac/i82860_edac.c index bfb7ae02e37..52596e75f9c 100644 --- a/drivers/edac/i82860_edac.c +++ b/drivers/edac/i82860_edac.c @@ -253,7 +253,7 @@ static struct pci_driver i82860_driver = { .id_table = i82860_pci_tbl, }; -int __init i82860_init(void) +static int __init i82860_init(void) { int pci_rc; diff --git a/drivers/edac/i82875p_edac.c b/drivers/edac/i82875p_edac.c index 79d14dfbcbd..009c08fe5d6 100644 --- a/drivers/edac/i82875p_edac.c +++ b/drivers/edac/i82875p_edac.c @@ -483,7 +483,7 @@ static struct pci_driver i82875p_driver = { }; -int __init i82875p_init(void) +static int __init i82875p_init(void) { int pci_rc; diff --git a/drivers/edac/r82600_edac.c b/drivers/edac/r82600_edac.c index b6399a54256..e90892831b9 100644 --- a/drivers/edac/r82600_edac.c +++ b/drivers/edac/r82600_edac.c @@ -381,7 +381,7 @@ static struct pci_driver r82600_driver = { }; -int __init r82600_init(void) +static int __init r82600_init(void) { return pci_register_driver(&r82600_driver); } diff --git a/include/asm-i386/atomic.h b/include/asm-i386/atomic.h index e2c00c95a5e..de649d3aa2d 100644 --- a/include/asm-i386/atomic.h +++ b/include/asm-i386/atomic.h @@ -255,17 +255,5 @@ __asm__ __volatile__(LOCK "orl %0,%1" \ #define smp_mb__before_atomic_inc() barrier() #define smp_mb__after_atomic_inc() barrier() -/* ECC atomic, DMA, SMP and interrupt safe scrub function */ - -static __inline__ void atomic_scrub(unsigned long *virt_addr, u32 size) -{ - u32 i; - for (i = 0; i < size / 4; i++, virt_addr++) - /* Very carefully read and write to memory atomically - * so we are interrupt, DMA and SMP safe. - */ - __asm__ __volatile__("lock; addl $0, %0"::"m"(*virt_addr)); -} - #include #endif diff --git a/include/asm-i386/edac.h b/include/asm-i386/edac.h new file mode 100644 index 00000000000..3e7dd0ab68c --- /dev/null +++ b/include/asm-i386/edac.h @@ -0,0 +1,18 @@ +#ifndef ASM_EDAC_H +#define ASM_EDAC_H + +/* ECC atomic, DMA, SMP and interrupt safe scrub function */ + +static __inline__ void atomic_scrub(void *va, u32 size) +{ + unsigned long *virt_addr = va; + u32 i; + + for (i = 0; i < size / 4; i++, virt_addr++) + /* Very carefully read and write to memory atomically + * so we are interrupt, DMA and SMP safe. + */ + __asm__ __volatile__("lock; addl $0, %0"::"m"(*virt_addr)); +} + +#endif diff --git a/include/asm-x86_64/atomic.h b/include/asm-x86_64/atomic.h index 4048508c4f4..4b5cd553e77 100644 --- a/include/asm-x86_64/atomic.h +++ b/include/asm-x86_64/atomic.h @@ -426,17 +426,5 @@ __asm__ __volatile__(LOCK "orl %0,%1" \ #define smp_mb__before_atomic_inc() barrier() #define smp_mb__after_atomic_inc() barrier() -/* ECC atomic, DMA, SMP and interrupt safe scrub function */ - -static __inline__ void atomic_scrub(u32 *virt_addr, u32 size) -{ - u32 i; - for (i = 0; i < size / 4; i++, virt_addr++) - /* Very carefully read and write to memory atomically - * so we are interrupt, DMA and SMP safe. - */ - __asm__ __volatile__("lock; addl $0, %0"::"m"(*virt_addr)); -} - #include #endif diff --git a/include/asm-x86_64/edac.h b/include/asm-x86_64/edac.h new file mode 100644 index 00000000000..cad1cd42b4e --- /dev/null +++ b/include/asm-x86_64/edac.h @@ -0,0 +1,18 @@ +#ifndef ASM_EDAC_H +#define ASM_EDAC_H + +/* ECC atomic, DMA, SMP and interrupt safe scrub function */ + +static __inline__ void atomic_scrub(void *va, u32 size) +{ + unsigned int *virt_addr = va; + u32 i; + + for (i = 0; i < size / 4; i++, virt_addr++) + /* Very carefully read and write to memory atomically + * so we are interrupt, DMA and SMP safe. + */ + __asm__ __volatile__("lock; addl $0, %0"::"m"(*virt_addr)); +} + +#endif -- cgit v1.2.3 From 648bf4fb21f5ce993403901ed82399abaa150b63 Mon Sep 17 00:00:00 2001 From: mark gross Date: Sun, 15 Jan 2006 17:37:30 -0800 Subject: [PATCH] tlclk driver update some driver clean ups, and a re-posting of changes that are needed to match the updated TPS. Signed-off-by: Mark Gross Signed-off-by: Linus Torvalds --- drivers/char/tlclk.c | 93 ++++++++++++++++++++++++++++++++-------------------- 1 file changed, 58 insertions(+), 35 deletions(-) diff --git a/drivers/char/tlclk.c b/drivers/char/tlclk.c index bc56df8a347..4c272189cd4 100644 --- a/drivers/char/tlclk.c +++ b/drivers/char/tlclk.c @@ -34,7 +34,6 @@ #include /* printk() */ #include /* everything... */ #include /* error codes */ -#include /* udelay */ #include #include #include @@ -156,6 +155,8 @@ This directory exports the following interfaces. There operation is documented in the MCPBL0010 TPS under the Telecom Clock API section, 11.4. alarms : current_ref : +received_ref_clk3a : +received_ref_clk3b : enable_clk3a_output : enable_clk3b_output : enable_clka0_output : @@ -165,7 +166,7 @@ enable_clkb1_output : filter_select : hardware_switching : hardware_switching_mode : -interrupt_switch : +telclock_version : mode_select : refalign : reset : @@ -173,7 +174,6 @@ select_amcb1_transmit_clock : select_amcb2_transmit_clock : select_redundant_clock : select_ref_frequency : -test_mode : All sysfs interfaces are integers in hex format, i.e echo 99 > refalign has the same effect as echo 0x99 > refalign. @@ -226,7 +226,7 @@ static int tlclk_release(struct inode *inode, struct file *filp) return 0; } -ssize_t tlclk_read(struct file *filp, char __user *buf, size_t count, +static ssize_t tlclk_read(struct file *filp, char __user *buf, size_t count, loff_t *f_pos) { if (count < sizeof(struct tlclk_alarms)) @@ -242,7 +242,7 @@ ssize_t tlclk_read(struct file *filp, char __user *buf, size_t count, return sizeof(struct tlclk_alarms); } -ssize_t tlclk_write(struct file *filp, const char __user *buf, size_t count, +static ssize_t tlclk_write(struct file *filp, const char __user *buf, size_t count, loff_t *f_pos) { return 0; @@ -278,21 +278,21 @@ static ssize_t show_current_ref(struct device *d, static DEVICE_ATTR(current_ref, S_IRUGO, show_current_ref, NULL); -static ssize_t show_interrupt_switch(struct device *d, +static ssize_t show_telclock_version(struct device *d, struct device_attribute *attr, char *buf) { unsigned long ret_val; unsigned long flags; spin_lock_irqsave(&event_lock, flags); - ret_val = inb(TLCLK_REG6); + ret_val = inb(TLCLK_REG5); spin_unlock_irqrestore(&event_lock, flags); return sprintf(buf, "0x%lX\n", ret_val); } -static DEVICE_ATTR(interrupt_switch, S_IRUGO, - show_interrupt_switch, NULL); +static DEVICE_ATTR(telclock_version, S_IRUGO, + show_telclock_version, NULL); static ssize_t show_alarms(struct device *d, struct device_attribute *attr, char *buf) @@ -309,6 +309,50 @@ static ssize_t show_alarms(struct device *d, static DEVICE_ATTR(alarms, S_IRUGO, show_alarms, NULL); +static ssize_t store_received_ref_clk3a(struct device *d, + struct device_attribute *attr, const char *buf, size_t count) +{ + unsigned long tmp; + unsigned char val; + unsigned long flags; + + sscanf(buf, "%lX", &tmp); + dev_dbg(d, ": tmp = 