From d5121ffebc38a16b2419b664e466a2f3e5c7b457 Mon Sep 17 00:00:00 2001 From: Wei Yongjun <weiyongjun1@huawei.com> Date: Thu, 18 Jul 2019 09:27:10 +0000 Subject: [PATCH 01/20] RDMA/siw: Fix error return code in siw_init_module() Fix to return a negative error code from the error handling case instead of 0, as done elsewhere in this function. Fixes: bdcf26bf9b3a ("rdma/siw: network and RDMA core interface") Link: https://lore.kernel.org/r/20190718092710.85709-1-weiyongjun1@huawei.com Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com> Reviewed-by: Bernard Metzler <bmt@zurich.ibm.com> Reviewed-by: Jason Gunthorpe <jgg@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/sw/siw/siw_main.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/infiniband/sw/siw/siw_main.c b/drivers/infiniband/sw/siw/siw_main.c index f55c4e80aea4..d0f140daf659 100644 --- a/drivers/infiniband/sw/siw/siw_main.c +++ b/drivers/infiniband/sw/siw/siw_main.c @@ -612,6 +612,7 @@ static __init int siw_init_module(void) if (!siw_create_tx_threads()) { pr_info("siw: Could not start any TX thread\n"); + rv = -ENOMEM; goto out_error; } /* From 60c3becfd1a138fdcfe48f2a5ef41ef0078d481e Mon Sep 17 00:00:00 2001 From: Xi Wang <wangxi11@huawei.com> Date: Thu, 11 Jul 2019 09:32:17 +0800 Subject: [PATCH 02/20] RDMA/hns: Fix sg offset non-zero issue When run perftest in many times, the system will report a BUG as follows: BUG: Bad rss-counter state mm:(____ptrval____) idx:0 val:-1 BUG: Bad rss-counter state mm:(____ptrval____) idx:1 val:1 We tested with different kernel version and found it started from the the following commit: commit d10bcf947a3e ("RDMA/umem: Combine contiguous PAGE_SIZE regions in SGEs") In this commit, the sg->offset is always 0 when sg_set_page() is called in ib_umem_get() and the drivers are not allowed to change the sgl, otherwise it will get bad page descriptor when unfolding SGEs in __ib_umem_release() as sg_page_count() will get wrong result while sgl->offset is not 0. However, there is a weird sgl usage in the current hns driver, the driver modified sg->offset after calling ib_umem_get(), which caused we iterate past the wrong number of pages in for_each_sg_page iterator. This patch fixes it by correcting the non-standard sgl usage found in the hns_roce_db_map_user() function. Fixes: d10bcf947a3e ("RDMA/umem: Combine contiguous PAGE_SIZE regions in SGEs") Fixes: 0425e3e6e0c7 ("RDMA/hns: Support flush cqe for hip08 in kernel space") Link: https://lore.kernel.org/r/1562808737-45723-1-git-send-email-oulijun@huawei.com Signed-off-by: Xi Wang <wangxi11@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/hw/hns/hns_roce_db.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_db.c b/drivers/infiniband/hw/hns/hns_roce_db.c index 627aa46ef683..c00714c2f16a 100644 --- a/drivers/infiniband/hw/hns/hns_roce_db.c +++ b/drivers/infiniband/hw/hns/hns_roce_db.c @@ -12,13 +12,15 @@ int hns_roce_db_map_user(struct hns_roce_ucontext *context, struct ib_udata *udata, unsigned long virt, struct hns_roce_db *db) { + unsigned long page_addr = virt & PAGE_MASK; struct hns_roce_user_db_page *page; + unsigned int offset; int ret = 0; mutex_lock(&context->page_mutex); list_for_each_entry(page, &context->page_list, list) - if (page->user_virt == (virt & PAGE_MASK)) + if (page->user_virt == page_addr) goto found; page = kmalloc(sizeof(*page), GFP_KERNEL); @@ -28,8 +30,8 @@ int hns_roce_db_map_user(struct hns_roce_ucontext *context, } refcount_set(&page->refcount, 1); - page->user_virt = (virt & PAGE_MASK); - page->umem = ib_umem_get(udata, virt & PAGE_MASK, PAGE_SIZE, 0, 0); + page->user_virt = page_addr; + page->umem = ib_umem_get(udata, page_addr, PAGE_SIZE, 0, 0); if (IS_ERR(page->umem)) { ret = PTR_ERR(page->umem); kfree(page); @@ -39,10 +41,9 @@ int hns_roce_db_map_user(struct hns_roce_ucontext *context, list_add(&page->list, &context->page_list); found: - db->dma = sg_dma_address(page->umem->sg_head.sgl) + - (virt & ~PAGE_MASK); - page->umem->sg_head.sgl->offset = virt & ~PAGE_MASK; - db->virt_addr = sg_virt(page->umem->sg_head.sgl); + offset = virt - page_addr; + db->dma = sg_dma_address(page->umem->sg_head.sgl) + offset; + db->virt_addr = sg_virt(page->umem->sg_head.sgl) + offset; db->u.user_page = page; refcount_inc(&page->refcount); From cd48a82087231fdba0e77521102386c6ed0168d6 Mon Sep 17 00:00:00 2001 From: John Fleck <john.fleck@intel.com> Date: Mon, 15 Jul 2019 12:45:21 -0400 Subject: [PATCH 03/20] IB/hfi1: Check for error on call to alloc_rsm_map_table The call to alloc_rsm_map_table does not check if the kmalloc fails. Check for a NULL on alloc, and bail if it fails. Fixes: 372cc85a13c9 ("IB/hfi1: Extract RSM map table init from QOS") Link: https://lore.kernel.org/r/20190715164521.74174.27047.stgit@awfm-01.aw.intel.com Cc: <stable@vger.kernel.org> Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com> Signed-off-by: John Fleck <john.fleck@intel.com> Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/hw/hfi1/chip.