1
linux/fs/inode.c

1444 lines
37 KiB
C
Raw Normal View History

/*
* linux/fs/inode.c
*
* (C) 1997 Linus Torvalds
*/
#include <linux/fs.h>
#include <linux/mm.h>
#include <linux/dcache.h>
#include <linux/init.h>
#include <linux/quotaops.h>
#include <linux/slab.h>
#include <linux/writeback.h>
#include <linux/module.h>
#include <linux/backing-dev.h>
#include <linux/wait.h>
#include <linux/hash.h>
#include <linux/swap.h>
#include <linux/security.h>
#include <linux/pagemap.h>
#include <linux/cdev.h>
#include <linux/bootmem.h>
#include <linux/inotify.h>
#include <linux/mount.h>
/*
* This is needed for the following functions:
* - inode_has_buffers
* - invalidate_inode_buffers
* - invalidate_bdev
*
* FIXME: remove all knowledge of the buffer layer from this file
*/
#include <linux/buffer_head.h>
/*
* New inode.c implementation.
*
* This implementation has the basic premise of trying
* to be extremely low-overhead and SMP-safe, yet be
* simple enough to be "obviously correct".
*
* Famous last words.
*/
/* inode dynamic allocation 1999, Andrea Arcangeli <andrea@suse.de> */
/* #define INODE_PARANOIA 1 */
/* #define INODE_DEBUG 1 */
/*
* Inode lookup is no longer as critical as it used to be:
* most of the lookups are going to be through the dcache.
*/
#define I_HASHBITS i_hash_shift
#define I_HASHMASK i_hash_mask
static unsigned int i_hash_mask __read_mostly;
static unsigned int i_hash_shift __read_mostly;
/*
* Each inode can be on two separate lists. One is
* the hash list of the inode, used for lookups. The
* other linked list is the "type" list:
* "in_use" - valid inode, i_count > 0, i_nlink > 0
* "dirty" - as "in_use" but also dirty
* "unused" - valid inode, i_count = 0
*
* A "dirty" list is maintained for each super block,
* allowing for low-overhead inode sync() operations.
*/
LIST_HEAD(inode_in_use);
LIST_HEAD(inode_unused);
static struct hlist_head *inode_hashtable __read_mostly;
/*
* A simple spinlock to protect the list manipulations.
*
* NOTE! You also have to own the lock if you change
* the i_state of an inode while it is in use..
*/
DEFINE_SPINLOCK(inode_lock);
/*
* iprune_mutex provides exclusion between the kswapd or try_to_free_pages
* icache shrinking path, and the umount path. Without this exclusion,
* by the time prune_icache calls iput for the inode whose pages it has
* been invalidating, or by the time it calls clear_inode & destroy_inode
* from its final dispose_list, the struct super_block they refer to
* (for inode->i_sb->s_op) may already have been freed and reused.
*/
static DEFINE_MUTEX(iprune_mutex);
/*
* Statistics gathering..
*/
struct inodes_stat_t inodes_stat;
static struct kmem_cache * inode_cachep __read_mostly;
static struct inode *alloc_inode(struct super_block *sb)
{
static const struct address_space_operations empty_aops;
static struct inode_operations empty_iops;
static const struct file_operations empty_fops;
struct inode *inode;
if (sb->s_op->alloc_inode)
inode = sb->s_op->alloc_inode(sb);
else
inode = (struct inode *) kmem_cache_alloc(inode_cachep, GFP_KERNEL);
if (inode) {
struct address_space * const mapping = &inode->i_data;
inode->i_sb = sb;
inode->i_blkbits = sb->s_blocksize_bits;
inode->i_flags = 0;
atomic_set(&inode->i_count, 1);
inode->i_op = &empty_iops;
inode->i_fop = &empty_fops;
inode->i_nlink = 1;
atomic_set(&inode->i_writecount, 0);
inode->i_size = 0;
inode->i_blocks = 0;
inode->i_bytes = 0;
inode->i_generation = 0;
#ifdef CONFIG_QUOTA
memset(&inode->i_dquot, 0, sizeof(inode->i_dquot));
#endif
inode->i_pipe = NULL;
inode->i_bdev = NULL;
inode->i_cdev = NULL;
inode->i_rdev = 0;
inode->dirtied_when = 0;
if (security_inode_alloc(inode)) {
if (inode->i_sb->s_op->destroy_inode)
inode->i_sb->s_op->destroy_inode(inode);
else
kmem_cache_free(inode_cachep, (inode));
return NULL;
}
mapping->a_ops = &empty_aops;
mapping->host = inode;
mapping->flags = 0;
mapping_set_gfp_mask(mapping, GFP_HIGHUSER);
mapping->assoc_mapping = NULL;
mapping->backing_dev_info = &default_backing_dev_info;
/*
* If the block_device provides a backing_dev_info for client
* inodes then use that. Otherwise the inode share the bdev's
* backing_dev_info.
*/
if (sb->s_bdev) {
struct backing_dev_info *bdi;
bdi = sb->s_bdev->bd_inode_backing_dev_info;
if (!bdi)
bdi = sb->s_bdev->bd_inode->i_mapping->backing_dev_info;
mapping->backing_dev_info = bdi;
}
inode->i_private = NULL;
inode->i_mapping = mapping;
}
return inode;
}
void destroy_inode(struct inode *inode)
{
BUG_ON(inode_has_buffers(inode));
security_inode_free(inode);
if (inode->i_sb->s_op->destroy_inode)
inode->i_sb->s_op->destroy_inode(inode);
else
kmem_cache_free(inode_cachep, (inode));
}
/*
* These are initializations that only need to be done
* once, because the fields are idempotent across use
* of the inode, so let the slab aware of that.
*/
void inode_init_once(struct inode *inode)
{
memset(inode, 0, sizeof(*inode));
INIT_HLIST_NODE(&inode->i_hash);
INIT_LIST_HEAD(&inode->i_dentry);
INIT_LIST_HEAD(&inode->i_devices);
mutex_init(&inode->i_mutex);
init_rwsem(&inode->i_alloc_sem);
INIT_RADIX_TREE(&inode->i_data.page_tree, GFP_ATOMIC);
rwlock_init(&inode->i_data.tree_lock);
spin_lock_init(&inode->i_data.i_mmap_lock);
INIT_LIST_HEAD(&inode->i_data.private_list);
spin_lock_init(&inode->i_data.private_lock);
INIT_RAW_PRIO_TREE_ROOT(&inode->i_data.i_mmap);
INIT_LIST_HEAD(&inode->i_data.i_mmap_nonlinear);
spin_lock_init(&inode->i_lock);
i_size_ordered_init(inode);
#ifdef CONFIG_INOTIFY
INIT_LIST_HEAD(&inode->inotify_watches);
mutex_init(&inode->inotify_mutex);
#endif
}
EXPORT_SYMBOL(inode_init_once);
static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags)
{
struct inode * inode = (struct inode *) foo;
if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) ==
SLAB_CTOR_CONSTRUCTOR)
inode_init_once(inode);
}
/*
* inode_lock must be held
*/
void __iget(struct inode * inode)
{
if (atomic_read(&inode->i_count)) {
atomic_inc(&inode->i_count);
return;
}
atomic_inc(&inode->i_count);
if (!(inode->i_state & (I_DIRTY|I_LOCK)))
list_move(&inode->i_list, &inode_in_use);
inodes_stat.nr_unused--;
}
/**
* clear_inode - clear an inode
* @inode: inode to clear
*
* This is called by the filesystem to tell us
* that the inode is no longer useful. We just
* terminate it with extreme prejudice.
*/
void clear_inode(struct inode *inode)
{
might_sleep();
invalidate_inode_buffers(inode);
BUG_ON(inode->i_data.nrpages);
BUG_ON(!(inode->i_state & I_FREEING));
BUG_ON(inode->i_state & I_CLEAR);
wait_on_inode(inode);
DQUOT_DROP(inode);
if (inode->i_sb && inode->i_sb->s_op->clear_inode)
inode->i_sb->s_op->clear_inode(inode);
if (S_ISBLK(inode->i_mode) && inode->i_bdev)
bd_forget(inode);
if (S_ISCHR(inode->i_mode) && inode->i_cdev)
cd_forget(inode);
inode->i_state = I_CLEAR;
}
EXPORT_SYMBOL(clear_inode);
/*
* dispose_list - dispose of the contents of a local list
* @head: the head of the list to free
*
* Dispose-list gets a local list with local inodes in it, so it doesn't
* need to worry about list corruption and SMP locks.
