Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To:     linux-kernel@vger.kernel.org
Cc:     Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        stable@vger.kernel.org,
        Jonathan Nicklin <jnicklin@blockbridge.com>,
        Sagi Grimberg <sagi@grimberg.me>,
        Christoph Hellwig <hch@lst.de>, Sasha Levin <sashal@kernel.org>
Subject: [PATCH 5.15 092/121] nvme-tcp: fix regression that causes sporadic requests to time out
Date:   Tue, 13 Sep 2022 16:04:43 +0200
Message-Id: <20220913140401.315082269@linuxfoundation.org>
In-Reply-To: <20220913140357.323297659@linuxfoundation.org>
References: <20220913140357.323297659@linuxfoundation.org>
User-Agent: quilt/0.67
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Precedence: bulk

From: Sagi Grimberg <sagi@grimberg.me>

[ Upstream commit 3770a42bb8ceb856877699257a43c0585a5d2996 ]

When we queue requests, we strive to batch as much as possible and also
signal the network stack that more data is about to be sent over a socket
with MSG_SENDPAGE_NOTLAST. This flag looks at the pending requests queued
as well as queue->more_requests that is derived from the block layer
last-in-batch indication.

We set more_request=true when we flush the request directly from
.queue_rq submission context (in nvme_tcp_send_all), however this is
wrongly assuming that no other requests may be queued during the
execution of nvme_tcp_send_all.

Due to this, a race condition may happen where:

 1. request X is queued as !last-in-batch
 2. request X submission context calls nvme_tcp_send_all directly
 3. nvme_tcp_send_all is preempted and schedules to a different cpu
 4. request Y is queued as last-in-batch
 5. nvme_tcp_send_all context sends request X+Y, however signals for
    both MSG_SENDPAGE_NOTLAST because queue->more_requests=true.

==> none of the requests is pushed down to the wire as the network
stack is waiting for more data, both requests timeout.

To fix this, we eliminate queue->more_requests and only rely on
the queue req_list and send_list to be not-empty.

Fixes: 122e5b9f3d37 ("nvme-tcp: optimize network stack with setting msg flags according to batch size")
Reported-by: Jonathan Nicklin <jnicklin@blockbridge.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Tested-by: Jonathan Nicklin <jnicklin@blockbridge.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/nvme/host/tcp.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c
index 2c6e031135716..96d8d7844e846 100644
--- a/drivers/nvme/host/tcp.c
+++ b/drivers/nvme/host/tcp.c
@@ -119,7 +119,6 @@ struct nvme_tcp_queue {
 	struct mutex		send_mutex;
 	struct llist_head	req_list;
 	struct list_head	send_list;
-	bool			more_requests;
 
 	/* recv state */
 	void			*pdu;
@@ -315,7 +314,7 @@ static inline void nvme_tcp_send_all(struct nvme_tcp_queue *queue)
 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;
+		!llist_empty(&queue->req_list);
 }
 
 static inline void nvme_tcp_queue_request(struct nvme_tcp_request *req,
@@ -334,9 +333,7 @@ static inline void nvme_tcp_queue_request(struct nvme_tcp_request *req,
 	 */
 	if (queue->io_cpu == raw_smp_processor_id() &&
 	    sync && empty && mutex_trylock(&queue->send_mutex)) {
-		queue->more_requests = !last;
 		nvme_tcp_send_all(queue);
-		queue->more_requests = false;
 		mutex_unlock(&queue->send_mutex);
 	}
 
-- 
2.35.1