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/hfi1/chip.c b/drivers/infiniband/hw/hfi1/chip.c index d5b643a1d9fd..67052dc3100c 100644 --- a/drivers/infiniband/hw/hfi1/chip.c +++ b/drivers/infiniband/hw/hfi1/chip.c @@ -14452,7 +14452,7 @@ void hfi1_deinit_vnic_rsm(struct hfi1_devdata *dd) clear_rcvctrl(dd, RCV_CTRL_RCV_RSM_ENABLE_SMASK); } -static void init_rxe(struct hfi1_devdata *dd) +static int init_rxe(struct hfi1_devdata *dd) { struct rsm_map_table *rmt; u64 val; @@ -14461,6 +14461,9 @@ static void init_rxe(struct hfi1_devdata *dd) write_csr(dd, RCV_ERR_MASK, ~0ull); rmt = alloc_rsm_map_table(dd); + if (!rmt) + return -ENOMEM; + /* set up QOS, including the QPN map table */ init_qos(dd, rmt); init_fecn_handling(dd, rmt); @@ -14487,6 +14490,7 @@ static void init_rxe(struct hfi1_devdata *dd) val |= ((4ull & RCV_BYPASS_HDR_SIZE_MASK) << RCV_BYPASS_HDR_SIZE_SHIFT); write_csr(dd, RCV_BYPASS, val); + return 0; } static void init_other(struct hfi1_devdata *dd) @@ -15024,7 +15028,10 @@ int hfi1_init_dd(struct hfi1_devdata *dd) goto bail_cleanup; /* set initial RXE CSRs */ - init_rxe(dd); + ret = init_rxe(dd); + if (ret) + goto bail_cleanup; + /* set initial TXE CSRs */ init_txe(dd); /* set initial non-RXE, non-TXE CSRs */ From 2b74c878b0eae4c32629c2d5ba69a29f69048313 Mon Sep 17 00:00:00 2001 From: Kaike Wan <kaike.wan@intel.com> Date: Mon, 15 Jul 2019 12:45:28 -0400 Subject: [PATCH 04/20] IB/hfi1: Unreserve a flushed OPFN request When an OPFN request is flushed, the request is completed without unreserving itself from the send queue. Subsequently, when a new request is post sent, the following warning will be triggered: WARNING: CPU: 4 PID: 8130 at rdmavt/qp.c:1761 rvt_post_send+0x72a/0x880 [rdmavt] Call Trace: [<ffffffffbbb61e41>] dump_stack+0x19/0x1b [<ffffffffbb497688>] __warn+0xd8/0x100 [<ffffffffbb4977cd>] warn_slowpath_null+0x1d/0x20 [<ffffffffc01c941a>] rvt_post_send+0x72a/0x880 [rdmavt] [<ffffffffbb4dcabe>] ? account_entity_dequeue+0xae/0xd0 [<ffffffffbb61d645>] ? __kmalloc+0x55/0x230 [<ffffffffc04e1a4c>] ib_uverbs_post_send+0x37c/0x5d0 [ib_uverbs] [<ffffffffc04e5e36>] ? rdma_lookup_put_uobject+0x26/0x60 [ib_uverbs] [<ffffffffc04dbce6>] ib_uverbs_write+0x286/0x460 [ib_uverbs] [<ffffffffbb6f9457>] ? security_file_permission+0x27/0xa0 [<ffffffffbb641650>] vfs_write+0xc0/0x1f0 [<ffffffffbb64246f>] SyS_write+0x7f/0xf0 [<ffffffffbbb74ddb>] system_call_fastpath+0x22/0x27 This patch fixes the problem by moving rvt_qp_wqe_unreserve() into rvt_qp_complete_swqe() to simplify the code and make it less error-prone. Fixes: ca95f802ef51 ("IB/hfi1: Unreserve a reserved request when it is completed") Link: https://lore.kernel.org/r/20190715164528.74174.31364.stgit@awfm-01.aw.intel.com Cc: <stable@vger.kernel.org> Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com> Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com> Signed-off-by: Kaike Wan <kaike.wan@intel.com> Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/hw/hfi1/rc.c | 2 -- include/rdma/rdmavt_qp.h | 9 ++++----- 2 files changed, 4 insertions(+), 7 deletions(-) diff --git a/drivers/infiniband/hw/hfi1/rc.c b/drivers/infiniband/hw/hfi1/rc.c index 0477c14633ab..024a7c2b6124 100644 --- a/drivers/infiniband/hw/hfi1/rc.c +++ b/drivers/infiniband/hw/hfi1/rc.c @@ -1835,7 +1835,6 @@ void hfi1_rc_send_complete(struct rvt_qp *qp, struct hfi1_opa_header *opah) cmp_psn(qp->s_sending_psn, qp->s_sending_hpsn) <= 0) break; trdma_clean_swqe(qp, wqe); - rvt_qp_wqe_unreserve(qp, wqe); trace_hfi1_qp_send_completion(qp, wqe, qp->s_last); rvt_qp_complete_swqe(qp, wqe, @@ -1882,7 +1881,6 @@ struct rvt_swqe *do_rc_completion(struct rvt_qp *qp, if (cmp_psn(wqe->lpsn, qp->s_sending_psn) < 0 || cmp_psn(qp->s_sending_psn, qp->s_sending_hpsn) > 0) { trdma_clean_swqe(qp, wqe); - rvt_qp_wqe_unreserve(qp, wqe); trace_hfi1_qp_send_completion(qp, wqe, qp->s_last); rvt_qp_complete_swqe(qp, wqe, diff --git a/include/rdma/rdmavt_qp.h b/include/rdma/rdmavt_qp.h index 0eeea520a853..e06c77d76463 100644 --- a/include/rdma/rdmavt_qp.h +++ b/include/rdma/rdmavt_qp.h @@ -608,7 +608,7 @@ static inline void rvt_qp_wqe_reserve( /** * rvt_qp_wqe_unreserve - clean reserved operation * @qp - the rvt qp - * @wqe - the send wqe + * @flags - send wqe flags * * This decrements the reserve use count. * @@ -620,11 +620,9 @@ static inline void rvt_qp_wqe_reserve( * the compiler does not juggle the order of the s_last * ring index and the decrementing of s_reserved_used. */ -static inline void rvt_qp_wqe_unreserve( - struct rvt_qp *qp, - struct rvt_swqe *wqe) +static inline void rvt_qp_wqe_unreserve(struct rvt_qp *qp, int flags) { - if (unlikely(wqe->wr.send_flags & RVT_SEND_RESERVE_USED)) { + if (unlikely(flags & RVT_SEND_RESERVE_USED)) { atomic_dec(&qp->s_reserved_used); /* insure no compiler re-order up to s_last change */ smp_mb__after_atomic(); @@ -853,6 +851,7 @@ rvt_qp_complete_swqe(struct rvt_qp *qp, u32 byte_len, last; int flags = wqe->wr.send_flags; + rvt_qp_wqe_unreserve(qp, flags); rvt_put_qp_swqe(qp, wqe); need_completion = From dc25b239ebeaa3c58e5ceaa732140427d386aa16 Mon Sep 17 00:00:00 2001 From: Kaike Wan <kaike.wan@intel.com> Date: Mon, 15 Jul 2019 12:45:34 -0400 Subject: [PATCH 05/20] IB/hfi1: Field not zero-ed when allocating TID flow memory The field flow->resync_npkts is added for TID RDMA WRITE request and zero-ed when a TID RDMA WRITE RESP packet is received by the requester. This field is used to rewind a request during retry in the function hfi1_tid_rdma_restart_req() shared by both TID RDMA WRITE and TID RDMA READ requests. Therefore, when a TID RDMA READ request is retried, this field may not be initialized at all, which causes the retry to start at an incorrect psn, leading to the drop of the retry request by the responder. This patch fixes the problem by zeroing out the field when the flow memory is allocated. Fixes: 838b6fd2d9ca ("IB/hfi1: TID RDMA RcvArray programming and TID allocation") Cc: <stable@vger.kernel.org> Link: https://lore.kernel.org/r/20190715164534.74174.6177.stgit@awfm-01.aw.intel.com Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com> Signed-off-by: Kaike Wan <kaike.wan@intel.com> Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/hw/hfi1/tid_rdma.