Unverified Commit a1fd264a authored by openeuler-ci-bot's avatar openeuler-ci-bot Committed by Gitee
Browse files

!13860 [2024.12.2]bugfix rdma to OLK-5.10 

Merge Pull Request from: @zhaojiahui111 
 
| commitID     | commit message                                                             | conflict |
|--------------|----------------------------------------------------------------------------|----|
| 571e4ab8a45e | RDMA/hns: Fix an AEQE overflow error caused by untimely update of eq_db_ci | NA  |
| 323275ac2ff1 | RDMA/hns: Fix cpu stuck caused by printings during reset                   | NA  |
 
 
Link:https://gitee.com/openeuler/kernel/pulls/13860

 

Reviewed-by: default avatarLi Nan <linan122@huawei.com>
Signed-off-by: default avatarLi Nan <linan122@huawei.com>
parents a2b4d661 80a0ecef

drivers/infiniband/hw/hns/hns_roce_cq.c

+2 −5
Original line number Diff line number Diff line
@@ -233,12 +233,9 @@ static void free_cqc(struct hns_roce_dev *hr_dev, struct hns_roce_cq *hr_cq) 


	ret = hns_roce_destroy_hw_ctx(hr_dev, HNS_ROCE_CMD_DESTROY_CQC,

				      hr_cq->cqn);

	if (ret) {

		hr_cq->delayed_destroy_flag = true;

		dev_err_ratelimited(dev,

				    "DESTROY_CQ failed (%d) for CQN %06lx\n",

	if (ret)

		dev_err_ratelimited(dev, "DESTROY_CQ failed (%d) for CQN %06lx\n",

				    ret, hr_cq->cqn);

	}



	xa_erase_irq(&cq_table->array, hr_cq->cqn);

drivers/infiniband/hw/hns/hns_roce_device.h

+1 −0
Original line number Diff line number Diff line
@@ -1538,6 +1538,7 @@ void hns_roce_cq_completion(struct hns_roce_dev *hr_dev, u32 cqn); 
void hns_roce_cq_event(struct hns_roce_dev *hr_dev, u32 cqn, int event_type);

void flush_cqe(struct hns_roce_dev *dev, struct hns_roce_qp *qp);

void hns_roce_qp_event(struct hns_roce_dev *hr_dev, u32 qpn, int event_type);

void hns_roce_flush_cqe(struct hns_roce_dev *hr_dev, u32 qpn);

void hns_roce_srq_event(struct hns_roce_dev *hr_dev, u32 srqn, int event_type);

u8 hns_get_gid_index(struct hns_roce_dev *hr_dev, u8 port, int gid_index);

void hns_roce_handle_device_err(struct hns_roce_dev *hr_dev);

drivers/infiniband/hw/hns/hns_roce_hem.c

+2 −3
Original line number Diff line number Diff line
@@ -713,8 +713,7 @@ void hns_roce_table_put(struct hns_roce_dev *hr_dev, 


	ret = hr_dev->hw->clear_hem(hr_dev, table, obj, HEM_HOP_STEP_DIRECT);

	if (ret)

		dev_warn_ratelimited(dev,

			"failed to clear HEM base address, ret = %d.\n",

		dev_warn_ratelimited(dev, "failed to clear HEM base address, ret = %d.\n",

				     ret);



	hns_roce_free_hem(hr_dev, table->hem[i]);

drivers/infiniband/hw/hns/hns_roce_hw_v2.c

+28 −36
Original line number Diff line number Diff line
@@ -451,17 +451,10 @@ static int check_send_valid(struct hns_roce_dev *hr_dev, 


	if (unlikely(hr_qp->state == IB_QPS_RESET ||

		     hr_qp->state == IB_QPS_INIT ||

		     hr_qp->state == IB_QPS_RTR)) {

		ibdev_err_ratelimited(ibdev,

				      "failed to post WQE, QP state %u!\n",

				      hr_qp->state);

		     hr_qp->state == IB_QPS_RTR))

		return -EINVAL;

	} else if (unlikely(hr_dev->state >= HNS_ROCE_DEVICE_STATE_RST_DOWN)) {

		ibdev_err_ratelimited(ibdev,

				      "failed to post WQE, dev state %d!\n",

				      hr_dev->state);

	else if (unlikely(hr_dev->state >= HNS_ROCE_DEVICE_STATE_RST_DOWN))

		return -EIO;

	}



	if (check_dca_attach_enable(hr_qp)) {

		ret = dca_attach_qp_buf(hr_dev, hr_qp);
@@ -3158,7 +3151,7 @@ static int free_mr_modify_rsv_qp(struct hns_roce_dev *hr_dev, 
	ret = hr_dev->hw->modify_qp(&hr_qp->ibqp, attr, mask, IB_QPS_INIT,

