This is a note to let you know that I've just added the patch titled nvme-tcp: fix io_work priority inversion to the 5.14-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary The filename of the patch is: nvme-tcp-fix-io_work-priority-inversion.patch and it can be found in the queue-5.14 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable@xxxxxxxxxxxxxxx> know about it. commit cf44fd56a5dbf25702cafd2470d67626c4a19aa3 Author: Keith Busch <kbusch@xxxxxxxxxx> Date: Thu Sep 9 08:54:52 2021 -0700 nvme-tcp: fix io_work priority inversion [ Upstream commit 70f437fb4395ad4d1d16fab9a1ad9fbc9fc0579b ] Dispatching requests inline with the .queue_rq() call may block while holding the send_mutex. If the tcp io_work also happens to schedule, it may see the req_list is non-empty, leaving "pending" true and remaining in TASK_RUNNING. Since io_work is of higher scheduling priority, the .queue_rq task may not get a chance to run, blocking forward progress and leading to io timeouts. Instead of checking for pending requests within io_work, let the queueing restart io_work outside the send_mutex lock if there is more work to be done. Fixes: a0fdd1418007f ("nvme-tcp: rerun io_work if req_list is not empty") Reported-by: Samuel Jones <sjones@xxxxxxxxxxxxx> Signed-off-by: Keith Busch <kbusch@xxxxxxxxxx> Reviewed-by: Sagi Grimberg <sagi@xxxxxxxxxxx> Signed-off-by: Christoph Hellwig <hch@xxxxxx> Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx> diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c index 48b70e5235a3..19a711395cdc 100644 --- a/drivers/nvme/host/tcp.c +++ b/drivers/nvme/host/tcp.c @@ -273,6 +273,12 @@ 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) +{ + return !list_empty(&queue->send_list) || + !llist_empty(&queue->req_list) || queue->more_requests; +} + static inline void nvme_tcp_queue_request(struct nvme_tcp_request *req, bool sync, bool last) { @@ -293,9 +299,10 @@ static inline void nvme_tcp_queue_request(struct nvme_tcp_request *req, nvme_tcp_send_all(queue); queue->more_requests = false; mutex_unlock(&queue->send_mutex); - } else if (last) { - queue_work_on(queue->io_cpu, nvme_tcp_wq, &queue->io_work); } + + if (last && nvme_tcp_queue_more(queue)) + queue_work_on(queue->io_cpu, nvme_tcp_wq, &queue->io_work); } static void nvme_tcp_process_req_list(struct nvme_tcp_queue *queue) @@ -893,12 +900,6 @@ done: read_unlock_bh(&sk->sk_callback_lock); } -static inline bool nvme_tcp_queue_more(struct nvme_tcp_queue *queue) -{ - return !list_empty(&queue->send_list) || - !llist_empty(&queue->req_list) || queue->more_requests; -} - static inline void nvme_tcp_done_send_req(struct nvme_tcp_queue *queue) { queue->request = NULL; @@ -1132,8 +1133,7 @@ static void nvme_tcp_io_work(struct work_struct *w) pending = true; else if (unlikely(result < 0)) break; - } else - pending = !llist_empty(&queue->req_list); + } result = nvme_tcp_try_recv(queue); if (result > 0)