From 5b6f1eb97d462a45be3b30759758b5fdbb562c8c Mon Sep 17 00:00:00 2001 From: Alain Knaff Date: Mon, 10 Nov 2008 17:08:08 -0800 Subject: vfs: lseek(fd, 0, SEEK_CUR) race condition This patch fixes a race condition in lseek. While it is expected that unpredictable behaviour may result while repositioning the offset of a file descriptor concurrently with reading/writing to the same file descriptor, this should not happen when merely *reading* the file descriptor's offset. Unfortunately, the only portable way in Unix to read a file descriptor's offset is lseek(fd, 0, SEEK_CUR); however executing this concurrently with read/write may mess up the position. [with fixes from akpm] Signed-off-by: Alain Knaff Signed-off-by: Andrew Morton Signed-off-by: Al Viro --- fs/read_write.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'fs') diff --git a/fs/read_write.c b/fs/read_write.c index 969a6d9c020..5cc6924eb15 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -50,6 +50,14 @@ generic_file_llseek_unlocked(struct file *file, loff_t offset, int origin) offset += inode->i_size; break; case SEEK_CUR: + /* + * Here we special-case the lseek(fd, 0, SEEK_CUR) + * position-querying operation. Avoid rewriting the "same" + * f_pos value back to the file because a concurrent read(), + * write() or lseek() might have altered it + */ + if (offset == 0) + return file->f_pos; offset += file->f_pos; break; } @@ -105,6 +113,10 @@ loff_t default_llseek(struct file *file, loff_t offset, int origin) offset += i_size_read(file->f_path.dentry->d_inode); break; case SEEK_CUR: + if (offset == 0) { + retval = file->f_pos; + goto out; + } offset += file->f_pos; } retval = -EINVAL; @@ -115,6 +127,7 @@ loff_t default_llseek(struct file *file, loff_t offset, int origin) } retval = offset; } +out: unlock_kernel(); return retval; } -- cgit v1.2.3 From c765d479037808532310212e9b3fa95760e975f2 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 4 Dec 2008 09:50:55 -0500 Subject: affs: do not zero ->i_op it is already set to empty table and should never be NULL Signed-off-by: Al Viro --- fs/affs/inode.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/affs/inode.c b/fs/affs/inode.c index 415d9c67ac1..3c4ec7d864c 100644 --- a/fs/affs/inode.c +++ b/fs/affs/inode.c @@ -119,8 +119,7 @@ struct inode *affs_iget(struct super_block *sb, unsigned long ino) goto bad_inode; #else inode->i_mode |= S_IFDIR; - inode->i_op = NULL; - inode->i_fop = NULL; + /* ... and leave ->i_op and ->i_fop pointing to empty */ break; #endif case ST_LINKFILE: -- cgit v1.2.3 From 261964c60ff6524076d439da9386d4782729c4d9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 4 Dec 2008 09:57:25 -0500 Subject: isofs check for NULL ->i_op in root directory is dead code for one thing it never happens, for another we check that inode is a directory right after that place anyway (and we'd already checked that reading it from disk has not failed). Signed-off-by: Al Viro --- fs/isofs/inode.c | 6 ------ 1 file changed, 6 deletions(-) (limited to 'fs') diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index 3f8af0f1505..6147ec3643a 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -855,10 +855,6 @@ root_found: } sbi->s_joliet_level = joliet_level; - /* check the root inode */ - if (!inode->i_op) - goto out_bad_root; - /* Make sure the root inode is a directory */ if (!S_ISDIR(inode->i_mode)) { printk(KERN_WARNING @@ -886,8 +882,6 @@ root_found: /* * Display error messages and free resources. */ -out_bad_root: - printk(KERN_WARNING "%s: root inode not initialized\n", __func__); out_iput: iput(inode); goto out_no_inode; -- cgit v1.2.3 From 9742df331deb3fce95b321f38d4ea0c4e75edb63 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 4 Dec 2008 09:59:23 -0500 Subject: ntfs: don't NULL i_op it's already set to empty table (and no, ntfs doesn't have any explicit checks for NULL ->i_op or NULL ->i_fop) Signed-off-by: Al Viro --- fs/ntfs/inode.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'fs') diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c index e9da092e277..86bef156cf0 100644 --- a/fs/ntfs/inode.c +++ b/fs/ntfs/inode.c @@ -1406,9 +1406,6 @@ static int ntfs_read_locked_attr_inode(struct inode *base_vi, struct inode *vi) ni->allocated_size = sle64_to_cpu( a->data.non_resident.allocated_size); } - /* Setup the operations for this attribute inode. */ - vi->i_op = NULL; - vi->i_fop = NULL; if (NInoMstProtected(ni)) vi->i_mapping->a_ops = &ntfs_mst_aops; else -- cgit v1.2.3 From acfa4380efe77e290d3a96b11cd4c9f24f4fbb18 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 4 Dec 2008 10:06:33 -0500 Subject: inode->i_op is never NULL We used to have rather