From 3606b99971b66b57a68c952619b43c12f90d2c70 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Fri, 26 Jul 2013 10:00:45 -0700 Subject: RDMA/ocrdma: Remove unused include I'd like to remove rdma/ib_cache.h some day, so let's avoid proliferating uses of it unnecessarily. Signed-off-by: Roland Dreier --- drivers/infiniband/hw/ocrdma/ocrdma_ah.c | 1 - 1 file changed, 1 deletion(-) (limited to 'drivers') diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_ah.c b/drivers/infiniband/hw/ocrdma/ocrdma_ah.c index a877a8ed7907..f4c587c68f64 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_ah.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_ah.c @@ -29,7 +29,6 @@ #include #include -#include #include "ocrdma.h" #include "ocrdma_verbs.h" -- cgit v1.2.3-58-ga151 From ae1fe07f3f422c62a0baf6a777141d7110e8a111 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 25 Jul 2013 19:48:32 +0300 Subject: RDMA/cxgb4: Fix stack info leak in c4iw_create_qp() "uresp.ma_sync_key" doesn't get set on this path so we leak 8 bytes of data. Signed-off-by: Dan Carpenter Acked-by: Steve Wise Signed-off-by: Roland Dreier --- drivers/infiniband/hw/cxgb4/qp.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'drivers') diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c index 232040447e8a..a4975e1654a6 100644 --- a/drivers/infiniband/hw/cxgb4/qp.c +++ b/drivers/infiniband/hw/cxgb4/qp.c @@ -1657,6 +1657,8 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs, if (mm5) { uresp.ma_sync_key = ucontext->key; ucontext->key += PAGE_SIZE; + } else { + uresp.ma_sync_key = 0; } uresp.sq_key = ucontext->key; ucontext->key += PAGE_SIZE; -- cgit v1.2.3-58-ga151 From 63ea37495714849a5a7986244d9d656e0af14961 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 29 Jul 2013 22:34:29 +0300 Subject: RDMA/ocrdma: Fix several stack info leaks A grab bag of places which don't properly initialize stack data. I removed one place which cleared ".rsvd" because it's not needed now that I have added a memset() earlier in the function. Signed-off-by: Dan Carpenter Signed-off-by: Roland Dreier --- drivers/infiniband/hw/ocrdma/ocrdma_verbs.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'drivers') diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c index dcfbab177faa..f36630e4b6be 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c @@ -242,6 +242,7 @@ struct ib_ucontext *ocrdma_alloc_ucontext(struct ib_device *ibdev, memset(ctx->ah_tbl.va, 0, map_len); ctx->ah_tbl.len = map_len; + memset(&resp, 0, sizeof(resp)); resp.ah_tbl_len = ctx->ah_tbl.len; resp.ah_tbl_page = ctx->ah_tbl.pa; @@ -253,7 +254,6 @@ struct ib_ucontext *ocrdma_alloc_ucontext(struct ib_device *ibdev, resp.wqe_size = dev->attr.wqe_size; resp.rqe_size = dev->attr.rqe_size; resp.dpp_wqe_size = dev->attr.wqe_size; - resp.rsvd = 0; memcpy(resp.fw_ver, dev->attr.fw_ver, sizeof(resp.fw_ver)); status = ib_copy_to_udata(udata, &resp, sizeof(resp)); @@ -338,6 +338,7 @@ static int ocrdma_copy_pd_uresp(struct ocrdma_pd *pd, struct ocrdma_alloc_pd_uresp rsp; struct ocrdma_ucontext *uctx = get_ocrdma_ucontext(ib_ctx); + memset(&rsp, 0, sizeof(rsp)); rsp.id = pd->id; rsp.dpp_enabled = pd->dpp_enabled; db_page_addr = pd->dev->nic_info.unmapped_db + @@ -692,6 +693,7 @@ static int ocrdma_copy_cq_uresp(struct ocrdma_cq *cq, struct ib_udata *udata, struct ocrdma_ucontext *uctx; struct ocrdma_create_cq_uresp uresp; + memset(&uresp, 0, sizeof(uresp)); uresp.cq_id = cq->id; uresp.page_size = cq->len; uresp.num_pages = 1; @@ -1460,6 +1462,7 @@ static int ocrdma_copy_srq_uresp(struct ocrdma_srq *srq, struct ib_udata *udata) int status; struct ocrdma_create_srq_uresp uresp; + memset(&uresp, 0, sizeof(uresp)); uresp.rq_dbid = srq->rq.dbid; uresp.num_rq_pages = 1; uresp.rq_page_addr[0] = srq->rq.pa; -- cgit v1.2.3-58-ga151 From 604296303f5f2698e4682c8aefb413554bc4b12a Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 25 Jul 2013 20:04:59 +0300 Subject: RDMA/nes: Fix info leaks in nes_create_qp() and nes_create_cq() We pass a few bytes of uninitialized stack memory to the user here. Signed-off-by: Dan Carpenter Signed-off-by: Roland Dreier --- drivers/infiniband/hw/nes/nes_verbs.