From 074b85175a43a23fdbde60f55feea636e0bf0f85 Mon Sep 17 00:00:00 2001 From: Dimitri Sivanich Date: Wed, 8 Feb 2012 12:39:07 -0800 Subject: vfs: fix panic in __d_lookup() with high dentry hashtable counts When the number of dentry cache hash table entries gets too high (2147483648 entries), as happens by default on a 16TB system, use of a signed integer in the dcache_init() initialization loop prevents the dentry_hashtable from getting initialized, causing a panic in __d_lookup(). Fix this in dcache_init() and similar areas. Signed-off-by: Dimitri Sivanich Acked-by: David S. Miller Cc: Al Viro Signed-off-by: Andrew Morton Signed-off-by: Al Viro --- fs/inode.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs/inode.c') diff --git a/fs/inode.c b/fs/inode.c index fb10d86ffad..d3ebdbe723d 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -1651,7 +1651,7 @@ __setup("ihash_entries=", set_ihash_entries); */ void __init inode_init_early(void) { - int loop; + unsigned int loop; /* If hashes are distributed across NUMA nodes, defer * hash allocation until vmalloc space is available. @@ -1669,13 +1669,13 @@ void __init inode_init_early(void) &i_hash_mask, 0); - for (loop = 0; loop < (1 << i_hash_shift); loop++) + for (loop = 0; loop < (1U << i_hash_shift); loop++) INIT_HLIST_HEAD(&inode_hashtable[loop]); } void __init inode_init(void) { - int loop; + unsigned int loop; /* inode slab cache */ inode_cachep = kmem_cache_create("inode_cache", @@ -1699,7 +1699,7 @@ void __init inode_init(void) &i_hash_mask, 0); - for (loop = 0; loop < (1 << i_hash_shift); loop++) + for (loop = 0; loop < (1U << i_hash_shift); loop++) INIT_HLIST_HEAD(&inode_hashtable[loop]); } -- cgit v1.2.3-70-g09d2 From 978d6d8c4574098050b22281b9ed06818c0b23ca Mon Sep 17 00:00:00 2001 From: Tyler Hicks Date: Mon, 12 Dec 2011 10:02:30 -0600 Subject: vfs: Correctly set the dir i_mutex lockdep class 9a7aa12f3911853a introduced additional logic around setting the i_mutex lockdep class for directory inodes. The idea was that some filesystems may want their own special lockdep class for different directory inodes and calling unlock_new_inode() should not clobber one of those special classes. I believe that the added conditional, around the *negated* return value of lockdep_match_class(), caused directory inodes to be placed in the wrong lockdep class. inode_init_always() sets the i_mutex lockdep class with i_mutex_key for all inodes. If the filesystem did not change the class during inode initialization, then the conditional mentioned above was false and the directory inode was incorrectly left in the non-directory lockdep class. If the filesystem did set a special lockdep class, then the conditional mentioned above was true and that class was clobbered with i_mutex_dir_key. This patch removes the negation from the conditional so that the i_mutex lockdep class is properly set for directory inodes. Special classes are preserved and directory inodes with unmodified classes are set with i_mutex_dir_key. Signed-off-by: Tyler Hicks Reviewed-by: Jan Kara Signed-off-by: Al Viro --- fs/inode.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs/inode.c') diff --git a/fs/inode.c b/fs/inode.c index d3ebdbe723d..8affbc9c211 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -938,8 +938,7 @@ void lockdep_annotate_inode_mutex_key(struct inode *inode) struct file_system_type *type = inode->i_sb->s_type; /* Set new key only if filesystem hasn't already changed it */ - if (!lockdep_match_class(&inode->i_mutex, - &type->i_mutex_key)) { + if (lockdep_match_class(&inode->i_mutex, &type->i_mutex_key)) { /* * ensure nobody is actually holding i_mutex */ -- cgit v1.2.3-70-g09d2 From 310fa7a36722017088af123043ebd231cd6bc559 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 10 Mar 2012 17:07:28 -0500 Subject: restore smp_mb() in unlock_new_inode() wait_on_inode() doesn't have ->i_lock Signed-off-by: Al Viro --- fs/inode.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs/inode.c') diff --git a/fs/inode.c b/fs/inode.c index 8affbc9c211..83ab215baab 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -965,6 +965,7 @@ void unlock_new_inode(struct inode *inode) spin_lock(&inode->i_lock); WARN_ON(!