driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I9SI44

---------------------------------------------------------------------- During the reset, issuing the cmdq command to destroy resources such as qp, cq, and mr may fail, and the destruction failure information is printed. When a large number of resources fail to be destroyed, printk() causes multiple threads to compete for a global log buffer spinlock. As a result, the CPU performance deteriorates and the CPU is stuck. To fix it, use a function like printk_limit_rate() to limit the print frequency of printk().

Fixes: 9a4435375cd1 ("IB/hns: Add driver files for hns RoCE driver") Fixes: c7bcb13442e1 ("RDMA/hns: Add SRQ support for hip08 kernel mode") Fixes: 626903e9355b ("RDMA/hns: Add support for reporting wc as software mode") Fixes: 6eef524201de ("RDMA/hns: Replace not intuitive function/macro names") Fixes: 6f5f556d3795 ("RDMA/hns: Use the reserved loopback QPs to free MR before destroying MPT") Fixes: 357f34294686 ("RDMA/hns: Simplify the state judgment code of qp") Fixes: 926a01dc000d ("RDMA/hns: Add QP operations support for hip08 SoC") Fixes: 0273952c5e6e ("RDMA/hns: Add method to detach WQE buffer") Fixes: 9c9edf689a60 ("RDMA/hns: Remove redundant DFX file and DFX ops structure") Signed-off-by: Chengchang Tang tangchengchang@huawei.com Signed-off-by: Guofeng Yue yueguofeng@h-partners.com --- drivers/infiniband/hw/hns/hns_roce_cq.c | 5 ++- drivers/infiniband/hw/hns/hns_roce_hem.c | 5 ++- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 50 +++++++++++++--------- drivers/infiniband/hw/hns/hns_roce_mr.c | 2 +- drivers/infiniband/hw/hns/hns_roce_srq.c | 5 ++- 5 files changed, 39 insertions(+), 28 deletions(-)

diff --git a/drivers/infiniband/hw/hns/hns_roce_cq.c b/drivers/infiniband/hw/hns/hns_roce_cq.c index 48c25b30a395..695be916c91d 100644 --- a/drivers/infiniband/hw/hns/hns_roce_cq.c +++ b/drivers/infiniband/hw/hns/hns_roce_cq.c @@ -179,8 +179,9 @@ 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) - dev_err(dev, "DESTROY_CQ failed (%d) for CQN %06lx\n", ret, - hr_cq->cqn); + dev_err_ratelimited(dev, + "DESTROY_CQ failed (%d) for CQN %06lx\n", + ret, hr_cq->cqn); if (ret == -EBUSY) hr_cq->delayed_destroy_flag = true;

diff --git a/drivers/infiniband/hw/hns/hns_roce_hem.c b/drivers/infiniband/hw/hns/hns_roce_hem.c index 709a446465a4..ac8aaddee248 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hem.c +++ b/drivers/infiniband/hw/hns/hns_roce_hem.c @@ -712,8 +712,9 @@ 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(dev, "failed to clear HEM base address, ret = %d.\n", - ret); + dev_warn_ratelimited(dev, + "failed to clear HEM base address, ret = %d.\n", + ret);

hns_roce_free_hem(hr_dev, table->hem[i]); table->hem[i] = NULL; diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 11800d07e11c..44c20f0e6c10 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -451,19 +451,21 @@ 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(ibdev, "failed to post WQE, QP state %u!\n", - hr_qp->state); + ibdev_err_ratelimited(ibdev, + "failed to post WQE, QP state %u!\n", + hr_qp->state); return -EINVAL; } else if (unlikely(hr_dev->state >= HNS_ROCE_DEVICE_STATE_RST_DOWN)) { - ibdev_err(ibdev, "failed to post WQE, dev state %d!\n", - hr_dev->state); + ibdev_err_ratelimited(ibdev, + "failed to post WQE, dev state %d!\n", + hr_dev->state); return -EIO; }

if (check_dca_attach_enable(hr_qp)) { ret = dca_attach_qp_buf(hr_dev, hr_qp); if (unlikely(ret)) { - ibdev_err(ibdev, + ibdev_err_ratelimited(ibdev, "failed to attach DCA for QP-%ld send!\n", hr_qp->qpn); return ret; @@ -3813,8 +3815,9 @@ 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(ibdev, "failed to post wqe for free mr, ret = %d.\n", - ret); + ibdev_err_ratelimited(ibdev, + "failed to post wqe for free mr, ret = %d.\n", + ret); return ret; }