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'drivers') diff --git a/drivers/infiniband/hw/nes/nes_verbs.c b/drivers/infiniband/hw/nes/nes_verbs.c index 8f67fe2e91e6..5b53ca5a2284 100644 --- a/drivers/infiniband/hw/nes/nes_verbs.c +++ b/drivers/infiniband/hw/nes/nes_verbs.c @@ -1384,6 +1384,7 @@ static struct ib_qp *nes_create_qp(struct ib_pd *ibpd, if (ibpd->uobject) { uresp.mmap_sq_db_index = nesqp->mmap_sq_db_index; + uresp.mmap_rq_db_index = 0; uresp.actual_sq_size = sq_size; uresp.actual_rq_size = rq_size; uresp.qp_id = nesqp->hwqp.qp_id; @@ -1767,7 +1768,7 @@ static struct ib_cq *nes_create_cq(struct ib_device *ibdev, int entries, resp.cq_id = nescq->hw_cq.cq_number; resp.cq_size = nescq->hw_cq.cq_size; resp.mmap_db_index = 0; - if (ib_copy_to_udata(udata, &resp, sizeof resp)) { + if (ib_copy_to_udata(udata, &resp, sizeof resp - sizeof resp.reserved)) { nes_free_resource(nesadapter, nesadapter->allocated_cqs, cq_num); kfree(nescq); return ERR_PTR(-EFAULT); -- cgit v1.2.3-58-ga151 From 246fcdbc9d6c7ee1b12e299427efbc2ab49e40b5 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 29 Jul 2013 22:19:14 +0300 Subject: RDMA/cxgb3: Fix stack info leak in iwch_create_cq() The "uresp.reserved" field isn't initialized on this path so it could leak uninitialized stack information to the user. Signed-off-by: Dan Carpenter Acked-by: Steve Wise Signed-off-by: Roland Dreier --- drivers/infiniband/hw/cxgb3/iwch_provider.c | 1 + 1 file changed, 1 insertion(+) (limited to 'drivers') diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.c b/drivers/infiniband/hw/cxgb3/iwch_provider.c index e87f2201b220..d2283837d451 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_provider.c +++ b/drivers/infiniband/hw/cxgb3/iwch_provider.c @@ -226,6 +226,7 @@ static struct ib_cq *iwch_create_cq(struct ib_device *ibdev, int entries, int ve mm->len = PAGE_ALIGN(((1UL << uresp.size_log2) + 1) * sizeof(struct t3_cqe)); uresp.memsize = mm->len; + uresp.reserved = 0; resplen = sizeof uresp; } if (ib_copy_to_udata(udata, &uresp, resplen)) { -- cgit v1.2.3-58-ga151 From b268e4db3d04ed83ecc76b0454b12a8884445f88 Mon Sep 17 00:00:00 2001 From: Mike Marciniszyn Date: Fri, 12 Jul 2013 09:24:56 -0400 Subject: IB/qib: Add err_decode() call for ring dump Commit 0b3ddf380ca7 ("Log all SDMA errors unconditionally") missed part of the patch. This also corrects a format warning when dma_addr_t is 32 bits on a 64 bit system. Signed-off-by: Mike Marciniszyn Signed-off-by: Roland Dreier --- drivers/infiniband/hw/qib/qib_iba7322.c | 2 ++ drivers/infiniband/hw/qib/qib_sdma.c | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'drivers') diff --git a/drivers/infiniband/hw/qib/qib_iba7322.c b/drivers/infiniband/hw/qib/qib_iba7322.c index 21e8b09d4bf8..016e7429adf6 100644 --- a/drivers/infiniband/hw/qib/qib_iba7322.c +++ b/drivers/infiniband/hw/qib/qib_iba7322.c @@ -1596,6 +1596,8 @@ static void sdma_7322_p_errors(struct qib_pportdata *ppd, u64 errs) struct qib_devdata *dd = ppd->dd; errs &= QIB_E_P_SDMAERRS; + err_decode(ppd->cpspec->sdmamsgbuf, sizeof(ppd->cpspec->sdmamsgbuf), + errs, qib_7322p_error_msgs); if (errs & QIB_E_P_SDMAUNEXPDATA) qib_dev_err(dd, "IB%u:%u SDmaUnexpData\n", dd->unit, diff --git a/drivers/infiniband/hw/qib/qib_sdma.c b/drivers/infiniband/hw/qib/qib_sdma.c index 32162d355370..9b5322d8cd5a 100644 --- a/drivers/infiniband/hw/qib/qib_sdma.c +++ b/drivers/infiniband/hw/qib/qib_sdma.c @@ -717,7 +717,7 @@ void dump_sdma_state(struct qib_pportdata *ppd) struct qib_sdma_txreq *txp, *txpnext; __le64 *descqp; u64 desc[2]; - dma_addr_t addr; + u64 addr; u16 gen, dwlen, dwoffset; u16 head, tail, cnt; -- cgit v1.2.3-58-ga151 From 3c93f039d29326954b2c503a71e1eab9ae49063b Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Tue, 30 Jul 2013 15:48:35 -0700 Subject: Revert "RDMA/nes: Fix compilation error when nes_debug is enabled" This reverts commit bca1935ccdec, which removes variables nes_tcp_state_str and nes_iwarp_state_str, assuming that they aren't defined. However, they are defined within a #ifdef NES_DEBUG statement, which if enabled causes "defined but not used" compiler warning, when the variables are removed. Signed-off-by: Tatyana Nikolova Reported-by: Stephen Rothwell Signed-off-by: Roland Dreier --- drivers/infiniband/hw/nes/nes_hw.