[PATCH 00/12] Few more patches from the centralization set

[PATCH 00/12] Few more patches from the centralization set

[Sagi Grimberg]

Mostly nvme-rdma. I'm trying to work my way incrementally so I can
start moving logic to nvme core in steps.

In the next set, I'll start detaching transport specific from
the reset, delete and error recovery flows.

In the set after that, I'll start adding the new transport ctrl->ops
and move some logic to the core.

Currently the new ctrl ops I'm shooting for are:
* ->alloc_admin_queue()
* ->free_admin_queue()
* ->start_admin_queue()
* ->stop_admin_queue()
* ->alloc_io_queues()	/* admin/io separation is due to how pci releases queues */
* ->free_io_queues()
* ->start_io_queues()
* ->stop_io_queues()
* ->alloc_tagset()
* ->free_tagset()
* ->transport_enable()	/* for pcie mostly */
* ->verify_ctrl()	/* for post features sanity checks */
* ->fast_fail_io()	/* for fc aborts handling */

Feedback and discussion is welcome.

Sagi Grimberg (12):
  nvme-rdma: move nvme_rdma_configure_admin_queue code location
  nvme: Add admin_tagset pointer to nvme_ctrl
  nvme-rdma: move tagset allocation to a dedicated routine
  nvme-rdma: disable the controller on resets
  nvme-rdma: don't free tagset on resets
  nvme-rdma: reuse configure/destroy_admin_queue
  nvme-rdma: introduce configure/destroy io queues
  nvme-rdma: stop queues instead of simply flipping their state
  nvme-rdma: rename nvme_rdma_init_queue to nvme_rdma_alloc_queue
  nvme-rdma: introduce nvme_rdma_start_queue
  nvme-rdma: cleanup error path in controller reset
  nvme-rdma: call ops->reg_read64 instead of nvmf_reg_read64

 drivers/nvme/host/fc.c     |   1 +
 drivers/nvme/host/nvme.h   |   1 +
 drivers/nvme/host/pci.c    |   1 +
 drivers/nvme/host/rdma.c   | 545 +++++++++++++++++++++++----------------------
 drivers/nvme/target/loop.c |   1 +
 5 files changed, 284 insertions(+), 265 deletions(-)

-- 
2.7.4

[PATCH 01/12] nvme-rdma: move nvme_rdma_configure_admin_queue code location

[Sagi Grimberg]

We will call it from other places so avoid having to
forward declare it. Also move it next to
nvme_rdma_destroy_admin_queue.

Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
---
 drivers/nvme/host/rdma.c | 185 ++++++++++++++++++++++++-----------------------
 1 file changed, 94 insertions(+), 91 deletions(-)

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index 4e25acc54e09..65e35824ceb6 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -152,6 +152,9 @@ static int nvme_rdma_cm_handler(struct rdma_cm_id *cm_id,
 		struct rdma_cm_event *event);
 static void nvme_rdma_recv_done(struct ib_cq *cq, struct ib_wc *wc);
 
+static const struct blk_mq_ops nvme_rdma_mq_ops;
+static const struct blk_mq_ops nvme_rdma_admin_mq_ops;
+
 /* XXX: really should move to a generic header sooner or later.. */
 static inline void put_unaligned_le24(u32 val, u8 *p)
 {
@@ -662,6 +665,97 @@ static void nvme_rdma_destroy_admin_queue(struct nvme_rdma_ctrl *ctrl)
 	nvme_rdma_dev_put(ctrl->device);
 }
 
+static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl)
+{
+	int error;
+
+	error = nvme_rdma_init_queue(ctrl, 0, NVME_AQ_DEPTH);
+	if (error)
+		return error;
+
+	ctrl->device = ctrl->queues[0].device;
+
+	/*
+	 * We need a reference on the device as long as the tag_set is alive,
+	 * as the MRs in the request structures need a valid ib_device.
+	 */
+	error = -EINVAL;
+	if (!nvme_rdma_dev_get(ctrl->device))
+		goto out_free_queue;
+
+	ctrl->max_fr_pages = min_t(u32, NVME_RDMA_MAX_SEGMENTS,
+		ctrl->device->dev->attrs.max_fast_reg_page_list_len);
+
+	memset(&ctrl->admin_tag_set, 0, sizeof(ctrl->admin_tag_set));
+	ctrl->admin_tag_set.ops = &nvme_rdma_admin_mq_ops;
+	ctrl->admin_tag_set.queue_depth = NVME_RDMA_AQ_BLKMQ_DEPTH;
+	ctrl->admin_tag_set.reserved_tags = 2; /* connect + keep-alive */
+	ctrl->admin_tag_set.numa_node = NUMA_NO_NODE;
+	ctrl->admin_tag_set.cmd_size = sizeof(struct nvme_rdma_request) +
+		SG_CHUNK_SIZE * sizeof(struct scatterlist);
+	ctrl->admin_tag_set.driver_data = ctrl;
+	ctrl->admin_tag_set.nr_hw_queues = 1;
+	ctrl->admin_tag_set.timeout = ADMIN_TIMEOUT;
+
+	error = blk_mq_alloc_tag_set(&ctrl->admin_tag_set);
+	if (error)
+		goto out_put_dev;
+
+	ctrl->ctrl.admin_q = blk_mq_init_queue(&ctrl->admin_tag_set);
+	if (IS_ERR(ctrl->ctrl.admin_q)) {
+		error = PTR_ERR(ctrl->ctrl.admin_q);
+		goto out_free_tagset;
+	}
+
+	error = nvmf_connect_admin_queue(&ctrl->ctrl);
+	if (error)
+		goto out_cleanup_queue;
+
+	set_bit(NVME_RDMA_Q_LIVE, &ctrl->queues[0].flags);
+
+	error = nvmf_reg_read64(&ctrl->ctrl, NVME_REG_CAP,
+			&ctrl->ctrl.cap);
+	if (error) {
+		dev_err(ctrl->ctrl.device,
+			"prop_get NVME_REG_CAP failed\n");
+		goto out_cleanup_queue;
+	}
+
+	ctrl->ctrl.sqsize =
+		min_t(int, NVME_CAP_MQES(ctrl->ctrl.cap), ctrl->ctrl.sqsize);
+
+	error = nvme_enable_ctrl(&ctrl->ctrl, ctrl->ctrl.cap);
+	if (error)
+		goto out_cleanup_queue;
+
+	ctrl->ctrl.max_hw_sectors =
+		(ctrl->max_fr_pages - 1) << (PAGE_SHIFT - 9);
+
+	error = nvme_init_identify(&ctrl->ctrl);
+	if (error)
+		goto out_cleanup_queue;
+
+	error = nvme_rdma_alloc_qe(ctrl->queues[0].device->dev,
+			&ctrl->async_event_sqe, sizeof(struct nvme_command),
+			DMA_TO_DEVICE);
+	if (error)
+		goto out_cleanup_queue;
+
+	return 0;
+
+out_cleanup_queue:
+	blk_cleanup_queue(ctrl->ctrl.admin_q);
+out_free_tagset:
+	/* disconnect and drain the queue before freeing the tagset */
+	nvme_rdma_stop_queue(&ctrl->queues[0]);
+	blk_mq_free_tag_set(&ctrl->admin_tag_set);
+out_put_dev:
+	nvme_rdma_dev_put(ctrl->device);
+out_free_queue:
+	nvme_rdma_free_queue(&ctrl->queues[0]);
+	return error;
+}
+
 static void nvme_rdma_free_ctrl(struct nvme_ctrl *nctrl)
 {
 	struct nvme_rdma_ctrl *ctrl = to_rdma_ctrl(nctrl);
@@ -1534,97 +1628,6 @@ static const struct blk_mq_ops nvme_rdma_admin_mq_ops = {
 	.timeout	= nvme_rdma_timeout,
 };
 
