Skip to contentSkip to conversation

nvme: revalidate paths during rescan

PATCH Applied Daniel Wagner 3 participants 7 messages Aug 11, 2021 – Aug 24, 2021 NVMe

v3 of 4 Applied Latest is v4 →

lore
Share & tools
lore
#0 lore raw

PATCH v3 1/1 nvme: revalidate paths during rescan

From: Hannes Reinecke <[email protected]>

When triggering a rescan due to a namespace resize we will be receiving AENs on every controller, triggering a rescan of all attached namespaces. If multipath is active only the current path and the ns_head disk will be updated, the other paths will still refer to the old size until AENs for the remaining controllers are received.

If I/O comes in before that it might be routed to one of the old paths, triggering an I/O failure with 'access beyond end of device'. With this patch the old paths are skipped from multipath path selection until the controller serving these paths has been rescanned.

[dwagner: - introduce NVME_NS_READY flag instead of NVME_NS_INVALIDATE
- use 'revalidate' instead of 'invalidate' which follows the zoned device code path.]

---
v3:
- Renamed nvme_mpath_invalidated_paths to nvme_mpath_revalidate_paths()
- Replaced NVME_NS_INVALIDATE with NVME_NS_READY
v2:
- https://lore.kernel.org/linux-nvme/[email protected]/
- removed churn from failed rebase.
v1:
- https://lore.kernel.org/linux-nvme/[email protected]/

drivers/nvme/host/core.c | 3 +++

 drivers/nvme/host/multipath.c | 17 ++++++++++++++++-
 drivers/nvme/host/nvme.h      |  5 +++++
 3 files changed, 24 insertions(+), 1 deletion(-)
drivers/nvme/host/core.c +3 -0
index 2f0cbaba12ac..54aafde4f556 100644
@@ -1878,6 +1878,7 @@ static int nvme_update_ns_info(struct nvme_ns *ns, struct nvme_id_ns *id)
 			goto out_unfreeze;
 	}
 
+	set_bit(NVME_NS_READY, &ns->flags);
 	blk_mq_unfreeze_queue(ns->disk->queue);
 
 	if (blk_queue_is_zoned(ns->queue)) {
@@ -1889,6 +1890,7 @@ static int nvme_update_ns_info(struct nvme_ns *ns, struct nvme_id_ns *id)
 	if (nvme_ns_head_multipath(ns->head)) {
 		blk_mq_freeze_queue(ns->head->disk->queue);
 		nvme_update_disk_info(ns->head->disk, ns, id);
+		nvme_mpath_revalidate_paths(ns);
 		blk_stack_limits(&ns->head->disk->queue->limits,
 				 &ns->queue->limits, 0);
 		blk_queue_update_readahead(ns->head->disk->queue);
@@ -3816,6 +3818,7 @@ static void nvme_ns_remove(struct nvme_ns *ns)
 	if (test_and_set_bit(NVME_NS_REMOVING, &ns->flags))
 		return;
 
+	clear_bit(NVME_NS_READY, &ns->flags);
 	set_capacity(ns->disk, 0);
 	nvme_fault_inject_fini(&ns->fault_inject);
 
drivers/nvme/host/multipath.c +16 -1
index 3f32c5e86bfc..d390f14b8bb6 100644
@@ -147,6 +147,21 @@ void nvme_mpath_clear_ctrl_paths(struct nvme_ctrl *ctrl)
 	mutex_unlock(&ctrl->scan_lock);
 }
 
+void nvme_mpath_revalidate_paths(struct nvme_ns *ns)
+{
+	struct nvme_ns_head *head = ns->head;
+	sector_t capacity = get_capacity(head->disk);
+	int node;
+
+	for_each_node(node)
+		rcu_assign_pointer(head->current_path[node], NULL);
+
+	list_for_each_entry_rcu(ns, &head->list, siblings) {
+		if (capacity != get_capacity(ns->disk))
+			clear_bit(NVME_NS_READY, &ns->flags);
+	}
+}
+
 static bool nvme_path_is_disabled(struct nvme_ns *ns)
 {
 	/*
@@ -158,7 +173,7 @@ static bool nvme_path_is_disabled(struct nvme_ns *ns)
 	    ns->ctrl->state != NVME_CTRL_DELETING)
 		return true;
 	if (test_bit(NVME_NS_ANA_PENDING, &ns->flags) ||
-	    test_bit(NVME_NS_REMOVING, &ns->flags))
+	    !test_bit(NVME_NS_READY, &ns->flags))
 		return true;
 	return false;
 }
drivers/nvme/host/nvme.h +5 -0
index 26511794629b..6c67dac79168 100644
@@ -468,6 +468,7 @@ struct nvme_ns {
 #define NVME_NS_DEAD     	1
 #define NVME_NS_ANA_PENDING	2
 #define NVME_NS_FORCE_RO	3
+#define NVME_NS_READY		4
 
