Received-SPF: pass (google.com: domain of linux-kernel+bounces-98413-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
From: "brookxu.cn" <brookxu.cn@gmail.com>
To: kbusch@kernel.org,
	axboe@kernel.dk,
	hch@lst.de,
	sagi@grimberg.me
Cc: kch@nvidia.com,
	linux-nvme@lists.infradead.org,
	linux-kernel@vger.kernel.org
Subject: [PATCH v4] nvme: fix reconnection fail due to reserved tag allocation
Date: Mon, 11 Mar 2024 10:09:27 +0800
Message-Id: <20240311020927.85482-1-brookxu.cn@gmail.com>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

From: Chunguang Xu <chunguang.xu@shopee.com>

We found a issue on production environment while using NVMe over RDMA,
admin_q reconnect failed forever while remote target and network is ok.
After dig into it, we found it may caused by a ABBA deadlock due to tag
allocation. In my case, the tag was hold by a keep alive request
waiting inside admin_q, as we quiesced admin_q while reset ctrl, so the
request maked as idle and will not process before reset success. As
fabric_q shares tagset with admin_q, while reconnect remote target, we
need a tag for connect command, but the only one reserved tag was held
by keep alive command which waiting inside admin_q. As a result, we
failed to reconnect admin_q forever. In order to fix this issue, I
think we should keep two reserved tags for admin queue.

Fixes: ed01fee283a0 ("nvme-fabrics: only reserve a single tag")
Signed-off-by: Chunguang Xu <chunguang.xu@shopee.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
v4: move init value and the comments to the real users
v3: rearrange commit log, no code change
v2: keep two reserved tags for admin_q instead of drop keep alive request

drivers/nvme/host/core.c    | 6 ++++--
 drivers/nvme/host/fabrics.h | 7 -------
 2 files changed, 4 insertions(+), 9 deletions(-)

diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
index 0a96362912ce..fe3627c5bdc9 100644
--- a/drivers/nvme/host/core.c
+++ b/drivers/nvme/host/core.c
@@ -4359,7 +4359,8 @@ int nvme_alloc_admin_tag_set(struct nvme_ctrl *ctrl, struct blk_mq_tag_set *set,
 	set->ops = ops;
 	set->queue_depth = NVME_AQ_MQ_TAG_DEPTH;
 	if (ctrl->ops->flags & NVME_F_FABRICS)
-		set->reserved_tags = NVMF_RESERVED_TAGS;
+		/* Reserved for fabric connect and keep alive */
+		set->reserved_tags = 2;
 	set->numa_node = ctrl->numa_node;
 	set->flags = BLK_MQ_F_NO_SCHED;
 	if (ctrl->ops->flags & NVME_F_BLOCKING)
@@ -4428,7 +4429,8 @@ int nvme_alloc_io_tag_set(struct nvme_ctrl *ctrl, struct blk_mq_tag_set *set,
 	if (ctrl->quirks & NVME_QUIRK_SHARED_TAGS)
 		set->reserved_tags = NVME_AQ_DEPTH;
 	else if (ctrl->ops->flags & NVME_F_FABRICS)
-		set->reserved_tags = NVMF_RESERVED_TAGS;
+		/* Reserved for fabric connect */
+		set->reserved_tags = 1;
 	set->numa_node = ctrl->numa_node;
 	set->flags = BLK_MQ_F_SHOULD_MERGE;
 	if (ctrl->ops->flags & NVME_F_BLOCKING)
diff --git a/drivers/nvme/host/fabrics.h b/drivers/nvme/host/fabrics.h
index 06cc54851b1b..37c974c38dcb 100644
--- a/drivers/nvme/host/fabrics.h
+++ b/drivers/nvme/host/fabrics.h
@@ -18,13 +18,6 @@
 /* default is -1: the fail fast mechanism is disabled  */
 #define NVMF_DEF_FAIL_FAST_TMO		-1
 
-/*
- * Reserved one command for internal usage.  This command is used for sending
- * the connect command, as well as for the keep alive command on the admin
- * queue once live.
- */
-#define NVMF_RESERVED_TAGS	1
-
 /*
  * Define a host as seen by the target.  We allocate one at boot, but also
  * allow the override it when creating controllers.  This is both to provide
-- 
2.25.1