0x%lX\n", tmp); + + val = (unsigned char)tmp; + spin_lock_irqsave(&event_lock, flags); + SET_PORT_BITS(TLCLK_REG1, 0xef, val); + spin_unlock_irqrestore(&event_lock, flags); + + return strnlen(buf, count); +} + +static DEVICE_ATTR(received_ref_clk3a, S_IWUGO, NULL, + store_received_ref_clk3a); + + +static ssize_t store_received_ref_clk3b(struct device *d, + struct device_attribute *attr, const char *buf, size_t count) +{ + unsigned long tmp; + unsigned char val; + unsigned long flags; + + sscanf(buf, "%lX", &tmp); + dev_dbg(d, ": tmp = 0x%lX\n", tmp); + + val = (unsigned char)tmp; + spin_lock_irqsave(&event_lock, flags); + SET_PORT_BITS(TLCLK_REG1, 0xef, val << 1); + spin_unlock_irqrestore(&event_lock, flags); + + return strnlen(buf, count); +} + +static DEVICE_ATTR(received_ref_clk3b, S_IWUGO, NULL, + store_received_ref_clk3b); + + static ssize_t store_enable_clk3b_output(struct device *d, struct device_attribute *attr, const char *buf, size_t count) { @@ -436,26 +480,6 @@ static ssize_t store_enable_clka0_output(struct device *d, static DEVICE_ATTR(enable_clka0_output, S_IWUGO, NULL, store_enable_clka0_output); -static ssize_t store_test_mode(struct device *d, - struct device_attribute *attr, const char *buf, size_t count) -{ - unsigned long flags; - unsigned long tmp; - unsigned char val; - - sscanf(buf, "%lX", &tmp); - dev_dbg(d, "tmp = 0x%lX\n", tmp); - - val = (unsigned char)tmp; - spin_lock_irqsave(&event_lock, flags); - SET_PORT_BITS(TLCLK_REG4, 0xfd, 2); - spin_unlock_irqrestore(&event_lock, flags); - - return strnlen(buf, count); -} - -static DEVICE_ATTR(test_mode, S_IWUGO, NULL, store_test_mode); - static ssize_t store_select_amcb2_transmit_clock(struct device *d, struct device_attribute *attr, const char *buf, size_t count) { @@ -475,7 +499,7 @@ static ssize_t store_select_amcb2_transmit_clock(struct device *d, SET_PORT_BITS(TLCLK_REG3, 0xc7, 0x38); switch (val) { case CLK_8_592MHz: - SET_PORT_BITS(TLCLK_REG0, 0xfc, 1); + SET_PORT_BITS(TLCLK_REG0, 0xfc, 2); break; case CLK_11_184MHz: SET_PORT_BITS(TLCLK_REG0, 0xfc, 0); @@ -484,7 +508,7 @@ static ssize_t store_select_amcb2_transmit_clock(struct device *d, SET_PORT_BITS(TLCLK_REG0, 0xfc, 3); break; case CLK_44_736MHz: - SET_PORT_BITS(TLCLK_REG0, 0xfc, 2); + SET_PORT_BITS(TLCLK_REG0, 0xfc, 1); break; } } else @@ -653,9 +677,7 @@ static ssize_t store_refalign (struct device *d, dev_dbg(d, "tmp = 0x%lX\n", tmp); spin_lock_irqsave(&event_lock, flags); SET_PORT_BITS(TLCLK_REG0, 0xf7, 0); - udelay(2); SET_PORT_BITS(TLCLK_REG0, 0xf7, 0x08); - udelay(2); SET_PORT_BITS(TLCLK_REG0, 0xf7, 0); spin_unlock_irqrestore(&event_lock, flags); @@ -706,15 +728,16 @@ static DEVICE_ATTR(reset, S_IWUGO, NULL, store_reset); static struct attribute *tlclk_sysfs_entries[] = { &dev_attr_current_ref.attr, - &dev_attr_interrupt_switch.attr, + &dev_attr_telclock_version.attr, &dev_attr_alarms.attr, + &dev_attr_received_ref_clk3a.attr, + &dev_attr_received_ref_clk3b.attr, &dev_attr_enable_clk3a_output.attr, &dev_attr_enable_clk3b_output.attr, &dev_attr_enable_clkb1_output.attr, &dev_attr_enable_clka1_output.attr, &dev_attr_enable_clkb0_output.attr, &dev_attr_enable_clka0_output.attr, - &dev_attr_test_mode.attr, &dev_attr_select_amcb1_transmit_clock.attr, &dev_attr_select_amcb2_transmit_clock.attr, &dev_attr_select_redundant_clock.attr, -- cgit v1.2.3 From f7111ceb5266750db2a1d193b98fb6a3d9b5a56a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 18 Jan 2006 21:57:37 -0800 Subject: [SPARC]: sparc32 needs PROMDEV_{I,O}RSC defines too. Signed-off-by: David S. Miller --- include/asm-sparc/oplib.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/asm-sparc/oplib.h b/include/asm-sparc/oplib.h index 95944556d8b..d0d76b30eb4 100644 --- a/include/asm-sparc/oplib.h +++ b/include/asm-sparc/oplib.h @@ -164,6 +164,7 @@ enum prom_input_device { PROMDEV_IKBD, /* input from keyboard */ PROMDEV_ITTYA, /* input from ttya */ PROMDEV_ITTYB, /* input from ttyb */ + PROMDEV_IRSC, /* input from rsc */ PROMDEV_I_UNK, }; @@ -175,6 +176,7 @@ enum prom_output_device { PROMDEV_OSCREEN, /* to screen */ PROMDEV_OTTYA, /* to ttya */ PROMDEV_OTTYB, /* to ttyb */ + PROMDEV_ORSC, /* to rsc */ PROMDEV_O_UNK, }; -- cgit v1.2.3 From 2d7d5f05111a9d913131a2764d8b20157f8f758d Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 19 Jan 2006 02:42:49 -0800 Subject: [SPARC]: Add support for *at(), ppoll, and pselect syscalls. This also includes by necessity _TIF_RESTORE_SIGMASK support, which actually resulted in a lot of cleanups. The sparc signal handling code is quite a mess and I should clean it up some day. Signed-off-by: David S. Miller --- arch/sparc/kernel/entry.S | 56 ------------- arch/sparc/kernel/rtrap.S | 9 +-- arch/sparc/kernel/signal.c | 117 ++++++++-------------------- arch/sparc/kernel/sparc_ksyms.c | 2 - arch/sparc/kernel/systbls.S | 10 ++- arch/sparc64/kernel/entry.S | 23 ------ arch/sparc64/kernel/rtrap.S | 33 ++++---- arch/sparc64/kernel/signal.c | 151 ++++++++++-------------------------- arch/sparc64/kernel/signal32.c | 122 +++++------------------------ arch/sparc64/kernel/sparc64_ksyms.c | 4 +- arch/sparc64/kernel/systbls.S | 21 ++++- arch/sparc64/solaris/entry64.S | 2 + include/asm-sparc/thread_info.h | 5 +- include/asm-sparc/unistd.h | 22 +++++- include/asm-sparc64/thread_info.h | 6 +- include/asm-sparc64/unistd.h | 23 +++++- 16 files changed, 184 insertions(+), 422 deletions(-) diff --git a/arch/sparc/kernel/entry.S b/arch/sparc/kernel/entry.S index 03ecb4e4614..c51d08d218e 100644 --- a/arch/sparc/kernel/entry.S +++ b/arch/sparc/kernel/entry.S @@ -1276,62 +1276,6 @@ sys_sigstack: call do_sys_sigstack mov %l5, %o7 - .align 4 - .globl sys_sigpause -sys_sigpause: - /* Note: %o0 already has correct value... */ - call do_sigpause - add %sp, STACKFRAME_SZ, %o1 - - ld [%curptr + TI_FLAGS], %l5 - andcc %l5, _TIF_SYSCALL_TRACE, %g0 - be 1f - nop - - call syscall_trace - nop - -1: - /* We are returning to a signal handler. */ - RESTORE_ALL - - .align 4 - .globl sys_sigsuspend -sys_sigsuspend: - call do_sigsuspend - add %sp, STACKFRAME_SZ, %o0 - - ld [%curptr + TI_FLAGS], %l5 - andcc %l5, _TIF_SYSCALL_TRACE, %g0 - be 1f - nop - - call syscall_trace - nop - -1: - /* We are returning to a signal handler. */ - RESTORE_ALL - - .align 4 - .globl sys_rt_sigsuspend -sys_rt_sigsuspend: - /* Note: %o0, %o1 already have correct value... */ - call do_rt_sigsuspend - add %sp, STACKFRAME_SZ, %o2 - - ld [%curptr + TI_FLAGS], %l5 - andcc %l5, _TIF_SYSCALL_TRACE, %g0 - be 1f - nop - - call syscall_trace - nop - -1: - /* We are returning to a signal handler. */ - RESTORE_ALL - .align 4 .globl sys_sigreturn sys_sigreturn: diff --git a/arch/sparc/kernel/rtrap.S b/arch/sparc/kernel/rtrap.S index f7460d897e7..77ca6fd8125 100644 --- a/arch/sparc/kernel/rtrap.S +++ b/arch/sparc/kernel/rtrap.S @@ -68,15 +68,14 @@ ret_trap_lockless_ipi: ld [%curptr + TI_FLAGS], %g2 signal_p: - andcc %g2, (_TIF_NOTIFY_RESUME|_TIF_SIGPENDING), %g0 + andcc %g2, (_TIF_SIGPENDING|_TIF_RESTORE_SIGMASK), %g0 bz,a ret_trap_continue ld [%sp + STACKFRAME_SZ + PT_PSR], %t_psr - clr %o0 - mov %l5, %o2 - mov %l6, %o3 + mov %l5, %o1 + mov %l6, %o2 call do_signal - add %sp, STACKFRAME_SZ, %o1 ! pt_regs ptr + add %sp, STACKFRAME_SZ, %o0 ! pt_regs ptr /* Fall through. */ ld [%sp + STACKFRAME_SZ + PT_PSR], %t_psr diff --git a/arch/sparc/kernel/signal.c b/arch/sparc/kernel/signal.c index 5f34d7dc2b8..0748d8147bb 100644 --- a/arch/sparc/kernel/signal.c +++ b/arch/sparc/kernel/signal.c @@ -35,9 +35,6 @@ extern void fpsave(unsigned long *fpregs, unsigned long *fsr, void *fpqueue, unsigned long *fpqdepth); extern void fpload(unsigned long *fpregs, unsigned long *fsr); -asmlinkage int do_signal(sigset_t *oldset, struct pt_regs * regs, - unsigned long orig_o0, int restart_syscall); - /* Signal frames: the original one (compatible with SunOS): * * Set up a signal frame... Make the stack look the way SunOS @@ -95,98 +92,30 @@ struct rt_signal_frame { #define NF_ALIGNEDSZ (((sizeof(struct new_signal_frame) + 7) & (~7))) #define RT_ALIGNEDSZ (((sizeof(struct rt_signal_frame) + 7) & (~7))) -/* - * atomically swap in the new signal mask, and wait for a signal. - * This is really tricky on the Sparc, watch out... - */ -asmlinkage void _sigpause_common(old_sigset_t set, struct pt_regs *regs) +static int _sigpause_common(old_sigset_t set) { - sigset_t saveset; - set &= _BLOCKABLE; spin_lock_irq(¤t->sighand->siglock); - saveset = current->blocked; + current->saved_sigmask = current->blocked; siginitset(¤t->blocked, set); recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); - regs->pc = regs->npc; - regs->npc += 4; - - /* Condition codes and return value where set here for sigpause, - * and so got used by setup_frame, which again causes sigreturn() - * to return -EINTR. - */ - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - /* - * Return -EINTR and set condition code here, - * so the interrupted system call actually returns - * these. - */ - regs->psr |= PSR_C; - regs->u_regs[UREG_I0] = EINTR; - if (do_signal(&saveset, regs, 0, 0)) - return; - } -} + current->state = TASK_INTERRUPTIBLE; + schedule(); + set_thread_flag(TIF_RESTORE_SIGMASK); -asmlinkage void do_sigpause(unsigned int set, struct pt_regs *regs) -{ - _sigpause_common(set, regs); + return -ERESTARTNOHAND; } -asmlinkage void do_sigsuspend (struct pt_regs *regs) +asmlinkage int sys_sigpause(unsigned int set) { - _sigpause_common(regs->u_regs[UREG_I0], regs); + return _sigpause_common(set); } -asmlinkage void do_rt_sigsuspend(sigset_t __user *uset, size_t sigsetsize, - struct pt_regs *regs) +asmlinkage int sys_sigsuspend(old_sigset_t set) { - sigset_t oldset, set; - - /* XXX: Don't preclude handling different sized sigset_t's. */ - if (sigsetsize != sizeof(sigset_t)) { - regs->psr |= PSR_C; - regs->u_regs[UREG_I0] = EINVAL; - return; - } - - if (copy_from_user(&set, uset, sizeof(set))) { - regs->psr |= PSR_C; - regs->u_regs[UREG_I0] = EFAULT; - return; - } - - sigdelsetmask(&set, ~_BLOCKABLE); - spin_lock_irq(¤t->sighand->siglock); - oldset = current->blocked; - current->blocked = set; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - regs->pc = regs->npc; - regs->npc += 4; - - /* Condition codes and return value where set here for sigpause, - * and so got used by setup_frame, which again causes sigreturn() - * to return -EINTR. - */ - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - /* - * Return -EINTR and set condition code here, - * so the interrupted system call actually returns - * these. - */ - regs->psr |= PSR_C; - regs->u_regs[UREG_I0] = EINTR; - if (do_signal(&oldset, regs, 0, 0)) - return; - } + return _sigpause_common(set); } static inline int @@ -1067,13 +996,13 @@ static inline void syscall_restart(unsigned long orig_i0, struct pt_regs *regs, * want to handle. Thus you cannot kill init even with a SIGKILL even by * mistake. */ -asmlinkage int do_signal(sigset_t *oldset, struct pt_regs * regs, - unsigned long orig_i0, int restart_syscall) +asmlinkage void do_signal(struct pt_regs * regs, unsigned long orig_i0, int restart_syscall) { siginfo_t info; struct sparc_deliver_cookie cookie; struct k_sigaction ka; int signr; + sigset_t *oldset; /* * XXX Disable svr4 signal handling until solaris emulation works. @@ -1089,7 +1018,9 @@ asmlinkage int do_signal(sigset_t *oldset, struct pt_regs * regs, cookie.restart_syscall = restart_syscall; cookie.orig_i0 = orig_i0; - if (!oldset) + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + oldset = ¤t->saved_sigmask; + else oldset = ¤t->blocked; signr = get_signal_to_deliver(&info, &ka, regs, &cookie); @@ -1098,7 +1029,14 @@ asmlinkage int do_signal(sigset_t *oldset, struct pt_regs * regs, syscall_restart(cookie.orig_i0, regs, &ka.sa); handle_signal(signr, &ka, &info, oldset, regs, svr4_signal); - return 1; + /* a signal was successfully delivered; the saved + * sigmask will have been stored in the signal frame, + * and will be restored by sigreturn, so we can simply + * clear the TIF_RESTORE_SIGMASK flag. + */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + clear_thread_flag(TIF_RESTORE_SIGMASK); + return; } if (cookie.restart_syscall && (regs->u_regs[UREG_I0] == ERESTARTNOHAND || @@ -1115,7 +1053,14 @@ asmlinkage int do_signal(sigset_t *oldset, struct pt_regs * regs, regs->pc -= 4; regs->npc -= 4; } - return 0; + + /* if there's no signal to deliver, we just put the saved sigmask + * back + */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) { + clear_thread_flag(TIF_RESTORE_SIGMASK); + sigprocmask(SIG_SETMASK, ¤t->saved_sigmask, NULL); + } } asmlinkage int diff --git a/arch/sparc/kernel/sparc_ksyms.c b/arch/sparc/kernel/sparc_ksyms.c index 0b0d492c953..19b25399d7e 100644 --- a/arch/sparc/kernel/sparc_ksyms.c +++ b/arch/sparc/kernel/sparc_ksyms.c @@ -66,7 +66,6 @@ struct poll { extern int svr4_getcontext (svr4_ucontext_t *, struct pt_regs *); extern int svr4_setcontext (svr4_ucontext_t *, struct pt_regs *); -void _sigpause_common (unsigned int set, struct pt_regs *); extern void (*__copy_1page)(void *, const void *); extern void __memmove(void *, const void *, __kernel_size_t); extern void (*bzero_1page)(void *); @@ -227,7 +226,6 @@ EXPORT_SYMBOL(kunmap_atomic); /* Solaris/SunOS binary compatibility */ EXPORT_SYMBOL(svr4_setcontext); EXPORT_SYMBOL(svr4_getcontext); -EXPORT_SYMBOL(_sigpause_common); EXPORT_SYMBOL(dump_thread); diff --git a/arch/sparc/kernel/systbls.S b/arch/sparc/kernel/systbls.S index e457a40838f..6877ae4cd1d 100644 --- a/arch/sparc/kernel/systbls.S +++ b/arch/sparc/kernel/systbls.S @@ -75,7 +75,10 @@ sys_call_table: /*265*/ .long sys_timer_delete, sys_timer_create, sys_nis_syscall, sys_io_setup, sys_io_destroy /*270*/ .long sys_io_submit, sys_io_cancel, sys_io_getevents, sys_mq_open, sys_mq_unlink /*275*/ .long sys_mq_timedsend, sys_mq_timedreceive, sys_mq_notify, sys_mq_getsetattr, sys_waitid -/*280*/ .long sys_ni_syscall, sys_add_key, sys_request_key, sys_keyctl +/*280*/ .long sys_ni_syscall, sys_add_key, sys_request_key, sys_keyctl, sys_openat +/*285*/ .long sys_mkdirat, sys_mknodat, sys_fchownat, sys_futimesat, sys_newfstatat +/*290*/ .long sys_unlinkat, sys_renameat, sys_linkat, sys_symlinkat, sys_readlinkat +/*295*/ .long sys_fchmodat, sys_faccessat, sys_pselect6, sys_ppoll #ifdef CONFIG_SUNOS_EMUL /* Now the SunOS syscall table. */ @@ -181,6 +184,11 @@ sunos_sys_table: .long sunos_nosys, sunos_nosys, sunos_nosys .long sunos_nosys /*280*/ .long sunos_nosys, sunos_nosys, sunos_nosys + .long sunos_nosys, sunos_nosys, sunos_nosys + .long sunos_nosys, sunos_nosys, sunos_nosys .long sunos_nosys +/*290*/ .long sunos_nosys, sunos_nosys, sunos_nosys + .long sunos_nosys, sunos_nosys, sunos_nosys + .long sunos_nosys, sunos_nosys, sunos_nosys #endif diff --git a/arch/sparc64/kernel/entry.S b/arch/sparc64/kernel/entry.S index 71000299188..e50e56e4ab6 100644 --- a/arch/sparc64/kernel/entry.S +++ b/arch/sparc64/kernel/entry.S @@ -1416,7 +1416,6 @@ execve_merge: add %sp, PTREGS_OFF, %o0 .globl sys_pipe, sys_sigpause, sys_nis_syscall - .globl sys_sigsuspend, sys_rt_sigsuspend .globl sys_rt_sigreturn .globl sys_ptrace .globl sys_sigaltstack @@ -1440,28 +1439,6 @@ sys32_sigaltstack: mov %i6, %o2 #endif .align 32 -sys_sigsuspend: add %sp, PTREGS_OFF, %o0 - call do_sigsuspend - add %o7, 1f-.-4, %o7 - nop -sys_rt_sigsuspend: /* NOTE: %o0,%o1 have a correct value already */ - add %sp, PTREGS_OFF, %o2 - call do_rt_sigsuspend - add %o7, 1f-.-4, %o7 - nop -#ifdef CONFIG_COMPAT - .globl sys32_rt_sigsuspend -sys32_rt_sigsuspend: /* NOTE: %o0,%o1 have a correct value already */ - srl %o0, 0, %o0 - add %sp, PTREGS_OFF, %o2 - call do_rt_sigsuspend32 - add %o7, 1f-.-4, %o7 -#endif - /* NOTE: %o0 has a correct value already */ -sys_sigpause: add %sp, PTREGS_OFF, %o1 - call do_sigpause - add %o7, 1f-.-4, %o7 - nop #ifdef CONFIG_COMPAT .globl sys32_sigreturn sys32_sigreturn: diff --git a/arch/sparc64/kernel/rtrap.S b/arch/sparc64/kernel/rtrap.S index 090dcca00d2..b80eba0081c 100644 --- a/arch/sparc64/kernel/rtrap.S +++ b/arch/sparc64/kernel/rtrap.S @@ -53,14 +53,13 @@ __handle_user_windows: wrpr %g0, RTRAP_PSTATE_IRQOFF, %pstate ldx [%g6 + TI_FLAGS], %l0 -1: andcc %l0, (_TIF_NOTIFY_RESUME | _TIF_SIGPENDING), %g0 +1: andcc %l0, (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK), %g0 be,pt %xcc, __handle_user_windows_continue nop - clr %o0 - mov %l5, %o2 - mov %l6, %o3 - add %sp, PTREGS_OFF, %o1 - mov %l0, %o4 + mov %l5, %o1 + mov %l6, %o2 + add %sp, PTREGS_OFF, %o0 + mov %l0, %o3 call do_notify_resume wrpr %g0, RTRAP_PSTATE, %pstate @@ -96,15 +95,14 @@ __handle_perfctrs: wrpr %g0, RTRAP_PSTATE, %pstate wrpr %g0, RTRAP_PSTATE_IRQOFF, %pstate ldx [%g6 + TI_FLAGS], %l0 -1: andcc %l0, (_TIF_NOTIFY_RESUME | _TIF_SIGPENDING), %g0 +1: andcc %l0, (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK), %g0 be,pt %xcc, __handle_perfctrs_continue sethi %hi(TSTATE_PEF), %o0 - clr %o0 - mov %l5, %o2 - mov %l6, %o3 - add %sp, PTREGS_OFF, %o1 - mov %l0, %o4 + mov %l5, %o1 + mov %l6, %o2 + add %sp, PTREGS_OFF, %o0 + mov %l0, %o3 call do_notify_resume wrpr %g0, RTRAP_PSTATE, %pstate @@ -129,11 +127,10 @@ __handle_userfpu: ba,a,pt %xcc, __handle_userfpu_continue __handle_signal: - clr %o0 - mov %l5, %o2 - mov %l6, %o3 - add %sp, PTREGS_OFF, %o1 - mov %l0, %o4 + mov %l5, %o1 + mov %l6, %o2 + add %sp, PTREGS_OFF, %o0 + mov %l0, %o3 call do_notify_resume wrpr %g0, RTRAP_PSTATE, %pstate wrpr %g0, RTRAP_PSTATE_IRQOFF, %pstate @@ -200,7 +197,7 @@ __handle_preemption_continue: andcc %l1, %o0, %g0 andcc %l0, _TIF_NEED_RESCHED, %g0 bne,pn %xcc, __handle_preemption - andcc %l0, (_TIF_NOTIFY_RESUME | _TIF_SIGPENDING), %g0 + andcc %l0, (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK), %g0 bne,pn %xcc, __handle_signal __handle_signal_continue: ldub [%g6 + TI_WSAVED], %o2 diff --git a/arch/sparc64/kernel/signal.c b/arch/sparc64/kernel/signal.c index 60f5dfabb1e..ca11a4c457d 100644 --- a/arch/sparc64/kernel/signal.c +++ b/arch/sparc64/kernel/signal.c @@ -36,9 +36,6 @@ #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP))) -static int do_signal(sigset_t *oldset, struct pt_regs * regs, - unsigned long orig_o0, int ret_from_syscall); - /* {set, get}context() needed for 64-bit SparcLinux userland. */ asmlinkage void sparc64_set_context(struct pt_regs *regs) { @@ -242,114 +239,29 @@ struct rt_signal_frame { /* Align macros */ #define RT_ALIGNEDSZ (((sizeof(struct rt_signal_frame) + 7) & (~7))) -/* - * atomically swap in the new signal mask, and wait for a signal. - * This is really tricky on the Sparc, watch out... - */ -asmlinkage void _sigpause_common(old_sigset_t set, struct pt_regs *regs) +static long _sigpause_common(old_sigset_t set) { - sigset_t saveset; - -#ifdef CONFIG_SPARC32_COMPAT - if (test_thread_flag(TIF_32BIT)) { - extern asmlinkage void _sigpause32_common(compat_old_sigset_t, - struct pt_regs *); - _sigpause32_common(set, regs); - return; - } -#endif set &= _BLOCKABLE; spin_lock_irq(¤t->sighand->siglock); - saveset = current->blocked; + current->saved_sigmask = current->blocked; siginitset(¤t->blocked, set); recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); - - if (test_thread_flag(TIF_32BIT)) { - regs->tpc = (regs->tnpc & 0xffffffff); - regs->tnpc = (regs->tnpc + 4) & 0xffffffff; - } else { - regs->tpc = regs->tnpc; - regs->tnpc += 4; - } - /* Condition codes and return value where set here for sigpause, - * and so got used by setup_frame, which again causes sigreturn() - * to return -EINTR. - */ - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - /* - * Return -EINTR and set condition code here, - * so the interrupted system call actually returns - * these. - */ - regs->tstate |= (TSTATE_ICARRY|TSTATE_XCARRY); - regs->u_regs[UREG_I0] = EINTR; - if (do_signal(&saveset, regs, 0, 0)) - return; - } + current->state = TASK_INTERRUPTIBLE; + schedule(); + set_thread_flag(TIF_RESTORE_SIGMASK); + return -ERESTARTNOHAND; } -asmlinkage void do_sigpause(unsigned int set, struct pt_regs *regs) +asmlinkage long sys_sigpause(unsigned int set) { - _sigpause_common(set, regs); + return _sigpause_common(set); } -asmlinkage void do_sigsuspend(struct pt_regs *regs) +asmlinkage long sys_sigsuspend(old_sigset_t set) { - _sigpause_common(regs->u_regs[UREG_I0], regs); -} - -asmlinkage void do_rt_sigsuspend(sigset_t __user *uset, size_t sigsetsize, struct pt_regs *regs) -{ - sigset_t oldset, set; - - /* XXX: Don't preclude handling different sized sigset_t's. */ - if (sigsetsize != sizeof(sigset_t)) { - regs->tstate |= (TSTATE_ICARRY|TSTATE_XCARRY); - regs->u_regs[UREG_I0] = EINVAL; - return; - } - if (copy_from_user(&set, uset, sizeof(set))) { - regs->tstate |= (TSTATE_ICARRY|TSTATE_XCARRY); - regs->u_regs[UREG_I0] = EFAULT; - return; - } - - sigdelsetmask(&set, ~_BLOCKABLE); - spin_lock_irq(¤t->sighand->siglock); - oldset = current->blocked; - current->blocked = set; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - if (test_thread_flag(TIF_32BIT)) { - regs->tpc = (regs->tnpc & 0xffffffff); - regs->tnpc = (regs->tnpc + 4) & 0xffffffff; - } else { - regs->tpc = regs->tnpc; - regs->tnpc += 4; - } - - /* Condition codes and return value where set here for sigpause, - * and so got used by setup_frame, which again causes sigreturn() - * to return -EINTR. - */ - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - /* - * Return -EINTR and set condition code here, - * so the interrupted system call actually returns - * these. - */ - regs->tstate |= (TSTATE_ICARRY|TSTATE_XCARRY); - regs->u_regs[UREG_I0] = EINTR; - if (do_signal(&oldset, regs, 0, 0)) - return; - } + return _sigpause_common(set); } static inline int @@ -607,26 +519,29 @@ static inline void syscall_restart(unsigned long orig_i0, struct pt_regs *regs, * want to handle. Thus you cannot kill init even with a SIGKILL even by * mistake. */ -static int do_signal(sigset_t *oldset, struct pt_regs * regs, - unsigned long orig_i0, int restart_syscall) +static void do_signal(struct pt_regs *regs, unsigned long orig_i0, int restart_syscall) { siginfo_t info; struct signal_deliver_cookie cookie; struct k_sigaction ka; int signr; + sigset_t *oldset; cookie.restart_syscall = restart_syscall; cookie.orig_i0 = orig_i0; - if (!oldset) + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + oldset = ¤t->saved_sigmask; + else oldset = ¤t->blocked; #ifdef CONFIG_SPARC32_COMPAT if (test_thread_flag(TIF_32BIT)) { - extern int do_signal32(sigset_t *, struct pt_regs *, - unsigned long, int); - return do_signal32(oldset, regs, orig_i0, - cookie.restart_syscall); + extern void do_signal32(sigset_t *, struct pt_regs *, + unsigned long, int); + do_signal32(oldset, regs, orig_i0, + cookie.restart_syscall); + return; } #endif @@ -635,7 +550,15 @@ static int do_signal(sigset_t *oldset, struct pt_regs * regs, if (cookie.restart_syscall) syscall_restart(orig_i0, regs, &ka.sa); handle_signal(signr, &ka, &info, oldset, regs); - return 1; + + /* a signal was successfully delivered; the saved + * sigmask will have been stored in the signal frame, + * and will be restored by sigreturn, so we can simply + * clear the TIF_RESTORE_SIGMASK flag. + */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + clear_thread_flag(TIF_RESTORE_SIGMASK); + return; } if (cookie.restart_syscall && (regs->u_regs[UREG_I0] == ERESTARTNOHAND || @@ -652,15 +575,21 @@ static int do_signal(sigset_t *oldset, struct pt_regs * regs, regs->tpc -= 4; regs->tnpc -= 4; } - return 0; + + /* if there's no signal to deliver, we just put the saved sigmask + * back + */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) { + clear_thread_flag(TIF_RESTORE_SIGMASK); + sigprocmask(SIG_SETMASK, ¤t->saved_sigmask, NULL); + } } -void do_notify_resume(sigset_t *oldset, struct pt_regs *regs, - unsigned long orig_i0, int restart_syscall, +void do_notify_resume(struct pt_regs *regs, unsigned long orig_i0, int restart_syscall, unsigned long thread_info_flags) { - if (thread_info_flags & _TIF_SIGPENDING) - do_signal(oldset, regs, orig_i0, restart_syscall); + if (thread_info_flags & (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK)) + do_signal(regs, orig_i0, restart_syscall); } void ptrace_signal_deliver(struct pt_regs *regs, void *cookie) diff --git a/arch/sparc64/kernel/signal32.c b/arch/sparc64/kernel/signal32.c index 009a86e5ded..708ba9b42cd 100644 --- a/arch/sparc64/kernel/signal32.c +++ b/arch/sparc64/kernel/signal32.c @@ -32,9 +32,6 @@ #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP))) -int do_signal32(sigset_t *oldset, struct pt_regs *regs, - unsigned long orig_o0, int ret_from_syscall); - /* Signal frames: the original one (compatible with SunOS): * * Set up a signal frame... Make the stack look the way SunOS @@ -226,102 +223,6 @@ int copy_siginfo_from_user32(siginfo_t *to, compat_siginfo_t __user *from) return 0; } -/* - * atomically swap in the new signal mask, and wait for a signal. - * This is really tricky on the Sparc, watch out... - */ -asmlinkage void _sigpause32_common(compat_old_sigset_t set, struct pt_regs *regs) -{ - sigset_t saveset; - - set &= _BLOCKABLE; - spin_lock_irq(¤t->sighand->siglock); - saveset = current->blocked; - siginitset(¤t->blocked, set); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - regs->tpc = regs->tnpc; - regs->tnpc += 4; - if (test_thread_flag(TIF_32BIT)) { - regs->tpc &= 0xffffffff; - regs->tnpc &= 0xffffffff; - } - - /* Condition codes and return value where set here for sigpause, - * and so got used by setup_frame, which again causes sigreturn() - * to return -EINTR. - */ - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - /* - * Return -EINTR and set condition code here, - * so the interrupted system call actually returns - * these. - */ - regs->tstate |= TSTATE_ICARRY; - regs->u_regs[UREG_I0] = EINTR; - if (do_signal32(&saveset, regs, 0, 0)) - return; - } -} - -asmlinkage void do_rt_sigsuspend32(u32 uset, size_t sigsetsize, struct pt_regs *regs) -{ - sigset_t oldset, set; - compat_sigset_t set32; - - /* XXX: Don't preclude handling different sized sigset_t's. */ - if (((compat_size_t)sigsetsize) != sizeof(sigset_t)) { - regs->tstate |= TSTATE_ICARRY; - regs->u_regs[UREG_I0] = EINVAL; - return; - } - if (copy_from_user(&set32, compat_ptr(uset), sizeof(set32))) { - regs->tstate |= TSTATE_ICARRY; - regs->u_regs[UREG_I0] = EFAULT; - return; - } - switch (_NSIG_WORDS) { - case 4: set.sig[3] = set32.sig[6] + (((long)set32.sig[7]) << 32); - case 3: set.sig[2] = set32.sig[4] + (((long)set32.sig[5]) << 32); - case 2: set.sig[1] = set32.sig[2] + (((long)set32.sig[3]) << 32); - case 1: set.sig[0] = set32.sig[0] + (((long)set32.sig[1]) << 32); - } - sigdelsetmask(&set, ~_BLOCKABLE); - spin_lock_irq(¤t->sighand->siglock); - oldset = current->blocked; - current->blocked = set; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - regs->tpc = regs->tnpc; - regs->tnpc += 4; - if (test_thread_flag(TIF_32BIT)) { - regs->tpc &= 0xffffffff; - regs->tnpc &= 0xffffffff; - } - - /* Condition codes and return value where set here for sigpause, - * and so got used by setup_frame, which again causes sigreturn() - * to return -EINTR. - */ - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - /* - * Return -EINTR and set condition code here, - * so the interrupted system call actually returns - * these. - */ - regs->tstate |= TSTATE_ICARRY; - regs->u_regs[UREG_I0] = EINTR; - if (do_signal32(&oldset, regs, 0, 0)) - return; - } -} - static int restore_fpu_state32(struct pt_regs *regs, __siginfo_fpu_t __user *fpu) { unsigned long *fpregs = current_thread_info()->fpregs; @@ -1362,8 +1263,8 @@ static inline void syscall_restart32(unsigned long orig_i0, struct pt_regs *regs * want to handle. Thus you cannot kill init even with a SIGKILL even by * mistake. */ -int do_signal32(sigset_t *oldset, struct pt_regs * regs, - unsigned long orig_i0, int restart_syscall) +void do_signal32(sigset_t *oldset, struct pt_regs * regs, + unsigned long orig_i0, int restart_syscall) { siginfo_t info; struct signal_deliver_cookie cookie; @@ -1380,7 +1281,15 @@ int do_signal32(sigset_t *oldset, struct pt_regs * regs, syscall_restart32(orig_i0, regs, &ka.sa); handle_signal32(signr, &ka, &info, oldset, regs, svr4_signal); - return 1; + + /* a signal was successfully delivered; the saved + * sigmask will have been stored in the signal frame, + * and will be restored by sigreturn, so we can simply + * clear the TIF_RESTORE_SIGMASK flag. + */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + clear_thread_flag(TIF_RESTORE_SIGMASK); + return; } if (cookie.restart_syscall && (regs->u_regs[UREG_I0] == ERESTARTNOHAND || @@ -1397,7 +1306,14 @@ int do_signal32(sigset_t *oldset, struct pt_regs * regs, regs->tpc -= 4; regs->tnpc -= 4; } - return 0; + + /* if there's no signal to deliver, we just put the saved sigmask + * back + */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) { + clear_thread_flag(TIF_RESTORE_SIGMASK); + sigprocmask(SIG_SETMASK, ¤t->saved_sigmask, NULL); + } } struct sigstack32 { diff --git a/arch/sparc64/kernel/sparc64_ksyms.c b/arch/sparc64/kernel/sparc64_ksyms.c index d177d7e5c9d..3c06bfb92a8 100644 --- a/arch/sparc64/kernel/sparc64_ksyms.c +++ b/arch/sparc64/kernel/sparc64_ksyms.c @@ -69,7 +69,6 @@ struct poll { extern void die_if_kernel(char *str, struct pt_regs *regs); extern pid_t kernel_thread(int (*fn)(void *), void * arg, unsigned long flags); -void _sigpause_common (unsigned int set, struct pt_regs *); extern void *__bzero(void *, size_t); extern void *__memscan_zero(void *, size_t); extern void *__memscan_generic(void *, int, size_t); @@ -236,9 +235,10 @@ EXPORT_SYMBOL(pci_dma_supported); /* I/O device mmaping on Sparc64. */ EXPORT_SYMBOL(io_remap_pfn_range); +#ifdef CONFIG_COMPAT /* Solaris/SunOS binary compatibility */ -EXPORT_SYMBOL(_sigpause_common); EXPORT_SYMBOL(verify_compat_iovec); +#endif EXPORT_SYMBOL(dump_fpu); EXPORT_SYMBOL(pte_alloc_one_kernel); diff --git a/arch/sparc64/kernel/systbls.S b/arch/sparc64/kernel/systbls.S index 98d24bc0004..5ed1a17cd5b 100644 --- a/arch/sparc64/kernel/systbls.S +++ b/arch/sparc64/kernel/systbls.S @@ -41,7 +41,7 @@ sys_call_table32: /*90*/ .word sys_dup2, sys_setfsuid, compat_sys_fcntl, sys32_select, sys_setfsgid .word sys_fsync, sys32_setpriority, sys_nis_syscall, sys_nis_syscall, sys_nis_syscall /*100*/ .word sys32_getpriority, sys32_rt_sigreturn, sys32_rt_sigaction, sys32_rt_sigprocmask, sys32_rt_sigpending - .word compat_sys_rt_sigtimedwait, sys32_rt_sigqueueinfo, sys32_rt_sigsuspend, sys_setresuid, sys_getresuid + .word compat_sys_rt_sigtimedwait, sys32_rt_sigqueueinfo, compat_sys_rt_sigsuspend, sys_setresuid, sys_getresuid /*110*/ .word sys_setresgid, sys_getresgid, sys_setregid, sys_nis_syscall, sys_nis_syscall .word sys32_getgroups, sys32_gettimeofday, sys32_getrusage, sys_nis_syscall, sys_getcwd /*120*/ .word compat_sys_readv, compat_sys_writev, sys32_settimeofday, sys32_fchown16, sys_fchmod @@ -76,7 +76,10 @@ sys_call_table32: .word sys_timer_delete, compat_sys_timer_create, sys_ni_syscall, compat_sys_io_setup, sys_io_destroy /*270*/ .word sys32_io_submit, sys_io_cancel, compat_sys_io_getevents, sys32_mq_open, sys_mq_unlink .word compat_sys_mq_timedsend, compat_sys_mq_timedreceive, compat_sys_mq_notify, compat_sys_mq_getsetattr, compat_sys_waitid -/*280*/ .word sys_ni_syscall, sys_add_key, sys_request_key, sys_keyctl +/*280*/ .word sys_ni_syscall, sys_add_key, sys_request_key, sys_keyctl, compat_sys_openat + .word sys_mkdirat, sys_mknodat, sys_fchownat, sys_futimesat, compat_sys_newfstatat +/*285*/ .word sys_unlinkat, sys_renameat, sys_linkat, sys_symlinkat, sys_readlinkat + .word sys_fchmodat, sys_faccessat, compat_sys_pselect6, compat_sys_ppoll #endif /* CONFIG_COMPAT */ @@ -142,7 +145,10 @@ sys_call_table: .word sys_timer_delete, sys_timer_create, sys_ni_syscall, sys_io_setup, sys_io_destroy /*270*/ .word sys_io_submit, sys_io_cancel, sys_io_getevents, sys_mq_open, sys_mq_unlink .word sys_mq_timedsend, sys_mq_timedreceive, sys_mq_notify, sys_mq_getsetattr, sys_waitid -/*280*/ .word sys_nis_syscall, sys_add_key, sys_request_key, sys_keyctl +/*280*/ .word sys_nis_syscall, sys_add_key, sys_request_key, sys_keyctl, sys_openat + .word sys_mkdirat, sys_mknodat, sys_fchownat, sys_futimesat, compat_sys_newfstatat +/*285*/ .word sys_unlinkat, sys_renameat, sys_linkat, sys_symlinkat, sys_readlinkat + .word sys_fchmodat, sys_faccessat, sys_pselect6, sys_ppoll #if defined(CONFIG_SUNOS_EMUL) || defined(CONFIG_SOLARIS_EMUL) || \ defined(CONFIG_SOLARIS_EMUL_MODULE) @@ -239,13 +245,20 @@ sunos_sys_table: /*250*/ .word sunos_nosys, sunos_nosys, sunos_nosys .word sunos_nosys, sunos_nosys, sunos_nosys .word sunos_nosys, sunos_nosys, sunos_nosys + .word sunos_nosys +/*260*/ .word sunos_nosys, sunos_nosys, sunos_nosys .word sunos_nosys, sunos_nosys, sunos_nosys .word sunos_nosys, sunos_nosys, sunos_nosys + .word sunos_nosys +/*270*/ .word sunos_nosys, sunos_nosys, sunos_nosys .word sunos_nosys, sunos_nosys, sunos_nosys .word sunos_nosys, sunos_nosys, sunos_nosys + .word sunos_nosys +/*280*/ .word sunos_nosys, sunos_nosys, sunos_nosys .word sunos_nosys, sunos_nosys, sunos_nosys .word sunos_nosys, sunos_nosys, sunos_nosys + .word sunos_nosys +/*290*/ .word sunos_nosys, sunos_nosys, sunos_nosys .word sunos_nosys, sunos_nosys, sunos_nosys .word sunos_nosys, sunos_nosys, sunos_nosys - .word sunos_nosys #endif diff --git a/arch/sparc64/solaris/entry64.S b/arch/sparc64/solaris/entry64.S index 4b6ae583c0a..eb314ed23cd 100644 --- a/arch/sparc64/solaris/entry64.S +++ b/arch/sparc64/solaris/entry64.S @@ -180,6 +180,8 @@ solaris_sigsuspend: nop call sys_sigsuspend stx %o0, [%sp + PTREGS_OFF + PT_V9_I0] + b,pt %xcc, ret_from_solaris + nop .globl solaris_getpid solaris_getpid: diff --git a/include/asm-sparc/thread_info.h b/include/asm-sparc/thread_info.h index 65f060b040a..91b9f5888c8 100644 --- a/include/asm-sparc/thread_info.h +++ b/include/asm-sparc/thread_info.h @@ -128,9 +128,10 @@ BTFIXUPDEF_CALL(void, free_thread_info, struct thread_info *) * thread information flag bit numbers */ #define TIF_SYSCALL_TRACE 0 /* syscall trace active */ -#define TIF_NOTIFY_RESUME 1 /* resumption notification requested */ +/* flag bit 1 is available */ #define TIF_SIGPENDING 2 /* signal pending */ #define TIF_NEED_RESCHED 3 /* rescheduling necessary */ +#define TIF_RESTORE_SIGMASK 4 /* restore signal mask in do_signal() */ #define TIF_USEDFPU 