1 files changed, 79 insertions, 48 deletions

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index 11dde0d83044..9bd3646568d0 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -18,7 +18,7 @@
 #include <linux/mutex.h>
 #include <linux/scatterlist.h>
 #include <linux/nvme.h>
-#include <asm/unaligned.h>
+#include <linux/unaligned.h>
 
 #include <rdma/ib_verbs.h>
 #include <rdma/rdma_cm.h>
@@ -221,7 +221,7 @@ static struct nvme_rdma_qe *nvme_rdma_alloc_ring(struct ib_device *ibdev,
 
 	/*
 	 * Bind the CQEs (post recv buffers) DMA mapping to the RDMA queue
-	 * lifetime. It's safe, since any chage in the underlying RDMA device
+	 * lifetime. It's safe, since any change in the underlying RDMA device
 	 * will issue error recovery and queue re-creation.
 	 */
 	for (i = 0; i < ib_queue_size; i++) {
@@ -800,7 +800,7 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl,
 
 	/*
 	 * Bind the async event SQE DMA mapping to the admin queue lifetime.
-	 * It's safe, since any chage in the underlying RDMA device will issue
+	 * It's safe, since any change in the underlying RDMA device will issue
 	 * error recovery and queue re-creation.
 	 */
 	error = nvme_rdma_alloc_qe(ctrl->device->dev, &ctrl->async_event_sqe,
@@ -982,7 +982,8 @@ free_ctrl:
 	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);
 
@@ -992,7 +993,7 @@ static void nvme_rdma_reconnect_or_remove(struct nvme_rdma_ctrl *ctrl)
 		return;
 	}
 
-	if (nvmf_should_reconnect(&ctrl->ctrl)) {
+	if (nvmf_should_reconnect(&ctrl->ctrl, status)) {
 		dev_info(ctrl->ctrl.device, "Reconnecting in %d seconds...\n",
 			ctrl->ctrl.opts->reconnect_delay);
 		queue_delayed_work(nvme_wq, &ctrl->reconnect_work,
@@ -1006,6 +1007,7 @@ static int nvme_rdma_setup_ctrl(struct nvme_rdma_ctrl *ctrl, bool new)
 {
 	int ret;
 	bool changed;
+	u16 max_queue_size;
 
 	ret = nvme_rdma_configure_admin_queue(ctrl, new);
 	if (ret)
@@ -1017,7 +1019,7 @@ static int nvme_rdma_setup_ctrl(struct nvme_rdma_ctrl *ctrl, bool new)
 		goto destroy_admin;
 	}
 
-	if (!(ctrl->ctrl.sgls & (1 << 2))) {
+	if (!(ctrl->ctrl.sgls & NVME_CTRL_SGLS_KSDBDS)) {
 		ret = -EOPNOTSUPP;
 		dev_err(ctrl->ctrl.device,
 			"Mandatory keyed sgls are not supported!\n");
@@ -1030,11 +1032,16 @@ static int nvme_rdma_setup_ctrl(struct nvme_rdma_ctrl *ctrl, bool new)
 			ctrl->ctrl.opts->queue_size, ctrl->ctrl.sqsize + 1);
 	}
 
-	if (ctrl->ctrl.sqsize + 1 > NVME_RDMA_MAX_QUEUE_SIZE) {
+	if (ctrl->ctrl.max_integrity_segments)
+		max_queue_size = NVME_RDMA_MAX_METADATA_QUEUE_SIZE;
+	else
+		max_queue_size = NVME_RDMA_MAX_QUEUE_SIZE;
+
+	if (ctrl->ctrl.sqsize + 1 > max_queue_size) {
 		dev_warn(ctrl->ctrl.device,
-			"ctrl sqsize %u > max queue size %u, clamping down\n",
-			ctrl->ctrl.sqsize + 1, NVME_RDMA_MAX_QUEUE_SIZE);
-		ctrl->ctrl.sqsize = NVME_RDMA_MAX_QUEUE_SIZE - 1;
+			 "ctrl sqsize %u > max queue size %u, clamping down\n",
+			 ctrl->ctrl.sqsize + 1, max_queue_size);
+		ctrl->ctrl.sqsize = max_queue_size - 1;
 	}
 
 	if (ctrl->ctrl.sqsize + 1 > ctrl->ctrl.maxcmd) {
@@ -1044,7 +1051,7 @@ static int nvme_rdma_setup_ctrl(struct nvme_rdma_ctrl *ctrl, bool new)
 		ctrl->ctrl.sqsize = ctrl->ctrl.maxcmd - 1;
 	}
 