-static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl)
-{
-	int error;
-
-	error = nvme_rdma_init_queue(ctrl, 0, NVME_AQ_DEPTH);
-	if (error)
-		return error;
-
-	ctrl->device = ctrl->queues[0].device;
-
-	/*
-	 * We need a reference on the device as long as the tag_set is alive,
-	 * as the MRs in the request structures need a valid ib_device.
-	 */
-	error = -EINVAL;
-	if (!nvme_rdma_dev_get(ctrl->device))
-		goto out_free_queue;
-
-	ctrl->max_fr_pages = min_t(u32, NVME_RDMA_MAX_SEGMENTS,
-		ctrl->device->dev->attrs.max_fast_reg_page_list_len);
-
-	memset(&ctrl->admin_tag_set, 0, sizeof(ctrl->admin_tag_set));
-	ctrl->admin_tag_set.ops = &nvme_rdma_admin_mq_ops;
-	ctrl->admin_tag_set.queue_depth = NVME_RDMA_AQ_BLKMQ_DEPTH;
-	ctrl->admin_tag_set.reserved_tags = 2; /* connect + keep-alive */
-	ctrl->admin_tag_set.numa_node = NUMA_NO_NODE;
-	ctrl->admin_tag_set.cmd_size = sizeof(struct nvme_rdma_request) +
-		SG_CHUNK_SIZE * sizeof(struct scatterlist);
-	ctrl->admin_tag_set.driver_data = ctrl;
-	ctrl->admin_tag_set.nr_hw_queues = 1;
-	ctrl->admin_tag_set.timeout = ADMIN_TIMEOUT;
-
-	error = blk_mq_alloc_tag_set(&ctrl->admin_tag_set);
-	if (error)
-		goto out_put_dev;
-
-	ctrl->ctrl.admin_q = blk_mq_init_queue(&ctrl->admin_tag_set);
-	if (IS_ERR(ctrl->ctrl.admin_q)) {
-		error = PTR_ERR(ctrl->ctrl.admin_q);
-		goto out_free_tagset;
-	}
-
-	error = nvmf_connect_admin_queue(&ctrl->ctrl);
-	if (error)
-		goto out_cleanup_queue;
-
-	set_bit(NVME_RDMA_Q_LIVE, &ctrl->queues[0].flags);
-
-	error = nvmf_reg_read64(&ctrl->ctrl, NVME_REG_CAP,
-			&ctrl->ctrl.cap);
-	if (error) {
-		dev_err(ctrl->ctrl.device,
-			"prop_get NVME_REG_CAP failed\n");
-		goto out_cleanup_queue;
-	}
-
-	ctrl->ctrl.sqsize =
-		min_t(int, NVME_CAP_MQES(ctrl->ctrl.cap), ctrl->ctrl.sqsize);
-
-	error = nvme_enable_ctrl(&ctrl->ctrl, ctrl->ctrl.cap);
-	if (error)
-		goto out_cleanup_queue;
-
-	ctrl->ctrl.max_hw_sectors =
-		(ctrl->max_fr_pages - 1) << (PAGE_SHIFT - 9);
-
-	error = nvme_init_identify(&ctrl->ctrl);
-	if (error)
-		goto out_cleanup_queue;
-
-	error = nvme_rdma_alloc_qe(ctrl->queues[0].device->dev,
-			&ctrl->async_event_sqe, sizeof(struct nvme_command),
-			DMA_TO_DEVICE);
-	if (error)
-		goto out_cleanup_queue;
-
-	return 0;
-
-out_cleanup_queue:
-	blk_cleanup_queue(ctrl->ctrl.admin_q);
-out_free_tagset:
-	/* disconnect and drain the queue before freeing the tagset */
-	nvme_rdma_stop_queue(&ctrl->queues[0]);
-	blk_mq_free_tag_set(&ctrl->admin_tag_set);
-out_put_dev:
-	nvme_rdma_dev_put(ctrl->device);
-out_free_queue:
-	nvme_rdma_free_queue(&ctrl->queues[0]);
-	return error;
-}
-
 static void nvme_rdma_shutdown_ctrl(struct nvme_rdma_ctrl *ctrl)
 {
 	cancel_work_sync(&ctrl->err_work);
-- 
2.7.4

[PATCH 02/12] nvme: Add admin_tagset pointer to nvme_ctrl

[Sagi Grimberg]

Will be used when we centralize control flows. only
rdma for now.

Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
---
 drivers/nvme/host/fc.c     | 1 +
 drivers/nvme/host/nvme.h   | 1 +
 drivers/nvme/host/pci.c    | 1 +
 drivers/nvme/host/rdma.c   | 1 +
 drivers/nvme/target/loop.c | 1 +
 5 files changed, 5 insertions(+)

diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c
index d666ada39a9b..ac8ae67cdc07 100644
--- a/drivers/nvme/host/fc.c
+++ b/drivers/nvme/host/fc.c
@@ -2730,6 +2730,7 @@ nvme_fc_init_ctrl(struct device *dev, struct nvmf_ctrl_options *opts,
 	ret = blk_mq_alloc_tag_set(&ctrl->admin_tag_set);
 	if (ret)
 		goto out_free_queues;
+	ctrl->ctrl.admin_tagset = &ctrl->admin_tag_set;
 
 	ctrl->ctrl.admin_q = blk_mq_init_queue(&ctrl->admin_tag_set);
 	if (IS_ERR(ctrl->ctrl.admin_q)) {
diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h
index 8f2a168ddc01..0a43f3e1a52a 100644
--- a/drivers/nvme/host/nvme.h
+++ b/drivers/nvme/host/nvme.h
@@ -125,6 +125,7 @@ struct nvme_ctrl {
 	struct kref kref;
 	int instance;
 	struct blk_mq_tag_set *tagset;
+	struct blk_mq_tag_set *admin_tagset;
 	struct list_head namespaces;
 	struct mutex namespaces_mutex;
 	struct device *device;	/* char device */
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index 73fddf2c99f9..81f7ccc58d70 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -1340,6 +1340,7 @@ static int nvme_alloc_admin_tags(struct nvme_dev *dev)
 
 		if (blk_mq_alloc_tag_set(&dev->admin_tagset))
 			return -ENOMEM;
+		dev->ctrl.admin_tagset = &dev->admin_tagset;
 
 		dev->ctrl.admin_q = blk_mq_init_queue(&dev->admin_tagset);
 		if (IS_ERR(dev->ctrl.admin_q)) {
diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index 65e35824ceb6..564a15af7c34 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -700,6 +700,7 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl)
 	error = blk_mq_alloc_tag_set(&ctrl->admin_tag_set);
 	if (error)
 		goto out_put_dev;
+	ctrl->ctrl.admin_tagset = &ctrl->admin_tag_set;
 
 	ctrl->ctrl.admin_q = blk_mq_init_queue(&ctrl->admin_tag_set);
 	if (IS_ERR(ctrl->ctrl.admin_q)) {
diff --git a/drivers/nvme/target/loop.c b/drivers/nvme/target/loop.c
index 717ed7ddb2f6..92628c432926 100644
--- a/drivers/nvme/target/loop.c
+++ b/drivers/nvme/target/loop.c
@@ -375,6 +375,7 @@ static int nvme_loop_configure_admin_queue(struct nvme_loop_ctrl *ctrl)
 	error = blk_mq_alloc_tag_set(&ctrl->admin_tag_set);
 	if (error)
 		goto out_free_sq;
+	ctrl->ctrl.admin_tagset = &ctrl->admin_tag_set;
 
 	ctrl->ctrl.admin_q = blk_mq_init_queue(&ctrl->admin_tag_set);
 	if (IS_ERR(ctrl->ctrl.admin_q)) {
-- 
2.7.4