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/infiniband/hw/hfi1/tid_rdma.c b/drivers/infiniband/hw/hfi1/tid_rdma.c index 92acccaaaa86..7fcbeee84293 100644 --- a/drivers/infiniband/hw/hfi1/tid_rdma.c +++ b/drivers/infiniband/hw/hfi1/tid_rdma.c @@ -1620,6 +1620,7 @@ static int hfi1_kern_exp_rcv_alloc_flows(struct tid_rdma_request *req, flows[i].req = req; flows[i].npagesets = 0; flows[i].pagesets[0].mapped = 0; + flows[i].resync_npkts = 0; } req->flows = flows; return 0; From f4d46119f214f9a7620b0d18b153d7e0e8c90b4f Mon Sep 17 00:00:00 2001 From: Kaike Wan <kaike.wan@intel.com> Date: Mon, 15 Jul 2019 12:45:40 -0400 Subject: [PATCH 06/20] IB/hfi1: Drop all TID RDMA READ RESP packets after r_next_psn When a TID sequence error occurs while receiving TID RDMA READ RESP packets, all packets after flow->flow_state.r_next_psn should be dropped, including those response packets for subsequent segments. The current implementation will drop the subsequent response packets for the segment to complete next, but may accept packets for subsequent segments and therefore mistakenly advance the r_next_psn fields for the corresponding software flows. This may result in failures to complete subsequent segments after the current segment is completed. The fix is to only use the flow pointed by req->clear_tail for checking KDETH PSN instead of finding a flow from the request's flow array. Fixes: b885d5be9ca1 ("IB/hfi1: Unify the software PSN check for TID RDMA READ/WRITE") Cc: <stable@vger.kernel.org> Link: https://lore.kernel.org/r/20190715164540.74174.54702.stgit@awfm-01.aw.intel.com Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com> Signed-off-by: Kaike Wan <kaike.wan@intel.com> Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/hw/hfi1/tid_rdma.c | 42 +-------------------------- 1 file changed, 1 insertion(+), 41 deletions(-) diff --git a/drivers/infiniband/hw/hfi1/tid_rdma.c b/drivers/infiniband/hw/hfi1/tid_rdma.c index 7fcbeee84293..996fc298207e 100644 --- a/drivers/infiniband/hw/hfi1/tid_rdma.c +++ b/drivers/infiniband/hw/hfi1/tid_rdma.c @@ -1674,34 +1674,6 @@ static struct tid_rdma_flow *find_flow_ib(struct tid_rdma_request *req, return NULL; } -static struct tid_rdma_flow * -__find_flow_ranged(struct tid_rdma_request *req, u16 head, u16 tail, - u32 psn, u16 *fidx) -{ - for ( ; CIRC_CNT(head, tail, MAX_FLOWS); - tail = CIRC_NEXT(tail, MAX_FLOWS)) { - struct tid_rdma_flow *flow = &req->flows[tail]; - u32 spsn, lpsn; - - spsn = full_flow_psn(flow, flow->flow_state.spsn); - lpsn = full_flow_psn(flow, flow->flow_state.lpsn); - - if (cmp_psn(psn, spsn) >= 0 && cmp_psn(psn, lpsn) <= 0) { - if (fidx) - *fidx = tail; - return flow; - } - } - return NULL; -} - -static struct tid_rdma_flow *find_flow(struct tid_rdma_request *req, - u32 psn, u16 *fidx) -{ - return __find_flow_ranged(req, req->setup_head, req->clear_tail, psn, - fidx); -} - /* TID RDMA READ functions */ u32 hfi1_build_tid_rdma_read_packet(struct rvt_swqe *wqe, struct ib_other_headers *ohdr, u32 *bth1, @@ -2789,19 +2761,7 @@ static bool handle_read_kdeth_eflags(struct hfi1_ctxtdata *rcd, * to prevent continuous Flow Sequence errors for any * packets that could be still in the fabric. */ - flow = find_flow(req, psn, NULL); - if (!flow) { - /* - * We can't find the IB PSN matching the - * received KDETH PSN. The only thing we can - * do at this point is report the error to - * the QP. - */ - hfi1_kern_read_tid_flow_free(qp); - spin_unlock(&qp->s_lock); - rvt_rc_error(qp, IB_WC_LOC_QP_OP_ERR); - return ret; - } + flow = &req->flows[req->clear_tail]; if (priv->s_flags & HFI1_R_TID_SW_PSN) { diff = cmp_psn(psn, flow->flow_state.r_next_psn); From c56b593d2af4cbd189c6af5fd6790728fade80cc Mon Sep 17 00:00:00 2001 From: Selvin Xavier <selvin.xavier@broadcom.com> Date: Mon, 15 Jul 2019 05:19:13 -0400 Subject: [PATCH 07/20] RDMA/bnxt_re: Honor vlan_id in GID entry comparison A GID entry consists of GID, vlan, netdev and smac. Extend GID duplicate check comparisons to consider vlan_id as well to support IPv6 VLAN based link local addresses. Introduce a new structure (bnxt_qplib_gid_info) to hold gid and vlan_id information. The issue is discussed in the following thread https://lore.kernel.org/r/AM0PR05MB4866CFEDCDF3CDA1D7D18AA5D1F20@AM0PR05MB4866.eurprd05.prod.outlook.com Fixes: 823b23da7113 ("IB/core: Allow vlan link local address based RoCE GIDs") Cc: <stable@vger.kernel.org> # v5.2+ Link: https://lore.kernel.org/r/20190715091913.15726-1-selvin.xavier@broadcom.com Reported-by: Yi Zhang <yi.zhang@redhat.com> Co-developed-by: Parav Pandit <parav@mellanox.com> Signed-off-by: Parav Pandit <parav@mellanox.com> Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com> Tested-by: Yi Zhang <yi.zhang@redhat.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/hw/bnxt_re/ib_verbs.c | 7 +++++-- drivers/infiniband/hw/bnxt_re/qplib_res.c | 13 +++++++++---- drivers/infiniband/hw/bnxt_re/qplib_res.h | 2 +- drivers/infiniband/hw/bnxt_re/qplib_sp.c | 14 +++++++++----- drivers/infiniband/hw/bnxt_re/qplib_sp.h | 7 ++++++- 5 files changed, 30 insertions(+), 13 deletions(-) diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.c b/drivers/infiniband/hw/bnxt_re/ib_verbs.c index a91653aabf38..098ab883733e 100644 --- a/drivers/infiniband/hw/bnxt_re/ib_verbs.c +++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.c @@ -308,6 +308,7 @@ int bnxt_re_del_gid(const struct ib_gid_attr *attr, void **context) struct bnxt_re_dev *rdev = to_bnxt_re_dev(attr->device, ibdev); struct bnxt_qplib_sgid_tbl *sgid_tbl = &rdev->qplib_res.sgid_tbl; struct bnxt_qplib_gid *gid_to_del; + u16 vlan_id = 0xFFFF; /* Delete the entry from the hardware */ ctx = *context; @@ -317,7 +318,8 @@ int bnxt_re_del_gid(const struct ib_gid_attr *attr, void **context) if (sgid_tbl && sgid_tbl->active) { if (ctx->idx >= sgid_tbl->max) return -EINVAL; - gid_to_del = &sgid_tbl->tbl[ctx->idx]; + gid_to_del = &sgid_tbl->tbl[ctx->idx].gid; + vlan_id = sgid_tbl->tbl[ctx->idx].vlan_id; /* DEL_GID is called in WQ context(netdevice_event_work_handler) * or via the ib_unregister_device path. In the former case QP1 * may not be destroyed yet, in which case just return as FW @@ -335,7 +337,8 @@ int