From c9c6cac0c2bdbda42e7b804838648d0bc60ddb13 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 16 Feb 2011 15:15:47 -0500 Subject: kill path_lookup() all remaining callers pass LOOKUP_PARENT to it, so flags argument can die; renamed to kern_path_parent() Signed-off-by: Al Viro --- include/linux/namei.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/namei.h b/include/linux/namei.h index f276d4fa01fc..58ce3433d4ec 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -72,7 +72,7 @@ extern int user_path_at(int, const char __user *, unsigned, struct path *); extern int kern_path(const char *, unsigned, struct path *); -extern int path_lookup(const char *, unsigned, struct nameidata *); +extern int kern_path_parent(const char *, struct nameidata *); extern int vfs_path_lookup(struct dentry *, struct vfsmount *, const char *, unsigned int, struct nameidata *); -- cgit v1.2.3 From 16c2cd7179881d5dd87779512ca5a0d657c64f62 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 22 Feb 2011 15:50:10 -0500 Subject: untangle the "need_reval_dot" mess instead of ad-hackery around need_reval_dot(), do the following: set a flag (LOOKUP_JUMPED) in the beginning of path, on absolute symlink traversal, on ".." and on procfs-style symlinks. Clear on normal components, leave unchanged on ".". Non-nested callers of link_path_walk() call handle_reval_path(), which checks that flag is set and that fs does want the final revalidate thing, then does ->d_revalidate(). In link_path_walk() all the return_reval stuff is gone. Signed-off-by: Al Viro --- fs/namei.c | 107 +++++++++++++++++++++----------------------------- include/linux/namei.h | 2 + 2 files changed, 46 insertions(+), 63 deletions(-) (limited to 'include') diff --git a/fs/namei.c b/fs/namei.c index 4521b5ff7c93..450b686e9682 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -613,19 +613,8 @@ do_revalidate_rcu(struct dentry *dentry, struct nameidata *nd) return dentry; } -static inline int need_reval_dot(struct dentry *dentry) -{ - if (likely(!(dentry->d_flags & DCACHE_OP_REVALIDATE))) - return 0; - - if (likely(!(dentry->d_sb->s_type->fs_flags & FS_REVAL_DOT))) - return 0; - - return 1; -} - /* - * force_reval_path - force revalidation of a dentry + * handle_reval_path - force revalidation of a dentry * * In some situations the path walking code will trust dentries without * revalidating them. This causes problems for filesystems that depend on @@ -639,27 +628,28 @@ static inline int need_reval_dot(struct dentry *dentry) * invalidate the dentry. It's up to the caller to handle putting references * to the path if necessary. */ -static int -force_reval_path(struct path *path, struct nameidata *nd) +static inline int handle_reval_path(struct nameidata *nd) { + struct dentry *dentry = nd->path.dentry; int status; - struct dentry *dentry = path->dentry; - /* - * only check on filesystems where it's possible for the dentry to - * become stale. - */ - if (!need_reval_dot(dentry)) + if (likely(!(nd->flags & LOOKUP_JUMPED))) + return 0; + + if (likely(!(dentry->d_flags & DCACHE_OP_REVALIDATE))) return 0; + if (likely(!(dentry->d_sb->s_type->fs_flags & FS_REVAL_DOT))) + return 0; + + /* Note: we do not d_invalidate() */ status = d_revalidate(dentry, nd); if (status > 0) return 0; - if (!status) { - d_invalidate(dentry); + if (!status) status = -ESTALE; - } + return status; } @@ -728,6 +718,7 @@ static __always_inline int __vfs_follow_link(struct nameidata *nd, const char *l path_put(&nd->path); nd->path = nd->root; path_get(&nd->root); + nd->flags |= LOOKUP_JUMPED; } nd->inode = nd->path.dentry->d_inode; @@ -779,11 +770,8 @@ __do_follow_link(const struct path *link, struct nameidata *nd, void **p) error = 0; if (s) error = __vfs_follow_link(nd, s); - else if (nd->last_type == LAST_BIND) { - error = force_reval_path(&nd->path, nd); - if (error) - path_put(&nd->path); - } + else if (nd->last_type == LAST_BIND) + nd->flags |= LOOKUP_JUMPED; } return error; } @@ -1351,7 +1339,7 @@ static int link_path_walk(const char *name, struct nameidata *nd) while (*name=='/') name++; if (!*name) - goto return_reval; + goto return_base; if (nd->depth) lookup_flags = LOOKUP_FOLLOW | (nd->flags & LOOKUP_CONTINUE); @@ -1385,12 +1373,16 @@ static int link_path_walk(const char *name, struct nameidata *nd) type = LAST_NORM; if (this.name[0] == '.') switch (this.len) { case 2: - if (this.name[1] == '.') + if (this.name[1] == '.') { type = LAST_DOTDOT; + nd->flags |= LOOKUP_JUMPED; + } break; case 1: type = LAST_DOT; } + if (likely(type == LAST_NORM)) + nd->flags &= ~LOOKUP_JUMPED; /* remove trailing slashes? */ if (!c) @@ -1456,7 +1448,7 @@ last_component: } else follow_dotdot(nd); } - goto return_reval; + goto return_base; } err = do_lookup(nd, &this, &next, &inode); if (err) @@ -1483,24 +1475,6 @@ last_component: lookup_parent: nd->last = this; nd->last_type = type; - if (type == LAST_NORM) - goto return_base; -return_reval: - /* - * We bypassed the ordinary revalidation routines. - * We may need to check the cached dentry for staleness. - */ - if (need_reval_dot(nd->path.dentry)) { - if (nameidata_drop_rcu_last_maybe(nd)) - return -ECHILD; - /* Note: we do not d_invalidate() */ - err = d_revalidate(nd->path.dentry, nd); - if (!err) - err = -ESTALE; - if (err < 0) - break; - return 0; - } return_base: if (nameidata_drop_rcu_last_maybe(nd)) return -ECHILD; @@ -1523,7 +1497,7 @@ static int path_init(int dfd, const char *name, unsigned int flags, struct namei struct file *file; nd->last_type = LAST_ROOT; /* if there are only slashes... */ - nd->flags = flags; + nd->flags = flags | LOOKUP_JUMPED; nd->depth = 0; nd->root.mnt = NULL; nd->file = NULL; @@ -1630,6 +1604,9 @@ static int path_lookupat(int dfd, const char *name, br_read_unlock(vfsmount_lock); } + if (!retval) + retval = handle_reval_path(nd); + if (nd->file) { fput(nd->file); nd->file = NULL; @@ -1690,7 +1667,7 @@ int vfs_path_lookup(struct dentry *dentry, struct vfsmount *mnt, /* same as do_path_lookup */ nd->last_type = LAST_ROOT; - nd->flags = flags; + nd->flags = flags | LOOKUP_JUMPED; nd->depth = 0; nd->path.dentry = dentry; @@ -1703,6 +1680,8 @@ int vfs_path_lookup(struct dentry *dentry, struct vfsmount *mnt, current->total_link_count = 0; result = link_path_walk(name, nd); + if (!result) + result = handle_reval_path(nd); if (result == -ESTALE) { /* nd->path had been dropped */ current->total_link_count = 0; @@ -1710,8 +1689,11 @@ int vfs_path_lookup(struct dentry *dentry, struct vfsmount *mnt, nd->path.mnt = mnt; nd->inode = dentry->d_inode; path_get(&nd->path); - nd->flags |= LOOKUP_REVAL; + nd->flags = flags | LOOKUP_JUMPED | LOOKUP_REVAL; + result = link_path_walk(name, nd); + if (!result) + result = handle_reval_path(nd); } if (unlikely(!result && !audit_dummy_context() && nd->path.dentry && nd->inode)) @@ -2198,30 +2180,29 @@ static struct file *do_last(struct nameidata *nd, struct path *path, { struct dentry *dir = nd->path.dentry; struct file *filp; - int error = -EISDIR; + int error; switch (nd->last_type) { case LAST_DOTDOT: follow_dotdot(nd); dir = nd->path.dentry; case LAST_DOT: - if (need_reval_dot(dir)) { - int status = d_revalidate(nd->path.dentry, nd); - if (!status) - status = -ESTALE; - if (status < 0) { - error = status; - goto exit; - } - } /* fallthrough */ case LAST_ROOT: + error = handle_reval_path(nd); + if (error) + goto exit; + error = -EISDIR; goto exit; case LAST_BIND: + error = handle_reval_path(nd); + if (error) + goto exit; audit_inode(pathname, dir); goto ok; } + error = -EISDIR; /* trailing slashes? */ if (nd->last.name[nd->last.len]) goto exit; @@ -2422,7 +2403,7 @@ reval: /* * We have the parent and last component. */ - nd.flags = flags; + nd.flags = (nd.flags & ~LOOKUP_PARENT) | flags; filp = do_last(&nd, &path, open_flag, acc_mode, mode, pathname); while (unlikely(!filp)) { /* trailing symlink */ struct path link = path; diff --git a/include/linux/namei.h b/include/linux/namei.h index 58ce3433d4ec..265378a707bd 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -63,6 +63,8 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND}; #define LOOKUP_EXCL 0x0400 #define LOOKUP_RENAME_TARGET 0x0800 +#define LOOKUP_JUMPED 0x1000 + extern int user_path_at(int, const char __user *, unsigned, struct path *); #define user_path(name, path) user_path_at(AT_FDCWD, name, LOOKUP_FOLLOW, path) -- cgit v1.2.3 From 47c805dc2d2dff686962f5f0baa6bac2d703ba19 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 23 Feb 2011 17:44:09 -0500 Subject: switch do_filp_open() to struct open_flags take calculation of open_flags by open(2) arguments into new helper in fs/open.c, move filp_open() over there, have it and do_sys_open() use that helper, switch exec.c callers of do_filp_open() to explicit (and constant) struct open_flags. Signed-off-by: Al Viro --- fs/exec.c | 18 +++++++---- fs/internal.h | 8 +++++ fs/namei.c | 88 ++++++------------------------------------------------ fs/open.c | 73 +++++++++++++++++++++++++++++++++++++++++++- include/linux/fs.h | 2 -- 5 files changed, 101 insertions(+), 88 deletions(-) (limited to 'include') diff --git a/fs/exec.c b/fs/exec.c index 52a447d9b6ab..ba99e1abb1aa 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -115,13 +115,16 @@ SYSCALL_DEFINE1(uselib, const char __user *, library) struct file *file; char *tmp = getname(library); int error = PTR_ERR(tmp); + static const struct open_flags uselib_flags = { + .open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC, + .acc_mode = MAY_READ | MAY_EXEC | MAY_OPEN, + .intent = LOOKUP_OPEN + }; if (IS_ERR(tmp)) goto out; - file = do_filp_open(AT_FDCWD, tmp, - O_LARGEFILE | O_RDONLY | __FMODE_EXEC, 0, - MAY_READ | MAY_EXEC | MAY_OPEN); + file = do_filp_open(AT_FDCWD, tmp, &uselib_flags, LOOKUP_FOLLOW); putname(tmp); error = PTR_ERR(file); if (IS_ERR(file)) @@ -721,10 +724,13 @@ struct file *open_exec(const char *name) { struct file *file; int err; + static const struct open_flags open_exec_flags = { + .open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC, + .acc_mode = MAY_EXEC | MAY_OPEN, + .intent = LOOKUP_OPEN + }; - file = do_filp_open(AT_FDCWD, name, - O_LARGEFILE | O_RDONLY | __FMODE_EXEC, 0, - MAY_EXEC | MAY_OPEN); + file = do_filp_open(AT_FDCWD, name, &open_exec_flags, LOOKUP_FOLLOW); if (IS_ERR(file)) goto out; diff --git a/fs/internal.h b/fs/internal.h index 9b976b57d7fe..6fdbdf2c6047 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -106,6 +106,14 @@ extern void put_super(struct super_block *sb); struct nameidata; extern struct file *nameidata_to_filp(struct nameidata *); extern void release_open_intent(struct nameidata *); +struct open_flags { + int open_flag; + int mode; + int acc_mode; + int intent; +}; +extern struct file *do_filp_open(int dfd, const char *pathname, + const struct open_flags *op, int lookup_flags); /* * inode.c diff --git a/fs/namei.c b/fs/namei.c index 5e4206f45371..9c7fa946abe1 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -2169,13 +2169,6 @@ exit: return ERR_PTR(error); } -struct open_flags { - int open_flag; - int mode; - int acc_mode; - int intent; -}; - /* * Handle O_CREAT case for do_filp_open */ @@ -2305,74 +2298,28 @@ exit: * open_to_namei_flags() for more details. */ struct file *do_filp_open(int dfd, const char *pathname, - int open_flag, int mode, int acc_mode) + const struct open_flags *op, int flags) { struct file *filp; struct nameidata nd; int error; struct path path; int count = 0; - int flag = open_to_namei_flags(open_flag); - int flags = 0; - struct open_flags op; - - if (!