From 752343be63d90c84d275f046e43371febe217863 Mon Sep 17 00:00:00 2001 From: Rasmus Villemoes Date: Thu, 29 Oct 2015 12:01:41 +0100 Subject: fs/file.c: __const_max is actually __const_min :-) 7f4b36f9bb930 "get rid of files_defer_init()" inexplicably changed a min() to a __const_max() - but the __const_max macro actually gives the minimum... So no functional change, just less confusing naming. Signed-off-by: Rasmus Villemoes Signed-off-by: Al Viro --- fs/file.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/file.c b/fs/file.c index 39f8f15921da..1aed0add16a2 100644 --- a/fs/file.c +++ b/fs/file.c @@ -25,9 +25,9 @@ int sysctl_nr_open __read_mostly = 1024*1024; int sysctl_nr_open_min = BITS_PER_LONG; -/* our max() is unusable in constant expressions ;-/ */ -#define __const_max(x, y) ((x) < (y) ? (x) : (y)) -int sysctl_nr_open_max = __const_max(INT_MAX, ~(size_t)0/sizeof(void *)) & +/* our min() is unusable in constant expressions ;-/ */ +#define __const_min(x, y) ((x) < (y) ? (x) : (y)) +int sysctl_nr_open_max = __const_min(INT_MAX, ~(size_t)0/sizeof(void *)) & -BITS_PER_LONG; static void *alloc_fdmem(size_t size) -- cgit v1.2.1 From 25ab4c9b1ccb64b1433cecd3f19f28fe300c1576 Mon Sep 17 00:00:00 2001 From: Yaowei Bai Date: Tue, 17 Nov 2015 14:40:10 +0800 Subject: fs/namespace.c: path_is_under can be boolean This patch makes path_is_under return bool to improve readability due to this particular function only using either one or zero as its return value. No functional change. Signed-off-by: Yaowei Bai Signed-off-by: Al Viro --- fs/namespace.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/namespace.c b/fs/namespace.c index 0570729c87fd..b27156f2e68b 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -2939,9 +2939,9 @@ bool is_path_reachable(struct mount *mnt, struct dentry *dentry, return &mnt->mnt == root->mnt && is_subdir(dentry, root->dentry); } -int path_is_under(struct path *path1, struct path *path2) +bool path_is_under(struct path *path1, struct path *path2) { - int res; + bool res; read_seqlock_excl(&mount_lock); res = is_path_reachable(real_mount(path1->mnt), path1->dentry, path2); read_sequnlock_excl(&mount_lock); -- cgit v1.2.1 From a6e5787fc8fc9c88290a7bceed07aa4d14029fa7 Mon Sep 17 00:00:00 2001 From: Yaowei Bai Date: Tue, 17 Nov 2015 14:40:11 +0800 Subject: fs/dcache.c: is_subdir can be boolean This patch makes is_subdir return bool to improve readability due to this particular function only using either one or zero as its return value. No functional change. Signed-off-by: Yaowei Bai Signed-off-by: Al Viro --- fs/dcache.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/dcache.c b/fs/dcache.c index 5c33aeb0f68f..670f7896945b 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -3303,18 +3303,18 @@ out: * @new_dentry: new dentry * @old_dentry: old dentry * - * Returns 1 if new_dentry is a subdirectory of the parent (at any depth). - * Returns 0 otherwise. + * Returns true if new_dentry is a subdirectory of the parent (at any depth). + * Returns false otherwise. * Caller must ensure that "new_dentry" is pinned before calling is_subdir() */ -int is_subdir(struct dentry *new_dentry, struct dentry *old_dentry) +bool is_subdir(struct dentry *new_dentry, struct dentry *old_dentry) { - int result; + bool result; unsigned seq; if (new_dentry == old_dentry) - return 1; + return true; do { /* for restarting inner loop in case of seq retry */ @@ -3325,9 +3325,9 @@ int is_subdir(struct dentry *new_dentry, struct dentry *old_dentry) */ rcu_read_lock(); if (d_ancestor(old_dentry, new_dentry)) - result = 1; + result = true; else - result = 0; + result = false; rcu_read_unlock(); } while (read_seqretry(&rename_lock, seq)); -- cgit v1.2.1 From 0e3ef1fe453c9f29c30d040cd2559c5cac724e93 Mon Sep 17 00:00:00 2001 From: Yaowei Bai Date: Thu, 19 Nov 2015 21:00:11 +0800 Subject: fs/bad_inode.c: is_bad_inode can be boolean This patch makes is_bad_inode return bool to improve readability due to this particular function only using either one or zero as its return value. No functional change. Signed-off-by: Yaowei Bai Signed-off-by: Al Viro --- fs/bad_inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bad_inode.c b/fs/bad_inode.c index 861b1e1c4777..103f5d7c3083 100644 --- a/fs/bad_inode.c +++ b/fs/bad_inode.c @@ -192,7 +192,7 @@ EXPORT_SYMBOL(make_bad_inode); * Returns true if the inode in question has been marked as bad. */ -int is_bad_inode(struct inode *inode) +bool is_bad_inode(struct inode *inode) { return (inode->i_op == &bad_inode_ops); } -- cgit v1.2.1 From 5d9f3c7b620f6d1d9555223817bdfddfbd4b93a0 Mon Sep 17 00:00:00 2001 From: "Dmitry V. Levin" Date: Thu, 19 Nov 2015 00:57:52 +0300 Subject: vfs: show_vfsmnt: remove redundant initialization of error code As err variable is now always checked right after the first assignment, its initialization is redundant and could be safely removed. Signed-off-by: Dmitry V. Levin Signed-off-by: Al Viro --- fs/proc_namespace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/proc_namespace.c b/fs/proc_namespace.c index 8ebd9a334085..1d984b5e489e 100644 --- a/fs/proc_namespace.c +++ b/fs/proc_namespace.c @@ -95,9 +95,9 @@ static int show_vfsmnt(struct seq_file *m, struct vfsmount *mnt) { struct proc_mounts *p = m->private; struct mount *r = real_mount(mnt); - int err = 0; struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt }; struct super_block *sb = mnt_path.dentry->d_sb; + int err; if (sb->s_op->show_devname) { err = sb->s_op->show_devname(m, mnt_path.dentry); -- cgit v1.2.1 From 6ce4bca0adfde9ee404ce659d110f2bdeff9c13b Mon Sep 17 00:00:00 2001 From: "Dmitry V. Levin" Date: Thu, 19 Nov 2015 00:58:20 +0300 Subject: vfs: show_mountinfo: cleanup error code checks Check err variable right after each assignment. This change makes initialization of err redundant, so remove the initialization. Signed-off-by: Dmitry V. Levin Signed-off-by: Al Viro --- fs/proc_namespace.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/proc_namespace.c b/fs/proc_namespace.c index 1d984b5e489e..93637588edf7 100644 --- a/fs/proc_namespace.c +++ b/fs/proc_namespace.c @@ -131,16 +131,17 @@ static int show_mountinfo(struct seq_file *m, struct vfsmount *mnt) struct mount *r = real_mount(mnt); struct super_block *sb = mnt->mnt_sb; struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt }; - int err = 0; + int err; seq_printf(m, "%i %i %u:%u ", r->mnt_id, r->mnt_parent->mnt_id, MAJOR(sb->s_dev), MINOR(sb->s_dev)); - if (sb->s_op->show_path) + if (sb->s_op->show_path) { err = sb->s_op->show_path(m, mnt->mnt_root); - else + if (err) + goto out; + } else { seq_dentry(m, mnt->mnt_root, " \t\n\\"); - if (err) - goto out; + } seq_putc(m, ' '); /* mountpoints outside of chroot jail will give SEQ_SKIP on this */ @@ -168,12 +169,13 @@ static int show_mountinfo(struct seq_file *m, struct vfsmount *mnt) seq_puts(m, " - "); show_type(m, sb); seq_putc(m, ' '); - if (sb->s_op->show_devname) + if (sb->s_op->show_devname) { err = sb->s_op->show_devname(m, mnt->mnt_root); - else + if (err) + goto out; + } else { mangle(m, r->mnt_devname ? r->mnt_devname : "none"); - if (err) - goto out; + } seq_puts(m, sb->s_flags & MS_RDONLY ? " ro" : " rw"); err = show_sb_opts(m, sb); if (err) -- cgit v1.2.1 From b896fb35ca904fe5b7765fcd81a45f09a62e8d03 Mon Sep 17 00:00:00 2001 From: "Dmitry V. Levin" Date: Thu, 19 Nov 2015 00:58:32 +0300 Subject: vfs: show_vfsstat: remove redundant initialization and check of error code As err variable is now always checked right after each assignment, its initialization is redundant and could be safely removed. For the same reason, the last check of err is also redundant and could be removed as well. Signed-off-by: Dmitry V. Levin Signed-off-by: Al Viro --- fs/proc_namespace.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/proc_namespace.c b/fs/proc_namespace.c index 93637588edf7..2256e7e23e67 100644 --- a/fs/proc_namespace.c +++ b/fs/proc_namespace.c @@ -193,7 +193,7 @@ static int show_vfsstat(struct seq_file *m, struct vfsmount *mnt) struct mount *r = real_mount(mnt); struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt }; struct super_block *sb = mnt_path.dentry->d_sb; - int err = 0; + int err; /* device */ if (sb->s_op->show_devname) { @@ -222,8 +222,7 @@ static int show_vfsstat(struct seq_file *m, struct vfsmount *mnt) /* optional statistics */ if (sb->s_op->show_stats) { seq_putc(m, ' '); - if (!err) - err = sb->s_op->show_stats(m, mnt_path.dentry); + err = sb->s_op->show_stats(m, mnt_path.dentry); } seq_putc(m, '\n'); -- cgit v1.2.1 From 0125f504ed320826a78b43d636de957d703465a8 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Sat, 21 Nov 2015 16:15:37 +0100 Subject: adfs: constify adfs_dir_ops structures The adfs_dir_ops structures are never modified, so declare them as const. Done with the help of Coccinelle. Signed-off-by: Julia Lawall Signed-off-by: Al Viro --- fs/adfs/adfs.h | 6 +++--- fs/adfs/dir.c | 6 +++--- fs/adfs/dir_f.c | 2 +- fs/adfs/dir_fplus.c | 2 +- 4 files changed, 8 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/adfs/adfs.h b/fs/adfs/adfs.h index 24575d9d882d..ea4aba56f29d 100644 --- a/fs/adfs/adfs.h +++ b/fs/adfs/adfs.h @@ -45,7 +45,7 @@ struct adfs_dir_ops; struct adfs_sb_info { union { struct { struct adfs_discmap *s_map; /* bh list containing map */ - struct adfs_dir_ops *s_dir; /* directory operations */ + const struct adfs_dir_ops *s_dir; /* directory operations */ }; struct rcu_head rcu; /* used only at shutdown time */ }; @@ -168,8 +168,8 @@ void __adfs_error(struct super_block *sb, const char *function, extern const struct inode_operations adfs_dir_inode_operations; extern const struct file_operations adfs_dir_operations; extern const struct dentry_operations adfs_dentry_operations; -extern struct adfs_dir_ops adfs_f_dir_ops; -extern struct adfs_dir_ops adfs_fplus_dir_ops; +extern const struct adfs_dir_ops adfs_f_dir_ops; +extern const struct adfs_dir_ops adfs_fplus_dir_ops; extern int adfs_dir_update(struct super_block *sb, struct object_info *obj, int wait); diff --git a/fs/adfs/dir.c b/fs/adfs/dir.c index 51c279a29845..fd4cf2c48e48 100644 --- a/fs/adfs/dir.c +++ b/fs/adfs/dir.c @@ -21,7 +21,7 @@ adfs_readdir(struct file *file, struct dir_context *ctx) { struct inode *inode = file_inode(file); struct super_block *sb = inode->i_sb; - struct