#include <linux/slab.h>
#include <linux/init.h>
#include <linux/smp_lock.h>
+#include <linux/hash.h>
#include <linux/cache.h>
#include <linux/module.h>
#include <linux/mount.h>
#include <linux/security.h>
#include <linux/seqlock.h>
#include <linux/swap.h>
+#include <linux/bootmem.h>
-#define DCACHE_PARANOIA 1
/* #define DCACHE_DEBUG 1 */
+int sysctl_vfs_cache_pressure = 100;
+
spinlock_t dcache_lock __cacheline_aligned_in_smp = SPIN_LOCK_UNLOCKED;
seqlock_t rename_lock __cacheline_aligned_in_smp = SEQLOCK_UNLOCKED;
static kmem_cache_t *dentry_cache;
+#define DNAME_INLINE_LEN (sizeof(struct dentry)-offsetof(struct dentry,d_iname))
+
/*
* This is the single most critical data structure when it comes
* to the dcache: the hashtable for lookups. Somebody should try
.age_limit = 45,
};
-static void d_callback(void *arg)
+static void d_callback(struct rcu_head *head)
{
- struct dentry * dentry = (struct dentry *)arg;
+ struct dentry * dentry = container_of(head, struct dentry, d_rcu);
- if (dname_external(dentry)) {
- kfree(dentry->d_qstr);
- }
+ if (dname_external(dentry))
+ kfree(dentry->d_name.name);
kmem_cache_free(dentry_cache, dentry);
}
{
if (dentry->d_op && dentry->d_op->d_release)
dentry->d_op->d_release(dentry);
- call_rcu(&dentry->d_rcu, d_callback, dentry);
+ call_rcu(&dentry->d_rcu, d_callback);
}
/*
return;
repeat:
+ if (atomic_read(&dentry->d_count) == 1)
+ might_sleep();
if (!atomic_dec_and_lock(&dentry->d_count, &dcache_lock))
return;
if (d_unhashed(dentry))
goto kill_it;
if (list_empty(&dentry->d_lru)) {
- dentry->d_vfs_flags |= DCACHE_REFERENCED;
+ dentry->d_flags |= DCACHE_REFERENCED;
list_add(&dentry->d_lru, &dentry_unused);
dentry_stat.nr_unused++;
}
static inline struct dentry * __dget_locked(struct dentry *dentry)
{
atomic_inc(&dentry->d_count);
- if (atomic_read(&dentry->d_count) == 1) {
+ if (!list_empty(&dentry->d_lru)) {
dentry_stat.nr_unused--;
list_del_init(&dentry->d_lru);
}
* any other hashed alias over that one.
*/
-struct dentry * d_find_alias(struct inode *inode)
+static struct dentry * __d_find_alias(struct inode *inode, int want_discon)
{
struct list_head *head, *next, *tmp;
struct dentry *alias, *discon_alias=NULL;
- spin_lock(&dcache_lock);
head = &inode->i_dentry;
next = inode->i_dentry.next;
while (next != head) {
if (!d_unhashed(alias)) {
if (alias->d_flags & DCACHE_DISCONNECTED)
discon_alias = alias;
- else {
+ else if (!want_discon) {
__dget_locked(alias);
- spin_unlock(&dcache_lock);
return alias;
}
}
}
if (discon_alias)
__dget_locked(discon_alias);
- spin_unlock(&dcache_lock);
return discon_alias;
}
+struct dentry * d_find_alias(struct inode *inode)
+{
+ struct dentry *de;
+ spin_lock(&dcache_lock);
+ de = __d_find_alias(inode, 0);
+ spin_unlock(&dcache_lock);
+ return de;
+}
+
/*
* Try to kill dentries associated with this inode.