*/
static void dispose_list(struct list_head *head)
{
int nr_disposed = 0;
while (!list_empty(head)) {
struct inode *inode;
inode = list_entry(head->next, struct inode, i_list);
list_del(&inode->i_list);
if (inode->i_data.nrpages)
truncate_inode_pages(&inode->i_data, 0);
clear_inode(inode);
[PATCH] bugfix: two read_inode() calls without clear_inode() call between Bug symptoms ~~~~~~~~~~~~ For the same inode VFS calls read_inode() twice and doesn't call clear_inode() between the two read_inode() invocations. Bug description ~~~~~~~~~~~~~~~ Suppose we have an inode which has zero reference count but is still in the inode cache. Suppose kswapd invokes shrink_icache_memory() to free some RAM. In prune_icache() inodes are removed from i_hash. prune_icache () is then going to call clear_inode(), but drops the inode_lock spinlock before this. If in this moment another task calls iget() for an inode which was just removed from i_hash by prune_icache(), then iget() invokes read_inode() for this inode, because it is *already removed* from i_hash. The end result is: we call iget(#N) then iput(#N); inode #N has zero i_count now and is in the inode cache; kswapd starts. kswapd removes the inode #N from i_hash ans is preempted; we call iget(#N) again; read_inode() is invoked as the result; but we expect clear_inode() before. Fix ~~~~~~~ To fix the bug I remove inodes from i_hash later, when clear_inode() is actually called. I remove them from i_hash under spinlock protection. Since the i_state is set to I_FREEING, it is safe to do this. The others will sleep waiting for the inode state change. I also postpone removing inodes from i_sb_list. It is not compulsory to do so but I do it for readability reasons. Inodes are added/removed to the lists together everywhere in the code and there is no point to change this rule. This is harmless because the only user of i_sb_list which somehow may interfere with me (invalidate_list()) is excluded by the iprune_sem mutex. The same race is possible in invalidate_list() so I do the same for it. Acked-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-07-12 13:58:12 -07:00
spin_lock(&inode_lock);
hlist_del_init(&inode->i_hash);
list_del_init(&inode->i_sb_list);
spin_unlock(&inode_lock);
wake_up_inode(inode);
destroy_inode(inode);
nr_disposed++;
}
spin_lock(&inode_lock);
inodes_stat.nr_inodes -= nr_disposed;
spin_unlock(&inode_lock);
}
/*
* Invalidate all inodes for a device.
*/
static int invalidate_list(struct list_head *head, struct list_head *dispose)
{
struct list_head *next;
int busy = 0, count = 0;
next = head->next;
for (;;) {
struct list_head * tmp = next;
struct inode * inode;
/*
* We can reschedule here without worrying about the list's
* consistency because the per-sb list of inodes must not
* change during umount anymore, and because iprune_mutex keeps
* shrink_icache_memory() away.
*/
cond_resched_lock(&inode_lock);
next = next->next;
if (tmp == head)
break;
inode = list_entry(tmp, struct inode, i_sb_list);
invalidate_inode_buffers(inode);
if (!atomic_read(&inode->i_count)) {
list_move(&inode->i_list, dispose);
inode->i_state |= I_FREEING;
count++;
continue;
}
busy = 1;
}
/* only unused inodes may be cached with i_count zero */
inodes_stat.nr_unused -= count;
return busy;
}
/**
* invalidate_inodes - discard the inodes on a device
* @sb: superblock
*
* Discard all of the inodes for a given superblock. If the discard
* fails because there are busy inodes then a non zero value is returned.
* If the discard is successful all the inodes have been discarded.
*/
int invalidate_inodes(struct super_block * sb)
{
int busy;
LIST_HEAD(throw_away);
mutex_lock(&iprune_mutex);
spin_lock(&inode_lock);
inotify_unmount_inodes(&sb->s_inodes);
busy = invalidate_list(&sb->s_inodes, &throw_away);
spin_unlock(&inode_lock);
dispose_list(&throw_away);
mutex_unlock(&iprune_mutex);
return busy;
}
EXPORT_SYMBOL(invalidate_inodes);
static int can_unuse(struct inode *inode)
{
if (inode->i_state)
return 0;
if (inode_has_buffers(inode))
return 0;
if (atomic_read(&inode->i_count))
return 0;
if (inode->i_data.nrpages)
return 0;
return 1;
}
/*
* Scan `goal' inodes on the unused list for freeable ones. They are moved to
* a temporary list and then are freed outside inode_lock by dispose_list().
*
* Any inodes which are pinned purely because of attached pagecache have their
* pagecache removed. We expect the final iput() on that inode to add it to
* the front of the inode_unused list. So look for it there and if the
* inode is still freeable, proceed. The right inode is found 99.9% of the
* time in testing on a 4-way.
*
* If the inode has metadata buffers attached to mapping->private_list then
* try to remove them.
*/
static void prune_icache(int nr_to_scan)
{
LIST_HEAD(freeable);
int nr_pruned = 0;
int nr_scanned;
unsigned long reap = 0;
mutex_lock(&iprune_mutex);
spin_lock(&inode_lock);
for (nr_scanned = 0; nr_scanned < nr_to_scan; nr_scanned++) {
struct inode *inode;
if (list_empty(&inode_unused))
break;
inode = list_entry(inode_unused.prev, struct inode, i_list);
if (inode->i_state || atomic_read(&inode->i_count)) {
list_move(&inode->i_list, &inode_unused);
continue;
}
if (inode_has_buffers(inode) || inode->i_data.nrpages) {
__iget(inode);
spin_unlock(&inode_lock);
if (remove_inode_buffers(inode))
reap += invalidate_inode_pages(&inode->i_data);
iput(inode);
spin_lock(&inode_lock);
if (inode != list_entry(inode_unused.next,
struct inode, i_list))
continue; /* wrong inode or list_empty */
if (!can_unuse(inode))
continue;
}
list_move(&inode->i_list, &freeable);
inode->i_state |= I_FREEING;
nr_pruned++;
}
inodes_stat.nr_unused -= nr_pruned;
[PATCH] Light weight event counters The remaining counters in page_state after the zoned VM counter patches have been applied are all just for show in /proc/vmstat. They have no essential function for the VM. We use a simple increment of per cpu variables. In order to avoid the most severe races we disable preempt. Preempt does not prevent the race between an increment and an interrupt handler incrementing the same statistics counter. However, that race is exceedingly rare, we may only loose one increment or so and there is no requirement (at least not in kernel) that the vm event counters have to be accurate. In the non preempt case this results in a simple increment for each counter. For many architectures this will be reduced by the compiler to a single instruction. This single instruction is atomic for i386 and x86_64. And therefore even the rare race condition in an interrupt is avoided for both architectures in most cases. The patchset also adds an off switch for embedded systems that allows a building of linux kernels without these counters. The implementation of these counters is through inline code that hopefully results in only a single instruction increment instruction being emitted (i386, x86_64) or in the increment being hidden though instruction concurrency (EPIC architectures such as ia64 can get that done). Benefits: - VM event counter operations usually reduce to a single inline instruction on i386 and x86_64. - No interrupt disable, only preempt disable for the preempt case. Preempt disable can also be avoided by moving the counter into a spinlock. - Handling is similar to zoned VM counters. - Simple and easily extendable. - Can be omitted to reduce memory use for embedded use. References: RFC http://marc.theaimsgroup.com/?l=linux-kernel&m=113512330605497&w=2 RFC http://marc.theaimsgroup.com/?l=linux-kernel&m=114988082814934&w=2 local_t http://marc.theaimsgroup.com/?l=linux-kernel&m=114991748606690&w=2 V2 http://marc.theaimsgroup.com/?t=115014808400007&r=1&w=2 V3 http://marc.theaimsgroup.com/?l=linux-kernel&m=115024767022346&w=2 V4 http://marc.theaimsgroup.com/?l=linux-kernel&m=115047968808926&w=2 Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-30 01:55:45 -07:00
if (current_is_kswapd())
__count_vm_events(KSWAPD_INODESTEAL, reap);
else
__count_vm_events(PGINODESTEAL, reap);
spin_unlock(&inode_lock);
dispose_list(&freeable);
mutex_unlock(&iprune_mutex);
}
/*
* shrink_icache_memory() will attempt to reclaim some unused inodes. Here,
* "unused" means that no dentries are referring to the inodes: the files are
* not open and the dcache references to those inodes have already been
* reclaimed.
*
* This function is passed the number of inodes to scan, and it returns the
* total number of remaining possibly-reclaimable inodes.
*/
static int shrink_icache_memory(int nr, gfp_t gfp_mask)
{
if (nr) {
/*
* Nasty deadlock avoidance. We may hold various FS locks,
* and we don't want to recurse into the FS that called us
* in clear_inode() and friends..
*/
if (!(gfp_mask & __GFP_FS))
return -1;
prune_icache(nr);
}
return (inodes_stat.nr_unused / 100) * sysctl_vfs_cache_pressure;
}
static void __wait_on_freeing_inode(struct inode *inode);
/*
* Called with the inode lock held.
* NOTE: we are not increasing the inode-refcount, you must call __iget()
* by hand after calling find_inode now! This simplifies iunique and won't
* add any additional branch in the common code.
*/
static struct inode * find_inode(struct super_block * sb, struct hlist_head *head, int (*test)(struct inode *, void *), void *data)
{
struct hlist_node *node;
struct inode * inode = NULL;
repeat:
hlist_for_each (node, head) {
inode = hlist_entry(node, struct inode, i_hash);
if (inode->i_sb != sb)
continue;
if (!test(inode, data))
continue;
if (inode->i_state & (I_FREEING|I_CLEAR|I_WILL_FREE)) {
__wait_on_freeing_inode(inode);
goto repeat;
}
break;
}
return node ? inode : NULL;
}
/*
* find_inode_fast is the fast path version of find_inode, see the comment at
* iget_locked for details.
