1 files changed, 77 insertions, 73 deletions

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index 6d4119dfbdaa..a7f7d0ae3331 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -19,6 +19,7 @@
 #include <linux/string.h>
 #include <linux/atomic.h>
 #include <linux/blk-mq.h>
+#include <linux/blk-mq-rdma.h>
 #include <linux/types.h>
 #include <linux/list.h>
 #include <linux/mutex.h>
@@ -86,7 +87,7 @@ enum nvme_rdma_queue_flags {
 
 struct nvme_rdma_queue {
 	struct nvme_rdma_qe	*rsp_ring;
-	u8			sig_count;
+	atomic_t		sig_count;
 	int			queue_size;
 	size_t			cmnd_capsule_len;
 	struct nvme_rdma_ctrl	*ctrl;
@@ -103,7 +104,6 @@ struct nvme_rdma_queue {
 struct nvme_rdma_ctrl {
 	/* read only in the hot path */
 	struct nvme_rdma_queue	*queues;
-	u32			queue_count;
 
 	/* other member variables */
 	struct blk_mq_tag_set	tag_set;
@@ -119,7 +119,6 @@ struct nvme_rdma_ctrl {
 	struct blk_mq_tag_set	admin_tag_set;
 	struct nvme_rdma_device	*device;
 
-	u64			cap;
 	u32			max_fr_pages;
 
 	struct sockaddr_storage addr;
@@ -274,9 +273,6 @@ static int nvme_rdma_reinit_request(void *data, struct request *rq)
 	struct nvme_rdma_request *req = blk_mq_rq_to_pdu(rq);
 	int ret = 0;
 
-	if (!req->mr->need_inval)
-		goto out;
-
 	ib_dereg_mr(req->mr);
 
 	req->mr = ib_alloc_mr(dev->pd, IB_MR_TYPE_MEM_REG,
@@ -349,7 +345,7 @@ static int nvme_rdma_init_hctx(struct blk_mq_hw_ctx *hctx, void *data,
 	struct nvme_rdma_ctrl *ctrl = data;
 	struct nvme_rdma_queue *queue = &ctrl->queues[hctx_idx + 1];
 
-	BUG_ON(hctx_idx >= ctrl->queue_count);
+	BUG_ON(hctx_idx >= ctrl->ctrl.queue_count);
 
 	hctx->driver_data = queue;
 	return 0;
@@ -468,14 +464,10 @@ static int nvme_rdma_create_queue_ib(struct nvme_rdma_queue *queue)
 	ibdev = queue->device->dev;
 
 	/*
-	 * The admin queue is barely used once the controller is live, so don't
-	 * bother to spread it out.
+	 * Spread I/O queues completion vectors according their queue index.
+	 * Admin queues can always go on completion vector 0.
 	 */
-	if (idx == 0)
-		comp_vector = 0;
-	else
-		comp_vector = idx % ibdev->num_comp_vectors;
-
+	comp_vector = idx == 0 ? idx : idx - 1;
 
 	/* +1 for ib_stop_cq */
 	queue->ib_cq = ib_alloc_cq(ibdev, queue,
@@ -525,6 +517,7 @@ static int nvme_rdma_init_queue(struct nvme_rdma_ctrl *ctrl,
 		queue->cmnd_capsule_len = sizeof(struct nvme_command);
 
 	queue->queue_size = queue_size;
+	atomic_set(&queue->sig_count, 0);
 
 	queue->cm_id = rdma_create_id(&init_net, nvme_rdma_cm_handler, queue,
 			RDMA_PS_TCP, IB_QPT_RC);
@@ -587,7 +580,7 @@ static void nvme_rdma_free_io_queues(struct nvme_rdma_ctrl *ctrl)
 {
 	int i;
 
-	for (i = 1; i < ctrl->queue_count; i++)
+	for (i = 1; i < ctrl->ctrl.queue_count; i++)
 		nvme_rdma_stop_and_free_queue(&ctrl->queues[i]);
 }
 
@@ -595,7 +588,7 @@ static int nvme_rdma_connect_io_queues(struct nvme_rdma_ctrl *ctrl)
 {
 	int i, ret = 0;
 
-	for (i = 1; i < ctrl->queue_count; i++) {
+	for (i = 1; i < ctrl->ctrl.queue_count; i++) {
 		ret = nvmf_connect_io_queue(&ctrl->ctrl, i);
 		if (ret) {
 			dev_info(ctrl->ctrl.device,
@@ -615,22 +608,32 @@ out_free_queues:
 static int nvme_rdma_init_io_queues(struct nvme_rdma_ctrl *ctrl)
 {
 	struct nvmf_ctrl_options *opts = ctrl->ctrl.opts;
+	struct ib_device *ibdev = ctrl->device->dev;
 	unsigned int nr_io_queues;
 	int i, ret;
 
 	nr_io_queues = min(opts->nr_io_queues, num_online_cpus());
+
+	/*
+	 * we map queues according to the device irq vectors for
+	 * optimal locality so we don't need more queues than
+	 * completion vectors.
+	 */
+	nr_io_queues = min_t(unsigned int, nr_io_queues,
+				ibdev->num_comp_vectors);
+
 	ret = nvme_set_queue_count(&ctrl->ctrl, &nr_io_queues);
 	if (ret)
 		return ret;
 
