[PATCH OLK-5.10 0/8] Some bug fix patches for RDMA/hns to olk-5.10
From: Xinghai Cen <cenxinghai@h-partners.com> Some bug fix patches for RDMA/hns to olk-5.10: Junxian Huang (7): RDMA/hns: Fix DCA error path in alloc_wqe_buf() RDMA/hns: Fix unmatched condition in error path of alloc_user_qp_db() RDMA/hns: Fix invalid sq params not being blocked RDMA/hns: Fix a missing rollback in error path of hns_roce_create_qp_common() RDMA/hns: Fix missing xa_destroy() RDMA/hns: Fix wrong value of max_sge_rd RDMA/hns: Fix udca not unregistered when reset entry allocation failed Xinghai Cen (1): RDMA/hns: Fix default congestion control algorithm not set for kernel QP drivers/infiniband/hw/hns/hns_roce_alloc.c | 4 +++- drivers/infiniband/hw/hns/hns_roce_cq.c | 1 + drivers/infiniband/hw/hns/hns_roce_main.c | 12 +++++------ drivers/infiniband/hw/hns/hns_roce_qp.c | 23 +++++++++++++--------- 4 files changed, 24 insertions(+), 16 deletions(-) -- 2.33.0
driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IBSE4C ---------------------------------------------------------------------- When hns_roce_map_dca_safe_page() fails in alloc_wqe_buf(), hns_roce_disable_dca() should be called in error path. Fixes: 36f678e7fa0f ("RDMA/hns: Fix possible RAS when DCA is not attached") Signed-off-by: Junxian Huang <huangjunxian6@hisilicon.com> Signed-off-by: Xinghai Cen <cenxinghai@h-partners.com> --- drivers/infiniband/hw/hns/hns_roce_qp.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_qp.c b/drivers/infiniband/hw/hns/hns_roce_qp.c index cd3a67e66eff..f444787f0598 100644 --- a/drivers/infiniband/hw/hns/hns_roce_qp.c +++ b/drivers/infiniband/hw/hns/hns_roce_qp.c @@ -866,12 +866,13 @@ static int alloc_wqe_buf(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp, if (IS_ERR(hr_qp->mtr)) { ret = PTR_ERR(hr_qp->mtr); ibdev_err(ibdev, "failed to create WQE mtr, ret = %d.\n", ret); - if (dca_en) - hns_roce_disable_dca(hr_dev, hr_qp, udata); } else if (dca_en) { ret = hns_roce_map_dca_safe_page(hr_dev, hr_qp); } + if (ret && dca_en) + hns_roce_disable_dca(hr_dev, hr_qp, udata); + return ret; } -- 2.33.0
driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IBSE4C ---------------------------------------------------------------------- Currently the condition of unmapping sdb in error path is not exactly the same as the condition of mapping in alloc_user_qp_db(). This may cause a problem of unmapping an unmapped db in some case, such as when the QP is XRC TGT. Unified the two conditions. Fixes: 90ae0b57e4a5 ("RDMA/hns: Combine enable flags of qp") Signed-off-by: Junxian Huang <huangjunxian6@hisilicon.com> Signed-off-by: Xinghai Cen <cenxinghai@h-partners.com> --- drivers/infiniband/hw/hns/hns_roce_qp.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_qp.c b/drivers/infiniband/hw/hns/hns_roce_qp.c index f444787f0598..fe25ce02c0bf 100644 --- a/drivers/infiniband/hw/hns/hns_roce_qp.c +++ b/drivers/infiniband/hw/hns/hns_roce_qp.c @@ -1011,12 +1011,14 @@ static int alloc_user_qp_db(struct hns_roce_dev *hr_dev, struct hns_roce_ib_create_qp *ucmd, struct hns_roce_ib_create_qp_resp *resp) { + bool has_sdb = user_qp_has_sdb(hr_dev, init_attr, udata, resp, ucmd); struct hns_roce_ucontext *uctx = rdma_udata_to_drv_context(udata, struct hns_roce_ucontext, ibucontext); + bool has_rdb = user_qp_has_rdb(hr_dev, init_attr, udata, resp); struct ib_device *ibdev = &hr_dev->ib_dev; int ret; - if (user_qp_has_sdb(hr_dev, init_attr, udata, resp, ucmd)) { + if (has_sdb) { ret = hns_roce_db_map_user(uctx, ucmd->sdb_addr, &hr_qp->sdb); if (ret) { ibdev_err(ibdev, @@ -1027,7 +1029,7 @@ static int alloc_user_qp_db(struct hns_roce_dev *hr_dev, hr_qp->en_flags |= HNS_ROCE_QP_CAP_SQ_RECORD_DB; } - if (user_qp_has_rdb(hr_dev, init_attr, udata, resp)) { + if (has_rdb) { ret = hns_roce_db_map_user(uctx, ucmd->db_addr, &hr_qp->rdb); if (ret) { ibdev_err(ibdev, @@ -1041,7 +1043,7 @@ static int alloc_user_qp_db(struct hns_roce_dev *hr_dev, return 0; err_sdb: - if (hr_qp->en_flags & HNS_ROCE_QP_CAP_SQ_RECORD_DB) + if (has_sdb) hns_roce_db_unmap_user(uctx, &hr_qp->sdb, false); err_out: return ret; -- 2.33.0
driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IBSE4C ---------------------------------------------------------------------- SQ params from userspace are checked in by set_user_sq_size(). But when the check fails, the function doesn't return but instead keep running and overwrite 'ret'. As a result, the invalid params will not get blocked actually. Add a return right after the failed check. Besides, to keep coding style unified, move default_congest_type() before set_kernel_sq_size(). Fixes: 7a5b94597d47 ("RDMA/hns: Fix congestions control algorithm type for UD") Fixes: 09f1b7cb29b2 ("RDMA/hns: Support congestion control algorithm configuration at QP granularity") Signed-off-by: Junxian Huang <huangjunxian6@hisilicon.com> Signed-off-by: Xinghai Cen <cenxinghai@h-partners.com> --- drivers/infiniband/hw/hns/hns_roce_qp.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_qp.c b/drivers/infiniband/hw/hns/hns_roce_qp.c index fe25ce02c0bf..1f5c1acfce06 100644 --- a/drivers/infiniband/hw/hns/hns_roce_qp.c +++ b/drivers/infiniband/hw/hns/hns_roce_qp.c @@ -1326,17 +1326,17 @@ static int set_qp_param(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp, ibucontext); hr_qp->config = uctx->config; ret = set_user_sq_size(hr_dev, &init_attr->cap, hr_qp, ucmd); - if (ret) + if (ret) { ibdev_err(ibdev, "Failed to set user SQ size, ret = %d\n", ret); + return ret; + } ret = set_uqp_create_flag_param(hr_dev, hr_qp, init_attr, ucmd); if (ret) return ret; ret = set_congest_param(hr_dev, hr_qp, ucmd); - if (ret) - return ret; } else { if (init_attr->create_flags & IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK) { -- 2.33.0
driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IBSE4C ---------------------------------------------------------------------- When ib_copy_to_udata() fails in hns_roce_create_qp_common(), hns_roce_qp_remove() should be called in the error path to clean up resources in hns_roce_qp_store(). Fixes: 0f00571f9433 ("RDMA/hns: Use new SQ doorbell register for HIP09") Signed-off-by: Junxian Huang <huangjunxian6@hisilicon.com> Signed-off-by: Xinghai Cen <cenxinghai@h-partners.com> --- drivers/infiniband/hw/hns/hns_roce_qp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_qp.c b/drivers/infiniband/hw/hns/hns_roce_qp.c index 1f5c1acfce06..8144dbab3e32 100644 --- a/drivers/infiniband/hw/hns/hns_roce_qp.c +++ b/drivers/infiniband/hw/hns/hns_roce_qp.c @@ -1434,7 +1434,7 @@ static int hns_roce_create_qp_common(struct hns_roce_dev *hr_dev, min(udata->outlen, sizeof(resp))); if (ret) { ibdev_err(ibdev, "copy qp resp failed!\n"); - goto err_store; + goto err_flow_ctrl; } } -- 2.33.0
driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IBSE4C ---------------------------------------------------------------------- Add xa_destroy() for xarray in driver. Fixes: 5c1f167af112 ("RDMA/hns: Init SRQ table for hip08") Fixes: 27e19f451089 ("RDMA/hns: Convert cq_table to XArray") Fixes: 736b5a70db98 ("RDMA/hns: Convert qp_table_tree to XArray") Signed-off-by: Junxian Huang <huangjunxian6@hisilicon.com> Signed-off-by: Xinghai Cen <cenxinghai@h-partners.com> --- drivers/infiniband/hw/hns/hns_roce_alloc.c | 4 +++- drivers/infiniband/hw/hns/hns_roce_cq.c | 1 + drivers/infiniband/hw/hns/hns_roce_qp.c | 1 + 3 files changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_alloc.c b/drivers/infiniband/hw/hns/hns_roce_alloc.c index b3edc09c2dc7..27de813498b1 100644 --- a/drivers/infiniband/hw/hns/hns_roce_alloc.c +++ b/drivers/infiniband/hw/hns/hns_roce_alloc.c @@ -169,8 +169,10 @@ void hns_roce_cleanup_bitmap(struct hns_roce_dev *hr_dev) if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_XRC) ida_destroy(&hr_dev->xrcd_ida.ida); - if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_SRQ) + if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_SRQ) { ida_destroy(&hr_dev->srq_table.srq_ida.ida); + xa_destroy(&hr_dev->srq_table.xa); + } hns_roce_cleanup_qp_table(hr_dev); hns_roce_cleanup_cq_table(hr_dev); ida_destroy(&hr_dev->mr_table.mtpt_ida.ida); diff --git a/drivers/infiniband/hw/hns/hns_roce_cq.c b/drivers/infiniband/hw/hns/hns_roce_cq.c index a2f1e722d019..06f349bfc213 100644 --- a/drivers/infiniband/hw/hns/hns_roce_cq.c +++ b/drivers/infiniband/hw/hns/hns_roce_cq.c @@ -722,5 +722,6 @@ void hns_roce_cleanup_cq_table(struct hns_roce_dev *hr_dev) for (i = 0; i < HNS_ROCE_CQ_BANK_NUM; i++) ida_destroy(&hr_dev->cq_table.bank[i].ida); + xa_destroy(&hr_dev->cq_table.array); mutex_destroy(&hr_dev->cq_table.bank_mutex); } diff --git a/drivers/infiniband/hw/hns/hns_roce_qp.c b/drivers/infiniband/hw/hns/hns_roce_qp.c index 8144dbab3e32..71288df24b08 100644 --- a/drivers/infiniband/hw/hns/hns_roce_qp.c +++ b/drivers/infiniband/hw/hns/hns_roce_qp.c @@ -1860,6 +1860,7 @@ void hns_roce_cleanup_qp_table(struct hns_roce_dev *hr_dev) for (i = 0; i < HNS_ROCE_QP_BANK_NUM; i++) ida_destroy(&hr_dev->qp_table.bank[i].ida); xa_destroy(&hr_dev->qp_table.dip_xa); + xa_destroy(&hr_dev->qp_table_xa); mutex_destroy(&hr_dev->qp_table.bank_mutex); mutex_destroy(&hr_dev->qp_table.scc_mutex); } -- 2.33.0
driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IBSE4C ---------------------------------------------------------------------- There is no sge difference between READ and non-READ operations in hns RoCE. Set max_sge_rd to the same value as max_send_sge. Fixes: 9a4435375cd1 ("IB/hns: Add driver files for hns RoCE driver") Signed-off-by: Junxian Huang <huangjunxian6@hisilicon.com> Signed-off-by: Xinghai Cen <cenxinghai@h-partners.com> --- drivers/infiniband/hw/hns/hns_roce_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c index 13cace92f7ff..61586142426c 100644 --- a/drivers/infiniband/hw/hns/hns_roce_main.c +++ b/drivers/infiniband/hw/hns/hns_roce_main.c @@ -279,7 +279,7 @@ static int hns_roce_query_device(struct ib_device *ib_dev, IB_DEVICE_RC_RNR_NAK_GEN; props->max_send_sge = hr_dev->caps.max_sq_sg; props->max_recv_sge = hr_dev->caps.max_rq_sg; - props->max_sge_rd = 1; + props->max_sge_rd = hr_dev->caps.max_sq_sg; props->max_cq = hr_dev->caps.num_cqs; props->max_cqe = hr_dev->caps.max_cqes; props->max_mr = hr_dev->caps.num_mtpts; -- 2.33.0
driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IBSE4C ---------------------------------------------------------------------- hns_roce_unregister_udca() is not called when hns_roce_alloc_reset_entry() failed. Reorder the sequence of error path in hns_roce_alloc_ucontext() to fix it. Besides, to keep the coding constant, reorder hns_roce_dealloc_ucontext() accordingly as well. Fixes: 754c3a8adaa5 ("RDMA/hns: Use one CQ bank per context") Fixes: e8b1fec497a0 ("RDMA/hns: Kernel notify usr space to stop ring db") Signed-off-by: Junxian Huang <huangjunxian6@hisilicon.com> Signed-off-by: Xinghai Cen <cenxinghai@h-partners.com> --- drivers/infiniband/hw/hns/hns_roce_main.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c index 61586142426c..720e463ccea4 100644 --- a/drivers/infiniband/hw/hns/hns_roce_main.c +++ b/drivers/infiniband/hw/hns/hns_roce_main.c @@ -633,10 +633,10 @@ static int hns_roce_alloc_ucontext(struct ib_ucontext *uctx, return 0; error_fail_copy_to_udata: - hns_roce_unregister_udca(hr_dev, context); hns_roce_dealloc_reset_entry(context); error_fail_reset_entry: + hns_roce_unregister_udca(hr_dev, context); if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_CQ_RECORD_DB || hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_QP_RECORD_DB) mutex_destroy(&context->page_mutex); @@ -656,20 +656,20 @@ static void hns_roce_dealloc_ucontext(struct ib_ucontext *ibcontext) struct hns_roce_ucontext *context = to_hr_ucontext(ibcontext); struct hns_roce_dev *hr_dev = to_hr_dev(ibcontext->device); + hns_roce_put_cq_bankid_for_uctx(context); + hns_roce_unregister_uctx_debugfs(context); + mutex_lock(&hr_dev->uctx_list_mutex); list_del(&context->list); mutex_unlock(&hr_dev->uctx_list_mutex); - hns_roce_unregister_uctx_debugfs(context); - + hns_roce_dealloc_reset_entry(context); hns_roce_unregister_udca(hr_dev, context); if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_CQ_RECORD_DB || hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_QP_RECORD_DB) mutex_destroy(&context->page_mutex); - hns_roce_put_cq_bankid_for_uctx(context); hns_roce_dealloc_uar_entry(context); - hns_roce_dealloc_reset_entry(context); ida_free(&hr_dev->uar_ida.ida, (int)context->uar.logic_idx); } -- 2.33.0
From: Xinghai Cen <cenxinghai@h-partners.com> driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IBSE4C ---------------------------------------------------------------------- default_congest_type() should be called to set the default congestion control algorithm for kernel QP. Fixes: 09f1b7cb29b2 ("RDMA/hns: Support congestion control algorithm configuration at QP granularity") Signed-off-by: Xinghai Cen <cenxinghai@h-partners.com> --- drivers/infiniband/hw/hns/hns_roce_qp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/infiniband/hw/hns/hns_roce_qp.c b/drivers/infiniband/hw/hns/hns_roce_qp.c index 71288df24b08..8f4ca0f53af2 100644 --- a/drivers/infiniband/hw/hns/hns_roce_qp.c +++ b/drivers/infiniband/hw/hns/hns_roce_qp.c @@ -1351,6 +1351,7 @@ static int set_qp_param(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp, if (hr_dev->pci_dev->revision >= PCI_REVISION_ID_HIP09) hr_qp->config = HNS_ROCE_EXSGE_FLAGS; + default_congest_type(hr_dev, hr_qp); ret = set_kernel_sq_size(hr_dev, &init_attr->cap, hr_qp); if (ret) ibdev_err(ibdev, "Failed to set kernel SQ size, ret = %d\n", -- 2.33.0
反馈： 您发送到kernel@openeuler.org的补丁/补丁集，已成功转换为PR！ PR链接地址： https://gitee.com/openeuler/kernel/pulls/15447 邮件列表地址：https://mailweb.openeuler.org/archives/list/kernel@openeuler.org/message/OVF... FeedBack: The patch(es) which you have sent to kernel@openeuler.org mailing list has been converted to a pull request successfully! Pull request link: https://gitee.com/openeuler/kernel/pulls/15447 Mailing list address: https://mailweb.openeuler.org/archives/list/kernel@openeuler.org/message/OVF...
participants (2)
-
Junxian Huang -
patchwork bot