 	struct cdev		cdev;
 	struct device		cdev_device;
@@ -760,6 +761,7 @@ void nvme_mpath_init_ctrl(struct nvme_ctrl *ctrl);
 void nvme_mpath_uninit(struct nvme_ctrl *ctrl);
 void nvme_mpath_stop(struct nvme_ctrl *ctrl);
 bool nvme_mpath_clear_current_path(struct nvme_ns *ns);
+void nvme_mpath_revalidate_paths(struct nvme_ns *ns);
 void nvme_mpath_clear_ctrl_paths(struct nvme_ctrl *ctrl);
 void nvme_mpath_shutdown_disk(struct nvme_ns_head *head);
 
@@ -807,6 +809,9 @@ static inline bool nvme_mpath_clear_current_path(struct nvme_ns *ns)
 {
 	return false;
 }
+static inline void nvme_mpath_revalidate_paths(struct nvme_ns *ns)
+{
+}
 static inline void nvme_mpath_clear_ctrl_paths(struct nvme_ctrl *ctrl)
 {
 }
Signature · 1 line
2.29.2
#1 lore raw
Quoted 8 lines · Daniel Wagner

From: Hannes Reinecke <[email protected]>

When triggering a rescan due to a namespace resize we will be receiving AENs on every controller, triggering a rescan of all attached namespaces. If multipath is active only the current path and the ns_head disk will be updated, the other paths will still refer to the old size until AENs for the remaining controllers are received.

Quoted 4 lines

If I/O comes in before that it might be routed to one of the old paths, triggering an I/O failure with 'access beyond end of device'. With this patch the old paths are skipped from multipath path selection until the controller serving these paths has been rescanned.

ping

maintainer
#3 lore raw
In reply to Daniel Wagner
Quoted 73 lines · Daniel Wagner

From: Hannes Reinecke <[email protected]>

When triggering a rescan due to a namespace resize we will be receiving AENs on every controller, triggering a rescan of all attached namespaces. If multipath is active only the current path and the ns_head disk will be updated, the other paths will still refer to the old size until AENs for the remaining controllers are received.

If I/O comes in before that it might be routed to one of the old paths, triggering an I/O failure with 'access beyond end of device'. With this patch the old paths are skipped from multipath path selection until the controller serving these paths has been rescanned.

[dwagner: - introduce NVME_NS_READY flag instead of NVME_NS_INVALIDATE
- use 'revalidate' instead of 'invalidate' which follows the zoned device code path.]

---
v3:
- Renamed nvme_mpath_invalidated_paths to nvme_mpath_revalidate_paths()
- Replaced NVME_NS_INVALIDATE with NVME_NS_READY
v2:
- https://lore.kernel.org/linux-nvme/[email protected]/
- removed churn from failed rebase.
v1:
- https://lore.kernel.org/linux-nvme/[email protected]/

drivers/nvme/host/core.c | 3 +++

  drivers/nvme/host/multipath.c | 17 ++++++++++++++++-
  drivers/nvme/host/nvme.h      |  5 +++++
  3 files changed, 24 insertions(+), 1 deletion(-)
drivers/nvme/host/core.c +3 -0
index 2f0cbaba12ac..54aafde4f556 100644
@@ -1878,6 +1878,7 @@ static int nvme_update_ns_info(struct nvme_ns *ns, struct nvme_id_ns *id)
  			goto out_unfreeze;
  	}
 
+	set_bit(NVME_NS_READY, &ns->flags);
  	blk_mq_unfreeze_queue(ns->disk->queue);
 
  	if (blk_queue_is_zoned(ns->queue)) {
@@ -1889,6 +1890,7 @@ static int nvme_update_ns_info(struct nvme_ns *ns, struct nvme_id_ns *id)
  	if (nvme_ns_head_multipath(ns->head)) {
  		blk_mq_freeze_queue(ns->head->disk->queue);
  		nvme_update_disk_info(ns->head->disk, ns, id);
+		nvme_mpath_revalidate_paths(ns);
  		blk_stack_limits(&ns->head->disk->queue->limits,
  				 &ns->queue->limits, 0);
  		blk_queue_update_readahead(ns->head->disk->queue);
@@ -3816,6 +3818,7 @@ static void nvme_ns_remove(struct nvme_ns *ns)
  	if (test_and_set_bit(NVME_NS_REMOVING, &ns->flags))
  		return;
 