8 /* FPU was used by this task * this quantum (SMP) */ #define TIF_POLLING_NRFLAG 9 /* true if poll_idle() is polling @@ -139,9 +140,9 @@ BTFIXUPDEF_CALL(void, free_thread_info, struct thread_info *) /* as above, but as bit values */ #define _TIF_SYSCALL_TRACE (1< Date: Wed, 18 Jan 2006 23:46:51 -0800 Subject: [IA64] sem2mutex: arch/ia64/ia32/sys_ia32.c Migrate arch/ia64/ia32/sys_ia32 to using a mutex for mmap protection. Signed-off-by: Jes Sorensen Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Tony Luck --- arch/ia64/ia32/sys_ia32.c | 28 ++++++++++++---------------- 1 file changed, 12 insertions(+), 16 deletions(-) diff --git a/arch/ia64/ia32/sys_ia32.c b/arch/ia64/ia32/sys_ia32.c index 3945d378bd7..70dba1f0e2e 100644 --- a/arch/ia64/ia32/sys_ia32.c +++ b/arch/ia64/ia32/sys_ia32.c @@ -52,9 +52,9 @@ #include #include #include +#include #include -#include #include #include #include @@ -86,7 +86,7 @@ * while doing so. */ /* XXX make per-mm: */ -static DECLARE_MUTEX(ia32_mmap_sem); +static DEFINE_MUTEX(ia32_mmap_mutex); asmlinkage long sys32_execve (char __user *name, compat_uptr_t __user *argv, compat_uptr_t __user *envp, @@ -895,11 +895,11 @@ ia32_do_mmap (struct file *file, unsigned long addr, unsigned long len, int prot prot = get_prot32(prot); #if PAGE_SHIFT > IA32_PAGE_SHIFT - down(&ia32_mmap_sem); + mutex_lock(&ia32_mmap_mutex); { addr = emulate_mmap(file, addr, len, prot, flags, offset); } - up(&ia32_mmap_sem); + mutex_unlock(&ia32_mmap_mutex); #else down_write(¤t->mm->mmap_sem); { @@ -1000,11 +1000,9 @@ sys32_munmap (unsigned int start, unsigned int len) if (start >= end) return 0; - down(&ia32_mmap_sem); - { - ret = sys_munmap(start, end - start); - } - up(&ia32_mmap_sem); + mutex_lock(&ia32_mmap_mutex); + ret = sys_munmap(start, end - start); + mutex_unlock(&ia32_mmap_mutex); #endif return ret; } @@ -1056,7 +1054,7 @@ sys32_mprotect (unsigned int start, unsigned int len, int prot) if (retval < 0) return retval; - down(&ia32_mmap_sem); + mutex_lock(&ia32_mmap_mutex); { if (offset_in_page(start)) { /* start address is 4KB aligned but not page aligned. */ @@ -1080,7 +1078,7 @@ sys32_mprotect (unsigned int start, unsigned int len, int prot) retval = sys_mprotect(start, end - start, prot); } out: - up(&ia32_mmap_sem); + mutex_unlock(&ia32_mmap_mutex); return retval; #endif } @@ -1124,11 +1122,9 @@ sys32_mremap (unsigned int addr, unsigned int old_len, unsigned int new_len, old_len = PAGE_ALIGN(old_end) - addr; new_len = PAGE_ALIGN(new_end) - addr; - down(&ia32_mmap_sem); - { - ret = sys_mremap(addr, old_len, new_len, flags, new_addr); - } - up(&ia32_mmap_sem); + mutex_lock(&ia32_mmap_mutex); + ret = sys_mremap(addr, old_len, new_len, flags, new_addr); + mutex_unlock(&ia32_mmap_mutex); if ((ret >= 0) && (old_len < new_len)) { /* mremap expanded successfully */ -- cgit v1.2.3 From 60f1c4443c4d391d8dfbe709f13296067b6c8021 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Wed, 18 Jan 2006 23:46:52 -0800 Subject: [IA64] sem2mutex: arch/ia64/kernel/perfmon.c Migrate perfmon from using an old semaphore to a completion handler. Signed-off-by: Jes Sorensen Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Tony Luck --- arch/ia64/kernel/perfmon.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c index 2ea4b39efff..9c5194b385d 100644 --- a/arch/ia64/kernel/perfmon.c +++ b/arch/ia64/kernel/perfmon.c @@ -40,6 +40,7 @@ #include #include #include +#include #include #include @@ -286,7 +287,7 @@ typedef struct pfm_context { unsigned long ctx_ovfl_regs[4]; /* which registers overflowed (notification) */ - struct semaphore ctx_restart_sem; /* use for blocking notification mode */ + struct completion ctx_restart_done; /* use for blocking notification mode */ unsigned long ctx_used_pmds[4]; /* bitmask of PMD used */ unsigned long ctx_all_pmds[4]; /* bitmask of all accessible PMDs */ @@ -1991,7 +1992,7 @@ pfm_close(struct inode *inode, struct file *filp) /* * force task to wake up from MASKED state */ - up(&ctx->ctx_restart_sem); + complete(&ctx->ctx_restart_done); DPRINT(("waking up ctx_state=%d\n", state)); @@ -2706,7 +2707,7 @@ pfm_context_create(pfm_context_t *ctx, void *arg, int count, struct pt_regs *reg /* * init restart semaphore to locked */ - sema_init(&ctx->ctx_restart_sem, 0); + init_completion(&ctx->ctx_restart_done); /* * activation is used in SMP only @@ -3687,7 +3688,7 @@ pfm_restart(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) */ if (CTX_OVFL_NOBLOCK(ctx) == 0 && state == PFM_CTX_MASKED) { DPRINT(("unblocking [%d] \n", task->pid)); - up(&ctx->ctx_restart_sem); + complete(&ctx->ctx_restart_done); } else { DPRINT(("[%d] armed exit trap\n", task->pid)); @@ -5089,7 +5090,7 @@ pfm_handle_work(void) * may go through without blocking on SMP systems * if restart has been received already by the time we call down() */ - ret = down_interruptible(&ctx->ctx_restart_sem); + ret = wait_for_completion_interruptible(&ctx->ctx_restart_done); DPRINT(("after block sleeping ret=%d\n", ret)); -- cgit v1.2.3 From 386d1d50c8eef254653b1015fde06622ef38ba76 Mon Sep 17 00:00:00 2001 From: John Hawkes Date: Wed, 18 Jan 2006 23:46:53 -0800 Subject: [IA64] eliminate softlockup warning Fix an unnecessary softlockup watchdog warning in the ia64 uncached_build_memmap() that occurs occasionally at 256p and always at 512p. The problem occurs at boot time. Signed-off-by: John Hawkes Signed-off-by: Andrew Morton Signed-off-by: Tony Luck --- arch/ia64/kernel/uncached.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/ia64/kernel/uncached.c b/arch/ia64/kernel/uncached.c index b631cf86ed4..fcd2bad0286 100644 --- a/arch/ia64/kernel/uncached.c +++ b/arch/ia64/kernel/uncached.c @@ -210,6 +210,7 @@ uncached_build_memmap(unsigned long start, unsigned long end, void *arg) dprintk(KERN_ERR "uncached_build_memmap(%lx %lx)\n", start, end); + touch_softlockup_watchdog(); memset((char *)start, 0, length); node = paddr_to_nid(start - __IA64_UNCACHED_OFFSET); -- cgit v1.2.3 From 4f2d7680cb1ac5c5a70f3ba2447d5aa5c0a1643a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 19 Jan 2006 16:58:37 -0800 Subject: [NETFILTER] x_tables: Make XT_ALIGN align as strictly as necessary. Or else we break on ppc32 and other 32-bit platforms. Based upon a patch from Harald Welte. Signed-off-by: David S. Miller --- include/linux/netfilter/x_tables.