(open_flag & O_CREAT)) - mode = 0; - - /* Must never be set by userspace */ - open_flag &= ~FMODE_NONOTIFY; - - /* - * O_SYNC is implemented as __O_SYNC|O_DSYNC. As many places only - * check for O_DSYNC if the need any syncing at all we enforce it's - * always set instead of having to deal with possibly weird behaviour - * for malicious applications setting only __O_SYNC. - */ - if (open_flag & __O_SYNC) - open_flag |= O_DSYNC; - - op.open_flag = open_flag; - - if (!acc_mode) - acc_mode = MAY_OPEN | ACC_MODE(open_flag); - - /* O_TRUNC implies we need access checks for write permissions */ - if (open_flag & O_TRUNC) - acc_mode |= MAY_WRITE; - - /* Allow the LSM permission hook to distinguish append - access from general write access. */ - if (open_flag & O_APPEND) - acc_mode |= MAY_APPEND; - - op.acc_mode = acc_mode; - - op.intent = LOOKUP_OPEN; - if (open_flag & O_CREAT) { - op.intent |= LOOKUP_CREATE; - if (open_flag & O_EXCL) - op.intent |= LOOKUP_EXCL; - } - - if (open_flag & O_DIRECTORY) - flags |= LOOKUP_DIRECTORY; - if (!(open_flag & O_NOFOLLOW)) - flags |= LOOKUP_FOLLOW; filp = get_empty_filp(); if (!filp) return ERR_PTR(-ENFILE); - filp->f_flags = open_flag; + filp->f_flags = op->open_flag; nd.intent.open.file = filp; - nd.intent.open.flags = flag; - nd.intent.open.create_mode = mode; + nd.intent.open.flags = open_to_namei_flags(op->open_flag); + nd.intent.open.create_mode = op->mode; - if (open_flag & O_CREAT) + if (op->open_flag & O_CREAT) goto creat; /* !O_CREAT, simple open */ - error = do_path_lookup(dfd, pathname, flags | op.intent, &nd); + error = do_path_lookup(dfd, pathname, flags | op->intent, &nd); if (unlikely(error)) goto out_filp2; error = -ELOOP; @@ -2386,7 +2333,7 @@ struct file *do_filp_open(int dfd, const char *pathname, goto out_path2; } audit_inode(pathname, nd.path.dentry); - filp = finish_open(&nd, open_flag, acc_mode); + filp = finish_open(&nd, op->open_flag, op->acc_mode); out2: release_open_intent(&nd); return filp; @@ -2416,7 +2363,7 @@ reval: /* * We have the parent and last component. */ - filp = do_last(&nd, &path, &op, pathname); + filp = do_last(&nd, &path, op, pathname); while (unlikely(!filp)) { /* trailing symlink */ struct path link = path; struct inode *linki = link.dentry->d_inode; @@ -2443,7 +2390,7 @@ reval: if (unlikely(error)) filp = ERR_PTR(error); else - filp = do_last(&nd, &path, &op, pathname); + filp = do_last(&nd, &path, op, pathname); if (!IS_ERR(cookie) && linki->i_op->put_link) linki->i_op->put_link(link.dentry, &nd, cookie); path_put(&link); @@ -2465,23 +2412,6 @@ out_filp: goto out; } -/** - * filp_open - open file and return file pointer - * - * @filename: path to open - * @flags: open flags as per the open(2) second argument - * @mode: mode for the new file if O_CREAT is set, else ignored - * - * This is the helper to open a file from kernelspace if you really - * have to. But in generally you should not do this, so please move - * along, nothing to see here.. - */ -struct file *filp_open(const char *filename, int flags, int mode) -{ - return do_filp_open(AT_FDCWD, filename, flags, mode, 0); -} -EXPORT_SYMBOL(filp_open); - /** * lookup_create - lookup a dentry, creating it if it doesn't exist * @nd: nameidata info diff --git a/fs/open.c b/fs/open.c index b47aab39c057..d05e18c60bae 100644 --- a/fs/open.c +++ b/fs/open.c @@ -890,15 +890,86 @@ void fd_install(unsigned int fd, struct file *file) EXPORT_SYMBOL(fd_install); +static inline int build_open_flags(int flags, int mode, struct open_flags *op) +{ + int lookup_flags = 0; + int acc_mode; + + if (!(flags & O_CREAT)) + mode = 0; + op->mode = mode; + + /* Must never be set by userspace */ + flags &= ~FMODE_NONOTIFY; + + /* + * O_SYNC is implemented as __O_SYNC|O_DSYNC. As many places only + * check for O_DSYNC if the need any syncing at all we enforce it's + * always set instead of having to deal with possibly weird behaviour + * for malicious applications setting only __O_SYNC. + */ + if (flags & __O_SYNC) + flags |= O_DSYNC; + + op->open_flag = flags; + + acc_mode = MAY_OPEN | ACC_MODE(flags); + + /* O_TRUNC implies we need access checks for write permissions */ + if (flags & O_TRUNC) + acc_mode |= MAY_WRITE; + + /* Allow the LSM permission hook to distinguish append + access from general write access. */ + if (flags & O_APPEND) + acc_mode |= MAY_APPEND; + + op->acc_mode = acc_mode; + + op->intent = LOOKUP_OPEN; + if (flags & O_CREAT) { + op->intent |= LOOKUP_CREATE; + if (flags & O_EXCL) + op->intent |= LOOKUP_EXCL; + } + + if (flags & O_DIRECTORY) + lookup_flags |= LOOKUP_DIRECTORY; + if (!(flags & O_NOFOLLOW)) + lookup_flags |= LOOKUP_FOLLOW; + return lookup_flags; +} + +/** + * filp_open - open file and return file pointer + * + * @filename: path to open + * @flags: open flags as per the open(2) second argument + * @mode: mode for the new file if O_CREAT is set, else ignored + * + * This is the helper to open a file from kernelspace if you really + * have to. But in generally you should not do this, so please move + * along, nothing to see here.. + */ +struct file *filp_open(const char *filename, int flags, int mode) +{ + struct open_flags op; + int lookup = build_open_flags(flags, mode, &op); + return do_filp_open(AT_FDCWD, filename, &op, lookup); +} +EXPORT_SYMBOL(filp_open); + long do_sys_open(int dfd, const char __user *filename, int flags, int mode) { + struct open_flags op; + int lookup = build_open_flags(flags, mode, &op); char *tmp = getname(filename); int fd = PTR_ERR(tmp); if (!IS_ERR(tmp)) { fd = get_unused_fd_flags(flags); if (fd >= 0) { - struct file *f = do_filp_open(dfd, tmp, flags, mode, 0); + struct file *f = do_filp_open(dfd, tmp, &op, lookup); if (IS_ERR(f)) { put_unused_fd(fd); fd = PTR_ERR(f); diff --git a/include/linux/fs.h b/include/linux/fs.h index e38b50a4b9d2..9c75714f92c1 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2205,8 +2205,6 @@ extern struct file *create_read_pipe(struct file *f, int flags); extern struct file *create_write_pipe(int flags); extern void free_write_pipe(struct file *); -extern struct file *do_filp_open(int dfd, const char *pathname, - int open_flag, int mode, int acc_mode); extern int may_open(struct path *, int, int); extern int kernel_read(struct file *, loff_t, char *, unsigned long); -- cgit v1.2.3 From 70e9b3571107b88674cd55ae4bed33f76261e7d3 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 5 Mar 2011 21:12:22 -0500 Subject: get rid of nd->file Don't stash the struct file * used as starting point of walk in nameidata; pass file ** to path_init() instead. Signed-off-by: Al Viro --- fs/namei.c | 15 +++++++-------- include/linux/namei.h | 1 - 2 files changed, 7 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/fs/namei.c b/fs/namei.c index 85f6e39b4034..a260a306daf5 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1498,7 +1498,8 @@ lookup_parent: return err; } -static int path_init(int dfd, const char *name, unsigned int flags, struct nameidata *nd) +static int path_init(int dfd, const char *name, unsigned int flags, + struct nameidata *nd, struct file **fp) { int retval = 0; int fput_needed; @@ -1508,7 +1509,6 @@ static int path_init(int dfd, const char *name, unsigned int flags, struct namei nd->flags = flags | LOOKUP_JUMPED; nd->depth = 0; nd->root.mnt = NULL; - nd->file = NULL; if (*name=='/') { if (flags & LOOKUP_RCU) { @@ -1557,7 +1557,7 @@ static int path_init(int dfd, const char *name, unsigned int flags, struct namei nd->path = file->f_path; if (flags & LOOKUP_RCU) { if (fput_needed) - nd->file = file; + *fp = file; nd->seq = __read_seqcount_begin(&nd->path.dentry->d_seq); br_read_lock(vfsmount_lock); rcu_read_lock(); @@ -1580,6 +1580,7 @@ out_fail: static int path_lookupat(int dfd, const char *name, unsigned int flags, struct nameidata *nd) { + struct file *base = NULL; int retval; /* @@ -1596,7 +1597,7 @@ static int path_lookupat(int dfd, const char *name, * be handled by restarting a traditional ref-walk (which will always * be able to complete). */ - retval = path_init(dfd, name, flags, nd); + retval = path_init(dfd, name, flags, nd, &base); if (unlikely(retval)) return retval; @@ -1614,10 +1615,8 @@ static int path_lookupat(int dfd, const char *name, if (!retval) retval = handle_reval_path(nd); - if (nd->file) { - fput(nd->file); - nd->file = NULL; - } + if (base) + fput(base); if (nd->root.mnt) { path_put(&nd->root); diff --git a/include/linux/namei.h b/include/linux/namei.h index 265378a707bd..72ffd62ac736 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -19,7 +19,6 @@ struct nameidata { struct path path; struct qstr last; struct path root; - struct file *file; struct inode *inode; /* path.dentry.d_inode */ unsigned int flags; unsigned seq; -- cgit v1.2.3 From 5b6ca027d85b7438c84b78a54ccdc2e53f2909cd Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 9 Mar 2011 23:04:47 -0500 Subject: reduce vfs_path_lookup() to do_path_lookup() New lookup flag: LOOKUP_ROOT. nd->root is set (and held) by caller, path_init() starts walking from that place and all pathname resolution machinery never drops nd->root if that flag is set. That turns vfs_path_lookup() into a special case of do_path_lookup() *and* gets us down to 3 callers of link_path_walk(), making it finally feasible to rip the handling of trailing symlink out of link_path_walk(). That will not only simply the living hell out of it, but make life much simpler for unionfs merge. Trailing symlink handling will become iterative, which is a good thing for stack footprint in a lot of situations as well. Signed-off-by: Al Viro --- fs/namei.c | 95 +++++++++++++++++++++++---------------------------- include/linux/namei.h | 1 + 2 files changed, 44 insertions(+), 52 deletions(-) (limited to 'include') diff --git a/fs/namei.c b/fs/namei.c index 0bebd13e5cb7..8ee7785d5642 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -401,9 +401,11 @@ static int nameidata_drop_rcu(struct nameidata *nd) { struct fs_struct *fs = current->fs; struct dentry *dentry = nd->path.dentry; + int want_root = 0; BUG_ON(!(nd->flags & LOOKUP_RCU)); - if (nd->root.mnt) { + if (nd->root.mnt && !(nd->flags & LOOKUP_ROOT)) { + want_root = 1; spin_lock(&fs->lock); if (nd->root.mnt != fs->root.mnt || nd->root.dentry != fs->root.dentry) @@ -414,7 +416,7 @@ static int nameidata_drop_rcu(struct nameidata *nd) goto err; BUG_ON(nd->inode != dentry->d_inode); spin_unlock(&dentry->d_lock); - if (nd->root.mnt) { + if (want_root) { path_get(&nd->root); spin_unlock(&fs->lock); } @@ -427,7 +429,7 @@ static int nameidata_drop_rcu(struct nameidata *nd) err: spin_unlock(&dentry->d_lock); err_root: - if (nd->root.mnt) + if (want_root) spin_unlock(&fs->lock); return -ECHILD; } @@ -454,9 +456,11 @@ static int nameidata_dentry_drop_rcu(struct nameidata *nd, struct dentry *dentry { struct fs_struct *fs = current->fs; struct dentry *parent = nd->path.dentry; + int want_root = 0; BUG_ON(!