				    IB_QPS_INIT, NULL);

	if (ret) {

		ibdev_err(ibdev, "failed to modify qp to init, ret = %d.\n",

		ibdev_err_ratelimited(ibdev, "failed to modify qp to init, ret = %d.\n",

				      ret);

		return ret;

	}
@@ -3848,8 +3841,7 @@ static int free_mr_post_send_lp_wqe(struct hns_roce_qp *hr_qp) 


	ret = hns_roce_v2_post_send(&hr_qp->ibqp, send_wr, &bad_wr);

	if (ret) {

		ibdev_err_ratelimited(ibdev,

			"failed to post wqe for free mr, ret = %d.\n",

		ibdev_err_ratelimited(ibdev, "failed to post wqe for free mr, ret = %d.\n",

				      ret);

		return ret;

	}
@@ -5683,11 +5675,8 @@ static int hns_roce_v2_set_abs_fields(struct ib_qp *ibqp, 
	struct hns_roce_dev *hr_dev = to_hr_dev(ibqp->device);

	int ret = 0;



	if (!check_qp_state(cur_state, new_state)) {

		ibdev_err_ratelimited(&hr_dev->ib_dev,

				      "Illegal state for QP!\n");

	if (!check_qp_state(cur_state, new_state))

		return -EINVAL;

	}



	if (cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) {

		memset(qpc_mask, 0, hr_dev->caps.qpc_sz);
@@ -6267,13 +6256,11 @@ int hns_roce_v2_destroy_qp_common(struct hns_roce_dev *hr_dev, 
		/* Modify qp to reset before destroying qp */

		ret = hns_roce_v2_modify_qp(&hr_qp->ibqp, NULL, 0,

					    hr_qp->state, IB_QPS_RESET, udata);

		if (ret) {

			hr_qp->delayed_destroy_flag = true;

		if (ret)

			ibdev_err_ratelimited(ibdev,

					      "failed to modify QP to RST, ret = %d.\n",

					      ret);

	}

	}



	send_cq = hr_qp->ibqp.send_cq ? to_hr_cq(hr_qp->ibqp.send_cq) : NULL;

	recv_cq = hr_qp->ibqp.recv_cq ? to_hr_cq(hr_qp->ibqp.recv_cq) : NULL;
@@ -6780,7 +6767,12 @@ static void hns_roce_irq_work_handle(struct work_struct *work) 
	case HNS_ROCE_EVENT_TYPE_PATH_MIG_FAILED:

	case HNS_ROCE_EVENT_TYPE_COMM_EST:

	case HNS_ROCE_EVENT_TYPE_SQ_DRAINED:

	case HNS_ROCE_EVENT_TYPE_WQ_CATAS_ERROR:

	case HNS_ROCE_EVENT_TYPE_SRQ_LAST_WQE_REACH:

	case HNS_ROCE_EVENT_TYPE_INV_REQ_LOCAL_WQ_ERROR:

	case HNS_ROCE_EVENT_TYPE_LOCAL_WQ_ACCESS_ERROR:

	case HNS_ROCE_EVENT_TYPE_XRCD_VIOLATION:

	case HNS_ROCE_EVENT_TYPE_INVALID_XRCETH:

		hns_roce_qp_event(hr_dev, queue_num, event_type);

		break;

	case HNS_ROCE_EVENT_TYPE_SRQ_LIMIT_REACH:
@@ -6879,7 +6871,7 @@ static irqreturn_t hns_roce_v2_aeq_int(struct hns_roce_dev *hr_dev, 
		case HNS_ROCE_EVENT_TYPE_LOCAL_WQ_ACCESS_ERROR:

		case HNS_ROCE_EVENT_TYPE_XRCD_VIOLATION:

		case HNS_ROCE_EVENT_TYPE_INVALID_XRCETH:

			hns_roce_qp_event(hr_dev, queue_num, event_type);

			hns_roce_flush_cqe(hr_dev, queue_num);

			break;

		case HNS_ROCE_EVENT_TYPE_MB:

			hns_roce_cmd_event(hr_dev,

drivers/infiniband/hw/hns/hns_roce_mr.c

+2 −4
Original line number Diff line number Diff line
@@ -152,12 +152,10 @@ static void hns_roce_mr_free(struct hns_roce_dev *hr_dev, struct hns_roce_mr *mr 
		ret = hns_roce_destroy_hw_ctx(hr_dev, HNS_ROCE_CMD_DESTROY_MPT,

					      key_to_hw_index(mr->key) &

					      (hr_dev->caps.num_mtpts - 1));

		if (ret) {

			mr->delayed_destroy_flag = true;

		if (ret)

			ibdev_warn_ratelimited(ibdev, "failed to destroy mpt, ret = %d.\n",

					       ret);

	}

	}



	free_mr_pbl(hr_dev, mr);

	free_mr_key(hr_dev, mr);