From 123df2944c436c80640c4281c5bc9c7950b18687 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 25 Dec 2009 04:57:57 -0500 Subject: Lose the new_name argument of fsnotify_move() it's always new_dentry->d_name.name Signed-off-by: Al Viro --- include/linux/fsnotify.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h index 936f9aa8bb97..2d755c49c324 100644 --- a/include/linux/fsnotify.h +++ b/include/linux/fsnotify.h @@ -65,7 +65,7 @@ static inline void fsnotify_link_count(struct inode *inode) * fsnotify_move - file old_name at old_dir was moved to new_name at new_dir */ static inline void fsnotify_move(struct inode *old_dir, struct inode *new_dir, - const char *old_name, const char *new_name, + const char *old_name, int isdir, struct inode *target, struct dentry *moved) { struct inode *source = moved->d_inode; @@ -73,6 +73,7 @@ static inline void fsnotify_move(struct inode *old_dir, struct inode *new_dir, u32 fs_cookie = fsnotify_get_cookie(); __u32 old_dir_mask = (FS_EVENT_ON_CHILD | FS_MOVED_FROM); __u32 new_dir_mask = (FS_EVENT_ON_CHILD | FS_MOVED_TO); + const char *new_name = moved->d_name.name; if (old_dir == new_dir) old_dir_mask |= FS_DN_RENAME; -- cgit v1.2.3-71-gd317 From cccc6bba3f771ef29b33e4f79e70ebc3dba245b0 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 25 Dec 2009 05:07:33 -0500 Subject: Lose the first argument of audit_inode_child() it's always equal to ->d_name.name of the second argument Signed-off-by: Al Viro --- fs/namei.c | 2 +- include/linux/audit.h | 11 +++++------ include/linux/fsnotify.h | 8 ++++---- kernel/auditsc.c | 7 ++----- 4 files changed, 12 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/fs/namei.c b/fs/namei.c index f69df876fac3..865282f8e012 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1337,7 +1337,7 @@ static int may_delete(struct inode *dir,struct dentry *victim,int isdir) return -ENOENT; BUG_ON(victim->d_parent->d_inode != dir); - audit_inode_child(victim->d_name.name, victim, dir); + audit_inode_child(victim, dir); error = inode_permission(dir, MAY_WRITE | MAY_EXEC); if (error) diff --git a/include/linux/audit.h b/include/linux/audit.h index 3c7a358241a7..f391d45c8aea 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -424,7 +424,7 @@ extern void audit_syscall_exit(int failed, long return_code); extern void __audit_getname(const char *name); extern void audit_putname(const char *name); extern void __audit_inode(const char *name, const struct dentry *dentry); -extern void __audit_inode_child(const char *dname, const struct dentry *dentry, +extern void __audit_inode_child(const struct dentry *dentry, const struct inode *parent); extern void __audit_ptrace(struct task_struct *t); @@ -442,11 +442,10 @@ static inline void audit_inode(const char *name, const struct dentry *dentry) { if (unlikely(!audit_dummy_context())) __audit_inode(name, dentry); } -static inline void audit_inode_child(const char *dname, - const struct dentry *dentry, +static inline void audit_inode_child(const struct dentry *dentry, const struct inode *parent) { if (unlikely(!audit_dummy_context())) - __audit_inode_child(dname, dentry, parent); + __audit_inode_child(dentry, parent); } void audit_core_dumps(long signr); @@ -544,9 +543,9 @@ extern int audit_signals; #define audit_getname(n) do { ; } while (0) #define audit_putname(n) do { ; } while (0) #define __audit_inode(n,d) do { ; } while (0) -#define __audit_inode_child(d,i,p) do { ; } while (0) +#define __audit_inode_child(i,p) do { ; } while (0) #define audit_inode(n,d) do { ; } while (0) -#define audit_inode_child(d,i,p) do { ; } while (0) +#define audit_inode_child(i,p) do { ; } while (0) #define audit_core_dumps(i) do { ; } while (0) #define auditsc_get_stamp(c,t,s) (0) #define audit_get_loginuid(t) (-1) diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h index 2d755c49c324..df8fd9a3b214 100644 --- a/include/linux/fsnotify.h +++ b/include/linux/fsnotify.h @@ -104,7 +104,7 @@ static inline void fsnotify_move(struct inode *old_dir, struct inode *new_dir, inotify_inode_queue_event(source, IN_MOVE_SELF, 0, NULL, NULL); fsnotify(source, FS_MOVE_SELF, moved->d_inode, FSNOTIFY_EVENT_INODE, NULL, 0); } - audit_inode_child(new_name, moved, new_dir); + audit_inode_child(moved, new_dir); } /* @@ -147,7 +147,7 @@ static inline void fsnotify_create(struct inode *inode, struct dentry *dentry) { inotify_inode_queue_event(inode, IN_CREATE, 