diff options
| author | Linus Torvalds | 2017-05-03 21:45:55 +0200 |
|---|---|---|
| committer | Linus Torvalds | 2017-05-03 21:45:55 +0200 |
| commit | 1684096b1ed813f621fb6cbd06e72235c1c2a0ca (patch) | |
| tree | 13a228c35d6344f5d23b2c195aa3b026e42aac4b /drivers/infiniband/hw | |
| parent | Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dto... (diff) | |
| parent | infiniband: avoid dereferencing uninitialized dst on error path (diff) | |
| download | kernel-qcow2-linux-1684096b1ed813f621fb6cbd06e72235c1c2a0ca.tar.gz kernel-qcow2-linux-1684096b1ed813f621fb6cbd06e72235c1c2a0ca.tar.xz kernel-qcow2-linux-1684096b1ed813f621fb6cbd06e72235c1c2a0ca.zip | |
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma
Pull rdma updates from Doug Ledford:
"More exchaustive description of primary updates in this release:
- Lots of driver fixes and misc fixes across the board.
- I had to base on a net-next tree because the IPoIB Accelorator
patches needed it.
Unfortunately, it was known to Mellanox that there would need to be
an IPoIB accelorator patch to the net tree (which left some
functions turned off by an #ifdef construct to avoid warnings about
defined but unused functions), then one to the RDMA tree, then a
fixup that went back and re-enabled the functions in the net tree
and enabled their use in the rdma tree
Also, a sparse fix was sent to the net tree after I did my pull,
and the fixup patch conflicts quite directly with that sparse fix,
so I'm going to submit the fixup patch towards the end of the merge
window by itself and based upon your master branch at the time.
- Two separate rounds of hfi1 fixes, one that got dropped from last
release because it came in just a day or two before the end of the
merge window and then the one from this release cycle.
Of note is that I now have a third series that just landed from
Intel yesterday. It is not included in this pull request, but I may
submit it by the end of the week. I'll talk to Intel about
improving the timing of thier submissions for my workflow.
- Changes to our idr usage in the RDMA subsystem that will tie into
our cgroup management and also into the upcoming changes for the
RDMA kernel<->userspace API.
- Addition of support for a netdev to be tied to an RDMA device at
the core level
- Addition of the VNIC driver from Intel.
While IPoIB provides IP over InfiniBand (and *only* IP, no lower
layer protocol headers are allowed or supported), the VNIC driver
presents a virtual Ethernet device with support for things like
varying Ethertypes, VLANs, priorities and other features of
Ethernet.
The virtual devices are centrally managed by the OPA fabric
manager, making this (for the time being) a strictly OPA specific
feature.
- Improvements to the On-Demand Paging support in the RDMA subsystem.
- Addition of three significant OPA changes.
While we added OPA support some time ago (via the hfi1 driver), the
RDMA subsystem has so far glossed over the areas where OPA and
InfiniBand differ.
With this release we are starting to add support for the OPA
extensions into the RDMA core in the following area: Extended port
information for OPA is now supported, extended Address Handle
attributes for OPA are now supported, and extended SA Queries to
get OPA specific subnet information is now supported.
Concise summary from the tag:
- idr usage and locking changes
- build fix for hns
- ipoib debug path record file fix
- hfi1 updates
- core RDMA netdev addition
- Intel VNIC driver addition
- Enhanced accelerators for IPoIB addition
- Debug cleanups in cxgb3/4
- Trivial cleanups from SF Markus Elfring
- Misc rxe fixes from Mellanox
- Misc ipoib fixes from Mellanox
- Lots of mlx4/mlx5 changes from Mellanox
- Misc fixes across the RDMA subsystem
- ODP paging fixes and improvements
- qedr updates
- hfi1 updates
- OPA port info patches
- OPA AH patches
- OPA SA Query patches"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma: (191 commits)
infiniband: avoid dereferencing uninitialized dst on error path
IB/SA: Add OPA addr header
IB/mlx5: Add port_xmit_wait to counter registers read
IB/ocrdma: fix out of bounds access to local buffer
IB/mlx4: Fix incorrect order of formal and actual parameters
IB/mlx4: Change flush logic so it adheres to the variable name
mlx5: Fix mlx5_ib_map_mr_sg mr length
IB/rxe: Don't clamp residual length to mtu
IB/SA: Add support to query OPA path records
IB/SA: Add OPA path record type
IB/SA: Split struct sa_path_rec based on IB and ROCE specific fields
IB/SA: Introduce path record specific types
IB/SA: Rename ib_sa_path_rec to sa_path_rec
IB/CM: Add braces when using sizeof
IB/core: Define 'opa' rdma_ah_attr type
IB/core: Define 'ib' and 'roce' rdma_ah_attr types
IB/core: Use rdma_ah_attr accessor functions
IB/core: Add accessor functions for rdma_ah_attr fields
IB/PVRDMA: Rename ib_ah_attr related functions
IB/mthca: Rename to_ib_ah_attr to to_rdma_ah_attr
...
Diffstat (limited to 'drivers/infiniband/hw')
136 files changed, 5628 insertions, 2698 deletions
diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.c b/drivers/infiniband/hw/bnxt_re/ib_verbs.c index 33af2e3de399..7ba9e699d7ab 100644 --- a/drivers/infiniband/hw/bnxt_re/ib_verbs.c +++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.c @@ -524,19 +524,20 @@ int bnxt_re_destroy_ah(struct ib_ah *ib_ah) } struct ib_ah *bnxt_re_create_ah(struct ib_pd *ib_pd, - struct ib_ah_attr *ah_attr, + struct rdma_ah_attr *ah_attr, struct ib_udata *udata) { struct bnxt_re_pd *pd = container_of(ib_pd, struct bnxt_re_pd, ib_pd); struct bnxt_re_dev *rdev = pd->rdev; struct bnxt_re_ah *ah; + const struct ib_global_route *grh = rdma_ah_read_grh(ah_attr); int rc; u16 vlan_tag; u8 nw_type; struct ib_gid_attr sgid_attr; - if (!(ah_attr->ah_flags & IB_AH_GRH)) { + if (!(rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH)) { dev_err(rdev_to_dev(rdev), "Failed to alloc AH: GRH not set"); return ERR_PTR(-EINVAL); } @@ -548,33 +549,33 @@ struct ib_ah *bnxt_re_create_ah(struct ib_pd *ib_pd, ah->qplib_ah.pd = &pd->qplib_pd; /* Supply the configuration for the HW */ - memcpy(ah->qplib_ah.dgid.data, ah_attr->grh.dgid.raw, + memcpy(ah->qplib_ah.dgid.data, grh->dgid.raw, sizeof(union ib_gid)); /* * If RoCE V2 is enabled, stack will have two entries for * each GID entry. Avoiding this duplicte entry in HW. Dividing * the GID index by 2 for RoCE V2 */ - ah->qplib_ah.sgid_index = ah_attr->grh.sgid_index / 2; - ah->qplib_ah.host_sgid_index = ah_attr->grh.sgid_index; - ah->qplib_ah.traffic_class = ah_attr->grh.traffic_class; - ah->qplib_ah.flow_label = ah_attr->grh.flow_label; - ah->qplib_ah.hop_limit = ah_attr->grh.hop_limit; - ah->qplib_ah.sl = ah_attr->sl; + ah->qplib_ah.sgid_index = grh->sgid_index / 2; + ah->qplib_ah.host_sgid_index = grh->sgid_index; + ah->qplib_ah.traffic_class = grh->traffic_class; + ah->qplib_ah.flow_label = grh->flow_label; + ah->qplib_ah.hop_limit = grh->hop_limit; + ah->qplib_ah.sl = rdma_ah_get_sl(ah_attr); if (ib_pd->uobject && !rdma_is_multicast_addr((struct in6_addr *) - ah_attr->grh.dgid.raw) && + grh->dgid.raw) && !rdma_link_local_addr((struct in6_addr *) - ah_attr->grh.dgid.raw)) { + grh->dgid.raw)) { union ib_gid sgid; rc = ib_get_cached_gid(&rdev->ibdev, 1, - ah_attr->grh.sgid_index, &sgid, + grh->sgid_index, &sgid, &sgid_attr); if (rc) { dev_err(rdev_to_dev(rdev), "Failed to query gid at index %d", - ah_attr->grh.sgid_index); + grh->sgid_index); goto fail; } if (sgid_attr.ndev) { @@ -595,8 +596,8 @@ struct ib_ah *bnxt_re_create_ah(struct ib_pd *ib_pd, ah->qplib_ah.nw_type = CMDQ_CREATE_AH_TYPE_V1; break; } - rc = rdma_addr_find_l2_eth_by_grh(&sgid, &ah_attr->grh.dgid, - ah_attr->dmac, &vlan_tag, + rc = rdma_addr_find_l2_eth_by_grh(&sgid, &grh->dgid, + ah_attr->roce.dmac, &vlan_tag, &sgid_attr.ndev->ifindex, NULL); if (rc) { @@ -605,7 +606,7 @@ struct ib_ah *bnxt_re_create_ah(struct ib_pd *ib_pd, } } - memcpy(ah->qplib_ah.dmac, ah_attr->dmac, ETH_ALEN); + memcpy(ah->qplib_ah.dmac, ah_attr->roce.dmac, ETH_ALEN); rc = bnxt_qplib_create_ah(&rdev->qplib_res, &ah->qplib_ah); if (rc) { dev_err(rdev_to_dev(rdev), "Failed to allocate HW AH"); @@ -634,24 +635,24 @@ fail: return ERR_PTR(rc); } -int bnxt_re_modify_ah(struct ib_ah *ib_ah, struct ib_ah_attr *ah_attr) +int bnxt_re_modify_ah(struct ib_ah *ib_ah, struct rdma_ah_attr *ah_attr) { return 0; } -int bnxt_re_query_ah(struct ib_ah *ib_ah, struct ib_ah_attr *ah_attr) +int bnxt_re_query_ah(struct ib_ah *ib_ah, struct rdma_ah_attr *ah_attr) { struct bnxt_re_ah *ah = container_of(ib_ah, struct bnxt_re_ah, ib_ah); - memcpy(ah_attr->grh.dgid.raw, ah->qplib_ah.dgid.data, - sizeof(union ib_gid)); - ah_attr->grh.sgid_index = ah->qplib_ah.host_sgid_index; - ah_attr->grh.traffic_class = ah->qplib_ah.traffic_class; - ah_attr->sl = ah->qplib_ah.sl; - memcpy(ah_attr->dmac, ah->qplib_ah.dmac, ETH_ALEN); - ah_attr->ah_flags = IB_AH_GRH; - ah_attr->port_num = 1; - ah_attr->static_rate = 0; + ah_attr->type = ib_ah->type; + rdma_ah_set_sl(ah_attr, ah->qplib_ah.sl); + memcpy(ah_attr->roce.dmac, ah->qplib_ah.dmac, ETH_ALEN); + rdma_ah_set_grh(ah_attr, NULL, 0, + ah->qplib_ah.host_sgid_index, + 0, ah->qplib_ah.traffic_class); + rdma_ah_set_dgid_raw(ah_attr, ah->qplib_ah.dgid.data); + rdma_ah_set_port_num(ah_attr, 1); + rdma_ah_set_static_rate(ah_attr, 0); return 0; } @@ -692,9 +693,9 @@ int bnxt_re_destroy_qp(struct ib_qp *ib_qp) kfree(rdev->qp1_sqp); } - if (qp->rumem && !IS_ERR(qp->rumem)) + if (!IS_ERR_OR_NULL(qp->rumem)) ib_umem_release(qp->rumem); - if (qp->sumem && !IS_ERR(qp->sumem)) + if (!IS_ERR_OR_NULL(qp->sumem)) ib_umem_release(qp->sumem); mutex_lock(&rdev->qp_lock); @@ -1258,6 +1259,9 @@ int bnxt_re_modify_qp(struct ib_qp *ib_qp, struct ib_qp_attr *qp_attr, qp->qplib_qp.qkey = qp_attr->qkey; } if (qp_attr_mask & IB_QP_AV) { + const struct ib_global_route *grh = + rdma_ah_read_grh(&qp_attr->ah_attr); + qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_DGID | CMDQ_MODIFY_QP_MODIFY_MASK_FLOW_LABEL | CMDQ_MODIFY_QP_MODIFY_MASK_SGID_INDEX | @@ -1265,25 +1269,23 @@ int bnxt_re_modify_qp(struct ib_qp *ib_qp, struct ib_qp_attr *qp_attr, CMDQ_MODIFY_QP_MODIFY_MASK_TRAFFIC_CLASS | CMDQ_MODIFY_QP_MODIFY_MASK_DEST_MAC | CMDQ_MODIFY_QP_MODIFY_MASK_VLAN_ID; - memcpy(qp->qplib_qp.ah.dgid.data, qp_attr->ah_attr.grh.dgid.raw, + memcpy(qp->qplib_qp.ah.dgid.data, grh->dgid.raw, sizeof(qp->qplib_qp.ah.dgid.data)); - qp->qplib_qp.ah.flow_label = qp_attr->ah_attr.grh.flow_label; + qp->qplib_qp.ah.flow_label = grh->flow_label; /* If RoCE V2 is enabled, stack will have two entries for * each GID entry. Avoiding this duplicte entry in HW. Dividing * the GID index by 2 for RoCE V2 */ - qp->qplib_qp.ah.sgid_index = - qp_attr->ah_attr.grh.sgid_index / 2; - qp->qplib_qp.ah.host_sgid_index = - qp_attr->ah_attr.grh.sgid_index; - qp->qplib_qp.ah.hop_limit = qp_attr->ah_attr.grh.hop_limit; - qp->qplib_qp.ah.traffic_class = - qp_attr->ah_attr.grh.traffic_class; - qp->qplib_qp.ah.sl = qp_attr->ah_attr.sl; - ether_addr_copy(qp->qplib_qp.ah.dmac, qp_attr->ah_attr.dmac); + qp->qplib_qp.ah.sgid_index = grh->sgid_index / 2; + qp->qplib_qp.ah.host_sgid_index = grh->sgid_index; + qp->qplib_qp.ah.hop_limit = grh->hop_limit; + qp->qplib_qp.ah.traffic_class = grh->traffic_class; + qp->qplib_qp.ah.sl = rdma_ah_get_sl(&qp_attr->ah_attr); + ether_addr_copy(qp->qplib_qp.ah.dmac, + qp_attr->ah_attr.roce.dmac); status = ib_get_cached_gid(&rdev->ibdev, 1, - qp_attr->ah_attr.grh.sgid_index, + grh->sgid_index, &sgid, &sgid_attr); if (!status && sgid_attr.ndev) { memcpy(qp->qplib_qp.smac, sgid_attr.ndev->dev_addr, @@ -1423,14 +1425,14 @@ int bnxt_re_query_qp(struct ib_qp *ib_qp, struct ib_qp_attr *qp_attr, qp_attr->qp_access_flags = __to_ib_access_flags(qplib_qp.access); qp_attr->pkey_index = qplib_qp.pkey_index; qp_attr->qkey = qplib_qp.qkey; - memcpy(qp_attr->ah_attr.grh.dgid.raw, qplib_qp.ah.dgid.data, - sizeof(qplib_qp.ah.dgid.data)); - qp_attr->ah_attr.grh.flow_label = qplib_qp.ah.flow_label; - qp_attr->ah_attr.grh.sgid_index = qplib_qp.ah.host_sgid_index; - qp_attr->ah_attr.grh.hop_limit = qplib_qp.ah.hop_limit; - qp_attr->ah_attr.grh.traffic_class = qplib_qp.ah.traffic_class; - qp_attr->ah_attr.sl = qplib_qp.ah.sl; - ether_addr_copy(qp_attr->ah_attr.dmac, qplib_qp.ah.dmac); + qp_attr->ah_attr.type = RDMA_AH_ATTR_TYPE_ROCE; + rdma_ah_set_grh(&qp_attr->ah_attr, NULL, qplib_qp.ah.flow_label, + qplib_qp.ah.host_sgid_index, + qplib_qp.ah.hop_limit, + qplib_qp.ah.traffic_class); + rdma_ah_set_dgid_raw(&qp_attr->ah_attr, qplib_qp.ah.dgid.data); + rdma_ah_set_sl(&qp_attr->ah_attr, qplib_qp.ah.sl); + ether_addr_copy(qp_attr->ah_attr.roce.dmac, qplib_qp.ah.dmac); qp_attr->path_mtu = __to_ib_mtu(qplib_qp.path_mtu); qp_attr->timeout = qplib_qp.timeout; qp_attr->retry_cnt = qplib_qp.retry_cnt; @@ -2116,7 +2118,7 @@ int bnxt_re_destroy_cq(struct ib_cq *ib_cq) dev_err(rdev_to_dev(rdev), "Failed to destroy HW CQ"); return rc; } - if (cq->umem && !IS_ERR(cq->umem)) + if (!IS_ERR_OR_NULL(cq->umem)) ib_umem_release(cq->umem); if (cq) { @@ -2818,7 +2820,7 @@ int bnxt_re_dereg_mr(struct ib_mr *ib_mr) { struct bnxt_re_mr *mr = container_of(ib_mr, struct bnxt_re_mr, ib_mr); struct bnxt_re_dev *rdev = mr->rdev; - int rc = 0; + int rc; if (mr->npages && mr->pages) { rc = bnxt_qplib_free_fast_reg_page_list(&rdev->qplib_res, @@ -2829,7 +2831,7 @@ int bnxt_re_dereg_mr(struct ib_mr *ib_mr) } rc = bnxt_qplib_free_mrw(&rdev->qplib_res, &mr->qplib_mr); - if (!IS_ERR(mr->ib_umem) && mr->ib_umem) + if (!IS_ERR_OR_NULL(mr->ib_umem)) ib_umem_release(mr->ib_umem); kfree(mr); @@ -3016,7 +3018,7 @@ struct ib_mr *bnxt_re_reg_user_mr(struct ib_pd *ib_pd, u64 start, u64 length, struct bnxt_re_mr *mr; struct ib_umem *umem; u64 *pbl_tbl, *pbl_tbl_orig; - int i, umem_pgs, pages, page_shift, rc; + int i, umem_pgs, pages, rc; struct scatterlist *sg; int entry; @@ -3062,22 +3064,22 @@ struct ib_mr *bnxt_re_reg_user_mr(struct ib_pd *ib_pd, u64 start, u64 length, } pbl_tbl_orig = pbl_tbl; - page_shift = ilog2(umem->page_size); if (umem->hugetlb) { dev_err(rdev_to_dev(rdev), "umem hugetlb not supported!"); rc = -EFAULT; goto fail; } - if (umem->page_size != PAGE_SIZE) { - dev_err(rdev_to_dev(rdev), "umem page size unsupported!"); + + if (umem->page_shift != PAGE_SHIFT) { + dev_err(rdev_to_dev(rdev), "umem page shift unsupported!"); rc = -EFAULT; goto fail; } /* Map umem buf ptrs to the PBL */ for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) { - pages = sg_dma_len(sg) >> page_shift; + pages = sg_dma_len(sg) >> umem->page_shift; for (i = 0; i < pages; i++, pbl_tbl++) - *pbl_tbl = sg_dma_address(sg) + (i << page_shift); + *pbl_tbl = sg_dma_address(sg) + (i << umem->page_shift); } rc = bnxt_qplib_reg_mr(&rdev->qplib_res, &mr->qplib_mr, pbl_tbl_orig, umem_pgs, false); diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.h b/drivers/infiniband/hw/bnxt_re/ib_verbs.h index b4084c252f06..5c3d71765454 100644 --- a/drivers/infiniband/hw/bnxt_re/ib_verbs.h +++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.h @@ -150,10 +150,10 @@ struct ib_pd *bnxt_re_alloc_pd(struct ib_device *ibdev, struct ib_udata *udata); int bnxt_re_dealloc_pd(struct ib_pd *pd); struct ib_ah *bnxt_re_create_ah(struct ib_pd *pd, - struct ib_ah_attr *ah_attr, + struct rdma_ah_attr *ah_attr, struct ib_udata *udata); -int bnxt_re_modify_ah(struct ib_ah *ah, struct ib_ah_attr *ah_attr); -int bnxt_re_query_ah(struct ib_ah *ah, struct ib_ah_attr *ah_attr); +int bnxt_re_modify_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr); +int bnxt_re_query_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr); int bnxt_re_destroy_ah(struct ib_ah *ah); struct ib_qp *bnxt_re_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *qp_init_attr, diff --git a/drivers/infiniband/hw/cxgb3/cxio_dbg.c b/drivers/infiniband/hw/cxgb3/cxio_dbg.c index 445e89e5e7cf..97dbe728520a 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_dbg.c +++ b/drivers/infiniband/hw/cxgb3/cxio_dbg.c @@ -51,17 +51,18 @@ void cxio_dump_tpt(struct cxio_rdev *rdev, u32 stag) m->mem_id = MEM_PMRX; m->addr = (stag>>8) * 32 + rdev->rnic_info.tpt_base; m->len = size; - PDBG("%s TPT addr 0x%x len %d\n", __func__, m->addr, m->len); + pr_debug("%s TPT addr 0x%x len %d\n", __func__, m->addr, m->len); rc = rdev->t3cdev_p->ctl(rdev->t3cdev_p, RDMA_GET_MEM, m); if (rc) { - PDBG("%s toectl returned error %d\n", __func__, rc); + pr_debug("%s toectl returned error %d\n", __func__, rc); kfree(m); return; } data = (u64 *)m->buf; while (size > 0) { - PDBG("TPT %08x: %016llx\n", m->addr, (unsigned long long) *data); + pr_debug("TPT %08x: %016llx\n", + m->addr, (unsigned long long)*data); size -= 8; data++; m->addr += 8; @@ -87,18 +88,19 @@ void cxio_dump_pbl(struct cxio_rdev *rdev, u32 pbl_addr, uint len, u8 shift) m->mem_id = MEM_PMRX; m->addr = pbl_addr; m->len = size; - PDBG("%s PBL addr 0x%x len %d depth %d\n", - __func__, m->addr, m->len, npages); + pr_debug("%s PBL addr 0x%x len %d depth %d\n", + __func__, m->addr, m->len, npages); rc = rdev->t3cdev_p->ctl(rdev->t3cdev_p, RDMA_GET_MEM, m); if (rc) { - PDBG("%s toectl returned error %d\n", __func__, rc); + pr_debug("%s toectl returned error %d\n", __func__, rc); kfree(m); return; } data = (u64 *)m->buf; while (size > 0) { - PDBG("PBL %08x: %016llx\n", m->addr, (unsigned long long) *data); + pr_debug("PBL %08x: %016llx\n", + m->addr, (unsigned long long)*data); size -= 8; data++; m->addr += 8; @@ -114,8 +116,8 @@ void cxio_dump_wqe(union t3_wr *wqe) if (size == 0) size = 8; while (size > 0) { - PDBG("WQE %p: %016llx\n", data, - (unsigned long long) be64_to_cpu(*data)); + pr_debug("WQE %p: %016llx\n", + data, (unsigned long long)be64_to_cpu(*data)); size--; data++; } @@ -127,8 +129,8 @@ void cxio_dump_wce(struct t3_cqe *wce) int size = sizeof(*wce); while (size > 0) { - PDBG("WCE %p: %016llx\n", data, - (unsigned long long) be64_to_cpu(*data)); + pr_debug("WCE %p: %016llx\n", + data, (unsigned long long)be64_to_cpu(*data)); size -= 8; data++; } @@ -148,17 +150,18 @@ void cxio_dump_rqt(struct cxio_rdev *rdev, u32 hwtid, int nents) m->mem_id = MEM_PMRX; m->addr = ((hwtid)<<10) + rdev->rnic_info.rqt_base; m->len = size; - PDBG("%s RQT addr 0x%x len %d\n", __func__, m->addr, m->len); + pr_debug("%s RQT addr 0x%x len %d\n", __func__, m->addr, m->len); rc = rdev->t3cdev_p->ctl(rdev->t3cdev_p, RDMA_GET_MEM, m); if (rc) { - PDBG("%s toectl returned error %d\n", __func__, rc); + pr_debug("%s toectl returned error %d\n", __func__, rc); kfree(m); return; } data = (u64 *)m->buf; while (size > 0) { - PDBG("RQT %08x: %016llx\n", m->addr, (unsigned long long) *data); + pr_debug("RQT %08x: %016llx\n", + m->addr, (unsigned long long)*data); size -= 8; data++; m->addr += 8; @@ -180,10 +183,10 @@ void cxio_dump_tcb(struct cxio_rdev *rdev, u32 hwtid) m->mem_id = MEM_CM; m->addr = hwtid * size; m->len = size; - PDBG("%s TCB %d len %d\n", __func__, m->addr, m->len); + pr_debug("%s TCB %d len %d\n", __func__, m->addr, m->len); rc = rdev->t3cdev_p->ctl(rdev->t3cdev_p, RDMA_GET_MEM, m); if (rc) { - PDBG("%s toectl returned error %d\n", __func__, rc); + pr_debug("%s toectl returned error %d\n", __func__, rc); kfree(m); return; } diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.c b/drivers/infiniband/hw/cxgb3/cxio_hal.c index ada2e5009c86..558d6a03375d 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_hal.c +++ b/drivers/infiniband/hw/cxgb3/cxio_hal.c @@ -110,8 +110,7 @@ int cxio_hal_cq_op(struct cxio_rdev *rdev_p, struct t3_cq *cq, while (!CQ_VLD_ENTRY(rptr, cq->size_log2, cqe)) { udelay(1); if (i++ > 1000000) { - printk(KERN_ERR "%s: stalled rnic\n", - rdev_p->dev_name); + pr_err("%s: stalled rnic\n", rdev_p->dev_name); return -EIO; } } @@ -140,7 +139,7 @@ static int cxio_hal_clear_qp_ctx(struct cxio_rdev *rdev_p, u32 qpid) struct t3_modify_qp_wr *wqe; struct sk_buff *skb = alloc_skb(sizeof(*wqe), GFP_KERNEL); if (!skb) { - PDBG("%s alloc_skb failed\n", __func__); + pr_debug("%s alloc_skb failed\n", __func__); return -ENOMEM; } wqe = (struct t3_modify_qp_wr *) skb_put(skb, sizeof(*wqe)); @@ -230,7 +229,7 @@ static u32 get_qpid(struct cxio_rdev *rdev_p, struct cxio_ucontext *uctx) } out: mutex_unlock(&uctx->lock); - PDBG("%s qpid 0x%x\n", __func__, qpid); + pr_debug("%s qpid 0x%x\n", __func__, qpid); return qpid; } @@ -242,7 +241,7 @@ static void put_qpid(struct cxio_rdev *rdev_p, u32 qpid, entry = kmalloc(sizeof *entry, GFP_KERNEL); if (!entry) return; - PDBG("%s qpid 0x%x\n", __func__, qpid); + pr_debug("%s qpid 0x%x\n", __func__, qpid); entry->qpid = qpid; mutex_lock(&uctx->lock); list_add_tail(&entry->entry, &uctx->qpids); @@ -306,8 +305,8 @@ int cxio_create_qp(struct cxio_rdev *rdev_p, u32 kernel_domain, wq->udb = (u64)rdev_p->rnic_info.udbell_physbase + (wq->qpid << rdev_p->qpshift); wq->rdev = rdev_p; - PDBG("%s qpid 0x%x doorbell 0x%p udb 0x%llx\n", __func__, - wq->qpid, wq->doorbell, (unsigned long long) wq->udb); + pr_debug("%s qpid 0x%x doorbell 0x%p udb 0x%llx\n", + __func__, wq->qpid, wq->doorbell, (unsigned long long)wq->udb); return 0; err4: kfree(wq->sq); @@ -351,8 +350,8 @@ static void insert_recv_cqe(struct t3_wq *wq, struct t3_cq *cq) { struct t3_cqe cqe; - PDBG("%s wq %p cq %p sw_rptr 0x%x sw_wptr 0x%x\n", __func__, - wq, cq, cq->sw_rptr, cq->sw_wptr); + pr_debug("%s wq %p cq %p sw_rptr 0x%x sw_wptr 0x%x\n", __func__, + wq, cq, cq->sw_rptr, cq->sw_wptr); memset(&cqe, 0, sizeof(cqe)); cqe.header = cpu_to_be32(V_CQE_STATUS(TPT_ERR_SWFLUSH) | V_CQE_OPCODE(T3_SEND) | @@ -370,11 +369,11 @@ int cxio_flush_rq(struct t3_wq *wq, struct t3_cq *cq, int count) u32 ptr; int flushed = 0; - PDBG("%s wq %p cq %p\n", __func__, wq, cq); + pr_debug("%s wq %p cq %p\n", __func__, wq, cq); /* flush RQ */ - PDBG("%s rq_rptr %u rq_wptr %u skip count %u\n", __func__, - wq->rq_rptr, wq->rq_wptr, count); + pr_debug("%s rq_rptr %u rq_wptr %u skip count %u\n", __func__, + wq->rq_rptr, wq->rq_wptr, count); ptr = wq->rq_rptr + count; while (ptr++ != wq->rq_wptr) { insert_recv_cqe(wq, cq); @@ -388,8 +387,8 @@ static void insert_sq_cqe(struct t3_wq *wq, struct t3_cq *cq, { struct t3_cqe cqe; - PDBG("%s wq %p cq %p sw_rptr 0x%x sw_wptr 0x%x\n", __func__, - wq, cq, cq->sw_rptr, cq->sw_wptr); + pr_debug("%s wq %p cq %p sw_rptr 0x%x sw_wptr 0x%x\n", __func__, + wq, cq, cq->sw_rptr, cq->sw_wptr); memset(&cqe, 0, sizeof(cqe)); cqe.header = cpu_to_be32(V_CQE_STATUS(TPT_ERR_SWFLUSH) | V_CQE_OPCODE(sqp->opcode) | @@ -429,11 +428,11 @@ void cxio_flush_hw_cq(struct t3_cq *cq) { struct t3_cqe *cqe, *swcqe; - PDBG("%s cq %p cqid 0x%x\n", __func__, cq, cq->cqid); + pr_debug("%s cq %p cqid 0x%x\n", __func__, cq, cq->cqid); cqe = cxio_next_hw_cqe(cq); while (cqe) { - PDBG("%s flushing hwcq rptr 0x%x to swcq wptr 0x%x\n", - __func__, cq->rptr, cq->sw_wptr); + pr_debug("%s flushing hwcq rptr 0x%x to swcq wptr 0x%x\n", + __func__, cq->rptr, cq->sw_wptr); swcqe = cq->sw_queue + Q_PTR2IDX(cq->sw_wptr, cq->size_log2); *swcqe = *cqe; swcqe->header |= cpu_to_be32(V_CQE_SWCQE(1)); @@ -476,7 +475,7 @@ void cxio_count_scqes(struct t3_cq *cq, struct t3_wq *wq, int *count) (*count)++; ptr++; } - PDBG("%s cq %p count %d\n", __func__, cq, *count); + pr_debug("%s cq %p count %d\n", __func__, cq, *count); } void cxio_count_rcqes(struct t3_cq *cq, struct t3_wq *wq, int *count) @@ -485,7 +484,7 @@ void cxio_count_rcqes(struct t3_cq *cq, struct t3_wq *wq, int *count) u32 ptr; *count = 0; - PDBG("%s count zero %d\n", __func__, *count); + pr_debug("%s count zero %d\n", __func__, *count); ptr = cq->sw_rptr; while (!Q_EMPTY(ptr, cq->sw_wptr)) { cqe = cq->sw_queue + (Q_PTR2IDX(ptr, cq->size_log2)); @@ -494,7 +493,7 @@ void cxio_count_rcqes(struct t3_cq *cq, struct t3_wq *wq, int *count) (*count)++; ptr++; } - PDBG("%s cq %p count %d\n", __func__, cq, *count); + pr_debug("%s cq %p count %d\n", __func__, cq, *count); } static int cxio_hal_init_ctrl_cq(struct cxio_rdev *rdev_p) @@ -521,12 +520,12 @@ static int cxio_hal_init_ctrl_qp(struct cxio_rdev *rdev_p) skb = alloc_skb(sizeof(*wqe), GFP_KERNEL); if (!skb) { - PDBG("%s alloc_skb failed\n", __func__); + pr_debug("%s alloc_skb failed\n", __func__); return -ENOMEM; } err = cxio_hal_init_ctrl_cq(rdev_p); if (err) { - PDBG("%s err %d initializing ctrl_cq\n", __func__, err); + pr_debug("%s err %d initializing ctrl_cq\n", __func__, err); goto err; } rdev_p->ctrl_qp.workq = dma_alloc_coherent( @@ -536,7 +535,7 @@ static int cxio_hal_init_ctrl_qp(struct cxio_rdev *rdev_p) &(rdev_p->ctrl_qp.dma_addr), GFP_KERNEL); if (!rdev_p->ctrl_qp.workq) { - PDBG("%s dma_alloc_coherent failed\n", __func__); + pr_debug("%s dma_alloc_coherent failed\n", __func__); err = -ENOMEM; goto err; } @@ -571,9 +570,9 @@ static int cxio_hal_init_ctrl_qp(struct cxio_rdev *rdev_p) wqe->sge_cmd = cpu_to_be64(sge_cmd); wqe->ctx1 = cpu_to_be64(ctx1); wqe->ctx0 = cpu_to_be64(ctx0); - PDBG("CtrlQP dma_addr 0x%llx workq %p size %d\n", - (unsigned long long) rdev_p->ctrl_qp.dma_addr, - rdev_p->ctrl_qp.workq, 1 << T3_CTRL_QP_SIZE_LOG2); + pr_debug("CtrlQP dma_addr 0x%llx workq %p size %d\n", + (unsigned long long)rdev_p->ctrl_qp.dma_addr, + rdev_p->ctrl_qp.workq, 1 << T3_CTRL_QP_SIZE_LOG2); skb->priority = CPL_PRIORITY_CONTROL; return iwch_cxgb3_ofld_send(rdev_p->t3cdev_p, skb); err: @@ -605,26 +604,26 @@ static int cxio_hal_ctrl_qp_write_mem(struct cxio_rdev *rdev_p, u32 addr, u64 utx_cmd; addr &= 0x7FFFFFF; nr_wqe = len % 96 ? len / 96 + 1 : len / 96; /* 96B max per WQE */ - PDBG("%s wptr 0x%x rptr 0x%x len %d, nr_wqe %d data %p addr 0x%0x\n", - __func__, rdev_p->ctrl_qp.wptr, rdev_p->ctrl_qp.rptr, len, - nr_wqe, data, addr); + pr_debug("%s wptr 0x%x rptr 0x%x len %d, nr_wqe %d data %p addr 0x%0x\n", + __func__, rdev_p->ctrl_qp.wptr, rdev_p->ctrl_qp.rptr, len, + nr_wqe, data, addr); utx_len = 3; /* in 32B unit */ for (i = 0; i < nr_wqe; i++) { if (Q_FULL(rdev_p->ctrl_qp.rptr, rdev_p->ctrl_qp.wptr, T3_CTRL_QP_SIZE_LOG2)) { - PDBG("%s ctrl_qp full wtpr 0x%0x rptr 0x%0x, " - "wait for more space i %d\n", __func__, - rdev_p->ctrl_qp.wptr, rdev_p->ctrl_qp.rptr, i); + pr_debug("%s ctrl_qp full wtpr 0x%0x rptr 0x%0x, wait for more space i %d\n", + __func__, + rdev_p->ctrl_qp.wptr, rdev_p->ctrl_qp.rptr, i); if (wait_event_interruptible(rdev_p->ctrl_qp.waitq, !Q_FULL(rdev_p->ctrl_qp.rptr, rdev_p->ctrl_qp.wptr, T3_CTRL_QP_SIZE_LOG2))) { - PDBG("%s ctrl_qp workq interrupted\n", - __func__); + pr_debug("%s ctrl_qp workq interrupted\n", + __func__); return -ERESTARTSYS; } - PDBG("%s ctrl_qp wakeup, continue posting work request " - "i %d\n", __func__, i); + pr_debug("%s ctrl_qp wakeup, continue posting work request i %d\n", + __func__, i); } wqe = (__be64 *)(rdev_p->ctrl_qp.workq + (rdev_p->ctrl_qp.wptr % (1 << T3_CTRL_QP_SIZE_LOG2))); @@ -645,7 +644,7 @@ static int cxio_hal_ctrl_qp_write_mem(struct cxio_rdev *rdev_p, u32 addr, if ((i != 0) && (i % (((1 << T3_CTRL_QP_SIZE_LOG2)) >> 1) == 0)) { flag = T3_COMPLETION_FLAG; - PDBG("%s force completion at i %d\n", __func__, i); + pr_debug("%s force completion at i %d\n", __func__, i); } /* build the utx mem command */ @@ -717,8 +716,8 @@ static int __cxio_tpt_op(struct cxio_rdev *rdev_p, u32 reset_tpt_entry, return -ENOMEM; *stag = (stag_idx << 8) | ((*stag) & 0xFF); } - PDBG("%s stag_state 0x%0x type 0x%0x pdid 0x%0x, stag_idx 0x%x\n", - __func__, stag_state, type, pdid, stag_idx); + pr_debug("%s stag_state 0x%0x type 0x%0x pdid 0x%0x, stag_idx 0x%x\n", + __func__, stag_state, type, pdid, stag_idx); mutex_lock(&rdev_p->ctrl_qp.lock); @@ -767,9 +766,9 @@ int cxio_write_pbl(struct cxio_rdev *rdev_p, __be64 *pbl, u32 wptr; int err; - PDBG("%s *pdb_addr 0x%x, pbl_base 0x%x, pbl_size %d\n", - __func__, pbl_addr, rdev_p->rnic_info.pbl_base, - pbl_size); + pr_debug("%s *pdb_addr 0x%x, pbl_base 0x%x, pbl_size %d\n", + __func__, pbl_addr, rdev_p->rnic_info.pbl_base, + pbl_size); mutex_lock(&rdev_p->ctrl_qp.lock); err = cxio_hal_ctrl_qp_write_mem(rdev_p, pbl_addr >> 5, pbl_size << 3, @@ -837,7 +836,7 @@ int cxio_rdma_init(struct cxio_rdev *rdev_p, struct t3_rdma_init_attr *attr) struct sk_buff *skb = alloc_skb(sizeof(*wqe), GFP_ATOMIC); if (!skb) return -ENOMEM; - PDBG("%s rdev_p %p\n", __func__, rdev_p); + pr_debug("%s rdev_p %p\n", __func__, rdev_p); wqe = (struct t3_rdma_init_wr *) __skb_put(skb, sizeof(*wqe)); wqe->wrh.op_seop_flags = cpu_to_be32(V_FW_RIWR_OP(T3_WR_INIT)); wqe->wrh.gen_tid_len = cpu_to_be32(V_FW_RIWR_TID(attr->tid) | @@ -880,22 +879,20 @@ static int cxio_hal_ev_handler(struct t3cdev *t3cdev_p, struct sk_buff *skb) static int cnt; struct cxio_rdev *rdev_p = NULL; struct respQ_msg_t *rsp_msg = (struct respQ_msg_t *) skb->data; - PDBG("%d: %s cq_id 0x%x cq_ptr 0x%x genbit %0x overflow %0x an %0x" - " se %0x notify %0x cqbranch %0x creditth %0x\n", - cnt, __func__, RSPQ_CQID(rsp_msg), RSPQ_CQPTR(rsp_msg), - RSPQ_GENBIT(rsp_msg), RSPQ_OVERFLOW(rsp_msg), RSPQ_AN(rsp_msg), - RSPQ_SE(rsp_msg), RSPQ_NOTIFY(rsp_msg), RSPQ_CQBRANCH(rsp_msg), - RSPQ_CREDIT_THRESH(rsp_msg)); - PDBG("CQE: QPID 0x%0x genbit %0x type 0x%0x status 0x%0x opcode %d " - "len 0x%0x wrid_hi_stag 0x%x wrid_low_msn 0x%x\n", - CQE_QPID(rsp_msg->cqe), CQE_GENBIT(rsp_msg->cqe), - CQE_TYPE(rsp_msg->cqe), CQE_STATUS(rsp_msg->cqe), - CQE_OPCODE(rsp_msg->cqe), CQE_LEN(rsp_msg->cqe), - CQE_WRID_HI(rsp_msg->cqe), CQE_WRID_LOW(rsp_msg->cqe)); + pr_debug("%d: %s cq_id 0x%x cq_ptr 0x%x genbit %0x overflow %0x an %0x se %0x notify %0x cqbranch %0x creditth %0x\n", + cnt, __func__, RSPQ_CQID(rsp_msg), RSPQ_CQPTR(rsp_msg), + RSPQ_GENBIT(rsp_msg), RSPQ_OVERFLOW(rsp_msg), RSPQ_AN(rsp_msg), + RSPQ_SE(rsp_msg), RSPQ_NOTIFY(rsp_msg), RSPQ_CQBRANCH(rsp_msg), + RSPQ_CREDIT_THRESH(rsp_msg)); + pr_debug("CQE: QPID 0x%0x genbit %0x type 0x%0x status 0x%0x opcode %d len 0x%0x wrid_hi_stag 0x%x wrid_low_msn 0x%x\n", + CQE_QPID(rsp_msg->cqe), CQE_GENBIT(rsp_msg->cqe), + CQE_TYPE(rsp_msg->cqe), CQE_STATUS(rsp_msg->cqe), + CQE_OPCODE(rsp_msg->cqe), CQE_LEN(rsp_msg->cqe), + CQE_WRID_HI(rsp_msg->cqe), CQE_WRID_LOW(rsp_msg->cqe)); rdev_p = (struct cxio_rdev *)t3cdev_p->ulp; if (!rdev_p) { - PDBG("%s called by t3cdev %p with null ulp\n", __func__, - t3cdev_p); + pr_debug("%s called by t3cdev %p with null ulp\n", __func__, + t3cdev_p); return 0; } if (CQE_QPID(rsp_msg->cqe) == T3_CTRL_QP_ID) { @@ -934,13 +931,13 @@ int cxio_rdev_open(struct cxio_rdev *rdev_p) strncpy(rdev_p->dev_name, rdev_p->t3cdev_p->name, T3_MAX_DEV_NAME_LEN); } else { - PDBG("%s t3cdev_p or dev_name must be set\n", __func__); + pr_debug("%s t3cdev_p or dev_name must be set\n", __func__); return -EINVAL; } list_add_tail(&rdev_p->entry, &rdev_list); - PDBG("%s opening rnic dev %s\n", __func__, rdev_p->dev_name); + pr_debug("%s opening rnic dev %s\n", __func__, rdev_p->dev_name); memset(&rdev_p->ctrl_qp, 0, sizeof(rdev_p->ctrl_qp)); if (!rdev_p->t3cdev_p) rdev_p->t3cdev_p = dev2t3cdev(netdev_p); @@ -949,13 +946,12 @@ int cxio_rdev_open(struct cxio_rdev *rdev_p) err = rdev_p->t3cdev_p->ctl(rdev_p->t3cdev_p, GET_EMBEDDED_INFO, &(rdev_p->fw_info)); if (err) { - printk(KERN_ERR "%s t3cdev_p(%p)->ctl returned error %d.\n", - __func__, rdev_p->t3cdev_p, err); + pr_err("%s t3cdev_p(%p)->ctl returned error %d\n", + __func__, rdev_p->t3cdev_p, err); goto err1; } if (G_FW_VERSION_MAJOR(rdev_p->fw_info.fw_vers) != CXIO_FW_MAJ) { - printk(KERN_ERR MOD "fatal firmware version mismatch: " - "need version %u but adapter has version %u\n", + pr_err("fatal firmware version mismatch: need version %u but adapter has version %u\n", CXIO_FW_MAJ, G_FW_VERSION_MAJOR(rdev_p->fw_info.fw_vers)); err = -EINVAL; @@ -965,15 +961,15 @@ int cxio_rdev_open(struct cxio_rdev *rdev_p) err = rdev_p->t3cdev_p->ctl(rdev_p->t3cdev_p, RDMA_GET_PARAMS, &(rdev_p->rnic_info)); if (err) { - printk(KERN_ERR "%s t3cdev_p(%p)->ctl returned error %d.\n", - __func__, rdev_p->t3cdev_p, err); + pr_err("%s t3cdev_p(%p)->ctl returned error %d\n", + __func__, rdev_p->t3cdev_p, err); goto err1; } err = rdev_p->t3cdev_p->ctl(rdev_p->t3cdev_p, GET_PORTS, &(rdev_p->port_info)); if (err) { - printk(KERN_ERR "%s t3cdev_p(%p)->ctl returned error %d.\n", - __func__, rdev_p->t3cdev_p, err); + pr_err("%s t3cdev_p(%p)->ctl returned error %d\n", + __func__, rdev_p->t3cdev_p, err); goto err1; } @@ -988,42 +984,39 @@ int cxio_rdev_open(struct cxio_rdev *rdev_p) PAGE_SHIFT)); rdev_p->qpnr = rdev_p->rnic_info.udbell_len >> PAGE_SHIFT; rdev_p->qpmask = (65536 >> ilog2(rdev_p->qpnr)) - 1; - PDBG("%s rnic %s info: tpt_base 0x%0x tpt_top 0x%0x num stags %d " - "pbl_base 0x%0x pbl_top 0x%0x rqt_base 0x%0x, rqt_top 0x%0x\n", - __func__, rdev_p->dev_name, rdev_p->rnic_info.tpt_base, - rdev_p->rnic_info.tpt_top, cxio_num_stags(rdev_p), - rdev_p->rnic_info.pbl_base, - rdev_p->rnic_info.pbl_top, rdev_p->rnic_info.rqt_base, - rdev_p->rnic_info.rqt_top); - PDBG("udbell_len 0x%0x udbell_physbase 0x%lx kdb_addr %p qpshift %lu " - "qpnr %d qpmask 0x%x\n", - rdev_p->rnic_info.udbell_len, - rdev_p->rnic_info.udbell_physbase, rdev_p->rnic_info.kdb_addr, - rdev_p->qpshift, rdev_p->qpnr, rdev_p->qpmask); + pr_debug("%s rnic %s info: tpt_base 0x%0x tpt_top 0x%0x num stags %d pbl_base 0x%0x pbl_top 0x%0x rqt_base 0x%0x, rqt_top 0x%0x\n", + __func__, rdev_p->dev_name, rdev_p->rnic_info.tpt_base, + rdev_p->rnic_info.tpt_top, cxio_num_stags(rdev_p), + rdev_p->rnic_info.pbl_base, + rdev_p->rnic_info.pbl_top, rdev_p->rnic_info.rqt_base, + rdev_p->rnic_info.rqt_top); + pr_debug("udbell_len 0x%0x udbell_physbase 0x%lx kdb_addr %p qpshift %lu qpnr %d qpmask 0x%x\n", + rdev_p->rnic_info.udbell_len, + rdev_p->rnic_info.udbell_physbase, rdev_p->rnic_info.kdb_addr, + rdev_p->qpshift, rdev_p->qpnr, rdev_p->qpmask); err = cxio_hal_init_ctrl_qp(rdev_p); if (err) { - printk(KERN_ERR "%s error %d initializing ctrl_qp.\n", - __func__, err); + pr_err("%s error %d initializing ctrl_qp\n", __func__, err); goto err1; } err = cxio_hal_init_resource(rdev_p, cxio_num_stags(rdev_p), 0, 0, T3_MAX_NUM_QP, T3_MAX_NUM_CQ, T3_MAX_NUM_PD); if (err) { - printk(KERN_ERR "%s error %d initializing hal resources.\n", + pr_err("%s error %d initializing hal resources\n", __func__, err); goto err2; } err = cxio_hal_pblpool_create(rdev_p); if (err) { - printk(KERN_ERR "%s error %d initializing pbl mem pool.\n", + pr_err("%s error %d initializing pbl mem pool\n", __func__, err); goto err3; } err = cxio_hal_rqtpool_create(rdev_p); if (err) { - printk(KERN_ERR "%s error %d initializing rqt mem pool.\n", + pr_err("%s error %d initializing rqt mem pool\n", __func__, err); goto err4; } @@ -1086,9 +1079,9 @@ static void flush_completed_wrs(struct t3_wq *wq, struct t3_cq *cq) /* * Insert this completed cqe into the swcq. */ - PDBG("%s moving cqe into swcq sq idx %ld cq idx %ld\n", - __func__, Q_PTR2IDX(ptr, wq->sq_size_log2), - Q_PTR2IDX(cq->sw_wptr, cq->size_log2)); + pr_debug("%s moving cqe into swcq sq idx %ld cq idx %ld\n", + __func__, Q_PTR2IDX(ptr, wq->sq_size_log2), + Q_PTR2IDX(cq->sw_wptr, cq->size_log2)); sqp->cqe.header |= htonl(V_CQE_SWCQE(1)); *(cq->sw_queue + Q_PTR2IDX(cq->sw_wptr, cq->size_log2)) = sqp->cqe; @@ -1154,12 +1147,11 @@ int cxio_poll_cq(struct t3_wq *wq, struct t3_cq *cq, struct t3_cqe *cqe, *credit = 0; hw_cqe = cxio_next_cqe(cq); - PDBG("%s CQE OOO %d qpid 0x%0x genbit %d type %d status 0x%0x" - " opcode 0x%0x len 0x%0x wrid_hi_stag 0x%x wrid_low_msn 0x%x\n", - __func__, CQE_OOO(*hw_cqe), CQE_QPID(*hw_cqe), - CQE_GENBIT(*hw_cqe), CQE_TYPE(*hw_cqe), CQE_STATUS(*hw_cqe), - CQE_OPCODE(*hw_cqe), CQE_LEN(*hw_cqe), CQE_WRID_HI(*hw_cqe), - CQE_WRID_LOW(*hw_cqe)); + pr_debug("%s CQE OOO %d qpid 0x%0x genbit %d type %d status 0x%0x opcode 0x%0x len 0x%0x wrid_hi_stag 0x%x wrid_low_msn 0x%x\n", + __func__, CQE_OOO(*hw_cqe), CQE_QPID(*hw_cqe), + CQE_GENBIT(*hw_cqe), CQE_TYPE(*hw_cqe), CQE_STATUS(*hw_cqe), + CQE_OPCODE(*hw_cqe), CQE_LEN(*hw_cqe), CQE_WRID_HI(*hw_cqe), + CQE_WRID_LOW(*hw_cqe)); /* * skip cqe's not affiliated with a QP. @@ -1278,9 +1270,10 @@ int cxio_poll_cq(struct t3_wq *wq, struct t3_cq *cq, struct t3_cqe *cqe, if (!SW_CQE(*hw_cqe) && (CQE_WRID_SQ_WPTR(*hw_cqe) != wq->sq_rptr)) { struct t3_swsq *sqp; - PDBG("%s out of order completion going in swsq at idx %ld\n", - __func__, - Q_PTR2IDX(CQE_WRID_SQ_WPTR(*hw_cqe), wq->sq_size_log2)); + pr_debug("%s out of order completion going in swsq at idx %ld\n", + __func__, + Q_PTR2IDX(CQE_WRID_SQ_WPTR(*hw_cqe), + wq->sq_size_log2)); sqp = wq->sq + Q_PTR2IDX(CQE_WRID_SQ_WPTR(*hw_cqe), wq->sq_size_log2); sqp->cqe = *hw_cqe; @@ -1298,13 +1291,13 @@ proc_cqe: */ if (SQ_TYPE(*hw_cqe)) { wq->sq_rptr = CQE_WRID_SQ_WPTR(*hw_cqe); - PDBG("%s completing sq idx %ld\n", __func__, - Q_PTR2IDX(wq->sq_rptr, wq->sq_size_log2)); + pr_debug("%s completing sq idx %ld\n", __func__, + Q_PTR2IDX(wq->sq_rptr, wq->sq_size_log2)); *cookie = wq->sq[Q_PTR2IDX(wq->sq_rptr, wq->sq_size_log2)].wr_id; wq->sq_rptr++; } else { - PDBG("%s completing rq idx %ld\n", __func__, - Q_PTR2IDX(wq->rq_rptr, wq->rq_size_log2)); + pr_debug("%s completing rq idx %ld\n", __func__, + Q_PTR2IDX(wq->rq_rptr, wq->rq_size_log2)); *cookie = wq->rq[Q_PTR2IDX(wq->rq_rptr, wq->rq_size_log2)].wr_id; if (wq->rq[Q_PTR2IDX(wq->rq_rptr, wq->rq_size_log2)].pbl_addr) cxio_hal_pblpool_free(wq->rdev, @@ -1322,12 +1315,12 @@ flush_wq: skip_cqe: if (SW_CQE(*hw_cqe)) { - PDBG("%s cq %p cqid 0x%x skip sw cqe sw_rptr 0x%x\n", - __func__, cq, cq->cqid, cq->sw_rptr); + pr_debug("%s cq %p cqid 0x%x skip sw cqe sw_rptr 0x%x\n", + __func__, cq, cq->cqid, cq->sw_rptr); ++cq->sw_rptr; } else { - PDBG("%s cq %p cqid 0x%x skip hw cqe rptr 0x%x\n", - __func__, cq, cq->cqid, cq->rptr); + pr_debug("%s cq %p cqid 0x%x skip hw cqe rptr 0x%x\n", + __func__, cq, cq->cqid, cq->rptr); ++cq->rptr; /* diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.h b/drivers/infiniband/hw/cxgb3/cxio_hal.h index 78fbe9ffe7f0..7e70c5492262 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_hal.h +++ b/drivers/infiniband/hw/cxgb3/cxio_hal.h @@ -196,8 +196,11 @@ int cxio_poll_cq(struct t3_wq *wq, struct t3_cq *cq, struct t3_cqe *cqe, u8 *cqe_flushed, u64 *cookie, u32 *credit); int iwch_cxgb3_ofld_send(struct t3cdev *tdev, struct sk_buff *skb); -#define MOD "iw_cxgb3: " -#define PDBG(fmt, args...) pr_debug(MOD fmt, ## args) +#ifdef pr_fmt +#undef pr_fmt +#endif + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #ifdef DEBUG void cxio_dump_tpt(struct cxio_rdev *rev, u32 stag); diff --git a/drivers/infiniband/hw/cxgb3/cxio_resource.c b/drivers/infiniband/hw/cxgb3/cxio_resource.c index c40088ecf9f3..c6e7bc4420b6 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_resource.c +++ b/drivers/infiniband/hw/cxgb3/cxio_resource.c @@ -209,13 +209,13 @@ u32 cxio_hal_get_qpid(struct cxio_hal_resource *rscp) { u32 qpid = cxio_hal_get_resource(&rscp->qpid_fifo, &rscp->qpid_fifo_lock); - PDBG("%s qpid 0x%x\n", __func__, qpid); + pr_debug("%s qpid 0x%x\n", __func__, qpid); return qpid; } void cxio_hal_put_qpid(struct cxio_hal_resource *rscp, u32 qpid) { - PDBG("%s qpid 0x%x\n", __func__, qpid); + pr_debug("%s qpid 0x%x\n", __func__, qpid); cxio_hal_put_resource(&rscp->qpid_fifo, &rscp->qpid_fifo_lock, qpid); } @@ -257,13 +257,13 @@ void cxio_hal_destroy_resource(struct cxio_hal_resource *rscp) u32 cxio_hal_pblpool_alloc(struct cxio_rdev *rdev_p, int size) { unsigned long addr = gen_pool_alloc(rdev_p->pbl_pool, size); - PDBG("%s addr 0x%x size %d\n", __func__, (u32)addr, size); + pr_debug("%s addr 0x%x size %d\n", __func__, (u32)addr, size); return (u32)addr; } void cxio_hal_pblpool_free(struct cxio_rdev *rdev_p, u32 addr, int size) { - PDBG("%s addr 0x%x size %d\n", __func__, addr, size); + pr_debug("%s addr 0x%x size %d\n", __func__, addr, size); gen_pool_free(rdev_p->pbl_pool, (unsigned long)addr, size); } @@ -282,17 +282,18 @@ int cxio_hal_pblpool_create(struct cxio_rdev *rdev_p) pbl_chunk = min(rdev_p->rnic_info.pbl_top - pbl_start + 1, pbl_chunk); if (gen_pool_add(rdev_p->pbl_pool, pbl_start, pbl_chunk, -1)) { - PDBG("%s failed to add PBL chunk (%x/%x)\n", - __func__, pbl_start, pbl_chunk); + pr_debug("%s failed to add PBL chunk (%x/%x)\n", + __func__, pbl_start, pbl_chunk); if (pbl_chunk <= 1024 << MIN_PBL_SHIFT) { - printk(KERN_WARNING MOD "%s: Failed to add all PBL chunks (%x/%x)\n", - __func__, pbl_start, rdev_p->rnic_info.pbl_top - pbl_start); + pr_warn("%s: Failed to add all PBL chunks (%x/%x)\n", + __func__, pbl_start, + rdev_p->rnic_info.pbl_top - pbl_start); return 0; } pbl_chunk >>= 1; } else { - PDBG("%s added PBL chunk (%x/%x)\n", - __func__, pbl_start, pbl_chunk); + pr_debug("%s added PBL chunk (%x/%x)\n", + __func__, pbl_start, pbl_chunk); pbl_start += pbl_chunk; } } @@ -315,13 +316,13 @@ void cxio_hal_pblpool_destroy(struct cxio_rdev *rdev_p) u32 cxio_hal_rqtpool_alloc(struct cxio_rdev *rdev_p, int size) { unsigned long addr = gen_pool_alloc(rdev_p->rqt_pool, size << 6); - PDBG("%s addr 0x%x size %d\n", __func__, (u32)addr, size << 6); + pr_debug("%s addr 0x%x size %d\n", __func__, (u32)addr, size << 6); return (u32)addr; } void cxio_hal_rqtpool_free(struct cxio_rdev *rdev_p, u32 addr, int size) { - PDBG("%s addr 0x%x size %d\n", __func__, addr, size << 6); + pr_debug("%s addr 0x%x size %d\n", __func__, addr, size << 6); gen_pool_free(rdev_p->rqt_pool, (unsigned long)addr, size << 6); } diff --git a/drivers/infiniband/hw/cxgb3/iwch.c b/drivers/infiniband/hw/cxgb3/iwch.c index b3e11329801d..47b2ce2ef203 100644 --- a/drivers/infiniband/hw/cxgb3/iwch.c +++ b/drivers/infiniband/hw/cxgb3/iwch.c @@ -105,7 +105,7 @@ static void iwch_db_drop_task(struct work_struct *work) static void rnic_init(struct iwch_dev *rnicp) { - PDBG("%s iwch_dev %p\n", __func__, rnicp); + pr_debug("%s iwch_dev %p\n", __func__, rnicp); idr_init(&rnicp->cqidr); idr_init(&rnicp->qpidr); idr_init(&rnicp->mmidr); @@ -145,12 +145,11 @@ static void open_rnic_dev(struct t3cdev *tdev) { struct iwch_dev *rnicp; - PDBG("%s t3cdev %p\n", __func__, tdev); - printk_once(KERN_INFO MOD "Chelsio T3 RDMA Driver - version %s\n", - DRV_VERSION); + pr_debug("%s t3cdev %p\n", __func__, tdev); + pr_info_once("Chelsio T3 RDMA Driver - version %s\n", DRV_VERSION); rnicp = (struct iwch_dev *)ib_alloc_device(sizeof(*rnicp)); if (!rnicp) { - printk(KERN_ERR MOD "Cannot allocate ib device\n"); + pr_err("Cannot allocate ib device\n"); return; } rnicp->rdev.ulp = rnicp; @@ -160,7 +159,7 @@ static void open_rnic_dev(struct t3cdev *tdev) if (cxio_rdev_open(&rnicp->rdev)) { mutex_unlock(&dev_mutex); - printk(KERN_ERR MOD "Unable to open CXIO rdev\n"); + pr_err("Unable to open CXIO rdev\n"); ib_dealloc_device(&rnicp->ibdev); return; } @@ -171,18 +170,18 @@ static void open_rnic_dev(struct t3cdev *tdev) mutex_unlock(&dev_mutex); if (iwch_register_device(rnicp)) { - printk(KERN_ERR MOD "Unable to register device\n"); + pr_err("Unable to register device\n"); close_rnic_dev(tdev); } - printk(KERN_INFO MOD "Initialized device %s\n", - pci_name(rnicp->rdev.rnic_info.pdev)); + pr_info("Initialized device %s\n", + pci_name(rnicp->rdev.rnic_info.pdev)); return; } static void close_rnic_dev(struct t3cdev *tdev) { struct iwch_dev *dev, *tmp; - PDBG("%s t3cdev %p\n", __func__, tdev); + pr_debug("%s t3cdev %p\n", __func__, tdev); mutex_lock(&dev_mutex); list_for_each_entry_safe(dev, tmp, &dev_list, entry) { if (dev->rdev.t3cdev_p == tdev) { diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.c b/drivers/infiniband/hw/cxgb3/iwch_cm.c index 65ee64400deb..b61630eba912 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_cm.c +++ b/drivers/infiniband/hw/cxgb3/iwch_cm.c @@ -112,9 +112,9 @@ static void connect_reply_upcall(struct iwch_ep *ep, int status); static void start_ep_timer(struct iwch_ep *ep) { - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); if (timer_pending(&ep->timer)) { - PDBG("%s stopped / restarted timer ep %p\n", __func__, ep); + pr_debug("%s stopped / restarted timer ep %p\n", __func__, ep); del_timer_sync(&ep->timer); } else get_ep(&ep->com); @@ -126,7 +126,7 @@ static void start_ep_timer(struct iwch_ep *ep) static void stop_ep_timer(struct iwch_ep *ep) { - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); if (!timer_pending(&ep->timer)) { WARN(1, "%s timer stopped when its not running! ep %p state %u\n", __func__, ep, ep->com.state); @@ -227,13 +227,13 @@ int iwch_resume_tid(struct iwch_ep *ep) static void set_emss(struct iwch_ep *ep, u16 opt) { - PDBG("%s ep %p opt %u\n", __func__, ep, opt); + pr_debug("%s ep %p opt %u\n", __func__, ep, opt); ep->emss = T3C_DATA(ep->com.tdev)->mtus[G_TCPOPT_MSS(opt)] - 40; if (G_TCPOPT_TSTAMP(opt)) ep->emss -= 12; if (ep->emss < 128) ep->emss = 128; - PDBG("emss=%d\n", ep->emss); + pr_debug("emss=%d\n", ep->emss); } static enum iwch_ep_state state_read(struct iwch_ep_common *epc) @@ -257,7 +257,7 @@ static void state_set(struct iwch_ep_common *epc, enum iwch_ep_state new) unsigned long flags; spin_lock_irqsave(&epc->lock, flags); - PDBG("%s - %s -> %s\n", __func__, states[epc->state], states[new]); + pr_debug("%s - %s -> %s\n", __func__, states[epc->state], states[new]); __state_set(epc, new); spin_unlock_irqrestore(&epc->lock, flags); return; @@ -273,7 +273,7 @@ static void *alloc_ep(int size, gfp_t gfp) spin_lock_init(&epc->lock); init_waitqueue_head(&epc->waitq); } - PDBG("%s alloc ep %p\n", __func__, epc); + pr_debug("%s alloc ep %p\n", __func__, epc); return epc; } @@ -282,7 +282,8 @@ void __free_ep(struct kref *kref) struct iwch_ep *ep; ep = container_of(container_of(kref, struct iwch_ep_common, kref), struct iwch_ep, com); - PDBG("%s ep %p state %s\n", __func__, ep, states[state_read(&ep->com)]); + pr_debug("%s ep %p state %s\n", + __func__, ep, states[state_read(&ep->com)]); if (test_bit(RELEASE_RESOURCES, &ep->com.flags)) { cxgb3_remove_tid(ep->com.tdev, (void *)ep, ep->hwtid); dst_release(ep->dst); @@ -293,7 +294,7 @@ void __free_ep(struct kref *kref) static void release_ep_resources(struct iwch_ep *ep) { - PDBG("%s ep %p tid %d\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %d\n", __func__, ep, ep->hwtid); set_bit(RELEASE_RESOURCES, &ep->com.flags); put_ep(&ep->com); } @@ -358,7 +359,7 @@ static unsigned int find_best_mtu(const struct t3c_data *d, unsigned short mtu) static void arp_failure_discard(struct t3cdev *dev, struct sk_buff *skb) { - PDBG("%s t3cdev %p\n", __func__, dev); + pr_debug("%s t3cdev %p\n", __func__, dev); kfree_skb(skb); } @@ -367,7 +368,7 @@ static void arp_failure_discard(struct t3cdev *dev, struct sk_buff *skb) */ static void act_open_req_arp_failure(struct t3cdev *dev, struct sk_buff *skb) { - printk(KERN_ERR MOD "ARP failure during connect\n"); + pr_err("ARP failure during connect\n"); kfree_skb(skb); } @@ -379,7 +380,7 @@ static void abort_arp_failure(struct t3cdev *dev, struct sk_buff *skb) { struct cpl_abort_req *req = cplhdr(skb); - PDBG("%s t3cdev %p\n", __func__, dev); + pr_debug("%s t3cdev %p\n", __func__, dev); req->cmd = CPL_ABORT_NO_RST; iwch_cxgb3_ofld_send(dev, skb); } @@ -389,10 +390,10 @@ static int send_halfclose(struct iwch_ep *ep, gfp_t gfp) struct cpl_close_con_req *req; struct sk_buff *skb; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); skb = get_skb(NULL, sizeof(*req), gfp); if (!skb) { - printk(KERN_ERR MOD "%s - failed to alloc skb\n", __func__); + pr_err("%s - failed to alloc skb\n", __func__); return -ENOMEM; } skb->priority = CPL_PRIORITY_DATA; @@ -408,11 +409,10 @@ static int send_abort(struct iwch_ep *ep, struct sk_buff *skb, gfp_t gfp) { struct cpl_abort_req *req; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); skb = get_skb(skb, sizeof(*req), gfp); if (!skb) { - printk(KERN_ERR MOD "%s - failed to alloc skb.\n", - __func__); + pr_err("%s - failed to alloc skb\n", __func__); return -ENOMEM; } skb->priority = CPL_PRIORITY_DATA; @@ -434,12 +434,11 @@ static int send_connect(struct iwch_ep *ep) unsigned int mtu_idx; int wscale; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); skb = get_skb(NULL, sizeof(*req), GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "%s - failed to alloc skb.