bnxt_re_del_gid(const struct ib_gid_attr *attr, void **context) } ctx->refcnt--; if (!ctx->refcnt) { - rc = bnxt_qplib_del_sgid(sgid_tbl, gid_to_del, true); + rc = bnxt_qplib_del_sgid(sgid_tbl, gid_to_del, + vlan_id, true); if (rc) { dev_err(rdev_to_dev(rdev), "Failed to remove GID: %#x", rc); diff --git a/drivers/infiniband/hw/bnxt_re/qplib_res.c b/drivers/infiniband/hw/bnxt_re/qplib_res.c index 37928b1111df..bdbde8e22420 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_res.c +++ b/drivers/infiniband/hw/bnxt_re/qplib_res.c @@ -488,7 +488,7 @@ static int bnxt_qplib_alloc_sgid_tbl(struct bnxt_qplib_res *res, struct bnxt_qplib_sgid_tbl *sgid_tbl, u16 max) { - sgid_tbl->tbl = kcalloc(max, sizeof(struct bnxt_qplib_gid), GFP_KERNEL); + sgid_tbl->tbl = kcalloc(max, sizeof(*sgid_tbl->tbl), GFP_KERNEL); if (!sgid_tbl->tbl) return -ENOMEM; @@ -526,9 +526,10 @@ static void bnxt_qplib_cleanup_sgid_tbl(struct bnxt_qplib_res *res, for (i = 0; i < sgid_tbl->max; i++) { if (memcmp(&sgid_tbl->tbl[i], &bnxt_qplib_gid_zero, sizeof(bnxt_qplib_gid_zero))) - bnxt_qplib_del_sgid(sgid_tbl, &sgid_tbl->tbl[i], true); + bnxt_qplib_del_sgid(sgid_tbl, &sgid_tbl->tbl[i].gid, + sgid_tbl->tbl[i].vlan_id, true); } - memset(sgid_tbl->tbl, 0, sizeof(struct bnxt_qplib_gid) * sgid_tbl->max); + memset(sgid_tbl->tbl, 0, sizeof(*sgid_tbl->tbl) * sgid_tbl->max); memset(sgid_tbl->hw_id, -1, sizeof(u16) * sgid_tbl->max); memset(sgid_tbl->vlan, 0, sizeof(u8) * sgid_tbl->max); sgid_tbl->active = 0; @@ -537,7 +538,11 @@ static void bnxt_qplib_cleanup_sgid_tbl(struct bnxt_qplib_res *res, static void bnxt_qplib_init_sgid_tbl(struct bnxt_qplib_sgid_tbl *sgid_tbl, struct net_device *netdev) { - memset(sgid_tbl->tbl, 0, sizeof(struct bnxt_qplib_gid) * sgid_tbl->max); + u32 i; + + for (i = 0; i < sgid_tbl->max; i++) + sgid_tbl->tbl[i].vlan_id = 0xffff; + memset(sgid_tbl->hw_id, -1, sizeof(u16) * sgid_tbl->max); } diff --git a/drivers/infiniband/hw/bnxt_re/qplib_res.h b/drivers/infiniband/hw/bnxt_re/qplib_res.h index 30c42c92fac7..fbda11a7ab1a 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_res.h +++ b/drivers/infiniband/hw/bnxt_re/qplib_res.h @@ -111,7 +111,7 @@ struct bnxt_qplib_pd_tbl { }; struct bnxt_qplib_sgid_tbl { - struct bnxt_qplib_gid *tbl; + struct bnxt_qplib_gid_info *tbl; u16 *hw_id; u16 max; u16 active; diff --git a/drivers/infiniband/hw/bnxt_re/qplib_sp.c b/drivers/infiniband/hw/bnxt_re/qplib_sp.c index 48793d3512ac..40296b97d21e 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_sp.c +++ b/drivers/infiniband/hw/bnxt_re/qplib_sp.c @@ -213,12 +213,12 @@ int bnxt_qplib_get_sgid(struct bnxt_qplib_res *res, index, sgid_tbl->max); return -EINVAL; } - memcpy(gid, &sgid_tbl->tbl[index], sizeof(*gid)); + memcpy(gid, &sgid_tbl->tbl[index].gid, sizeof(*gid)); return 0; } int bnxt_qplib_del_sgid(struct bnxt_qplib_sgid_tbl *sgid_tbl, - struct bnxt_qplib_gid *gid, bool update) + struct bnxt_qplib_gid *gid, u16 vlan_id, bool update) { struct bnxt_qplib_res *res = to_bnxt_qplib(sgid_tbl, struct bnxt_qplib_res, @@ -236,7 +236,8 @@ int bnxt_qplib_del_sgid(struct bnxt_qplib_sgid_tbl *sgid_tbl, return -ENOMEM; } for (index = 0; index < sgid_tbl->max; index++) { - if (!memcmp(&sgid_tbl->tbl[index], gid, sizeof(*gid))) + if (!memcmp(&sgid_tbl->tbl[index].gid, gid, sizeof(*gid)) && + vlan_id == sgid_tbl->tbl[index].vlan_id) break; } if (index == sgid_tbl->max) { @@ -262,8 +263,9 @@ int bnxt_qplib_del_sgid(struct bnxt_qplib_sgid_tbl *sgid_tbl, if (rc) return rc; } - memcpy(&sgid_tbl->tbl[index], &bnxt_qplib_gid_zero, + memcpy(&sgid_tbl->tbl[index].gid, &bnxt_qplib_gid_zero, sizeof(bnxt_qplib_gid_zero)); + sgid_tbl->tbl[index].vlan_id = 0xFFFF; sgid_tbl->vlan[index] = 0; sgid_tbl->active--; dev_dbg(&res->pdev->dev, @@ -296,7 +298,8 @@ int bnxt_qplib_add_sgid(struct bnxt_qplib_sgid_tbl *sgid_tbl, } free_idx = sgid_tbl->max; for (i = 0; i < sgid_tbl->max; i++) { - if (!memcmp(&sgid_tbl->tbl[i], gid, sizeof(*gid))) { + if (!memcmp(&sgid_tbl->tbl[i], gid, sizeof(*gid)) && + sgid_tbl->tbl[i].vlan_id == vlan_id) { dev_dbg(&res->pdev->dev, "SGID entry already exist in entry %d!\n", i); *index = i; @@ -351,6 +354,7 @@ int bnxt_qplib_add_sgid(struct bnxt_qplib_sgid_tbl *sgid_tbl, } /* Add GID to the sgid_tbl */ memcpy(&sgid_tbl->tbl[free_idx], gid, sizeof(*gid)); + sgid_tbl->tbl[free_idx].vlan_id = vlan_id; sgid_tbl->active++; if (vlan_id != 0xFFFF) sgid_tbl->vlan[free_idx] = 1; diff --git a/drivers/infiniband/hw/bnxt_re/qplib_sp.h b/drivers/infiniband/hw/bnxt_re/qplib_sp.h index 0ec3b12b0bcd..13d9432d5ce2 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_sp.h +++ b/drivers/infiniband/hw/bnxt_re/qplib_sp.h @@ -84,6 +84,11 @@ struct bnxt_qplib_gid { u8 data[16]; }; +struct bnxt_qplib_gid_info { + struct bnxt_qplib_gid gid; + u16 vlan_id; +}; + struct bnxt_qplib_ah { struct bnxt_qplib_gid dgid; struct bnxt_qplib_pd *pd; @@ -221,7 +226,7 @@ int bnxt_qplib_get_sgid(struct bnxt_qplib_res *res, struct bnxt_qplib_sgid_tbl *sgid_tbl, int index, struct bnxt_qplib_gid *gid); int bnxt_qplib_del_sgid(struct bnxt_qplib_sgid_tbl *sgid_tbl, - struct bnxt_qplib_gid *gid, bool update); + struct bnxt_qplib_gid *gid, u16 vlan_id, bool update); int bnxt_qplib_add_sgid(struct bnxt_qplib_sgid_tbl *sgid_tbl, struct bnxt_qplib_gid *gid, u8 *mac, u16 vlan_id, bool update, u32 *index); From b7f406bb883ba7ac3222298f6b44cebc4cfe2dde Mon Sep 17 00:00:00 2001 From: Chuhong Yuan <hslester96@gmail.com> Date: Wed, 17 Jul 2019 16:21:01 +0800 Subject: [PATCH 08/20] IB/mlx5: Replace kfree with kvfree Memory allocated by kvzalloc should not be freed by kfree(), use kvfree() instead. Fixes: 813e90b1aeaa ("IB/mlx5: Add advise_mr() support") Link: https://lore.kernel.org/r/20190717082101.14196-1-hslester96@gmail.com Signed-off-by: Chuhong Yuan <hslester96@gmail.com> Reviewed-by: Jason