* WARNING: you must own a reference to inode.
continue;
}
/* If the dentry was recently referenced, don't free it. */
- if (dentry->d_vfs_flags & DCACHE_REFERENCED) {
- dentry->d_vfs_flags &= ~DCACHE_REFERENCED;
+ if (dentry->d_flags & DCACHE_REFERENCED) {
+ dentry->d_flags &= ~DCACHE_REFERENCED;
list_add(&dentry->d_lru, &dentry_unused);
dentry_stat.nr_unused++;
spin_unlock(&dentry->d_lock);
*
* Prune the dentries that are anonymous
*
- * parsing d_hash list does not read_barrier_depends() as it
+ * parsing d_hash list does not hlist_for_each_rcu() as it
* done under dcache_lock.
*
*/
struct dentry *this = hlist_entry(lp, struct dentry, d_hash);
if (!list_empty(&this->d_lru)) {
dentry_stat.nr_unused--;
- list_del(&this->d_lru);
+ list_del_init(&this->d_lru);
}
/*
}
/*
- * This is called from kswapd when we think we need some more memory.
+ * Scan `nr' dentries and return the number which remain.
+ *
+ * We need to avoid reentering the filesystem if the caller is performing a
+ * GFP_NOFS allocation attempt. One example deadlock is:
+ *
+ * ext2_new_block->getblk->GFP->shrink_dcache_memory->prune_dcache->
+ * prune_one_dentry->dput->dentry_iput->iput->inode->i_sb->s_op->put_inode->
+ * ext2_discard_prealloc->ext2_free_blocks->lock_super->DEADLOCK.
+ *
+ * In this case we return -1 to tell the caller that we baled.
*/
static int shrink_dcache_memory(int nr, unsigned int gfp_mask)
{
if (nr) {
- /*
- * Nasty deadlock avoidance.
- *
- * ext2_new_block->getblk->GFP->shrink_dcache_memory->
- * prune_dcache->prune_one_dentry->dput->dentry_iput->iput->
- * inode->i_sb->s_op->put_inode->ext2_discard_prealloc->
- * ext2_free_blocks->lock_super->DEADLOCK.
- *
- * We should make sure we don't hold the superblock lock over
- * block allocations, but for now:
- */
- if (gfp_mask & __GFP_FS)
- prune_dcache(nr);
+ if (!(gfp_mask & __GFP_FS))
+ return -1;
+ prune_dcache(nr);
}
- return dentry_stat.nr_unused;
+ return (dentry_stat.nr_unused / 100) * sysctl_vfs_cache_pressure;
}
-#define NAME_ALLOC_LEN(len) ((len+16) & ~15)
-
/**
* d_alloc - allocate a dcache entry
* @parent: parent of entry to allocate
* copied and the copy passed in may be reused after this call.
*/
-struct dentry * d_alloc(struct dentry * parent, const struct qstr *name)
+struct dentry *d_alloc(struct dentry * parent, const struct qstr *name)
{
- char * str;
struct dentry *dentry;
- struct qstr * qstr;
+ char *dname;
dentry = kmem_cache_alloc(dentry_cache, GFP_KERNEL);
if (!dentry)
return NULL;
if (name->len > DNAME_INLINE_LEN-1) {
- qstr = kmalloc(sizeof(*qstr) + NAME_ALLOC_LEN(name->len),
- GFP_KERNEL);
- if (!qstr) {
+ dname = kmalloc(name->len + 1, GFP_KERNEL);
+ if (!dname) {
kmem_cache_free(dentry_cache, dentry);
return NULL;
}
- qstr->name = qstr->name_str;
- qstr->len = name->len;
- qstr->hash = name->hash;
- dentry->d_qstr = qstr;
- str = qstr->name_str;
} else {
- dentry->d_qstr = &dentry->d_name;
- str = dentry->d_iname;
+ dname = dentry->d_iname;
}
+ dentry->d_name.name = dname;
- memcpy(str, name->name, name->len);
- str[name->len] = 0;
+ dentry->d_name.len = name->len;
+ dentry->d_name.hash = name->hash;
