[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <lsq.1570043211.946262289@decadent.org.uk>
Date: Wed, 02 Oct 2019 20:06:51 +0100
From: Ben Hutchings <ben@...adent.org.uk>
To: linux-kernel@...r.kernel.org, stable@...r.kernel.org
CC: akpm@...ux-foundation.org, Denis Kirjanov <kda@...ux-powerpc.org>,
"Omar Sandoval" <osandov@...com>,
"Paul E. McKenney" <paulmck@...ux.ibm.com>,
"Peter Zijlstra" <peterz@...radead.org>,
linux-block@...r.kernel.org,
"Andrea Parri" <andrea.parri@...rulasolutions.com>,
"Jens Axboe" <axboe@...nel.dk>, "Ming Lei" <ming.lei@...hat.com>
Subject: [PATCH 3.16 22/87] sbitmap: fix improper use of smp_mb__before_atomic()
3.16.75-rc1 review patch. If anyone has any objections, please let me know.
------------------
From: Andrea Parri <andrea.parri@...rulasolutions.com>
commit a0934fd2b1208458e55fc4b48f55889809fce666 upstream.
This barrier only applies to the read-modify-write operations; in
particular, it does not apply to the atomic_set() primitive.
Replace the barrier with an smp_mb().
Fixes: 6c0ca7ae292ad ("sbitmap: fix wakeup hang after sbq resize")
Reported-by: "Paul E. McKenney" <paulmck@...ux.ibm.com>
Reported-by: Peter Zijlstra <peterz@...radead.org>
Signed-off-by: Andrea Parri <andrea.parri@...rulasolutions.com>
Reviewed-by: Ming Lei <ming.lei@...hat.com>
Cc: Jens Axboe <axboe@...nel.dk>
Cc: Omar Sandoval <osandov@...com>
Cc: Ming Lei <ming.lei@...hat.com>
Cc: linux-block@...r.kernel.org
Cc: "Paul E. McKenney" <paulmck@...ux.ibm.com>
Cc: Peter Zijlstra <peterz@...radead.org>
Signed-off-by: Jens Axboe <axboe@...nel.dk>
[bwh: Backported to 3.16: adjust filename]
Signed-off-by: Ben Hutchings <ben@...adent.org.uk>
---
block/blk-mq-tag.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
--- a/block/blk-mq-tag.c
+++ b/block/blk-mq-tag.c
@@ -499,7 +499,7 @@ static void bt_update_count(struct blk_m
* Pairs with the memory barrier in bt_clear_tag() to ensure
* that the batch size is updated before the wait counts.
*/
- smp_mb__before_atomic();
+ smp_mb();
for (i = 0; i < BT_WAIT_QUEUES; i++)
atomic_set(&bt->bs[i].wait_cnt, 1);
}
Powered by blists - more mailing lists