(nd->flags & LOOKUP_RCU)); - if (nd->root.mnt) { + if (nd->root.mnt && !(nd->flags & LOOKUP_ROOT)) { + want_root = 1; spin_lock(&fs->lock); if (nd->root.mnt != fs->root.mnt || nd->root.dentry != fs->root.dentry) @@ -476,7 +480,7 @@ static int nameidata_dentry_drop_rcu(struct nameidata *nd, struct dentry *dentry parent->d_count++; spin_unlock(&dentry->d_lock); spin_unlock(&parent->d_lock); - if (nd->root.mnt) { + if (want_root) { path_get(&nd->root); spin_unlock(&fs->lock); } @@ -490,7 +494,7 @@ err: spin_unlock(&dentry->d_lock); spin_unlock(&parent->d_lock); err_root: - if (nd->root.mnt) + if (want_root) spin_unlock(&fs->lock); return -ECHILD; } @@ -501,7 +505,8 @@ static inline int nameidata_dentry_drop_rcu_maybe(struct nameidata *nd, struct d if (nd->flags & LOOKUP_RCU) { if (unlikely(nameidata_dentry_drop_rcu(nd, dentry))) { nd->flags &= ~LOOKUP_RCU; - nd->root.mnt = NULL; + if (!(nd->flags & LOOKUP_ROOT)) + nd->root.mnt = NULL; rcu_read_unlock(); br_read_unlock(vfsmount_lock); return -ECHILD; @@ -525,7 +530,8 @@ static int nameidata_drop_rcu_last(struct nameidata *nd) BUG_ON(!(nd->flags & LOOKUP_RCU)); nd->flags &= ~LOOKUP_RCU; - nd->root.mnt = NULL; + if (!(nd->flags & LOOKUP_ROOT)) + nd->root.mnt = NULL; spin_lock(&dentry->d_lock); if (!__d_rcu_to_refcount(dentry, nd->seq)) goto err_unlock; @@ -1053,7 +1059,8 @@ static int follow_dotdot_rcu(struct nameidata *nd) failed: nd->flags &= ~LOOKUP_RCU; - nd->root.mnt = NULL; + if (!(nd->flags & LOOKUP_ROOT)) + nd->root.mnt = NULL; rcu_read_unlock(); br_read_unlock(vfsmount_lock); return -ECHILD; @@ -1310,7 +1317,8 @@ static void terminate_walk(struct nameidata *nd) path_put(&nd->path); } else { nd->flags &= ~LOOKUP_RCU; - nd->root.mnt = NULL; + if (!(nd->flags & LOOKUP_ROOT)) + nd->root.mnt = NULL; rcu_read_unlock(); br_read_unlock(vfsmount_lock); } @@ -1477,6 +1485,25 @@ static int path_init(int dfd, const char *name, unsigned int flags, nd->last_type = LAST_ROOT; /* if there are only slashes... */ nd->flags = flags | LOOKUP_JUMPED; nd->depth = 0; + if (flags & LOOKUP_ROOT) { + struct inode *inode = nd->root.dentry->d_inode; + if (!inode->i_op->lookup) + return -ENOTDIR; + retval = inode_permission(inode, MAY_EXEC); + if (retval) + return retval; + nd->path = nd->root; + nd->inode = inode; + if (flags & LOOKUP_RCU) { + br_read_lock(vfsmount_lock); + rcu_read_lock(); + nd->seq = __read_seqcount_begin(&nd->path.dentry->d_seq); + } else { + path_get(&nd->path); + } + return 0; + } + nd->root.mnt = NULL; if (*name=='/') { @@ -1587,7 +1614,7 @@ static int path_lookupat(int dfd, const char *name, if (base) fput(base); - if (nd->root.mnt) { + if (nd->root.mnt && !(nd->flags & LOOKUP_ROOT)) { path_put(&nd->root); nd->root.mnt = NULL; } @@ -1638,46 +1665,10 @@ int vfs_path_lookup(struct dentry *dentry, struct vfsmount *mnt, const char *name, unsigned int flags, struct nameidata *nd) { - int result; - - /* same as do_path_lookup */ - nd->last_type = LAST_ROOT; - nd->flags = flags | LOOKUP_JUMPED; - nd->depth = 0; - - nd->path.dentry = dentry; - nd->path.mnt = mnt; - path_get(&nd->path); - nd->root = nd->path; - path_get(&nd->root); - nd->inode = nd->path.dentry->d_inode; - - current->total_link_count = 0; - - result = link_path_walk(name, nd); - if (!result) - result = handle_reval_path(nd); - if (result == -ESTALE) { - /* nd->path had been dropped */ - current->total_link_count = 0; - nd->path.dentry = dentry; - nd->path.mnt = mnt; - nd->inode = dentry->d_inode; - path_get(&nd->path); - nd->flags = flags | LOOKUP_JUMPED | LOOKUP_REVAL; - - result = link_path_walk(name, nd); - if (!result) - result = handle_reval_path(nd); - } - if (unlikely(!result && !audit_dummy_context() && nd->path.dentry && - nd->inode)) - audit_inode(name, nd->path.dentry); - - path_put(&nd->root); - nd->root.mnt = NULL; - - return result; + nd->root.dentry = dentry; + nd->root.mnt = mnt; + /* the first argument of do_path_lookup() is ignored with LOOKUP_ROOT */ + return do_path_lookup(AT_FDCWD, name, flags | LOOKUP_ROOT, nd); } static struct dentry *__lookup_hash(struct qstr *name, @@ -2320,7 +2311,7 @@ static struct file *path_openat(int dfd, const char *pathname, path_put(&link); } out: - if (nd.root.mnt) + if (nd.root.mnt && !(nd.flags & LOOKUP_ROOT)) path_put(&nd.root); if (base) fput(base); diff --git a/include/linux/namei.h b/include/linux/namei.h index 72ffd62ac736..83cd6e5cd7dc 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -63,6 +63,7 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND}; #define LOOKUP_RENAME_TARGET 0x0800 #define LOOKUP_JUMPED 0x1000 +#define LOOKUP_ROOT 0x2000 extern int user_path_at(int, const char __user *, unsigned, struct path *); -- cgit v1.2.3 From 73d049a40fc6269189c4e2ba6792cb5dd054883c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 11 Mar 2011 12:08:24 -0500 Subject: open-style analog of vfs_path_lookup() new function: file_open_root(dentry, mnt, name, flags) opens the file vfs_path_lookup would arrive to. Note that name can be empty; in that case the usual requirement that dentry should be a directory is lifted. open-coded equivalents switched to it, may_open() got down exactly one caller and became static. Signed-off-by: Al Viro --- arch/um/drivers/mconsole_kern.c | 21 ++--------- fs/internal.h | 2 ++ fs/namei.c | 80 ++++++++++++++++++++++++++--------------- fs/nfsctl.c | 21 +++-------- fs/open.c | 14 ++++++++ include/linux/fs.h | 4 +-- kernel/sysctl_binary.c | 19 +--------- 7 files changed, 77 insertions(+), 84 deletions(-) (limited to 'include') diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c index 975613b23dcf..c70e047eed72 100644 --- a/arch/um/drivers/mconsole_kern.c +++ b/arch/um/drivers/mconsole_kern.c @@ -124,35 +124,18 @@ void mconsole_log(struct mc_request *req) #if 0 void mconsole_proc(struct mc_request *req) { - struct nameidata nd; struct vfsmount *mnt = current->nsproxy->pid_ns->proc_mnt; struct file *file; - int n, err; + int n; char *ptr = req->request.data, *buf; mm_segment_t old_fs = get_fs(); ptr += strlen("proc"); ptr = skip_spaces(ptr); - err = vfs_path_lookup(mnt->mnt_root, mnt, ptr, LOOKUP_FOLLOW, &nd); - if (err) { - mconsole_reply(req, "Failed to look up file", 1, 0); - goto out; - } - - err = may_open(&nd.path, MAY_READ, O_RDONLY); - if (result) { - mconsole_reply(req, "Failed to open file", 1, 0); - path_put(&nd.path); - goto out; - } - - file = dentry_open(nd.path.dentry, nd.path.mnt, O_RDONLY, - current_cred()); - err = PTR_ERR(file); + file = file_open_root(mnt->mnt_root, mnt, ptr, O_RDONLY); if (IS_ERR(file)) { mconsole_reply(req, "Failed to open file", 1, 0); - path_put(&nd.path); goto out; } diff --git a/fs/internal.h b/fs/internal.h index 6fdbdf2c6047..52abc5287f50 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -114,6 +114,8 @@ struct open_flags { }; extern struct file *do_filp_open(int dfd, const char *pathname, const struct open_flags *op, int lookup_flags); +extern struct file *do_file_open_root(struct dentry *, struct vfsmount *, + const char *, const struct open_flags *, int lookup_flags); /* * inode.c diff --git a/fs/namei.c b/fs/namei.c index 8ee7785d5642..abc8d2df121c 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1487,11 +1487,13 @@ static int path_init(int dfd, const char *name, unsigned int flags, nd->depth = 0; if (flags & LOOKUP_ROOT) { struct inode *inode = nd->root.dentry->d_inode; - if (!inode->i_op->lookup) - return -ENOTDIR; - retval = inode_permission(inode, MAY_EXEC); - if (retval) - return retval; + if (*name) { + if (!inode->i_op->lookup) + return -ENOTDIR; + retval = inode_permission(inode, MAY_EXEC); + if (retval) + return retval; + } nd->path = nd->root; nd->inode = inode; if (flags & LOOKUP_RCU) { @@ -1937,7 +1939,7 @@ int vfs_create(struct inode *dir, struct dentry *dentry, int mode, return error; } -int may_open(struct path *path, int acc_mode, int flag) +static int may_open(struct path *path, int acc_mode, int flag) { struct dentry *dentry = path->dentry; struct inode *inode = dentry->d_inode; @@ -2250,11 +2252,10 @@ exit: } static struct file *path_openat(int dfd, const char *pathname, - const struct open_flags *op, int flags) + struct nameidata *nd, const struct open_flags *op, int flags) { struct file *base = NULL; struct file *filp; - struct nameidata nd; struct path path; int count = 0; int error; @@ -2264,27 +2265,27 @@ static struct file *path_openat(int dfd, const char *pathname, return ERR_PTR(-ENFILE); filp->f_flags = op->open_flag; - nd.intent.open.file = filp; - nd.intent.open.flags = open_to_namei_flags(op->open_flag); - nd.intent.open.create_mode = op->mode; + nd->intent.open.file = filp; + nd->intent.open.flags = open_to_namei_flags(op->open_flag); + nd->intent.open.create_mode = op->mode; - error = path_init(dfd, pathname, flags | LOOKUP_PARENT, &nd, &base); + error = path_init(dfd, pathname, flags | LOOKUP_PARENT, nd, &base); if (unlikely(error)) goto out_filp; current->total_link_count = 0; - error = link_path_walk(pathname, &nd); + error = link_path_walk(pathname, nd); if (unlikely(error)) goto out_filp; - filp = do_last(&nd, &path, op, pathname); + filp = do_last(nd, &path, op, pathname); while (unlikely(!filp)) { /* trailing symlink */ struct path link = path; struct inode *linki = link.dentry->d_inode; void *cookie; - if (!(nd.flags & LOOKUP_FOLLOW) || count++ == 32) { - path_put_conditional(&path, &nd); - path_put(&nd.path); + if (!(nd->flags & LOOKUP_FOLLOW) || count++ == 32) { + path_put_conditional(&path, nd); + path_put(&nd->path); filp = ERR_PTR(-ELOOP); break; } @@ -2299,23 +2300,23 @@ static struct file *path_openat(int dfd, const char *pathname, * have to putname() it when we are done. Procfs-like symlinks * just set LAST_BIND. */ - nd.flags |= LOOKUP_PARENT; - nd.flags &= ~(LOOKUP_OPEN|LOOKUP_CREATE|LOOKUP_EXCL); - error = __do_follow_link(&link, &nd, &cookie); + nd->flags |= LOOKUP_PARENT; + nd->flags &= ~(LOOKUP_OPEN|LOOKUP_CREATE|LOOKUP_EXCL); + error = __do_follow_link(&link, nd, &cookie); if (unlikely(error)) filp = ERR_PTR(error); else - filp = do_last(&nd, &path, op, pathname); + filp = do_last(nd, &path, op, pathname); if (!IS_ERR(cookie) && linki->i_op->put_link) - linki->i_op->put_link(link.dentry, &nd, cookie); + linki->i_op->put_link(link.dentry, nd, cookie); path_put(&link); } out: - if (nd.root.mnt && !(nd.flags & LOOKUP_ROOT)) - path_put(&nd.root); + if (nd->root.mnt && !