adfs_dir_ops *ops = ADFS_SB(sb)->s_dir; + const struct adfs_dir_ops *ops = ADFS_SB(sb)->s_dir; struct object_info obj; struct adfs_dir dir; int ret = 0; @@ -69,7 +69,7 @@ adfs_dir_update(struct super_block *sb, struct object_info *obj, int wait) { int ret = -EINVAL; #ifdef CONFIG_ADFS_FS_RW - struct adfs_dir_ops *ops = ADFS_SB(sb)->s_dir; + const struct adfs_dir_ops *ops = ADFS_SB(sb)->s_dir; struct adfs_dir dir; printk(KERN_INFO "adfs_dir_update: object %06X in dir %06X\n", @@ -129,7 +129,7 @@ static int adfs_dir_lookup_byname(struct inode *inode, struct qstr *name, struct object_info *obj) { struct super_block *sb = inode->i_sb; - struct adfs_dir_ops *ops = ADFS_SB(sb)->s_dir; + const struct adfs_dir_ops *ops = ADFS_SB(sb)->s_dir; struct adfs_dir dir; int ret; diff --git a/fs/adfs/dir_f.c b/fs/adfs/dir_f.c index 4bbe853ee50a..0fbfd0b04ae0 100644 --- a/fs/adfs/dir_f.c +++ b/fs/adfs/dir_f.c @@ -476,7 +476,7 @@ adfs_f_free(struct adfs_dir *dir) dir->sb = NULL; } -struct adfs_dir_ops adfs_f_dir_ops = { +const struct adfs_dir_ops adfs_f_dir_ops = { .read = adfs_f_read, .setpos = adfs_f_setpos, .getnext = adfs_f_getnext, diff --git a/fs/adfs/dir_fplus.c b/fs/adfs/dir_fplus.c index 82d14cdf70f9..c92cfb638c18 100644 --- a/fs/adfs/dir_fplus.c +++ b/fs/adfs/dir_fplus.c @@ -256,7 +256,7 @@ adfs_fplus_free(struct adfs_dir *dir) dir->sb = NULL; } -struct adfs_dir_ops adfs_fplus_dir_ops = { +const struct adfs_dir_ops adfs_fplus_dir_ops = { .read = adfs_fplus_read, .setpos = adfs_fplus_setpos, .getnext = adfs_fplus_getnext, -- cgit v1.2.1 From 03927c8acb63100046260711c06ba28b6b5936fb Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 25 Nov 2015 16:22:25 +0100 Subject: coredump: Use 64bit time for unix time of coredump struct timeval on 32-bit systems will have its tv_sec value overflow in year 2038 and beyond. Use a 64 bit value to print time of the coredump in seconds. ktime_get_real_seconds is chosen here for efficiency reasons. Suggested by: Arnd Bergmann Signed-off-by: Tina Ruchandani Signed-off-by: Arnd Bergmann Signed-off-by: Al Viro --- fs/coredump.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/coredump.c b/fs/coredump.c index 1777331eee76..b3c153ca435d 100644 --- a/fs/coredump.c +++ b/fs/coredump.c @@ -32,6 +32,7 @@ #include #include #include +#include #include #include @@ -232,9 +233,10 @@ static int format_corename(struct core_name *cn, struct coredump_params *cprm) break; /* UNIX time of coredump */ case 't': { - struct timeval tv; - do_gettimeofday(&tv); - err = cn_printf(cn, "%lu", tv.tv_sec); + time64_t time; + + time = ktime_get_real_seconds(); + err = cn_printf(cn, "%lld", time); break; } /* hostname */ -- cgit v1.2.1 From 57e3715cfa3fb01581555934d7191f8eabf740f4 Mon Sep 17 00:00:00 2001 From: Mike Marshall Date: Mon, 30 Nov 2015 11:11:59 -0500 Subject: typo in fs/namei.c comment Signed-off-by: Al Viro --- fs/namei.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/namei.c b/fs/namei.c index 0c3974cd3ecd..e818ed135df0 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -657,7 +657,7 @@ static bool legitimize_links(struct nameidata *nd) * Path walking has 2 modes, rcu-walk and ref-walk (see * Documentation/filesystems/path-lookup.txt). In situations when we can't * continue in RCU mode, we attempt to drop out of rcu-walk mode and grab - * normal reference counts on dentries and vfsmounts to transition to rcu-walk + * normal reference counts on dentries and vfsmounts to transition to ref-walk * mode. Refcounts are grabbed at the last known good point before rcu-walk * got stuck, so ref-walk may continue from there. If this is not successful * (eg. a seqcount has changed), then failure is returned and it's up to caller -- cgit v1.2.1 From 886f56f970f9c1563503cd71a572ba082b6a035b Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 5 Dec 2015 03:56:06 -0500 Subject: f2fs: it's umode_t, not mode_t... Signed-off-by: Al Viro --- fs/f2fs/f2fs.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 9db5500d63d9..ec6067c33a3f 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -1602,13 +1602,11 @@ static inline bool is_dot_dotdot(const struct qstr *str) static inline bool f2fs_may_extent_tree(struct inode *inode) { - mode_t mode = inode->i_mode; - if (!test_opt(F2FS_I_SB(inode), EXTENT_CACHE) || is_inode_flag_set(F2FS_I(inode), FI_NO_EXTENT)) return false; - return S_ISREG(mode); + return S_ISREG(inode->i_mode); } static inline void *f2fs_kvmalloc(size_t size, gfp_t flags) @@ -2121,7 +2119,7 @@ static inline int f2fs_sb_has_crypto(struct super_block *sb) static inline bool f2fs_may_encrypt(struct inode *inode) { #ifdef CONFIG_F2FS_FS_ENCRYPTION - mode_t mode = inode->i_mode; + umode_t mode = inode->i_mode; return (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode)); #else -- cgit v1.2.1 From 0e81ba231239b58c8b17dca18c6345c13e793824 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 5 Dec 2015 18:23:48 -0500 Subject: don't opencode iget_failed() Signed-off-by: Al Viro --- fs/ecryptfs/inode.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index e2e47ba5d313..92c5dae576b1 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -282,9 +282,7 @@ ecryptfs_create(struct inode *directory_inode, struct dentry *ecryptfs_dentry, if (rc) { ecryptfs_do_unlink(directory_inode, ecryptfs_dentry, ecryptfs_inode); - make_bad_inode(ecryptfs_inode); - unlock_new_inode(ecryptfs_inode); - iput(ecryptfs_inode); + iget_failed(ecryptfs_inode); goto out; } unlock_new_inode(ecryptfs_inode); -- cgit v1.2.1 From 9e6697e26f9888cdb6088664d31c3772b0dff0a4 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 5 Dec 2015 20:07:21 -0500 Subject: namei.c: fold set_root_rcu() into set_root() Signed-off-by: Al Viro --- fs/namei.c | 44 ++++++++++++++++++++------------------------ 1 file changed, 20 insertions(+), 24 deletions(-) (limited to 'fs') diff --git a/fs/namei.c b/fs/namei.c index e818ed135df0..f89fe5f7eac3 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -806,20 +806,20 @@ static int complete_walk(struct nameidata *nd) } static void set_root(struct nameidata *nd) -{ - get_fs_root(current->fs, &nd->root); -} - -static void set_root_rcu(struct nameidata *nd) { struct fs_struct *fs = current->fs; - unsigned seq; - do { - seq = read_seqcount_begin(&fs->seq); - nd->root = fs->root; - nd->root_seq = __read_seqcount_begin(&nd->root.dentry->d_seq); - } while (read_seqcount_retry(&fs->seq, seq)); + if (nd->flags & LOOKUP_RCU) { + unsigned seq; + + do { + seq = read_seqcount_begin(&fs->seq); + nd->root = fs->root; + nd->root_seq = __read_seqcount_begin(&nd->root.dentry->d_seq); + } while (read_seqcount_retry(&fs->seq, seq)); + } else { + get_fs_root(fs, &nd->root); + } } static void path_put_conditional(struct path *path, struct nameidata *nd) @@ -1015,10 +1015,10 @@ const char *get_link(struct nameidata *nd) } } if (*res == '/') { + if (!nd->root.mnt) + set_root(nd); if (nd->flags & LOOKUP_RCU) { struct dentry *d; - if (!nd->root.mnt) - set_root_rcu(nd); nd->path = nd->root; d = nd->path.dentry; nd->inode = d->d_inode; @@ -1026,8 +1026,6 @@ const char *get_link(struct nameidata *nd) if (unlikely(read_seqcount_retry(&d->d_seq, nd->seq))) return ERR_PTR(-ECHILD); } else { - if (!nd->root.mnt) - set_root(nd); path_put(&nd->path); nd->path = nd->root; path_get(&nd->root); @@ -1294,8 +1292,6 @@ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path, static int follow_dotdot_rcu(struct nameidata *nd) { struct inode *inode = nd->inode; - if (!nd->root.mnt) - set_root_rcu(nd); while (1) { if (path_equal(&nd->path, &nd->root)) @@ -1415,9 +1411,6 @@ static void follow_mount(struct path *path) static int follow_dotdot(struct nameidata *nd) { - if (!nd->root.mnt) - set_root(nd); - while(1) { struct dentry *old = nd->path.dentry; @@ -1655,6 +1648,8 @@ static inline int may_lookup(struct nameidata *nd) static inline int handle_dots(struct nameidata *nd, int type) { if (type == LAST_DOTDOT) { + if (!nd->root.mnt) + set_root(nd); if (nd->flags & LOOKUP_RCU) { return follow_dotdot_rcu(nd); } else @@ -2023,15 +2018,16 @@ static const char *path_init(struct nameidata *nd, unsigned flags) nd->m_seq = read_seqbegin(&mount_lock); if (*s == '/') { - if (flags & LOOKUP_RCU) { + if (flags & LOOKUP_RCU) rcu_read_lock(); - set_root_rcu(nd); + set_root(nd); + if (flags & LOOKUP_RCU) { nd->seq = nd->root_seq; + nd->path = nd->root; } else { - set_root(nd); path_get(&nd->root); + nd->path = nd->root; } - nd->path = nd->root; } else if (nd->dfd == AT_FDCWD) { if (flags & LOOKUP_RCU) { struct fs_struct *fs = current->fs; -- cgit v1.2.1 From ef55d91700d54f29b9ac301658b5b8f377ef3206 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 5 Dec 2015 20:25:06 -0500 Subject: path_init(): set nd->inode earlier in cwd-relative case that allows to kill the recheck of nd->seq on the way out in this case, and this check on the way out is left only for absolute pathnames. Signed-off-by: Al Viro --- fs/namei.c | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/namei.c b/fs/namei.c index f89fe5f7eac3..a08018b1485c 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -2028,6 +2028,15 @@ static const char *path_init(struct nameidata *nd, unsigned flags) path_get(&nd->root); nd->path = nd->root; } + nd->inode = nd->path.dentry->d_inode; + if (!(flags & LOOKUP_RCU)) + return s; + if (likely(!read_seqcount_retry(&nd->path.dentry->d_seq, nd->seq))) + return s; + if (!(nd->flags & LOOKUP_ROOT)) + nd->root.mnt = NULL; + rcu_read_unlock(); + return ERR_PTR(-ECHILD); } else if (nd->dfd == AT_FDCWD) { if (flags & LOOKUP_RCU) { struct fs_struct *fs = current->fs; @@ -2038,11 +2047,14 @@ static const char *path_init(struct nameidata *nd, unsigned flags) do { seq = read_seqcount_begin(&fs->seq); nd->path = fs->pwd; + nd->inode = nd->path.dentry->d_inode; nd->seq = __read_seqcount_begin(&nd->path.dentry->d_seq); } while (read_seqcount_retry(&fs->seq, seq)); } else { get_fs_pwd(current->fs, &nd->path); + nd->inode = nd->path.dentry->d_inode; } + return s; } else { /* Caller must check execute permissions on the starting path component */ struct fd f = fdget_raw(nd->dfd); @@ -2072,16 +2084,6 @@ static const char *path_init(struct nameidata *nd, unsigned flags) fdput(f); return s; } - - nd->inode = nd->path.dentry->d_inode; - if (!(flags & LOOKUP_RCU)) - return s; - if (likely(!read_seqcount_retry(&nd->path.dentry->d_seq, nd->seq))) - return s; - if (!