wake up s_wait_unfrozen when ->freeze_fs fails
[linux-2.6.git] / fs / super.c
index 444da95..6015c02 100644 (file)
 LIST_HEAD(super_blocks);
 DEFINE_SPINLOCK(sb_lock);
 
+/*
+ * One thing we have to be careful of with a per-sb shrinker is that we don't
+ * drop the last active reference to the superblock from within the shrinker.
+ * If that happens we could trigger unregistering the shrinker from within the
+ * shrinker path and that leads to deadlock on the shrinker_rwsem. Hence we
+ * take a passive reference to the superblock to avoid this from occurring.
+ */
+static int prune_super(struct shrinker *shrink, struct shrink_control *sc)
+{
+       struct super_block *sb;
+       int     fs_objects = 0;
+       int     total_objects;
+
+       sb = container_of(shrink, struct super_block, s_shrink);
+
+       /*
+        * 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 (sc->nr_to_scan && !(sc->gfp_mask & __GFP_FS))
+               return -1;
+
+       if (!grab_super_passive(sb))
+               return !sc->nr_to_scan ? 0 : -1;
+
+       if (sb->s_op && sb->s_op->nr_cached_objects)
+               fs_objects = sb->s_op->nr_cached_objects(sb);
+
+       total_objects = sb->s_nr_dentry_unused +
+                       sb->s_nr_inodes_unused + fs_objects + 1;
+
+       if (sc->nr_to_scan) {
+               int     dentries;
+               int     inodes;
+
+               /* proportion the scan between the caches */
+               dentries = (sc->nr_to_scan * sb->s_nr_dentry_unused) /
+                                                       total_objects;
+               inodes = (sc->nr_to_scan * sb->s_nr_inodes_unused) /
+                                                       total_objects;
+               if (fs_objects)
+                       fs_objects = (sc->nr_to_scan * fs_objects) /
+                                                       total_objects;
+               /*
+                * prune the dcache first as the icache is pinned by it, then
+                * prune the icache, followed by the filesystem specific caches
+                */
+               prune_dcache_sb(sb, dentries);
+               prune_icache_sb(sb, inodes);
+
+               if (fs_objects && sb->s_op->free_cached_objects) {
+                       sb->s_op->free_cached_objects(sb, fs_objects);
+                       fs_objects = sb->s_op->nr_cached_objects(sb);
+               }
+               total_objects = sb->s_nr_dentry_unused +
+                               sb->s_nr_inodes_unused + fs_objects;
+       }
+
+       total_objects = (total_objects / 100) * sysctl_vfs_cache_pressure;
+       drop_super(sb);
+       return total_objects;
+}
+
 /**
  *     alloc_super     -       create new superblock
  *     @type:  filesystem type superblock should belong to
@@ -73,10 +136,13 @@ static struct super_block *alloc_super(struct file_system_type *type)
                INIT_LIST_HEAD(&s->s_files);
 #endif
                s->s_bdi = &default_backing_dev_info;
-               INIT_LIST_HEAD(&s->s_instances);
+               INIT_HLIST_NODE(&s->s_instances);
                INIT_HLIST_BL_HEAD(&s->s_anon);
                INIT_LIST_HEAD(&s->s_inodes);
                INIT_LIST_HEAD(&s->s_dentry_lru);
+               INIT_LIST_HEAD(&s->s_inode_lru);
+               spin_lock_init(&s->s_inode_lru_lock);
+               INIT_LIST_HEAD(&s->s_mounts);
                init_rwsem(&s->s_umount);
                mutex_init(&s->s_lock);
                lockdep_set_class(&s->s_umount, &type->s_umount_key);
@@ -114,6 +180,10 @@ static struct super_block *alloc_super(struct file_system_type *type)
                s->s_op = &default_op;
                s->s_time_gran = 1000000000;
                s->cleancache_poolid = -1;
+
+               s->s_shrink.seeks = DEFAULT_SEEKS;
+               s->s_shrink.shrink = prune_super;
+               s->s_shrink.batch = 1024;
        }
 out:
        return s;
@@ -131,6 +201,7 @@ static inline void destroy_super(struct super_block *s)
        free_percpu(s->s_files);
 #endif
        security_sb_free(s);
+       WARN_ON(!list_empty(&s->s_mounts));
        kfree(s->s_subtype);
        kfree(s->s_options);
        kfree(s);
@@ -141,7 +212,7 @@ static inline void destroy_super(struct super_block *s)
 /*
  * Drop a superblock's refcount.  The caller must hold sb_lock.
  */
-void __put_super(struct super_block *sb)
+static void __put_super(struct super_block *sb)
 {
        if (!--sb->s_count) {
                list_del_init(&sb->s_list);
@@ -156,7 +227,7 @@ void __put_super(struct super_block *sb)
  *     Drops a temporary reference, frees superblock if there's no
  *     references left.
  */
-void put_super(struct super_block *sb)
+static void put_super(struct super_block *sb)
 {
        spin_lock(&sb_lock);
        __put_super(sb);
@@ -181,6 +252,10 @@ void deactivate_locked_super(struct super_block *s)
        if (atomic_dec_and_test(&s->s_active)) {
                cleancache_flush_fs(s);
                fs->kill_sb(s);
+
+               /* caches are now gone, we can safely kill the shrinker now */
+               unregister_shrinker(&s->s_shrink);
+
                /*
                 * We need to call rcu_barrier so all the delayed rcu free
                 * inodes are flushed before we release the fs module.
@@ -241,17 +316,48 @@ static int grab_super(struct super_block *s) __releases(sb_lock)
 }
 