-	if (ctrl->ctrl.sgls & (1 << 20))
+	if (ctrl->ctrl.sgls & NVME_CTRL_SGLS_SAOS)
 		ctrl->use_inline_data = true;
 
 	if (ctrl->ctrl.queue_count > 1) {
@@ -1084,13 +1091,7 @@ destroy_io:
 	}
 destroy_admin:
 	nvme_stop_keep_alive(&ctrl->ctrl);
-	nvme_quiesce_admin_queue(&ctrl->ctrl);
-	blk_sync_queue(ctrl->ctrl.admin_q);
-	nvme_rdma_stop_queue(&ctrl->queues[0]);
-	nvme_cancel_admin_tagset(&ctrl->ctrl);
-	if (new)
-		nvme_remove_admin_tag_set(&ctrl->ctrl);
-	nvme_rdma_destroy_admin_queue(ctrl);
+	nvme_rdma_teardown_admin_queue(ctrl, new);
 	return ret;
 }
 
@@ -1098,10 +1099,12 @@ 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",
@@ -1112,9 +1115,9 @@ static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work)
 	return;
 
 requeue:
-	dev_info(ctrl->ctrl.device, "Failed reconnect attempt %d\n",
-			ctrl->ctrl.nr_reconnects);
-	nvme_rdma_reconnect_or_remove(ctrl);
+	dev_info(ctrl->ctrl.device, "Failed reconnect attempt %d/%d\n",
+		 ctrl->ctrl.nr_reconnects, ctrl->ctrl.opts->max_reconnects);
+	nvme_rdma_reconnect_or_remove(ctrl, ret);
 }
 
 static void nvme_rdma_error_recovery_work(struct work_struct *work)
@@ -1139,7 +1142,7 @@ static void nvme_rdma_error_recovery_work(struct work_struct *work)
 		return;
 	}
 
-	nvme_rdma_reconnect_or_remove(ctrl);
+	nvme_rdma_reconnect_or_remove(ctrl, 0);
 }
 
 static void nvme_rdma_error_recovery(struct nvme_rdma_ctrl *ctrl)
@@ -1354,8 +1357,8 @@ static void nvme_rdma_set_sig_domain(struct blk_integrity *bi,
 	if (control & NVME_RW_PRINFO_PRCHK_REF)
 		domain->sig.dif.ref_remap = true;
 
-	domain->sig.dif.app_tag = le16_to_cpu(cmd->rw.apptag);
-	domain->sig.dif.apptag_check_mask = le16_to_cpu(cmd->rw.appmask);
+	domain->sig.dif.app_tag = le16_to_cpu(cmd->rw.lbat);
+	domain->sig.dif.apptag_check_mask = le16_to_cpu(cmd->rw.lbatm);
 	domain->sig.dif.app_escape = true;
 	if (pi_type == NVME_NS_DPS_PI_TYPE3)
 		domain->sig.dif.ref_escape = true;
@@ -1410,6 +1413,8 @@ static int nvme_rdma_map_sg_pi(struct nvme_rdma_queue *queue,
 	struct nvme_ns *ns = rq->q->queuedata;
 	struct bio *bio = rq->bio;
 	struct nvme_keyed_sgl_desc *sg = &c->common.dptr.ksgl;
+	struct blk_integrity *bi = blk_get_integrity(bio->bi_bdev->bd_disk);
+	u32 xfer_len;
 	int nr;
 
 	req->mr = ib_mr_pool_get(queue->qp, &queue->qp->sig_mrs);
@@ -1422,8 +1427,7 @@ static int nvme_rdma_map_sg_pi(struct nvme_rdma_queue *queue,
 	if (unlikely(nr))
 		goto mr_put;
 
-	nvme_rdma_set_sig_attrs(blk_get_integrity(bio->bi_bdev->bd_disk), c,
-				req->mr->sig_attrs, ns->head->pi_type);
+	nvme_rdma_set_sig_attrs(bi, c, req->mr->sig_attrs, ns->head->pi_type);
 	nvme_rdma_set_prot_checks(c, &req->mr->sig_attrs->check_mask);
 
 	ib_update_fast_reg_key(req->mr, ib_inc_rkey(req->mr->rkey));
@@ -1441,7 +1445,11 @@ static int nvme_rdma_map_sg_pi(struct nvme_rdma_queue *queue,
 		     IB_ACCESS_REMOTE_WRITE;
 
 	sg->addr = cpu_to_le64(req->mr->iova);
-	put_unaligned_le24(req->mr->length, sg->length);
+	xfer_len = req->mr->length;
+	/* Check if PI is added by the HW */
+	if (!pi_count)
+		xfer_len += (xfer_len >> bi->interval_exp) * ns->head->pi_size;
+	put_unaligned_le24(xfer_len, sg->length);
 	put_unaligned_le32(req->mr->rkey, sg->key);
 	sg->type = NVME_KEY_SGL_FMT_DATA_DESC << 4;
 
@@ -1468,8 +1476,7 @@ static int nvme_rdma_dma_map_req(struct ib_device *ibdev, struct request *rq,
 	if (ret)
 		return -ENOMEM;
 
-	req->data_sgl.nents = blk_rq_map_sg(rq->q, rq,
-					    req->data_sgl.sg_table.sgl);
+	req->data_sgl.nents = blk_rq_map_sg(rq, req->data_sgl.sg_table.sgl);
 
 	*count = ib_dma_map_sg(ibdev, req->data_sgl.sg_table.sgl,
 			       req->data_sgl.nents, rq_dma_dir(rq));
@@ -1482,7 +1489,7 @@ static int nvme_rdma_dma_map_req(struct ib_device *ibdev, struct request *rq,
 		req->metadata_sgl->sg_table.sgl =
 			(struct scatterlist *)(req->metadata_sgl + 1);
 		ret = sg_alloc_table_chained(&req->metadata_sgl->sg_table,
-				blk_rq_count_integrity_sg(rq->q, rq->bio),
+				rq->nr_integrity_segments,
 				req->metadata_sgl->sg_table.sgl,
 				NVME_INLINE_METADATA_SG_CNT);
 		if (unlikely(ret)) {
@@ -1490,8 +1497,8 @@ static int nvme_rdma_dma_map_req(struct ib_device *ibdev, struct request *rq,
 			goto out_unmap_sg;
 		}
 
-		req->metadata_sgl->nents = blk_rq_map_integrity_sg(rq->q,
-				rq->bio, req->metadata_sgl->sg_table.sgl);
+		req->metadata_sgl->nents = blk_rq_map_integrity_sg(rq,
+				req->metadata_sgl->sg_table.sgl);
 		*pi_count = ib_dma_map_sg(ibdev,
 					  req->metadata_sgl->sg_table.sgl,
 					  req->metadata_sgl->nents,
@@ -1862,6 +1869,8 @@ static int nvme_rdma_route_resolved(struct nvme_rdma_queue *queue)
 		 */
 		priv.hrqsize = cpu_to_le16(queue->queue_size);
 		priv.hsqsize = cpu_to_le16(queue->ctrl->ctrl.sqsize);
+		/* cntlid should only be set when creating an I/O queue */
+		priv.cntlid = cpu_to_le16(ctrl->ctrl.cntlid);
 	}
 
 	ret = rdma_connect_locked(queue->cm_id, &param);
@@ -1946,14 +1955,13 @@ static enum blk_eh_timer_return nvme_rdma_timeout(struct request *rq)
 	struct nvme_rdma_request *req = blk_mq_rq_to_pdu(rq);
 	struct nvme_rdma_queue *queue = req->queue;
 	struct nvme_rdma_ctrl *ctrl = queue->ctrl;
-	u8 opcode = req->req.cmd->common.opcode;
-	u8 fctype = req->req.cmd->fabrics.fctype;
+	struct nvme_command *cmd = req->req.cmd;
 	int qid = nvme_rdma_queue_idx(queue);
 