0, dentry->d_name.name, dentry->d_inode); - audit_inode_child(dentry->d_name.name, dentry, inode); + audit_inode_child(dentry, inode); fsnotify(inode, FS_CREATE, dentry->d_inode, FSNOTIFY_EVENT_INODE, dentry->d_name.name, 0); } @@ -162,7 +162,7 @@ static inline void fsnotify_link(struct inode *dir, struct inode *inode, struct inotify_inode_queue_event(dir, IN_CREATE, 0, new_dentry->d_name.name, inode); fsnotify_link_count(inode); - audit_inode_child(new_dentry->d_name.name, new_dentry, dir); + audit_inode_child(new_dentry, dir); fsnotify(dir, FS_CREATE, inode, FSNOTIFY_EVENT_INODE, new_dentry->d_name.name, 0); } @@ -176,7 +176,7 @@ static inline void fsnotify_mkdir(struct inode *inode, struct dentry *dentry) struct inode *d_inode = dentry->d_inode; inotify_inode_queue_event(inode, mask, 0, dentry->d_name.name, d_inode); - audit_inode_child(dentry->d_name.name, dentry, inode); + audit_inode_child(dentry, inode); fsnotify(inode, mask, d_inode, FSNOTIFY_EVENT_INODE, dentry->d_name.name, 0); } diff --git a/kernel/auditsc.c b/kernel/auditsc.c index fc0f928167e7..f3a461c0970a 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -1988,7 +1988,6 @@ void __audit_inode(const char *name, const struct dentry *dentry) /** * audit_inode_child - collect inode info for created/removed objects - * @dname: inode's dentry name * @dentry: dentry being audited * @parent: inode of dentry parent * @@ -2000,13 +1999,14 @@ void __audit_inode(const char *name, const struct dentry *dentry) * must be hooked prior, in order to capture the target inode during * unsuccessful attempts. */ -void __audit_inode_child(const char *dname, const struct dentry *dentry, +void __audit_inode_child(const struct dentry *dentry, const struct inode *parent) { int idx; struct audit_context *context = current->audit_context; const char *found_parent = NULL, *found_child = NULL; const struct inode *inode = dentry->d_inode; + const char *dname = dentry->d_name.name; int dirlen = 0; if (!context->in_syscall) @@ -2014,9 +2014,6 @@ void __audit_inode_child(const char *dname, const struct dentry *dentry, if (inode) handle_one(inode); - /* determine matching parent */ - if (!dname) - goto add_names; /* parent is more likely, look for it first */ for (idx = 0; idx < context->name_count; idx++) { -- cgit v1.2.3-71-gd317 From c44dcc56d2b5c79ba3063d20f76e5347e2e418f6 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 11 Feb 2010 02:24:46 -0500 Subject: switch inotify_user to anon_inode Signed-off-by: Al Viro --- fs/notify/inotify/inotify_user.c | 59 ++++------------------------------------ include/linux/magic.h | 1 - 2 files changed, 6 insertions(+), 54 deletions(-) (limited to 'include/linux') diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c index a94e8bd8eb1f..472cdf29ef82 100644 --- a/fs/notify/inotify/inotify_user.c +++ b/fs/notify/inotify/inotify_user.c @@ -29,14 +29,12 @@ #include /* module_init */ #include #include /* roundup() */ -#include /* superblock magic number */ -#include /* mntget */ #include /* LOOKUP_FOLLOW */ -#include /* struct path */ #include /* struct user */ #include /* struct kmem_cache */ #include #include +#include #include #include #include @@ -45,8 +43,6 @@ #include -static struct vfsmount *inotify_mnt __read_mostly; - /* these are configurable via /proc/sys/fs/inotify/ */ static int inotify_max_user_instances __read_mostly; static int inotify_max_queued_events __read_mostly; @@ -645,9 +641,7 @@ SYSCALL_DEFINE1(inotify_init1, int, flags) { struct fsnotify_group *group; struct user_struct *user; - struct file *filp; - struct path path; - int fd, ret; + int ret; /* Check the IN_* constants for consistency. */ BUILD_BUG_ON(IN_CLOEXEC != O_CLOEXEC); @@ -656,10 +650,6 @@ SYSCALL_DEFINE1(inotify_init1, int, flags) if (flags & ~(IN_CLOEXEC | IN_NONBLOCK)) return -EINVAL; - fd = get_unused_fd_flags(flags & O_CLOEXEC); - if (fd < 0) - return fd; - user = get_current_user(); if (unlikely(atomic_read(&user->inotify_devs) >= inotify_max_user_instances)) { @@ -676,27 +666,14 @@ SYSCALL_DEFINE1(inotify_init1, int, flags) atomic_inc(&user->inotify_devs); - path.mnt = inotify_mnt; - path.dentry = inotify_mnt->mnt_root; - path_get(&path); - filp = alloc_file(&path, FMODE_READ, &inotify_fops); - if (!filp) - goto Enfile; + ret = anon_inode_getfd("inotify", &inotify_fops, group, + O_RDONLY | flags); + if (ret >= 0) + return ret; - filp->f_flags = O_RDONLY | (flags & O_NONBLOCK); - filp->private_data = group; - - fd_install(fd, filp); - - return fd; - -Enfile: - ret = -ENFILE; - path_put(&path); atomic_dec(&user->inotify_devs); out_free_uid: free_uid(user); - put_unused_fd(fd); return ret; } @@ -783,20 +760,6 @@ out: return ret; } -static int -inotify_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *data, struct vfsmount *mnt) -{ - return get_sb_pseudo(fs_type, "inotify", NULL, - INOTIFYFS_SUPER_MAGIC, mnt); -} - -static struct file_system_type inotify_fs_type = { - .name = "inotifyfs", - .get_sb = inotify_get_sb, - .kill_sb = kill_anon_super, -}; - /* * inotify_user_setup - Our initialization function. Note that we cannnot return * error because we have compiled-in VFS hooks. So an (unlikely) failure here @@ -804,16 +767,6 @@ static struct file_system_type inotify_fs_type = { */ static int __init inotify_user_setup(void) { - int ret; - - ret = register_filesystem(&inotify_fs_type); - if (unlikely(ret)) - panic("inotify: register_filesystem returned %d!\n", ret); - - inotify_mnt = kern_mount(&inotify_fs_type); - if (IS_ERR(inotify_mnt)) - panic("inotify: kern_mount ret %ld!\n", PTR_ERR(inotify_mnt)); - inotify_inode_mark_cachep = KMEM_CACHE(inotify_inode_mark_entry, SLAB_PANIC); event_priv_cachep = KMEM_CACHE(inotify_event_private_data, SLAB_PANIC); diff --git a/include/linux/magic.h b/include/linux/magic.h index 76285e01b39e..eb9800f05782 100644 --- a/include/linux/magic.h +++ b/include/linux/magic.h @@ -52,7 +52,6 @@ #define CGROUP_SUPER_MAGIC 0x27e0eb #define FUTEXFS_SUPER_MAGIC 0xBAD1DEA -#define INOTIFYFS_SUPER_MAGIC 0x2BAD1DEA #define STACK_END_MAGIC 0x57AC6E9D -- cgit v1.2.3-71-gd317 From 193cf4b99113a4550598ba9e8343e591fc062e23 Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Tue, 12 Jan 2010 16:18:08 +0200 Subject: libfs: Unexport and kill simple_prepare_write Remove the EXPORT_UNUSED_SYMBOL of simple_prepare_write Collapse simple_prepare_write into it's only caller, though making it simpler and clearer to understand. Signed-off-by: Boaz Harrosh Signed-off-by: Al Viro --- fs/libfs.c | 22 ++++++---------------- include/linux/fs.h | 2 -- 2 files changed, 6 insertions(+), 18 deletions(-) (limited to 'include/linux') diff --git a/fs/libfs.c b/fs/libfs.c index cd88abdcb436..9e50bcf55857 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -338,28 +338,14 @@ int simple_readpage(struct file *file, struct page *page) return 0; } -int simple_prepare_write(struct file *file, struct page *page, - unsigned from, unsigned to) -{ - if (!PageUptodate(page)) { - if (to - from != PAGE_CACHE_SIZE) - zero_user_segments(page, - 0, from, - to, PAGE_CACHE_SIZE); - } - return 0; -} - int simple_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata) { struct page *page; pgoff_t index; - unsigned from; index = pos >> PAGE_CACHE_SHIFT; - from = pos & (PAGE_CACHE_SIZE - 1); page = grab_cache_page_write_begin(mapping, index, flags); if (!page) @@ -367,7 +353,12 @@ int simple_write_begin(struct file *file, struct address_space *mapping, *pagep = page; - return simple_prepare_write(file, page, from, from+len); + if (!PageUptodate(page) && (len != PAGE_CACHE_SIZE)) { + unsigned from = pos & (PAGE_CACHE_SIZE - 1); + + zero_user_segments(page, 0, from, from + len, PAGE_CACHE_SIZE); + } + return 0; } /** @@ -864,7 +855,6 @@ EXPORT_SYMBOL(simple_getattr); EXPORT_SYMBOL(simple_link); EXPORT_SYMBOL(simple_lookup); EXPORT_SYMBOL(simple_pin_fs); -EXPORT_UNUSED_SYMBOL(simple_prepare_write); EXPORT_SYMBOL(simple_readpage); EXPORT_SYMBOL(simple_release_fs); EXPORT_SYMBOL(simple_rename); diff --git a/include/linux/fs.h b/include/linux/fs.h index ebb1cd5bc241..2b124c825e38 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2340,8 +2340,6 @@ extern int simple_rename(struct inode *, struct dentry *, struct inode *, struct extern int simple_sync_file(struct file *, struct dentry *, int); extern int simple_empty(struct dentry *); extern int simple_readpage(struct file *file, struct page *page); -extern int simple_prepare_write(struct file *file, struct page *page, - unsigned offset, unsigned to); extern int simple_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata); -- cgit v1.2.3-71-gd317 From 270ba5f7c5dac0bfb564aa35a536fb31ad4075bd Mon Sep 17 00:00:00 2001 From: Richard Kennedy Date: Tue, 26 Jan 2010 14:12:43 +0000 Subject: fs: re-order super_block to remove 16 bytes of padding on 64bit builds re-order structure super_block to remove 16 bytes of alignment padding on 64bit builds. This shrinks the size of super_block from 712 to 696 bytes so requiring one fewer 64 byte cache lines. Signed-off-by: Richard Kennedy ----- patch against 2.6.33-rc5 compiled & tested on x86_64 AMDX2 desktop machine. I've been running with this patch applied for several weeks with no problems. regards Richard Signed-off-by: Al Viro --- include/linux/fs.h | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 2b124c825e38..aa76dae673eb 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1314,9 +1314,9 @@ extern spinlock_t sb_lock; struct super_block { struct list_head s_list; /* Keep this first */ dev_t s_dev; /* search index; _not_ kdev_t */ - unsigned long s_blocksize; - unsigned char s_blocksize_bits; unsigned char s_dirt; + unsigned char s_blocksize_bits; + unsigned long s_blocksize; loff_t s_maxbytes; /* Max file size */ struct file_system_type *s_type; const struct super_operations *s_op; @@ -1357,16 +1357,16 @@ struct super_block { void *s_fs_info; /* Filesystem private info */ fmode_t s_mode; + /* Granularity of c/m/atime in ns. + Cannot be worse than a second */ + u32 s_time_gran; + /* * The next field is for VFS *only*. No filesystems have any business * even looking at it. You had been warned. */ struct mutex s_vfs_rename_mutex; /* Kludge */ - /* Granularity of c/m/atime in ns. - Cannot be worse than a second */ - u32 s_time_gran; - /* * Filesystem subtype. If non-empty the filesystem type field * in /proc/mounts will be "type.subtype" -- cgit v1.2.3-71-gd317 From 2ecdc82ef0b03e67ce5ecee79d0d108177a704df Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 26 Jan 2010 17:27:20 +0100 Subject: kill unused invalidate_inode_pages helper No one is calling this anymore as everyone has switched to invalidate_mapping_pages long time ago. Also update a few references to it in comments. nfs has two more, but I can't easily figure what they are actually referring to, so I left them as-is. Signed-off-by: Christoph Hellwig Signed-off-by: Al Viro --- drivers/usb/gadget/f_mass_storage.c | 2 +- drivers/usb/gadget/file_storage.c | 2 +- include/linux/fs.h | 6 ------ mm/filemap.c | 2 +- 4 files changed, 3 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/drivers/usb/gadget/f_mass_storage.c b/drivers/usb/gadget/f_mass_storage.c index a37640eba434..77fcd1b697e8 100644 --- a/drivers/usb/gadget/f_mass_storage.c +++ b/drivers/usb/gadget/f_mass_storage.c @@ -1041,7 +1041,7 @@ static void invalidate_sub(struct fsg_lun *curlun) unsigned long rc; rc = invalidate_mapping_pages(inode->i_mapping, 0, -1); - VLDBG(curlun, "invalidate_inode_pages -> %ld\n", rc); + VLDBG(curlun, "invalidate_mapping_pages -> %ld\n", rc); } static int do_verify(struct fsg_common *common) diff --git a/drivers/usb/gadget/file_storage.c b/drivers/usb/gadget/file_storage.c index 29dfb0277ffb..7dcdbda49cac 100644 --- a/drivers/usb/gadget/file_storage.c +++ b/drivers/usb/gadget/file_storage.c @@ -1448,7 +1448,7 @@ static void invalidate_sub(struct fsg_lun *curlun) unsigned long rc; rc = invalidate_mapping_pages(inode->i_mapping, 0, -1); - VLDBG(curlun, "invalidate_inode_pages -> %ld\n", rc); + VLDBG(curlun, "invalidate_mapping_pages -> %ld\n", rc); } static int do_verify(struct fsg_dev *fsg) diff --git a/include/linux/fs.h b/include/linux/fs.h index aa76dae673eb..d443c9dd3caa 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2058,12 +2058,6 @@ extern int invalidate_inodes(struct super_block *); unsigned long invalidate_mapping_pages(struct address_space *mapping, pgoff_t start, pgoff_t end); -static inline unsigned long __deprecated -invalidate_inode_pages(struct address_space *mapping) -{ - return invalidate_mapping_pages(mapping, 0, ~0UL); -} - static inline void invalidate_remote_inode(struct inode *inode) { if (S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || diff --git a/mm/filemap.c b/mm/filemap.c index 698ea80f2102..148b52a5bb7e 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1117,7 +1117,7 @@ readpage: if (!PageUptodate(page)) { if (page->mapping == NULL) { /* - * invalidate_inode_pages got it + * invalidate_mapping_pages got it */ unlock_page(page); page_cache_release(page); -- cgit v1.2.3-71-gd317 From 495d6c9c6595ec7b37910dfd42634839431d21fd Mon Sep 17 00:00:00 2001 From: Valerie Aurora Date: Tue, 26 Jan 2010 14:20:47 -0500 Subject: VFS: Clean up shared mount flag propagation The handling of mount flags in set_mnt_shared() got a little tangled up during previous cleanups, with the following problems: * MNT_PNODE_MASK is defined as a literal constant when it should be a bitwise xor of other MNT_* flags * set_mnt_shared() clears and then sets MNT_SHARED (part of MNT_PNODE_MASK) * MNT_PNODE_MASK