[PATCH 03/12] nvme-rdma: move tagset allocation to a dedicated routine

[Sagi Grimberg]

We always pair tagset allocation with rdma device reference
and it shares some code, centralize it with an argument if its
an admin or IO tagset.

Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
---
 drivers/nvme/host/rdma.c | 131 +++++++++++++++++++++++++++--------------------
 1 file changed, 75 insertions(+), 56 deletions(-)

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index 564a15af7c34..7ec9adef6015 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -655,14 +655,79 @@ static int nvme_rdma_init_io_queues(struct nvme_rdma_ctrl *ctrl)
 	return ret;
 }
 
+static void nvme_rdma_free_tagset(struct nvme_ctrl *nctrl, bool admin)
+{
+	struct nvme_rdma_ctrl *ctrl = to_rdma_ctrl(nctrl);
+	struct blk_mq_tag_set *set = admin ?
+			&ctrl->admin_tag_set : &ctrl->tag_set;
+
+	blk_mq_free_tag_set(set);
+	nvme_rdma_dev_put(ctrl->device);
+}
+
+static struct blk_mq_tag_set *nvme_rdma_alloc_tagset(struct nvme_ctrl *nctrl,
+		bool admin)
+{
+	struct nvme_rdma_ctrl *ctrl = to_rdma_ctrl(nctrl);
+	struct blk_mq_tag_set *set;
+	int ret;
+
+	if (admin) {
+		set = &ctrl->admin_tag_set;
+		memset(set, 0, sizeof(*set));
+		set->ops = &nvme_rdma_admin_mq_ops;
+		set->queue_depth = NVME_RDMA_AQ_BLKMQ_DEPTH;
+		set->reserved_tags = 2; /* connect + keep-alive */
+		set->numa_node = NUMA_NO_NODE;
+		set->cmd_size = sizeof(struct nvme_rdma_request) +
+			SG_CHUNK_SIZE * sizeof(struct scatterlist);
+		set->driver_data = ctrl;
+		set->nr_hw_queues = 1;
+		set->timeout = ADMIN_TIMEOUT;
+	} else {
+		set = &ctrl->tag_set;
+		memset(set, 0, sizeof(*set));
+		set->ops = &nvme_rdma_mq_ops;
+		set->queue_depth = nctrl->opts->queue_size;
+		set->reserved_tags = 1; /* fabric connect */
+		set->numa_node = NUMA_NO_NODE;
+		set->flags = BLK_MQ_F_SHOULD_MERGE;
+		set->cmd_size = sizeof(struct nvme_rdma_request) +
+			SG_CHUNK_SIZE * sizeof(struct scatterlist);
+		set->driver_data = ctrl;
+		set->nr_hw_queues = nctrl->queue_count - 1;
+		set->timeout = NVME_IO_TIMEOUT;
+	}
+
+	ret = blk_mq_alloc_tag_set(set);
+	if (ret)
+		goto out;
+
+	/*
+	 * We need a reference on the device as long as the tag_set is alive,
+	 * as the MRs in the request structures need a valid ib_device.
+	 */
+	ret = nvme_rdma_dev_get(ctrl->device);
+	if (!ret) {
+		ret = -EINVAL;
+		goto out_free_tagset;
+	}
+
+	return set;
+
+out_free_tagset:
+	blk_mq_free_tag_set(set);
+out:
+	return ERR_PTR(ret);
+}
+
 static void nvme_rdma_destroy_admin_queue(struct nvme_rdma_ctrl *ctrl)
 {
 	nvme_rdma_free_qe(ctrl->queues[0].device->dev, &ctrl->async_event_sqe,
 			sizeof(struct nvme_command), DMA_TO_DEVICE);
 	nvme_rdma_stop_and_free_queue(&ctrl->queues[0]);
 	blk_cleanup_queue(ctrl->ctrl.admin_q);
-	blk_mq_free_tag_set(&ctrl->admin_tag_set);
-	nvme_rdma_dev_put(ctrl->device);
+	nvme_rdma_free_tagset(&ctrl->ctrl, true);
 }
 
 static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl)
@@ -675,32 +740,12 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl)
 
 	ctrl->device = ctrl->queues[0].device;
 
-	/*
-	 * We need a reference on the device as long as the tag_set is alive,
-	 * as the MRs in the request structures need a valid ib_device.
-	 */
-	error = -EINVAL;
-	if (!nvme_rdma_dev_get(ctrl->device))
-		goto out_free_queue;
-
 	ctrl->max_fr_pages = min_t(u32, NVME_RDMA_MAX_SEGMENTS,
 		ctrl->device->dev->attrs.max_fast_reg_page_list_len);
 
-	memset(&ctrl->admin_tag_set, 0, sizeof(ctrl->admin_tag_set));
-	ctrl->admin_tag_set.ops = &nvme_rdma_admin_mq_ops;
-	ctrl->admin_tag_set.queue_depth = NVME_RDMA_AQ_BLKMQ_DEPTH;
-	ctrl->admin_tag_set.reserved_tags = 2; /* connect + keep-alive */
-	ctrl->admin_tag_set.numa_node = NUMA_NO_NODE;
-	ctrl->admin_tag_set.cmd_size = sizeof(struct nvme_rdma_request) +
-		SG_CHUNK_SIZE * sizeof(struct scatterlist);
-	ctrl->admin_tag_set.driver_data = ctrl;
-	ctrl->admin_tag_set.nr_hw_queues = 1;
-	ctrl->admin_tag_set.timeout = ADMIN_TIMEOUT;
-
-	error = blk_mq_alloc_tag_set(&ctrl->admin_tag_set);
-	if (error)
-		goto out_put_dev;
-	ctrl->ctrl.admin_tagset = &ctrl->admin_tag_set;
+	ctrl->ctrl.admin_tagset = nvme_rdma_alloc_tagset(&ctrl->ctrl, true);
+	if (IS_ERR(ctrl->ctrl.admin_tagset))
+		goto out_free_queue;
 