\n", - __func__); + pr_err("%s - failed to alloc skb\n", __func__); return -ENOMEM; } mtu_idx = find_best_mtu(T3C_DATA(ep->com.tdev), dst_mtu(ep->dst)); @@ -478,7 +477,7 @@ static void send_mpa_req(struct iwch_ep *ep, struct sk_buff *skb) struct mpa_message *mpa; int len; - PDBG("%s ep %p pd_len %d\n", __func__, ep, ep->plen); + pr_debug("%s ep %p pd_len %d\n", __func__, ep, ep->plen); BUG_ON(skb_cloned(skb)); @@ -538,13 +537,13 @@ static int send_mpa_reject(struct iwch_ep *ep, const void *pdata, u8 plen) struct mpa_message *mpa; struct sk_buff *skb; - PDBG("%s ep %p plen %d\n", __func__, ep, plen); + pr_debug("%s ep %p plen %d\n", __func__, ep, plen); mpalen = sizeof(*mpa) + plen; skb = get_skb(NULL, mpalen + sizeof(*req), GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "%s - cannot alloc skb!\n", __func__); + pr_err("%s - cannot alloc skb!\n", __func__); return -ENOMEM; } skb_reserve(skb, sizeof(*req)); @@ -587,13 +586,13 @@ static int send_mpa_reply(struct iwch_ep *ep, const void *pdata, u8 plen) int len; struct sk_buff *skb; - PDBG("%s ep %p plen %d\n", __func__, ep, plen); + pr_debug("%s ep %p plen %d\n", __func__, ep, plen); mpalen = sizeof(*mpa) + plen; skb = get_skb(NULL, mpalen + sizeof(*req), GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "%s - cannot alloc skb!\n", __func__); + pr_err("%s - cannot alloc skb!\n", __func__); return -ENOMEM; } skb->priority = CPL_PRIORITY_DATA; @@ -636,7 +635,7 @@ static int act_establish(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) struct cpl_act_establish *req = cplhdr(skb); unsigned int tid = GET_TID(req); - PDBG("%s ep %p tid %d\n", __func__, ep, tid); + pr_debug("%s ep %p tid %d\n", __func__, ep, tid); dst_confirm(ep->dst); @@ -660,7 +659,7 @@ static int act_establish(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) static void abort_connection(struct iwch_ep *ep, struct sk_buff *skb, gfp_t gfp) { - PDBG("%s ep %p\n", __FILE__, ep); + pr_debug("%s ep %p\n", __FILE__, ep); state_set(&ep->com, ABORTING); send_abort(ep, skb, gfp); } @@ -669,12 +668,12 @@ static void close_complete_upcall(struct iwch_ep *ep) { struct iw_cm_event event; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_CLOSE; if (ep->com.cm_id) { - PDBG("close complete delivered ep %p cm_id %p tid %d\n", - ep, ep->com.cm_id, ep->hwtid); + pr_debug("close complete delivered ep %p cm_id %p tid %d\n", + ep, ep->com.cm_id, ep->hwtid); ep->com.cm_id->event_handler(ep->com.cm_id, &event); ep->com.cm_id->rem_ref(ep->com.cm_id); ep->com.cm_id = NULL; @@ -686,12 +685,12 @@ static void peer_close_upcall(struct iwch_ep *ep) { struct iw_cm_event event; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_DISCONNECT; if (ep->com.cm_id) { - PDBG("peer close delivered ep %p cm_id %p tid %d\n", - ep, ep->com.cm_id, ep->hwtid); + pr_debug("peer close delivered ep %p cm_id %p tid %d\n", + ep, ep->com.cm_id, ep->hwtid); ep->com.cm_id->event_handler(ep->com.cm_id, &event); } } @@ -700,13 +699,13 @@ static void peer_abort_upcall(struct iwch_ep *ep) { struct iw_cm_event event; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_CLOSE; event.status = -ECONNRESET; if (ep->com.cm_id) { - PDBG("abort delivered ep %p cm_id %p tid %d\n", ep, - ep->com.cm_id, ep->hwtid); + pr_debug("abort delivered ep %p cm_id %p tid %d\n", ep, + ep->com.cm_id, ep->hwtid); ep->com.cm_id->event_handler(ep->com.cm_id, &event); ep->com.cm_id->rem_ref(ep->com.cm_id); ep->com.cm_id = NULL; @@ -718,7 +717,7 @@ static void connect_reply_upcall(struct iwch_ep *ep, int status) { struct iw_cm_event event; - PDBG("%s ep %p status %d\n", __func__, ep, status); + pr_debug("%s ep %p status %d\n", __func__, ep, status); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_CONNECT_REPLY; event.status = status; @@ -732,8 +731,8 @@ static void connect_reply_upcall(struct iwch_ep *ep, int status) event.private_data = ep->mpa_pkt + sizeof(struct mpa_message); } if (ep->com.cm_id) { - PDBG("%s ep %p tid %d status %d\n", __func__, ep, - ep->hwtid, status); + pr_debug("%s ep %p tid %d status %d\n", __func__, ep, + ep->hwtid, status); ep->com.cm_id->event_handler(ep->com.cm_id, &event); } if (status < 0) { @@ -747,7 +746,7 @@ static void connect_request_upcall(struct iwch_ep *ep) { struct iw_cm_event event; - PDBG("%s ep %p tid %d\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %d\n", __func__, ep, ep->hwtid); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_CONNECT_REQUEST; memcpy(&event.local_addr, &ep->com.local_addr, @@ -776,7 +775,7 @@ static void established_upcall(struct iwch_ep *ep) { struct iw_cm_event event; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_ESTABLISHED; /* @@ -785,7 +784,7 @@ static void established_upcall(struct iwch_ep *ep) */ event.ird = event.ord = 8; if (ep->com.cm_id) { - PDBG("%s ep %p tid %d\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %d\n", __func__, ep, ep->hwtid); ep->com.cm_id->event_handler(ep->com.cm_id, &event); } } @@ -795,10 +794,10 @@ static int update_rx_credits(struct iwch_ep *ep, u32 credits) struct cpl_rx_data_ack *req; struct sk_buff *skb; - PDBG("%s ep %p credits %u\n", __func__, ep, credits); + pr_debug("%s ep %p credits %u\n", __func__, ep, credits); skb = get_skb(NULL, sizeof(*req), GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "update_rx_credits - cannot alloc skb!\n"); + pr_err("update_rx_credits - cannot alloc skb!\n"); return 0; } @@ -819,7 +818,7 @@ static void process_mpa_reply(struct iwch_ep *ep, struct sk_buff *skb) enum iwch_qp_attr_mask mask; int err; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); /* * Stop mpa timer. If it expired, then the state has @@ -906,10 +905,10 @@ static void process_mpa_reply(struct iwch_ep *ep, struct sk_buff *skb) ep->mpa_attr.recv_marker_enabled = markers_enabled; ep->mpa_attr.xmit_marker_enabled = mpa->flags & MPA_MARKERS ? 1 : 0; ep->mpa_attr.version = mpa_rev; - PDBG("%s - crc_enabled=%d, recv_marker_enabled=%d, " - "xmit_marker_enabled=%d, version=%d\n", __func__, - ep->mpa_attr.crc_enabled, ep->mpa_attr.recv_marker_enabled, - ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version); + pr_debug("%s - crc_enabled=%d, recv_marker_enabled=%d, xmit_marker_enabled=%d, version=%d\n", + __func__, + ep->mpa_attr.crc_enabled, ep->mpa_attr.recv_marker_enabled, + ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version); attrs.mpa_attr = ep->mpa_attr; attrs.max_ird = ep->ird; @@ -944,7 +943,7 @@ static void process_mpa_request(struct iwch_ep *ep, struct sk_buff *skb) struct mpa_message *mpa; u16 plen; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); /* * Stop mpa timer. If it expired, then the state has @@ -964,7 +963,7 @@ static void process_mpa_request(struct iwch_ep *ep, struct sk_buff *skb) return; } - PDBG("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__); + pr_debug("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__); /* * Copy the new data into our accumulation buffer. @@ -979,7 +978,7 @@ static void process_mpa_request(struct iwch_ep *ep, struct sk_buff *skb) */ if (ep->mpa_pkt_len < sizeof(*mpa)) return; - PDBG("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__); + pr_debug("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__); mpa = (struct mpa_message *) ep->mpa_pkt; /* @@ -1029,10 +1028,10 @@ static void process_mpa_request(struct iwch_ep *ep, struct sk_buff *skb) ep->mpa_attr.recv_marker_enabled = markers_enabled; ep->mpa_attr.xmit_marker_enabled = mpa->flags & MPA_MARKERS ? 1 : 0; ep->mpa_attr.version = mpa_rev; - PDBG("%s - crc_enabled=%d, recv_marker_enabled=%d, " - "xmit_marker_enabled=%d, version=%d\n", __func__, - ep->mpa_attr.crc_enabled, ep->mpa_attr.recv_marker_enabled, - ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version); + pr_debug("%s - crc_enabled=%d, recv_marker_enabled=%d, xmit_marker_enabled=%d, version=%d\n", + __func__, + ep->mpa_attr.crc_enabled, ep->mpa_attr.recv_marker_enabled, + ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version); state_set(&ep->com, MPA_REQ_RCVD); @@ -1047,7 +1046,7 @@ static int rx_data(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) struct cpl_rx_data *hdr = cplhdr(skb); unsigned int dlen = ntohs(hdr->len); - PDBG("%s ep %p dlen %u\n", __func__, ep, dlen); + pr_debug("%s ep %p dlen %u\n", __func__, ep, dlen); skb_pull(skb, sizeof(*hdr)); skb_trim(skb, dlen); @@ -1065,8 +1064,7 @@ static int rx_data(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) case MPA_REP_SENT: break; default: - printk(KERN_ERR MOD "%s Unexpected streaming data." - " ep %p state %d tid %d\n", + pr_err("%s Unexpected streaming data. ep %p state %d tid %d\n", __func__, ep, state_read(&ep->com), ep->hwtid); /* @@ -1095,11 +1093,11 @@ static int tx_ack(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) unsigned long flags; int post_zb = 0; - PDBG("%s ep %p credits %u\n", __func__, ep, credits); + pr_debug("%s ep %p credits %u\n", __func__, ep, credits); if (credits == 0) { - PDBG("%s 0 credit ack ep %p state %u\n", - __func__, ep, state_read(&ep->com)); + pr_debug("%s 0 credit ack ep %p state %u\n", + __func__, ep, state_read(&ep->com)); return CPL_RET_BUF_DONE; } @@ -1107,24 +1105,24 @@ static int tx_ack(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) BUG_ON(credits != 1); dst_confirm(ep->dst); if (!ep->mpa_skb) { - PDBG("%s rdma_init wr_ack ep %p state %u\n", - __func__, ep, ep->com.state); + pr_debug("%s rdma_init wr_ack ep %p state %u\n", + __func__, ep, ep->com.state); if (ep->mpa_attr.initiator) { - PDBG("%s initiator ep %p state %u\n", - __func__, ep, ep->com.state); + pr_debug("%s initiator ep %p state %u\n", + __func__, ep, ep->com.state); if (peer2peer && ep->com.state == FPDU_MODE) post_zb = 1; } else { - PDBG("%s responder ep %p state %u\n", - __func__, ep, ep->com.state); + pr_debug("%s responder ep %p state %u\n", + __func__, ep, ep->com.state); if (ep->com.state == MPA_REQ_RCVD) { ep->com.rpl_done = 1; wake_up(&ep->com.waitq); } } } else { - PDBG("%s lsm ack ep %p state %u freeing skb\n", - __func__, ep, ep->com.state); + pr_debug("%s lsm ack ep %p state %u freeing skb\n", + __func__, ep, ep->com.state); kfree_skb(ep->mpa_skb); ep->mpa_skb = NULL; } @@ -1140,7 +1138,7 @@ static int abort_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) unsigned long flags; int release = 0; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); BUG_ON(!ep); /* @@ -1159,8 +1157,7 @@ static int abort_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) release = 1; break; default: - printk(KERN_ERR "%s ep %p state %d\n", - __func__, ep, ep->com.state); + pr_err("%s ep %p state %d\n", __func__, ep, ep->com.state); break; } spin_unlock_irqrestore(&ep->com.lock, flags); @@ -1184,8 +1181,8 @@ static int act_open_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) struct iwch_ep *ep = ctx; struct cpl_act_open_rpl *rpl = cplhdr(skb); - PDBG("%s ep %p status %u errno %d\n", __func__, ep, rpl->status, - status2errno(rpl->status)); + pr_debug("%s ep %p status %u errno %d\n", __func__, ep, rpl->status, + status2errno(rpl->status)); connect_reply_upcall(ep, status2errno(rpl->status)); state_set(&ep->com, DEAD); if (ep->com.tdev->type != T3A && act_open_has_tid(rpl->status)) @@ -1202,10 +1199,10 @@ static int listen_start(struct iwch_listen_ep *ep) struct sk_buff *skb; struct cpl_pass_open_req *req; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); skb = get_skb(NULL, sizeof(*req), GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "t3c_listen_start failed to alloc skb!\n"); + pr_err("t3c_listen_start failed to alloc skb!\n"); return -ENOMEM; } @@ -1230,8 +1227,8 @@ static int pass_open_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) struct iwch_listen_ep *ep = ctx; struct cpl_pass_open_rpl *rpl = cplhdr(skb); - PDBG("%s ep %p status %d error %d\n", __func__, ep, - rpl->status, status2errno(rpl->status)); + pr_debug("%s ep %p status %d error %d\n", __func__, ep, + rpl->status, status2errno(rpl->status)); ep->com.rpl_err = status2errno(rpl->status); ep->com.rpl_done = 1; wake_up(&ep->com.waitq); @@ -1244,10 +1241,10 @@ static int listen_stop(struct iwch_listen_ep *ep) struct sk_buff *skb; struct cpl_close_listserv_req *req; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); skb = get_skb(NULL, sizeof(*req), GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "%s - failed to alloc skb\n", __func__); + pr_err("%s - failed to alloc skb\n", __func__); return -ENOMEM; } req = (struct cpl_close_listserv_req *) skb_put(skb, sizeof(*req)); @@ -1264,7 +1261,7 @@ static int close_listsrv_rpl(struct t3cdev *tdev, struct sk_buff *skb, struct iwch_listen_ep *ep = ctx; struct cpl_close_listserv_rpl *rpl = cplhdr(skb); - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); ep->com.rpl_err = status2errno(rpl->status); ep->com.rpl_done = 1; wake_up(&ep->com.waitq); @@ -1278,7 +1275,7 @@ static void accept_cr(struct iwch_ep *ep, __be32 peer_ip, struct sk_buff *skb) u32 opt0h, opt0l, opt2; int wscale; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); BUG_ON(skb_cloned(skb)); skb_trim(skb, sizeof(*rpl)); skb_get(skb); @@ -1312,8 +1309,8 @@ static void accept_cr(struct iwch_ep *ep, __be32 peer_ip, struct sk_buff *skb) static void reject_cr(struct t3cdev *tdev, u32 hwtid, __be32 peer_ip, struct sk_buff *skb) { - PDBG("%s t3cdev %p tid %u peer_ip %x\n", __func__, tdev, hwtid, - peer_ip); + pr_debug("%s t3cdev %p tid %u peer_ip %x\n", __func__, tdev, hwtid, + peer_ip); BUG_ON(skb_cloned(skb)); skb_trim(skb, sizeof(struct cpl_tid_release)); skb_get(skb); @@ -1347,11 +1344,10 @@ static int pass_accept_req(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) struct rtable *rt; struct iff_mac tim; - PDBG("%s parent ep %p tid %u\n", __func__, parent_ep, hwtid); + pr_debug("%s parent ep %p tid %u\n", __func__, parent_ep, hwtid); if (state_read(&parent_ep->com) != LISTEN) { - printk(KERN_ERR "%s - listening ep not in LISTEN\n", - __func__); + pr_err("%s - listening ep not in LISTEN\n", __func__); goto reject; } @@ -1361,8 +1357,7 @@ static int pass_accept_req(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) tim.mac_addr = req->dst_mac; tim.vlan_tag = ntohs(req->vlan_tag); if (tdev->ctl(tdev, GET_IFF_FROM_MAC, &tim) < 0 || !tim.dev) { - printk(KERN_ERR "%s bad dst mac %pM\n", - __func__, req->dst_mac); + pr_err("%s bad dst mac %pM\n", __func__, req->dst_mac); goto reject; } @@ -1373,22 +1368,19 @@ static int pass_accept_req(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) req->local_port, req->peer_port, G_PASS_OPEN_TOS(ntohl(req->tos_tid))); if (!rt) { - printk(KERN_ERR MOD "%s - failed to find dst entry!\n", - __func__); + pr_err("%s - failed to find dst entry!\n", __func__); goto reject; } dst = &rt->dst; l2t = t3_l2t_get(tdev, dst, NULL, &req->peer_ip); if (!l2t) { - printk(KERN_ERR MOD "%s - failed to allocate l2t entry!\n", - __func__); + pr_err("%s - failed to allocate l2t entry!\n", __func__); dst_release(dst); goto reject; } child_ep = alloc_ep(sizeof(*child_ep), GFP_KERNEL); if (!child_ep) { - printk(KERN_ERR MOD "%s - failed to allocate ep entry!\n", - __func__); + pr_err("%s - failed to allocate ep entry!\n", __func__); l2t_release(tdev, l2t); dst_release(dst); goto reject; @@ -1423,7 +1415,7 @@ static int pass_establish(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) struct iwch_ep *ep = ctx; struct cpl_pass_establish *req = cplhdr(skb); - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); ep->snd_seq = ntohl(req->snd_isn); ep->rcv_seq = ntohl(req->rcv_isn); @@ -1444,7 +1436,7 @@ static int peer_close(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) int disconnect = 1; int release = 0; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); dst_confirm(ep->dst); spin_lock_irqsave(&ep->com.lock, flags); @@ -1467,14 +1459,14 @@ static int peer_close(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) __state_set(&ep->com, CLOSING); ep->com.rpl_done = 1; ep->com.rpl_err = -ECONNRESET; - PDBG("waking up ep %p\n", ep); + pr_debug("waking up ep %p\n", ep); wake_up(&ep->com.waitq); break; case MPA_REP_SENT: __state_set(&ep->com, CLOSING); ep->com.rpl_done = 1; ep->com.rpl_err = -ECONNRESET; - PDBG("waking up ep %p\n", ep); + pr_debug("waking up ep %p\n", ep); wake_up(&ep->com.waitq); break; case FPDU_MODE: @@ -1539,8 +1531,8 @@ static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) unsigned long flags; if (is_neg_adv_abort(req->status)) { - PDBG("%s neg_adv_abort ep %p tid %d\n", __func__, ep, - ep->hwtid); + pr_debug("%s neg_adv_abort ep %p tid %d\n", __func__, ep, + ep->hwtid); t3_l2t_send_event(ep->com.tdev, ep->l2t); return CPL_RET_BUF_DONE; } @@ -1554,7 +1546,7 @@ static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) } spin_lock_irqsave(&ep->com.lock, flags); - PDBG("%s ep %p state %u\n", __func__, ep, ep->com.state); + pr_debug("%s ep %p state %u\n", __func__, ep, ep->com.state); switch (ep->com.state) { case CONNECTING: break; @@ -1568,7 +1560,7 @@ static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) case MPA_REP_SENT: ep->com.rpl_done = 1; ep->com.rpl_err = -ECONNRESET; - PDBG("waking up ep %p\n", ep); + pr_debug("waking up ep %p\n", ep); wake_up(&ep->com.waitq); break; case MPA_REQ_RCVD: @@ -1581,7 +1573,7 @@ static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) */ ep->com.rpl_done = 1; ep->com.rpl_err = -ECONNRESET; - PDBG("waking up ep %p\n", ep); + pr_debug("waking up ep %p\n", ep); wake_up(&ep->com.waitq); break; case MORIBUND: @@ -1595,16 +1587,14 @@ static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) ep->com.qp, IWCH_QP_ATTR_NEXT_STATE, &attrs, 1); if (ret) - printk(KERN_ERR MOD - "%s - qp <- error failed!\n", - __func__); + pr_err("%s - qp <- error failed!\n", __func__); } peer_abort_upcall(ep); break; case ABORTING: break; case DEAD: - PDBG("%s PEER_ABORT IN DEAD STATE!!!!\n", __func__); + pr_debug("%s PEER_ABORT IN DEAD STATE!!!!\n", __func__); spin_unlock_irqrestore(&ep->com.lock, flags); return CPL_RET_BUF_DONE; default: @@ -1620,8 +1610,7 @@ static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) rpl_skb = get_skb(skb, sizeof(*rpl), GFP_KERNEL); if (!rpl_skb) { - printk(KERN_ERR MOD "%s - cannot allocate skb!\n", - __func__); + pr_err("%s - cannot allocate skb!\n", __func__); release = 1; goto out; } @@ -1645,7 +1634,7 @@ static int close_con_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) unsigned long flags; int release = 0; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); BUG_ON(!ep); /* The cm_id may be null if we failed to connect */ @@ -1699,9 +1688,9 @@ static int terminate(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) if (state_read(&ep->com) != FPDU_MODE) return CPL_RET_BUF_DONE; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); skb_pull(skb, sizeof(struct cpl_rdma_terminate)); - PDBG("%s saving %d bytes of term msg\n", __func__, skb->len); + pr_debug("%s saving %d bytes of term msg\n", __func__, skb->len); skb_copy_from_linear_data(skb, ep->com.qp->attr.terminate_buffer, skb->len); ep->com.qp->attr.terminate_msg_len = skb->len; @@ -1714,12 +1703,12 @@ static int ec_status(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) struct cpl_rdma_ec_status *rep = cplhdr(skb); struct iwch_ep *ep = ctx; - PDBG("%s ep %p tid %u status %d\n", __func__, ep, ep->hwtid, - rep->status); + pr_debug("%s ep %p tid %u status %d\n", __func__, ep, ep->hwtid, + rep->status); if (rep->status) { struct iwch_qp_attributes attrs; - printk(KERN_ERR MOD "%s BAD CLOSE - Aborting tid %u\n", + pr_err("%s BAD CLOSE - Aborting tid %u\n", __func__, ep->hwtid); stop_ep_timer(ep); attrs.next_state = IWCH_QP_STATE_ERROR; @@ -1739,8 +1728,8 @@ static void ep_timeout(unsigned long arg) int abort = 1; spin_lock_irqsave(&ep->com.lock, flags); - PDBG("%s ep %p tid %u state %d\n", __func__, ep, ep->hwtid, - ep->com.state); + pr_debug("%s ep %p tid %u state %d\n", __func__, ep, ep->hwtid, + ep->com.state); switch (ep->com.state) { case MPA_REQ_SENT: __state_set(&ep->com, ABORTING); @@ -1774,7 +1763,7 @@ int iwch_reject_cr(struct iw_cm_id *cm_id, const void *pdata, u8 pdata_len) { int err; struct iwch_ep *ep = to_ep(cm_id); - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); if (state_read(&ep->com) == DEAD) { put_ep(&ep->com); @@ -1800,7 +1789,7 @@ int iwch_accept_cr(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) struct iwch_dev *h = to_iwch_dev(cm_id->device); struct iwch_qp *qp = get_qhp(h, conn_param->qpn); - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); if (state_read(&ep->com) == DEAD) { err = -ECONNRESET; goto err; @@ -1826,7 +1815,7 @@ int iwch_accept_cr(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) if (peer2peer && ep->ird == 0) ep->ird = 1; - PDBG("%s %d ird %d ord %d\n", __func__, __LINE__, ep->ird, ep->ord); + pr_debug("%s %d ird %d ord %d\n", __func__, __LINE__, ep->ird, ep->ord); /* bind QP to EP and move to RTS */ attrs.mpa_attr = ep->mpa_attr; @@ -1907,7 +1896,7 @@ int iwch_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) ep = alloc_ep(sizeof(*ep), GFP_KERNEL); if (!ep) { - printk(KERN_ERR MOD "%s - cannot alloc ep.\n", __func__); + pr_err("%s - cannot alloc ep\n", __func__); err = -ENOMEM; goto out; } @@ -1928,15 +1917,15 @@ int iwch_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) ep->com.cm_id = cm_id; ep->com.qp = get_qhp(h, conn_param->qpn); BUG_ON(!ep->com.qp); - PDBG("%s qpn 0x%x qp %p cm_id %p\n", __func__, conn_param->qpn, - ep->com.qp, cm_id); + pr_debug("%s qpn 0x%x qp %p cm_id %p\n", __func__, conn_param->qpn, + ep->com.qp, cm_id); /* * Allocate an active TID to initiate a TCP connection. */ ep->atid = cxgb3_alloc_atid(h->rdev.t3cdev_p, &t3c_client, ep); if (ep->atid == -1) { - printk(KERN_ERR MOD "%s - cannot alloc atid.\n", __func__); + pr_err("%s - cannot alloc atid\n", __func__); err = -ENOMEM; goto fail2; } @@ -1946,7 +1935,7 @@ int iwch_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) raddr->sin_addr.s_addr, laddr->sin_port, raddr->sin_port, IPTOS_LOWDELAY); if (!rt) { - printk(KERN_ERR MOD "%s - cannot find route.\n", __func__); + pr_err("%s - cannot find route\n", __func__); err = -EHOSTUNREACH; goto fail3; } @@ -1954,7 +1943,7 @@ int iwch_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) ep->l2t = t3_l2t_get(ep->com.tdev, ep->dst, NULL, &raddr->sin_addr.s_addr); if (!ep->l2t) { - printk(KERN_ERR MOD "%s - cannot alloc l2e.\n", __func__); + pr_err("%s - cannot alloc l2e\n", __func__); err = -ENOMEM; goto fail4; } @@ -1999,11 +1988,11 @@ int iwch_create_listen(struct iw_cm_id *cm_id, int backlog) ep = alloc_ep(sizeof(*ep), GFP_KERNEL); if (!ep) { - printk(KERN_ERR MOD "%s - cannot alloc ep.\n", __func__); + pr_err("%s - cannot alloc ep\n", __func__); err = -ENOMEM; goto fail1; } - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); ep->com.tdev = h->rdev.t3cdev_p; cm_id->add_ref(cm_id); ep->com.cm_id = cm_id; @@ -2016,7 +2005,7 @@ int iwch_create_listen(struct iw_cm_id *cm_id, int backlog) */ ep->stid = cxgb3_alloc_stid(h->rdev.t3cdev_p, &t3c_client, ep); if (ep->stid == -1) { - printk(KERN_ERR MOD "%s - cannot alloc atid.\n", __func__); + pr_err("%s - cannot alloc atid\n", __func__); err = -ENOMEM; goto fail2; } @@ -2048,7 +2037,7 @@ int iwch_destroy_listen(struct iw_cm_id *cm_id) int err; struct iwch_listen_ep *ep = to_listen_ep(cm_id); - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); might_sleep(); state_set(&ep->com, DEAD); @@ -2077,8 +2066,8 @@ int iwch_ep_disconnect(struct iwch_ep *ep, int abrupt, gfp_t gfp) spin_lock_irqsave(&ep->com.lock, flags); - PDBG("%s ep %p state %s, abrupt %d\n", __func__, ep, - states[ep->com.state], abrupt); + pr_debug("%s ep %p state %s, abrupt %d\n", __func__, ep, + states[ep->com.state], abrupt); tdev = (struct t3cdev *)ep->com.tdev; rdev = (struct cxio_rdev *)tdev->ulp; @@ -2115,8 +2104,8 @@ int iwch_ep_disconnect(struct iwch_ep *ep, int abrupt, gfp_t gfp) case MORIBUND: case ABORTING: case DEAD: - PDBG("%s ignoring disconnect ep %p state %u\n", - __func__, ep, ep->com.state); + pr_debug("%s ignoring disconnect ep %p state %u\n", + __func__, ep, ep->com.state); break; default: BUG(); @@ -2145,8 +2134,8 @@ int iwch_ep_redirect(void *ctx, struct dst_entry *old, struct dst_entry *new, if (ep->dst != old) return 0; - PDBG("%s ep %p redirect to dst %p l2t %p\n", __func__, ep, new, - l2t); + pr_debug("%s ep %p redirect to dst %p l2t %p\n", __func__, ep, new, + l2t); dst_hold(new); l2t_release(ep->com.tdev, ep->l2t); ep->l2t = l2t; @@ -2225,8 +2214,8 @@ static int set_tcb_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) struct cpl_set_tcb_rpl *rpl = cplhdr(skb); if (rpl->status != CPL_ERR_NONE) { - printk(KERN_ERR MOD "Unexpected SET_TCB_RPL status %u " - "for tid %u\n", rpl->status, GET_TID(rpl)); + pr_err("Unexpected SET_TCB_RPL status %u for tid %u\n", + rpl->status, GET_TID(rpl)); } return CPL_RET_BUF_DONE; } diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.h b/drivers/infiniband/hw/cxgb3/iwch_cm.h index e66e75921797..cc7fe644d260 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_cm.h +++ b/drivers/infiniband/hw/cxgb3/iwch_cm.h @@ -53,17 +53,17 @@ #define MPA_MARKERS 0x80 #define MPA_FLAGS_MASK 0xE0 -#define put_ep(ep) { \ - PDBG("put_ep (via %s:%u) ep %p refcnt %d\n", __func__, __LINE__, \ - ep, kref_read(&((ep)->kref))); \ - WARN_ON(kref_read(&((ep)->kref)) < 1); \ - kref_put(&((ep)->kref), __free_ep); \ +#define put_ep(ep) { \ + pr_debug("put_ep (via %s:%u) ep %p refcnt %d\n", \ + __func__, __LINE__, ep, kref_read(&((ep)->kref))); \ + WARN_ON(kref_read(&((ep)->kref)) < 1); \ + kref_put(&((ep)->kref), __free_ep); \ } -#define get_ep(ep) { \ - PDBG("get_ep (via %s:%u) ep %p, refcnt %d\n", __func__, __LINE__, \ - ep, kref_read(&((ep)->kref))); \ - kref_get(&((ep)->kref)); \ +#define get_ep(ep) { \ + pr_debug("get_ep (via %s:%u) ep %p, refcnt %d\n", \ + __func__, __LINE__, ep, kref_read(&((ep)->kref))); \ + kref_get(&((ep)->kref)); \ } struct mpa_message { diff --git a/drivers/infiniband/hw/cxgb3/iwch_cq.c b/drivers/infiniband/hw/cxgb3/iwch_cq.c index 97fbfd2c298e..dd5348e48806 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_cq.c +++ b/drivers/infiniband/hw/cxgb3/iwch_cq.c @@ -67,8 +67,8 @@ static int iwch_poll_cq_one(struct iwch_dev *rhp, struct iwch_cq *chp, ret = cxio_poll_cq(wq, &(chp->cq), &cqe, &cqe_flushed, &cookie, &credit); if (t3a_device(chp->rhp) && credit) { - PDBG("%s updating %d cq credits on id %d\n", __func__, - credit, chp->cq.cqid); + pr_debug("%s updating %d cq credits on id %d\n", __func__, + credit, chp->cq.cqid); cxio_hal_cq_op(&rhp->rdev, &chp->cq, CQ_CREDIT_UPDATE, credit); } @@ -83,11 +83,11 @@ static int iwch_poll_cq_one(struct iwch_dev *rhp, struct iwch_cq *chp, wc->vendor_err = CQE_STATUS(cqe); wc->wc_flags = 0; - PDBG("%s qpid 0x%x type %d opcode %d status 0x%x wrid hi 0x%x " - "lo 0x%x cookie 0x%llx\n", __func__, - CQE_QPID(cqe), CQE_TYPE(cqe), - CQE_OPCODE(cqe), CQE_STATUS(cqe), CQE_WRID_HI(cqe), - CQE_WRID_LOW(cqe), (unsigned long long) cookie); + pr_debug("%s qpid 0x%x type %d opcode %d status 0x%x wrid hi 0x%x lo 0x%x cookie 0x%llx\n", + __func__, + CQE_QPID(cqe), CQE_TYPE(cqe), + CQE_OPCODE(cqe), CQE_STATUS(cqe), CQE_WRID_HI(cqe), + CQE_WRID_LOW(cqe), (unsigned long long)cookie); if (CQE_TYPE(cqe) == 0) { if (!CQE_STATUS(cqe)) @@ -122,8 +122,7 @@ static int iwch_poll_cq_one(struct iwch_dev *rhp, struct iwch_cq *chp, wc->opcode = IB_WC_REG_MR; break; default: - printk(KERN_ERR MOD "Unexpected opcode %d " - "in the CQE received for QPID=0x%0x\n", + pr_err("Unexpected opcode %d in the CQE received for QPID=0x%0x\n", CQE_OPCODE(cqe), CQE_QPID(cqe)); ret = -EINVAL; goto out; @@ -177,8 +176,8 @@ static int iwch_poll_cq_one(struct iwch_dev *rhp, struct iwch_cq *chp, wc->status = IB_WC_WR_FLUSH_ERR; break; default: - printk(KERN_ERR MOD "Unexpected cqe_status 0x%x for " - "QPID=0x%0x\n", CQE_STATUS(cqe), CQE_QPID(cqe)); + pr_err("Unexpected cqe_status 0x%x for QPID=0x%0x\n", + CQE_STATUS(cqe), CQE_QPID(cqe)); ret = -EINVAL; } } diff --git a/drivers/infiniband/hw/cxgb3/iwch_ev.c b/drivers/infiniband/hw/cxgb3/iwch_ev.c index abcc9e76962b..4a0c82a8fb60 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_ev.c +++ b/drivers/infiniband/hw/cxgb3/iwch_ev.c @@ -52,7 +52,7 @@ static void post_qp_event(struct iwch_dev *rnicp, struct iwch_cq *chp, qhp = get_qhp(rnicp, CQE_QPID(rsp_msg->cqe)); if (!qhp) { - printk(KERN_ERR "%s unaffiliated error 0x%x qpid 0x%x\n", + pr_err("%s unaffiliated error 0x%x qpid 0x%x\n", __func__, CQE_STATUS(rsp_msg->cqe), CQE_QPID(rsp_msg->cqe)); spin_unlock(&rnicp->lock); @@ -61,15 +61,16 @@ static void post_qp_event(struct iwch_dev *rnicp, struct iwch_cq *chp, if ((qhp->attr.state == IWCH_QP_STATE_ERROR) || (qhp->attr.state == IWCH_QP_STATE_TERMINATE)) { - PDBG("%s AE received after RTS - " - "qp state %d qpid 0x%x status 0x%x\n", __func__, - qhp->attr.state, qhp->wq.qpid, CQE_STATUS(rsp_msg->cqe)); + pr_debug("%s AE received after RTS - qp state %d qpid 0x%x status 0x%x\n", + __func__, + qhp->attr.state, qhp->wq.qpid, + CQE_STATUS(rsp_msg->cqe)); spin_unlock(&rnicp->lock); return; } - printk(KERN_ERR "%s - AE qpid 0x%x opcode %d status 0x%x " - "type %d wrid.hi 0x%x wrid.lo 0x%x \n", __func__, + pr_err("%s - AE qpid 0x%x opcode %d status 0x%x type %d wrid.hi 0x%x wrid.lo 0x%x\n", + __func__, CQE_QPID(rsp_msg->cqe), CQE_OPCODE(rsp_msg->cqe), CQE_STATUS(rsp_msg->cqe), CQE_TYPE(rsp_msg->cqe), CQE_WRID_HI(rsp_msg->cqe), CQE_WRID_LOW(rsp_msg->cqe)); @@ -117,8 +118,7 @@ void iwch_ev_dispatch(struct cxio_rdev *rdev_p, struct sk_buff *skb) chp = get_chp(rnicp, cqid); qhp = get_qhp(rnicp, CQE_QPID(rsp_msg->cqe)); if (!chp || !qhp) { - printk(KERN_ERR MOD "BAD AE cqid 0x%x qpid 0x%x opcode %d " - "status 0x%x type %d wrid.hi 0x%x wrid.lo 0x%x \n", + pr_err("BAD AE cqid 0x%x qpid 0x%x opcode %d status 0x%x type %d wrid.hi 0x%x wrid.lo 0x%x\n", cqid, CQE_QPID(rsp_msg->cqe), CQE_OPCODE(rsp_msg->cqe), CQE_STATUS(rsp_msg->cqe), CQE_TYPE(rsp_msg->cqe), CQE_WRID_HI(rsp_msg->cqe), @@ -137,12 +137,12 @@ void iwch_ev_dispatch(struct cxio_rdev *rdev_p, struct sk_buff *skb) if ((CQE_OPCODE(rsp_msg->cqe) == T3_TERMINATE) && (CQE_STATUS(rsp_msg->cqe) == 0)) { if (SQ_TYPE(rsp_msg->cqe)) { - PDBG("%s QPID 0x%x ep %p disconnecting\n", - __func__, qhp->wq.qpid, qhp->ep); + pr_debug("%s QPID 0x%x ep %p disconnecting\n", + __func__, qhp->wq.qpid, qhp->ep); iwch_ep_disconnect(qhp->ep, 0, GFP_ATOMIC); } else { - PDBG("%s post REQ_ERR AE QPID 0x%x\n", __func__, - qhp->wq.qpid); + pr_debug("%s post REQ_ERR AE QPID 0x%x\n", __func__, + qhp->wq.qpid); post_qp_event(rnicp, chp, rsp_msg, IB_EVENT_QP_REQ_ERR, 0); iwch_ep_disconnect(qhp->ep, 0, GFP_ATOMIC); @@ -218,7 +218,7 @@ void iwch_ev_dispatch(struct cxio_rdev *rdev_p, struct sk_buff *skb) break; default: - printk(KERN_ERR MOD "Unknown T3 status 0x%x QPID 0x%x\n", + pr_err("Unknown T3 status 0x%x QPID 0x%x\n", CQE_STATUS(rsp_msg->cqe), qhp->wq.qpid); post_qp_event(rnicp, chp, rsp_msg, IB_EVENT_QP_FATAL, 1); break; diff --git a/drivers/infiniband/hw/cxgb3/iwch_mem.c b/drivers/infiniband/hw/cxgb3/iwch_mem.c index 1d04c872c9d5..12886b1b4b10 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_mem.c +++ b/drivers/infiniband/hw/cxgb3/iwch_mem.c @@ -48,7 +48,7 @@ static int iwch_finish_mem_reg(struct iwch_mr *mhp, u32 stag) mhp->attr.stag = stag; mmid = stag >> 8; mhp->ibmr.rkey = mhp->ibmr.lkey = stag; - PDBG("%s mmid 0x%x mhp %p\n", __func__, mmid, mhp); + pr_debug("%s mmid 0x%x mhp %p\n", __func__, mmid, mhp); return insert_handle(mhp->rhp, &mhp->rhp->mmidr, mhp, mmid); } diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.c b/drivers/infiniband/hw/cxgb3/iwch_provider.c index 86ecd3ea6a4b..29d30744d6c9 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_provider.c +++ b/drivers/infiniband/hw/cxgb3/iwch_provider.c @@ -62,7 +62,7 @@ #include "common.h" static struct ib_ah *iwch_ah_create(struct ib_pd *pd, - struct ib_ah_attr *ah_attr, + struct rdma_ah_attr *ah_attr, struct ib_udata *udata) { return ERR_PTR(-ENOSYS); @@ -103,7 +103,7 @@ static int iwch_dealloc_ucontext(struct ib_ucontext *context) struct iwch_ucontext *ucontext = to_iwch_ucontext(context); struct iwch_mm_entry *mm, *tmp; - PDBG("%s context %p\n", __func__, context); + pr_debug("%s context %p\n", __func__, context); list_for_each_entry_safe(mm, tmp, &ucontext->mmaps, entry) kfree(mm); cxio_release_ucontext(&rhp->rdev, &ucontext->uctx); @@ -117,7 +117,7 @@ static struct ib_ucontext *iwch_alloc_ucontext(struct ib_device *ibdev, struct iwch_ucontext *context; struct iwch_dev *rhp = to_iwch_dev(ibdev); - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); context = kzalloc(sizeof(*context), GFP_KERNEL); if (!context) return ERR_PTR(-ENOMEM); @@ -131,7 +131,7 @@ static int iwch_destroy_cq(struct ib_cq *ib_cq) { struct iwch_cq *chp; - PDBG("%s ib_cq %p\n", __func__, ib_cq); + pr_debug("%s ib_cq %p\n", __func__, ib_cq); chp = to_iwch_cq(ib_cq); remove_handle(chp->rhp, &chp->rhp->cqidr, chp->cq.cqid); @@ -157,7 +157,7 @@ static struct ib_cq *iwch_create_cq(struct ib_device *ibdev, static int warned; size_t resplen; - PDBG("%s ib_dev %p entries %d\n", __func__, ibdev, entries); + pr_debug("%s ib_dev %p entries %d\n", __func__, ibdev, entries); if (attr->flags) return ERR_PTR(-EINVAL); @@ -227,8 +227,7 @@ static struct ib_cq *iwch_create_cq(struct ib_device *ibdev, mm->addr = virt_to_phys(chp->cq.queue); if (udata->outlen < sizeof uresp) { if (!warned++) - printk(KERN_WARNING MOD "Warning - " - "downlevel libcxgb3 (non-fatal).\n"); + pr_warn("Warning - downlevel libcxgb3 (non-fatal)\n"); mm->len = PAGE_ALIGN((1UL << uresp.size_log2) * sizeof(struct t3_cqe)); resplen = sizeof(struct iwch_create_cq_resp_v0); @@ -246,9 +245,9 @@ static struct ib_cq *iwch_create_cq(struct ib_device *ibdev, } insert_mmap(ucontext, mm); } - PDBG("created cqid 0x%0x chp %p size 0x%0x, dma_addr 0x%0llx\n", - chp->cq.cqid, chp, (1 << chp->cq.size_log2), - (unsigned long long) chp->cq.dma_addr); + pr_debug("created cqid 0x%0x chp %p size 0x%0x, dma_addr 0x%0llx\n", + chp->cq.cqid, chp, (1 << chp->cq.size_log2), + (unsigned long long)chp->cq.dma_addr); return &chp->ibcq; } @@ -259,7 +258,7 @@ static int iwch_resize_cq(struct ib_cq *cq, int cqe, struct ib_udata *udata) struct t3_cq oldcq, newcq; int ret; - PDBG("%s ib_cq %p cqe %d\n", __func__, cq, cqe); + pr_debug("%s ib_cq %p cqe %d\n", __func__, cq, cqe); /* We don't downsize... */ if (cqe <= cq->cqe) @@ -306,8 +305,7 @@ static int iwch_resize_cq(struct ib_cq *cq, int cqe, struct ib_udata *udata) oldcq.cqid = newcq.cqid; ret = cxio_destroy_cq(&chp->rhp->rdev, &oldcq); if (ret) { - printk(KERN_ERR MOD "%s - cxio_destroy_cq failed %d\n", - __func__, ret); + pr_err("%s - cxio_destroy_cq failed %d\n", __func__, ret); } /* add user hooks here */ @@ -342,12 +340,11 @@ static int iwch_arm_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags flags) chp->cq.rptr = rptr; } else spin_lock_irqsave(&chp->lock, flag); - PDBG("%s rptr 0x%x\n", __func__, chp->cq.rptr); + pr_debug("%s rptr 0x%x\n", __func__, chp->cq.rptr); err = cxio_hal_cq_op(&rhp->rdev, &chp->cq, cq_op, 0); spin_unlock_irqrestore(&chp->lock, flag); if (err < 0) - printk(KERN_ERR MOD "Error %d rearming CQID 0x%x\n", err, - chp->cq.cqid); + pr_err("Error %d rearming CQID 0x%x\n", err, chp->cq.cqid); if (err > 0 && !(flags & IB_CQ_REPORT_MISSED_EVENTS)) err = 0; return err; @@ -363,8 +360,8 @@ static int iwch_mmap(struct ib_ucontext *context, struct vm_area_struct *vma) struct iwch_ucontext *ucontext; u64 addr; - PDBG("%s pgoff 0x%lx key 0x%x len %d\n", __func__, vma->vm_pgoff, - key, len); + pr_debug("%s pgoff 0x%lx key 0x%x len %d\n", __func__, vma->vm_pgoff, + key, len); if (vma->vm_start & (PAGE_SIZE-1)) { return -EINVAL; @@ -416,7 +413,7 @@ static int iwch_deallocate_pd(struct ib_pd *pd) php = to_iwch_pd(pd); rhp = php->rhp; - PDBG("%s ibpd %p pdid 0x%x\n", __func__, pd, php->pdid); + pr_debug("%s ibpd %p pdid 0x%x\n", __func__, pd, php->pdid); cxio_hal_put_pdid(rhp->rdev.rscp, php->pdid); kfree(php); return 0; @@ -430,7 +427,7 @@ static struct ib_pd *iwch_allocate_pd(struct ib_device *ibdev, u32 pdid; struct iwch_dev *rhp; - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); rhp = (struct iwch_dev *) ibdev; pdid = cxio_hal_get_pdid(rhp->rdev.rscp); if (!pdid) @@ -448,7 +445,7 @@ static struct ib_pd *iwch_allocate_pd(struct ib_device *ibdev, return ERR_PTR(-EFAULT); } } - PDBG("%s pdid 0x%0x ptr 0x%p\n", __func__, pdid, php); + pr_debug("%s pdid 0x%0x ptr 0x%p\n", __func__, pdid, php); return &php->ibpd; } @@ -458,7 +455,7 @@ static int iwch_dereg_mr(struct ib_mr *ib_mr) struct iwch_mr *mhp; u32 mmid; - PDBG("%s ib_mr %p\n", __func__, ib_mr); + pr_debug("%s ib_mr %p\n", __func__, ib_mr); mhp = to_iwch_mr(ib_mr); kfree(mhp->pages); @@ -472,7 +469,7 @@ static int iwch_dereg_mr(struct ib_mr *ib_mr) kfree((void *) (unsigned long) mhp->kva); if (mhp->umem) ib_umem_release(mhp->umem); - PDBG("%s mmid 0x%x ptr %p\n", __func__, mmid, mhp); + pr_debug("%s mmid 0x%x ptr %p\n", __func__, mmid, mhp); kfree(mhp); return 0; } @@ -487,13 +484,13 @@ static struct ib_mr *iwch_get_dma_mr(struct ib_pd *pd, int acc) __be64 *page_list; int shift = 26, npages, ret, i; - PDBG("%s ib_pd %p\n", __func__, pd); + pr_debug("%s ib_pd %p\n", __func__, pd); /* * T3 only supports 32 bits of size. */ if (sizeof(phys_addr_t) > 4) { - pr_warn_once(MOD "Cannot support dma_mrs on this platform.\n"); + pr_warn_once("Cannot support dma_mrs on this platform\n"); return ERR_PTR(-ENOTSUPP); } @@ -518,8 +515,8 @@ static struct ib_mr *iwch_get_dma_mr(struct ib_pd *pd, int acc) for (i = 0; i < npages; i++) page_list[i] = cpu_to_be64((u64)i << shift); - PDBG("%s mask 0x%llx shift %d len %lld pbl_size %d\n", - __func__, mask, shift, total_size, npages); + pr_debug("%s mask 0x%llx shift %d len %lld pbl_size %d\n", + __func__, mask, shift, total_size, npages); ret = iwch_alloc_pbl(mhp, npages); if (ret) { @@ -567,7 +564,7 @@ static struct ib_mr *iwch_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, struct iwch_mr *mhp; struct iwch_reg_user_mr_resp uresp; struct scatterlist *sg; - PDBG("%s ib_pd %p\n", __func__, pd); + pr_debug("%s ib_pd %p\n", __func__, pd); php = to_iwch_pd(pd); rhp = php->rhp; @@ -584,7 +581,7 @@ static struct ib_mr *iwch_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, return ERR_PTR(err); } - shift = ffs(mhp->umem->page_size) - 1; + shift = mhp->umem->page_shift; n = mhp->umem->nmap; @@ -604,7 +601,7 @@ static struct ib_mr *iwch_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, len = sg_dma_len(sg) >> shift; for (k = 0; k < len; ++k) { pages[i++] = cpu_to_be64(sg_dma_address(sg) + - mhp->umem->page_size * k); + (k << shift)); if (i == PAGE_SIZE / sizeof *pages) { err = iwch_write_pbl(mhp, pages, i, n); if (err) @@ -637,8 +634,8 @@ pbl_done: if (udata && !t3a_device(rhp)) { uresp.pbl_addr = (mhp->attr.pbl_addr - rhp->rdev.rnic_info.pbl_base) >> 3; - PDBG("%s user resp pbl_addr 0x%x\n", __func__, - uresp.pbl_addr); + pr_debug("%s user resp pbl_addr 0x%x\n", __func__, + uresp.pbl_addr); if (ib_copy_to_udata(udata, &uresp, sizeof (uresp))) { iwch_dereg_mr(&mhp->ibmr); @@ -692,7 +689,7 @@ static struct ib_mw *iwch_alloc_mw(struct ib_pd *pd, enum ib_mw_type type, kfree(mhp); return ERR_PTR(-ENOMEM); } - PDBG("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); + pr_debug("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); return &(mhp->ibmw); } @@ -707,7 +704,7 @@ static int iwch_dealloc_mw(struct ib_mw *mw) mmid = (mw->rkey) >> 8; cxio_deallocate_window(&rhp->rdev, mhp->attr.stag); remove_handle(rhp, &rhp->mmidr, mmid); - PDBG("%s ib_mw %p mmid 0x%x ptr %p\n", __func__, mw, mmid, mhp); + pr_debug("%s ib_mw %p mmid 0x%x ptr %p\n", __func__, mw, mmid, mhp); kfree(mhp); return 0; } @@ -757,7 +754,7 @@ static struct ib_mr *iwch_alloc_mr(struct ib_pd *pd, if (insert_handle(rhp, &rhp->mmidr, mhp, mmid)) goto err3; - PDBG("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); + pr_debug("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); return &(mhp->ibmr); err3: cxio_dereg_mem(&rhp->rdev, stag, mhp->attr.pbl_size, @@ -818,8 +815,8 @@ static int iwch_destroy_qp(struct ib_qp *ib_qp) cxio_destroy_qp(&rhp->rdev, &qhp->wq, ucontext ? &ucontext->uctx : &rhp->rdev.uctx); - PDBG("%s ib_qp %p qpid 0x%0x qhp %p\n", __func__, - ib_qp, qhp->wq.qpid, qhp); + pr_debug("%s ib_qp %p qpid 0x%0x qhp %p\n", __func__, + ib_qp, qhp->wq.qpid, qhp); kfree(qhp); return 0; } @@ -837,7 +834,7 @@ static struct ib_qp *iwch_create_qp(struct ib_pd *pd, int wqsize, sqsize, rqsize; struct iwch_ucontext *ucontext; - PDBG("%s ib_pd %p\n", __func__, pd); + pr_debug("%s ib_pd %p\n", __func__, pd); if (attrs->qp_type != IB_QPT_RC) return ERR_PTR(-EINVAL); php = to_iwch_pd(pd); @@ -878,8 +875,8 @@ static struct ib_qp *iwch_create_qp(struct ib_pd *pd, if (!ucontext && wqsize < (rqsize + (2 * sqsize))) wqsize = roundup_pow_of_two(rqsize + roundup_pow_of_two(attrs->cap.max_send_wr * 2)); - PDBG("%s wqsize %d sqsize %d rqsize %d\n", __func__, - wqsize, sqsize, rqsize); + pr_debug("%s wqsize %d sqsize %d rqsize %d\n", __func__, + wqsize, sqsize, rqsize); qhp = kzalloc(sizeof(*qhp), GFP_KERNEL); if (!qhp) return ERR_PTR(-ENOMEM); @@ -974,11 +971,10 @@ static struct ib_qp *iwch_create_qp(struct ib_pd *pd, } qhp->ibqp.qp_num = qhp->wq.qpid; init_timer(&(qhp->timer)); - PDBG("%s sq_num_entries %d, rq_num_entries %d " - "qpid 0x%0x qhp %p dma_addr 0x%llx size %d rq_addr 0x%x\n", - __func__, qhp->attr.sq_num_entries, qhp->attr.rq_num_entries, - qhp->wq.qpid, qhp, (unsigned long long) qhp->wq.dma_addr, - 1 << qhp->wq.size_log2, qhp->wq.rq_addr); + pr_debug("%s sq_num_entries %d, rq_num_entries %d qpid 0x%0x qhp %p dma_addr 0x%llx size %d rq_addr 0x%x\n", + __func__, qhp->attr.sq_num_entries, qhp->attr.rq_num_entries, + qhp->wq.qpid, qhp, (unsigned long long)qhp->wq.dma_addr, + 1 << qhp->wq.size_log2, qhp->wq.rq_addr); return &qhp->ibqp; } @@ -990,7 +986,7 @@ static int iwch_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, enum iwch_qp_attr_mask mask = 0; struct iwch_qp_attributes attrs; - PDBG("%s ib_qp %p\n", __func__, ibqp); + pr_debug("%s ib_qp %p\n", __func__, ibqp); /* iwarp does not support the RTR state */ if ((attr_mask & IB_QP_STATE) && (attr->qp_state == IB_QPS_RTR)) @@ -1023,20 +1019,20 @@ static int iwch_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, void iwch_qp_add_ref(struct ib_qp *qp) { - PDBG("%s ib_qp %p\n", __func__, qp); + pr_debug("%s ib_qp %p\n", __func__, qp); atomic_inc(&(to_iwch_qp(qp)->refcnt)); } void iwch_qp_rem_ref(struct ib_qp *qp) { - PDBG("%s ib_qp %p\n", __func__, qp); + pr_debug("%s ib_qp %p\n", __func__, qp); if (atomic_dec_and_test(&(to_iwch_qp(qp)->refcnt))) wake_up(&(to_iwch_qp(qp)->wait)); } static struct ib_qp *iwch_get_qp(struct ib_device *dev, int qpn) { - PDBG("%s ib_dev %p qpn 0x%x\n", __func__, dev, qpn); + pr_debug("%s ib_dev %p qpn 0x%x\n", __func__, dev, qpn); return (struct ib_qp *)get_qhp(to_iwch_dev(dev), qpn); } @@ -1044,7 +1040,7 @@ static struct ib_qp *iwch_get_qp(struct ib_device *dev, int qpn) static int iwch_query_pkey(struct ib_device *ibdev, u8 port, u16 index, u16 * pkey) { - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); *pkey = 0; return 0; } @@ -1054,8 +1050,8 @@ static int iwch_query_gid(struct ib_device *ibdev, u8 port, { struct iwch_dev *dev; - PDBG("%s ibdev %p, port %d, index %d, gid %p\n", - __func__, ibdev, port, index, gid); + pr_debug("%s ibdev %p, port %d, index %d, gid %p\n", + __func__, ibdev, port, index, gid); dev = to_iwch_dev(ibdev); BUG_ON(port == 0 || port > 2); memset(&(gid->raw[0]), 0, sizeof(gid->raw)); @@ -1090,7 +1086,7 @@ static int iwch_query_device(struct ib_device *ibdev, struct ib_device_attr *pro struct iwch_dev *dev; - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); if (uhw->inlen || uhw->outlen) return -EINVAL; @@ -1128,7 +1124,7 @@ static int iwch_query_port(struct ib_device *ibdev, struct net_device *netdev; struct in_device *inetdev; - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); dev = to_iwch_dev(ibdev); netdev = dev->rdev.port_info.lldevs[port-1]; @@ -1171,7 +1167,7 @@ static ssize_t show_rev(struct device *dev, struct device_attribute *attr, { struct iwch_dev *iwch_dev = container_of(dev, struct iwch_dev, ibdev.dev); - PDBG("%s dev 0x%p\n", __func__, dev); + pr_debug("%s dev 0x%p\n", __func__, dev); return sprintf(buf, "%d\n", iwch_dev->rdev.t3cdev_p->type); } @@ -1183,7 +1179,7 @@ static ssize_t show_hca(struct device *dev, struct device_attribute *attr, struct ethtool_drvinfo info; struct net_device *lldev = iwch_dev->rdev.t3cdev_p->lldev; - PDBG("%s dev 0x%p\n", __func__, dev); + pr_debug("%s dev 0x%p\n", __func__, dev); lldev->ethtool_ops->get_drvinfo(lldev, &info); return sprintf(buf, "%s\n", info.driver); } @@ -1193,7 +1189,7 @@ static ssize_t show_board(struct device *dev, struct device_attribute *attr, { struct iwch_dev *iwch_dev = container_of(dev, struct iwch_dev, ibdev.dev); - PDBG("%s dev 0x%p\n", __func__, dev); + pr_debug("%s dev 0x%p\n", __func__, dev); return sprintf(buf, "%x.%x\n", iwch_dev->rdev.rnic_info.pdev->vendor, iwch_dev->rdev.rnic_info.pdev->device); } @@ -1278,7 +1274,7 @@ static int iwch_get_mib(struct ib_device *ibdev, struct rdma_hw_stats *stats, if (port != 0 || !stats) return -ENOSYS; - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); dev = to_iwch_dev(ibdev); ret = dev->rdev.t3cdev_p->ctl(dev->rdev.t3cdev_p, RDMA_GET_MIB, &m); if (ret) @@ -1348,7 +1344,7 @@ static void get_dev_fw_ver_str(struct ib_device *ibdev, char *str, struct ethtool_drvinfo info; struct net_device *lldev = iwch_dev->rdev.t3cdev_p->lldev; - PDBG("%s dev 0x%p\n", __func__, iwch_dev); + pr_debug("%s dev 0x%p\n", __func__, iwch_dev); lldev->ethtool_ops->get_drvinfo(lldev, &info); snprintf(str, str_len, "%s", info.fw_version); } @@ -1358,7 +1354,7 @@ int iwch_register_device(struct iwch_dev *dev) int ret; int i; - PDBG("%s iwch_dev %p\n", __func__, dev); + pr_debug("%s iwch_dev %p\n", __func__, dev); strlcpy(dev->ibdev.name, "cxgb3_%d", IB_DEVICE_NAME_MAX); memset(&dev->ibdev.node_guid, 0, sizeof(dev->ibdev.node_guid)); memcpy(&dev->ibdev.node_guid, dev->rdev.t3cdev_p->lldev->dev_addr, 6); @@ -1469,7 +1465,7 @@ void iwch_unregister_device(struct iwch_dev *dev) { int i; - PDBG("%s iwch_dev %p\n", __func__, dev); + pr_debug("%s iwch_dev %p\n", __func__, dev); for (i = 0; i < ARRAY_SIZE(iwch_class_attributes); ++i) device_remove_file(&dev->ibdev.dev, iwch_class_attributes[i]); diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.h b/drivers/infiniband/hw/cxgb3/iwch_provider.h index 252c464a09f6..9e216edec4c0 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_provider.h +++ b/drivers/infiniband/hw/cxgb3/iwch_provider.h @@ -217,8 +217,9 @@ static inline struct iwch_mm_entry *remove_mmap(struct iwch_ucontext *ucontext, if (mm->key == key && mm->len == len) { list_del_init(&mm->entry); spin_unlock(&ucontext->mmap_lock); - PDBG("%s key 0x%x addr 0x%llx len %d\n", __func__, - key, (unsigned long long) mm->addr, mm->len); + pr_debug("%s key 0x%x addr 0x%llx len %d\n", + __func__, key, + (unsigned long long)mm->addr, mm->len); return mm; } } @@ -230,8 +231,8 @@ static inline void insert_mmap(struct iwch_ucontext *ucontext, struct iwch_mm_entry *mm) { spin_lock(&ucontext->mmap_lock); - PDBG("%s key 0x%x addr 0x%llx len %d\n", __func__, - mm->key, (unsigned long long) mm->addr, mm->len); + pr_debug("%s key 0x%x addr 0x%llx len %d\n", + __func__, mm->key, (unsigned long long)mm->addr, mm->len); list_add_tail(&mm->entry, &ucontext->mmaps); spin_unlock(&ucontext->mmap_lock); } diff --git a/drivers/infiniband/hw/cxgb3/iwch_qp.c b/drivers/infiniband/hw/cxgb3/iwch_qp.c index a9194db7f9b8..ba6d5d281b03 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_qp.c +++ b/drivers/infiniband/hw/cxgb3/iwch_qp.c @@ -208,30 +208,30 @@ static int iwch_sgl2pbl_map(struct iwch_dev *rhp, struct ib_sge *sg_list, mhp = get_mhp(rhp, (sg_list[i].lkey) >> 8); if (!mhp) { - PDBG("%s %d\n", __func__, __LINE__); + pr_debug("%s %d\n", __func__, __LINE__); return -EIO; } if (!mhp->attr.state) { - PDBG("%s %d\n", __func__, __LINE__); + pr_debug("%s %d\n", __func__, __LINE__); return -EIO; } if (mhp->attr.zbva) { - PDBG("%s %d\n", __func__, __LINE__); + pr_debug("%s %d\n", __func__, __LINE__); return -EIO; } if (sg_list[i].addr < mhp->attr.va_fbo) { - PDBG("%s %d\n", __func__, __LINE__); + pr_debug("%s %d\n", __func__, __LINE__); return -EINVAL; } if (sg_list[i].addr + ((u64) sg_list[i].length) < sg_list[i].addr) { - PDBG("%s %d\n", __func__, __LINE__); + pr_debug("%s %d\n", __func__, __LINE__); return -EINVAL; } if (sg_list[i].addr + ((u64) sg_list[i].length) > mhp->attr.va_fbo + ((u64) mhp->attr.len)) { - PDBG("%s %d\n", __func__, __LINE__); + pr_debug("%s %d\n", __func__, __LINE__); return -EINVAL; } offset = sg_list[i].addr - mhp->attr.va_fbo; @@ -427,8 +427,8 @@ int iwch_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, err = build_inv_stag(wqe, wr, &t3_wr_flit_cnt); break; default: - PDBG("%s post of type=%d TBD!