-	ctrl->queue_count = nr_io_queues + 1;
-	if (ctrl->queue_count < 2)
+	ctrl->ctrl.queue_count = nr_io_queues + 1;
+	if (ctrl->ctrl.queue_count < 2)
 		return 0;
 
 	dev_info(ctrl->ctrl.device,
 		"creating %d I/O queues.\n", nr_io_queues);
 
-	for (i = 1; i < ctrl->queue_count; i++) {
+	for (i = 1; i < ctrl->ctrl.queue_count; i++) {
 		ret = nvme_rdma_init_queue(ctrl, i,
 					   ctrl->ctrl.opts->queue_size);
 		if (ret) {
@@ -705,7 +708,7 @@ static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work)
 
 	++ctrl->ctrl.nr_reconnects;
 
-	if (ctrl->queue_count > 1) {
+	if (ctrl->ctrl.queue_count > 1) {
 		nvme_rdma_free_io_queues(ctrl);
 
 		ret = blk_mq_reinit_tagset(&ctrl->tag_set);
@@ -729,13 +732,11 @@ static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work)
 
 	set_bit(NVME_RDMA_Q_LIVE, &ctrl->queues[0].flags);
 
-	ret = nvme_enable_ctrl(&ctrl->ctrl, ctrl->cap);
+	ret = nvme_enable_ctrl(&ctrl->ctrl, ctrl->ctrl.cap);
 	if (ret)
 		goto requeue;
 
-	nvme_start_keep_alive(&ctrl->ctrl);
-
-	if (ctrl->queue_count > 1) {
+	if (ctrl->ctrl.queue_count > 1) {
 		ret = nvme_rdma_init_io_queues(ctrl);
 		if (ret)
 			goto requeue;
@@ -743,16 +744,16 @@ static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work)
 		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);
 	WARN_ON_ONCE(!changed);
 	ctrl->ctrl.nr_reconnects = 0;
 
-	if (ctrl->queue_count > 1) {
-		nvme_queue_scan(&ctrl->ctrl);
-		nvme_queue_async_events(&ctrl->ctrl);
-	}
+	nvme_start_ctrl(&ctrl->ctrl);
 
 	dev_info(ctrl->ctrl.device, "Successfully reconnected\n");
 
@@ -770,17 +771,17 @@ static void nvme_rdma_error_recovery_work(struct work_struct *work)
 			struct nvme_rdma_ctrl, err_work);
 	int i;
 
-	nvme_stop_keep_alive(&ctrl->ctrl);
+	nvme_stop_ctrl(&ctrl->ctrl);
 
-	for (i = 0; i < ctrl->queue_count; i++)
+	for (i = 0; i < ctrl->ctrl.queue_count; i++)
 		clear_bit(NVME_RDMA_Q_LIVE, &ctrl->queues[i].flags);
 
-	if (ctrl->queue_count > 1)
+	if (ctrl->ctrl.queue_count > 1)
 		nvme_stop_queues(&ctrl->ctrl);
-	blk_mq_stop_hw_queues(ctrl->ctrl.admin_q);
+	blk_mq_quiesce_queue(ctrl->ctrl.admin_q);
 
 	/* We must take care of fastfail/requeue all our inflight requests */
-	if (ctrl->queue_count > 1)
+	if (ctrl->ctrl.queue_count > 1)
 		blk_mq_tagset_busy_iter(&ctrl->tag_set,
 					nvme_cancel_request, &ctrl->ctrl);
 	blk_mq_tagset_busy_iter(&ctrl->admin_tag_set,
@@ -790,7 +791,7 @@ static void nvme_rdma_error_recovery_work(struct work_struct *work)
 	 * queues are not a live anymore, so restart the queues to fail fast
 	 * new IO
 	 */
-	blk_mq_start_stopped_hw_queues(ctrl->ctrl.admin_q, true);
+	blk_mq_unquiesce_queue(ctrl->ctrl.admin_q);
 	nvme_start_queues(&ctrl->ctrl);
 
 	nvme_rdma_reconnect_or_remove(ctrl);
@@ -926,7 +927,11 @@ static int nvme_rdma_map_sg_fr(struct nvme_rdma_queue *queue,
 	struct nvme_keyed_sgl_desc *sg = &c->common.dptr.ksgl;
 	int nr;
 
