[PATCH AUTOSEL 5.4 37/40] nvme-tcp: fix controller reset hang during traffic

[PATCH AUTOSEL 5.4 37/40] nvme-tcp: fix controller reset hang during traffic

[Sasha Levin]

From: Sagi Grimberg <sagi@grimberg.me>

[ Upstream commit 2875b0aecabe2f081a8432e2bc85b85df0529490 ]

commit fe35ec58f0d3 ("block: update hctx map when use multiple maps")
exposed an issue where we may hang trying to wait for queue freeze
during I/O. We call blk_mq_update_nr_hw_queues which in case of multiple
queue maps (which we have now for default/read/poll) is attempting to
freeze the queue. However we never started queue freeze when starting the
reset, which means that we have inflight pending requests that entered the
queue that we will not complete once the queue is quiesced.

So start a freeze before we quiesce the queue, and unfreeze the queue
after we successfully connected the I/O queues (and make sure to call
blk_mq_update_nr_hw_queues only after we are sure that the queue was
already frozen).

This follows to how the pci driver handles resets.

Fixes: fe35ec58f0d3 ("block: update hctx map when use multiple maps")
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/nvme/host/tcp.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c
index 53e113a18a549..0166ff0e4738e 100644
--- a/drivers/nvme/host/tcp.c
+++ b/drivers/nvme/host/tcp.c
@@ -1684,15 +1684,20 @@ static int nvme_tcp_configure_io_queues(struct nvme_ctrl *ctrl, bool new)
 			ret = PTR_ERR(ctrl->connect_q);
 			goto out_free_tag_set;
 		}
-	} else {
-		blk_mq_update_nr_hw_queues(ctrl->tagset,
-			ctrl->queue_count - 1);
 	}
 
 	ret = nvme_tcp_start_io_queues(ctrl);
 	if (ret)
 		goto out_cleanup_connect_q;
 
+	if (!new) {
+		nvme_start_queues(ctrl);
+		nvme_wait_freeze(ctrl);
+		blk_mq_update_nr_hw_queues(ctrl->tagset,
+			ctrl->queue_count - 1);
+		nvme_unfreeze(ctrl);
+	}
+
 	return 0;
 
 out_cleanup_connect_q:
@@ -1797,6 +1802,7 @@ static void nvme_tcp_teardown_io_queues(struct nvme_ctrl *ctrl,
 {
 	if (ctrl->queue_count <= 1)
 		return;
+	nvme_start_freeze(ctrl);
 	nvme_stop_queues(ctrl);
 	nvme_tcp_stop_io_queues(ctrl);
 	if (ctrl->tagset) {
-- 
2.25.1


_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme

[PATCH AUTOSEL 5.4 38/40] nvme-rdma: fix controller reset hang during traffic

[Sasha Levin]

From: Sagi Grimberg <sagi@grimberg.me>

[ Upstream commit 9f98772ba307dd89a3d17dc2589f213d3972fc64 ]

commit fe35ec58f0d3 ("block: update hctx map when use multiple maps")
exposed an issue where we may hang trying to wait for queue freeze
during I/O. We call blk_mq_update_nr_hw_queues which in case of multiple
queue maps (which we have now for default/read/poll) is attempting to
freeze the queue. However we never started queue freeze when starting the
reset, which means that we have inflight pending requests that entered the
queue that we will not complete once the queue is quiesced.

So start a freeze before we quiesce the queue, and unfreeze the queue
after we successfully connected the I/O queues (and make sure to call
blk_mq_update_nr_hw_queues only after we are sure that the queue was
already frozen).

This follows to how the pci driver handles resets.