*/
static struct inode * find_inode_fast(struct super_block * sb, struct hlist_head *head, unsigned long ino)
{
struct hlist_node *node;
struct inode * inode = NULL;
repeat:
hlist_for_each (node, head) {
inode = hlist_entry(node, struct inode, i_hash);
if (inode->i_ino != ino)
continue;
if (inode->i_sb != sb)
continue;
if (inode->i_state & (I_FREEING|I_CLEAR|I_WILL_FREE)) {
__wait_on_freeing_inode(inode);
goto repeat;
}
break;
}
return node ? inode : NULL;
}
/**
* new_inode - obtain an inode
* @sb: superblock
*
* Allocates a new inode for given superblock.
*/
struct inode *new_inode(struct super_block *sb)
{
static unsigned long last_ino;
struct inode * inode;
spin_lock_prefetch(&inode_lock);
inode = alloc_inode(sb);
if (inode) {
spin_lock(&inode_lock);
inodes_stat.nr_inodes++;
list_add(&inode->i_list, &inode_in_use);
list_add(&inode->i_sb_list, &sb->s_inodes);
inode->i_ino = ++last_ino;
inode->i_state = 0;
spin_unlock(&inode_lock);
}
return inode;
}
EXPORT_SYMBOL(new_inode);
void unlock_new_inode(struct inode *inode)
{
/*
* This is special! We do not need the spinlock
* when clearing I_LOCK, because we're guaranteed
* that nobody else tries to do anything about the
* state of the inode when it is locked, as we
* just created it (so there can be no old holders
* that haven't tested I_LOCK).
*/
inode->i_state &= ~(I_LOCK|I_NEW);
wake_up_inode(inode);
}
EXPORT_SYMBOL(unlock_new_inode);
/*
* This is called without the inode lock held.. Be careful.
*
* We no longer cache the sb_flags in i_flags - see fs.h
* -- rmk@arm.uk.linux.org
*/
static struct inode * get_new_inode(struct super_block *sb, struct hlist_head *head, int (*test)(struct inode *, void *), int (*set)(struct inode *, void *), void *data)
{
struct inode * inode;
inode = alloc_inode(sb);
if (inode) {
struct inode * old;
spin_lock(&inode_lock);
/* We released the lock, so.. */
old = find_inode(sb, head, test, data);
if (!old) {
if (set(inode, data))
goto set_failed;
inodes_stat.nr_inodes++;
list_add(&inode->i_list, &inode_in_use);
list_add(&inode->i_sb_list, &sb->s_inodes);
hlist_add_head(&inode->i_hash, head);
inode->i_state = I_LOCK|I_NEW;
spin_unlock(&inode_lock);
/* Return the locked inode with I_NEW set, the
* caller is responsible for filling in the contents
*/
return inode;
}
/*
* Uhhuh, somebody else created the same inode under
* us. Use the old inode instead of the one we just
* allocated.
*/
__iget(old);
spin_unlock(&inode_lock);
destroy_inode(inode);
inode = old;
wait_on_inode(inode);
}
return inode;
set_failed:
spin_unlock(&inode_lock);
destroy_inode(inode);
return NULL;
}
/*
* get_new_inode_fast is the fast path version of get_new_inode, see the
* comment at iget_locked for details.
*/
static struct inode * get_new_inode_fast(struct super_block *sb, struct hlist_head *head, unsigned long ino)
{
struct inode * inode;
inode = alloc_inode(sb);
if (inode) {
struct inode * old;
spin_lock(&inode_lock);
/* We released the lock, so.. */
old = find_inode_fast(sb, head, ino);
if (!old) {
inode->i_ino = ino;
inodes_stat.nr_inodes++;
list_add(&inode->i_list, &inode_in_use);
list_add(&inode->i_sb_list, &sb->s_inodes);
hlist_add_head(&inode->i_hash, head);
inode->i_state = I_LOCK|I_NEW;
spin_unlock(&inode_lock);
/* Return the locked inode with I_NEW set, the
* caller is responsible for filling in the contents
*/
return inode;
}
/*
* Uhhuh, somebody else created the same inode under
* us. Use the old inode instead of the one we just
* allocated.
*/
__iget(old);
spin_unlock(&inode_lock);
destroy_inode(inode);
inode = old;
wait_on_inode(inode);
}
return inode;
}
static unsigned long hash(struct super_block *sb, unsigned long hashval)
{
unsigned long tmp;
tmp = (hashval * (unsigned long)sb) ^ (GOLDEN_RATIO_PRIME + hashval) /
L1_CACHE_BYTES;
tmp = tmp ^ ((tmp ^ GOLDEN_RATIO_PRIME) >> I_HASHBITS);
return tmp & I_HASHMASK;
}
/**
* iunique - get a unique inode number
* @sb: superblock
* @max_reserved: highest reserved inode number
*
* Obtain an inode number that is unique on the system for a given
* superblock. This is used by file systems that have no natural
* permanent inode numbering system. An inode number is returned that
* is higher than the reserved limit but unique.
*
* BUGS:
* With a large number of inodes live on the file system this function
* currently becomes quite slow.
*/
ino_t iunique(struct super_block *sb, ino_t max_reserved)
{
static ino_t counter;
struct inode *inode;
struct hlist_head * head;
ino_t res;
spin_lock(&inode_lock);
retry:
if (counter > max_reserved) {
head = inode_hashtable + hash(sb,counter);
res = counter++;
inode = find_inode_fast(sb, head, res);
if (!inode) {
spin_unlock(&inode_lock);
return res;
}
} else {
counter = max_reserved + 1;
}
goto retry;
}
EXPORT_SYMBOL(iunique);
struct inode *igrab(struct inode *inode)
{
spin_lock(&inode_lock);
if (!(inode->i_state & (I_FREEING|I_WILL_FREE)))
__iget(inode);
else
/*
* Handle the case where s_op->clear_inode is not been
* called yet, and somebody is calling igrab
* while the inode is getting freed.
*/
inode = NULL;
spin_unlock(&inode_lock);
return inode;
}
EXPORT_SYMBOL(igrab);
/**
* ifind - internal function, you want ilookup5() or iget5().
* @sb: super block of file system to search
* @head: the head of the list to search
* @test: callback used for comparisons between inodes
* @data: opaque data pointer to pass to @test
[PATCH] Fix soft lockup due to NTFS: VFS part and explanation Something has changed in the core kernel such that we now get concurrent inode write outs, one e.g via pdflush and one via sys_sync or whatever. This causes a nasty deadlock in ntfs. The only clean solution unfortunately requires a minor vfs api extension. First the deadlock analysis: Prerequisive knowledge: NTFS has a file $MFT (inode 0) loaded at mount time. The NTFS driver uses the page cache for storing the file contents as usual. More interestingly this file contains the table of on-disk inodes as a sequence of MFT_RECORDs. Thus NTFS driver accesses the on-disk inodes by accessing the MFT_RECORDs in the page cache pages of the loaded inode $MFT. The situation: VFS inode X on a mounted ntfs volume is dirty. For same inode X, the ntfs_inode is dirty and thus corresponding on-disk inode, which is as explained above in a dirty PAGE_CACHE_PAGE belonging to the table of inodes ($MFT, inode 0). What happens: Process 1: sys_sync()/umount()/whatever... calls __sync_single_inode() for $MFT -> do_writepages() -> write_page for the dirty page containing the on-disk inode X, the page is now locked -> ntfs_write_mst_block() which clears PageUptodate() on the page to prevent anyone else getting hold of it whilst it does the write out (this is necessary as the on-disk inode needs "fixups" applied before the write to disk which are removed again after the write and PageUptodate is then set again). It then analyses the page looking for dirty on-disk inodes and when it finds one it calls ntfs_may_write_mft_record() to see if it is safe to write this on-disk inode. This then calls ilookup5() to check if the corresponding VFS inode is in icache(). This in turn calls ifind() which waits on the inode lock via wait_on_inode whilst holding the global inode_lock. Process 2: pdflush results in a call to __sync_single_inode for the same VFS inode X on the ntfs volume. This locks the inode (I_LOCK) then calls write-inode -> ntfs_write_inode -> map_mft_record() -> read_cache_page() of the page (in page cache of table of inodes $MFT, inode 0) containing the on-disk inode. This page has PageUptodate() clear because of Process 1 (see above) so read_cache_page() blocks when tries to take the page lock for the page so it can call ntfs_read_page(). Thus Process 1 is holding the page lock on the page containing the on-disk inode X and it is waiting on the inode X to be unlocked in ifind() so it can write the page out and then unlock the page. And Process 2 is holding the inode lock on inode X and is waiting for the page to be unlocked so it can call ntfs_readpage() or discover that Process 1 set PageUptodate() again and use the page. Thus we have a deadlock due to ifind() waiting on the inode lock. The only sensible solution: NTFS does not care whether the VFS inode is locked or not when it calls ilookup5() (it doesn't use the VFS inode at all, it just uses it to find the corresponding ntfs_inode which is of course attached to the VFS inode (both are one single struct); and it uses the ntfs_inode which is subject to its own locking so I_LOCK is irrelevant) hence we want a modified ilookup5_nowait() which is the same as ilookup5() but it does not wait on the inode lock. Without such functionality I would have to keep my own ntfs_inode cache in the NTFS driver just so I can find ntfs_inodes independent of their VFS inodes which would be slow, memory and cpu cycle wasting, and incredibly stupid given the icache already exists in the VFS. Below is a patch that does the ilookup5_nowait() implementation in fs/inode.c and exports it. ilookup5_nowait.diff: Introduce ilookup5_nowait() which is basically the same as ilookup5() but it does not wait on the inode's lock (i.e. it omits the wait_on_inode() done in ifind()). This is needed to avoid a nasty deadlock in NTFS. Signed-off-by: Anton Altaparmakov <aia21@cantab.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-07-13 01:10:44 -07:00
* @wait: if true wait for the inode to be unlocked, if false do not
*
* ifind() searches for the inode specified by @data in the inode
* cache. This is a generalized version of ifind_fast() for file systems where
* the inode number is not sufficient for unique identification of an inode.