(inode->i_state & I_NEW)); inode->i_state &= ~I_NEW; + smp_mb(); wake_up_bit(&inode->i_state, __I_NEW); spin_unlock(&inode->i_lock); } -- cgit v1.2.3-70-g09d2 From 9bcb4b733c22b7dbc4cf847e707ac98f751e9180 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 12 Feb 2012 19:43:17 -0500 Subject: vfs: turn generic_drop_inode() into static inline Once upon a time it used to be much bigger, but these days there's no point whatsoever keeping it in fs/inode.c, especially since it's not even needed as initializer for ->drop_inode() - it's the default and leaving ->drop_inode NULL will do just as well. Signed-off-by: Al Viro --- fs/inode.c | 11 ----------- include/linux/fs.h | 5 ++++- 2 files changed, 4 insertions(+), 12 deletions(-) (limited to 'fs/inode.c') diff --git a/fs/inode.c b/fs/inode.c index 83ab215baab..92de04b0baa 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -1368,17 +1368,6 @@ int generic_delete_inode(struct inode *inode) } EXPORT_SYMBOL(generic_delete_inode); -/* - * Normal UNIX filesystem behaviour: delete the - * inode when the usage count drops to zero, and - * i_nlink is zero. - */ -int generic_drop_inode(struct inode *inode) -{ - return !inode->i_nlink || inode_unhashed(inode); -} -EXPORT_SYMBOL_GPL(generic_drop_inode); - /* * Called when we're dropping the last reference * to an inode. diff --git a/include/linux/fs.h b/include/linux/fs.h index 0ab89426ba5..b89dc4d60d9 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2305,7 +2305,10 @@ extern struct inode * igrab(struct inode *); extern ino_t iunique(struct super_block *, ino_t); extern int inode_needs_sync(struct inode *inode); extern int generic_delete_inode(struct inode *inode); -extern int generic_drop_inode(struct inode *inode); +static inline int generic_drop_inode(struct inode *inode) +{ + return !inode->i_nlink || inode_unhashed(inode); +} extern struct inode *ilookup5_nowait(struct super_block *sb, unsigned long hashval, int (*test)(struct inode *, void *), -- cgit v1.2.3-70-g09d2 From 68ac1234fb949b66941d94dce4157742799fc581 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 15 Mar 2012 08:21:57 -0400 Subject: switch touch_atime to struct path Signed-off-by: Al Viro --- fs/cachefiles/namei.c | 3 ++- fs/ecryptfs/file.c | 9 ++++----- fs/inode.c | 5 +++-- fs/namei.c | 2 +- fs/nfsd/vfs.c | 11 ++++++----- fs/stat.c | 2 +- include/linux/fs.h | 4 ++-- net/unix/af_unix.c | 4 ++-- 8 files changed, 21 insertions(+), 19 deletions(-) (limited to 'fs/inode.c') diff --git a/fs/cachefiles/namei.c b/fs/cachefiles/namei.c index a0358c2189c..7f0771d3894 100644 --- a/fs/cachefiles/namei.c +++ b/fs/cachefiles/namei.c @@ -646,7 +646,8 @@ lookup_again: * (this is used to keep track of culling, and atimes are only * updated by read, write and readdir but not lookup or * open) */ - touch_atime(cache->mnt, next); + path.dentry = next; + touch_atime(&path); } /* open a file interface onto a data file */ diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index d3f95f941c4..2b17f2f9b12 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -48,8 +48,7 @@ static ssize_t ecryptfs_read_update_atime(struct kiocb *iocb, unsigned long nr_segs, loff_t pos) { ssize_t rc; - struct dentry *lower_dentry; - struct vfsmount *lower_vfsmount; + struct path lower; struct file *file = iocb->ki_filp; rc = generic_file_aio_read(iocb, iov, nr_segs, pos); @@ -60,9 +59,9 @@ static ssize_t ecryptfs_read_update_atime(struct kiocb *iocb, if (-EIOCBQUEUED == rc) rc = wait_on_sync_kiocb(iocb); if (rc >= 0) { - lower_dentry = ecryptfs_dentry_to_lower(file->f_path.dentry); - lower_vfsmount = ecryptfs_dentry_to_lower_mnt(file->f_path.dentry); - touch_atime(lower_vfsmount, lower_dentry); + lower.dentry = ecryptfs_dentry_to_lower(file->f_path.dentry); + lower.mnt = ecryptfs_dentry_to_lower_mnt(file->f_path.dentry); + touch_atime(&lower); } return rc; } diff --git a/fs/inode.c b/fs/inode.c index 92de04b0baa..8b612813a6a 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -1499,9 +1499,10 @@ static int relatime_need_update(struct vfsmount *mnt, struct inode *inode, * This function automatically handles read only file systems and media, * as well as the "noatime" flag and inode specific "noatime" markers. */ -void touch_atime(struct vfsmount *mnt, struct dentry *dentry) +void touch_atime(struct path *path) { - struct inode *inode = dentry->d_inode; + struct vfsmount *mnt = path->mnt; + struct inode *inode = path->dentry->d_inode; struct timespec now; if (inode->i_flags & S_NOATIME) diff --git a/fs/namei.c b/fs/namei.c index a0b82762e8f..0ccc74ee92f 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -642,7 +642,7 @@ follow_link(struct path *link, struct nameidata *nd, void **p) cond_resched(); current->total_link_count++; - touch_atime(link->mnt, dentry); + touch_atime(link); nd_set_link(nd, NULL); error = security_inode_follow_link(link->dentry, nd); diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index edf6d3ed877..e59f71d0cf7 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1541,30 +1541,31 @@ do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, __be32 nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp) { - struct dentry *dentry; struct inode *inode; mm_segment_t oldfs; __be32 err; int host_err; + struct path path; err = fh_verify(rqstp, fhp, S_IFLNK, NFSD_MAY_NOP); if (err) goto out; - dentry = fhp->fh_dentry; - inode = dentry->d_inode; + path.mnt = fhp->fh_export->ex_path.mnt; + path.dentry = fhp->fh_dentry; + inode = path.dentry->d_inode; err = nfserr_inval; if (!inode->i_op->readlink) goto out; - touch_atime(fhp->fh_export->ex_path.mnt, dentry); + touch_atime(&path); /* N.B. Why does this call need a get_fs()?? * Remove the set_fs and watch the fireworks:-) --okir */ oldfs = get_fs(); set_fs(KERNEL_DS); - host_err = inode->i_op->readlink(dentry, buf, *lenp); + host_err = inode->i_op->readlink(path.dentry, buf, *lenp); set_fs(oldfs); if (host_err < 0) diff --git a/fs/stat.c b/fs/stat.c index 8806b8997d2..86f13563a46 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -307,7 +307,7 @@ SYSCALL_DEFINE4(readlinkat, int, dfd, const char __user *, pathname, if (inode->i_op->readlink) { error = security_inode_readlink(path.dentry); if (!error) { - touch_atime(path.mnt, path.dentry); + touch_atime(&path); error = inode->i_op->readlink(path.dentry, buf, bufsiz); } diff --git a/include/linux/fs.h b/include/linux/fs.h index b89dc4d60d9..9bbe1a9ac43 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1812,11 +1812,11 @@ static inline void inode_inc_iversion(struct inode *inode) spin_unlock(&inode->i_lock); } -extern void touch_atime(struct vfsmount *mnt, struct dentry *dentry); +extern void touch_atime(struct path *); static inline void file_accessed(struct file *file) { if (!(file->f_flags & O_NOATIME)) - touch_atime(file->f_path.mnt, file->f_path.dentry); + touch_atime(&file->f_path); } int sync_inode(struct inode *inode, struct writeback_control *wbc); diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index ef4b780ef63..081679444a6 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -771,7 +771,7 @@ static struct sock *unix_find_other(struct net *net, goto put_fail; if (u->sk_type == type) - touch_atime(path.mnt, path.dentry); + touch_atime(&path); path_put(&path); @@ -787,7 +787,7 @@ static struct sock *unix_find_other(struct net *net, struct dentry *dentry; dentry = unix_sk(u)->path.dentry; if (dentry) - touch_atime(unix_sk(u)->path.mnt, dentry); + touch_atime(&unix_sk(u)->path); } else goto fail; } -- cgit v1.2.3-70-g09d2 From e59cc473cc603d562f2c80c12c943ef2a8cde6b2 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 7 Dec 2011 13:17:19 -0500 Subject: trim includes in inode.c Signed-off-by: Al Viro --- fs/inode.c | 12 +----------- 1 file changed, 1 insertion(+), 11 deletions(-) (limited to 'fs/inode.c') diff --git a/fs/inode.c b/fs/inode.c index 8b612813a6a..9f4f5fecc09 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -2,29 +2,19 @@ * (C) 1997 Linus Torvalds * (C) 1999 Andrea Arcangeli (dynamic inode allocation) */ +#include #include #include -#include -#include -#include -#include -#include #include -#include -#include #include #include #include -#include #include #include #include #include -#include #include #include -#include -#include #include /* for inode_has_buffers */ #include #include "internal.h" -- cgit v1.2.3-70-g09d2