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers') diff --git a/drivers/infiniband/hw/nes/nes_hw.c b/drivers/infiniband/hw/nes/nes_hw.c index 418004c93feb..90200245c5eb 100644 --- a/drivers/infiniband/hw/nes/nes_hw.c +++ b/drivers/infiniband/hw/nes/nes_hw.c @@ -3570,10 +3570,10 @@ static void nes_process_iwarp_aeqe(struct nes_device *nesdev, tcp_state = (aeq_info & NES_AEQE_TCP_STATE_MASK) >> NES_AEQE_TCP_STATE_SHIFT; iwarp_state = (aeq_info & NES_AEQE_IWARP_STATE_MASK) >> NES_AEQE_IWARP_STATE_SHIFT; nes_debug(NES_DBG_AEQ, "aeid = 0x%04X, qp-cq id = %d, aeqe = %p," - " Tcp state = %d, iWARP state = %d\n", + " Tcp state = %s, iWARP state = %s\n", async_event_id, le32_to_cpu(aeqe->aeqe_words[NES_AEQE_COMP_QP_CQ_ID_IDX]), aeqe, - tcp_state, iwarp_state); + nes_tcp_state_str[tcp_state], nes_iwarp_state_str[iwarp_state]); aeqe_cq_id = le32_to_cpu(aeqe->aeqe_words[NES_AEQE_COMP_QP_CQ_ID_IDX]); if (aeq_info & NES_AEQE_QP) { -- cgit v1.2.3-58-ga151 From 3eac103f83ab0c2013b734981d92d56be51a1a9c Mon Sep 17 00:00:00 2001 From: Jack Morgenstein Date: Thu, 18 Jul 2013 14:02:30 +0300 Subject: IB/mlx4: Use default pkey when creating tunnel QPs When creating tunnel QPs for special QP tunneling, look for the default pkey in the slave's virtual pkey table. If it is present, use the real pkey index where the default pkey is located. If the default pkey is not found in the pkey table, use the real pkey index which is stored at index 0 in the slave's virtual pkey table (this is the current behavior). This change is required to support cloud computing, where the paravirtualized index of the default pkey is moved to index 1 or higher. The pkey at paravirtualized index 0 is used for the default IPoIB interface created by the VF. Its possible for the pkey value at paravirtualized index 0 to be invalid (zero) at VF probe time (pkey index 0 is mapped to real pkey index 127, which contains pkey = 0). At some point after the VF probe, the cloud computing interface at the hypervisor maps virtual index 0 for the VF to the pkey index containing the pkey that IPoIB will use in its operation. However, when the tunnel QP is created, the pkey at the slave's virtual index 0 is still mapped to the invalid pkey index, so tunnel QP creation fails. This commit causes the hypervisor to search for the default pkey in the slave's pkey table -- and this pkey is present in the table (at index > 0) at tunnel QP creation time, so that the tunnel QP creation will succeed. Signed-off-by: Jack Morgenstein Signed-off-by: Or Gerlitz Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx4/mad.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'drivers') diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c index 4d599cedbb0b..f2a3f48107e7 100644 --- a/drivers/infiniband/hw/mlx4/mad.c +++ b/drivers/infiniband/hw/mlx4/mad.c @@ -1511,8 +1511,14 @@ static int create_pv_sqp(struct mlx4_ib_demux_pv_ctx *ctx, memset(&attr, 0, sizeof attr); attr.qp_state = IB_QPS_INIT; - attr.pkey_index = - to_mdev(ctx->ib_dev)->pkeys.virt2phys_pkey[ctx->slave][ctx->port - 1][0]; + ret = 0; + if (create_tun) + ret = find_slave_port_pkey_ix(to_mdev(ctx->ib_dev), ctx->slave, + ctx->port, IB_DEFAULT_PKEY_FULL, + &attr.pkey_index); + if (ret || !create_tun) + attr.pkey_index = + to_mdev(ctx->ib_dev)->pkeys.virt2phys_pkey[ctx->slave][ctx->port - 1][0]; attr.qkey = IB_QP1_QKEY; attr.port_num = ctx->port; ret = ib_modify_qp(tun_qp->qp, &attr, qp_attr_mask_INIT); -- cgit v1.2.3-58-ga151 From 281d1a9211986cfdccbefcd6128049a0a0d473b2 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 30 Jul 2013 07:54:26 +0800 Subject: IB/mlx5: Fix error return code in init_one() Fix to return a negative error code from the error handling case instead of 0, as done elsewhere in this function. Signed-off-by: Wei Yongjun Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx5/main.