could use a comment in mount.h * MNT_PNODE_MASK is a terrible name, change to MNT_SHARED_MASK This patch fixes these problems. Signed-off-by: Al Viro --- fs/namespace.c | 2 +- fs/pnode.h | 2 +- include/linux/mount.h | 11 ++++++++++- 3 files changed, 12 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/fs/namespace.c b/fs/namespace.c index 25c1dcf9e9eb..d25d4602ab50 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1538,7 +1538,7 @@ static int do_remount(struct path *path, int flags, int mnt_flags, err = do_remount_sb(sb, flags, data, 0); if (!err) { spin_lock(&vfsmount_lock); - mnt_flags |= path->mnt->mnt_flags & MNT_PNODE_MASK; + mnt_flags |= path->mnt->mnt_flags & MNT_PROPAGATION_MASK; path->mnt->mnt_flags = mnt_flags; spin_unlock(&vfsmount_lock); } diff --git a/fs/pnode.h b/fs/pnode.h index 6c7ef3252a26..1ea4ae1efcd3 100644 --- a/fs/pnode.h +++ b/fs/pnode.h @@ -25,7 +25,7 @@ static inline void set_mnt_shared(struct vfsmount *mnt) { - mnt->mnt_flags &= ~MNT_PNODE_MASK; + mnt->mnt_flags &= ~MNT_SHARED_MASK; mnt->mnt_flags |= MNT_SHARED; } diff --git a/include/linux/mount.h b/include/linux/mount.h index 5d5275364867..375d43a5d802 100644 --- a/include/linux/mount.h +++ b/include/linux/mount.h @@ -34,7 +34,16 @@ struct mnt_namespace; #define MNT_SHARED 0x1000 /* if the vfsmount is a shared mount */ #define MNT_UNBINDABLE 0x2000 /* if the vfsmount is a unbindable mount */ -#define MNT_PNODE_MASK 0x3000 /* propagation flag mask */ +/* + * MNT_SHARED_MASK is the set of flags that should be cleared when a + * mount becomes shared. Currently, this is only the flag that says a + * mount cannot be bind mounted, since this is how we create a mount + * that shares events with another mount. If you add a new MNT_* + * flag, consider how it interacts with shared mounts. + */ +#define MNT_SHARED_MASK (MNT_UNBINDABLE) +#define MNT_PROPAGATION_MASK (MNT_SHARED | MNT_UNBINDABLE) + struct vfsmount { struct list_head mnt_hash; -- cgit v1.2.3-71-gd317 From 2096f759abcb42200a81d776f597362fd9265024 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 30 Jan 2010 13:16:21 -0500 Subject: New helper: path_is_under(path1, path2) Analog of is_subdir for vfsmount,dentry pairs, moved from audit_tree.c Signed-off-by: Al Viro --- fs/dcache.c | 24 ++++++++++++++++++++++++ include/linux/fs.h | 1 + kernel/audit_tree.c | 51 ++++++++++++--------------------------------------- 3 files changed, 37 insertions(+), 39 deletions(-) (limited to 'include/linux') diff --git a/fs/dcache.c b/fs/dcache.c index 4365998b8df4..74da947b160b 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -2191,6 +2191,30 @@ int is_subdir(struct dentry *new_dentry, struct dentry *old_dentry) return result; } +int path_is_under(struct path *path1, struct path *path2) +{ + struct vfsmount *mnt = path1->mnt; + struct dentry *dentry = path1->dentry; + int res; + spin_lock(&vfsmount_lock); + if (mnt != path2->mnt) { + for (;;) { + if (mnt->mnt_parent == mnt) { + spin_unlock(&vfsmount_lock); + return 0; + } + if (mnt->mnt_parent == path2->mnt) + break; + mnt = mnt->mnt_parent; + } + dentry = mnt->mnt_mountpoint; + } + res = is_subdir(dentry, path2->dentry); + spin_unlock(&vfsmount_lock); + return res; +} +EXPORT_SYMBOL(path_is_under); + void d_genocide(struct dentry *root) { struct dentry *this_parent = root; diff --git a/include/linux/fs.h b/include/linux/fs.h index d443c9dd3caa..8d53bc17f93f 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2126,6 +2126,7 @@ extern struct file * open_exec(const char *); /* fs/dcache.c -- generic fs support functions */ extern int is_subdir(struct dentry *, struct dentry *); +extern int path_is_under(struct path *, struct path *); extern ino_t find_inode_number(struct dentry *, struct qstr *); #include diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c index 4b05bd9479db..f09b42d9c32d 100644 --- a/kernel/audit_tree.c +++ b/kernel/audit_tree.c @@ -603,22 +603,6 @@ skip_it: mutex_unlock(&audit_filter_mutex); } -static int is_under(struct vfsmount *mnt, struct dentry *dentry, - struct path *path) -{ - if (mnt != path->mnt) { - for (;;) { - if (mnt->mnt_parent == mnt) - return 0; - if (mnt->mnt_parent == path->mnt) - break; - mnt = mnt->mnt_parent; - } - dentry = mnt->mnt_mountpoint; - } - return is_subdir(dentry, path->dentry); -} - int audit_make_tree(struct audit_krule *rule, char *pathname, u32 op) { @@ -714,29 +698,24 @@ int audit_tag_tree(char *old, char *new) { struct list_head cursor, barrier; int