Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/torvalds/linux-2.6 into common/serial-rework

Conflicts:
	arch/sh/kernel/cpu/sh2/setup-sh7619.c
	arch/sh/kernel/cpu/sh2a/setup-mxg.c
	arch/sh/kernel/cpu/sh2a/setup-sh7201.c
	arch/sh/kernel/cpu/sh2a/setup-sh7203.c
	arch/sh/kernel/cpu/sh2a/setup-sh7206.c
	arch/sh/kernel/cpu/sh3/setup-sh7705.c
	arch/sh/kernel/cpu/sh3/setup-sh770x.c
	arch/sh/kernel/cpu/sh3/setup-sh7710.c
	arch/sh/kernel/cpu/sh3/setup-sh7720.c
	arch/sh/kernel/cpu/sh4/setup-sh4-202.c
	arch/sh/kernel/cpu/sh4/setup-sh7750.c
	arch/sh/kernel/cpu/sh4/setup-sh7760.c
	arch/sh/kernel/cpu/sh4a/setup-sh7343.c
	arch/sh/kernel/cpu/sh4a/setup-sh7366.c
	arch/sh/kernel/cpu/sh4a/setup-sh7722.c
	arch/sh/kernel/cpu/sh4a/setup-sh7723.c
	arch/sh/kernel/cpu/sh4a/setup-sh7724.c
	arch/sh/kernel/cpu/sh4a/setup-sh7763.c
	arch/sh/kernel/cpu/sh4a/setup-sh7770.c
	arch/sh/kernel/cpu/sh4a/setup-sh7780.c
	arch/sh/kernel/cpu/sh4a/setup-sh7785.c
	arch/sh/kernel/cpu/sh4a/setup-sh7786.c
	arch/sh/kernel/cpu/sh4a/setup-shx3.c
	arch/sh/kernel/cpu/sh5/setup-sh5.c
	drivers/serial/sh-sci.c
	drivers/serial/sh-sci.h
	include/linux/serial_sci.h

1 files changed, 366 insertions, 230 deletions

diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index d28d29c95f7c..51d2e4de34eb 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -27,6 +27,8 @@
 #include <linux/kthread.h>
 #include <linux/freezer.h>
 #include <linux/crc32c.h>
+#include <linux/slab.h>
+#include <linux/migrate.h>
 #include "compat.h"
 #include "ctree.h"
 #include "disk-io.h"
@@ -41,8 +43,7 @@
 
 static struct extent_io_ops btree_extent_io_ops;
 static void end_workqueue_fn(struct btrfs_work *work);
-
-static atomic_t btrfs_bdi_num = ATOMIC_INIT(0);
+static void free_fs_root(struct btrfs_root *root);
 
 /*
  * end_io_wq structs are used to do processing in task context when an IO is
@@ -74,6 +75,11 @@ struct async_submit_bio {
         int rw;
         int mirror_num;
         unsigned long bio_flags;
+        /*
+         * bio_offset is optional, can be used if the pages in the bio
+         * can't tell us where in the file the bio should go
+         */
+        u64 bio_offset;
         struct btrfs_work work;
 };
 
@@ -123,15 +129,15 @@ static struct extent_map *btree_get_extent(struct inode *inode,
         struct extent_map *em;
         int ret;
 
-        spin_lock(&em_tree->lock);
+        read_lock(&em_tree->lock);
         em = lookup_extent_mapping(em_tree, start, len);
         if (em) {
                 em->bdev =
                         BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev;
-                spin_unlock(&em_tree->lock);
+                read_unlock(&em_tree->lock);
                 goto out;
         }
-        spin_unlock(&em_tree->lock);
+        read_unlock(&em_tree->lock);
 
         em = alloc_extent_map(GFP_NOFS);
         if (!em) {
@@ -144,7 +150,7 @@ static struct extent_map *btree_get_extent(struct inode *inode,
         em->block_start = 0;
         em->bdev = BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev;
 
-        spin_lock(&em_tree->lock);
+        write_lock(&em_tree->lock);
         ret = add_extent_mapping(em_tree, em);
         if (ret == -EEXIST) {
                 u64 failed_start = em->start;
@@ -163,7 +169,7 @@ static struct extent_map *btree_get_extent(struct inode *inode,
                 free_extent_map(em);
                 em = NULL;
         }
-        spin_unlock(&em_tree->lock);
+        write_unlock(&em_tree->lock);
 
         if (ret)
                 em = ERR_PTR(ret);
@@ -262,13 +268,15 @@ static int csum_tree_block(struct btrfs_root *root, struct extent_buffer *buf,
 static int verify_parent_transid(struct extent_io_tree *io_tree,
                                  struct extent_buffer *eb, u64 parent_transid)
 {
+        struct extent_state *cached_state = NULL;
         int ret;
 
         if (!parent_transid || btrfs_header_generation(eb) == parent_transid)
                 return 0;
 
-        lock_extent(io_tree, eb->start, eb->start + eb->len - 1, GFP_NOFS);
-        if (extent_buffer_uptodate(io_tree, eb) &&
+        lock_extent_bits(io_tree, eb->start, eb->start + eb->len - 1,
+                         0, &cached_state, GFP_NOFS);
+        if (extent_buffer_uptodate(io_tree, eb, cached_state) &&
             btrfs_header_generation(eb) == parent_transid) {
                 ret = 0;
                 goto out;
@@ -281,10 +289,10 @@ static int verify_parent_transid(struct extent_io_tree *io_tree,
                        (unsigned long long)btrfs_header_generation(eb));
         }
         ret = 1;
-        clear_extent_buffer_uptodate(io_tree, eb);
+        clear_extent_buffer_uptodate(io_tree, eb, &cached_state);
 out:
-        unlock_extent(io_tree, eb->start, eb->start + eb->len - 1,
-                      GFP_NOFS);
+        unlock_extent_cached(io_tree, eb->start, eb->start + eb->len - 1,
+                             &cached_state, GFP_NOFS);
         return ret;
 }
 
@@ -331,7 +339,6 @@ static int csum_dirty_buffer(struct btrfs_root *root, struct page *page)
         struct extent_io_tree *tree;
         u64 start = (u64)page->index << PAGE_CACHE_SHIFT;
         u64 found_start;
-        int found_level;
         unsigned long len;
         struct extent_buffer *eb;
         int ret;
@@ -349,6 +356,8 @@ static int csum_dirty_buffer(struct btrfs_root *root, struct page *page)
         ret = btree_read_extent_buffer_pages(root, eb, start + PAGE_CACHE_SIZE,
                                              btrfs_header_generation(eb));
         BUG_ON(ret);
+        WARN_ON(!btrfs_header_flag(eb, BTRFS_HEADER_FLAG_WRITTEN));
+
         found_start = btrfs_header_bytenr(eb);
         if (found_start != start) {
                 WARN_ON(1);
@@ -362,8 +371,6 @@ static int csum_dirty_buffer(struct btrfs_root *root, struct page *page)
                 WARN_ON(1);
                 goto err;
         }
-        found_level = btrfs_header_level(eb);
-
         csum_tree_block(root, eb, 0);
 err:
         free_extent_buffer(eb);
@@ -473,10 +480,13 @@ static void end_workqueue_bio(struct bio *bio, int err)
         end_io_wq->work.func = end_workqueue_fn;
         end_io_wq->work.flags = 0;
 