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'drivers') diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 8000fff4d444..b1cbf338bcf6 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -1426,7 +1426,8 @@ static int init_one(struct pci_dev *pdev, if (err) goto err_eqs; - if (ib_register_device(&dev->ib_dev, NULL)) + err = ib_register_device(&dev->ib_dev, NULL); + if (err) goto err_rsrc; err = create_umr_res(dev); @@ -1434,8 +1435,9 @@ static int init_one(struct pci_dev *pdev, goto err_dev; for (i = 0; i < ARRAY_SIZE(mlx5_class_attributes); i++) { - if (device_create_file(&dev->ib_dev.dev, - mlx5_class_attributes[i])) + err = device_create_file(&dev->ib_dev.dev, + mlx5_class_attributes[i]); + if (err) goto err_umrc; } -- cgit v1.2.3-58-ga151 From 92b0ca7cb149d20bd97a2ad88d860745045a77dc Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 25 Jul 2013 20:04:36 +0300 Subject: IB/mlx5: Fix stack info leak in mlx5_ib_alloc_ucontext() We don't set "resp.reserved". Since it's at the end of the struct that means we don't have to copy it to the user. Signed-off-by: Dan Carpenter Acked-by: Eli Cohen Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx5/main.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'drivers') diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index b1cbf338bcf6..3f831de9a4d8 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -619,7 +619,8 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev, resp.tot_uuars = req.total_num_uuars; resp.num_ports = dev->mdev.caps.num_ports; - err = ib_copy_to_udata(udata, &resp, sizeof(resp)); + err = ib_copy_to_udata(udata, &resp, + sizeof(resp) - sizeof(resp.reserved)); if (err) goto out_uars; -- cgit v1.2.3-58-ga151 From 11940c8728b8e4bd67d0afacefbb9391ff8f4201 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 22 Jul 2013 11:02:01 +0300 Subject: mlx5_core: Fix use after free in mlx5_cmd_comp_handler() We can't dereference "ent" after passing it to free_cmd(). Signed-off-by: Dan Carpenter Acked-by: Eli Cohen Signed-off-by: Roland Dreier --- drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'drivers') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index 205753a04cfc..40374063c01e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -1113,7 +1113,13 @@ void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, unsigned long vector) for (i = 0; i < (1 << cmd->log_sz); i++) { if (test_bit(i, &vector)) { + struct semaphore *sem; + ent = cmd->ent_arr[i]; + if (ent->page_queue) + sem = &cmd->pages_sem; + else + sem = &cmd->sem; ktime_get_ts(&ent->ts2); memcpy(ent->out->first.data, ent->lay->out, sizeof(ent->lay->out)); dump_command(dev, ent, 0); @@ -1136,10 +1142,7 @@ void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, unsigned long vector) } else { complete(&ent->done); } - if (ent->page_queue) - up(&cmd->pages_sem); - else - up(&cmd->sem); + up(sem); } } } -- cgit v1.2.3-58-ga151 From cd23b14b654769db83c9684ae1ba32c0e066670f Mon Sep 17 00:00:00 2001 From: Eli Cohen Date: Thu, 18 Jul 2013 15:31:08 +0300 Subject: mlx5_core: Implement new initialization sequence Introduce enbale_hca and disable_hca commands to signify when the driver starts or ceases to operate on the device. In addition the driver will use boot and init pages count; boot pages is required to allow firmware to complete boot commands and the other to complete init hca. Command interface revision is bumped to 4 to enforce using supported firmware. This patch breaks compatibility with old versions of firmware (< 4); however, the first GA firmware we will publish will support version 4 so this should not be a problem. Signed-off-by: Eli Cohen Signed-off-by: Roland Dreier --- drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 8 ++- drivers/net/ethernet/mellanox/mlx5/core/main.c | 69 +++++++++++++++++++--- .../net/ethernet/mellanox/mlx5/core/pagealloc.c | 20 +++++-- include/linux/mlx5/device.h | 20 +++++++ include/linux/mlx5/driver.h | 4 +- 5 files changed, 106 insertions(+), 15 deletions(-) (limited to 'drivers') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index 40374063c01e..c571de85d0f9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -46,7 +46,7 @@ #include "mlx5_core.h" enum { - CMD_IF_REV = 3, + CMD_IF_REV = 4, }; enum { @@ -282,6 +282,12 @@ const char *mlx5_command_str(int command) case MLX5_CMD_OP_TEARDOWN_HCA: return "TEARDOWN_HCA"; + case MLX5_CMD_OP_ENABLE_HCA: + return "MLX5_CMD_OP_ENABLE_HCA"; + + case MLX5_CMD_OP_DISABLE_HCA: + return "MLX5_CMD_OP_DISABLE_HCA"; + case MLX5_CMD_OP_QUERY_PAGES: return "QUERY_PAGES"; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 