Merge git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable

* git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable:
  Btrfs: try to free metadata pages when we free btree blocks
  Btrfs: add extra flushing for renames and truncates
  Btrfs: make sure btrfs_update_delayed_ref doesn't increase ref_mod
  Btrfs: optimize fsyncs on old files
  Btrfs: tree logging unlink/rename fixes
  Btrfs: Make sure i_nlink doesn't hit zero too soon during log replay
  Btrfs: limit balancing work while flushing delayed refs
  Btrfs: readahead checksums during btrfs_finish_ordered_io
  Btrfs: leave btree locks spinning more often
  Btrfs: Only let very young transactions grow during commit
  Btrfs: Check for a blocking lock before taking the spin
  Btrfs: reduce stack in cow_file_range
  Btrfs: reduce stalls during transaction commit
  Btrfs: process the delayed reference queue in clusters
  Btrfs: try to cleanup delayed refs while freeing extents
  Btrfs: reduce stack usage in some crucial tree balancing functions
  Btrfs: do extent allocation and reference count updates in the background
  Btrfs: don't preallocate metadata blocks during btrfs_search_slot

1 files changed, 57 insertions, 24 deletions

diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 6ec80c0fc869..92d73929d381 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -668,14 +668,31 @@ static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio,
 static int btree_writepage(struct page *page, struct writeback_control *wbc)
 {
         struct extent_io_tree *tree;
+        struct btrfs_root *root = BTRFS_I(page->mapping->host)->root;
+        struct extent_buffer *eb;
+        int was_dirty;
+
         tree = &BTRFS_I(page->mapping->host)->io_tree;
+        if (!(current->flags & PF_MEMALLOC)) {
+                return extent_write_full_page(tree, page,
+                                              btree_get_extent, wbc);
+        }
 
-        if (current->flags & PF_MEMALLOC) {
-                redirty_page_for_writepage(wbc, page);
-                unlock_page(page);
-                return 0;
+        redirty_page_for_writepage(wbc, page);
+        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);
+        if (!was_dirty) {
+                spin_lock(&root->fs_info->delalloc_lock);
+                root->fs_info->dirty_metadata_bytes += PAGE_CACHE_SIZE;
+                spin_unlock(&root->fs_info->delalloc_lock);
         }
-        return extent_write_full_page(tree, page, btree_get_extent, wbc);
+        free_extent_buffer(eb);
+
+        unlock_page(page);
+        return 0;
 }
 
 static int btree_writepages(struct address_space *mapping,
@@ -684,15 +701,15 @@ static int btree_writepages(struct address_space *mapping,
         struct extent_io_tree *tree;
         tree = &BTRFS_I(mapping->host)->io_tree;
         if (wbc->sync_mode == WB_SYNC_NONE) {
+                struct btrfs_root *root = BTRFS_I(mapping->host)->root;
                 u64 num_dirty;
-                u64 start = 0;
                 unsigned long thresh = 32 * 1024 * 1024;
 
                 if (wbc->for_kupdate)
                         return 0;
 
-                num_dirty = count_range_bits(tree, &start, (u64)-1,
-                                             thresh, EXTENT_DIRTY);
+                /* this is a bit racy, but that's ok */
+                num_dirty = root->fs_info->dirty_metadata_bytes;
                 if (num_dirty < thresh)
                         return 0;
         }
@@ -859,9 +876,17 @@ int clean_tree_block(struct btrfs_trans_handle *trans, struct btrfs_root *root,
             root->fs_info->running_transaction->transid) {
                 btrfs_assert_tree_locked(buf);
 
-                /* ugh, clear_extent_buffer_dirty can be expensive */
-                btrfs_set_lock_blocking(buf);
+                if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &buf->bflags)) {
+                        spin_lock(&root->fs_info->delalloc_lock);
+                        if (root->fs_info->dirty_metadata_bytes >= buf->len)
+                                root->fs_info->dirty_metadata_bytes -= buf->len;
+                        else
+                                WARN_ON(1);
+                        spin_unlock(&root->fs_info->delalloc_lock);
+                }
 
+                /* ugh, clear_extent_buffer_dirty needs to lock the page */
+                btrfs_set_lock_blocking(buf);
                 clear_extent_buffer_dirty(&BTRFS_I(btree_inode)->io_tree,
                                           buf);
         }
@@ -1471,12 +1496,6 @@ static int transaction_kthread(void *arg)
                 vfs_check_frozen(root->fs_info->sb, SB_FREEZE_WRITE);
                 mutex_lock(&root->fs_info->transaction_kthread_mutex);
 