-        if (bio->bi_rw & (1 << BIO_RW)) {
-                if (end_io_wq->metadata)
+        if (bio->bi_rw & REQ_WRITE) {
+                if (end_io_wq->metadata == 1)
                         btrfs_queue_worker(&fs_info->endio_meta_write_workers,
                                            &end_io_wq->work);
+                else if (end_io_wq->metadata == 2)
+                        btrfs_queue_worker(&fs_info->endio_freespace_worker,
+                                           &end_io_wq->work);
                 else
                         btrfs_queue_worker(&fs_info->endio_write_workers,
                                            &end_io_wq->work);
@@ -490,6 +500,13 @@ static void end_workqueue_bio(struct bio *bio, int err)
         }
 }
 
+/*
+ * For the metadata arg you want
+ *
+ * 0 - if data
+ * 1 - if normal metadta
+ * 2 - if writing to the free space cache area
+ */
 int btrfs_bio_wq_end_io(struct btrfs_fs_info *info, struct bio *bio,
                         int metadata)
 {
@@ -526,13 +543,12 @@ int btrfs_congested_async(struct btrfs_fs_info *info, int iodone)
 
 static void run_one_async_start(struct btrfs_work *work)
 {
-        struct btrfs_fs_info *fs_info;
         struct async_submit_bio *async;
 
         async = container_of(work, struct  async_submit_bio, work);
-        fs_info = BTRFS_I(async->inode)->root->fs_info;
         async->submit_bio_start(async->inode, async->rw, async->bio,
-                               async->mirror_num, async->bio_flags);
+                               async->mirror_num, async->bio_flags,
+                               async->bio_offset);
 }
 
 static void run_one_async_done(struct btrfs_work *work)
@@ -554,7 +570,8 @@ static void run_one_async_done(struct btrfs_work *work)
                 wake_up(&fs_info->async_submit_wait);
 
         async->submit_bio_done(async->inode, async->rw, async->bio,
-                               async->mirror_num, async->bio_flags);
+                               async->mirror_num, async->bio_flags,
+                               async->bio_offset);
 }
 
 static void run_one_async_free(struct btrfs_work *work)
@@ -568,6 +585,7 @@ static void run_one_async_free(struct btrfs_work *work)
 int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode,
                         int rw, struct bio *bio, int mirror_num,
                         unsigned long bio_flags,
+                        u64 bio_offset,
                         extent_submit_bio_hook_t *submit_bio_start,
                         extent_submit_bio_hook_t *submit_bio_done)
 {
@@ -590,10 +608,11 @@ int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode,
 
         async->work.flags = 0;
         async->bio_flags = bio_flags;
+        async->bio_offset = bio_offset;
 
         atomic_inc(&fs_info->nr_async_submits);
 
-        if (rw & (1 << BIO_RW_SYNCIO))
+        if (rw & REQ_SYNC)
                 btrfs_set_work_high_prio(&async->work);
 
         btrfs_queue_worker(&fs_info->workers, &async->work);
@@ -625,7 +644,8 @@ static int btree_csum_one_bio(struct bio *bio)
 
 static int __btree_submit_bio_start(struct inode *inode, int rw,
                                     struct bio *bio, int mirror_num,
-                                    unsigned long bio_flags)
+                                    unsigned long bio_flags,
+                                    u64 bio_offset)
 {
         /*
          * when we're called for a write, we're already in the async
@@ -636,7 +656,8 @@ static int __btree_submit_bio_start(struct inode *inode, int rw,
 }
 
 static int __btree_submit_bio_done(struct inode *inode, int rw, struct bio *bio,
-                                 int mirror_num, unsigned long bio_flags)
+                                 int mirror_num, unsigned long bio_flags,
+                                 u64 bio_offset)
 {
         /*
          * when we're called for a write, we're already in the async
@@ -646,7 +667,8 @@ static int __btree_submit_bio_done(struct inode *inode, int rw, struct bio *bio,
 }
 
 static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio,
-                                 int mirror_num, unsigned long bio_flags)
+                                 int mirror_num, unsigned long bio_flags,
+                                 u64 bio_offset)
 {
         int ret;
 
@@ -654,7 +676,7 @@ static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio,
                                           bio, 1);
         BUG_ON(ret);
 
-        if (!(rw & (1 << BIO_RW))) {
+        if (!(rw & REQ_WRITE)) {
                 /*
                  * called for a read, do the setup so that checksum validation
                  * can happen in the async kernel threads
@@ -669,10 +691,32 @@ static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio,
          */
         return btrfs_wq_submit_bio(BTRFS_I(inode)->root->fs_info,
                                    inode, rw, bio, mirror_num, 0,
+                                   bio_offset,
                                    __btree_submit_bio_start,
                                    __btree_submit_bio_done);
 }
 
+#ifdef CONFIG_MIGRATION
+static int btree_migratepage(struct address_space *mapping,
+                        struct page *newpage, struct page *page)
+{
+        /*
+         * we can't safely write a btree page from here,
+         * we haven't done the locking hook
+         */
+        if (PageDirty(page))
+                return -EAGAIN;
+        /*
+         * Buffers may be managed in a filesystem specific way.
+         * We must have no buffers or drop them.
+         */
+        if (page_has_private(page) &&
+            !try_to_release_page(page, GFP_KERNEL))
+                return -EAGAIN;
+        return migrate_page(mapping, newpage, page);
+}
+#endif
+
 static int btree_writepage(struct page *page, struct writeback_control *wbc)
 {
         struct extent_io_tree *tree;
@@ -687,8 +731,7 @@ static int btree_writepage(struct page *page, struct writeback_control *wbc)
         }
 
         redirty_page_for_writepage(wbc, page);
-        eb = btrfs_find_tree_block(root, page_offset(page),
-                                      PAGE_CACHE_SIZE);
+        eb = btrfs_find_tree_block(root, page_offset(page), PAGE_CACHE_SIZE);
         WARN_ON(!eb);
 
         was_dirty = test_and_set_bit(EXTENT_BUFFER_DIRTY, &eb->bflags);
@@ -772,13 +815,16 @@ static void btree_invalidatepage(struct page *page, unsigned long offset)
         }
 }
 
-static struct address_space_operations btree_aops = {
+static const struct address_space_operations btree_aops = {
         .readpage       = btree_readpage,
         .writepage      = btree_writepage,
         .writepages     = btree_writepages,
         .releasepage    = btree_releasepage,
         .invalidatepage = btree_invalidatepage,
         .sync_page      = block_sync_page,
+#ifdef CONFIG_MIGRATION
+        .migratepage    = btree_migratepage,
+#endif
 };
 
 int readahead_tree_block(struct btrfs_root *root, u64 bytenr, u32 blocksize,
@@ -821,26 +867,22 @@ struct extent_buffer *btrfs_find_create_tree_block(struct btrfs_root *root,
 
 int btrfs_write_tree_block(struct extent_buffer *buf)
 {
-        return btrfs_fdatawrite_range(buf->first_page->mapping, buf->start,
-                                      buf->start + buf->len - 1, WB_SYNC_ALL);
+        return filemap_fdatawrite_range(buf->first_page->mapping, buf->start,
+                                        buf->start + buf->len - 1);
 }
 
 int btrfs_wait_tree_block_writeback(struct extent_buffer *buf)
 {
-        return btrfs_wait_on_page_writeback_range(buf->first_page->mapping,
-                                  buf->start, buf->start + buf->len - 1);
+        return filemap_fdatawait_range(buf->first_page->mapping,
+                                       buf->start, buf->start + buf->len - 1);
 }
 
 struct extent_buffer *read_tree_block(struct btrfs_root *root, u64 bytenr,
                                       u32 blocksize, u64 parent_transid)
 {
         struct extent_buffer *buf = NULL;
-        struct inode *btree_inode = root->fs_info->btree_inode;
-        struct extent_io_tree *io_tree;
         int ret;
 