12242de2b0e3..b47739b0b5f6 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -249,6 +249,44 @@ static int set_hca_ctrl(struct mlx5_core_dev *dev) return err; } +static int mlx5_core_enable_hca(struct mlx5_core_dev *dev) +{ + int err; + struct mlx5_enable_hca_mbox_in in; + struct mlx5_enable_hca_mbox_out out; + + memset(&in, 0, sizeof(in)); + memset(&out, 0, sizeof(out)); + in.hdr.opcode = cpu_to_be16(MLX5_CMD_OP_ENABLE_HCA); + err = mlx5_cmd_exec(dev, &in, sizeof(in), &out, sizeof(out)); + if (err) + return err; + + if (out.hdr.status) + return mlx5_cmd_status_to_err(&out.hdr); + + return 0; +} + +static int mlx5_core_disable_hca(struct mlx5_core_dev *dev) +{ + int err; + struct mlx5_disable_hca_mbox_in in; + struct mlx5_disable_hca_mbox_out out; + + memset(&in, 0, sizeof(in)); + memset(&out, 0, sizeof(out)); + in.hdr.opcode = cpu_to_be16(MLX5_CMD_OP_DISABLE_HCA); + err = mlx5_cmd_exec(dev, &in, sizeof(in), &out, sizeof(out)); + if (err) + return err; + + if (out.hdr.status) + return mlx5_cmd_status_to_err(&out.hdr); + + return 0; +} + int mlx5_dev_init(struct mlx5_core_dev *dev, struct pci_dev *pdev) { struct mlx5_priv *priv = &dev->priv; @@ -304,28 +342,41 @@ int mlx5_dev_init(struct mlx5_core_dev *dev, struct pci_dev *pdev) } mlx5_pagealloc_init(dev); + + err = mlx5_core_enable_hca(dev); + if (err) { + dev_err(&pdev->dev, "enable hca failed\n"); + goto err_pagealloc_cleanup; + } + + err = mlx5_satisfy_startup_pages(dev, 1); + if (err) { + dev_err(&pdev->dev, "failed to allocate boot pages\n"); + goto err_disable_hca; + } + err = set_hca_ctrl(dev); if (err) { dev_err(&pdev->dev, "set_hca_ctrl failed\n"); - goto err_pagealloc_cleanup; + goto reclaim_boot_pages; } err = handle_hca_cap(dev); if (err) { dev_err(&pdev->dev, "handle_hca_cap failed\n"); - goto err_pagealloc_cleanup; + goto reclaim_boot_pages; } - err = mlx5_satisfy_startup_pages(dev); + err = mlx5_satisfy_startup_pages(dev, 0); if (err) { - dev_err(&pdev->dev, "failed to allocate startup pages\n"); - goto err_pagealloc_cleanup; + dev_err(&pdev->dev, "failed to allocate init pages\n"); + goto reclaim_boot_pages; } err = mlx5_pagealloc_start(dev); if (err) { dev_err(&pdev->dev, "mlx5_pagealloc_start failed\n"); - goto err_reclaim_pages; + goto reclaim_boot_pages; } err = mlx5_cmd_init_hca(dev); @@ -396,9 +447,12 @@ err_stop_poll: err_pagealloc_stop: mlx5_pagealloc_stop(dev); -err_reclaim_pages: +reclaim_boot_pages: mlx5_reclaim_startup_pages(dev); +err_disable_hca: + mlx5_core_disable_hca(dev); + err_pagealloc_cleanup: mlx5_pagealloc_cleanup(dev); mlx5_cmd_cleanup(dev); @@ -434,6 +488,7 @@ void mlx5_dev_cleanup(struct mlx5_core_dev *dev) mlx5_cmd_teardown_hca(dev); mlx5_pagealloc_stop(dev); mlx5_reclaim_startup_pages(dev); + mlx5_core_disable_hca(dev); mlx5_pagealloc_cleanup(dev); mlx5_cmd_cleanup(dev); iounmap(dev->iseg); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c index f0bf46339b28..4a3e137931a3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c @@ -64,7 +64,7 @@ struct mlx5_query_pages_inbox { struct mlx5_query_pages_outbox { struct mlx5_outbox_hdr hdr; - u8 reserved[2]; + __be16 num_boot_pages; __be16 func_id; __be16 init_pages; __be16 num_pages; @@ -146,7 +146,7 @@ static struct page *remove_page(struct mlx5_core_dev *dev, u64 addr) } static int mlx5_cmd_query_pages(struct mlx5_core_dev *dev, u16 *func_id, - s16 *pages, s16 *init_pages) + s16 *pages, s16 *init_pages, u16 *boot_pages) { struct mlx5_query_pages_inbox in; struct mlx5_query_pages_outbox out; @@ -164,8 +164,13 @@ static int mlx5_cmd_query_pages(struct mlx5_core_dev *dev, u16 *func_id, if (pages) *pages = be16_to_cpu(out.num_pages); + if (init_pages) *init_pages = be16_to_cpu(out.init_pages); + + if (boot_pages) + *boot_pages = be16_to_cpu(out.num_boot_pages); + *func_id = be16_to_cpu(out.func_id); return err; @@ -357,19 +362,22 @@ void mlx5_core_req_pages_handler(struct mlx5_core_dev *dev, u16 func_id, queue_work(dev->priv.pg_wq, &req->work); } -int mlx5_satisfy_startup_pages(struct mlx5_core_dev *dev) +int mlx5_satisfy_startup_pages(struct mlx5_core_dev *dev, int boot) { + u16 uninitialized_var(boot_pages); s16 uninitialized_var(init_pages); u16 uninitialized_var(func_id); int