Btrfs: Data ordered fixes

* In btrfs_delete_inode, wait for ordered extents after calling
truncate_inode_pages.  This is much faster, and more correct

* Properly clear our the PageChecked bit everywhere we redirty the page.

* Change the writepage fixup handler to lock the page range and check to
see if an ordered extent had been inserted since the improperly dirtied
page was discovered

* Wait for ordered extents outside the transaction.  This isn't required
for locking rules but does improve transaction latencies

* Reduce contention on the alloc_mutex by dropping it while incrementing
refs on a node/leaf and while dropping refs on a leaf.

Signed-off-by: Chris Mason <chris.mason@oracle.com>

1 files changed, 18 insertions, 5 deletions

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 50ee4befac8e..8fb6dc25e7a5 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -418,7 +418,7 @@ void btrfs_writepage_fixup_worker(struct btrfs_work *work)
 
         fixup = container_of(work, struct btrfs_writepage_fixup, work);
         page = fixup->page;
-
+again:
         lock_page(page);
         if (!page->mapping || !PageDirty(page) || !PageChecked(page)) {
                 ClearPageChecked(page);
@@ -430,9 +430,21 @@ void btrfs_writepage_fixup_worker(struct btrfs_work *work)
         page_end = page_offset(page) + PAGE_CACHE_SIZE - 1;
 
         lock_extent(&BTRFS_I(inode)->io_tree, page_start, page_end, GFP_NOFS);
-        ordered = btrfs_lookup_ordered_extent(inode, page_start);
-        if (ordered)
+
+        /* already ordered? We're done */
+        if (test_range_bit(&BTRFS_I(inode)->io_tree, page_start, page_end,
+                             EXTENT_ORDERED, 0)) {
                 goto out;
+        }
+
+        ordered = btrfs_lookup_ordered_extent(inode, page_start);
+        if (ordered) {
+                unlock_extent(&BTRFS_I(inode)->io_tree, page_start,
+                              page_end, GFP_NOFS);
+                unlock_page(page);
+                btrfs_start_ordered_extent(inode, ordered, 1);
+                goto again;
+        }
 
         set_extent_delalloc(&BTRFS_I(inode)->io_tree, page_start, page_end,
                             GFP_NOFS);
@@ -1465,11 +1477,11 @@ void btrfs_delete_inode(struct inode *inode)
         unsigned long nr;
         int ret;
 
-        btrfs_wait_ordered_range(inode, 0, (u64)-1);
         truncate_inode_pages(&inode->i_data, 0);
         if (is_bad_inode(inode)) {
                 goto no_delete;
         }
+        btrfs_wait_ordered_range(inode, 0, (u64)-1);
 
         btrfs_i_size_write(inode, 0);
         trans = btrfs_start_transaction(root, 1);
@@ -2707,6 +2719,7 @@ static void btrfs_invalidatepage(struct page *page, unsigned long offset)
                  1, 1, GFP_NOFS);
         __btrfs_releasepage(page, GFP_NOFS);
 
+        ClearPageChecked(page);
         if (PagePrivate(page)) {
                 invalidate_extent_lru(tree, page_offset(page),
                                       PAGE_CACHE_SIZE);
@@ -2818,10 +2831,10 @@ static void btrfs_truncate(struct inode *inode)
                 return;
 
         btrfs_truncate_page(inode->i_mapping, inode->i_size);
+        btrfs_wait_ordered_range(inode, inode->i_size & (~mask), (u64)-1);
 
         trans = btrfs_start_transaction(root, 1);
         btrfs_set_trans_block_group(trans, inode);
-        btrfs_wait_ordered_range(inode, inode->i_size & (~mask), (u64)-1);
         btrfs_i_size_write(inode, inode->i_size);
 
         /* FIXME, add redo link to tree so we don't leak on crash */