+ memcpy(dname, name->name, name->len);
+ dname[name->len] = 0;
atomic_set(&dentry->d_count, 1);
- dentry->d_vfs_flags = DCACHE_UNHASHED;
+ dentry->d_flags = DCACHE_UNHASHED;
dentry->d_lock = SPIN_LOCK_UNLOCKED;
- dentry->d_flags = 0;
dentry->d_inode = NULL;
dentry->d_parent = NULL;
- dentry->d_move_count = 0;
dentry->d_sb = NULL;
- dentry->d_name.name = str;
- dentry->d_name.len = name->len;
- dentry->d_name.hash = name->hash;
dentry->d_op = NULL;
dentry->d_fsdata = NULL;
dentry->d_mounted = 0;
struct dentry *res = NULL;
if (root_inode) {
- static const struct qstr name = { .name = "/", .len = 1, .hash = 0 };
+ static const struct qstr name = { .name = "/", .len = 1 };
+
res = d_alloc(NULL, &name);
if (res) {
res->d_sb = root_inode->i_sb;
return res;
}
-static inline struct hlist_head * d_hash(struct dentry * parent, unsigned long hash)
+static inline struct hlist_head *d_hash(struct dentry *parent,
+ unsigned long hash)
{
- hash += (unsigned long) parent / L1_CACHE_BYTES;
- hash = hash ^ (hash >> D_HASHBITS);
+ hash += ((unsigned long) parent ^ GOLDEN_RATIO_PRIME) / L1_CACHE_BYTES;
+ hash = hash ^ ((hash ^ GOLDEN_RATIO_PRIME) >> D_HASHBITS);
return dentry_hashtable + (hash & D_HASHMASK);
}
struct dentry * d_alloc_anon(struct inode *inode)
{
- static const struct qstr anonstring = { "", 0, 0};
+ static const struct qstr anonstring = { .name = "" };
struct dentry *tmp;
struct dentry *res;
tmp->d_parent = tmp; /* make sure dput doesn't croak */
spin_lock(&dcache_lock);
- if (S_ISDIR(inode->i_mode) && !list_empty(&inode->i_dentry)) {
- /* A directory can only have one dentry.
- * This (now) has one, so use it.
- */
- res = list_entry(inode->i_dentry.next, struct dentry, d_alias);
- __dget_locked(res);
- } else {
+ res = __d_find_alias(inode, 0);
+ if (!res) {
/* attach a disconnected dentry */
res = tmp;
tmp = NULL;
- if (res) {
- spin_lock(&res->d_lock);
- res->d_sb = inode->i_sb;
- res->d_parent = res;
- res->d_inode = inode;
- res->d_bucket = d_hash(res, res->d_name.hash);
- res->d_flags |= DCACHE_DISCONNECTED;
- res->d_vfs_flags &= ~DCACHE_UNHASHED;
- list_add(&res->d_alias, &inode->i_dentry);
- hlist_add_head(&res->d_hash, &inode->i_sb->s_anon);
- spin_unlock(&res->d_lock);
- }
+ spin_lock(&res->d_lock);
+ res->d_sb = inode->i_sb;
+ res->d_parent = res;
+ res->d_inode = inode;
+
+ /*
+ * Set d_bucket to an "impossible" bucket address so
+ * that d_move() doesn't get a false positive
+ */
+ res->d_bucket = NULL;
+ res->d_flags |= DCACHE_DISCONNECTED;
+ res->d_flags &= ~DCACHE_UNHASHED;
+ list_add(&res->d_alias, &inode->i_dentry);
+ hlist_add_head(&res->d_hash, &inode->i_sb->s_anon);
+ spin_unlock(&res->d_lock);
+
inode = NULL; /* don't drop reference */
}
spin_unlock(&dcache_lock);
* DCACHE_DISCONNECTED), then d_move that in place of the given dentry
* and return it, else simply d_add the inode to the dentry and return NULL.
*
- * This is (will be) needed in the lookup routine of any filesystem that is exportable
+ * This is needed in the lookup routine of any filesystem that is exportable
* (via knfsd) so that we can build dcache paths to directories effectively.