err; - err = mlx5_cmd_query_pages(dev, &func_id, NULL, &init_pages); + err = mlx5_cmd_query_pages(dev, &func_id, NULL, &init_pages, + &boot_pages); if (err) return err; - mlx5_core_dbg(dev, "requested %d init pages for func_id 0x%x\n", init_pages, func_id); - return give_pages(dev, func_id, init_pages, 0); + mlx5_core_dbg(dev, "requested %d init pages and %d boot pages for func_id 0x%x\n", + init_pages, boot_pages, func_id); + return give_pages(dev, func_id, boot ? boot_pages : init_pages, 0); } static int optimal_reclaimed_pages(void) diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h index 8de8d8f22384..737685e9e852 100644 --- a/include/linux/mlx5/device.h +++ b/include/linux/mlx5/device.h @@ -690,6 +690,26 @@ struct mlx5_query_cq_mbox_out { __be64 pas[0]; }; +struct mlx5_enable_hca_mbox_in { + struct mlx5_inbox_hdr hdr; + u8 rsvd[8]; +}; + +struct mlx5_enable_hca_mbox_out { + struct mlx5_outbox_hdr hdr; + u8 rsvd[8]; +}; + +struct mlx5_disable_hca_mbox_in { + struct mlx5_inbox_hdr hdr; + u8 rsvd[8]; +}; + +struct mlx5_disable_hca_mbox_out { + struct mlx5_outbox_hdr hdr; + u8 rsvd[8]; +}; + struct mlx5_eq_context { u8 status; u8 ec_oi; diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index f22e4419839b..2aa258b0ced1 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -101,6 +101,8 @@ enum { MLX5_CMD_OP_QUERY_ADAPTER = 0x101, MLX5_CMD_OP_INIT_HCA = 0x102, MLX5_CMD_OP_TEARDOWN_HCA = 0x103, + MLX5_CMD_OP_ENABLE_HCA = 0x104, + MLX5_CMD_OP_DISABLE_HCA = 0x105, MLX5_CMD_OP_QUERY_PAGES = 0x107, MLX5_CMD_OP_MANAGE_PAGES = 0x108, MLX5_CMD_OP_SET_HCA_CAP = 0x109, @@ -690,7 +692,7 @@ int mlx5_pagealloc_start(struct mlx5_core_dev *dev); void mlx5_pagealloc_stop(struct mlx5_core_dev *dev); void mlx5_core_req_pages_handler(struct mlx5_core_dev *dev, u16 func_id, s16 npages); -int mlx5_satisfy_startup_pages(struct mlx5_core_dev *dev); +int mlx5_satisfy_startup_pages(struct mlx5_core_dev *dev, int boot); int mlx5_reclaim_startup_pages(struct mlx5_core_dev *dev); void mlx5_register_debugfs(void); void mlx5_unregister_debugfs(void); -- cgit v1.2.3-58-ga151 From 618af3846be1d29118b00529e68971e8c95d5b72 Mon Sep 17 00:00:00 2001 From: Andi Shyti Date: Tue, 16 Jul 2013 15:35:01 +0200 Subject: mlx5_core: Variable may be used uninitialized In the sq_overhead() function, if qp_typ is equal to IB_QPT_RC, size will be used uninitialized. Signed-off-by: Andi Shyti Acked-by: Eli Cohen Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx5/qp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers') diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index 16ac54c9819f..045f8cdbd303 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -199,7 +199,7 @@ static int set_rq_size(struct mlx5_ib_dev *dev, struct ib_qp_cap *cap, static int sq_overhead(enum ib_qp_type qp_type) { - int size; + int size = 0; switch (qp_type) { case IB_QPT_XRC_INI: -- cgit v1.2.3-58-ga151 From ef5ed4166ffc192434a9d7324339ed1d308fc2dc Mon Sep 17 00:00:00 2001 From: Jack Morgenstein Date: Thu, 18 Jul 2013 14:02:29 +0300 Subject: IB/core: Create QP1 using the pkey index which contains the default pkey Currently, QP1 is created using pkey index 0. This patch simply looks for the index containing the default pkey, rather than hard-coding pkey index 0. This change will have no effect in native mode, since QP0 and QP1 are created before the SM configures the port, so pkey table will still be the default table defined by the IB Spec, in C10-123: "If non-volatile storage is not used to hold P_Key Table contents, then if a PM (Partition Manager) is not present, and prior to PM initialization of the P_Key Table, the P_Key Table must act as if it contains a single valid entry, at P_Key_ix = 0, containing the default partition key. All other entries in the P_Key Table must be invalid." Thus, in the native mode case, the driver will find the default pkey at index 0 (so it will be no different than the hard-coding). However, in SR-IOV mode, for VFs, the pkey table may be paravirtualized, so that the VF's pkey index zero may not necessarily be mapped to the real pkey index 0. For VFs, therefore, it is important to find the virtual index which maps to the real default pkey. This commit does the following for QP1 creation: 1. Find the pkey index containing the default pkey, and use that index if found. ib_find_pkey() returns the index of the limited-membership default pkey (0x7FFF) if the full-member default pkey is not in the table. 