@@ -3853,7 +3856,7 @@ static void free_mr_send_cmd_to_hw(struct hns_roce_dev *hr_dev)

ret = free_mr_post_send_lp_wqe(hr_qp); if (ret) { - ibdev_err(ibdev, + ibdev_err_ratelimited(ibdev, "failed to send wqe (qp:0x%lx) for free mr, ret = %d.\n", hr_qp->qpn, ret); break; @@ -3866,14 +3869,14 @@ static void free_mr_send_cmd_to_hw(struct hns_roce_dev *hr_dev) while (cqe_cnt) { npolled = hns_roce_v2_poll_cq(&free_mr->rsv_cq->ib_cq, cqe_cnt, wc); if (npolled < 0) { - ibdev_err(ibdev, + ibdev_err_ratelimited(ibdev, "failed to poll cqe for free mr, remain %d cqe.\n", cqe_cnt); goto out; }

if (time_after(jiffies, end)) { - ibdev_err(ibdev, + ibdev_err_ratelimited(ibdev, "failed to poll cqe for free mr and timeout, remain %d cqe.\n", cqe_cnt); goto out; @@ -5535,7 +5538,8 @@ static int hns_roce_v2_set_abs_fields(struct ib_qp *ibqp, int ret = 0;

if (!check_qp_state(cur_state, new_state)) { - ibdev_err(&hr_dev->ib_dev, "Illegal state for QP!\n"); + ibdev_err_ratelimited(&hr_dev->ib_dev, + "Illegal state for QP!\n"); return -EINVAL; }

@@ -5795,7 +5799,7 @@ static int hns_roce_v2_modify_qp(struct ib_qp *ibqp, /* SW pass context to HW */ ret = hns_roce_v2_qp_modify(hr_dev, context, qpc_mask, hr_qp); if (ret) { - ibdev_err(ibdev, "failed to modify QP, ret = %d.\n", ret); + ibdev_err_ratelimited(ibdev, "failed to modify QP, ret = %d.\n", ret); goto out; }

@@ -5980,7 +5984,9 @@ static int hns_roce_v2_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr,

ret = hns_roce_v2_query_qpc(hr_dev, hr_qp->qpn, &context); if (ret) { - ibdev_err(ibdev, "failed to query QPC, ret = %d.\n", ret); + ibdev_err_ratelimited(ibdev, + "failed to query QPC, ret = %d.\n", + ret); ret = -EINVAL; goto out; } @@ -5988,7 +5994,7 @@ static int hns_roce_v2_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, state = hr_reg_read(&context, QPC_QP_ST); tmp_qp_state = to_ib_qp_st((enum hns_roce_v2_qp_state)state); if (tmp_qp_state == -1) { - ibdev_err(ibdev, "Illegal ib_qp_state\n"); + ibdev_err_ratelimited(ibdev, "Illegal ib_qp_state\n"); ret = -EINVAL; goto out; } @@ -6070,7 +6076,9 @@ static bool hns_roce_v2_chk_dca_buf_inactive(struct hns_roce_dev *hr_dev,

ret = hns_roce_v2_query_qpc(hr_dev, hr_qp->qpn, &context); if (ret) { - ibdev_err(ibdev, "failed to query DCA QPC, ret = %d.\n", ret); + ibdev_err_ratelimited(ibdev, + "failed to query DCA QPC, ret = %d.\n", + ret); return false; }

@@ -6114,7 +6122,7 @@ int hns_roce_v2_destroy_qp_common(struct hns_roce_dev *hr_dev, ret = hns_roce_v2_modify_qp(&hr_qp->ibqp, NULL, 0, hr_qp->state, IB_QPS_RESET, udata); if (ret) - ibdev_err(ibdev, + ibdev_err_ratelimited(ibdev, "failed to modify QP to RST, ret = %d.\n", ret); } @@ -6152,7 +6160,7 @@ int hns_roce_v2_destroy_qp(struct ib_qp *ibqp, struct ib_udata *udata)

ret = hns_roce_v2_destroy_qp_common(hr_dev, hr_qp, udata); if (ret) - ibdev_err(&hr_dev->ib_dev, + ibdev_err_ratelimited(&hr_dev->ib_dev, "failed to destroy QP, QPN = 0x%06lx, ret = %d.\n", hr_qp->qpn, ret);

@@ -6451,7 +6459,7 @@ static int hns_roce_v2_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period) HNS_ROCE_CMD_MODIFY_CQC, hr_cq->cqn); hns_roce_free_cmd_mailbox(hr_dev, mailbox); if (ret) - ibdev_err(&hr_dev->ib_dev, + ibdev_err_ratelimited(&hr_dev->ib_dev, "failed to process cmd when modifying CQ, ret = %d.\n", ret);

