[03/18] xfs: evict CoW fork extents when performing finsert/fcollapse

Commit Message

Christoph Hellwig Oct. 19, 2017, 2:21 p.m. UTC

From: "Darrick J. Wong" <[e-mail]>

commit 3af423b03435c81036fa710623d3ae92fbe346a3 upstream.

When we perform an finsert/fcollapse operation, cancel all the CoW
extents for the affected file offset range so that they don't end up
pointing to the wrong blocks.

Reported-by: Amir Goldstein <[e-mail]>
Reviewed-by: Carlos Maiolino <[e-mail]>
Reviewed-by: Christoph Hellwig <[e-mail]>
Signed-off-by: Darrick J. Wong <[e-mail]>
---
 fs/xfs/xfs_bmap_util.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

Patch

diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c
index 3e9b7a4fb8fd..1de4e96df15a 100644
--- a/fs/xfs/xfs_bmap_util.c
+++ b/fs/xfs/xfs_bmap_util.c
@@ -1459,7 +1459,19 @@  xfs_shift_file_space(
 		return error;
 
 	/*
-	 * The extent shiting code works on extent granularity. So, if
+	 * Clean out anything hanging around in the cow fork now that
+	 * we've flushed all the dirty data out to disk to avoid having
+	 * CoW extents at the wrong offsets.
+	 */
+	if (xfs_is_reflink_inode(ip)) {
+		error = xfs_reflink_cancel_cow_range(ip, offset, NULLFILEOFF,
+				true);
+		if (error)
+			return error;
+	}
+
+	/*
+	 * The extent shifting code works on extent granularity. So, if
 	 * stop_fsb is not the starting block of extent, we need to split
 	 * the extent at stop_fsb.
 	 */