2. If neither form of the default pkey is found, use pkey index 0 (previous behavior). Signed-off-by: Jack Morgenstein Signed-off-by: Or Gerlitz Reviewed-by: Sean Hefty Signed-off-by: Roland Dreier --- drivers/infiniband/core/mad.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'drivers') diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c index dc3fd1e8af07..4c837e66516b 100644 --- a/drivers/infiniband/core/mad.c +++ b/drivers/infiniband/core/mad.c @@ -2663,6 +2663,7 @@ static int ib_mad_port_start(struct ib_mad_port_private *port_priv) int ret, i; struct ib_qp_attr *attr; struct ib_qp *qp; + u16 pkey_index; attr = kmalloc(sizeof *attr, GFP_KERNEL); if (!attr) { @@ -2670,6 +2671,11 @@ static int ib_mad_port_start(struct ib_mad_port_private *port_priv) return -ENOMEM; } + ret = ib_find_pkey(port_priv->device, port_priv->port_num, + IB_DEFAULT_PKEY_FULL, &pkey_index); + if (ret) + pkey_index = 0; + for (i = 0; i < IB_MAD_QPS_CORE; i++) { qp = port_priv->qp_info[i].qp; if (!qp) @@ -2680,7 +2686,7 @@ static int ib_mad_port_start(struct ib_mad_port_private *port_priv) * one is needed for the Reset to Init transition */ attr->qp_state = IB_QPS_INIT; - attr->pkey_index = 0; + attr->pkey_index = pkey_index; attr->qkey = (qp->qp_num == 0) ? 0 : IB_QP1_QKEY; ret = ib_modify_qp(qp, attr, IB_QP_STATE | IB_QP_PKEY_INDEX | IB_QP_QKEY); -- cgit v1.2.3-58-ga151 From 3d790a4c269aefef8f6d32efb07333bc65bd43fe Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Thu, 18 Jul 2013 14:02:31 +0300 Subject: IPoIB: Make sure child devices use valid/proper pkeys Make sure that the IB invalid pkey (0x0000 or 0x8000) isn't used for child devices. Also, make sure to always set the full membership bit for the pkey of devices created by rtnl link ops. Signed-off-by: Or Gerlitz Signed-off-by: Roland Dreier --- drivers/infiniband/ulp/ipoib/ipoib_main.c | 2 +- drivers/infiniband/ulp/ipoib/ipoib_netlink.c | 9 +++++++++ 2 files changed, 10 insertions(+), 1 deletion(-) (limited to 'drivers') diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index b6e049a3c7a8..c6f71a88c55c 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -1461,7 +1461,7 @@ static ssize_t create_child(struct device *dev, if (sscanf(buf, "%i", &pkey) != 1) return -EINVAL; - if (pkey < 0 || pkey > 0xffff) + if (pkey <= 0 || pkey > 0xffff || pkey == 0x8000) return -EINVAL; /* diff --git a/drivers/infiniband/ulp/ipoib/ipoib_netlink.c b/drivers/infiniband/ulp/ipoib/ipoib_netlink.c index 74685936c948..f81abe16cf09 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_netlink.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_netlink.c @@ -119,6 +119,15 @@ static int ipoib_new_child_link(struct net *src_net, struct net_device *dev, } else child_pkey = nla_get_u16(data[IFLA_IPOIB_PKEY]); + if (child_pkey == 0 || child_pkey == 0x8000) + return -EINVAL; + + /* + * Set the full membership bit, so that we join the right + * broadcast group, etc. + */ + child_pkey |= 0x8000; + err = __ipoib_vlan_add(ppriv, netdev_priv(dev), child_pkey, IPOIB_RTNL_CHILD); if (!err && data) -- cgit v1.2.3-58-ga151 From c2904141696ee19551f1553944446f23cdd5d95e Mon Sep 17 00:00:00 2001 From: Erez Shitrit Date: Thu, 18 Jul 2013 14:02:32 +0300 Subject: IPoIB: Fix pkey change flow for virtualization environments IPoIB's required behaviour w.r.t to the pkey used by the device is the following: - For "parent" interfaces (e.g ib0, ib1, etc) who are created automatically as a result of hot-plug events from the IB core, the driver needs to take whatever pkey vlaue it finds in index 0, and stick to that index. - For child interfaces (e.g ib0.8001, etc) created by admin directive, the driver needs to use and stick to the value provided during its creation. In SR-IOV environment its possible for the VF probe to take place before the cloud management software provisions the suitable pkey for the VF in the paravirtualed PKEY table index 0. When this is the case, the VF IB stack will find in index 0 an invalide pkey, which is