video: tegra: dc: Add quick for Vizio P series
[linux-3.10.git] / fs / super.c
index f31ef82..eb03b48 100644 (file)
  *  Heavily rewritten for 'one fs - one tree' dcache architecture. AV, Mar 2000
  */
 
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/slab.h>
-#include <linux/init.h>
-#include <linux/smp_lock.h>
 #include <linux/acct.h>
 #include <linux/blkdev.h>
-#include <linux/quotaops.h>
-#include <linux/namei.h>
-#include <linux/buffer_head.h>         /* for fsync_super() */
 #include <linux/mount.h>
 #include <linux/security.h>
-#include <linux/syscalls.h>
-#include <linux/vfs.h>
 #include <linux/writeback.h>           /* for the emergency remount stuff */
 #include <linux/idr.h>
-#include <linux/kobject.h>
 #include <linux/mutex.h>
-#include <linux/file.h>
-#include <asm/uaccess.h>
+#include <linux/backing-dev.h>
+#include <linux/rculist_bl.h>
+#include <linux/cleancache.h>
+#include <linux/fsnotify.h>
+#include <linux/lockdep.h>
 #include "internal.h"
 
 
 LIST_HEAD(super_blocks);
 DEFINE_SPINLOCK(sb_lock);
 
+static char *sb_writers_name[SB_FREEZE_LEVELS] = {
+       "sb_writers",
+       "sb_pagefaults",
+       "sb_internal",
+};
+
+/*
+ * 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 -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 (!total_objects)
+               total_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;
+}
+
+static int init_sb_writers(struct super_block *s, struct file_system_type *type)
+{
+       int err;
+       int i;
+
+       for (i = 0; i < SB_FREEZE_LEVELS; i++) {
+               err = percpu_counter_init(&s->s_writers.counter[i], 0);
+               if (err < 0)
+                       goto err_out;
+               lockdep_init_map(&s->s_writers.lock_map[i], sb_writers_name[i],
+                                &type->s_writers_key[i], 0);
+       }
+       init_waitqueue_head(&s->s_writers.wait);
+       init_waitqueue_head(&s->s_writers.wait_unfrozen);
+       return 0;
+err_out:
+       while (--i >= 0)
+               percpu_counter_destroy(&s->s_writers.counter[i]);
+       return err;
+}
+
+static void destroy_sb_writers(struct super_block *s)
+{
+       int i;
+
+       for (i = 0; i < SB_FREEZE_LEVELS; i++)
+               percpu_counter_destroy(&s->s_writers.counter[i]);
+}
+
 /**
  *     alloc_super     -       create new superblock
  *     @type:  filesystem type superblock should belong to
+ *     @flags: the mount flags
  *
  *     Allocates and initializes a new &struct super_block.  alloc_super()
  *     returns a pointer new superblock or %NULL if allocation had failed.
  */
-static struct super_block *alloc_super(struct file_system_type *type)
+static struct super_block *alloc_super(struct file_system_type *type, int flags)
 {
        struct super_block *s = kzalloc(sizeof(struct super_block),  GFP_USER);
-       static struct super_operations default_op;
+       static const struct super_operations default_op;
 
        if (s) {
-               if (security_sb_alloc(s)) {
-                       kfree(s);
-                       s = NULL;
-                       goto out;
-               }
-               INIT_LIST_HEAD(&s->s_dirty);
-               INIT_LIST_HEAD(&s->s_io);
-               INIT_LIST_HEAD(&s->s_more_io);
-               INIT_LIST_HEAD(&s->s_files);
-               INIT_LIST_HEAD(&s->s_instances);
-               INIT_HLIST_HEAD(&s->s_anon);
+               if (security_sb_alloc(s))
+                       goto out_free_sb;
+
+               if (init_sb_writers(s, type))
+                       goto err_out;
+               s->s_flags = flags;
+               s->s_bdi = &default_backing_dev_info;
+               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);
                /*
-                * The locking rules for s_lock are up to the
-                * filesystem. For example ext3fs has different
-                * lock ordering than usbfs:
+                * sget() can have s_umount recursion.
+                *
+                * When it cannot find a suitable sb, it allocates a new
+                * one (this one), and tries again to find a suitable old
+                * one.
+                *
+                * In case that succeeds, it will acquire the s_umount
+                * lock of the old one. Since these are clearly distrinct
+                * locks, and this object isn't exposed yet, there's no
+                * risk of deadlocks.
+                *
+                * Annotate this by putting this lock in a different
+                * subclass.
                 */
-               lockdep_set_class(&s->s_lock, &type->s_lock_key);
-               down_write(&s->s_umount);
-               s->s_count = S_BIAS;
+               down_write_nested(&s->s_umount, SINGLE_DEPTH_NESTING);
+               s->s_count = 1;
                atomic_set(&s->s_active, 1);
                mutex_init(&s->s_vfs_rename_mutex);
+               lockdep_set_class(&s->s_vfs_rename_mutex, &type->s_vfs_rename_key);
                mutex_init(&s->s_dquot.dqio_mutex);
                mutex_init(&s->s_dquot.dqonoff_mutex);
                init_rwsem(&s->s_dquot.dqptr_sem);
-               init_waitqueue_head(&s->s_wait_unfrozen);
                s->s_maxbytes = MAX_NON_LFS;
-               s->dq_op = sb_dquot_ops;
-               s->s_qcop = sb_quotactl_ops;
                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;
+err_out:
+       security_sb_free(s);
+       destroy_sb_writers(s);
+out_free_sb:
+       kfree(s);
+       s = NULL;
+       goto out;
 }
 
 /**
@@ -106,7 +221,9 @@ out:
  */
 static inline void destroy_super(struct super_block *s)
 {
+       destroy_sb_writers(s);
        security_sb_free(s);
+       WARN_ON(!list_empty(&s->s_mounts));
        kfree(s->s_subtype);
        kfree(s->s_options);
        kfree(s);
@@ -115,39 +232,14 @@ static inline void destroy_super(struct super_block *s)
 /* Superblock refcounting  */
 
 /*
- * Drop a superblock's refcount.  Returns non-zero if the superblock was
- * destroyed.  The caller must hold sb_lock.
+ * Drop a superblock's refcount.  The caller must hold sb_lock.
  */
-static int __put_super(struct super_block *sb)
+static void __put_super(struct super_block *sb)
 {
-       int ret = 0;
-
        if (!--sb->s_count) {
+               list_del_init(&sb->s_list);
                destroy_super(sb);
-               ret = 1;
        }
-       return ret;
-}
-
-/*
- * Drop a superblock's refcount.
- * Returns non-zero if the superblock is about to be destroyed and
- * at least is already removed from super_blocks list, so if we are
- * making a loop through super blocks then we need to restart.
- * The caller must hold sb_lock.
- */
-int __put_super_and_need_restart(struct super_block *sb)
-{
-       /* check for race with generic_shutdown_super() */
-       if (list_empty(&sb->s_list)) {
-               /* super block is removed, need to restart... */
-               __put_super(sb);
-               return 1;
-       }
-       /* can't be the last, since s_list is still in use */
-       sb->s_count--;
-       BUG_ON(sb->s_count == 0);
-       return 0;
 }
 