 	ctrl->ctrl.admin_q = blk_mq_init_queue(&ctrl->admin_tag_set);
 	if (IS_ERR(ctrl->ctrl.admin_q)) {
@@ -749,9 +794,7 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl)
 out_free_tagset:
 	/* disconnect and drain the queue before freeing the tagset */
 	nvme_rdma_stop_queue(&ctrl->queues[0]);
-	blk_mq_free_tag_set(&ctrl->admin_tag_set);
-out_put_dev:
-	nvme_rdma_dev_put(ctrl->device);
+	nvme_rdma_free_tagset(&ctrl->ctrl, true);
 out_free_queue:
 	nvme_rdma_free_queue(&ctrl->queues[0]);
 	return error;
@@ -1661,8 +1704,7 @@ static void __nvme_rdma_remove_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown)
 	nvme_uninit_ctrl(&ctrl->ctrl);
 	if (ctrl->ctrl.tagset) {
 		blk_cleanup_queue(ctrl->ctrl.connect_q);
-		blk_mq_free_tag_set(&ctrl->tag_set);
-		nvme_rdma_dev_put(ctrl->device);
+		nvme_rdma_free_tagset(&ctrl->ctrl, false);
 	}
 
 	nvme_put_ctrl(&ctrl->ctrl);
@@ -1781,31 +1823,10 @@ static int nvme_rdma_create_io_queues(struct nvme_rdma_ctrl *ctrl)
 	if (ret)
 		return ret;
 
-	/*
-	 * We need a reference on the device as long as the tag_set is alive,
-	 * as the MRs in the request structures need a valid ib_device.
-	 */
-	ret = -EINVAL;
-	if (!nvme_rdma_dev_get(ctrl->device))
+	ctrl->ctrl.tagset = nvme_rdma_alloc_tagset(&ctrl->ctrl, true);
+	if (IS_ERR(ctrl->ctrl.tagset))
 		goto out_free_io_queues;
 
-	memset(&ctrl->tag_set, 0, sizeof(ctrl->tag_set));
-	ctrl->tag_set.ops = &nvme_rdma_mq_ops;
-	ctrl->tag_set.queue_depth = ctrl->ctrl.opts->queue_size;
-	ctrl->tag_set.reserved_tags = 1; /* fabric connect */
-	ctrl->tag_set.numa_node = NUMA_NO_NODE;
-	ctrl->tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
-	ctrl->tag_set.cmd_size = sizeof(struct nvme_rdma_request) +
-		SG_CHUNK_SIZE * sizeof(struct scatterlist);
-	ctrl->tag_set.driver_data = ctrl;
-	ctrl->tag_set.nr_hw_queues = ctrl->ctrl.queue_count - 1;
-	ctrl->tag_set.timeout = NVME_IO_TIMEOUT;
-
-	ret = blk_mq_alloc_tag_set(&ctrl->tag_set);
-	if (ret)
-		goto out_put_dev;
-	ctrl->ctrl.tagset = &ctrl->tag_set;
-
 	ctrl->ctrl.connect_q = blk_mq_init_queue(&ctrl->tag_set);
 	if (IS_ERR(ctrl->ctrl.connect_q)) {
 		ret = PTR_ERR(ctrl->ctrl.connect_q);
@@ -1821,9 +1842,7 @@ static int nvme_rdma_create_io_queues(struct nvme_rdma_ctrl *ctrl)
 out_cleanup_connect_q:
 	blk_cleanup_queue(ctrl->ctrl.connect_q);
 out_free_tag_set:
-	blk_mq_free_tag_set(&ctrl->tag_set);
-out_put_dev:
-	nvme_rdma_dev_put(ctrl->device);
+	nvme_rdma_free_tagset(&ctrl->ctrl, false);
 out_free_io_queues:
 	nvme_rdma_free_io_queues(ctrl);
 	return ret;
-- 
2.7.4

[PATCH 04/12] nvme-rdma: disable the controller on resets

[Sagi Grimberg]

mimic pci driver. Also like pci, some controllers might treat
shutdown more heavily.

Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
---
 drivers/nvme/host/rdma.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index 7ec9adef6015..496e5b686290 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -1672,7 +1672,7 @@ static const struct blk_mq_ops nvme_rdma_admin_mq_ops = {
 	.timeout	= nvme_rdma_timeout,
 };
 
-static void nvme_rdma_shutdown_ctrl(struct nvme_rdma_ctrl *ctrl)
+static void nvme_rdma_shutdown_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown)
 {
 	cancel_work_sync(&ctrl->err_work);
 	cancel_delayed_work_sync(&ctrl->reconnect_work);
@@ -1684,8 +1684,10 @@ static void nvme_rdma_shutdown_ctrl(struct nvme_rdma_ctrl *ctrl)
 		nvme_rdma_free_io_queues(ctrl);
 	}
 
-	if (test_bit(NVME_RDMA_Q_LIVE, &ctrl->queues[0].flags))
+	if (shutdown)
 		nvme_shutdown_ctrl(&ctrl->ctrl);
+	else
+		nvme_disable_ctrl(&ctrl->ctrl, ctrl->ctrl.cap);
 
 	blk_mq_quiesce_queue(ctrl->ctrl.admin_q);
 	blk_mq_tagset_busy_iter(&ctrl->admin_tag_set,
@@ -1699,7 +1701,7 @@ static void __nvme_rdma_remove_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown)
 	nvme_stop_ctrl(&ctrl->ctrl);
 	nvme_remove_namespaces(&ctrl->ctrl);
 	if (shutdown)
-		nvme_rdma_shutdown_ctrl(ctrl);
+		nvme_rdma_shutdown_ctrl(ctrl, shutdown);
 
 	nvme_uninit_ctrl(&ctrl->ctrl);
 	if (ctrl->ctrl.tagset) {
@@ -1763,7 +1765,7 @@ static void nvme_rdma_reset_ctrl_work(struct work_struct *work)
 	bool changed;
 
 	nvme_stop_ctrl(&ctrl->ctrl);
-	nvme_rdma_shutdown_ctrl(ctrl);
+	nvme_rdma_shutdown_ctrl(ctrl, false);
 
 	ret = nvme_rdma_configure_admin_queue(ctrl);
 	if (ret) {
-- 
2.7.4

[PATCH 05/12] nvme-rdma: don't free tagset on resets

[Sagi Grimberg]

We're not supposed to do that.

Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
---
 drivers/nvme/host/rdma.c | 48 +++++++++++++++++++++++++++++-------------------
 1 file changed, 29 insertions(+), 19 deletions(-)

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index 496e5b686290..a7cc06173ead 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -721,16 +721,20 @@ static struct blk_mq_tag_set *nvme_rdma_alloc_tagset(struct nvme_ctrl *nctrl,
 	return ERR_PTR(ret);
 }
 
-static void nvme_rdma_destroy_admin_queue(struct nvme_rdma_ctrl *ctrl)
+static void nvme_rdma_destroy_admin_queue(struct nvme_rdma_ctrl *ctrl,
+		bool remove)
 {
 	nvme_rdma_free_qe(ctrl->queues[0].device->dev, &ctrl->async_event_sqe,
 			sizeof(struct nvme_command), DMA_TO_DEVICE);
 	nvme_rdma_stop_and_free_queue(&ctrl->queues[0]);
-	blk_cleanup_queue(ctrl->ctrl.admin_q);
-	nvme_rdma_free_tagset(&ctrl->ctrl, true);
+	if (remove) {
+		blk_cleanup_queue(ctrl->ctrl.admin_q);
+		nvme_rdma_free_tagset(&ctrl->ctrl, true);
+	}
 }
 