(nd->flags & LOOKUP_ROOT)) - nd->root.mnt = NULL; - rcu_read_unlock(); - return ERR_PTR(-ECHILD); } static const char *trailing_symlink(struct nameidata *nd) -- cgit v1.2.1 From 248fb5b9557aa117f0b8c68b8cf2ce436e4d839d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 5 Dec 2015 20:51:58 -0500 Subject: namei.c: take "jump to root" into a new helper ... and use it both in path_init() (for absolute pathnames) and get_link() (for absolute symlinks). Signed-off-by: Al Viro --- fs/namei.c | 54 ++++++++++++++++++++++++++---------------------------- 1 file changed, 26 insertions(+), 28 deletions(-) (limited to 'fs') diff --git a/fs/namei.c b/fs/namei.c index a08018b1485c..0baf64b116bd 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -841,6 +841,26 @@ static inline void path_to_nameidata(const struct path *path, nd->path.dentry = path->dentry; } +static int nd_jump_root(struct nameidata *nd) +{ + if (nd->flags & LOOKUP_RCU) { + struct dentry *d; + nd->path = nd->root; + d = nd->path.dentry; + nd->inode = d->d_inode; + nd->seq = nd->root_seq; + if (unlikely(read_seqcount_retry(&d->d_seq, nd->seq))) + return -ECHILD; + } else { + path_put(&nd->path); + nd->path = nd->root; + path_get(&nd->path); + nd->inode = nd->path.dentry->d_inode; + } + nd->flags |= LOOKUP_JUMPED; + return 0; +} + /* * Helper to directly jump to a known parsed path from ->follow_link, * caller must have taken a reference to path beforehand. @@ -1017,21 +1037,8 @@ const char *get_link(struct nameidata *nd) if (*res == '/') { if (!nd->root.mnt) set_root(nd); - if (nd->flags & LOOKUP_RCU) { - struct dentry *d; - nd->path = nd->root; - d = nd->path.dentry; - nd->inode = d->d_inode; - nd->seq = nd->root_seq; - if (unlikely(read_seqcount_retry(&d->d_seq, nd->seq))) - return ERR_PTR(-ECHILD); - } else { - path_put(&nd->path); - nd->path = nd->root; - path_get(&nd->root); - nd->inode = nd->path.dentry->d_inode; - } - nd->flags |= LOOKUP_JUMPED; + if (unlikely(nd_jump_root(nd))) + return ERR_PTR(-ECHILD); while (unlikely(*++res == '/')) ; } @@ -2015,26 +2022,17 @@ static const char *path_init(struct nameidata *nd, unsigned flags) } nd->root.mnt = NULL; + nd->path.mnt = NULL; + nd->path.dentry = NULL; nd->m_seq = read_seqbegin(&mount_lock); if (*s == '/') { if (flags & LOOKUP_RCU) rcu_read_lock(); set_root(nd); - if (flags & LOOKUP_RCU) { - nd->seq = nd->root_seq; - nd->path = nd->root; - } else { - path_get(&nd->root); - nd->path = nd->root; - } - nd->inode = nd->path.dentry->d_inode; - if (!(flags & LOOKUP_RCU)) - return s; - if (likely(!read_seqcount_retry(&nd->path.dentry->d_seq, nd->seq))) + if (likely(!nd_jump_root(nd))) return s; - if (!(nd->flags & LOOKUP_ROOT)) - nd->root.mnt = NULL; + nd->root.mnt = NULL; rcu_read_unlock(); return ERR_PTR(-ECHILD); } else if (nd->dfd == AT_FDCWD) { -- cgit v1.2.1 From e1a63bbc40c00d5198b1c1d133b139e962f5e872 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 5 Dec 2015 21:06:33 -0500 Subject: restore_nameidata(): no need to clear now->stack microoptimization: in all callers *now is in the frame we are about to leave. Signed-off-by: Al Viro --- fs/namei.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/namei.c b/fs/namei.c index 0baf64b116bd..9e102aca3480 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -534,10 +534,8 @@ static void restore_nameidata(void) current->nameidata = old; if (old) old->total_link_count = now->total_link_count; - if (now->stack != now->internal) { + if (now->stack != now->internal) kfree(now->stack); - now->stack = now->internal; - } } static int __nd_alloc_stack(struct nameidata *nd) -- cgit v1.2.1 From b25472f9b96159cc0b9b7ed449448805973cd789 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 5 Dec 2015 22:04:48 -0500 Subject: new helpers: no_seek_end_llseek{,_size}() Signed-off-by: Al Viro --- fs/read_write.c | 39 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+) (limited to 'fs') diff --git a/fs/read_write.c b/fs/read_write.c index 819ef3faf1bb..acb171331278 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -170,6 +170,45 @@ loff_t fixed_size_llseek(struct file *file, loff_t offset, int whence, loff_t si } EXPORT_SYMBOL(fixed_size_llseek); +/** + * no_seek_end_llseek - llseek implementation for fixed-sized devices + * @file: file structure to seek on + * @offset: file offset to seek to + * @whence: type of seek + * + */ +loff_t no_seek_end_llseek(struct file *file, loff_t offset, int whence) +{ + switch (whence) { + case SEEK_SET: case SEEK_CUR: + return generic_file_llseek_size(file, offset, whence, + ~0ULL, 0); + default: + return -EINVAL; + } +} +EXPORT_SYMBOL(no_seek_end_llseek); + +/** + * no_seek_end_llseek_size - llseek implementation for fixed-sized devices + * @file: file structure to seek on + * @offset: file offset to seek to + * @whence: type of seek + * @size: maximal offset allowed + * + */ +loff_t no_seek_end_llseek_size(struct file *file, loff_t offset, int whence, loff_t size) +{ + switch (whence) { + case SEEK_SET: case SEEK_CUR: + return generic_file_llseek_size(file, offset, whence, + size, 0); + default: + return -EINVAL; + } +} +EXPORT_SYMBOL(no_seek_end_llseek_size); + /** * noop_llseek - No Operation Performed llseek implementation * @file: file structure to seek on -- cgit v1.2.1 From 16e5c1fc36040e592128a164499bc25eb138a80f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 24 Dec 2015 00:06:05 -0500 Subject: convert a bunch of open-coded instances of memdup_user_nul() A _lot_ of ->write() instances were open-coding it; some are converted to memdup_user_nul(), a lot more remain... Signed-off-by: Al Viro --- fs/afs/proc.c | 25 ++++++------------------- fs/cachefiles/daemon.c | 12 +++--------- fs/dlm/user.c | 11 +++-------- 3 files changed, 12 insertions(+), 36 deletions(-) (limited to 'fs') diff --git a/fs/afs/proc.c b/fs/afs/proc.c index 24a905b076fd..2853b4095344 100644 --- a/fs/afs/proc.c +++ b/fs/afs/proc.c @@ -230,14 +230,9 @@ static ssize_t afs_proc_cells_write(struct file *file, const char __user *buf, if (size <= 1 || size >= PAGE_SIZE) return -EINVAL; - kbuf = kmalloc(size + 1, GFP_KERNEL); - if (!kbuf) - return -ENOMEM; - - ret = -EFAULT; - if (copy_from_user(kbuf, buf, size) != 0) - goto done; - kbuf[size] = 0; + kbuf = memdup_user_nul(buf, size); + if (IS_ERR(kbuf)) + return PTR_ERR(kbuf); /* trim to first NL */ name = memchr(kbuf, '\n', size); @@ -315,15 +310,9 @@ static ssize_t afs_proc_rootcell_write(struct file *file, if (size <= 1 || size >= PAGE_SIZE) return -EINVAL; - ret = -ENOMEM; - kbuf = kmalloc(size + 1, GFP_KERNEL); - if (!kbuf) - goto nomem; - - ret = -EFAULT; - if (copy_from_user(kbuf, buf, size) != 0) - goto infault; - kbuf[size] = 0; + kbuf = memdup_user_nul(buf, size); + if (IS_ERR(kbuf)) + return PTR_ERR(kbuf); /* trim to first NL */ s = memchr(kbuf, '\n', size); @@ -337,9 +326,7 @@ static ssize_t afs_proc_rootcell_write(struct file *file, if (ret >= 0) ret = size; /* consume everything, always */ -infault: kfree(kbuf); -nomem: _leave(" = %d", ret); return ret; } diff --git a/fs/cachefiles/daemon.c b/fs/cachefiles/daemon.c index f601def05bdf..452e98dd7560 100644 --- a/fs/cachefiles/daemon.c +++ b/fs/cachefiles/daemon.c @@ -226,15 +226,9 @@ static ssize_t cachefiles_daemon_write(struct file *file, return -EOPNOTSUPP; /* drag the command string into the kernel so we can parse it */ - data = kmalloc(datalen + 1, GFP_KERNEL); - if (!data) - return -ENOMEM; - - ret = -EFAULT; - if (copy_from_user(data, _data, datalen) != 0) - goto error; - - data[datalen] = '\0'; + data = memdup_user_nul(_data, datalen); + if (IS_ERR(data)) + return PTR_ERR(data); ret = -EINVAL; if (memchr(data, '\0', datalen)) diff --git a/fs/dlm/user.c b/fs/dlm/user.c index 173b3873a4f4..1925d6d222b8 100644 --- a/fs/dlm/user.c +++ b/fs/dlm/user.c @@ -515,14 +515,9 @@ static ssize_t device_write(struct file *file, const char __user *buf, if (count > sizeof(struct dlm_write_request) + DLM_RESNAME_MAXLEN) return -EINVAL; - kbuf = kzalloc(count + 1, GFP_NOFS); - if (!kbuf) - return -ENOMEM; - - if (copy_from_user(kbuf, buf, count)) { - error = -EFAULT; - goto out_free; - } + kbuf = memdup_user_nul(buf, count); + if (!IS_ERR(kbuf)) + return PTR_ERR(kbuf); if (check_version(kbuf)) { error = -EBADE; -- cgit v1.2.1 From bb646cdb12e75d82258c2f2e7746d5952d3e321a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 24 Dec 2015 00:16:30 -0500 Subject: proc_pid_attr_write(): switch to memdup_user() Signed-off-by: Al Viro --- fs/proc/base.c | 17 +++++++---------- 1 file changed, 7 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/proc/base.c b/fs/proc/base.c index 4bd5d3118acd..1b0f470a3e35 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -2359,7 +2359,7 @@ static ssize_t proc_pid_attr_write(struct file * file, const char __user * buf, size_t count, loff_t *ppos) { struct inode * inode = file_inode(file); - char *page; + void *page; ssize_t length; struct task_struct *task = get_proc_task(inode); @@ -2374,14 +2374,11 @@ static ssize_t proc_pid_attr_write(struct file * file, const char __user * buf, if (*ppos != 0) goto out; - length = -ENOMEM; - page = (char*)__get_free_page(GFP_TEMPORARY); - if (!page) + page = memdup_user(buf, count); + if (IS_ERR(page)) { + length = PTR_ERR(page); goto out; - - length = -EFAULT; - if (copy_from_user(page, buf, count)) - goto out_free; + } /* Guard against adverse ptrace interaction */ length = mutex_lock_interruptible(&task->signal->cred_guard_mutex); @@ -2390,10 +2387,10 @@ static ssize_t proc_pid_attr_write(struct file * file, const char __user * buf, length = security_setprocattr(task, (char*)file->f_path.dentry->d_name.name, - (void*)page, count); + page, count); mutex_unlock(&task->signal->cred_guard_mutex); out_free: - free_page((unsigned long) page); + kfree(page); out: put_task_struct(task); out_no_task: -- cgit v1.2.1 From b40ef8696fbbb1107fbe5f4afc21c357f16e5ffc Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 14 Dec 2015 18:44:44 -0500 Subject: saner calling conventions for copy_mount_options() let it just return NULL, pointer to kernel copy or ERR_PTR(). Signed-off-by: Al Viro --- fs/compat.c | 21 ++++++++++----------- fs/internal.h | 2 +- fs/namespace.c | 35 +++++++++++++++++------------------ 3 files changed, 28 insertions(+), 30 deletions(-) (limited to 'fs') diff --git a/fs/compat.c b/fs/compat.c index 6fd272d455e4..a71936a3f4cb 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -792,7 +792,7 @@ COMPAT_SYSCALL_DEFINE5(mount, const char __user *, dev_name, const void __user *, data) { char *kernel_type; - unsigned long data_page; + void *options; char *kernel_dev; int retval; @@ -806,26 +806,25 @@ COMPAT_SYSCALL_DEFINE5(mount, const char __user *, dev_name, if (IS_ERR(kernel_dev)) goto out1; - retval = copy_mount_options(data, &data_page); - if (retval < 0) + options = copy_mount_options(data); + retval = PTR_ERR(options); + if (IS_ERR(options)) goto out2; - retval = -EINVAL; - - if (kernel_type && data_page) { + if (kernel_type && options) { if (!strcmp(kernel_type, NCPFS_NAME)) { - do_ncp_super_data_conv((void *)data_page); + do_ncp_super_data_conv(options); } else if (!strcmp(kernel_type, NFS4_NAME)) { - if (do_nfs4_super_data_conv((void *) data_page)) + retval = -EINVAL; + if (do_nfs4_super_data_conv(options)) goto out3; } } - retval = do_mount(kernel_dev, dir_name, kernel_type, - flags, (void*)data_page); + retval = do_mount(kernel_dev, dir_name, kernel_type, flags, options); out3: - free_page(data_page); + kfree(options); out2: kfree(kernel_dev); out1: diff --git a/fs/internal.h b/fs/internal.h index 71859c4d0b41..aa81316aaf47 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -55,7 +55,7 @@ extern int vfs_path_lookup(struct dentry *, struct vfsmount *, /* * namespace.c */ -extern int copy_mount_options(const void __user *, unsigned long *); +extern void *copy_mount_options(const void __user *); extern char *copy_mount_string(const void __user *); extern struct vfsmount *lookup_mnt(struct path *); diff --git a/fs/namespace.c b/fs/namespace.c index b27156f2e68b..33a75c8e9fcb 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -2601,18 +2601,18 @@ static long exact_copy_from_user(void *to, const void __user * from, return n; } -int copy_mount_options(const void __user * data, unsigned long *where) +void *copy_mount_options(const void __user * data) { int i; - unsigned long page; unsigned long size; + char *copy; - *where = 0; if (!data) - return 0; + return NULL; - if (!(page = __get_free_page(GFP_KERNEL))) - return -ENOMEM; + copy = kmalloc(PAGE_SIZE, GFP_KERNEL); + if (!copy) + return ERR_PTR(-ENOMEM); /* We only care that *some* data at the address the user * gave us is valid. Just in case, we'll zero @@ -2623,15 +2623,14 @@ int copy_mount_options(const void __user * data, unsigned long *where) if (size > PAGE_SIZE) size = PAGE_SIZE; - i = size - exact_copy_from_user((void *)page, data, size); + i = size - exact_copy_from_user(copy, data, size); if (!i) { - free_page(page); - return -EFAULT; + kfree(copy); + return ERR_PTR(-EFAULT); } if (i != PAGE_SIZE) - memset((char *)page + i, 0, PAGE_SIZE - i); - *where = page; - return 0; + memset(copy + i, 0, PAGE_SIZE - i); + return copy; } char *copy_mount_string(const void __user *data) @@ -2896,7 +2895,7 @@ SYSCALL_DEFINE5(mount, char __user *, dev_name, char __user *, dir_name, int ret; char *kernel_type; char *kernel_dev; - unsigned long data_page; + void *options; kernel_type = copy_mount_string(type); ret = PTR_ERR(kernel_type); @@ -2908,14 +2907,14 @@ SYSCALL_DEFINE5(mount, char __user *, dev_name, char __user *, dir_name, if (IS_ERR(kernel_dev)) goto out_dev; - ret = copy_mount_options(data, &data_page); - if (ret < 0) + options = copy_mount_options(data); + ret = PTR_ERR(options); + if (IS_ERR(options)) goto out_data; - ret = do_mount(kernel_dev, dir_name, kernel_type, flags, - (void *) data_page); + ret = do_mount(kernel_dev, dir_name, kernel_type, flags, options); - free_page(data_page); + kfree(options); out_data: kfree(kernel_dev); out_dev: -- cgit v1.2.1 From 62fb4a155f745285d9b1640c3ef53bf90c12f17c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 26 Dec 2015 22:33:24 -0500 Subject: don't carry MAY_OPEN in op->acc_mode Signed-off-by: Al Viro --- fs/exec.c | 4 ++-- fs/namei.c | 21 +++++++++------------ fs/open.c | 5 +---- 3 files changed, 12 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/exec.c b/fs/exec.c index b06623a9347f..828ec5f07de0 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -119,7 +119,7 @@ SYSCALL_DEFINE1(uselib, const char __user *, 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, + .acc_mode = MAY_READ | MAY_EXEC, .intent = LOOKUP_OPEN, .lookup_flags = LOOKUP_FOLLOW, }; @@ -763,7 +763,7 @@ static struct file *do_open_execat(int fd, struct filename *name, int flags) int err; struct open_flags open_exec_flags = { .open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC, - .acc_mode = MAY_EXEC | MAY_OPEN, + .acc_mode = MAY_EXEC, .intent = LOOKUP_OPEN, .lookup_flags = LOOKUP_FOLLOW, }; diff --git a/fs/namei.c b/fs/namei.c index 9e102aca3480..45c702edce3c 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -2663,10 +2663,6 @@ static int may_open(struct path *path, int acc_mode, int flag) struct inode *inode = dentry->d_inode; int error; - /* O_PATH? */ - if (!acc_mode) - return 0; - if (!inode) return -ENOENT; @@ -2688,7 +2684,7 @@ static int may_open(struct path *path, int acc_mode, int flag) break; } - error = inode_permission(inode, acc_mode); + error = inode_permission(inode, MAY_OPEN | acc_mode); if (error) return error; @@ -2880,7 +2876,7 @@ static int atomic_open(struct nameidata *nd, struct dentry *dentry, if (*opened & FILE_CREATED) { WARN_ON(!(open_flag & O_CREAT)); fsnotify_create(dir, dentry); - acc_mode = MAY_OPEN; + acc_mode = 0; } error = may_open(&file->f_path, acc_mode, open_flag); if (error) @@ -3093,7 +3089,7 @@ retry_lookup: /* Don't check for write permission, don't truncate */ open_flag &= ~O_TRUNC; will_truncate = false; - acc_mode = MAY_OPEN; + acc_mode = 0; path_to_nameidata(&path, nd); goto finish_open_created; } @@ -3177,10 +3173,11 @@ finish_open: got_write = true; } finish_open_created: - error = may_open(&nd->path, acc_mode, open_flag); - if (error) - goto out; - + if (likely(!(open_flag & O_PATH))) { + error = may_open(&nd->path, acc_mode, open_flag); + if (error) + goto out; + } BUG_ON(*opened & FILE_OPENED); /* once it's opened, it's opened */ error = vfs_open(&nd->path, file, current_cred()); if (!error) { @@ -3267,7 +3264,7 @@ static int do_tmpfile(struct nameidata *nd, unsigned flags, goto out2; audit_inode(nd->name, child, 0); /* Don't check for other permissions, the inode was just created */ - error = may_open(&path, MAY_OPEN, op->open_flag); + error = may_open(&path, 0, op->open_flag); if (error) goto out2; file->f_path.mnt = path.mnt; diff --git a/fs/open.c b/fs/open.c index b6f1e96a7c0b..b25b1542c530 100644 --- a/fs/open.c +++ b/fs/open.c @@ -887,7 +887,7 @@ EXPORT_SYMBOL(dentry_open); static inline int build_open_flags(int flags, umode_t mode, struct open_flags *op) { int lookup_flags = 0; - int acc_mode; + int acc_mode = ACC_MODE(flags); if (flags & (O_CREAT | __O_TMPFILE)) op->mode = (mode & S_IALLUGO) | S_IFREG; @@ -909,7 +909,6 @@ static inline int build_open_flags(int flags, umode_t mode, struct open_flags *o if (flags & __O_TMPFILE) { if ((flags & O_TMPFILE_MASK) != O_TMPFILE) return -EINVAL; - acc_mode = MAY_OPEN | ACC_MODE(flags); if (!(acc_mode & MAY_WRITE)) return -EINVAL; } else if (flags & O_PATH) { @@ -919,8 +918,6 @@ static inline int build_open_flags(int flags, umode_t mode, struct open_flags *o */ flags &= O_DIRECTORY | O_NOFOLLOW | O_PATH; acc_mode = 0; - } else { - acc_mode = MAY_OPEN | ACC_MODE(flags); } op->open_flag = flags; -- cgit v1.2.1 From 4e728cf8ffa2891913dd48e179c132fb9fa07e0e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 29 Dec 2015 15:38:25 -0500 Subject: hpfs: missing endianness annotation Signed-off-by: Al Viro --- fs/hpfs/map.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/hpfs/map.c b/fs/hpfs/map.c index a69bbc1e87f8..a136929189f0 100644 --- a/fs/hpfs/map.c +++ b/fs/hpfs/map.c @@ -133,7 +133,7 @@ __le32 *hpfs_load_bitmap_directory(struct super_block *s, secno bmp) void hpfs_load_hotfix_map(struct super_block *s, struct hpfs_spare_block *spareblock) { struct quad_buffer_head qbh; - u32 *directory; + __le32 *directory; u32 n_hotfixes, n_used_hotfixes; unsigned i; -- cgit v1.2.1 From 76e8d7cb71d415581402dbc5c5bbbbcc28ca0e47 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 2 Jan 2016 14:22:04 -0500 Subject: jfs: microoptimize get_zeroed_page / virt_to_page get_zeroed_page does alloc_page and returns page_address of the result; subsequent virt_to_page will recover the page, but since the caller needs both page and its page_address() anyway, why bother going through that wrapper at all? Signed-off-by: Al Viro --- fs/jfs/jfs_logmgr.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c index a69bdf2a1085..a270cb7ff4e0 100644 --- a/fs/jfs/jfs_logmgr.c +++ b/fs/jfs/jfs_logmgr.c @@ -1835,17 +1835,16 @@ static int lbmLogInit(struct jfs_log * log) for (i = 0; i < LOGPAGES;) { char *buffer; uint offset; - struct page *page; + struct page *page = alloc_page(GFP_KERNEL | __GFP_ZERO); - buffer = (char *) get_zeroed_page(GFP_KERNEL); - if (buffer == NULL) + if (!page) goto error; - page = virt_to_page(buffer); + buffer = page_address(page); for (offset = 0; offset < PAGE_SIZE; offset += LOGPSIZE) { lbuf = kmalloc(sizeof(struct lbuf), GFP_KERNEL); if (lbuf == NULL) { if (offset == 0) - free_page((unsigned long) buffer); + __free_page(page); goto error; } if (offset) /* we already have one reference */ -- cgit v1.2.1 From 80f8dccf95147f9668eee021425ecf3ff6432c51 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 2 Jan 2016 14:29:23 -0500 Subject: HFS wants 8Kb per-superblock allocation; just use kmalloc() ... rather than play with __get_free_pages() (and figuring out the allocation order, etc.) Signed-off-by: Al Viro --- fs/hfs/mdb.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/hfs/mdb.c b/fs/hfs/mdb.c index aa3f0d6d043c..a3ec3ae7d347 100644 --- a/fs/hfs/mdb.c +++ b/fs/hfs/mdb.c @@ -166,7 +166,7 @@ int hfs_mdb_get(struct super_block *sb) pr_warn("continuing without an alternate MDB\n"); } - HFS_SB(sb)->bitmap = (__be32 *)__get_free_pages(GFP_KERNEL, PAGE_SIZE < 8192 ? 1 : 0); + HFS_SB(sb)->bitmap = kmalloc(8192, GFP_KERNEL); if (!HFS_SB(sb)->bitmap) goto out; @@ -360,7 +360,7 @@ void hfs_mdb_put(struct super_block *sb) unload_nls(HFS_SB(sb)->nls_io); unload_nls(HFS_SB(sb)->nls_disk); - free_pages((unsigned long)HFS_SB(sb)->bitmap, PAGE_SIZE < 8192 ? 