*
* If the inode is in the cache, the inode is returned with an incremented
* reference count.
*
* Otherwise NULL is returned.
*
* Note, @test is called with the inode_lock held, so can't sleep.
*/
static struct inode *ifind(struct super_block *sb,
struct hlist_head *head, int (*test)(struct inode *, void *),
[PATCH] Fix soft lockup due to NTFS: VFS part and explanation Something has changed in the core kernel such that we now get concurrent inode write outs, one e.g via pdflush and one via sys_sync or whatever. This causes a nasty deadlock in ntfs. The only clean solution unfortunately requires a minor vfs api extension. First the deadlock analysis: Prerequisive knowledge: NTFS has a file $MFT (inode 0) loaded at mount time. The NTFS driver uses the page cache for storing the file contents as usual. More interestingly this file contains the table of on-disk inodes as a sequence of MFT_RECORDs. Thus NTFS driver accesses the on-disk inodes by accessing the MFT_RECORDs in the page cache pages of the loaded inode $MFT. The situation: VFS inode X on a mounted ntfs volume is dirty. For same inode X, the ntfs_inode is dirty and thus corresponding on-disk inode, which is as explained above in a dirty PAGE_CACHE_PAGE belonging to the table of inodes ($MFT, inode 0). What happens: Process 1: sys_sync()/umount()/whatever... calls __sync_single_inode() for $MFT -> do_writepages() -> write_page for the dirty page containing the on-disk inode X, the page is now locked -> ntfs_write_mst_block() which clears PageUptodate() on the page to prevent anyone else getting hold of it whilst it does the write out (this is necessary as the on-disk inode needs "fixups" applied before the write to disk which are removed again after the write and PageUptodate is then set again). It then analyses the page looking for dirty on-disk inodes and when it finds one it calls ntfs_may_write_mft_record() to see if it is safe to write this on-disk inode. This then calls ilookup5() to check if the corresponding VFS inode is in icache(). This in turn calls ifind() which waits on the inode lock via wait_on_inode whilst holding the global inode_lock. Process 2: pdflush results in a call to __sync_single_inode for the same VFS inode X on the ntfs volume. This locks the inode (I_LOCK) then calls write-inode -> ntfs_write_inode -> map_mft_record() -> read_cache_page() of the page (in page cache of table of inodes $MFT, inode 0) containing the on-disk inode. This page has PageUptodate() clear because of Process 1 (see above) so read_cache_page() blocks when tries to take the page lock for the page so it can call ntfs_read_page(). Thus Process 1 is holding the page lock on the page containing the on-disk inode X and it is waiting on the inode X to be unlocked in ifind() so it can write the page out and then unlock the page. And Process 2 is holding the inode lock on inode X and is waiting for the page to be unlocked so it can call ntfs_readpage() or discover that Process 1 set PageUptodate() again and use the page. Thus we have a deadlock due to ifind() waiting on the inode lock. The only sensible solution: NTFS does not care whether the VFS inode is locked or not when it calls ilookup5() (it doesn't use the VFS inode at all, it just uses it to find the corresponding ntfs_inode which is of course attached to the VFS inode (both are one single struct); and it uses the ntfs_inode which is subject to its own locking so I_LOCK is irrelevant) hence we want a modified ilookup5_nowait() which is the same as ilookup5() but it does not wait on the inode lock. Without such functionality I would have to keep my own ntfs_inode cache in the NTFS driver just so I can find ntfs_inodes independent of their VFS inodes which would be slow, memory and cpu cycle wasting, and incredibly stupid given the icache already exists in the VFS. Below is a patch that does the ilookup5_nowait() implementation in fs/inode.c and exports it. ilookup5_nowait.diff: Introduce ilookup5_nowait() which is basically the same as ilookup5() but it does not wait on the inode's lock (i.e. it omits the wait_on_inode() done in ifind()). This is needed to avoid a nasty deadlock in NTFS. Signed-off-by: Anton Altaparmakov <aia21@cantab.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-07-13 01:10:44 -07:00
void *data, const int wait)
{
struct inode *inode;
spin_lock(&inode_lock);
inode = find_inode(sb, head, test, data);
if (inode) {
__iget(inode);
spin_unlock(&inode_lock);
[PATCH] Fix soft lockup due to NTFS: VFS part and explanation Something has changed in the core kernel such that we now get concurrent inode write outs, one e.g via pdflush and one via sys_sync or whatever. This causes a nasty deadlock in ntfs. The only clean solution unfortunately requires a minor vfs api extension. First the deadlock analysis: Prerequisive knowledge: NTFS has a file $MFT (inode 0) loaded at mount time. The NTFS driver uses the page cache for storing the file contents as usual. More interestingly this file contains the table of on-disk inodes as a sequence of MFT_RECORDs. Thus NTFS driver accesses the on-disk inodes by accessing the MFT_RECORDs in the page cache pages of the loaded inode $MFT. The situation: VFS inode X on a mounted ntfs volume is dirty. For same inode X, the ntfs_inode is dirty and thus corresponding on-disk inode, which is as explained above in a dirty PAGE_CACHE_PAGE belonging to the table of inodes ($MFT, inode 0). What happens: Process 1: sys_sync()/umount()/whatever... calls __sync_single_inode() for $MFT -> do_writepages() -> write_page for the dirty page containing the on-disk inode X, the page is now locked -> ntfs_write_mst_block() which clears PageUptodate() on the page to prevent anyone else getting hold of it whilst it does the write out (this is necessary as the on-disk inode needs "fixups" applied before the write to disk which are removed again after the write and PageUptodate is then set again). It then analyses the page looking for dirty on-disk inodes and when it finds one it calls ntfs_may_write_mft_record() to see if it is safe to write this on-disk inode. This then calls ilookup5() to check if the corresponding VFS inode is in icache(). This in turn calls ifind() which waits on the inode lock via wait_on_inode whilst holding the global inode_lock. Process 2: pdflush results in a call to __sync_single_inode for the same VFS inode X on the ntfs volume. This locks the inode (I_LOCK) then calls write-inode -> ntfs_write_inode -> map_mft_record() -> read_cache_page() of the page (in page cache of table of inodes $MFT, inode 0) containing the on-disk inode. This page has PageUptodate() clear because of Process 1 (see above) so read_cache_page() blocks when tries to take the page lock for the page so it can call ntfs_read_page(). Thus Process 1 is holding the page lock on the page containing the on-disk inode X and it is waiting on the inode X to be unlocked in ifind() so it can write the page out and then unlock the page. And Process 2 is holding the inode lock on inode X and is waiting for the page to be unlocked so it can call ntfs_readpage() or discover that Process 1 set PageUptodate() again and use the page. Thus we have a deadlock due to ifind() waiting on the inode lock. The only sensible solution: NTFS does not care whether the VFS inode is locked or not when it calls ilookup5() (it doesn't use the VFS inode at all, it just uses it to find the corresponding ntfs_inode which is of course attached to the VFS inode (both are one single struct); and it uses the ntfs_inode which is subject to its own locking so I_LOCK is irrelevant) hence we want a modified ilookup5_nowait() which is the same as ilookup5() but it does not wait on the inode lock. Without such functionality I would have to keep my own ntfs_inode cache in the NTFS driver just so I can find ntfs_inodes independent of their VFS inodes which would be slow, memory and cpu cycle wasting, and incredibly stupid given the icache already exists in the VFS. Below is a patch that does the ilookup5_nowait() implementation in fs/inode.c and exports it. ilookup5_nowait.diff: Introduce ilookup5_nowait() which is basically the same as ilookup5() but it does not wait on the inode's lock (i.e. it omits the wait_on_inode() done in ifind()). This is needed to avoid a nasty deadlock in NTFS. Signed-off-by: Anton Altaparmakov <aia21@cantab.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-07-13 01:10:44 -07:00
if (likely(wait))
wait_on_inode(inode);
return inode;
}
spin_unlock(&inode_lock);
return NULL;
}
/**
* ifind_fast - internal function, you want ilookup() or iget().