*
* If a dentry was found and moved, then it is returned. Otherwise NULL
{
struct dentry *new = NULL;
- if (inode && S_ISDIR(inode->i_mode)) {
+ if (inode) {
spin_lock(&dcache_lock);
- if (!list_empty(&inode->i_dentry)) {
- new = list_entry(inode->i_dentry.next, struct dentry, d_alias);
- __dget_locked(new);
+ new = __d_find_alias(inode, 1);
+ if (new) {
+ BUG_ON(!(new->d_flags & DCACHE_DISCONNECTED));
spin_unlock(&dcache_lock);
security_d_instantiate(new, inode);
d_rehash(dentry);
*
* __d_lookup is dcache_lock free. The hash list is protected using RCU.
* Memory barriers are used while updating and doing lockless traversal.
- * To avoid races with d_move while rename is happening, d_move_count is
- * used.
+ * To avoid races with d_move while rename is happening, d_lock is used.
*
* Overflows in memcmp(), while d_move, are avoided by keeping the length
* and name pointer in one structure pointed by d_qstr.
* lookup is going on.
*
* dentry_unused list is not updated even if lookup finds the required dentry
- * in there. It is updated in places such as prune_dcache, shrink_dcache_sb and
- * select_parent. This laziness saves lookup from dcache_lock acquisition.
+ * in there. It is updated in places such as prune_dcache, shrink_dcache_sb,
+ * select_parent and __dget_locked. This laziness saves lookup from dcache_lock
+ * acquisition.
*
* d_lookup() is protected against the concurrent renames in some unrelated
* directory using the seqlockt_t rename_lock.
rcu_read_lock();
- hlist_for_each (node, head) {
+ hlist_for_each_rcu(node, head) {
struct dentry *dentry;
- unsigned long move_count;
- struct qstr * qstr;
+ struct qstr *qstr;
- smp_read_barrier_depends();
dentry = hlist_entry(node, struct dentry, d_hash);
- /* if lookup ends up in a different bucket
- * due to concurrent rename, fail it
- */
- if (unlikely(dentry->d_bucket != head))
- break;
-
- /*
- * We must take a snapshot of d_move_count followed by
- * read memory barrier before any search key comparison
- */
- move_count = dentry->d_move_count;
smp_rmb();
if (dentry->d_name.hash != hash)
if (dentry->d_parent != parent)
continue;
- qstr = dentry->d_qstr;
- smp_read_barrier_depends();
+ spin_lock(&dentry->d_lock);
+
+ /*
+ * If lookup ends up in a different bucket due to concurrent
+ * rename, fail it
+ */
+ if (unlikely(dentry->d_bucket != head))
+ goto terminate;
+
+ /*
+ * Recheck the dentry after taking the lock - d_move may have
+ * changed things. Don't bother checking the hash because we're
+ * about to compare the whole name anyway.
+ */
+ if (dentry->d_parent != parent)
+ goto next;
+
+ qstr = rcu_dereference(&dentry->d_name);
if (parent->d_op && parent->d_op->d_compare) {
if (parent->d_op->d_compare(parent, qstr, name))
- continue;
+ goto next;
} else {
if (qstr->len != len)
- continue;
+ goto next;
if (memcmp(qstr->name, str, len))
- continue;
+ goto next;
}
- spin_lock(&dentry->d_lock);
- /*
- * If dentry is moved, fail the lookup
- */
- if (likely(move_count == dentry->d_move_count)) {
- if (!d_unhashed(dentry)) {
- atomic_inc(&dentry->d_count);
- found = dentry;
- }
+
+ if (!d_unhashed(dentry)) {
+ atomic_inc(&dentry->d_count);
+ found = dentry;
}
+terminate:
spin_unlock(&dentry->d_lock);
break;
+next:
+ spin_unlock(&dentry->d_lock);
}
rcu_read_unlock();
spin_lock(&dcache_lock);
base = d_hash(dparent, dentry->d_name.hash);
hlist_for_each(lhp,base) {
- /* read_barrier_depends() not required for d_hash list
+ /* hlist_for_each_rcu() not required for d_hash list
* as it is parsed under dcache_lock
*/
if (dentry == hlist_entry(lhp, struct dentry, d_hash)) {
void d_rehash(struct dentry * entry)
{
struct hlist_head *list = d_hash(entry->d_parent, entry->d_name.hash);
+
spin_lock(&dcache_lock);
- entry->d_vfs_flags &= ~DCACHE_UNHASHED;
+ spin_lock(&entry->d_lock);
+ entry->d_flags &= ~DCACHE_UNHASHED;
+ spin_unlock(&entry->d_lock);
entry->d_bucket = list;
hlist_add_head_rcu(&entry->d_hash, list);
spin_unlock(&dcache_lock);
* then no longer matches the actual (corrupted) string of the target.