\n", __func__, - wr->opcode); + pr_debug("%s post of type=%d TBD!\n", __func__, + wr->opcode); err = -EINVAL; } if (err) @@ -444,10 +444,10 @@ int iwch_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, Q_GENBIT(qhp->wq.wptr, qhp->wq.size_log2), 0, t3_wr_flit_cnt, (wr_cnt == 1) ? T3_SOPEOP : T3_SOP); - PDBG("%s cookie 0x%llx wq idx 0x%x swsq idx %ld opcode %d\n", - __func__, (unsigned long long) wr->wr_id, idx, - Q_PTR2IDX(qhp->wq.sq_wptr, qhp->wq.sq_size_log2), - sqp->opcode); + pr_debug("%s cookie 0x%llx wq idx 0x%x swsq idx %ld opcode %d\n", + __func__, (unsigned long long)wr->wr_id, idx, + Q_PTR2IDX(qhp->wq.sq_wptr, qhp->wq.sq_size_log2), + sqp->opcode); wr = wr->next; num_wrs--; qhp->wq.wptr += wr_cnt; @@ -508,9 +508,9 @@ int iwch_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr, build_fw_riwrh((void *) wqe, T3_WR_RCV, T3_COMPLETION_FLAG, Q_GENBIT(qhp->wq.wptr, qhp->wq.size_log2), 0, sizeof(struct t3_receive_wr) >> 3, T3_SOPEOP); - PDBG("%s cookie 0x%llx idx 0x%x rq_wptr 0x%x rw_rptr 0x%x " - "wqe %p \n", __func__, (unsigned long long) wr->wr_id, - idx, qhp->wq.rq_wptr, qhp->wq.rq_rptr, wqe); + pr_debug("%s cookie 0x%llx idx 0x%x rq_wptr 0x%x rw_rptr 0x%x wqe %p\n", + __func__, (unsigned long long)wr->wr_id, + idx, qhp->wq.rq_wptr, qhp->wq.rq_rptr, wqe); ++(qhp->wq.rq_wptr); ++(qhp->wq.wptr); wr = wr->next; @@ -664,10 +664,10 @@ int iwch_post_zb_read(struct iwch_ep *ep) struct sk_buff *skb; u8 flit_cnt = sizeof(struct t3_rdma_read_wr) >> 3; - PDBG("%s enter\n", __func__); + pr_debug("%s enter\n", __func__); skb = alloc_skb(40, GFP_KERNEL); if (!skb) { - printk(KERN_ERR "%s cannot send zb_read!!\n", __func__); + pr_err("%s cannot send zb_read!!\n", __func__); return -ENOMEM; } wqe = (union t3_wr *)skb_put(skb, sizeof(struct t3_rdma_read_wr)); @@ -696,10 +696,10 @@ int iwch_post_terminate(struct iwch_qp *qhp, struct respQ_msg_t *rsp_msg) struct terminate_message *term; struct sk_buff *skb; - PDBG("%s %d\n", __func__, __LINE__); + pr_debug("%s %d\n", __func__, __LINE__); skb = alloc_skb(40, GFP_ATOMIC); if (!skb) { - printk(KERN_ERR "%s cannot send TERMINATE!\n", __func__); + pr_err("%s cannot send TERMINATE!\n", __func__); return -ENOMEM; } wqe = (union t3_wr *)skb_put(skb, 40); @@ -729,7 +729,7 @@ static void __flush_qp(struct iwch_qp *qhp, struct iwch_cq *rchp, int flushed; - PDBG("%s qhp %p rchp %p schp %p\n", __func__, qhp, rchp, schp); + pr_debug("%s qhp %p rchp %p schp %p\n", __func__, qhp, rchp, schp); /* take a ref on the qhp since we must release the lock */ atomic_inc(&qhp->refcnt); spin_unlock(&qhp->lock); @@ -807,7 +807,7 @@ u16 iwch_rqes_posted(struct iwch_qp *qhp) count++; wqe++; } - PDBG("%s qhp %p count %u\n", __func__, qhp, count); + pr_debug("%s qhp %p count %u\n", __func__, qhp, count); return count; } @@ -854,12 +854,12 @@ static int rdma_init(struct iwch_dev *rhp, struct iwch_qp *qhp, } else init_attr.rtr_type = 0; init_attr.irs = qhp->ep->rcv_seq; - PDBG("%s init_attr.rq_addr 0x%x init_attr.rq_size = %d " - "flags 0x%x qpcaps 0x%x\n", __func__, - init_attr.rq_addr, init_attr.rq_size, - init_attr.flags, init_attr.qpcaps); + pr_debug("%s init_attr.rq_addr 0x%x init_attr.rq_size = %d flags 0x%x qpcaps 0x%x\n", + __func__, + init_attr.rq_addr, init_attr.rq_size, + init_attr.flags, init_attr.qpcaps); ret = cxio_rdma_init(&rhp->rdev, &init_attr); - PDBG("%s ret %d\n", __func__, ret); + pr_debug("%s ret %d\n", __func__, ret); return ret; } @@ -877,9 +877,9 @@ int iwch_modify_qp(struct iwch_dev *rhp, struct iwch_qp *qhp, int free = 0; struct iwch_ep *ep = NULL; - PDBG("%s qhp %p qpid 0x%x ep %p state %d -> %d\n", __func__, - qhp, qhp->wq.qpid, qhp->ep, qhp->attr.state, - (mask & IWCH_QP_ATTR_NEXT_STATE) ? attrs->next_state : -1); + pr_debug("%s qhp %p qpid 0x%x ep %p state %d -> %d\n", __func__, + qhp, qhp->wq.qpid, qhp->ep, qhp->attr.state, + (mask & IWCH_QP_ATTR_NEXT_STATE) ? attrs->next_state : -1); spin_lock_irqsave(&qhp->lock, flag); @@ -1034,16 +1034,15 @@ int iwch_modify_qp(struct iwch_dev *rhp, struct iwch_qp *qhp, goto err; break; default: - printk(KERN_ERR "%s in a bad state %d\n", - __func__, qhp->attr.state); + pr_err("%s in a bad state %d\n", __func__, qhp->attr.state); ret = -EINVAL; goto err; break; } goto out; err: - PDBG("%s disassociating ep %p qpid 0x%x\n", __func__, qhp->ep, - qhp->wq.qpid); + pr_debug("%s disassociating ep %p qpid 0x%x\n", __func__, qhp->ep, + qhp->wq.qpid); /* disassociate the LLP connection */ qhp->attr.llp_stream_handle = NULL; @@ -1077,6 +1076,6 @@ out: if (free) put_ep(&ep->com); - PDBG("%s exit state %d\n", __func__, qhp->attr.state); + pr_debug("%s exit state %d\n", __func__, qhp->attr.state); return ret; } diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index 03a1b0e64fc3..b6fe45924c6e 100644 --- a/drivers/infiniband/hw/cxgb4/cm.c +++ b/drivers/infiniband/hw/cxgb4/cm.c @@ -101,7 +101,7 @@ MODULE_PARM_DESC(enable_tcp_window_scaling, int c4iw_debug; module_param(c4iw_debug, int, 0644); -MODULE_PARM_DESC(c4iw_debug, "Enable debug logging (default=0)"); +MODULE_PARM_DESC(c4iw_debug, "obsolete"); static int peer2peer = 1; module_param(peer2peer, int, 0644); @@ -180,7 +180,7 @@ static void ref_qp(struct c4iw_ep *ep) static void start_ep_timer(struct c4iw_ep *ep) { - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); if (timer_pending(&ep->timer)) { pr_err("%s timer already started! ep %p\n", __func__, ep); @@ -196,7 +196,7 @@ static void start_ep_timer(struct c4iw_ep *ep) static int stop_ep_timer(struct c4iw_ep *ep) { - PDBG("%s ep %p stopping\n", __func__, ep); + pr_debug("%s ep %p stopping\n", __func__, ep); del_timer_sync(&ep->timer); if (!test_and_set_bit(TIMEOUT, &ep->com.flags)) { c4iw_put_ep(&ep->com); @@ -212,7 +212,7 @@ static int c4iw_l2t_send(struct c4iw_rdev *rdev, struct sk_buff *skb, if (c4iw_fatal_error(rdev)) { kfree_skb(skb); - PDBG("%s - device in error state - dropping\n", __func__); + pr_debug("%s - device in error state - dropping\n", __func__); return -EIO; } error = cxgb4_l2t_send(rdev->lldi.ports[0], skb, l2e); @@ -229,7 +229,7 @@ int c4iw_ofld_send(struct c4iw_rdev *rdev, struct sk_buff *skb) if (c4iw_fatal_error(rdev)) { kfree_skb(skb); - PDBG("%s - device in error state - dropping\n", __func__); + pr_debug("%s - device in error state - dropping\n", __func__); return -EIO; } error = cxgb4_ofld_send(rdev->lldi.ports[0], skb); @@ -263,10 +263,10 @@ static void set_emss(struct c4iw_ep *ep, u16 opt) if (ep->emss < 128) ep->emss = 128; if (ep->emss & 7) - PDBG("Warning: misaligned mtu idx %u mss %u emss=%u\n", - TCPOPT_MSS_G(opt), ep->mss, ep->emss); - PDBG("%s mss_idx %u mss %u emss=%u\n", __func__, TCPOPT_MSS_G(opt), - ep->mss, ep->emss); + pr_debug("Warning: misaligned mtu idx %u mss %u emss=%u\n", + TCPOPT_MSS_G(opt), ep->mss, ep->emss); + pr_debug("%s mss_idx %u mss %u emss=%u\n", __func__, TCPOPT_MSS_G(opt), + ep->mss, ep->emss); } static enum c4iw_ep_state state_read(struct c4iw_ep_common *epc) @@ -287,7 +287,7 @@ static void __state_set(struct c4iw_ep_common *epc, enum c4iw_ep_state new) static void state_set(struct c4iw_ep_common *epc, enum c4iw_ep_state new) { mutex_lock(&epc->mutex); - PDBG("%s - %s -> %s\n", __func__, states[epc->state], states[new]); + pr_debug("%s - %s -> %s\n", __func__, states[epc->state], states[new]); __state_set(epc, new); mutex_unlock(&epc->mutex); return; @@ -322,7 +322,7 @@ static void *alloc_ep(int size, gfp_t gfp) mutex_init(&epc->mutex); c4iw_init_wr_wait(&epc->wr_wait); } - PDBG("%s alloc ep %p\n", __func__, epc); + pr_debug("%s alloc ep %p\n", __func__, epc); return epc; } @@ -384,7 +384,7 @@ void _c4iw_free_ep(struct kref *kref) struct c4iw_ep *ep; ep = container_of(kref, struct c4iw_ep, com.kref); - PDBG("%s ep %p state %s\n", __func__, ep, states[ep->com.state]); + pr_debug("%s ep %p state %s\n", __func__, ep, states[ep->com.state]); if (test_bit(QP_REFERENCED, &ep->com.flags)) deref_qp(ep); if (test_bit(RELEASE_RESOURCES, &ep->com.flags)) { @@ -467,7 +467,7 @@ static struct net_device *get_real_dev(struct net_device *egress_dev) static void arp_failure_discard(void *handle, struct sk_buff *skb) { - pr_err(MOD "ARP failure\n"); + pr_err("ARP failure\n"); kfree_skb(skb); } @@ -528,7 +528,7 @@ static void pass_accept_rpl_arp_failure(void *handle, struct sk_buff *skb) { struct c4iw_ep *ep = handle; - pr_err(MOD "ARP failure during accept - tid %u -dropping connection\n", + pr_err("ARP failure during accept - tid %u - dropping connection\n", ep->hwtid); __state_set(&ep->com, DEAD); @@ -542,7 +542,7 @@ static void act_open_req_arp_failure(void *handle, struct sk_buff *skb) { struct c4iw_ep *ep = handle; - printk(KERN_ERR MOD "ARP failure during connect\n"); + pr_err("ARP failure during connect\n"); connect_reply_upcall(ep, -EHOSTUNREACH); __state_set(&ep->com, DEAD); if (ep->com.remote_addr.ss_family == AF_INET6) { @@ -567,7 +567,7 @@ static void abort_arp_failure(void *handle, struct sk_buff *skb) struct c4iw_rdev *rdev = &ep->com.dev->rdev; struct cpl_abort_req *req = cplhdr(skb); - PDBG("%s rdev %p\n", __func__, rdev); + pr_debug("%s rdev %p\n", __func__, rdev); req->cmd = CPL_ABORT_NO_RST; ret = c4iw_ofld_send(rdev, skb); if (ret) { @@ -642,7 +642,7 @@ static int send_halfclose(struct c4iw_ep *ep) struct sk_buff *skb = skb_dequeue(&ep->com.ep_skb_list); u32 wrlen = roundup(sizeof(struct cpl_close_con_req), 16); - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); if (WARN_ON(!skb)) return -ENOMEM; @@ -657,7 +657,7 @@ static int send_abort(struct c4iw_ep *ep) u32 wrlen = roundup(sizeof(struct cpl_abort_req), 16); struct sk_buff *req_skb = skb_dequeue(&ep->com.ep_skb_list); - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); if (WARN_ON(!req_skb)) return -ENOMEM; @@ -720,12 +720,11 @@ static int send_connect(struct c4iw_ep *ep) roundup(sizev4, 16) : roundup(sizev6, 16); - PDBG("%s ep %p atid %u\n", __func__, ep, ep->atid); + pr_debug("%s ep %p atid %u\n", __func__, ep, ep->atid); skb = get_skb(NULL, wrlen, GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "%s - failed to alloc skb.\n", - __func__); + pr_err("%s - failed to alloc skb\n", __func__); return -ENOMEM; } set_wr_txq(skb, CPL_PRIORITY_SETUP, ep->ctrlq_idx); @@ -822,13 +821,13 @@ static int send_connect(struct c4iw_ep *ep) t5req->params = cpu_to_be64(FILTER_TUPLE_V(params)); t5req->rsvd = cpu_to_be32(isn); - PDBG("%s snd_isn %u\n", __func__, t5req->rsvd); + pr_debug("%s snd_isn %u\n", __func__, t5req->rsvd); t5req->opt2 = cpu_to_be32(opt2); } else { t6req->params = cpu_to_be64(FILTER_TUPLE_V(params)); t6req->rsvd = cpu_to_be32(isn); - PDBG("%s snd_isn %u\n", __func__, t6req->rsvd); + pr_debug("%s snd_isn %u\n", __func__, t6req->rsvd); t6req->opt2 = cpu_to_be32(opt2); } } @@ -877,13 +876,13 @@ static int send_connect(struct c4iw_ep *ep) t5req6->params = cpu_to_be64(FILTER_TUPLE_V(params)); t5req6->rsvd = cpu_to_be32(isn); - PDBG("%s snd_isn %u\n", __func__, t5req6->rsvd); + pr_debug("%s snd_isn %u\n", __func__, t5req6->rsvd); t5req6->opt2 = cpu_to_be32(opt2); } else { t6req6->params = cpu_to_be64(FILTER_TUPLE_V(params)); t6req6->rsvd = cpu_to_be32(isn); - PDBG("%s snd_isn %u\n", __func__, t6req6->rsvd); + pr_debug("%s snd_isn %u\n", __func__, t6req6->rsvd); t6req6->opt2 = cpu_to_be32(opt2); } @@ -907,7 +906,8 @@ static int send_mpa_req(struct c4iw_ep *ep, struct sk_buff *skb, struct mpa_message *mpa; struct mpa_v2_conn_params mpa_v2_params; - PDBG("%s ep %p tid %u pd_len %d\n", __func__, ep, ep->hwtid, ep->plen); + pr_debug("%s ep %p tid %u pd_len %d\n", + __func__, ep, ep->hwtid, ep->plen); BUG_ON(skb_cloned(skb)); @@ -961,8 +961,8 @@ static int send_mpa_req(struct c4iw_ep *ep, struct sk_buff *skb, if (mpa_rev_to_use == 2) { mpa->private_data_size = htons(ntohs(mpa->private_data_size) + sizeof (struct mpa_v2_conn_params)); - PDBG("%s initiator ird %u ord %u\n", __func__, ep->ird, - ep->ord); + pr_debug("%s initiator ird %u ord %u\n", __func__, ep->ird, + ep->ord); mpa_v2_params.ird = htons((u16)ep->ird); mpa_v2_params.ord = htons((u16)ep->ord); @@ -1014,7 +1014,8 @@ static int send_mpa_reject(struct c4iw_ep *ep, const void *pdata, u8 plen) struct sk_buff *skb; struct mpa_v2_conn_params mpa_v2_params; - PDBG("%s ep %p tid %u pd_len %d\n", __func__, ep, ep->hwtid, ep->plen); + pr_debug("%s ep %p tid %u pd_len %d\n", + __func__, ep, ep->hwtid, ep->plen); mpalen = sizeof(*mpa) + plen; if (ep->mpa_attr.version == 2 && ep->mpa_attr.enhanced_rdma_conn) @@ -1023,7 +1024,7 @@ static int send_mpa_reject(struct c4iw_ep *ep, const void *pdata, u8 plen) skb = get_skb(NULL, wrlen, GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "%s - cannot alloc skb!\n", __func__); + pr_err("%s - cannot alloc skb!\n", __func__); return -ENOMEM; } set_wr_txq(skb, CPL_PRIORITY_DATA, ep->txq_idx); @@ -1094,7 +1095,8 @@ static int send_mpa_reply(struct c4iw_ep *ep, const void *pdata, u8 plen) struct sk_buff *skb; struct mpa_v2_conn_params mpa_v2_params; - PDBG("%s ep %p tid %u pd_len %d\n", __func__, ep, ep->hwtid, ep->plen); + pr_debug("%s ep %p tid %u pd_len %d\n", + __func__, ep, ep->hwtid, ep->plen); mpalen = sizeof(*mpa) + plen; if (ep->mpa_attr.version == 2 && ep->mpa_attr.enhanced_rdma_conn) @@ -1103,7 +1105,7 @@ static int send_mpa_reply(struct c4iw_ep *ep, const void *pdata, u8 plen) skb = get_skb(NULL, wrlen, GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "%s - cannot alloc skb!\n", __func__); + pr_err("%s - cannot alloc skb!\n", __func__); return -ENOMEM; } set_wr_txq(skb, CPL_PRIORITY_DATA, ep->txq_idx); @@ -1185,8 +1187,8 @@ static int act_establish(struct c4iw_dev *dev, struct sk_buff *skb) ep = lookup_atid(t, atid); - PDBG("%s ep %p tid %u snd_isn %u rcv_isn %u\n", __func__, ep, tid, - be32_to_cpu(req->snd_isn), be32_to_cpu(req->rcv_isn)); + pr_debug("%s ep %p tid %u snd_isn %u rcv_isn %u\n", __func__, ep, tid, + be32_to_cpu(req->snd_isn), be32_to_cpu(req->rcv_isn)); mutex_lock(&ep->com.mutex); dst_confirm(ep->dst); @@ -1229,13 +1231,13 @@ static void close_complete_upcall(struct c4iw_ep *ep, int status) { struct iw_cm_event event; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_CLOSE; event.status = status; if (ep->com.cm_id) { - PDBG("close complete delivered ep %p cm_id %p tid %u\n", - ep, ep->com.cm_id, ep->hwtid); + pr_debug("close complete delivered ep %p cm_id %p tid %u\n", + ep, ep->com.cm_id, ep->hwtid); ep->com.cm_id->event_handler(ep->com.cm_id, &event); deref_cm_id(&ep->com); set_bit(CLOSE_UPCALL, &ep->com.history); @@ -1246,12 +1248,12 @@ static void peer_close_upcall(struct c4iw_ep *ep) { struct iw_cm_event event; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_DISCONNECT; if (ep->com.cm_id) { - PDBG("peer close delivered ep %p cm_id %p tid %u\n", - ep, ep->com.cm_id, ep->hwtid); + pr_debug("peer close delivered ep %p cm_id %p tid %u\n", + ep, ep->com.cm_id, ep->hwtid); ep->com.cm_id->event_handler(ep->com.cm_id, &event); set_bit(DISCONN_UPCALL, &ep->com.history); } @@ -1261,13 +1263,13 @@ static void peer_abort_upcall(struct c4iw_ep *ep) { struct iw_cm_event event; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_CLOSE; event.status = -ECONNRESET; if (ep->com.cm_id) { - PDBG("abort delivered ep %p cm_id %p tid %u\n", ep, - ep->com.cm_id, ep->hwtid); + pr_debug("abort delivered ep %p cm_id %p tid %u\n", ep, + ep->com.cm_id, ep->hwtid); ep->com.cm_id->event_handler(ep->com.cm_id, &event); deref_cm_id(&ep->com); set_bit(ABORT_UPCALL, &ep->com.history); @@ -1278,7 +1280,8 @@ static void connect_reply_upcall(struct c4iw_ep *ep, int status) { struct iw_cm_event event; - PDBG("%s ep %p tid %u status %d\n", __func__, ep, ep->hwtid, status); + pr_debug("%s ep %p tid %u status %d\n", + __func__, ep, ep->hwtid, status); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_CONNECT_REPLY; event.status = status; @@ -1307,8 +1310,8 @@ static void connect_reply_upcall(struct c4iw_ep *ep, int status) } } - PDBG("%s ep %p tid %u status %d\n", __func__, ep, - ep->hwtid, status); + pr_debug("%s ep %p tid %u status %d\n", __func__, ep, + ep->hwtid, status); set_bit(CONN_RPL_UPCALL, &ep->com.history); ep->com.cm_id->event_handler(ep->com.cm_id, &event); @@ -1321,7 +1324,7 @@ static int connect_request_upcall(struct c4iw_ep *ep) struct iw_cm_event event; int ret; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_CONNECT_REQUEST; memcpy(&event.local_addr, &ep->com.local_addr, @@ -1358,13 +1361,13 @@ static void established_upcall(struct c4iw_ep *ep) { struct iw_cm_event event; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_ESTABLISHED; event.ird = ep->ord; event.ord = ep->ird; if (ep->com.cm_id) { - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); ep->com.cm_id->event_handler(ep->com.cm_id, &event); set_bit(ESTAB_UPCALL, &ep->com.history); } @@ -1376,10 +1379,11 @@ static int update_rx_credits(struct c4iw_ep *ep, u32 credits) u32 wrlen = roundup(sizeof(struct cpl_rx_data_ack), 16); u32 credit_dack; - PDBG("%s ep %p tid %u credits %u\n", __func__, ep, ep->hwtid, credits); + pr_debug("%s ep %p tid %u credits %u\n", + __func__, ep, ep->hwtid, credits); skb = get_skb(NULL, wrlen, GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "update_rx_credits - cannot alloc skb!\n"); + pr_err("update_rx_credits - cannot alloc skb!\n"); return 0; } @@ -1427,7 +1431,7 @@ static int process_mpa_reply(struct c4iw_ep *ep, struct sk_buff *skb) int err; int disconnect = 0; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); /* * If we get more than the supported amount of private data @@ -1454,8 +1458,8 @@ static int process_mpa_reply(struct c4iw_ep *ep, struct sk_buff *skb) /* Validate MPA header. */ if (mpa->revision > mpa_rev) { - printk(KERN_ERR MOD "%s MPA version mismatch. Local = %d," - " Received = %d\n", __func__, mpa_rev, mpa->revision); + pr_err("%s MPA version mismatch. Local = %d, Received = %d\n", + __func__, mpa_rev, mpa->revision); err = -EPROTO; goto err_stop_timer; } @@ -1525,8 +1529,9 @@ static int process_mpa_reply(struct c4iw_ep *ep, struct sk_buff *skb) MPA_V2_IRD_ORD_MASK; resp_ord = ntohs(mpa_v2_params->ord) & MPA_V2_IRD_ORD_MASK; - PDBG("%s responder ird %u ord %u ep ird %u ord %u\n", - __func__, resp_ird, resp_ord, ep->ird, ep->ord); + pr_debug("%s responder ird %u ord %u ep ird %u ord %u\n", + __func__, + resp_ird, resp_ord, ep->ird, ep->ord); /* * This is a double-check. Ideally, below checks are @@ -1570,12 +1575,11 @@ static int process_mpa_reply(struct c4iw_ep *ep, struct sk_buff *skb) if (peer2peer) ep->mpa_attr.p2p_type = p2p_type; - PDBG("%s - crc_enabled=%d, recv_marker_enabled=%d, " - "xmit_marker_enabled=%d, version=%d p2p_type=%d local-p2p_type = " - "%d\n", __func__, ep->mpa_attr.crc_enabled, - ep->mpa_attr.recv_marker_enabled, - ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version, - ep->mpa_attr.p2p_type, p2p_type); + pr_debug("%s - crc_enabled=%d, recv_marker_enabled=%d, xmit_marker_enabled=%d, version=%d p2p_type=%d local-p2p_type = %d\n", + __func__, ep->mpa_attr.crc_enabled, + ep->mpa_attr.recv_marker_enabled, + ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version, + ep->mpa_attr.p2p_type, p2p_type); /* * If responder's RTR does not match with that of initiator, assign @@ -1610,7 +1614,7 @@ static int process_mpa_reply(struct c4iw_ep *ep, struct sk_buff *skb) * supports, generate TERM message */ if (rtr_mismatch) { - printk(KERN_ERR "%s: RTR mismatch, sending TERM\n", __func__); + pr_err("%s: RTR mismatch, sending TERM\n", __func__); attrs.layer_etype = LAYER_MPA | DDP_LLP; attrs.ecode = MPA_NOMATCH_RTR; attrs.next_state = C4IW_QP_STATE_TERMINATE; @@ -1629,8 +1633,7 @@ static int process_mpa_reply(struct c4iw_ep *ep, struct sk_buff *skb) * initiator ORD. */ if (insuff_ird) { - printk(KERN_ERR "%s: Insufficient IRD, sending TERM\n", - __func__); + pr_err("%s: Insufficient IRD, sending TERM\n", __func__); attrs.layer_etype = LAYER_MPA | DDP_LLP; attrs.ecode = MPA_INSUFF_IRD; attrs.next_state = C4IW_QP_STATE_TERMINATE; @@ -1669,7 +1672,7 @@ static int process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb) struct mpa_v2_conn_params *mpa_v2_params; u16 plen; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); /* * If we get more than the supported amount of private data @@ -1678,7 +1681,7 @@ static int process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb) if (ep->mpa_pkt_len + skb->len > sizeof(ep->mpa_pkt)) goto err_stop_timer; - PDBG("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__); + pr_debug("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__); /* * Copy the new data into our accumulation buffer. @@ -1694,15 +1697,15 @@ static int process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb) if (ep->mpa_pkt_len < sizeof(*mpa)) return 0; - PDBG("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__); + pr_debug("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__); mpa = (struct mpa_message *) ep->mpa_pkt; /* * Validate MPA Header. */ if (mpa->revision > mpa_rev) { - printk(KERN_ERR MOD "%s MPA version mismatch. Local = %d," - " Received = %d\n", __func__, mpa_rev, mpa->revision); + pr_err("%s MPA version mismatch. Local = %d, Received = %d\n", + __func__, mpa_rev, mpa->revision); goto err_stop_timer; } @@ -1757,8 +1760,8 @@ static int process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb) MPA_V2_IRD_ORD_MASK; ep->ord = min_t(u32, ep->ord, cur_max_read_depth(ep->com.dev)); - PDBG("%s initiator ird %u ord %u\n", __func__, ep->ird, - ep->ord); + pr_debug("%s initiator ird %u ord %u\n", + __func__, ep->ird, ep->ord); if (ntohs(mpa_v2_params->ird) & MPA_V2_PEER2PEER_MODEL) if (peer2peer) { if (ntohs(mpa_v2_params->ord) & @@ -1775,11 +1778,11 @@ static int process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb) if (peer2peer) ep->mpa_attr.p2p_type = p2p_type; - PDBG("%s - crc_enabled=%d, recv_marker_enabled=%d, " - "xmit_marker_enabled=%d, version=%d p2p_type=%d\n", __func__, - ep->mpa_attr.crc_enabled, ep->mpa_attr.recv_marker_enabled, - ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version, - ep->mpa_attr.p2p_type); + pr_debug("%s - crc_enabled=%d, recv_marker_enabled=%d, xmit_marker_enabled=%d, version=%d p2p_type=%d\n", + __func__, + ep->mpa_attr.crc_enabled, ep->mpa_attr.recv_marker_enabled, + ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version, + ep->mpa_attr.p2p_type); __state_set(&ep->com, MPA_REQ_RCVD); @@ -1815,7 +1818,7 @@ static int rx_data(struct c4iw_dev *dev, struct sk_buff *skb) ep = get_ep_from_tid(dev, tid); if (!ep) return 0; - PDBG("%s ep %p tid %u dlen %u\n", __func__, ep, ep->hwtid, dlen); + pr_debug("%s ep %p tid %u dlen %u\n", __func__, ep, ep->hwtid, dlen); skb_pull(skb, sizeof(*hdr)); skb_trim(skb, dlen); mutex_lock(&ep->com.mutex); @@ -1866,10 +1869,10 @@ static int abort_rpl(struct c4iw_dev *dev, struct sk_buff *skb) ep = get_ep_from_tid(dev, tid); if (!ep) { - printk(KERN_WARNING MOD "Abort rpl to freed endpoint\n"); + pr_warn("Abort rpl to freed endpoint\n"); return 0; } - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); mutex_lock(&ep->com.mutex); switch (ep->com.state) { case ABORTING: @@ -1878,8 +1881,7 @@ static int abort_rpl(struct c4iw_dev *dev, struct sk_buff *skb) release = 1; break; default: - printk(KERN_ERR "%s ep %p state %d\n", - __func__, ep, ep->com.state); + pr_err("%s ep %p state %d\n", __func__, ep, ep->com.state); break; } mutex_unlock(&ep->com.mutex); @@ -1995,7 +1997,8 @@ static void set_tcp_window(struct c4iw_ep *ep, struct port_info *pi) { ep->snd_win = snd_win; ep->rcv_win = rcv_win; - PDBG("%s snd_win %d rcv_win %d\n", __func__, ep->snd_win, ep->rcv_win); + pr_debug("%s snd_win %d rcv_win %d\n", + __func__, ep->snd_win, ep->rcv_win); } #define ACT_OPEN_RETRY_COUNT 2 @@ -2100,7 +2103,7 @@ static int c4iw_reconnect(struct c4iw_ep *ep) int iptype; __u8 *ra; - PDBG("%s qp %p cm_id %p\n", __func__, ep->com.qp, ep->com.cm_id); + pr_debug("%s qp %p cm_id %p\n", __func__, ep->com.qp, ep->com.cm_id); init_timer(&ep->timer); c4iw_init_wr_wait(&ep->com.wr_wait); @@ -2124,7 +2127,7 @@ static int c4iw_reconnect(struct c4iw_ep *ep) */ ep->atid = cxgb4_alloc_atid(ep->com.dev->rdev.lldi.tids, ep); if (ep->atid == -1) { - pr_err("%s - cannot alloc atid.\n", __func__); + pr_err("%s - cannot alloc atid\n", __func__); err = -ENOMEM; goto fail2; } @@ -2151,7 +2154,7 @@ static int c4iw_reconnect(struct c4iw_ep *ep) ra = (__u8 *)&raddr6->sin6_addr; } if (!ep->dst) { - pr_err("%s - cannot find route.\n", __func__); + pr_err("%s - cannot find route\n", __func__); err = -EHOSTUNREACH; goto fail3; } @@ -2159,13 +2162,13 @@ static int c4iw_reconnect(struct c4iw_ep *ep) ep->com.dev->rdev.lldi.adapter_type, ep->com.cm_id->tos); if (err) { - pr_err("%s - cannot alloc l2e.\n", __func__); + pr_err("%s - cannot alloc l2e\n", __func__); goto fail4; } - PDBG("%s txq_idx %u tx_chan %u smac_idx %u rss_qid %u l2t_idx %u\n", - __func__, ep->txq_idx, ep->tx_chan, ep->smac_idx, ep->rss_qid, - ep->l2t->idx); + pr_debug("%s txq_idx %u tx_chan %u smac_idx %u rss_qid %u l2t_idx %u\n", + __func__, ep->txq_idx, ep->tx_chan, ep->smac_idx, ep->rss_qid, + ep->l2t->idx); state_set(&ep->com, CONNECTING); ep->tos = ep->com.cm_id->tos; @@ -2215,12 +2218,12 @@ static int act_open_rpl(struct c4iw_dev *dev, struct sk_buff *skb) la6 = (struct sockaddr_in6 *)&ep->com.local_addr; ra6 = (struct sockaddr_in6 *)&ep->com.remote_addr; - PDBG("%s ep %p atid %u status %u errno %d\n", __func__, ep, atid, - status, status2errno(status)); + pr_debug("%s ep %p atid %u status %u errno %d\n", __func__, ep, atid, + status, status2errno(status)); if (cxgb_is_neg_adv(status)) { - PDBG("%s Connection problems for atid %u status %u (%s)\n", - __func__, atid, status, neg_adv_str(status)); + pr_debug("%s Connection problems for atid %u status %u (%s)\n", + __func__, atid, status, neg_adv_str(status)); ep->stats.connect_neg_adv++; mutex_lock(&dev->rdev.stats.lock); dev->rdev.stats.neg_adv++; @@ -2315,11 +2318,11 @@ static int pass_open_rpl(struct c4iw_dev *dev, struct sk_buff *skb) struct c4iw_listen_ep *ep = get_ep_from_stid(dev, stid); if (!ep) { - PDBG("%s stid %d lookup failure!\n", __func__, stid); + pr_debug("%s stid %d lookup failure!\n", __func__, stid); goto out; } - PDBG("%s ep %p status %d error %d\n", __func__, ep, - rpl->status, status2errno(rpl->status)); + pr_debug("%s ep %p status %d error %d\n", __func__, ep, + rpl->status, status2errno(rpl->status)); c4iw_wake_up(&ep->com.wr_wait, status2errno(rpl->status)); c4iw_put_ep(&ep->com); out: @@ -2332,7 +2335,7 @@ static int close_listsrv_rpl(struct c4iw_dev *dev, struct sk_buff *skb) unsigned int stid = GET_TID(rpl); struct c4iw_listen_ep *ep = get_ep_from_stid(dev, stid); - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); c4iw_wake_up(&ep->com.wr_wait, status2errno(rpl->status)); c4iw_put_ep(&ep->com); return 0; @@ -2350,7 +2353,7 @@ static int accept_cr(struct c4iw_ep *ep, struct sk_buff *skb, int win; enum chip_type adapter_type = ep->com.dev->rdev.lldi.adapter_type; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); BUG_ON(skb_cloned(skb)); skb_get(skb); @@ -2421,7 +2424,7 @@ static int accept_cr(struct c4iw_ep *ep, struct sk_buff *skb, if (peer2peer) isn += 4; rpl5->iss = cpu_to_be32(isn); - PDBG("%s iss %u\n", __func__, be32_to_cpu(rpl5->iss)); + pr_debug("%s iss %u\n", __func__, be32_to_cpu(rpl5->iss)); } rpl->opt0 = cpu_to_be64(opt0); @@ -2434,7 +2437,7 @@ static int accept_cr(struct c4iw_ep *ep, struct sk_buff *skb, static void reject_cr(struct c4iw_dev *dev, u32 hwtid, struct sk_buff *skb) { - PDBG("%s c4iw_dev %p tid %u\n", __func__, dev, hwtid); + pr_debug("%s c4iw_dev %p tid %u\n", __func__, dev, hwtid); BUG_ON(skb_cloned(skb)); skb_trim(skb, sizeof(struct cpl_tid_release)); release_tid(&dev->rdev, hwtid, skb); @@ -2460,12 +2463,13 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb) parent_ep = (struct c4iw_ep *)get_ep_from_stid(dev, stid); if (!parent_ep) { - PDBG("%s connect request on invalid stid %d\n", __func__, stid); + pr_debug("%s connect request on invalid stid %d\n", + __func__, stid); goto reject; } if (state_read(&parent_ep->com) != LISTEN) { - PDBG("%s - listening ep not in LISTEN\n", __func__); + pr_debug("%s - listening ep not in LISTEN\n", __func__); goto reject; } @@ -2474,18 +2478,18 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb) /* Find output route */ if (iptype == 4) { - PDBG("%s parent ep %p hwtid %u laddr %pI4 raddr %pI4 lport %d rport %d peer_mss %d\n" - , __func__, parent_ep, hwtid, - local_ip, peer_ip, ntohs(local_port), - ntohs(peer_port), peer_mss); + pr_debug("%s parent ep %p hwtid %u laddr %pI4 raddr %pI4 lport %d rport %d peer_mss %d\n" + , __func__, parent_ep, hwtid, + local_ip, peer_ip, ntohs(local_port), + ntohs(peer_port), peer_mss); dst = cxgb_find_route(&dev->rdev.lldi, get_real_dev, *(__be32 *)local_ip, *(__be32 *)peer_ip, local_port, peer_port, tos); } else { - PDBG("%s parent ep %p hwtid %u laddr %pI6 raddr %pI6 lport %d rport %d peer_mss %d\n" - , __func__, parent_ep, hwtid, - local_ip, peer_ip, ntohs(local_port), - ntohs(peer_port), peer_mss); + pr_debug("%s parent ep %p hwtid %u laddr %pI6 raddr %pI6 lport %d rport %d peer_mss %d\n" + , __func__, parent_ep, hwtid, + local_ip, peer_ip, ntohs(local_port), + ntohs(peer_port), peer_mss); dst = cxgb_find_route6(&dev->rdev.lldi, get_real_dev, local_ip, peer_ip, local_port, peer_port, PASS_OPEN_TOS_G(ntohl(req->tos_stid)), @@ -2493,15 +2497,13 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb) &parent_ep->com.local_addr)->sin6_scope_id); } if (!dst) { - printk(KERN_ERR MOD "%s - failed to find dst entry!\n", - __func__); + pr_err("%s - failed to find dst entry!\n", __func__); goto reject; } child_ep = alloc_ep(sizeof(*child_ep), GFP_KERNEL); if (!child_ep) { - printk(KERN_ERR MOD "%s - failed to allocate ep entry!\n", - __func__); + pr_err("%s - failed to allocate ep entry!\n", __func__); dst_release(dst); goto reject; } @@ -2509,8 +2511,7 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb) err = import_ep(child_ep, iptype, peer_ip, dst, dev, false, parent_ep->com.dev->rdev.lldi.adapter_type, tos); if (err) { - printk(KERN_ERR MOD "%s - failed to allocate l2t entry!\n", - __func__); + pr_err("%s - failed to allocate l2t entry!\n", __func__); dst_release(dst); kfree(child_ep); goto reject; @@ -2571,8 +2572,8 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb) child_ep->dst = dst; child_ep->hwtid = hwtid; - PDBG("%s tx_chan %u smac_idx %u rss_qid %u\n", __func__, - child_ep->tx_chan, child_ep->smac_idx, child_ep->rss_qid); + pr_debug("%s tx_chan %u smac_idx %u rss_qid %u\n", __func__, + child_ep->tx_chan, child_ep->smac_idx, child_ep->rss_qid); init_timer(&child_ep->timer); cxgb4_insert_tid(t, child_ep, hwtid); @@ -2607,12 +2608,12 @@ static int pass_establish(struct c4iw_dev *dev, struct sk_buff *skb) int ret; ep = get_ep_from_tid(dev, tid); - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); ep->snd_seq = be32_to_cpu(req->snd_isn); ep->rcv_seq = be32_to_cpu(req->rcv_isn); - PDBG("%s ep %p hwtid %u tcp_opt 0x%02x\n", __func__, ep, tid, - ntohs(req->tcp_opt)); + pr_debug("%s ep %p hwtid %u tcp_opt 0x%02x\n", __func__, ep, tid, + ntohs(req->tcp_opt)); set_emss(ep, ntohs(req->tcp_opt)); @@ -2644,7 +2645,7 @@ static int peer_close(struct c4iw_dev *dev, struct sk_buff *skb) if (!ep) return 0; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); dst_confirm(ep->dst); set_bit(PEER_CLOSE, &ep->com.history); @@ -2666,12 +2667,12 @@ static int peer_close(struct c4iw_dev *dev, struct sk_buff *skb) * in rdma connection migration (see c4iw_accept_cr()). */ __state_set(&ep->com, CLOSING); - PDBG("waking up ep %p tid %u\n", ep, ep->hwtid); + pr_debug("waking up ep %p tid %u\n", ep, ep->hwtid); c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET); break; case MPA_REP_SENT: __state_set(&ep->com, CLOSING); - PDBG("waking up ep %p tid %u\n", ep, ep->hwtid); + pr_debug("waking up ep %p tid %u\n", ep, ep->hwtid); c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET); break; case FPDU_MODE: @@ -2735,17 +2736,17 @@ static int peer_abort(struct c4iw_dev *dev, struct sk_buff *skb) return 0; if (cxgb_is_neg_adv(req->status)) { - PDBG("%s Negative advice on abort- tid %u status %d (%s)\n", - __func__, ep->hwtid, req->status, - neg_adv_str(req->status)); + pr_debug("%s Negative advice on abort- tid %u status %d (%s)\n", + __func__, ep->hwtid, req->status, + neg_adv_str(req->status)); ep->stats.abort_neg_adv++; mutex_lock(&dev->rdev.stats.lock); dev->rdev.stats.neg_adv++; mutex_unlock(&dev->rdev.stats.lock); goto deref_ep; } - PDBG("%s ep %p tid %u state %u\n", __func__, ep, ep->hwtid, - ep->com.state); + pr_debug("%s ep %p tid %u state %u\n", __func__, ep, ep->hwtid, + ep->com.state); set_bit(PEER_ABORT, &ep->com.history); /* @@ -2777,8 +2778,8 @@ static int peer_abort(struct c4iw_dev *dev, struct sk_buff *skb) * do some housekeeping so as to re-initiate the * connection */ - PDBG("%s: mpa_rev=%d. Retrying with mpav1\n", __func__, - mpa_rev); + pr_debug("%s: mpa_rev=%d. Retrying with mpav1\n", + __func__, mpa_rev); ep->retry_with_mpa_v1 = 1; } break; @@ -2797,16 +2798,14 @@ static int peer_abort(struct c4iw_dev *dev, struct sk_buff *skb) ep->com.qp, C4IW_QP_ATTR_NEXT_STATE, &attrs, 1); if (ret) - printk(KERN_ERR MOD - "%s - qp <- error failed!\n", - __func__); + pr_err("%s - qp <- error failed!\n", __func__); } peer_abort_upcall(ep); break; case ABORTING: break; case DEAD: - PDBG("%s PEER_ABORT IN DEAD STATE!!!!\n", __func__); + pr_debug("%s PEER_ABORT IN DEAD STATE!!!!\n", __func__); mutex_unlock(&ep->com.mutex); goto deref_ep; default: @@ -2870,7 +2869,7 @@ static int close_con_rpl(struct c4iw_dev *dev, struct sk_buff *skb) if (!ep) return 0; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); BUG_ON(!ep); /* The cm_id may be null if we failed to connect */ @@ -2918,13 +2917,13 @@ static int terminate(struct c4iw_dev *dev, struct sk_buff *skb) BUG_ON(!ep); if (ep && ep->com.qp) { - printk(KERN_WARNING MOD "TERM received tid %u qpid %u\n", tid, - ep->com.qp->wq.sq.qid); + pr_warn("TERM received tid %u qpid %u\n", + tid, ep->com.qp->wq.sq.qid); attrs.next_state = C4IW_QP_STATE_TERMINATE; c4iw_modify_qp(ep->com.qp->rhp, ep->com.qp, C4IW_QP_ATTR_NEXT_STATE, &attrs, 1); } else - printk(KERN_WARNING MOD "TERM received tid %u no ep/qp\n", tid); + pr_warn("TERM received tid %u no ep/qp\n", tid); c4iw_put_ep(&ep->com); return 0; @@ -2946,18 +2945,19 @@ static int fw4_ack(struct c4iw_dev *dev, struct sk_buff *skb) ep = get_ep_from_tid(dev, tid); if (!ep) return 0; - PDBG("%s ep %p tid %u credits %u\n", __func__, ep, ep->hwtid, credits); + pr_debug("%s ep %p tid %u credits %u\n", + __func__, ep, ep->hwtid, credits); if (credits == 0) { - PDBG("%s 0 credit ack ep %p tid %u state %u\n", - __func__, ep, ep->hwtid, state_read(&ep->com)); + pr_debug("%s 0 credit ack ep %p tid %u state %u\n", + __func__, ep, ep->hwtid, state_read(&ep->com)); goto out; } dst_confirm(ep->dst); if (ep->mpa_skb) { - PDBG("%s last streaming msg ack ep %p tid %u state %u " - "initiator %u freeing skb\n", __func__, ep, ep->hwtid, - state_read(&ep->com), ep->mpa_attr.initiator ? 1 : 0); + pr_debug("%s last streaming msg ack ep %p tid %u state %u initiator %u freeing skb\n", + __func__, ep, ep->hwtid, + state_read(&ep->com), ep->mpa_attr.initiator ? 1 : 0); mutex_lock(&ep->com.mutex); kfree_skb(ep->mpa_skb); ep->mpa_skb = NULL; @@ -2975,7 +2975,7 @@ int c4iw_reject_cr(struct iw_cm_id *cm_id, const void *pdata, u8 pdata_len) int abort; struct c4iw_ep *ep = to_ep(cm_id); - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); mutex_lock(&ep->com.mutex); if (ep->com.state != MPA_REQ_RCVD) { @@ -3006,7 +3006,7 @@ int c4iw_accept_cr(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) struct c4iw_qp *qp = get_qhp(h, conn_param->qpn); int abort = 0; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); mutex_lock(&ep->com.mutex); if (ep->com.state != MPA_REQ_RCVD) { @@ -3059,7 +3059,7 @@ int c4iw_accept_cr(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) ep->ird = 1; } - PDBG("%s %d ird %d ord %d\n", __func__, __LINE__, ep->ird, ep->ord); + pr_debug("%s %d ird %d ord %d\n", __func__, __LINE__, ep->ird, ep->ord); ep->com.cm_id = cm_id; ref_cm_id(&ep->com); @@ -3188,7 +3188,7 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) } ep = alloc_ep(sizeof(*ep), GFP_KERNEL); if (!ep) { - printk(KERN_ERR MOD "%s - cannot alloc ep.\n", __func__); + pr_err("%s - cannot alloc ep\n", __func__); err = -ENOMEM; goto out; } @@ -3215,20 +3215,20 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) ep->com.dev = dev; ep->com.qp = get_qhp(dev, conn_param->qpn); if (!ep->com.qp) { - PDBG("%s qpn 0x%x not found!\n", __func__, conn_param->qpn); + pr_debug("%s qpn 0x%x not found!\n", __func__, conn_param->qpn); err = -EINVAL; goto fail2; } ref_qp(ep); - PDBG("%s qpn 0x%x qp %p cm_id %p\n", __func__, conn_param->qpn, - ep->com.qp, cm_id); + pr_debug("%s qpn 0x%x qp %p cm_id %p\n", __func__, conn_param->qpn, + ep->com.qp, cm_id); /* * Allocate an active TID to initiate a TCP connection. */ ep->atid = cxgb4_alloc_atid(dev->rdev.lldi.tids, ep); if (ep->atid == -1) { - printk(KERN_ERR MOD "%s - cannot alloc atid.\n", __func__); + pr_err("%s - cannot alloc atid\n", __func__); err = -ENOMEM; goto fail2; } @@ -3258,9 +3258,9 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) } /* find a route */ - PDBG("%s saddr %pI4 sport 0x%x raddr %pI4 rport 0x%x\n", - __func__, &laddr->sin_addr, ntohs(laddr->sin_port), - ra, ntohs(raddr->sin_port)); + pr_debug("%s saddr %pI4 sport 0x%x raddr %pI4 rport 0x%x\n", + __func__, &laddr->sin_addr, ntohs(laddr->sin_port), + ra, ntohs(raddr->sin_port)); ep->dst = cxgb_find_route(&dev->rdev.lldi, get_real_dev, laddr->sin_addr.s_addr, raddr->sin_addr.s_addr, @@ -3280,10 +3280,10 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) } /* find a route */ - PDBG("%s saddr %pI6 sport 0x%x raddr %pI6 rport 0x%x\n", - __func__, laddr6->sin6_addr.s6_addr, - ntohs(laddr6->sin6_port), - raddr6->sin6_addr.s6_addr, ntohs(raddr6->sin6_port)); + pr_debug("%s saddr %pI6 sport 0x%x raddr %pI6 rport 0x%x\n", + __func__, laddr6->sin6_addr.s6_addr, + ntohs(laddr6->sin6_port), + raddr6->sin6_addr.s6_addr, ntohs(raddr6->sin6_port)); ep->dst = cxgb_find_route6(&dev->rdev.lldi, get_real_dev, laddr6->sin6_addr.s6_addr, raddr6->sin6_addr.s6_addr, @@ -3292,7 +3292,7 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) raddr6->sin6_scope_id); } if (!ep->dst) { - printk(KERN_ERR MOD "%s - cannot find route.\n", __func__); + pr_err("%s - cannot find route\n", __func__); err = -EHOSTUNREACH; goto fail3; } @@ -3300,13 +3300,13 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) err = import_ep(ep, iptype, ra, ep->dst, ep->com.dev, true, ep->com.dev->rdev.lldi.adapter_type, cm_id->tos); if (err) { - printk(KERN_ERR MOD "%s - cannot alloc l2e.\n", __func__); + pr_err("%s - cannot alloc l2e\n", __func__); goto fail4; } - PDBG("%s txq_idx %u tx_chan %u smac_idx %u rss_qid %u l2t_idx %u\n", - __func__, ep->txq_idx, ep->tx_chan, ep->smac_idx, ep->rss_qid, - ep->l2t->idx); + pr_debug("%s txq_idx %u tx_chan %u smac_idx %u rss_qid %u l2t_idx %u\n", + __func__, ep->txq_idx, ep->tx_chan, ep->smac_idx, ep->rss_qid, + ep->l2t->idx); state_set(&ep->com, CONNECTING); ep->tos = cm_id->tos; @@ -3414,12 +3414,12 @@ int c4iw_create_listen(struct iw_cm_id *cm_id, int backlog) ep = alloc_ep(sizeof(*ep), GFP_KERNEL); if (!ep) { - printk(KERN_ERR MOD "%s - cannot alloc ep.\n", __func__); + pr_err("%s - cannot alloc ep\n", __func__); err = -ENOMEM; goto fail1; } skb_queue_head_init(&ep->com.ep_skb_list); - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); ep->com.cm_id = cm_id; ref_cm_id(&ep->com); ep->com.dev = dev; @@ -3439,7 +3439,7 @@ int c4iw_create_listen(struct iw_cm_id *cm_id, int backlog) cm_id->m_local_addr.ss_family, ep); if (ep->stid == -1) { - printk(KERN_ERR MOD "%s - cannot alloc stid.\n", __func__); + pr_err("%s - cannot alloc stid\n", __func__); err = -ENOMEM; goto fail2; } @@ -3473,7 +3473,7 @@ int c4iw_destroy_listen(struct iw_cm_id *cm_id) int err; struct c4iw_listen_ep *ep = to_listen_ep(cm_id); - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); might_sleep(); state_set(&ep->com, DEAD); @@ -3514,8 +3514,8 @@ int c4iw_ep_disconnect(struct c4iw_ep *ep, int abrupt, gfp_t gfp) mutex_lock(&ep->com.mutex); - PDBG("%s ep %p state %s, abrupt %d\n", __func__, ep, - states[ep->com.state], abrupt); + pr_debug("%s ep %p state %s, abrupt %d\n", __func__, ep, + states[ep->com.state], abrupt); /* * Ref the ep here in case we have fatal errors causing the @@ -3568,8 +3568,8 @@ int c4iw_ep_disconnect(struct c4iw_ep *ep, int abrupt, gfp_t gfp) case MORIBUND: case ABORTING: case DEAD: - PDBG("%s ignoring disconnect ep %p state %u\n", - __func__, ep, ep->com.state); + pr_debug("%s ignoring disconnect ep %p state %u\n", + __func__, ep, ep->com.state); break; default: BUG(); @@ -3600,8 +3600,7 @@ int c4iw_ep_disconnect(struct c4iw_ep *ep, int abrupt, gfp_t gfp) C4IW_QP_ATTR_NEXT_STATE, &attrs, 1); if (ret) - pr_err(MOD - "%s - qp <- error failed!\n", + pr_err("%s - qp <- error failed!\n", __func__); } fatal = 1; @@ -3674,7 +3673,7 @@ static void passive_ofld_conn_reply(struct c4iw_dev *dev, struct sk_buff *skb, rpl_skb = (struct sk_buff *)(unsigned long)req->cookie; BUG_ON(!rpl_skb); if (req->retval) { - PDBG("%s passive open failure %d\n", __func__, req->retval); + pr_debug("%s passive open failure %d\n", __func__, req->retval); mutex_lock(&dev->rdev.stats.lock); dev->rdev.stats.pas_ofld_conn_fails++; mutex_unlock(&dev->rdev.stats.lock); @@ -3800,6 +3799,8 @@ static void send_fw_pass_open_req(struct c4iw_dev *dev, struct sk_buff *skb, int ret; req_skb = alloc_skb(sizeof(struct fw_ofld_connection_wr), GFP_KERNEL); + if (!req_skb) + return; req = (struct fw_ofld_connection_wr *)__skb_put(req_skb, sizeof(*req)); memset(req, 0, sizeof(*req)); req->op_compl = htonl(WR_OP_V(FW_OFLD_CONNECTION_WR) | FW_WR_COMPL_F); @@ -3890,7 +3891,8 @@ static int rx_pkt(struct c4iw_dev *dev, struct sk_buff *skb) lep = (struct c4iw_ep *)get_ep_from_stid(dev, stid); if (!lep) { - PDBG("%s connect request on invalid stid %d\n", __func__, stid); + pr_debug("%s connect request on invalid stid %d\n", + __func__, stid); goto reject; } @@ -3927,9 +3929,9 @@ static int rx_pkt(struct c4iw_dev *dev, struct sk_buff *skb) skb_set_transport_header(skb, (void *)tcph - (void *)rss); skb_get(skb); - PDBG("%s lip 0x%x lport %u pip 0x%x pport %u tos %d\n", __func__, - ntohl(iph->daddr), ntohs(tcph->dest), ntohl(iph->saddr), - ntohs(tcph->source), iph->tos); + pr_debug("%s lip 0x%x lport %u pip 0x%x pport %u tos %d\n", __func__, + ntohl(iph->daddr), ntohs(tcph->dest), ntohl(iph->saddr), + ntohs(tcph->source), iph->tos); dst = cxgb_find_route(&dev->rdev.lldi, get_real_dev, iph->daddr, iph->saddr, tcph->dest, @@ -4026,8 +4028,8 @@ static void process_timeout(struct c4iw_ep *ep) int abort = 1; mutex_lock(&ep->com.mutex); - PDBG("%s ep %p tid %u state %d\n", __func__, ep, ep->hwtid, - ep->com.state); + pr_debug("%s ep %p tid %u state %d\n", __func__, ep, ep->hwtid, + ep->com.state); set_bit(TIMEDOUT, &ep->com.history); switch (ep->com.state) { case MPA_REQ_SENT: @@ -4157,8 +4159,8 @@ static int set_tcb_rpl(struct c4iw_dev *dev, struct sk_buff *skb) struct cpl_set_tcb_rpl *rpl = cplhdr(skb); if (rpl->status != CPL_ERR_NONE) { - printk(KERN_ERR MOD "Unexpected SET_TCB_RPL status %u " - "for tid %u\n", rpl->status, GET_TID(rpl)); + pr_err("Unexpected SET_TCB_RPL status %u for tid %u\n", + rpl->status, GET_TID(rpl)); } kfree_skb(skb); return 0; @@ -4170,13 +4172,13 @@ static int fw6_msg(struct c4iw_dev *dev, struct sk_buff *skb) struct c4iw_wr_wait *wr_waitp; int ret; - PDBG("%s type %u\n", __func__, rpl->type); + pr_debug("%s type %u\n", __func__, rpl->type); switch (rpl->type) { case FW6_TYPE_WR_RPL: ret = (int)((be64_to_cpu(rpl->data[0]) >> 8) & 0xff); wr_waitp = (struct c4iw_wr_wait *)(__force unsigned long) rpl->data[1]; - PDBG("%s wr_waitp %p ret %u\n", __func__, wr_waitp, ret); + pr_debug("%s wr_waitp %p ret %u\n", __func__, wr_waitp, ret); if (wr_waitp) c4iw_wake_up(wr_waitp, ret ? -ret : 0); kfree_skb(skb); @@ -4186,8 +4188,8 @@ static int fw6_msg(struct c4iw_dev *dev, struct sk_buff *skb) sched(dev, skb); break; default: - printk(KERN_ERR MOD "%s unexpected fw6 msg type %u\n", __func__, - rpl->type); + pr_err("%s unexpected fw6 msg type %u\n", + __func__, rpl->type); kfree_skb(skb); break; } @@ -4203,19 +4205,18 @@ static int peer_abort_intr(struct c4iw_dev *dev, struct sk_buff *skb) ep = get_ep_from_tid(dev, tid); /* This EP will be dereferenced in peer_abort() */ if (!ep) { - printk(KERN_WARNING MOD - "Abort on non-existent endpoint, tid %d\n", tid); + pr_warn("Abort on non-existent endpoint, tid %d\n", tid); kfree_skb(skb); return 0; } if (cxgb_is_neg_adv(req->status)) { - PDBG("%s Negative advice on abort- tid %u status %d (%s)\n", - __func__, ep->hwtid, req->status, - neg_adv_str(req->status)); + pr_debug("%s Negative advice on abort- tid %u status %d (%s)\n", + __func__, ep->hwtid, req->status, + neg_adv_str(req->status)); goto out; } - PDBG("%s ep %p tid %u state %u\n", __func__, ep, ep->hwtid, - ep->com.state); + pr_debug("%s ep %p tid %u state %u\n", __func__, ep, ep->hwtid, + ep->com.state); c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET); out: diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c index bec82a600d77..14de5bde1b63 100644 --- a/drivers/infiniband/hw/cxgb4/cq.c +++ b/drivers/infiniband/hw/cxgb4/cq.c @@ -146,7 +146,7 @@ static int create_cq(struct c4iw_rdev *rdev, struct t4_cq *cq, ret = c4iw_ofld_send(rdev, skb); if (ret) goto err4; - PDBG("%s wait_event wr_wait %p\n", __func__, &wr_wait); + pr_debug("%s wait_event wr_wait %p\n", __func__, &wr_wait); ret = c4iw_wait_for_reply(rdev, &wr_wait, 0, 0, __func__); if (ret) goto err4; @@ -159,7 +159,7 @@ static int create_cq(struct c4iw_rdev *rdev, struct t4_cq *cq, &cq->bar2_qid, user ? &cq->bar2_pa : NULL); if (user && !cq->bar2_pa) { - pr_warn(MOD "%s: cqid %u not in BAR2 range.