1 : 0); + kfree(HFS_SB(sb)->bitmap); kfree(HFS_SB(sb)); sb->s_fs_info = NULL; } -- cgit v1.2.1 From ccec5ee302d5cbd0273eb1c79bc935a8e3f873c6 Mon Sep 17 00:00:00 2001 From: Mateusz Guzik Date: Wed, 6 Jan 2016 06:41:53 +0100 Subject: poll: plug an unused argument to do_poll Number of fds is already known based on passed list. No functional changes. Signed-off-by: Mateusz Guzik Signed-off-by: Al Viro --- fs/select.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/select.c b/fs/select.c index 015547330e88..79d0d4953cad 100644 --- a/fs/select.c +++ b/fs/select.c @@ -778,8 +778,8 @@ static inline unsigned int do_pollfd(struct pollfd *pollfd, poll_table *pwait, return mask; } -static int do_poll(unsigned int nfds, struct poll_list *list, - struct poll_wqueues *wait, struct timespec *end_time) +static int do_poll(struct poll_list *list, struct poll_wqueues *wait, + struct timespec *end_time) { poll_table* pt = &wait->pt; ktime_t expire, *to = NULL; @@ -908,7 +908,7 @@ int do_sys_poll(struct pollfd __user *ufds, unsigned int nfds, } poll_initwait(&table); - fdcount = do_poll(nfds, head, &table, end_time); + fdcount = do_poll(head, &table, end_time); poll_freewait(&table); for (walk = head; walk; walk = walk->next) { -- cgit v1.2.1 From 424081f3c881ca3aef50cfa571e91863c10d952a Mon Sep 17 00:00:00 2001 From: Dmitry Monakhov Date: Mon, 13 Apr 2015 16:31:34 +0400 Subject: fs: use gendisk->disk_name where possible gendisk with part==0 is obviously gendisk->disk_name. Signed-off-by: Dmitry Monakhov Signed-off-by: Al Viro --- fs/block_dev.c | 12 +++--------- 1 file changed, 3 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/block_dev.c b/fs/block_dev.c index 44d4a1e9244e..01b8e0d4b4ff 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -1042,12 +1042,9 @@ EXPORT_SYMBOL_GPL(bd_unlink_disk_holder); static void flush_disk(struct block_device *bdev, bool kill_dirty) { if (__invalidate_device(bdev, kill_dirty)) { - char name[BDEVNAME_SIZE] = ""; - - if (bdev->bd_disk) - disk_name(bdev->bd_disk, 0, name); printk(KERN_WARNING "VFS: busy inodes on changed media or " - "resized disk %s\n", name); + "resized disk %s\n", + bdev->bd_disk ? bdev->bd_disk->disk_name : ""); } if (!bdev->bd_disk) @@ -1071,12 +1068,9 @@ void check_disk_size_change(struct gendisk *disk, struct block_device *bdev) disk_size = (loff_t)get_capacity(disk) << 9; bdev_size = i_size_read(bdev->bd_inode); if (disk_size != bdev_size) { - char name[BDEVNAME_SIZE]; - - disk_name(disk, 0, name); printk(KERN_INFO "%s: detected capacity change from %lld to %lld\n", - name, bdev_size, disk_size); + disk->disk_name, bdev_size, disk_size); i_size_write(bdev->bd_inode, disk_size); flush_disk(bdev, false); } -- cgit v1.2.1 From a1c6f05733c27ba7067c06c095f49e8732a5ae17 Mon Sep 17 00:00:00 2001 From: Dmitry Monakhov Date: Mon, 13 Apr 2015 16:31:37 +0400 Subject: fs: use block_device name vsprintf helper Signed-off-by: Dmitry Monakhov Signed-off-by: Al Viro --- fs/btrfs/super.c | 4 +--- fs/buffer.c | 21 ++++++--------------- fs/ext2/xattr.c | 6 ++---- fs/ext4/page-io.c | 5 ++--- fs/ext4/xattr.c | 6 ++---- fs/f2fs/debug.c | 6 ++---- fs/gfs2/ops_fstype.c | 4 +--- fs/jbd2/transaction.c | 6 ++---- fs/minix/itree_v1.c | 9 ++++----- fs/minix/itree_v2.c | 9 ++++----- fs/nilfs2/super.c | 6 ++---- fs/reiserfs/journal.c | 24 ++++++++++-------------- fs/reiserfs/prints.c | 9 +++------ fs/reiserfs/procfs.c | 5 ++--- fs/squashfs/super.c | 7 +++---- fs/super.c | 4 +--- fs/xfs/xfs_buf.c | 8 ++------ 17 files changed, 49 insertions(+), 90 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 24154e422945..a0434c179ea9 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -1514,9 +1514,7 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, if ((flags ^ s->s_flags) & MS_RDONLY) error = -EBUSY; } else { - char b[BDEVNAME_SIZE]; - - strlcpy(s->s_id, bdevname(bdev, b), sizeof(s->s_id)); + snprintf(s->s_id, sizeof(s->s_id), "%pg", bdev); btrfs_sb(s)->bdev_holder = fs_type; error = btrfs_fill_super(s, fs_devices, data, flags & MS_SILENT ? 1 : 0); diff --git a/fs/buffer.c b/fs/buffer.c index 4f4cd959da7c..e1632abb4ca9 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -134,13 +134,10 @@ __clear_page_buffers(struct page *page) static void buffer_io_error(struct buffer_head *bh, char *msg) { - char b[BDEVNAME_SIZE]; - if (!test_bit(BH_Quiet, &bh->b_state)) printk_ratelimited(KERN_ERR - "Buffer I/O error on dev %s, logical block %llu%s\n", - bdevname(bh->b_bdev, b), - (unsigned long long)bh->b_blocknr, msg); + "Buffer I/O error on dev %pg, logical block %llu%s\n", + bh->b_bdev, (unsigned long long)bh->b_blocknr, msg); } /* @@ -237,15 +234,13 @@ __find_get_block_slow(struct block_device *bdev, sector_t block) * elsewhere, don't buffer_error if we had some unmapped buffers */ if (all_mapped) { - char b[BDEVNAME_SIZE]; - printk("__find_get_block_slow() failed. " "block=%llu, b_blocknr=%llu\n", (unsigned long long)block, (unsigned long long)bh->b_blocknr); printk("b_state=0x%08lx, b_size=%zu\n", bh->b_state, bh->b_size); - printk("device %s blocksize: %d\n", bdevname(bdev, b), + printk("device %pg blocksize: %d\n", bdev, 1 << bd_inode->i_blkbits); } out_unlock: @@ -531,10 +526,8 @@ repeat: static void do_thaw_one(struct super_block *sb, void *unused) { - char b[BDEVNAME_SIZE]; while (sb->s_bdev && !thaw_bdev(sb->s_bdev, sb)) - printk(KERN_WARNING "Emergency Thaw on %s\n", - bdevname(sb->s_bdev, b)); + printk(KERN_WARNING "Emergency Thaw on %pg\n", sb->s_bdev); } static void do_thaw_all(struct work_struct *work) @@ -1074,12 +1067,10 @@ grow_buffers(struct block_device *bdev, sector_t block, int size, gfp_t gfp) * pagecache index. (this comparison is done using sector_t types). */ if (unlikely(index != block >> sizebits)) { - char b[BDEVNAME_SIZE]; - printk(KERN_ERR "%s: requested out-of-range block %llu for " - "device %s\n", + "device %pg\n", __func__, (unsigned long long)block, - bdevname(bdev, b)); + bdev); return -EIO; } diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c index fa70848afa8f..a21c259b5fde 100644 --- a/fs/ext2/xattr.c +++ b/fs/ext2/xattr.c @@ -77,10 +77,8 @@ printk("\n"); \ } while (0) # define ea_bdebug(bh, f...) do { \ - char b[BDEVNAME_SIZE]; \ - printk(KERN_DEBUG "block %s:%lu: ", \ - bdevname(bh->b_bdev, b), \ - (unsigned long) bh->b_blocknr); \ + printk(KERN_DEBUG "block %pg:%lu: ", \ + bh->b_bdev, (unsigned long) bh->b_blocknr); \ printk(f); \ printk("\n"); \ } while (0) diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c index 17fbe3882b8e..090b3498638e 100644 --- a/fs/ext4/page-io.c +++ b/fs/ext4/page-io.c @@ -52,9 +52,8 @@ void ext4_exit_pageio(void) */ static void buffer_io_error(struct buffer_head *bh) { - char b[BDEVNAME_SIZE]; - printk_ratelimited(KERN_ERR "Buffer I/O error on device %s, logical block %llu\n", - bdevname(bh->b_bdev, b), + printk_ratelimited(KERN_ERR "Buffer I/O error on device %pg, logical block %llu\n", + bh->b_bdev, (unsigned long long)bh->b_blocknr); } diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 6b6b3e751f8c..ec4e50ce0733 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -68,10 +68,8 @@ printk("\n"); \ } while (0) # define ea_bdebug(bh, f...) do { \ - char b[BDEVNAME_SIZE]; \ - printk(KERN_DEBUG "block %s:%lu: ", \ - bdevname(bh->b_bdev, b), \ - (unsigned long) bh->b_blocknr); \ + printk(KERN_DEBUG "block %pg:%lu: ", \ + bh->b_bdev, (unsigned long) bh->b_blocknr); \ printk(f); \ printk("\n"); \ } while (0) diff --git a/fs/f2fs/debug.c b/fs/f2fs/debug.c index 478e5d54154f..ad1b18a7705b 100644 --- a/fs/f2fs/debug.c +++ b/fs/f2fs/debug.c @@ -211,12 +211,10 @@ static int stat_show(struct seq_file *s, void *v) mutex_lock(&f2fs_stat_mutex); list_for_each_entry(si, &f2fs_stat_list, stat_list) { - char devname[BDEVNAME_SIZE]; - update_general_status(si->sbi); - seq_printf(s, "\n=====[ partition info(%s). #%d ]=====\n", - bdevname(si->sbi->sb->s_bdev, devname), i++); + seq_printf(s, "\n=====[ partition info(%pg). #%d ]=====\n", + si->sbi->sb->s_bdev, i++); seq_printf(s, "[SB: 1] [CP: 2] [SIT: %d] [NAT: %d] ", si->sit_area_segs, si->nat_area_segs); seq_printf(s, "[SSA: %d] [MAIN: %d", diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index baab99b69d8a..001c66641243 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -1315,9 +1315,7 @@ static struct dentry *gfs2_mount(struct file_system_type *fs_type, int flags, if ((flags ^ s->s_flags) & MS_RDONLY) goto error_super; } else { - char b[BDEVNAME_SIZE]; - - strlcpy(s->s_id, bdevname(bdev, b), sizeof(s->s_id)); + snprintf(s->s_id, sizeof(s->s_id), "%pg", bdev); sb_set_blocksize(s, block_size(bdev)); error = fill_super(s, &args, flags & MS_SILENT ? 1 : 0); if (error) diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index ca181e81c765..081dff087fc0 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -764,13 +764,11 @@ void jbd2_journal_unlock_updates (journal_t *journal) static void warn_dirty_buffer(struct buffer_head *bh) { - char b[BDEVNAME_SIZE]; - printk(KERN_WARNING - "JBD2: Spotted dirty metadata buffer (dev = %s, blocknr = %llu). " + "JBD2: Spotted dirty metadata buffer (dev = %pg, blocknr = %llu). " "There's a risk of filesystem corruption in case of system " "crash.