schizophrenic set of checks for NULL ->i_op even though it had been eliminated years ago. You'd need to go out of your way to set it to NULL explicitly _and_ a bunch of code would die on such inodes anyway. After killing two remaining places that still did that bogosity, all that crap can go away. Signed-off-by: Al Viro --- fs/cifs/inode.c | 2 +- fs/ecryptfs/inode.c | 3 +-- fs/namei.c | 32 +++++++++++++------------------- fs/nfsd/vfs.c | 8 ++++---- fs/open.c | 2 +- fs/stat.c | 2 +- fs/xattr.c | 2 +- 7 files changed, 22 insertions(+), 29 deletions(-) (limited to 'fs') diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index f247da9f4ed..5ab9896fdcb 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -1641,7 +1641,7 @@ do_expand: i_size_write(inode, offset); spin_unlock(&inode->i_lock); out_truncate: - if (inode->i_op && inode->i_op->truncate) + if (inode->i_op->truncate) inode->i_op->truncate(inode); return 0; out_sig: diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 5e78fc17988..0111906a887 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -612,8 +612,7 @@ ecryptfs_readlink(struct dentry *dentry, char __user * buf, int bufsiz) struct ecryptfs_crypt_stat *crypt_stat; lower_dentry = ecryptfs_dentry_to_lower(dentry); - if (!lower_dentry->d_inode->i_op || - !lower_dentry->d_inode->i_op->readlink) { + if (!lower_dentry->d_inode->i_op->readlink) { rc = -EINVAL; goto out; } diff --git a/fs/namei.c b/fs/namei.c index dd5c9f0bf82..1f6656c3d1b 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -257,7 +257,7 @@ int inode_permission(struct inode *inode, int mask) return -EACCES; } - if (inode->i_op && inode->i_op->permission) + if (inode->i_op->permission) retval = inode->i_op->permission(inode, mask); else retval = generic_permission(inode, mask, NULL); @@ -432,7 +432,7 @@ static int exec_permission_lite(struct inode *inode) { umode_t mode = inode->i_mode; - if (inode->i_op && inode->i_op->permission) + if (inode->i_op->permission) return -EAGAIN; if (current_fsuid() == inode->i_uid) @@ -908,9 +908,6 @@ static int __link_path_walk(const char *name, struct nameidata *nd) inode = next.dentry->d_inode; if (!inode) goto out_dput; - err = -ENOTDIR; - if (!inode->i_op) - goto out_dput; if (inode->i_op->follow_link) { err = do_follow_link(&next, nd); @@ -920,9 +917,6 @@ static int __link_path_walk(const char *name, struct nameidata *nd) inode = nd->path.dentry->d_inode; if (!inode) break; - err = -ENOTDIR; - if (!inode->i_op) - break; } else path_to_nameidata(&next, nd); err = -ENOTDIR; @@ -961,7 +955,7 @@ last_component: break; inode = next.dentry->d_inode; if ((lookup_flags & LOOKUP_FOLLOW) - && inode && inode->i_op && inode->i_op->follow_link) { + && inode && inode->i_op->follow_link) { err = do_follow_link(&next, nd); if (err) goto return_err; @@ -973,7 +967,7 @@ last_component: break; if (lookup_flags & LOOKUP_DIRECTORY) { err = -ENOTDIR; - if (!inode->i_op || !inode->i_op->lookup) + if (!inode->i_op->lookup) break; } goto return_base; @@ -1469,7 +1463,7 @@ int vfs_create(struct inode *dir, struct dentry *dentry, int mode, if (error) return error; - if (!dir->i_op || !dir->i_op->create) + if (!dir->i_op->create) return -EACCES; /* shouldn't it be ENOSYS? */ mode &= S_IALLUGO; mode |= S_IFREG; @@ -1752,7 +1746,7 @@ do_last: error = -ENOENT; if (!path.dentry->d_inode) goto exit_dput; - if (path.dentry->d_inode->i_op && path.dentry->d_inode->i_op->follow_link) + if (path.dentry->d_inode->i_op->follow_link) goto do_link; path_to_nameidata(&path, &nd); @@ -1933,7 +1927,7 @@ int vfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) if ((S_ISCHR(mode) || S_ISBLK(mode)) && !capable(CAP_MKNOD)) return -EPERM; - if (!dir->i_op || !dir->i_op->mknod) + if (!dir->i_op->mknod) return -EPERM; error = devcgroup_inode_mknod(mode, dev); @@ -2035,7 +2029,7 @@ int vfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) if (error) return error; - if (!dir->i_op || !dir->i_op->mkdir) + if (!dir->i_op->mkdir) return -EPERM; mode &= (S_IRWXUGO|S_ISVTX); @@ -2126,7 +2120,7 @@ int vfs_rmdir(struct inode *dir, struct dentry *dentry) if (error) return error; - if (!dir->i_op || !dir->i_op->rmdir) + if (!dir->i_op->rmdir) return -EPERM; DQUOT_INIT(dir); @@ -2213,7 +2207,7 @@ int vfs_unlink(struct inode *dir, struct dentry *dentry) if (error) return error; - if (!dir->i_op || !dir->i_op->unlink) + if (!dir->i_op->unlink) return -EPERM; DQUOT_INIT(dir); @@ -2320,7 +2314,7 @@ int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname) if (error) return error; - if (!dir->i_op || !dir->i_op->symlink) + if (!dir->i_op->symlink) return -EPERM; error = security_inode_symlink(dir, dentry, oldname); @@ -2401,7 +2395,7 @@ int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_de */ if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) return -EPERM; - if (!dir->i_op || !dir->i_op->link) + if (!dir->i_op->link) return -EPERM; if (S_ISDIR(inode->i_mode)) return -EPERM; @@ -2608,7 +2602,7 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (error) return error; - if (!old_dir->i_op || !old_dir->i_op->rename) + if (!old_dir->i_op->rename) return -EPERM; DQUOT_INIT(old_dir); diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index d1c5f787b36..5245a396500 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1211,7 +1211,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, dirp = dentry->d_inode; err = nfserr_notdir; - if(!dirp->i_op || !dirp->i_op->lookup) + if (!dirp->i_op->lookup) goto out; /* * Check whether the response file handle has been verified yet. @@ -1347,7 +1347,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, /* Get all the sanity checks out of the way before * we lock the parent. */ err = nfserr_notdir; - if(!dirp->i_op || !dirp->i_op->lookup) + if (!dirp->i_op->lookup) goto out; fh_lock_nested(fhp, I_MUTEX_PARENT); @@ -1482,7 +1482,7 @@ nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp) inode = dentry->d_inode; err = nfserr_inval; - if (!inode->i_op || !inode->i_op->readlink) + if (!inode->i_op->readlink) goto out; touch_atime(fhp->fh_export->ex_path.mnt, dentry); @@ -2162,7 +2162,7 @@ nfsd_set_posix_acl(struct svc_fh *fhp, int type, struct posix_acl *acl) size_t size; int error; - if (!IS_POSIXACL(inode) || !inode->i_op || + if (!IS_POSIXACL(inode) || !inode->i_op->setxattr || !inode->i_op->removexattr) return -EOPNOTSUPP; switch(type) { diff --git a/fs/open.c b/fs/open.c index 1cd7d40e999..d882fd2351d 100644 --- a/fs/open.c +++ b/fs/open.c @@ -412,7 +412,7 @@ asmlinkage long sys_fallocate(int fd, int mode, loff_t offset, loff_t len) if (((offset + len) > inode->i_sb->s_maxbytes) || ((offset + len) < 0)) goto out_fput; - if (inode->i_op && inode->i_op->fallocate) + if (inode->i_op->fallocate) ret = inode->i_op->fallocate(inode, mode, offset, len); else ret = -EOPNOTSUPP; diff --git a/fs/stat.c b/fs/stat.c index 7c46fbeb8b7..7e12a6f8279 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -305,7 +305,7 @@ asmlinkage long sys_readlinkat(int dfd, const char __user *pathname, struct inode *inode = path.dentry->d_inode; error = -EINVAL; - if (inode->i_op && inode->i_op->readlink) { + if (inode->i_op->readlink) { error = security_inode_readlink(path.dentry); if (!error) { touch_atime(path.mnt, path.dentry); diff --git a/fs/xattr.c b/fs/xattr.c index 468377e6653..237804cd6b5 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -175,7 +175,7 @@ vfs_listxattr(struct dentry *d, char *list, size_t size) if (error) return error; error = -EOPNOTSUPP; - if (d->d_inode->i_op && d->d_inode->i_op->listxattr) { + if (d->d_inode->i_op->listxattr) { error = d->d_inode->i_op->listxattr(d, list, size); } else { error = security_inode_listsecurity(d->d_inode, list, size); -- cgit v1.2.3 From 56ff5efad96182f4d3cb3dc6b07396762c658f16 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 9 Dec 2008 09:34:39 -0500 Subject: zero i_uid/i_gid on inode allocation ... and don't bother in callers. Don't bother with zeroing i_blocks, while we are at it - it's already been zeroed. i_mode is not worth the effort; it has no common default value. Signed-off-by: Al Viro --- fs/autofs/inode.c | 2 -- fs/autofs4/inode.c | 4 ---- fs/binfmt_misc.c | 3 --- fs/configfs/inode.c | 3 --- fs/cramfs/inode.c | 2 -- fs/debugfs/inode.c | 3 --- fs/devpts/inode.c | 4 ---- fs/hugetlbfs/inode.c | 1 - fs/inode.c | 2 ++ fs/libfs.c | 5 ----- fs/ocfs2/dlm/dlmfs.c | 2 -- fs/omfs/inode.c | 1 - fs/openpromfs/inode.c | 3 --- fs/proc/base.c | 4 ---- fs/proc/proc_sysctl.c | 1 - fs/ramfs/inode.c | 1 - fs/romfs/inode.c | 1 - fs/sysfs/inode.c | 3 --- 18 files changed, 2 insertions(+), 43 deletions(-) (limited to 'fs') diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c index c773680d5c6..e1734f2d6e2 100644 --- a/fs/autofs/inode.c +++ b/fs/autofs/inode.c @@ -251,13 +251,11 @@ struct inode *autofs_iget(struct super_block *sb, unsigned long ino) inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO; inode->i_nlink = 2; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; - inode->i_blocks = 0; if (ino == AUTOFS_ROOT_INO) { inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR; inode->i_op = &autofs_root_inode_operations; inode->i_fop = &autofs_root_operations; - inode->i_uid = inode->i_gid = 0; /* Changed in read_super */ goto done; } diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index 7b19802cfef..cfc23e53b6f 