\n", + pr_warn("%s: cqid %u not in BAR2 range\n", pci_name(rdev->lldi.pdev), cq->cqid); ret = -EINVAL; goto err4; @@ -180,8 +180,8 @@ static void insert_recv_cqe(struct t4_wq *wq, struct t4_cq *cq) { struct t4_cqe cqe; - PDBG("%s wq %p cq %p sw_cidx %u sw_pidx %u\n", __func__, - wq, cq, cq->sw_cidx, cq->sw_pidx); + pr_debug("%s wq %p cq %p sw_cidx %u sw_pidx %u\n", __func__, + wq, cq, cq->sw_cidx, cq->sw_pidx); memset(&cqe, 0, sizeof(cqe)); cqe.header = cpu_to_be32(CQE_STATUS_V(T4_ERR_SWFLUSH) | CQE_OPCODE_V(FW_RI_SEND) | @@ -199,8 +199,8 @@ int c4iw_flush_rq(struct t4_wq *wq, struct t4_cq *cq, int count) int in_use = wq->rq.in_use - count; BUG_ON(in_use < 0); - PDBG("%s wq %p cq %p rq.in_use %u skip count %u\n", __func__, - wq, cq, wq->rq.in_use, count); + pr_debug("%s wq %p cq %p rq.in_use %u skip count %u\n", __func__, + wq, cq, wq->rq.in_use, count); while (in_use--) { insert_recv_cqe(wq, cq); flushed++; @@ -213,8 +213,8 @@ static void insert_sq_cqe(struct t4_wq *wq, struct t4_cq *cq, { struct t4_cqe cqe; - PDBG("%s wq %p cq %p sw_cidx %u sw_pidx %u\n", __func__, - wq, cq, cq->sw_cidx, cq->sw_pidx); + pr_debug("%s wq %p cq %p sw_cidx %u sw_pidx %u\n", __func__, + wq, cq, cq->sw_cidx, cq->sw_pidx); memset(&cqe, 0, sizeof(cqe)); cqe.header = cpu_to_be32(CQE_STATUS_V(T4_ERR_SWFLUSH) | CQE_OPCODE_V(swcqe->opcode) | @@ -283,8 +283,8 @@ static void flush_completed_wrs(struct t4_wq *wq, struct t4_cq *cq) /* * Insert this completed cqe into the swcq. */ - PDBG("%s moving cqe into swcq sq idx %u cq idx %u\n", - __func__, cidx, cq->sw_pidx); + pr_debug("%s moving cqe into swcq sq idx %u cq idx %u\n", + __func__, cidx, cq->sw_pidx); swsqe->cqe.header |= htonl(CQE_SWCQE_V(1)); cq->sw_queue[cq->sw_pidx] = swsqe->cqe; t4_swcq_produce(cq); @@ -339,7 +339,7 @@ void c4iw_flush_hw_cq(struct c4iw_cq *chp) struct t4_swsqe *swsqe; int ret; - PDBG("%s cqid 0x%x\n", __func__, chp->cq.cqid); + pr_debug("%s cqid 0x%x\n", __func__, chp->cq.cqid); ret = t4_next_hw_cqe(&chp->cq, &hw_cqe); /* @@ -432,7 +432,7 @@ void c4iw_count_rcqes(struct t4_cq *cq, struct t4_wq *wq, int *count) u32 ptr; *count = 0; - PDBG("%s count zero %d\n", __func__, *count); + pr_debug("%s count zero %d\n", __func__, *count); ptr = cq->sw_cidx; while (ptr != cq->sw_pidx) { cqe = &cq->sw_queue[ptr]; @@ -442,7 +442,7 @@ void c4iw_count_rcqes(struct t4_cq *cq, struct t4_wq *wq, int *count) if (++ptr == cq->size) ptr = 0; } - PDBG("%s cq %p count %d\n", __func__, cq, *count); + pr_debug("%s cq %p count %d\n", __func__, cq, *count); } /* @@ -473,12 +473,11 @@ static int poll_cq(struct t4_wq *wq, struct t4_cq *cq, struct t4_cqe *cqe, if (ret) return ret; - PDBG("%s CQE OVF %u qpid 0x%0x genbit %u type %u status 0x%0x" - " opcode 0x%0x len 0x%0x wrid_hi_stag 0x%x wrid_low_msn 0x%x\n", - __func__, CQE_OVFBIT(hw_cqe), CQE_QPID(hw_cqe), - CQE_GENBIT(hw_cqe), CQE_TYPE(hw_cqe), CQE_STATUS(hw_cqe), - CQE_OPCODE(hw_cqe), CQE_LEN(hw_cqe), CQE_WRID_HI(hw_cqe), - CQE_WRID_LOW(hw_cqe)); + pr_debug("%s CQE OVF %u qpid 0x%0x genbit %u type %u status 0x%0x opcode 0x%0x len 0x%0x wrid_hi_stag 0x%x wrid_low_msn 0x%x\n", + __func__, CQE_OVFBIT(hw_cqe), CQE_QPID(hw_cqe), + CQE_GENBIT(hw_cqe), CQE_TYPE(hw_cqe), CQE_STATUS(hw_cqe), + CQE_OPCODE(hw_cqe), CQE_LEN(hw_cqe), CQE_WRID_HI(hw_cqe), + CQE_WRID_LOW(hw_cqe)); /* * skip cqe's not affiliated with a QP. @@ -606,8 +605,8 @@ static int poll_cq(struct t4_wq *wq, struct t4_cq *cq, struct t4_cqe *cqe, if (!SW_CQE(hw_cqe) && (CQE_WRID_SQ_IDX(hw_cqe) != wq->sq.cidx)) { struct t4_swsqe *swsqe; - PDBG("%s out of order completion going in sw_sq at idx %u\n", - __func__, CQE_WRID_SQ_IDX(hw_cqe)); + pr_debug("%s out of order completion going in sw_sq at idx %u\n", + __func__, CQE_WRID_SQ_IDX(hw_cqe)); swsqe = &wq->sq.sw_sq[CQE_WRID_SQ_IDX(hw_cqe)]; swsqe->cqe = *hw_cqe; swsqe->complete = 1; @@ -641,13 +640,13 @@ proc_cqe: BUG_ON(wq->sq.in_use <= 0 && wq->sq.in_use >= wq->sq.size); wq->sq.cidx = (uint16_t)idx; - PDBG("%s completing sq idx %u\n", __func__, wq->sq.cidx); + pr_debug("%s completing sq idx %u\n", __func__, wq->sq.cidx); *cookie = wq->sq.sw_sq[wq->sq.cidx].wr_id; if (c4iw_wr_log) c4iw_log_wr_stats(wq, hw_cqe); t4_sq_consume(wq); } else { - PDBG("%s completing rq idx %u\n", __func__, wq->rq.cidx); + pr_debug("%s completing rq idx %u\n", __func__, wq->rq.cidx); *cookie = wq->rq.sw_rq[wq->rq.cidx].wr_id; BUG_ON(t4_rq_empty(wq)); if (c4iw_wr_log) @@ -664,12 +663,12 @@ flush_wq: skip_cqe: if (SW_CQE(hw_cqe)) { - PDBG("%s cq %p cqid 0x%x skip sw cqe cidx %u\n", - __func__, cq, cq->cqid, cq->sw_cidx); + pr_debug("%s cq %p cqid 0x%x skip sw cqe cidx %u\n", + __func__, cq, cq->cqid, cq->sw_cidx); t4_swcq_consume(cq); } else { - PDBG("%s cq %p cqid 0x%x skip hw cqe cidx %u\n", - __func__, cq, cq->cqid, cq->cidx); + pr_debug("%s cq %p cqid 0x%x skip hw cqe cidx %u\n", + __func__, cq, cq->cqid, cq->cidx); t4_hwcq_consume(cq); } return ret; @@ -715,10 +714,12 @@ static int c4iw_poll_cq_one(struct c4iw_cq *chp, struct ib_wc *wc) wc->vendor_err = CQE_STATUS(&cqe); wc->wc_flags = 0; - PDBG("%s qpid 0x%x type %d opcode %d status 0x%x len %u wrid hi 0x%x " - "lo 0x%x cookie 0x%llx\n", __func__, CQE_QPID(&cqe), - CQE_TYPE(&cqe), CQE_OPCODE(&cqe), CQE_STATUS(&cqe), CQE_LEN(&cqe), - CQE_WRID_HI(&cqe), CQE_WRID_LOW(&cqe), (unsigned long long)cookie); + pr_debug("%s qpid 0x%x type %d opcode %d status 0x%x len %u wrid hi 0x%x lo 0x%x cookie 0x%llx\n", + __func__, CQE_QPID(&cqe), + CQE_TYPE(&cqe), CQE_OPCODE(&cqe), + CQE_STATUS(&cqe), CQE_LEN(&cqe), + CQE_WRID_HI(&cqe), CQE_WRID_LOW(&cqe), + (unsigned long long)cookie); if (CQE_TYPE(&cqe) == 0) { if (!CQE_STATUS(&cqe)) @@ -766,8 +767,7 @@ static int c4iw_poll_cq_one(struct c4iw_cq *chp, struct ib_wc *wc) wc->opcode = IB_WC_SEND; break; default: - printk(KERN_ERR MOD "Unexpected opcode %d " - "in the CQE received for QPID=0x%0x\n", + pr_err("Unexpected opcode %d in the CQE received for QPID=0x%0x\n", CQE_OPCODE(&cqe), CQE_QPID(&cqe)); ret = -EINVAL; goto out; @@ -822,8 +822,7 @@ static int c4iw_poll_cq_one(struct c4iw_cq *chp, struct ib_wc *wc) wc->status = IB_WC_WR_FLUSH_ERR; break; default: - printk(KERN_ERR MOD - "Unexpected cqe_status 0x%x for QPID=0x%0x\n", + pr_err("Unexpected cqe_status 0x%x for QPID=0x%0x\n", CQE_STATUS(&cqe), CQE_QPID(&cqe)); wc->status = IB_WC_FATAL_ERR; } @@ -860,7 +859,7 @@ int c4iw_destroy_cq(struct ib_cq *ib_cq) struct c4iw_cq *chp; struct c4iw_ucontext *ucontext; - PDBG("%s ib_cq %p\n", __func__, ib_cq); + pr_debug("%s ib_cq %p\n", __func__, ib_cq); chp = to_c4iw_cq(ib_cq); remove_handle(chp->rhp, &chp->rhp->cqidr, chp->cq.cqid); @@ -892,7 +891,7 @@ struct ib_cq *c4iw_create_cq(struct ib_device *ibdev, size_t memsize, hwentries; struct c4iw_mm_entry *mm, *mm2; - PDBG("%s ib_dev %p entries %d\n", __func__, ibdev, entries); + pr_debug("%s ib_dev %p entries %d\n", __func__, ibdev, entries); if (attr->flags) return ERR_PTR(-EINVAL); @@ -998,9 +997,9 @@ struct ib_cq *c4iw_create_cq(struct ib_device *ibdev, mm2->len = PAGE_SIZE; insert_mmap(ucontext, mm2); } - PDBG("%s cqid 0x%0x chp %p size %u memsize %zu, dma_addr 0x%0llx\n", - __func__, chp->cq.cqid, chp, chp->cq.size, - chp->cq.memsize, (unsigned long long) chp->cq.dma_addr); + pr_debug("%s cqid 0x%0x chp %p size %u memsize %zu, dma_addr 0x%0llx\n", + __func__, chp->cq.cqid, chp, chp->cq.size, + chp->cq.memsize, (unsigned long long)chp->cq.dma_addr); return &chp->ibcq; err6: kfree(mm2); diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c index 4e4f1a732b01..329fb65e8fb0 100644 --- a/drivers/infiniband/hw/cxgb4/device.c +++ b/drivers/infiniband/hw/cxgb4/device.c @@ -334,7 +334,7 @@ static int qp_release(struct inode *inode, struct file *file) { struct c4iw_debugfs_data *qpd = file->private_data; if (!qpd) { - printk(KERN_INFO "%s null qpd?\n", __func__); + pr_info("%s null qpd?\n", __func__); return 0; } vfree(qpd->buf); @@ -422,7 +422,7 @@ static int stag_release(struct inode *inode, struct file *file) { struct c4iw_debugfs_data *stagd = file->private_data; if (!stagd) { - printk(KERN_INFO "%s null stagd?\n", __func__); + pr_info("%s null stagd?\n", __func__); return 0; } vfree(stagd->buf); @@ -796,15 +796,14 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev) * cqid and qpid range must match for now. */ if (rdev->lldi.udb_density != rdev->lldi.ucq_density) { - pr_err(MOD "%s: unsupported udb/ucq densities %u/%u\n", + pr_err("%s: unsupported udb/ucq densities %u/%u\n", pci_name(rdev->lldi.pdev), rdev->lldi.udb_density, rdev->lldi.ucq_density); return -EINVAL; } if (rdev->lldi.vr->qp.start != rdev->lldi.vr->cq.start || rdev->lldi.vr->qp.size != rdev->lldi.vr->cq.size) { - pr_err(MOD "%s: unsupported qp and cq id ranges " - "qp start %u size %u cq start %u size %u\n", + pr_err("%s: unsupported qp and cq id ranges qp start %u size %u cq start %u size %u\n", pci_name(rdev->lldi.pdev), rdev->lldi.vr->qp.start, rdev->lldi.vr->qp.size, rdev->lldi.vr->cq.size, rdev->lldi.vr->cq.size); @@ -813,23 +812,20 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev) rdev->qpmask = rdev->lldi.udb_density - 1; rdev->cqmask = rdev->lldi.ucq_density - 1; - PDBG("%s dev %s stag start 0x%0x size 0x%0x num stags %d " - "pbl start 0x%0x size 0x%0x rq start 0x%0x size 0x%0x " - "qp qid start %u size %u cq qid start %u size %u\n", - __func__, pci_name(rdev->lldi.pdev), rdev->lldi.vr->stag.start, - rdev->lldi.vr->stag.size, c4iw_num_stags(rdev), - rdev->lldi.vr->pbl.start, - rdev->lldi.vr->pbl.size, rdev->lldi.vr->rq.start, - rdev->lldi.vr->rq.size, - rdev->lldi.vr->qp.start, - rdev->lldi.vr->qp.size, - rdev->lldi.vr->cq.start, - rdev->lldi.vr->cq.size); - PDBG("udb %pR db_reg %p gts_reg %p " - "qpmask 0x%x cqmask 0x%x\n", - &rdev->lldi.pdev->resource[2], - rdev->lldi.db_reg, rdev->lldi.gts_reg, - rdev->qpmask, rdev->cqmask); + pr_debug("%s dev %s stag start 0x%0x size 0x%0x num stags %d pbl start 0x%0x size 0x%0x rq start 0x%0x size 0x%0x qp qid start %u size %u cq qid start %u size %u\n", + __func__, pci_name(rdev->lldi.pdev), rdev->lldi.vr->stag.start, + rdev->lldi.vr->stag.size, c4iw_num_stags(rdev), + rdev->lldi.vr->pbl.start, + rdev->lldi.vr->pbl.size, rdev->lldi.vr->rq.start, + rdev->lldi.vr->rq.size, + rdev->lldi.vr->qp.start, + rdev->lldi.vr->qp.size, + rdev->lldi.vr->cq.start, + rdev->lldi.vr->cq.size); + pr_debug("udb %pR db_reg %p gts_reg %p qpmask 0x%x cqmask 0x%x\n", + &rdev->lldi.pdev->resource[2], + rdev->lldi.db_reg, rdev->lldi.gts_reg, + rdev->qpmask, rdev->cqmask); if (c4iw_num_stags(rdev) == 0) return -EINVAL; @@ -843,22 +839,22 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev) err = c4iw_init_resource(rdev, c4iw_num_stags(rdev), T4_MAX_NUM_PD); if (err) { - printk(KERN_ERR MOD "error %d initializing resources\n", err); + pr_err("error %d initializing resources\n", err); return err; } err = c4iw_pblpool_create(rdev); if (err) { - printk(KERN_ERR MOD "error %d initializing pbl pool\n", err); + pr_err("error %d initializing pbl pool\n", err); goto destroy_resource; } err = c4iw_rqtpool_create(rdev); if (err) { - printk(KERN_ERR MOD "error %d initializing rqt pool\n", err); + pr_err("error %d initializing rqt pool\n", err); goto destroy_pblpool; } err = c4iw_ocqp_pool_create(rdev); if (err) { - printk(KERN_ERR MOD "error %d initializing ocqp pool\n", err); + pr_err("error %d initializing ocqp pool\n", err); goto destroy_rqtpool; } rdev->status_page = (struct t4_dev_status_page *) @@ -936,7 +932,7 @@ static void c4iw_dealloc(struct uld_ctx *ctx) static void c4iw_remove(struct uld_ctx *ctx) { - PDBG("%s c4iw_dev %p\n", __func__, ctx->dev); + pr_debug("%s c4iw_dev %p\n", __func__, ctx->dev); c4iw_unregister_device(ctx->dev); c4iw_dealloc(ctx); } @@ -954,25 +950,25 @@ static struct c4iw_dev *c4iw_alloc(const struct cxgb4_lld_info *infop) int ret; if (!rdma_supported(infop)) { - printk(KERN_INFO MOD "%s: RDMA not supported on this device.\n", - pci_name(infop->pdev)); + pr_info("%s: RDMA not supported on this device\n", + pci_name(infop->pdev)); return ERR_PTR(-ENOSYS); } if (!ocqp_supported(infop)) - pr_info("%s: On-Chip Queues not supported on this device.\n", + pr_info("%s: On-Chip Queues not supported on this device\n", pci_name(infop->pdev)); devp = (struct c4iw_dev *)ib_alloc_device(sizeof(*devp)); if (!devp) { - printk(KERN_ERR MOD "Cannot allocate ib device\n"); + pr_err("Cannot allocate ib device\n"); return ERR_PTR(-ENOMEM); } devp->rdev.lldi = *infop; /* init various hw-queue params based on lld info */ - PDBG("%s: Ing. padding boundary is %d, egrsstatuspagesize = %d\n", - __func__, devp->rdev.lldi.sge_ingpadboundary, - devp->rdev.lldi.sge_egrstatuspagesize); + pr_debug("%s: Ing. padding boundary is %d, egrsstatuspagesize = %d\n", + __func__, devp->rdev.lldi.sge_ingpadboundary, + devp->rdev.lldi.sge_egrstatuspagesize); devp->rdev.hw_queue.t4_eq_status_entries = devp->rdev.lldi.sge_ingpadboundary > 64 ? 2 : 1; @@ -1000,7 +996,7 @@ static struct c4iw_dev *c4iw_alloc(const struct cxgb4_lld_info *infop) devp->rdev.bar2_kva = ioremap_wc(devp->rdev.bar2_pa, pci_resource_len(devp->rdev.lldi.pdev, 2)); if (!devp->rdev.bar2_kva) { - pr_err(MOD "Unable to ioremap BAR2\n"); + pr_err("Unable to ioremap BAR2\n"); ib_dealloc_device(&devp->ibdev); return ERR_PTR(-EINVAL); } @@ -1012,20 +1008,19 @@ static struct c4iw_dev *c4iw_alloc(const struct cxgb4_lld_info *infop) devp->rdev.oc_mw_kva = ioremap_wc(devp->rdev.oc_mw_pa, devp->rdev.lldi.vr->ocq.size); if (!devp->rdev.oc_mw_kva) { - pr_err(MOD "Unable to ioremap onchip mem\n"); + pr_err("Unable to ioremap onchip mem\n"); ib_dealloc_device(&devp->ibdev); return ERR_PTR(-EINVAL); } } - PDBG(KERN_INFO MOD "ocq memory: " - "hw_start 0x%x size %u mw_pa 0x%lx mw_kva %p\n", - devp->rdev.lldi.vr->ocq.start, devp->rdev.lldi.vr->ocq.size, - devp->rdev.oc_mw_pa, devp->rdev.oc_mw_kva); + pr_debug("ocq memory: hw_start 0x%x size %u mw_pa 0x%lx mw_kva %p\n", + devp->rdev.lldi.vr->ocq.start, devp->rdev.lldi.vr->ocq.size, + devp->rdev.oc_mw_pa, devp->rdev.oc_mw_kva); ret = c4iw_rdev_open(&devp->rdev); if (ret) { - printk(KERN_ERR MOD "Unable to open CXIO rdev err %d\n", ret); + pr_err("Unable to open CXIO rdev err %d\n", ret); ib_dealloc_device(&devp->ibdev); return ERR_PTR(ret); } @@ -1071,17 +1066,17 @@ static void *c4iw_uld_add(const struct cxgb4_lld_info *infop) } ctx->lldi = *infop; - PDBG("%s found device %s nchan %u nrxq %u ntxq %u nports %u\n", - __func__, pci_name(ctx->lldi.pdev), - ctx->lldi.nchan, ctx->lldi.nrxq, - ctx->lldi.ntxq, ctx->lldi.nports); + pr_debug("%s found device %s nchan %u nrxq %u ntxq %u nports %u\n", + __func__, pci_name(ctx->lldi.pdev), + ctx->lldi.nchan, ctx->lldi.nrxq, + ctx->lldi.ntxq, ctx->lldi.nports); mutex_lock(&dev_mutex); list_add_tail(&ctx->entry, &uld_ctx_list); mutex_unlock(&dev_mutex); for (i = 0; i < ctx->lldi.nrxq; i++) - PDBG("rxqid[%u] %u\n", i, ctx->lldi.rxq_ids[i]); + pr_debug("rxqid[%u] %u\n", i, ctx->lldi.rxq_ids[i]); out: return ctx; } @@ -1138,8 +1133,7 @@ static inline int recv_rx_pkt(struct c4iw_dev *dev, const struct pkt_gl *gl, goto out; if (c4iw_handlers[opcode] == NULL) { - pr_info("%s no handler opcode 0x%x...\n", __func__, - opcode); + pr_info("%s no handler opcode 0x%x...\n", __func__, opcode); kfree_skb(skb); goto out; } @@ -1176,13 +1170,11 @@ static int c4iw_uld_rx_handler(void *handle, const __be64 *rsp, if (recv_rx_pkt(dev, gl, rsp)) return 0; - pr_info("%s: unexpected FL contents at %p, " \ - "RSS %#llx, FL %#llx, len %u\n", - pci_name(ctx->lldi.pdev), gl->va, - (unsigned long long)be64_to_cpu(*rsp), - (unsigned long long)be64_to_cpu( - *(__force __be64 *)gl->va), - gl->tot_len); + pr_info("%s: unexpected FL contents at %p, RSS %#llx, FL %#llx, len %u\n", + pci_name(ctx->lldi.pdev), gl->va, + be64_to_cpu(*rsp), + be64_to_cpu(*(__force __be64 *)gl->va), + gl->tot_len); return 0; } else { @@ -1195,8 +1187,7 @@ static int c4iw_uld_rx_handler(void *handle, const __be64 *rsp, if (c4iw_handlers[opcode]) { c4iw_handlers[opcode](dev, skb); } else { - pr_info("%s no handler opcode 0x%x...\n", __func__, - opcode); + pr_info("%s no handler opcode 0x%x...\n", __func__, opcode); kfree_skb(skb); } @@ -1209,17 +1200,16 @@ static int c4iw_uld_state_change(void *handle, enum cxgb4_state new_state) { struct uld_ctx *ctx = handle; - PDBG("%s new_state %u\n", __func__, new_state); + pr_debug("%s new_state %u\n", __func__, new_state); switch (new_state) { case CXGB4_STATE_UP: - printk(KERN_INFO MOD "%s: Up\n", pci_name(ctx->lldi.pdev)); + pr_info("%s: Up\n", pci_name(ctx->lldi.pdev)); if (!ctx->dev) { int ret; ctx->dev = c4iw_alloc(&ctx->lldi); if (IS_ERR(ctx->dev)) { - printk(KERN_ERR MOD - "%s: initialization failed: %ld\n", + pr_err("%s: initialization failed: %ld\n", pci_name(ctx->lldi.pdev), PTR_ERR(ctx->dev)); ctx->dev = NULL; @@ -1227,22 +1217,19 @@ static int c4iw_uld_state_change(void *handle, enum cxgb4_state new_state) } ret = c4iw_register_device(ctx->dev); if (ret) { - printk(KERN_ERR MOD - "%s: RDMA registration failed: %d\n", + pr_err("%s: RDMA registration failed: %d\n", pci_name(ctx->lldi.pdev), ret); c4iw_dealloc(ctx); } } break; case CXGB4_STATE_DOWN: - printk(KERN_INFO MOD "%s: Down\n", - pci_name(ctx->lldi.pdev)); + pr_info("%s: Down\n", pci_name(ctx->lldi.pdev)); if (ctx->dev) c4iw_remove(ctx); break; case CXGB4_STATE_START_RECOVERY: - printk(KERN_INFO MOD "%s: Fatal Error\n", - pci_name(ctx->lldi.pdev)); + pr_info("%s: Fatal Error\n", pci_name(ctx->lldi.pdev)); if (ctx->dev) { struct ib_event event; @@ -1255,8 +1242,7 @@ static int c4iw_uld_state_change(void *handle, enum cxgb4_state new_state) } break; case CXGB4_STATE_DETACH: - printk(KERN_INFO MOD "%s: Detach\n", - pci_name(ctx->lldi.pdev)); + pr_info("%s: Detach\n", pci_name(ctx->lldi.pdev)); if (ctx->dev) c4iw_remove(ctx); break; @@ -1406,9 +1392,7 @@ static void recover_lost_dbs(struct uld_ctx *ctx, struct qp_list *qp_list) t4_sq_host_wq_pidx(&qp->wq), t4_sq_wq_size(&qp->wq)); if (ret) { - pr_err(MOD "%s: Fatal error - " - "DB overflow recovery failed - " - "error syncing SQ qid %u\n", + pr_err("%s: Fatal error - DB overflow recovery failed - error syncing SQ qid %u\n", pci_name(ctx->lldi.pdev), qp->wq.sq.qid); spin_unlock(&qp->lock); spin_unlock_irq(&qp->rhp->lock); @@ -1422,9 +1406,7 @@ static void recover_lost_dbs(struct uld_ctx *ctx, struct qp_list *qp_list) t4_rq_wq_size(&qp->wq)); if (ret) { - pr_err(MOD "%s: Fatal error - " - "DB overflow recovery failed - " - "error syncing RQ qid %u\n", + pr_err("%s: Fatal error - DB overflow recovery failed - error syncing RQ qid %u\n", pci_name(ctx->lldi.pdev), qp->wq.rq.qid); spin_unlock(&qp->lock); spin_unlock_irq(&qp->rhp->lock); @@ -1455,7 +1437,7 @@ static void recover_queues(struct uld_ctx *ctx) /* flush the SGE contexts */ ret = cxgb4_flush_eq_cache(ctx->dev->rdev.lldi.ports[0]); if (ret) { - printk(KERN_ERR MOD "%s: Fatal error - DB overflow recovery failed\n", + pr_err("%s: Fatal error - DB overflow recovery failed\n", pci_name(ctx->lldi.pdev)); return; } @@ -1513,8 +1495,8 @@ static int c4iw_uld_control(void *handle, enum cxgb4_control control, ...) mutex_unlock(&ctx->dev->rdev.stats.lock); break; default: - printk(KERN_WARNING MOD "%s: unknown control cmd %u\n", - pci_name(ctx->lldi.pdev), control); + pr_warn("%s: unknown control cmd %u\n", + pci_name(ctx->lldi.pdev), control); break; } return 0; @@ -1543,8 +1525,7 @@ static int __init c4iw_init_module(void) c4iw_debugfs_root = debugfs_create_dir(DRV_NAME, NULL); if (!c4iw_debugfs_root) - printk(KERN_WARNING MOD - "could not create debugfs entry, continuing\n"); + pr_warn("could not create debugfs entry, continuing\n"); cxgb4_register_uld(CXGB4_ULD_RDMA, &c4iw_uld_info); diff --git a/drivers/infiniband/hw/cxgb4/ev.c b/drivers/infiniband/hw/cxgb4/ev.c index bdfac2ccb704..8f963df0bffc 100644 --- a/drivers/infiniband/hw/cxgb4/ev.c +++ b/drivers/infiniband/hw/cxgb4/ev.c @@ -47,17 +47,16 @@ static void print_tpte(struct c4iw_dev *dev, u32 stag) "%s cxgb4_read_tpte err %d\n", __func__, ret); return; } - PDBG("stag idx 0x%x valid %d key 0x%x state %d pdid %d " - "perm 0x%x ps %d len 0x%llx va 0x%llx\n", - stag & 0xffffff00, - FW_RI_TPTE_VALID_G(ntohl(tpte.valid_to_pdid)), - FW_RI_TPTE_STAGKEY_G(ntohl(tpte.valid_to_pdid)), - FW_RI_TPTE_STAGSTATE_G(ntohl(tpte.valid_to_pdid)), - FW_RI_TPTE_PDID_G(ntohl(tpte.valid_to_pdid)), - FW_RI_TPTE_PERM_G(ntohl(tpte.locread_to_qpid)), - FW_RI_TPTE_PS_G(ntohl(tpte.locread_to_qpid)), - ((u64)ntohl(tpte.len_hi) << 32) | ntohl(tpte.len_lo), - ((u64)ntohl(tpte.va_hi) << 32) | ntohl(tpte.va_lo_fbo)); + pr_debug("stag idx 0x%x valid %d key 0x%x state %d pdid %d perm 0x%x ps %d len 0x%llx va 0x%llx\n", + stag & 0xffffff00, + FW_RI_TPTE_VALID_G(ntohl(tpte.valid_to_pdid)), + FW_RI_TPTE_STAGKEY_G(ntohl(tpte.valid_to_pdid)), + FW_RI_TPTE_STAGSTATE_G(ntohl(tpte.valid_to_pdid)), + FW_RI_TPTE_PDID_G(ntohl(tpte.valid_to_pdid)), + FW_RI_TPTE_PERM_G(ntohl(tpte.locread_to_qpid)), + FW_RI_TPTE_PS_G(ntohl(tpte.locread_to_qpid)), + ((u64)ntohl(tpte.len_hi) << 32) | ntohl(tpte.len_lo), + ((u64)ntohl(tpte.va_hi) << 32) | ntohl(tpte.va_lo_fbo)); } static void dump_err_cqe(struct c4iw_dev *dev, struct t4_cqe *err_cqe) @@ -71,9 +70,9 @@ static void dump_err_cqe(struct c4iw_dev *dev, struct t4_cqe *err_cqe) CQE_STATUS(err_cqe), CQE_TYPE(err_cqe), ntohl(err_cqe->len), CQE_WRID_HI(err_cqe), CQE_WRID_LOW(err_cqe)); - PDBG("%016llx %016llx %016llx %016llx\n", - be64_to_cpu(p[0]), be64_to_cpu(p[1]), be64_to_cpu(p[2]), - be64_to_cpu(p[3])); + pr_debug("%016llx %016llx %016llx %016llx\n", + be64_to_cpu(p[0]), be64_to_cpu(p[1]), be64_to_cpu(p[2]), + be64_to_cpu(p[3])); /* * Ingress WRITE and READ_RESP errors provide @@ -124,8 +123,7 @@ void c4iw_ev_dispatch(struct c4iw_dev *dev, struct t4_cqe *err_cqe) spin_lock_irq(&dev->lock); qhp = get_qhp(dev, CQE_QPID(err_cqe)); if (!qhp) { - printk(KERN_ERR MOD "BAD AE qpid 0x%x opcode %d " - "status 0x%x type %d wrid.hi 0x%x wrid.lo 0x%x\n", + pr_err("BAD AE qpid 0x%x opcode %d status 0x%x type %d wrid.hi 0x%x wrid.lo 0x%x\n", CQE_QPID(err_cqe), CQE_OPCODE(err_cqe), CQE_STATUS(err_cqe), CQE_TYPE(err_cqe), CQE_WRID_HI(err_cqe), @@ -140,8 +138,7 @@ void c4iw_ev_dispatch(struct c4iw_dev *dev, struct t4_cqe *err_cqe) cqid = qhp->attr.rcq; chp = get_chp(dev, cqid); if (!chp) { - printk(KERN_ERR MOD "BAD AE cqid 0x%x qpid 0x%x opcode %d " - "status 0x%x type %d wrid.hi 0x%x wrid.lo 0x%x\n", + pr_err("BAD AE cqid 0x%x qpid 0x%x opcode %d status 0x%x type %d wrid.hi 0x%x wrid.lo 0x%x\n", cqid, CQE_QPID(err_cqe), CQE_OPCODE(err_cqe), CQE_STATUS(err_cqe), CQE_TYPE(err_cqe), CQE_WRID_HI(err_cqe), @@ -165,7 +162,7 @@ void c4iw_ev_dispatch(struct c4iw_dev *dev, struct t4_cqe *err_cqe) /* Completion Events */ case T4_ERR_SUCCESS: - printk(KERN_ERR MOD "AE with status 0!\n"); + pr_err("AE with status 0!\n"); break; case T4_ERR_STAG: @@ -207,7 +204,7 @@ void c4iw_ev_dispatch(struct c4iw_dev *dev, struct t4_cqe *err_cqe) break; default: - printk(KERN_ERR MOD "Unknown T4 status 0x%x QPID 0x%x\n", + pr_err("Unknown T4 status 0x%x QPID 0x%x\n", CQE_STATUS(err_cqe), qhp->wq.sq.qid); post_qp_event(dev, chp, qhp, err_cqe, IB_EVENT_QP_FATAL); break; @@ -237,7 +234,7 @@ int c4iw_ev_handler(struct c4iw_dev *dev, u32 qid) if (atomic_dec_and_test(&chp->refcnt)) wake_up(&chp->wait); } else { - PDBG("%s unknown cqid 0x%x\n", __func__, qid); + pr_debug("%s unknown cqid 0x%x\n", __func__, qid); spin_unlock_irqrestore(&dev->lock, flag); } return 0; diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h index 5846c47c8d55..819a30635d53 100644 --- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h +++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h @@ -64,12 +64,11 @@ #define DRV_NAME "iw_cxgb4" #define MOD DRV_NAME ":" -extern int c4iw_debug; -#define PDBG(fmt, args...) \ -do { \ - if (c4iw_debug) \ - printk(MOD fmt, ## args); \ -} while (0) +#ifdef pr_fmt +#undef pr_fmt +#endif + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include "t4.h" @@ -231,15 +230,15 @@ static inline int c4iw_wait_for_reply(struct c4iw_rdev *rdev, ret = wait_for_completion_timeout(&wr_waitp->completion, C4IW_WR_TO); if (!ret) { - PDBG("%s - Device %s not responding (disabling device) - tid %u qpid %u\n", - func, pci_name(rdev->lldi.pdev), hwtid, qpid); + pr_debug("%s - Device %s not responding (disabling device) - tid %u qpid %u\n", + func, pci_name(rdev->lldi.pdev), hwtid, qpid); rdev->flags |= T4_FATAL_ERROR; wr_waitp->ret = -EIO; } out: if (wr_waitp->ret) - PDBG("%s: FW reply %d tid %u qpid %u\n", - pci_name(rdev->lldi.pdev), wr_waitp->ret, hwtid, qpid); + pr_debug("%s: FW reply %d tid %u qpid %u\n", + pci_name(rdev->lldi.pdev), wr_waitp->ret, hwtid, qpid); return wr_waitp->ret; } @@ -538,8 +537,9 @@ static inline struct c4iw_mm_entry *remove_mmap(struct c4iw_ucontext *ucontext, if (mm->key == key && mm->len == len) { list_del_init(&mm->entry); spin_unlock(&ucontext->mmap_lock); - PDBG("%s key 0x%x addr 0x%llx len %d\n", __func__, - key, (unsigned long long) mm->addr, mm->len); + pr_debug("%s key 0x%x addr 0x%llx len %d\n", + __func__, key, + (unsigned long long)mm->addr, mm->len); return mm; } } @@ -551,8 +551,8 @@ static inline void insert_mmap(struct c4iw_ucontext *ucontext, struct c4iw_mm_entry *mm) { spin_lock(&ucontext->mmap_lock); - PDBG("%s key 0x%x addr 0x%llx len %d\n", __func__, - mm->key, (unsigned long long) mm->addr, mm->len); + pr_debug("%s key 0x%x addr 0x%llx len %d\n", + __func__, mm->key, (unsigned long long)mm->addr, mm->len); list_add_tail(&mm->entry, &ucontext->mmaps); spin_unlock(&ucontext->mmap_lock); } @@ -670,17 +670,19 @@ enum c4iw_mmid_state { #define MPA_V2_RDMA_READ_RTR 0x4000 #define MPA_V2_IRD_ORD_MASK 0x3FFF -#define c4iw_put_ep(ep) { \ - PDBG("put_ep (via %s:%u) ep %p refcnt %d\n", __func__, __LINE__, \ - ep, kref_read(&((ep)->kref))); \ - WARN_ON(kref_read(&((ep)->kref)) < 1); \ - kref_put(&((ep)->kref), _c4iw_free_ep); \ +#define c4iw_put_ep(ep) { \ + pr_debug("put_ep (via %s:%u) ep %p refcnt %d\n", \ + __func__, __LINE__, \ + ep, kref_read(&((ep)->kref))); \ + WARN_ON(kref_read(&((ep)->kref)) < 1); \ + kref_put(&((ep)->kref), _c4iw_free_ep); \ } -#define c4iw_get_ep(ep) { \ - PDBG("get_ep (via %s:%u) ep %p, refcnt %d\n", __func__, __LINE__, \ - ep, kref_read(&((ep)->kref))); \ - kref_get(&((ep)->kref)); \ +#define c4iw_get_ep(ep) { \ + pr_debug("get_ep (via %s:%u) ep %p, refcnt %d\n", \ + __func__, __LINE__, \ + ep, kref_read(&((ep)->kref))); \ + kref_get(&((ep)->kref)); \ } void _c4iw_free_ep(struct kref *kref); diff --git a/drivers/infiniband/hw/cxgb4/mem.c b/drivers/infiniband/hw/cxgb4/mem.c index 410408f886c1..3ee7f43e419a 100644 --- a/drivers/infiniband/hw/cxgb4/mem.c +++ b/drivers/infiniband/hw/cxgb4/mem.c @@ -38,9 +38,9 @@ #include "iw_cxgb4.h" -int use_dsgl = 0; +int use_dsgl = 1; module_param(use_dsgl, int, 0644); -MODULE_PARM_DESC(use_dsgl, "Use DSGL for PBL/FastReg (default=0)"); +MODULE_PARM_DESC(use_dsgl, "Use DSGL for PBL/FastReg (default=1) (DEPRECATED)"); #define T4_ULPTX_MIN_IO 32 #define C4IW_MAX_INLINE_SIZE 96 @@ -125,7 +125,7 @@ static int _c4iw_write_mem_inline(struct c4iw_rdev *rdev, u32 addr, u32 len, cmd |= cpu_to_be32(T5_ULP_MEMIO_IMM_F); addr &= 0x7FFFFFF; - PDBG("%s addr 0x%x len %u\n", __func__, addr, len); + pr_debug("%s addr 0x%x len %u\n", __func__, addr, len); num_wqe = DIV_ROUND_UP(len, C4IW_MAX_INLINE_SIZE); c4iw_init_wr_wait(&wr_wait); for (i = 0; i < num_wqe; i++) { @@ -231,13 +231,11 @@ out: static int write_adapter_mem(struct c4iw_rdev *rdev, u32 addr, u32 len, void *data, struct sk_buff *skb) { - if (is_t5(rdev->lldi.adapter_type) && use_dsgl) { + if (rdev->lldi.ulptx_memwrite_dsgl && use_dsgl) { if (len > inline_threshold) { if (_c4iw_write_mem_dma(rdev, addr, len, data, skb)) { - printk_ratelimited(KERN_WARNING - "%s: dma map" - " failure (non fatal)\n", - pci_name(rdev->lldi.pdev)); + pr_warn_ratelimited("%s: dma map failure (non fatal)\n", + pci_name(rdev->lldi.pdev)); return _c4iw_write_mem_inline(rdev, addr, len, data, skb); } else { @@ -289,8 +287,8 @@ static int write_tpt_entry(struct c4iw_rdev *rdev, u32 reset_tpt_entry, mutex_unlock(&rdev->stats.lock); *stag = (stag_idx << 8) | (atomic_inc_return(&key) & 0xff); } - PDBG("%s stag_state 0x%0x type 0x%0x pdid 0x%0x, stag_idx 0x%x\n", - __func__, stag_state, type, pdid, stag_idx); + pr_debug("%s stag_state 0x%0x type 0x%0x pdid 0x%0x, stag_idx 0x%x\n", + __func__, stag_state, type, pdid, stag_idx); /* write TPT entry */ if (reset_tpt_entry) @@ -331,9 +329,9 @@ static int write_pbl(struct c4iw_rdev *rdev, __be64 *pbl, { int err; - PDBG("%s *pdb_addr 0x%x, pbl_base 0x%x, pbl_size %d\n", - __func__, pbl_addr, rdev->lldi.vr->pbl.start, - pbl_size); + pr_debug("%s *pdb_addr 0x%x, pbl_base 0x%x, pbl_size %d\n", + __func__, pbl_addr, rdev->lldi.vr->pbl.start, + pbl_size); err = write_adapter_mem(rdev, pbl_addr >> 5, pbl_size << 3, pbl, NULL); return err; @@ -376,7 +374,7 @@ static int finish_mem_reg(struct c4iw_mr *mhp, u32 stag) mhp->attr.stag = stag; mmid = stag >> 8; mhp->ibmr.rkey = mhp->ibmr.lkey = stag; - PDBG("%s mmid 0x%x mhp %p\n", __func__, mmid, mhp); + pr_debug("%s mmid 0x%x mhp %p\n", __func__, mmid, mhp); return insert_handle(mhp->rhp, &mhp->rhp->mmidr, mhp, mmid); } @@ -426,7 +424,7 @@ struct ib_mr *c4iw_get_dma_mr(struct ib_pd *pd, int acc) int ret; u32 stag = T4_STAG_UNSET; - PDBG("%s ib_pd %p\n", __func__, pd); + pr_debug("%s ib_pd %p\n", __func__, pd); php = to_c4iw_pd(pd); rhp = php->rhp; @@ -483,7 +481,7 @@ struct ib_mr *c4iw_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, struct c4iw_pd *php; struct c4iw_mr *mhp; - PDBG("%s ib_pd %p\n", __func__, pd); + pr_debug("%s ib_pd %p\n", __func__, pd); if (length == ~0ULL) return ERR_PTR(-EINVAL); @@ -517,7 +515,7 @@ struct ib_mr *c4iw_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, return ERR_PTR(err); } - shift = ffs(mhp->umem->page_size) - 1; + shift = mhp->umem->page_shift; n = mhp->umem->nmap; err = alloc_pbl(mhp, n); @@ -536,7 +534,7 @@ struct ib_mr *c4iw_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, len = sg_dma_len(sg) >> shift; for (k = 0; k < len; ++k) { pages[i++] = cpu_to_be64(sg_dma_address(sg) + - mhp->umem->page_size * k); + (k << shift)); if (i == PAGE_SIZE / sizeof *pages) { err = write_pbl(&mhp->rhp->rdev, pages, @@ -620,7 +618,7 @@ struct ib_mw *c4iw_alloc_mw(struct ib_pd *pd, enum ib_mw_type type, ret = -ENOMEM; goto dealloc_win; } - PDBG("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); + pr_debug("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); return &(mhp->ibmw); dealloc_win: @@ -645,7 +643,7 @@ int c4iw_dealloc_mw(struct ib_mw *mw) deallocate_window(&rhp->rdev, mhp->attr.stag, mhp->dereg_skb); kfree_skb(mhp->dereg_skb); kfree(mhp); - PDBG("%s ib_mw %p mmid 0x%x ptr %p\n", __func__, mw, mmid, mhp); + pr_debug("%s ib_mw %p mmid 0x%x ptr %p\n", __func__, mw, mmid, mhp); return 0; } @@ -703,7 +701,7 @@ struct ib_mr *c4iw_alloc_mr(struct ib_pd *pd, goto err3; } - PDBG("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); + pr_debug("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); return &(mhp->ibmr); err3: dereg_mem(&rhp->rdev, stag, mhp->attr.pbl_size, @@ -748,7 +746,7 @@ int c4iw_dereg_mr(struct ib_mr *ib_mr) struct c4iw_mr *mhp; u32 mmid; - PDBG("%s ib_mr %p\n", __func__, ib_mr); + pr_debug("%s ib_mr %p\n", __func__, ib_mr); mhp = to_c4iw_mr(ib_mr); rhp = mhp->rhp; @@ -766,7 +764,7 @@ int c4iw_dereg_mr(struct ib_mr *ib_mr) kfree((void *) (unsigned long) mhp->kva); if (mhp->umem) ib_umem_release(mhp->umem); - PDBG("%s mmid 0x%x ptr %p\n", __func__, mmid, mhp); + pr_debug("%s mmid 0x%x ptr %p\n", __func__, mmid, mhp); kfree(mhp); return 0; } diff --git a/drivers/infiniband/hw/cxgb4/provider.c b/drivers/infiniband/hw/cxgb4/provider.c index df64417ab6f2..0771e9a4d061 100644 --- a/drivers/infiniband/hw/cxgb4/provider.c +++ b/drivers/infiniband/hw/cxgb4/provider.c @@ -59,7 +59,7 @@ module_param(fastreg_support, int, 0644); MODULE_PARM_DESC(fastreg_support, "Advertise fastreg support (default=1)"); static struct ib_ah *c4iw_ah_create(struct ib_pd *pd, - struct ib_ah_attr *ah_attr, + struct rdma_ah_attr *ah_attr, struct ib_udata *udata) { @@ -102,7 +102,7 @@ void _c4iw_free_ucontext(struct kref *kref) ucontext = container_of(kref, struct c4iw_ucontext, kref); rhp = to_c4iw_dev(ucontext->ibucontext.device); - PDBG("%s ucontext %p\n", __func__, ucontext); + pr_debug("%s ucontext %p\n", __func__, ucontext); list_for_each_entry_safe(mm, tmp, &ucontext->mmaps, entry) kfree(mm); c4iw_release_dev_ucontext(&rhp->rdev, &ucontext->uctx); @@ -113,7 +113,7 @@ static int c4iw_dealloc_ucontext(struct ib_ucontext *context) { struct c4iw_ucontext *ucontext = to_c4iw_ucontext(context); - PDBG("%s context %p\n", __func__, context); + pr_debug("%s context %p\n", __func__, context); c4iw_put_ucontext(ucontext); return 0; } @@ -123,12 +123,11 @@ static struct ib_ucontext *c4iw_alloc_ucontext(struct ib_device *ibdev, { struct c4iw_ucontext *context; struct c4iw_dev *rhp = to_c4iw_dev(ibdev); - static int warned; struct c4iw_alloc_ucontext_resp uresp; int ret = 0; struct c4iw_mm_entry *mm = NULL; - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); context = kzalloc(sizeof(*context), GFP_KERNEL); if (!context) { ret = -ENOMEM; @@ -141,8 +140,7 @@ static struct ib_ucontext *c4iw_alloc_ucontext(struct ib_device *ibdev, kref_init(&context->kref); if (udata->outlen < sizeof(uresp) - sizeof(uresp.reserved)) { - if (!warned++) - pr_err(MOD "Warning - downlevel libcxgb4 (non-fatal), device status page disabled."); + pr_err_once("Warning - downlevel libcxgb4 (non-fatal), device status page disabled\n"); rhp->rdev.flags |= T4_STATUS_PAGE_DISABLED; } else { mm = kmalloc(sizeof(*mm), GFP_KERNEL); @@ -187,8 +185,8 @@ static int c4iw_mmap(struct ib_ucontext *context, struct vm_area_struct *vma) struct c4iw_ucontext *ucontext; u64 addr; - PDBG("%s pgoff 0x%lx key 0x%x len %d\n", __func__, vma->vm_pgoff, - key, len); + pr_debug("%s pgoff 0x%lx key 0x%x len %d\n", __func__, vma->vm_pgoff, + key, len); if (vma->vm_start & (PAGE_SIZE-1)) return -EINVAL; @@ -253,7 +251,7 @@ static int c4iw_deallocate_pd(struct ib_pd *pd) php = to_c4iw_pd(pd); rhp = php->rhp; - PDBG("%s ibpd %p pdid 0x%x\n", __func__, pd, php->pdid); + pr_debug("%s ibpd %p pdid 0x%x\n", __func__, pd, php->pdid); c4iw_put_resource(&rhp->rdev.resource.pdid_table, php->pdid); mutex_lock(&rhp->rdev.stats.lock); rhp->rdev.stats.pd.cur--; @@ -270,7 +268,7 @@ static struct ib_pd *c4iw_allocate_pd(struct ib_device *ibdev, u32 pdid; struct c4iw_dev *rhp; - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); rhp = (struct c4iw_dev *) ibdev; pdid = c4iw_get_resource(&rhp->rdev.resource.pdid_table); if (!pdid) @@ -293,14 +291,14 @@ static struct ib_pd *c4iw_allocate_pd(struct ib_device *ibdev, if (rhp->rdev.stats.pd.cur > rhp->rdev.stats.pd.max) rhp->rdev.stats.pd.max = rhp->rdev.stats.pd.cur; mutex_unlock(&rhp->rdev.stats.lock); - PDBG("%s pdid 0x%0x ptr 0x%p\n", __func__, pdid, php); + pr_debug("%s pdid 0x%0x ptr 0x%p\n", __func__, pdid, php); return &php->ibpd; } static int c4iw_query_pkey(struct ib_device *ibdev, u8 port, u16 index, u16 *pkey) { - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); *pkey = 0; return 0; } @@ -310,8 +308,8 @@ static int c4iw_query_gid(struct ib_device *ibdev, u8 port, int index, { struct c4iw_dev *dev; - PDBG("%s ibdev %p, port %d, index %d, gid %p\n", - __func__, ibdev, port, index, gid); + pr_debug("%s ibdev %p, port %d, index %d, gid %p\n", + __func__, ibdev, port, index, gid); dev = to_c4iw_dev(ibdev); BUG_ON(port == 0); memset(&(gid->raw[0]), 0, sizeof(gid->raw)); @@ -325,7 +323,7 @@ static int c4iw_query_device(struct ib_device *ibdev, struct ib_device_attr *pro struct c4iw_dev *dev; - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); if (uhw->inlen || uhw->outlen) return -EINVAL; @@ -366,7 +364,7 @@ static int c4iw_query_port(struct ib_device *ibdev, u8 port, struct net_device *netdev; struct in_device *inetdev; - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); dev = to_c4iw_dev(ibdev); netdev = dev->rdev.lldi.ports[port-1]; @@ -408,7 +406,7 @@ static ssize_t show_rev(struct device *dev, struct device_attribute *attr, { struct c4iw_dev *c4iw_dev = container_of(dev, struct c4iw_dev, ibdev.dev); - PDBG("%s dev 0x%p\n", __func__, dev); + pr_debug("%s dev 0x%p\n", __func__, dev); return sprintf(buf, "%d\n", CHELSIO_CHIP_RELEASE(c4iw_dev->rdev.lldi.adapter_type)); } @@ -421,7 +419,7 @@ static ssize_t show_hca(struct device *dev, struct device_attribute *attr, struct ethtool_drvinfo info; struct net_device *lldev = c4iw_dev->rdev.lldi.ports[0]; - PDBG("%s dev 0x%p\n", __func__, dev); + pr_debug("%s dev 0x%p\n", __func__, dev); lldev->ethtool_ops->get_drvinfo(lldev, &info); return sprintf(buf, "%s\n", info.driver); } @@ -431,7 +429,7 @@ static ssize_t show_board(struct device *dev, struct device_attribute *attr, { struct c4iw_dev *c4iw_dev = container_of(dev, struct c4iw_dev, ibdev.dev); - PDBG("%s dev 0x%p\n", __func__, dev); + pr_debug("%s dev 0x%p\n", __func__, dev); return sprintf(buf, "%x.%x\n", c4iw_dev->rdev.lldi.pdev->vendor, c4iw_dev->rdev.lldi.pdev->device); } @@ -524,7 +522,7 @@ static void get_dev_fw_str(struct ib_device *dev, char *str, { struct c4iw_dev *c4iw_dev = container_of(dev, struct c4iw_dev, ibdev); - PDBG("%s dev 0x%p\n", __func__, dev); + pr_debug("%s dev 0x%p\n", __func__, dev); snprintf(str, str_len, "%u.%u.%u.%u", FW_HDR_FW_VER_MAJOR_G(c4iw_dev->rdev.lldi.fw_vers), @@ -538,7 +536,7 @@ int c4iw_register_device(struct c4iw_dev *dev) int ret; int i; - PDBG("%s c4iw_dev %p\n", __func__, dev); + pr_debug("%s c4iw_dev %p\n", __func__, dev); BUG_ON(!dev->rdev.lldi.ports[0]); strlcpy(dev->ibdev.name, "cxgb4_%d", IB_DEVICE_NAME_MAX); memset(&dev->ibdev.node_guid, 0, sizeof(dev->ibdev.node_guid)); @@ -648,7 +646,7 @@ void c4iw_unregister_device(struct c4iw_dev *dev) { int i; - PDBG("%s c4iw_dev %p\n", __func__, dev); + pr_debug("%s c4iw_dev %p\n", __func__, dev); for (i = 0; i < ARRAY_SIZE(c4iw_class_attributes); ++i) device_remove_file(&dev->ibdev.dev, c4iw_class_attributes[i]); diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c index d4fd2f5c8326..8e4154b4253e 100644 --- a/drivers/infiniband/hw/cxgb4/qp.c +++ b/drivers/infiniband/hw/cxgb4/qp.c @@ -254,11 +254,11 @@ static int create_qp(struct c4iw_rdev *rdev, struct t4_wq *wq, ret = -ENOMEM; goto free_sq; } - PDBG("%s sq base va 0x%p pa 0x%llx rq base va 0x%p pa 0x%llx\n", - __func__, wq->sq.queue, - (unsigned long long)virt_to_phys(wq->sq.queue), - wq->rq.queue, - (unsigned long long)virt_to_phys(wq->rq.queue)); + pr_debug("%s sq base va 0x%p pa 0x%llx rq base va 0x%p pa 0x%llx\n", + __func__, wq->sq.queue, + (unsigned long long)virt_to_phys(wq->sq.queue), + wq->rq.queue, + (unsigned long long)virt_to_phys(wq->rq.queue)); memset(wq->rq.queue, 0, wq->rq.memsize); dma_unmap_addr_set(&wq->rq, mapping, wq->rq.dma_addr); @@ -275,7 +275,7 @@ static int create_qp(struct c4iw_rdev *rdev, struct t4_wq *wq, * User mode must have bar2 access. */ if (user && (!wq->sq.bar2_pa || !wq->rq.bar2_pa)) { - pr_warn(MOD "%s: sqid %u or rqid %u not in BAR2 range.\n", + pr_warn("%s: sqid %u or rqid %u not in BAR2 range\n", pci_name(rdev->lldi.pdev), wq->sq.qid, wq->rq.qid); goto free_dma; } @@ -362,9 +362,9 @@ static int create_qp(struct c4iw_rdev *rdev, struct t4_wq *wq, if (ret) goto free_dma; - PDBG("%s sqid 0x%x rqid 0x%x kdb 0x%p sq_bar2_addr %p rq_bar2_addr %p\n", - __func__, wq->sq.qid, wq->rq.qid, wq->db, - wq->sq.bar2_va, wq->rq.bar2_va); + pr_debug("%s sqid 0x%x rqid 0x%x kdb 0x%p sq_bar2_addr %p rq_bar2_addr %p\n", + __func__, wq->sq.qid, wq->rq.qid, wq->db, + wq->sq.bar2_va, wq->rq.bar2_va); return 0; free_dma: @@ -725,7 +725,7 @@ static void free_qp_work(struct work_struct *work) ucontext = qhp->ucontext; rhp = qhp->rhp; - PDBG("%s qhp %p ucontext %p\n", __func__, qhp, ucontext); + pr_debug("%s qhp %p ucontext %p\n", __func__, qhp, ucontext); destroy_qp(&rhp->rdev, &qhp->wq, ucontext ? &ucontext->uctx : &rhp->rdev.uctx); @@ -739,19 +739,19 @@ static void queue_qp_free(struct kref *kref) struct c4iw_qp *qhp; qhp = container_of(kref, struct c4iw_qp, kref); - PDBG("%s qhp %p\n", __func__, qhp); + pr_debug("%s qhp %p\n", __func__, qhp); queue_work(qhp->rhp->rdev.free_workq, &qhp->free_work); } void c4iw_qp_add_ref(struct ib_qp *qp) { - PDBG("%s ib_qp %p\n", __func__, qp); + pr_debug("%s ib_qp %p\n", __func__, qp); kref_get(&to_c4iw_qp(qp)->kref); } void c4iw_qp_rem_ref(struct ib_qp *qp) { - PDBG("%s ib_qp %p\n", __func__, qp); + pr_debug("%s ib_qp %p\n", __func__, qp); kref_put(&to_c4iw_qp(qp)->kref, queue_qp_free); } @@ -959,8 +959,8 @@ int c4iw_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, c4iw_invalidate_mr(qhp->rhp, wr->ex.invalidate_rkey); break; default: - PDBG("%s post of type=%d TBD!\n", __func__, - wr->opcode); + pr_debug("%s post of type=%d TBD!\n", __func__, + wr->opcode); err = -EINVAL; } if (err) { @@ -981,9 +981,10 @@ int c4iw_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, init_wr_hdr(wqe, qhp->wq.sq.pidx, fw_opcode, fw_flags, len16); - PDBG("%s cookie 0x%llx pidx 0x%x opcode 0x%x read_len %u\n", - __func__, (unsigned long long)wr->wr_id, qhp->wq.sq.pidx, - swsqe->opcode, swsqe->read_len); + pr_debug("%s cookie 0x%llx pidx 0x%x opcode 0x%x read_len %u\n", + __func__, + (unsigned long long)wr->wr_id, qhp->wq.sq.pidx, + swsqe->opcode, swsqe->read_len); wr = wr->next; num_wrs--; t4_sq_produce(&qhp->wq, len16); @@ -1057,8 +1058,9 @@ int c4iw_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr, wqe->recv.r2[1] = 0; wqe->recv.r2[2] = 0; wqe->recv.len16 = len16; - PDBG("%s cookie 0x%llx pidx %u\n", __func__, - (unsigned long long) wr->wr_id, qhp->wq.rq.pidx); + pr_debug("%s cookie 0x%llx pidx %u\n", + __func__, + (unsigned long long)wr->wr_id, qhp->wq.rq.pidx); t4_rq_produce(&qhp->wq, len16); idx += DIV_ROUND_UP(len16*16, T4_EQ_ENTRY_SIZE); wr = wr->next; @@ -1217,8 +1219,8 @@ static void post_terminate(struct c4iw_qp *qhp, struct t4_cqe *err_cqe, struct sk_buff *skb; struct terminate_message *term; - PDBG("%s qhp %p qid 0x%x tid %u\n", __func__, qhp, qhp->wq.sq.qid, - qhp->ep->hwtid); + pr_debug("%s qhp %p qid 0x%x tid %u\n", __func__, qhp, qhp->wq.sq.qid, + qhp->ep->hwtid); skb = skb_dequeue(&qhp->ep->com.ep_skb_list); if (WARN_ON(!skb)) @@ -1254,7 +1256,7 @@ static void __flush_qp(struct c4iw_qp *qhp, struct c4iw_cq *rchp, int rq_flushed, sq_flushed; unsigned long flag; - PDBG("%s qhp %p rchp %p schp %p\n", __func__, qhp, rchp, schp); + pr_debug("%s qhp %p rchp %p schp %p\n", __func__, qhp, rchp, schp); /* locking hierarchy: cq lock first, then qp lock. */ spin_lock_irqsave(&rchp->lock, flag); @@ -1339,8 +1341,8 @@ static int rdma_fini(struct c4iw_dev *rhp, struct c4iw_qp *qhp, int ret; struct sk_buff *skb; - PDBG("%s qhp %p qid 0x%x tid %u\n", __func__, qhp, qhp->wq.sq.qid, - ep->hwtid); + pr_debug("%s qhp %p qid 0x%x tid %u\n", __func__, qhp, qhp->wq.sq.qid, + ep->hwtid); skb = skb_dequeue(&ep->com.ep_skb_list); if (WARN_ON(!skb)) @@ -1366,13 +1368,13 @@ static int rdma_fini(struct c4iw_dev *rhp, struct c4iw_qp *qhp, ret = c4iw_wait_for_reply(&rhp->rdev, &ep->com.wr_wait, qhp->ep->hwtid, qhp->wq.sq.qid, __func__); out: - PDBG("%s ret %d\n", __func__, ret); + pr_debug("%s ret %d\n", __func__, ret); return ret; } static void build_rtr_msg(u8 p2p_type, struct fw_ri_init *init) { - PDBG("%s p2p_type = %d\n", __func__, p2p_type); + pr_debug("%s p2p_type = %d\n", __func__, p2p_type); memset(&init->u, 0, sizeof init->u); switch (p2p_type) { case FW_RI_INIT_P2PTYPE_RDMA_WRITE: @@ -1401,8 +1403,8 @@ static int rdma_init(struct c4iw_dev *rhp, struct c4iw_qp *qhp) int ret; struct sk_buff *skb; - PDBG("%s qhp %p qid 0x%x tid %u ird %u ord %u\n", __func__, qhp, - qhp->wq.sq.qid, qhp->ep->hwtid, qhp->ep->ird, qhp->ep->ord); + pr_debug("%s qhp %p qid 0x%x tid %u ird %u ord %u\n", __func__, qhp, + qhp->wq.sq.qid, qhp->ep->hwtid, qhp->ep->ird, qhp->ep->ord); skb = alloc_skb(sizeof *wqe, GFP_KERNEL); if (!skb) { @@ -1474,7 +1476,7 @@ static int rdma_init(struct c4iw_dev *rhp, struct c4iw_qp *qhp) err1: free_ird(rhp, qhp->attr.max_ird); out: - PDBG("%s ret %d\n", __func__, ret); + pr_debug("%s ret %d\n", __func__, ret); return ret; } @@ -1491,9 +1493,10 @@ int c4iw_modify_qp(struct c4iw_dev *rhp, struct c4iw_qp *qhp, int free = 0; struct c4iw_ep *ep = NULL; - PDBG("%s qhp %p sqid 0x%x rqid 0x%x ep %p state %d -> %d\n", __func__, - qhp, qhp->wq.sq.qid, qhp->wq.rq.qid, qhp->ep, qhp->attr.state, - (mask & C4IW_QP_ATTR_NEXT_STATE) ? attrs->next_state : -1); + pr_debug("%s qhp %p sqid 0x%x rqid 0x%x ep %p state %d -> %d\n", + __func__, + qhp, qhp->wq.sq.qid, qhp->wq.rq.qid, qhp->ep, qhp->attr.state, + (mask & C4IW_QP_ATTR_NEXT_STATE) ? attrs->next_state : -1); mutex_lock(&qhp->mutex); @@ -1671,16 +1674,15 @@ int c4iw_modify_qp(struct c4iw_dev *rhp, struct c4iw_qp *qhp, goto err; break; default: - printk(KERN_ERR "%s in a bad state %d\n", - __func__, qhp->attr.state); + pr_err("%s in a bad state %d\n", __func__, qhp->attr.state); ret = -EINVAL; goto err; break; } goto out; err: - PDBG("%s disassociating ep %p qpid 0x%x\n", __func__, qhp->ep, - qhp->wq.sq.qid); + pr_debug("%s disassociating ep %p qpid 0x%x\n", __func__, qhp->ep, + qhp->wq.sq.qid); /* disassociate the LLP connection */ qhp->attr.llp_stream_handle = NULL; @@ -1716,7 +1718,7 @@ out: */ if (free) c4iw_put_ep(&ep->com); - PDBG("%s exit state %d\n", __func__, qhp->attr.state); + pr_debug("%s exit state %d\n", __func__, qhp->attr.state); return ret; } @@ -1746,7 +1748,7 @@ int c4iw_destroy_qp(struct ib_qp *ib_qp) c4iw_qp_rem_ref(ib_qp); - PDBG("%s ib_qp %p qpid 0x%0x\n", __func__, ib_qp, qhp->wq.sq.qid); + pr_debug("%s ib_qp %p qpid 0x%0x\n", __func__, ib_qp, qhp->wq.sq.qid); return 0; } @@ -1765,7 +1767,7 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs, struct c4iw_mm_entry *sq_key_mm, *rq_key_mm = NULL, *sq_db_key_mm; struct c4iw_mm_entry *rq_db_key_mm = NULL, *ma_sync_key_mm = NULL; - PDBG("%s ib_pd %p\n", __func__, pd); + pr_debug("%s ib_pd %p\n", __func__, pd); if (attrs->qp_type != IB_QPT_RC) return ERR_PTR(-EINVAL); @@ -1936,11 +1938,11 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs, qhp->ibqp.qp_num = qhp->wq.sq.qid; init_timer(&(qhp->timer)); INIT_LIST_HEAD(&qhp->db_fc_entry); - PDBG("%s sq id %u size %u memsize %zu num_entries %u " - "rq id %u size %u memsize %zu num_entries %u\n", __func__, - qhp->wq.sq.qid, qhp->wq.sq.size, qhp->wq.sq.memsize, - attrs->cap.max_send_wr, qhp->wq.rq.qid, qhp->wq.rq.size, - qhp->wq.rq.memsize, attrs->cap.max_recv_wr); + pr_debug("%s sq id %u size %u memsize %zu num_entries %u rq id %u size %u memsize %zu num_entries %u\n", + __func__, + qhp->wq.sq.qid, qhp->wq.sq.size, qhp->wq.sq.memsize, + attrs->cap.max_send_wr, qhp->wq.rq.qid, qhp->wq.rq.size, + qhp->wq.rq.memsize, attrs->cap.max_recv_wr); return &qhp->ibqp; err8: kfree(ma_sync_key_mm); @@ -1970,7 +1972,7 @@ int c4iw_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, enum c4iw_qp_attr_mask mask = 0; struct c4iw_qp_attributes attrs; - PDBG("%s ib_qp %p\n", __func__, ibqp); + pr_debug("%s ib_qp %p\n", __func__, ibqp); /* iwarp does not support the RTR state */ if ((attr_mask & IB_QP_STATE) && (attr->qp_state == IB_QPS_RTR)) @@ -2016,7 +2018,7 @@ int c4iw_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, struct ib_qp *c4iw_get_qp(struct ib_device *dev, int qpn) { - PDBG("%s ib_dev %p qpn 0x%x\n", __func__, dev, qpn); + pr_debug("%s ib_dev %p qpn 0x%x\n", __func__, dev, qpn); return (struct ib_qp *)get_qhp(to_c4iw_dev(dev), qpn); } diff --git a/drivers/infiniband/hw/cxgb4/resource.c b/drivers/infiniband/hw/cxgb4/resource.c index 67df71a7012e..8ff0cbe5cb16 100644 --- a/drivers/infiniband/hw/cxgb4/resource.c +++ b/drivers/infiniband/hw/cxgb4/resource.c @@ -90,7 +90,7 @@ u32 c4iw_get_resource(struct c4iw_id_table *id_table) void c4iw_put_resource(struct c4iw_id_table *id_table, u32 entry) { - PDBG("%s entry 0x%x\n", __func__, entry); + pr_debug("%s entry 0x%x\n", __func__, entry); c4iw_id_free(id_table, entry); } @@ -141,7 +141,7 @@ u32 c4iw_get_cqid(struct c4iw_rdev *rdev, struct c4iw_dev_ucontext *uctx) } out: mutex_unlock(&uctx->lock); - PDBG("%s qid 0x%x\n", __func__, qid); + pr_debug("%s qid 0x%x\n", __func__, qid); mutex_lock(&rdev->stats.lock); if (rdev->stats.qid.cur > rdev->stats.qid.max) rdev->stats.qid.max = rdev->stats.qid.cur; @@ -157,7 +157,7 @@ void c4iw_put_cqid(struct c4iw_rdev *rdev, u32 qid, entry = kmalloc(sizeof *entry, GFP_KERNEL); if (!entry) return; - PDBG("%s qid 0x%x\n", __func__, qid); + pr_debug("%s qid 0x%x\n", __func__, qid); entry->qid = qid; mutex_lock(&uctx->lock); list_add_tail(&entry->entry, &uctx->cqids); @@ -215,7 +215,7 @@ u32 c4iw_get_qpid(struct c4iw_rdev *rdev, struct c4iw_dev_ucontext *uctx) } out: mutex_unlock(&uctx->lock); - PDBG("%s qid 0x%x\n", __func__, qid); + pr_debug("%s qid 0x%x\n", __func__, qid); mutex_lock(&rdev->stats.lock); if (rdev->stats.qid.cur > rdev->stats.qid.max) rdev->stats.qid.max = rdev->stats.qid.cur; @@ -231,7 +231,7 @@ void c4iw_put_qpid(struct c4iw_rdev *rdev, u32 qid, entry = kmalloc(sizeof *entry, GFP_KERNEL); if (!entry) return; - PDBG("%s qid 0x%x\n", __func__, qid); + pr_debug("%s qid 0x%x\n", __func__, qid); entry->qid = qid; mutex_lock(&uctx->lock); list_add_tail(&entry->entry, &uctx->qpids); @@ -254,7 +254,7 @@ void c4iw_destroy_resource(struct c4iw_resource *rscp) u32 c4iw_pblpool_alloc(struct c4iw_rdev *rdev, int size) { unsigned long addr = gen_pool_alloc(rdev->pbl_pool, size); - PDBG("%s addr 0x%x size %d\n", __func__, (u32)addr, size); + pr_debug("%s addr 0x%x size %d\n", __func__, (u32)addr, size); mutex_lock(&rdev->stats.lock); if (addr) { rdev->stats.pbl.cur += roundup(size, 1 << MIN_PBL_SHIFT); @@ -268,7 +268,7 @@ u32 c4iw_pblpool_alloc(struct c4iw_rdev *rdev, int size) void c4iw_pblpool_free(struct c4iw_rdev *rdev, u32 addr, int size) { - PDBG("%s addr 0x%x size %d\n", __func__, addr, size); + pr_debug("%s addr 0x%x size %d\n", __func__, addr, size); mutex_lock(&rdev->stats.lock); rdev->stats.pbl.cur -= roundup(size, 1 << MIN_PBL_SHIFT); mutex_unlock(&rdev->stats.lock); @@ -290,19 +290,17 @@ int c4iw_pblpool_create(struct c4iw_rdev *rdev) while (pbl_start < pbl_top) { pbl_chunk = min(pbl_top - pbl_start + 1, pbl_chunk); if (gen_pool_add(rdev->pbl_pool, pbl_start, pbl_chunk, -1)) { - PDBG("%s failed to add PBL chunk (%x/%x)\n", - __func__, pbl_start, pbl_chunk); + pr_debug("%s failed to add PBL chunk (%x/%x)\n", + __func__, pbl_start, pbl_chunk); if (pbl_chunk <= 1024 << MIN_PBL_SHIFT) { - printk(KERN_WARNING MOD - "Failed to add all PBL chunks (%x/%x)\n", - pbl_start, - pbl_top - pbl_start); + pr_warn("Failed to add all PBL chunks (%x/%x)\n", + pbl_start, pbl_top - pbl_start); return 0; } pbl_chunk >>= 1; } else { - PDBG("%s added PBL chunk (%x/%x)\n", - __func__, pbl_start, pbl_chunk); + pr_debug("%s added PBL chunk (%x/%x)\n", + __func__, pbl_start, pbl_chunk); pbl_start += pbl_chunk; } } @@ -324,9 +322,9 @@ void c4iw_pblpool_destroy(struct c4iw_rdev *rdev) u32 c4iw_rqtpool_alloc(struct c4iw_rdev *rdev, int size) { unsigned long addr = gen_pool_alloc(rdev->rqt_pool, size << 6); - PDBG("%s addr 0x%x size %d\n", __func__, (u32)addr, size << 6); + pr_debug("%s addr 0x%x size %d\n", __func__, (u32)addr, size << 6); if (!addr) - pr_warn_ratelimited(MOD "%s: Out of RQT memory\n", + pr_warn_ratelimited("%s: Out of RQT memory\n", pci_name(rdev->lldi.pdev)); mutex_lock(&rdev->stats.lock); if (addr) { @@ -341,7 +339,7 @@ u32 c4iw_rqtpool_alloc(struct c4iw_rdev *rdev, int size) void c4iw_rqtpool_free(struct c4iw_rdev *rdev, u32 addr, int size) { - PDBG("%s addr 0x%x size %d\n", __func__, addr, size << 6); + pr_debug("%s addr 0x%x size %d\n", __func__, addr, size << 6); mutex_lock(&rdev->stats.lock); rdev->stats.rqt.cur -= roundup(size << 6, 1 << MIN_RQT_SHIFT); mutex_unlock(&rdev->stats.lock); @@ -363,18 +361,17 @@ int c4iw_rqtpool_create(struct c4iw_rdev *rdev) while (rqt_start < rqt_top) { rqt_chunk = min(rqt_top - rqt_start + 1, rqt_chunk); if (gen_pool_add(rdev->rqt_pool, rqt_start, rqt_chunk, -1)) { - PDBG("%s failed to add RQT chunk (%x/%x)\n", - __func__, rqt_start, rqt_chunk); + pr_debug("%s failed to add RQT chunk (%x/%x)\n", + __func__, rqt_start, rqt_chunk); if (rqt_chunk <= 1024 << MIN_RQT_SHIFT) { - printk(KERN_WARNING MOD - "Failed to add all RQT chunks (%x/%x)\n", - rqt_start, rqt_top - rqt_start); + pr_warn("Failed to add all RQT chunks (%x/%x)\n", + rqt_start, rqt_top - rqt_start); return 0; } rqt_chunk >>= 1; } else { - PDBG("%s added RQT chunk (%x/%x)\n", - __func__, rqt_start, rqt_chunk); + pr_debug("%s added RQT chunk (%x/%x)\n", + __func__, rqt_start, rqt_chunk); rqt_start += rqt_chunk; } } @@ -394,7 +391,7 @@ void c4iw_rqtpool_destroy(struct c4iw_rdev *rdev) u32 c4iw_ocqp_pool_alloc(struct c4iw_rdev *rdev, int size) { unsigned long addr = gen_pool_alloc(rdev->ocqp_pool, size); - PDBG("%s addr 0x%x size %d\n", __func__, (u32)addr, size); + pr_debug("%s addr 0x%x size %d\n", __func__, (u32)addr, size); if (addr) { mutex_lock(&rdev->stats.lock); rdev->stats.ocqp.cur += roundup(size, 1 << MIN_OCQP_SHIFT); @@ -407,7 +404,7 @@ u32 c4iw_ocqp_pool_alloc(struct c4iw_rdev *rdev, int size) void c4iw_ocqp_pool_free(struct c4iw_rdev *rdev, u32 addr, int size) { - PDBG("%s addr 0x%x size %d\n", __func__, addr, size); + pr_debug("%s addr 0x%x size %d\n", __func__, addr, size); mutex_lock(&rdev->stats.lock); rdev->stats.ocqp.cur -= roundup(size, 1 << MIN_OCQP_SHIFT); mutex_unlock(&rdev->stats.lock); @@ -429,18 +426,17 @@ int c4iw_ocqp_pool_create(struct c4iw_rdev *rdev) while (start < top) { chunk = min(top - start + 1, chunk); if (gen_pool_add(rdev->ocqp_pool, start, chunk, -1)) { - PDBG("%s failed to add OCQP chunk (%x/%x)\n", - __func__, start, chunk); + pr_debug("%s failed to add OCQP chunk (%x/%x)\n", + __func__, start, chunk); if (chunk <= 1024 << MIN_OCQP_SHIFT) { - printk(KERN_WARNING MOD - "Failed to add all OCQP chunks (%x/%x)\n", - start, top - start); + pr_warn("Failed to add all OCQP chunks (%x/%x)\n", + start, top - start); return 0; } chunk >>= 1; } else { - PDBG("%s added OCQP chunk (%x/%x)\n", - __func__, start, chunk); + pr_debug("%s added OCQP chunk (%x/%x)\n", + __func__, start, chunk); start += chunk; } } diff --git a/drivers/infiniband/hw/cxgb4/t4.h b/drivers/infiniband/hw/cxgb4/t4.h index 640d22148a3e..e765c00303cd 100644 --- a/drivers/infiniband/hw/cxgb4/t4.h +++ b/drivers/infiniband/hw/cxgb4/t4.h @@ -466,14 +466,14 @@ static inline void t4_ring_sq_db(struct t4_wq *wq, u16 inc, union t4_wr *wqe) wmb(); if (wq->sq.bar2_va) { if (inc == 1 && wq->sq.bar2_qid == 0 && wqe) { - PDBG("%s: WC wq->sq.pidx = %d\n", - __func__, wq->sq.pidx); + pr_debug("%s: WC wq->sq.pidx = %d\n", + __func__, wq->sq.pidx); pio_copy((u64 __iomem *) (wq->sq.bar2_va + SGE_UDB_WCDOORBELL), (u64 *)wqe); } else { - PDBG("%s: DB wq->sq.pidx = %d\n", - __func__, wq->sq.pidx); + pr_debug("%s: DB wq->sq.pidx = %d\n", + __func__, wq->sq.pidx); writel(PIDX_T5_V(inc) | QID_V(wq->sq.bar2_qid), wq->sq.bar2_va + SGE_UDB_KDOORBELL); } @@ -493,14 +493,14 @@ static inline void t4_ring_rq_db(struct t4_wq *wq, u16 inc, wmb(); if (wq->rq.bar2_va) { if (inc == 1 && wq->rq.bar2_qid == 0 && wqe) { - PDBG("%s: WC wq->rq.pidx = %d\n", - __func__, wq->rq.pidx); + pr_debug("%s: WC wq->rq.pidx = %d\n", + __func__, wq->rq.pidx); pio_copy((u64 __iomem *) (wq->rq.bar2_va + SGE_UDB_WCDOORBELL), (void *)wqe); } else { - PDBG("%s: DB wq->rq.pidx = %d\n", - __func__, wq->rq.pidx); + pr_debug("%s: DB wq->rq.pidx = %d\n", + __func__, wq->rq.pidx); writel(PIDX_T5_V(inc) | QID_V(wq->rq.bar2_qid), wq->rq.bar2_va + SGE_UDB_KDOORBELL); } @@ -601,7 +601,8 @@ static inline void t4_swcq_produce(struct t4_cq *cq) { cq->sw_in_use++; if (cq->sw_in_use == cq->size) { - PDBG("%s cxgb4 sw cq overflow cqid %u\n", __func__, cq->cqid); + pr_debug("%s cxgb4 sw cq overflow cqid %u\n", + __func__, cq->cqid); cq->error = 1; BUG_ON(1); } @@ -656,7 +657,7 @@ static inline int t4_next_hw_cqe(struct t4_cq *cq, struct t4_cqe **cqe) if (cq->queue[prev_cidx].bits_type_ts != cq->bits_type_ts) { ret = -EOVERFLOW; cq->error = 1; - printk(KERN_ERR MOD "cq overflow cqid %u\n", cq->cqid); + pr_err("cq overflow cqid %u\n", cq->cqid); BUG_ON(1); } else if (t4_valid_cqe(cq, &cq->queue[cq->cidx])) { @@ -672,7 +673,8 @@ static inline int t4_next_hw_cqe(struct t4_cq *cq, struct t4_cqe **cqe) static inline struct t4_cqe *t4_next_sw_cqe(struct t4_cq *cq) { if (cq->sw_in_use == cq->size) { - PDBG("%s cxgb4 sw cq overflow cqid %u\n", __func__, cq->cqid); + pr_debug("%s cxgb4 sw cq overflow cqid %u\n", + __func__, cq->cqid); cq->error = 1; BUG_ON(1); return NULL; diff --git a/drivers/infiniband/hw/hfi1/Makefile b/drivers/infiniband/hw/hfi1/Makefile index 0cf97a09b64b..88085f65432e 100644 --- a/drivers/infiniband/hw/hfi1/Makefile +++ b/drivers/infiniband/hw/hfi1/Makefile @@ -12,7 +12,7 @@ hfi1-y := affinity.o chip.o device.o driver.o efivar.o \ init.o intr.o mad.o mmu_rb.o pcie.o pio.o pio_copy.o platform.o \ qp.o qsfp.o rc.o ruc.o sdma.o sysfs.o trace.o \ uc.o ud.o user_exp_rcv.o user_pages.o user_sdma.o verbs.o \ - verbs_txreq.o + verbs_txreq.o vnic_main.o vnic_sdma.o hfi1-$(CONFIG_DEBUG_FS) += debugfs.o CFLAGS_trace.o = -I$(src) diff --git a/drivers/infiniband/hw/hfi1/aspm.h b/drivers/infiniband/hw/hfi1/aspm.h index 0d58fe3b49b5..794e6814a531 100644 --- a/drivers/infiniband/hw/hfi1/aspm.h +++ b/drivers/infiniband/hw/hfi1/aspm.h @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -229,14 +229,17 @@ static inline void aspm_ctx_timer_function(unsigned long data) spin_unlock_irqrestore(&rcd->aspm_lock, flags); } -/* Disable interrupt processing for verbs contexts when PSM contexts are open */ +/* + * Disable interrupt processing for verbs contexts when PSM or VNIC contexts + * are open. + */ static inline void aspm_disable_all(struct hfi1_devdata *dd) { struct hfi1_ctxtdata *rcd; unsigned long flags; unsigned i; - for (i = 0; i < dd->first_user_ctxt; i++) { + for (i = 0; i < dd->first_dyn_alloc_ctxt; i++) { rcd = dd->rcd[i]; del_timer_sync(&rcd->aspm_timer); spin_lock_irqsave(&rcd->aspm_lock, flags); @@ -260,7 +263,7 @@ static inline void aspm_enable_all(struct hfi1_devdata *dd) if (aspm_mode != ASPM_MODE_DYNAMIC) return; - for (i = 0; i < dd->first_user_ctxt; i++) { + for (i = 0; i < dd->first_dyn_alloc_ctxt; i++) { rcd = dd->rcd[i]; spin_lock_irqsave(&rcd->aspm_lock, flags); rcd->aspm_intr_enable = true; @@ -276,7 +279,7 @@ static inline void aspm_ctx_init(struct hfi1_ctxtdata *rcd) (unsigned long)rcd); rcd->aspm_intr_supported = rcd->dd->aspm_supported && aspm_mode == ASPM_MODE_DYNAMIC && - rcd->ctxt < rcd->dd->first_user_ctxt; + rcd->ctxt < rcd->dd->first_dyn_alloc_ctxt; } static inline void aspm_init(struct hfi1_devdata *dd) @@ -286,7 +289,7 @@ static inline void aspm_init(struct hfi1_devdata *dd) spin_lock_init(&dd->aspm_lock); dd->aspm_supported = aspm_hw_l1_supported(dd); - for (i = 0; i < dd->first_user_ctxt; i++) + for (i = 0; i < dd->first_dyn_alloc_ctxt; i++) aspm_ctx_init(dd->rcd[i]); /* Start with ASPM disabled */ diff --git a/drivers/infiniband/hw/hfi1/chip.c b/drivers/infiniband/hw/hfi1/chip.c index 121a4c920f1b..0f6916d2d549 100644 --- a/drivers/infiniband/hw/hfi1/chip.c +++ b/drivers/infiniband/hw/hfi1/chip.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015 - 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -64,6 +64,7 @@ #include "platform.h" #include "aspm.h" #include "affinity.h" +#include "debugfs.h" #define NUM_IB_PORTS 1 @@ -125,9 +126,16 @@ struct flag_table { #define DEFAULT_KRCVQS 2 #define MIN_KERNEL_KCTXTS 2 #define FIRST_KERNEL_KCTXT 1 -/* sizes for both the QP and RSM map tables */ -#define NUM_MAP_ENTRIES 256 -#define NUM_MAP_REGS 32 + +/* + * RSM instance allocation + * 0 - Verbs + * 1 - User Fecn Handling + * 2 - Vnic + */ +#define RSM_INS_VERBS 0 +#define RSM_INS_FECN 1 +#define RSM_INS_VNIC 2 /* Bit offset into the GUID which carries HFI id information */ #define GUID_HFI_INDEX_SHIFT 39 @@ -138,8 +146,7 @@ struct flag_table { #define is_emulator_p(dd) ((((dd)->irev) & 0xf) == 3) #define is_emulator_s(dd) ((((dd)->irev) & 0xf) == 4) -/* RSM fields */ - +/* RSM fields for Verbs */ /* packet type */ #define IB_PACKET_TYPE 2ull #define QW_SHIFT 6ull @@ -169,6 +176,28 @@ struct flag_table { /* QPN[m+n:1] QW 1, OFFSET 1 */ #define QPN_SELECT_OFFSET ((1ull << QW_SHIFT) | (1ull)) +/* RSM fields for Vnic */ +/* L2_TYPE: QW 0, OFFSET 61 - for match */ +#define L2_TYPE_QW 0ull +#define L2_TYPE_BIT_OFFSET 61ull +#define L2_TYPE_OFFSET(off) ((L2_TYPE_QW << QW_SHIFT) | (off)) +#define L2_TYPE_MATCH_OFFSET L2_TYPE_OFFSET(L2_TYPE_BIT_OFFSET) +#define L2_TYPE_MASK 3ull +#define L2_16B_VALUE 2ull + +/* L4_TYPE QW 1, OFFSET 0 - for match */ +#define L4_TYPE_QW 1ull +#define L4_TYPE_BIT_OFFSET 0ull +#define L4_TYPE_OFFSET(off) ((L4_TYPE_QW << QW_SHIFT) | (off)) +#define L4_TYPE_MATCH_OFFSET L4_TYPE_OFFSET(L4_TYPE_BIT_OFFSET) +#define L4_16B_TYPE_MASK 0xFFull +#define L4_16B_ETH_VALUE 0x78ull + +/* 16B VESWID - for select */ +#define L4_16B_HDR_VESWID_OFFSET ((2 << QW_SHIFT) | (16ull)) +/* 16B ENTROPY - for select */ +#define L2_16B_ENTROPY_OFFSET ((1 << QW_SHIFT) | (32ull)) + /* defines to build power on SC2VL table */ #define SC2VL_VAL( \ num, \ @@ -1045,6 +1074,8 @@ static void dc_start(struct hfi1_devdata *); static int qos_rmt_entries(struct hfi1_devdata *dd, unsigned int *mp, unsigned int *np); static void clear_full_mgmt_pkey(struct hfi1_pportdata *ppd); +static int wait_link_transfer_active(struct hfi1_devdata *dd, int wait_ms); +static void clear_rsm_rule(struct hfi1_devdata *dd, u8 rule_index); /* * Error interrupt table entry. This is used as input to the interrupt @@ -6379,18 +6410,17 @@ static void lcb_shutdown(struct hfi1_devdata *dd, int abort) * * The expectation is that the caller of this routine would have taken * care of properly transitioning the link into the correct state. + * NOTE: the caller needs to acquire the dd->dc8051_lock lock + * before calling this function. */ -static void dc_shutdown(struct hfi1_devdata *dd) +static void _dc_shutdown(struct hfi1_devdata *dd) { - unsigned long flags; + lockdep_assert_held(&dd->dc8051_lock); - spin_lock_irqsave(&dd->dc8051_lock, flags); - if (dd->dc_shutdown) { - spin_unlock_irqrestore(&dd->dc8051_lock, flags); + if (dd->dc_shutdown) return; - } + dd->dc_shutdown = 1; - spin_unlock_irqrestore(&dd->dc8051_lock, flags); /* Shutdown the LCB */ lcb_shutdown(dd, 1); /* @@ -6401,35 +6431,45 @@ static void dc_shutdown(struct hfi1_devdata *dd) write_csr(dd, DC_DC8051_CFG_RST, 0x1); } +static void dc_shutdown(struct hfi1_devdata *dd) +{ + mutex_lock(&dd->dc8051_lock); + _dc_shutdown(dd); + mutex_unlock(&dd->dc8051_lock); +} + /* * Calling this after the DC has been brought out of reset should not * do any damage. + * NOTE: the caller needs to acquire the dd->dc8051_lock lock + * before calling this function. */ -static void dc_start(struct hfi1_devdata *dd) +static void _dc_start(struct hfi1_devdata *dd) { - unsigned long flags; - int ret; + lockdep_assert_held(&dd->dc8051_lock); - spin_lock_irqsave(&dd->dc8051_lock, flags); if (!dd->dc_shutdown) - goto done; - spin_unlock_irqrestore(&dd->dc8051_lock, flags); + return; + /* Take the 8051 out of reset */ write_csr(dd, DC_DC8051_CFG_RST, 0ull); /* Wait until 8051 is ready */ - ret = wait_fm_ready(dd, TIMEOUT_8051_START); - if (ret) { + if (wait_fm_ready(dd, TIMEOUT_8051_START)) dd_dev_err(dd, "%s: timeout starting 8051 firmware\n", __func__); - } + /* Take away reset for LCB and RX FPE (set in lcb_shutdown). */ write_csr(dd, DCC_CFG_RESET, 0x10); /* lcb_shutdown() with abort=1 does not restore these */ write_csr(dd, DC_LCB_ERR_EN, dd->lcb_err_en); - spin_lock_irqsave(&dd->dc8051_lock, flags); dd->dc_shutdown = 0; -done: - spin_unlock_irqrestore(&dd->dc8051_lock, flags); +} + +static void dc_start(struct hfi1_devdata *dd) +{ + mutex_lock(&dd->dc8051_lock); + _dc_start(dd); + mutex_unlock(&dd->dc8051_lock); } /* @@ -6701,7 +6741,13 @@ static void rxe_kernel_unfreeze(struct hfi1_devdata *dd) int i; /* enable all kernel contexts */ - for (i = 0; i < dd->n_krcv_queues; i++) { + for (i = 0; i < dd->num_rcv_contexts; i++) { + struct hfi1_ctxtdata *rcd = dd->rcd[i]; + + /* Ensure all non-user contexts(including vnic) are enabled */ + if (!rcd || !rcd->sc || (rcd->sc->type == SC_USER)) + continue; + rcvmask = HFI1_RCVCTRL_CTXT_ENB; /* HFI1_RCVCTRL_TAILUPD_[ENB|DIS] needs to be set explicitly */ rcvmask |= HFI1_CAP_KGET_MASK(dd->rcd[i]->flags, DMA_RTAIL) ? @@ -7077,7 +7123,7 @@ static void add_full_mgmt_pkey(struct hfi1_pportdata *ppd) { struct hfi1_devdata *dd = ppd->dd; - /* Sanity check - ppd->pkeys[2] should be 0, or already initalized */ + /* Sanity check - ppd->pkeys[2] should be 0, or already initialized */ if (!((ppd->pkeys[2] == 0) || (ppd->pkeys[2] == FULL_MGMT_P_KEY))) dd_dev_warn(dd, "%s pkey[2] already set to 0x%x, resetting it to 0x%x\n", __func__, ppd->pkeys[2], FULL_MGMT_P_KEY); @@ -7165,7 +7211,7 @@ static void get_link_widths(struct hfi1_devdata *dd, u16 *tx_width, * set the max_rate field in handle_verify_cap until v0.19. */ if ((dd->icode == ICODE_RTL_SILICON) && - (dd->dc8051_ver < dc8051_ver(0, 19))) { + (dd->dc8051_ver < dc8051_ver(0, 19, 0))) { /* max_rate: 0 = 12.5G, 1 = 25G */ switch (max_rate) { case 0: @@ -7277,15 +7323,6 @@ void handle_verify_cap(struct work_struct *work) lcb_shutdown(dd, 0); adjust_lcb_for_fpga_serdes(dd); - /* - * These are now valid: - * remote VerifyCap fields in the general LNI config - * CSR DC8051_STS_REMOTE_GUID - * CSR DC8051_STS_REMOTE_NODE_TYPE - * CSR DC8051_STS_REMOTE_FM_SECURITY - * CSR DC8051_STS_REMOTE_PORT_NO - */ - read_vc_remote_phy(dd, &power_management, &continious); read_vc_remote_fabric(dd, &vau, &z, &vcu, &vl15buf, &partner_supported_crc); @@ -7350,7 +7387,7 @@ void handle_verify_cap(struct work_struct *work) } ppd->link_speed_active = 0; /* invalid value */ - if (dd->dc8051_ver < dc8051_ver(0, 20)) { + if (dd->dc8051_ver < dc8051_ver(0, 20, 0)) { /* remote_tx_rate: 0 = 12.5G, 1 = 25G */ switch (remote_tx_rate) { case 0: @@ -7416,20 +7453,6 @@ void handle_verify_cap(struct work_struct *work) write_csr(dd, DC_LCB_ERR_EN, 0); /* mask LCB errors */ set_8051_lcb_access(dd); - ppd->neighbor_guid = - read_csr(dd, DC_DC8051_STS_REMOTE_GUID); - ppd->neighbor_port_number = read_csr(dd, DC_DC8051_STS_REMOTE_PORT_NO) & - DC_DC8051_STS_REMOTE_PORT_NO_VAL_SMASK; - ppd->neighbor_type = - read_csr(dd, DC_DC8051_STS_REMOTE_NODE_TYPE) & - DC_DC8051_STS_REMOTE_NODE_TYPE_VAL_MASK; - ppd->neighbor_fm_security = - read_csr(dd, DC_DC8051_STS_REMOTE_FM_SECURITY) & - DC_DC8051_STS_LOCAL_FM_SECURITY_DISABLED_MASK; - dd_dev_info(dd, - "Neighbor Guid: %llx Neighbor type %d MgmtAllowed %d FM security bypass %d\n", - ppd->neighbor_guid, ppd->neighbor_type, - ppd->mgmt_allowed, ppd->neighbor_fm_security); if (ppd->mgmt_allowed) add_full_mgmt_pkey(ppd); @@ -7897,6 +7920,9 @@ static void handle_dcc_err(struct hfi1_devdata *dd, u32 unused, u64 reg) reg &= ~DCC_ERR_FLG_EN_CSR_ACCESS_BLOCKED_HOST_SMASK; } + if (unlikely(hfi1_dbg_fault_suppress_err(&dd->verbs_dev))) + reg &= ~DCC_ERR_FLG_LATE_EBP_ERR_SMASK; + /* report any remaining errors */ if (reg) dd_dev_info_ratelimited(dd, "DCC Error: %s\n", @@ -7995,7 +8021,9 @@ static void is_rcv_avail_int(struct hfi1_devdata *dd, unsigned int source) if (likely(source < dd->num_rcv_contexts)) { rcd = dd->rcd[source]; if (rcd) { - if (source < dd->first_user_ctxt) + /* Check for non-user contexts, including vnic */ + if ((source < dd->first_dyn_alloc_ctxt) || + (rcd->sc && (rcd->sc->type == SC_KERNEL))) rcd->do_interrupt(rcd, 0); else handle_user_interrupt(rcd); @@ -8023,7 +8051,8 @@ static void is_rcv_urgent_int(struct hfi1_devdata *dd, unsigned int source) rcd = dd->rcd[source]; if (rcd) { /* only pay attention to user urgent interrupts */ - if (source >= dd->first_user_ctxt) + if ((source >= dd->first_dyn_alloc_ctxt) && + (!rcd->sc || (rcd->sc->type == SC_USER))) handle_user_interrupt(rcd); return; /* OK */ } @@ -8156,10 +8185,10 @@ static irqreturn_t sdma_interrupt(int irq, void *data) /* handle the interrupt(s) */ sdma_engine_interrupt(sde, status); - } else + } else { dd_dev_err(dd, "SDMA engine %u interrupt, but no status bits set\n", sde->this_idx); - + } return IRQ_HANDLED; } @@ -8344,6 +8373,52 @@ static int read_lcb_via_8051(struct hfi1_devdata *dd, u32 addr, u64 *data) } /* + * Provide a cache for some of the LCB registers in case the LCB is + * unavailable. + * (The LCB is unavailable in certain link states, for example.) + */ +struct lcb_datum { + u32 off; + u64 val; +}; + +static struct lcb_datum lcb_cache[] = { + { DC_LCB_ERR_INFO_RX_REPLAY_CNT, 0}, + { DC_LCB_ERR_INFO_SEQ_CRC_CNT, 0 }, + { DC_LCB_ERR_INFO_REINIT_FROM_PEER_CNT, 0 }, +}; + +static void update_lcb_cache(struct hfi1_devdata *dd) +{ + int i; + int ret; + u64 val; + + for (i = 0; i < ARRAY_SIZE(lcb_cache); i++) { + ret = read_lcb_csr(dd, lcb_cache[i].off, &val); + + /* Update if we get good data */ + if (likely(ret != -EBUSY)) + lcb_cache[i].val = val; + } +} + +static int read_lcb_cache(u32 off, u64 *val) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(lcb_cache); i++) { + if (lcb_cache[i].off == off) { + *val = lcb_cache[i].val; + return 0; + } + } + + pr_warn("%s bad offset 0x%x\n", __func__, off); + return -1; +} + +/* * Read an LCB CSR. Access may not be in host control, so check. * Return 0 on success, -EBUSY on failure. */ @@ -8354,9 +8429,13 @@ int read_lcb_csr(struct hfi1_devdata *dd, u32 addr, u64 *data) /* if up, go through the 8051 for the value */ if (ppd->host_link_state & HLS_UP) return read_lcb_via_8051(dd, addr, data); - /* if going up or down, no access */ - if (ppd->host_link_state & (HLS_GOING_UP | HLS_GOING_OFFLINE)) - return -EBUSY; + /* if going up or down, check the cache, otherwise, no access */ + if (ppd->host_link_state & (HLS_GOING_UP | HLS_GOING_OFFLINE)) { + if (read_lcb_cache(addr, data)) + return -EBUSY; + return 0; + } + /* otherwise, host has access */ *data = read_csr(dd, addr); return 0; @@ -8371,7 +8450,7 @@ static int write_lcb_via_8051(struct hfi1_devdata *dd, u32 addr, u64 data) int ret; if (dd->icode == ICODE_FUNCTIONAL_SIMULATOR || - (dd->dc8051_ver < dc8051_ver(0, 20))) { + (dd->dc8051_ver < dc8051_ver(0, 20, 0))) { if (acquire_lcb_access(dd, 0) == 0) { write_csr(dd, addr, data); release_lcb_access(dd, 0); @@ -8420,16 +8499,11 @@ static int do_8051_command( { u64 reg, completed; int return_code; - unsigned long flags; unsigned long timeout; hfi1_cdbg(DC8051, "type %d, data 0x%012llx", type, in_data); - /* - * Alternative to holding the lock for a long time: - * - keep busy wait - have other users bounce off - */ - spin_lock_irqsave(&dd->dc8051_lock, flags); + mutex_lock(&dd->dc8051_lock); /* We can't send any commands to the 8051 if it's in reset */ if (dd->dc_shutdown) { @@ -8455,10 +8529,8 @@ static int do_8051_command( return_code = -ENXIO; goto fail; } - spin_unlock_irqrestore(&dd->dc8051_lock, flags); - dc_shutdown(dd); - dc_start(dd); - spin_lock_irqsave(&dd->dc8051_lock, flags); + _dc_shutdown(dd); + _dc_start(dd); } /* @@ -8539,8 +8611,7 @@ static int do_8051_command( write_csr(dd, DC_DC8051_CFG_HOST_CMD_0, 0); fail: - spin_unlock_irqrestore(&dd->dc8051_lock, flags); - + mutex_unlock(&dd->dc8051_lock); return return_code; } @@ -8677,13 +8748,20 @@ static void read_remote_device_id(struct hfi1_devdata *dd, u16 *device_id, & REMOTE_DEVICE_REV_MASK; } -void read_misc_status(struct hfi1_devdata *dd, u8 *ver_a, u8 *ver_b) +void read_misc_status(struct hfi1_devdata *dd, u8 *ver_major, u8 *ver_minor, + u8 *ver_patch) { u32 frame; read_8051_config(dd, MISC_STATUS, GENERAL_CONFIG, &frame); - *ver_a = (frame >> STS_FM_VERSION_A_SHIFT) & STS_FM_VERSION_A_MASK; - *ver_b = (frame >> STS_FM_VERSION_B_SHIFT) & STS_FM_VERSION_B_MASK; + *ver_major = (frame >> STS_FM_VERSION_MAJOR_SHIFT) & + STS_FM_VERSION_MAJOR_MASK; + *ver_minor = (frame >> STS_FM_VERSION_MINOR_SHIFT) & + STS_FM_VERSION_MINOR_MASK; + + read_8051_config(dd, VERSION_PATCH, GENERAL_CONFIG, &frame); + *ver_patch = (frame >> STS_FM_VERSION_PATCH_SHIFT) & + STS_FM_VERSION_PATCH_MASK; } static void read_vc_remote_phy(struct hfi1_devdata *dd, u8 *power_management, @@ -8891,8 +8969,6 @@ int send_idle_sma(struct hfi1_devdata *dd, u64 message) */ static int do_quick_linkup(struct hfi1_devdata *dd) { - u64 reg; - unsigned long timeout; int ret; lcb_shutdown(dd, 0); @@ -8915,19 +8991,9 @@ static int do_quick_linkup(struct hfi1_devdata *dd) write_csr(dd, DC_LCB_CFG_RUN, 1ull << DC_LCB_CFG_RUN_EN_SHIFT); - /* watch LCB_STS_LINK_TRANSFER_ACTIVE */ - timeout = jiffies + msecs_to_jiffies(10); - while (1) { - reg = read_csr(dd, DC_LCB_STS_LINK_TRANSFER_ACTIVE); - if (reg) - break; - if (time_after(jiffies, timeout)) { - dd_dev_err(dd, - "timeout waiting for LINK_TRANSFER_ACTIVE\n"); - return -ETIMEDOUT; - } - udelay(2); - } + ret = wait_link_transfer_active(dd, 10); + if (ret) + return ret; write_csr(dd, DC_LCB_CFG_ALLOW_LINK_UP, 1ull << DC_LCB_CFG_ALLOW_LINK_UP_VAL_SHIFT); @@ -9091,7 +9157,7 @@ static int set_local_link_attributes(struct hfi1_pportdata *ppd) if (ret) goto set_local_link_attributes_fail; - if (dd->dc8051_ver < dc8051_ver(0, 20)) { + if (dd->dc8051_ver < dc8051_ver(0, 20, 0)) { /* set the tx rate to the fastest enabled */ if (ppd->link_speed_enabled & OPA_LINK_SPEED_25G) ppd->local_tx_rate = 1; @@ -9274,7 +9340,7 @@ static int handle_qsfp_error_conditions(struct hfi1_pportdata *ppd, if ((qsfp_interrupt_status[0] & QSFP_HIGH_TEMP_ALARM) || (qsfp_interrupt_status[0] & QSFP_HIGH_TEMP_WARNING)) - dd_dev_info(dd, "%s: QSFP cable on fire\n", + dd_dev_info(dd, "%s: QSFP cable temperature too high\n", __func__); if ((qsfp_interrupt_status[0] & QSFP_LOW_TEMP_ALARM) || @@ -9494,8 +9560,11 @@ static int test_qsfp_read(struct hfi1_pportdata *ppd) int ret; u8 status; - /* report success if not a QSFP */ - if (ppd->port_type != PORT_TYPE_QSFP) + /* + * Report success if not a QSFP or, if it is a QSFP, but the cable is + * not present + */ + if (ppd->port_type != PORT_TYPE_QSFP || !qsfp_mod_present(ppd)) return 0; /* read byte 2, the status byte */ @@ -10082,6 +10151,64 @@ static void check_lni_states(struct hfi1_pportdata *ppd) decode_state_complete(ppd, last_remote_state, "received"); } +/* wait for wait_ms for LINK_TRANSFER_ACTIVE to go to 1 */ +static int wait_link_transfer_active(struct hfi1_devdata *dd, int wait_ms) +{ + u64 reg; + unsigned long timeout; + + /* watch LCB_STS_LINK_TRANSFER_ACTIVE */ + timeout = jiffies + msecs_to_jiffies(wait_ms); + while (1) { + reg = read_csr(dd, DC_LCB_STS_LINK_TRANSFER_ACTIVE); + if (reg) + break; + if (time_after(jiffies, timeout)) { + dd_dev_err(dd, + "timeout waiting for LINK_TRANSFER_ACTIVE\n"); + return -ETIMEDOUT; + } + udelay(2); + } + return 0; +} + +/* called when the logical link state is not down as it should be */ +static void force_logical_link_state_down(struct hfi1_pportdata *ppd) +{ + struct hfi1_devdata *dd = ppd->dd; + + /* + * Bring link up in LCB loopback + */ + write_csr(dd, DC_LCB_CFG_TX_FIFOS_RESET, 1); + write_csr(dd, DC_LCB_CFG_IGNORE_LOST_RCLK, + DC_LCB_CFG_IGNORE_LOST_RCLK_EN_SMASK); + + write_csr(dd, DC_LCB_CFG_LANE_WIDTH, 0); + write_csr(dd, DC_LCB_CFG_REINIT_AS_SLAVE, 0); + write_csr(dd, DC_LCB_CFG_CNT_FOR_SKIP_STALL, 0x110); + write_csr(dd, DC_LCB_CFG_LOOPBACK, 0x2); + + write_csr(dd, DC_LCB_CFG_TX_FIFOS_RESET, 0); + (void)read_csr(dd, DC_LCB_CFG_TX_FIFOS_RESET); + udelay(3); + write_csr(dd, DC_LCB_CFG_ALLOW_LINK_UP, 1); + write_csr(dd, DC_LCB_CFG_RUN, 1ull << DC_LCB_CFG_RUN_EN_SHIFT); + + wait_link_transfer_active(dd, 100); + + /* + * Bring the link down again. + */ + write_csr(dd, DC_LCB_CFG_TX_FIFOS_RESET, 1); + write_csr(dd, DC_LCB_CFG_ALLOW_LINK_UP, 0); + write_csr(dd, DC_LCB_CFG_IGNORE_LOST_RCLK, 0); + + /* call again to adjust ppd->statusp, if needed */ + get_logical_state(ppd); +} + /* * Helper for set_link_state(). Do not call except from that routine. * Expects ppd->hls_mutex to be held. @@ -10098,6 +10225,8 @@ static int goto_offline(struct hfi1_pportdata *ppd, u8 rem_reason) int do_transition; int do_wait; + update_lcb_cache(dd); + previous_state = ppd->host_link_state; ppd->host_link_state = HLS_GOING_OFFLINE; pstate = read_physical_state(dd); @@ -10135,15 +10264,18 @@ static int goto_offline(struct hfi1_pportdata *ppd, u8 rem_reason) return ret; } - /* make sure the logical state is also down */ - wait_logical_linkstate(ppd, IB_PORT_DOWN, 1000); - /* * Now in charge of LCB - must be after the physical state is * offline.quiet and before host_link_state is changed. */ set_host_lcb_access(dd); write_csr(dd, DC_LCB_ERR_EN, ~0ull); /* watch LCB errors */ + + /* make sure the logical state is also down */ + ret = wait_logical_linkstate(ppd, IB_PORT_DOWN, 1000); + if (ret) + force_logical_link_state_down(ppd); + ppd->host_link_state = HLS_LINK_COOLDOWN; /* LCB access allowed */ if (ppd->port_type == PORT_TYPE_QSFP && @@ -10380,11 +10512,8 @@ int set_link_state(struct hfi1_pportdata *ppd, u32 state) goto unexpected; } - ppd->host_link_state = HLS_UP_INIT; ret = wait_logical_linkstate(ppd, IB_PORT_INIT, 1000); if (ret) { - /* logical state didn't change, stay at going_up */ - ppd->host_link_state = HLS_GOING_UP; dd_dev_err(dd, "%s: logical state did not change to INIT\n", __func__); @@ -10398,6 +10527,7 @@ int set_link_state(struct hfi1_pportdata *ppd, u32 state) add_rcvctrl(dd, RCV_CTRL_RCV_PORT_ENABLE_SMASK); handle_linkup_change(dd, 1); + ppd->host_link_state = HLS_UP_INIT; } break; case HLS_UP_ARMED: @@ -11853,6 +11983,10 @@ static void free_cntrs(struct hfi1_devdata *dd) dd->scntrs = NULL; kfree(dd->cntrnames); dd->cntrnames = NULL; + if (dd->update_cntr_wq) { + destroy_workqueue(dd->update_cntr_wq); + dd->update_cntr_wq = NULL; + } } static u64 read_dev_port_cntr(struct hfi1_devdata *dd, struct cntr_entry *entry, @@ -12008,7 +12142,7 @@ u64 write_port_cntr(struct hfi1_pportdata *ppd, int index, int vl, u64 data) return write_dev_port_cntr(ppd->dd, entry, sval, ppd, vl, data); } -static void update_synth_timer(unsigned long opaque) +static void do_update_synth_timer(struct work_struct *work) { u64 cur_tx; u64 cur_rx; @@ -12017,8 +12151,8 @@ static void update_synth_timer(unsigned long opaque) int i, j, vl; struct hfi1_pportdata *ppd; struct cntr_entry *entry; - - struct hfi1_devdata *dd = (struct hfi1_devdata *)opaque; + struct hfi1_devdata *dd = container_of(work, struct hfi1_devdata, + update_cntr_work); /* * Rather than keep beating on the CSRs pick a minimal set that we can @@ -12101,7 +12235,13 @@ static void update_synth_timer(unsigned long opaque) } else { hfi1_cdbg(CNTR, "[%d] No update necessary", dd->unit); } +} +static void update_synth_timer(unsigned long opaque) +{ + struct hfi1_devdata *dd = (struct hfi1_devdata *)opaque; + + queue_work(dd->update_cntr_wq, &dd->update_cntr_work); mod_timer(&dd->synth_stats_timer, jiffies + HZ * SYNTH_CNT_TIME); } @@ -12337,6 +12477,13 @@ static int init_cntrs(struct hfi1_devdata *dd) if (init_cpu_counters(dd)) goto bail; + dd->update_cntr_wq = alloc_ordered_workqueue("hfi1_update_cntr_%d", + WQ_MEM_RECLAIM, dd->unit); + if (!dd->update_cntr_wq) + goto bail; + + INIT_WORK(&dd->update_cntr_work, do_update_synth_timer); + mod_timer(&dd->synth_stats_timer, jiffies + HZ * SYNTH_CNT_TIME); return 0; bail: @@ -12726,7 +12873,10 @@ static int request_msix_irqs(struct hfi1_devdata *dd) first_sdma = last_general; last_sdma = first_sdma + dd->num_sdma; first_rx = last_sdma; - last_rx = first_rx + dd->n_krcv_queues; + last_rx = first_rx + dd->n_krcv_queues + HFI1_NUM_VNIC_CTXT; + + /* VNIC MSIx interrupts get mapped when VNIC contexts are created */ + dd->first_dyn_msix_idx = first_rx + dd->n_krcv_queues; /* * Sanity check - the code expects all SDMA chip source @@ -12740,7 +12890,7 @@ static int request_msix_irqs(struct hfi1_devdata *dd) const char *err_info; irq_handler_t handler; irq_handler_t thread = NULL; - void *arg; + void *arg = NULL; int idx; struct hfi1_ctxtdata *rcd = NULL; struct sdma_engine *sde = NULL; @@ -12767,24 +12917,25 @@ static int request_msix_irqs(struct hfi1_devdata *dd) } else if (first_rx <= i && i < last_rx) { idx = i - first_rx; rcd = dd->rcd[idx]; - /* no interrupt if no rcd */ - if (!rcd) - continue; - /* - * Set the interrupt register and mask for this - * context's interrupt. - */ - rcd->ireg = (IS_RCVAVAIL_START + idx) / 64; - rcd->imask = ((u64)1) << - ((IS_RCVAVAIL_START + idx) % 64); - handler = receive_context_interrupt; - thread = receive_context_thread; - arg = rcd; - snprintf(me->name, sizeof(me->name), - DRIVER_NAME "_%d kctxt%d", dd->unit, idx); - err_info = "receive context"; - remap_intr(dd, IS_RCVAVAIL_START + idx, i); - me->type = IRQ_RCVCTXT; + if (rcd) { + /* + * Set the interrupt register and mask for this + * context's interrupt. + */ + rcd->ireg = (IS_RCVAVAIL_START + idx) / 64; + rcd->imask = ((u64)1) << + ((IS_RCVAVAIL_START + idx) % 64); + handler = receive_context_interrupt; + thread = receive_context_thread; + arg = rcd; + snprintf(me->name, sizeof(me->name), + DRIVER_NAME "_%d kctxt%d", + dd->unit, idx); + err_info = "receive context"; + remap_intr(dd, IS_RCVAVAIL_START + idx, i); + me->type = IRQ_RCVCTXT; + rcd->msix_intr = i; + } } else { /* not in our expected range - complain, then * ignore it @@ -12822,6 +12973,84 @@ static int request_msix_irqs(struct hfi1_devdata *dd) return ret; } +void hfi1_vnic_synchronize_irq(struct hfi1_devdata *dd) +{ + int i; + + if (!dd->num_msix_entries) { + synchronize_irq(dd->pcidev->irq); + return; + } + + for (i = 0; i < dd->vnic.num_ctxt; i++) { + struct hfi1_ctxtdata *rcd = dd->vnic.ctxt[i]; + struct hfi1_msix_entry *me = &dd->msix_entries[rcd->msix_intr]; + + synchronize_irq(me->msix.vector); + } +} + +void hfi1_reset_vnic_msix_info(struct hfi1_ctxtdata *rcd) +{ + struct hfi1_devdata *dd = rcd->dd; + struct hfi1_msix_entry *me = &dd->msix_entries[rcd->msix_intr]; + + if (!me->arg) /* => no irq, no affinity */ + return; + + hfi1_put_irq_affinity(dd, me); + free_irq(me->msix.vector, me->arg); + + me->arg = NULL; +} + +void hfi1_set_vnic_msix_info(struct hfi1_ctxtdata *rcd) +{ + struct hfi1_devdata *dd = rcd->dd; + struct hfi1_msix_entry *me; + int idx = rcd->ctxt; + void *arg = rcd; + int ret; + + rcd->msix_intr = dd->vnic.msix_idx++; + me = &dd->msix_entries[rcd->msix_intr]; + + /* + * Set the interrupt register and mask for this + * context's interrupt. + */ + rcd->ireg = (IS_RCVAVAIL_START + idx) / 64; + rcd->imask = ((u64)1) << + ((IS_RCVAVAIL_START + idx) % 64); + + snprintf(me->name, sizeof(me->name), + DRIVER_NAME "_%d kctxt%d", dd->unit, idx); + me->name[sizeof(me->name) - 1] = 0; + me->type = IRQ_RCVCTXT; + + remap_intr(dd, IS_RCVAVAIL_START + idx, rcd->msix_intr); + + ret = request_threaded_irq(me->msix.vector, receive_context_interrupt, + receive_context_thread, 0, me->name, arg); + if (ret) { + dd_dev_err(dd, "vnic irq request (vector %d, idx %d) fail %d\n", + me->msix.vector, idx, ret); + return; + } + /* + * assign arg after request_irq call, so it will be + * cleaned up + */ + me->arg = arg; + + ret = hfi1_get_irq_affinity(dd, me); + if (ret) { + dd_dev_err(dd, + "unable to pin IRQ %d\n", ret); + free_irq(me->msix.vector, me->arg); + } +} + /* * Set the general handler to accept all interrupts, remap all * chip interrupts back to MSI-X 0. @@ -12853,7 +13082,7 @@ static int set_up_interrupts(struct hfi1_devdata *dd) * N interrupts - one per used SDMA engine * M interrupt - one per kernel receive context */ - total = 1 + dd->num_sdma + dd->n_krcv_queues; + total = 1 + dd->num_sdma + dd->n_krcv_queues + HFI1_NUM_VNIC_CTXT; entries = kcalloc(total, sizeof(*entries), GFP_KERNEL); if (!entries) { @@ -12918,7 +13147,8 @@ fail: * * num_rcv_contexts - number of contexts being used * n_krcv_queues - number of kernel contexts - * first_user_ctxt - first non-kernel context in array of contexts + * first_dyn_alloc_ctxt - first dynamically allocated context + * in array of contexts * freectxts - number of free user contexts * num_send_contexts - number of PIO send contexts being used */ @@ -12995,10 +13225,14 @@ static int set_up_context_variables(struct hfi1_devdata *dd) total_contexts = num_kernel_contexts + num_user_contexts; } - /* the first N are kernel contexts, the rest are user contexts */ + /* Accommodate VNIC contexts */ + if ((total_contexts + HFI1_NUM_VNIC_CTXT) <= dd->chip_rcv_contexts) + total_contexts += HFI1_NUM_VNIC_CTXT; + + /* the first N are kernel contexts, the rest are user/vnic contexts */ dd->num_rcv_contexts = total_contexts; dd->n_krcv_queues = num_kernel_contexts; - dd->first_user_ctxt = num_kernel_contexts; + dd->first_dyn_alloc_ctxt = num_kernel_contexts; dd->num_user_contexts = num_user_contexts; dd->freectxts = num_user_contexts; dd_dev_info(dd, @@ -13454,11 +13688,8 @@ static void reset_rxe_csrs(struct hfi1_devdata *dd) write_csr(dd, RCV_COUNTER_ARRAY32 + (8 * i), 0); for (i = 0; i < RXE_NUM_64_BIT_COUNTERS; i++) write_csr(dd, RCV_COUNTER_ARRAY64 + (8 * i), 0); - for (i = 0; i < RXE_NUM_RSM_INSTANCES; i++) { - write_csr(dd, RCV_RSM_CFG + (8 * i), 0); - write_csr(dd, RCV_RSM_SELECT + (8 * i), 0); - write_csr(dd, RCV_RSM_MATCH + (8 * i), 0); - } + for (i = 0; i < RXE_NUM_RSM_INSTANCES; i++) + clear_rsm_rule(dd, i); for (i = 0; i < 32; i++) write_csr(dd, RCV_RSM_MAP_TABLE + (8 * i), 0); @@ -13817,6 +14048,16 @@ static void add_rsm_rule(struct hfi1_devdata *dd, u8 rule_index, (u64)rrd->value2 << RCV_RSM_MATCH_VALUE2_SHIFT); } +/* + * Clear a receive side mapping rule. + */ +static void clear_rsm_rule(struct hfi1_devdata *dd, u8 rule_index) +{ + write_csr(dd, RCV_RSM_CFG + (8 * rule_index), 0); + write_csr(dd, RCV_RSM_SELECT + (8 * rule_index), 0); + write_csr(dd, RCV_RSM_MATCH + (8 * rule_index), 0); +} + /* return the number of RSM map table entries that will be used for QOS */ static int qos_rmt_entries(struct hfi1_devdata *dd, unsigned int *mp, unsigned int *np) @@ -13932,7 +14173,7 @@ static void init_qos(struct hfi1_devdata *dd, struct rsm_map_table *rmt) rrd.value2 = LRH_SC_VALUE; /* add rule 0 */ - add_rsm_rule(dd, 0, &rrd); + add_rsm_rule(dd, RSM_INS_VERBS, &rrd); /* mark RSM map entries as used */ rmt->used += rmt_entries; @@ -13962,7 +14203,7 @@ static void init_user_fecn_handling(struct hfi1_devdata *dd, /* * RSM will extract the destination context as an index into the * map table. The destination contexts are a sequential block - * in the range first_user_ctxt...num_rcv_contexts-1 (inclusive). + * in the range first_dyn_alloc_ctxt...num_rcv_contexts-1 (inclusive). * Map entries are accessed as offset + extracted value. Adjust * the added offset so this sequence can be placed anywhere in * the table - as long as the entries themselves do not wrap. @@ -13970,9 +14211,9 @@ static void init_user_fecn_handling(struct hfi1_devdata *dd, * start with that to allow for a "negative" offset. */ offset = (u8)(NUM_MAP_ENTRIES + (int)rmt->used - - (int)dd->first_user_ctxt); + (int)dd->first_dyn_alloc_ctxt); - for (i = dd->first_user_ctxt, idx = rmt->used; + for (i = dd->first_dyn_alloc_ctxt, idx = rmt->used; i < dd->num_rcv_contexts; i++, idx++) { /* replace with identity mapping */ regoff = (idx % 8) * 8; @@ -14006,11 +14247,84 @@ static void init_user_fecn_handling(struct hfi1_devdata *dd, rrd.value2 = 1; /* add rule 1 */ - add_rsm_rule(dd, 1, &rrd); + add_rsm_rule(dd, RSM_INS_FECN, &rrd); rmt->used += dd->num_user_contexts; } +/* Initialize RSM for VNIC */ +void hfi1_init_vnic_rsm(struct hfi1_devdata *dd) +{ + u8 i, j; + u8 ctx_id = 0; + u64 reg; + u32 regoff; + struct rsm_rule_data rrd; + + if (hfi1_vnic_is_rsm_full(dd, NUM_VNIC_MAP_ENTRIES)) { + dd_dev_err(dd, "Vnic RSM disabled, rmt entries used = %d\n", + dd->vnic.rmt_start); + return; + } + + dev_dbg(&(dd)->pcidev->dev, "Vnic rsm start = %d, end %d\n", + dd->vnic.rmt_start, + dd->vnic.rmt_start + NUM_VNIC_MAP_ENTRIES); + + /* Update RSM mapping table, 32 regs, 256 entries - 1 ctx per byte */ + regoff = RCV_RSM_MAP_TABLE + (dd->vnic.rmt_start / 8) * 8; + reg = read_csr(dd, regoff); + for (i = 0; i < NUM_VNIC_MAP_ENTRIES; i++) { + /* Update map register with vnic context */ + j = (dd->vnic.rmt_start + i) % 8; + reg &= ~(0xffllu << (j * 8)); + reg |= (u64)dd->vnic.ctxt[ctx_id++]->ctxt << (j * 8); + /* Wrap up vnic ctx index */ + ctx_id %= dd->vnic.num_ctxt; + /* Write back map register */ + if (j == 7 || ((i + 1) == NUM_VNIC_MAP_ENTRIES)) { + dev_dbg(&(dd)->pcidev->dev, + "Vnic rsm map reg[%d] =0x%llx\n", + regoff - RCV_RSM_MAP_TABLE, reg); + + write_csr(dd, regoff, reg); + regoff += 8; + if (i < (NUM_VNIC_MAP_ENTRIES - 1)) + reg = read_csr(dd, regoff); + } + } + + /* Add rule for vnic */ + rrd.offset = dd->vnic.rmt_start; + rrd.pkt_type = 4; + /* Match 16B packets */ + rrd.field1_off = L2_TYPE_MATCH_OFFSET; + rrd.mask1 = L2_TYPE_MASK; + rrd.value1 = L2_16B_VALUE; + /* Match ETH L4 packets */ + rrd.field2_off = L4_TYPE_MATCH_OFFSET; + rrd.mask2 = L4_16B_TYPE_MASK; + rrd.value2 = L4_16B_ETH_VALUE; + /* Calc context from veswid and entropy */ + rrd.index1_off = L4_16B_HDR_VESWID_OFFSET; + rrd.index1_width = ilog2(NUM_VNIC_MAP_ENTRIES); + rrd.index2_off = L2_16B_ENTROPY_OFFSET; + rrd.index2_width = ilog2(NUM_VNIC_MAP_ENTRIES); + add_rsm_rule(dd, RSM_INS_VNIC, &rrd); + + /* Enable RSM if not already enabled */ + add_rcvctrl(dd, RCV_CTRL_RCV_RSM_ENABLE_SMASK); +} + +void hfi1_deinit_vnic_rsm(struct hfi1_devdata *dd) +{ + clear_rsm_rule(dd, RSM_INS_VNIC); + + /* Disable RSM if used only by vnic */ + if (dd->vnic.rmt_start == 0) + clear_rcvctrl(dd, RCV_CTRL_RCV_RSM_ENABLE_SMASK); +} + static void init_rxe(struct hfi1_devdata *dd) { struct rsm_map_table *rmt; @@ -14023,6 +14337,8 @@ static void init_rxe(struct hfi1_devdata *dd) init_qos(dd, rmt); init_user_fecn_handling(dd, rmt); complete_rsm_map_table(dd, rmt); + /* record number of used rsm map entries for vnic */ + dd->vnic.rmt_start = rmt->used; kfree(rmt); /* diff --git a/drivers/infiniband/hw/hfi1/chip.h b/drivers/infiniband/hw/hfi1/chip.h index 043fd21dc5f3..b9dbf16d7703 100644 --- a/drivers/infiniband/hw/hfi1/chip.h +++ b/drivers/infiniband/hw/hfi1/chip.h @@ -1,7 +1,7 @@ #ifndef _CHIP_H #define _CHIP_H /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015 - 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -394,7 +394,8 @@ #define LAST_REMOTE_STATE_COMPLETE 0x13 #define LINK_QUALITY_INFO 0x14 #define REMOTE_DEVICE_ID 0x15 -#define LINK_DOWN_REASON 0x16 +#define LINK_DOWN_REASON 0x16 /* first byte of offset 0x16 */ +#define VERSION_PATCH 0x16 /* last byte of offset 0x16 */ /* 8051 lane specific register field IDs */ #define TX_EQ_SETTINGS 0x00 @@ -524,10 +525,12 @@ enum { #define SUPPORTED_CRCS (CAP_CRC_14B | CAP_CRC_48B) /* misc status version fields */ -#define STS_FM_VERSION_A_SHIFT 16 -#define STS_FM_VERSION_A_MASK 0xff -#define STS_FM_VERSION_B_SHIFT 24 -#define STS_FM_VERSION_B_MASK 0xff +#define STS_FM_VERSION_MINOR_SHIFT 16 +#define STS_FM_VERSION_MINOR_MASK 0xff +#define STS_FM_VERSION_MAJOR_SHIFT 24 +#define STS_FM_VERSION_MAJOR_MASK 0xff +#define STS_FM_VERSION_PATCH_SHIFT 24 +#define STS_FM_VERSION_PATCH_MASK 0xff /* LCB_CFG_CRC_MODE TX_VAL and RX_VAL CRC mode values */ #define LCB_CRC_16B 0x0 /* 16b CRC */ @@ -698,7 +701,8 @@ void fabric_serdes_reset(struct hfi1_devdata *dd); int read_8051_data(struct hfi1_devdata *dd, u32 addr, u32 len, u64 *result); /* chip.c */ -void read_misc_status(struct hfi1_devdata *dd, u8 *ver_a, u8 *ver_b); +void read_misc_status(struct hfi1_devdata *dd, u8 *ver_major, u8 *ver_minor, + u8 *ver_patch); void read_guid(struct hfi1_devdata *dd); int wait_fm_ready(struct hfi1_devdata *dd, u32 mstimeout); void set_link_down_reason(struct hfi1_pportdata *ppd, u8 lcl_reason, @@ -1358,6 +1362,8 @@ int hfi1_clear_ctxt_jkey(struct hfi1_devdata *dd, unsigned ctxt); int hfi1_set_ctxt_pkey(struct hfi1_devdata *dd, unsigned ctxt, u16 pkey); int hfi1_clear_ctxt_pkey(struct hfi1_devdata *dd, unsigned ctxt); void hfi1_read_link_quality(struct hfi1_devdata *dd, u8 *link_quality); +void hfi1_init_vnic_rsm(struct hfi1_devdata *dd); +void hfi1_deinit_vnic_rsm(struct hfi1_devdata *dd); /* * Interrupt source table. diff --git a/drivers/infiniband/hw/hfi1/common.h b/drivers/infiniband/hw/hfi1/common.h index 1b783bbee4bb..995d62c7f9a7 100644 --- a/drivers/infiniband/hw/hfi1/common.h +++ b/drivers/infiniband/hw/hfi1/common.h @@ -331,12 +331,15 @@ struct diag_pkt { #define FULL_MGMT_P_KEY 0xFFFF #define DEFAULT_P_KEY LIM_MGMT_P_KEY -#define HFI1_FECN_SHIFT 31 -#define HFI1_FECN_MASK 1 -#define HFI1_FECN_SMASK BIT(HFI1_FECN_SHIFT) -#define HFI1_BECN_SHIFT 30 -#define HFI1_BECN_MASK 1 -#define HFI1_BECN_SMASK BIT(HFI1_BECN_SHIFT) + +/** + * 0xF8 - 4 bits of multicast range and 1 bit for collective range + * Example: For 24 bit LID space, + * Multicast range: 0xF00000 to 0xF7FFFF + * Collective range: 0xF80000 to 0xFFFFFE + */ +#define HFI1_MCAST_NR 0x4 /* Number of top bits set */ +#define HFI1_COLLECTIVE_NR 0x1 /* Number of bits after MCAST_NR */ #define HFI1_PSM_IOC_BASE_SEQ 0x0 diff --git a/drivers/infiniband/hw/hfi1/debugfs.c b/drivers/infiniband/hw/hfi1/debugfs.c index 7fe9dd885746..e9fa3c293e42 100644 --- a/drivers/infiniband/hw/hfi1/debugfs.c +++ b/drivers/infiniband/hw/hfi1/debugfs.c @@ -1,6 +1,6 @@ #ifdef CONFIG_DEBUG_FS /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -51,8 +51,12 @@ #include <linux/export.h> #include <linux/module.h> #include <linux/string.h> +#include <linux/types.h> +#include <linux/ratelimit.h> +#include <linux/fault-inject.h> #include "hfi.h" +#include "trace.h" #include "debugfs.h" #include "device.h" #include "qp.h" @@ -170,7 +174,7 @@ static int _opcode_stats_seq_show(struct seq_file *s, void *v) struct hfi1_ibdev *ibd = (struct hfi1_ibdev *)s->private; struct hfi1_devdata *dd = dd_from_dev(ibd); - for (j = 0; j < dd->first_user_ctxt; j++) { + for (j = 0; j < dd->first_dyn_alloc_ctxt; j++) { if (!dd->rcd[j]) continue; n_packets += dd->rcd[j]->opstats->stats[i].n_packets; @@ -196,7 +200,7 @@ static void *_ctx_stats_seq_start(struct seq_file *s, loff_t *pos) if (!