Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (61 commits)
  ext4: Documention update for new ordered mode and delayed allocation
  ext4: do not set extents feature from the kernel
  ext4: Don't allow nonextenst mount option for large filesystem
  ext4: Enable delalloc by default.
  ext4: delayed allocation i_blocks fix for stat
  ext4: fix delalloc i_disksize early update issue
  ext4: Handle page without buffers in ext4_*_writepage()
  ext4: Add ordered mode support for delalloc
  ext4: Invert lock ordering of page_lock and transaction start in delalloc
  mm: Add range_cont mode for writeback
  ext4: delayed allocation ENOSPC handling
  percpu_counter: new function percpu_counter_sum_and_set
  ext4: Add delayed allocation support in data=writeback mode
  vfs: add hooks for ext4's delayed allocation support
  jbd2: Remove data=ordered mode support using jbd buffer heads
  ext4: Use new framework for data=ordered mode in JBD2
  jbd2: Implement data=ordered mode handling via inodes
  vfs: export filemap_fdatawrite_range()
  ext4: Fix lock inversion in ext4_ext_truncate()
  ext4: Invert the locking order of page_lock and transaction start
  ...

1 files changed, 16 insertions, 3 deletions

diff --git a/fs/buffer.c b/fs/buffer.c
index 0f51c0f7c266..5fa1512cd9a2 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -1691,11 +1691,13 @@ static int __block_write_full_page(struct inode *inode, struct page *page,
                          */
                         clear_buffer_dirty(bh);
                         set_buffer_uptodate(bh);
-                } else if (!buffer_mapped(bh) && buffer_dirty(bh)) {
+                } else if ((!buffer_mapped(bh) || buffer_delay(bh)) &&
+                           buffer_dirty(bh)) {
                         WARN_ON(bh->b_size != blocksize);
                         err = get_block(inode, block, bh, 1);
                         if (err)
                                 goto recover;
+                        clear_buffer_delay(bh);
                         if (buffer_new(bh)) {
                                 /* blockdev mappings never come here */
                                 clear_buffer_new(bh);
@@ -1774,7 +1776,8 @@ recover:
         bh = head;
         /* Recovery: lock and submit the mapped buffers */
         do {
-                if (buffer_mapped(bh) && buffer_dirty(bh)) {
+                if (buffer_mapped(bh) && buffer_dirty(bh) &&
+                    !buffer_delay(bh)) {
                         lock_buffer(bh);
                         mark_buffer_async_write(bh);
                 } else {
@@ -2061,6 +2064,7 @@ int generic_write_end(struct file *file, struct address_space *mapping,
                         struct page *page, void *fsdata)
 {
         struct inode *inode = mapping->host;
+        int i_size_changed = 0;
 
         copied = block_write_end(file, mapping, pos, len, copied, page, fsdata);
 
@@ -2073,12 +2077,21 @@ int generic_write_end(struct file *file, struct address_space *mapping,
          */
         if (pos+copied > inode->i_size) {
                 i_size_write(inode, pos+copied);
-                mark_inode_dirty(inode);
+                i_size_changed = 1;
         }
 
         unlock_page(page);
         page_cache_release(page);
 
+        /*
+         * Don't mark the inode dirty under page lock. First, it unnecessarily
+         * makes the holding time of page lock longer. Second, it forces lock
+         * ordering of page lock and transaction start for journaling
+         * filesystems.
+         */
+        if (i_size_changed)
+                mark_inode_dirty(inode);
+
         return copied;
 }
 EXPORT_SYMBOL(generic_write_end);