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -455,11 +455,7 @@ struct inode *autofs4_get_inode(struct super_block *sb, if (sb->s_root) { inode->i_uid = sb->s_root->d_inode->i_uid; inode->i_gid = sb->s_root->d_inode->i_gid; - } else { - inode->i_uid = 0; - inode->i_gid = 0; } - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; if (S_ISDIR(inf->mode)) { diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index f2744ab4e5b..e1158cb4fbd 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -496,9 +496,6 @@ static struct inode *bm_get_inode(struct super_block *sb, int mode) if (inode) { inode->i_mode = mode; - inode->i_uid = 0; - inode->i_gid = 0; - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = current_fs_time(inode->i_sb); } diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c index 4803ccc9448..5d349d38e05 100644 --- a/fs/configfs/inode.c +++ b/fs/configfs/inode.c @@ -117,8 +117,6 @@ int configfs_setattr(struct dentry * dentry, struct iattr * iattr) static inline void set_default_inode_attr(struct inode * inode, mode_t mode) { inode->i_mode = mode; - inode->i_uid = 0; - inode->i_gid = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; } @@ -136,7 +134,6 @@ struct inode * configfs_new_inode(mode_t mode, struct configfs_dirent * sd) { struct inode * inode = new_inode(configfs_sb); if (inode) { - inode->i_blocks = 0; inode->i_mapping->a_ops = &configfs_aops; inode->i_mapping->backing_dev_info = &configfs_backing_dev_info; inode->i_op = &configfs_inode_operations; diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index f40423eb1a1..a07338d2d14 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -83,8 +83,6 @@ static struct inode *get_cramfs_inode(struct super_block *sb, inode->i_op = &page_symlink_inode_operations; inode->i_data.a_ops = &cramfs_aops; } else { - inode->i_size = 0; - inode->i_blocks = 0; init_special_inode(inode, inode->i_mode, old_decode_dev(cramfs_inode->size)); } diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index 3dbe2169cf3..81ae9ea3c6e 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -37,9 +37,6 @@ static struct inode *debugfs_get_inode(struct super_block *sb, int mode, dev_t d if (inode) { inode->i_mode = mode; - inode->i_uid = 0; - inode->i_gid = 0; - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; switch (mode & S_IFMT) { default: diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c index fff96e152c0..5f3231b9633 100644 --- a/fs/devpts/inode.c +++ b/fs/devpts/inode.c @@ -189,8 +189,6 @@ static int mknod_ptmx(struct super_block *sb) } inode->i_ino = 2; - inode->i_uid = inode->i_gid = 0; - inode->i_blocks = 0; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; mode = S_IFCHR|opts->ptmxmode; @@ -300,8 +298,6 @@ devpts_fill_super(struct super_block *s, void *data, int silent) goto free_fsi; inode->i_ino = 1; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; - inode->i_blocks = 0; - inode->i_uid = inode->i_gid = 0; inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR; inode->i_op = &simple_dir_inode_operations; inode->i_fop = &simple_dir_operations; diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 7d479ce3ace..0ab0c6f5f43 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -506,7 +506,6 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb, uid_t uid, inode->i_mode = mode; inode->i_uid = uid; inode->i_gid = gid; - inode->i_blocks = 0; inode->i_mapping->a_ops = &hugetlbfs_aops; inode->i_mapping->backing_dev_info =&hugetlbfs_backing_dev_info; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; diff --git a/fs/inode.c b/fs/inode.c index 7de1cda9248..bd48e5e6d3e 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -131,6 +131,8 @@ struct inode *inode_init_always(struct super_block *sb, struct inode *inode) inode->i_op = &empty_iops; inode->i_fop = &empty_fops; inode->i_nlink = 1; + inode->i_uid = 0; + inode->i_gid = 0; atomic_set(&inode->i_writecount, 0); inode->i_size = 0; inode->i_blocks = 0; diff --git a/fs/libfs.c b/fs/libfs.c index e960a832190..7de05f7ce74 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -231,7 +231,6 @@ int get_sb_pseudo(struct file_system_type *fs_type, char *name, */ root->i_ino = 1; root->i_mode = S_IFDIR | S_IRUSR | S_IWUSR; - root->i_uid = root->i_gid = 0; root->i_atime = root->i_mtime = root->i_ctime = CURRENT_TIME; dentry = d_alloc(NULL, &d_name); if (!dentry) { @@ -436,8 +435,6 @@ int simple_fill_super(struct super_block *s, int magic, struct tree_descr *files */ inode->i_ino = 1; inode->i_mode = S_IFDIR | 0755; - inode->i_uid = inode->i_gid = 0; - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; inode->i_op = &simple_dir_inode_operations; inode->i_fop = &simple_dir_operations; @@ -464,8 +461,6 @@ int simple_fill_super(struct super_block *s, int magic, struct tree_descr *files if (!inode) goto out; inode->i_mode = S_IFREG | files->mode; - inode->i_uid = inode->i_gid = 0; - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; inode->i_fop = files->ops; inode->i_ino = i; diff --git a/fs/ocfs2/dlm/dlmfs.c b/fs/ocfs2/dlm/dlmfs.c index 6f7a77d5402..1c9efb406a9 100644 --- a/fs/ocfs2/dlm/dlmfs.c +++ b/fs/ocfs2/dlm/dlmfs.c @@ -341,7 +341,6 @@ static struct inode *dlmfs_get_root_inode(struct super_block *sb) inode->i_mode = mode; inode->i_uid = current_fsuid(); inode->i_gid = current_fsgid(); - inode->i_blocks = 0; inode->i_mapping->backing_dev_info = &dlmfs_backing_dev_info; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; inc_nlink(inode); @@ -367,7 +366,6 @@ static struct inode *dlmfs_get_inode(struct inode *parent, inode->i_mode = mode; inode->i_uid = current_fsuid(); inode->i_gid = current_fsgid(); - inode->i_blocks = 0; inode->i_mapping->backing_dev_info = &dlmfs_backing_dev_info; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; diff --git a/fs/omfs/inode.c b/fs/omfs/inode.c index 6afe57c84f8..633e9dc972b 100644 --- a/fs/omfs/inode.c +++ b/fs/omfs/inode.c @@ -39,7 +39,6 @@ struct inode *omfs_new_inode(struct inode *dir, int mode) inode->i_mode = mode; inode->i_uid = current_fsuid(); inode->i_gid = current_fsgid(); - inode->i_blocks = 0; inode->i_mapping->a_ops = &omfs_aops; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index d41bdc784de..ffcd04f0012 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -256,9 +256,6 @@ found: break; } - inode->i_gid = 0; - inode->i_uid = 0; - d_add(dentry, inode); return NULL; } diff --git a/fs/proc/base.c b/fs/proc/base.c index cad92c1ac2b..10fd5223d60 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1426,8 +1426,6 @@ static struct inode *proc_pid_make_inode(struct super_block * sb, struct task_st if (!ei->pid) goto out_unlock; - inode->i_uid = 0; - inode->i_gid = 0; if (task_dumpable(task)) { rcu_read_lock(); cred = __task_cred(task); @@ -2349,8 +2347,6 @@ static struct dentry *proc_base_instantiate(struct inode *dir, if (!ei->pid) goto out_iput; - inode->i_uid = 0; - inode->i_gid = 0; inode->i_mode = p->mode; if (S_ISDIR(inode->i_mode)) inode->i_nlink = 2; diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index 06ed10b7da9..94fcfff6863 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -31,7 +31,6 @@ static struct inode *proc_sys_make_inode(struct super_block *sb, inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; inode->i_flags |= S_PRIVATE; /* tell selinux to ignore this inode */ inode->i_mode = table->mode; - inode->i_uid = inode->i_gid = 0; if (!table->child) { inode->i_mode |= S_IFREG; inode->i_op = &proc_sys_inode_operations; diff --git a/fs/ramfs/inode.c b/fs/ramfs/inode.c index a83a3518ae3..b7e6ac706b8 100644 --- a/fs/ramfs/inode.c +++ b/fs/ramfs/inode.c @@ -57,7 +57,6 @@ struct inode *ramfs_get_inode(struct super_block *sb, int mode, dev_t dev) inode->i_mode = mode; inode->i_uid = current_fsuid(); inode->i_gid = current_fsgid(); - inode->i_blocks = 0; inode->i_mapping->a_ops = &ramfs_aops; inode->i_mapping->backing_dev_info = &ramfs_backing_dev_info; mapping_set_gfp_mask(inode->i_mapping, GFP_HIGHUSER); diff --git a/fs/romfs/inode.c b/fs/romfs/inode.c index 60d2f822e87..c97d4c93171 100644 --- a/fs/romfs/inode.c +++ b/fs/romfs/inode.c @@ -524,7 +524,6 @@ romfs_iget(struct super_block *sb, unsigned long ino) i->i_size = be32_to_cpu(ri.size); i->i_mtime.tv_sec = i->i_atime.tv_sec = i->i_ctime.tv_sec = 0; i->i_mtime.tv_nsec = i->i_atime.tv_nsec = i->i_ctime.tv_nsec = 0; - i->i_uid = i->i_gid = 0; /* Precalculate the data offset */ ino = romfs_strnlen(i, ino+ROMFH_SIZE, ROMFS_MAXFN); diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c index eb53c632f85..dfa3d94cfc7 100644 --- a/fs/sysfs/inode.c +++ b/fs/sysfs/inode.c @@ -107,8 +107,6 @@ int sysfs_setattr(struct dentry * dentry, struct iattr * iattr) static inline void set_default_inode_attr(struct inode * inode, mode_t mode) { inode->i_mode = mode; - inode->i_uid = 0; - inode->i_gid = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; } @@ -149,7 +147,6 @@ static void sysfs_init_inode(struct sysfs_dirent *sd, struct inode *inode) { struct bin_attribute *bin_attr; - inode->i_blocks = 0; inode->i_mapping->a_ops = &sysfs_aops; inode->i_mapping->backing_dev_info = &sysfs_backing_dev_info; inode->i_op = &sysfs_inode_operations; -- cgit v1.2.3 From 6110e3abbff8b785907d4db50240e63c1be726e3 Mon Sep 17 00:00:00 2001 From: Eric Paris Date: Wed, 17 Dec 2008 13:53:20 -0500 Subject: sys_execve and sys_uselib do not call into fsnotify sys_execve and sys_uselib do not call into fsnotify so inotify does not get open events for these types of syscalls. This patch simply makes the requisite fsnotify calls. Signed-off-by: Eric Paris Signed-off-by: Al Viro --- fs/exec.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'fs') diff --git a/fs/exec.c b/fs/exec.c index 3ef9cf9b187..9c33f542dc7 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -51,6 +51,7 @@ #include #include #include +#include #include #include @@ -132,6 +133,8 @@ asmlinkage long sys_uselib(const char __user * library) if (IS_ERR(file)) goto out; + fsnotify_open(file->f_path.dentry); + error = -ENOEXEC; if(file->f_op) { struct linux_binfmt * fmt; @@ -684,6 +687,8 @@ struct file *open_exec(const char *name) if (IS_ERR(file)) return file; + fsnotify_open(file->f_path.dentry); + err = deny_write_access(file); if (err) { fput(file); -- cgit v1.2.3 From 4c728ef583b3d82266584da5cb068294c09df31e Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 22 Dec 2008 21:11:15 +0100 Subject: add a vfs_fsync helper Fsync currently has a fdatawrite/fdatawait pair around the method call, and a mutex_lock/unlock of the inode mutex. All callers of fsync have to duplicate this, but we have a few and most of them don't quite get it right. This patch adds a new vfs_fsync that takes care of this. It's a little more complicated as usual as ->fsync might get a NULL file pointer and just a dentry from nfsd, but otherwise gets afile and we want to take the mapping and file operations from it when it is there. Notes on the fsync callers: - ecryptfs wasn't calling filemap_fdatawrite / filemap_fdatawait on the lower file - coda wasn't calling filemap_fdatawrite / filemap_fdatawait on the host file, and returning 0 when ->fsync was missing - shm wasn't calling either filemap_fdatawrite / filemap_fdatawait nor taking i_mutex. Now given that shared memory doesn't have disk backing not doing anything in fsync seems fine and I left it out of the vfs_fsync conversion for now, but in that case we might just not pass it through to the lower file at all but just call the no-op simple_sync_file directly. [and now actually export vfs_fsync] Signed-off-by: Christoph Hellwig Signed-off-by: Al Viro --- fs/coda/file.c | 12 ++---------- fs/ecryptfs/file.c | 15 +++------------ fs/nfsd/vfs.c | 35 +++-------------------------------- fs/sync.c | 48 +++++++++++++++++++++++++++++++++++++----------- 4 files changed, 45 insertions(+), 65 deletions(-) (limited to 'fs') diff --git a/fs/coda/file.c b/fs/coda/file.c index 466303db2df..6a347fbc998 100644 --- a/fs/coda/file.c +++ b/fs/coda/file.c @@ -201,8 +201,7 @@ int coda_release(struct inode *coda_inode, struct file *coda_file) int coda_fsync(struct file *coda_file, struct dentry *coda_dentry, int datasync) { struct file *host_file; - struct dentry *host_dentry; - struct inode *host_inode, *coda_inode = coda_dentry->d_inode; + struct inode *coda_inode = coda_dentry->d_inode; struct coda_file_info *cfi; int err = 0; @@ -214,14 +213,7 @@ int coda_fsync(struct file *coda_file, struct dentry *coda_dentry, int datasync) BUG_ON(!cfi || cfi->cfi_magic != CODA_MAGIC); host_file = cfi->cfi_container; - if (host_file->f_op && host_file->f_op->fsync) { - host_dentry = host_file->f_path.dentry; - host_inode = host_dentry->d_inode; - mutex_lock(&host_inode->i_mutex); - err = host_file->f_op->fsync(host_file, host_dentry, datasync); - mutex_unlock(&host_inode->i_mutex); - } - + err = vfs_fsync(host_file, host_file->f_path.dentry, datasync); if ( !err && !datasync ) { lock_kernel(); err = venus_fsync(coda_inode->i_sb, coda_i2f(coda_inode)); diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index eb3dc4c7ac0..71383437122 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -275,18 +275,9 @@ static int ecryptfs_release(struct inode *inode, struct file *file) static int ecryptfs_fsync(struct file *file, struct dentry *dentry, int datasync) { - struct file *lower_file = ecryptfs_file_to_lower(file); - struct dentry *lower_dentry = ecryptfs_dentry_to_lower(dentry); - struct inode *lower_inode = lower_dentry->d_inode; - int rc = -EINVAL; - - if (lower_inode->i_fop->fsync) { - mutex_lock(&lower_inode->i_mutex); - rc = lower_inode->i_fop->fsync(lower_file, lower_dentry, - datasync); - mutex_unlock(&lower_inode->i_mutex); - } - return rc; + return vfs_fsync(ecryptfs_file_to_lower(file), + ecryptfs_dentry_to_lower(dentry), + datasync); } static int ecryptfs_fasync(int fd, struct file *file, int flag) diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 5245a396500..44aa92aba89 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -744,45 +744,16 @@ nfsd_close(struct file *filp) fput(filp); } -/* - * Sync a file - * As this calls fsync (not fdatasync) there is no need for a write_inode - * after it. - */ -static inline int nfsd_dosync(struct file *filp, struct dentry *dp, - const struct file_operations *fop) -{ - struct inode *inode = dp->d_inode; - int (*fsync) (struct file *, struct dentry *, int); - int err; - - 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 err; -} - - static int nfsd_sync(struct file *filp) { - int err; - struct inode *inode = filp->f_path.dentry->d_inode; - dprintk("nfsd: sync file %s\n", filp->f_path.dentry->d_name.name); - mutex_lock(&inode->i_mutex); - err=nfsd_dosync(filp, filp->f_path.dentry, filp->f_op); - mutex_unlock(&inode->i_mutex); - - return err; + return vfs_fsync(filp, filp->f_path.dentry, 0); } int -nfsd_sync_dir(struct dentry *dp) +nfsd_sync_dir(struct dentry *dentry) { - return nfsd_dosync(NULL, dp, dp->d_inode->i_fop); + return vfs_fsync(NULL, dentry, 0); } /* diff --git a/fs/sync.c b/fs/sync.c index 2967562d416..0921d6d4b5e 100644 --- a/fs/sync.c +++ b/fs/sync.c @@ -75,14 +75,39 @@ int file_fsync(struct file *filp, struct dentry *dentry, int datasync) return ret; } -long do_fsync(struct file *file, int datasync) +/** + * vfs_fsync - perform a fsync or fdatasync on a file + * @file: file to sync + * @dentry: dentry of @file + * @data: only perform a fdatasync operation + * + * Write back data and metadata for @file to disk. If @datasync is + * set only metadata needed to access modified file data is written. + * + * In case this function is called from nfsd @file may be %NULL and + * only @dentry is set. This can only happen when the filesystem + * implements the export_operations API. + */ +int vfs_fsync(struct file *file, struct dentry *dentry, int datasync) { - int ret; - int err; - struct address_space *mapping = file->f_mapping; + const struct file_operations *fop; + struct address_space *mapping; + int err, ret; + + /* + * Get mapping and operations from the file in case we have + * as file, or get the default values for them in case we + * don't have a struct file available. Damn nfsd.. + */ + if (file) { + mapping = file->f_mapping; + fop = file->f_op; + } else { + mapping = dentry->d_inode->i_mapping; + fop = dentry->d_inode->i_fop; + } - if (!file->f_op || !file->f_op->fsync) { - /* Why? We can still call filemap_fdatawrite */ + if (!fop || !fop->fsync) { ret = -EINVAL; goto out; } @@ -94,7 +119,7 @@ long do_fsync(struct file *file, int datasync) * livelocks in fsync_buffers_list(). */ mutex_lock(&mapping->host->i_mutex); - err = file->f_op->fsync(file, file->f_path.dentry, datasync); + err = fop->fsync(file, dentry, datasync); if (!ret) ret = err; mutex_unlock(&mapping->host->i_mutex); @@ -104,15 +129,16 @@ long do_fsync(struct file *file, int datasync) out: return ret; } +EXPORT_SYMBOL(vfs_fsync); -static long __do_fsync(unsigned int fd, int datasync) +static int do_fsync(unsigned int fd, int datasync) { struct file *file; int ret = -EBADF; file = fget(fd); if (file) { - ret = do_fsync(file, datasync); + ret = vfs_fsync(file, file->f_path.dentry, datasync); fput(file); } return ret; @@ -120,12 +146,12 @@ static long __do_fsync(unsigned int fd, int datasync) asmlinkage long sys_fsync(unsigned int fd) { - return __do_fsync(fd, 0); + return do_fsync(fd, 0); } asmlinkage long sys_fdatasync(unsigned int fd) { - return __do_fsync(fd, 1); + return do_fsync(fd, 1); } /* -- cgit v1.2.3 From d8e9650dff48055057253ca30933605bd7d0733b Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Thu, 25 Dec 2008 13:32:15 +0800 Subject: vfs: remove duplicate code in get_fs_type() save 14 bytes: text data bss dec hex filename 1354 32 4 1390 56e fs/filesystems.o.before text data bss dec hex filename 1340 32 4 1376 560 fs/filesystems.o Signed-off-by: Li Zefan Signed-off-by: Al Viro --- fs/filesystems.c | 23 +++++++++++++---------- 1 file changed, 13 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/filesystems.c b/fs/filesystems.c index d0e20ced62d..d488dcd7f2b 100644 --- a/fs/filesystems.c +++ b/fs/filesystems.c @@ -253,24 +253,27 @@ static int __init proc_filesystems_init(void) module_init(proc_filesystems_init); #endif -struct file_system_type *get_fs_type(const char *name) +static struct file_system_type *__get_fs_type(const char *name, int len) { struct file_system_type *fs; - const char *dot = strchr(name, '.'); - unsigned len = dot ? dot - name : strlen(name); read_lock(&file_systems_lock); fs = *(find_filesystem(name, len)); if (fs && !try_module_get(fs->owner)) fs = NULL; read_unlock(&file_systems_lock); - if (!fs && (request_module("%.