* The hash value has to match the hash queue that the dentry is on..
*/
-static inline void switch_names(struct dentry * dentry, struct dentry * target)
+static void switch_names(struct dentry *dentry, struct dentry *target)
{
- const unsigned char *old_name, *new_name;
- struct qstr *old_qstr, *new_qstr;
-
- memcpy(dentry->d_iname, target->d_iname, DNAME_INLINE_LEN);
- old_qstr = target->d_qstr;
- old_name = target->d_name.name;
- new_qstr = dentry->d_qstr;
- new_name = dentry->d_name.name;
- if (old_name == target->d_iname) {
- old_name = dentry->d_iname;
- old_qstr = &dentry->d_name;
- }
- if (new_name == dentry->d_iname) {
- new_name = target->d_iname;
- new_qstr = &target->d_name;
+ if (dname_external(target)) {
+ if (dname_external(dentry)) {
+ /*
+ * Both external: swap the pointers
+ */
+ do_switch(target->d_name.name, dentry->d_name.name);
+ } else {
+ /*
+ * dentry:internal, target:external. Steal target's
+ * storage and make target internal.
+ */
+ dentry->d_name.name = target->d_name.name;
+ target->d_name.name = target->d_iname;
+ }
+ } else {
+ if (dname_external(dentry)) {
+ /*
+ * dentry:external, target:internal. Give dentry's
+ * storage to target and make dentry internal
+ */
+ memcpy(dentry->d_iname, target->d_name.name,
+ target->d_name.len + 1);
+ target->d_name.name = dentry->d_name.name;
+ dentry->d_name.name = dentry->d_iname;
+ } else {
+ /*
+ * Both are internal. Just copy target to dentry
+ */
+ memcpy(dentry->d_iname, target->d_name.name,
+ target->d_name.len + 1);
+ }
}
- target->d_name.name = new_name;
- dentry->d_name.name = old_name;
- target->d_qstr = new_qstr;
- dentry->d_qstr = old_qstr;
}
/*
}
/* Move the dentry to the target hash queue, if on different bucket */
- if (dentry->d_vfs_flags & DCACHE_UNHASHED)
+ if (dentry->d_flags & DCACHE_UNHASHED)
goto already_unhashed;
if (dentry->d_bucket != target->d_bucket) {
hlist_del_rcu(&dentry->d_hash);
already_unhashed:
dentry->d_bucket = target->d_bucket;
hlist_add_head_rcu(&dentry->d_hash, target->d_bucket);
- dentry->d_vfs_flags &= ~DCACHE_UNHASHED;
+ dentry->d_flags &= ~DCACHE_UNHASHED;
}
/* Unhash the target: dput() will then get rid of it */
}
list_add(&dentry->d_child, &dentry->d_parent->d_subdirs);
- dentry->d_move_count++;
spin_unlock(&target->d_lock);
spin_unlock(&dentry->d_lock);
write_sequnlock(&rename_lock);
char *res;
struct vfsmount *rootmnt;
struct dentry *root;
+
read_lock(¤t->fs->lock);
rootmnt = mntget(current->fs->rootmnt);
root = dget(current->fs->root);
}
__setup("dhash_entries=", set_dhash_entries);
-static void __init dcache_init(unsigned long mempages)
+static void __init dcache_init_early(void)
{
- struct hlist_head *d;
- unsigned long order;
- unsigned int nr_hash;
- int i;
+ int loop;
+
+ dentry_hashtable =