-        io_tree = &BTRFS_I(btree_inode)->io_tree;
-
         buf = btrfs_find_create_tree_block(root, bytenr, blocksize);
         if (!buf)
                 return NULL;
@@ -891,22 +933,27 @@ static int __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
         root->stripesize = stripesize;
         root->ref_cows = 0;
         root->track_dirty = 0;
+        root->in_radix = 0;
+        root->orphan_item_inserted = 0;
+        root->orphan_cleanup_state = 0;
 
         root->fs_info = fs_info;
         root->objectid = objectid;
         root->last_trans = 0;
-        root->highest_inode = 0;
-        root->last_inode_alloc = 0;
+        root->highest_objectid = 0;
         root->name = NULL;
         root->in_sysfs = 0;
-        root->inode_tree.rb_node = NULL;
+        root->inode_tree = RB_ROOT;
+        root->block_rsv = NULL;
+        root->orphan_block_rsv = NULL;
 
         INIT_LIST_HEAD(&root->dirty_list);
         INIT_LIST_HEAD(&root->orphan_list);
         INIT_LIST_HEAD(&root->root_list);
         spin_lock_init(&root->node_lock);
-        spin_lock_init(&root->list_lock);
+        spin_lock_init(&root->orphan_lock);
         spin_lock_init(&root->inode_lock);
+        spin_lock_init(&root->accounting_lock);
         mutex_init(&root->objectid_mutex);
         mutex_init(&root->log_mutex);
         init_waitqueue_head(&root->log_writer_wait);
@@ -917,6 +964,7 @@ static int __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
         atomic_set(&root->log_writers, 0);
         root->log_batch = 0;
         root->log_transid = 0;
+        root->last_log_commit = 0;
         extent_io_tree_init(&root->dirty_log_pages,
                              fs_info->btree_inode->i_mapping, GFP_NOFS);
 
@@ -927,7 +975,6 @@ static int __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
         root->defrag_trans_start = fs_info->generation;
         init_completion(&root->kobj_unregister);
         root->defrag_running = 0;
-        root->defrag_level = 0;
         root->root_key.objectid = objectid;
         root->anon_super.s_root = NULL;
         root->anon_super.s_dev = 0;
@@ -952,50 +999,19 @@ static int find_and_setup_root(struct btrfs_root *tree_root,
                      root, fs_info, objectid);
         ret = btrfs_find_last_root(tree_root, objectid,
                                    &root->root_item, &root->root_key);
+        if (ret > 0)
+                return -ENOENT;
         BUG_ON(ret);
 
         generation = btrfs_root_generation(&root->root_item);
         blocksize = btrfs_level_size(root, btrfs_root_level(&root->root_item));
         root->node = read_tree_block(root, btrfs_root_bytenr(&root->root_item),
                                      blocksize, generation);
-        root->commit_root = btrfs_root_node(root);
-        BUG_ON(!root->node);
-        return 0;
-}
-
-int btrfs_free_log_root_tree(struct btrfs_trans_handle *trans,
-                             struct btrfs_fs_info *fs_info)
-{
-        struct extent_buffer *eb;
-        struct btrfs_root *log_root_tree = fs_info->log_root_tree;
-        u64 start = 0;
-        u64 end = 0;
-        int ret;
-
-        if (!log_root_tree)
-                return 0;
-
-        while (1) {
-                ret = find_first_extent_bit(&log_root_tree->dirty_log_pages,
-                                    0, &start, &end, EXTENT_DIRTY);
-                if (ret)
-                        break;
-
-                clear_extent_dirty(&log_root_tree->dirty_log_pages,
-                                   start, end, GFP_NOFS);
+        if (!root->node || !btrfs_buffer_uptodate(root->node, generation)) {
+                free_extent_buffer(root->node);
+                return -EIO;
         }
-        eb = fs_info->log_root_tree->node;
-
-        WARN_ON(btrfs_header_level(eb) != 0);
-        WARN_ON(btrfs_header_nritems(eb) != 0);
-
-        ret = btrfs_free_reserved_extent(fs_info->tree_root,
-                                eb->start, eb->len);
-        BUG_ON(ret);
-
-        free_extent_buffer(eb);
-        kfree(fs_info->log_root_tree);
-        fs_info->log_root_tree = NULL;
+        root->commit_root = btrfs_root_node(root);
         return 0;
 }
 
@@ -1085,6 +1101,7 @@ int btrfs_add_log_tree(struct btrfs_trans_handle *trans,
         WARN_ON(root->log_root);
         root->log_root = log_root;
         root->log_transid = 0;
+        root->last_log_commit = 0;
         return 0;
 }
 
@@ -1095,7 +1112,6 @@ struct btrfs_root *btrfs_read_fs_root_no_radix(struct btrfs_root *tree_root,
         struct btrfs_fs_info *fs_info = tree_root->fs_info;
         struct btrfs_path *path;
         struct extent_buffer *l;
-        u64 highest_inode;
         u64 generation;
         u32 blocksize;
         int ret = 0;
@@ -1110,7 +1126,7 @@ struct btrfs_root *btrfs_read_fs_root_no_radix(struct btrfs_root *tree_root,
                         kfree(root);
                         return ERR_PTR(ret);
                 }
-                goto insert;
+                goto out;
         }
 
         __setup_root(tree_root->nodesize, tree_root->leafsize,
@@ -1120,39 +1136,30 @@ struct btrfs_root *btrfs_read_fs_root_no_radix(struct btrfs_root *tree_root,
         path = btrfs_alloc_path();
         BUG_ON(!path);
         ret = btrfs_search_slot(NULL, tree_root, location, path, 0, 0);
-        if (ret != 0) {
-                if (ret > 0)
-                        ret = -ENOENT;
-                goto out;
+        if (ret == 0) {
+                l = path->nodes[0];
+                read_extent_buffer(l, &root->root_item,
+                                btrfs_item_ptr_offset(l, path->slots[0]),
+                                sizeof(root->root_item));
+                memcpy(&root->root_key, location, sizeof(*location));
         }
-        l = path->nodes[0];
-        read_extent_buffer(l, &root->root_item,
-               btrfs_item_ptr_offset(l, path->slots[0]),
-               sizeof(root->root_item));
-        memcpy(&root->root_key, location, sizeof(*location));
-        ret = 0;
-out:
-        btrfs_release_path(root, path);
         btrfs_free_path(path);
         if (ret) {
-                kfree(root);
+                if (ret > 0)
+                        ret = -ENOENT;
                 return ERR_PTR(ret);
         }
+
         generation = btrfs_root_generation(&root->root_item);
         blocksize = btrfs_level_size(root, btrfs_root_level(&root->root_item));
         root->node = read_tree_block(root, btrfs_root_bytenr(&root->root_item),
                                      blocksize, generation);
         root->commit_root = btrfs_root_node(root);
         BUG_ON(!root->node);
-insert:
-        if (location->objectid != BTRFS_TREE_LOG_OBJECTID) {
+out:
+        if (location->objectid != BTRFS_TREE_LOG_OBJECTID)
                 root->ref_cows = 1;
-                ret = btrfs_find_highest_inode(root, &highest_inode);
-                if (ret == 0) {
-                        root->highest_inode = highest_inode;
-                        root->last_inode_alloc = highest_inode;
-                }
-        }
+
         return root;
 }
 
