[regression] ocfs2: do *not* increment ->ki_pos twice
authorAl Viro <viro@zeniv.linux.org.uk>
Wed, 8 Apr 2015 19:41:17 +0000 (15:41 -0400)
committerAl Viro <viro@zeniv.linux.org.uk>
Wed, 8 Apr 2015 20:58:59 +0000 (16:58 -0400)
generic_file_direct_write() already does that.  Broken by
"ocfs2: do not fallback to buffer I/O write if appending"

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
fs/ocfs2/file.c

index 46e0d4e857c7f493f512196603d3725ca8d3dfaa..099972490f392dc618a6844972f15874ed1e4573 100644 (file)
@@ -2394,7 +2394,6 @@ relock:
                /*
                 * for completing the rest of the request.
                 */
-               *ppos += written;
                count -= written;
                written_buffered = generic_perform_write(file, from, *ppos);
                /*