-	nr = ib_map_mr_sg(req->mr, req->sg_table.sgl, count, NULL, PAGE_SIZE);
+	/*
+	 * Align the MR to a 4K page size to match the ctrl page size and
+	 * the block virtual boundary.
+	 */
+	nr = ib_map_mr_sg(req->mr, req->sg_table.sgl, count, NULL, SZ_4K);
 	if (nr < count) {
 		if (nr < 0)
 			return nr;
@@ -1008,17 +1013,16 @@ static void nvme_rdma_send_done(struct ib_cq *cq, struct ib_wc *wc)
 		nvme_rdma_wr_error(cq, wc, "SEND");
 }
 
-static inline int nvme_rdma_queue_sig_limit(struct nvme_rdma_queue *queue)
+/*
+ * We want to signal completion at least every queue depth/2.  This returns the
+ * largest power of two that is not above half of (queue size + 1) to optimize
+ * (avoid divisions).
+ */
+static inline bool nvme_rdma_queue_sig_limit(struct nvme_rdma_queue *queue)
 {
-	int sig_limit;
+	int limit = 1 << ilog2((queue->queue_size + 1) / 2);
 
-	/*
-	 * We signal completion every queue depth/2 and also handle the
-	 * degenerated case of a  device with queue_depth=1, where we
-	 * would need to signal every message.
-	 */
-	sig_limit = max(queue->queue_size / 2, 1);
-	return (++queue->sig_count % sig_limit) == 0;
+	return (atomic_inc_return(&queue->sig_count) & (limit - 1)) == 0;
 }
 
 static int nvme_rdma_post_send(struct nvme_rdma_queue *queue,
@@ -1505,6 +1509,13 @@ static void nvme_rdma_complete_rq(struct request *rq)
 	nvme_complete_rq(rq);
 }
 
+static int nvme_rdma_map_queues(struct blk_mq_tag_set *set)
+{
+	struct nvme_rdma_ctrl *ctrl = set->driver_data;
+
+	return blk_mq_rdma_map_queues(set, ctrl->device->dev, 0);
+}
+
 static const struct blk_mq_ops nvme_rdma_mq_ops = {
 	.queue_rq	= nvme_rdma_queue_rq,
 	.complete	= nvme_rdma_complete_rq,
@@ -1514,6 +1525,7 @@ static const struct blk_mq_ops nvme_rdma_mq_ops = {
 	.init_hctx	= nvme_rdma_init_hctx,
 	.poll		= nvme_rdma_poll,
 	.timeout	= nvme_rdma_timeout,
+	.map_queues	= nvme_rdma_map_queues,
 };
 
 static const struct blk_mq_ops nvme_rdma_admin_mq_ops = {
@@ -1574,7 +1586,8 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl)
 
 	set_bit(NVME_RDMA_Q_LIVE, &ctrl->queues[0].flags);
 
-	error = nvmf_reg_read64(&ctrl->ctrl, NVME_REG_CAP, &ctrl->cap);
+	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");
@@ -1582,14 +1595,14 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl)
 	}
 
 	ctrl->ctrl.sqsize =
-		min_t(int, NVME_CAP_MQES(ctrl->cap), ctrl->ctrl.sqsize);
+		min_t(int, NVME_CAP_MQES(ctrl->ctrl.cap), ctrl->ctrl.sqsize);
 
-	error = nvme_enable_ctrl(&ctrl->ctrl, ctrl->cap);
+	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);
+		(ctrl->max_fr_pages - 1) << (ilog2(SZ_4K) - 9);
 
 	error = nvme_init_identify(&ctrl->ctrl);
 	if (error)
@@ -1601,8 +1614,6 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl)
 	if (error)
 		goto out_cleanup_queue;
 
-	nvme_start_keep_alive(&ctrl->ctrl);
-
 	return 0;
 
 out_cleanup_queue:
@@ -1620,11 +1631,10 @@ out_free_queue:
 
 static void nvme_rdma_shutdown_ctrl(struct nvme_rdma_ctrl *ctrl)
 {
-	nvme_stop_keep_alive(&ctrl->ctrl);
 	cancel_work_sync(&ctrl->err_work);
 	cancel_delayed_work_sync(&ctrl->reconnect_work);
 
-	if (ctrl->queue_count > 1) {
+	if (ctrl->ctrl.queue_count > 1) {
 		nvme_stop_queues(&ctrl->ctrl);
 		blk_mq_tagset_busy_iter(&ctrl->tag_set,
 					nvme_cancel_request, &ctrl->ctrl);
@@ -1634,18 +1644,21 @@ static void nvme_rdma_shutdown_ctrl(struct nvme_rdma_ctrl *ctrl)
 	if (test_bit(NVME_RDMA_Q_LIVE, &ctrl->queues[0].flags))
 		nvme_shutdown_ctrl(&ctrl->ctrl);
 