@@ -1187,9 +1194,11 @@ struct btrfs_root *btrfs_read_fs_root_no_name(struct btrfs_fs_info *fs_info,
                 return fs_info->dev_root;
         if (location->objectid == BTRFS_CSUM_TREE_OBJECTID)
                 return fs_info->csum_root;
-
+again:
+        spin_lock(&fs_info->fs_roots_radix_lock);
         root = radix_tree_lookup(&fs_info->fs_roots_radix,
                                  (unsigned long)location->objectid);
+        spin_unlock(&fs_info->fs_roots_radix_lock);
         if (root)
                 return root;
 
@@ -1199,27 +1208,53 @@ struct btrfs_root *btrfs_read_fs_root_no_name(struct btrfs_fs_info *fs_info,
 
         set_anon_super(&root->anon_super, NULL);
 
+        if (btrfs_root_refs(&root->root_item) == 0) {
+                ret = -ENOENT;
+                goto fail;
+        }
+
+        ret = btrfs_find_orphan_item(fs_info->tree_root, location->objectid);
+        if (ret < 0)
+                goto fail;
+        if (ret == 0)
+                root->orphan_item_inserted = 1;
+
+        ret = radix_tree_preload(GFP_NOFS & ~__GFP_HIGHMEM);
+        if (ret)
+                goto fail;
+
+        spin_lock(&fs_info->fs_roots_radix_lock);
         ret = radix_tree_insert(&fs_info->fs_roots_radix,
                                 (unsigned long)root->root_key.objectid,
                                 root);
+        if (ret == 0)
+                root->in_radix = 1;
+
+        spin_unlock(&fs_info->fs_roots_radix_lock);
+        radix_tree_preload_end();
         if (ret) {
-                free_extent_buffer(root->node);
-                kfree(root);
-                return ERR_PTR(ret);
-        }
-        if (!(fs_info->sb->s_flags & MS_RDONLY)) {
-                ret = btrfs_find_dead_roots(fs_info->tree_root,
-                                            root->root_key.objectid);
-                BUG_ON(ret);
-                btrfs_orphan_cleanup(root);
+                if (ret == -EEXIST) {
+                        free_fs_root(root);
+                        goto again;
+                }
+                goto fail;
         }
+
+        ret = btrfs_find_dead_roots(fs_info->tree_root,
+                                    root->root_key.objectid);
+        WARN_ON(ret);
         return root;
+fail:
+        free_fs_root(root);
+        return ERR_PTR(ret);
 }
 
 struct btrfs_root *btrfs_read_fs_root(struct btrfs_fs_info *fs_info,
                                       struct btrfs_key *location,
                                       const char *name, int namelen)
 {
+        return btrfs_read_fs_root_no_name(fs_info, location);
+#if 0
         struct btrfs_root *root;
         int ret;
 
@@ -1236,7 +1271,7 @@ struct btrfs_root *btrfs_read_fs_root(struct btrfs_fs_info *fs_info,
                 kfree(root);
                 return ERR_PTR(ret);
         }
-#if 0
+
         ret = btrfs_sysfs_add_root(root);
         if (ret) {
                 free_extent_buffer(root->node);
@@ -1244,9 +1279,9 @@ struct btrfs_root *btrfs_read_fs_root(struct btrfs_fs_info *fs_info,
                 kfree(root);
                 return ERR_PTR(ret);
         }
-#endif
         root->in_sysfs = 1;
         return root;
+#endif
 }
 
 static int btrfs_congested_fn(void *congested_data, int bdi_bits)
@@ -1325,9 +1360,9 @@ static void btrfs_unplug_io_fn(struct backing_dev_info *bdi, struct page *page)
         offset = page_offset(page);
 
         em_tree = &BTRFS_I(inode)->extent_tree;
-        spin_lock(&em_tree->lock);
+        read_lock(&em_tree->lock);
         em = lookup_extent_mapping(em_tree, offset, PAGE_CACHE_SIZE);
-        spin_unlock(&em_tree->lock);
+        read_unlock(&em_tree->lock);
         if (!em) {
                 __unplug_io_fn(bdi, page);
                 return;
@@ -1353,12 +1388,7 @@ static int setup_bdi(struct btrfs_fs_info *info, struct backing_dev_info *bdi)
         int err;
 
         bdi->capabilities = BDI_CAP_MAP_COPY;
-        err = bdi_init(bdi);
-        if (err)
-                return err;
-
-        err = bdi_register(bdi, NULL, "btrfs-%d",
-                                atomic_inc_return(&btrfs_bdi_num));
+        err = bdi_setup_and_register(bdi, "btrfs", BDI_CAP_MAP_COPY);
         if (err)
                 return err;
 
@@ -1377,7 +1407,6 @@ static int bio_ready_for_csum(struct bio *bio)
         u64 start = 0;
         struct page *page;
         struct extent_io_tree *io_tree = NULL;
-        struct btrfs_fs_info *info = NULL;
         struct bio_vec *bvec;
         int i;
         int ret;
@@ -1396,7 +1425,6 @@ static int bio_ready_for_csum(struct bio *bio)
                 buf_len = page->private >> 2;
                 start = page_offset(page) + bvec->bv_offset;
                 io_tree = &BTRFS_I(page->mapping->host)->io_tree;
-                info = BTRFS_I(page->mapping->host)->root->fs_info;
         }
         /* are we fully contained in this bio? */
         if (buf_len <= length)
@@ -1427,7 +1455,7 @@ static void end_workqueue_fn(struct btrfs_work *work)
          * ram and up to date before trying to verify things.  For
          * blocksize <= pagesize, it is basically a noop
          */
-        if (!(bio->bi_rw & (1 << BIO_RW)) && end_io_wq->metadata &&
+        if (!(bio->bi_rw & REQ_WRITE) && end_io_wq->metadata &&
             !bio_ready_for_csum(bio)) {
                 btrfs_queue_worker(&fs_info->endio_meta_workers,
                                    &end_io_wq->work);
@@ -1445,23 +1473,21 @@ static int cleaner_kthread(void *arg)
         struct btrfs_root *root = arg;
 
         do {
-                smp_mb();
-                if (root->fs_info->closing)
-                        break;
-
                 vfs_check_frozen(root->fs_info->sb, SB_FREEZE_WRITE);
-                mutex_lock(&root->fs_info->cleaner_mutex);
-                btrfs_clean_old_snapshots(root);
-                mutex_unlock(&root->fs_info->cleaner_mutex);
+
+                if (!(root->fs_info->sb->s_flags & MS_RDONLY) &&
+                    mutex_trylock(&root->fs_info->cleaner_mutex)) {
+                        btrfs_run_delayed_iputs(root);
+                        btrfs_clean_old_snapshots(root);
+                        mutex_unlock(&root->fs_info->cleaner_mutex);
+                }
 
                 if (freezing(current)) {
                         refrigerator();
                 } else {
-                        smp_mb();
-                        if (root->fs_info->closing)
-                                break;
                         set_current_state(TASK_INTERRUPTIBLE);
-                        schedule();
+                        if (!kthread_should_stop())
+                                schedule();
                         __set_current_state(TASK_RUNNING);
                 }
         } while (!kthread_should_stop());
@@ -1473,36 +1499,40 @@ static int transaction_kthread(void *arg)
         struct btrfs_root *root = arg;
         struct btrfs_trans_handle *trans;
         struct btrfs_transaction *cur;
+        u64 transid;
         unsigned long now;
         unsigned long delay;
         int ret;
 
         do {
-                smp_mb();
-                if (root->fs_info->closing)
-                        break;
-
                 delay = HZ * 30;
                 vfs_check_frozen(root->fs_info->sb, SB_FREEZE_WRITE);
                 mutex_lock(&root->fs_info->transaction_kthread_mutex);
 