+	clear_bit(NVME_NS_READY, &ns->flags);
  	set_capacity(ns->disk, 0);
  	nvme_fault_inject_fini(&ns->fault_inject);
 
drivers/nvme/host/multipath.c +3 -0
index 3f32c5e86bfc..d390f14b8bb6 100644
@@ -147,6 +147,21 @@ void nvme_mpath_clear_ctrl_paths(struct nvme_ctrl *ctrl)
  	mutex_unlock(&ctrl->scan_lock);
  }
 
+void nvme_mpath_revalidate_paths(struct nvme_ns *ns)
+{
+	struct nvme_ns_head *head = ns->head;

drivers/nvme/host/multipath.c void nvme_mpath_clear_ctrl_paths(struct nvme_ctrl *ctrl) · line 162

Quoted 10 lines
drivers/nvme/host/multipath.c +10 -0
+	sector_t capacity = get_capacity(head->disk);
+	int node;
+
+	for_each_node(node)
+		rcu_assign_pointer(head->current_path[node], NULL);
+
+	list_for_each_entry_rcu(ns, &head->list, siblings) {
+		if (capacity != get_capacity(ns->disk))
+			clear_bit(NVME_NS_READY, &ns->flags);
+	}

Shouldn't the null setting to current_path come after we clear NVME_NS_READY on the ns? Otherwise we may still submit and current_path will be populated with the ns again...

#4 lore raw
Quoted 10 lines · Sagi Grimberg
Quoted 10 lines
drivers/nvme/host/multipath.c +10 -0
+void nvme_mpath_revalidate_paths(struct nvme_ns *ns)
+{
+	struct nvme_ns_head *head = ns->head;
+	sector_t capacity = get_capacity(head->disk);
+	int node;
+
+	for_each_node(node)
+		rcu_assign_pointer(head->current_path[node], NULL);
+
+	list_for_each_entry_rcu(ns, &head->list, siblings) {
Quoted 8 lines

drivers/nvme/host/multipath.c void nvme_mpath_clear_ctrl_paths(struct nvme_ctrl *ctrl) · line 162

Quoted 3 lines
drivers/nvme/host/multipath.c +3 -0
+		if (capacity != get_capacity(ns->disk))
+			clear_bit(NVME_NS_READY, &ns->flags);
+	}

Shouldn't the null setting to current_path come after we clear NVME_NS_READY on the ns? Otherwise we may still submit and current_path will be populated with the ns again...

Ahh, I got it this time! Yes, you are right. I think Christoph has dropped this patch from the nvme-5.15 queue anyway. I'll resend a new version with the order changed.

Daniel

#5 lore raw
Quoted 20 lines · Daniel Wagner
Quoted 10 lines · Sagi Grimberg
Quoted 10 lines
drivers/nvme/host/multipath.c +10 -0
+void nvme_mpath_revalidate_paths(struct nvme_ns *ns)
+{
+	struct nvme_ns_head *head = ns->head;
+	sector_t capacity = get_capacity(head->disk);
+	int node;
+
+	for_each_node(node)
+		rcu_assign_pointer(head->current_path[node], NULL);
+
+	list_for_each_entry_rcu(ns, &head->list, siblings) {
Quoted 8 lines

drivers/nvme/host/multipath.c void nvme_mpath_clear_ctrl_paths(struct nvme_ctrl *ctrl) · line 162

Quoted 3 lines
drivers/nvme/host/multipath.c +3 -0
+		if (capacity != get_capacity(ns->disk))
+			clear_bit(NVME_NS_READY, &ns->flags);
+	}

Shouldn't the null setting to current_path come after we clear NVME_NS_READY on the ns? Otherwise we may still submit and current_path will be populated with the ns again...

Quoted 3 lines

Ahh, I got it this time! Yes, you are right. I think Christoph has dropped this patch from the nvme-5.15 queue anyway. I'll resend a new version with the order changed.

I am blind. The patch is queued up in nvme-5.15.

@Christoph, do you want me to send fix on top of this patch or do you want an updated version of this patch?

#6 lore raw
Quoted 4 lines · Daniel Wagner

I am blind. The patch is queued up in nvme-5.15.

@Christoph, do you want me to send fix on top of this patch or do you want an updated version of this patch?

Either way is fine.