-	blk_mq_stop_hw_queues(ctrl->ctrl.admin_q);
+	blk_mq_quiesce_queue(ctrl->ctrl.admin_q);
 	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);
 }
 
 static void __nvme_rdma_remove_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown)
 {
-	nvme_uninit_ctrl(&ctrl->ctrl);
+	nvme_stop_ctrl(&ctrl->ctrl);
+	nvme_remove_namespaces(&ctrl->ctrl);
 	if (shutdown)
 		nvme_rdma_shutdown_ctrl(ctrl);
 
+	nvme_uninit_ctrl(&ctrl->ctrl);
 	if (ctrl->ctrl.tagset) {
 		blk_cleanup_queue(ctrl->ctrl.connect_q);
 		blk_mq_free_tag_set(&ctrl->tag_set);
@@ -1707,6 +1720,7 @@ static void nvme_rdma_reset_ctrl_work(struct work_struct *work)
 	int ret;
 	bool changed;
 
+	nvme_stop_ctrl(&ctrl->ctrl);
 	nvme_rdma_shutdown_ctrl(ctrl);
 
 	ret = nvme_rdma_configure_admin_queue(ctrl);
@@ -1716,7 +1730,7 @@ static void nvme_rdma_reset_ctrl_work(struct work_struct *work)
 		goto del_dead_ctrl;
 	}
 
-	if (ctrl->queue_count > 1) {
+	if (ctrl->ctrl.queue_count > 1) {
 		ret = blk_mq_reinit_tagset(&ctrl->tag_set);
 		if (ret)
 			goto del_dead_ctrl;
@@ -1728,16 +1742,15 @@ static void nvme_rdma_reset_ctrl_work(struct work_struct *work)
 		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);
 	WARN_ON_ONCE(!changed);
 
-	if (ctrl->queue_count > 1) {
-		nvme_start_queues(&ctrl->ctrl);
-		nvme_queue_scan(&ctrl->ctrl);
-		nvme_queue_async_events(&ctrl->ctrl);
-	}
+	nvme_start_ctrl(&ctrl->ctrl);
 
 	return;
 
@@ -1785,7 +1798,7 @@ static int nvme_rdma_create_io_queues(struct nvme_rdma_ctrl *ctrl)
 	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->queue_count - 1;
+	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);
@@ -1863,12 +1876,12 @@ static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev,
 	INIT_WORK(&ctrl->delete_work, nvme_rdma_del_ctrl_work);
 	INIT_WORK(&ctrl->ctrl.reset_work, nvme_rdma_reset_ctrl_work);
 
-	ctrl->queue_count = opts->nr_io_queues + 1; /* +1 for admin queue */
+	ctrl->ctrl.queue_count = opts->nr_io_queues + 1; /* +1 for admin queue */
 	ctrl->ctrl.sqsize = opts->queue_size - 1;
 	ctrl->ctrl.kato = opts->kato;
 
 	ret = -ENOMEM;
-	ctrl->queues = kcalloc(ctrl->queue_count, sizeof(*ctrl->queues),
+	ctrl->queues = kcalloc(ctrl->ctrl.queue_count, sizeof(*ctrl->queues),
 				GFP_KERNEL);
 	if (!ctrl->queues)
 		goto out_uninit_ctrl;
@@ -1925,15 +1938,11 @@ static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev,
 	list_add_tail(&ctrl->list, &nvme_rdma_ctrl_list);
 	mutex_unlock(&nvme_rdma_ctrl_mutex);
 
-	if (opts->nr_io_queues) {
-		nvme_queue_scan(&ctrl->ctrl);
-		nvme_queue_async_events(&ctrl->ctrl);
-	}
+	nvme_start_ctrl(&ctrl->ctrl);
 
 	return &ctrl->ctrl;
 
 out_remove_admin_queue:
-	nvme_stop_keep_alive(&ctrl->ctrl);
 	nvme_rdma_destroy_admin_queue(ctrl);
 out_kfree_queues:
 	kfree(ctrl->queues);
@@ -1956,10 +1965,6 @@ static struct nvmf_transport_ops nvme_rdma_transport = {
 	.create_ctrl	= nvme_rdma_create_ctrl,
 };
 
-static void nvme_rdma_add_one(struct ib_device *ib_device)
-{
-}
-
 static void nvme_rdma_remove_one(struct ib_device *ib_device, void *client_data)
 {
 	struct nvme_rdma_ctrl *ctrl;
@@ -1981,7 +1986,6 @@ static void nvme_rdma_remove_one(struct ib_device *ib_device, void *client_data)
 
 static struct ib_client nvme_rdma_ib_client = {
 	.name   = "nvme_rdma",
-	.add = nvme_rdma_add_one,
 	.remove = nvme_rdma_remove_one
 };