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index 472f0483480..59ff6c430cf 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -19,7 +19,7 @@ struct xt_get_revision /* For standard target */ #define XT_RETURN (-NF_REPEAT - 1) -#define XT_ALIGN(s) (((s) + (__alignof__(void *)-1)) & ~(__alignof__(void *)-1)) +#define XT_ALIGN(s) (((s) + (__alignof__(u_int64_t)-1)) & ~(__alignof__(u_int64_t)-1)) /* Standard return verdict, or do jump. */ #define XT_STANDARD_TARGET "" -- cgit v1.2.3 From 7e732bfc5570b8f9bb5f155cf36e94b2e7d6bf6a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 19 Jan 2006 16:40:42 -0800 Subject: [PATCH] Fix regression added by ppoll/pselect code. The compat layer timeout handling changes in: 9f72949f679df06021c9e43886c9191494fdb007 are busted. This is most easily seen with an X application that uses sub-second select/poll timeout such as emacs. You hit a key and it takes a second or so before the app responds. The two ROUND_UP() calls upon entry are using {tv,ts}_sec where it should instead be using {tv_usec,ts_nsec}, which perfectly explains the observed incorrect behavior. Another bug shot down with git bisect. Signed-off-by: David S. Miller Signed-off-by: Linus Torvalds --- fs/compat.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/compat.c b/fs/compat.c index 18b21b4c9e3..ff0bafcff72 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -1743,7 +1743,7 @@ asmlinkage long compat_sys_select(int n, compat_ulong_t __user *inp, if ((u64)tv.tv_sec >= (u64)MAX_INT64_SECONDS) timeout = -1; /* infinite */ else { - timeout = ROUND_UP(tv.tv_sec, 1000000/HZ); + timeout = ROUND_UP(tv.tv_usec, 1000000/HZ); timeout += tv.tv_sec * HZ; } } @@ -1884,7 +1884,7 @@ asmlinkage long compat_sys_ppoll(struct pollfd __user *ufds, /* We assume that ts.tv_sec is always lower than the number of seconds that can be expressed in an s64. Otherwise the compiler bitches at us */ - timeout = ROUND_UP(ts.tv_sec, 1000000000/HZ); + timeout = ROUND_UP(ts.tv_nsec, 1000000000/HZ); timeout += ts.tv_sec * HZ; } -- cgit v1.2.3 From 3ee68c4af3fd7228c1be63254b9f884614f9ebb2 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 20 Jan 2006 01:49:15 -0800 Subject: [SPARC64]: Use compat_sys_futimesat in 32-bit syscall table. Signed-off-by: David S. Miller --- arch/sparc64/kernel/systbls.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/sparc64/kernel/systbls.S b/arch/sparc64/kernel/systbls.S index 5ed1a17cd5b..bf0fc5bfbfb 100644 --- a/arch/sparc64/kernel/systbls.S +++ b/arch/sparc64/kernel/systbls.S @@ -77,7 +77,7 @@ sys_call_table32: /*270*/ .word sys32_io_submit, sys_io_cancel, compat_sys_io_getevents, sys32_mq_open, sys_mq_unlink .word compat_sys_mq_timedsend, compat_sys_mq_timedreceive, compat_sys_mq_notify, compat_sys_mq_getsetattr, compat_sys_waitid /*280*/ .word sys_ni_syscall, sys_add_key, sys_request_key, sys_keyctl, compat_sys_openat - .word sys_mkdirat, sys_mknodat, sys_fchownat, sys_futimesat, compat_sys_newfstatat + .word sys_mkdirat, sys_mknodat, sys_fchownat, compat_sys_futimesat, compat_sys_newfstatat /*285*/ .word sys_unlinkat, sys_renameat, sys_linkat, sys_symlinkat, sys_readlinkat .word sys_fchmodat, sys_faccessat, compat_sys_pselect6, compat_sys_ppoll -- cgit v1.2.3 From 6a986ce45d45b099ddf676c340267765e76db91e Mon Sep 17 00:00:00 2001 From: Eric Sesterhenn Date: Fri, 20 Jan 2006 23:30:01 +0300 Subject: [PATCH] bonding: fix ->get_settings error checking Since get_settings() returns a signed int and it gets checked for < 0 to catch an error, res should be a signed int too. Signed-off-by: Eric Sesterhenn Signed-off-by: Alexey Dobriyan Signed-off-by: Jeff Garzik --- drivers/net/bonding/bond_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 2582d98ef5c..4ff006c3762 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -576,7 +576,7 @@ static int bond_update_speed_duplex(struct slave *slave) slave->duplex = DUPLEX_FULL; if (slave_dev->ethtool_ops) { - u32 res; + int res; if (!slave_dev->ethtool_ops->get_settings) { return -1; -- cgit v1.2.3 From 6f9d47220eb2d1b17a0a3ecaf1b564ff95b8393d Mon Sep 17 00:00:00 2001 From: Eric Sesterhenn Date: Fri, 20 Jan 2006 23:32:56 +0300 Subject: [PATCH] acenic: fix checking of read_eeprom_byte() return values tmp in ace_init is u32 thus rendering read_eeprom_byte() return values checks useless. Signed-off-by: Eric Sesterhenn Signed-off-by: Alexey Dobriyan Signed-off-by: Jeff Garzik --- drivers/net/acenic.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/acenic.c b/drivers/net/acenic.c index b8953de5664..b508812e97a 100644 --- a/drivers/net/acenic.c +++ b/drivers/net/acenic.c @@ -1002,6 +1002,8 @@ static int __devinit ace_init(struct net_device *dev) mac1 = 0; for(i = 0; i < 4; i++) { + int tmp; + mac1 = mac1 << 8; tmp = read_eeprom_byte(dev, 0x8c+i); if (tmp < 0) { @@ -1012,6 +1014,8 @@ static int __devinit ace_init(struct net_device *dev) } mac2 = 0; for(i = 4; i < 8; i++) { + int tmp; + mac2 = mac2 << 8; tmp = read_eeprom_byte(dev, 0x8c+i); if (tmp < 0) { -- cgit v1.2.3 From c35ca399e09828f3f6b40c0007a95a6582d90347 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 20 Jan 2006 21:13:17 -0800 Subject: [PATCH] b44: fix laptop carrier detect On my laptop, the b44 device is created and the carrier state defaults to ON when created by alloc_etherdev. This means tools like NetworkManager see the carrier as On and try and bring the device up. The correct thing to do is mark the carrier as Off when device is created. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik --- drivers/net/b44.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/b44.c b/drivers/net/b44.c index df9d6e80c4f..c3267e4e1bb 100644 --- a/drivers/net/b44.c +++ b/drivers/net/b44.c @@ -1399,7 +1399,6 @@ static int b44_open(struct net_device *dev) b44_init_rings(bp); b44_init_hw(bp); - netif_carrier_off(dev); b44_check_phy(bp); err = request_irq(dev->irq, b44_interrupt, SA_SHIRQ, dev->name, dev); @@ -1464,7 +1463,7 @@ static int b44_close(struct net_device *dev) #endif b44_halt(bp); b44_free_rings(bp); - netif_carrier_off(bp->dev); + netif_carrier_off(dev); spin_unlock_irq(&bp->lock); @@ -2000,6 +1999,8 @@ static int __devinit b44_init_one(struct pci_dev *pdev, dev->irq = pdev->irq; SET_ETHTOOL_OPS(dev, &b44_ethtool_ops); + netif_carrier_off(dev); + err = b44_get_invariants(bp); if (err) { printk(KERN_ERR PFX "Problem fetching invariants of chip, " -- cgit v1.2.3 From efd51b5c6798d103e3aa683464aebb2019b62119 Mon Sep 17 00:00:00 2001 From: Ananda Raju Date: Thu, 19 Jan 2006 14:11:54 -0500 Subject: [PATCH] s2io: scatter-gather fix There is a problem with fragmented skb in s2io driver version 2.0.9.4 available in 2.6.16-rc1 kernel. The adapter will fail to transmit if any scatter-gather skb arrives. This patch provides fix for the above described problem. Signed-off-by: Ananda Raju Signed-off-by: Jeff Garzik --- drivers/net/s2io.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c index 89c46787676..49b597cbc19 100644 --- a/drivers/net/s2io.c +++ b/drivers/net/s2io.c @@ -3586,7 +3586,7 @@ static int s2io_xmit(struct sk_buff *skb, struct net_device *dev) txdp->Buffer_Pointer = (u64) pci_map_page (sp->pdev, frag->page, frag->page_offset, frag->size, PCI_DMA_TODEVICE); - txdp->Control_1 |= TXD_BUFFER0_SIZE(frag->size); + txdp->Control_1 = TXD_BUFFER0_SIZE(frag->size); if (skb_shinfo(skb)->ufo_size) txdp->Control_1 |= TXD_UFO_EN; } -- cgit v1.2.3