 /**
@@ -166,25 +258,47 @@ static void put_super(struct super_block *sb)
 
 
 /**
- *     deactivate_super        -       drop an active reference to superblock
+ *     deactivate_locked_super -       drop an active reference to superblock
  *     @s: superblock to deactivate
  *
- *     Drops an active reference to superblock, acquiring a temprory one if
- *     there is no active references left.  In that case we lock superblock,
+ *     Drops an active reference to superblock, converting it into a temprory
+ *     one if there is no other active references left.  In that case we
  *     tell fs driver to shut it down and drop the temporary reference we
  *     had just acquired.
+ *
+ *     Caller holds exclusive lock on superblock; that lock is released.
  */
-void deactivate_super(struct super_block *s)
+void deactivate_locked_super(struct super_block *s)
 {
        struct file_system_type *fs = s->s_type;
-       if (atomic_dec_and_lock(&s->s_active, &sb_lock)) {
-               s->s_count -= S_BIAS-1;
-               spin_unlock(&sb_lock);
-               DQUOT_OFF(s, 0);
-               down_write(&s->s_umount);
+       if (atomic_dec_and_test(&s->s_active)) {
+               cleancache_invalidate_fs(s);
                fs->kill_sb(s);
+
+               /* caches are now gone, we can safely kill the shrinker now */
+               unregister_shrinker(&s->s_shrink);
                put_filesystem(fs);
                put_super(s);
+       } else {
+               up_write(&s->s_umount);
+       }
+}
+
+EXPORT_SYMBOL(deactivate_locked_super);
+
+/**
+ *     deactivate_super        -       drop an active reference to superblock
+ *     @s: superblock to deactivate
+ *
+ *     Variant of deactivate_locked_super(), except that superblock is *not*
+ *     locked by caller.  If we are going to drop the final active reference,
+ *     lock will be acquired prior to that.
+ */
+void deactivate_super(struct super_block *s)
+{
+        if (!atomic_add_unless(&s->s_active, -1, 1)) {
+               down_write(&s->s_umount);
+               deactivate_locked_super(s);
        }
 }
 
@@ -199,76 +313,55 @@ EXPORT_SYMBOL(deactivate_super);
  *     and want to turn it into a full-blown active reference.  grab_super()
  *     is called with sb_lock held and drops it.  Returns 1 in case of
  *     success, 0 if we had failed (superblock contents was already dead or
- *     dying when grab_super() had been called).
+ *     dying when grab_super() had been called).  Note that this is only
+ *     called for superblocks not in rundown mode (== ones still on ->fs_supers
+ *     of their type), so increment of ->s_count is OK here.
  */
 static int grab_super(struct super_block *s) __releases(sb_lock)
 {
        s->s_count++;
        spin_unlock(&sb_lock);
        down_write(&s->s_umount);
-       if (s->s_root) {
-               spin_lock(&sb_lock);
-               if (s->s_count > S_BIAS) {
-                       atomic_inc(&s->s_active);
-                       s->s_count--;
-                       spin_unlock(&sb_lock);
-                       return 1;
-               }
-               spin_unlock(&sb_lock);
+       if ((s->s_flags & MS_BORN) && atomic_inc_not_zero(&s->s_active)) {
+               put_super(s);
+               return 1;
        }
        up_write(&s->s_umount);
        put_super(s);
-       yield();
        return 0;
 }
 
 /*
- * Superblock locking.  We really ought to get rid of these two.
+ *     grab_super_passive - acquire a passive reference
+ *     @sb: 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.
  */
-void lock_super(struct super_block * sb)
+bool grab_super_passive(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);
-}
+       spin_lock(&sb_lock);
+       if (hlist_unhashed(&sb->s_instances)) {
+               spin_unlock(&sb_lock);
+               return false;
+       }
 
-EXPORT_SYMBOL(lock_super);
-EXPORT_SYMBOL(unlock_super);
+       sb->s_count++;
+       spin_unlock(&sb_lock);
 
-/*
- * Write out and wait upon all dirty data associated with this
- * superblock.  Filesystem data as well as the underlying block
- * device.  Takes the superblock lock.  Requires a second blkdev
- * flush by the caller to complete the operation.
- */
-void __fsync_super(struct super_block *sb)
-{
-       sync_inodes_sb(sb, 0);
-       DQUOT_SYNC(sb);
-       lock_super(sb);
-       if (sb->s_dirt && sb->s_op->write_super)
-               sb->s_op->write_super(sb);
-       unlock_super(sb);
-       if (sb->s_op->sync_fs)
-               sb->s_op->sync_fs(sb, 1);
-       sync_blockdev(sb->s_bdev);
-       sync_inodes_sb(sb, 1);
-}
+       if (down_read_trylock(&sb->s_umount)) {
+               if (sb->s_root && (sb->s_flags & MS_BORN))
+                       return true;
+               up_read(&sb->s_umount);
+       }
 
-/*
- * Write out and wait upon all dirty data associated with this
- * superblock.  Filesystem data as well as the underlying block
- * device.  Takes the superblock lock.
- */
-int fsync_super(struct super_block *sb)
-{
-       __fsync_super(sb);
-       return sync_blockdev(sb->s_bdev);
+       put_super(sb);
+       return false;
 }
 
 /**
@@ -291,32 +384,30 @@ void generic_shutdown_super(struct super_block *sb)
 
        if (sb->s_root) {
                shrink_dcache_for_umount(sb);
-               fsync_super(sb);
-               lock_super(sb);
+               sync_filesystem(sb);
                sb->s_flags &= ~MS_ACTIVE;
-               /* bad name - it should be evict_inodes() */
-               invalidate_inodes(sb);
-               lock_kernel();
 
-               if (sop->write_super && sb->s_dirt)
-                       sop->write_super(sb);
+               fsnotify_unmount_inodes(&sb->s_inodes);
+
+               evict_inodes(sb);
+
+               if (sb->s_dio_done_wq) {
+                       destroy_workqueue(sb->s_dio_done_wq);
+                       sb->s_dio_done_wq = NULL;
+               }
+
                if (sop->put_super)
                        sop->put_super(sb);
 