Gunthorpe <jgg@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/hw/mlx5/odp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c index 5b642d81e617..36ba901cc9a5 100644 --- a/drivers/infiniband/hw/mlx5/odp.c +++ b/drivers/infiniband/hw/mlx5/odp.c @@ -1771,7 +1771,7 @@ static void mlx5_ib_prefetch_mr_work(struct work_struct *work) num_pending_prefetch_dec(to_mdev(w->pd->device), w->sg_list, w->num_sge, 0); - kfree(w); + kvfree(w); } int mlx5_ib_advise_mr_prefetch(struct ib_pd *pd, @@ -1813,7 +1813,7 @@ int mlx5_ib_advise_mr_prefetch(struct ib_pd *pd, if (valid_req) queue_work(system_unbound_wq, &work->work); else - kfree(work); + kvfree(work); srcu_read_unlock(&dev->mr_srcu, srcu_key); From af0653d56657340a80622aeb96707f7fc8506225 Mon Sep 17 00:00:00 2001 From: Mao Wenan <maowenan@huawei.com> Date: Fri, 19 Jul 2019 09:29:38 +0800 Subject: [PATCH 09/20] RDMA/siw: Remove set but not used variables 'rv' Fixes gcc '-Wunused-but-set-variable' warning: drivers/infiniband/sw/siw/siw_cm.c: In function siw_cep_set_inuse: drivers/infiniband/sw/siw/siw_cm.c:223:6: warning: variable rv set but not used [-Wunused-but-set-variable] Fixes: 6c52fdc244b5 ("rdma/siw: connection management") Link: https://lore.kernel.org/r/20190719012938.100628-1-maowenan@huawei.com Signed-off-by: Mao Wenan <maowenan@huawei.com> Reviewed-by: Bernard Metzler <bmt@zurich.ibm.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/sw/siw/siw_cm.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/infiniband/sw/siw/siw_cm.c b/drivers/infiniband/sw/siw/siw_cm.c index a7cde98e73e8..9ce8a1b925d2 100644 --- a/drivers/infiniband/sw/siw/siw_cm.c +++ b/drivers/infiniband/sw/siw/siw_cm.c @@ -220,13 +220,12 @@ static void siw_put_work(struct siw_cm_work *work) static void siw_cep_set_inuse(struct siw_cep *cep) { unsigned long flags; - int rv; retry: spin_lock_irqsave(&cep->lock, flags); if (cep->in_use) { spin_unlock_irqrestore(&cep->lock, flags); - rv = wait_event_interruptible(cep->waitq, !cep->in_use); + wait_event_interruptible(cep->waitq, !cep->in_use); if (signal_pending(current)) flush_signals(current); goto retry; From 6a053953739d23694474a5f9c81d1a30093da81a Mon Sep 17 00:00:00 2001 From: Yishai Hadas <yishaih@mellanox.com> Date: Tue, 23 Jul 2019 09:57:25 +0300 Subject: [PATCH 10/20] IB/mlx5: Fix unreg_umr to ignore the mkey state Fix unreg_umr to ignore the mkey state and do not fail if was freed. This prevents a case that a user space application already changed the mkey state to free and then the UMR operation will fail leaving the mkey in an inappropriate state. Link: https://lore.kernel.org/r/20190723065733.4899-3-leon@kernel.org Cc: <stable@vger.kernel.org> # 3.19 Fixes: 968e78dd9644 ("IB/mlx5: Enhance UMR support to allow partial page table update") Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Reviewed-by: Artemy Kovalyov <artemyko@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Reviewed-by: Jason Gunthorpe <jgg@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/hw/mlx5/mlx5_ib.h | 1 + drivers/infiniband/hw/mlx5/mr.c | 4 ++-- drivers/infiniband/hw/mlx5/qp.c | 12 ++++++++---- 3 files changed, 11 insertions(+), 6 deletions(-) diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index c482f19958b3..f6a53455bf8b 100644 --- a/drivers/infiniband/hw/mlx5/mlx5_ib.h +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -481,6 +481,7 @@ struct mlx5_umr_wr { u64 length; int access_flags; u32 mkey; + u8 ignore_free_state:1; }; static inline const struct mlx5_umr_wr *umr_wr(const struct ib_send_wr *wr) diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c index 20ece6e0b2fc..266edaf8029d 100644 --- a/drivers/infiniband/hw/mlx5/mr.c +++ b/drivers/infiniband/hw/mlx5/mr.c @@ -1372,10 +1372,10 @@ static int unreg_umr(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr) if (mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) return 0; - umrwr.wr.send_flags = MLX5_IB_SEND_UMR_DISABLE_MR | - MLX5_IB_SEND_UMR_FAIL_IF_FREE; + umrwr.wr.send_flags = MLX5_IB_SEND_UMR_DISABLE_MR; umrwr.wr.opcode = MLX5_IB_WR_UMR; umrwr.mkey = mr->mmkey.key; + umrwr.ignore_free_state = 1; return mlx5_ib_post_send_wait(dev, &umrwr); } diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index 2a97619ed603..615cc6771516 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -4295,10 +4295,14 @@ static int set_reg_umr_segment(struct mlx5_ib_dev *dev, memset(umr, 0, sizeof(*umr)); - if (wr->send_flags & MLX5_IB_SEND_UMR_FAIL_IF_FREE) - umr->flags = MLX5_UMR_CHECK_FREE; /* fail if free */ - else - umr->flags = MLX5_UMR_CHECK_NOT_FREE; /* fail if not free */ + if (!umrwr->ignore_free_state) { + if (wr->send_flags & MLX5_IB_SEND_UMR_FAIL_IF_FREE) + /* fail if free */ + umr->flags = MLX5_UMR_CHECK_FREE; + else + /* fail if not free */ + umr->flags = MLX5_UMR_CHECK_NOT_FREE; + } umr->xlt_octowords = cpu_to_be16(get_xlt_octo(umrwr->xlt_size)); if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_XLT) { From afd1417404fba6dbfa6c0a8e5763bd348da682e4 Mon Sep 17 00:00:00 2001 From: Yishai Hadas <yishaih@mellanox.com> Date: Tue, 23 Jul 2019 09:57:26 +0300 Subject: [PATCH 11/20] IB/mlx5: Use direct mkey destroy command upon UMR unreg failure Use a direct firmware command to destroy the mkey in case the unreg UMR operation has failed. This prevents a case that a mkey will leak out from the cache post a failure to be destroyed by a UMR WR. In case the MR cache limit didn't reach a call to add another entry to the cache instead of the destroyed one is issued. In addition, replaced a warn message to WARN_ON() as this flow is fatal and can't happen unless some bug around. Link: https://lore.kernel.org/r/20190723065733.4899-4-leon@kernel.org Cc: <stable@vger.kernel.org> # 4.10 Fixes: 49780d42dfc9 ("IB/mlx5: Expose MR cache for mlx5_ib") Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Reviewed-by: Artemy Kovalyov <artemyko@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Reviewed-by: Jason Gunthorpe <jgg@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/hw/mlx5/mr.