failed = 0; - struct path path; + struct path path1, path2; struct vfsmount *tagged; struct list_head list; - struct vfsmount *mnt; - struct dentry *dentry; int err; - err = kern_path(new, 0, &path); + err = kern_path(new, 0, &path2); if (err) return err; - tagged = collect_mounts(&path); - path_put(&path); + tagged = collect_mounts(&path2); + path_put(&path2); if (!tagged) return -ENOMEM; - err = kern_path(old, 0, &path); + err = kern_path(old, 0, &path1); if (err) { drop_collected_mounts(tagged); return err; } - mnt = mntget(path.mnt); - dentry = dget(path.dentry); - path_put(&path); list_add_tail(&list, &tagged->mnt_list); @@ -747,6 +726,7 @@ int audit_tag_tree(char *old, char *new) while (cursor.next != &tree_list) { struct audit_tree *tree; struct vfsmount *p; + int good_one = 0; tree = container_of(cursor.next, struct audit_tree, list); get_tree(tree); @@ -754,23 +734,17 @@ int audit_tag_tree(char *old, char *new) list_add(&cursor, &tree->list); mutex_unlock(&audit_filter_mutex); - err = kern_path(tree->pathname, 0, &path); - if (err) { - put_tree(tree); - mutex_lock(&audit_filter_mutex); - continue; + err = kern_path(tree->pathname, 0, &path2); + if (!err) { + good_one = path_is_under(&path1, &path2); + path_put(&path2); } - spin_lock(&vfsmount_lock); - if (!is_under(mnt, dentry, &path)) { - spin_unlock(&vfsmount_lock); - path_put(&path); + if (!good_one) { put_tree(tree); mutex_lock(&audit_filter_mutex); continue; } - spin_unlock(&vfsmount_lock); - path_put(&path); list_for_each_entry(p, &list, mnt_list) { failed = tag_chunk(p->mnt_root->d_inode, tree); @@ -820,8 +794,7 @@ int audit_tag_tree(char *old, char *new) list_del(&cursor); list_del(&list); mutex_unlock(&audit_filter_mutex); - dput(dentry); - mntput(mnt); + path_put(&path1); drop_collected_mounts(tagged); return failed; } -- cgit v1.2.3-71-gd317 From 1f707137b55764740981d022d29c622832a61880 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 30 Jan 2010 22:51:25 -0500 Subject: new helper: iterate_mounts() apply function to vfsmounts in set returned by collect_mounts(), stop if it returns non-zero. Signed-off-by: Al Viro --- fs/namespace.c | 15 +++++++++++++++ include/linux/fs.h | 3 ++- kernel/audit_tree.c | 49 ++++++++++++++++--------------------------------- 3 files changed, 33 insertions(+), 34 deletions(-) (limited to 'include/linux') diff --git a/fs/namespace.c b/fs/namespace.c index d25d4602ab50..d5906c19e08e 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1246,6 +1246,21 @@ void drop_collected_mounts(struct vfsmount *mnt) release_mounts(&umount_list); } +int iterate_mounts(int (*f)(struct vfsmount *, void *), void *arg, + struct vfsmount *root) +{ + struct vfsmount *mnt; + int res = f(root, arg); + if (res) + return res; + list_for_each_entry(mnt, &root->mnt_list, mnt_list) { + res = f(mnt, arg); + if (res) + return res; + } + return 0; +} + static void cleanup_group_ids(struct vfsmount *mnt, struct vfsmount *end) { struct vfsmount *p; diff --git a/include/linux/fs.h b/include/linux/fs.h index 8d53bc17f93f..e764f247d0ab 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1794,7 +1794,8 @@ extern int may_umount(struct vfsmount *); extern long do_mount(char *, char *, char *, unsigned long, void *); extern struct vfsmount *collect_mounts(struct path *); extern void drop_collected_mounts(struct vfsmount *); - +extern int iterate_mounts(int (*)(struct vfsmount *, void *), void *, + struct vfsmount *); extern int vfs_statfs(struct dentry *, struct kstatfs *); extern int current_umask(void); diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c index f09b42d9c32d..028e85663f27 100644 --- a/kernel/audit_tree.c +++ b/kernel/audit_tree.c @@ -548,6 +548,11 @@ int audit_remove_tree_rule(struct audit_krule *rule) return 0; } +static int compare_root(struct vfsmount *mnt, void *arg) +{ + return mnt->mnt_root->d_inode == arg; +} + void audit_trim_trees(void) { struct list_head cursor; @@ -559,7 +564,6 @@ void audit_trim_trees(void) struct path path; struct vfsmount *root_mnt; struct node *node; - struct list_head list; int err; tree = container_of(cursor.next, struct audit_tree, list); @@ -577,24 +581,16 @@ void audit_trim_trees(void) if (!root_mnt) goto skip_it; - list_add_tail(&list, &root_mnt->mnt_list); spin_lock(&hash_lock); list_for_each_entry(node, &tree->chunks, list) { - struct audit_chunk *chunk = find_chunk(node); - struct inode *inode = chunk->watch.inode; - struct vfsmount *mnt; + struct inode *inode = find_chunk(node)->watch.inode; node->index |= 1U<<31; - list_for_each_entry(mnt, &list, mnt_list) { - if (mnt->mnt_root->d_inode == inode) { - node->index &= ~(1U<<31); - break; - } - } + if (iterate_mounts(compare_root, inode, root_mnt)) + node->index &= ~(1U<<31); } spin_unlock(&hash_lock); trim_marked(tree); put_tree(tree); - list_del_init(&list); drop_collected_mounts(root_mnt); skip_it: mutex_lock(&audit_filter_mutex); @@ -622,13 +618,17 @@ void audit_put_tree(struct audit_tree *tree) put_tree(tree); } +static int tag_mount(struct vfsmount *mnt, void *arg) +{ + return tag_chunk(mnt->mnt_root->d_inode, arg); +} + /* called with audit_filter_mutex */ int audit_add_tree_rule(struct audit_krule *rule) { struct audit_tree *seed = rule->tree, *tree; struct path path; - struct vfsmount *mnt, *p; - struct list_head list; + struct vfsmount *mnt; int err; list_for_each_entry(tree, &tree_list, list) { @@ -654,16 +654,9 @@ int audit_add_tree_rule(struct audit_krule *rule) err = -ENOMEM; goto Err; } - list_add_tail(&list, &mnt->mnt_list); get_tree(tree); - list_for_each_entry(p, &list, mnt_list) { - err = tag_chunk(p->mnt_root->d_inode, tree); - if (err) - break; - } - - list_del(&list); + err = iterate_mounts(tag_mount, tree, mnt); drop_collected_mounts(mnt); if (!err) { @@ -700,7 +693,6 @@ int audit_tag_tree(char *old, char *new) int failed = 0; struct path path1, path2; struct vfsmount *tagged; - struct list_head list; int err; err = kern_path(new, 0, &path2); @@ -717,15 +709,12 @@ int audit_tag_tree(char *old, char *new) return err; } - list_add_tail(&list, &tagged->mnt_list); - mutex_lock(&audit_filter_mutex); list_add(&barrier, &tree_list); list_add(&cursor, &barrier); while (cursor.next != &tree_list) { struct audit_tree *tree; - struct vfsmount *p; int good_one = 0; tree = container_of(cursor.next, struct audit_tree, list); @@ -746,12 +735,7 @@ int audit_tag_tree(char *old, char *new) continue; } - list_for_each_entry(p, &list, mnt_list) { - failed = tag_chunk(p->mnt_root->d_inode, tree); - if (failed) - break; - } - + failed = iterate_mounts(tag_mount, tree, tagged); if (failed) { put_tree(tree); mutex_lock(&audit_filter_mutex); @@ -792,7 +776,6 @@ int audit_tag_tree(char *old, char *new) } list_del(&barrier); list_del(&cursor); - list_del(&list); mutex_unlock(&audit_filter_mutex); path_put(&path1); drop_collected_mounts(tagged); -- cgit v1.2.3-71-gd317 From 9f5596af44514f99e3a654a4f7cb813354b9e516 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 5 Feb 2010 00:40:25 -0500 Subject: take check for new events in namespace (guts of mounts_poll()) to namespace.c Signed-off-by: Al Viro --- fs/namespace.c | 15 +++++++++++++++ fs/proc/base.c | 10 ++-------- include/linux/mnt_namespace.h | 1 + 3 files changed, 18 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/fs/namespace.c b/fs/namespace.c index d5906c19e08e..970fe79d7867 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -737,6 +737,21 @@ static void m_stop(struct seq_file *m, void *v) up_read(&namespace_sem); } +int mnt_had_events(struct proc_mounts *p) +{ + struct mnt_namespace *ns = p->ns; + int res = 0; + + spin_lock(&vfsmount_lock); + if (p->event != ns->event) { + p->event = ns->event; + res = 1; + } + spin_unlock(&vfsmount_lock); + + return res; +} + struct proc_fs_info { int flag; const char *str; diff --git a/fs/proc/base.c b/fs/proc/base.c index 58324c299165..746895ddfda1 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -647,17 +647,11 @@ static int mounts_release(struct inode *inode, struct file *file) static unsigned mounts_poll(struct file *file, poll_table *wait) { struct proc_mounts *p = file->private_data; - struct mnt_namespace *ns = p->ns; unsigned res = POLLIN | POLLRDNORM; - poll_wait(file, &ns->poll, wait); - - spin_lock(&vfsmount_lock); - if (p->event != ns->event) { - p->event = ns->event; + poll_wait(file, &p->ns->poll, wait); + if (mnt_had_events(p)) res |= POLLERR | POLLPRI; - } - spin_unlock(&vfsmount_lock); return res; } diff --git a/include/linux/mnt_namespace.h b/include/linux/mnt_namespace.h index d74785c2393a..0b89efc6f215 100644 --- a/include/linux/mnt_namespace.h +++ b/include/linux/mnt_namespace.h @@ -35,6 +35,7 @@ static inline void get_mnt_ns(struct mnt_namespace *ns) extern const struct seq_operations mounts_op; extern const struct seq_operations mountinfo_op; extern const struct seq_operations mountstats_op; +extern int mnt_had_events(struct proc_mounts *); #endif #endif -- cgit v1.2.3-71-gd317 From 47cd813f2984569570021ce3d34cdf9cb20aa6a2 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 5 Feb 2010 02:01:14 -0500 Subject: Take vfsmount_lock to fs/internal.h no more users left outside of fs/*.c (and very few outside of fs/namespace.c, actually) Signed-off-by: Al Viro --- fs/internal.h | 2 ++ include/linux/mount.h | 1 - 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/fs/internal.h b/fs/internal.h index e96a1667d749..8a03a5447bdf 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -70,6 +70,8 @@ extern struct vfsmount *copy_tree(struct vfsmount *, struct dentry *, int); extern void __init mnt_init(void); +extern spinlock_t vfsmount_lock; + /* * fs_struct.c */ diff --git a/include/linux/mount.h b/include/linux/mount.h index 375d43a5d802..163896137ab5 100644 --- a/include/linux/mount.h +++ b/include/linux/mount.h @@ -132,7 +132,6 @@ extern int do_add_mount(struct vfsmount *newmnt, struct path *path, extern void mark_mounts_for_expiry(struct list_head *mounts); -extern spinlock_t vfsmount_lock; extern dev_t name_to_dev_t(char *name); #endif /* _LINUX_MOUNT_H */ -- cgit v1.2.3-71-gd317 From 8089352a13b785d4e0df63d87bd2b71c76bb9aee Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 5 Feb 2010 09:30:46 -0500 Subject: Mirror MS_KERNMOUNT in ->mnt_flags Signed-off-by: Al Viro --- fs/namespace.c | 2 +- fs/super.c | 3 +++ include/linux/mount.h | 2 ++ 3 files changed, 6 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/fs/namespace.c b/fs/namespace.c index b0b15cc2117c..ffa3843404e0 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1701,7 +1701,7 @@ int do_add_mount(struct vfsmount *newmnt, struct path *path, { int err; - mnt_flags &= ~(MNT_SHARED | MNT_WRITE_HOLD); + mnt_flags &= ~(MNT_SHARED | MNT_WRITE_HOLD | MNT_INTERNAL); down_write(&namespace_sem); /* Something was mounted here while we slept */ diff --git a/fs/super.c b/fs/super.c index 903896ec7c73..f35ac6022109 100644 --- a/fs/super.c +++ b/fs/super.c @@ -937,6 +937,9 @@ vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void if (!mnt) goto out; + if (flags & MS_KERNMOUNT) + mnt->mnt_flags = MNT_INTERNAL; + if (data && !(type->fs_flags & FS_BINARY_MOUNTDATA)) { secdata = alloc_secdata(); if (!secdata) diff --git a/include/linux/mount.h b/include/linux/mount.h index 163896137ab5..ca726ebf50a3 100644 --- a/include/linux/mount.h +++ b/include/linux/mount.h @@ -45,6 +45,8 @@ struct mnt_namespace; #define MNT_PROPAGATION_MASK (MNT_SHARED | MNT_UNBINDABLE) +#define MNT_INTERNAL 0x4000 + struct vfsmount { struct list_head mnt_hash; struct vfsmount *mnt_parent; /* fs we are mounted on */ -- cgit v1.2.3-71-gd317 From db1f05bb85d7966b9176e293f3ceead1cb8b5d79 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 10 Feb 2010 12:15:53 +0100 Subject: vfs: add NOFOLLOW flag to umount(2) Add a new UMOUNT_NOFOLLOW flag to umount(2). This is needed to prevent symlink attacks in unprivileged unmounts (fuse, samba, ncpfs). Additionally, return -EINVAL if an unknown flag is used (and specify an explicitly unused flag: UMOUNT_UNUSED). This makes it possible for the caller to determine if a flag is supported or not. CC: Eugene Teo CC: Michael Kerrisk Signed-off-by: Miklos Szeredi Signed-off-by: Al Viro --- fs/namespace.c | 9 ++++++++- include/linux/fs.h | 2 ++ 2 files changed, 10 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/fs/namespace.c b/fs/namespace.c index ffa3843404e0..8174c8ab5c70 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1136,8 +1136,15 @@ SYSCALL_DEFINE2(umount, char __user *, name, int, flags) { struct path path; int retval; + int lookup_flags = 0; - retval = user_path(name, &path); + if (flags & ~(MNT_FORCE | MNT_DETACH | MNT_EXPIRE | UMOUNT_NOFOLLOW)) + return -EINVAL; + + if (!(flags & UMOUNT_NOFOLLOW)) + lookup_flags |= LOOKUP_FOLLOW; + + retval = user_path_at(AT_FDCWD, name, lookup_flags, &path); if (retval) goto out; retval = -EINVAL; diff --git a/include/linux/fs.h b/include/linux/fs.h index e764f247d0ab..5b3182c7eb5f 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1305,6 +1305,8 @@ extern int send_sigurg(struct fown_struct *fown); #define MNT_FORCE 0x00000001 /* Attempt to forcibily umount */ #define MNT_DETACH 0x00000002 /* Just detach from the tree */ #define MNT_EXPIRE 0x00000004 /* Mark for expiry */ +#define UMOUNT_NOFOLLOW 0x00000008 /* Don't follow symlink on umount */ +#define UMOUNT_UNUSED 0x80000000 /* Flag guaranteed to be unused */ extern struct list_head super_blocks; extern spinlock_t sb_lock; -- cgit v1.2.3-71-gd317