-               /* Forget any remaining inodes */
-               if (invalidate_inodes(sb)) {
+               if (!list_empty(&sb->s_inodes)) {
                        printk("VFS: Busy inodes after unmount of %s. "
                           "Self-destruct in 5 seconds.  Have a nice day...\n",
                           sb->s_id);
                }
-
-               unlock_kernel();
-               unlock_super(sb);
        }
        spin_lock(&sb_lock);
        /* should be initialized for __put_super_and_need_restart() */
-       list_del_init(&sb->s_list);
-       list_del(&sb->s_instances);
+       hlist_del_init(&sb->s_instances);
        spin_unlock(&sb_lock);
        up_write(&sb->s_umount);
 }
@@ -328,11 +419,13 @@ EXPORT_SYMBOL(generic_shutdown_super);
  *     @type:  filesystem type superblock should belong to
  *     @test:  comparison callback
  *     @set:   setup callback
+ *     @flags: mount flags
  *     @data:  argument to each of them
  */
 struct super_block *sget(struct file_system_type *type,
                        int (*test)(struct super_block *,void *),
                        int (*set)(struct super_block *,void *),
+                       int flags,
                        void *data)
 {
        struct super_block *s = NULL;
@@ -342,19 +435,22 @@ struct super_block *sget(struct file_system_type *type,
 retry:
        spin_lock(&sb_lock);
        if (test) {
-               list_for_each_entry(old, &type->fs_supers, s_instances) {
+               hlist_for_each_entry(old, &type->fs_supers, s_instances) {
                        if (!test(old, data))
                                continue;
                        if (!grab_super(old))
                                goto retry;
-                       if (s)
+                       if (s) {
+                               up_write(&s->s_umount);
                                destroy_super(s);
+                               s = NULL;
+                       }
                        return old;
                }
        }
        if (!s) {
                spin_unlock(&sb_lock);
-               s = alloc_super(type);
+               s = alloc_super(type, flags);
                if (!s)
                        return ERR_PTR(-ENOMEM);
                goto retry;
@@ -363,15 +459,17 @@ retry:
        err = set(s, data);
        if (err) {
                spin_unlock(&sb_lock);
+               up_write(&s->s_umount);
                destroy_super(s);
                return ERR_PTR(err);
        }
        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;
 }
 
@@ -385,94 +483,76 @@ void drop_super(struct super_block *sb)
 
 EXPORT_SYMBOL(drop_super);
 
-static inline void write_super(struct super_block *sb)
-{
-       lock_super(sb);
-       if (sb->s_root && sb->s_dirt)
-               if (sb->s_op->write_super)
-                       sb->s_op->write_super(sb);
-       unlock_super(sb);
-}
-
-/*
- * Note: check the dirty flag before waiting, so we don't
- * hold up the sync while mounting a device. (The newly
- * mounted device won't need syncing.)
+/**
+ *     iterate_supers - call function for all active superblocks
+ *     @f: function to call
+ *     @arg: argument to pass to it
+ *
+ *     Scans the superblock list and calls given function, passing it
+ *     locked superblock and given argument.
  */
-void sync_supers(void)
+void iterate_supers(void (*f)(struct super_block *, void *), void *arg)
 {
-       struct super_block *sb;
+       struct super_block *sb, *p = NULL;
 
        spin_lock(&sb_lock);
-restart:
        list_for_each_entry(sb, &super_blocks, s_list) {
-               if (sb->s_dirt) {
-                       sb->s_count++;
-                       spin_unlock(&sb_lock);
-                       down_read(&sb->s_umount);
-                       write_super(sb);
-                       up_read(&sb->s_umount);
-                       spin_lock(&sb_lock);
-                       if (__put_super_and_need_restart(sb))
-                               goto restart;
-               }
+               if (hlist_unhashed(&sb->s_instances))
+                       continue;
+               sb->s_count++;
+               spin_unlock(&sb_lock);
+
+               down_read(&sb->s_umount);
+               if (sb->s_root && (sb->s_flags & MS_BORN))
+                       f(sb, arg);
+               up_read(&sb->s_umount);
+
+               spin_lock(&sb_lock);
+               if (p)
+                       __put_super(p);
+               p = sb;
        }
+       if (p)
+               __put_super(p);
        spin_unlock(&sb_lock);
 }
 
-/*
- * Call the ->sync_fs super_op against all filesystems which are r/w and
- * which implement it.
- *
- * This operation is careful to avoid the livelock which could easily happen
- * if two or more filesystems are being continuously dirtied.  s_need_sync_fs
- * is used only here.  We set it against all filesystems and then clear it as
- * we sync them.  So redirtied filesystems are skipped.
- *
- * But if process A is currently running sync_filesystems and then process B
- * calls sync_filesystems as well, process B will set all the s_need_sync_fs
- * flags again, which will cause process A to resync everything.  Fix that with
- * a local mutex.
+/**
+ *     iterate_supers_type - call function for superblocks of given type
+ *     @type: fs type
+ *     @f: function to call
+ *     @arg: argument to pass to it
  *
- * (Fabian) Avoid sync_fs with clean fs & wait mode 0
+ *     Scans the superblock list and calls given function, passing it
+ *     locked superblock and given argument.
  */
-void sync_filesystems(int wait)
+void iterate_supers_type(struct file_system_type *type,
+       void (*f)(struct super_block *, void *), void *arg)
 {
-       struct super_block *sb;
-       static DEFINE_MUTEX(mutex);
+       struct super_block *sb, *p = NULL;
 
-       mutex_lock(&mutex);             /* Could be down_interruptible */
        spin_lock(&sb_lock);
-       list_for_each_entry(sb, &super_blocks, s_list) {
-               if (!sb->s_op->sync_fs)
-                       continue;
-               if (sb->s_flags & MS_RDONLY)
-                       continue;
-               sb->s_need_sync_fs = 1;
-       }
-
-restart:
-       list_for_each_entry(sb, &super_blocks, s_list) {
-               if (!sb->s_need_sync_fs)
-                       continue;
-               sb->s_need_sync_fs = 0;
-               if (sb->s_flags & MS_RDONLY)
-                       continue;       /* hm.  Was remounted r/o meanwhile */
+       hlist_for_each_entry(sb, &type->fs_supers, s_instances) {
                sb->s_count++;
                spin_unlock(&sb_lock);
+
                down_read(&sb->s_umount);
-               if (sb->s_root && (wait || sb->s_dirt))
-                       sb->s_op->sync_fs(sb, wait);
+               if (sb->s_root && (sb->s_flags & MS_BORN))
+                       f(sb, arg);
                up_read(&sb->s_umount);
-               /* restart only when sb is no longer on the list */
+
                spin_lock(&sb_lock);
-               if (__put_super_and_need_restart(sb))
-                       goto restart;
+               if (p)
+                       __put_super(p);
+               p = sb;
        }
+       if (p)
+               __put_super(p);
        spin_unlock(&sb_lock);
-       mutex_unlock(&mutex);
 }
 