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c index 266edaf8029d..b83361aebf28 100644 --- a/drivers/infiniband/hw/mlx5/mr.c +++ b/drivers/infiniband/hw/mlx5/mr.c @@ -545,14 +545,17 @@ void mlx5_mr_cache_free(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr) return; c = order2idx(dev, mr->order); - if (c < 0 || c >= MAX_MR_CACHE_ENTRIES) { - mlx5_ib_warn(dev, "order %d, cache index %d\n", mr->order, c); + WARN_ON(c < 0 || c >= MAX_MR_CACHE_ENTRIES); + + if (unreg_umr(dev, mr)) { + mr->allocated_from_cache = false; + destroy_mkey(dev, mr); + ent = &cache->ent[c]; + if (ent->cur < ent->limit) + queue_work(cache->wq, &ent->work); return; } - if (unreg_umr(dev, mr)) - return; - ent = &cache->ent[c]; spin_lock_irq(&ent->lock); list_add_tail(&mr->list, &ent->head); From 9ec4483a3f0f71a228a5933bc040441322bfb090 Mon Sep 17 00:00:00 2001 From: Yishai Hadas <yishaih@mellanox.com> Date: Tue, 23 Jul 2019 09:57:27 +0300 Subject: [PATCH 12/20] IB/mlx5: Move MRs to a kernel PD when freeing them to the MR cache Fix unreg_umr to move the MR to a kernel owned PD (i.e. the UMR PD) which can't be accessed by userspace. This ensures that nothing can continue to access the MR once it has been placed in the kernels cache for reuse. MRs in the cache continue to have their HW state, including DMA tables, present. Even though the MR has been invalidated, changing the PD provides an additional layer of protection against use of the MR. Link: https://lore.kernel.org/r/20190723065733.4899-5-leon@kernel.org Cc: <stable@vger.kernel.org> # 3.10 Fixes: e126ba97dba9 ("mlx5: Add driver for Mellanox Connect-IB adapters") Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Reviewed-by: Artemy Kovalyov <artemyko@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Reviewed-by: Jason Gunthorpe <jgg@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/hw/mlx5/mr.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c index b83361aebf28..7274a9b9df58 100644 --- a/drivers/infiniband/hw/mlx5/mr.c +++ b/drivers/infiniband/hw/mlx5/mr.c @@ -1375,8 +1375,10 @@ static int unreg_umr(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr) if (mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) return 0; - umrwr.wr.send_flags = MLX5_IB_SEND_UMR_DISABLE_MR; + umrwr.wr.send_flags = MLX5_IB_SEND_UMR_DISABLE_MR | + MLX5_IB_SEND_UMR_UPDATE_PD_ACCESS; umrwr.wr.opcode = MLX5_IB_WR_UMR; + umrwr.pd = dev->umrc.pd; umrwr.mkey = mr->mmkey.key; umrwr.ignore_free_state = 1; From b9332dad987018745a0c0bb718d12dacfa760489 Mon Sep 17 00:00:00 2001 From: Yishai Hadas <yishaih@mellanox.com> Date: Tue, 23 Jul 2019 09:57:28 +0300 Subject: [PATCH 13/20] IB/mlx5: Fix clean_mr() to work in the expected order Any dma map underlying the MR should only be freed once the MR is fenced at the hardware. As of the above we first destroy the MKEY and just after that can safely call to dma_unmap_single(). Link: https://lore.kernel.org/r/20190723065733.4899-6-leon@kernel.org Cc: <stable@vger.kernel.org> # 4.3 Fixes: 8a187ee52b04 ("IB/mlx5: Support the new memory registration API") Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Reviewed-by: Artemy Kovalyov <artemyko@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Reviewed-by: Jason Gunthorpe <jgg@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/hw/mlx5/mr.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c index 7274a9b9df58..2c77456f359f 100644 --- a/drivers/infiniband/hw/mlx5/mr.c +++ b/drivers/infiniband/hw/mlx5/mr.c @@ -1582,10 +1582,10 @@ static void clean_mr(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr) mr->sig = NULL; } - mlx5_free_priv_descs(mr); - - if (!allocated_from_cache) + if (!allocated_from_cache) { destroy_mkey(dev, mr); + mlx5_free_priv_descs(mr); + } } static void dereg_mr(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr) From 296e3a2aad09d328f22e54655c3d736033fe1ae8 Mon Sep 17 00:00:00 2001 From: Moni Shoua <monis@mellanox.com> Date: Tue, 23 Jul 2019 09:57:30 +0300 Subject: [PATCH 14/20] IB/mlx5: Prevent concurrent MR updates during invalidation The device requires that memory registration work requests that update the address translation table of a MR will be fenced if posted together. This scenario can happen when address ranges are invalidated by the mmu in separate concurrent calls to the invalidation callback. We prefer to block concurrent address updates for a single MR over fencing since making the decision if a WQE needs fencing will be more expensive and fencing all WQEs is a too radical choice. Further, it isn't clear that this code can even run safely concurrently, so a lock is a safer choice. Fixes: b4cfe447d47b ("IB/mlx5: Implement on demand paging by adding support for MMU notifiers") Link: https://lore.kernel.org/r/20190723065733.4899-8-leon@kernel.org Signed-off-by: Moni Shoua <monis@mellanox.com> Reviewed-by: Artemy Kovalyov <artemyko@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Reviewed-by: Jason Gunthorpe <jgg@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/hw/mlx5/odp.