\n", - bdevname(bh->b_bdev, b), (unsigned long long)bh->b_blocknr); + bh->b_bdev, (unsigned long long)bh->b_blocknr); } /* Call t_frozen trigger and copy buffer data into jh->b_frozen_data. */ diff --git a/fs/minix/itree_v1.c b/fs/minix/itree_v1.c index 282e15ad8cd8..46ca39d6c735 100644 --- a/fs/minix/itree_v1.c +++ b/fs/minix/itree_v1.c @@ -24,16 +24,15 @@ static inline block_t *i_data(struct inode *inode) static int block_to_path(struct inode * inode, long block, int offsets[DEPTH]) { int n = 0; - char b[BDEVNAME_SIZE]; if (block < 0) { - printk("MINIX-fs: block_to_path: block %ld < 0 on dev %s\n", - block, bdevname(inode->i_sb->s_bdev, b)); + printk("MINIX-fs: block_to_path: block %ld < 0 on dev %pg\n", + block, inode->i_sb->s_bdev); } else if (block >= (minix_sb(inode->i_sb)->s_max_size/BLOCK_SIZE)) { if (printk_ratelimit()) printk("MINIX-fs: block_to_path: " - "block %ld too big on dev %s\n", - block, bdevname(inode->i_sb->s_bdev, b)); + "block %ld too big on dev %pg\n", + block, inode->i_sb->s_bdev); } else if (block < 7) { offsets[n++] = block; } else if ((block -= 7) < 512) { diff --git a/fs/minix/itree_v2.c b/fs/minix/itree_v2.c index 78e2d93e5c83..1ee101352586 100644 --- a/fs/minix/itree_v2.c +++ b/fs/minix/itree_v2.c @@ -26,18 +26,17 @@ static inline block_t *i_data(struct inode *inode) static int block_to_path(struct inode * inode, long block, int offsets[DEPTH]) { int n = 0; - char b[BDEVNAME_SIZE]; struct super_block *sb = inode->i_sb; if (block < 0) { - printk("MINIX-fs: block_to_path: block %ld < 0 on dev %s\n", - block, bdevname(sb->s_bdev, b)); + printk("MINIX-fs: block_to_path: block %ld < 0 on dev %pg\n", + block, sb->s_bdev); } else if ((u64)block * (u64)sb->s_blocksize >= minix_sb(sb)->s_max_size) { if (printk_ratelimit()) printk("MINIX-fs: block_to_path: " - "block %ld too big on dev %s\n", - block, bdevname(sb->s_bdev, b)); + "block %ld too big on dev %pg\n", + block, sb->s_bdev); } else if (block < DIRCOUNT) { offsets[n++] = block; } else if ((block -= DIRCOUNT) < INDIRCOUNT(sb)) { diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c index 354013ea22ec..c7343844e6b6 100644 --- a/fs/nilfs2/super.c +++ b/fs/nilfs2/super.c @@ -1316,13 +1316,11 @@ nilfs_mount(struct file_system_type *fs_type, int flags, } if (!s->s_root) { - char b[BDEVNAME_SIZE]; - - s_new = true; + s_new = true; /* New superblock instance created */ s->s_mode = mode; - strlcpy(s->s_id, bdevname(sd.bdev, b), sizeof(s->s_id)); + snprintf(s->s_id, sizeof(s->s_id), "%pg", sd.bdev); sb_set_blocksize(s, block_size(sd.bdev)); err = nilfs_fill_super(s, data, flags & MS_SILENT ? 1 : 0); diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 9d6486d416a3..44c2bdced1c8 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -618,12 +618,10 @@ static void release_buffer_page(struct buffer_head *bh) static void reiserfs_end_buffer_io_sync(struct buffer_head *bh, int uptodate) { - char b[BDEVNAME_SIZE]; - if (buffer_journaled(bh)) { reiserfs_warning(NULL, "clm-2084", - "pinned buffer %lu:%s sent to disk", - bh->b_blocknr, bdevname(bh->b_bdev, b)); + "pinned buffer %lu:%pg sent to disk", + bh->b_blocknr, bh->b_bdev); } if (uptodate) set_buffer_uptodate(bh); @@ -2387,11 +2385,10 @@ static int journal_read(struct super_block *sb) int replay_count = 0; int continue_replay = 1; int ret; - char b[BDEVNAME_SIZE]; cur_dblock = SB_ONDISK_JOURNAL_1st_BLOCK(sb); - reiserfs_info(sb, "checking transaction log (%s)\n", - bdevname(journal->j_dev_bd, b)); + reiserfs_info(sb, "checking transaction log (%pg)\n", + journal->j_dev_bd); start = get_seconds(); /* @@ -2651,8 +2648,8 @@ static int journal_init_dev(struct super_block *super, set_blocksize(journal->j_dev_bd, super->s_blocksize); reiserfs_info(super, - "journal_init_dev: journal device: %s\n", - bdevname(journal->j_dev_bd, b)); + "journal_init_dev: journal device: %pg\n", + journal->j_dev_bd); return 0; } @@ -2724,7 +2721,6 @@ int journal_init(struct super_block *sb, const char *j_dev_name, struct reiserfs_journal_header *jh; struct reiserfs_journal *journal; struct reiserfs_journal_list *jl; - char b[BDEVNAME_SIZE]; int ret; journal = SB_JOURNAL(sb) = vzalloc(sizeof(struct reiserfs_journal)); @@ -2794,10 +2790,10 @@ int journal_init(struct super_block *sb, const char *j_dev_name, && (le32_to_cpu(jh->jh_journal.jp_journal_magic) != sb_jp_journal_magic(rs))) { reiserfs_warning(sb, "sh-460", - "journal header magic %x (device %s) does " + "journal header magic %x (device %pg) does " "not match to magic found in super block %x", jh->jh_journal.jp_journal_magic, - bdevname(journal->j_dev_bd, b), + journal->j_dev_bd, sb_jp_journal_magic(rs)); brelse(bhjh); goto free_and_return; @@ -2818,10 +2814,10 @@ int journal_init(struct super_block *sb, const char *j_dev_name, journal->j_max_trans_age = commit_max_age; } - reiserfs_info(sb, "journal params: device %s, size %u, " + reiserfs_info(sb, "journal params: device %pg, size %u, " "journal first block %u, max trans len %u, max batch %u, " "max commit age %u, max trans age %u\n", - bdevname(journal->j_dev_bd, b), + journal->j_dev_bd, SB_ONDISK_JOURNAL_SIZE(sb), SB_ONDISK_JOURNAL_1st_BLOCK(sb), journal->j_trans_max, diff --git a/fs/reiserfs/prints.c b/fs/reiserfs/prints.c index ae1dc841db3a..4f3f928076f3 100644 --- a/fs/reiserfs/prints.c +++ b/fs/reiserfs/prints.c @@ -139,11 +139,9 @@ static void sprintf_block_head(char *buf, struct buffer_head *bh) static void sprintf_buffer_head(char *buf, struct buffer_head *bh) { - char b[BDEVNAME_SIZE]; - sprintf(buf, - "dev %s, size %zd, blocknr %llu, count %d, state 0x%lx, page %p, (%s, %s, %s)", - bdevname(bh->b_bdev, b), bh->b_size, + "dev %pg, size %zd, blocknr %llu, count %d, state 0x%lx, page %p, (%s, %s, %s)", + bh->b_bdev, bh->b_size, (unsigned long long)bh->b_blocknr, atomic_read(&(bh->b_count)), bh->b_state, bh->b_page, buffer_uptodate(bh) ? "UPTODATE" : "!UPTODATE", @@ -530,7 +528,6 @@ static int print_super_block(struct buffer_head *bh) (struct reiserfs_super_block *)(bh->b_data); int skipped, data_blocks; char *version; - char b[BDEVNAME_SIZE]; if (is_reiserfs_3_5(rs)) { version = "3.5"; @@ -543,7 +540,7 @@ static int print_super_block(struct buffer_head *bh) return 1; } - printk("%s\'s super block is in block %llu\n", bdevname(bh->b_bdev, b), + printk("%pg\'s super block is in block %llu\n", bh->b_bdev, (unsigned long long)bh->b_blocknr); printk("Reiserfs version %s\n", version); printk("Block count %u\n", sb_block_count(rs)); diff --git a/fs/reiserfs/procfs.c b/fs/reiserfs/procfs.c index 621b9f381fe1..fe999157dd97 100644 --- a/fs/reiserfs/procfs.c +++ b/fs/reiserfs/procfs.c @@ -303,11 +303,10 @@ static int show_journal(struct seq_file *m, void *unused) struct reiserfs_sb_info *r = REISERFS_SB(sb); struct reiserfs_super_block *rs = r->s_rs; struct journal_params *jp = &rs->s_v1.s_journal; - char b[BDEVNAME_SIZE]; seq_printf(m, /* on-disk fields */ "jp_journal_1st_block: \t%i\n" - "jp_journal_dev: \t%s[%x]\n" + "jp_journal_dev: \t%pg[%x]\n" "jp_journal_size: \t%i\n" "jp_journal_trans_max: \t%i\n" "jp_journal_magic: \t%i\n" @@ -348,7 +347,7 @@ static int show_journal(struct seq_file *m, void *unused) "prepare: \t%12lu\n" "prepare_retry: \t%12lu\n", DJP(jp_journal_1st_block), - bdevname(SB_JOURNAL(sb)->j_dev_bd, b), + SB_JOURNAL(sb)->j_dev_bd, DJP(jp_journal_dev), DJP(jp_journal_size), DJP(jp_journal_trans_max), diff --git a/fs/squashfs/super.c b/fs/squashfs/super.c index 5056babe00df..dded920cbc8f 100644 --- a/fs/squashfs/super.c +++ b/fs/squashfs/super.c @@ -80,7 +80,6 @@ static int squashfs_fill_super(struct super_block *sb, void *data, int silent) { struct squashfs_sb_info *msblk; struct squashfs_super_block *sblk = NULL; - char b[BDEVNAME_SIZE]; struct inode *root; long long root_inode; unsigned short flags; @@ -124,8 +123,8 @@ static int squashfs_fill_super(struct super_block *sb, void *data, int silent) sb->s_magic = le32_to_cpu(sblk->s_magic); if (sb->s_magic != SQUASHFS_MAGIC) { if (!silent) - ERROR("Can't find a SQUASHFS superblock on %s\n", - bdevname(sb->s_bdev, b)); + ERROR("Can't find a SQUASHFS superblock on %pg\n", + sb->s_bdev); goto failed_mount; } @@ -178,7 +177,7 @@ static int squashfs_fill_super(struct super_block *sb, void *data, int silent) msblk->inodes = le32_to_cpu(sblk->inodes); flags = le16_to_cpu(sblk->flags); - TRACE("Found valid superblock on %s\n", bdevname(sb->s_bdev, b)); + TRACE("Found valid superblock on %pg\n", sb->s_bdev); TRACE("Inodes are %scompressed\n", SQUASHFS_UNCOMPRESSED_INODES(flags) ? "un" : ""); TRACE("Data is %scompressed\n", SQUASHFS_UNCOMPRESSED_DATA(flags) diff --git a/fs/super.c b/fs/super.c index 954aeb80e202..cc658a20a29e 100644 --- a/fs/super.c +++ b/fs/super.c @@ -1012,10 +1012,8 @@ struct dentry *mount_bdev(struct file_system_type *fs_type, blkdev_put(bdev, mode); down_write(&s->s_umount); } else { - char b[BDEVNAME_SIZE]; - s->s_mode = mode; - strlcpy(s->s_id, bdevname(bdev, b), sizeof(s->s_id)); + snprintf(s->s_id, sizeof(s->s_id), "%pg", bdev); sb_set_blocksize(s, block_size(bdev)); error = fill_super(s, data, flags & MS_SILENT ? 1 : 0); if (error) { diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index 3243cdf97f33..ace91e7c713e 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -1632,13 +1632,9 @@ xfs_setsize_buftarg( btp->bt_meta_sectormask = sectorsize - 1; if (set_blocksize(btp->bt_bdev, sectorsize)) { - char name[BDEVNAME_SIZE]; - - bdevname(btp->bt_bdev, name); - xfs_warn(btp->bt_mount, - "Cannot set_blocksize to %u on device %s", - sectorsize, name); + "Cannot set_blocksize to %u on device %pg", + sectorsize, btp->bt_bdev); return -EINVAL; } -- cgit v1.2.1 From 263a3df18fa2c078ab3e8b14fece6cfa6053e86a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 7 Jan 2016 10:04:37 -0500 Subject: nbd: use ->compat_ioctl() Signed-off-by: Al Viro --- fs/compat_ioctl.c | 11 ----------- 1 file changed, 11 deletions(-) (limited to 'fs') diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index 9144b779d10e..c9fca9c33ff1 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -1305,12 +1305,6 @@ COMPATIBLE_IOCTL(PCIIOC_CONTROLLER) COMPATIBLE_IOCTL(PCIIOC_MMAP_IS_IO) COMPATIBLE_IOCTL(PCIIOC_MMAP_IS_MEM) COMPATIBLE_IOCTL(PCIIOC_WRITE_COMBINE) -/* NBD */ -COMPATIBLE_IOCTL(NBD_DO_IT) -COMPATIBLE_IOCTL(NBD_CLEAR_SOCK) -COMPATIBLE_IOCTL(NBD_CLEAR_QUE) -COMPATIBLE_IOCTL(NBD_PRINT_DEBUG) -COMPATIBLE_IOCTL(NBD_DISCONNECT) /* i2c */ COMPATIBLE_IOCTL(I2C_SLAVE) COMPATIBLE_IOCTL(I2C_SLAVE_FORCE) @@ -1529,11 +1523,6 @@ static long do_ioctl_trans(unsigned int cmd, case KDSKBMETA: case KDSKBLED: case KDSETLED: - /* NBD */ - case NBD_SET_SOCK: - case NBD_SET_BLKSIZE: - case NBD_SET_SIZE: - case NBD_SET_SIZE_BLOCKS: return vfs_ioctl(file, cmd, arg); } -- cgit v1.2.1 From 0b2a6f231dcbc7bd543a5aaa23b0ea76ce16e585 Mon Sep 17 00:00:00 2001 From: Richard Weinberger Date: Sat, 2 Jan 2016 23:09:47 +0100 Subject: fs: xattr: Use kvfree() ... instead of open coding it. Signed-off-by: Richard Weinberger Signed-off-by: Al Viro --- fs/xattr.c | 38 ++++++++++++++------------------------ 1 file changed, 14 insertions(+), 24 deletions(-) (limited to 'fs') diff --git a/fs/xattr.c b/fs/xattr.c index 9b932b95d74e..9e8a6529dfc5 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -324,7 +324,6 @@ setxattr(struct dentry *d, const char __user *name, const void __user *value, { int error; void *kvalue = NULL; - void *vvalue = NULL; /* If non-NULL, we used vmalloc() */ char