 /*
+ *     grab_super_passive - acquire a passive reference
+ *     @s: reference we are trying to grab
+ *
+ *     Tries to acquire a passive reference. This is used in places where we
+ *     cannot take an active reference but we need to ensure that the
+ *     superblock does not go away while we are working on it. It returns
+ *     false if a reference was not gained, and returns true with the s_umount
+ *     lock held in read mode if a reference is gained. On successful return,
+ *     the caller must drop the s_umount lock and the passive reference when
+ *     done.
+ */
+bool grab_super_passive(struct super_block *sb)
+{
+       spin_lock(&sb_lock);
+       if (hlist_unhashed(&sb->s_instances)) {
+               spin_unlock(&sb_lock);
+               return false;
+       }
+
+       sb->s_count++;
+       spin_unlock(&sb_lock);
+
+       if (down_read_trylock(&sb->s_umount)) {
+               if (sb->s_root && (sb->s_flags & MS_BORN))
+                       return true;
+               up_read(&sb->s_umount);
+       }
+
+       put_super(sb);
+       return false;
+}
+
+/*
  * Superblock locking.  We really ought to get rid of these two.
  */
 void lock_super(struct super_block * sb)
 {
-       get_fs_excl();
        mutex_lock(&sb->s_lock);
 }
 
 void unlock_super(struct super_block * sb)
 {
-       put_fs_excl();
        mutex_unlock(&sb->s_lock);
 }
 
@@ -276,11 +382,9 @@ void generic_shutdown_super(struct super_block *sb)
 {
        const struct super_operations *sop = sb->s_op;
 
-
        if (sb->s_root) {
                shrink_dcache_for_umount(sb);
                sync_filesystem(sb);
-               get_fs_excl();
                sb->s_flags &= ~MS_ACTIVE;
 
                fsnotify_unmount_inodes(&sb->s_inodes);
@@ -295,11 +399,10 @@ void generic_shutdown_super(struct super_block *sb)
                           "Self-destruct in 5 seconds.  Have a nice day...\n",
                           sb->s_id);
                }
-               put_fs_excl();
        }
        spin_lock(&sb_lock);
        /* should be initialized for __put_super_and_need_restart() */
-       list_del_init(&sb->s_instances);
+       hlist_del_init(&sb->s_instances);
        spin_unlock(&sb_lock);
        up_write(&sb->s_umount);
 }
@@ -319,13 +422,14 @@ struct super_block *sget(struct file_system_type *type,
                        void *data)
 {
        struct super_block *s = NULL;
+       struct hlist_node *node;
        struct super_block *old;
        int err;
 
 retry:
        spin_lock(&sb_lock);
        if (test) {
-               list_for_each_entry(old, &type->fs_supers, s_instances) {
+               hlist_for_each_entry(old, node, &type->fs_supers, s_instances) {
                        if (!test(old, data))
                                continue;
                        if (!grab_super(old))
@@ -361,9 +465,10 @@ retry:
        s->s_type = type;
        strlcpy(s->s_id, type->name, sizeof(s->s_id));
        list_add_tail(&s->s_list, &super_blocks);
-       list_add(&s->s_instances, &type->fs_supers);
+       hlist_add_head(&s->s_instances, &type->fs_supers);
        spin_unlock(&sb_lock);
        get_filesystem(type);
+       register_shrinker(&s->s_shrink);
        return s;
 }
 