-static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl)
+static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl,
+		bool new)
 {
 	int error;
 
@@ -743,14 +747,20 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl)
 	ctrl->max_fr_pages = min_t(u32, NVME_RDMA_MAX_SEGMENTS,
 		ctrl->device->dev->attrs.max_fast_reg_page_list_len);
 
-	ctrl->ctrl.admin_tagset = nvme_rdma_alloc_tagset(&ctrl->ctrl, true);
-	if (IS_ERR(ctrl->ctrl.admin_tagset))
-		goto out_free_queue;
+	if (new) {
+		ctrl->ctrl.admin_tagset = nvme_rdma_alloc_tagset(&ctrl->ctrl, true);
+		if (IS_ERR(ctrl->ctrl.admin_tagset))
+			goto out_free_queue;
 
-	ctrl->ctrl.admin_q = blk_mq_init_queue(&ctrl->admin_tag_set);
-	if (IS_ERR(ctrl->ctrl.admin_q)) {
-		error = PTR_ERR(ctrl->ctrl.admin_q);
-		goto out_free_tagset;
+		ctrl->ctrl.admin_q = blk_mq_init_queue(&ctrl->admin_tag_set);
+		if (IS_ERR(ctrl->ctrl.admin_q)) {
+			error = PTR_ERR(ctrl->ctrl.admin_q);
+			goto out_free_tagset;
+		}
+	} else {
+		error = blk_mq_reinit_tagset(&ctrl->admin_tag_set);
+		if (error)
+			goto out_free_queue;
 	}
 
 	error = nvmf_connect_admin_queue(&ctrl->ctrl);
@@ -790,11 +800,11 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl)
 	return 0;
 
 out_cleanup_queue:
-	blk_cleanup_queue(ctrl->ctrl.admin_q);
+	if (new)
+		blk_cleanup_queue(ctrl->ctrl.admin_q);
 out_free_tagset:
-	/* disconnect and drain the queue before freeing the tagset */
-	nvme_rdma_stop_queue(&ctrl->queues[0]);
-	nvme_rdma_free_tagset(&ctrl->ctrl, true);
+	if (new)
+		nvme_rdma_free_tagset(&ctrl->ctrl, true);
 out_free_queue:
 	nvme_rdma_free_queue(&ctrl->queues[0]);
 	return error;
@@ -1693,7 +1703,7 @@ static void nvme_rdma_shutdown_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown)
 	blk_mq_tagset_busy_iter(&ctrl->admin_tag_set,
 				nvme_cancel_request, &ctrl->ctrl);
 	blk_mq_unquiesce_queue(ctrl->ctrl.admin_q);
-	nvme_rdma_destroy_admin_queue(ctrl);
+	nvme_rdma_destroy_admin_queue(ctrl, shutdown);
 }
 
 static void __nvme_rdma_remove_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown)
@@ -1767,7 +1777,7 @@ static void nvme_rdma_reset_ctrl_work(struct work_struct *work)
 	nvme_stop_ctrl(&ctrl->ctrl);
 	nvme_rdma_shutdown_ctrl(ctrl, false);
 
-	ret = nvme_rdma_configure_admin_queue(ctrl);
+	ret = nvme_rdma_configure_admin_queue(ctrl, false);
 	if (ret) {
 		/* ctrl is already shutdown, just remove the ctrl */
 		INIT_WORK(&ctrl->delete_work, nvme_rdma_remove_ctrl_work);
@@ -1907,7 +1917,7 @@ static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev,
 	if (!ctrl->queues)
 		goto out_uninit_ctrl;
 
-	ret = nvme_rdma_configure_admin_queue(ctrl);
+	ret = nvme_rdma_configure_admin_queue(ctrl, true);
 	if (ret)
 		goto out_kfree_queues;
 
@@ -1964,7 +1974,7 @@ static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev,
 	return &ctrl->ctrl;
 
 out_remove_admin_queue:
-	nvme_rdma_destroy_admin_queue(ctrl);
+	nvme_rdma_destroy_admin_queue(ctrl, true);
 out_kfree_queues:
 	kfree(ctrl->queues);
 out_uninit_ctrl:
-- 
2.7.4

[PATCH 06/12] nvme-rdma: reuse configure/destroy_admin_queue

[Sagi Grimberg]

No need to open-code it.

Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
---
 drivers/nvme/host/rdma.c | 19 ++-----------------
 1 file changed, 2 insertions(+), 17 deletions(-)

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index a7cc06173ead..2d51badc4347 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -864,23 +864,8 @@ static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work)
 			goto requeue;
 	}
 
-	nvme_rdma_stop_and_free_queue(&ctrl->queues[0]);
-
-	ret = blk_mq_reinit_tagset(&ctrl->admin_tag_set);
-	if (ret)
-		goto requeue;
-
-	ret = nvme_rdma_init_queue(ctrl, 0, NVME_AQ_DEPTH);
-	if (ret)
-		goto requeue;
-
-	ret = nvmf_connect_admin_queue(&ctrl->ctrl);
-	if (ret)
-		goto requeue;
-
-	set_bit(NVME_RDMA_Q_LIVE, &ctrl->queues[0].flags);
-
-	ret = nvme_enable_ctrl(&ctrl->ctrl, ctrl->ctrl.cap);
+	nvme_rdma_destroy_admin_queue(ctrl, false);
+	ret = nvme_rdma_configure_admin_queue(ctrl, false);
 	if (ret)
 		goto requeue;
 
-- 
2.7.4

[PATCH 07/12] nvme-rdma: introduce configure/destroy io queues

[Sagi Grimberg]

Make a symmetrical handling with admin queue.

Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
---
 drivers/nvme/host/rdma.c | 159 ++++++++++++++++++++++++-----------------------
 1 file changed, 80 insertions(+), 79 deletions(-)

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index 2d51badc4347..685ad6acd7cd 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -561,30 +561,36 @@ static int nvme_rdma_init_queue(struct nvme_rdma_ctrl *ctrl,
 
 static void nvme_rdma_stop_queue(struct nvme_rdma_queue *queue)
 {
+	if (!test_and_clear_bit(NVME_RDMA_Q_LIVE, &queue->flags))
+		return;
+
 	rdma_disconnect(queue->cm_id);
 	ib_drain_qp(queue->qp);
 }
 
 static void nvme_rdma_free_queue(struct nvme_rdma_queue *queue)
 {
+	if (test_and_set_bit(NVME_RDMA_Q_DELETING, &queue->flags))
+		return;
+
 	nvme_rdma_destroy_queue_ib(queue);
 	rdma_destroy_id(queue->cm_id);
 }
 
-static void nvme_rdma_stop_and_free_queue(struct nvme_rdma_queue *queue)
+static void nvme_rdma_free_io_queues(struct nvme_rdma_ctrl *ctrl)
 {
-	if (test_and_set_bit(NVME_RDMA_Q_DELETING, &queue->flags))
-		return;
-	nvme_rdma_stop_queue(queue);
-	nvme_rdma_free_queue(queue);
+	int i;
+
+	for (i = 1; i < ctrl->ctrl.queue_count; i++)
+		nvme_rdma_free_queue(&ctrl->queues[i]);
 }
 