Fixes: fe35ec58f0d3 ("block: update hctx map when use multiple maps")
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/nvme/host/rdma.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index cd0d499781908..d0336545e1fe0 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -890,15 +890,20 @@ static int nvme_rdma_configure_io_queues(struct nvme_rdma_ctrl *ctrl, bool new)
 			ret = PTR_ERR(ctrl->ctrl.connect_q);
 			goto out_free_tag_set;
 		}
-	} else {
-		blk_mq_update_nr_hw_queues(&ctrl->tag_set,
-			ctrl->ctrl.queue_count - 1);
 	}
 
 	ret = nvme_rdma_start_io_queues(ctrl);
 	if (ret)
 		goto out_cleanup_connect_q;
 
+	if (!new) {
+		nvme_start_queues(&ctrl->ctrl);
+		nvme_wait_freeze(&ctrl->ctrl);
+		blk_mq_update_nr_hw_queues(ctrl->ctrl.tagset,
+			ctrl->ctrl.queue_count - 1);
+		nvme_unfreeze(&ctrl->ctrl);
+	}
+
 	return 0;
 
 out_cleanup_connect_q:
@@ -931,6 +936,7 @@ static void nvme_rdma_teardown_io_queues(struct nvme_rdma_ctrl *ctrl,
 		bool remove)
 {
 	if (ctrl->ctrl.queue_count > 1) {
+		nvme_start_freeze(&ctrl->ctrl);
 		nvme_stop_queues(&ctrl->ctrl);
 		nvme_rdma_stop_io_queues(ctrl);
 		if (ctrl->ctrl.tagset) {
-- 
2.25.1


_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme

[PATCH AUTOSEL 5.4 39/40] nvme-multipath: fix logic for non-optimized paths

[Sasha Levin]

From: Martin Wilck <mwilck@suse.com>

[ Upstream commit 3f6e3246db0e6f92e784965d9d0edb8abe6c6b74 ]

Handle the special case where we have exactly one optimized path,
which we should keep using in this case.

Fixes: 75c10e732724 ("nvme-multipath: round-robin I/O policy")
Signed off-by: Martin Wilck <mwilck@suse.com>
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/nvme/host/multipath.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c
index 5433aa2f76017..38d25d7c6bca3 100644
--- a/drivers/nvme/host/multipath.c
+++ b/drivers/nvme/host/multipath.c
@@ -249,6 +249,12 @@ static struct nvme_ns *nvme_round_robin_path(struct nvme_ns_head *head,
 			fallback = ns;
 	}
 
+	/* No optimized path found, re-check the current path */
+	if (!nvme_path_is_disabled(old) &&
+	    old->ana_state == NVME_ANA_OPTIMIZED) {
+		found = old;
+		goto out;
+	}
 	if (!fallback)
 		return NULL;
 	found = fallback;
-- 
2.25.1


_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme

[PATCH AUTOSEL 5.4 40/40] nvme-multipath: do not fall back to __nvme_find_path() for non-optimized paths

[Sasha Levin]

From: Hannes Reinecke <hare@suse.de>

[ Upstream commit fbd6a42d8932e172921c7de10468a2e12c34846b ]

When nvme_round_robin_path() finds a valid namespace we should be using it;
falling back to __nvme_find_path() for non-optimized paths will cause the
result from nvme_round_robin_path() to be ignored for non-optimized paths.

Fixes: 75c10e732724 ("nvme-multipath: round-robin I/O policy")
Signed-off-by: Martin Wilck <mwilck@suse.com>
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/nvme/host/multipath.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c
index 38d25d7c6bca3..484aad0d0c9c6 100644
--- a/drivers/nvme/host/multipath.c
+++ b/drivers/nvme/host/multipath.c
@@ -275,10 +275,13 @@ inline struct nvme_ns *nvme_find_path(struct nvme_ns_head *head)
 	struct nvme_ns *ns;
 
 	ns = srcu_dereference(head->current_path[node], &head->srcu);
-	if (READ_ONCE(head->subsys->iopolicy) == NVME_IOPOLICY_RR && ns)
-		ns = nvme_round_robin_path(head, node, ns);
-	if (unlikely(!ns || !nvme_path_is_optimized(ns)))
-		ns = __nvme_find_path(head, node);
+	if (unlikely(!ns))
+		return __nvme_find_path(head, node);
+
+	if (READ_ONCE(head->subsys->iopolicy) == NVME_IOPOLICY_RR)
+		return nvme_round_robin_path(head, node, ns);
+	if (unlikely(!nvme_path_is_optimized(ns)))
+		return __nvme_find_path(head, node);
 	return ns;
 }
 
-- 
2.25.1


_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme