Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma

Pull rdma updates from Jason Gunthorpe:
 "A very quiet cycle with few notable changes. Mostly the usual list of
  one or two patches to drivers changing something that isn't quite rc
  worthy. The subsystem seems to be seeing a larger number of rework and
  cleanup style patches right now, I feel that several vendors are
  prepping their drivers for new silicon.

  Summary:

   - Driver updates and cleanup for qedr, bnxt_re, hns, siw, mlx5, mlx4,
     rxe, i40iw

   - Larger series doing cleanup and rework for hns and hfi1.

   - Some general reworking of the CM code to make it a little more
     understandable

   - Unify the different code paths connected to the uverbs FD scheme

   - New UAPI ioctls conversions for get context and get async fd

   - Trace points for CQ and CM portions of the RDMA stack

   - mlx5 driver support for virtio-net formatted rings as RDMA raw
     ethernet QPs

   - verbs support for setting the PCI-E relaxed ordering bit on DMA
     traffic connected to a MR

   - A couple of bug fixes that came too late to make rc7"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (108 commits)
  RDMA/core: Make the entire API tree static
  RDMA/efa: Mask access flags with the correct optional range
  RDMA/cma: Fix unbalanced cm_id reference count during address resolve
  RDMA/umem: Fix ib_umem_find_best_pgsz()
  IB/mlx4: Fix leak in id_map_find_del
  IB/opa_vnic: Spelling correction of 'erorr' to 'error'
  IB/hfi1: Fix logical condition in msix_request_irq
  RDMA/cm: Remove CM message structs
  RDMA/cm: Use IBA functions for complex structure members
  RDMA/cm: Use IBA functions for simple structure members
  RDMA/cm: Use IBA functions for swapping get/set acessors
  RDMA/cm: Use IBA functions for simple get/set acessors
  RDMA/cm: Add SET/GET implementations to hide IBA wire format
  RDMA/cm: Add accessors for CM_REQ transport_type
  IB/mlx5: Return the administrative GUID if exists
  RDMA/core: Ensure that rdma_user_mmap_entry_remove() is a fence
  IB/mlx4: Fix memory leak in add_gid error flow
  IB/mlx5: Expose RoCE accelerator counters
  RDMA/mlx5: Set relaxed ordering when requested
  RDMA/core: Add the core support field to METHOD_GET_CONTEXT
  ...

1 files changed, 76 insertions, 30 deletions

diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c
index 854ef6e74788..d0031d559213 100644
--- a/drivers/infiniband/hw/hns/hns_roce_main.c
+++ b/drivers/infiniband/hw/hns/hns_roce_main.c
@@ -90,7 +90,7 @@ static int hns_roce_add_gid(const struct ib_gid_attr *attr, void **context)
 static int hns_roce_del_gid(const struct ib_gid_attr *attr, void **context)
 {
 	struct hns_roce_dev *hr_dev = to_hr_dev(attr->device);
-	struct ib_gid_attr zattr = { };
+	struct ib_gid_attr zattr = {};
 	u8 port = attr->port_num - 1;
 	int ret;
 
@@ -210,7 +210,7 @@ static int hns_roce_query_device(struct ib_device *ib_dev,
 	props->max_pkeys = 1;
 	props->local_ca_ack_delay = hr_dev->caps.local_ca_ack_delay;
 	if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_SRQ) {
-		props->max_srq = hr_dev->caps.max_srqs;
+		props->max_srq = hr_dev->caps.num_srqs;
 		props->max_srq_wr = hr_dev->caps.max_srq_wrs;
 		props->max_srq_sge = hr_dev->caps.max_srq_sges;
 	}
@@ -259,11 +259,12 @@ static int hns_roce_query_port(struct ib_device *ib_dev, u8 port_num,
 
 	mtu = iboe_get_mtu(net_dev->mtu);
 	props->active_mtu = mtu ? min(props->max_mtu, mtu) : IB_MTU_256;
-	props->state = (netif_running(net_dev) && netif_carrier_ok(net_dev)) ?
-			IB_PORT_ACTIVE : IB_PORT_DOWN;
-	props->phys_state = (props->state == IB_PORT_ACTIVE) ?
-			     IB_PORT_PHYS_STATE_LINK_UP :
-			     IB_PORT_PHYS_STATE_DISABLED;
+	props->state = netif_running(net_dev) && netif_carrier_ok(net_dev) ?
+			       IB_PORT_ACTIVE :
+			       IB_PORT_DOWN;
+	props->phys_state = props->state == IB_PORT_ACTIVE ?
+				    IB_PORT_PHYS_STATE_LINK_UP :
+				    IB_PORT_PHYS_STATE_DISABLED;
 
 	spin_unlock_irqrestore(&hr_dev->iboe.lock, flags);
 
@@ -481,13 +482,13 @@ static int hns_roce_register_device(struct hns_roce_dev *hr_dev)
 
 	ib_dev = &hr_dev->ib_dev;
 
-	ib_dev->node_type		= RDMA_NODE_IB_CA;
-	ib_dev->dev.parent		= dev;
+	ib_dev->node_type = RDMA_NODE_IB_CA;
+	ib_dev->dev.parent = dev;
 
-	ib_dev->phys_port_cnt		= hr_dev->caps.num_ports;
-	ib_dev->local_dma_lkey		= hr_dev->caps.reserved_lkey;
-	ib_dev->num_comp_vectors	= hr_dev->caps.num_comp_vectors;
-	ib_dev->uverbs_cmd_mask		=
+	ib_dev->phys_port_cnt = hr_dev->caps.num_ports;
+	ib_dev->local_dma_lkey = hr_dev->caps.reserved_lkey;
+	ib_dev->num_comp_vectors = hr_dev->caps.num_comp_vectors;
+	ib_dev->uverbs_cmd_mask =
 		(1ULL << IB_USER_VERBS_CMD_GET_CONTEXT) |
 		(1ULL << IB_USER_VERBS_CMD_QUERY_DEVICE) |
 		(1ULL << IB_USER_VERBS_CMD_QUERY_PORT) |
@@ -503,8 +504,7 @@ static int hns_roce_register_device(struct hns_roce_dev *hr_dev)
 		(1ULL << IB_USER_VERBS_CMD_QUERY_QP) |
 		(1ULL << IB_USER_VERBS_CMD_DESTROY_QP);
 
-	ib_dev->uverbs_ex_cmd_mask |=
-		(1ULL << IB_USER_VERBS_EX_CMD_MODIFY_CQ);
+	ib_dev->uverbs_ex_cmd_mask |= (1ULL << IB_USER_VERBS_EX_CMD_MODIFY_CQ);
 
 	if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_REREG_MR) {
 		ib_dev->uverbs_cmd_mask |= (1ULL << IB_USER_VERBS_CMD_REREG_MR);
@@ -589,11 +589,13 @@ static int hns_roce_init_hem(struct hns_roce_dev *hr_dev)
 
 	if (hns_roce_check_whether_mhop(hr_dev, HEM_TYPE_CQE)) {
 		ret = hns_roce_init_hem_table(hr_dev,
-				      &hr_dev->mr_table.mtt_cqe_table,
-				      HEM_TYPE_CQE, hr_dev->caps.mtt_entry_sz,
-				      hr_dev->caps.num_cqe_segs, 1);
+					      &hr_dev->mr_table.mtt_cqe_table,
+					      HEM_TYPE_CQE,
+					      hr_dev->caps.mtt_entry_sz,
+					      hr_dev->caps.num_cqe_segs, 1);
 		if (ret) {
-			dev_err(dev, "Failed to init MTT CQE context memory, aborting.\n");
+			dev_err(dev,
+				"Failed to init CQE context memory, aborting.\n");
 			goto err_unmap_cqe;
 		}
 	}
@@ -633,7 +635,7 @@ static int hns_roce_init_hem(struct hns_roce_dev *hr_dev)
 					      hr_dev->caps.num_qps, 1);
 		if (ret) {
 			dev_err(dev,
-			       "Failed to init trrl_table memory, aborting.\n");
+				"Failed to init trrl_table memory, aborting.\n");
 			goto err_unmap_irrl;
 		}
 	}
@@ -653,7 +655,7 @@ static int hns_roce_init_hem(struct hns_roce_dev *hr_dev)
 					      hr_dev->caps.num_srqs, 1);
 		if (ret) {
 			dev_err(dev,
-			      "Failed to init SRQ context memory, aborting.\n");
+				"Failed to init SRQ context memory, aborting.\n");
 			goto err_unmap_cq;
 		}
 	}
@@ -692,33 +694,31 @@ static int hns_roce_init_hem(struct hns_roce_dev *hr_dev)
 					      hr_dev->caps.num_qps, 1);
 		if (ret) {
 			dev_err(dev,
-			      "Failed to init SCC context memory, aborting.\n");
+				"Failed to init SCC context memory, aborting.\n");
 			goto err_unmap_idx;
 		}
 	}
 
 	if (hr_dev->caps.qpc_timer_entry_sz) {
-		ret = hns_roce_init_hem_table(hr_dev,
-					      &hr_dev->qpc_timer_table,
+		ret = hns_roce_init_hem_table(hr_dev, &hr_dev->qpc_timer_table,
 					      HEM_TYPE_QPC_TIMER,
 					      hr_dev->caps.qpc_timer_entry_sz,
 					      hr_dev->caps.num_qpc_timer, 1);
 		if (ret) {
 			dev_err(dev,
-			      "Failed to init QPC timer memory, aborting.\n");
+				"Failed to init QPC timer memory, aborting.\n");
 			goto err_unmap_ctx;
 		}
 	}
 
 	if (hr_dev->caps.cqc_timer_entry_sz) {
-		ret = hns_roce_init_hem_table(hr_dev,
-					      &hr_dev->cqc_timer_table,
+		ret = hns_roce_init_hem_table(hr_dev, &hr_dev->cqc_timer_table,
 					      HEM_TYPE_CQC_TIMER,
 					      hr_dev->caps.cqc_timer_entry_sz,
 					      hr_dev->caps.num_cqc_timer, 1);
 		if (ret) {
 			dev_err(dev,
-			      "Failed to init CQC timer memory, aborting.\n");
+				"Failed to init CQC timer memory, aborting.\n");
 			goto err_unmap_qpc_timer;
 		}
 	}
@@ -727,8 +727,7 @@ static int hns_roce_init_hem(struct hns_roce_dev *hr_dev)
 
 err_unmap_qpc_timer:
 	if (hr_dev->caps.qpc_timer_entry_sz)
-		hns_roce_cleanup_hem_table(hr_dev,
-					   &hr_dev->qpc_timer_table);
+		hns_roce_cleanup_hem_table(hr_dev, &hr_dev->qpc_timer_table);
 
 err_unmap_ctx:
 	if (hr_dev->caps.sccc_entry_sz)
@@ -863,6 +862,50 @@ err_uar_table_free:
 	return ret;
 }
 
+static void check_and_get_armed_cq(struct list_head *cq_list, struct ib_cq *cq)
+{
+	struct hns_roce_cq *hr_cq = to_hr_cq(cq);
+	unsigned long flags;
+
+	spin_lock_irqsave(&hr_cq->lock, flags);
+	if (cq->comp_handler) {
+		if (!hr_cq->is_armed) {
+			hr_cq->is_armed = 1;
+			list_add_tail(&hr_cq->node, cq_list);
+		}
+	}
+	spin_unlock_irqrestore(&hr_cq->lock, flags);
+}
+
+void hns_roce_handle_device_err(struct hns_roce_dev *hr_dev)
+{
+	struct hns_roce_qp *hr_qp;
+	struct hns_roce_cq *hr_cq;
+	struct list_head cq_list;
+	unsigned long flags_qp;
+	unsigned long flags;
+
+	INIT_LIST_HEAD(&cq_list);
+
+	spin_lock_irqsave(&hr_dev->qp_list_lock, flags);
+	list_for_each_entry(hr_qp, &hr_dev->qp_list, node) {
+		spin_lock_irqsave(&hr_qp->sq.lock, flags_qp);
+		if (hr_qp->sq.tail != hr_qp->sq.head)
+			check_and_get_armed_cq(&cq_list, hr_qp->ibqp.send_cq);
+		spin_unlock_irqrestore(&hr_qp->sq.lock, flags_qp);
+
+		spin_lock_irqsave(&hr_qp->rq.lock, flags_qp);
+		if ((!hr_qp->ibqp.srq) && (hr_qp->rq.tail != hr_qp->rq.head))
+			check_and_get_armed_cq(&cq_list, hr_qp->ibqp.recv_cq);
+		spin_unlock_irqrestore(&hr_qp->rq.lock, flags_qp);
+	}
+
+	list_for_each_entry(hr_cq, &cq_list, node)
+		hns_roce_cq_completion(hr_dev, hr_cq->cqn);
+
+	spin_unlock_irqrestore(&hr_dev->qp_list_lock, flags);
+}
+
 int hns_roce_init(struct hns_roce_dev *hr_dev)
 {
 	int ret;
@@ -933,6 +976,9 @@ int hns_roce_init(struct hns_roce_dev *hr_dev)
 		}
 	}
 
+	INIT_LIST_HEAD(&hr_dev->qp_list);
+	spin_lock_init(&hr_dev->qp_list_lock);
+
 	ret = hns_roce_register_device(hr_dev);
 	if (ret)
 		goto error_failed_register_device;