Unverified Commit 6ad884d5 authored by openeuler-ci-bot's avatar openeuler-ci-bot Committed by Gitee
Browse files

!14109 RDMA/hns: backport some bugfix from maillist linux 

Merge Pull Request from: @ci-robot 
 
PR sync from: Chengchang Tang <tangchengchang@huawei.com>
https://mailweb.openeuler.org/hyperkitty/list/kernel@openeuler.org/message/AZ5FFXINGXSCPF5QN437P4XOJSDHKUTQ/ 
From: Xinghai Cen <cenxinghai@h-partners.com>

[RDMA/hns] backport some bugfix from maillist linux:

Xinghai Cen (1):
  Revert "RDMA/hns: Fix an AEQE overflow error caused by untimely update
    of eq_db_ci"

wenglianfa (2):
  RDMA/hns: Fix flush cqe error when racing with destroy qp
  RDMA/hns: Fix an AEQE overflow error caused by untimely update of
    eq_db_ci


-- 
2.33.0
 
https://gitee.com/openeuler/kernel/issues/IB8NKF 
 
Link:https://gitee.com/openeuler/kernel/pulls/14109

 

Reviewed-by: default avatarChengchang Tang <tangchengchang@huawei.com>
Reviewed-by: default avatarLi Nan <linan122@huawei.com>
Signed-off-by: default avatarLi Nan <linan122@huawei.com>
parents 36ade961 9a72d9ef

drivers/infiniband/hw/hns/hns_roce_device.h

+1 −0
Original line number Diff line number Diff line
@@ -1540,6 +1540,7 @@ 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_hw_v2.c

+11 −1
Original line number Diff line number Diff line
@@ -6332,6 +6332,11 @@ int hns_roce_v2_destroy_qp(struct ib_qp *ibqp, struct ib_udata *udata) 
	struct hns_roce_qp *hr_qp = to_hr_qp(ibqp);

	unsigned long flags;

	int ret;

	/* Make sure flush_cqe() is completed */

	spin_lock_irqsave(&hr_qp->flush_lock, flags);

	set_bit(HNS_ROCE_STOP_FLUSH_FLAG, &hr_qp->flush_flag);

	spin_unlock_irqrestore(&hr_qp->flush_lock, flags);

	flush_work(&hr_qp->flush_work.work);



	if (hr_qp->congest_type == HNS_ROCE_CONGEST_TYPE_DIP)

		put_dip_ctx_idx(hr_dev, hr_qp);
@@ -6787,7 +6792,12 @@ 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:
@@ -6886,7 +6896,7 @@ 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_qp.c

+35 −19
Original line number Diff line number Diff line
@@ -39,6 +39,25 @@ 
#include "hns_roce_hem.h"

#include "hns_roce_dca.h"



static struct hns_roce_qp *hns_roce_qp_lookup(struct hns_roce_dev *hr_dev,

					     u32 qpn)

{

	struct device *dev = hr_dev->dev;

	struct hns_roce_qp *qp;

	unsigned long flags;



	xa_lock_irqsave(&hr_dev->qp_table_xa, flags);

	qp = __hns_roce_qp_lookup(hr_dev, qpn);

	if (qp)

		refcount_inc(&qp->refcount);

	xa_unlock_irqrestore(&hr_dev->qp_table_xa, flags);



	if (!qp)

		dev_warn(dev, "async event for bogus QP %08x\n", qpn);



	return qp;

}



static void flush_work_handle(struct work_struct *work)

{

	struct hns_roce_work *flush_work = container_of(work,
@@ -102,31 +121,28 @@ 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)

{

	struct device *dev = hr_dev->dev;

	struct hns_roce_qp *qp;



	xa_lock(&hr_dev->qp_table_xa);

	qp = __hns_roce_qp_lookup(hr_dev, qpn);

	if (qp)

		refcount_inc(&qp->refcount);

	xa_unlock(&hr_dev->qp_table_xa);



	if (!qp) {

		dev_warn(dev, "Async event for bogus QP %08x\n", qpn);

	qp = hns_roce_qp_lookup(hr_dev, qpn);

	if (!qp)

		return;

	}



	if (event_type == HNS_ROCE_EVENT_TYPE_WQ_CATAS_ERROR ||

	    event_type == HNS_ROCE_EVENT_TYPE_INV_REQ_LOCAL_WQ_ERROR ||

	    event_type == HNS_ROCE_EVENT_TYPE_LOCAL_WQ_ACCESS_ERROR ||

	    event_type == HNS_ROCE_EVENT_TYPE_XRCD_VIOLATION ||

	    event_type == HNS_ROCE_EVENT_TYPE_INVALID_XRCETH) {

		qp->state = IB_QPS_ERR;

	qp->event(qp, (enum hns_roce_event)event_type);



		flush_cqe(hr_dev, qp);

	if (refcount_dec_and_test(&qp->refcount))

		complete(&qp->free);

}



	qp->event(qp, (enum hns_roce_event)event_type);

void hns_roce_flush_cqe(struct hns_roce_dev *hr_dev, u32 qpn)

{

	struct hns_roce_qp *qp;



	qp = hns_roce_qp_lookup(hr_dev, qpn);

	if (!qp)

		return;



	qp->state = IB_QPS_ERR;

	flush_cqe(hr_dev, qp);



	if (refcount_dec_and_test(&qp->refcount))

		complete(&qp->free);