Received-SPF: pass (google.com: domain of linux-kernel+bounces-91859-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
From: Daniel Wagner <dwagner@suse.de>
To: James Smart <james.smart@broadcom.com>
Cc: Keith Busch <kbusch@kernel.org>,
	Christoph Hellwig <hch@lst.de>,
	Sagi Grimberg <sagi@grimberg.me>,
	Hannes Reinecke <hare@suse.de>,
	linux-nvme@lists.infradead.org,
	linux-kernel@vger.kernel.org,
	Daniel Wagner <dwagner@suse.de>
Subject: [PATCH v3 2/2] nvme-rdma: short-circuit reconnect retries
Date: Tue,  5 Mar 2024 09:00:05 +0100
Message-ID: <20240305080005.3638-3-dwagner@suse.de>
In-Reply-To: <20240305080005.3638-1-dwagner@suse.de>
References: <20240305080005.3638-1-dwagner@suse.de>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

From: Hannes Reinecke <hare@suse.de>

Returning an nvme status from nvme_rdma_setup_ctrl() indicates
that the association was established and we have received a status
from the controller; consequently we should honour the DNR bit.
If not any future reconnect attempts will just return the same error, so
we can short-circuit the reconnect attempts and fail the connection
directly.

Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Daniel Wagner <dwagner@suse.de>
---
 drivers/nvme/host/rdma.c | 22 +++++++++++++++-------
 1 file changed, 15 insertions(+), 7 deletions(-)

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index d3747795ad80..7e556e10caba 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -982,9 +982,11 @@ static void nvme_rdma_free_ctrl(struct nvme_ctrl *nctrl)
 	kfree(ctrl);
 }
 
-static void nvme_rdma_reconnect_or_remove(struct nvme_rdma_ctrl *ctrl)
+static void nvme_rdma_reconnect_or_remove(struct nvme_rdma_ctrl *ctrl,
+		int status)
 {
 	enum nvme_ctrl_state state = nvme_ctrl_state(&ctrl->ctrl);
+	bool recon = status > 0 && (status & NVME_SC_DNR) ? false : true;
 
 	/* If we are resetting/deleting then do nothing */
 	if (state != NVME_CTRL_CONNECTING) {
@@ -992,12 +994,14 @@ static void nvme_rdma_reconnect_or_remove(struct nvme_rdma_ctrl *ctrl)
 		return;
 	}
 
-	if (nvmf_should_reconnect(&ctrl->ctrl)) {
+	if (recon && nvmf_should_reconnect(&ctrl->ctrl)) {
 		dev_info(ctrl->ctrl.device, "Reconnecting in %d seconds...\n",
 			ctrl->ctrl.opts->reconnect_delay);
 		queue_delayed_work(nvme_wq, &ctrl->reconnect_work,
 				ctrl->ctrl.opts->reconnect_delay * HZ);
 	} else {
+		dev_info(ctrl->ctrl.device, "Removing controller (%d)...\n",
+			 status);
 		nvme_delete_ctrl(&ctrl->ctrl);
 	}
 }
@@ -1098,10 +1102,12 @@ static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work)
 {
 	struct nvme_rdma_ctrl *ctrl = container_of(to_delayed_work(work),
 			struct nvme_rdma_ctrl, reconnect_work);
+	int ret;
 
 	++ctrl->ctrl.nr_reconnects;
 
-	if (nvme_rdma_setup_ctrl(ctrl, false))
+	ret = nvme_rdma_setup_ctrl(ctrl, false);
+	if (ret)
 		goto requeue;
 
 	dev_info(ctrl->ctrl.device, "Successfully reconnected (%d attempts)\n",
@@ -1114,7 +1120,7 @@ static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work)
 requeue:
 	dev_info(ctrl->ctrl.device, "Failed reconnect attempt %d\n",
 			ctrl->ctrl.nr_reconnects);
-	nvme_rdma_reconnect_or_remove(ctrl);
+	nvme_rdma_reconnect_or_remove(ctrl, ret);
 }
 
 static void nvme_rdma_error_recovery_work(struct work_struct *work)
@@ -1139,7 +1145,7 @@ static void nvme_rdma_error_recovery_work(struct work_struct *work)
 		return;
 	}
 
-	nvme_rdma_reconnect_or_remove(ctrl);
+	nvme_rdma_reconnect_or_remove(ctrl, -ENOTCONN);
 }
 
 static void nvme_rdma_error_recovery(struct nvme_rdma_ctrl *ctrl)
@@ -2163,6 +2169,7 @@ static void nvme_rdma_reset_ctrl_work(struct work_struct *work)
 {
 	struct nvme_rdma_ctrl *ctrl =
 		container_of(work, struct nvme_rdma_ctrl, ctrl.reset_work);
+	int ret;
 
 	nvme_stop_ctrl(&ctrl->ctrl);
 	nvme_rdma_shutdown_ctrl(ctrl, false);
@@ -2173,14 +2180,15 @@ static void nvme_rdma_reset_ctrl_work(struct work_struct *work)
 		return;
 	}
 
-	if (nvme_rdma_setup_ctrl(ctrl, false))
+	ret = nvme_rdma_setup_ctrl(ctrl, false);
+	if (ret)
 		goto out_fail;
 
 	return;
 
 out_fail:
 	++ctrl->ctrl.nr_reconnects;
-	nvme_rdma_reconnect_or_remove(ctrl);
+	nvme_rdma_reconnect_or_remove(ctrl, ret);
 }
 
 static const struct nvme_ctrl_ops nvme_rdma_ctrl_ops = {
-- 
2.44.0