-                if (root->fs_info->total_ref_cache_size > 20 * 1024 * 1024) {
-                        printk(KERN_INFO "btrfs: total reference cache "
-                               "size %llu\n",
-                               root->fs_info->total_ref_cache_size);
-                }
-
                 mutex_lock(&root->fs_info->trans_mutex);
                 cur = root->fs_info->running_transaction;
                 if (!cur) {
@@ -1493,6 +1512,7 @@ static int transaction_kthread(void *arg)
                 mutex_unlock(&root->fs_info->trans_mutex);
                 trans = btrfs_start_transaction(root, 1);
                 ret = btrfs_commit_transaction(trans, root);
+
 sleep:
                 wake_up_process(root->fs_info->cleaner_kthread);
                 mutex_unlock(&root->fs_info->transaction_kthread_mutex);
@@ -1552,6 +1572,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
         INIT_LIST_HEAD(&fs_info->dead_roots);
         INIT_LIST_HEAD(&fs_info->hashers);
         INIT_LIST_HEAD(&fs_info->delalloc_inodes);
+        INIT_LIST_HEAD(&fs_info->ordered_operations);
         spin_lock_init(&fs_info->delalloc_lock);
         spin_lock_init(&fs_info->new_trans_lock);
         spin_lock_init(&fs_info->ref_cache_lock);
@@ -1611,10 +1632,6 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 
         extent_io_tree_init(&fs_info->pinned_extents,
                              fs_info->btree_inode->i_mapping, GFP_NOFS);
-        extent_io_tree_init(&fs_info->pending_del,
-                             fs_info->btree_inode->i_mapping, GFP_NOFS);
-        extent_io_tree_init(&fs_info->extent_ins,
-                             fs_info->btree_inode->i_mapping, GFP_NOFS);
         fs_info->do_barriers = 1;
 
         INIT_LIST_HEAD(&fs_info->dead_reloc_roots);
@@ -1627,9 +1644,9 @@ struct btrfs_root *open_ctree(struct super_block *sb,
         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->extent_ins_mutex);
         mutex_init(&fs_info->pinned_mutex);
         mutex_init(&fs_info->chunk_mutex);
         mutex_init(&fs_info->transaction_kthread_mutex);
@@ -2358,8 +2375,7 @@ void btrfs_mark_buffer_dirty(struct extent_buffer *buf)
         struct btrfs_root *root = BTRFS_I(buf->first_page->mapping->host)->root;
         u64 transid = btrfs_header_generation(buf);
         struct inode *btree_inode = root->fs_info->btree_inode;
-
-        btrfs_set_lock_blocking(buf);
+        int was_dirty;
 
         btrfs_assert_tree_locked(buf);
         if (transid != root->fs_info->generation) {
@@ -2370,7 +2386,13 @@ void btrfs_mark_buffer_dirty(struct extent_buffer *buf)
                         (unsigned long long)root->fs_info->generation);
                 WARN_ON(1);
         }
-        set_extent_buffer_dirty(&BTRFS_I(btree_inode)->io_tree, buf);
+        was_dirty = set_extent_buffer_dirty(&BTRFS_I(btree_inode)->io_tree,
+                                            buf);
+        if (!was_dirty) {
+                spin_lock(&root->fs_info->delalloc_lock);
+                root->fs_info->dirty_metadata_bytes += buf->len;
+                spin_unlock(&root->fs_info->delalloc_lock);
+        }
 }
 
 void btrfs_btree_balance_dirty(struct btrfs_root *root, unsigned long nr)
@@ -2410,6 +2432,7 @@ int btrfs_read_buffer(struct extent_buffer *buf, u64 parent_transid)
 int btree_lock_page_hook(struct page *page)
 {
         struct inode *inode = page->mapping->host;
+        struct btrfs_root *root = BTRFS_I(inode)->root;
         struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
         struct extent_buffer *eb;
         unsigned long len;
@@ -2425,6 +2448,16 @@ int btree_lock_page_hook(struct page *page)
 
         btrfs_tree_lock(eb);
         btrfs_set_header_flag(eb, BTRFS_HEADER_FLAG_WRITTEN);
+
+        if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &eb->bflags)) {
+                spin_lock(&root->fs_info->delalloc_lock);
+                if (root->fs_info->dirty_metadata_bytes >= eb->len)
+                        root->fs_info->dirty_metadata_bytes -= eb->len;
+                else
+                        WARN_ON(1);
+                spin_unlock(&root->fs_info->delalloc_lock);
+        }
+
         btrfs_tree_unlock(eb);
         free_extent_buffer(eb);
 out: