[PATCHv3] nvme-tcp: strict pdu pacing to avoid send stalls on TLS

[PATCHv3] nvme-tcp: strict pdu pacing to avoid send stalls on TLS

[Hannes Reinecke]

TLS requires a strict pdu pacing via MSG_EOR to signal the end
of a record and subsequent encryption. If we do not set MSG_EOR
at the end of a sequence the record won't be closed, encryption
doesn't start, and we end up with a send stall as the message
will never be passed on to the TCP layer.
So do not check for the queue status when TLS is enabled but
rather make the MSG_MORE setting dependent on the current
request only.

Signed-off-by: Hannes Reinecke <hare@kernel.org>
---
 drivers/nvme/host/tcp.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c
index fdbcdcedcee9..28bc2f373cfa 100644
--- a/drivers/nvme/host/tcp.c
+++ b/drivers/nvme/host/tcp.c
@@ -360,12 +360,18 @@ static inline void nvme_tcp_send_all(struct nvme_tcp_queue *queue)
 	} while (ret > 0);
 }
 
-static inline bool nvme_tcp_queue_more(struct nvme_tcp_queue *queue)
+static inline bool nvme_tcp_queue_has_pending(struct nvme_tcp_queue *queue)
 {
 	return !list_empty(&queue->send_list) ||
 		!llist_empty(&queue->req_list);
 }
 
+static inline bool nvme_tcp_queue_more(struct nvme_tcp_queue *queue)
+{
+	return !nvme_tcp_tls(&queue->ctrl->ctrl) &&
+		nvme_tcp_queue_has_pending(queue);
+}
+
 static inline void nvme_tcp_queue_request(struct nvme_tcp_request *req,
 		bool sync, bool last)
 {
@@ -386,7 +392,7 @@ static inline void nvme_tcp_queue_request(struct nvme_tcp_request *req,
 		mutex_unlock(&queue->send_mutex);
 	}
 
-	if (last && nvme_tcp_queue_more(queue))
+	if (last && nvme_tcp_queue_has_pending(queue))
 		queue_work_on(queue->io_cpu, nvme_tcp_wq, &queue->io_work);
 }
 
-- 
2.35.3

Re: [PATCHv3] nvme-tcp: strict pdu pacing to avoid send stalls on TLS

[Sagi Grimberg]

Reviewed-by: Sagi Grimberg <sagi@grimberg.me>

Re: [PATCHv3] nvme-tcp: strict pdu pacing to avoid send stalls on TLS

[Keith Busch]

On Thu, Apr 18, 2024 at 12:39:45PM +0200, Hannes Reinecke wrote:
> TLS requires a strict pdu pacing via MSG_EOR to signal the end
> of a record and subsequent encryption. If we do not set MSG_EOR
> at the end of a sequence the record won't be closed, encryption
> doesn't start, and we end up with a send stall as the message
> will never be passed on to the TCP layer.
> So do not check for the queue status when TLS is enabled but
> rather make the MSG_MORE setting dependent on the current
> request only.

Thanks, applied to nvme-6.9.