@@ -6477,9 +6485,9 @@ static int hns_roce_v2_query_cqc(struct hns_roce_dev *hr_dev, u32 cqn, ret = hns_roce_cmd_mbox(hr_dev, 0, mailbox->dma, HNS_ROCE_CMD_QUERY_CQC, cqn); if (ret) { - ibdev_err(&hr_dev->ib_dev, - "failed to process cmd when querying CQ, ret = %d.\n", - ret); + ibdev_err_ratelimited(&hr_dev->ib_dev, + "failed to process cmd when querying CQ, ret = %d.\n", + ret); goto err_mailbox; }

diff --git a/drivers/infiniband/hw/hns/hns_roce_mr.c b/drivers/infiniband/hw/hns/hns_roce_mr.c index 68049467a721..5af4e365afa5 100644 --- a/drivers/infiniband/hw/hns/hns_roce_mr.c +++ b/drivers/infiniband/hw/hns/hns_roce_mr.c @@ -151,7 +151,7 @@ static void hns_roce_mr_free(struct hns_roce_dev *hr_dev, struct hns_roce_mr *mr key_to_hw_index(mr->key) & (hr_dev->caps.num_mtpts - 1)); if (ret) - ibdev_warn(ibdev, "failed to destroy mpt, ret = %d.\n", + ibdev_warn_ratelimited(ibdev, "failed to destroy mpt, ret = %d.\n", ret); if (ret == -EBUSY) mr->delayed_destroy_flag = true; diff --git a/drivers/infiniband/hw/hns/hns_roce_srq.c b/drivers/infiniband/hw/hns/hns_roce_srq.c index 4779232ee318..1e57395f1148 100644 --- a/drivers/infiniband/hw/hns/hns_roce_srq.c +++ b/drivers/infiniband/hw/hns/hns_roce_srq.c @@ -162,8 +162,9 @@ static void free_srqc(struct hns_roce_dev *hr_dev, struct hns_roce_srq *srq) ret = hns_roce_destroy_hw_ctx(hr_dev, HNS_ROCE_CMD_DESTROY_SRQ, srq->srqn); if (ret) - dev_err(hr_dev->dev, "DESTROY_SRQ failed (%d) for SRQN %06lx\n", - ret, srq->srqn); + dev_err_ratelimited(hr_dev->dev, + "DESTROY_SRQ failed (%d) for SRQN %06lx\n", + ret, srq->srqn); if (ret == -EBUSY) srq->delayed_destroy_flag = true;

driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I9SI44

---------------------------------------------------------------------- Currently, the driver does not actively read the default parameters of scc, that is, the default parameters are all 0. Therefore, if the user directly modifies any parameter value without actively reading the scc default parameters, other parameters will become 0.

This patch will read and save the default value of scc immediately after the driver registers sysfs to avoid the above problems.

Fixes: 523f34d81ea7 ("RDMA/hns: Support congestion control algorithm parameter configuration") Signed-off-by: Chengchang Tang tangchengchang@huawei.com Signed-off-by: Guofeng Yue yueguofeng@h-partners.com --- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 3 +++ drivers/infiniband/hw/hns/hns_roce_sysfs.c | 20 +++++++++++++++++++- 2 files changed, 22 insertions(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 00a13b36ce2d..5df6a07cc8f7 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -7382,6 +7382,9 @@ static int hns_roce_v2_query_scc_param(struct hns_roce_dev *hr_dev, struct hns_roce_port *pdata; int ret;

+ if (hr_dev->pci_dev->revision <= PCI_REVISION_ID_HIP08) + return -EOPNOTSUPP; + if (port_num > hr_dev->caps.num_ports) { ibdev_err_ratelimited(&hr_dev->ib_dev, "invalid port num %u.\n", port_num); diff --git a/drivers/infiniband/hw/hns/hns_roce_sysfs.c b/drivers/infiniband/hw/hns/hns_roce_sysfs.c index a51423b94dc1..f92e5c13ab92 100644 --- a/drivers/infiniband/hw/hns/hns_roce_sysfs.c +++ b/drivers/infiniband/hw/hns/hns_roce_sysfs.c @@ -26,6 +26,22 @@ static void scc_param_config_work(struct work_struct *work) scc_param->algo_type); }

+static void get_default_scc_param(struct hns_roce_dev *hr_dev, + struct hns_roce_port *pdata) +{ + int ret; + int i; + + for (i = 0; i < HNS_ROCE_SCC_ALGO_TOTAL; i++) { + pdata->scc_param[i].timestamp = jiffies; + ret = hr_dev->hw->query_scc_param(hr_dev, pdata->port_num, i); + if (ret && ret != -EOPNOTSUPP) + ibdev_warn_ratelimited(&hr_dev->ib_dev, + "failed to get default parameters of scc algo %d, ret = %d.\n", + i, ret); + } +} + static int alloc_scc_param(struct hns_roce_dev *hr_dev, struct hns_roce_port *pdata) { @@ -39,7 +55,6 @@ static int alloc_scc_param(struct hns_roce_dev *hr_dev,

for (i = 0; i < HNS_ROCE_SCC_ALGO_TOTAL; i++) { scc_param[i].algo_type = i; - scc_param[i].timestamp = jiffies; scc_param[i].hr_dev = hr_dev; scc_param[i].port_num = pdata->port_num; INIT_DELAYED_WORK(&scc_param[i].scc_cfg_dwork, @@ -47,6 +62,9 @@ static int alloc_scc_param(struct hns_roce_dev *hr_dev, }

pdata->scc_param = scc_param; + + get_default_scc_param(hr_dev, pdata); + return 0; }