*pos) return SEQ_START_TOKEN; - if (*pos >= dd->first_user_ctxt) + if (*pos >= dd->first_dyn_alloc_ctxt) return NULL; return pos; } @@ -210,7 +214,7 @@ static void *_ctx_stats_seq_next(struct seq_file *s, void *v, loff_t *pos) return pos; ++*pos; - if (*pos >= dd->first_user_ctxt) + if (*pos >= dd->first_dyn_alloc_ctxt) return NULL; return pos; } @@ -1063,6 +1067,222 @@ DEBUGFS_SEQ_FILE_OPS(sdma_cpu_list); DEBUGFS_SEQ_FILE_OPEN(sdma_cpu_list) DEBUGFS_FILE_OPS(sdma_cpu_list); +#ifdef CONFIG_FAULT_INJECTION +static void *_fault_stats_seq_start(struct seq_file *s, loff_t *pos) +{ + struct hfi1_opcode_stats_perctx *opstats; + + if (*pos >= ARRAY_SIZE(opstats->stats)) + return NULL; + return pos; +} + +static void *_fault_stats_seq_next(struct seq_file *s, void *v, loff_t *pos) +{ + struct hfi1_opcode_stats_perctx *opstats; + + ++*pos; + if (*pos >= ARRAY_SIZE(opstats->stats)) + return NULL; + return pos; +} + +static void _fault_stats_seq_stop(struct seq_file *s, void *v) +{ +} + +static int _fault_stats_seq_show(struct seq_file *s, void *v) +{ + loff_t *spos = v; + loff_t i = *spos, j; + u64 n_packets = 0, n_bytes = 0; + struct hfi1_ibdev *ibd = (struct hfi1_ibdev *)s->private; + struct hfi1_devdata *dd = dd_from_dev(ibd); + + for (j = 0; j < dd->first_dyn_alloc_ctxt; j++) { + if (!dd->rcd[j]) + continue; + n_packets += dd->rcd[j]->opstats->stats[i].n_packets; + n_bytes += dd->rcd[j]->opstats->stats[i].n_bytes; + } + if (!n_packets && !n_bytes) + return SEQ_SKIP; + if (!ibd->fault_opcode->n_rxfaults[i] && + !ibd->fault_opcode->n_txfaults[i]) + return SEQ_SKIP; + seq_printf(s, "%02llx %llu/%llu (faults rx:%llu faults: tx:%llu)\n", i, + (unsigned long long)n_packets, + (unsigned long long)n_bytes, + (unsigned long long)ibd->fault_opcode->n_rxfaults[i], + (unsigned long long)ibd->fault_opcode->n_txfaults[i]); + return 0; +} + +DEBUGFS_SEQ_FILE_OPS(fault_stats); +DEBUGFS_SEQ_FILE_OPEN(fault_stats); +DEBUGFS_FILE_OPS(fault_stats); + +static void fault_exit_opcode_debugfs(struct hfi1_ibdev *ibd) +{ + debugfs_remove_recursive(ibd->fault_opcode->dir); + kfree(ibd->fault_opcode); + ibd->fault_opcode = NULL; +} + +static int fault_init_opcode_debugfs(struct hfi1_ibdev *ibd) +{ + struct dentry *parent = ibd->hfi1_ibdev_dbg; + + ibd->fault_opcode = kzalloc(sizeof(*ibd->fault_opcode), GFP_KERNEL); + if (!ibd->fault_opcode) + return -ENOMEM; + + ibd->fault_opcode->attr.interval = 1; + ibd->fault_opcode->attr.require_end = ULONG_MAX; + ibd->fault_opcode->attr.stacktrace_depth = 32; + ibd->fault_opcode->attr.dname = NULL; + ibd->fault_opcode->attr.verbose = 0; + ibd->fault_opcode->fault_by_opcode = false; + ibd->fault_opcode->opcode = 0; + ibd->fault_opcode->mask = 0xff; + + ibd->fault_opcode->dir = + fault_create_debugfs_attr("fault_opcode", + parent, + &ibd->fault_opcode->attr); + if (IS_ERR(ibd->fault_opcode->dir)) { + kfree(ibd->fault_opcode); + return -ENOENT; + } + + DEBUGFS_SEQ_FILE_CREATE(fault_stats, ibd->fault_opcode->dir, ibd); + if (!debugfs_create_bool("fault_by_opcode", 0600, + ibd->fault_opcode->dir, + &ibd->fault_opcode->fault_by_opcode)) + goto fail; + if (!debugfs_create_x8("opcode", 0600, ibd->fault_opcode->dir, + &ibd->fault_opcode->opcode)) + goto fail; + if (!debugfs_create_x8("mask", 0600, ibd->fault_opcode->dir, + &ibd->fault_opcode->mask)) + goto fail; + + return 0; +fail: + fault_exit_opcode_debugfs(ibd); + return -ENOMEM; +} + +static void fault_exit_packet_debugfs(struct hfi1_ibdev *ibd) +{ + debugfs_remove_recursive(ibd->fault_packet->dir); + kfree(ibd->fault_packet); + ibd->fault_packet = NULL; +} + +static int fault_init_packet_debugfs(struct hfi1_ibdev *ibd) +{ + struct dentry *parent = ibd->hfi1_ibdev_dbg; + + ibd->fault_packet = kzalloc(sizeof(*ibd->fault_packet), GFP_KERNEL); + if (!ibd->fault_packet) + return -ENOMEM; + + ibd->fault_packet->attr.interval = 1; + ibd->fault_packet->attr.require_end = ULONG_MAX; + ibd->fault_packet->attr.stacktrace_depth = 32; + ibd->fault_packet->attr.dname = NULL; + ibd->fault_packet->attr.verbose = 0; + ibd->fault_packet->fault_by_packet = false; + + ibd->fault_packet->dir = + fault_create_debugfs_attr("fault_packet", + parent, + &ibd->fault_opcode->attr); + if (IS_ERR(ibd->fault_packet->dir)) { + kfree(ibd->fault_packet); + return -ENOENT; + } + + if (!debugfs_create_bool("fault_by_packet", 0600, + ibd->fault_packet->dir, + &ibd->fault_packet->fault_by_packet)) + goto fail; + if (!debugfs_create_u64("fault_stats", 0400, + ibd->fault_packet->dir, + &ibd->fault_packet->n_faults)) + goto fail; + + return 0; +fail: + fault_exit_packet_debugfs(ibd); + return -ENOMEM; +} + +static void fault_exit_debugfs(struct hfi1_ibdev *ibd) +{ + fault_exit_opcode_debugfs(ibd); + fault_exit_packet_debugfs(ibd); +} + +static int fault_init_debugfs(struct hfi1_ibdev *ibd) +{ + int ret = 0; + + ret = fault_init_opcode_debugfs(ibd); + if (ret) + return ret; + + ret = fault_init_packet_debugfs(ibd); + if (ret) + fault_exit_opcode_debugfs(ibd); + + return ret; +} + +bool hfi1_dbg_fault_suppress_err(struct hfi1_ibdev *ibd) +{ + return ibd->fault_suppress_err; +} + +bool hfi1_dbg_fault_opcode(struct rvt_qp *qp, u32 opcode, bool rx) +{ + bool ret = false; + struct hfi1_ibdev *ibd = to_idev(qp->ibqp.device); + + if (!ibd->fault_opcode || !ibd->fault_opcode->fault_by_opcode) + return false; + if (ibd->fault_opcode->opcode != (opcode & ibd->fault_opcode->mask)) + return false; + ret = should_fail(&ibd->fault_opcode->attr, 1); + if (ret) { + trace_hfi1_fault_opcode(qp, opcode); + if (rx) + ibd->fault_opcode->n_rxfaults[opcode]++; + else + ibd->fault_opcode->n_txfaults[opcode]++; + } + return ret; +} + +bool hfi1_dbg_fault_packet(struct hfi1_packet *packet) +{ + struct rvt_dev_info *rdi = &packet->rcd->ppd->dd->verbs_dev.rdi; + struct hfi1_ibdev *ibd = dev_from_rdi(rdi); + bool ret = false; + + if (!ibd->fault_packet || !ibd->fault_packet->fault_by_packet) + return false; + + ret = should_fail(&ibd->fault_packet->attr, 1); + if (ret) { + ++ibd->fault_packet->n_faults; + trace_hfi1_fault_packet(packet); + } + return ret; +} +#endif + void hfi1_dbg_ibdev_init(struct hfi1_ibdev *ibd) { char name[sizeof("port0counters") + 1]; @@ -1112,12 +1332,22 @@ void hfi1_dbg_ibdev_init(struct hfi1_ibdev *ibd) !port_cntr_ops[i].ops.write ? S_IRUGO : S_IRUGO | S_IWUSR); } + +#ifdef CONFIG_FAULT_INJECTION + debugfs_create_bool("fault_suppress_err", 0600, + ibd->hfi1_ibdev_dbg, + &ibd->fault_suppress_err); + fault_init_debugfs(ibd); +#endif } void hfi1_dbg_ibdev_exit(struct hfi1_ibdev *ibd) { if (!hfi1_dbg_root) goto out; +#ifdef CONFIG_FAULT_INJECTION + fault_exit_debugfs(ibd); +#endif debugfs_remove(ibd->hfi1_ibdev_link); debugfs_remove_recursive(ibd->hfi1_ibdev_dbg); out: diff --git a/drivers/infiniband/hw/hfi1/debugfs.h b/drivers/infiniband/hw/hfi1/debugfs.h index b6fb6814f1b8..38c38a98156d 100644 --- a/drivers/infiniband/hw/hfi1/debugfs.h +++ b/drivers/infiniband/hw/hfi1/debugfs.h @@ -53,23 +53,79 @@ void hfi1_dbg_ibdev_init(struct hfi1_ibdev *ibd); void hfi1_dbg_ibdev_exit(struct hfi1_ibdev *ibd); void hfi1_dbg_init(void); void hfi1_dbg_exit(void); + +#ifdef CONFIG_FAULT_INJECTION +#include <linux/fault-inject.h> +struct fault_opcode { + struct fault_attr attr; + struct dentry *dir; + bool fault_by_opcode; + u64 n_rxfaults[256]; + u64 n_txfaults[256]; + u8 opcode; + u8 mask; +}; + +struct fault_packet { + struct fault_attr attr; + struct dentry *dir; + bool fault_by_packet; + u64 n_faults; +}; + +bool hfi1_dbg_fault_opcode(struct rvt_qp *qp, u32 opcode, bool rx); +bool hfi1_dbg_fault_packet(struct hfi1_packet *packet); +bool hfi1_dbg_fault_suppress_err(struct hfi1_ibdev *ibd); +#else +static inline bool hfi1_dbg_fault_packet(struct hfi1_packet *packet) +{ + return false; +} + +static inline bool hfi1_dbg_fault_opcode(struct rvt_qp *qp, + u32 opcode, bool rx) +{ + return false; +} + +static inline bool hfi1_dbg_fault_suppress_err(struct hfi1_ibdev *ibd) +{ + return false; +} +#endif + #else static inline void hfi1_dbg_ibdev_init(struct hfi1_ibdev *ibd) { } -void hfi1_dbg_ibdev_exit(struct hfi1_ibdev *ibd) +static inline void hfi1_dbg_ibdev_exit(struct hfi1_ibdev *ibd) +{ +} + +static inline void hfi1_dbg_init(void) { } -void hfi1_dbg_init(void) +static inline void hfi1_dbg_exit(void) { } -void hfi1_dbg_exit(void) +static inline bool hfi1_dbg_fault_packet(struct hfi1_packet *packet) { + return false; } +static inline bool hfi1_dbg_fault_opcode(struct rvt_qp *qp, + u32 opcode, bool rx) +{ + return false; +} + +static inline bool hfi1_dbg_fault_suppress_err(struct hfi1_ibdev *ibd) +{ + return false; +} #endif #endif /* _HFI1_DEBUGFS_H */ diff --git a/drivers/infiniband/hw/hfi1/driver.c b/drivers/infiniband/hw/hfi1/driver.c index 3881c951f6af..527895487175 100644 --- a/drivers/infiniband/hw/hfi1/driver.c +++ b/drivers/infiniband/hw/hfi1/driver.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -59,6 +59,8 @@ #include "trace.h" #include "qp.h" #include "sdma.h" +#include "debugfs.h" +#include "vnic.h" #undef pr_fmt #define pr_fmt(fmt) DRIVER_NAME ": " fmt @@ -283,7 +285,7 @@ static void rcv_hdrerr(struct hfi1_ctxtdata *rcd, struct hfi1_pportdata *ppd, { struct ib_header *rhdr = packet->hdr; u32 rte = rhf_rcv_type_err(packet->rhf); - int lnh = be16_to_cpu(rhdr->lrh[0]) & 3; + int lnh = ib_get_lnh(rhdr); struct hfi1_ibport *ibp = rcd_to_iport(rcd); struct hfi1_devdata *dd = ppd->dd; struct rvt_dev_info *rdi = &dd->verbs_dev.rdi; @@ -295,7 +297,7 @@ static void rcv_hdrerr(struct hfi1_ctxtdata *rcd, struct hfi1_pportdata *ppd, /* For TIDERR and RC QPs preemptively schedule a NAK */ struct ib_other_headers *ohdr = NULL; u32 tlen = rhf_pkt_len(packet->rhf); /* in bytes */ - u16 lid = be16_to_cpu(rhdr->lrh[1]); + u16 lid = ib_get_dlid(rhdr); u32 qp_num; u32 rcv_flags = 0; @@ -396,7 +398,7 @@ static void rcv_hdrerr(struct hfi1_ctxtdata *rcd, struct hfi1_pportdata *ppd, u16 rlid; u8 svc_type, sl, sc5; - sc5 = hdr2sc(rhdr, packet->rhf); + sc5 = hfi1_9B_get_sc5(rhdr, packet->rhf); sl = ibp->sc_to_sl[sc5]; lqpn = be32_to_cpu(bth[1]) & RVT_QPN_MASK; @@ -414,7 +416,7 @@ static void rcv_hdrerr(struct hfi1_ctxtdata *rcd, struct hfi1_pportdata *ppd, svc_type = IB_CC_SVCTYPE_UD; break; case IB_QPT_UC: - rlid = be16_to_cpu(rhdr->lrh[3]); + rlid = ib_get_slid(rhdr); rqpn = qp->remote_qpn; svc_type = IB_CC_SVCTYPE_UC; break; @@ -460,7 +462,7 @@ void hfi1_process_ecn_slowpath(struct rvt_qp *qp, struct hfi1_packet *pkt, struct ib_other_headers *ohdr = pkt->ohdr; struct ib_grh *grh = NULL; u32 rqpn = 0, bth1; - u16 rlid, dlid = be16_to_cpu(hdr->lrh[1]); + u16 rlid, dlid = ib_get_dlid(hdr); u8 sc, svc_type; bool is_mcast = false; @@ -471,19 +473,19 @@ void hfi1_process_ecn_slowpath(struct rvt_qp *qp, struct hfi1_packet *pkt, case IB_QPT_SMI: case IB_QPT_GSI: case IB_QPT_UD: - rlid = be16_to_cpu(hdr->lrh[3]); + rlid = ib_get_slid(hdr); rqpn = be32_to_cpu(ohdr->u.ud.deth[1]) & RVT_QPN_MASK; svc_type = IB_CC_SVCTYPE_UD; is_mcast = (dlid > be16_to_cpu(IB_MULTICAST_LID_BASE)) && (dlid != be16_to_cpu(IB_LID_PERMISSIVE)); break; case IB_QPT_UC: - rlid = qp->remote_ah_attr.dlid; + rlid = rdma_ah_get_dlid(&qp->remote_ah_attr); rqpn = qp->remote_qpn; svc_type = IB_CC_SVCTYPE_UC; break; case IB_QPT_RC: - rlid = qp->remote_ah_attr.dlid; + rlid = rdma_ah_get_dlid(&qp->remote_ah_attr); rqpn = qp->remote_qpn; svc_type = IB_CC_SVCTYPE_RC; break; @@ -491,16 +493,16 @@ void hfi1_process_ecn_slowpath(struct rvt_qp *qp, struct hfi1_packet *pkt, return; } - sc = hdr2sc(hdr, pkt->rhf); + sc = hfi1_9B_get_sc5(hdr, pkt->rhf); bth1 = be32_to_cpu(ohdr->bth[1]); - if (do_cnp && (bth1 & HFI1_FECN_SMASK)) { + if (do_cnp && (bth1 & IB_FECN_SMASK)) { u16 pkey = (u16)be32_to_cpu(ohdr->bth[0]); return_cnp(ibp, qp, rqpn, pkey, dlid, rlid, sc, grh); } - if (!is_mcast && (bth1 & HFI1_BECN_SMASK)) { + if (!is_mcast && (bth1 & IB_BECN_SMASK)) { struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); u32 lqpn = bth1 & RVT_QPN_MASK; u8 sl = ibp->sc_to_sl[sc]; @@ -621,8 +623,7 @@ static void __prescan_rxq(struct hfi1_packet *packet) packet->hdr = hfi1_get_msgheader(dd, rhf_addr); hdr = packet->hdr; - - lnh = be16_to_cpu(hdr->lrh[0]) & 3; + lnh = ib_get_lnh(hdr); if (lnh == HFI1_LRH_BTH) { packet->ohdr = &hdr->u.oth; @@ -634,7 +635,7 @@ static void __prescan_rxq(struct hfi1_packet *packet) } bth1 = be32_to_cpu(packet->ohdr->bth[1]); - is_ecn = !!(bth1 & (HFI1_FECN_SMASK | HFI1_BECN_SMASK)); + is_ecn = !!(bth1 & (IB_FECN_SMASK | IB_BECN_SMASK)); if (!is_ecn) goto next; @@ -652,7 +653,7 @@ static void __prescan_rxq(struct hfi1_packet *packet) rcu_read_unlock(); /* turn off BECN, FECN */ - bth1 &= ~(HFI1_FECN_SMASK | HFI1_BECN_SMASK); + bth1 &= ~(IB_FECN_SMASK | IB_BECN_SMASK); packet->ohdr->bth[1] = cpu_to_be32(bth1); next: update_ps_mdata(&mdata, rcd); @@ -872,20 +873,42 @@ bail: return last; } -static inline void set_all_nodma_rtail(struct hfi1_devdata *dd) +static inline void set_nodma_rtail(struct hfi1_devdata *dd, u8 ctxt) { int i; - for (i = HFI1_CTRL_CTXT + 1; i < dd->first_user_ctxt; i++) + /* + * For dynamically allocated kernel contexts (like vnic) switch + * interrupt handler only for that context. Otherwise, switch + * interrupt handler for all statically allocated kernel contexts. + */ + if (ctxt >= dd->first_dyn_alloc_ctxt) { + dd->rcd[ctxt]->do_interrupt = + &handle_receive_interrupt_nodma_rtail; + return; + } + + for (i = HFI1_CTRL_CTXT + 1; i < dd->first_dyn_alloc_ctxt; i++) dd->rcd[i]->do_interrupt = &handle_receive_interrupt_nodma_rtail; } -static inline void set_all_dma_rtail(struct hfi1_devdata *dd) +static inline void set_dma_rtail(struct hfi1_devdata *dd, u8 ctxt) { int i; - for (i = HFI1_CTRL_CTXT + 1; i < dd->first_user_ctxt; i++) + /* + * For dynamically allocated kernel contexts (like vnic) switch + * interrupt handler only for that context. Otherwise, switch + * interrupt handler for all statically allocated kernel contexts. + */ + if (ctxt >= dd->first_dyn_alloc_ctxt) { + dd->rcd[ctxt]->do_interrupt = + &handle_receive_interrupt_dma_rtail; + return; + } + + for (i = HFI1_CTRL_CTXT + 1; i < dd->first_dyn_alloc_ctxt; i++) dd->rcd[i]->do_interrupt = &handle_receive_interrupt_dma_rtail; } @@ -895,8 +918,13 @@ void set_all_slowpath(struct hfi1_devdata *dd) int i; /* HFI1_CTRL_CTXT must always use the slow path interrupt handler */ - for (i = HFI1_CTRL_CTXT + 1; i < dd->first_user_ctxt; i++) - dd->rcd[i]->do_interrupt = &handle_receive_interrupt; + for (i = HFI1_CTRL_CTXT + 1; i < dd->num_rcv_contexts; i++) { + struct hfi1_ctxtdata *rcd = dd->rcd[i]; + + if ((i < dd->first_dyn_alloc_ctxt) || + (rcd && rcd->sc && (rcd->sc->type == SC_KERNEL))) + rcd->do_interrupt = &handle_receive_interrupt; + } } static inline int set_armed_to_active(struct hfi1_ctxtdata *rcd, @@ -908,7 +936,8 @@ static inline int set_armed_to_active(struct hfi1_ctxtdata *rcd, packet->rhf_addr); u8 etype = rhf_rcv_type(packet->rhf); - if (etype == RHF_RCV_TYPE_IB && hdr2sc(hdr, packet->rhf) != 0xf) { + if (etype == RHF_RCV_TYPE_IB && + hfi1_9B_get_sc5(hdr, packet->rhf) != 0xf) { int hwstate = read_logical_state(dd); if (hwstate != LSTATE_ACTIVE) { @@ -1006,7 +1035,7 @@ int handle_receive_interrupt(struct hfi1_ctxtdata *rcd, int thread) last = RCV_PKT_DONE; if (needset) { dd_dev_info(dd, "Switching to NO_DMA_RTAIL\n"); - set_all_nodma_rtail(dd); + set_nodma_rtail(dd, rcd->ctxt); needset = 0; } } else { @@ -1028,7 +1057,7 @@ int handle_receive_interrupt(struct hfi1_ctxtdata *rcd, int thread) if (needset) { dd_dev_info(dd, "Switching to DMA_RTAIL\n"); - set_all_dma_rtail(dd); + set_dma_rtail(dd, rcd->ctxt); needset = 0; } } @@ -1077,10 +1106,10 @@ void receive_interrupt_work(struct work_struct *work) set_link_state(ppd, HLS_UP_ACTIVE); /* - * Interrupt all kernel contexts that could have had an - * interrupt during auto activation. + * Interrupt all statically allocated kernel contexts that could + * have had an interrupt during auto activation. */ - for (i = HFI1_CTRL_CTXT; i < dd->first_user_ctxt; i++) + for (i = HFI1_CTRL_CTXT; i < dd->first_dyn_alloc_ctxt; i++) force_recv_intr(dd->rcd[i]); } @@ -1294,7 +1323,8 @@ int hfi1_reset_device(int unit) spin_lock_irqsave(&dd->uctxt_lock, flags); if (dd->rcd) - for (i = dd->first_user_ctxt; i < dd->num_rcv_contexts; i++) { + for (i = dd->first_dyn_alloc_ctxt; + i < dd->num_rcv_contexts; i++) { if (!dd->rcd[i] || !dd->rcd[i]->cnt) continue; spin_unlock_irqrestore(&dd->uctxt_lock, flags); @@ -1354,6 +1384,9 @@ void handle_eflags(struct hfi1_packet *packet) */ int process_receive_ib(struct hfi1_packet *packet) { + if (unlikely(hfi1_dbg_fault_packet(packet))) + return RHF_RCV_CONTINUE; + trace_hfi1_rcvhdr(packet->rcd->ppd->dd, packet->rcd->ctxt, rhf_err_flags(packet->rhf), @@ -1363,6 +1396,11 @@ int process_receive_ib(struct hfi1_packet *packet) packet->updegr, rhf_egr_index(packet->rhf)); + if (unlikely( + (hfi1_dbg_fault_suppress_err(&packet->rcd->dd->verbs_dev) && + (packet->rhf & RHF_DC_ERR)))) + return RHF_RCV_CONTINUE; + if (unlikely(rhf_err_flags(packet->rhf))) { handle_eflags(packet); return RHF_RCV_CONTINUE; @@ -1372,15 +1410,31 @@ int process_receive_ib(struct hfi1_packet *packet) return RHF_RCV_CONTINUE; } +static inline bool hfi1_is_vnic_packet(struct hfi1_packet *packet) +{ + /* Packet received in VNIC context via RSM */ + if (packet->rcd->is_vnic) + return true; + + if ((HFI1_GET_L2_TYPE(packet->ebuf) == OPA_VNIC_L2_TYPE) && + (HFI1_GET_L4_TYPE(packet->ebuf) == OPA_VNIC_L4_ETHR)) + return true; + + return false; +} + int process_receive_bypass(struct hfi1_packet *packet) { struct hfi1_devdata *dd = packet->rcd->dd; - if (unlikely(rhf_err_flags(packet->rhf))) + if (unlikely(rhf_err_flags(packet->rhf))) { handle_eflags(packet); + } else if (hfi1_is_vnic_packet(packet)) { + hfi1_vnic_bypass_rcv(packet); + return RHF_RCV_CONTINUE; + } - dd_dev_err(dd, - "Bypass packets are not supported in normal operation. Dropping\n"); + dd_dev_err(dd, "Unsupported bypass packet. Dropping\n"); incr_cntr64(&dd->sw_rcv_bypass_packet_errors); if (!(dd->err_info_rcvport.status_and_code & OPA_EI_STATUS_SMASK)) { u64 *flits = packet->ebuf; @@ -1398,6 +1452,12 @@ int process_receive_bypass(struct hfi1_packet *packet) int process_receive_error(struct hfi1_packet *packet) { + /* KHdrHCRCErr -- KDETH packet with a bad HCRC */ + if (unlikely( + hfi1_dbg_fault_suppress_err(&packet->rcd->dd->verbs_dev) && + rhf_rcv_type_err(packet->rhf) == 3)) + return RHF_RCV_CONTINUE; + handle_eflags(packet); if (unlikely(rhf_err_flags(packet->rhf))) @@ -1409,6 +1469,8 @@ int process_receive_error(struct hfi1_packet *packet) int kdeth_process_expected(struct hfi1_packet *packet) { + if (unlikely(hfi1_dbg_fault_packet(packet))) + return RHF_RCV_CONTINUE; if (unlikely(rhf_err_flags(packet->rhf))) handle_eflags(packet); @@ -1421,6 +1483,8 @@ int kdeth_process_eager(struct hfi1_packet *packet) { if (unlikely(rhf_err_flags(packet->rhf))) handle_eflags(packet); + if (unlikely(hfi1_dbg_fault_packet(packet))) + return RHF_RCV_CONTINUE; dd_dev_err(packet->rcd->dd, "Unhandled eager packet received. Dropping.\n"); diff --git a/drivers/infiniband/hw/hfi1/file_ops.c b/drivers/infiniband/hw/hfi1/file_ops.c index f78c739b330a..3d9bce4bfcc7 100644 --- a/drivers/infiniband/hw/hfi1/file_ops.c +++ b/drivers/infiniband/hw/hfi1/file_ops.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -586,8 +586,8 @@ static int hfi1_file_mmap(struct file *fp, struct vm_area_struct *vma) * knows where it's own bitmap is within the page. */ memaddr = (unsigned long)(dd->events + - ((uctxt->ctxt - dd->first_user_ctxt) * - HFI1_MAX_SHARED_CTXTS)) & PAGE_MASK; + ((uctxt->ctxt - dd->first_dyn_alloc_ctxt) * + HFI1_MAX_SHARED_CTXTS)) & PAGE_MASK; memlen = PAGE_SIZE; /* * v3.7 removes VM_RESERVED but the effect is kept by @@ -597,6 +597,10 @@ static int hfi1_file_mmap(struct file *fp, struct vm_area_struct *vma) vmf = 1; break; case STATUS: + if (flags & (unsigned long)(VM_WRITE | VM_EXEC)) { + ret = -EPERM; + goto done; + } memaddr = kvirt_to_phys((void *)dd->status); memlen = PAGE_SIZE; flags |= VM_IO | VM_DONTEXPAND; @@ -756,7 +760,7 @@ static int hfi1_file_close(struct inode *inode, struct file *fp) * Clear any left over, unhandled events so the next process that * gets this context doesn't get confused. */ - ev = dd->events + ((uctxt->ctxt - dd->first_user_ctxt) * + ev = dd->events + ((uctxt->ctxt - dd->first_dyn_alloc_ctxt) * HFI1_MAX_SHARED_CTXTS) + fdata->subctxt; *ev = 0; @@ -909,12 +913,18 @@ static int find_shared_ctxt(struct file *fp, if (!(dd && (dd->flags & HFI1_PRESENT) && dd->kregbase)) continue; - for (i = dd->first_user_ctxt; i < dd->num_rcv_contexts; i++) { + for (i = dd->first_dyn_alloc_ctxt; + i < dd->num_rcv_contexts; i++) { struct hfi1_ctxtdata *uctxt = dd->rcd[i]; /* Skip ctxts which are not yet open */ if (!uctxt || !uctxt->cnt) continue; + + /* Skip dynamically allocted kernel contexts */ + if (uctxt->sc && (uctxt->sc->type == SC_KERNEL)) + continue; + /* Skip ctxt if it doesn't match the requested one */ if (memcmp(uctxt->uuid, uinfo->uuid, sizeof(uctxt->uuid)) || @@ -960,7 +970,8 @@ static int allocate_ctxt(struct file *fp, struct hfi1_devdata *dd, return -EIO; } - for (ctxt = dd->first_user_ctxt; ctxt < dd->num_rcv_contexts; ctxt++) + for (ctxt = dd->first_dyn_alloc_ctxt; + ctxt < dd->num_rcv_contexts; ctxt++) if (!dd->rcd[ctxt]) break; @@ -1306,7 +1317,7 @@ static int get_base_info(struct file *fp, void __user *ubase, __u32 len) */ binfo.user_regbase = HFI1_MMAP_TOKEN(UREGS, uctxt->ctxt, fd->subctxt, 0); - offset = offset_in_page((((uctxt->ctxt - dd->first_user_ctxt) * + offset = offset_in_page((((uctxt->ctxt - dd->first_dyn_alloc_ctxt) * HFI1_MAX_SHARED_CTXTS) + fd->subctxt) * sizeof(*dd->events)); binfo.events_bufbase = HFI1_MMAP_TOKEN(EVENTS, uctxt->ctxt, @@ -1400,12 +1411,12 @@ int hfi1_set_uevent_bits(struct hfi1_pportdata *ppd, const int evtbit) } spin_lock_irqsave(&dd->uctxt_lock, flags); - for (ctxt = dd->first_user_ctxt; ctxt < dd->num_rcv_contexts; + for (ctxt = dd->first_dyn_alloc_ctxt; ctxt < dd->num_rcv_contexts; ctxt++) { uctxt = dd->rcd[ctxt]; if (uctxt) { unsigned long *evs = dd->events + - (uctxt->ctxt - dd->first_user_ctxt) * + (uctxt->ctxt - dd->first_dyn_alloc_ctxt) * HFI1_MAX_SHARED_CTXTS; int i; /* @@ -1477,7 +1488,7 @@ static int user_event_ack(struct hfi1_ctxtdata *uctxt, int subctxt, if (!dd->events) return 0; - evs = dd->events + ((uctxt->ctxt - dd->first_user_ctxt) * + evs = dd->events + ((uctxt->ctxt - dd->first_dyn_alloc_ctxt) * HFI1_MAX_SHARED_CTXTS) + subctxt; for (i = 0; i <= _HFI1_MAX_EVENT_BIT; i++) { diff --git a/drivers/infiniband/hw/hfi1/firmware.c b/drivers/infiniband/hw/hfi1/firmware.c index 0dd50cdb039a..4042c11b2742 100644 --- a/drivers/infiniband/hw/hfi1/firmware.c +++ b/drivers/infiniband/hw/hfi1/firmware.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015 - 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -1004,7 +1004,9 @@ static int load_8051_firmware(struct hfi1_devdata *dd, { u64 reg; int ret; - u8 ver_a, ver_b; + u8 ver_major; + u8 ver_minor; + u8 ver_patch; /* * DC Reset sequence @@ -1073,10 +1075,10 @@ static int load_8051_firmware(struct hfi1_devdata *dd, return -ETIMEDOUT; } - read_misc_status(dd, &ver_a, &ver_b); - dd_dev_info(dd, "8051 firmware version %d.%d\n", - (int)ver_b, (int)ver_a); - dd->dc8051_ver = dc8051_ver(ver_b, ver_a); + read_misc_status(dd, &ver_major, &ver_minor, &ver_patch); + dd_dev_info(dd, "8051 firmware version %d.%d.%d\n", + (int)ver_major, (int)ver_minor, (int)ver_patch); + dd->dc8051_ver = dc8051_ver(ver_major, ver_minor, ver_patch); return 0; } diff --git a/drivers/infiniband/hw/hfi1/hfi.h b/drivers/infiniband/hw/hfi1/hfi.h index 0808e3c3ba39..f06674317abf 100644 --- a/drivers/infiniband/hw/hfi1/hfi.h +++ b/drivers/infiniband/hw/hfi1/hfi.h @@ -1,7 +1,7 @@ #ifndef _HFI1_KERNEL_H #define _HFI1_KERNEL_H /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -54,6 +54,7 @@ #include <linux/list.h> #include <linux/scatterlist.h> #include <linux/slab.h> +#include <linux/idr.h> #include <linux/io.h> #include <linux/fs.h> #include <linux/completion.h> @@ -66,6 +67,7 @@ #include <linux/i2c-algo-bit.h> #include <rdma/ib_hdrs.h> #include <linux/rhashtable.h> +#include <linux/netdevice.h> #include <rdma/rdma_vt.h> #include "chip_registers.h" @@ -278,6 +280,8 @@ struct hfi1_ctxtdata { struct hfi1_devdata *dd; /* so functions that need physical port can get it easily */ struct hfi1_pportdata *ppd; + /* associated msix interrupt */ + u32 msix_intr; /* A page of memory for rcvhdrhead, rcvegrhead, rcvegrtail * N */ void *subctxt_uregbase; /* An array of pages for the eager receive buffers * N */ @@ -337,6 +341,12 @@ struct hfi1_ctxtdata { * packets with the wrong interrupt handler. */ int (*do_interrupt)(struct hfi1_ctxtdata *rcd, int threaded); + + /* Indicates that this is vnic context */ + bool is_vnic; + + /* vnic queue index this context is mapped to */ + u8 vnic_q_idx; }; /* @@ -474,7 +484,7 @@ struct rvt_sge_state; #define HFI1_PART_ENFORCE_OUT 0x2 /* how often we check for synthetic counter wrap around */ -#define SYNTH_CNT_TIME 2 +#define SYNTH_CNT_TIME 3 /* Counter flags */ #define CNTR_NORMAL 0x0 /* Normal counters, just read register */ @@ -808,6 +818,32 @@ struct hfi1_asic_data { struct hfi1_i2c_bus *i2c_bus1; }; +/* sizes for both the QP and RSM map tables */ +#define NUM_MAP_ENTRIES 256 +#define NUM_MAP_REGS 32 + +/* + * Number of VNIC contexts used. Ensure it is less than or equal to + * max queues supported by VNIC (HFI1_VNIC_MAX_QUEUE). + */ +#define HFI1_NUM_VNIC_CTXT 8 + +/* Number of VNIC RSM entries */ +#define NUM_VNIC_MAP_ENTRIES 8 + +/* Virtual NIC information */ +struct hfi1_vnic_data { + struct hfi1_ctxtdata *ctxt[HFI1_NUM_VNIC_CTXT]; + struct kmem_cache *txreq_cache; + u8 num_vports; + struct idr vesw_idr; + u8 rmt_start; + u8 num_ctxt; + u32 msix_idx; +}; + +struct hfi1_vnic_vport_info; + /* device data struct now contains only "general per-device" info. * fields related to a physical IB port are in a hfi1_pportdata struct. */ @@ -926,8 +962,9 @@ struct hfi1_devdata { spinlock_t rcvctrl_lock; /* protect changes to RcvCtrl */ /* around rcd and (user ctxts) ctxt_cnt use (intr vs free) */ spinlock_t uctxt_lock; /* rcd and user context changes */ - /* exclusive access to 8051 */ - spinlock_t dc8051_lock; + struct mutex dc8051_lock; /* exclusive access to 8051 */ + struct workqueue_struct *update_cntr_wq; + struct work_struct update_cntr_work; /* exclusive access to 8051 memory */ spinlock_t dc8051_memlock; int dc8051_timed_out; /* remember if the 8051 timed out */ @@ -1020,7 +1057,7 @@ struct hfi1_devdata { u8 qos_shift; u16 irev; /* implementation revision */ - u16 dc8051_ver; /* 8051 firmware version */ + u32 dc8051_ver; /* 8051 firmware version */ spinlock_t hfi1_diag_trans_lock; /* protect diag observer ops */ struct platform_config platform_config; @@ -1031,6 +1068,7 @@ struct hfi1_devdata { /* MSI-X information */ struct hfi1_msix_entry *msix_entries; u32 num_msix_entries; + u32 first_dyn_msix_idx; /* INTx information */ u32 requested_intx_irq; /* did we request one? */ @@ -1115,6 +1153,9 @@ struct hfi1_devdata { send_routine process_dma_send; void (*pio_inline_send)(struct hfi1_devdata *dd, struct pio_buf *pbuf, u64 pbc, const void *from, size_t count); + int (*process_vnic_dma_send)(struct hfi1_devdata *dd, u8 q_idx, + struct hfi1_vnic_vport_info *vinfo, + struct sk_buff *skb, u64 pbc, u8 plen); /* hfi1_pportdata, points to array of (physical) port-specific * data structs, indexed by pidx (0..n-1) */ @@ -1126,8 +1167,8 @@ struct hfi1_devdata { u16 flags; /* Number of physical ports available */ u8 num_pports; - /* Lowest context number which can be used by user processes */ - u8 first_user_ctxt; + /* Lowest context number which can be used by user processes or VNIC */ + u8 first_dyn_alloc_ctxt; /* adding a new field here would make it part of this cacheline */ /* seqlock for sc2vl */ @@ -1167,15 +1208,24 @@ struct hfi1_devdata { bool eprom_available; /* true if EPROM is available for this device */ bool aspm_supported; /* Does HW support ASPM */ bool aspm_enabled; /* ASPM state: enabled/disabled */ - struct rhashtable sdma_rht; + struct rhashtable *sdma_rht; struct kobject kobj; + + /* vnic data */ + struct hfi1_vnic_data vnic; }; +static inline bool hfi1_vnic_is_rsm_full(struct hfi1_devdata *dd, int spare) +{ + return (dd->vnic.rmt_start + spare) > NUM_MAP_ENTRIES; +} + /* 8051 firmware version helper */ -#define dc8051_ver(a, b) ((a) << 8 | (b)) -#define dc8051_ver_maj(a) ((a & 0xff00) >> 8) -#define dc8051_ver_min(a) (a & 0x00ff) +#define dc8051_ver(a, b, c) ((a) << 16 | (b) << 8 | (c)) +#define dc8051_ver_maj(a) (((a) & 0xff0000) >> 16) +#define dc8051_ver_min(a) (((a) & 0x00ff00) >> 8) +#define dc8051_ver_patch(a) ((a) & 0x0000ff) /* f_put_tid types */ #define PT_EXPECTED 0 @@ -1235,6 +1285,9 @@ int handle_receive_interrupt(struct hfi1_ctxtdata *, int); int handle_receive_interrupt_nodma_rtail(struct hfi1_ctxtdata *, int); int handle_receive_interrupt_dma_rtail(struct hfi1_ctxtdata *, int); void set_all_slowpath(struct hfi1_devdata *dd); +void hfi1_vnic_synchronize_irq(struct hfi1_devdata *dd); +void hfi1_set_vnic_msix_info(struct hfi1_ctxtdata *rcd); +void hfi1_reset_vnic_msix_info(struct hfi1_ctxtdata *rcd); extern const struct pci_device_id hfi1_pci_tbl[]; @@ -1254,16 +1307,24 @@ int hfi1_reset_device(int); /* return the driver's idea of the logical OPA port state */ static inline u32 driver_lstate(struct hfi1_pportdata *ppd) { - return ppd->lstate; /* use the cached value */ + /* + * The driver does some processing from the time the logical + * link state is at INIT to the time the SM can be notified + * as such. Return IB_PORT_DOWN until the software state + * is ready. + */ + if (ppd->lstate == IB_PORT_INIT && !(ppd->host_link_state & HLS_UP)) + return IB_PORT_DOWN; + else + return ppd->lstate; } void receive_interrupt_work(struct work_struct *work); /* extract service channel from header and rhf */ -static inline int hdr2sc(struct ib_header *hdr, u64 rhf) +static inline int hfi1_9B_get_sc5(struct ib_header *hdr, u64 rhf) { - return ((be16_to_cpu(hdr->lrh[0]) >> 12) & 0xf) | - ((!!(rhf_dc_info(rhf))) << 4); + return ib_get_sc(hdr) | ((!!(rhf_dc_info(rhf))) << 4); } #define HFI1_JKEY_WIDTH 16 @@ -1597,9 +1658,9 @@ static inline bool process_ecn(struct rvt_qp *qp, struct hfi1_packet *pkt, u32 bth1; bth1 = be32_to_cpu(ohdr->bth[1]); - if (unlikely(bth1 & (HFI1_BECN_SMASK | HFI1_FECN_SMASK))) { + if (unlikely(bth1 & (IB_BECN_SMASK | IB_FECN_SMASK))) { hfi1_process_ecn_slowpath(qp, pkt, do_cnp); - return bth1 & HFI1_FECN_SMASK; + return !!(bth1 & IB_FECN_SMASK); } return false; } diff --git a/drivers/infiniband/hw/hfi1/init.c b/drivers/infiniband/hw/hfi1/init.c index f40864e9a3b2..4d6b9f82efa3 100644 --- a/drivers/infiniband/hw/hfi1/init.c +++ b/drivers/infiniband/hw/hfi1/init.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -65,6 +65,7 @@ #include "verbs.h" #include "aspm.h" #include "affinity.h" +#include "vnic.h" #undef pr_fmt #define pr_fmt(fmt) DRIVER_NAME ": " fmt @@ -139,7 +140,7 @@ int hfi1_create_ctxts(struct hfi1_devdata *dd) goto nomem; /* create one or more kernel contexts */ - for (i = 0; i < dd->first_user_ctxt; ++i) { + for (i = 0; i < dd->first_dyn_alloc_ctxt; ++i) { struct hfi1_pportdata *ppd; struct hfi1_ctxtdata *rcd; @@ -214,9 +215,9 @@ struct hfi1_ctxtdata *hfi1_create_ctxtdata(struct hfi1_pportdata *ppd, u32 ctxt, u32 base; if (dd->rcv_entries.nctxt_extra > - dd->num_rcv_contexts - dd->first_user_ctxt) + dd->num_rcv_contexts - dd->first_dyn_alloc_ctxt) kctxt_ngroups = (dd->rcv_entries.nctxt_extra - - (dd->num_rcv_contexts - dd->first_user_ctxt)); + (dd->num_rcv_contexts - dd->first_dyn_alloc_ctxt)); rcd = kzalloc_node(sizeof(*rcd), GFP_KERNEL, numa); if (rcd) { u32 rcvtids, max_entries; @@ -238,27 +239,29 @@ struct hfi1_ctxtdata *hfi1_create_ctxtdata(struct hfi1_pportdata *ppd, u32 ctxt, * Calculate the context's RcvArray entry starting point. * We do this here because we have to take into account all * the RcvArray entries that previous context would have - * taken and we have to account for any extra groups - * assigned to the kernel or user contexts. + * taken and we have to account for any extra groups assigned + * to the static (kernel) or dynamic (vnic/user) contexts. */ - if (ctxt < dd->first_user_ctxt) { + if (ctxt < dd->first_dyn_alloc_ctxt) { if (ctxt < kctxt_ngroups) { base = ctxt * (dd->rcv_entries.ngroups + 1); rcd->rcv_array_groups++; - } else + } else { base = kctxt_ngroups + (ctxt * dd->rcv_entries.ngroups); + } } else { - u16 ct = ctxt - dd->first_user_ctxt; + u16 ct = ctxt - dd->first_dyn_alloc_ctxt; base = ((dd->n_krcv_queues * dd->rcv_entries.ngroups) + kctxt_ngroups); if (ct < dd->rcv_entries.nctxt_extra) { base += ct * (dd->rcv_entries.ngroups + 1); rcd->rcv_array_groups++; - } else + } else { base += dd->rcv_entries.nctxt_extra + (ct * dd->rcv_entries.ngroups); + } } rcd->eager_base = base * dd->rcv_entries.group_size; @@ -322,7 +325,8 @@ struct hfi1_ctxtdata *hfi1_create_ctxtdata(struct hfi1_pportdata *ppd, u32 ctxt, } rcd->egrbufs.rcvtid_size = HFI1_MAX_EAGER_BUFFER_SIZE; - if (ctxt < dd->first_user_ctxt) { /* N/A for PSM contexts */ + /* Applicable only for statically created kernel contexts */ + if (ctxt < dd->first_dyn_alloc_ctxt) { rcd->opstats = kzalloc_node(sizeof(*rcd->opstats), GFP_KERNEL, numa); if (!rcd->opstats) @@ -482,6 +486,9 @@ void hfi1_init_pportdata(struct pci_dev *pdev, struct hfi1_pportdata *ppd, default_pkey_idx = 1; ppd->pkeys[default_pkey_idx] = DEFAULT_P_KEY; + ppd->part_enforce |= HFI1_PART_ENFORCE_IN; + ppd->part_enforce |= HFI1_PART_ENFORCE_OUT; + if (loopback) { hfi1_early_err(&pdev->dev, "Faking data partition 0x8001 in idx %u\n", @@ -585,7 +592,7 @@ static void enable_chip(struct hfi1_devdata *dd) * Enable kernel ctxts' receive and receive interrupt. * Other ctxts done as user opens and initializes them. */ - for (i = 0; i < dd->first_user_ctxt; ++i) { + for (i = 0; i < dd->first_dyn_alloc_ctxt; ++i) { rcvmask = HFI1_RCVCTRL_CTXT_ENB | HFI1_RCVCTRL_INTRAVAIL_ENB; rcvmask |= HFI1_CAP_KGET_MASK(dd->rcd[i]->flags, DMA_RTAIL) ? HFI1_RCVCTRL_TAILUPD_ENB : HFI1_RCVCTRL_TAILUPD_DIS; @@ -679,6 +686,7 @@ int hfi1_init(struct hfi1_devdata *dd, int reinit) dd->process_pio_send = hfi1_verbs_send_pio; dd->process_dma_send = hfi1_verbs_send_dma; dd->pio_inline_send = pio_copy; + dd->process_vnic_dma_send = hfi1_vnic_send_dma; if (is_ax(dd)) { atomic_set(&dd->drop_packet, DROP_PACKET_ON); @@ -714,7 +722,7 @@ int hfi1_init(struct hfi1_devdata *dd, int reinit) } /* dd->rcd can be NULL if early initialization failed */ - for (i = 0; dd->rcd && i < dd->first_user_ctxt; ++i) { + for (i = 0; dd->rcd && i < dd->first_dyn_alloc_ctxt; ++i) { /* * Set up the (kernel) rcvhdr queue and egr TIDs. If doing * re-init, the simplest way to handle this is to free @@ -1078,11 +1086,11 @@ struct hfi1_devdata *hfi1_alloc_devdata(struct pci_dev *pdev, size_t extra) spin_lock_init(&dd->uctxt_lock); spin_lock_init(&dd->hfi1_diag_trans_lock); spin_lock_init(&dd->sc_init_lock); - spin_lock_init(&dd->dc8051_lock); spin_lock_init(&dd->dc8051_memlock); seqlock_init(&dd->sc2vl_lock); spin_lock_init(&dd->sde_map_lock); spin_lock_init(&dd->pio_map_lock); + mutex_init(&dd->dc8051_lock); init_waitqueue_head(&dd->event_queue); dd->int_counter = alloc_percpu(u64); @@ -1425,6 +1433,16 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) /* First, lock the non-writable module parameters */ HFI1_CAP_LOCK(); + /* Validate dev ids */ + if (!(ent->device == PCI_DEVICE_ID_INTEL0 || + ent->device == PCI_DEVICE_ID_INTEL1)) { + hfi1_early_err(&pdev->dev, + "Failing on unknown Intel deviceid 0x%x\n", + ent->device); + ret = -ENODEV; + goto bail; + } + /* Validate some global module parameters */ ret = init_validate_rcvhdrcnt(&pdev->dev, rcvhdrcnt); if (ret) @@ -1470,15 +1488,6 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (ret) goto bail; - if (!(ent->device == PCI_DEVICE_ID_INTEL0 || - ent->device == PCI_DEVICE_ID_INTEL1)) { - hfi1_early_err(&pdev->dev, - "Failing on unknown Intel deviceid 0x%x\n", - ent->device); - ret = -ENODEV; - goto clean_bail; - } - /* * Do device-specific initialization, function table setup, dd * allocation, etc. @@ -1497,6 +1506,9 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) /* do the generic initialization */ initfail = hfi1_init(dd, 0); + /* setup vnic */ + hfi1_vnic_setup(dd); + ret = hfi1_register_ib_device(dd); /* @@ -1530,6 +1542,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) hfi1_device_remove(dd); if (!ret) hfi1_unregister_ib_device(dd); + hfi1_vnic_cleanup(dd); postinit_cleanup(dd); if (initfail) ret = initfail; @@ -1574,6 +1587,9 @@ static void remove_one(struct pci_dev *pdev) /* unregister from IB core */ hfi1_unregister_ib_device(dd); + /* cleanup vnic */ + hfi1_vnic_cleanup(dd); + /* * Disable the IB link, disable interrupts on the device, * clear dma engines, etc. @@ -1613,8 +1629,11 @@ int hfi1_create_rcvhdrq(struct hfi1_devdata *dd, struct hfi1_ctxtdata *rcd) amt = PAGE_ALIGN(rcd->rcvhdrq_cnt * rcd->rcvhdrqentsize * sizeof(u32)); - gfp_flags = (rcd->ctxt >= dd->first_user_ctxt) ? - GFP_USER : GFP_KERNEL; + if ((rcd->ctxt < dd->first_dyn_alloc_ctxt) || + (rcd->sc && (rcd->sc->type == SC_KERNEL))) + gfp_flags = GFP_KERNEL; + else + gfp_flags = GFP_USER; rcd->rcvhdrq = dma_zalloc_coherent( &dd->pcidev->dev, amt, &rcd->rcvhdrq_dma, gfp_flags | __GFP_COMP); diff --git a/drivers/infiniband/hw/hfi1/intr.c b/drivers/infiniband/hw/hfi1/intr.c index 65348d16ab2f..232014d46f79 100644 --- a/drivers/infiniband/hw/hfi1/intr.c +++ b/drivers/infiniband/hw/hfi1/intr.c @@ -131,19 +131,24 @@ void handle_linkup_change(struct hfi1_devdata *dd, u32 linkup) if (quick_linkup || dd->icode == ICODE_FUNCTIONAL_SIMULATOR) { set_up_vl15(dd, dd->vau, dd->vl15_init); assign_remote_cm_au_table(dd, dd->vcu); - ppd->neighbor_guid = - read_csr(dd, DC_DC8051_STS_REMOTE_GUID); - ppd->neighbor_type = - read_csr(dd, DC_DC8051_STS_REMOTE_NODE_TYPE) & - DC_DC8051_STS_REMOTE_NODE_TYPE_VAL_MASK; - ppd->neighbor_port_number = - read_csr(dd, DC_DC8051_STS_REMOTE_PORT_NO) & - DC_DC8051_STS_REMOTE_PORT_NO_VAL_SMASK; - dd_dev_info(dd, "Neighbor GUID: %llx Neighbor type %d\n", - ppd->neighbor_guid, - ppd->neighbor_type); } + ppd->neighbor_guid = + read_csr(dd, DC_DC8051_STS_REMOTE_GUID); + ppd->neighbor_type = + read_csr(dd, DC_DC8051_STS_REMOTE_NODE_TYPE) & + DC_DC8051_STS_REMOTE_NODE_TYPE_VAL_MASK; + ppd->neighbor_port_number = + read_csr(dd, DC_DC8051_STS_REMOTE_PORT_NO) & + DC_DC8051_STS_REMOTE_PORT_NO_VAL_SMASK; + ppd->neighbor_fm_security = + read_csr(dd, DC_DC8051_STS_REMOTE_FM_SECURITY) & + DC_DC8051_STS_LOCAL_FM_SECURITY_DISABLED_MASK; + dd_dev_info(dd, + "Neighbor Guid %llx, Type %d, Port Num %d\n", + ppd->neighbor_guid, ppd->neighbor_type, + ppd->neighbor_port_number); + /* physical link went up */ ppd->linkup = 1; ppd->offline_disabled_reason = diff --git a/drivers/infiniband/hw/hfi1/mad.c b/drivers/infiniband/hw/hfi1/mad.c index 09cda3c35e82..5977673a52d4 100644 --- a/drivers/infiniband/hw/hfi1/mad.c +++ b/drivers/infiniband/hw/hfi1/mad.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -53,6 +53,7 @@ #include "mad.h" #include "trace.h" #include "qp.h" +#include "vnic.h" /* the reset value from the FM is supposed to be 0xffff, handle both */ #define OPA_LINK_WIDTH_RESET_OLD 0x0fff @@ -650,9 +651,11 @@ static int __subn_get_opa_portinfo(struct opa_smp *smp, u32 am, u8 *data, OPA_PI_MASK_PORT_ACTIVE_OPTOMIZE : 0); pi->port_packet_format.supported = - cpu_to_be16(OPA_PORT_PACKET_FORMAT_9B); + cpu_to_be16(OPA_PORT_PACKET_FORMAT_9B | + OPA_PORT_PACKET_FORMAT_16B); pi->port_packet_format.enabled = - cpu_to_be16(OPA_PORT_PACKET_FORMAT_9B); + cpu_to_be16(OPA_PORT_PACKET_FORMAT_9B | + OPA_PORT_PACKET_FORMAT_16B); /* flit_control.interleave is (OPA V1, version .76): * bits use @@ -701,7 +704,13 @@ static int __subn_get_opa_portinfo(struct opa_smp *smp, u32 am, u8 *data, buffer_units |= (dd->vl15_init << 11) & OPA_PI_MASK_BUF_UNIT_VL15_INIT; pi->buffer_units = cpu_to_be32(buffer_units); - pi->opa_cap_mask = cpu_to_be16(OPA_CAP_MASK3_IsSharedSpaceSupported); + pi->opa_cap_mask = cpu_to_be16(OPA_CAP_MASK3_IsSharedSpaceSupported | + OPA_CAP_MASK3_IsEthOnFabricSupported); + /* Driver does not support mcast/collective configuration */ + pi->opa_cap_mask &= + cpu_to_be16(~OPA_CAP_MASK3_IsAddrRangeConfigSupported); + pi->collectivemask_multicastmask = ((HFI1_COLLECTIVE_NR & 0x7) + << 3 | (HFI1_MCAST_NR & 0x7)); /* HFI supports a replay buffer 128 LTPs in size */ pi->replay_depth.buffer = 0x80; @@ -1146,16 +1155,6 @@ static int __subn_set_opa_portinfo(struct opa_smp *smp, u32 am, u8 *data, ppd->linkinit_reason = (pi->partenforce_filterraw & OPA_PI_MASK_LINKINIT_REASON); - /* enable/disable SW pkey checking as per FM control */ - if (pi->partenforce_filterraw & OPA_PI_MASK_PARTITION_ENFORCE_IN) - ppd->part_enforce |= HFI1_PART_ENFORCE_IN; - else - ppd->part_enforce &= ~HFI1_PART_ENFORCE_IN; - - if (pi->partenforce_filterraw & OPA_PI_MASK_PARTITION_ENFORCE_OUT) - ppd->part_enforce |= HFI1_PART_ENFORCE_OUT; - else - ppd->part_enforce &= ~HFI1_PART_ENFORCE_OUT; /* Must be a valid unicast LID address. */ if ((smlid == 0 && ls_old > IB_PORT_INIT) || @@ -1167,9 +1166,9 @@ static int __subn_set_opa_portinfo(struct opa_smp *smp, u32 am, u8 *data, spin_lock_irqsave(&ibp->rvp.lock, flags); if (ibp->rvp.sm_ah) { if (smlid != ibp->rvp.sm_lid) - ibp->rvp.sm_ah->attr.dlid = smlid; + rdma_ah_set_dlid(&ibp->rvp.sm_ah->attr, smlid); if (msl != ibp->rvp.sm_sl) - ibp->rvp.sm_ah->attr.sl = msl; + rdma_ah_set_sl(&ibp->rvp.sm_ah->attr, msl); } spin_unlock_irqrestore(&ibp->rvp.lock, flags); if (smlid != ibp->rvp.sm_lid) @@ -1465,25 +1464,15 @@ static int __subn_set_opa_pkeytable(struct opa_smp *smp, u32 am, u8 *data, return __subn_get_opa_pkeytable(smp, am, data, ibdev, port, resp_len); } -static int get_sc2vlt_tables(struct hfi1_devdata *dd, void *data) -{ - u64 *val = data; - - *val++ = read_csr(dd, SEND_SC2VLT0); - *val++ = read_csr(dd, SEND_SC2VLT1); - *val++ = read_csr(dd, SEND_SC2VLT2); - *val++ = read_csr(dd, SEND_SC2VLT3); - return 0; -} - #define ILLEGAL_VL 12 /* * filter_sc2vlt changes mappings to VL15 to ILLEGAL_VL (except * for SC15, which must map to VL15). If we don't remap things this * way it is possible for VL15 counters to increment when we try to * send on a SC which is mapped to an invalid VL. + * When getting the table convert ILLEGAL_VL back to VL15. */ -static void filter_sc2vlt(void *data) +static void filter_sc2vlt(void *data, bool set) { int i; u8 *pd = data; @@ -1491,8 +1480,14 @@ static void filter_sc2vlt(void *data) for (i = 0; i < OPA_MAX_SCS; i++) { if (i == 15) continue; - if ((pd[i] & 0x1f) == 0xf) - pd[i] = ILLEGAL_VL; + + if (set) { + if ((pd[i] & 0x1f) == 0xf) + pd[i] = ILLEGAL_VL; + } else { + if ((pd[i] & 0x1f) == ILLEGAL_VL) + pd[i] = 0xf; + } } } @@ -1500,7 +1495,7 @@ static int set_sc2vlt_tables(struct hfi1_devdata *dd, void *data) { u64 *val = data; - filter_sc2vlt(data); + filter_sc2vlt(data, true); write_csr(dd, SEND_SC2VLT0, *val++); write_csr(dd, SEND_SC2VLT1, *val++); @@ -1512,6 +1507,19 @@ static int set_sc2vlt_tables(struct hfi1_devdata *dd, void *data) return 0; } +static int get_sc2vlt_tables(struct hfi1_devdata *dd, void *data) +{ + u64 *val = (u64 *)data; + + *val++ = read_csr(dd, SEND_SC2VLT0); + *val++ = read_csr(dd, SEND_SC2VLT1); + *val++ = read_csr(dd, SEND_SC2VLT2); + *val++ = read_csr(dd, SEND_SC2VLT3); + + filter_sc2vlt((u64 *)data, false); + return 0; +} + static int __subn_get_opa_sl_to_sc(struct opa_smp *smp, u32 am, u8 *data, struct ib_device *ibdev, u8 port, u32 *resp_len) @@ -1986,31 +1994,6 @@ struct opa_pma_mad { u8 data[2024]; } __packed; -struct opa_class_port_info { - u8 base_version; - u8 class_version; - __be16 cap_mask; - __be32 cap_mask2_resp_time; - - u8 redirect_gid[16]; - __be32 redirect_tc_fl; - __be32 redirect_lid; - __be32 redirect_sl_qp; - __be32 redirect_qkey; - - u8 trap_gid[16]; - __be32 trap_tc_fl; - __be32 trap_lid; - __be32 trap_hl_qp; - __be32 trap_qkey; - - __be16 trap_pkey; - __be16 redirect_pkey; - - u8 trap_sl_rsvd; - u8 reserved[3]; -} __packed; - struct opa_port_status_req { __u8 port_num; __u8 reserved[3]; diff --git a/drivers/infiniband/hw/hfi1/pcie.c b/drivers/infiniband/hw/hfi1/pcie.c index 0829fce06172..e39e01b79382 100644 --- a/drivers/infiniband/hw/hfi1/pcie.c +++ b/drivers/infiniband/hw/hfi1/pcie.c @@ -583,7 +583,7 @@ pci_mmio_enabled(struct pci_dev *pdev) if (words == ~0ULL) ret = PCI_ERS_RESULT_NEED_RESET; dd_dev_info(dd, - "HFI1 mmio_enabled function called, read wordscntr %Lx, returning %d\n", + "HFI1 mmio_enabled function called, read wordscntr %llx, returning %d\n", words, ret); } return ret; diff --git a/drivers/infiniband/hw/hfi1/pio.c b/drivers/infiniband/hw/hfi1/pio.c index 615be68e40b3..ed72b5aca139 100644 --- a/drivers/infiniband/hw/hfi1/pio.c +++ b/drivers/infiniband/hw/hfi1/pio.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -703,6 +703,7 @@ struct send_context *sc_alloc(struct hfi1_devdata *dd, int type, { struct send_context_info *sci; struct send_context *sc = NULL; + int req_type = type; dma_addr_t dma; unsigned long flags; u64 reg; @@ -729,6 +730,13 @@ struct send_context *sc_alloc(struct hfi1_devdata *dd, int type, return NULL; } + /* + * VNIC contexts are dynamically allocated. + * Hence, pick a user context for VNIC. + */ + if (type == SC_VNIC) + type = SC_USER; + spin_lock_irqsave(&dd->sc_lock, flags); ret = sc_hw_alloc(dd, type, &sw_index, &hw_context); if (ret) { @@ -738,6 +746,15 @@ struct send_context *sc_alloc(struct hfi1_devdata *dd, int type, return NULL; } + /* + * VNIC contexts are used by kernel driver. + * Hence, mark them as kernel contexts. + */ + if (req_type == SC_VNIC) { + dd->send_contexts[sw_index].type = SC_KERNEL; + type = SC_KERNEL; + } + sci = &dd->send_contexts[sw_index]; sci->sc = sc; diff --git a/drivers/infiniband/hw/hfi1/pio.h b/drivers/infiniband/hw/hfi1/pio.h index 867e5ffc3595..99ca5edb0b43 100644 --- a/drivers/infiniband/hw/hfi1/pio.h +++ b/drivers/infiniband/hw/hfi1/pio.h @@ -1,7 +1,7 @@ #ifndef _PIO_H #define _PIO_H /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -54,6 +54,12 @@ #define SC_USER 3 /* must be the last one: it may take all left */ #define SC_MAX 4 /* count of send context types */ +/* + * SC_VNIC types are allocated (dynamically) from the user context pool, + * (SC_USER) and used by kernel driver as kernel contexts (SC_KERNEL). + */ +#define SC_VNIC SC_MAX + /* invalid send context index */ #define INVALID_SCI 0xff @@ -195,7 +201,7 @@ struct sc_config_sizes { * | mask | --/ |--------------------| * |--------------------------| -/ | * | * | actual_vls (max 8) | -/ |--------------------| - * |--------------------------| --/ | ksc[n] -> sc n | + * |--------------------------| --/ | ksc[n-1] -> sc n | * | vls (max 8) | -/ +--------------------+ * |--------------------------| --/ * | map[0] |-/ @@ -208,21 +214,21 @@ struct sc_config_sizes { * |--------------------------| |--------------------| * | map[vls - 1] |- | * | * +--------------------------+ \- |--------------------| - * \- | ksc[m] -> sc m+n | + * \- | ksc[m-1] -> sc m+n | * \ +--------------------+ * \- * \ - * \- +--------------------+ - * \- | mask | - * \ |--------------------| - * \- | ksc[0] -> sc 1+m+n | - * \- |--------------------| - * >| ksc[1] -> sc 2+m+n | - * |--------------------| - * | * | - * |--------------------| - * | ksc[o] -> sc o+m+n | - * +--------------------+ + * \- +----------------------+ + * \- | mask | + * \ |----------------------| + * \- | ksc[0] -> sc 1+m+n | + * \- |----------------------| + * >| ksc[1] -> sc 2+m+n | + * |----------------------| + * | * | + * |----------------------| + * | ksc[o-1] -> sc o+m+n | + * +----------------------+ * */ diff --git a/drivers/infiniband/hw/hfi1/qp.c b/drivers/infiniband/hw/hfi1/qp.c index c4ebd097b20f..4573e4c9f35c 100644 --- a/drivers/infiniband/hw/hfi1/qp.c +++ b/drivers/infiniband/hw/hfi1/qp.c @@ -294,7 +294,7 @@ int hfi1_check_send_wqe(struct rvt_qp *qp, ah = ibah_to_rvtah(wqe->ud_wr.ah); if (wqe->length > (1 << ah->log_pmtu)) return -EINVAL; - if (ibp->sl_to_sc[ah->attr.sl] == 0xf) + if (ibp->sl_to_sc[rdma_ah_get_sl(&ah->attr)] == 0xf) return -EINVAL; default: break; @@ -631,8 +631,8 @@ void qp_iter_print(struct seq_file *s, struct qp_iter *iter) qp->s_tail, qp->s_head, qp->s_size, qp->s_avail, qp->remote_qpn, - qp->remote_ah_attr.dlid, - qp->remote_ah_attr.sl, + rdma_ah_get_dlid(&qp->remote_ah_attr), + rdma_ah_get_sl(&qp->remote_ah_attr), qp->pmtu, qp->s_retry, qp->s_retry_cnt, @@ -748,7 +748,7 @@ void hfi1_migrate_qp(struct rvt_qp *qp) qp->s_mig_state = IB_MIG_MIGRATED; qp->remote_ah_attr = qp->alt_ah_attr; - qp->port_num = qp->alt_ah_attr.port_num; + qp->port_num = rdma_ah_get_port_num(&qp->alt_ah_attr); qp->s_pkey_index = qp->s_alt_pkey_index; qp->s_flags |= RVT_S_AHG_CLEAR; priv->s_sc = ah_to_sc(qp->ibqp.device, &qp->remote_ah_attr); @@ -778,7 +778,7 @@ u32 mtu_from_qp(struct rvt_dev_info *rdi, struct rvt_qp *qp, u32 pmtu) u8 sc, vl; ibp = &dd->pport[qp->port_num - 1].ibport_data; - sc = ibp->sl_to_sc[qp->remote_ah_attr.sl]; + sc = ibp->sl_to_sc[rdma_ah_get_sl(&qp->remote_ah_attr)]; vl = sc_to_vlt(dd, sc); mtu = verbs_mtu_enum_to_int(qp->ibqp.device, pmtu); @@ -861,7 +861,7 @@ void hfi1_error_port_qps(struct hfi1_ibport *ibp, u8 sl) if (qp->port_num == ppd->port && (qp->ibqp.qp_type == IB_QPT_UC || qp->ibqp.qp_type == IB_QPT_RC) && - qp->remote_ah_attr.sl == sl && + rdma_ah_get_sl(&qp->remote_ah_attr) == sl && (ib_rvt_state_ops[qp->state] & RVT_POST_SEND_OK)) { spin_lock_irq(&qp->r_lock); diff --git a/drivers/infiniband/hw/hfi1/rc.c b/drivers/infiniband/hw/hfi1/rc.c index 7382be11afca..75a729cd0c3d 100644 --- a/drivers/infiniband/hw/hfi1/rc.c +++ b/drivers/infiniband/hw/hfi1/rc.c @@ -274,7 +274,7 @@ int hfi1_make_rc_req(struct rvt_qp *qp, struct hfi1_pkt_state *ps) goto bail_no_tx; ohdr = &ps->s_txreq->phdr.hdr.u.oth; - if (qp->remote_ah_attr.ah_flags & IB_AH_GRH) + if (rdma_ah_get_ah_flags(&qp->remote_ah_attr) & IB_AH_GRH) ohdr = &ps->s_txreq->phdr.hdr.u.l.oth; /* Sending responses has higher priority over sending requests. */ @@ -744,9 +744,10 @@ void hfi1_send_rc_ack(struct hfi1_ctxtdata *rcd, struct rvt_qp *qp, /* Construct the header */ /* header size in 32-bit words LRH+BTH+AETH = (8+12+4)/4 */ hwords = 6; - if (unlikely(qp->remote_ah_attr.ah_flags & IB_AH_GRH)) { + if (unlikely(rdma_ah_get_ah_flags(&qp->remote_ah_attr) & IB_AH_GRH)) { hwords += hfi1_make_grh(ibp, &hdr.u.l.grh, - &qp->remote_ah_attr.grh, hwords, 0); + rdma_ah_read_grh(&qp->remote_ah_attr), + hwords, 0); ohdr = &hdr.u.l.oth; lrh0 = HFI1_LRH_GRH; } else { @@ -763,17 +764,19 @@ void hfi1_send_rc_ack(struct hfi1_ctxtdata *rcd, struct rvt_qp *qp, IB_AETH_CREDIT_SHIFT)); else ohdr->u.aeth = rvt_compute_aeth(qp); - sc5 = ibp->sl_to_sc[qp->remote_ah_attr.sl]; + sc5 = ibp->sl_to_sc[rdma_ah_get_sl(&qp->remote_ah_attr)]; /* set PBC_DC_INFO bit (aka SC[4]) in pbc_flags */ pbc_flags |= ((!!