*s", len, name) == 0)) { - read_lock(&file_systems_lock); - fs = *(find_filesystem(name, len)); - if (fs && !try_module_get(fs->owner)) - fs = NULL; - read_unlock(&file_systems_lock); - } + return fs; +} + +struct file_system_type *get_fs_type(const char *name) +{ + struct file_system_type *fs; + const char *dot = strchr(name, '.'); + int len = dot ? dot - name : strlen(name); + + fs = __get_fs_type(name, len); + if (!fs && (request_module("%.*s", len, name) == 0)) + fs = __get_fs_type(name, len); if (dot && fs && !(fs->fs_flags & FS_HAS_SUBTYPE)) { put_filesystem(fs); -- cgit v1.2.3 From 5b45d96bf963afeb931a75faf02fb424e446e5a9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 29 Dec 2008 07:40:31 -0500 Subject: fix the treatment of jfs special inodes We used to put them on a single list, without any locking. Racy. Signed-off-by: Al Viro --- fs/jfs/jfs_imap.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c index d6363d8309d..0f94381ca6d 100644 --- a/fs/jfs/jfs_imap.c +++ b/fs/jfs/jfs_imap.c @@ -58,9 +58,9 @@ /* * __mark_inode_dirty expects inodes to be hashed. Since we don't want - * special inodes in the fileset inode space, we hash them to a dummy head + * special inodes in the fileset inode space, we make them appear hashed, + * but do not put on any lists. */ -static HLIST_HEAD(aggregate_hash); /* * imap locks @@ -496,7 +496,11 @@ struct inode *diReadSpecial(struct super_block *sb, ino_t inum, int secondary) /* release the page */ release_metapage(mp); - hlist_add_head(&ip->i_hash, &aggregate_hash); + /* + * that will look hashed, but won't be on any list; hlist_del() + * will work fine and require no locking. + */ + ip->i_hash.pprev = &ip->i_hash.next; return (ip); } -- cgit v1.2.3 From 2f1169e2dc0c70e213f79ada88a10912cc2fbe94 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 2 Jan 2009 08:16:51 -0500 Subject: fix breakage in reiserfs_new_inode() now that we use ih.key earlier, we need to do all its setup early enough Signed-off-by: Al Viro --- fs/reiserfs/inode.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 145c2d3e5e0..1306d4f0f44 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -1782,6 +1782,12 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th, goto out_bad_inode; } args.objectid = inode->i_ino = le32_to_cpu(ih.ih_key.k_objectid); + if (old_format_only(sb)) + make_le_item_head(&ih, NULL, KEY_FORMAT_3_5, SD_OFFSET, + TYPE_STAT_DATA, SD_V1_SIZE, MAX_US_INT); + else + make_le_item_head(&ih, NULL, KEY_FORMAT_3_6, SD_OFFSET, + TYPE_STAT_DATA, SD_SIZE, MAX_US_INT); memcpy(INODE_PKEY(inode), &(ih.ih_key), KEY_SIZE); args.dirid = le32_to_cpu(ih.ih_key.k_dir_id); if (insert_inode_locked4(inode, args.objectid, @@ -1834,13 +1840,6 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th, reiserfs_init_acl_default(inode); reiserfs_init_xattr_rwsem(inode); - if (old_format_only(sb)) - make_le_item_head(&ih, NULL, KEY_FORMAT_3_5, SD_OFFSET, - TYPE_STAT_DATA, SD_V1_SIZE, MAX_US_INT); - else - make_le_item_head(&ih, NULL, KEY_FORMAT_3_6, SD_OFFSET, - TYPE_STAT_DATA, SD_SIZE, MAX_US_INT); - /* key to search for correct place for new stat data */ _make_cpu_key(&key, KEY_FORMAT_3_6, le32_to_cpu(ih.ih_key.k_dir_id), le32_to_cpu(ih.ih_key.k_objectid), SD_OFFSET, -- cgit v1.2.3 From 4ae8978cf92a96257cd8998a49e781be83571d64 Mon Sep 17 00:00:00 2001 From: Michael Kerrisk Date: Mon, 5 Jan 2009 07:19:16 -0500 Subject: inotify: fix type errors in interfaces The problems lie in the types used for some inotify interfaces, both at the kernel level and at the glibc level. This mail addresses the kernel problem. I will follow up with some suggestions for glibc changes. For the sys_inotify_rm_watch() interface, the type of the 'wd' argument is currently 'u32', it should be '__s32' . That is Robert's suggestion, and is consistent with the other declarations of watch descriptors in the kernel source, in particular, the inotify_event structure in include/linux/inotify.h: struct inotify_event { __s32 wd; /* watch descriptor */ __u32 mask; /* watch mask */ __u32 cookie; /* cookie to synchronize two events */ __u32 len; /* length (including nulls) of name */ char name[0]; /* stub for possible name */ }; The patch makes the changes needed for inotify_rm_watch(). Signed-off-by: Michael Kerrisk Cc: Robert Love Cc: Vegard Nossum Cc: Ulrich Drepper Signed-off-by: Andrew Morton Signed-off-by: Al Viro --- fs/notify/inotify/inotify_user.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c index 400f8064a54..81b8644b013 100644 --- a/fs/notify/inotify/inotify_user.c +++ b/fs/notify/inotify/inotify_user.c @@ -704,7 +704,7 @@ fput_and_out: return ret; } -asmlinkage long sys_inotify_rm_watch(int fd, u32 wd) +asmlinkage long sys_inotify_rm_watch(int fd, __s32 wd) { struct file *filp; struct inotify_device *dev; -- cgit v1.2.3