-                mutex_lock(&root->fs_info->trans_mutex);
+                spin_lock(&root->fs_info->new_trans_lock);
                 cur = root->fs_info->running_transaction;
                 if (!cur) {
-                        mutex_unlock(&root->fs_info->trans_mutex);
+                        spin_unlock(&root->fs_info->new_trans_lock);
                         goto sleep;
                 }
 
                 now = get_seconds();
-                if (now < cur->start_time || now - cur->start_time < 30) {
-                        mutex_unlock(&root->fs_info->trans_mutex);
+                if (!cur->blocked &&
+                    (now < cur->start_time || now - cur->start_time < 30)) {
+                        spin_unlock(&root->fs_info->new_trans_lock);
                         delay = HZ * 5;
                         goto sleep;
                 }
-                mutex_unlock(&root->fs_info->trans_mutex);
-                trans = btrfs_start_transaction(root, 1);
-                ret = btrfs_commit_transaction(trans, root);
+                transid = cur->transid;
+                spin_unlock(&root->fs_info->new_trans_lock);
 
+                trans = btrfs_join_transaction(root, 1);
+                if (transid == trans->transid) {
+                        ret = btrfs_commit_transaction(trans, root);
+                        BUG_ON(ret);
+                } else {
+                        btrfs_end_transaction(trans, root);
+                }
 sleep:
                 wake_up_process(root->fs_info->cleaner_kthread);
                 mutex_unlock(&root->fs_info->transaction_kthread_mutex);
@@ -1510,10 +1540,10 @@ sleep:
                 if (freezing(current)) {
                         refrigerator();
                 } else {
-                        if (root->fs_info->closing)
-                                break;
                         set_current_state(TASK_INTERRUPTIBLE);
-                        schedule_timeout(delay);
+                        if (!kthread_should_stop() &&
+                            !btrfs_transaction_blocked(root->fs_info))
+                                schedule_timeout(delay);
                         __set_current_state(TASK_RUNNING);
                 }
         } while (!kthread_should_stop());
@@ -1537,10 +1567,8 @@ struct btrfs_root *open_ctree(struct super_block *sb,
                                                  GFP_NOFS);
         struct btrfs_root *csum_root = kzalloc(sizeof(struct btrfs_root),
                                                  GFP_NOFS);
-        struct btrfs_root *tree_root = kzalloc(sizeof(struct btrfs_root),
-                                               GFP_NOFS);
-        struct btrfs_fs_info *fs_info = kzalloc(sizeof(*fs_info),
-                                                GFP_NOFS);
+        struct btrfs_root *tree_root = btrfs_sb(sb);
+        struct btrfs_fs_info *fs_info = tree_root->fs_info;
         struct btrfs_root *chunk_root = kzalloc(sizeof(struct btrfs_root),
                                                 GFP_NOFS);
         struct btrfs_root *dev_root = kzalloc(sizeof(struct btrfs_root),
@@ -1557,15 +1585,38 @@ struct btrfs_root *open_ctree(struct super_block *sb,
                 err = -ENOMEM;
                 goto fail;
         }
-        INIT_RADIX_TREE(&fs_info->fs_roots_radix, GFP_NOFS);
+
+        ret = init_srcu_struct(&fs_info->subvol_srcu);
+        if (ret) {
+                err = ret;
+                goto fail;
+        }
+
+        ret = setup_bdi(fs_info, &fs_info->bdi);
+        if (ret) {
+                err = ret;
+                goto fail_srcu;
+        }
+
+        fs_info->btree_inode = new_inode(sb);
+        if (!fs_info->btree_inode) {
+                err = -ENOMEM;
+                goto fail_bdi;
+        }
+
+        INIT_RADIX_TREE(&fs_info->fs_roots_radix, GFP_ATOMIC);
         INIT_LIST_HEAD(&fs_info->trans_list);
         INIT_LIST_HEAD(&fs_info->dead_roots);
+        INIT_LIST_HEAD(&fs_info->delayed_iputs);
         INIT_LIST_HEAD(&fs_info->hashers);
         INIT_LIST_HEAD(&fs_info->delalloc_inodes);
         INIT_LIST_HEAD(&fs_info->ordered_operations);
+        INIT_LIST_HEAD(&fs_info->caching_block_groups);
         spin_lock_init(&fs_info->delalloc_lock);
         spin_lock_init(&fs_info->new_trans_lock);
         spin_lock_init(&fs_info->ref_cache_lock);
+        spin_lock_init(&fs_info->fs_roots_radix_lock);
+        spin_lock_init(&fs_info->delayed_iput_lock);
 
         init_completion(&fs_info->kobj_unregister);
         fs_info->tree_root = tree_root;
@@ -1577,19 +1628,20 @@ struct btrfs_root *open_ctree(struct super_block *sb,
         INIT_LIST_HEAD(&fs_info->dirty_cowonly_roots);
         INIT_LIST_HEAD(&fs_info->space_info);
         btrfs_mapping_init(&fs_info->mapping_tree);
+        btrfs_init_block_rsv(&fs_info->global_block_rsv);
+        btrfs_init_block_rsv(&fs_info->delalloc_block_rsv);
+        btrfs_init_block_rsv(&fs_info->trans_block_rsv);
+        btrfs_init_block_rsv(&fs_info->chunk_block_rsv);
+        btrfs_init_block_rsv(&fs_info->empty_block_rsv);
+        INIT_LIST_HEAD(&fs_info->durable_block_rsv_list);
+        mutex_init(&fs_info->durable_block_rsv_mutex);
         atomic_set(&fs_info->nr_async_submits, 0);
         atomic_set(&fs_info->async_delalloc_pages, 0);
         atomic_set(&fs_info->async_submit_draining, 0);
         atomic_set(&fs_info->nr_async_bios, 0);
         fs_info->sb = sb;
-        fs_info->max_extent = (u64)-1;
         fs_info->max_inline = 8192 * 1024;
-        if (setup_bdi(fs_info, &fs_info->bdi))
-                goto fail_bdi;
-        fs_info->btree_inode = new_inode(sb);
-        fs_info->btree_inode->i_ino = 1;
-        fs_info->btree_inode->i_nlink = 1;
-        fs_info->metadata_ratio = 8;
+        fs_info->metadata_ratio = 0;
 
         fs_info->thread_pool_size = min_t(unsigned long,
                                           num_online_cpus() + 2, 8);
@@ -1599,7 +1651,10 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 
         sb->s_blocksize = 4096;
         sb->s_blocksize_bits = blksize_bits(4096);
+        sb->s_bdi = &fs_info->bdi;
 