+EXPORT_SYMBOL(iterate_supers_type);
+
 /**
  *     get_super - get the superblock of a device
  *     @bdev: device to get the superblock for
@@ -481,7 +561,7 @@ restart:
  *     mounted on the device given. %NULL is returned if no match is found.
  */
 
-struct super_block * get_super(struct block_device *bdev)
+struct super_block *get_super(struct block_device *bdev)
 {
        struct super_block *sb;
 
@@ -491,17 +571,20 @@ struct super_block * get_super(struct block_device *bdev)
        spin_lock(&sb_lock);
 rescan:
        list_for_each_entry(sb, &super_blocks, s_list) {
+               if (hlist_unhashed(&sb->s_instances))
+                       continue;
                if (sb->s_bdev == bdev) {
                        sb->s_count++;
                        spin_unlock(&sb_lock);
                        down_read(&sb->s_umount);
-                       if (sb->s_root)
+                       /* still alive? */
+                       if (sb->s_root && (sb->s_flags & MS_BORN))
                                return sb;
                        up_read(&sb->s_umount);
-                       /* restart only when sb is no longer on the list */
+                       /* nope, got unmounted */
                        spin_lock(&sb_lock);
-                       if (__put_super_and_need_restart(sb))
-                               goto rescan;
+                       __put_super(sb);
+                       goto rescan;
                }
        }
        spin_unlock(&sb_lock);
@@ -509,94 +592,88 @@ rescan:
 }
 
 EXPORT_SYMBOL(get_super);
+
+/**
+ *     get_super_thawed - get thawed superblock of a device
+ *     @bdev: device to get the superblock for
+ *
+ *     Scans the superblock list and finds the superblock of the file system
+ *     mounted on the device. The superblock is returned once it is thawed
+ *     (or immediately if it was not frozen). %NULL is returned if no match
+ *     is found.
+ */
+struct super_block *get_super_thawed(struct block_device *bdev)
+{
+       while (1) {
+               struct super_block *s = get_super(bdev);
+               if (!s || s->s_writers.frozen == SB_UNFROZEN)
+                       return s;
+               up_read(&s->s_umount);
+               wait_event(s->s_writers.wait_unfrozen,
+                          s->s_writers.frozen == SB_UNFROZEN);
+               put_super(s);
+       }
+}
+EXPORT_SYMBOL(get_super_thawed);
+
+/**
+ * get_active_super - get an active reference to the superblock of a device
+ * @bdev: device to get the superblock for
+ *
+ * Scans the superblock list and finds the superblock of the file system
+ * mounted on the device given.  Returns the superblock with an active
+ * reference or %NULL if none was found.
+ */
+struct super_block *get_active_super(struct block_device *bdev)
+{
+       struct super_block *sb;
+
+       if (!bdev)
+               return NULL;
+
+restart:
+       spin_lock(&sb_lock);
+       list_for_each_entry(sb, &super_blocks, s_list) {
+               if (hlist_unhashed(&sb->s_instances))
+                       continue;
+               if (sb->s_bdev == bdev) {
+                       if (!grab_super(sb))
+                               goto restart;
+                       up_write(&sb->s_umount);
+                       return sb;
+               }
+       }
+       spin_unlock(&sb_lock);
+       return NULL;
+}
  
-struct super_block * user_get_super(dev_t dev)
+struct super_block *user_get_super(dev_t dev)
 {
        struct super_block *sb;
 
        spin_lock(&sb_lock);
 rescan:
        list_for_each_entry(sb, &super_blocks, s_list) {
+               if (hlist_unhashed(&sb->s_instances))
+                       continue;
                if (sb->s_dev ==  dev) {
                        sb->s_count++;
                        spin_unlock(&sb_lock);
                        down_read(&sb->s_umount);
-                       if (sb->s_root)
+                       /* still alive? */
+                       if (sb->s_root && (sb->s_flags & MS_BORN))
                                return sb;
                        up_read(&sb->s_umount);
-                       /* restart only when sb is no longer on the list */
+                       /* nope, got unmounted */
                        spin_lock(&sb_lock);
-                       if (__put_super_and_need_restart(sb))
-                               goto rescan;
+                       __put_super(sb);
+                       goto rescan;
                }
        }
        spin_unlock(&sb_lock);
        return NULL;
 }
 
-asmlinkage long sys_ustat(unsigned dev, struct ustat __user * ubuf)
-{
-        struct super_block *s;
-        struct ustat tmp;
-        struct kstatfs sbuf;
-       int err = -EINVAL;
-
-        s = user_get_super(new_decode_dev(dev));
-        if (s == NULL)
-                goto out;
-       err = vfs_statfs(s->s_root, &sbuf);
-       drop_super(s);
-       if (err)
-               goto out;
-
-        memset(&tmp,0,sizeof(struct ustat));
-        tmp.f_tfree = sbuf.f_bfree;
-        tmp.f_tinode = sbuf.f_ffree;
-
-        err = copy_to_user(ubuf,&tmp,sizeof(struct ustat)) ? -EFAULT : 0;
-out:
-       return err;
-}
-
-/**
- *     mark_files_ro - mark all files read-only
- *     @sb: superblock in question
- *
- *     All files are marked read-only.  We don't care about pending
- *     delete files so this should be used in 'force' mode only.
- */
-
-static void mark_files_ro(struct super_block *sb)
-{
-       struct file *f;
-
-retry:
-       file_list_lock();
-       list_for_each_entry(f, &sb->s_files, f_u.fu_list) {
-               struct vfsmount *mnt;
-               if (!S_ISREG(f->f_path.dentry->d_inode->i_mode))
-                      continue;
-               if (!file_count(f))
-                       continue;
-               if (!(f->f_mode & FMODE_WRITE))
-                       continue;
-               f->f_mode &= ~FMODE_WRITE;
-               if (file_check_writeable(f) != 0)
-                       continue;
-               file_release_write(f);
-               mnt = mntget(f->f_path.mnt);
-               file_list_unlock();
-               /*
-                * This can sleep, so we can't hold
-                * the file_list_lock() spinlock.
-                */
-               mnt_drop_write(mnt);
-               mntput(mnt);
-               goto retry;
-       }
-       file_list_unlock();
-}
-
 /**
  *     do_remount_sb - asks filesystem to change mount options.
  *     @sb:    superblock in question
@@ -609,72 +686,108 @@ retry:
 int do_remount_sb(struct super_block *sb, int flags, void *data, int force)
 {
        int retval;
-       int remount_rw;
-       
+       int remount_ro;
+
+       if (sb->s_writers.frozen != SB_UNFROZEN)
+               return -EBUSY;
+
 #ifdef CONFIG_BLOCK
        if (!(flags & MS_RDONLY) && bdev_read_only(sb->s_bdev))
                return -EACCES;
 #endif
+
        if (flags & MS_RDONLY)
                acct_auto_close(sb);
        shrink_dcache_sb(sb);
-       fsync_super(sb);
+       sync_filesystem(sb);
+
+       remount_ro = (flags & MS_RDONLY) && !(sb->s_flags & MS_RDONLY);
 