@@ -395,14 +500,14 @@ void sync_supers(void)
 
        spin_lock(&sb_lock);
        list_for_each_entry(sb, &super_blocks, s_list) {
-               if (list_empty(&sb->s_instances))
+               if (hlist_unhashed(&sb->s_instances))
                        continue;
                if (sb->s_op->write_super && sb->s_dirt) {
                        sb->s_count++;
                        spin_unlock(&sb_lock);
 
                        down_read(&sb->s_umount);
-                       if (sb->s_root && sb->s_dirt)
+                       if (sb->s_root && sb->s_dirt && (sb->s_flags & MS_BORN))
                                sb->s_op->write_super(sb);
                        up_read(&sb->s_umount);
 
@@ -431,13 +536,13 @@ void iterate_supers(void (*f)(struct super_block *, void *), void *arg)
 
        spin_lock(&sb_lock);
        list_for_each_entry(sb, &super_blocks, s_list) {
-               if (list_empty(&sb->s_instances))
+               if (hlist_unhashed(&sb->s_instances))
                        continue;
                sb->s_count++;
                spin_unlock(&sb_lock);
 
                down_read(&sb->s_umount);
-               if (sb->s_root)
+               if (sb->s_root && (sb->s_flags & MS_BORN))
                        f(sb, arg);
                up_read(&sb->s_umount);
 
@@ -464,14 +569,15 @@ void iterate_supers_type(struct file_system_type *type,
        void (*f)(struct super_block *, void *), void *arg)
 {
        struct super_block *sb, *p = NULL;
+       struct hlist_node *node;
 
        spin_lock(&sb_lock);
-       list_for_each_entry(sb, &type->fs_supers, s_instances) {
+       hlist_for_each_entry(sb, node, &type->fs_supers, s_instances) {
                sb->s_count++;
                spin_unlock(&sb_lock);
 
                down_read(&sb->s_umount);
-               if (sb->s_root)
+               if (sb->s_root && (sb->s_flags & MS_BORN))
                        f(sb, arg);
                up_read(&sb->s_umount);
 
@@ -505,14 +611,14 @@ struct super_block *get_super(struct block_device *bdev)
        spin_lock(&sb_lock);
 rescan:
        list_for_each_entry(sb, &super_blocks, s_list) {
-               if (list_empty(&sb->s_instances))
+               if (hlist_unhashed(&sb->s_instances))
                        continue;
                if (sb->s_bdev == bdev) {
                        sb->s_count++;
                        spin_unlock(&sb_lock);
                        down_read(&sb->s_umount);
                        /* still alive? */
-                       if (sb->s_root)
+                       if (sb->s_root && (sb->s_flags & MS_BORN))
                                return sb;
                        up_read(&sb->s_umount);
                        /* nope, got unmounted */
@@ -545,7 +651,7 @@ struct super_block *get_active_super(struct block_device *bdev)
 restart:
        spin_lock(&sb_lock);
        list_for_each_entry(sb, &super_blocks, s_list) {
-               if (list_empty(&sb->s_instances))
+               if (hlist_unhashed(&sb->s_instances))
                        continue;
                if (sb->s_bdev == bdev) {
                        if (grab_super(sb)) /* drops sb_lock */
@@ -565,14 +671,14 @@ struct super_block *user_get_super(dev_t dev)
        spin_lock(&sb_lock);
 rescan:
        list_for_each_entry(sb, &super_blocks, s_list) {
-               if (list_empty(&sb->s_instances))
+               if (hlist_unhashed(&sb->s_instances))
                        continue;
                if (sb->s_dev ==  dev) {
                        sb->s_count++;
                        spin_unlock(&sb_lock);
                        down_read(&sb->s_umount);
                        /* still alive? */
-                       if (sb->s_root)
+                       if (sb->s_root && (sb->s_flags & MS_BORN))
                                return sb;
                        up_read(&sb->s_umount);
                        /* nope, got unmounted */
@@ -617,18 +723,29 @@ int do_remount_sb(struct super_block *sb, int flags, void *data, int force)
        /* If we are remounting RDONLY and current sb is read/write,
           make sure there are no rw files opened */
        if (remount_ro) {
-               if (force)
+               if (force) {
                        mark_files_ro(sb);
-               else if (!fs_may_remount_ro(sb))
-                       return -EBUSY;
+               } else {
+                       retval = sb_prepare_remount_readonly(sb);
+                       if (retval)
+                               return retval;
+               }
        }
 
        if (sb->s_op->remount_fs) {
                retval = sb->s_op->remount_fs(sb, &flags, data);
-               if (retval)
-                       return retval;
+               if (retval) {
+                       if (!force)
+                               goto cancel_readonly;
+                       /* If forced remount, go ahead despite any errors */
+                       WARN(1, "forced remount of a %s fs returned %i\n",
+                            sb->s_type->name, retval);
+               }
        }
        sb->s_flags = (sb->s_flags & ~MS_RMT_MASK) | (flags & MS_RMT_MASK);
+       /* Needs to be ordered wrt mnt_is_readonly() */
+       smp_wmb();
+       sb->s_readonly_remount = 0;
 