kname[XATTR_NAME_MAX + 1]; if (flags & ~(XATTR_CREATE|XATTR_REPLACE)) @@ -341,10 +340,9 @@ setxattr(struct dentry *d, const char __user *name, const void __user *value, return -E2BIG; kvalue = kmalloc(size, GFP_KERNEL | __GFP_NOWARN); if (!kvalue) { - vvalue = vmalloc(size); - if (!vvalue) + kvalue = vmalloc(size); + if (!kvalue) return -ENOMEM; - kvalue = vvalue; } if (copy_from_user(kvalue, value, size)) { error = -EFAULT; @@ -357,10 +355,8 @@ setxattr(struct dentry *d, const char __user *name, const void __user *value, error = vfs_setxattr(d, kname, kvalue, size, flags); out: - if (vvalue) - vfree(vvalue); - else - kfree(kvalue); + kvfree(kvalue); + return error; } @@ -428,7 +424,6 @@ getxattr(struct dentry *d, const char __user *name, void __user *value, { ssize_t error; void *kvalue = NULL; - void *vvalue = NULL; char kname[XATTR_NAME_MAX + 1]; error = strncpy_from_user(kname, name, sizeof(kname)); @@ -442,10 +437,9 @@ getxattr(struct dentry *d, const char __user *name, void __user *value, size = XATTR_SIZE_MAX; kvalue = kzalloc(size, GFP_KERNEL | __GFP_NOWARN); if (!kvalue) { - vvalue = vmalloc(size); - if (!vvalue) + kvalue = vmalloc(size); + if (!kvalue) return -ENOMEM; - kvalue = vvalue; } } @@ -461,10 +455,9 @@ getxattr(struct dentry *d, const char __user *name, void __user *value, than XATTR_SIZE_MAX bytes. Not possible. */ error = -E2BIG; } - if (vvalue) - vfree(vvalue); - else - kfree(kvalue); + + kvfree(kvalue); + return error; } @@ -521,17 +514,15 @@ listxattr(struct dentry *d, char __user *list, size_t size) { ssize_t error; char *klist = NULL; - char *vlist = NULL; /* If non-NULL, we used vmalloc() */ if (size) { if (size > XATTR_LIST_MAX) size = XATTR_LIST_MAX; klist = kmalloc(size, __GFP_NOWARN | GFP_KERNEL); if (!klist) { - vlist = vmalloc(size); - if (!vlist) + klist = vmalloc(size); + if (!klist) return -ENOMEM; - klist = vlist; } } @@ -544,10 +535,9 @@ listxattr(struct dentry *d, char __user *list, size_t size) than XATTR_LIST_MAX bytes. Not possible. */ error = -E2BIG; } - if (vlist) - vfree(vlist); - else - kfree(klist); + + kvfree(klist); + return error; } -- cgit v1.2.1 From 90330e689c32e5105265c461c54af6ecec3373fa Mon Sep 17 00:00:00 2001 From: Abhi Das Date: Fri, 18 Dec 2015 14:11:36 -0600 Subject: fs: __generic_file_splice_read retry lookup on AOP_TRUNCATED_PAGE During testing, I discovered that __generic_file_splice_read() returns 0 (EOF) when aops->readpage fails with AOP_TRUNCATED_PAGE on the first page of a single/multi-page splice read operation. This EOF return code causes the userspace test to (correctly) report a zero-length read error when it was expecting otherwise. The current strategy of returning a partial non-zero read when ->readpage returns AOP_TRUNCATED_PAGE works only when the failed page is not the first of the lot being processed. This patch attempts to retry lookup and call ->readpage again on pages that had previously failed with AOP_TRUNCATED_PAGE. With this patch, my tests pass and I haven't noticed any unwanted side effects. This version removes the thrice-retry loop and instead indefinitely retries lookups on AOP_TRUNCATED_PAGE errors from ->readpage. This behavior is now similar to do_generic_file_read(). Signed-off-by: Abhi Das Reviewed-by: Jan Kara Cc: Bob Peterson Cc: Al Viro Signed-off-by: Al Viro --- fs/splice.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/splice.c b/fs/splice.c index 4cf700d50b40..82bc0d64fc38 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -415,6 +415,7 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, */ if (!page->mapping) { unlock_page(page); +retry_lookup: page = find_or_create_page(mapping, index, mapping_gfp_mask(mapping)); @@ -439,13 +440,10 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, error = mapping->a_ops->readpage(in, page); if (unlikely(error)) { /* - * We really should re-lookup the page here, - * but it complicates things a lot. Instead - * lets just do what we already stored, and - * we'll get it the next time we are called. + * Re-lookup the page */ if (error == AOP_TRUNCATED_PAGE) - error = 0; + goto retry_lookup; break; } -- cgit v1.2.1 From 0dbf5f20652108106cb822ad7662c786baaa03ff Mon Sep 17 00:00:00 2001 From: Stanislav Kinsburskiy Date: Tue, 15 Dec 2015 19:41:31 +0400 Subject: fcntl: allow to set O_DIRECT flag on pipe With packetized mode for pipes, it's not possible to set O_DIRECT on pipe file via sys_fcntl, because of unsupported sanity checks. Ability to set this flag will be used by CRIU to migrate packetized pipes. v2: Fixed typos and mode variable to check. Signed-off-by: Stanislav Kinsburskiy Signed-off-by: Al Viro --- fs/fcntl.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/fcntl.c b/fs/fcntl.c index ee85cd4e136a..350a2c8cfd28 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -51,7 +51,8 @@ static int setfl(int fd, struct file * filp, unsigned long arg) if (arg & O_NDELAY) arg |= O_NONBLOCK; - if (arg & O_DIRECT) { + /* Pipe packetized mode is controlled by O_DIRECT flag */ + if (!S_ISFIFO(filp->f_inode->i_mode) && (arg & O_DIRECT)) { if (!filp->f_mapping || !filp->f_mapping->a_ops || !filp->f_mapping->a_ops->direct_IO) return -EINVAL; -- cgit v1.2.1 From bc51b2a9193f917f2065313d9971d7d5e867bc81 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Fri, 11 Dec 2015 17:03:26 +0100 Subject: logfs: constify logfs_block_ops structures The logfs_block_ops structures are never modified, so declare them as const. Done with the help of Coccinelle. Signed-off-by: Julia Lawall Signed-off-by: Al Viro --- fs/logfs/logfs.h | 4 ++-- fs/logfs/readwrite.c | 4 ++-- fs/logfs/segment.c | 2 +- 3 files changed, 5 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/logfs/logfs.h b/fs/logfs/logfs.h index 5f0937609465..5731361eba46 100644 --- a/fs/logfs/logfs.h +++ b/fs/logfs/logfs.h @@ -302,7 +302,7 @@ struct logfs_block { struct inode *inode; struct logfs_transaction *ta; unsigned long alias_map[LOGFS_BLOCK_FACTOR / BITS_PER_LONG]; - struct logfs_block_ops *ops; + const struct logfs_block_ops *ops; int full; int partial; int reserved_bytes; @@ -579,7 +579,7 @@ int logfs_exist_block(struct inode *inode, u64 bix); int get_page_reserve(struct inode *inode, struct page *page); void logfs_get_wblocks(struct super_block *sb, struct page *page, int lock); void logfs_put_wblocks(struct super_block *sb, struct page *page, int lock); -extern struct logfs_block_ops indirect_block_ops; +extern const struct logfs_block_ops indirect_block_ops; /* segment.c */ int logfs_erase_segment(struct super_block *sb, u32 ofs, int ensure_erase); diff --git a/fs/logfs/readwrite.c b/fs/logfs/readwrite.c index 380d86e1ab45..20973c9e52f8 100644 --- a/fs/logfs/readwrite.c +++ b/fs/logfs/readwrite.c @@ -569,13 +569,13 @@ static void indirect_free_block(struct super_block *sb, } -static struct logfs_block_ops inode_block_ops = { +static const struct logfs_block_ops inode_block_ops = { .write_block = inode_write_block, .free_block = inode_free_block, .write_alias = inode_write_alias, }; -struct logfs_block_ops indirect_block_ops = { +const struct logfs_block_ops indirect_block_ops = { .write_block = indirect_write_block, .free_block = indirect_free_block, .write_alias = indirect_write_alias, diff --git a/fs/logfs/segment.c b/fs/logfs/segment.c index 6de0fbfc6c00..d270e4b2ab6b 100644 --- a/fs/logfs/segment.c +++ b/fs/logfs/segment.c @@ -197,7 +197,7 @@ static int btree_write_alias(struct super_block *sb, struct logfs_block *block, return 0; } -static struct logfs_block_ops btree_block_ops = { +static const struct logfs_block_ops btree_block_ops = { .write_block = btree_write_block, .free_block = __free_block, .write_alias = btree_write_alias, -- cgit v1.2.1 From 3cc4a84e026e8d61b7ffe4a7367ed09a555f2c5b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 3 Dec 2015 11:12:07 -0800 Subject: proc: add a reschedule point in proc_readfd_common() User can pass an arbitrary large buffer to getdents(). It is typically a 32KB buffer used by libc scandir() implementation. When scanning /proc/{pid}/fd, we can hold cpu way too long, so add a cond_resched() to be kind with other tasks. We've seen latencies of more than 50ms on real workloads. Signed-off-by: Eric Dumazet Cc: Alexander Viro Signed-off-by: Al Viro --- fs/proc/fd.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/proc/fd.c b/fs/proc/fd.c index 3c2a915c695a..56afa5ef08f2 100644 --- a/fs/proc/fd.c +++ b/fs/proc/fd.c @@ -258,6 +258,7 @@ static int proc_readfd_common(struct file *file, struct dir_context *ctx, name, len, instantiate, p, (void *)(unsigned long)fd)) goto out_fd_loop; + cond_resched(); rcu_read_lock(); } rcu_read_unlock(); -- cgit v1.2.1 From 8f5fed1e917588f946ad8882bd47a4093db0ff4c Mon Sep 17 00:00:00 2001 From: Sasha Levin Date: Thu, 7 Jan 2016 17:49:51 -0500 Subject: fs/9p: use fscache mutex rather than spinlock We may sleep inside a the lock, so use a mutex rather than spinlock. Signed-off-by: Sasha Levin Signed-off-by: Al Viro --- fs/9p/cache.c | 8 ++++---- fs/9p/v9fs.h | 2 +- fs/9p/vfs_inode.c | 2 +- 3 files changed, 6 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/9p/cache.c b/fs/9p/cache.c index a69260f27555..103ca5e1267b 100644 --- a/fs/9p/cache.c +++ b/fs/9p/cache.c @@ -243,14 +243,14 @@ void v9fs_cache_inode_set_cookie(struct inode *inode, struct file *filp) if (!v9inode->fscache) return; - spin_lock(&v9inode->fscache_lock); + mutex_lock(&v9inode->fscache_lock); if ((filp->f_flags & O_ACCMODE) != O_RDONLY) v9fs_cache_inode_flush_cookie(inode); else v9fs_cache_inode_get_cookie(inode); - spin_unlock(&v9inode->fscache_lock); + mutex_unlock(&v9inode->fscache_lock); } void v9fs_cache_inode_reset_cookie(struct inode *inode) @@ -264,7 +264,7 @@ void v9fs_cache_inode_reset_cookie(struct inode *inode) old = v9inode->fscache; - spin_lock(&v9inode->fscache_lock); + mutex_lock(&v9inode->fscache_lock); fscache_relinquish_cookie(v9inode->fscache, 1); v9ses = v9fs_inode2v9ses(inode); @@ -274,7 +274,7 @@ void v9fs_cache_inode_reset_cookie(struct inode *inode) p9_debug(P9_DEBUG_FSC, "inode %p revalidating cookie old %p new %p\n", inode, old, v9inode->fscache); - spin_unlock(&v9inode->fscache_lock); + mutex_unlock(&v9inode->fscache_lock); } int __v9fs_fscache_release_page(struct page *page, gfp_t gfp) diff --git a/fs/9p/v9fs.h b/fs/9p/v9fs.h index 0923f2cf3c80..6877050384a1 