diff mbox

[6/7] IB/rxe: Avoid missed completions in the CM/MAD

Message ID 1479479809-10798-6-git-send-email-andrew.boyer@dell.com (mailing list archive)
State Superseded
Headers show

Commit Message

Andrew Boyer Nov. 18, 2016, 2:36 p.m. UTC
The MAD code uses the IB_CQ_REPORT_MISSED_EVENTS flag to avoid a
race between posting CQEs and arming the CQ. Without this fix, the
last completion might be left on the CQ, hanging the kthread
waiting on MAD to complete.
See ib_cq_poll_work().

The console backtraces look like this:

[ 4199.911284] Call Trace:
[ 4199.911401]  [<ffffffff9657fe95>] schedule+0x35/0x80
[ 4199.911556]  [<ffffffff965830df>] schedule_timeout+0x22f/0x2c0
[ 4199.911727]  [<ffffffff9657f7a8>] ? __schedule+0x368/0xa20
[ 4199.911891]  [<ffffffff96580903>] wait_for_completion+0xb3/0x130
[ 4199.912067]  [<ffffffff960a17e0>] ? wake_up_q+0x70/0x70
[ 4199.912243]  [<ffffffffc074a06d>] cm_destroy_id+0x13d/0x450 [ib_cm]
[ 4199.912422]  [<ffffffff961615d5>] ? printk+0x57/0x73
[ 4199.912578]  [<ffffffffc074a390>] ib_destroy_cm_id+0x10/0x20 [ib_cm]
[ 4199.912759]  [<ffffffffc076098c>] rdma_destroy_id+0xac/0x340 [rdma_cm]
[ 4199.912941]  [<ffffffffc076f2cc>] 0xffffffffc076f2cc

Peek at the CQ after arming it so that we can return a hint.

Signed-off-by: Andrew Boyer <andrew.boyer@dell.com>
---
 drivers/infiniband/sw/rxe/rxe_verbs.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

Comments

Moni Shoua Nov. 22, 2016, 5:14 p.m. UTC | #1
On Fri, Nov 18, 2016 at 4:36 PM, Andrew Boyer <andrew.boyer@dell.com> wrote:
> The MAD code uses the IB_CQ_REPORT_MISSED_EVENTS flag to avoid a
> race between posting CQEs and arming the CQ. Without this fix, the
> last completion might be left on the CQ, hanging the kthread
> waiting on MAD to complete.
> See ib_cq_poll_work().
>
Looks OK but I would edit the commit message a bit. This fix is
relevant not only for MAD and not only for workqueue polling context.
For example, iSER allocates CQ with SOFTIRQ polling context and is
also exposed to this bug (see ib_poll_handler)
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.c b/drivers/infiniband/sw/rxe/rxe_verbs.c
index 19841c8..de39b0a 100644
--- a/drivers/infiniband/sw/rxe/rxe_verbs.c
+++ b/drivers/infiniband/sw/rxe/rxe_verbs.c
@@ -1007,11 +1007,19 @@  static int rxe_peek_cq(struct ib_cq *ibcq, int wc_cnt)
 static int rxe_req_notify_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags flags)
 {
 	struct rxe_cq *cq = to_rcq(ibcq);
+	unsigned long irq_flags;
+	int ret = 0;
 
+	spin_lock_irqsave(&cq->cq_lock, irq_flags);
 	if (cq->notify != IB_CQ_NEXT_COMP)
 		cq->notify = flags & IB_CQ_SOLICITED_MASK;
 
-	return 0;
+	if ((flags & IB_CQ_REPORT_MISSED_EVENTS) && !queue_empty(cq->queue))
+		ret = 1;
+
+	spin_unlock_irqrestore(&cq->cq_lock, irq_flags);
+
+	return ret;
 }
 
 static struct ib_mr *rxe_get_dma_mr(struct ib_pd *ibpd, int access)