        /*
         * Some filesystems modify their metadata via some other path than the
@@ -641,6 +758,10 @@ int do_remount_sb(struct super_block *sb, int flags, void *data, int force)
        if (remount_ro && sb->s_bdev)
                invalidate_bdev(sb->s_bdev);
        return 0;
+
+cancel_readonly:
+       sb->s_readonly_remount = 0;
+       return retval;
 }
 
 static void do_emergency_remount(struct work_struct *work)
@@ -649,12 +770,13 @@ static void do_emergency_remount(struct work_struct *work)
 
        spin_lock(&sb_lock);
        list_for_each_entry(sb, &super_blocks, s_list) {
-               if (list_empty(&sb->s_instances))
+               if (hlist_unhashed(&sb->s_instances))
                        continue;
                sb->s_count++;
                spin_unlock(&sb_lock);
                down_write(&sb->s_umount);
-               if (sb->s_root && sb->s_bdev && !(sb->s_flags & MS_RDONLY)) {
+               if (sb->s_root && sb->s_bdev && (sb->s_flags & MS_BORN) &&
+                   !(sb->s_flags & MS_RDONLY)) {
                        /*
                         * What lock protects sb->s_flags??
                         */
@@ -693,7 +815,7 @@ static DEFINE_IDA(unnamed_dev_ida);
 static DEFINE_SPINLOCK(unnamed_dev_lock);/* protects the above */
 static int unnamed_dev_start = 0; /* don't bother trying below it */
 
-int set_anon_super(struct super_block *s, void *data)
+int get_anon_bdev(dev_t *p)
 {
        int dev;
        int error;
@@ -720,24 +842,38 @@ int set_anon_super(struct super_block *s, void *data)
                spin_unlock(&unnamed_dev_lock);
                return -EMFILE;
        }
-       s->s_dev = MKDEV(0, dev & MINORMASK);
-       s->s_bdi = &noop_backing_dev_info;
+       *p = MKDEV(0, dev & MINORMASK);
        return 0;
 }
+EXPORT_SYMBOL(get_anon_bdev);
 
-EXPORT_SYMBOL(set_anon_super);
-
-void kill_anon_super(struct super_block *sb)
+void free_anon_bdev(dev_t dev)
 {
-       int slot = MINOR(sb->s_dev);
-
-       generic_shutdown_super(sb);
+       int slot = MINOR(dev);
        spin_lock(&unnamed_dev_lock);
        ida_remove(&unnamed_dev_ida, slot);
        if (slot < unnamed_dev_start)
                unnamed_dev_start = slot;
        spin_unlock(&unnamed_dev_lock);
 }
+EXPORT_SYMBOL(free_anon_bdev);
+
+int set_anon_super(struct super_block *s, void *data)
+{
+       int error = get_anon_bdev(&s->s_dev);
+       if (!error)
+               s->s_bdi = &noop_backing_dev_info;
+       return error;
+}
+
+EXPORT_SYMBOL(set_anon_super);
+
+void kill_anon_super(struct super_block *sb)
+{
+       dev_t dev = sb->s_dev;
+       generic_shutdown_super(sb);
+       free_anon_bdev(dev);
+}
 
 EXPORT_SYMBOL(kill_anon_super);
 
@@ -1023,6 +1159,11 @@ int freeze_super(struct super_block *sb)
                return -EBUSY;
        }
 
+       if (!(sb->s_flags & MS_BORN)) {
+               up_write(&sb->s_umount);
+               return 0;       /* sic - it's "nothing to do" */
+       }
+
        if (sb->s_flags & MS_RDONLY) {
                sb->s_frozen = SB_FREEZE_TRANS;
                smp_wmb();
@@ -1045,6 +1186,8 @@ int freeze_super(struct super_block *sb)
                        printk(KERN_ERR
                                "VFS:Filesystem freeze failed\n");
                        sb->s_frozen = SB_UNFROZEN;
+                       smp_wmb();
+                       wake_up(&sb->s_wait_unfrozen);
                        deactivate_locked_super(sb);
                        return ret;
                }