 	dev_warn(ctrl->ctrl.device,
 		 "I/O tag %d (%04x) opcode %#x (%s) QID %d timeout\n",
-		 rq->tag, nvme_cid(rq), opcode,
-		 nvme_opcode_str(qid, opcode, fctype), qid);
+		 rq->tag, nvme_cid(rq), cmd->common.opcode,
+		 nvme_fabrics_opcode_str(qid, cmd), qid);
 
 	if (nvme_ctrl_state(&ctrl->ctrl) != NVME_CTRL_LIVE) {
 		/*
@@ -2159,6 +2167,7 @@ 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);
@@ -2169,14 +2178,15 @@ 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 = {
@@ -2186,6 +2196,7 @@ static const struct nvme_ctrl_ops nvme_rdma_ctrl_ops = {
 	.reg_read32		= nvmf_reg_read32,
 	.reg_read64		= nvmf_reg_read64,
 	.reg_write32		= nvmf_reg_write32,
+	.subsystem_reset	= nvmf_subsystem_reset,
 	.free_ctrl		= nvme_rdma_free_ctrl,
 	.submit_async_event	= nvme_rdma_submit_async_event,
 	.delete_ctrl		= nvme_rdma_delete_ctrl,
@@ -2222,12 +2233,11 @@ nvme_rdma_existing_controller(struct nvmf_ctrl_options *opts)
 	return found;
 }
 
-static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev,
+static struct nvme_rdma_ctrl *nvme_rdma_alloc_ctrl(struct device *dev,
 		struct nvmf_ctrl_options *opts)
 {
 	struct nvme_rdma_ctrl *ctrl;
 	int ret;
-	bool changed;
 
 	ctrl = kzalloc(sizeof(*ctrl), GFP_KERNEL);
 	if (!ctrl)
@@ -2289,6 +2299,30 @@ static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev,
 	if (ret)
 		goto out_kfree_queues;
 
+	return ctrl;
+
+out_kfree_queues:
+	kfree(ctrl->queues);
+out_free_ctrl:
+	kfree(ctrl);
+	return ERR_PTR(ret);
+}
+
+static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev,
+		struct nvmf_ctrl_options *opts)
+{
+	struct nvme_rdma_ctrl *ctrl;
+	bool changed;
+	int ret;
+
+	ctrl = nvme_rdma_alloc_ctrl(dev, opts);
+	if (IS_ERR(ctrl))
+		return ERR_CAST(ctrl);
+
+	ret = nvme_add_ctrl(&ctrl->ctrl);
+	if (ret)
+		goto out_put_ctrl;
+
 	changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING);
 	WARN_ON_ONCE(!changed);
 
@@ -2296,8 +2330,8 @@ static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev,
 	if (ret)
 		goto out_uninit_ctrl;
 
-	dev_info(ctrl->ctrl.device, "new ctrl: NQN \"%s\", addr %pISpcs\n",
-		nvmf_ctrl_subsysnqn(&ctrl->ctrl), &ctrl->addr);
+	dev_info(ctrl->ctrl.device, "new ctrl: NQN \"%s\", addr %pISpcs, hostnqn: %s\n",
+		nvmf_ctrl_subsysnqn(&ctrl->ctrl), &ctrl->addr, opts->host->nqn);
 
 	mutex_lock(&nvme_rdma_ctrl_mutex);
 	list_add_tail(&ctrl->list, &nvme_rdma_ctrl_list);
@@ -2307,15 +2341,11 @@ static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev,
 
 out_uninit_ctrl:
 	nvme_uninit_ctrl(&ctrl->ctrl);
+out_put_ctrl:
 	nvme_put_ctrl(&ctrl->ctrl);
 	if (ret > 0)
 		ret = -EIO;
 	return ERR_PTR(ret);
-out_kfree_queues:
-	kfree(ctrl->queues);
-out_free_ctrl:
-	kfree(ctrl);
-	return ERR_PTR(ret);
 }
 
 static struct nvmf_transport_ops nvme_rdma_transport = {
@@ -2400,4 +2430,5 @@ static void __exit nvme_rdma_cleanup_module(void)
 module_init(nvme_rdma_init_module);
 module_exit(nvme_rdma_cleanup_module);
 
+MODULE_DESCRIPTION("NVMe host RDMA transport driver");
 MODULE_LICENSE("GPL v2");