        /* If we are remounting RDONLY and current sb is read/write,
           make sure there are no rw files opened */
-       if ((flags & MS_RDONLY) && !(sb->s_flags & MS_RDONLY)) {
-               if (force)
-                       mark_files_ro(sb);
-               else if (!fs_may_remount_ro(sb))
-                       return -EBUSY;
-               retval = DQUOT_OFF(sb, 1);
-               if (retval < 0 && retval != -ENOSYS)
-                       return -EBUSY;
+       if (remount_ro) {
+               if (force) {
+                       sb->s_readonly_remount = 1;
+                       smp_wmb();
+               } else {
+                       retval = sb_prepare_remount_readonly(sb);
+                       if (retval)
+                               return retval;
+               }
        }
-       remount_rw = !(flags & MS_RDONLY) && (sb->s_flags & MS_RDONLY);
 
        if (sb->s_op->remount_fs) {
-               lock_super(sb);
                retval = sb->s_op->remount_fs(sb, &flags, data);
-               unlock_super(sb);
-               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);
-       if (remount_rw)
-               DQUOT_ON_REMOUNT(sb);
+       /* 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
+        * bdev buffer cache (eg. use a private mapping, or directories in
+        * pagecache, etc). Also file data modifications go via their own
+        * mappings. So If we try to mount readonly then copy the filesystem
+        * from bdev, we could get stale data, so invalidate it to give a best
+        * effort at coherency.
+        */
+       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(unsigned long foo)
+static void do_emergency_remount(struct work_struct *work)
 {
-       struct super_block *sb;
+       struct super_block *sb, *p = NULL;
 
        spin_lock(&sb_lock);
        list_for_each_entry(sb, &super_blocks, s_list) {
+               if (hlist_unhashed(&sb->s_instances))
+                       continue;
                sb->s_count++;
                spin_unlock(&sb_lock);
-               down_read(&sb->s_umount);
-               if (sb->s_root && sb->s_bdev && !(sb->s_flags & MS_RDONLY)) {
+               down_write(&sb->s_umount);
+               if (sb->s_root && sb->s_bdev && (sb->s_flags & MS_BORN) &&
+                   !(sb->s_flags & MS_RDONLY)) {
                        /*
-                        * ->remount_fs needs lock_kernel().
-                        *
                         * What lock protects sb->s_flags??
                         */
-                       lock_kernel();
                        do_remount_sb(sb, MS_RDONLY, NULL, 1);
-                       unlock_kernel();
                }
-               drop_super(sb);
+               up_write(&sb->s_umount);
                spin_lock(&sb_lock);
+               if (p)
+                       __put_super(p);
+               p = sb;
        }
+       if (p)
+               __put_super(p);
        spin_unlock(&sb_lock);
+       kfree(work);
        printk("Emergency Remount complete\n");
 }
 
 void emergency_remount(void)
 {
-       pdflush_operation(do_emergency_remount, 0);
+       struct work_struct *work;
+
+       work = kmalloc(sizeof(*work), GFP_ATOMIC);
+       if (work) {
+               INIT_WORK(work, do_emergency_remount);
+               schedule_work(work);
+       }
 }
 
 /*
@@ -684,8 +797,9 @@ void emergency_remount(void)
 
 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;
@@ -694,7 +808,9 @@ int set_anon_super(struct super_block *s, void *data)
        if (ida_pre_get(&unnamed_dev_ida, GFP_ATOMIC) == 0)
                return -ENOMEM;
        spin_lock(&unnamed_dev_lock);
-       error = ida_get_new(&unnamed_dev_ida, &dev);
+       error = ida_get_new_above(&unnamed_dev_ida, unnamed_dev_start, &dev);
+       if (!error)
+               unnamed_dev_start = dev + 1;
        spin_unlock(&unnamed_dev_lock);
        if (error == -EAGAIN)
                /* We raced and lost with another CPU. */
@@ -702,26 +818,45 @@ int set_anon_super(struct super_block *s, void *data)
        else if (error)
                return -EAGAIN;
 
-       if ((dev & MAX_ID_MASK) == (1 << MINORBITS)) {
+       if (dev == (1 << MINORBITS)) {
                spin_lock(&unnamed_dev_lock);
                ida_remove(&unnamed_dev_ida, dev);
+               if (unnamed_dev_start > dev)
+                       unnamed_dev_start = dev;
                spin_unlock(&unnamed_dev_lock);
                return -EMFILE;
        }
-       s->s_dev = MKDEV(0, dev & MINORMASK);
+       *p = MKDEV(0, dev & MINORMASK);
        return 0;
 }
+EXPORT_SYMBOL(get_anon_bdev);
+
+void free_anon_bdev(dev_t dev)
+{
+       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)
 {
-       int slot = MINOR(sb->s_dev);
-
+       dev_t dev = sb->s_dev;
        generic_shutdown_super(sb);
-       spin_lock(&unnamed_dev_lock);
-       ida_remove(&unnamed_dev_ida, slot);
-       spin_unlock(&unnamed_dev_lock);
+       free_anon_bdev(dev);
 }
 
 EXPORT_SYMBOL(kill_anon_super);
@@ -735,11 +870,53 @@ void kill_litter_super(struct super_block *sb)
 
 EXPORT_SYMBOL(kill_litter_super);
 
+static int ns_test_super(struct super_block *sb, void *data)
+{
+       return sb->s_fs_info == data;
+}
+
+static int ns_set_super(struct super_block *sb, void *data)
+{
+       sb->s_fs_info = data;
+       return set_anon_super(sb, NULL);
+}
+
+struct dentry *mount_ns(struct file_system_type *fs_type, int flags,
+       void *data, int (*fill_super)(struct super_block *, void *, int))
+{
+       struct super_block *sb;
+
+       sb = sget(fs_type, ns_test_super, ns_set_super, flags, data);
+       if (IS_ERR(sb))
+               return ERR_CAST(sb);
+
+       if (!sb->s_root) {
+               int err;
+               err = fill_super(sb, data, flags & MS_SILENT ? 1 : 0);
+               if (err) {
+                       deactivate_locked_super(sb);
+                       return ERR_PTR(err);
+               }
+
+               sb->s_flags |= MS_ACTIVE;
+       }
+
+       return dget(sb->s_root);
+}
+
+EXPORT_SYMBOL(mount_ns);
+
 #ifdef CONFIG_BLOCK
 static int set_bdev_super(struct super_block *s, void *data)
 {
        s->s_bdev = data;
        s->s_dev = s->s_bdev->bd_dev;
+
+       /*
+        * We set the bdi here to the queue backing, file systems can
+        * overwrite this in ->fill_super()
+        */
+       s->s_bdi = &bdev_get_queue(s->s_bdev)->backing_dev_info;
        return 0;
 }
 