-static void nvme_rdma_free_io_queues(struct nvme_rdma_ctrl *ctrl)
+static void nvme_rdma_stop_io_queues(struct nvme_rdma_ctrl *ctrl)
 {
 	int i;
 
 	for (i = 1; i < ctrl->ctrl.queue_count; i++)
-		nvme_rdma_stop_and_free_queue(&ctrl->queues[i]);
+		nvme_rdma_stop_queue(&ctrl->queues[i]);
 }
 
 static int nvme_rdma_connect_io_queues(struct nvme_rdma_ctrl *ctrl)
@@ -596,15 +602,15 @@ static int nvme_rdma_connect_io_queues(struct nvme_rdma_ctrl *ctrl)
 		if (ret) {
 			dev_info(ctrl->ctrl.device,
 				"failed to connect i/o queue: %d\n", ret);
-			goto out_free_queues;
+			goto out_stop_queues;
 		}
 		set_bit(NVME_RDMA_Q_LIVE, &ctrl->queues[i].flags);
 	}
 
 	return 0;
 
-out_free_queues:
-	nvme_rdma_free_io_queues(ctrl);
+out_stop_queues:
+	nvme_rdma_stop_io_queues(ctrl);
 	return ret;
 }
 
@@ -650,7 +656,7 @@ static int nvme_rdma_init_io_queues(struct nvme_rdma_ctrl *ctrl)
 
 out_free_queues:
 	for (i--; i >= 1; i--)
-		nvme_rdma_stop_and_free_queue(&ctrl->queues[i]);
+		nvme_rdma_free_queue(&ctrl->queues[i]);
 
 	return ret;
 }
@@ -726,11 +732,12 @@ static void nvme_rdma_destroy_admin_queue(struct nvme_rdma_ctrl *ctrl,
 {
 	nvme_rdma_free_qe(ctrl->queues[0].device->dev, &ctrl->async_event_sqe,
 			sizeof(struct nvme_command), DMA_TO_DEVICE);
-	nvme_rdma_stop_and_free_queue(&ctrl->queues[0]);
+	nvme_rdma_stop_queue(&ctrl->queues[0]);
 	if (remove) {
 		blk_cleanup_queue(ctrl->ctrl.admin_q);
 		nvme_rdma_free_tagset(&ctrl->ctrl, true);
 	}
+	nvme_rdma_free_queue(&ctrl->queues[0]);
 }
 
 static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl,
@@ -810,6 +817,61 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl,
 	return error;
 }
 
+static void nvme_rdma_destroy_io_queues(struct nvme_rdma_ctrl *ctrl,
+		bool remove)
+{
+	nvme_rdma_stop_io_queues(ctrl);
+	if (remove) {
+		blk_cleanup_queue(ctrl->ctrl.connect_q);
+		nvme_rdma_free_tagset(&ctrl->ctrl, false);
+	}
+	nvme_rdma_free_io_queues(ctrl);
+}
+
+static int nvme_rdma_configure_io_queues(struct nvme_rdma_ctrl *ctrl, bool new)
+{
+	int ret;
+
+	ret = nvme_rdma_init_io_queues(ctrl);
+	if (ret)
+		return ret;
+
+	if (new) {
+		ctrl->ctrl.tagset = nvme_rdma_alloc_tagset(&ctrl->ctrl, false);
+		if (IS_ERR(ctrl->ctrl.tagset))
+			goto out_free_io_queues;
+
+		ctrl->ctrl.connect_q = blk_mq_init_queue(&ctrl->tag_set);
+		if (IS_ERR(ctrl->ctrl.connect_q)) {
+			ret = PTR_ERR(ctrl->ctrl.connect_q);
+			goto out_free_tag_set;
+		}
+	} else {
+		ret = blk_mq_reinit_tagset(&ctrl->tag_set);
+		if (ret)
+			goto out_free_io_queues;
+
+		blk_mq_update_nr_hw_queues(&ctrl->tag_set,
+			ctrl->ctrl.queue_count - 1);
+	}
+
+	ret = nvme_rdma_connect_io_queues(ctrl);
+	if (ret)
+		goto out_cleanup_connect_q;
+
+	return 0;
+
+out_cleanup_connect_q:
+	if (new)
+		blk_cleanup_queue(ctrl->ctrl.connect_q);
+out_free_tag_set:
+	if (new)
+		nvme_rdma_free_tagset(&ctrl->ctrl, false);
+out_free_io_queues:
+	nvme_rdma_free_io_queues(ctrl);
+	return ret;
+}
+
 static void nvme_rdma_free_ctrl(struct nvme_ctrl *nctrl)
 {
 	struct nvme_rdma_ctrl *ctrl = to_rdma_ctrl(nctrl);
@@ -856,13 +918,8 @@ static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work)
 
 	++ctrl->ctrl.nr_reconnects;
 
-	if (ctrl->ctrl.queue_count > 1) {
-		nvme_rdma_free_io_queues(ctrl);
-
-		ret = blk_mq_reinit_tagset(&ctrl->tag_set);
-		if (ret)
-			goto requeue;
-	}
+	if (ctrl->ctrl.queue_count > 1)
+		nvme_rdma_destroy_io_queues(ctrl, false);
 
 	nvme_rdma_destroy_admin_queue(ctrl, false);
 	ret = nvme_rdma_configure_admin_queue(ctrl, false);
@@ -870,16 +927,9 @@ static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work)
 		goto requeue;
 
 	if (ctrl->ctrl.queue_count > 1) {
-		ret = nvme_rdma_init_io_queues(ctrl);
+		ret = nvme_rdma_configure_io_queues(ctrl, false);
 		if (ret)
 			goto requeue;
-
-		ret = nvme_rdma_connect_io_queues(ctrl);
-		if (ret)
-			goto requeue;
-
-		blk_mq_update_nr_hw_queues(&ctrl->tag_set,
-				ctrl->ctrl.queue_count - 1);
 	}
 
 	changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE);
@@ -1676,7 +1726,7 @@ static void nvme_rdma_shutdown_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown)
 		nvme_stop_queues(&ctrl->ctrl);
 		blk_mq_tagset_busy_iter(&ctrl->tag_set,
 					nvme_cancel_request, &ctrl->ctrl);
-		nvme_rdma_free_io_queues(ctrl);
+		nvme_rdma_destroy_io_queues(ctrl, shutdown);
 	}
 
 	if (shutdown)
@@ -1699,11 +1749,6 @@ static void __nvme_rdma_remove_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown)
 		nvme_rdma_shutdown_ctrl(ctrl, shutdown);
 
 	nvme_uninit_ctrl(&ctrl->ctrl);
-	if (ctrl->ctrl.tagset) {
-		blk_cleanup_queue(ctrl->ctrl.connect_q);
-		nvme_rdma_free_tagset(&ctrl->ctrl, false);
-	}
-
 	nvme_put_ctrl(&ctrl->ctrl);
 }
 
@@ -1770,20 +1815,9 @@ static void nvme_rdma_reset_ctrl_work(struct work_struct *work)
 	}
 
 	if (ctrl->ctrl.queue_count > 1) {
-		ret = blk_mq_reinit_tagset(&ctrl->tag_set);
-		if (ret)
-			goto del_dead_ctrl;
-
-		ret = nvme_rdma_init_io_queues(ctrl);
+		ret = nvme_rdma_configure_io_queues(ctrl, false);
 		if (ret)
 			goto del_dead_ctrl;
-
-		ret = nvme_rdma_connect_io_queues(ctrl);
-		if (ret)
-			goto del_dead_ctrl;
-
-		blk_mq_update_nr_hw_queues(&ctrl->tag_set,
-				ctrl->ctrl.queue_count - 1);
 	}
 
 	changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE);
@@ -1812,39 +1846,6 @@ static const struct nvme_ctrl_ops nvme_rdma_ctrl_ops = {
 	.get_address		= nvmf_get_address,
 };
 
-static int nvme_rdma_create_io_queues(struct nvme_rdma_ctrl *ctrl)
-{
-	int ret;
-
-	ret = nvme_rdma_init_io_queues(ctrl);
-	if (ret)
-		return ret;
-
-	ctrl->ctrl.tagset = nvme_rdma_alloc_tagset(&ctrl->ctrl, true);
-	if (IS_ERR(ctrl->ctrl.tagset))
-		goto out_free_io_queues;
-
-	ctrl->ctrl.connect_q = blk_mq_init_queue(&ctrl->tag_set);
-	if (IS_ERR(ctrl->ctrl.connect_q)) {
-		ret = PTR_ERR(ctrl->ctrl.connect_q);
-		goto out_free_tag_set;
-	}
-
-	ret = nvme_rdma_connect_io_queues(ctrl);
-	if (ret)
-		goto out_cleanup_connect_q;
-
-	return 0;
-
-out_cleanup_connect_q:
-	blk_cleanup_queue(ctrl->ctrl.connect_q);
-out_free_tag_set:
-	nvme_rdma_free_tagset(&ctrl->ctrl, false);
-out_free_io_queues:
-	nvme_rdma_free_io_queues(ctrl);
-	return ret;
-}
-
 static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev,
 		struct nvmf_ctrl_options *opts)
 {
@@ -1937,7 +1938,7 @@ static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev,
 	}
 
 	if (opts->nr_io_queues) {
-		ret = nvme_rdma_create_io_queues(ctrl);
+		ret = nvme_rdma_configure_io_queues(ctrl, true);
 		if (ret)
 			goto out_remove_admin_queue;
 	}
-- 
2.7.4

[PATCH 08/12] nvme-rdma: stop queues instead of simply flipping their state

[Sagi Grimberg]

If we move the queues from LIVE state, we might as well stop
them (drain for rdma). Do it after we stop the sw request queues
to prevent a stray request sneaking in .queue_rq after we stop
the queue.

Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
---
 drivers/nvme/host/rdma.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index 685ad6acd7cd..346c35d5bf6f 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -952,16 +952,15 @@ static void nvme_rdma_error_recovery_work(struct work_struct *work)
 {
 	struct nvme_rdma_ctrl *ctrl = container_of(work,
 			struct nvme_rdma_ctrl, err_work);
-	int i;
 
 	nvme_stop_ctrl(&ctrl->ctrl);
 
-	for (i = 0; i < ctrl->ctrl.queue_count; i++)
-		clear_bit(NVME_RDMA_Q_LIVE, &ctrl->queues[i].flags);
-
-	if (ctrl->ctrl.queue_count > 1)
+	if (ctrl->ctrl.queue_count > 1) {
 		nvme_stop_queues(&ctrl->ctrl);
+		nvme_rdma_stop_io_queues(ctrl);
+	}
 	blk_mq_quiesce_queue(ctrl->ctrl.admin_q);
+	nvme_rdma_stop_queue(&ctrl->queues[0]);
 
 	/* We must take care of fastfail/requeue all our inflight requests */
 	if (ctrl->ctrl.queue_count > 1)
-- 
2.7.4

[PATCH 09/12] nvme-rdma: rename nvme_rdma_init_queue to nvme_rdma_alloc_queue

[Sagi Grimberg]

Give it a name symmetric to nvme_rdma_free_queue. Also pass in
the ctrl sqsize+1 and not the opts queue_size. And suppress
failure message that has prior verbosity.

Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
---
 drivers/nvme/host/rdma.c | 17 +++++++----------
 1 file changed, 7 insertions(+), 10 deletions(-)

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index 346c35d5bf6f..6f5f50ed7aa8 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -503,7 +503,7 @@ static int nvme_rdma_create_queue_ib(struct nvme_rdma_queue *queue)
 	return ret;
 }
 
-static int nvme_rdma_init_queue(struct nvme_rdma_ctrl *ctrl,
+static int nvme_rdma_alloc_queue(struct nvme_rdma_ctrl *ctrl,
 		int idx, size_t queue_size)
 {
 	struct nvme_rdma_queue *queue;
@@ -614,7 +614,7 @@ static int nvme_rdma_connect_io_queues(struct nvme_rdma_ctrl *ctrl)
 	return ret;
 }
 
-static int nvme_rdma_init_io_queues(struct nvme_rdma_ctrl *ctrl)
+static int nvme_rdma_alloc_io_queues(struct nvme_rdma_ctrl *ctrl)
 {
 	struct nvmf_ctrl_options *opts = ctrl->ctrl.opts;
 	struct ib_device *ibdev = ctrl->device->dev;
@@ -643,13 +643,10 @@ static int nvme_rdma_init_io_queues(struct nvme_rdma_ctrl *ctrl)
 		"creating %d I/O queues.\n", nr_io_queues);
 
 	for (i = 1; i < ctrl->ctrl.queue_count; i++) {
-		ret = nvme_rdma_init_queue(ctrl, i,
-					   ctrl->ctrl.opts->queue_size);
-		if (ret) {
-			dev_info(ctrl->ctrl.device,
-				"failed to initialize i/o queue: %d\n", ret);
+		ret = nvme_rdma_alloc_queue(ctrl, i,
+				ctrl->ctrl.sqsize + 1);
+		if (ret)
 			goto out_free_queues;
-		}
 	}
 
 	return 0;
@@ -745,7 +742,7 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl,
 {
 	int error;
 
-	error = nvme_rdma_init_queue(ctrl, 0, NVME_AQ_DEPTH);
+	error = nvme_rdma_alloc_queue(ctrl, 0, NVME_AQ_DEPTH);
 	if (error)
 		return error;
 
@@ -832,7 +829,7 @@ static int nvme_rdma_configure_io_queues(struct nvme_rdma_ctrl *ctrl, bool new)
 {
 	int ret;
 
-	ret = nvme_rdma_init_io_queues(ctrl);
+	ret = nvme_rdma_alloc_io_queues(ctrl);
 	if (ret)
 		return ret;
 
-- 
2.7.4

[PATCH 10/12] nvme-rdma: introduce nvme_rdma_start_queue

[Sagi Grimberg]

This should pair with nvme_rdma_stop_queue. While this
is not a complete 1x1 reverse, it still pairs up pretty
well because in fabrics we don't have a disconnect capsule
(yet) but we simply teardown the transport association.

Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
---
 drivers/nvme/host/rdma.c | 36 ++++++++++++++++++++++++------------
 1 file changed, 24 insertions(+), 12 deletions(-)

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index 6f5f50ed7aa8..8f8a4ecef82f 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -593,24 +593,38 @@ static void nvme_rdma_stop_io_queues(struct nvme_rdma_ctrl *ctrl)
 		nvme_rdma_stop_queue(&ctrl->queues[i]);
 }
 
-static int nvme_rdma_connect_io_queues(struct nvme_rdma_ctrl *ctrl)
+static int nvme_rdma_start_queue(struct nvme_rdma_ctrl *ctrl, int idx)
+{
+	int ret;
+
+	if (idx)
+		ret = nvmf_connect_io_queue(&ctrl->ctrl, idx);
+	else
+		ret = nvmf_connect_admin_queue(&ctrl->ctrl);
+
+	if (!ret)
+		set_bit(NVME_RDMA_Q_LIVE, &ctrl->queues[idx].flags);
+	else
+		dev_info(ctrl->ctrl.device,
+			"failed to connect queue: %d ret=%d\n", idx, ret);
+	return ret;
+}
+
+static int nvme_rdma_start_io_queues(struct nvme_rdma_ctrl *ctrl)
 {
 	int i, ret = 0;
 
 	for (i = 1; i < ctrl->ctrl.queue_count; i++) {
-		ret = nvmf_connect_io_queue(&ctrl->ctrl, i);
-		if (ret) {
-			dev_info(ctrl->ctrl.device,
-				"failed to connect i/o queue: %d\n", ret);
+		ret = nvme_rdma_start_queue(ctrl, i);
+		if (ret)
 			goto out_stop_queues;
-		}
-		set_bit(NVME_RDMA_Q_LIVE, &ctrl->queues[i].flags);
 	}
 
 	return 0;
 
 out_stop_queues:
-	nvme_rdma_stop_io_queues(ctrl);
+	for (i--; i >= 1; i--)
+		nvme_rdma_stop_queue(&ctrl->queues[i]);
 	return ret;
 }
 
@@ -767,12 +781,10 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl,
 			goto out_free_queue;
 	}
 
-	error = nvmf_connect_admin_queue(&ctrl->ctrl);
+	error = nvme_rdma_start_queue(ctrl, 0);
 	if (error)
 		goto out_cleanup_queue;
 
-	set_bit(NVME_RDMA_Q_LIVE, &ctrl->queues[0].flags);
-
 	error = nvmf_reg_read64(&ctrl->ctrl, NVME_REG_CAP,
 			&ctrl->ctrl.cap);
 	if (error) {
@@ -852,7 +864,7 @@ static int nvme_rdma_configure_io_queues(struct nvme_rdma_ctrl *ctrl, bool new)
 			ctrl->ctrl.queue_count - 1);
 	}
 
-	ret = nvme_rdma_connect_io_queues(ctrl);
+	ret = nvme_rdma_start_io_queues(ctrl);
 	if (ret)
 		goto out_cleanup_connect_q;
 
-- 
2.7.4

[PATCH 11/12] nvme-rdma: cleanup error path in controller reset

[Sagi Grimberg]

No need to queue an extra work to indirect controller
removal, just call a ctrl remove routine.

Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
---
 drivers/nvme/host/rdma.c | 32 +++++++++-----------------------
 1 file changed, 9 insertions(+), 23 deletions(-)

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index 8f8a4ecef82f..bd0d9f7a04ec 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -1749,13 +1749,10 @@ static void nvme_rdma_shutdown_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown)
 	nvme_rdma_destroy_admin_queue(ctrl, shutdown);
 }
 
-static void __nvme_rdma_remove_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown)
+static void nvme_rdma_remove_ctrl(struct nvme_rdma_ctrl *ctrl)
 {
-	nvme_stop_ctrl(&ctrl->ctrl);
 	nvme_remove_namespaces(&ctrl->ctrl);
-	if (shutdown)
-		nvme_rdma_shutdown_ctrl(ctrl, shutdown);
-
+	nvme_rdma_shutdown_ctrl(ctrl, true);
 	nvme_uninit_ctrl(&ctrl->ctrl);
 	nvme_put_ctrl(&ctrl->ctrl);
 }
@@ -1765,7 +1762,8 @@ static void nvme_rdma_del_ctrl_work(struct work_struct *work)
 	struct nvme_rdma_ctrl *ctrl = container_of(work,
 				struct nvme_rdma_ctrl, delete_work);
 
-	__nvme_rdma_remove_ctrl(ctrl, true);
+	nvme_stop_ctrl(&ctrl->ctrl);
+	nvme_rdma_remove_ctrl(ctrl);
 }
 
 static int __nvme_rdma_del_ctrl(struct nvme_rdma_ctrl *ctrl)
@@ -1797,14 +1795,6 @@ static int nvme_rdma_del_ctrl(struct nvme_ctrl *nctrl)
 	return ret;
 }
 
-static void nvme_rdma_remove_ctrl_work(struct work_struct *work)
-{
-	struct nvme_rdma_ctrl *ctrl = container_of(work,
-				struct nvme_rdma_ctrl, delete_work);
-
-	__nvme_rdma_remove_ctrl(ctrl, false);
-}
-
 static void nvme_rdma_reset_ctrl_work(struct work_struct *work)
 {
 	struct nvme_rdma_ctrl *ctrl =
@@ -1816,16 +1806,13 @@ static void nvme_rdma_reset_ctrl_work(struct work_struct *work)
 	nvme_rdma_shutdown_ctrl(ctrl, false);
 
 	ret = nvme_rdma_configure_admin_queue(ctrl, false);
-	if (ret) {
-		/* ctrl is already shutdown, just remove the ctrl */
-		INIT_WORK(&ctrl->delete_work, nvme_rdma_remove_ctrl_work);
-		goto del_dead_ctrl;
-	}
+	if (ret)
+		goto out_fail;
 
 	if (ctrl->ctrl.queue_count > 1) {
 		ret = nvme_rdma_configure_io_queues(ctrl, false);
 		if (ret)
-			goto del_dead_ctrl;
+			goto out_fail;
 	}
 
 	changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE);
@@ -1835,10 +1822,9 @@ static void nvme_rdma_reset_ctrl_work(struct work_struct *work)
 
 	return;
 
-del_dead_ctrl:
-	/* Deleting this dead controller... */
+out_fail:
 	dev_warn(ctrl->ctrl.device, "Removing after reset failure\n");
-	WARN_ON(!queue_work(nvme_wq, &ctrl->delete_work));
+	nvme_rdma_remove_ctrl(ctrl);
 }
 
 static const struct nvme_ctrl_ops nvme_rdma_ctrl_ops = {
-- 
2.7.4

[PATCH 12/12] nvme-rdma: call ops->reg_read64 instead of nvmf_reg_read64

[Sagi Grimberg]

Make it a generic routine.

Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
---
 drivers/nvme/host/rdma.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index bd0d9f7a04ec..000e2f31e8ca 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -785,7 +785,7 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl,
 	if (error)
 		goto out_cleanup_queue;
 
-	error = nvmf_reg_read64(&ctrl->ctrl, NVME_REG_CAP,
+	error = ctrl->ctrl.ops->reg_read64(&ctrl->ctrl, NVME_REG_CAP,
 			&ctrl->ctrl.cap);
 	if (error) {
 		dev_err(ctrl->ctrl.device,
-- 
2.7.4

[PATCH 00/12] Few more patches from the centralization set

[Christoph Hellwig]

Thanks Sagi,

I've applied the series to the nvme-4.14 branch.