all zeros. Moreover, the cloud managment can assign the pkey value at index 0 at any time of the guest life cycle. The correct behavior for IPoIB to address these requirements for parent interfaces is to use PKEY_CHANGE event as trigger to optionally re-init the device pkey value and re-create all the relevant resources accordingly, if the value of the pkey in index 0 has changed (from invalid to valid or from valid value X to invalid value Y). This patch enhances the heavy flushing code which is triggered by pkey change event, to behave correctly for parent devices. For child devices, the code remains the same, namely chases pkey value and not index. Signed-off-by: Erez Shitrit Signed-off-by: Or Gerlitz Signed-off-by: Roland Dreier --- drivers/infiniband/ulp/ipoib/ipoib_ib.c | 76 +++++++++++++++++++++++++++------ 1 file changed, 63 insertions(+), 13 deletions(-) (limited to 'drivers') diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c index 2cfa76f5d99e..196b1d13cbcb 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c @@ -932,12 +932,47 @@ int ipoib_ib_dev_init(struct net_device *dev, struct ib_device *ca, int port) return 0; } +/* + * Takes whatever value which is in pkey index 0 and updates priv->pkey + * returns 0 if the pkey value was changed. + */ +static inline int update_parent_pkey(struct ipoib_dev_priv *priv) +{ + int result; + u16 prev_pkey; + + prev_pkey = priv->pkey; + result = ib_query_pkey(priv->ca, priv->port, 0, &priv->pkey); + if (result) { + ipoib_warn(priv, "ib_query_pkey port %d failed (ret = %d)\n", + priv->port, result); + return result; + } + + priv->pkey |= 0x8000; + + if (prev_pkey != priv->pkey) { + ipoib_dbg(priv, "pkey changed from 0x%x to 0x%x\n", + prev_pkey, priv->pkey); + /* + * Update the pkey in the broadcast address, while making sure to set + * the full membership bit, so that we join the right broadcast group. + */ + priv->dev->broadcast[8] = priv->pkey >> 8; + priv->dev->broadcast[9] = priv->pkey & 0xff; + return 0; + } + + return 1; +} + static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, enum ipoib_flush_level level) { struct ipoib_dev_priv *cpriv; struct net_device *dev = priv->dev; u16 new_index; + int result; mutex_lock(&priv->vlan_mutex); @@ -951,6 +986,10 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, mutex_unlock(&priv->vlan_mutex); if (!test_bit(IPOIB_FLAG_INITIALIZED, &priv->flags)) { + /* for non-child devices must check/update the pkey value here */ + if (level == IPOIB_FLUSH_HEAVY && + !test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags)) + update_parent_pkey(priv); ipoib_dbg(priv, "Not flushing - IPOIB_FLAG_INITIALIZED not set.\n"); return; } @@ -961,21 +1000,32 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, } if (level == IPOIB_FLUSH_HEAVY) { - if (ib_find_pkey(priv->ca, priv->port, priv->pkey, &new_index)) { - clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags); - ipoib_ib_dev_down(dev, 0); - ipoib_ib_dev_stop(dev, 0); - if (ipoib_pkey_dev_delay_open(dev)) + /* child devices chase their origin pkey value, while non-child + * (parent) devices should always takes what present in pkey index 0 + */ + if (test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags)) { + if (ib_find_pkey(priv->ca, priv->port, priv->pkey, &new_index)) { + clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags); + ipoib_ib_dev_down(dev, 0); + ipoib_ib_dev_stop(dev, 0); + if (ipoib_pkey_dev_delay_open(dev)) + return; + } + /* restart QP only if P_Key index is changed */ + if (test_and_set_bit(IPOIB_PKEY_ASSIGNED, &priv->flags) && + new_index == priv->pkey_index) { + ipoib_dbg(priv, "Not flushing - P_Key index not changed.\n"); return; + } + priv->pkey_index = new_index; + } else { + result = update_parent_pkey(priv); + /* restart QP only if P_Key value changed */ + if (result) { + ipoib_dbg(priv, "Not flushing - P_Key value not changed.\n"); + return; + } } - - /* restart QP only if P_Key index is changed */ - if (test_and_set_bit(IPOIB_PKEY_ASSIGNED, &priv->flags) && - new_index == priv->pkey_index) { - ipoib_dbg(priv, "Not flushing - P_Key index not changed.\n"); - return; - } - priv->pkey_index = new_index; } if (level == IPOIB_FLUSH_LIGHT) { -- cgit v1.2.3-58-ga151