(nd->flags & LOOKUP_ROOT)) + path_put(&nd->root); if (base) fput(base); - release_open_intent(&nd); + release_open_intent(nd); return filp; out_filp: @@ -2326,16 +2327,39 @@ out_filp: struct file *do_filp_open(int dfd, const char *pathname, const struct open_flags *op, int flags) { + struct nameidata nd; struct file *filp; - filp = path_openat(dfd, pathname, op, flags | LOOKUP_RCU); + filp = path_openat(dfd, pathname, &nd, op, flags | LOOKUP_RCU); if (unlikely(filp == ERR_PTR(-ECHILD))) - filp = path_openat(dfd, pathname, op, flags); + filp = path_openat(dfd, pathname, &nd, op, flags); if (unlikely(filp == ERR_PTR(-ESTALE))) - filp = path_openat(dfd, pathname, op, flags | LOOKUP_REVAL); + filp = path_openat(dfd, pathname, &nd, op, flags | LOOKUP_REVAL); return filp; } +struct file *do_file_open_root(struct dentry *dentry, struct vfsmount *mnt, + const char *name, const struct open_flags *op, int flags) +{ + struct nameidata nd; + struct file *file; + + nd.root.mnt = mnt; + nd.root.dentry = dentry; + + flags |= LOOKUP_ROOT; + + if (dentry->d_inode->i_op->follow_link) + return ERR_PTR(-ELOOP); + + file = path_openat(-1, name, &nd, op, flags | LOOKUP_RCU); + if (unlikely(file == ERR_PTR(-ECHILD))) + file = path_openat(-1, name, &nd, op, flags); + if (unlikely(file == ERR_PTR(-ESTALE))) + file = path_openat(-1, name, &nd, op, flags | LOOKUP_REVAL); + return file; +} + /** * lookup_create - lookup a dentry, creating it if it doesn't exist * @nd: nameidata info diff --git a/fs/nfsctl.c b/fs/nfsctl.c index bf9cbd242ddd..124e8fcb0dd6 100644 --- a/fs/nfsctl.c +++ b/fs/nfsctl.c @@ -22,30 +22,17 @@ static struct file *do_open(char *name, int flags) { - struct nameidata nd; struct vfsmount *mnt; - int error; + struct file *file; mnt = do_kern_mount("nfsd", 0, "nfsd", NULL); if (IS_ERR(mnt)) return (struct file *)mnt; - error = vfs_path_lookup(mnt->mnt_root, mnt, name, 0, &nd); - mntput(mnt); /* drop do_kern_mount reference */ - if (error) - return ERR_PTR(error); - - if (flags == O_RDWR) - error = may_open(&nd.path, MAY_READ|MAY_WRITE, flags); - else - error = may_open(&nd.path, MAY_WRITE, flags); + file = file_open_root(mnt->mnt_root, mnt, name, flags); - if (!error) - return dentry_open(nd.path.dentry, nd.path.mnt, flags, - current_cred()); - - path_put(&nd.path); - return ERR_PTR(error); + mntput(mnt); /* drop do_kern_mount reference */ + return file; } static struct { diff --git a/fs/open.c b/fs/open.c index d05e18c60bae..48afc5c139d2 100644 --- a/fs/open.c +++ b/fs/open.c @@ -959,6 +959,20 @@ struct file *filp_open(const char *filename, int flags, int mode) } EXPORT_SYMBOL(filp_open); +struct file *file_open_root(struct dentry *dentry, struct vfsmount *mnt, + const char *filename, int flags) +{ + struct open_flags op; + int lookup = build_open_flags(flags, 0, &op); + if (flags & O_CREAT) + return ERR_PTR(-EINVAL); + if (!filename && (flags & O_DIRECTORY)) + if (!dentry->d_inode->i_op->lookup) + return ERR_PTR(-ENOTDIR); + return do_file_open_root(dentry, mnt, filename, &op, lookup); +} +EXPORT_SYMBOL(file_open_root); + long do_sys_open(int dfd, const char __user *filename, int flags, int mode) { struct open_flags op; diff --git a/include/linux/fs.h b/include/linux/fs.h index 9c75714f92c1..bf5c3c896072 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1990,6 +1990,8 @@ extern int do_fallocate(struct file *file, int mode, loff_t offset, extern long do_sys_open(int dfd, const char __user *filename, int flags, int mode); extern struct file *filp_open(const char *, int, int); +extern struct file *file_open_root(struct dentry *, struct vfsmount *, + const char *, int); extern struct file * dentry_open(struct dentry *, struct vfsmount *, int, const struct cred *); extern int filp_close(struct file *, fl_owner_t id); @@ -2205,8 +2207,6 @@ extern struct file *create_read_pipe(struct file *f, int flags); extern struct file *create_write_pipe(int flags); extern void free_write_pipe(struct file *); -extern int may_open(struct path *, int, int); - extern int kernel_read(struct file *, loff_t, char *, unsigned long); extern struct file * open_exec(const char *); diff --git a/kernel/sysctl_binary.c b/kernel/sysctl_binary.c index b875bedf7c9a..3b8e028b9601 100644 --- a/kernel/sysctl_binary.c +++ b/kernel/sysctl_binary.c @@ -1321,13 +1321,11 @@ static ssize_t binary_sysctl(const int *name, int nlen, void __user *oldval, size_t oldlen, void __user *newval, size_t newlen) { const struct bin_table *table = NULL; - struct nameidata nd; struct vfsmount *mnt; struct file *file; ssize_t result; char *pathname; int flags; - int acc_mode; pathname = sysctl_getname(name, nlen, &table); result = PTR_ERR(pathname); @@ -1337,28 +1335,17 @@ static ssize_t binary_sysctl(const int *name, int nlen, /* How should the sysctl be accessed? */ if (oldval && oldlen && newval && newlen) { flags = O_RDWR; - acc_mode = MAY_READ | MAY_WRITE; } else if (newval && newlen) { flags = O_WRONLY; - acc_mode = MAY_WRITE; } else if (oldval && oldlen) { flags = O_RDONLY; - acc_mode = MAY_READ; } else { result = 0; goto out_putname; } mnt = current->nsproxy->pid_ns->proc_mnt; - result = vfs_path_lookup(mnt->mnt_root, mnt, pathname, 0, &nd); - if (result) - goto out_putname; - - result = may_open(&nd.path, acc_mode, flags); - if (result) - goto out_putpath; - - file = dentry_open(nd.path.dentry, nd.path.mnt, flags, current_cred()); + file = file_open_root(mnt->mnt_root, mnt, pathname, flags); result = PTR_ERR(file); if (IS_ERR(file)) goto out_putname; @@ -1370,10 +1357,6 @@ out_putname: putname(pathname); out: return result; - -out_putpath: - path_put(&nd.path); - goto out_putname; } -- cgit v1.2.3 From c8b91accfa1059d5565443193d89572eca2f5dd6 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 12 Mar 2011 10:41:39 -0500 Subject: clean statfs-like syscalls up New helpers: user_statfs() and fd_statfs(), taking userland pathname and descriptor resp. and filling struct kstatfs. Syscalls of statfs family (native, compat and foreign - osf and hpux on alpha and parisc resp.) switched to those. Removes some boilerplate code, simplifies cleanup on errors... Signed-off-by: Al Viro --- arch/alpha/kernel/osf_sys.c | 36 ++------- arch/parisc/hpux/sys_hpux.c | 65 ++++++---------- fs/compat.c | 48 +++--------- fs/statfs.c | 176 ++++++++++++++++++++------------------------ include/linux/fs.h | 2 + 5 files changed, 123 insertions(+), 204 deletions(-) (limited to 'include') diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index fe698b5045e9..376f22130791 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -230,44 +230,24 @@ linux_to_osf_statfs(struct kstatfs *linux_stat, struct osf_statfs __user *osf_st return copy_to_user(osf_stat, &tmp_stat, bufsiz) ? -EFAULT : 0; } -static int -do_osf_statfs(struct path *path, struct osf_statfs __user *buffer, - unsigned long bufsiz) +SYSCALL_DEFINE3(osf_statfs, const char __user *, pathname, + struct osf_statfs __user *, buffer, unsigned long, bufsiz) { struct kstatfs linux_stat; - int error = vfs_statfs(path, &linux_stat); + int error = user_statfs(pathname, &linux_stat); if (!error) error = linux_to_osf_statfs(&linux_stat, buffer, bufsiz); return error; } -SYSCALL_DEFINE3(osf_statfs, const char __user *, pathname, - struct osf_statfs __user *, buffer, unsigned long, bufsiz) -{ - struct path path; - int retval; - - retval = user_path(pathname, &path); - if (!retval) { - retval = do_osf_statfs(&path, buffer, bufsiz); - path_put(&path); - } - return retval; -} - SYSCALL_DEFINE3(osf_fstatfs, unsigned long, fd, struct osf_statfs __user *, buffer, unsigned long, bufsiz) { - struct file *file; - int retval; - - retval = -EBADF; - file = fget(fd); - if (file) { - retval = do_osf_statfs(&file->f_path, buffer, bufsiz); - fput(file); - } - return retval; + struct kstatfs linux_stat; + int error = fd_statfs(fd, &linux_stat); + if (!error) + error = linux_to_osf_statfs(&linux_stat, buffer, bufsiz); + return error; } /* diff --git a/arch/parisc/hpux/sys_hpux.c b/arch/parisc/hpux/sys_hpux.c index 30394081d9b6..6ab9580b0b00 100644 --- a/arch/parisc/hpux/sys_hpux.c +++ b/arch/parisc/hpux/sys_hpux.c @@ -185,26 +185,21 @@ struct hpux_statfs { int16_t f_pad; }; -static int do_statfs_hpux(struct path *path, struct hpux_statfs *buf) +static int do_statfs_hpux(struct kstatfs *st, struct hpux_statfs __user *p) { - struct kstatfs st; - int retval; - - retval = vfs_statfs(path, &st); - if (retval) - return retval; - - memset(buf, 0, sizeof(*buf)); - buf->f_type = st.f_type; - buf->f_bsize = st.f_bsize; - buf->f_blocks = st.f_blocks; - buf->f_bfree = st.f_bfree; - buf->f_bavail = st.f_bavail; - buf->f_files = st.f_files; - buf->f_ffree = st.f_ffree; - buf->f_fsid[0] = st.f_fsid.val[0]; - buf->f_fsid[1] = st.f_fsid.val[1]; - + struct hpux_statfs buf; + memset(&buf, 0, sizeof(buf)); + buf.f_type = st->f_type; + buf.f_bsize = st->f_bsize; + buf.f_blocks = st->f_blocks; + buf.f_bfree = st->f_bfree; + buf.f_bavail = st->f_bavail; + buf.f_files = st->f_files; + buf.f_ffree = st->f_ffree; + buf.f_fsid[0] = st->f_fsid.val[0]; + buf.f_fsid[1] = st->f_fsid.val[1]; + if (copy_to_user(p, &buf, sizeof(buf))) + return -EFAULT; return 0; } @@ -212,35 +207,19 @@ static int do_statfs_hpux(struct path *path, struct hpux_statfs *buf) asmlinkage long hpux_statfs(const char __user *pathname, struct hpux_statfs __user *buf) { - struct path path; - int error; - - error = user_path(pathname, &path); - if (!error) { - struct hpux_statfs tmp; - error = do_statfs_hpux(&path, &tmp); - if (!error && copy_to_user(buf, &tmp, sizeof(tmp))) - error = -EFAULT; - path_put(&path); - } + struct kstatfs st; + int error = user_statfs(pathname, &st); + if (!error) + error = do_statfs_hpux(&st, buf); return error; } asmlinkage long hpux_fstatfs(unsigned int fd, struct hpux_statfs __user * buf) { - struct file *file; - struct hpux_statfs tmp; - int error; - - error = -EBADF; - file = fget(fd); - if (!file) - goto out; - error = do_statfs_hpux(&file->f_path, &tmp); - if (!error && copy_to_user(buf, &tmp, sizeof(tmp))) - error = -EFAULT; - fput(file); - out: + struct kstatfs st; + int error = fd_statfs(fd, &st); + if (!error) + error = do_statfs_hpux(&st, buf); return error; } diff --git a/fs/compat.c b/fs/compat.c index 691c3fd8ce1d..a071775f3bb3 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -262,35 +262,19 @@ static int put_compat_statfs(struct compat_statfs __user *ubuf, struct kstatfs * */ asmlinkage long compat_sys_statfs(const char __user *pathname, struct compat_statfs __user *buf) { - struct path path; - int error; - - error = user_path(pathname, &path); - if (!error) { - struct kstatfs tmp; - error = vfs_statfs(&path, &tmp); - if (!error) - error = put_compat_statfs(buf, &tmp); - path_put(&path); - } + struct kstatfs tmp; + int error = user_statfs(pathname, &tmp); + if (!error) + error = put_compat_statfs(buf, &tmp); return error; } asmlinkage long compat_sys_fstatfs(unsigned int fd, struct compat_statfs __user *buf) { - struct file * file; struct kstatfs tmp; - int error; - - error = -EBADF; - file = fget(fd); - if (!file) - goto out; - error = vfs_statfs(&file->f_path, &tmp); + int error = fd_statfs(fd, &tmp); if (!error) error = put_compat_statfs(buf, &tmp); - fput(file); -out: return error; } @@ -329,41 +313,29 @@ static int put_compat_statfs64(struct compat_statfs64 __user *ubuf, struct kstat asmlinkage long compat_sys_statfs64(const char __user *pathname, compat_size_t sz, struct compat_statfs64 __user *buf) { - struct path path; + struct kstatfs tmp; int error; if (sz != sizeof(*buf)) return -EINVAL; - error = user_path(pathname, &path); - if (!error) { - struct kstatfs tmp; - error = vfs_statfs(&path, &tmp); - if (!error) - error = put_compat_statfs64(buf, &tmp); - path_put(&path); - } + error = user_statfs(pathname, &tmp); + if (!error) + error = put_compat_statfs64(buf, &tmp); return error; } asmlinkage long compat_sys_fstatfs64(unsigned int fd, compat_size_t sz, struct compat_statfs64 __user *buf) { - struct file * file; struct kstatfs tmp; int error; if (sz != sizeof(*buf)) return -EINVAL; - error = -EBADF; - file = fget(fd); - if (!file) - goto out; - error = vfs_statfs(&file->f_path, &tmp); + error = fd_statfs(fd, &tmp); if (!error) error = put_compat_statfs64(buf, &tmp); - fput(file); -out: return error; } diff --git a/fs/statfs.c b/fs/statfs.c index 30ea8c8a996b..8244924dec55 100644 --- a/fs/statfs.c +++ b/fs/statfs.c @@ -73,149 +73,135 @@ int vfs_statfs(struct path *path, struct kstatfs *buf) } EXPORT_SYMBOL(vfs_statfs); -static int do_statfs_native(struct path *path, struct statfs *buf) +int user_statfs(const char __user *pathname, struct kstatfs *st) { - struct kstatfs st; - int retval; + struct path path; + int error = user_path(pathname, &path); + if (!error) { + error = vfs_statfs(&path, st); + path_put(&path); + } + return error; +} - retval = vfs_statfs(path, &st); - if (retval) - return retval; +int fd_statfs(int fd, struct kstatfs *st) +{ + struct file *file = fget(fd); + int error = -EBADF; + if (file) { + error = vfs_statfs(&file->f_path, st); + fput(file); + } + return error; +} - if (sizeof(*buf) == sizeof(st)) - memcpy(buf, &st, sizeof(st)); +static int do_statfs_native(struct kstatfs *st, struct statfs __user *p) +{ + struct statfs buf; + + if (sizeof(buf) == sizeof(*st)) + memcpy(&buf, st, sizeof(*st)); else { - if (sizeof buf->f_blocks == 4) { - if ((st.f_blocks | st.f_bfree | st.f_bavail | - st.f_bsize | st.f_frsize) & + if (sizeof buf.f_blocks == 4) { + if ((st->f_blocks | st->f_bfree | st->f_bavail | + st->f_bsize | st->f_frsize) & 0xffffffff00000000ULL) return -EOVERFLOW; /* * f_files and f_ffree may be -1; it's okay to stuff * that into 32 bits */ - if (st.f_files != -1 && - (st.f_files & 0xffffffff00000000ULL)) + if (st->f_files != -1 && + (st->f_files & 0xffffffff00000000ULL)) return -EOVERFLOW; - if (st.f_ffree != -1 && - (st.f_ffree & 0xffffffff00000000ULL)) + if (st->f_ffree != -1 && + (st->f_ffree & 0xffffffff00000000ULL)) return -EOVERFLOW; } - buf->f_type = st.f_type; - buf->f_bsize = st.f_bsize; - buf->f_blocks = st.f_blocks; - buf->f_bfree = st.f_bfree; - buf->f_bavail = st.f_bavail; - buf->f_files = st.f_files; - buf->f_ffree = st.f_ffree; - buf->f_fsid = st.f_fsid; - buf->f_namelen = st.f_namelen; - buf->f_frsize = st.f_frsize; - buf->f_flags = st.f_flags; - memset(buf->f_spare, 0, sizeof(buf->f_spare)); + buf.f_type = st->f_type; + buf.f_bsize = st->f_bsize; + buf.f_blocks = st->f_blocks; + buf.f_bfree = st->f_bfree; + buf.f_bavail = st->f_bavail; + buf.f_files = st->f_files; + buf.f_ffree = st->f_ffree; + buf.f_fsid = st->f_fsid; + buf.f_namelen = st->f_namelen; + buf.f_frsize = st->f_frsize; + buf.f_flags = st->f_flags; + memset(buf.f_spare, 0, sizeof(buf.f_spare)); } + if (copy_to_user(p, &buf, sizeof(buf))) + return -EFAULT; return 0; } -static int do_statfs64(struct path *path, struct statfs64 *buf) +static int do_statfs64(struct kstatfs *st, struct statfs64 __user *p) { - struct kstatfs st; - int retval; - - retval = vfs_statfs(path, &st); - if (retval) - return retval; - - if (sizeof(*buf) == sizeof(st)) - memcpy(buf, &st, sizeof(st)); + struct statfs64 buf; + if (sizeof(buf) == sizeof(*st)) + memcpy(&buf, st, sizeof(*st)); else { - buf->f_type = st.f_type; - buf->f_bsize = st.f_bsize; - buf->f_blocks = st.f_blocks; - buf->f_bfree = st.f_bfree; - buf->f_bavail = st.f_bavail; - buf->f_files = st.f_files; - buf->f_ffree = st.f_ffree; - buf->f_fsid = st.f_fsid; - buf->f_namelen = st.f_namelen; - buf->f_frsize = st.f_frsize; - buf->f_flags = st.f_flags; - memset(buf->f_spare, 0, sizeof(buf->f_spare)); + buf.f_type = st->f_type; + buf.f_bsize = st->f_bsize; + buf.f_blocks = st->f_blocks; + buf.f_bfree = st->f_bfree; + buf.f_bavail = st->f_bavail; + buf.f_files = st->f_files; + buf.f_ffree = st->f_ffree; + buf.f_fsid = st->f_fsid; + buf.f_namelen = st->f_namelen; + buf.f_frsize = st->f_frsize; + buf.f_flags = st->f_flags; + memset(buf.f_spare, 0, sizeof(buf.f_spare)); } + if (copy_to_user(p, &buf, sizeof(buf))) + return -EFAULT; return 0; } SYSCALL_DEFINE2(statfs, const char __user *, pathname, struct statfs __user *, buf) { - struct path path; - int error; - - error = user_path(pathname, &path); - if (!error) { - struct statfs tmp; - error = do_statfs_native(&path, &tmp); - if (!error && copy_to_user(buf, &tmp, sizeof(tmp))) - error = -EFAULT; - path_put(&path); - } + struct kstatfs st; + int error = user_statfs(pathname, &st); + if (!error) + error = do_statfs_native(&st, buf); return error; } SYSCALL_DEFINE3(statfs64, const char __user *, pathname, size_t, sz, struct statfs64 __user *, buf) { - struct path path; - long error; - + struct kstatfs st; + int error; if (sz != sizeof(*buf)) return -EINVAL; - error = user_path(pathname, &path); - if (!error) { - struct statfs64 tmp; - error = do_statfs64(&path, &tmp); - if (!error && copy_to_user(buf, &tmp, sizeof(tmp))) - error = -EFAULT; - path_put(&path); - } + error = user_statfs(pathname, &st); + if (!error) + error = do_statfs64(&st, buf); return error; } SYSCALL_DEFINE2(fstatfs, unsigned int, fd, struct statfs __user *, buf) { - struct file *file; - struct statfs tmp; - int error; - - error = -EBADF; - file = fget(fd); - if (!file) - goto out; - error = do_statfs_native(&file->f_path, &tmp); - if (!error && copy_to_user(buf, &tmp, sizeof(tmp))) - error = -EFAULT; - fput(file); -out: + struct kstatfs st; + int error = fd_statfs(fd, &st); + if (!error) + error = do_statfs_native(&st, buf); return error; } SYSCALL_DEFINE3(fstatfs64, unsigned int, fd, size_t, sz, struct statfs64 __user *, buf) { - struct file *file; - struct statfs64 tmp; + struct kstatfs st; int error; if (sz != sizeof(*buf)) return -EINVAL; - error = -EBADF; - file = fget(fd); - if (!file) - goto out; - error = do_statfs64(&file->f_path, &tmp); - if (!error && copy_to_user(buf, &tmp, sizeof(tmp))) - error = -EFAULT; - fput(file); -out: + error = fd_statfs(fd, &st); + if (!error) + error = do_statfs64(&st, buf); return error; } diff --git a/include/linux/fs.h b/include/linux/fs.h index bf5c3c896072..b7178b05cf3a 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1874,6 +1874,8 @@ extern void drop_collected_mounts(struct vfsmount *); extern int iterate_mounts(int (*)(struct vfsmount *, void *), void *, struct vfsmount *); extern int vfs_statfs(struct path *, struct kstatfs *); +extern int user_statfs(const char __user *, struct kstatfs *); +extern int fd_statfs(int, struct kstatfs *); extern int statfs_by_dentry(struct dentry *, struct kstatfs *); extern int freeze_super(struct super_block *super); extern int thaw_super(struct super_block *super); -- cgit v1.2.3 From 5fe0c2378884e68beb532f5890cc0e3539ac747b Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Sat, 29 Jan 2011 18:43:25 +0530 Subject: exportfs: Return the minimum required handle size The exportfs encode handle function should return the minimum required handle size. This helps user to find out the handle size by passing 0 handle size in the first step and then redoing to the call again with the returned handle size value. Acked-by: Serge Hallyn Signed-off-by: Aneesh Kumar K.V Signed-off-by: Al Viro --- fs/btrfs/export.c | 8 ++++++-- fs/exportfs/expfs.c | 9 +++++++-- fs/fat/inode.c | 4 +++- fs/fuse/inode.c | 4 +++- fs/gfs2/export.c | 8 ++++++-- fs/isofs/export.c | 8 ++++++-- fs/ocfs2/export.c | 8 ++++++-- fs/reiserfs/inode.c | 7 ++++++- fs/udf/namei.c | 7 ++++++- fs/xfs/linux-2.6/xfs_export.c | 4 +++- include/linux/exportfs.h | 6 ++++-- mm/shmem.c | 4 +++- 12 files changed, 59 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c index ff27d7a477b2..b4ffad859adb 100644 --- a/fs/btrfs/export.c +++ b/fs/btrfs/export.c @@ -21,9 +21,13 @@ static int btrfs_encode_fh(struct dentry *dentry, u32 *fh, int *max_len, int len = *max_len; int type; - if ((len < BTRFS_FID_SIZE_NON_CONNECTABLE) || - (connectable && len < BTRFS_FID_SIZE_CONNECTABLE)) + if (connectable && (len < BTRFS_FID_SIZE_CONNECTABLE)) { + *max_len = BTRFS_FID_SIZE_CONNECTABLE; return 255; + } else if (len < BTRFS_FID_SIZE_NON_CONNECTABLE) { + *max_len = BTRFS_FID_SIZE_NON_CONNECTABLE; + return 255; + } len = BTRFS_FID_SIZE_NON_CONNECTABLE; type = FILEID_BTRFS_WITHOUT_PARENT; diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c index 4b6825740dd5..cfe55731b6dc 100644 --- a/fs/exportfs/expfs.c +++ b/fs/exportfs/expfs.c @@ -320,9 +320,14 @@ static int export_encode_fh(struct dentry *dentry, struct fid *fid, struct inode * inode = dentry->d_inode; int len = *max_len; int type = FILEID_INO32_GEN; - - if (len < 2 || (connectable && len < 4)) + + if (connectable && (len < 4)) { + *max_len = 4; + return 255; + } else if (len < 2) { + *max_len = 2; return 255; + } len = 2; fid->i32.ino = inode->i_ino; diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 86753fe10bd1..0e277ec4b612 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -757,8 +757,10 @@ fat_encode_fh(struct dentry *de, __u32 *fh, int *lenp, int connectable) struct inode *inode = de->d_inode; u32 ipos_h, ipos_m, ipos_l; - if (len < 5) + if (len < 5) { + *lenp = 5; return 255; /* no room */ + } ipos_h = MSDOS_I(inode)->i_pos >> 8; ipos_m = (MSDOS_I(inode)->i_pos & 0xf0) << 24; diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 9e3f68cc1bd1..051b1a084528 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -637,8 +637,10 @@ static int fuse_encode_fh(struct dentry *dentry, u32 *fh, int *max_len, u64 nodeid; u32 generation; - if (*max_len < len) + if (*max_len < len) { + *max_len = len; return 255; + } nodeid = get_fuse_inode(inode)->nodeid; generation = inode->i_generation; diff --git a/fs/gfs2/export.c b/fs/gfs2/export.c index 9023db8184f9..b5a5e60df0d5 100644 --- a/fs/gfs2/export.c +++ b/fs/gfs2/export.c @@ -36,9 +36,13 @@ static int gfs2_encode_fh(struct dentry *dentry, __u32 *p, int *len, struct super_block *sb = inode->i_sb; struct gfs2_inode *ip = GFS2_I(inode); - if (*len < GFS2_SMALL_FH_SIZE || - (connectable && *len < GFS2_LARGE_FH_SIZE)) + if (connectable && (*len < GFS2_LARGE_FH_SIZE)) { + *len = GFS2_LARGE_FH_SIZE; return 255; + } else if (*len < GFS2_SMALL_FH_SIZE) { + *len = GFS2_SMALL_FH_SIZE; + return 255; + } fh[0] = cpu_to_be32(ip->i_no_formal_ino >> 32); fh[1] = cpu_to_be32(ip->i_no_formal_ino & 0xFFFFFFFF); diff --git a/fs/isofs/export.c b/fs/isofs/export.c index ed752cb38474..dd4687ff30d0 100644 --- a/fs/isofs/export.c +++ b/fs/isofs/export.c @@ -124,9 +124,13 @@ isofs_export_encode_fh(struct dentry *dentry, * offset of the inode and the upper 16 bits of fh32[1] to * hold the offset of the parent. */ - - if (len < 3 || (connectable && len < 5)) + if (connectable && (len < 5)) { + *max_len = 5; + return 255; + } else if (len < 3) { + *max_len = 3; return 255; + } len = 3; fh32[0] = ei->i_iget5_block; diff --git a/fs/ocfs2/export.c b/fs/ocfs2/export.c index 5dbc3062b4fd..254652a9b542 100644 --- a/fs/ocfs2/export.c +++ b/fs/ocfs2/export.c @@ -197,8 +197,12 @@ static int ocfs2_encode_fh(struct dentry *dentry, u32 *fh_in, int *max_len, dentry->d_name.len, dentry->d_name.name, fh, len, connectable); - if (len < 3 || (connectable && len < 6)) { - mlog(ML_ERROR, "fh buffer is too small for encoding\n"); + if (connectable && (len < 6)) { + *max_len = 6; + type = 255; + goto bail; + } else if (len < 3) { + *max_len = 3; type = 255; goto bail; } diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 0bae036831e2..1bba24bad820 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -1593,8 +1593,13 @@ int reiserfs_encode_fh(struct dentry *dentry, __u32 * data, int *lenp, struct inode *inode = dentry->d_inode; int maxlen = *lenp; - if (maxlen < 3) + if (need_parent && (maxlen < 5)) { + *lenp = 5; return 255; + } else if (maxlen < 3) { + *lenp = 3; + return 255; + } data[0] = inode->i_ino; data[1] = le32_to_cpu(INODE_PKEY(inode)->k_dir_id); diff --git a/fs/udf/namei.c b/fs/udf/namei.c index b7c338d5e9df..f1dce848ef96 100644 --- a/fs/udf/namei.c +++ b/fs/udf/namei.c @@ -1286,8 +1286,13 @@ static int udf_encode_fh(struct dentry *de, __u32 *fh, int *lenp, struct fid *fid = (struct fid *)fh; int type = FILEID_UDF_WITHOUT_PARENT; - if (len < 3 || (connectable && len < 5)) + if (connectable && (len < 5)) { + *lenp = 5; + return 255; + } else if (len < 3) { + *lenp = 3; return 255; + } *lenp = 3; fid->udf.block = location.logicalBlockNum; diff --git a/fs/xfs/linux-2.6/xfs_export.c b/fs/xfs/linux-2.6/xfs_export.c index fc0114da7fdd..f4f878fc0083 100644 --- a/fs/xfs/linux-2.6/xfs_export.c +++ b/fs/xfs/linux-2.6/xfs_export.c @@ -89,8 +89,10 @@ xfs_fs_encode_fh( * seven combinations work. The real answer is "don't use v2". */ len = xfs_fileid_length(fileid_type); - if (*max_len < len) + if (*max_len < len) { + *max_len = len; return 255; + } *max_len = len; switch (fileid_type) { diff --git a/include/linux/exportfs.h b/include/linux/exportfs.h index 28028988c862..65afdfd31b7b 100644 --- a/include/linux/exportfs.h +++ b/include/linux/exportfs.h @@ -121,8 +121,10 @@ struct fid { * set, the encode_fh() should store sufficient information so that a good * attempt can be made to find not only the file but also it's place in the * filesystem. This typically means storing a reference to de->d_parent in - * the filehandle fragment. encode_fh() should return the number of bytes - * stored or a negative error code such as %-ENOSPC + * the filehandle fragment. encode_fh() should return the fileid_type on + * success and on error returns 255 (if the space needed to encode fh is + * greater than @max_len*4 bytes). On error @max_len contains the minimum + * size(in 4 byte unit) needed to encode the file handle. * * fh_to_dentry: * @fh_to_dentry is given a &struct super_block (@sb) and a file handle diff --git a/mm/shmem.c b/mm/shmem.c index 5ee67c990602..3437b65d6d6e 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -2144,8 +2144,10 @@ static int shmem_encode_fh(struct dentry *dentry, __u32 *fh, int *len, { struct inode *inode = dentry->d_inode; - if (*len < 3) + if (*len < 3) { + *len = 3; return 255; + } if (inode_unhashed(inode)) { /* Unfortunately insert_inode_hash is not idempotent, -- cgit v1.2.3 From f52e0c11305aa09ed56cad97ffc8f0cdc3d78b5d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 14 Mar 2011 18:56:51 -0400 Subject: New AT_... flag: AT_EMPTY_PATH For name_to_handle_at(2) we'll want both ...at()-style syscall that would be usable for non-directory descriptors (with empty relative pathname). Introduce new flag (AT_EMPTY_PATH) to deal with that and corresponding LOOKUP_EMPTY; teach user_path_at() and path_init() to deal with the latter. Signed-off-by: Al Viro --- fs/namei.c | 29 +++++++++++++++++++---------- include/linux/fcntl.h | 1 + include/linux/namei.h | 1 + 3 files changed, 21 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/fs/namei.c b/fs/namei.c index abc8d2df121c..83e92bab79a6 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -136,7 +136,7 @@ static int do_getname(const char __user *filename, char *page) return retval; } -char * getname(const char __user * filename) +static char *getname_flags(const char __user * filename, int flags) { char *tmp, *result; @@ -147,14 +147,21 @@ char * getname(const char __user * filename) result = tmp; if (retval < 0) { - __putname(tmp); - result = ERR_PTR(retval); + if (retval != -ENOENT || !(flags & LOOKUP_EMPTY)) { + __putname(tmp); + result = ERR_PTR(retval); + } } } audit_getname(result); return result; } +char *getname(const char __user * filename) +{ + return getname_flags(filename, 0); +} + #ifdef CONFIG_AUDITSYSCALL void putname(const char *name) { @@ -1544,13 +1551,15 @@ static int path_init(int dfd, const char *name, unsigned int flags, dentry = file->f_path.dentry; - retval = -ENOTDIR; - if (!S_ISDIR(dentry->d_inode->i_mode)) - goto fput_fail; + if (*name) { + retval = -ENOTDIR; + if (!S_ISDIR(dentry->d_inode->i_mode)) + goto fput_fail; - retval = file_permission(file, MAY_EXEC); - if (retval) - goto fput_fail; + retval = file_permission(file, MAY_EXEC); + if (retval) + goto fput_fail; + } nd->path = file->f_path; if (flags & LOOKUP_RCU) { @@ -1759,7 +1768,7 @@ int user_path_at(int dfd, const char __user *name, unsigned flags, struct path *path) { struct nameidata nd; - char *tmp = getname(name); + char *tmp = getname_flags(name, flags); int err = PTR_ERR(tmp); if (!IS_ERR(tmp)) { diff --git a/include/linux/fcntl.h b/include/linux/fcntl.h index a562fa5fb4e3..f550f894ba15 100644 --- a/include/linux/fcntl.h +++ b/include/linux/fcntl.h @@ -46,6 +46,7 @@ unlinking file. */ #define AT_SYMLINK_FOLLOW 0x400 /* Follow symbolic links. */ #define AT_NO_AUTOMOUNT 0x800 /* Suppress terminal automount traversal */ +#define AT_EMPTY_PATH 0x1000 /* Allow empty relative pathname */ #ifdef __KERNEL__ diff --git a/include/linux/namei.h b/include/linux/namei.h index 83cd6e5cd7dc..9c8603872c36 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -64,6 +64,7 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND}; #define LOOKUP_JUMPED 0x1000 #define LOOKUP_ROOT 0x2000 +#define LOOKUP_EMPTY 0x4000 extern int user_path_at(int, const char __user *, unsigned, struct path *); -- cgit v1.2.3 From 990d6c2d7aee921e3bce22b2d6a750fd552262be Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Sat, 29 Jan 2011 18:43:26 +0530 Subject: vfs: Add name to file handle conversion support The syscall also return mount id which can be used to lookup file system specific information such as uuid in /proc//mountinfo Signed-off-by: Aneesh Kumar K.V Signed-off-by: Al Viro --- fs/Kconfig | 2 +- fs/Makefile | 2 + fs/fhandle.c | 107 +++++++++++++++++++++++++++++++++++++++++++++++ include/linux/exportfs.h | 3 ++ include/linux/fs.h | 7 ++++ include/linux/syscalls.h | 5 ++- init/Kconfig | 12 ++++++ kernel/sys_ni.c | 3 ++ 8 files changed, 139 insertions(+), 2 deletions(-) create mode 100644 fs/fhandle.c (limited to 'include') diff --git a/fs/Kconfig b/fs/Kconfig index 3db9caa57edc..7cb53aafac1e 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -47,7 +47,7 @@ config FS_POSIX_ACL def_bool n config EXPORTFS - tristate + bool config FILE_LOCKING bool "Enable POSIX file locking API" if EXPERT diff --git a/fs/Makefile b/fs/Makefile index a7f7cef0c0c8..ba01202844c5 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -48,6 +48,8 @@ obj-$(CONFIG_FS_POSIX_ACL) += posix_acl.o xattr_acl.o obj-$(CONFIG_NFS_COMMON) += nfs_common/ obj-$(CONFIG_GENERIC_ACL) += generic_acl.o +obj-$(CONFIG_FHANDLE) += fhandle.o + obj-y += quota/ obj-$(CONFIG_PROC_FS) += proc/ diff --git a/fs/fhandle.c b/fs/fhandle.c new file mode 100644 index 000000000000..9f79e743a840 --- /dev/null +++ b/fs/fhandle.c @@ -0,0 +1,107 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include "internal.h" + +static long do_sys_name_to_handle(struct path *path, + struct file_handle __user *ufh, + int __user *mnt_id) +{ + long retval; + struct file_handle f_handle; + int handle_dwords, handle_bytes; + struct file_handle *handle = NULL; + + /* + * We need t make sure wether the file system + * support decoding of the file handle + */ + if (!path->mnt->mnt_sb->s_export_op || + !path->mnt->mnt_sb->s_export_op->fh_to_dentry) + return -EOPNOTSUPP; + + if (copy_from_user(&f_handle, ufh, sizeof(struct file_handle))) + return -EFAULT; + + if (f_handle.handle_bytes > MAX_HANDLE_SZ) + return -EINVAL; + + handle = kmalloc(sizeof(struct file_handle) + f_handle.handle_bytes, + GFP_KERNEL); + if (!handle) + return -ENOMEM; + + /* convert handle size to multiple of sizeof(u32) */ + handle_dwords = f_handle.handle_bytes >> 2; + + /* we ask for a non connected handle */ + retval = exportfs_encode_fh(path->dentry, + (struct fid *)handle->f_handle, + &handle_dwords, 0); + handle->handle_type = retval; + /* convert handle size to bytes */ + handle_bytes = handle_dwords * sizeof(u32); + handle->handle_bytes = handle_bytes; + if ((handle->handle_bytes > f_handle.handle_bytes) || + (retval == 255) || (retval == -ENOSPC)) { + /* As per old exportfs_encode_fh documentation + * we could return ENOSPC to indicate overflow + * But file system returned 255 always. So handle + * both the values + */ + /* + * set the handle size to zero so we copy only + * non variable part of the file_handle + */ + handle_bytes = 0; + retval = -EOVERFLOW; + } else + retval = 0; + /* copy the mount id */ + if (copy_to_user(mnt_id, &path->mnt->mnt_id, sizeof(*mnt_id)) || + copy_to_user(ufh, handle, + sizeof(struct file_handle) + handle_bytes)) + retval = -EFAULT; + kfree(handle); + return retval; +} + +/** + * sys_name_to_handle_at: convert name to handle + * @dfd: directory relative to which name is interpreted if not absolute + * @name: name that should be converted to handle. + * @handle: resulting file handle + * @mnt_id: mount id of the file system containing the file + * @flag: flag value to indicate whether to follow symlink or not + * + * @handle->handle_size indicate the space available to store the + * variable part of the file handle in bytes. If there is not + * enough space, the field is updated to return the minimum + * value required. + */ +SYSCALL_DEFINE5(name_to_handle_at, int, dfd, const char __user *, name, + struct file_handle __user *, handle, int __user *, mnt_id, + int, flag) +{ + struct path path; + int lookup_flags; + int err; + + if ((flag & ~(AT_SYMLINK_FOLLOW | AT_EMPTY_PATH)) != 0) + return -EINVAL; + + lookup_flags = (flag & AT_SYMLINK_FOLLOW) ? LOOKUP_FOLLOW : 0; + if (flag & AT_EMPTY_PATH) + lookup_flags |= LOOKUP_EMPTY; + err = user_path_at(dfd, name, lookup_flags, &path); + if (!err) { + err = do_sys_name_to_handle(&path, handle, mnt_id); + path_put(&path); + } + return err; +} diff --git a/include/linux/exportfs.h b/include/linux/exportfs.h index 65afdfd31b7b..33a42f24b275 100644 --- a/include/linux/exportfs.h +++ b/include/linux/exportfs.h @@ -8,6 +8,9 @@ struct inode; struct super_block; struct vfsmount; +/* limit the handle size to NFSv4 handle size now */ +#define MAX_HANDLE_SZ 128 + /* * The fileid_type identifies how the file within the filesystem is encoded. * In theory this is freely set and parsed by the filesystem, but we try to diff --git a/include/linux/fs.h b/include/linux/fs.h index b7178b05cf3a..3f64630c0e10 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -978,6 +978,13 @@ struct file { #endif }; +struct file_handle { + __u32 handle_bytes; + int handle_type; + /* file identifier */ + unsigned char f_handle[0]; +}; + #define get_file(x) atomic_long_inc(&(x)->f_count) #define fput_atomic(x) atomic_long_add_unless(&(x)->f_count, -1, 1) #define file_count(x) atomic_long_read(&(x)->f_count) diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 98664db1be47..970112613fb4 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -62,6 +62,7 @@ struct robust_list_head; struct getcpu_cache; struct old_linux_dirent; struct perf_event_attr; +struct file_handle; #include #include @@ -832,5 +833,7 @@ asmlinkage long sys_mmap_pgoff(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long pgoff); asmlinkage long sys_old_mmap(struct mmap_arg_struct __user *arg); - +asmlinkage long sys_name_to_handle_at(int dfd, const char __user *name, + struct file_handle __user *handle, + int __user *mnt_id, int flag); #endif diff --git a/init/Kconfig b/init/Kconfig index be788c0957d4..e72fa17fe559 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -287,6 +287,18 @@ config BSD_PROCESS_ACCT_V3 for processing it. A preliminary version of these tools is available at . +config FHANDLE + bool "open by fhandle syscalls" + select EXPORTFS + help + If you say Y here, a user level program will be able to map + file names to handle and then later use the handle for + different file system operations. This is useful in implementing + userspace file servers, which now track files using handles instead + of names. The handle would remain the same even if file names + get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2) + syscalls. + config TASKSTATS bool "Export task/process statistics through netlink (EXPERIMENTAL)" depends on NET diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index c782fe9924c7..4e013439ac28 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -186,3 +186,6 @@ cond_syscall(sys_perf_event_open); /* fanotify! */ cond_syscall(sys_fanotify_init); cond_syscall(sys_fanotify_mark); + +/* open by handle */ +cond_syscall(sys_name_to_handle_at); -- cgit v1.2.3 From becfd1f37544798cbdfd788f32c827160fab98c1 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Sat, 29 Jan 2011 18:43:26 +0530 Subject: vfs: Add open by file handle support [AV: duplicate of open() guts removed; file_open_root() used instead] Signed-off-by: Aneesh Kumar K.V Signed-off-by: Al Viro --- fs/compat.c | 13 ++++ fs/exportfs/expfs.c | 2 + fs/fhandle.c | 158 +++++++++++++++++++++++++++++++++++++++++++++++ fs/internal.h | 3 + include/linux/syscalls.h | 3 + kernel/sys_ni.c | 2 + 6 files changed, 181 insertions(+) (limited to 'include') diff --git a/fs/compat.c b/fs/compat.c index a071775f3bb3..c6d31a3bab88 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -2284,3 +2284,16 @@ asmlinkage long compat_sys_timerfd_gettime(int ufd, } #endif /* CONFIG_TIMERFD */ + +#ifdef CONFIG_FHANDLE +/* + * Exactly like fs/open.c:sys_open_by_handle_at(), except that it + * doesn't set the O_LARGEFILE flag. + */ +asmlinkage long +compat_sys_open_by_handle_at(int mountdirfd, + struct file_handle __user *handle, int flags) +{ + return do_handle_open(mountdirfd, handle, flags); +} +#endif diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c index cfe55731b6dc..b05acb796135 100644 --- a/fs/exportfs/expfs.c +++ b/fs/exportfs/expfs.c @@ -374,6 +374,8 @@ struct dentry *exportfs_decode_fh(struct vfsmount *mnt, struct fid *fid, /* * Try to get any dentry for the given file handle from the filesystem. */ + if (!nop || !nop->fh_to_dentry) + return ERR_PTR(-ESTALE); result = nop->fh_to_dentry(mnt->mnt_sb, fid, fh_len, fileid_type); if (!result) result = ERR_PTR(-ESTALE); diff --git a/fs/fhandle.c b/fs/fhandle.c index 9f79e743a840..bf93ad2bee07 100644 --- a/fs/fhandle.c +++ b/fs/fhandle.c @@ -5,6 +5,8 @@ #include #include #include +#include +#include #include #include "internal.h" @@ -105,3 +107,159 @@ SYSCALL_DEFINE5(name_to_handle_at, int, dfd, const char __user *, name, } return err; } + +static struct vfsmount *get_vfsmount_from_fd(int fd) +{ + struct path path; + + if (fd == AT_FDCWD) { + struct fs_struct *fs = current->fs; + spin_lock(&fs->lock); + path = fs->pwd; + mntget(path.mnt); + spin_unlock(&fs->lock); + } else { + int fput_needed; + struct file *file = fget_light(fd, &fput_needed); + if (!file) + return ERR_PTR(-EBADF); + path = file->f_path; + mntget(path.mnt); + fput_light(file, fput_needed); + } + return path.mnt; +} + +static int vfs_dentry_acceptable(void *context, struct dentry *dentry) +{ + return 1; +} + +static int do_handle_to_path(int mountdirfd, struct file_handle *handle, + struct path *path) +{ + int retval = 0; + int handle_dwords; + + path->mnt = get_vfsmount_from_fd(mountdirfd); + if (IS_ERR(path->mnt)) { + retval = PTR_ERR(path->mnt); + goto out_err; + } + /* change the handle size to multiple of sizeof(u32) */ + handle_dwords = handle->handle_bytes >> 2; + path->dentry = exportfs_decode_fh(path->mnt, + (struct fid *)handle->f_handle, + handle_dwords, handle->handle_type, + vfs_dentry_acceptable, NULL); + if (IS_ERR(path->dentry)) { + retval = PTR_ERR(path->dentry); + goto out_mnt; + } + return 0; +out_mnt: + mntput(path->mnt); +out_err: + return retval; +} + +static int handle_to_path(int mountdirfd, struct file_handle __user *ufh, + struct path *path) +{ + int retval = 0; + struct file_handle f_handle; + struct file_handle *handle = NULL; + + /* + * With handle we don't look at the execute bit on the + * the directory. Ideally we would like CAP_DAC_SEARCH. + * But we don't have that + */ + if (!capable(CAP_DAC_READ_SEARCH)) { + retval = -EPERM; + goto out_err; + } + if (copy_from_user(&f_handle, ufh, sizeof(struct file_handle))) { + retval = -EFAULT; + goto out_err; + } + if ((f_handle.handle_bytes > MAX_HANDLE_SZ) || + (f_handle.handle_bytes == 0)) { + retval = -EINVAL; + goto out_err; + } + handle = kmalloc(sizeof(struct file_handle) + f_handle.handle_bytes, + GFP_KERNEL); + if (!handle) { + retval = -ENOMEM; + goto out_err; + } + /* copy the full handle */ + if (copy_from_user(handle, ufh, + sizeof(struct file_handle) + + f_handle.handle_bytes)) { + retval = -EFAULT; + goto out_handle; + } + + retval = do_handle_to_path(mountdirfd, handle, path); + +out_handle: + kfree(handle); +out_err: + return retval; +} + +long do_handle_open(int mountdirfd, + struct file_handle __user *ufh, int open_flag) +{ + long retval = 0; + struct path path; + struct file *file; + int fd; + + retval = handle_to_path(mountdirfd, ufh, &path); + if (retval) + return retval; + + fd = get_unused_fd_flags(open_flag); + if (fd < 0) { + path_put(&path); + return fd; + } + file = file_open_root(path.dentry, path.mnt, "", open_flag); + if (IS_ERR(file)) { + put_unused_fd(fd); + retval = PTR_ERR(file); + } else { + retval = fd; + fsnotify_open(file); + fd_install(fd, file); + } + path_put(&path); + return retval; +} + +/** + * sys_open_by_handle_at: Open the file handle + * @mountdirfd: directory file descriptor + * @handle: file handle to be opened + * @flag: open flags. + * + * @mountdirfd indicate the directory file descriptor + * of the mount point. file handle is decoded relative + * to the vfsmount pointed by the @mountdirfd. @flags + * value is same as the open(2) flags. + */ +SYSCALL_DEFINE3(open_by_handle_at, int, mountdirfd, + struct file_handle __user *, handle, + int, flags) +{ + long ret; + + if (force_o_largefile()) + flags |= O_LARGEFILE; + + ret = do_handle_open(mountdirfd, handle, flags); + return ret; +} diff --git a/fs/internal.h b/fs/internal.h index 52abc5287f50..f3d15de44b15 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -117,6 +117,9 @@ extern struct file *do_filp_open(int dfd, const char *pathname, extern struct file *do_file_open_root(struct dentry *, struct vfsmount *, const char *, const struct open_flags *, int lookup_flags); +extern long do_handle_open(int mountdirfd, + struct file_handle __user *ufh, int open_flag); + /* * inode.c */ diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 970112613fb4..2d9b79c0f224 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -836,4 +836,7 @@ asmlinkage long sys_old_mmap(struct mmap_arg_struct __user *arg); asmlinkage long sys_name_to_handle_at(int dfd, const char __user *name, struct file_handle __user *handle, int __user *mnt_id, int flag); +asmlinkage long sys_open_by_handle_at(int mountdirfd, + struct file_handle __user *handle, + int flags); #endif diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index 4e013439ac28..25cc41cd8f33 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -189,3 +189,5 @@ cond_syscall(sys_fanotify_mark); /* open by handle */ cond_syscall(sys_name_to_handle_at); +cond_syscall(sys_open_by_handle_at); +cond_syscall(compat_sys_open_by_handle_at); -- cgit v1.2.3 From a51571ccb8be1b88aea502ebba8350519682c16d Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Sat, 29 Jan 2011 18:43:38 +0530 Subject: unistd.h: Add new syscalls numbers to asm-generic Signed-off-by: Aneesh Kumar K.V Signed-off-by: Al Viro --- include/asm-generic/unistd.