100644 --- a/fs/9p/v9fs.h +++ b/fs/9p/v9fs.h @@ -123,7 +123,7 @@ struct v9fs_session_info { struct v9fs_inode { #ifdef CONFIG_9P_FSCACHE - spinlock_t fscache_lock; + struct mutex fscache_lock; struct fscache_cookie *fscache; #endif struct p9_qid qid; diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 511078586fa1..fbdb29ba62af 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -244,7 +244,7 @@ struct inode *v9fs_alloc_inode(struct super_block *sb) return NULL; #ifdef CONFIG_9P_FSCACHE v9inode->fscache = NULL; - spin_lock_init(&v9inode->fscache_lock); + mutex_init(&v9inode->fscache_lock); #endif v9inode->writeback_fid = NULL; v9inode->cache_validity = 0; -- cgit v1.2.1 From db39c16724d019029d7533561754d92bef1b389a Mon Sep 17 00:00:00 2001 From: DengChao Date: Thu, 12 Nov 2015 21:40:41 +0800 Subject: fs:affs:Replace time_t with time64_t The affs code uses "time_t" and "get_seconds()". This will cause problems on 32-bit architectures in 2038 when time_t overflows. This patch replaces them with "time64_t" and "ktime_get_real_seconds()". This patch introduces expensive 64-bit divsion in "secs_to_datestamp()", considering this function is not called so often, the cost should be acceptable. Reviewed-by: Arnd Bergmann Signed-off-by: DengChao Signed-off-by: Al Viro --- fs/affs/affs.h | 2 +- fs/affs/amigaffs.c | 13 +++++++------ fs/affs/super.c | 2 +- 3 files changed, 9 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/affs/affs.h b/fs/affs/affs.h index c69a87eaf57d..cc2b2efc9211 100644 --- a/fs/affs/affs.h +++ b/fs/affs/affs.h @@ -138,7 +138,7 @@ extern int affs_remove_hash(struct inode *dir, struct buffer_head *rem_bh); extern int affs_remove_header(struct dentry *dentry); extern u32 affs_checksum_block(struct super_block *sb, struct buffer_head *bh); extern void affs_fix_checksum(struct super_block *sb, struct buffer_head *bh); -extern void secs_to_datestamp(time_t secs, struct affs_date *ds); +extern void secs_to_datestamp(time64_t secs, struct affs_date *ds); extern umode_t prot_to_mode(u32 prot); extern void mode_to_prot(struct inode *inode); __printf(3, 4) diff --git a/fs/affs/amigaffs.c b/fs/affs/amigaffs.c index 5fa92bc790ef..d6c7a51c93e4 100644 --- a/fs/affs/amigaffs.c +++ b/fs/affs/amigaffs.c @@ -8,6 +8,7 @@ * Please send bug reports to: hjw@zvw.de */ +#include #include "affs.h" /* @@ -366,22 +367,22 @@ affs_fix_checksum(struct super_block *sb, struct buffer_head *bh) } void -secs_to_datestamp(time_t secs, struct affs_date *ds) +secs_to_datestamp(time64_t secs, struct affs_date *ds) { u32 days; u32 minute; + s32 rem; secs -= sys_tz.tz_minuteswest * 60 + ((8 * 365 + 2) * 24 * 60 * 60); if (secs < 0) secs = 0; - days = secs / 86400; - secs -= days * 86400; - minute = secs / 60; - secs -= minute * 60; + days = div_s64_rem(secs, 86400, &rem); + minute = rem / 60; + rem -= minute * 60; ds->days = cpu_to_be32(days); ds->mins = cpu_to_be32(minute); - ds->ticks = cpu_to_be32(secs * 50); + ds->ticks = cpu_to_be32(rem * 50); } umode_t diff --git a/fs/affs/super.c b/fs/affs/super.c index 5b50c4ca43a7..8836df5f1e11 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -32,7 +32,7 @@ affs_commit_super(struct super_block *sb, int wait) struct affs_root_tail *tail = AFFS_ROOT_TAIL(sb, bh); lock_buffer(bh); - secs_to_datestamp(get_seconds(), &tail->disk_change); + secs_to_datestamp(ktime_get_real_seconds(), &tail->disk_change); affs_fix_checksum(sb, bh); unlock_buffer(bh); -- cgit v1.2.1 From bbddca8e8fac07ece3938e03526b5d00fa791a4c Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 7 Jan 2016 16:08:20 -0500 Subject: nfsd: don't hold i_mutex over userspace upcalls We need information about exports when crossing mountpoints during lookup or NFSv4 readdir. If we don't already have that information cached, we may have to ask (and wait for) rpc.mountd. In both cases we currently hold the i_mutex on the parent of the directory we're asking rpc.mountd about. We've seen situations where rpc.mountd performs some operation on that directory that tries to take the i_mutex again, resulting in deadlock. With some care, we may be able to avoid that in rpc.mountd. But it seems better just to avoid holding a mutex while waiting on userspace. It appears that lookup_one_len is pretty much the only operation that needs the i_mutex. So we could just drop the i_mutex elsewhere and do something like mutex_lock() lookup_one_len() mutex_unlock() In many cases though the lookup would have been cached and not required the i_mutex, so it's more efficient to create a lookup_one_len() variant that only takes the i_mutex when necessary. Signed-off-by: NeilBrown Signed-off-by: J. Bruce Fields Signed-off-by: Al Viro --- fs/namei.c | 71 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ fs/nfsd/nfs3xdr.c | 2 +- fs/nfsd/nfs4xdr.c | 8 +++---- fs/nfsd/vfs.c | 23 +++++++----------- 4 files changed, 85 insertions(+), 19 deletions(-) (limited to 'fs') diff --git a/fs/namei.c b/fs/namei.c index 45c702edce3c..1067f7a0287a 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -2272,6 +2272,8 @@ EXPORT_SYMBOL(vfs_path_lookup); * * Note that this routine is purely a helper for filesystem usage and should * not be called by generic code. + * + * The caller must hold base->i_mutex. */ struct dentry *lookup_one_len(const char *name, struct dentry *base, int len) { @@ -2315,6 +2317,75 @@ struct dentry *lookup_one_len(const char *name, struct dentry *base, int len) } EXPORT_SYMBOL(lookup_one_len); +/** + * lookup_one_len_unlocked - filesystem helper to lookup single pathname component + * @name: pathname component to lookup + * @base: base directory to lookup from + * @len: maximum length @len should be interpreted to + * + * Note that this routine is purely a helper for filesystem usage and should + * not be called by generic code. + * + * Unlike lookup_one_len, it should be called without the parent + * i_mutex held, and will take the i_mutex itself if necessary. + */ +struct dentry *lookup_one_len_unlocked(const char *name, + struct dentry *base, int len) +{ + struct qstr this; + unsigned int c; + int err; + struct dentry *ret; + + this.name = name; + this.len = len; + this.hash = full_name_hash(name, len); + if (!len) + return ERR_PTR(-EACCES); + + if (unlikely(name[0] == '.')) { + if (len < 2 || (len == 2 && name[1] == '.')) + return ERR_PTR(-EACCES); + } + + while (len--) { + c = *(const unsigned char *)name++; + if (c == '/' || c == '\0') + return ERR_PTR(-EACCES); + } + /* + * See if the low-level filesystem might want + * to use its own hash.. + */ + if (base->d_flags & DCACHE_OP_HASH) { + int err = base->d_op->d_hash(base, &this); + if (err < 0) + return ERR_PTR(err); + } + + err = inode_permission(base->d_inode, MAY_EXEC); + if (err) + return ERR_PTR(err); + + /* + * __d_lookup() is used to try to get a quick answer and avoid the + * mutex. A false-negative does no harm. + */ + ret = __d_lookup(base, &this); + if (ret && unlikely(ret->d_flags & DCACHE_OP_REVALIDATE)) { + dput(ret); + ret = NULL; + } + if (ret) + return ret; + + mutex_lock(&base->d_inode->i_mutex); + ret = __lookup_hash(&this, base, 0); + mutex_unlock(&base->d_inode->i_mutex); + return ret; +} +EXPORT_SYMBOL(lookup_one_len_unlocked); + int user_path_at_empty(int dfd, const char __user *name, unsigned flags, struct path *path, int *empty) { diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index 00575d776d91..2246454dec76 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -823,7 +823,7 @@ compose_entry_fh(struct nfsd3_readdirres *cd, struct svc_fh *fhp, } else dchild = dget(dparent); } else - dchild = lookup_one_len(name, dparent, namlen); + dchild = lookup_one_len_unlocked(name, dparent, namlen); if (IS_ERR(dchild)) return rv; if (d_mountpoint(dchild)) diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 51c9e9ca39a4..325521ce389a 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -2838,14 +2838,14 @@ nfsd4_encode_dirent_fattr(struct xdr_stream *xdr, struct nfsd4_readdir *cd, __be32 nfserr; int ignore_crossmnt = 0; - dentry = lookup_one_len(name, cd->rd_fhp->fh_dentry, namlen); + dentry = lookup_one_len_unlocked(name, cd->rd_fhp->fh_dentry, namlen); if (IS_ERR(dentry)) return nfserrno(PTR_ERR(dentry)); if (d_really_is_negative(dentry)) { /* - * nfsd_buffered_readdir drops the i_mutex between - * readdir and calling this callback, leaving a window - * where this directory entry could have gone away. + * we're not holding the i_mutex here, so there's + * a window where this directory entry could have gone + * away. */ dput(dentry); return nfserr_noent; diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 994d66fbb446..4212aaacbb55 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -217,10 +217,16 @@ nfsd_lookup_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp, host_err = PTR_ERR(dentry); if (IS_ERR(dentry)) goto out_nfserr; - /* - * check if we have crossed a mount point ... - */ if (nfsd_mountpoint(dentry, exp)) { + /* + * We don't need the i_mutex after all. It's + * still possible we could open this (regular + * files can be mountpoints too), but the + * i_mutex is just there to prevent renames of + * something that we might be about to delegate, + * and a mountpoint won't be renamed: + */ + fh_unlock(fhp); if ((host_err = nfsd_cross_mnt(rqstp, &dentry, &exp))) { dput(dentry); goto out_nfserr; @@ -1809,7 +1815,6 @@ static __be32 nfsd_buffered_readdir(struct file *file, nfsd_filldir_t func, offset = *offsetp; while (1) { - struct inode *dir_inode = file_inode(file); unsigned int reclen; cdp->err = nfserr_eof; /* will be cleared on successful read */ @@ -1828,15 +1833,6 @@ static __be32 nfsd_buffered_readdir(struct file *file, nfsd_filldir_t func, if (!size) break; - /* - * Various filldir functions may end up calling back into - * lookup_one_len() and the file system's ->lookup() method. - * These expect i_mutex to be held, as it would within readdir. - */ - host_err = mutex_lock_killable(&dir_inode->i_mutex); - if (host_err) - break; - de = (struct buffered_dirent *)buf.dirent; while (size > 0) { offset = de->offset; @@ -1853,7 +1849,6 @@ static __be32 nfsd_buffered_readdir(struct file *file, nfsd_filldir_t func, size -= reclen; de = (struct buffered_dirent *)((char *)de + reclen); } - mutex_unlock(&dir_inode->i_mutex); if (size > 0) /* We bailed out early */ break; -- cgit v1.2.1