[PATCH -stable] workqueue: UNBOUND -> REBIND morphing inrebind_workers() should be atomic

From: Tejun Heo
Date: Tue Sep 04 2012 - 19:58:11 EST


From: Lai Jiangshan <laijs@xxxxxxxxxxxxxx>

The compiler may compile the following code into TWO write/modify
instructions.

worker->flags &= ~WORKER_UNBOUND;
worker->flags |= WORKER_REBIND;

so the other CPU may temporarily see worker->flags which doesn't have
either WORKER_UNBOUND or WORKER_REBIND set and perform local wakeup
prematurely.

Fix it by using single explicit assignment via ACCESS_ONCE().

Because idle workers have another WORKER_NOT_RUNNING flag, this bug
doesn't exist for them; however, update it to use the same pattern for
consistency.

tj: Applied the change to idle workers too and updated comments and
patch description a bit.

stable: Idle worker rebinding doesn't apply for -stable and
WORKER_UNBOUND used to be WORKER_ROGUE. Updated accordingly.

Signed-off-by: Lai Jiangshan <laijs@xxxxxxxxxxxxxx>
Signed-off-by: Tejun Heo <tj@xxxxxxxxxx>
Cc: stable@xxxxxxxxxxxxxxx
---
kernel/workqueue.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 9a3128d..d7eb05a 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -3441,14 +3441,17 @@ static int __cpuinit trustee_thread(void *__gcwq)

for_each_busy_worker(worker, i, pos, gcwq) {
struct work_struct *rebind_work = &worker->rebind_work;
+ unsigned long worker_flags = worker->flags;

/*
* Rebind_work may race with future cpu hotplug
* operations. Use a separate flag to mark that
- * rebinding is scheduled.
+ * rebinding is scheduled. The morphing should
+ * be atomic.
*/
- worker->flags |= WORKER_REBIND;
- worker->flags &= ~WORKER_ROGUE;
+ worker_flags |= WORKER_REBIND;
+ worker_flags &= ~WORKER_ROGUE;
+ ACCESS_ONCE(worker->flags) = worker_flags;

/* queue rebind_work, wq doesn't matter, use the default one */
if (test_and_set_bit(WORK_STRUCT_PENDING_BIT,
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/