+ alloc_large_system_hash("Dentry cache",
+ sizeof(struct hlist_head),
+ dhash_entries,
+ 13,
+ 0,
+ &d_hash_shift,
+ &d_hash_mask);
+
+ for (loop = 0; loop < (1 << d_hash_shift); loop++)
+ INIT_HLIST_HEAD(&dentry_hashtable[loop]);
+}
+static void __init dcache_init(unsigned long mempages)
+{
/*
* A constructor could be added for stable state like the lists,
* but it is probably not worth it because of the cache nature
dentry_cache = kmem_cache_create("dentry_cache",
sizeof(struct dentry),
0,
- SLAB_RECLAIM_ACCOUNT,
+ SLAB_RECLAIM_ACCOUNT|SLAB_PANIC,
NULL, NULL);
- if (!dentry_cache)
- panic("Cannot create dentry cache");
set_shrinker(DEFAULT_SEEKS, shrink_dcache_memory);
-
- if (!dhash_entries)
- dhash_entries = PAGE_SHIFT < 13 ?
- mempages >> (13 - PAGE_SHIFT) :
- mempages << (PAGE_SHIFT - 13);
-
- dhash_entries *= sizeof(struct hlist_head);
- for (order = 0; ((1UL << order) << PAGE_SHIFT) < dhash_entries; order++)
- ;
-
- do {
- unsigned long tmp;
-
- nr_hash = (1UL << order) * PAGE_SIZE /
- sizeof(struct hlist_head);
- d_hash_mask = (nr_hash - 1);
-
- tmp = nr_hash;
- d_hash_shift = 0;
- while ((tmp >>= 1UL) != 0UL)
- d_hash_shift++;
-
- dentry_hashtable = (struct hlist_head *)
- __get_free_pages(GFP_ATOMIC, order);
- } while (dentry_hashtable == NULL && --order >= 0);
-
- printk(KERN_INFO "Dentry cache hash table entries: %d (order: %ld, %ld bytes)\n",
- nr_hash, order, (PAGE_SIZE << order));
-
- if (!dentry_hashtable)
- panic("Failed to allocate dcache hash table\n");
-
- d = dentry_hashtable;
- i = nr_hash;
- do {
- INIT_HLIST_HEAD(d);
- d++;
- i--;
- } while (i);
}
/* SLAB cache for __getname() consumers */
extern void bdev_cache_init(void);
extern void chrdev_init(void);
+void __init vfs_caches_init_early(void)
+{
+ dcache_init_early();
+ inode_init_early();
+}
+
void __init vfs_caches_init(unsigned long mempages)
{
unsigned long reserve;
/* Base hash sizes on available memory, with a reserve equal to
150% of current kernel size */
- reserve = (mempages - nr_free_pages()) * 3/2;
+ reserve = min((mempages - nr_free_pages()) * 3/2, mempages - 1);
mempages -= reserve;
- names_cachep = kmem_cache_create("names_cache",
- PATH_MAX, 0,
- SLAB_HWCACHE_ALIGN, NULL, NULL);
- if (!names_cachep)
- panic("Cannot create names SLAB cache");
-
- filp_cachep = kmem_cache_create("filp",
- sizeof(struct file), 0,
- SLAB_HWCACHE_ALIGN, filp_ctor, filp_dtor);
- if(!filp_cachep)
- panic("Cannot create filp SLAB cache");
+ names_cachep = kmem_cache_create("names_cache", PATH_MAX, 0,
+ SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL, NULL);
+
+ filp_cachep = kmem_cache_create("filp", sizeof(struct file), 0,
+ SLAB_HWCACHE_ALIGN|SLAB_PANIC, filp_ctor, filp_dtor);
dcache_init(mempages);
inode_init(mempages);
git://git.onelab.eu / linux-2.6.git / blobdiff