+        fs_info->btree_inode->i_ino = BTRFS_BTREE_INODE_OBJECTID;
+        fs_info->btree_inode->i_nlink = 1;
         /*
          * we set the i_size on the btree inode to the max possible int.
          * the real end of the address space is determined by all of
@@ -1618,39 +1673,45 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 
         BTRFS_I(fs_info->btree_inode)->io_tree.ops = &btree_extent_io_ops;
 
+        BTRFS_I(fs_info->btree_inode)->root = tree_root;
+        memset(&BTRFS_I(fs_info->btree_inode)->location, 0,
+               sizeof(struct btrfs_key));
+        BTRFS_I(fs_info->btree_inode)->dummy_inode = 1;
+        insert_inode_hash(fs_info->btree_inode);
+
         spin_lock_init(&fs_info->block_group_cache_lock);
-        fs_info->block_group_cache_tree.rb_node = NULL;
+        fs_info->block_group_cache_tree = RB_ROOT;
 
-        extent_io_tree_init(&fs_info->pinned_extents,
+        extent_io_tree_init(&fs_info->freed_extents[0],
+                             fs_info->btree_inode->i_mapping, GFP_NOFS);
+        extent_io_tree_init(&fs_info->freed_extents[1],
                              fs_info->btree_inode->i_mapping, GFP_NOFS);
+        fs_info->pinned_extents = &fs_info->freed_extents[0];
         fs_info->do_barriers = 1;
 
-        BTRFS_I(fs_info->btree_inode)->root = tree_root;
-        memset(&BTRFS_I(fs_info->btree_inode)->location, 0,
-               sizeof(struct btrfs_key));
-        insert_inode_hash(fs_info->btree_inode);
 
         mutex_init(&fs_info->trans_mutex);
         mutex_init(&fs_info->ordered_operations_mutex);
         mutex_init(&fs_info->tree_log_mutex);
-        mutex_init(&fs_info->drop_mutex);
         mutex_init(&fs_info->chunk_mutex);
         mutex_init(&fs_info->transaction_kthread_mutex);
         mutex_init(&fs_info->cleaner_mutex);
         mutex_init(&fs_info->volume_mutex);
-        mutex_init(&fs_info->tree_reloc_mutex);
+        init_rwsem(&fs_info->extent_commit_sem);
+        init_rwsem(&fs_info->cleanup_work_sem);
+        init_rwsem(&fs_info->subvol_sem);
 
         btrfs_init_free_cluster(&fs_info->meta_alloc_cluster);
         btrfs_init_free_cluster(&fs_info->data_alloc_cluster);
 
         init_waitqueue_head(&fs_info->transaction_throttle);
         init_waitqueue_head(&fs_info->transaction_wait);
+        init_waitqueue_head(&fs_info->transaction_blocked_wait);
         init_waitqueue_head(&fs_info->async_submit_wait);
 
         __setup_root(4096, 4096, 4096, 4096, tree_root,
                      fs_info, BTRFS_ROOT_TREE_OBJECTID);
 
-
         bh = btrfs_read_dev_super(fs_devices->latest_bdev);
         if (!bh)
                 goto fail_iput;
@@ -1698,20 +1759,21 @@ struct btrfs_root *open_ctree(struct super_block *sb,
                 goto fail_iput;
         }
 
-        /*
-         * we need to start all the end_io workers up front because the
-         * queue work function gets called at interrupt time, and so it
-         * cannot dynamically grow.
-         */
+        btrfs_init_workers(&fs_info->generic_worker,
+                           "genwork", 1, NULL);
+
         btrfs_init_workers(&fs_info->workers, "worker",
-                           fs_info->thread_pool_size);
+                           fs_info->thread_pool_size,
+                           &fs_info->generic_worker);
 
         btrfs_init_workers(&fs_info->delalloc_workers, "delalloc",
-                           fs_info->thread_pool_size);
+                           fs_info->thread_pool_size,
+                           &fs_info->generic_worker);
 
         btrfs_init_workers(&fs_info->submit_workers, "submit",
                            min_t(u64, fs_devices->num_devices,
-                           fs_info->thread_pool_size));
+                           fs_info->thread_pool_size),
+                           &fs_info->generic_worker);
 
         /* a higher idle thresh on the submit workers makes it much more
          * likely that bios will be send down in a sane order to the
@@ -1725,15 +1787,22 @@ struct btrfs_root *open_ctree(struct super_block *sb,
         fs_info->delalloc_workers.idle_thresh = 2;
         fs_info->delalloc_workers.ordered = 1;
 
-        btrfs_init_workers(&fs_info->fixup_workers, "fixup", 1);
+        btrfs_init_workers(&fs_info->fixup_workers, "fixup", 1,
+                           &fs_info->generic_worker);
         btrfs_init_workers(&fs_info->endio_workers, "endio",
-                           fs_info->thread_pool_size);
+                           fs_info->thread_pool_size,
+                           &fs_info->generic_worker);
         btrfs_init_workers(&fs_info->endio_meta_workers, "endio-meta",
-                           fs_info->thread_pool_size);
+                           fs_info->thread_pool_size,
+                           &fs_info->generic_worker);
         btrfs_init_workers(&fs_info->endio_meta_write_workers,
-                           "endio-meta-write", fs_info->thread_pool_size);
+                           "endio-meta-write", fs_info->thread_pool_size,
+                           &fs_info->generic_worker);
         btrfs_init_workers(&fs_info->endio_write_workers, "endio-write",
-                           fs_info->thread_pool_size);
+                           fs_info->thread_pool_size,
+                           &fs_info->generic_worker);
+        btrfs_init_workers(&fs_info->endio_freespace_worker, "freespace-write",
+                           1, &fs_info->generic_worker);
 
         /*
          * endios are largely parallel and should have a very
@@ -1742,20 +1811,19 @@ struct btrfs_root *open_ctree(struct super_block *sb,
         fs_info->endio_workers.idle_thresh = 4;
         fs_info->endio_meta_workers.idle_thresh = 4;
 
-        fs_info->endio_write_workers.idle_thresh = 64;
-        fs_info->endio_meta_write_workers.idle_thresh = 64;
+        fs_info->endio_write_workers.idle_thresh = 2;
+        fs_info->endio_meta_write_workers.idle_thresh = 2;
 
         btrfs_start_workers(&fs_info->workers, 1);
+        btrfs_start_workers(&fs_info->generic_worker, 1);
         btrfs_start_workers(&fs_info->submit_workers, 1);
         btrfs_start_workers(&fs_info->delalloc_workers, 1);
         btrfs_start_workers(&fs_info->fixup_workers, 1);
-        btrfs_start_workers(&fs_info->endio_workers, fs_info->thread_pool_size);
-        btrfs_start_workers(&fs_info->endio_meta_workers,
-                            fs_info->thread_pool_size);
-        btrfs_start_workers(&fs_info->endio_meta_write_workers,
-                            fs_info->thread_pool_size);
-        btrfs_start_workers(&fs_info->endio_write_workers,
-                            fs_info->thread_pool_size);
+        btrfs_start_workers(&fs_info->endio_workers, 1);
+        btrfs_start_workers(&fs_info->endio_meta_workers, 1);
+        btrfs_start_workers(&fs_info->endio_meta_write_workers, 1);
+        btrfs_start_workers(&fs_info->endio_write_workers, 1);
+        btrfs_start_workers(&fs_info->endio_freespace_worker, 1);
 
         fs_info->bdi.ra_pages *= btrfs_super_num_devices(disk_super);
         fs_info->bdi.ra_pages = max(fs_info->bdi.ra_pages,
@@ -1799,6 +1867,11 @@ struct btrfs_root *open_ctree(struct super_block *sb,
                                            btrfs_super_chunk_root(disk_super),
                                            blocksize, generation);
         BUG_ON(!chunk_root->node);
+        if (!test_bit(EXTENT_BUFFER_UPTODATE, &chunk_root->node->bflags)) {
+                printk(KERN_WARNING "btrfs: failed to read chunk root on %s\n",
+                       sb->s_id);
+                goto fail_chunk_root;
+        }
         btrfs_set_root_node(&chunk_root->root_item, chunk_root->node);
         chunk_root->commit_root = btrfs_root_node(chunk_root);
 
@@ -1826,6 +1899,11 @@ struct btrfs_root *open_ctree(struct super_block *sb,
                                           blocksize, generation);
         if (!tree_root->node)
                 goto fail_chunk_root;
+        if (!test_bit(EXTENT_BUFFER_UPTODATE, &tree_root->node->bflags)) {
+                printk(KERN_WARNING "btrfs: failed to read tree root on %s\n",
+                       sb->s_id);
+                goto fail_tree_root;
+        }
         btrfs_set_root_node(&tree_root->root_item, tree_root->node);
         tree_root->commit_root = btrfs_root_node(tree_root);
 
@@ -1848,17 +1926,22 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 
         csum_root->track_dirty = 1;
 
-        btrfs_read_block_groups(extent_root);
-
         fs_info->generation = generation;
         fs_info->last_trans_committed = generation;
         fs_info->data_alloc_profile = (u64)-1;
         fs_info->metadata_alloc_profile = (u64)-1;
         fs_info->system_alloc_profile = fs_info->metadata_alloc_profile;
+
+        ret = btrfs_read_block_groups(extent_root);
+        if (ret) {
+                printk(KERN_ERR "Failed to read block groups: %d\n", ret);
+                goto fail_block_groups;
+        }
+
         fs_info->cleaner_kthread = kthread_run(cleaner_kthread, tree_root,
                                                "btrfs-cleaner");
         if (IS_ERR(fs_info->cleaner_kthread))
-                goto fail_csum_root;
+                goto fail_block_groups;
 
         fs_info->transaction_kthread = kthread_run(transaction_kthread,
                                                    tree_root,
@@ -1887,8 +1970,11 @@ struct btrfs_root *open_ctree(struct super_block *sb,
                      btrfs_level_size(tree_root,
                                       btrfs_super_log_root_level(disk_super));
 
-                log_tree_root = kzalloc(sizeof(struct btrfs_root),
-                                                      GFP_NOFS);
+                log_tree_root = kzalloc(sizeof(struct btrfs_root), GFP_NOFS);
+                if (!log_tree_root) {
+                        err = -ENOMEM;
+                        goto fail_trans_kthread;
+                }
 
                 __setup_root(nodesize, leafsize, sectorsize, stripesize,
                              log_tree_root, fs_info, BTRFS_TREE_LOG_OBJECTID);
@@ -1905,9 +1991,20 @@ struct btrfs_root *open_ctree(struct super_block *sb,
                 }
         }
 
+        ret = btrfs_find_orphan_roots(tree_root);
+        BUG_ON(ret);
+
         if (!(sb->s_flags & MS_RDONLY)) {
-                ret = btrfs_recover_relocation(tree_root);
+                ret = btrfs_cleanup_fs_roots(fs_info);
                 BUG_ON(ret);
+
+                ret = btrfs_recover_relocation(tree_root);
+                if (ret < 0) {
+                        printk(KERN_WARNING
+                               "btrfs: failed to recover relocation\n");
+                        err = -EINVAL;
+                        goto fail_trans_kthread;
+                }
         }
 
         location.objectid = BTRFS_FS_TREE_OBJECTID;
@@ -1917,6 +2014,17 @@ struct btrfs_root *open_ctree(struct super_block *sb,
         fs_info->fs_root = btrfs_read_fs_root_no_name(fs_info, &location);
         if (!fs_info->fs_root)
                 goto fail_trans_kthread;
+        if (IS_ERR(fs_info->fs_root)) {
+                err = PTR_ERR(fs_info->fs_root);
+                goto fail_trans_kthread;
+        }
+
+        if (!(sb->s_flags & MS_RDONLY)) {
+                down_read(&fs_info->cleanup_work_sem);
+                btrfs_orphan_cleanup(fs_info->fs_root);
+                btrfs_orphan_cleanup(fs_info->tree_root);
+                up_read(&fs_info->cleanup_work_sem);
+        }
 
         return tree_root;
 
@@ -1932,7 +2040,8 @@ fail_cleaner:
         filemap_write_and_wait(fs_info->btree_inode->i_mapping);
         invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
 
-fail_csum_root:
+fail_block_groups:
+        btrfs_free_block_groups(fs_info);
         free_extent_buffer(csum_root->node);
         free_extent_buffer(csum_root->commit_root);
 fail_dev_root:
@@ -1948,6 +2057,7 @@ fail_chunk_root:
         free_extent_buffer(chunk_root->node);
         free_extent_buffer(chunk_root->commit_root);
 fail_sb_buffer:
+        btrfs_stop_workers(&fs_info->generic_worker);
         btrfs_stop_workers(&fs_info->fixup_workers);
         btrfs_stop_workers(&fs_info->delalloc_workers);
         btrfs_stop_workers(&fs_info->workers);
@@ -1955,6 +2065,7 @@ fail_sb_buffer:
         btrfs_stop_workers(&fs_info->endio_meta_workers);
         btrfs_stop_workers(&fs_info->endio_meta_write_workers);
         btrfs_stop_workers(&fs_info->endio_write_workers);
+        btrfs_stop_workers(&fs_info->endio_freespace_worker);
         btrfs_stop_workers(&fs_info->submit_workers);
 fail_iput:
         invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
@@ -1964,6 +2075,8 @@ fail_iput:
         btrfs_mapping_tree_free(&fs_info->mapping_tree);
 fail_bdi:
         bdi_destroy(&fs_info->bdi);
+fail_srcu:
+        cleanup_srcu_struct(&fs_info->subvol_srcu);
 fail:
         kfree(extent_root);
         kfree(tree_root);
@@ -1981,7 +2094,7 @@ static void btrfs_end_buffer_write_sync(struct buffer_head *bh, int uptodate)
         if (uptodate) {
                 set_buffer_uptodate(bh);
         } else {
-                if (!buffer_eopnotsupp(bh) && printk_ratelimit()) {
+                if (printk_ratelimit()) {
                         printk(KERN_WARNING "lost page write due to "
                                         "I/O error on %s\n",
                                        bdevname(bh->b_bdev, b));
@@ -2118,21 +2231,10 @@ static int write_dev_supers(struct btrfs_device *device,
                         bh->b_end_io = btrfs_end_buffer_write_sync;
                 }
 