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c index 36ba901cc9a5..81da82050d05 100644 --- a/drivers/infiniband/hw/mlx5/odp.c +++ b/drivers/infiniband/hw/mlx5/odp.c @@ -246,7 +246,7 @@ void mlx5_ib_invalidate_range(struct ib_umem_odp *umem_odp, unsigned long start, * overwrite the same MTTs. Concurent invalidations might race us, * but they will write 0s as well, so no difference in the end result. */ - + mutex_lock(&umem_odp->umem_mutex); for (addr = start; addr < end; addr += BIT(umem_odp->page_shift)) { idx = (addr - ib_umem_start(umem_odp)) >> umem_odp->page_shift; /* @@ -278,6 +278,7 @@ void mlx5_ib_invalidate_range(struct ib_umem_odp *umem_odp, unsigned long start, idx - blk_start_idx + 1, 0, MLX5_IB_UPD_XLT_ZAP | MLX5_IB_UPD_XLT_ATOMIC); + mutex_unlock(&umem_odp->umem_mutex); /* * We are now sure that the device will not access the * memory. We can safely unmap it, and mark it as dirty if From a379d1ce32f1c97b19d68cffbd195e025dbb9d43 Mon Sep 17 00:00:00 2001 From: Parav Pandit <parav@mellanox.com> Date: Tue, 23 Jul 2019 09:57:32 +0300 Subject: [PATCH 15/20] IB/core: Fix querying total rdma stats rdma_counter_init() may fail for a device. In such case while calculating total sum, ignore NULL hstats. This fixes below observed call trace. BUG: kernel NULL pointer dereference, address: 00000000000000a0 PGD 8000001009b30067 P4D 8000001009b30067 PUD 10549c9067 PMD 0 Oops: 0000 [#1] SMP PTI CPU: 55 PID: 20887 Comm: cat Kdump: loaded Not tainted 5.2.0-rc6-jdc+ #13 RIP: 0010:rdma_counter_get_hwstat_value+0xf2/0x150 [ib_core] Call Trace: show_hw_stats+0x5e/0x130 [ib_core] dev_attr_show+0x15/0x50 sysfs_kf_seq_show+0xc6/0x1a0 seq_read+0x132/0x370 vfs_read+0x89/0x140 ksys_read+0x5c/0xd0 do_syscall_64+0x5a/0x240 entry_SYSCALL_64_after_hwframe+0x49/0xbe Fixes: f34a55e497e8 ("RDMA/core: Get sum value of all counters when perform a sysfs stat read") Link: https://lore.kernel.org/r/20190723065733.4899-10-leon@kernel.org Signed-off-by: Parav Pandit <parav@mellanox.com> Reviewed-by: Mark Zhang <markz@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/core/counters.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/infiniband/core/counters.c b/drivers/infiniband/core/counters.c index 01faef7bc061..c7d445635476 100644 --- a/drivers/infiniband/core/counters.c +++ b/drivers/infiniband/core/counters.c @@ -393,6 +393,9 @@ u64 rdma_counter_get_hwstat_value(struct ib_device *dev, u8 port, u32 index) u64 sum; port_counter = &dev->port_data[port].port_counter; + if (!port_counter->hstats) + return 0; + sum = get_running_counters_hwstat_sum(dev, port, index); sum += port_counter->hstats->value[index]; From d191152f43a5869d7dbb50dd2a7a4b3b8b71f1f0 Mon Sep 17 00:00:00 2001 From: Parav Pandit <parav@mellanox.com> Date: Tue, 23 Jul 2019 09:57:33 +0300 Subject: [PATCH 16/20] IB/counters: Always initialize the port counter object Port counter objects should be initialized even if alloc_stats is unsupported, otherwise QP bind operations in user space can trigger a NULL pointer deference if they try to bind QP on RDMA device which doesn't support counters. Fixes: f34a55e497e8 ("RDMA/core: Get sum value of all counters when perform a sysfs stat read") Link: https://lore.kernel.org/r/20190723065733.4899-11-leon@kernel.org Signed-off-by: Parav Pandit <parav@mellanox.com> Reviewed-by: Mark Zhang <markz@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/core/counters.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/infiniband/core/counters.c b/drivers/infiniband/core/counters.c index c7d445635476..45d5164e9574 100644 --- a/drivers/infiniband/core/counters.c +++ b/drivers/infiniband/core/counters.c @@ -597,7 +597,7 @@ void rdma_counter_init(struct ib_device *dev) struct rdma_port_counter *port_counter; u32 port; - if (!dev->ops.alloc_hw_stats || !dev->port_data) + if (!dev->port_data) return; rdma_for_each_port(dev, port) { @@ -605,6 +605,9 @@ void rdma_counter_init(struct ib_device *dev) port_counter->mode.mode = RDMA_COUNTER_MODE_NONE; mutex_init(&port_counter->lock); + if (!dev->ops.alloc_hw_stats) + continue; + port_counter->hstats = dev->ops.alloc_hw_stats(dev, port); if (!port_counter->hstats) goto fail; @@ -627,9 +630,6 @@ void rdma_counter_release(struct ib_device *dev) struct rdma_port_counter *port_counter; u32 port; - if (!dev->ops.alloc_hw_stats) - return; - rdma_for_each_port(dev, port) { port_counter = &dev->port_data[port].port_counter; kfree(port_counter->hstats); From b7165bd0d6cbb93732559be6ea8774653b204480 Mon Sep 17 00:00:00 2001 From: Yishai Hadas <yishaih@mellanox.com> Date: Tue, 23 Jul 2019 09:57:29 +0300 Subject: [PATCH 17/20] IB/mlx5: Fix RSS Toeplitz setup to be aligned with the HW specification The specification for the Toeplitz function doesn't require to set the key explicitly to be symmetric. In case a symmetric functionality is required a symmetric key can be simply used. Wrongly forcing the algorithm to symmetric causes the wrong packet distribution and a performance degradation. Link: https://lore.kernel.org/r/20190723065733.4899-7-leon@kernel.org Cc: <stable@vger.kernel.org> # 4.7 Fixes: 28d6137008b2 ("IB/mlx5: Add RSS QP support") Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Reviewed-by: Alex Vainman <alexv@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- drivers/infiniband/hw/mlx5/qp.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index 615cc6771516..379328b2598f 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -1713,7 +1713,6 @@ static int create_rss_raw_qp_tir(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp, } MLX5_SET(tirc, tirc, rx_hash_fn, MLX5_RX_HASH_FN_TOEPLITZ); - MLX5_SET(tirc, tirc, rx_hash_symmetric, 1); memcpy(rss_key, ucmd.rx_hash_key, len); break; } From a07fc0bb483eb24444cebd59a8112ce6e6964c48 Mon Sep 17 00:00:00 2001 From: YueHaibing <yuehaibing@huawei.com> Date: Wed, 24 Jul 2019 14:54:43 +0800 Subject: [PATCH 18/20] RDMA/hns: Fix build error If INFINIBAND_HNS_HIP08 is selected and HNS3 is m, but INFINIBAND_HNS is y, building fails: drivers/infiniband/hw/hns/hns_roce_hw_v2.o: In function `hns_roce_hw_v2_exit': hns_roce_hw_v2.c:(.exit.text+0xd): undefined reference to `hnae3_unregister_client' drivers/infiniband/hw/hns/hns_roce_hw_v2.o: In function `hns_roce_hw_v2_init': hns_roce_hw_v2.c:(.init.text+0xd): undefined reference to `hnae3_register_client' Also if