* @sb: super block of file system to search
* @head: head of the list to search
* @ino: inode number to search for
*
* ifind_fast() searches for the inode @ino in the inode cache. This is for
* file systems where the inode number is sufficient for unique identification
* of an inode.
*
* If the inode is in the cache, the inode is returned with an incremented
* reference count.
*
* Otherwise NULL is returned.
*/
static struct inode *ifind_fast(struct super_block *sb,
struct hlist_head *head, unsigned long ino)
{
struct inode *inode;
spin_lock(&inode_lock);
inode = find_inode_fast(sb, head, ino);
if (inode) {
__iget(inode);
spin_unlock(&inode_lock);
wait_on_inode(inode);
return inode;
}
spin_unlock(&inode_lock);
return NULL;
}
/**
[PATCH] Fix soft lockup due to NTFS: VFS part and explanation Something has changed in the core kernel such that we now get concurrent inode write outs, one e.g via pdflush and one via sys_sync or whatever. This causes a nasty deadlock in ntfs. The only clean solution unfortunately requires a minor vfs api extension. First the deadlock analysis: Prerequisive knowledge: NTFS has a file $MFT (inode 0) loaded at mount time. The NTFS driver uses the page cache for storing the file contents as usual. More interestingly this file contains the table of on-disk inodes as a sequence of MFT_RECORDs. Thus NTFS driver accesses the on-disk inodes by accessing the MFT_RECORDs in the page cache pages of the loaded inode $MFT. The situation: VFS inode X on a mounted ntfs volume is dirty. For same inode X, the ntfs_inode is dirty and thus corresponding on-disk inode, which is as explained above in a dirty PAGE_CACHE_PAGE belonging to the table of inodes ($MFT, inode 0). What happens: Process 1: sys_sync()/umount()/whatever... calls __sync_single_inode() for $MFT -> do_writepages() -> write_page for the dirty page containing the on-disk inode X, the page is now locked -> ntfs_write_mst_block() which clears PageUptodate() on the page to prevent anyone else getting hold of it whilst it does the write out (this is necessary as the on-disk inode needs "fixups" applied before the write to disk which are removed again after the write and PageUptodate is then set again). It then analyses the page looking for dirty on-disk inodes and when it finds one it calls ntfs_may_write_mft_record() to see if it is safe to write this on-disk inode. This then calls ilookup5() to check if the corresponding VFS inode is in icache(). This in turn calls ifind() which waits on the inode lock via wait_on_inode whilst holding the global inode_lock. Process 2: pdflush results in a call to __sync_single_inode for the same VFS inode X on the ntfs volume. This locks the inode (I_LOCK) then calls write-inode -> ntfs_write_inode -> map_mft_record() -> read_cache_page() of the page (in page cache of table of inodes $MFT, inode 0) containing the on-disk inode. This page has PageUptodate() clear because of Process 1 (see above) so read_cache_page() blocks when tries to take the page lock for the page so it can call ntfs_read_page(). Thus Process 1 is holding the page lock on the page containing the on-disk inode X and it is waiting on the inode X to be unlocked in ifind() so it can write the page out and then unlock the page. And Process 2 is holding the inode lock on inode X and is waiting for the page to be unlocked so it can call ntfs_readpage() or discover that Process 1 set PageUptodate() again and use the page. Thus we have a deadlock due to ifind() waiting on the inode lock. The only sensible solution: NTFS does not care whether the VFS inode is locked or not when it calls ilookup5() (it doesn't use the VFS inode at all, it just uses it to find the corresponding ntfs_inode which is of course attached to the VFS inode (both are one single struct); and it uses the ntfs_inode which is subject to its own locking so I_LOCK is irrelevant) hence we want a modified ilookup5_nowait() which is the same as ilookup5() but it does not wait on the inode lock. Without such functionality I would have to keep my own ntfs_inode cache in the NTFS driver just so I can find ntfs_inodes independent of their VFS inodes which would be slow, memory and cpu cycle wasting, and incredibly stupid given the icache already exists in the VFS. Below is a patch that does the ilookup5_nowait() implementation in fs/inode.c and exports it. ilookup5_nowait.diff: Introduce ilookup5_nowait() which is basically the same as ilookup5() but it does not wait on the inode's lock (i.e. it omits the wait_on_inode() done in ifind()). This is needed to avoid a nasty deadlock in NTFS. Signed-off-by: Anton Altaparmakov <aia21@cantab.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-07-13 01:10:44 -07:00
* ilookup5_nowait - search for an inode in the inode cache
* @sb: super block of file system to search
* @hashval: hash value (usually inode number) to search for
* @test: callback used for comparisons between inodes
* @data: opaque data pointer to pass to @test
*
* ilookup5() uses ifind() to search for the inode specified by @hashval and
* @data in the inode cache. This is a generalized version of ilookup() for
* file systems where the inode number is not sufficient for unique
* identification of an inode.
*
* If the inode is in the cache, the inode is returned with an incremented
[PATCH] Fix soft lockup due to NTFS: VFS part and explanation Something has changed in the core kernel such that we now get concurrent inode write outs, one e.g via pdflush and one via sys_sync or whatever. This causes a nasty deadlock in ntfs. The only clean solution unfortunately requires a minor vfs api extension. First the deadlock analysis: Prerequisive knowledge: NTFS has a file $MFT (inode 0) loaded at mount time. The NTFS driver uses the page cache for storing the file contents as usual. More interestingly this file contains the table of on-disk inodes as a sequence of MFT_RECORDs. Thus NTFS driver accesses the on-disk inodes by accessing the MFT_RECORDs in the page cache pages of the loaded inode $MFT. The situation: VFS inode X on a mounted ntfs volume is dirty. For same inode X, the ntfs_inode is dirty and thus corresponding on-disk inode, which is as explained above in a dirty PAGE_CACHE_PAGE belonging to the table of inodes ($MFT, inode 0). What happens: Process 1: sys_sync()/umount()/whatever... calls __sync_single_inode() for $MFT -> do_writepages() -> write_page for the dirty page containing the on-disk inode X, the page is now locked -> ntfs_write_mst_block() which clears PageUptodate() on the page to prevent anyone else getting hold of it whilst it does the write out (this is necessary as the on-disk inode needs "fixups" applied before the write to disk which are removed again after the write and PageUptodate is then set again). It then analyses the page looking for dirty on-disk inodes and when it finds one it calls ntfs_may_write_mft_record() to see if it is safe to write this on-disk inode. This then calls ilookup5() to check if the corresponding VFS inode is in icache(). This in turn calls ifind() which waits on the inode lock via wait_on_inode whilst holding the global inode_lock. Process 2: pdflush results in a call to __sync_single_inode for the same VFS inode X on the ntfs volume. This locks the inode (I_LOCK) then calls write-inode -> ntfs_write_inode -> map_mft_record() -> read_cache_page() of the page (in page cache of table of inodes $MFT, inode 0) containing the on-disk inode. This page has PageUptodate() clear because of Process 1 (see above) so read_cache_page() blocks when tries to take the page lock for the page so it can call ntfs_read_page(). Thus Process 1 is holding the page lock on the page containing the on-disk inode X and it is waiting on the inode X to be unlocked in ifind() so it can write the page out and then unlock the page. And Process 2 is holding the inode lock on inode X and is waiting for the page to be unlocked so it can call ntfs_readpage() or discover that Process 1 set PageUptodate() again and use the page. Thus we have a deadlock due to ifind() waiting on the inode lock. The only sensible solution: NTFS does not care whether the VFS inode is locked or not when it calls ilookup5() (it doesn't use the VFS inode at all, it just uses it to find the corresponding ntfs_inode which is of course attached to the VFS inode (both are one single struct); and it uses the ntfs_inode which is subject to its own locking so I_LOCK is irrelevant) hence we want a modified ilookup5_nowait() which is the same as ilookup5() but it does not wait on the inode lock. Without such functionality I would have to keep my own ntfs_inode cache in the NTFS driver just so I can find ntfs_inodes independent of their VFS inodes which would be slow, memory and cpu cycle wasting, and incredibly stupid given the icache already exists in the VFS. Below is a patch that does the ilookup5_nowait() implementation in fs/inode.c and exports it. ilookup5_nowait.diff: Introduce ilookup5_nowait() which is basically the same as ilookup5() but it does not wait on the inode's lock (i.e. it omits the wait_on_inode() done in ifind()). This is needed to avoid a nasty deadlock in NTFS. Signed-off-by: Anton Altaparmakov <aia21@cantab.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-07-13 01:10:44 -07:00
* reference count. Note, the inode lock is not waited upon so you have to be
* very careful what you do with the returned inode. You probably should be
* using ilookup5() instead.
*
* Otherwise NULL is returned.
*
* Note, @test is called with the inode_lock held, so can't sleep.