-                if (i == last_barrier && do_barriers && device->barriers) {
-                        ret = submit_bh(WRITE_BARRIER, bh);
-                        if (ret == -EOPNOTSUPP) {
-                                printk("btrfs: disabling barriers on dev %s\n",
-                                       device->name);
-                                set_buffer_uptodate(bh);
-                                device->barriers = 0;
-                                /* one reference for submit_bh */
-                                get_bh(bh);
-                                lock_buffer(bh);
-                                ret = submit_bh(WRITE_SYNC, bh);
-                        }
-                } else {
+                if (i == last_barrier && do_barriers)
+                        ret = submit_bh(WRITE_FLUSH_FUA, bh);
+                else
                         ret = submit_bh(WRITE_SYNC, bh);
-                }
 
                 if (ret)
                         errors++;
@@ -2223,20 +2325,29 @@ int write_ctree_super(struct btrfs_trans_handle *trans,
 
 int btrfs_free_fs_root(struct btrfs_fs_info *fs_info, struct btrfs_root *root)
 {
-        WARN_ON(!RB_EMPTY_ROOT(&root->inode_tree));
+        spin_lock(&fs_info->fs_roots_radix_lock);
         radix_tree_delete(&fs_info->fs_roots_radix,
                           (unsigned long)root->root_key.objectid);
+        spin_unlock(&fs_info->fs_roots_radix_lock);
+
+        if (btrfs_root_refs(&root->root_item) == 0)
+                synchronize_srcu(&fs_info->subvol_srcu);
+
+        free_fs_root(root);
+        return 0;
+}
+
+static void free_fs_root(struct btrfs_root *root)
+{
+        WARN_ON(!RB_EMPTY_ROOT(&root->inode_tree));
         if (root->anon_super.s_dev) {
                 down_write(&root->anon_super.s_umount);
                 kill_anon_super(&root->anon_super);
         }
-        if (root->node)
-                free_extent_buffer(root->node);
-        if (root->commit_root)
-                free_extent_buffer(root->commit_root);
+        free_extent_buffer(root->node);
+        free_extent_buffer(root->commit_root);
         kfree(root->name);
         kfree(root);
-        return 0;
 }
 
 static int del_fs_roots(struct btrfs_fs_info *fs_info)
@@ -2245,6 +2356,20 @@ static int del_fs_roots(struct btrfs_fs_info *fs_info)
         struct btrfs_root *gang[8];
         int i;
 
+        while (!list_empty(&fs_info->dead_roots)) {
+                gang[0] = list_entry(fs_info->dead_roots.next,
+                                     struct btrfs_root, root_list);
+                list_del(&gang[0]->root_list);
+
+                if (gang[0]->in_radix) {
+                        btrfs_free_fs_root(fs_info, gang[0]);
+                } else {
+                        free_extent_buffer(gang[0]->node);
+                        free_extent_buffer(gang[0]->commit_root);
+                        kfree(gang[0]);
+                }
+        }
+
         while (1) {
                 ret = radix_tree_gang_lookup(&fs_info->fs_roots_radix,
                                              (void **)gang, 0,
@@ -2274,9 +2399,6 @@ int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info)
                 root_objectid = gang[ret - 1]->root_key.objectid + 1;
                 for (i = 0; i < ret; i++) {
                         root_objectid = gang[i]->root_key.objectid;
-                        ret = btrfs_find_dead_roots(fs_info->tree_root,
-                                                    root_objectid);
-                        BUG_ON(ret);
                         btrfs_orphan_cleanup(gang[i]);
                 }
                 root_objectid++;
@@ -2290,13 +2412,19 @@ int btrfs_commit_super(struct btrfs_root *root)
         int ret;
 
         mutex_lock(&root->fs_info->cleaner_mutex);
+        btrfs_run_delayed_iputs(root);
         btrfs_clean_old_snapshots(root);
         mutex_unlock(&root->fs_info->cleaner_mutex);
-        trans = btrfs_start_transaction(root, 1);
+
+        /* wait until ongoing cleanup work done */
+        down_write(&root->fs_info->cleanup_work_sem);
+        up_write(&root->fs_info->cleanup_work_sem);
+
+        trans = btrfs_join_transaction(root, 1);
         ret = btrfs_commit_transaction(trans, root);
         BUG_ON(ret);
         /* run commit again to drop the original snapshot */
-        trans = btrfs_start_transaction(root, 1);
+        trans = btrfs_join_transaction(root, 1);
         btrfs_commit_transaction(trans, root);
         ret = btrfs_write_and_wait_transaction(NULL, root);
         BUG_ON(ret);
@@ -2313,15 +2441,19 @@ int close_ctree(struct btrfs_root *root)
         fs_info->closing = 1;
         smp_mb();
 
-        kthread_stop(root->fs_info->transaction_kthread);
-        kthread_stop(root->fs_info->cleaner_kthread);
-
+        btrfs_put_block_group_cache(fs_info);
         if (!(fs_info->sb->s_flags & MS_RDONLY)) {
                 ret =  btrfs_commit_super(root);
                 if (ret)
                         printk(KERN_ERR "btrfs: commit super ret %d\n", ret);
         }
 
+        kthread_stop(root->fs_info->transaction_kthread);
+        kthread_stop(root->fs_info->cleaner_kthread);
+
+        fs_info->closing = 2;
+        smp_mb();
+
         if (fs_info->delalloc_bytes) {
                 printk(KERN_INFO "btrfs: at unmount delalloc count %llu\n",
                        (unsigned long long)fs_info->delalloc_bytes);
@@ -2348,6 +2480,7 @@ int close_ctree(struct btrfs_root *root)
 
         iput(fs_info->btree_inode);
 
+        btrfs_stop_workers(&fs_info->generic_worker);
         btrfs_stop_workers(&fs_info->fixup_workers);
         btrfs_stop_workers(&fs_info->delalloc_workers);
         btrfs_stop_workers(&fs_info->workers);
@@ -2355,12 +2488,14 @@ int close_ctree(struct btrfs_root *root)
         btrfs_stop_workers(&fs_info->endio_meta_workers);
         btrfs_stop_workers(&fs_info->endio_meta_write_workers);
         btrfs_stop_workers(&fs_info->endio_write_workers);
+        btrfs_stop_workers(&fs_info->endio_freespace_worker);
         btrfs_stop_workers(&fs_info->submit_workers);
 
         btrfs_close_devices(fs_info->fs_devices);
         btrfs_mapping_tree_free(&fs_info->mapping_tree);
 
         bdi_destroy(&fs_info->bdi);
+        cleanup_srcu_struct(&fs_info->subvol_srcu);
 
         kfree(fs_info->extent_root);
         kfree(fs_info->tree_root);
@@ -2375,7 +2510,8 @@ int btrfs_buffer_uptodate(struct extent_buffer *buf, u64 parent_transid)
         int ret;
         struct inode *btree_inode = buf->first_page->mapping->host;
 
-        ret = extent_buffer_uptodate(&BTRFS_I(btree_inode)->io_tree, buf);
+        ret = extent_buffer_uptodate(&BTRFS_I(btree_inode)->io_tree, buf,
+                                     NULL);
         if (!ret)
                 return ret;