(sc5 & 0x10)) << PBC_DC_INFO_SHIFT); - lrh0 |= (sc5 & 0xf) << 12 | (qp->remote_ah_attr.sl & 0xf) << 4; + lrh0 |= (sc5 & 0xf) << 12 | (rdma_ah_get_sl(&qp->remote_ah_attr) + & 0xf) << 4; hdr.lrh[0] = cpu_to_be16(lrh0); - hdr.lrh[1] = cpu_to_be16(qp->remote_ah_attr.dlid); + hdr.lrh[1] = cpu_to_be16(rdma_ah_get_dlid(&qp->remote_ah_attr)); hdr.lrh[2] = cpu_to_be16(hwords + SIZE_OF_CRC); - hdr.lrh[3] = cpu_to_be16(ppd->lid | qp->remote_ah_attr.src_path_bits); + hdr.lrh[3] = cpu_to_be16(ppd->lid | + rdma_ah_get_path_bits(&qp->remote_ah_attr)); ohdr->bth[0] = cpu_to_be32(bth0); ohdr->bth[1] = cpu_to_be32(qp->remote_qpn); - ohdr->bth[1] |= cpu_to_be32((!!is_fecn) << HFI1_BECN_SHIFT); + ohdr->bth[1] |= cpu_to_be32((!!is_fecn) << IB_BECN_SHIFT); ohdr->bth[2] = cpu_to_be32(mask_psn(qp->r_ack_psn)); /* Don't try to send ACKs if the link isn't ACTIVE */ @@ -994,12 +997,12 @@ void hfi1_rc_send_complete(struct rvt_qp *qp, struct ib_header *hdr) return; /* Find out where the BTH is */ - if ((be16_to_cpu(hdr->lrh[0]) & 3) == HFI1_LRH_BTH) + if (ib_get_lnh(hdr) == HFI1_LRH_BTH) ohdr = &hdr->u.oth; else ohdr = &hdr->u.l.oth; - opcode = be32_to_cpu(ohdr->bth[0]) >> 24; + opcode = ib_bth_get_opcode(ohdr); if (opcode >= OP(RDMA_READ_RESPONSE_FIRST) && opcode <= OP(ATOMIC_ACKNOWLEDGE)) { WARN_ON(!qp->s_rdma_ack_cnt); @@ -1028,13 +1031,17 @@ void hfi1_rc_send_complete(struct rvt_qp *qp, struct ib_header *hdr) cmp_psn(qp->s_sending_psn, qp->s_sending_hpsn) <= 0) break; s_last = qp->s_last; + trace_hfi1_qp_send_completion(qp, wqe, s_last); if (++s_last >= qp->s_size) s_last = 0; qp->s_last = s_last; /* see post_send() */ barrier(); rvt_put_swqe(wqe); - rvt_qp_swqe_complete(qp, wqe, IB_WC_SUCCESS); + rvt_qp_swqe_complete(qp, + wqe, + ib_hfi1_wc_opcode[wqe->wr.opcode], + IB_WC_SUCCESS); } /* * If we were waiting for sends to complete before re-sending, @@ -1076,12 +1083,16 @@ static struct rvt_swqe *do_rc_completion(struct rvt_qp *qp, rvt_put_swqe(wqe); s_last = qp->s_last; + trace_hfi1_qp_send_completion(qp, wqe, s_last); if (++s_last >= qp->s_size) s_last = 0; qp->s_last = s_last; /* see post_send() */ barrier(); - rvt_qp_swqe_complete(qp, wqe, IB_WC_SUCCESS); + rvt_qp_swqe_complete(qp, + wqe, + ib_hfi1_wc_opcode[wqe->wr.opcode], + IB_WC_SUCCESS); } else { struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); @@ -1092,10 +1103,11 @@ static struct rvt_swqe *do_rc_completion(struct rvt_qp *qp, */ if (ppd->dd->flags & HFI1_HAS_SEND_DMA) { struct sdma_engine *engine; + u8 sl = rdma_ah_get_sl(&qp->remote_ah_attr); u8 sc5; /* For now use sc to find engine */ - sc5 = ibp->sl_to_sc[qp->remote_ah_attr.sl]; + sc5 = ibp->sl_to_sc[sl]; engine = qp_to_sdma_engine(qp, sc5); sdma_engine_progress_schedule(engine); } @@ -1516,7 +1528,7 @@ read_middle: if (!do_rc_ack(qp, aeth, psn, opcode, 0, rcd)) goto ack_done; /* Get the number of bytes the message was padded by. */ - pad = (be32_to_cpu(ohdr->bth[0]) >> 20) & 3; + pad = ib_bth_get_pad(ohdr); /* * Check that the data size is >= 0 && <= pmtu. * Remember to account for ICRC (4). @@ -1540,7 +1552,7 @@ read_middle: if (unlikely(wqe->wr.opcode != IB_WR_RDMA_READ)) goto ack_op_err; /* Get the number of bytes the message was padded by. */ - pad = (be32_to_cpu(ohdr->bth[0]) >> 20) & 3; + pad = ib_bth_get_pad(ohdr); /* * Check that the data size is >= 1 && <= pmtu. * Remember to account for ICRC (4). @@ -1922,7 +1934,8 @@ void hfi1_rc_rcv(struct hfi1_packet *packet) int diff; struct ib_reth *reth; unsigned long flags; - int ret, is_fecn = 0; + int ret; + bool is_fecn = false; bool copy_last = false; u32 rkey; @@ -1934,7 +1947,7 @@ void hfi1_rc_rcv(struct hfi1_packet *packet) is_fecn = process_ecn(qp, packet, false); psn = be32_to_cpu(ohdr->bth[2]); - opcode = (bth0 >> 24) & 0xff; + opcode = ib_bth_get_opcode(ohdr); /* * Process responses (ACKs) before anything else. Note that the @@ -2065,7 +2078,7 @@ no_immediate_data: wc.ex.imm_data = 0; send_last: /* Get the number of bytes the message was padded by. */ - pad = (bth0 >> 20) & 3; + pad = ib_bth_get_pad(ohdr); /* Check for invalid length. */ /* LAST len should be >= 1 */ if (unlikely(tlen < (hdrsize + pad + 4))) @@ -2089,7 +2102,7 @@ send_last: wc.opcode = IB_WC_RECV; wc.qp = &qp->ibqp; wc.src_qp = qp->remote_qpn; - wc.slid = qp->remote_ah_attr.dlid; + wc.slid = rdma_ah_get_dlid(&qp->remote_ah_attr); /* * It seems that IB mandates the presence of an SL in a * work completion only for the UD transport (see section @@ -2101,7 +2114,7 @@ send_last: * * See also OPA Vol. 1, section 9.7.6, and table 9-17. */ - wc.sl = qp->remote_ah_attr.sl; + wc.sl = rdma_ah_get_sl(&qp->remote_ah_attr); /* zero fields that are N/A */ wc.vendor_err = 0; wc.pkey_index = 0; @@ -2378,7 +2391,7 @@ void hfi1_rc_hdrerr( return; psn = be32_to_cpu(ohdr->bth[2]); - opcode = (bth0 >> 24) & 0xff; + opcode = ib_bth_get_opcode(ohdr); /* Only deal with RDMA Writes for now */ if (opcode < IB_OPCODE_RC_RDMA_READ_RESPONSE_FIRST) { diff --git a/drivers/infiniband/hw/hfi1/ruc.c b/drivers/infiniband/hw/hfi1/ruc.c index aa15bcbfb079..891ba0a81bbd 100644 --- a/drivers/infiniband/hw/hfi1/ruc.c +++ b/drivers/infiniband/hw/hfi1/ruc.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015 - 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -219,72 +219,84 @@ int hfi1_ruc_check_hdr(struct hfi1_ibport *ibp, struct ib_header *hdr, { __be64 guid; unsigned long flags; - u8 sc5 = ibp->sl_to_sc[qp->remote_ah_attr.sl]; + u8 sc5 = ibp->sl_to_sc[rdma_ah_get_sl(&qp->remote_ah_attr)]; if (qp->s_mig_state == IB_MIG_ARMED && (bth0 & IB_BTH_MIG_REQ)) { if (!has_grh) { - if (qp->alt_ah_attr.ah_flags & IB_AH_GRH) + if (rdma_ah_get_ah_flags(&qp->alt_ah_attr) & + IB_AH_GRH) goto err; } else { - if (!(qp->alt_ah_attr.ah_flags & IB_AH_GRH)) + const struct ib_global_route *grh; + + if (!(rdma_ah_get_ah_flags(&qp->alt_ah_attr) & + IB_AH_GRH)) goto err; - guid = get_sguid(ibp, qp->alt_ah_attr.grh.sgid_index); + grh = rdma_ah_read_grh(&qp->alt_ah_attr); + guid = get_sguid(ibp, grh->sgid_index); if (!gid_ok(&hdr->u.l.grh.dgid, ibp->rvp.gid_prefix, guid)) goto err; if (!gid_ok( &hdr->u.l.grh.sgid, - qp->alt_ah_attr.grh.dgid.global.subnet_prefix, - qp->alt_ah_attr.grh.dgid.global.interface_id)) + grh->dgid.global.subnet_prefix, + grh->dgid.global.interface_id)) goto err; } - if (unlikely(rcv_pkey_check(ppd_from_ibp(ibp), (u16)bth0, - sc5, be16_to_cpu(hdr->lrh[3])))) { + if (unlikely(rcv_pkey_check(ppd_from_ibp(ibp), (u16)bth0, sc5, + ib_get_slid(hdr)))) { hfi1_bad_pqkey(ibp, OPA_TRAP_BAD_P_KEY, (u16)bth0, - (be16_to_cpu(hdr->lrh[0]) >> 4) & 0xF, + ib_get_sl(hdr), 0, qp->ibqp.qp_num, - be16_to_cpu(hdr->lrh[3]), - be16_to_cpu(hdr->lrh[1])); + ib_get_slid(hdr), + ib_get_dlid(hdr)); goto err; } /* Validate the SLID. See Ch. 9.6.1.5 and 17.2.8 */ - if (be16_to_cpu(hdr->lrh[3]) != qp->alt_ah_attr.dlid || - ppd_from_ibp(ibp)->port != qp->alt_ah_attr.port_num) + if (ib_get_slid(hdr) != + rdma_ah_get_dlid(&qp->alt_ah_attr) || + ppd_from_ibp(ibp)->port != + rdma_ah_get_port_num(&qp->alt_ah_attr)) goto err; spin_lock_irqsave(&qp->s_lock, flags); hfi1_migrate_qp(qp); spin_unlock_irqrestore(&qp->s_lock, flags); } else { if (!has_grh) { - if (qp->remote_ah_attr.ah_flags & IB_AH_GRH) + if (rdma_ah_get_ah_flags(&qp->remote_ah_attr) & + IB_AH_GRH) goto err; } else { - if (!(qp->remote_ah_attr.ah_flags & IB_AH_GRH)) + const struct ib_global_route *grh; + + if (!(rdma_ah_get_ah_flags(&qp->remote_ah_attr) & + IB_AH_GRH)) goto err; - guid = get_sguid(ibp, - qp->remote_ah_attr.grh.sgid_index); + grh = rdma_ah_read_grh(&qp->remote_ah_attr); + guid = get_sguid(ibp, grh->sgid_index); if (!gid_ok(&hdr->u.l.grh.dgid, ibp->rvp.gid_prefix, guid)) goto err; if (!gid_ok( &hdr->u.l.grh.sgid, - qp->remote_ah_attr.grh.dgid.global.subnet_prefix, - qp->remote_ah_attr.grh.dgid.global.interface_id)) + grh->dgid.global.subnet_prefix, + grh->dgid.global.interface_id)) goto err; } - if (unlikely(rcv_pkey_check(ppd_from_ibp(ibp), (u16)bth0, - sc5, be16_to_cpu(hdr->lrh[3])))) { + if (unlikely(rcv_pkey_check(ppd_from_ibp(ibp), (u16)bth0, sc5, + ib_get_slid(hdr)))) { hfi1_bad_pqkey(ibp, OPA_TRAP_BAD_P_KEY, (u16)bth0, - (be16_to_cpu(hdr->lrh[0]) >> 4) & 0xF, + ib_get_sl(hdr), 0, qp->ibqp.qp_num, - be16_to_cpu(hdr->lrh[3]), - be16_to_cpu(hdr->lrh[1])); + ib_get_slid(hdr), + ib_get_dlid(hdr)); goto err; } /* Validate the SLID. See Ch. 9.6.1.5 */ - if (be16_to_cpu(hdr->lrh[3]) != qp->remote_ah_attr.dlid || + if (ib_get_slid(hdr) != + rdma_ah_get_dlid(&qp->remote_ah_attr) || ppd_from_ibp(ibp)->port != qp->port_num) goto err; if (qp->s_mig_state == IB_MIG_REARM && @@ -542,8 +554,8 @@ do_write: wc.byte_len = wqe->length; wc.qp = &qp->ibqp; wc.src_qp = qp->remote_qpn; - wc.slid = qp->remote_ah_attr.dlid; - wc.sl = qp->remote_ah_attr.sl; + wc.slid = rdma_ah_get_dlid(&qp->remote_ah_attr); + wc.sl = rdma_ah_get_sl(&qp->remote_ah_attr); wc.port_num = 1; /* Signal completion event if the solicited bit is set. */ rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc, @@ -637,7 +649,7 @@ done: * Return the size of the header in 32 bit words. */ u32 hfi1_make_grh(struct hfi1_ibport *ibp, struct ib_grh *hdr, - struct ib_global_route *grh, u32 hwords, u32 nwords) + const struct ib_global_route *grh, u32 hwords, u32 nwords) { hdr->version_tclass_flow = cpu_to_be32((IB_GRH_VERSION << IB_GRH_VERSION_SHIFT) | @@ -731,15 +743,17 @@ void hfi1_make_ruc_header(struct rvt_qp *qp, struct ib_other_headers *ohdr, extra_bytes = -ps->s_txreq->s_cur_size & 3; nwords = (ps->s_txreq->s_cur_size + extra_bytes) >> 2; lrh0 = HFI1_LRH_BTH; - if (unlikely(qp->remote_ah_attr.ah_flags & IB_AH_GRH)) { - qp->s_hdrwords += hfi1_make_grh(ibp, - &ps->s_txreq->phdr.hdr.u.l.grh, - &qp->remote_ah_attr.grh, - qp->s_hdrwords, nwords); + if (unlikely(rdma_ah_get_ah_flags(&qp->remote_ah_attr) & IB_AH_GRH)) { + qp->s_hdrwords += + hfi1_make_grh(ibp, + &ps->s_txreq->phdr.hdr.u.l.grh, + rdma_ah_read_grh(&qp->remote_ah_attr), + qp->s_hdrwords, nwords); lrh0 = HFI1_LRH_GRH; middle = 0; } - lrh0 |= (priv->s_sc & 0xf) << 12 | (qp->remote_ah_attr.sl & 0xf) << 4; + lrh0 |= (priv->s_sc & 0xf) << 12 | + (rdma_ah_get_sl(&qp->remote_ah_attr) & 0xf) << 4; /* * reset s_ahg/AHG fields * @@ -763,11 +777,13 @@ void hfi1_make_ruc_header(struct rvt_qp *qp, struct ib_other_headers *ohdr, else qp->s_flags &= ~RVT_S_AHG_VALID; ps->s_txreq->phdr.hdr.lrh[0] = cpu_to_be16(lrh0); - ps->s_txreq->phdr.hdr.lrh[1] = cpu_to_be16(qp->remote_ah_attr.dlid); + ps->s_txreq->phdr.hdr.lrh[1] = + cpu_to_be16(rdma_ah_get_dlid(&qp->remote_ah_attr)); ps->s_txreq->phdr.hdr.lrh[2] = cpu_to_be16(qp->s_hdrwords + nwords + SIZE_OF_CRC); - ps->s_txreq->phdr.hdr.lrh[3] = cpu_to_be16(ppd_from_ibp(ibp)->lid | - qp->remote_ah_attr.src_path_bits); + ps->s_txreq->phdr.hdr.lrh[3] = + cpu_to_be16(ppd_from_ibp(ibp)->lid | + rdma_ah_get_path_bits(&qp->remote_ah_attr)); bth0 |= hfi1_get_pkey(ibp, qp->s_pkey_index); bth0 |= extra_bytes << 20; ohdr->bth[0] = cpu_to_be32(bth0); @@ -775,7 +791,7 @@ void hfi1_make_ruc_header(struct rvt_qp *qp, struct ib_other_headers *ohdr, if (qp->s_flags & RVT_S_ECN) { qp->s_flags &= ~RVT_S_ECN; /* we recently received a FECN, so return a BECN */ - bth1 |= (HFI1_BECN_MASK << HFI1_BECN_SHIFT); + bth1 |= (IB_BECN_MASK << IB_BECN_SHIFT); } ohdr->bth[1] = cpu_to_be32(bth1); ohdr->bth[2] = cpu_to_be32(bth2); @@ -784,23 +800,29 @@ void hfi1_make_ruc_header(struct rvt_qp *qp, struct ib_other_headers *ohdr, /* when sending, force a reschedule every one of these periods */ #define SEND_RESCHED_TIMEOUT (5 * HZ) /* 5s in jiffies */ +void hfi1_do_send_from_rvt(struct rvt_qp *qp) +{ + hfi1_do_send(qp, false); +} + void _hfi1_do_send(struct work_struct *work) { struct iowait *wait = container_of(work, struct iowait, iowork); struct rvt_qp *qp = iowait_to_qp(wait); - hfi1_do_send(qp); + hfi1_do_send(qp, true); } /** * hfi1_do_send - perform a send on a QP * @work: contains a pointer to the QP + * @in_thread: true if in a workqueue thread * * Process entries in the send work queue until credit or queue is * exhausted. Only allow one CPU to send a packet per QP. * Otherwise, two threads could send packets out of order. */ -void hfi1_do_send(struct rvt_qp *qp) +void hfi1_do_send(struct rvt_qp *qp, bool in_thread) { struct hfi1_pkt_state ps; struct hfi1_qp_priv *priv = qp->priv; @@ -815,9 +837,9 @@ void hfi1_do_send(struct rvt_qp *qp) switch (qp->ibqp.qp_type) { case IB_QPT_RC: - if (!loopback && ((qp->remote_ah_attr.dlid & ~((1 << ps.ppd->lmc - ) - 1)) == - ps.ppd->lid)) { + if (!loopback && ((rdma_ah_get_dlid(&qp->remote_ah_attr) & + ~((1 << ps.ppd->lmc) - 1)) == + ps.ppd->lid)) { ruc_loopback(qp); return; } @@ -825,9 +847,9 @@ void hfi1_do_send(struct rvt_qp *qp) timeout_int = (qp->timeout_jiffies); break; case IB_QPT_UC: - if (!loopback && ((qp->remote_ah_attr.dlid & ~((1 << ps.ppd->lmc - ) - 1)) == - ps.ppd->lid)) { + if (!loopback && ((rdma_ah_get_dlid(&qp->remote_ah_attr) & + ~((1 << ps.ppd->lmc) - 1)) == + ps.ppd->lid)) { ruc_loopback(qp); return; } @@ -868,8 +890,10 @@ void hfi1_do_send(struct rvt_qp *qp) qp->s_hdrwords = 0; /* allow other tasks to run */ if (unlikely(time_after(jiffies, timeout))) { - if (workqueue_congested(cpu, - ps.ppd->hfi1_wq)) { + if (!in_thread || + workqueue_congested( + cpu, + ps.ppd->hfi1_wq)) { spin_lock_irqsave( &qp->s_lock, ps.flags); @@ -882,11 +906,9 @@ void hfi1_do_send(struct rvt_qp *qp) *ps.ppd->dd->send_schedule); return; } - if (!irqs_disabled()) { - cond_resched(); - this_cpu_inc( - *ps.ppd->dd->send_schedule); - } + cond_resched(); + this_cpu_inc( + *ps.ppd->dd->send_schedule); timeout = jiffies + (timeout_int) / 8; } spin_lock_irqsave(&qp->s_lock, ps.flags); @@ -909,8 +931,10 @@ void hfi1_send_complete(struct rvt_qp *qp, struct rvt_swqe *wqe, last = qp->s_last; old_last = last; + trace_hfi1_qp_send_completion(qp, wqe, last); if (++last >= qp->s_size) last = 0; + trace_hfi1_qp_send_completion(qp, wqe, last); qp->s_last = last; /* See post_send() */ barrier(); @@ -920,7 +944,10 @@ void hfi1_send_complete(struct rvt_qp *qp, struct rvt_swqe *wqe, qp->ibqp.qp_type == IB_QPT_GSI) atomic_dec(&ibah_to_rvtah(wqe->ud_wr.ah)->refcount); - rvt_qp_swqe_complete(qp, wqe, status); + rvt_qp_swqe_complete(qp, + wqe, + ib_hfi1_wc_opcode[wqe->wr.opcode], + status); if (qp->s_acked == old_last) qp->s_acked = last; diff --git a/drivers/infiniband/hw/hfi1/sdma.c b/drivers/infiniband/hw/hfi1/sdma.c index 5cde1ecda0fe..bfd0d5187e9b 100644 --- a/drivers/infiniband/hw/hfi1/sdma.c +++ b/drivers/infiniband/hw/hfi1/sdma.c @@ -868,7 +868,7 @@ struct sdma_engine *sdma_select_user_engine(struct hfi1_devdata *dd, cpu_id = smp_processor_id(); rcu_read_lock(); - rht_node = rhashtable_lookup_fast(&dd->sdma_rht, &cpu_id, + rht_node = rhashtable_lookup_fast(dd->sdma_rht, &cpu_id, sdma_rht_params); if (rht_node && rht_node->map[vl]) { @@ -962,7 +962,12 @@ ssize_t sdma_set_cpu_to_sde_map(struct sdma_engine *sde, const char *buf, continue; } - rht_node = rhashtable_lookup_fast(&dd->sdma_rht, &cpu, + if (vl >= ARRAY_SIZE(rht_node->map)) { + ret = -EINVAL; + goto out; + } + + rht_node = rhashtable_lookup_fast(dd->sdma_rht, &cpu, sdma_rht_params); if (!rht_node) { rht_node = kzalloc(sizeof(*rht_node), GFP_KERNEL); @@ -982,7 +987,7 @@ ssize_t sdma_set_cpu_to_sde_map(struct sdma_engine *sde, const char *buf, rht_node->map[vl]->ctr = 1; rht_node->map[vl]->sde[0] = sde; - ret = rhashtable_insert_fast(&dd->sdma_rht, + ret = rhashtable_insert_fast(dd->sdma_rht, &rht_node->node, sdma_rht_params); if (ret) { @@ -1025,7 +1030,7 @@ ssize_t sdma_set_cpu_to_sde_map(struct sdma_engine *sde, const char *buf, if (cpumask_test_cpu(cpu, mask)) continue; - rht_node = rhashtable_lookup_fast(&dd->sdma_rht, &cpu, + rht_node = rhashtable_lookup_fast(dd->sdma_rht, &cpu, sdma_rht_params); if (rht_node) { bool empty = true; @@ -1049,7 +1054,7 @@ ssize_t sdma_set_cpu_to_sde_map(struct sdma_engine *sde, const char *buf, } if (empty) { - ret = rhashtable_remove_fast(&dd->sdma_rht, + ret = rhashtable_remove_fast(dd->sdma_rht, &rht_node->node, sdma_rht_params); WARN_ON(ret); @@ -1108,7 +1113,7 @@ void sdma_seqfile_dump_cpu_list(struct seq_file *s, struct sdma_rht_node *rht_node; int i, j; - rht_node = rhashtable_lookup_fast(&dd->sdma_rht, &cpuid, + rht_node = rhashtable_lookup_fast(dd->sdma_rht, &cpuid, sdma_rht_params); if (!rht_node) return; @@ -1322,6 +1327,12 @@ static void sdma_clean(struct hfi1_devdata *dd, size_t num_engines) synchronize_rcu(); kfree(dd->per_sdma); dd->per_sdma = NULL; + + if (dd->sdma_rht) { + rhashtable_free_and_destroy(dd->sdma_rht, sdma_rht_free, NULL); + kfree(dd->sdma_rht); + dd->sdma_rht = NULL; + } } /** @@ -1341,12 +1352,14 @@ int sdma_init(struct hfi1_devdata *dd, u8 port) { unsigned this_idx; struct sdma_engine *sde; + struct rhashtable *tmp_sdma_rht; u16 descq_cnt; void *curr_head; struct hfi1_pportdata *ppd = dd->pport + port; u32 per_sdma_credits; uint idle_cnt = sdma_idle_cnt; size_t num_engines = dd->chip_sdma_engines; + int ret = -ENOMEM; if (!HFI1_CAP_IS_KSET(SDMA)) { HFI1_CAP_CLEAR(SDMA_AHG); @@ -1378,7 +1391,7 @@ int sdma_init(struct hfi1_devdata *dd, u8 port) /* alloc memory for array of send engines */ dd->per_sdma = kcalloc(num_engines, sizeof(*dd->per_sdma), GFP_KERNEL); if (!dd->per_sdma) - return -ENOMEM; + return ret; idle_cnt = ns_to_cclock(dd, idle_cnt); if (!sdma_desct_intr) @@ -1507,18 +1520,27 @@ int sdma_init(struct hfi1_devdata *dd, u8 port) dd->flags |= HFI1_HAS_SEND_DMA; dd->flags |= idle_cnt ? HFI1_HAS_SDMA_TIMEOUT : 0; dd->num_sdma = num_engines; - if (sdma_map_init(dd, port, ppd->vls_operational, NULL)) + ret = sdma_map_init(dd, port, ppd->vls_operational, NULL); + if (ret < 0) goto bail; - if (rhashtable_init(&dd->sdma_rht, &sdma_rht_params)) + tmp_sdma_rht = kzalloc(sizeof(*tmp_sdma_rht), GFP_KERNEL); + if (!tmp_sdma_rht) { + ret = -ENOMEM; goto bail; + } + + ret = rhashtable_init(tmp_sdma_rht, &sdma_rht_params); + if (ret < 0) + goto bail; + dd->sdma_rht = tmp_sdma_rht; dd_dev_info(dd, "SDMA num_sdma: %u\n", dd->num_sdma); return 0; bail: sdma_clean(dd, num_engines); - return -ENOMEM; + return ret; } /** @@ -1604,7 +1626,6 @@ void sdma_exit(struct hfi1_devdata *dd) sdma_finalput(&sde->state); } sdma_clean(dd, dd->num_sdma); - rhashtable_free_and_destroy(&dd->sdma_rht, sdma_rht_free, NULL); } /* diff --git a/drivers/infiniband/hw/hfi1/sdma.h b/drivers/infiniband/hw/hfi1/sdma.h index 21f1e2834f37..64f10b8b5db8 100644 --- a/drivers/infiniband/hw/hfi1/sdma.h +++ b/drivers/infiniband/hw/hfi1/sdma.h @@ -966,34 +966,34 @@ void sdma_engine_interrupt(struct sdma_engine *sde, u64 status); * | mask | --/ |--------------------| * |--------------------------| -/ | * | * | actual_vls (max 8) | -/ |--------------------| - * |--------------------------| --/ | sde[n] -> eng n | + * |--------------------------| --/ | sde[n-1] -> eng n | * | vls (max 8) | -/ +--------------------+ * |--------------------------| --/ * | map[0] |-/ - * |--------------------------| +--------------------+ - * | map[1] |--- | mask | - * |--------------------------| \---- |--------------------| - * | * | \-- | sde[0] -> eng 1+n | - * | * | \---- |--------------------| - * | * | \->| sde[1] -> eng 2+n | - * |--------------------------| |--------------------| - * | map[vls - 1] |- | * | - * +--------------------------+ \- |--------------------| - * \- | sde[m] -> eng m+n | - * \ +--------------------+ + * |--------------------------| +---------------------+ + * | map[1] |--- | mask | + * |--------------------------| \---- |---------------------| + * | * | \-- | sde[0] -> eng 1+n | + * | * | \---- |---------------------| + * | * | \->| sde[1] -> eng 2+n | + * |--------------------------| |---------------------| + * | map[vls - 1] |- | * | + * +--------------------------+ \- |---------------------| + * \- | sde[m-1] -> eng m+n | + * \ +---------------------+ * \- * \ - * \- +--------------------+ - * \- | mask | - * \ |--------------------| - * \- | sde[0] -> eng 1+m+n| - * \- |--------------------| - * >| sde[1] -> eng 2+m+n| - * |--------------------| - * | * | - * |--------------------| - * | sde[o] -> eng o+m+n| - * +--------------------+ + * \- +----------------------+ + * \- | mask | + * \ |----------------------| + * \- | sde[0] -> eng 1+m+n | + * \- |----------------------| + * >| sde[1] -> eng 2+m+n | + * |----------------------| + * | * | + * |----------------------| + * | sde[o-1] -> eng o+m+n| + * +----------------------+ * */ diff --git a/drivers/infiniband/hw/hfi1/sysfs.c b/drivers/infiniband/hw/hfi1/sysfs.c index 919a5474e651..50d140d25e38 100644 --- a/drivers/infiniband/hw/hfi1/sysfs.c +++ b/drivers/infiniband/hw/hfi1/sysfs.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -542,7 +542,7 @@ static ssize_t show_nctxts(struct device *device, * give a more accurate picture of total contexts available. */ return scnprintf(buf, PAGE_SIZE, "%u\n", - min(dd->num_rcv_contexts - dd->first_user_ctxt, + min(dd->num_rcv_contexts - dd->first_dyn_alloc_ctxt, (u32)dd->sc_sizes[SC_USER].count)); } diff --git a/drivers/infiniband/hw/hfi1/trace.c b/drivers/infiniband/hw/hfi1/trace.c index e86798af6903..eafae487face 100644 --- a/drivers/infiniband/hw/hfi1/trace.c +++ b/drivers/infiniband/hw/hfi1/trace.c @@ -51,13 +51,12 @@ u8 ibhdr_exhdr_len(struct ib_header *hdr) { struct ib_other_headers *ohdr; u8 opcode; - u8 lnh = (u8)(be16_to_cpu(hdr->lrh[0]) & 3); - if (lnh == HFI1_LRH_BTH) + if (ib_get_lnh(hdr) == HFI1_LRH_BTH) ohdr = &hdr->u.oth; else ohdr = &hdr->u.l.oth; - opcode = be32_to_cpu(ohdr->bth[0]) >> 24; + opcode = ib_bth_get_opcode(ohdr); return hdr_len_by_opcode[opcode] == 0 ? 0 : hdr_len_by_opcode[opcode] - (12 + 8); } diff --git a/drivers/infiniband/hw/hfi1/trace_ibhdrs.h b/drivers/infiniband/hw/hfi1/trace_ibhdrs.h index 382fcda3a5f6..090f6b506953 100644 --- a/drivers/infiniband/hw/hfi1/trace_ibhdrs.h +++ b/drivers/infiniband/hw/hfi1/trace_ibhdrs.h @@ -139,11 +139,11 @@ DECLARE_EVENT_CLASS(hfi1_ibhdr_template, __entry->pkey = be32_to_cpu(ohdr->bth[0]) & 0xffff; __entry->f = - (be32_to_cpu(ohdr->bth[1]) >> HFI1_FECN_SHIFT) & - HFI1_FECN_MASK; + (be32_to_cpu(ohdr->bth[1]) >> IB_FECN_SHIFT) & + IB_FECN_MASK; __entry->b = - (be32_to_cpu(ohdr->bth[1]) >> HFI1_BECN_SHIFT) & - HFI1_BECN_MASK; + (be32_to_cpu(ohdr->bth[1]) >> IB_BECN_SHIFT) & + IB_BECN_MASK; __entry->qpn = be32_to_cpu(ohdr->bth[1]) & RVT_QPN_MASK; __entry->a = diff --git a/drivers/infiniband/hw/hfi1/trace_misc.h b/drivers/infiniband/hw/hfi1/trace_misc.h index d308454af7fd..deac77ddaeab 100644 --- a/drivers/infiniband/hw/hfi1/trace_misc.h +++ b/drivers/infiniband/hw/hfi1/trace_misc.h @@ -72,6 +72,54 @@ TRACE_EVENT(hfi1_interrupt, __entry->src) ); +#ifdef CONFIG_FAULT_INJECTION +TRACE_EVENT(hfi1_fault_opcode, + TP_PROTO(struct rvt_qp *qp, u8 opcode), + TP_ARGS(qp, opcode), + TP_STRUCT__entry(DD_DEV_ENTRY(dd_from_ibdev(qp->ibqp.device)) + __field(u32, qpn) + __field(u8, opcode) + ), + TP_fast_assign(DD_DEV_ASSIGN(dd_from_ibdev(qp->ibqp.device)) + __entry->qpn = qp->ibqp.qp_num; + __entry->opcode = opcode; + ), + TP_printk("[%s] qpn 0x%x opcode 0x%x", + __get_str(dev), __entry->qpn, __entry->opcode) +); + +TRACE_EVENT(hfi1_fault_packet, + TP_PROTO(struct hfi1_packet *packet), + TP_ARGS(packet), + TP_STRUCT__entry(DD_DEV_ENTRY(packet->rcd->ppd->dd) + __field(u64, eflags) + __field(u32, ctxt) + __field(u32, hlen) + __field(u32, tlen) + __field(u32, updegr) + __field(u32, etail) + ), + TP_fast_assign(DD_DEV_ASSIGN(packet->rcd->ppd->dd); + __entry->eflags = rhf_err_flags(packet->rhf); + __entry->ctxt = packet->rcd->ctxt; + __entry->hlen = packet->hlen; + __entry->tlen = packet->tlen; + __entry->updegr = packet->updegr; + __entry->etail = rhf_egr_index(packet->rhf); + ), + TP_printk( + "[%s] ctxt %d eflags 0x%llx hlen %d tlen %d updegr %d etail %d", + __get_str(dev), + __entry->ctxt, + __entry->eflags, + __entry->hlen, + __entry->tlen, + __entry->updegr, + __entry->etail + ) +); +#endif + #endif /* __HFI1_TRACE_MISC_H */ #undef TRACE_INCLUDE_PATH diff --git a/drivers/infiniband/hw/hfi1/trace_rc.h b/drivers/infiniband/hw/hfi1/trace_rc.h index 5ea5005f9f41..8ce476570462 100644 --- a/drivers/infiniband/hw/hfi1/trace_rc.h +++ b/drivers/infiniband/hw/hfi1/trace_rc.h @@ -1,5 +1,5 @@ /* -* Copyright(c) 2015, 2016 Intel Corporation. +* Copyright(c) 2015, 2016, 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -104,11 +104,6 @@ DEFINE_EVENT(hfi1_rc_template, hfi1_ack, TP_ARGS(qp, psn) ); -DEFINE_EVENT(hfi1_rc_template, hfi1_timeout, - TP_PROTO(struct rvt_qp *qp, u32 psn), - TP_ARGS(qp, psn) -); - DEFINE_EVENT(hfi1_rc_template, hfi1_rcv_error, TP_PROTO(struct rvt_qp *qp, u32 psn), TP_ARGS(qp, psn) diff --git a/drivers/infiniband/hw/hfi1/trace_tx.h b/drivers/infiniband/hw/hfi1/trace_tx.h index 415d6be42c5d..2c9ac57657d3 100644 --- a/drivers/infiniband/hw/hfi1/trace_tx.h +++ b/drivers/infiniband/hw/hfi1/trace_tx.h @@ -633,6 +633,49 @@ DEFINE_EVENT(hfi1_bct_template, bct_get, TP_PROTO(struct hfi1_devdata *dd, struct buffer_control *bc), TP_ARGS(dd, bc)); +TRACE_EVENT( + hfi1_qp_send_completion, + TP_PROTO(struct rvt_qp *qp, struct rvt_swqe *wqe, u32 idx), + TP_ARGS(qp, wqe, idx), + TP_STRUCT__entry( + DD_DEV_ENTRY(dd_from_ibdev(qp->ibqp.device)) + __field(struct rvt_swqe *, wqe) + __field(u64, wr_id) + __field(u32, qpn) + __field(u32, qpt) + __field(u32, length) + __field(u32, idx) + __field(u32, ssn) + __field(enum ib_wr_opcode, opcode) + __field(int, send_flags) + ), + TP_fast_assign( + DD_DEV_ASSIGN(dd_from_ibdev(qp->ibqp.device)) + __entry->wqe = wqe; + __entry->wr_id = wqe->wr.wr_id; + __entry->qpn = qp->ibqp.qp_num; + __entry->qpt = qp->ibqp.qp_type; + __entry->length = wqe->length; + __entry->idx = idx; + __entry->ssn = wqe->ssn; + __entry->opcode = wqe->wr.opcode; + __entry->send_flags = wqe->wr.send_flags; + ), + TP_printk( + "[%s] qpn 0x%x qpt %u wqe %p idx %u wr_id %llx length %u ssn %u opcode %x send_flags %x", + __get_str(dev), + __entry->qpn, + __entry->qpt, + __entry->wqe, + __entry->idx, + __entry->wr_id, + __entry->length, + __entry->ssn, + __entry->opcode, + __entry->send_flags + ) +); + #endif /* __HFI1_TRACE_TX_H */ #undef TRACE_INCLUDE_PATH diff --git a/drivers/infiniband/hw/hfi1/uc.c b/drivers/infiniband/hw/hfi1/uc.c index 4b2a8400c823..5da1e4546543 100644 --- a/drivers/infiniband/hw/hfi1/uc.c +++ b/drivers/infiniband/hw/hfi1/uc.c @@ -94,7 +94,7 @@ int hfi1_make_uc_req(struct rvt_qp *qp, struct hfi1_pkt_state *ps) } ohdr = &ps->s_txreq->phdr.hdr.u.oth; - if (qp->remote_ah_attr.ah_flags & IB_AH_GRH) + if (rdma_ah_get_ah_flags(&qp->remote_ah_attr) & IB_AH_GRH) ohdr = &ps->s_txreq->phdr.hdr.u.l.oth; /* Get the next send request. */ @@ -320,7 +320,7 @@ void hfi1_uc_rcv(struct hfi1_packet *packet) process_ecn(qp, packet, true); psn = be32_to_cpu(ohdr->bth[2]); - opcode = (bth0 >> 24) & 0xff; + opcode = ib_bth_get_opcode(ohdr); /* Compare the PSN verses the expected PSN. */ if (unlikely(cmp_psn(psn, qp->r_psn) != 0)) { @@ -433,7 +433,7 @@ no_immediate_data: wc.wc_flags = 0; send_last: /* Get the number of bytes the message was padded by. */ - pad = (be32_to_cpu(ohdr->bth[0]) >> 20) & 3; + pad = ib_bth_get_pad(ohdr); /* Check for invalid length. */ /* LAST len should be >= 1 */ if (unlikely(tlen < (hdrsize + pad + 4))) @@ -451,7 +451,7 @@ last_imm: wc.status = IB_WC_SUCCESS; wc.qp = &qp->ibqp; wc.src_qp = qp->remote_qpn; - wc.slid = qp->remote_ah_attr.dlid; + wc.slid = rdma_ah_get_dlid(&qp->remote_ah_attr); /* * It seems that IB mandates the presence of an SL in a * work completion only for the UD transport (see section @@ -463,7 +463,7 @@ last_imm: * * See also OPA Vol. 1, section 9.7.6, and table 9-17. */ - wc.sl = qp->remote_ah_attr.sl; + wc.sl = rdma_ah_get_sl(&qp->remote_ah_attr); /* zero fields that are N/A */ wc.vendor_err = 0; wc.pkey_index = 0; @@ -528,7 +528,7 @@ rdma_last_imm: wc.wc_flags = IB_WC_WITH_IMM; /* Get the number of bytes the message was padded by. */ - pad = (be32_to_cpu(ohdr->bth[0]) >> 20) & 3; + pad = ib_bth_get_pad(ohdr); /* Check for invalid length. */ /* LAST len should be >= 1 */ if (unlikely(tlen < (hdrsize + pad + 4))) @@ -555,7 +555,7 @@ rdma_last_imm: case OP(RDMA_WRITE_LAST): rdma_last: /* Get the number of bytes the message was padded by. */ - pad = (be32_to_cpu(ohdr->bth[0]) >> 20) & 3; + pad = ib_bth_get_pad(ohdr); /* Check for invalid length. */ /* LAST len should be >= 1 */ if (unlikely(tlen < (hdrsize + pad + 4))) diff --git a/drivers/infiniband/hw/hfi1/ud.c b/drivers/infiniband/hw/hfi1/ud.c index 13ea4eb6ef3d..6a4e95cefae5 100644 --- a/drivers/infiniband/hw/hfi1/ud.c +++ b/drivers/infiniband/hw/hfi1/ud.c @@ -68,7 +68,7 @@ static void ud_loopback(struct rvt_qp *sqp, struct rvt_swqe *swqe) struct hfi1_ibport *ibp = to_iport(sqp->ibqp.device, sqp->port_num); struct hfi1_pportdata *ppd; struct rvt_qp *qp; - struct ib_ah_attr *ah_attr; + struct rdma_ah_attr *ah_attr; unsigned long flags; struct rvt_sge_state ssge; struct rvt_sge *sge; @@ -103,17 +103,17 @@ static void ud_loopback(struct rvt_qp *sqp, struct rvt_swqe *swqe) if (qp->ibqp.qp_num > 1) { u16 pkey; u16 slid; - u8 sc5 = ibp->sl_to_sc[ah_attr->sl]; + u8 sc5 = ibp->sl_to_sc[rdma_ah_get_sl(ah_attr)]; pkey = hfi1_get_pkey(ibp, sqp->s_pkey_index); - slid = ppd->lid | (ah_attr->src_path_bits & + slid = ppd->lid | (rdma_ah_get_path_bits(ah_attr) & ((1 << ppd->lmc) - 1)); if (unlikely(ingress_pkey_check(ppd, pkey, sc5, qp->s_pkey_index, slid))) { hfi1_bad_pqkey(ibp, OPA_TRAP_BAD_P_KEY, pkey, - ah_attr->sl, + rdma_ah_get_sl(ah_attr), sqp->ibqp.qp_num, qp->ibqp.qp_num, - slid, ah_attr->dlid); + slid, rdma_ah_get_dlid(ah_attr)); goto drop; } } @@ -131,13 +131,13 @@ static void ud_loopback(struct rvt_qp *sqp, struct rvt_swqe *swqe) if (unlikely(qkey != qp->qkey)) { u16 lid; - lid = ppd->lid | (ah_attr->src_path_bits & + lid = ppd->lid | (rdma_ah_get_path_bits(ah_attr) & ((1 << ppd->lmc) - 1)); hfi1_bad_pqkey(ibp, OPA_TRAP_BAD_Q_KEY, qkey, - ah_attr->sl, + rdma_ah_get_sl(ah_attr), sqp->ibqp.qp_num, qp->ibqp.qp_num, lid, - ah_attr->dlid); + rdma_ah_get_dlid(ah_attr)); goto drop; } } @@ -183,11 +183,11 @@ static void ud_loopback(struct rvt_qp *sqp, struct rvt_swqe *swqe) goto bail_unlock; } - if (ah_attr->ah_flags & IB_AH_GRH) { + if (rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH) { struct ib_grh grh; - struct ib_global_route grd = ah_attr->grh; + const struct ib_global_route *grd = rdma_ah_read_grh(ah_attr); - hfi1_make_grh(ibp, &grh, &grd, 0, 0); + hfi1_make_grh(ibp, &grh, grd, 0, 0); hfi1_copy_sge(&qp->r_sge, &grh, sizeof(grh), true, false); wc.wc_flags |= IB_WC_GRH; @@ -243,12 +243,13 @@ static void ud_loopback(struct rvt_qp *sqp, struct rvt_swqe *swqe) } else { wc.pkey_index = 0; } - wc.slid = ppd->lid | (ah_attr->src_path_bits & ((1 << ppd->lmc) - 1)); + wc.slid = ppd->lid | (rdma_ah_get_path_bits(ah_attr) & + ((1 << ppd->lmc) - 1)); /* Check for loopback when the port lid is not set */ if (wc.slid == 0 && sqp->ibqp.qp_type == IB_QPT_GSI) wc.slid = be16_to_cpu(IB_LID_PERMISSIVE); - wc.sl = ah_attr->sl; - wc.dlid_path_bits = ah_attr->dlid & ((1 << ppd->lmc) - 1); + wc.sl = rdma_ah_get_sl(ah_attr); + wc.dlid_path_bits = rdma_ah_get_dlid(ah_attr) & ((1 << ppd->lmc) - 1); wc.port_num = qp->port_num; /* Signal completion event if the solicited bit is set. */ rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc, @@ -272,7 +273,7 @@ int hfi1_make_ud_req(struct rvt_qp *qp, struct hfi1_pkt_state *ps) { struct hfi1_qp_priv *priv = qp->priv; struct ib_other_headers *ohdr; - struct ib_ah_attr *ah_attr; + struct rdma_ah_attr *ah_attr; struct hfi1_pportdata *ppd; struct hfi1_ibport *ibp; struct rvt_swqe *wqe; @@ -319,9 +320,9 @@ int hfi1_make_ud_req(struct rvt_qp *qp, struct hfi1_pkt_state *ps) ibp = to_iport(qp->ibqp.device, qp->port_num); ppd = ppd_from_ibp(ibp); ah_attr = &ibah_to_rvtah(wqe->ud_wr.ah)->attr; - if (ah_attr->dlid < be16_to_cpu(IB_MULTICAST_LID_BASE) || - ah_attr->dlid == be16_to_cpu(IB_LID_PERMISSIVE)) { - lid = ah_attr->dlid & ~((1 << ppd->lmc) - 1); + if (rdma_ah_get_dlid(ah_attr) < be16_to_cpu(IB_MULTICAST_LID_BASE) || + rdma_ah_get_dlid(ah_attr) == be16_to_cpu(IB_LID_PERMISSIVE)) { + lid = rdma_ah_get_dlid(ah_attr) & ~((1 << ppd->lmc) - 1); if (unlikely(!loopback && (lid == ppd->lid || (lid == be16_to_cpu(IB_LID_PERMISSIVE) && @@ -356,7 +357,7 @@ int hfi1_make_ud_req(struct rvt_qp *qp, struct hfi1_pkt_state *ps) qp->s_hdrwords = 7; ps->s_txreq->s_cur_size = wqe->length; ps->s_txreq->ss = &qp->s_sge; - qp->s_srate = ah_attr->static_rate; + qp->s_srate = rdma_ah_get_static_rate(ah_attr); qp->srate_mbps = ib_rate_to_mbps(qp->s_srate); qp->s_wqe = wqe; qp->s_sge.sge = wqe->sg_list[0]; @@ -364,11 +365,11 @@ int hfi1_make_ud_req(struct rvt_qp *qp, struct hfi1_pkt_state *ps) qp->s_sge.num_sge = wqe->wr.num_sge; qp->s_sge.total_len = wqe->length; - if (ah_attr->ah_flags & IB_AH_GRH) { + if (rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH) { /* Header size in 32-bit words. */ qp->s_hdrwords += hfi1_make_grh(ibp, &ps->s_txreq->phdr.hdr.u.l.grh, - &ah_attr->grh, + rdma_ah_read_grh(ah_attr), qp->s_hdrwords, nwords); lrh0 = HFI1_LRH_GRH; ohdr = &ps->s_txreq->phdr.hdr.u.l.oth; @@ -388,8 +389,8 @@ int hfi1_make_ud_req(struct rvt_qp *qp, struct hfi1_pkt_state *ps) } else { bth0 = IB_OPCODE_UD_SEND_ONLY << 24; } - sc5 = ibp->sl_to_sc[ah_attr->sl]; - lrh0 |= (ah_attr->sl & 0xf) << 4; + sc5 = ibp->sl_to_sc[rdma_ah_get_sl(ah_attr)]; + lrh0 |= (rdma_ah_get_sl(ah_attr) & 0xf) << 4; if (qp->ibqp.qp_type == IB_QPT_SMI) { lrh0 |= 0xF000; /* Set VL (see ch. 13.5.3.1) */ priv->s_sc = 0xf; @@ -402,15 +403,17 @@ int hfi1_make_ud_req(struct rvt_qp *qp, struct hfi1_pkt_state *ps) priv->s_sendcontext = qp_to_send_context(qp, priv->s_sc); ps->s_txreq->psc = priv->s_sendcontext; ps->s_txreq->phdr.hdr.lrh[0] = cpu_to_be16(lrh0); - ps->s_txreq->phdr.hdr.lrh[1] = cpu_to_be16(ah_attr->dlid); + ps->s_txreq->phdr.hdr.lrh[1] = + cpu_to_be16(rdma_ah_get_dlid(ah_attr)); ps->s_txreq->phdr.hdr.lrh[2] = cpu_to_be16(qp->s_hdrwords + nwords + SIZE_OF_CRC); - if (ah_attr->dlid == be16_to_cpu(IB_LID_PERMISSIVE)) { + if (rdma_ah_get_dlid(ah_attr) == be16_to_cpu(IB_LID_PERMISSIVE)) { ps->s_txreq->phdr.hdr.lrh[3] = IB_LID_PERMISSIVE; } else { lid = ppd->lid; if (lid) { - lid |= ah_attr->src_path_bits & ((1 << ppd->lmc) - 1); + lid |= rdma_ah_get_path_bits(ah_attr) & + ((1 << ppd->lmc) - 1); ps->s_txreq->phdr.hdr.lrh[3] = cpu_to_be16(lid); } else { ps->s_txreq->phdr.hdr.lrh[3] = IB_LID_PERMISSIVE; @@ -537,7 +540,7 @@ void return_cnp(struct hfi1_ibport *ibp, struct rvt_qp *qp, u32 remote_qpn, bth0 = pkey | (IB_OPCODE_CNP << 24); ohdr->bth[0] = cpu_to_be32(bth0); - ohdr->bth[1] = cpu_to_be32(remote_qpn | (1 << HFI1_BECN_SHIFT)); + ohdr->bth[1] = cpu_to_be32(remote_qpn | (1 << IB_BECN_SHIFT)); ohdr->bth[2] = 0; /* PSN 0 */ hdr.lrh[0] = cpu_to_be16(lrh0); @@ -680,7 +683,7 @@ void hfi1_ud_rcv(struct hfi1_packet *packet) u32 tlen = packet->tlen; struct rvt_qp *qp = packet->qp; bool has_grh = rcv_flags & HFI1_HAS_GRH; - u8 sc5 = hdr2sc(hdr, packet->rhf); + u8 sc5 = hfi1_9B_get_sc5(hdr, packet->rhf); u32 bth1; u8 sl_from_sc, sl; u16 slid; @@ -688,18 +691,16 @@ void hfi1_ud_rcv(struct hfi1_packet *packet) qkey = be32_to_cpu(ohdr->u.ud.deth[0]); src_qp = be32_to_cpu(ohdr->u.ud.deth[1]) & RVT_QPN_MASK; - dlid = be16_to_cpu(hdr->lrh[1]); + dlid = ib_get_dlid(hdr); bth1 = be32_to_cpu(ohdr->bth[1]); - slid = be16_to_cpu(hdr->lrh[3]); - pkey = (u16)be32_to_cpu(ohdr->bth[0]); - sl = (be16_to_cpu(hdr->lrh[0]) >> 4) & 0xf; - extra_bytes = (be32_to_cpu(ohdr->bth[0]) >> 20) & 3; + slid = ib_get_slid(hdr); + pkey = ib_bth_get_pkey(ohdr); + opcode = ib_bth_get_opcode(ohdr); + sl = ib_get_sl(hdr); + extra_bytes = ib_bth_get_pad(ohdr); extra_bytes += (SIZE_OF_CRC << 2); sl_from_sc = ibp->sc_to_sl[sc5]; - opcode = be32_to_cpu(ohdr->bth[0]) >> 24; - opcode &= 0xff; - process_ecn(qp, packet, (opcode != IB_OPCODE_CNP)); /* * Get the number of bytes the message was padded by diff --git a/drivers/infiniband/hw/hfi1/user_exp_rcv.c b/drivers/infiniband/hw/hfi1/user_exp_rcv.c index 4a8295399e71..35c6e7ec8ad6 100644 --- a/drivers/infiniband/hw/hfi1/user_exp_rcv.c +++ b/drivers/infiniband/hw/hfi1/user_exp_rcv.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -200,8 +200,9 @@ int hfi1_user_exp_rcv_init(struct file *fp) if (!HFI1_CAP_UGET_MASK(uctxt->flags, TID_UNMAP)) { fd->invalid_tid_idx = 0; - fd->invalid_tids = kzalloc(uctxt->expected_count * - sizeof(u32), GFP_KERNEL); + fd->invalid_tids = kcalloc(uctxt->expected_count, + sizeof(*fd->invalid_tids), + GFP_KERNEL); if (!fd->invalid_tids) { ret = -ENOMEM; goto done; @@ -578,6 +579,9 @@ int hfi1_user_exp_rcv_clear(struct file *fp, struct hfi1_tid_info *tinfo) u32 *tidinfo; unsigned tididx; + if (unlikely(tinfo->tidcnt > fd->tid_used)) + return -EINVAL; + tidinfo = memdup_user((void __user *)(unsigned long)tinfo->tidlist, sizeof(tidinfo[0]) * tinfo->tidcnt); if (IS_ERR(tidinfo)) @@ -607,7 +611,7 @@ int hfi1_user_exp_rcv_invalid(struct file *fp, struct hfi1_tid_info *tinfo) struct hfi1_filedata *fd = fp->private_data; struct hfi1_ctxtdata *uctxt = fd->uctxt; unsigned long *ev = uctxt->dd->events + - (((uctxt->ctxt - uctxt->dd->first_user_ctxt) * + (((uctxt->ctxt - uctxt->dd->first_dyn_alloc_ctxt) * HFI1_MAX_SHARED_CTXTS) + fd->subctxt); u32 *array; int ret = 0; @@ -1011,8 +1015,8 @@ static int tid_rb_invalidate(void *arg, struct mmu_rb_node *mnode) * process in question. */ ev = uctxt->dd->events + - (((uctxt->ctxt - uctxt->dd->first_user_ctxt) * - HFI1_MAX_SHARED_CTXTS) + fdata->subctxt); + (((uctxt->ctxt - uctxt->dd->first_dyn_alloc_ctxt) * + HFI1_MAX_SHARED_CTXTS) + fdata->subctxt); set_bit(_HFI1_EVENT_TID_MMU_NOTIFY_BIT, ev); } fdata->invalid_tid_idx++; diff --git a/drivers/infiniband/hw/hfi1/user_pages.c b/drivers/infiniband/hw/hfi1/user_pages.c index 68295a12b771..e341e6dcc388 100644 --- a/drivers/infiniband/hw/hfi1/user_pages.c +++ b/drivers/infiniband/hw/hfi1/user_pages.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -73,7 +73,8 @@ bool hfi1_can_pin_pages(struct hfi1_devdata *dd, struct mm_struct *mm, { unsigned long ulimit = rlimit(RLIMIT_MEMLOCK), pinned, cache_limit, size = (cache_size * (1UL << 20)); /* convert to bytes */ - unsigned usr_ctxts = dd->num_rcv_contexts - dd->first_user_ctxt; + unsigned int usr_ctxts = + dd->num_rcv_contexts - dd->first_dyn_alloc_ctxt; bool can_lock = capable(CAP_IPC_LOCK); /* diff --git a/drivers/infiniband/hw/hfi1/user_sdma.c b/drivers/infiniband/hw/hfi1/user_sdma.c index e6811c4edc73..0749689d7643 100644 --- a/drivers/infiniband/hw/hfi1/user_sdma.c +++ b/drivers/infiniband/hw/hfi1/user_sdma.c @@ -376,7 +376,6 @@ int hfi1_user_sdma_alloc_queues(struct hfi1_ctxtdata *uctxt, struct file *fp) { struct hfi1_filedata *fd; int ret = 0; - unsigned memsize; char buf[64]; struct hfi1_devdata *dd; struct hfi1_user_sdma_comp_q *cq; @@ -401,13 +400,15 @@ int hfi1_user_sdma_alloc_queues(struct hfi1_ctxtdata *uctxt, struct file *fp) if (!pq) goto pq_nomem; - memsize = sizeof(*pq->reqs) * hfi1_sdma_comp_ring_size; - pq->reqs = kzalloc(memsize, GFP_KERNEL); + pq->reqs = kcalloc(hfi1_sdma_comp_ring_size, + sizeof(*pq->reqs), + GFP_KERNEL); if (!pq->reqs) goto pq_reqs_nomem; - memsize = BITS_TO_LONGS(hfi1_sdma_comp_ring_size) * sizeof(long); - pq->req_in_use = kzalloc(memsize, GFP_KERNEL); + pq->req_in_use = kcalloc(BITS_TO_LONGS(hfi1_sdma_comp_ring_size), + sizeof(*pq->req_in_use), + GFP_KERNEL); if (!pq->req_in_use) goto pq_reqs_no_in_use; @@ -442,8 +443,8 @@ int hfi1_user_sdma_alloc_queues(struct hfi1_ctxtdata *uctxt, struct file *fp) if (!cq) goto cq_nomem; - memsize = PAGE_ALIGN(sizeof(*cq->comps) * hfi1_sdma_comp_ring_size); - cq->comps = vmalloc_user(memsize); + cq->comps = vmalloc_user(PAGE_ALIGN(sizeof(*cq->comps) + * hfi1_sdma_comp_ring_size)); if (!cq->comps) goto cq_comps_nomem; @@ -704,7 +705,9 @@ int hfi1_user_sdma_process_request(struct file *fp, struct iovec *iovec, /* Save all the IO vector structures */ for (i = 0; i < req->data_iovs; i++) { INIT_LIST_HEAD(&req->iovs[i].list); - memcpy(&req->iovs[i].iov, iovec + idx++, sizeof(struct iovec)); + memcpy(&req->iovs[i].iov, + iovec + idx++, + sizeof(req->iovs[i].iov)); ret = pin_vector_pages(req, &req->iovs[i]); if (ret) { req->status = ret; @@ -1615,9 +1618,10 @@ static inline void set_comp_state(struct hfi1_user_sdma_pkt_q *pq, { hfi1_cdbg(SDMA, "[%u:%u:%u:%u] Setting completion status %u %d", pq->dd->unit, pq->ctxt, pq->subctxt, idx, state, ret); - cq->comps[idx].status = state; if (state == ERROR) cq->comps[idx].errcode = -ret; + smp_wmb(); /* make sure errcode is visible first */ + cq->comps[idx].status = state; trace_hfi1_sdma_user_completion(pq->dd, pq->ctxt, pq->subctxt, idx, state, ret); } diff --git a/drivers/infiniband/hw/hfi1/verbs.c b/drivers/infiniband/hw/hfi1/verbs.c index 222315fadab1..90e7b77d68e8 100644 --- a/drivers/infiniband/hw/hfi1/verbs.c +++ b/drivers/infiniband/hw/hfi1/verbs.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015 - 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -60,6 +60,8 @@ #include "trace.h" #include "qp.h" #include "verbs_txreq.h" +#include "debugfs.h" +#include "vnic.h" static unsigned int hfi1_lkey_table_size = 16; module_param_named(lkey_table_size, hfi1_lkey_table_size, uint, @@ -297,6 +299,22 @@ static inline bool wss_exceeds_threshold(void) } /* + * Translate ib_wr_opcode into ib_wc_opcode. + */ +const enum ib_wc_opcode ib_hfi1_wc_opcode[] = { + [IB_WR_RDMA_WRITE] = IB_WC_RDMA_WRITE, + [IB_WR_RDMA_WRITE_WITH_IMM] = IB_WC_RDMA_WRITE, + [IB_WR_SEND] = IB_WC_SEND, + [IB_WR_SEND_WITH_IMM] = IB_WC_SEND, + [IB_WR_RDMA_READ] = IB_WC_RDMA_READ, + [IB_WR_ATOMIC_CMP_AND_SWP] = IB_WC_COMP_SWAP, + [IB_WR_ATOMIC_FETCH_AND_ADD] = IB_WC_FETCH_ADD, + [IB_WR_SEND_WITH_INV] = IB_WC_SEND, + [IB_WR_LOCAL_INV] = IB_WC_LOCAL_INV, + [IB_WR_REG_MR] = IB_WC_REG_MR +}; + +/* * Length of header by opcode, 0 --> not supported */ const u8 hdr_len_by_opcode[256] = { @@ -501,6 +519,35 @@ static inline opcode_handler qp_ok(int opcode, struct hfi1_packet *packet) return NULL; } +static u64 hfi1_fault_tx(struct rvt_qp *qp, u8 opcode, u64 pbc) +{ +#ifdef CONFIG_FAULT_INJECTION + if ((opcode & IB_OPCODE_MSP) == IB_OPCODE_MSP) + /* + * In order to drop non-IB traffic we + * set PbcInsertHrc to NONE (0x2). + * The packet will still be delivered + * to the receiving node but a + * KHdrHCRCErr (KDETH packet with a bad + * HCRC) will be triggered and the + * packet will not be delivered to the + * correct context. + */ + pbc |= (u64)PBC_IHCRC_NONE << PBC_INSERT_HCRC_SHIFT; + else + /* + * In order to drop regular verbs + * traffic we set the PbcTestEbp + * flag. The packet will still be + * delivered to the receiving node but + * a 'late ebp error' will be + * triggered and will be dropped. + */ + pbc |= PBC_TEST_EBP; +#endif + return pbc; +} + /** * hfi1_ib_rcv - process an incoming packet * @packet: data packet information @@ -525,7 +572,7 @@ void hfi1_ib_rcv(struct hfi1_packet *packet) u16 lid; /* Check for GRH */ - lnh = be16_to_cpu(hdr->lrh[0]) & 3; + lnh = ib_get_lnh(hdr); if (lnh == HFI1_LRH_BTH) { packet->ohdr = &hdr->u.oth; } else if (lnh == HFI1_LRH_GRH) { @@ -544,12 +591,12 @@ void hfi1_ib_rcv(struct hfi1_packet *packet) trace_input_ibhdr(rcd->dd, hdr); - opcode = (be32_to_cpu(packet->ohdr->bth[0]) >> 24); + opcode = ib_bth_get_opcode(packet->ohdr); inc_opstats(tlen, &rcd->opstats->stats[opcode]); /* Get the destination QP number. */ qp_num = be32_to_cpu(packet->ohdr->bth[1]) & RVT_QPN_MASK; - lid = be16_to_cpu(hdr->lrh[1]); + lid = ib_get_dlid(hdr); if (unlikely((lid >= be16_to_cpu(IB_MULTICAST_LID_BASE)) && (lid != be16_to_cpu(IB_LID_PERMISSIVE)))) { struct rvt_mcast *mcast; @@ -557,7 +604,7 @@ void hfi1_ib_rcv(struct hfi1_packet *packet) if (lnh != HFI1_LRH_GRH) goto drop; - mcast = rvt_mcast_find(&ibp->rvp, &hdr->u.l.grh.dgid); + mcast = rvt_mcast_find(&ibp->rvp, &hdr->u.l.grh.dgid, lid); if (!mcast) goto drop; list_for_each_entry_rcu(p, &mcast->qp_list, list) { @@ -583,6 +630,11 @@ void hfi1_ib_rcv(struct hfi1_packet *packet) rcu_read_unlock(); goto drop; } + if (unlikely(hfi1_dbg_fault_opcode(packet->qp, opcode, + true))) { + rcu_read_unlock(); + goto drop; + } spin_lock_irqsave(&packet->qp->r_lock, flags); packet_handler = qp_ok(opcode, packet); if (likely(packet_handler)) @@ -781,7 +833,6 @@ static int build_verbs_tx_desc( if (ret) goto bail_txadd; } - /* add the ulp payload - if any. tx->ss can be NULL for acks */ if (tx->ss) ret = build_verbs_ulp_payload(sde, length, tx); @@ -800,7 +851,6 @@ int hfi1_verbs_send_dma(struct rvt_qp *qp, struct hfi1_pkt_state *ps, struct hfi1_ibdev *dev = ps->dev; struct hfi1_pportdata *ppd = ps->ppd; struct verbs_txreq *tx; - u64 pbc_flags = 0; u8 sc5 = priv->s_sc; int ret; @@ -809,12 +859,16 @@ int hfi1_verbs_send_dma(struct rvt_qp *qp, struct hfi1_pkt_state *ps, if (!sdma_txreq_built(&tx->txreq)) { if (likely(pbc == 0)) { u32 vl = sc_to_vlt(dd_from_ibdev(qp->ibqp.device), sc5); + u8 opcode = get_opcode(&tx->phdr.hdr); + /* No vl15 here */ /* set PBC_DC_INFO bit (aka SC[4]) in pbc_flags */ - pbc_flags |= (!!(sc5 & 0x10)) << PBC_DC_INFO_SHIFT; + pbc |= (!!(sc5 & 0x10)) << PBC_DC_INFO_SHIFT; + if (unlikely(hfi1_dbg_fault_opcode(qp, opcode, false))) + pbc = hfi1_fault_tx(qp, opcode, pbc); pbc = create_pbc(ppd, - pbc_flags, + pbc, qp->srate_mbps, vl, plen); @@ -917,7 +971,6 @@ int hfi1_verbs_send_pio(struct rvt_qp *qp, struct hfi1_pkt_state *ps, u32 plen = hdrwords + dwords + 2; /* includes pbc */ struct hfi1_pportdata *ppd = ps->ppd; u32 *hdr = (u32 *)&ps->s_txreq->phdr.hdr; - u64 pbc_flags = 0; u8 sc5; unsigned long flags = 0; struct send_context *sc; @@ -942,9 +995,14 @@ int hfi1_verbs_send_pio(struct rvt_qp *qp, struct hfi1_pkt_state *ps, if (likely(pbc == 0)) { u8 vl = sc_to_vlt(dd_from_ibdev(qp->ibqp.device), sc5); + struct verbs_txreq *tx = ps->s_txreq; + u8 opcode = get_opcode(&tx->phdr.hdr); + /* set PBC_DC_INFO bit (aka SC[4]) in pbc_flags */ - pbc_flags |= (!!(sc5 & 0x10)) << PBC_DC_INFO_SHIFT; - pbc = create_pbc(ppd, pbc_flags, qp->srate_mbps, vl, plen); + pbc |= (!!