Hi Mingming,
Some update for patches in the patch queue. I am sending it as a diff
against patch queue so that it makes it easier to apply
The first hunk is a missed return 0 as a part of patch
ext4_invalidate_pages_when_delalloc_alloc_fail.patch
diff --git a/ext4_invalidate_pages_when_delalloc_alloc_fail.patch b/ext4_invalidate_pages_when_delalloc_alloc_fail.patch
index f0722ab..d47d3bf 100644
--- a/ext4_invalidate_pages_when_delalloc_alloc_fail.patch
+++ b/ext4_invalidate_pages_when_delalloc_alloc_fail.patch
@@ -92,7 +92,7 @@ Index: linux-2.6.27-rc3/fs/ext4/inode.c
+ * will find the dirty page again
+ */
+ if (err == -EAGAIN)
-+ return;
++ return 0;
+ /*
+ * get block failure will cause us
+ * to loop in writepages. Because
diff --git a/fix-delalloc-release-block-reservation-for-truncate b/fix-delalloc-release-block-reservation-for-truncate
index 8dc891b..32f1063 100644
--- a/fix-delalloc-release-block-reservation-for-truncate
+++ b/fix-delalloc-release-block-reservation-for-truncate
@@ -48,8 +48,8 @@ Index: linux-2.6.27-rc3/fs/ext4/inode.c
+ * but since this function is called from invalidate
+ * page, it's harmless to return without any action
+ */
-+ printk(KERN_INFO "ext4 delalloc try to release %d reserved"
-+ "blocks for inode %lu, but there is no reserved"
++ printk(KERN_INFO "ext4 delalloc try to release %d reserved "
++ "blocks for inode %lu, but there is no reserved "
+ "data blocks\n", to_free, inode->i_ino);
+ spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
+ return;