*/
struct inode *ilookup5_nowait(struct super_block *sb, unsigned long hashval,
int (*test)(struct inode *, void *), void *data)
{
struct hlist_head *head = inode_hashtable + hash(sb, hashval);
return ifind(sb, head, test, data, 0);
}
EXPORT_SYMBOL(ilookup5_nowait);
/**
* ilookup5 - search for an inode in the inode cache
* @sb: super block of file system to search
* @hashval: hash value (usually inode number) to search for
* @test: callback used for comparisons between inodes
* @data: opaque data pointer to pass to @test
*
* ilookup5() uses ifind() to search for the inode specified by @hashval and
* @data in the inode cache. This is a generalized version of ilookup() for
* file systems where the inode number is not sufficient for unique
* identification of an inode.
*
* If the inode is in the cache, the inode lock is waited upon and the inode is
* returned with an incremented reference count.
*
* Otherwise NULL is returned.
*
* Note, @test is called with the inode_lock held, so can't sleep.
*/
struct inode *ilookup5(struct super_block *sb, unsigned long hashval,
int (*test)(struct inode *, void *), void *data)
{
struct hlist_head *head = inode_hashtable + hash(sb, hashval);
[PATCH] Fix soft lockup due to NTFS: VFS part and explanation Something has changed in the core kernel such that we now get concurrent inode write outs, one e.g via pdflush and one via sys_sync or whatever. This causes a nasty deadlock in ntfs. The only clean solution unfortunately requires a minor vfs api extension. First the deadlock analysis: Prerequisive knowledge: NTFS has a file $MFT (inode 0) loaded at mount time. The NTFS driver uses the page cache for storing the file contents as usual. More interestingly this file contains the table of on-disk inodes as a sequence of MFT_RECORDs. Thus NTFS driver accesses the on-disk inodes by accessing the MFT_RECORDs in the page cache pages of the loaded inode $MFT. The situation: VFS inode X on a mounted ntfs volume is dirty. For same inode X, the ntfs_inode is dirty and thus corresponding on-disk inode, which is as explained above in a dirty PAGE_CACHE_PAGE belonging to the table of inodes ($MFT, inode 0). What happens: Process 1: sys_sync()/umount()/whatever... calls __sync_single_inode() for $MFT -> do_writepages() -> write_page for the dirty page containing the on-disk inode X, the page is now locked -> ntfs_write_mst_block() which clears PageUptodate() on the page to prevent anyone else getting hold of it whilst it does the write out (this is necessary as the on-disk inode needs "fixups" applied before the write to disk which are removed again after the write and PageUptodate is then set again). It then analyses the page looking for dirty on-disk inodes and when it finds one it calls ntfs_may_write_mft_record() to see if it is safe to write this on-disk inode. This then calls ilookup5() to check if the corresponding VFS inode is in icache(). This in turn calls ifind() which waits on the inode lock via wait_on_inode whilst holding the global inode_lock. Process 2: pdflush results in a call to __sync_single_inode for the same VFS inode X on the ntfs volume. This locks the inode (I_LOCK) then calls write-inode -> ntfs_write_inode -> map_mft_record() -> read_cache_page() of the page (in page cache of table of inodes $MFT, inode 0) containing the on-disk inode. This page has PageUptodate() clear because of Process 1 (see above) so read_cache_page() blocks when tries to take the page lock for the page so it can call ntfs_read_page(). Thus Process 1 is holding the page lock on the page containing the on-disk inode X and it is waiting on the inode X to be unlocked in ifind() so it can write the page out and then unlock the page. And Process 2 is holding the inode lock on inode X and is waiting for the page to be unlocked so it can call ntfs_readpage() or discover that Process 1 set PageUptodate() again and use the page. Thus we have a deadlock due to ifind() waiting on the inode lock. The only sensible solution: NTFS does not care whether the VFS inode is locked or not when it calls ilookup5() (it doesn't use the VFS inode at all, it just uses it to find the corresponding ntfs_inode which is of course attached to the VFS inode (both are one single struct); and it uses the ntfs_inode which is subject to its own locking so I_LOCK is irrelevant) hence we want a modified ilookup5_nowait() which is the same as ilookup5() but it does not wait on the inode lock. Without such functionality I would have to keep my own ntfs_inode cache in the NTFS driver just so I can find ntfs_inodes independent of their VFS inodes which would be slow, memory and cpu cycle wasting, and incredibly stupid given the icache already exists in the VFS. Below is a patch that does the ilookup5_nowait() implementation in fs/inode.c and exports it. ilookup5_nowait.diff: Introduce ilookup5_nowait() which is basically the same as ilookup5() but it does not wait on the inode's lock (i.e. it omits the wait_on_inode() done in ifind()). This is needed to avoid a nasty deadlock in NTFS. Signed-off-by: Anton Altaparmakov <aia21@cantab.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-07-13 01:10:44 -07:00
return ifind(sb, head, test, data, 1);
}
EXPORT_SYMBOL(ilookup5);
/**
* ilookup - search for an inode in the inode cache
* @sb: super block of file system to search
* @ino: inode number to search for
*
* ilookup() uses ifind_fast() to search for the inode @ino in the inode cache.
* This is for file systems where the inode number is sufficient for unique
* identification of an inode.
*
* If the inode is in the cache, the inode is returned with an incremented
* reference count.
*
* Otherwise NULL is returned.
*/
struct inode *ilookup(struct super_block *sb, unsigned long ino)
{
struct hlist_head *head = inode_hashtable + hash(sb, ino);
return ifind_fast(sb, head, ino);
}
EXPORT_SYMBOL(ilookup);
/**
* iget5_locked - obtain an inode from a mounted file system
* @sb: super block of file system
* @hashval: hash value (usually inode number) to get
* @test: callback used for comparisons between inodes
* @set: callback used to initialize a new struct inode
* @data: opaque data pointer to pass to @test and @set
*
* This is iget() without the read_inode() portion of get_new_inode().
*
* iget5_locked() uses ifind() to search for the inode specified by @hashval
* and @data in the inode cache and if present it is returned with an increased
* reference count. This is a generalized version of iget_locked() for file
* systems where the inode number is not sufficient for unique identification
* of an inode.
*
* If the inode is not in cache, get_new_inode() is called to allocate a new
* inode and this is returned locked, hashed, and with the I_NEW flag set. The
* file system gets to fill it in before unlocking it via unlock_new_inode().
*
* Note both @test and @set are called with the inode_lock held, so can't sleep.
*/
struct inode *iget5_locked(struct super_block *sb, unsigned long hashval,
int (*test)(struct inode *, void *),
int (*set)(struct inode *, void *), void *data)
{
struct hlist_head *head = inode_hashtable + hash(sb, hashval);
struct inode *inode;
[PATCH] Fix soft lockup due to NTFS: VFS part and explanation Something has changed in the core kernel such that we now get concurrent inode write outs, one e.g via pdflush and one via sys_sync or whatever. This causes a nasty deadlock in ntfs. The only clean solution unfortunately requires a minor vfs api extension. First the deadlock analysis: Prerequisive knowledge: NTFS has a file $MFT (inode 0) loaded at mount time. The NTFS driver uses the page cache for storing the file contents as usual. More interestingly this file contains the table of on-disk inodes as a sequence of MFT_RECORDs. Thus NTFS driver accesses the on-disk inodes by accessing the MFT_RECORDs in the page cache pages of the loaded inode $MFT. The situation: VFS inode X on a mounted ntfs volume is dirty. For same inode X, the ntfs_inode is dirty and thus corresponding on-disk inode, which is as explained above in a dirty PAGE_CACHE_PAGE belonging to the table of inodes ($MFT, inode 0). What happens: Process 1: sys_sync()/umount()/whatever... calls __sync_single_inode() for $MFT -> do_writepages() -> write_page for the dirty page containing the on-disk inode X, the page is now locked -> ntfs_write_mst_block() which clears PageUptodate() on the page to prevent anyone else getting hold of it whilst it does the write out (this is necessary as the on-disk inode needs "fixups" applied before the write to disk which are removed again after the write and PageUptodate is then set again). It then analyses the page looking for dirty on-disk inodes and when it finds one it calls ntfs_may_write_mft_record() to see if it is safe to write this on-disk inode. This then calls ilookup5() to check if the corresponding VFS inode is in icache(). This in turn calls ifind() which waits on the inode lock via wait_on_inode whilst holding the global inode_lock. Process 2: pdflush results in a call to __sync_single_inode for the same VFS inode X on the ntfs volume. This locks the inode (I_LOCK) then calls write-inode -> ntfs_write_inode -> map_mft_record() -> read_cache_page() of the page (in page cache of table of inodes $MFT, inode 0) containing the on-disk inode. This page has PageUptodate() clear because of Process 1 (see above) so read_cache_page() blocks when tries to take the page lock for the page so it can call ntfs_read_page(). Thus Process 1 is holding the page lock on the page containing the on-disk inode X and it is waiting on the inode X to be unlocked in ifind() so it can write the page out and then unlock the page. And Process 2 is holding the inode lock on inode X and is waiting for the page to be unlocked so it can call ntfs_readpage() or discover that Process 1 set PageUptodate() again and use the page. Thus we have a deadlock due to ifind() waiting on the inode lock. The only sensible solution: NTFS does not care whether the VFS inode is locked or not when it calls ilookup5() (it doesn't use the VFS inode at all, it just uses it to find the corresponding ntfs_inode which is of course attached to the VFS inode (both are one single struct); and it uses the ntfs_inode which is subject to its own locking so I_LOCK is irrelevant) hence we want a modified ilookup5_nowait() which is the same as ilookup5() but it does not wait on the inode lock. Without such functionality I would have to keep my own ntfs_inode cache in the NTFS driver just so I can find ntfs_inodes independent of their VFS inodes which would be slow, memory and cpu cycle wasting, and incredibly stupid given the icache already exists in the VFS. Below is a patch that does the ilookup5_nowait() implementation in fs/inode.c and exports it. ilookup5_nowait.diff: Introduce ilookup5_nowait() which is basically the same as ilookup5() but it does not wait on the inode's lock (i.e. it omits the wait_on_inode() done in ifind()). This is needed to avoid a nasty deadlock in NTFS. Signed-off-by: Anton Altaparmakov <aia21@cantab.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-07-13 01:10:44 -07:00
inode = ifind(sb, head, test, data, 1);
if (inode)
return inode;
/*
* get_new_inode() will do the right thing, re-trying the search
* in case it had to block at any point.
*/
return get_new_inode(sb, head, test, set, data);
}
EXPORT_SYMBOL(iget5_locked);
/**
* iget_locked - obtain an inode from a mounted file system
* @sb: super block of file system
* @ino: inode number to get
*
* This is iget() without the read_inode() portion of get_new_inode_fast().
*
* iget_locked() uses ifind_fast() to search for the inode specified by @ino in
* the inode cache and if present it is returned with an increased reference
* count. This is for file systems where the inode number is sufficient for
* unique identification of an inode.
*
* If the inode is not in cache, get_new_inode_fast() is called to allocate a
* new inode and this is returned locked, hashed, and with the I_NEW flag set.
* The file system gets to fill it in before unlocking it via
* unlock_new_inode().
*/
struct inode *iget_locked(struct super_block *sb, unsigned long ino)
{
struct hlist_head *head = inode_hashtable + hash(sb, ino);
struct inode *inode;
inode = ifind_fast(sb, head, ino);
if (inode)
return inode;
/*
* get_new_inode_fast() will do the right thing, re-trying the search
* in case it had to block at any point.
*/
return get_new_inode_fast(sb, head, ino);
}
EXPORT_SYMBOL(iget_locked);
/**
* __insert_inode_hash - hash an inode
* @inode: unhashed inode
* @hashval: unsigned long value used to locate this object in the
* inode_hashtable.
*
* Add an inode to the inode hash for this superblock.
*/
void __insert_inode_hash(struct inode *inode, unsigned long hashval)
{
struct hlist_head *head = inode_hashtable + hash(inode->i_sb, hashval);
spin_lock(&inode_lock);
hlist_add_head(&inode->i_hash, head);
spin_unlock(&inode_lock);
}
EXPORT_SYMBOL(__insert_inode_hash);
/**
* remove_inode_hash - remove an inode from the hash
* @inode: inode to unhash
*
* Remove an inode from the superblock.
*/
void remove_inode_hash(struct inode *inode)
{
spin_lock(&inode_lock);
hlist_del_init(&inode->i_hash);
spin_unlock(&inode_lock);
}
EXPORT_SYMBOL(remove_inode_hash);
/*
* Tell the filesystem that this inode is no longer of any interest and should
* be completely destroyed.
*
* We leave the inode in the inode hash table until *after* the filesystem's
* ->delete_inode completes. This ensures that an iget (such as nfsd might
* instigate) will always find up-to-date information either in the hash or on
* disk.
*
* I_FREEING is set so that no-one will take a new reference to the inode while
* it is being deleted.
*/
void generic_delete_inode(struct inode *inode)
{
struct super_operations *op = inode->i_sb->s_op;
list_del_init(&inode->i_list);
list_del_init(&inode->i_sb_list);
inode->i_state |= I_FREEING;
inodes_stat.nr_inodes--;
spin_unlock(&inode_lock);
security_inode_delete(inode);
if (op->delete_inode) {
void (*delete)(struct inode *) = op->delete_inode;
if (!is_bad_inode(inode))
DQUOT_INIT(inode);
/* Filesystems implementing their own
* s_op->delete_inode are required to call
* truncate_inode_pages and clear_inode()
* internally */
delete(inode);
} else {
truncate_inode_pages(&inode->i_data, 0);
clear_inode(inode);
}
spin_lock(&inode_lock);
hlist_del_init(&inode->i_hash);
spin_unlock(&inode_lock);
wake_up_inode(inode);
BUG_ON(inode->i_state != I_CLEAR);
destroy_inode(inode);
}
EXPORT_SYMBOL(generic_delete_inode);
static void generic_forget_inode(struct inode *inode)
{
struct super_block *sb = inode->i_sb;
if (!hlist_unhashed(&inode->i_hash)) {
if (!(inode->i_state & (I_DIRTY|I_LOCK)))
list_move(&inode->i_list, &inode_unused);
inodes_stat.nr_unused++;
if (!sb || (sb->s_flags & MS_ACTIVE)) {
spin_unlock(&inode_lock);
return;
}
inode->i_state |= I_WILL_FREE;
spin_unlock(&inode_lock);
write_inode_now(inode, 1);
spin_lock(&inode_lock);
inode->i_state &= ~I_WILL_FREE;
inodes_stat.nr_unused--;
hlist_del_init(&inode->i_hash);
}
list_del_init(&inode->i_list);
list_del_init(&inode->i_sb_list);
inode->i_state |= I_FREEING;
inodes_stat.nr_inodes--;
spin_unlock(&inode_lock);
if (inode->i_data.nrpages)
truncate_inode_pages(&inode->i_data, 0);
clear_inode(inode);
[PATCH] fix nr_unused accounting, and avoid recursing in iput with I_WILL_FREE set list_move(&inode->i_list, &inode_in_use); } else { list_move(&inode->i_list, &inode_unused); + inodes_stat.nr_unused++; } } wake_up_inode(inode); Are you sure the above diff is correct? It was added somewhere between 2.6.5 and 2.6.8. I think it's wrong. The only way I can imagine the i_count to be zero in the above path, is that I_WILL_FREE is set. And if I_WILL_FREE is set, then we must not increase nr_unused. So I believe the above change is buggy and it will definitely overstate the number of unused inodes and it should be backed out. Note that __writeback_single_inode before calling __sync_single_inode, can drop the spinlock and we can have both the dirty and locked bitflags clear here: spin_unlock(&inode_lock); __wait_on_inode(inode); iput(inode); XXXXXXX spin_lock(&inode_lock); } use inode again here a construct like the above makes zero sense from a reference counting standpoint. Either we don't ever use the inode again after the iput, or the inode_lock should be taken _before_ executing the iput (i.e. a __iput would be required). Taking the inode_lock after iput means the iget was useless if we keep using the inode after the iput. So the only chance the 2.6 was safe to call __writeback_single_inode with the i_count == 0, is that I_WILL_FREE is set (I_WILL_FREE will prevent the VM to free the inode in XXXXX). Potentially calling the above iput with I_WILL_FREE was also wrong because it would recurse in iput_final (the second mainline bug). The below (untested) patch fixes the nr_unused accounting, avoids recursing in iput when I_WILL_FREE is set and makes sure (with the BUG_ON) that we don't corrupt memory and that all holders that don't set I_WILL_FREE, keeps a reference on the inode! Signed-off-by: Andrea Arcangeli <andrea@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-30 16:03:05 -07:00
wake_up_inode(inode);
destroy_inode(inode);
}
/*
* Normal UNIX filesystem behaviour: delete the
* inode when the usage count drops to zero, and
* i_nlink is zero.
*/
void generic_drop_inode(struct inode *inode)
{
if (!inode->i_nlink)
generic_delete_inode(inode);
else
generic_forget_inode(inode);
}
EXPORT_SYMBOL_GPL(generic_drop_inode);
/*
* Called when we're dropping the last reference
* to an inode.
*
* Call the FS "drop()" function, defaulting to
* the legacy UNIX filesystem behaviour..
*
* NOTE! NOTE! NOTE! We're called with the inode lock
* held, and the drop function is supposed to release
* the lock!
*/
static inline void iput_final(struct inode *inode)
{
struct super_operations *op = inode->i_sb->s_op;
void (*drop)(struct inode *) = generic_drop_inode;
if (op && op->drop_inode)
drop = op->drop_inode;
drop(inode);
}
/**
* iput - put an inode
* @inode: inode to put
*
* Puts an inode, dropping its usage count. If the inode use count hits
* zero, the inode is then freed and may also be destroyed.
*
* Consequently, iput() can sleep.
