[PATCH 4.9 50/94] nvme-rdma: support devices with queue size < 32

[Greg Kroah-Hartman]

4.9-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Marta Rybczynska <mrybczyn@xxxxxxxxx>

commit 0544f5494a03b8846db74e02be5685d1f32b06c9 upstream.

In the case of small NVMe-oF queue size (<32) we may enter a deadlock
caused by the fact that the IB completions aren't sent waiting for 32
and the send queue will fill up.

The error is seen as (using mlx5):
[ 2048.693355] mlx5_0:mlx5_ib_post_send:3765:(pid 7273):
[ 2048.693360] nvme nvme1: nvme_rdma_post_send failed with error code -12

This patch changes the way the signaling is done so that it depends on
the queue depth now. The magic define has been removed completely.

Signed-off-by: Marta Rybczynska <marta.rybczynska@xxxxxxxxx>
Signed-off-by: Samuel Jones <sjones@xxxxxxxxx>
Acked-by: Sagi Grimberg <sagi@xxxxxxxxxxx>
Signed-off-by: Christoph Hellwig <hch@xxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

---
 drivers/nvme/host/rdma.c |   18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -1011,6 +1011,19 @@ static void nvme_rdma_send_done(struct i
 		nvme_rdma_wr_error(cq, wc, "SEND");
 }
 
+static inline int nvme_rdma_queue_sig_limit(struct nvme_rdma_queue *queue)
+{
+	int sig_limit;
+
+	/*
+	 * We signal completion every queue depth/2 and also handle the
+	 * degenerated case of a  device with queue_depth=1, where we
+	 * would need to signal every message.
+	 */
+	sig_limit = max(queue->queue_size / 2, 1);
+	return (++queue->sig_count % sig_limit) == 0;
+}
+
 static int nvme_rdma_post_send(struct nvme_rdma_queue *queue,
 		struct nvme_rdma_qe *qe, struct ib_sge *sge, u32 num_sge,
 		struct ib_send_wr *first, bool flush)
@@ -1038,9 +1051,6 @@ static int nvme_rdma_post_send(struct nv
 	 * Would have been way to obvious to handle this in hardware or
 	 * at least the RDMA stack..
 	 *
-	 * This messy and racy code sniplet is copy and pasted from the iSER
-	 * initiator, and the magic '32' comes from there as well.
-	 *
 	 * Always signal the flushes. The magic request used for the flush
 	 * sequencer is not allocated in our driver's tagset and it's
 	 * triggered to be freed by blk_cleanup_queue(). So we need to
@@ -1048,7 +1058,7 @@ static int nvme_rdma_post_send(struct nv
 	 * embeded in request's payload, is not freed when __ib_process_cq()
 	 * calls wr_cqe->done().
 	 */
-	if ((++queue->sig_count % 32) == 0 || flush)
+	if (nvme_rdma_queue_sig_limit(queue) || flush)
 		wr.send_flags |= IB_SEND_SIGNALED;
 
 	if (first)