@@ -748,231 +925,452 @@ static int test_bdev_super(struct super_block *s, void *data)
        return (void *)s->s_bdev == data;
 }
 
-int get_sb_bdev(struct file_system_type *fs_type,
+struct dentry *mount_bdev(struct file_system_type *fs_type,
        int flags, const char *dev_name, void *data,
-       int (*fill_super)(struct super_block *, void *, int),
-       struct vfsmount *mnt)
+       int (*fill_super)(struct super_block *, void *, int))
 {
        struct block_device *bdev;
        struct super_block *s;
+       fmode_t mode = FMODE_READ | FMODE_EXCL;
        int error = 0;
 
-       bdev = open_bdev_excl(dev_name, flags, fs_type);
+       if (!(flags & MS_RDONLY))
+               mode |= FMODE_WRITE;
+
+       bdev = blkdev_get_by_path(dev_name, mode, fs_type);
        if (IS_ERR(bdev))
-               return PTR_ERR(bdev);
+               return ERR_CAST(bdev);
 
        /*
         * once the super is inserted into the list by sget, s_umount
         * will protect the lockfs code from trying to start a snapshot
         * while we are mounting
         */
-       down(&bdev->bd_mount_sem);
-       s = sget(fs_type, test_bdev_super, set_bdev_super, bdev);
-       up(&bdev->bd_mount_sem);
+       mutex_lock(&bdev->bd_fsfreeze_mutex);
+       if (bdev->bd_fsfreeze_count > 0) {
+               mutex_unlock(&bdev->bd_fsfreeze_mutex);
+               error = -EBUSY;
+               goto error_bdev;
+       }
+       s = sget(fs_type, test_bdev_super, set_bdev_super, flags | MS_NOSEC,
+                bdev);
+       mutex_unlock(&bdev->bd_fsfreeze_mutex);
        if (IS_ERR(s))
                goto error_s;
 
        if (s->s_root) {
                if ((flags ^ s->s_flags) & MS_RDONLY) {
-                       up_write(&s->s_umount);
-                       deactivate_super(s);
+                       deactivate_locked_super(s);
                        error = -EBUSY;
                        goto error_bdev;
                }
 
-               close_bdev_excl(bdev);
+               /*
+                * s_umount nests inside bd_mutex during
+                * __invalidate_device().  blkdev_put() acquires
+                * bd_mutex and can't be called under s_umount.  Drop
+                * s_umount temporarily.  This is safe as we're
+                * holding an active reference.
+                */
+               up_write(&s->s_umount);
+               blkdev_put(bdev, mode);
+               down_write(&s->s_umount);
        } else {
                char b[BDEVNAME_SIZE];
 
-               s->s_flags = flags;
+               s->s_mode = mode;
                strlcpy(s->s_id, bdevname(bdev, b), sizeof(s->s_id));
                sb_set_blocksize(s, block_size(bdev));
                error = fill_super(s, data, flags & MS_SILENT ? 1 : 0);
                if (error) {
-                       up_write(&s->s_umount);
-                       deactivate_super(s);
+                       deactivate_locked_super(s);
                        goto error;
                }
 
                s->s_flags |= MS_ACTIVE;
+               bdev->bd_super = s;
        }
 
-       return simple_set_mnt(mnt, s);
+       return dget(s->s_root);
 
 error_s:
        error = PTR_ERR(s);
 error_bdev:
-       close_bdev_excl(bdev);
+       blkdev_put(bdev, mode);
 error:
-       return error;
+       return ERR_PTR(error);
 }
-
-EXPORT_SYMBOL(get_sb_bdev);
+EXPORT_SYMBOL(mount_bdev);
 
 void kill_block_super(struct super_block *sb)
 {
        struct block_device *bdev = sb->s_bdev;
+       fmode_t mode = sb->s_mode;
 
+       bdev->bd_super = NULL;
        generic_shutdown_super(sb);
        sync_blockdev(bdev);
-       close_bdev_excl(bdev);
+       WARN_ON_ONCE(!(mode & FMODE_EXCL));
+       blkdev_put(bdev, mode | FMODE_EXCL);
 }
 
 EXPORT_SYMBOL(kill_block_super);
 #endif
 
-int get_sb_nodev(struct file_system_type *fs_type,
+struct dentry *mount_nodev(struct file_system_type *fs_type,
        int flags, void *data,
-       int (*fill_super)(struct super_block *, void *, int),
-       struct vfsmount *mnt)
+       int (*fill_super)(struct super_block *, void *, int))
 {
        int error;
-       struct super_block *s = sget(fs_type, NULL, set_anon_super, NULL);
+       struct super_block *s = sget(fs_type, NULL, set_anon_super, flags, NULL);
 
        if (IS_ERR(s))
-               return PTR_ERR(s);
-
-       s->s_flags = flags;
+               return ERR_CAST(s);
 
        error = fill_super(s, data, flags & MS_SILENT ? 1 : 0);
        if (error) {
-               up_write(&s->s_umount);
-               deactivate_super(s);
-               return error;
+               deactivate_locked_super(s);
+               return ERR_PTR(error);
        }
        s->s_flags |= MS_ACTIVE;
-       return simple_set_mnt(mnt, s);
+       return dget(s->s_root);
 }
-
-EXPORT_SYMBOL(get_sb_nodev);
+EXPORT_SYMBOL(mount_nodev);
 
 static int compare_single(struct super_block *s, void *p)
 {
        return 1;
 }
 
-int get_sb_single(struct file_system_type *fs_type,
+struct dentry *mount_single(struct file_system_type *fs_type,
        int flags, void *data,
-       int (*fill_super)(struct super_block *, void *, int),
-       struct vfsmount *mnt)
+       int (*fill_super)(struct super_block *, void *, int))
 {
        struct super_block *s;
        int error;
 
-       s = sget(fs_type, compare_single, set_anon_super, NULL);
+       s = sget(fs_type, compare_single, set_anon_super, flags, NULL);
        if (IS_ERR(s))
-               return PTR_ERR(s);
+               return ERR_CAST(s);
        if (!s->s_root) {
-               s->s_flags = flags;
                error = fill_super(s, data, flags & MS_SILENT ? 1 : 0);
                if (error) {
-                       up_write(&s->s_umount);
-                       deactivate_super(s);
-                       return error;
+                       deactivate_locked_super(s);
+                       return ERR_PTR(error);
                }
                s->s_flags |= MS_ACTIVE;
+       } else {
+               do_remount_sb(s, flags, data, 0);
        }
-       do_remount_sb(s, flags, data, 0);
-       return simple_set_mnt(mnt, s);
+       return dget(s->s_root);
 }
+EXPORT_SYMBOL(mount_single);
 