*/
void iput(struct inode *inode)
{
if (inode) {
struct super_operations *op = inode->i_sb->s_op;
BUG_ON(inode->i_state == I_CLEAR);
if (op && op->put_inode)
op->put_inode(inode);
if (atomic_dec_and_lock(&inode->i_count, &inode_lock))
iput_final(inode);
}
}
EXPORT_SYMBOL(iput);
/**
* bmap - find a block number in a file
* @inode: inode of file
* @block: block to find
*
* Returns the block number on the device holding the inode that
* is the disk block number for the block of the file requested.
* That is, asked for block 4 of inode 1 the function will return the
* disk block relative to the disk start that holds that block of the
* file.
*/
sector_t bmap(struct inode * inode, sector_t block)
{
sector_t res = 0;
if (inode->i_mapping->a_ops->bmap)
res = inode->i_mapping->a_ops->bmap(inode->i_mapping, block);
return res;
}
EXPORT_SYMBOL(bmap);
/**
* touch_atime - update the access time
* @mnt: mount the inode is accessed on
* @dentry: dentry accessed
*
* Update the accessed time on an inode and mark it for writeback.
* 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)
{
struct inode *inode = dentry->d_inode;
struct timespec now;
if (IS_RDONLY(inode))
return;
if (inode->i_flags & S_NOATIME)
return;
if (inode->i_sb->s_flags & MS_NOATIME)
return;
if ((inode->i_sb->s_flags & MS_NODIRATIME) && S_ISDIR(inode->i_mode))
return;
/*
* We may have a NULL vfsmount when coming from NFSD
*/
if (mnt) {
if (mnt->mnt_flags & MNT_NOATIME)
return;
if ((mnt->mnt_flags & MNT_NODIRATIME) && S_ISDIR(inode->i_mode))
return;
if (mnt->mnt_flags & MNT_RELATIME) {
/*
* With relative atime, only update atime if the
* previous atime is earlier than either the ctime or
* mtime.
*/
if (timespec_compare(&inode->i_mtime,
&inode->i_atime) < 0 &&
timespec_compare(&inode->i_ctime,
&inode->i_atime) < 0)
return;
}
}
now = current_fs_time(inode->i_sb);
if (timespec_equal(&inode->i_atime, &now))
return;
inode->i_atime = now;
mark_inode_dirty_sync(inode);
}
EXPORT_SYMBOL(touch_atime);
/**
* file_update_time - update mtime and ctime time
* @file: file accessed
*
* Update the mtime and ctime members of an inode and mark the inode
* for writeback. Note that this function is meant exclusively for
* usage in the file write path of filesystems, and filesystems may
* choose to explicitly ignore update via this function with the
* S_NOCTIME inode flag, e.g. for network filesystem where these
* timestamps are handled by the server.
*/
void file_update_time(struct file *file)
{
struct inode *inode = file->f_path.dentry->d_inode;
struct timespec now;
int sync_it = 0;
if (IS_NOCMTIME(inode))
return;
if (IS_RDONLY(inode))
return;
now = current_fs_time(inode->i_sb);
if (!timespec_equal(&inode->i_mtime, &now)) {
inode->i_mtime = now;
sync_it = 1;
}
if (!timespec_equal(&inode->i_ctime, &now)) {
inode->i_ctime = now;
sync_it = 1;
}
if (sync_it)
mark_inode_dirty_sync(inode);
}
EXPORT_SYMBOL(file_update_time);
int inode_needs_sync(struct inode *inode)
{
if (IS_SYNC(inode))
return 1;
if (S_ISDIR(inode->i_mode) && IS_DIRSYNC(inode))
return 1;
return 0;
}
EXPORT_SYMBOL(inode_needs_sync);
/*
* Quota functions that want to walk the inode lists..
*/
#ifdef CONFIG_QUOTA
void remove_dquot_ref(struct super_block *sb, int type,
struct list_head *tofree_head)
{
struct inode *inode;
if (!sb->dq_op)
return; /* nothing to do */
spin_lock(&inode_lock); /* This lock is for inodes code */
/*
* We don't have to lock against quota code - test IS_QUOTAINIT is
* just for speedup...
*/
list_for_each_entry(inode, &sb->s_inodes, i_sb_list)
if (!IS_NOQUOTA(inode))
remove_inode_dquot_ref(inode, type, tofree_head);
spin_unlock(&inode_lock);
}
#endif
int inode_wait(void *word)
{
schedule();
return 0;
}
/*
* If we try to find an inode in the inode hash while it is being
* deleted, we have to wait until the filesystem completes its
* deletion before reporting that it isn't found. This function waits
* until the deletion _might_ have completed. Callers are responsible
* to recheck inode state.
*
* It doesn't matter if I_LOCK is not set initially, a call to
* wake_up_inode() after removing from the hash list will DTRT.
*
* This is called with inode_lock held.
*/
static void __wait_on_freeing_inode(struct inode *inode)
{
wait_queue_head_t *wq;
DEFINE_WAIT_BIT(wait, &inode->i_state, __I_LOCK);
wq = bit_waitqueue(&inode->i_state, __I_LOCK);
prepare_to_wait(wq, &wait.wait, TASK_UNINTERRUPTIBLE);
spin_unlock(&inode_lock);
schedule();
finish_wait(wq, &wait.wait);
spin_lock(&inode_lock);
}
void wake_up_inode(struct inode *inode)
{
/*
* Prevent speculative execution through spin_unlock(&inode_lock);
*/
smp_mb();
wake_up_bit(&inode->i_state, __I_LOCK);
}
/*
* We rarely want to lock two inodes that do not have a parent/child
* relationship (such as directory, child inode) simultaneously. The
* vast majority of file systems should be able to get along fine
* without this. Do not use these functions except as a last resort.
*/
void inode_double_lock(struct inode *inode1, struct inode *inode2)
{
if (inode1 == NULL || inode2 == NULL || inode1 == inode2) {
if (inode1)
mutex_lock(&inode1->i_mutex);
else if (inode2)
mutex_lock(&inode2->i_mutex);
return;
}
if (inode1 < inode2) {
mutex_lock_nested(&inode1->i_mutex, I_MUTEX_PARENT);
mutex_lock_nested(&inode2->i_mutex, I_MUTEX_CHILD);
} else {
mutex_lock_nested(&inode2->i_mutex, I_MUTEX_PARENT);
mutex_lock_nested(&inode1->i_mutex, I_MUTEX_CHILD);
}
}
EXPORT_SYMBOL(inode_double_lock);
void inode_double_unlock(struct inode *inode1, struct inode *inode2)
{
if (inode1)
mutex_unlock(&inode1->i_mutex);
if (inode2 && inode2 != inode1)
mutex_unlock(&inode2->i_mutex);
}
EXPORT_SYMBOL(inode_double_unlock);
static __initdata unsigned long ihash_entries;
static int __init set_ihash_entries(char *str)
{
if (!str)
return 0;
ihash_entries = simple_strtoul(str, &str, 0);
return 1;
}
__setup("ihash_entries=", set_ihash_entries);
/*
* Initialize the waitqueues and inode hash table.
*/
void __init inode_init_early(void)
{
int loop;
/* If hashes are distributed across NUMA nodes, defer
* hash allocation until vmalloc space is available.
*/
if (hashdist)
return;
inode_hashtable =
alloc_large_system_hash("Inode-cache",
sizeof(struct hlist_head),
ihash_entries,
14,
HASH_EARLY,
&i_hash_shift,
&i_hash_mask,
0);
for (loop = 0; loop < (1 << i_hash_shift); loop++)
INIT_HLIST_HEAD(&inode_hashtable[loop]);
}
void __init inode_init(unsigned long mempages)
{
int loop;
/* inode slab cache */
inode_cachep = kmem_cache_create("inode_cache",
sizeof(struct inode),
0,
(SLAB_RECLAIM_ACCOUNT|SLAB_PANIC|
SLAB_MEM_SPREAD),
init_once,
NULL);
set_shrinker(DEFAULT_SEEKS, shrink_icache_memory);
/* Hash may have been set up in inode_init_early */
if (!hashdist)
return;
inode_hashtable =
alloc_large_system_hash("Inode-cache",
sizeof(struct hlist_head),
ihash_entries,
14,
0,
&i_hash_shift,
&i_hash_mask,
0);
for (loop = 0; loop < (1 << i_hash_shift); loop++)
INIT_HLIST_HEAD(&inode_hashtable[loop]);
}
void init_special_inode(struct inode *inode, umode_t mode, dev_t rdev)
{
inode->i_mode = mode;
if (S_ISCHR(mode)) {
inode->i_fop = &def_chr_fops;
inode->i_rdev = rdev;
} else if (S_ISBLK(mode)) {
inode->i_fop = &def_blk_fops;
inode->i_rdev = rdev;
} else if (S_ISFIFO(mode))
inode->i_fop = &def_fifo_fops;
else if (S_ISSOCK(mode))
inode->i_fop = &bad_sock_fops;
else
printk(KERN_DEBUG "init_special_inode: bogus i_mode (%o)\n",
mode);
}
EXPORT_SYMBOL(init_special_inode);