[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20190130203036.11220-1-agruenba@redhat.com>
Date:   Wed, 30 Jan 2019 21:30:36 +0100
From:   Andreas Gruenbacher <agruenba@...hat.com>
To:     Linus Torvalds <torvalds@...ux-foundation.org>
Cc:     Bob Peterson <rpeterso@...hat.com>, cluster-devel@...hat.com,
        linux-kernel@...r.kernel.org,
        Andreas Gruenbacher <agruenba@...hat.com>,
        stable@...r.kernel.org
Subject: [PATCH] gfs2: Revert "Fix loop in gfs2_rbm_find"
This reverts commit 2d29f6b96d8f80322ed2dd895bca590491c38d34.
It turns out that the fix can lead to a ~20 percent performance regression
in initial writes to the page cache according to iozone.  Let's revert this
for now to have more time for a proper fix.
Cc: stable@...r.kernel.org # v3.13+
Signed-off-by: Andreas Gruenbacher <agruenba@...hat.com>
Signed-off-by: Bob Peterson <rpeterso@...hat.com>
---
 fs/gfs2/rgrp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index 831d7cb5a49c4..17a8d3b439905 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -1780,9 +1780,9 @@ static int gfs2_rbm_find(struct gfs2_rbm *rbm, u8 state, u32 *minext,
 			goto next_iter;
 		}
 		if (ret == -E2BIG) {
-			n += rbm->bii - initial_bii;
 			rbm->bii = 0;
 			rbm->offset = 0;
+			n += (rbm->bii - initial_bii);
 			goto res_covered_end_of_rgrp;
 		}
 		return ret;
-- 
2.20.1
Powered by blists - more mailing lists