-EXPORT_SYMBOL(get_sb_single);
-
-struct vfsmount *
-vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void *data)
+struct dentry *
+mount_fs(struct file_system_type *type, int flags, const char *name, void *data)
 {
-       struct vfsmount *mnt;
+       struct dentry *root;
+       struct super_block *sb;
        char *secdata = NULL;
-       int error;
-
-       if (!type)
-               return ERR_PTR(-ENODEV);
-
-       error = -ENOMEM;
-       mnt = alloc_vfsmnt(name);
-       if (!mnt)
-               goto out;
+       int error = -ENOMEM;
 
        if (data && !(type->fs_flags & FS_BINARY_MOUNTDATA)) {
                secdata = alloc_secdata();
                if (!secdata)
-                       goto out_mnt;
+                       goto out;
 
                error = security_sb_copy_data(data, secdata);
                if (error)
                        goto out_free_secdata;
        }
 
-       error = type->get_sb(type, flags, name, data, mnt);
-       if (error < 0)
+       root = type->mount(type, flags, name, data);
+       if (IS_ERR(root)) {
+               error = PTR_ERR(root);
                goto out_free_secdata;
-       BUG_ON(!mnt->mnt_sb);
+       }
+       sb = root->d_sb;
+       BUG_ON(!sb);
+       WARN_ON(!sb->s_bdi);
+       WARN_ON(sb->s_bdi == &default_backing_dev_info);
+       sb->s_flags |= MS_BORN;
 
-       error = security_sb_kern_mount(mnt->mnt_sb, secdata);
-       if (error)
-               goto out_sb;
+       error = security_sb_kern_mount(sb, flags, secdata);
+       if (error)
+               goto out_sb;
 
-       mnt->mnt_mountpoint = mnt->mnt_root;
-       mnt->mnt_parent = mnt;
-       up_write(&mnt->mnt_sb->s_umount);
+       /*
+        * filesystems should never set s_maxbytes larger than MAX_LFS_FILESIZE
+        * but s_maxbytes was an unsigned long long for many releases. Throw
+        * this warning for a little while to try and catch filesystems that
+        * violate this rule.
+        */
+       WARN((sb->s_maxbytes < 0), "%s set sb->s_maxbytes to "
+               "negative value (%lld)\n", type->name, sb->s_maxbytes);
+
+       up_write(&sb->s_umount);
        free_secdata(secdata);
-       return mnt;
+       return root;
 out_sb:
-       dput(mnt->mnt_root);
-       up_write(&mnt->mnt_sb->s_umount);
-       deactivate_super(mnt->mnt_sb);
+       dput(root);
+       deactivate_locked_super(sb);
 out_free_secdata:
        free_secdata(secdata);
-out_mnt:
-       free_vfsmnt(mnt);
 out:
        return ERR_PTR(error);
 }
 
-EXPORT_SYMBOL_GPL(vfs_kern_mount);
+/*
+ * This is an internal function, please use sb_end_{write,pagefault,intwrite}
+ * instead.
+ */
+void __sb_end_write(struct super_block *sb, int level)
+{
+       percpu_counter_dec(&sb->s_writers.counter[level-1]);
+       /*
+        * Make sure s_writers are updated before we wake up waiters in
+        * freeze_super().
+        */
+       smp_mb();
+       if (waitqueue_active(&sb->s_writers.wait))
+               wake_up(&sb->s_writers.wait);
+       rwsem_release(&sb->s_writers.lock_map[level-1], 1, _RET_IP_);
+}
+EXPORT_SYMBOL(__sb_end_write);
 
-static struct vfsmount *fs_set_subtype(struct vfsmount *mnt, const char *fstype)
+#ifdef CONFIG_LOCKDEP
+/*
+ * We want lockdep to tell us about possible deadlocks with freezing but
+ * it's it bit tricky to properly instrument it. Getting a freeze protection
+ * works as getting a read lock but there are subtle problems. XFS for example
+ * gets freeze protection on internal level twice in some cases, which is OK
+ * only because we already hold a freeze protection also on higher level. Due
+ * to these cases we have to tell lockdep we are doing trylock when we
+ * already hold a freeze protection for a higher freeze level.
+ */
+static void acquire_freeze_lock(struct super_block *sb, int level, bool trylock,
+                               unsigned long ip)
 {
-       int err;
-       const char *subtype = strchr(fstype, '.');
-       if (subtype) {
-               subtype++;
-               err = -EINVAL;
-               if (!subtype[0])
-                       goto err;
-       } else
-               subtype = "";
+       int i;
+
+       if (!trylock) {
+               for (i = 0; i < level - 1; i++)
+                       if (lock_is_held(&sb->s_writers.lock_map[i])) {
+                               trylock = true;
+                               break;
+                       }
+       }
+       rwsem_acquire_read(&sb->s_writers.lock_map[level-1], 0, trylock, ip);
+}
+#endif
 
-       mnt->mnt_sb->s_subtype = kstrdup(subtype, GFP_KERNEL);
-       err = -ENOMEM;
-       if (!mnt->mnt_sb->s_subtype)
-               goto err;
-       return mnt;
+/*
+ * This is an internal function, please use sb_start_{write,pagefault,intwrite}
+ * instead.
+ */
+int __sb_start_write(struct super_block *sb, int level, bool wait)
+{
+retry:
+       if (unlikely(sb->s_writers.frozen >= level)) {
+               if (!wait)
+                       return 0;
+               wait_event(sb->s_writers.wait_unfrozen,
+                          sb->s_writers.frozen < level);
+       }
 
- err:
-       mntput(mnt);
-       return ERR_PTR(err);
+#ifdef CONFIG_LOCKDEP
+       acquire_freeze_lock(sb, level, !wait, _RET_IP_);
+#endif
+       percpu_counter_inc(&sb->s_writers.counter[level-1]);
+       /*
+        * Make sure counter is updated before we check for frozen.
+        * freeze_super() first sets frozen and then checks the counter.
+        */
+       smp_mb();
+       if (unlikely(sb->s_writers.frozen >= level)) {
+               __sb_end_write(sb, level);
+               goto retry;
+       }
+       return 1;
 }
+EXPORT_SYMBOL(__sb_start_write);
 