From: Junxian Huang huangjunxian6@hisilicon.com

driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I9SI44

---------------------------------------------------------------------- DCA mem size from userspace should be aligned to HNS_HW_PAGE_SIZE as the DCA buffer is managed in the form of pages. Otherwise it may lead to a mis-calculation of pages, which is basically dividing the size by HNS_HW_PAGE_SIZE. Besides, zero size is invalid since it will lead to the failure of page allocation in ib_umem_get().

Fixes: f44a2f97d82a ("RDMA/hns: Introduce DCA for RC QP") Signed-off-by: Junxian Huang huangjunxian6@hisilicon.com Signed-off-by: Guofeng Yue yueguofeng@h-partners.com --- drivers/infiniband/hw/hns/hns_roce_dca.c | 5 +++++ 1 file changed, 5 insertions(+)

diff --git a/drivers/infiniband/hw/hns/hns_roce_dca.c b/drivers/infiniband/hw/hns/hns_roce_dca.c index e88603c96ae4..ddc21f10b923 100644 --- a/drivers/infiniband/hw/hns/hns_roce_dca.c +++ b/drivers/infiniband/hw/hns/hns_roce_dca.c @@ -1529,6 +1529,11 @@ static int UVERBS_HANDLER(HNS_IB_METHOD_DCA_MEM_REG)( if (ret) return ret;

+ if (!init_attr.size) + return -EINVAL; + + init_attr.size = hr_hw_page_align(init_attr.size); + mem = alloc_dca_mem(to_hr_dca_ctx(hr_dev, uctx)); if (!mem) return -ENOMEM;

driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I9SI44

---------------------------------------------------------------------- xa_lock of qp_table_xa may be acquired in interrupt context. Replace xa_lock()/xa_unlock() with xa_lock_irq()/xa_unlock_irq() to avoid deadlock.

Fixes: a21781182f77 ("RDMA/hns: Add debugfs support for DCA") Signed-off-by: Chengchang Tang tangchengchang@huawei.com Signed-off-by: Guofeng Yue yueguofeng@h-partners.com --- drivers/infiniband/hw/hns/hns_roce_debugfs.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/infiniband/hw/hns/hns_roce_debugfs.c b/drivers/infiniband/hw/hns/hns_roce_debugfs.c index c8294a836ee3..85fa88e8a657 100644 --- a/drivers/infiniband/hw/hns/hns_roce_debugfs.c +++ b/drivers/infiniband/hw/hns/hns_roce_debugfs.c @@ -353,19 +353,19 @@ static void dca_stats_dev_qp_in_seqfile(struct hns_roce_dev *hr_dev, seq_printf(file, "%-10s %-10s %-10s %s\n", "QPN", "Size(kB)", "PID", "State");

- xa_lock(&hr_dev->qp_table_xa); + xa_lock_irq(&hr_dev->qp_table_xa); xa_for_each(&hr_dev->qp_table_xa, id, hr_qp) { stats.total_size = 0; dca_setup_qp_stats(hr_qp, &stats); if (!stats.total_size) continue;

- xa_unlock(&hr_dev->qp_table_xa); + xa_unlock_irq(&hr_dev->qp_table_xa); seq_printf(file, "%-10u %-10u %-10s %-s\n", stats.qpn, stats.total_size / KB, stats.name, stats.state); - xa_lock(&hr_dev->qp_table_xa); + xa_lock_irq(&hr_dev->qp_table_xa); } - xa_unlock(&hr_dev->qp_table_xa); + xa_unlock_irq(&hr_dev->qp_table_xa); }

static void dca_stats_ctx_qp_in_seqfile(struct hns_roce_dev *hr_dev, @@ -391,11 +391,11 @@ static void dca_stats_ctx_qp_in_seqfile(struct hns_roce_dev *hr_dev, dca_ctx_stats_qp(ctx, bitmap, nbits); for_each_set_bit(qpn, bitmap, nbits) { stats.total_size = 0; - xa_lock(&hr_dev->qp_table_xa); + xa_lock_irq(&hr_dev->qp_table_xa); hr_qp = __hns_roce_qp_lookup(hr_dev, qpn); if (hr_qp) dca_setup_qp_stats(hr_qp, &stats); - xa_unlock(&hr_dev->qp_table_xa); + xa_unlock_irq(&hr_dev->qp_table_xa); if (!stats.total_size) continue;