INFINIBAND_HNS_HIP06 is selected and HNS_DSAF is m, but INFINIBAND_HNS is y, building fails: drivers/infiniband/hw/hns/hns_roce_hw_v1.o: In function `hns_roce_v1_reset': hns_roce_hw_v1.c:(.text+0x39fa): undefined reference to `hns_dsaf_roce_reset' hns_roce_hw_v1.c:(.text+0x3a25): undefined reference to `hns_dsaf_roce_reset' Reported-by: Hulk Robot <hulkci@huawei.com> Fixes: dd74282df573 ("RDMA/hns: Initialize the PCI device for hip08 RoCE") Fixes: 08805fdbeb2d ("RDMA/hns: Split hw v1 driver from hns roce driver") Signed-off-by: YueHaibing <yuehaibing@huawei.com> Link: https://lore.kernel.org/r/20190724065443.53068-1-yuehaibing@huawei.com Signed-off-by: Doug Ledford <dledford@redhat.com> --- drivers/infiniband/hw/hns/Kconfig | 6 +++--- drivers/infiniband/hw/hns/Makefile | 8 ++------ 2 files changed, 5 insertions(+), 9 deletions(-) diff --git a/drivers/infiniband/hw/hns/Kconfig b/drivers/infiniband/hw/hns/Kconfig index 8bf847bcd8d3..54782197c717 100644 --- a/drivers/infiniband/hw/hns/Kconfig +++ b/drivers/infiniband/hw/hns/Kconfig @@ -1,6 +1,6 @@ # SPDX-License-Identifier: GPL-2.0-only config INFINIBAND_HNS - tristate "HNS RoCE Driver" + bool "HNS RoCE Driver" depends on NET_VENDOR_HISILICON depends on ARM64 || (COMPILE_TEST && 64BIT) ---help--- @@ -11,7 +11,7 @@ config INFINIBAND_HNS To compile HIP06 or HIP08 driver as module, choose M here. config INFINIBAND_HNS_HIP06 - bool "Hisilicon Hip06 Family RoCE support" + tristate "Hisilicon Hip06 Family RoCE support" depends on INFINIBAND_HNS && HNS && HNS_DSAF && HNS_ENET ---help--- RoCE driver support for Hisilicon RoCE engine in Hisilicon Hip06 and @@ -21,7 +21,7 @@ config INFINIBAND_HNS_HIP06 module will be called hns-roce-hw-v1 config INFINIBAND_HNS_HIP08 - bool "Hisilicon Hip08 Family RoCE support" + tristate "Hisilicon Hip08 Family RoCE support" depends on INFINIBAND_HNS && PCI && HNS3 ---help--- RoCE driver support for Hisilicon RoCE engine in Hisilicon Hip08 SoC. diff --git a/drivers/infiniband/hw/hns/Makefile b/drivers/infiniband/hw/hns/Makefile index e105945b94a1..449a2d81319d 100644 --- a/drivers/infiniband/hw/hns/Makefile +++ b/drivers/infiniband/hw/hns/Makefile @@ -9,12 +9,8 @@ hns-roce-objs := hns_roce_main.o hns_roce_cmd.o hns_roce_pd.o \ hns_roce_ah.o hns_roce_hem.o hns_roce_mr.o hns_roce_qp.o \ hns_roce_cq.o hns_roce_alloc.o hns_roce_db.o hns_roce_srq.o hns_roce_restrack.o -ifdef CONFIG_INFINIBAND_HNS_HIP06 hns-roce-hw-v1-objs := hns_roce_hw_v1.o $(hns-roce-objs) -obj-$(CONFIG_INFINIBAND_HNS) += hns-roce-hw-v1.o -endif +obj-$(CONFIG_INFINIBAND_HNS_HIP06) += hns-roce-hw-v1.o -ifdef CONFIG_INFINIBAND_HNS_HIP08 hns-roce-hw-v2-objs := hns_roce_hw_v2.o hns_roce_hw_v2_dfx.o $(hns-roce-objs) -obj-$(CONFIG_INFINIBAND_HNS) += hns-roce-hw-v2.o -endif +obj-$(CONFIG_INFINIBAND_HNS_HIP08) += hns-roce-hw-v2.o From 15fe6a8dcc3b48358c28e17b485fc837f9605ec4 Mon Sep 17 00:00:00 2001 From: Michal Kalderon <michal.kalderon@marvell.com> Date: Sun, 28 Jul 2019 14:13:38 +0300 Subject: [PATCH 19/20] RDMA/qedr: Fix the hca_type and hca_rev returned in device attributes There was a place holder for hca_type and vendor was returned in hca_rev. Fix the hca_rev to return the hw revision and fix the hca_type to return an informative string representing the hca. Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com> Link: https://lore.kernel.org/r/20190728111338.21930-1-michal.kalderon@marvell.com Signed-off-by: Doug Ledford <dledford@redhat.com> --- drivers/infiniband/hw/qedr/main.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/qedr/main.c b/drivers/infiniband/hw/qedr/main.c index 533157a2a3be..f97b3d65b30c 100644 --- a/drivers/infiniband/hw/qedr/main.c +++ b/drivers/infiniband/hw/qedr/main.c @@ -125,14 +125,20 @@ static ssize_t hw_rev_show(struct device *device, struct device_attribute *attr, struct qedr_dev *dev = rdma_device_to_drv_device(device, struct qedr_dev, ibdev); - return scnprintf(buf, PAGE_SIZE, "0x%x\n", dev->pdev->vendor); + return scnprintf(buf, PAGE_SIZE, "0x%x\n", dev->attr.hw_ver); } static DEVICE_ATTR_RO(hw_rev); static ssize_t hca_type_show(struct device *device, struct device_attribute *attr, char *buf) { - return scnprintf(buf, PAGE_SIZE, "%s\n", "HCA_TYPE_TO_SET"); + struct qedr_dev *dev = + rdma_device_to_drv_device(device, struct qedr_dev, ibdev); + + return scnprintf(buf, PAGE_SIZE, "FastLinQ QL%x %s\n", + dev->pdev->device, + rdma_protocol_iwarp(&dev->ibdev, 1) ? + "iWARP" : "RoCE"); } static DEVICE_ATTR_RO(hca_type); From 708637e65abd487ebb75fb55401c36a466c3135b Mon Sep 17 00:00:00 2001 From: Bernard Metzler <bmt@zurich.ibm.com> Date: Sat, 27 Jul 2019 12:38:32 +0200 Subject: [PATCH 20/20] Do not dereference 'siw_crypto_shash' before checking Reported-by: "Dan Carpenter" <dan.carpenter@oracle.com> Fixes: f29dd55b0236 ("rdma/siw: queue pair methods") Link: https://lore.kernel.org/r/OF61E386ED.49A73798-ON00258444.003BD6A6-00258444.003CC8D9@notes.na.collabserv.com Signed-off-by: Bernard Metzler <bmt@zurich.ibm.com> Signed-off-by: Doug Ledford <dledford@redhat.com> --- drivers/infiniband/sw/siw/siw_qp.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/sw/siw/siw_qp.c b/drivers/infiniband/sw/siw/siw_qp.c index 11383d9f95ef..e27bd5b35b96 100644 --- a/drivers/infiniband/sw/siw/siw_qp.c +++ b/drivers/infiniband/sw/siw/siw_qp.c @@ -220,12 +220,14 @@ static int siw_qp_enable_crc(struct siw_qp *qp) { struct siw_rx_stream *c_rx = &qp->rx_stream; struct siw_iwarp_tx *c_tx = &qp->tx_ctx; - int size = crypto_shash_descsize(siw_crypto_shash) + - sizeof(struct shash_desc); + int size; if (siw_crypto_shash == NULL) return -ENOENT; + size = crypto_shash_descsize(siw_crypto_shash) + + sizeof(struct shash_desc); + c_tx->mpa_crc_hd = kzalloc(size, GFP_KERNEL); c_rx->mpa_crc_hd = kzalloc(size, GFP_KERNEL); if (!c_tx->mpa_crc_hd || !c_rx->mpa_crc_hd) {