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-generic/unistd.h b/include/asm-generic/unistd.h index b969770196c2..57af0338d270 100644 --- a/include/asm-generic/unistd.h +++ b/include/asm-generic/unistd.h @@ -646,9 +646,13 @@ __SYSCALL(__NR_prlimit64, sys_prlimit64) __SYSCALL(__NR_fanotify_init, sys_fanotify_init) #define __NR_fanotify_mark 263 __SYSCALL(__NR_fanotify_mark, sys_fanotify_mark) +#define __NR_name_to_handle_at 264 +__SYSCALL(__NR_name_to_handle_at, sys_name_to_handle_at) +#define __NR_open_by_handle_at 265 +__SYSCALL(__NR_open_by_handle_at, sys_open_by_handle_at) #undef __NR_syscalls -#define __NR_syscalls 264 +#define __NR_syscalls 266 /* * All syscalls below here should go away really, -- cgit v1.2.3 From 93f1c20bc8cdb757be50566eff88d65c3b26881f Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Sat, 29 Jan 2011 18:43:38 +0530 Subject: vfs: Export file system uuid via /proc//mountinfo We add a per superblock uuid field. File systems should update the uuid in the fill_super callback Signed-off-by: Aneesh Kumar K.V Signed-off-by: Al Viro --- fs/namespace.c | 16 ++++++++++++++++ include/linux/fs.h | 1 + 2 files changed, 17 insertions(+) (limited to 'include') diff --git a/fs/namespace.c b/fs/namespace.c index d1edf26025dc..dffe6f49ab93 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1002,6 +1002,18 @@ const struct seq_operations mounts_op = { .show = show_vfsmnt }; +static int uuid_is_nil(u8 *uuid) +{ + int i; + u8 *cp = (u8 *)uuid; + + for (i = 0; i < 16; i++) { + if (*cp++) + return 0; + } + return 1; +} + static int show_mountinfo(struct seq_file *m, void *v) { struct proc_mounts *p = m->private; @@ -1040,6 +1052,10 @@ static int show_mountinfo(struct seq_file *m, void *v) if (IS_MNT_UNBINDABLE(mnt)) seq_puts(m, " unbindable"); + if (!uuid_is_nil(mnt->mnt_sb->s_uuid)) + /* print the uuid */ + seq_printf(m, " uuid:%pU", mnt->mnt_sb->s_uuid); + /* Filesystem specific data */ seq_puts(m, " - "); show_type(m, sb); diff --git a/include/linux/fs.h b/include/linux/fs.h index 3f64630c0e10..f2143e0942c2 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1408,6 +1408,7 @@ struct super_block { wait_queue_head_t s_wait_unfrozen; char s_id[32]; /* Informational name */ + u8 s_uuid[16]; /* UUID */ void *s_fs_info; /* Filesystem private info */ fmode_t s_mode; -- cgit v1.2.3 From 1abf0c718f15a56a0a435588d1b104c7a37dc9bd Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 13 Mar 2011 03:51:11 -0400 Subject: New kind of open files - "location only". New flag for open(2) - O_PATH. Semantics: * pathname is resolved, but the file itself is _NOT_ opened as far as filesystem is concerned. * almost all operations on the resulting descriptors shall fail with -EBADF. Exceptions are: 1) operations on descriptors themselves (i.e. close(), dup(), dup2(), dup3(), fcntl(fd, F_DUPFD), fcntl(fd, F_DUPFD_CLOEXEC, ...), fcntl(fd, F_GETFD), fcntl(fd, F_SETFD, ...)) 2) fcntl(fd, F_GETFL), for a common non-destructive way to check if descriptor is open 3) "dfd" arguments of ...at(2) syscalls, i.e. the starting points of pathname resolution * closing such descriptor does *NOT* affect dnotify or posix locks. * permissions are checked as usual along the way to file; no permission checks are applied to the file itself. Of course, giving such thing to syscall will result in permission checks (at the moment it means checking that starting point of ....at() is a directory and caller has exec permissions on it). fget() and fget_light() return NULL on such descriptors; use of fget_raw() and fget_raw_light() is needed to get them. That protects existing code from dealing with those things. There are two things still missing (they come in the next commits): one is handling of symlinks (right now we refuse to open them that way; see the next commit for semantics related to those) and another is descriptor passing via SCM_RIGHTS datagrams. Signed-off-by: Al Viro --- fs/fcntl.c | 37 ++++++++++++++++++++++++++----- fs/file_table.c | 53 ++++++++++++++++++++++++++++++++++++++++----- fs/namei.c | 2 +- fs/open.c | 35 +++++++++++++++++++++++++----- include/asm-generic/fcntl.h | 4 ++++ include/linux/file.h | 2 ++ include/linux/fs.h | 3 +++ 7 files changed, 119 insertions(+), 17 deletions(-) (limited to 'include') diff --git a/fs/fcntl.c b/fs/fcntl.c index cb1026181bdc..6c82e5bac039 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -131,7 +131,7 @@ SYSCALL_DEFINE2(dup2, unsigned int, oldfd, unsigned int, newfd) SYSCALL_DEFINE1(dup, unsigned int, fildes) { int ret = -EBADF; - struct file *file = fget(fildes); + struct file *file = fget_raw(fildes); if (file) { ret = get_unused_fd(); @@ -426,15 +426,35 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg, return err; } +static int check_fcntl_cmd(unsigned cmd) +{ + switch (cmd) { + case F_DUPFD: + case F_DUPFD_CLOEXEC: + case F_GETFD: + case F_SETFD: + case F_GETFL: + return 1; + } + return 0; +} + SYSCALL_DEFINE3(fcntl, unsigned int, fd, unsigned int, cmd, unsigned long, arg) { struct file *filp; long err = -EBADF; - filp = fget(fd); + filp = fget_raw(fd); if (!filp) goto out; + if (unlikely(filp->f_mode & FMODE_PATH)) { + if (!check_fcntl_cmd(cmd)) { + fput(filp); + goto out; + } + } + err = security_file_fcntl(filp, cmd, arg); if (err) { fput(filp); @@ -456,10 +476,17 @@ SYSCALL_DEFINE3(fcntl64, unsigned int, fd, unsigned int, cmd, long err; err = -EBADF; - filp = fget(fd); + filp = fget_raw(fd); if (!filp) goto out; + if (unlikely(filp->f_mode & FMODE_PATH)) { + if (!check_fcntl_cmd(cmd)) { + fput(filp); + goto out; + } + } + err = security_file_fcntl(filp, cmd, arg); if (err) { fput(filp); @@ -808,14 +835,14 @@ static int __init fcntl_init(void) * Exceptions: O_NONBLOCK is a two bit define on parisc; O_NDELAY * is defined as O_NONBLOCK on some platforms and not on others. */ - BUILD_BUG_ON(18 - 1 /* for O_RDONLY being 0 */ != HWEIGHT32( + BUILD_BUG_ON(19 - 1 /* for O_RDONLY being 0 */ != HWEIGHT32( O_RDONLY | O_WRONLY | O_RDWR | O_CREAT | O_EXCL | O_NOCTTY | O_TRUNC | O_APPEND | /* O_NONBLOCK | */ __O_SYNC | O_DSYNC | FASYNC | O_DIRECT | O_LARGEFILE | O_DIRECTORY | O_NOFOLLOW | O_NOATIME | O_CLOEXEC | - __FMODE_EXEC + __FMODE_EXEC | O_PATH )); fasync_cache = kmem_cache_create("fasync_cache", diff --git a/fs/file_table.c b/fs/file_table.c index eb36b6b17e26..3c16e1ca163e 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -276,11 +276,10 @@ struct file *fget(unsigned int fd) rcu_read_lock(); file = fcheck_files(files, fd); if (file) { - if (!atomic_long_inc_not_zero(&file->f_count)) { - /* File object ref couldn't be taken */ - rcu_read_unlock(); - return NULL; - } + /* File object ref couldn't be taken */ + if (file->f_mode & FMODE_PATH || + !atomic_long_inc_not_zero(&file->f_count)) + file = NULL; } rcu_read_unlock(); @@ -289,6 +288,23 @@ struct file *fget(unsigned int fd) EXPORT_SYMBOL(fget); +struct file *fget_raw(unsigned int fd) +{ + struct file *file; + struct files_struct *files = current->files; + + rcu_read_lock(); + file = fcheck_files(files, fd); + if (file) { + /* File object ref couldn't be taken */ + if (!atomic_long_inc_not_zero(&file->f_count)) + file = NULL; + } + rcu_read_unlock(); + + return file; +} + /* * Lightweight file lookup - no refcnt increment if fd table isn't shared. * @@ -310,6 +326,33 @@ struct file *fget_light(unsigned int fd, int *fput_needed) struct file *file; struct files_struct *files = current->files; + *fput_needed = 0; + if (atomic_read(&files->count) == 1) { + file = fcheck_files(files, fd); + if (file && (file->f_mode & FMODE_PATH)) + file = NULL; + } else { + rcu_read_lock(); + file = fcheck_files(files, fd); + if (file) { + if (!(file->f_mode & FMODE_PATH) && + atomic_long_inc_not_zero(&file->f_count)) + *fput_needed = 1; + else + /* Didn't get the reference, someone's freed */ + file = NULL; + } + rcu_read_unlock(); + } + + return file; +} + +struct file *fget_raw_light(unsigned int fd, int *fput_needed) +{ + struct file *file; + struct files_struct *files = current->files; + *fput_needed = 0; if (atomic_read(&files->count) == 1) { file = fcheck_files(files, fd); diff --git a/fs/namei.c b/fs/namei.c index 33be51a2ddb7..e1d9f90d9776 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1544,7 +1544,7 @@ static int path_init(int dfd, const char *name, unsigned int flags, } else { struct dentry *dentry; - file = fget_light(dfd, &fput_needed); + file = fget_raw_light(dfd, &fput_needed); retval = -EBADF; if (!file) goto out_fail; diff --git a/fs/open.c b/fs/open.c index 48afc5c139d2..14a51de01f54 100644 --- a/fs/open.c +++ b/fs/open.c @@ -669,11 +669,16 @@ static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt, int (*open)(struct inode *, struct file *), const struct cred *cred) { + static const struct file_operations empty_fops = {}; struct inode *inode; int error; f->f_mode = OPEN_FMODE(f->f_flags) | FMODE_LSEEK | FMODE_PREAD | FMODE_PWRITE; + + if (unlikely(f->f_flags & O_PATH)) + f->f_mode = FMODE_PATH; + inode = dentry->d_inode; if (f->f_mode & FMODE_WRITE) { error = __get_file_write_access(inode, mnt); @@ -687,9 +692,15 @@ static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt, f->f_path.dentry = dentry; f->f_path.mnt = mnt; f->f_pos = 0; - f->f_op = fops_get(inode->i_fop); file_sb_list_add(f, inode->i_sb); + if (unlikely(f->f_mode & FMODE_PATH)) { + f->f_op = &empty_fops; + return f; + } + + f->f_op = fops_get(inode->i_fop); + error = security_dentry_open(f, cred); if (error) goto cleanup_all; @@ -911,9 +922,18 @@ static inline int build_open_flags(int flags, int mode, struct open_flags *op) if (flags & __O_SYNC) flags |= O_DSYNC; - op->open_flag = flags; + /* + * If we have O_PATH in the open flag. Then we + * cannot have anything other than the below set of flags + */ + if (flags & O_PATH) { + flags &= O_DIRECTORY | O_NOFOLLOW | O_PATH; + acc_mode = 0; + } else { + acc_mode = MAY_OPEN | ACC_MODE(flags); + } - acc_mode = MAY_OPEN | ACC_MODE(flags); + op->open_flag = flags; /* O_TRUNC implies we need access checks for write permissions */ if (flags & O_TRUNC) @@ -926,7 +946,8 @@ static inline int build_open_flags(int flags, int mode, struct open_flags *op) op->acc_mode = acc_mode; - op->intent = LOOKUP_OPEN; + op->intent = flags & O_PATH ? 0 : LOOKUP_OPEN; + if (flags & O_CREAT) { op->intent |= LOOKUP_CREATE; if (flags & O_EXCL) @@ -1053,8 +1074,10 @@ int filp_close(struct file *filp, fl_owner_t id) if (filp->f_op && filp->f_op->flush) retval = filp->f_op->flush(filp, id); - dnotify_flush(filp, id); - locks_remove_posix(filp, id); + if (likely(!(filp->f_mode & FMODE_PATH))) { + dnotify_flush(filp, id); + locks_remove_posix(filp, id); + } fput(filp); return retval; } diff --git a/include/asm-generic/fcntl.h b/include/asm-generic/fcntl.h index 0fc16e3f0bfc..84793c7025e2 100644 --- a/include/asm-generic/fcntl.h +++ b/include/asm-generic/fcntl.h @@ -80,6 +80,10 @@ #define O_SYNC (__O_SYNC|O_DSYNC) #endif +#ifndef O_PATH +#define O_PATH 010000000 +#endif + #ifndef O_NDELAY #define O_NDELAY O_NONBLOCK #endif diff --git a/include/linux/file.h b/include/linux/file.h index e85baebf6279..21a79958541c 100644 --- a/include/linux/file.h +++ b/include/linux/file.h @@ -29,6 +29,8 @@ static inline void fput_light(struct file *file, int fput_needed) extern struct file *fget(unsigned int fd); extern struct file *fget_light(unsigned int fd, int *fput_needed); +extern struct file *fget_raw(unsigned int fd); +extern struct file *fget_raw_light(unsigned int fd, int *fput_needed); extern void set_close_on_exec(unsigned int fd, int flag); extern void put_filp(struct file *); extern int alloc_fd(unsigned start, unsigned flags); diff --git a/include/linux/fs.h b/include/linux/fs.h index f2143e0942c2..13df14e2c42e 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -102,6 +102,9 @@ struct inodes_stat_t { /* File is huge (eg. /dev/kmem): treat loff_t as unsigned */ #define FMODE_UNSIGNED_OFFSET ((__force fmode_t)0x2000) +/* File is opened with O_PATH; almost nothing can be done with it */ +#define FMODE_PATH ((__force fmode_t)0x4000) + /* File was opened by fanotify and shouldn't generate fanotify events */ #define FMODE_NONOTIFY ((__force fmode_t)0x1000000) -- cgit v1.2.3