-struct vfsmount *
-do_kern_mount(const char *fstype, int flags, const char *name, void *data)
+/**
+ * sb_wait_write - wait until all writers to given file system finish
+ * @sb: the super for which we wait
+ * @level: type of writers we wait for (normal vs page fault)
+ *
+ * This function waits until there are no writers of given type to given file
+ * system. Caller of this function should make sure there can be no new writers
+ * of type @level before calling this function. Otherwise this function can
+ * livelock.
+ */
+static void sb_wait_write(struct super_block *sb, int level)
 {
-       struct file_system_type *type = get_fs_type(fstype);
-       struct vfsmount *mnt;
-       if (!type)
-               return ERR_PTR(-ENODEV);
-       mnt = vfs_kern_mount(type, flags, name, data);
-       if (!IS_ERR(mnt) && (type->fs_flags & FS_HAS_SUBTYPE) &&
-           !mnt->mnt_sb->s_subtype)
-               mnt = fs_set_subtype(mnt, fstype);
-       put_filesystem(type);
-       return mnt;
+       s64 writers;
+
+       /*
+        * We just cycle-through lockdep here so that it does not complain
+        * about returning with lock to userspace
+        */
+       rwsem_acquire(&sb->s_writers.lock_map[level-1], 0, 0, _THIS_IP_);
+       rwsem_release(&sb->s_writers.lock_map[level-1], 1, _THIS_IP_);
+
+       do {
+               DEFINE_WAIT(wait);
+
+               /*
+                * We use a barrier in prepare_to_wait() to separate setting
+                * of frozen and checking of the counter
+                */
+               prepare_to_wait(&sb->s_writers.wait, &wait,
+                               TASK_UNINTERRUPTIBLE);
+
+               writers = percpu_counter_sum(&sb->s_writers.counter[level-1]);
+               if (writers)
+                       schedule();
+
+               finish_wait(&sb->s_writers.wait, &wait);
+       } while (writers);
 }
-EXPORT_SYMBOL_GPL(do_kern_mount);
 
-struct vfsmount *kern_mount_data(struct file_system_type *type, void *data)
+/**
+ * freeze_super - lock the filesystem and force it into a consistent state
+ * @sb: the super to lock
+ *
+ * Syncs the super to make sure the filesystem is consistent and calls the fs's
+ * freeze_fs.  Subsequent calls to this without first thawing the fs will return
+ * -EBUSY.
+ *
+ * During this function, sb->s_writers.frozen goes through these values:
+ *
+ * SB_UNFROZEN: File system is normal, all writes progress as usual.
+ *
+ * SB_FREEZE_WRITE: The file system is in the process of being frozen.  New
+ * writes should be blocked, though page faults are still allowed. We wait for
+ * all writes to complete and then proceed to the next stage.
+ *
+ * SB_FREEZE_PAGEFAULT: Freezing continues. Now also page faults are blocked
+ * but internal fs threads can still modify the filesystem (although they
+ * should not dirty new pages or inodes), writeback can run etc. After waiting
+ * for all running page faults we sync the filesystem which will clean all
+ * dirty pages and inodes (no new dirty pages or inodes can be created when
+ * sync is running).
+ *
+ * SB_FREEZE_FS: The file system is frozen. Now all internal sources of fs
+ * modification are blocked (e.g. XFS preallocation truncation on inode
+ * reclaim). This is usually implemented by blocking new transactions for
+ * filesystems that have them and need this additional guard. After all
+ * internal writers are finished we call ->freeze_fs() to finish filesystem
+ * freezing. Then we transition to SB_FREEZE_COMPLETE state. This state is
+ * mostly auxiliary for filesystems to verify they do not modify frozen fs.
+ *
+ * sb->s_writers.frozen is protected by sb->s_umount.
+ */
+int freeze_super(struct super_block *sb)
 {
-       return vfs_kern_mount(type, MS_KERNMOUNT, type->name, data);
+       int ret;
+
+       atomic_inc(&sb->s_active);
+       down_write(&sb->s_umount);
+       if (sb->s_writers.frozen != SB_UNFROZEN) {
+               deactivate_locked_super(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) {
+               /* Nothing to do really... */
+               sb->s_writers.frozen = SB_FREEZE_COMPLETE;
+               up_write(&sb->s_umount);
+               return 0;
+       }
+
+       /* From now on, no new normal writers can start */
+       sb->s_writers.frozen = SB_FREEZE_WRITE;
+       smp_wmb();
+
+       /* Release s_umount to preserve sb_start_write -> s_umount ordering */
+       up_write(&sb->s_umount);
+
+       sb_wait_write(sb, SB_FREEZE_WRITE);
+
+       /* Now we go and block page faults... */
+       down_write(&sb->s_umount);
+       sb->s_writers.frozen = SB_FREEZE_PAGEFAULT;
+       smp_wmb();
+
+       sb_wait_write(sb, SB_FREEZE_PAGEFAULT);
+
+       /* All writers are done so after syncing there won't be dirty data */
+       sync_filesystem(sb);
+
+       /* Now wait for internal filesystem counter */
+       sb->s_writers.frozen = SB_FREEZE_FS;
+       smp_wmb();
+       sb_wait_write(sb, SB_FREEZE_FS);
+
+       if (sb->s_op->freeze_fs) {
+               ret = sb->s_op->freeze_fs(sb);
+               if (ret) {
+                       printk(KERN_ERR
+                               "VFS:Filesystem freeze failed\n");
+                       sb->s_writers.frozen = SB_UNFROZEN;
+                       smp_wmb();
+                       wake_up(&sb->s_writers.wait_unfrozen);
+                       deactivate_locked_super(sb);
+                       return ret;
+               }
+       }
+       /*
+        * This is just for debugging purposes so that fs can warn if it
+        * sees write activity when frozen is set to SB_FREEZE_COMPLETE.
+        */
+       sb->s_writers.frozen = SB_FREEZE_COMPLETE;
+       up_write(&sb->s_umount);
+       return 0;
 }
+EXPORT_SYMBOL(freeze_super);
+
+/**
+ * thaw_super -- unlock filesystem
+ * @sb: the super to thaw
+ *
+ * Unlocks the filesystem and marks it writeable again after freeze_super().
+ */
+int thaw_super(struct super_block *sb)
+{
+       int error;
+
+       down_write(&sb->s_umount);
+       if (sb->s_writers.frozen == SB_UNFROZEN) {
+               up_write(&sb->s_umount);
+               return -EINVAL;
+       }
 
-EXPORT_SYMBOL_GPL(kern_mount_data);
+       if (sb->s_flags & MS_RDONLY)
+               goto out;
+
+       if (sb->s_op->unfreeze_fs) {
+               error = sb->s_op->unfreeze_fs(sb);
+               if (error) {
+                       printk(KERN_ERR
+                               "VFS:Filesystem thaw failed\n");
+                       up_write(&sb->s_umount);
+                       return error;
+               }
+       }
+
+out:
+       sb->s_writers.frozen = SB_UNFROZEN;
+       smp_wmb();
+       wake_up(&sb->s_writers.wait_unfrozen);
+       deactivate_locked_super(sb);
+
+       return 0;
+}
+EXPORT_SYMBOL(thaw_super);