(sc5 & 0x10)) << PBC_DC_INFO_SHIFT; + if (unlikely(hfi1_dbg_fault_opcode(qp, opcode, false))) + pbc = hfi1_fault_tx(qp, opcode, pbc); + pbc = create_pbc(ppd, pbc, qp->srate_mbps, vl, plen); } if (cb) iowait_pio_inc(&priv->s_iowait); @@ -1173,7 +1231,7 @@ int hfi1_verbs_send(struct rvt_qp *qp, struct hfi1_pkt_state *ps) hdr = &ps->s_txreq->phdr.hdr; /* locate the pkey within the headers */ - lnh = be16_to_cpu(hdr->lrh[0]) & 3; + lnh = ib_get_lnh(hdr); if (lnh == HFI1_LRH_GRH) ohdr = &hdr->u.l.oth; else @@ -1220,17 +1278,20 @@ int hfi1_verbs_send(struct rvt_qp *qp, struct hfi1_pkt_state *ps) static void hfi1_fill_device_attr(struct hfi1_devdata *dd) { struct rvt_dev_info *rdi = &dd->verbs_dev.rdi; - u16 ver = dd->dc8051_ver; + u32 ver = dd->dc8051_ver; memset(&rdi->dparms.props, 0, sizeof(rdi->dparms.props)); - rdi->dparms.props.fw_ver = ((u64)(dc8051_ver_maj(ver)) << 16) | - (u64)dc8051_ver_min(ver); + rdi->dparms.props.fw_ver = ((u64)(dc8051_ver_maj(ver)) << 32) | + ((u64)(dc8051_ver_min(ver)) << 16) | + (u64)dc8051_ver_patch(ver); + rdi->dparms.props.device_cap_flags = IB_DEVICE_BAD_PKEY_CNTR | IB_DEVICE_BAD_QKEY_CNTR | IB_DEVICE_SHUTDOWN_PORT | IB_DEVICE_SYS_IMAGE_GUID | IB_DEVICE_RC_RNR_NAK_GEN | IB_DEVICE_PORT_ACTIVE_EVENT | IB_DEVICE_SRQ_RESIZE | - IB_DEVICE_MEM_MGT_EXTENSIONS; + IB_DEVICE_MEM_MGT_EXTENSIONS | + IB_DEVICE_RDMA_NETDEV_OPA_VNIC; rdi->dparms.props.page_size_cap = PAGE_SIZE; rdi->dparms.props.vendor_id = dd->oui1 << 16 | dd->oui2 << 8 | dd->oui3; rdi->dparms.props.vendor_part_id = dd->pcidev->device; @@ -1398,14 +1459,14 @@ static int hfi1_get_guid_be(struct rvt_dev_info *rdi, struct rvt_ibport *rvp, /* * convert ah port,sl to sc */ -u8 ah_to_sc(struct ib_device *ibdev, struct ib_ah_attr *ah) +u8 ah_to_sc(struct ib_device *ibdev, struct rdma_ah_attr *ah) { - struct hfi1_ibport *ibp = to_iport(ibdev, ah->port_num); + struct hfi1_ibport *ibp = to_iport(ibdev, rdma_ah_get_port_num(ah)); - return ibp->sl_to_sc[ah->sl]; + return ibp->sl_to_sc[rdma_ah_get_sl(ah)]; } -static int hfi1_check_ah(struct ib_device *ibdev, struct ib_ah_attr *ah_attr) +static int hfi1_check_ah(struct ib_device *ibdev, struct rdma_ah_attr *ah_attr) { struct hfi1_ibport *ibp; struct hfi1_pportdata *ppd; @@ -1413,9 +1474,9 @@ static int hfi1_check_ah(struct ib_device *ibdev, struct ib_ah_attr *ah_attr) u8 sc5; /* test the mapping for validity */ - ibp = to_iport(ibdev, ah_attr->port_num); + ibp = to_iport(ibdev, rdma_ah_get_port_num(ah_attr)); ppd = ppd_from_ibp(ibp); - sc5 = ibp->sl_to_sc[ah_attr->sl]; + sc5 = ibp->sl_to_sc[rdma_ah_get_sl(ah_attr)]; dd = dd_from_ppd(ppd); if (sc_to_vlt(dd, sc5) > num_vls && sc_to_vlt(dd, sc5) != 0xf) return -EINVAL; @@ -1423,7 +1484,7 @@ static int hfi1_check_ah(struct ib_device *ibdev, struct ib_ah_attr *ah_attr) } static void hfi1_notify_new_ah(struct ib_device *ibdev, - struct ib_ah_attr *ah_attr, + struct rdma_ah_attr *ah_attr, struct rvt_ah *ah) { struct hfi1_ibport *ibp; @@ -1436,9 +1497,9 @@ static void hfi1_notify_new_ah(struct ib_device *ibdev, * done being setup. We can however modify things which we need to set. */ - ibp = to_iport(ibdev, ah_attr->port_num); + ibp = to_iport(ibdev, rdma_ah_get_port_num(ah_attr)); ppd = ppd_from_ibp(ibp); - sc5 = ibp->sl_to_sc[ah->attr.sl]; + sc5 = ibp->sl_to_sc[rdma_ah_get_sl(&ah->attr)]; dd = dd_from_ppd(ppd); ah->vl = sc_to_vlt(dd, sc5); if (ah->vl < num_vls || ah->vl == 15) @@ -1447,17 +1508,21 @@ static void hfi1_notify_new_ah(struct ib_device *ibdev, struct ib_ah *hfi1_create_qp0_ah(struct hfi1_ibport *ibp, u16 dlid) { - struct ib_ah_attr attr; + struct rdma_ah_attr attr; struct ib_ah *ah = ERR_PTR(-EINVAL); struct rvt_qp *qp0; + struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); + struct hfi1_devdata *dd = dd_from_ppd(ppd); + u8 port_num = ppd->port; memset(&attr, 0, sizeof(attr)); - attr.dlid = dlid; - attr.port_num = ppd_from_ibp(ibp)->port; + attr.type = rdma_ah_find_type(&dd->verbs_dev.rdi.ibdev, port_num); + rdma_ah_set_dlid(&attr, dlid); + rdma_ah_set_port_num(&attr, ppd_from_ibp(ibp)->port); rcu_read_lock(); qp0 = rcu_dereference(ibp->rvp.qp[0]); if (qp0) - ah = ib_create_ah(qp0->ibqp.pd, &attr); + ah = rdma_create_ah(qp0->ibqp.pd, &attr); rcu_read_unlock(); return ah; } @@ -1504,10 +1569,10 @@ static void hfi1_get_dev_fw_str(struct ib_device *ibdev, char *str, { struct rvt_dev_info *rdi = ib_to_rvt(ibdev); struct hfi1_ibdev *dev = dev_from_rdi(rdi); - u16 ver = dd_from_dev(dev)->dc8051_ver; + u32 ver = dd_from_dev(dev)->dc8051_ver; - snprintf(str, str_len, "%u.%u", dc8051_ver_maj(ver), - dc8051_ver_min(ver)); + snprintf(str, str_len, "%u.%u.%u", dc8051_ver_maj(ver), + dc8051_ver_min(ver), dc8051_ver_patch(ver)); } static const char * const driver_cntr_names[] = { @@ -1524,6 +1589,7 @@ static const char * const driver_cntr_names[] = { "DRIVER_EgrHdrFull" }; +static DEFINE_MUTEX(cntr_names_lock); /* protects the *_cntr_names bufers */ static const char **dev_cntr_names; static const char **port_cntr_names; static int num_driver_cntrs = ARRAY_SIZE(driver_cntr_names); @@ -1578,6 +1644,7 @@ static struct rdma_hw_stats *alloc_hw_stats(struct ib_device *ibdev, { int i, err; + mutex_lock(&cntr_names_lock); if (!cntr_names_initialized) { struct hfi1_devdata *dd = dd_from_ibdev(ibdev); @@ -1586,8 +1653,10 @@ static struct rdma_hw_stats *alloc_hw_stats(struct ib_device *ibdev, num_driver_cntrs, &num_dev_cntrs, &dev_cntr_names); - if (err) + if (err) { + mutex_unlock(&cntr_names_lock); return NULL; + } for (i = 0; i < num_driver_cntrs; i++) dev_cntr_names[num_dev_cntrs + i] = @@ -1601,10 +1670,12 @@ static struct rdma_hw_stats *alloc_hw_stats(struct ib_device *ibdev, if (err) { kfree(dev_cntr_names); dev_cntr_names = NULL; + mutex_unlock(&cntr_names_lock); return NULL; } cntr_names_initialized = 1; } + mutex_unlock(&cntr_names_lock); if (!port_num) return rdma_alloc_hw_stats_struct( @@ -1707,6 +1778,8 @@ int hfi1_register_ib_device(struct hfi1_devdata *dd) ibdev->modify_device = modify_device; ibdev->alloc_hw_stats = alloc_hw_stats; ibdev->get_hw_stats = get_hw_stats; + ibdev->alloc_rdma_netdev = hfi1_vnic_alloc_rn; + ibdev->free_rdma_netdev = hfi1_vnic_free_rn; /* keep process mad in the driver */ ibdev->process_mad = hfi1_process_mad; @@ -1751,7 +1824,7 @@ int hfi1_register_ib_device(struct hfi1_devdata *dd) dd->verbs_dev.rdi.driver_f.qp_priv_free = qp_priv_free; dd->verbs_dev.rdi.driver_f.free_all_qps = free_all_qps; dd->verbs_dev.rdi.driver_f.notify_qp_reset = notify_qp_reset; - dd->verbs_dev.rdi.driver_f.do_send = hfi1_do_send; + dd->verbs_dev.rdi.driver_f.do_send = hfi1_do_send_from_rvt; dd->verbs_dev.rdi.driver_f.schedule_send = hfi1_schedule_send; dd->verbs_dev.rdi.driver_f.schedule_send_no_lock = _hfi1_schedule_send; dd->verbs_dev.rdi.driver_f.get_pmtu_from_attr = get_pmtu_from_attr; @@ -1823,9 +1896,13 @@ void hfi1_unregister_ib_device(struct hfi1_devdata *dd) del_timer_sync(&dev->mem_timer); verbs_txreq_exit(dev); + mutex_lock(&cntr_names_lock); kfree(dev_cntr_names); kfree(port_cntr_names); + dev_cntr_names = NULL; + port_cntr_names = NULL; cntr_names_initialized = 0; + mutex_unlock(&cntr_names_lock); } void hfi1_cnp_rcv(struct hfi1_packet *packet) @@ -1840,12 +1917,12 @@ void hfi1_cnp_rcv(struct hfi1_packet *packet) switch (packet->qp->ibqp.qp_type) { case IB_QPT_UC: - rlid = qp->remote_ah_attr.dlid; + rlid = rdma_ah_get_dlid(&qp->remote_ah_attr); rqpn = qp->remote_qpn; svc_type = IB_CC_SVCTYPE_UC; break; case IB_QPT_RC: - rlid = qp->remote_ah_attr.dlid; + rlid = rdma_ah_get_dlid(&qp->remote_ah_attr); rqpn = qp->remote_qpn; svc_type = IB_CC_SVCTYPE_RC; break; @@ -1859,7 +1936,7 @@ void hfi1_cnp_rcv(struct hfi1_packet *packet) return; } - sc5 = hdr2sc(hdr, packet->rhf); + sc5 = hfi1_9B_get_sc5(hdr, packet->rhf); sl = ibp->sc_to_sl[sc5]; lqpn = qp->ibqp.qp_num; diff --git a/drivers/infiniband/hw/hfi1/verbs.h b/drivers/infiniband/hw/hfi1/verbs.h index 3a0b589e41c2..52ff275caf54 100644 --- a/drivers/infiniband/hw/hfi1/verbs.h +++ b/drivers/infiniband/hw/hfi1/verbs.h @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015 - 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -195,6 +195,11 @@ struct hfi1_ibdev { struct dentry *hfi1_ibdev_dbg; /* per HFI symlinks to above */ struct dentry *hfi1_ibdev_link; +#ifdef CONFIG_FAULT_INJECTION + struct fault_opcode *fault_opcode; + struct fault_packet *fault_packet; + bool fault_suppress_err; +#endif #endif }; @@ -303,7 +308,7 @@ void hfi1_rc_hdrerr( u32 rcv_flags, struct rvt_qp *qp); -u8 ah_to_sc(struct ib_device *ibdev, struct ib_ah_attr *ah_attr); +u8 ah_to_sc(struct ib_device *ibdev, struct rdma_ah_attr *ah_attr); struct ib_ah *hfi1_create_qp0_ah(struct hfi1_ibport *ibp, u16 dlid); @@ -342,7 +347,7 @@ int hfi1_ruc_check_hdr(struct hfi1_ibport *ibp, struct ib_header *hdr, int has_grh, struct rvt_qp *qp, u32 bth0); u32 hfi1_make_grh(struct hfi1_ibport *ibp, struct ib_grh *hdr, - struct ib_global_route *grh, u32 hwords, u32 nwords); + const struct ib_global_route *grh, u32 hwords, u32 nwords); void hfi1_make_ruc_header(struct rvt_qp *qp, struct ib_other_headers *ohdr, u32 bth0, u32 bth2, int middle, @@ -350,7 +355,9 @@ void hfi1_make_ruc_header(struct rvt_qp *qp, struct ib_other_headers *ohdr, void _hfi1_do_send(struct work_struct *work); -void hfi1_do_send(struct rvt_qp *qp); +void hfi1_do_send_from_rvt(struct rvt_qp *qp); + +void hfi1_do_send(struct rvt_qp *qp, bool in_thread); void hfi1_send_complete(struct rvt_qp *qp, struct rvt_swqe *wqe, enum ib_wc_status status); diff --git a/drivers/infiniband/hw/hfi1/vnic.h b/drivers/infiniband/hw/hfi1/vnic.h new file mode 100644 index 000000000000..e2c455299b53 --- /dev/null +++ b/drivers/infiniband/hw/hfi1/vnic.h @@ -0,0 +1,184 @@ +#ifndef _HFI1_VNIC_H +#define _HFI1_VNIC_H +/* + * Copyright(c) 2017 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ + +#include <rdma/opa_vnic.h> +#include "hfi.h" +#include "sdma.h" + +#define HFI1_VNIC_MAX_TXQ 16 +#define HFI1_VNIC_MAX_PAD 12 + +/* L2 header definitions */ +#define HFI1_L2_TYPE_OFFSET 0x7 +#define HFI1_L2_TYPE_SHFT 0x5 +#define HFI1_L2_TYPE_MASK 0x3 + +#define HFI1_GET_L2_TYPE(hdr) \ + ((*((u8 *)(hdr) + HFI1_L2_TYPE_OFFSET) >> HFI1_L2_TYPE_SHFT) & \ + HFI1_L2_TYPE_MASK) + +/* L4 type definitions */ +#define HFI1_L4_TYPE_OFFSET 8 + +#define HFI1_GET_L4_TYPE(data) \ + (*((u8 *)(data) + HFI1_L4_TYPE_OFFSET)) + +/* L4 header definitions */ +#define HFI1_VNIC_L4_HDR_OFFSET OPA_VNIC_L2_HDR_LEN + +#define HFI1_VNIC_GET_L4_HDR(data) \ + (*((u16 *)((u8 *)(data) + HFI1_VNIC_L4_HDR_OFFSET))) + +#define HFI1_VNIC_GET_VESWID(data) \ + (HFI1_VNIC_GET_L4_HDR(data) & 0xFFF) + +/* Service class */ +#define HFI1_VNIC_SC_OFFSET_LOW 6 +#define HFI1_VNIC_SC_OFFSET_HI 7 +#define HFI1_VNIC_SC_SHIFT 4 + +#define HFI1_VNIC_MAX_QUEUE 16 + +/** + * struct hfi1_vnic_sdma - VNIC per Tx ring SDMA information + * @dd - device data pointer + * @sde - sdma engine + * @vinfo - vnic info pointer + * @wait - iowait structure + * @stx - sdma tx request + * @state - vnic Tx ring SDMA state + * @q_idx - vnic Tx queue index + */ +struct hfi1_vnic_sdma { + struct hfi1_devdata *dd; + struct sdma_engine *sde; + struct hfi1_vnic_vport_info *vinfo; + struct iowait wait; + struct sdma_txreq stx; + unsigned int state; + u8 q_idx; +}; + +/** + * struct hfi1_vnic_rx_queue - HFI1 VNIC receive queue + * @idx: queue index + * @vinfo: pointer to vport information + * @netdev: network device + * @napi: netdev napi structure + * @skbq: queue of received socket buffers + */ +struct hfi1_vnic_rx_queue { + u8 idx; + struct hfi1_vnic_vport_info *vinfo; + struct net_device *netdev; + struct napi_struct napi; + struct sk_buff_head skbq; +}; + +/** + * struct hfi1_vnic_vport_info - HFI1 VNIC virtual port information + * @dd: device data pointer + * @netdev: net device pointer + * @flags: state flags + * @lock: vport lock + * @num_tx_q: number of transmit queues + * @num_rx_q: number of receive queues + * @vesw_id: virtual switch id + * @rxq: Array of receive queues + * @stats: per queue stats + * @sdma: VNIC SDMA structure per TXQ + */ +struct hfi1_vnic_vport_info { + struct hfi1_devdata *dd; + struct net_device *netdev; + unsigned long flags; + + /* Lock used around state updates */ + struct mutex lock; + + u8 num_tx_q; + u8 num_rx_q; + u16 vesw_id; + struct hfi1_vnic_rx_queue rxq[HFI1_NUM_VNIC_CTXT]; + + struct opa_vnic_stats stats[HFI1_VNIC_MAX_QUEUE]; + struct hfi1_vnic_sdma sdma[HFI1_VNIC_MAX_TXQ]; +}; + +#define v_dbg(format, arg...) \ + netdev_dbg(vinfo->netdev, format, ## arg) +#define v_err(format, arg...) \ + netdev_err(vinfo->netdev, format, ## arg) +#define v_info(format, arg...) \ + netdev_info(vinfo->netdev, format, ## arg) + +/* vnic hfi1 internal functions */ +void hfi1_vnic_setup(struct hfi1_devdata *dd); +void hfi1_vnic_cleanup(struct hfi1_devdata *dd); +int hfi1_vnic_txreq_init(struct hfi1_devdata *dd); +void hfi1_vnic_txreq_deinit(struct hfi1_devdata *dd); + +void hfi1_vnic_bypass_rcv(struct hfi1_packet *packet); +void hfi1_vnic_sdma_init(struct hfi1_vnic_vport_info *vinfo); +bool hfi1_vnic_sdma_write_avail(struct hfi1_vnic_vport_info *vinfo, + u8 q_idx); + +/* vnic rdma netdev operations */ +struct net_device *hfi1_vnic_alloc_rn(struct ib_device *device, + u8 port_num, + enum rdma_netdev_t type, + const char *name, + unsigned char name_assign_type, + void (*setup)(struct net_device *)); +void hfi1_vnic_free_rn(struct net_device *netdev); +int hfi1_vnic_send_dma(struct hfi1_devdata *dd, u8 q_idx, + struct hfi1_vnic_vport_info *vinfo, + struct sk_buff *skb, u64 pbc, u8 plen); + +#endif /* _HFI1_VNIC_H */ diff --git a/drivers/infiniband/hw/hfi1/vnic_main.c b/drivers/infiniband/hw/hfi1/vnic_main.c new file mode 100644 index 000000000000..392f4d57f3e3 --- /dev/null +++ b/drivers/infiniband/hw/hfi1/vnic_main.c @@ -0,0 +1,907 @@ +/* + * Copyright(c) 2017 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ + +/* + * This file contains HFI1 support for VNIC functionality + */ + +#include <linux/io.h> +#include <linux/if_vlan.h> + +#include "vnic.h" + +#define HFI_TX_TIMEOUT_MS 1000 + +#define HFI1_VNIC_RCV_Q_SIZE 1024 + +#define HFI1_VNIC_UP 0 + +static DEFINE_SPINLOCK(vport_cntr_lock); + +static int setup_vnic_ctxt(struct hfi1_devdata *dd, struct hfi1_ctxtdata *uctxt) +{ + unsigned int rcvctrl_ops = 0; + int ret; + + ret = hfi1_init_ctxt(uctxt->sc); + if (ret) + goto done; + + uctxt->do_interrupt = &handle_receive_interrupt; + + /* Now allocate the RcvHdr queue and eager buffers. */ + ret = hfi1_create_rcvhdrq(dd, uctxt); + if (ret) + goto done; + + ret = hfi1_setup_eagerbufs(uctxt); + if (ret) + goto done; + + set_bit(HFI1_CTXT_SETUP_DONE, &uctxt->event_flags); + + if (uctxt->rcvhdrtail_kvaddr) + clear_rcvhdrtail(uctxt); + + rcvctrl_ops = HFI1_RCVCTRL_CTXT_ENB; + rcvctrl_ops |= HFI1_RCVCTRL_INTRAVAIL_ENB; + + if (!HFI1_CAP_KGET_MASK(uctxt->flags, MULTI_PKT_EGR)) + rcvctrl_ops |= HFI1_RCVCTRL_ONE_PKT_EGR_ENB; + if (HFI1_CAP_KGET_MASK(uctxt->flags, NODROP_EGR_FULL)) + rcvctrl_ops |= HFI1_RCVCTRL_NO_EGR_DROP_ENB; + if (HFI1_CAP_KGET_MASK(uctxt->flags, NODROP_RHQ_FULL)) + rcvctrl_ops |= HFI1_RCVCTRL_NO_RHQ_DROP_ENB; + if (HFI1_CAP_KGET_MASK(uctxt->flags, DMA_RTAIL)) + rcvctrl_ops |= HFI1_RCVCTRL_TAILUPD_ENB; + + hfi1_rcvctrl(uctxt->dd, rcvctrl_ops, uctxt->ctxt); + + uctxt->is_vnic = true; +done: + return ret; +} + +static int allocate_vnic_ctxt(struct hfi1_devdata *dd, + struct hfi1_ctxtdata **vnic_ctxt) +{ + struct hfi1_ctxtdata *uctxt; + unsigned int ctxt; + int ret; + + if (dd->flags & HFI1_FROZEN) + return -EIO; + + for (ctxt = dd->first_dyn_alloc_ctxt; + ctxt < dd->num_rcv_contexts; ctxt++) + if (!dd->rcd[ctxt]) + break; + + if (ctxt == dd->num_rcv_contexts) + return -EBUSY; + + uctxt = hfi1_create_ctxtdata(dd->pport, ctxt, dd->node); + if (!uctxt) { + dd_dev_err(dd, "Unable to create ctxtdata, failing open\n"); + return -ENOMEM; + } + + uctxt->flags = HFI1_CAP_KGET(MULTI_PKT_EGR) | + HFI1_CAP_KGET(NODROP_RHQ_FULL) | + HFI1_CAP_KGET(NODROP_EGR_FULL) | + HFI1_CAP_KGET(DMA_RTAIL); + uctxt->seq_cnt = 1; + + /* Allocate and enable a PIO send context */ + uctxt->sc = sc_alloc(dd, SC_VNIC, uctxt->rcvhdrqentsize, + uctxt->numa_id); + + ret = uctxt->sc ? 0 : -ENOMEM; + if (ret) + goto bail; + + dd_dev_dbg(dd, "allocated vnic send context %u(%u)\n", + uctxt->sc->sw_index, uctxt->sc->hw_context); + ret = sc_enable(uctxt->sc); + if (ret) + goto bail; + + if (dd->num_msix_entries) + hfi1_set_vnic_msix_info(uctxt); + + hfi1_stats.sps_ctxts++; + dd_dev_dbg(dd, "created vnic context %d\n", uctxt->ctxt); + *vnic_ctxt = uctxt; + + return ret; +bail: + /* + * hfi1_free_ctxtdata() also releases send_context + * structure if uctxt->sc is not null + */ + dd->rcd[uctxt->ctxt] = NULL; + hfi1_free_ctxtdata(dd, uctxt); + dd_dev_dbg(dd, "vnic allocation failed. rc %d\n", ret); + return ret; +} + +static void deallocate_vnic_ctxt(struct hfi1_devdata *dd, + struct hfi1_ctxtdata *uctxt) +{ + unsigned long flags; + + dd_dev_dbg(dd, "closing vnic context %d\n", uctxt->ctxt); + flush_wc(); + + if (dd->num_msix_entries) + hfi1_reset_vnic_msix_info(uctxt); + + spin_lock_irqsave(&dd->uctxt_lock, flags); + /* + * Disable receive context and interrupt available, reset all + * RcvCtxtCtrl bits to default values. + */ + hfi1_rcvctrl(dd, HFI1_RCVCTRL_CTXT_DIS | + HFI1_RCVCTRL_TIDFLOW_DIS | + HFI1_RCVCTRL_INTRAVAIL_DIS | + HFI1_RCVCTRL_ONE_PKT_EGR_DIS | + HFI1_RCVCTRL_NO_RHQ_DROP_DIS | + HFI1_RCVCTRL_NO_EGR_DROP_DIS, uctxt->ctxt); + /* + * VNIC contexts are allocated from user context pool. + * Release them back to user context pool. + * + * Reset context integrity checks to default. + * (writes to CSRs probably belong in chip.c) + */ + write_kctxt_csr(dd, uctxt->sc->hw_context, SEND_CTXT_CHECK_ENABLE, + hfi1_pkt_default_send_ctxt_mask(dd, SC_USER)); + sc_disable(uctxt->sc); + + dd->send_contexts[uctxt->sc->sw_index].type = SC_USER; + spin_unlock_irqrestore(&dd->uctxt_lock, flags); + + dd->rcd[uctxt->ctxt] = NULL; + uctxt->event_flags = 0; + + hfi1_clear_tids(uctxt); + hfi1_clear_ctxt_pkey(dd, uctxt->ctxt); + + hfi1_stats.sps_ctxts--; + hfi1_free_ctxtdata(dd, uctxt); +} + +void hfi1_vnic_setup(struct hfi1_devdata *dd) +{ + idr_init(&dd->vnic.vesw_idr); +} + +void hfi1_vnic_cleanup(struct hfi1_devdata *dd) +{ + idr_destroy(&dd->vnic.vesw_idr); +} + +#define SUM_GRP_COUNTERS(stats, qstats, x_grp) do { \ + u64 *src64, *dst64; \ + for (src64 = &qstats->x_grp.unicast, \ + dst64 = &stats->x_grp.unicast; \ + dst64 <= &stats->x_grp.s_1519_max;) { \ + *dst64++ += *src64++; \ + } \ + } while (0) + +/* hfi1_vnic_update_stats - update statistics */ +static void hfi1_vnic_update_stats(struct hfi1_vnic_vport_info *vinfo, + struct opa_vnic_stats *stats) +{ + struct net_device *netdev = vinfo->netdev; + u8 i; + + /* add tx counters on different queues */ + for (i = 0; i < vinfo->num_tx_q; i++) { + struct opa_vnic_stats *qstats = &vinfo->stats[i]; + struct rtnl_link_stats64 *qnstats = &vinfo->stats[i].netstats; + + stats->netstats.tx_fifo_errors += qnstats->tx_fifo_errors; + stats->netstats.tx_carrier_errors += qnstats->tx_carrier_errors; + stats->tx_drop_state += qstats->tx_drop_state; + stats->tx_dlid_zero += qstats->tx_dlid_zero; + + SUM_GRP_COUNTERS(stats, qstats, tx_grp); + stats->netstats.tx_packets += qnstats->tx_packets; + stats->netstats.tx_bytes += qnstats->tx_bytes; + } + + /* add rx counters on different queues */ + for (i = 0; i < vinfo->num_rx_q; i++) { + struct opa_vnic_stats *qstats = &vinfo->stats[i]; + struct rtnl_link_stats64 *qnstats = &vinfo->stats[i].netstats; + + stats->netstats.rx_fifo_errors += qnstats->rx_fifo_errors; + stats->netstats.rx_nohandler += qnstats->rx_nohandler; + stats->rx_drop_state += qstats->rx_drop_state; + stats->rx_oversize += qstats->rx_oversize; + stats->rx_runt += qstats->rx_runt; + + SUM_GRP_COUNTERS(stats, qstats, rx_grp); + stats->netstats.rx_packets += qnstats->rx_packets; + stats->netstats.rx_bytes += qnstats->rx_bytes; + } + + stats->netstats.tx_errors = stats->netstats.tx_fifo_errors + + stats->netstats.tx_carrier_errors + + stats->tx_drop_state + stats->tx_dlid_zero; + stats->netstats.tx_dropped = stats->netstats.tx_errors; + + stats->netstats.rx_errors = stats->netstats.rx_fifo_errors + + stats->netstats.rx_nohandler + + stats->rx_drop_state + stats->rx_oversize + + stats->rx_runt; + stats->netstats.rx_dropped = stats->netstats.rx_errors; + + netdev->stats.tx_packets = stats->netstats.tx_packets; + netdev->stats.tx_bytes = stats->netstats.tx_bytes; + netdev->stats.tx_fifo_errors = stats->netstats.tx_fifo_errors; + netdev->stats.tx_carrier_errors = stats->netstats.tx_carrier_errors; + netdev->stats.tx_errors = stats->netstats.tx_errors; + netdev->stats.tx_dropped = stats->netstats.tx_dropped; + + netdev->stats.rx_packets = stats->netstats.rx_packets; + netdev->stats.rx_bytes = stats->netstats.rx_bytes; + netdev->stats.rx_fifo_errors = stats->netstats.rx_fifo_errors; + netdev->stats.multicast = stats->rx_grp.mcastbcast; + netdev->stats.rx_length_errors = stats->rx_oversize + stats->rx_runt; + netdev->stats.rx_errors = stats->netstats.rx_errors; + netdev->stats.rx_dropped = stats->netstats.rx_dropped; +} + +/* update_len_counters - update pkt's len histogram counters */ +static inline void update_len_counters(struct opa_vnic_grp_stats *grp, + int len) +{ + /* account for 4 byte FCS */ + if (len >= 1515) + grp->s_1519_max++; + else if (len >= 1020) + grp->s_1024_1518++; + else if (len >= 508) + grp->s_512_1023++; + else if (len >= 252) + grp->s_256_511++; + else if (len >= 124) + grp->s_128_255++; + else if (len >= 61) + grp->s_65_127++; + else + grp->s_64++; +} + +/* hfi1_vnic_update_tx_counters - update transmit counters */ +static void hfi1_vnic_update_tx_counters(struct hfi1_vnic_vport_info *vinfo, + u8 q_idx, struct sk_buff *skb, int err) +{ + struct ethhdr *mac_hdr = (struct ethhdr *)skb_mac_header(skb); + struct opa_vnic_stats *stats = &vinfo->stats[q_idx]; + struct opa_vnic_grp_stats *tx_grp = &stats->tx_grp; + u16 vlan_tci; + + stats->netstats.tx_packets++; + stats->netstats.tx_bytes += skb->len + ETH_FCS_LEN; + + update_len_counters(tx_grp, skb->len); + + /* rest of the counts are for good packets only */ + if (unlikely(err)) + return; + + if (is_multicast_ether_addr(mac_hdr->h_dest)) + tx_grp->mcastbcast++; + else + tx_grp->unicast++; + + if (!__vlan_get_tag(skb, &vlan_tci)) + tx_grp->vlan++; + else + tx_grp->untagged++; +} + +/* hfi1_vnic_update_rx_counters - update receive counters */ +static void hfi1_vnic_update_rx_counters(struct hfi1_vnic_vport_info *vinfo, + u8 q_idx, struct sk_buff *skb, int err) +{ + struct ethhdr *mac_hdr = (struct ethhdr *)skb->data; + struct opa_vnic_stats *stats = &vinfo->stats[q_idx]; + struct opa_vnic_grp_stats *rx_grp = &stats->rx_grp; + u16 vlan_tci; + + stats->netstats.rx_packets++; + stats->netstats.rx_bytes += skb->len + ETH_FCS_LEN; + + update_len_counters(rx_grp, skb->len); + + /* rest of the counts are for good packets only */ + if (unlikely(err)) + return; + + if (is_multicast_ether_addr(mac_hdr->h_dest)) + rx_grp->mcastbcast++; + else + rx_grp->unicast++; + + if (!__vlan_get_tag(skb, &vlan_tci)) + rx_grp->vlan++; + else + rx_grp->untagged++; +} + +/* This function is overloaded for opa_vnic specific implementation */ +static void hfi1_vnic_get_stats64(struct net_device *netdev, + struct rtnl_link_stats64 *stats) +{ + struct opa_vnic_stats *vstats = (struct opa_vnic_stats *)stats; + struct hfi1_vnic_vport_info *vinfo = opa_vnic_dev_priv(netdev); + + hfi1_vnic_update_stats(vinfo, vstats); +} + +static u64 create_bypass_pbc(u32 vl, u32 dw_len) +{ + u64 pbc; + + pbc = ((u64)PBC_IHCRC_NONE << PBC_INSERT_HCRC_SHIFT) + | PBC_INSERT_BYPASS_ICRC | PBC_CREDIT_RETURN + | PBC_PACKET_BYPASS + | ((vl & PBC_VL_MASK) << PBC_VL_SHIFT) + | (dw_len & PBC_LENGTH_DWS_MASK) << PBC_LENGTH_DWS_SHIFT; + + return pbc; +} + +/* hfi1_vnic_maybe_stop_tx - stop tx queue if required */ +static void hfi1_vnic_maybe_stop_tx(struct hfi1_vnic_vport_info *vinfo, + u8 q_idx) +{ + netif_stop_subqueue(vinfo->netdev, q_idx); + if (!hfi1_vnic_sdma_write_avail(vinfo, q_idx)) + return; + + netif_start_subqueue(vinfo->netdev, q_idx); +} + +static netdev_tx_t hfi1_netdev_start_xmit(struct sk_buff *skb, + struct net_device *netdev) +{ + struct hfi1_vnic_vport_info *vinfo = opa_vnic_dev_priv(netdev); + u8 pad_len, q_idx = skb->queue_mapping; + struct hfi1_devdata *dd = vinfo->dd; + struct opa_vnic_skb_mdata *mdata; + u32 pkt_len, total_len; + int err = -EINVAL; + u64 pbc; + + v_dbg("xmit: queue %d skb len %d\n", q_idx, skb->len); + if (unlikely(!netif_oper_up(netdev))) { + vinfo->stats[q_idx].tx_drop_state++; + goto tx_finish; + } + + /* take out meta data */ + mdata = (struct opa_vnic_skb_mdata *)skb->data; + skb_pull(skb, sizeof(*mdata)); + if (unlikely(mdata->flags & OPA_VNIC_SKB_MDATA_ENCAP_ERR)) { + vinfo->stats[q_idx].tx_dlid_zero++; + goto tx_finish; + } + + /* add tail padding (for 8 bytes size alignment) and icrc */ + pad_len = -(skb->len + OPA_VNIC_ICRC_TAIL_LEN) & 0x7; + pad_len += OPA_VNIC_ICRC_TAIL_LEN; + + /* + * pkt_len is how much data we have to write, includes header and data. + * total_len is length of the packet in Dwords plus the PBC should not + * include the CRC. + */ + pkt_len = (skb->len + pad_len) >> 2; + total_len = pkt_len + 2; /* PBC + packet */ + + pbc = create_bypass_pbc(mdata->vl, total_len); + + skb_get(skb); + v_dbg("pbc 0x%016llX len %d pad_len %d\n", pbc, skb->len, pad_len); + err = dd->process_vnic_dma_send(dd, q_idx, vinfo, skb, pbc, pad_len); + if (unlikely(err)) { + if (err == -ENOMEM) + vinfo->stats[q_idx].netstats.tx_fifo_errors++; + else if (err != -EBUSY) + vinfo->stats[q_idx].netstats.tx_carrier_errors++; + } + /* remove the header before updating tx counters */ + skb_pull(skb, OPA_VNIC_HDR_LEN); + + if (unlikely(err == -EBUSY)) { + hfi1_vnic_maybe_stop_tx(vinfo, q_idx); + dev_kfree_skb_any(skb); + return NETDEV_TX_BUSY; + } + +tx_finish: + /* update tx counters */ + hfi1_vnic_update_tx_counters(vinfo, q_idx, skb, err); + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; +} + +static u16 hfi1_vnic_select_queue(struct net_device *netdev, + struct sk_buff *skb, + void *accel_priv, + select_queue_fallback_t fallback) +{ + struct hfi1_vnic_vport_info *vinfo = opa_vnic_dev_priv(netdev); + struct opa_vnic_skb_mdata *mdata; + struct sdma_engine *sde; + + mdata = (struct opa_vnic_skb_mdata *)skb->data; + sde = sdma_select_engine_vl(vinfo->dd, mdata->entropy, mdata->vl); + return sde->this_idx; +} + +/* hfi1_vnic_decap_skb - strip OPA header from the skb (ethernet) packet */ +static inline int hfi1_vnic_decap_skb(struct hfi1_vnic_rx_queue *rxq, + struct sk_buff *skb) +{ + struct hfi1_vnic_vport_info *vinfo = rxq->vinfo; + int max_len = vinfo->netdev->mtu + VLAN_ETH_HLEN; + int rc = -EFAULT; + + skb_pull(skb, OPA_VNIC_HDR_LEN); + + /* Validate Packet length */ + if (unlikely(skb->len > max_len)) + vinfo->stats[rxq->idx].rx_oversize++; + else if (unlikely(skb->len < ETH_ZLEN)) + vinfo->stats[rxq->idx].rx_runt++; + else + rc = 0; + return rc; +} + +static inline struct sk_buff *hfi1_vnic_get_skb(struct hfi1_vnic_rx_queue *rxq) +{ + unsigned char *pad_info; + struct sk_buff *skb; + + skb = skb_dequeue(&rxq->skbq); + if (unlikely(!skb)) + return NULL; + + /* remove tail padding and icrc */ + pad_info = skb->data + skb->len - 1; + skb_trim(skb, (skb->len - OPA_VNIC_ICRC_TAIL_LEN - + ((*pad_info) & 0x7))); + + return skb; +} + +/* hfi1_vnic_handle_rx - handle skb receive */ +static void hfi1_vnic_handle_rx(struct hfi1_vnic_rx_queue *rxq, + int *work_done, int work_to_do) +{ + struct hfi1_vnic_vport_info *vinfo = rxq->vinfo; + struct sk_buff *skb; + int rc; + + while (1) { + if (*work_done >= work_to_do) + break; + + skb = hfi1_vnic_get_skb(rxq); + if (unlikely(!skb)) + break; + + rc = hfi1_vnic_decap_skb(rxq, skb); + /* update rx counters */ + hfi1_vnic_update_rx_counters(vinfo, rxq->idx, skb, rc); + if (unlikely(rc)) { + dev_kfree_skb_any(skb); + continue; + } + + skb_checksum_none_assert(skb); + skb->protocol = eth_type_trans(skb, rxq->netdev); + + napi_gro_receive(&rxq->napi, skb); + (*work_done)++; + } +} + +/* hfi1_vnic_napi - napi receive polling callback function */ +static int hfi1_vnic_napi(struct napi_struct *napi, int budget) +{ + struct hfi1_vnic_rx_queue *rxq = container_of(napi, + struct hfi1_vnic_rx_queue, napi); + struct hfi1_vnic_vport_info *vinfo = rxq->vinfo; + int work_done = 0; + + v_dbg("napi %d budget %d\n", rxq->idx, budget); + hfi1_vnic_handle_rx(rxq, &work_done, budget); + + v_dbg("napi %d work_done %d\n", rxq->idx, work_done); + if (work_done < budget) + napi_complete(napi); + + return work_done; +} + +void hfi1_vnic_bypass_rcv(struct hfi1_packet *packet) +{ + struct hfi1_devdata *dd = packet->rcd->dd; + struct hfi1_vnic_vport_info *vinfo = NULL; + struct hfi1_vnic_rx_queue *rxq; + struct sk_buff *skb; + int l4_type, vesw_id = -1; + u8 q_idx; + + l4_type = HFI1_GET_L4_TYPE(packet->ebuf); + if (likely(l4_type == OPA_VNIC_L4_ETHR)) { + vesw_id = HFI1_VNIC_GET_VESWID(packet->ebuf); + vinfo = idr_find(&dd->vnic.vesw_idr, vesw_id); + + /* + * In case of invalid vesw id, count the error on + * the first available vport. + */ + if (unlikely(!vinfo)) { + struct hfi1_vnic_vport_info *vinfo_tmp; + int id_tmp = 0; + + vinfo_tmp = idr_get_next(&dd->vnic.vesw_idr, &id_tmp); + if (vinfo_tmp) { + spin_lock(&vport_cntr_lock); + vinfo_tmp->stats[0].netstats.rx_nohandler++; + spin_unlock(&vport_cntr_lock); + } + } + } + + if (unlikely(!vinfo)) { + dd_dev_warn(dd, "vnic rcv err: l4 %d vesw id %d ctx %d\n", + l4_type, vesw_id, packet->rcd->ctxt); + return; + } + + q_idx = packet->rcd->vnic_q_idx; + rxq = &vinfo->rxq[q_idx]; + if (unlikely(!netif_oper_up(vinfo->netdev))) { + vinfo->stats[q_idx].rx_drop_state++; + skb_queue_purge(&rxq->skbq); + return; + } + + if (unlikely(skb_queue_len(&rxq->skbq) > HFI1_VNIC_RCV_Q_SIZE)) { + vinfo->stats[q_idx].netstats.rx_fifo_errors++; + return; + } + + skb = netdev_alloc_skb(vinfo->netdev, packet->tlen); + if (unlikely(!skb)) { + vinfo->stats[q_idx].netstats.rx_fifo_errors++; + return; + } + + memcpy(skb->data, packet->ebuf, packet->tlen); + skb_put(skb, packet->tlen); + skb_queue_tail(&rxq->skbq, skb); + + if (napi_schedule_prep(&rxq->napi)) { + v_dbg("napi %d scheduling\n", q_idx); + __napi_schedule(&rxq->napi); + } +} + +static int hfi1_vnic_up(struct hfi1_vnic_vport_info *vinfo) +{ + struct hfi1_devdata *dd = vinfo->dd; + struct net_device *netdev = vinfo->netdev; + int i, rc; + + /* ensure virtual eth switch id is valid */ + if (!vinfo->vesw_id) + return -EINVAL; + + rc = idr_alloc(&dd->vnic.vesw_idr, vinfo, vinfo->vesw_id, + vinfo->vesw_id + 1, GFP_NOWAIT); + if (rc < 0) + return rc; + + for (i = 0; i < vinfo->num_rx_q; i++) { + struct hfi1_vnic_rx_queue *rxq = &vinfo->rxq[i]; + + skb_queue_head_init(&rxq->skbq); + napi_enable(&rxq->napi); + } + + netif_carrier_on(netdev); + netif_tx_start_all_queues(netdev); + set_bit(HFI1_VNIC_UP, &vinfo->flags); + + return 0; +} + +static void hfi1_vnic_down(struct hfi1_vnic_vport_info *vinfo) +{ + struct hfi1_devdata *dd = vinfo->dd; + u8 i; + + clear_bit(HFI1_VNIC_UP, &vinfo->flags); + netif_carrier_off(vinfo->netdev); + netif_tx_disable(vinfo->netdev); + idr_remove(&dd->vnic.vesw_idr, vinfo->vesw_id); + + /* ensure irqs see the change */ + hfi1_vnic_synchronize_irq(dd); + + /* remove unread skbs */ + for (i = 0; i < vinfo->num_rx_q; i++) { + struct hfi1_vnic_rx_queue *rxq = &vinfo->rxq[i]; + + napi_disable(&rxq->napi); + skb_queue_purge(&rxq->skbq); + } +} + +static int hfi1_netdev_open(struct net_device *netdev) +{ + struct hfi1_vnic_vport_info *vinfo = opa_vnic_dev_priv(netdev); + int rc; + + mutex_lock(&vinfo->lock); + rc = hfi1_vnic_up(vinfo); + mutex_unlock(&vinfo->lock); + return rc; +} + +static int hfi1_netdev_close(struct net_device *netdev) +{ + struct hfi1_vnic_vport_info *vinfo = opa_vnic_dev_priv(netdev); + + mutex_lock(&vinfo->lock); + if (test_bit(HFI1_VNIC_UP, &vinfo->flags)) + hfi1_vnic_down(vinfo); + mutex_unlock(&vinfo->lock); + return 0; +} + +static int hfi1_vnic_allot_ctxt(struct hfi1_devdata *dd, + struct hfi1_ctxtdata **vnic_ctxt) +{ + int rc; + + rc = allocate_vnic_ctxt(dd, vnic_ctxt); + if (rc) { + dd_dev_err(dd, "vnic ctxt alloc failed %d\n", rc); + return rc; + } + + rc = setup_vnic_ctxt(dd, *vnic_ctxt); + if (rc) { + dd_dev_err(dd, "vnic ctxt setup failed %d\n", rc); + deallocate_vnic_ctxt(dd, *vnic_ctxt); + *vnic_ctxt = NULL; + } + + return rc; +} + +static int hfi1_vnic_init(struct hfi1_vnic_vport_info *vinfo) +{ + struct hfi1_devdata *dd = vinfo->dd; + int i, rc = 0; + + mutex_lock(&hfi1_mutex); + if (!dd->vnic.num_vports) { + rc = hfi1_vnic_txreq_init(dd); + if (rc) + goto txreq_fail; + + dd->vnic.msix_idx = dd->first_dyn_msix_idx; + } + + for (i = dd->vnic.num_ctxt; i < vinfo->num_rx_q; i++) { + rc = hfi1_vnic_allot_ctxt(dd, &dd->vnic.ctxt[i]); + if (rc) + break; + dd->vnic.ctxt[i]->vnic_q_idx = i; + } + + if (i < vinfo->num_rx_q) { + /* + * If required amount of contexts is not + * allocated successfully then remaining contexts + * are released. + */ + while (i-- > dd->vnic.num_ctxt) { + deallocate_vnic_ctxt(dd, dd->vnic.ctxt[i]); + dd->vnic.ctxt[i] = NULL; + } + goto alloc_fail; + } + + if (dd->vnic.num_ctxt != i) { + dd->vnic.num_ctxt = i; + hfi1_init_vnic_rsm(dd); + } + + dd->vnic.num_vports++; + hfi1_vnic_sdma_init(vinfo); +alloc_fail: + if (!dd->vnic.num_vports) + hfi1_vnic_txreq_deinit(dd); +txreq_fail: + mutex_unlock(&hfi1_mutex); + return rc; +} + +static void hfi1_vnic_deinit(struct hfi1_vnic_vport_info *vinfo) +{ + struct hfi1_devdata *dd = vinfo->dd; + int i; + + mutex_lock(&hfi1_mutex); + if (--dd->vnic.num_vports == 0) { + for (i = 0; i < dd->vnic.num_ctxt; i++) { + deallocate_vnic_ctxt(dd, dd->vnic.ctxt[i]); + dd->vnic.ctxt[i] = NULL; + } + hfi1_deinit_vnic_rsm(dd); + dd->vnic.num_ctxt = 0; + hfi1_vnic_txreq_deinit(dd); + } + mutex_unlock(&hfi1_mutex); +} + +static void hfi1_vnic_set_vesw_id(struct net_device *netdev, int id) +{ + struct hfi1_vnic_vport_info *vinfo = opa_vnic_dev_priv(netdev); + bool reopen = false; + + /* + * If vesw_id is being changed, and if the vnic port is up, + * reset the vnic port to ensure new vesw_id gets picked up + */ + if (id != vinfo->vesw_id) { + mutex_lock(&vinfo->lock); + if (test_bit(HFI1_VNIC_UP, &vinfo->flags)) { + hfi1_vnic_down(vinfo); + reopen = true; + } + + vinfo->vesw_id = id; + if (reopen) + hfi1_vnic_up(vinfo); + + mutex_unlock(&vinfo->lock); + } +} + +/* netdev ops */ +static const struct net_device_ops hfi1_netdev_ops = { + .ndo_open = hfi1_netdev_open, + .ndo_stop = hfi1_netdev_close, + .ndo_start_xmit = hfi1_netdev_start_xmit, + .ndo_select_queue = hfi1_vnic_select_queue, + .ndo_get_stats64 = hfi1_vnic_get_stats64, +}; + +struct net_device *hfi1_vnic_alloc_rn(struct ib_device *device, + u8 port_num, + enum rdma_netdev_t type, + const char *name, + unsigned char name_assign_type, + void (*setup)(struct net_device *)) +{ + struct hfi1_devdata *dd = dd_from_ibdev(device); + struct hfi1_vnic_vport_info *vinfo; + struct net_device *netdev; + struct rdma_netdev *rn; + int i, size, rc; + + if (!port_num || (port_num > dd->num_pports)) + return ERR_PTR(-EINVAL); + + if (type != RDMA_NETDEV_OPA_VNIC) + return ERR_PTR(-EOPNOTSUPP); + + size = sizeof(struct opa_vnic_rdma_netdev) + sizeof(*vinfo); + netdev = alloc_netdev_mqs(size, name, name_assign_type, setup, + dd->chip_sdma_engines, HFI1_NUM_VNIC_CTXT); + if (!netdev) + return ERR_PTR(-ENOMEM); + + rn = netdev_priv(netdev); + vinfo = opa_vnic_dev_priv(netdev); + vinfo->dd = dd; + vinfo->num_tx_q = dd->chip_sdma_engines; + vinfo->num_rx_q = HFI1_NUM_VNIC_CTXT; + vinfo->netdev = netdev; + rn->set_id = hfi1_vnic_set_vesw_id; + + netdev->features = NETIF_F_HIGHDMA | NETIF_F_SG; + netdev->hw_features = netdev->features; + netdev->vlan_features = netdev->features; + netdev->watchdog_timeo = msecs_to_jiffies(HFI_TX_TIMEOUT_MS); + netdev->netdev_ops = &hfi1_netdev_ops; + mutex_init(&vinfo->lock); + + for (i = 0; i < vinfo->num_rx_q; i++) { + struct hfi1_vnic_rx_queue *rxq = &vinfo->rxq[i]; + + rxq->idx = i; + rxq->vinfo = vinfo; + rxq->netdev = netdev; + netif_napi_add(netdev, &rxq->napi, hfi1_vnic_napi, 64); + } + + rc = hfi1_vnic_init(vinfo); + if (rc) + goto init_fail; + + return netdev; +init_fail: + mutex_destroy(&vinfo->lock); + free_netdev(netdev); + return ERR_PTR(rc); +} + +void hfi1_vnic_free_rn(struct net_device *netdev) +{ + struct hfi1_vnic_vport_info *vinfo = opa_vnic_dev_priv(netdev); + + hfi1_vnic_deinit(vinfo); + mutex_destroy(&vinfo->lock); + free_netdev(netdev); +} diff --git a/drivers/infiniband/hw/hfi1/vnic_sdma.c b/drivers/infiniband/hw/hfi1/vnic_sdma.c new file mode 100644 index 000000000000..51a817d3aa14 --- /dev/null +++ b/drivers/infiniband/hw/hfi1/vnic_sdma.c @@ -0,0 +1,323 @@ +/* + * Copyright(c) 2017 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ + +/* + * This file contains HFI1 support for VNIC SDMA functionality + */ + +#include "sdma.h" +#include "vnic.h" + +#define HFI1_VNIC_SDMA_Q_ACTIVE BIT(0) +#define HFI1_VNIC_SDMA_Q_DEFERRED BIT(1) + +#define HFI1_VNIC_TXREQ_NAME_LEN 32 +#define HFI1_VNIC_SDMA_DESC_WTRMRK 64 +#define HFI1_VNIC_SDMA_RETRY_COUNT 1 + +/* + * struct vnic_txreq - VNIC transmit descriptor + * @txreq: sdma transmit request + * @sdma: vnic sdma pointer + * @skb: skb to send + * @pad: pad buffer + * @plen: pad length + * @pbc_val: pbc value + * @retry_count: tx retry count + */ +struct vnic_txreq { + struct sdma_txreq txreq; + struct hfi1_vnic_sdma *sdma; + + struct sk_buff *skb; + unsigned char pad[HFI1_VNIC_MAX_PAD]; + u16 plen; + __le64 pbc_val; + + u32 retry_count; +}; + +static void vnic_sdma_complete(struct sdma_txreq *txreq, + int status) +{ + struct vnic_txreq *tx = container_of(txreq, struct vnic_txreq, txreq); + struct hfi1_vnic_sdma *vnic_sdma = tx->sdma; + + sdma_txclean(vnic_sdma->dd, txreq); + dev_kfree_skb_any(tx->skb); + kmem_cache_free(vnic_sdma->dd->vnic.txreq_cache, tx); +} + +static noinline int build_vnic_ulp_payload(struct sdma_engine *sde, + struct vnic_txreq *tx) +{ + int i, ret = 0; + + ret = sdma_txadd_kvaddr( + sde->dd, + &tx->txreq, + tx->skb->data, + skb_headlen(tx->skb)); + if (unlikely(ret)) + goto bail_txadd; + + for (i = 0; i < skb_shinfo(tx->skb)->nr_frags; i++) { + struct skb_frag_struct *frag = &skb_shinfo(tx->skb)->frags[i]; + + /* combine physically continuous fragments later? */ + ret = sdma_txadd_page(sde->dd, + &tx->txreq, + skb_frag_page(frag), + frag->page_offset, + skb_frag_size(frag)); + if (unlikely(ret)) + goto bail_txadd; + } + + if (tx->plen) + ret = sdma_txadd_kvaddr(sde->dd, &tx->txreq, + tx->pad + HFI1_VNIC_MAX_PAD - tx->plen, + tx->plen); + +bail_txadd: + return ret; +} + +static int build_vnic_tx_desc(struct sdma_engine *sde, + struct vnic_txreq *tx, + u64 pbc) +{ + int ret = 0; + u16 hdrbytes = 2 << 2; /* PBC */ + + ret = sdma_txinit_ahg( + &tx->txreq, + 0, + hdrbytes + tx->skb->len + tx->plen, + 0, + 0, + NULL, + 0, + vnic_sdma_complete); + if (unlikely(ret)) + goto bail_txadd; + + /* add pbc */ + tx->pbc_val = cpu_to_le64(pbc); + ret = sdma_txadd_kvaddr( + sde->dd, + &tx->txreq, + &tx->pbc_val, + hdrbytes); + if (unlikely(ret)) + goto bail_txadd; + + /* add the ulp payload */ + ret = build_vnic_ulp_payload(sde, tx); +bail_txadd: + return ret; +} + +/* setup the last plen bypes of pad */ +static inline void hfi1_vnic_update_pad(unsigned char *pad, u8 plen) +{ + pad[HFI1_VNIC_MAX_PAD - 1] = plen - OPA_VNIC_ICRC_TAIL_LEN; +} + +int hfi1_vnic_send_dma(struct hfi1_devdata *dd, u8 q_idx, + struct hfi1_vnic_vport_info *vinfo, + struct sk_buff *skb, u64 pbc, u8 plen) +{ + struct hfi1_vnic_sdma *vnic_sdma = &vinfo->sdma[q_idx]; + struct sdma_engine *sde = vnic_sdma->sde; + struct vnic_txreq *tx; + int ret = -ECOMM; + + if (unlikely(READ_ONCE(vnic_sdma->state) != HFI1_VNIC_SDMA_Q_ACTIVE)) + goto tx_err; + + if (unlikely(!sde || !sdma_running(sde))) + goto tx_err; + + tx = kmem_cache_alloc(dd->vnic.txreq_cache, GFP_ATOMIC); + if (unlikely(!tx)) { + ret = -ENOMEM; + goto tx_err; + } + + tx->sdma = vnic_sdma; + tx->skb = skb; + hfi1_vnic_update_pad(tx->pad, plen); + tx->plen = plen; + ret = build_vnic_tx_desc(sde, tx, pbc); + if (unlikely(ret)) + goto free_desc; + tx->retry_count = 0; + + ret = sdma_send_txreq(sde, &vnic_sdma->wait, &tx->txreq); + /* When -ECOMM, sdma callback will be called with ABORT status */ + if (unlikely(ret && unlikely(ret != -ECOMM))) + goto free_desc; + + return ret; + +free_desc: + sdma_txclean(dd, &tx->txreq); + kmem_cache_free(dd->vnic.txreq_cache, tx); +tx_err: + if (ret != -EBUSY) + dev_kfree_skb_any(skb); + return ret; +} + +/* + * hfi1_vnic_sdma_sleep - vnic sdma sleep function + * + * This function gets called from sdma_send_txreq() when there are not enough + * sdma descriptors available to send the packet. It adds Tx queue's wait + * structure to sdma engine's dmawait list to be woken up when descriptors + * become available. + */ +static int hfi1_vnic_sdma_sleep(struct sdma_engine *sde, + struct iowait *wait, + struct sdma_txreq *txreq, + unsigned int seq) +{ + struct hfi1_vnic_sdma *vnic_sdma = + container_of(wait, struct hfi1_vnic_sdma, wait); + struct hfi1_ibdev *dev = &vnic_sdma->dd->verbs_dev; + struct vnic_txreq *tx = container_of(txreq, struct vnic_txreq, txreq); + + if (sdma_progress(sde, seq, txreq)) + if (tx->retry_count++ < HFI1_VNIC_SDMA_RETRY_COUNT) + return -EAGAIN; + + vnic_sdma->state = HFI1_VNIC_SDMA_Q_DEFERRED; + write_seqlock(&dev->iowait_lock); + if (list_empty(&vnic_sdma->wait.list)) + list_add_tail(&vnic_sdma->wait.list, &sde->dmawait); + write_sequnlock(&dev->iowait_lock); + return -EBUSY; +} + +/* + * hfi1_vnic_sdma_wakeup - vnic sdma wakeup function + * + * This function gets called when SDMA descriptors becomes available and Tx + * queue's wait structure was previously added to sdma engine's dmawait list. + * It notifies the upper driver about Tx queue wakeup. + */ +static void hfi1_vnic_sdma_wakeup(struct iowait *wait, int reason) +{ + struct hfi1_vnic_sdma *vnic_sdma = + container_of(wait, struct hfi1_vnic_sdma, wait); + struct hfi1_vnic_vport_info *vinfo = vnic_sdma->vinfo; + + vnic_sdma->state = HFI1_VNIC_SDMA_Q_ACTIVE; + if (__netif_subqueue_stopped(vinfo->netdev, vnic_sdma->q_idx)) + netif_wake_subqueue(vinfo->netdev, vnic_sdma->q_idx); +}; + +inline bool hfi1_vnic_sdma_write_avail(struct hfi1_vnic_vport_info *vinfo, + u8 q_idx) +{ + struct hfi1_vnic_sdma *vnic_sdma = &vinfo->sdma[q_idx]; + + return (READ_ONCE(vnic_sdma->state) == HFI1_VNIC_SDMA_Q_ACTIVE); +} + +void hfi1_vnic_sdma_init(struct hfi1_vnic_vport_info *vinfo) +{ + int i; + + for (i = 0; i < vinfo->num_tx_q; i++) { + struct hfi1_vnic_sdma *vnic_sdma = &vinfo->sdma[i]; + + iowait_init(&vnic_sdma->wait, 0, NULL, hfi1_vnic_sdma_sleep, + hfi1_vnic_sdma_wakeup, NULL); + vnic_sdma->sde = &vinfo->dd->per_sdma[i]; + vnic_sdma->dd = vinfo->dd; + vnic_sdma->vinfo = vinfo; + vnic_sdma->q_idx = i; + vnic_sdma->state = HFI1_VNIC_SDMA_Q_ACTIVE; + + /* Add a free descriptor watermark for wakeups */ + if (vnic_sdma->sde->descq_cnt > HFI1_VNIC_SDMA_DESC_WTRMRK) { + INIT_LIST_HEAD(&vnic_sdma->stx.list); + vnic_sdma->stx.num_desc = HFI1_VNIC_SDMA_DESC_WTRMRK; + list_add_tail(&vnic_sdma->stx.list, + &vnic_sdma->wait.tx_head); + } + } +} + +static void hfi1_vnic_txreq_kmem_cache_ctor(void *obj) +{ + struct vnic_txreq *tx = (struct vnic_txreq *)obj; + + memset(tx, 0, sizeof(*tx)); +} + +int hfi1_vnic_txreq_init(struct hfi1_devdata *dd) +{ + char buf[HFI1_VNIC_TXREQ_NAME_LEN]; + + snprintf(buf, sizeof(buf), "hfi1_%u_vnic_txreq_cache", dd->unit); + dd->vnic.txreq_cache = kmem_cache_create(buf, + sizeof(struct vnic_txreq), + 0, SLAB_HWCACHE_ALIGN, + hfi1_vnic_txreq_kmem_cache_ctor); + if (!dd->vnic.txreq_cache) + return -ENOMEM; + return 0; +} + +void hfi1_vnic_txreq_deinit(struct hfi1_devdata *dd) +{ + kmem_cache_destroy(dd->vnic.txreq_cache); + dd->vnic.txreq_cache = NULL; +} diff --git a/drivers/infiniband/hw/hns/hns_roce_ah.c b/drivers/infiniband/hw/hns/hns_roce_ah.c index 0ac294db3b29..f78a733a63ec 100644 --- a/drivers/infiniband/hw/hns/hns_roce_ah.c +++ b/drivers/infiniband/hw/hns/hns_roce_ah.c @@ -39,7 +39,8 @@ #define HNS_ROCE_VLAN_SL_BIT_MASK 7 #define HNS_ROCE_VLAN_SL_SHIFT 13 -struct ib_ah *hns_roce_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *ah_attr, +struct ib_ah *hns_roce_create_ah(struct ib_pd *ibpd, + struct rdma_ah_attr *ah_attr, struct ib_udata *udata) { struct hns_roce_dev *hr_dev = to_hr_dev(ibpd->device); @@ -48,6 +49,7 @@ struct ib_ah *hns_roce_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *ah_attr, struct hns_roce_ah *ah; u16 vlan_tag = 0xffff; struct in6_addr in6; + const struct ib_global_route *grh = rdma_ah_read_grh(ah_attr); union ib_gid sgid; int ret; @@ -56,15 +58,20 @@ struct ib_ah *hns_roce_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *ah_attr, return ERR_PTR(-ENOMEM); /* Get mac address */ - memcpy(&in6, ah_attr->grh.dgid.raw, sizeof(ah_attr->grh.dgid.raw)); - if (rdma_is_multicast_addr(&in6)) + memcpy(&in6, grh->dgid.raw, sizeof(grh->dgid.raw)); + if (rdma_is_multicast_addr(&in6)) { rdma_get_mcast_mac(&in6, ah->av.mac); - else - memcpy(ah->av.mac, ah_attr->dmac, sizeof(ah_attr->dmac)); + } else { + u8 *dmac = rdma_ah_retrieve_dmac(ah_attr); + + if (!dmac) + return ERR_PTR(-EINVAL); + memcpy(ah->av.mac, dmac, ETH_ALEN); + } /* Get source gid */ - ret = ib_get_cached_gid(ibpd->device, ah_attr->port_num, - ah_attr->grh.sgid_index, &sgid, &gid_attr); + ret = ib_get_cached_gid(ibpd->device, rdma_ah_get_port_num(ah_attr), + grh->sgid_index, &sgid, &gid_attr); if (ret) { dev_err(dev, "get sgid failed! ret = %d\n", ret); kfree(ah); @@ -78,45 +85,46 @@ struct ib_ah *hns_roce_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *ah_attr, } if (vlan_tag < 0x1000) - vlan_tag |= (ah_attr->sl & HNS_ROCE_VLAN_SL_BIT_MASK) << + vlan_tag |= (rdma_ah_get_sl(ah_attr) & + HNS_ROCE_VLAN_SL_BIT_MASK) << HNS_ROCE_VLAN_SL_SHIFT; - ah->av.port_pd = cpu_to_be32(to_hr_pd(ibpd)->pdn | (ah_attr->port_num << + ah->av.port_pd = cpu_to_be32(to_hr_pd(ibpd)->pdn | + (rdma_ah_get_port_num(ah_attr) << HNS_ROCE_PORT_NUM_SHIFT)); - ah->av.gid_index = ah_attr->grh.sgid_index; + ah->av.gid_index = grh->sgid_index; ah->av.vlan = cpu_to_le16(vlan_tag); dev_dbg(dev, "gid_index = 0x%x,vlan = 0x%x\n", ah->av.gid_index, ah->av.vlan); - if (ah_attr->static_rate) + if (rdma_ah_get_static_rate(ah_attr)) ah->av.stat_rate = IB_RATE_10_GBPS; - memcpy(ah->av.dgid, ah_attr->grh.dgid.raw, HNS_ROCE_GID_SIZE); - ah->av.sl_tclass_flowlabel = cpu_to_le32(ah_attr->sl << + memcpy(ah->av.dgid, grh->dgid.raw, HNS_ROCE_GID_SIZE); + ah->av.sl_tclass_flowlabel = cpu_to_le32(rdma_ah_get_sl(ah_attr) << HNS_ROCE_SL_SHIFT); return &ah->ibah; } -int hns_roce_query_ah(struct ib_ah *ibah, struct ib_ah_attr *ah_attr) +int hns_roce_query_ah(struct ib_ah *ibah, struct rdma_ah_attr *ah_attr) { struct hns_roce_ah *ah = to_hr_ah(ibah); memset(ah_attr, 0, sizeof(*ah_attr)); - ah_attr->sl = le32_to_cpu(ah->av.sl_tclass_flowlabel) >> - HNS_ROCE_SL_SHIFT; - ah_attr->port_num = le32_to_cpu(ah->av.port_pd) >> - HNS_ROCE_PORT_NUM_SHIFT; - ah_attr->static_rate = ah->av.stat_rate; - ah_attr->ah_flags = IB_AH_GRH; - ah_attr->grh.traffic_class = le32_to_cpu(ah->av.sl_tclass_flowlabel) >> - HNS_ROCE_TCLASS_SHIFT; - ah_attr->grh.flow_label = le32_to_cpu(ah->av.sl_tclass_flowlabel) & - HNS_ROCE_FLOW_LABLE_MASK; - ah_attr->grh.hop_limit = ah->av.hop_limit; - ah_attr->grh.sgid_index = ah->av.gid_index; - memcpy(ah_attr->grh.dgid.raw, ah->av.dgid, HNS_ROCE_GID_SIZE); + rdma_ah_set_sl(ah_attr, (le32_to_cpu(ah->av.sl_tclass_flowlabel) >> + HNS_ROCE_SL_SHIFT)); + rdma_ah_set_port_num(ah_attr, (le32_to_cpu(ah->av.port_pd) >> + HNS_ROCE_PORT_NUM_SHIFT)); + rdma_ah_set_static_rate(ah_attr, ah->av.stat_rate); + rdma_ah_set_grh(ah_attr, NULL, + (le32_to_cpu(ah->av.sl_tclass_flowlabel) & + HNS_ROCE_FLOW_LABLE_MASK), ah->av.gid_index, + ah->av.hop_limit, + (le32_to_cpu(ah->av.sl_tclass_flowlabel) >> + HNS_ROCE_TCLASS_SHIFT)); + rdma_ah_set_dgid_raw(ah_attr, ah->av.dgid); return 0; } diff --git a/drivers/infiniband/hw/hns/hns_roce_cmd.c b/drivers/infiniband/hw/hns/hns_roce_cmd.c index 8c1f7a6f84d2..b94dcd823ad1 100644 --- a/drivers/infiniband/hw/hns/hns_roce_cmd.c +++ b/drivers/infiniband/hw/hns/hns_roce_cmd.c @@ -299,9 +299,9 @@ int hns_roce_cmd_use_events(struct hns_roce_dev *hr_dev) struct hns_roce_cmdq *hr_cmd = &hr_dev->cmd; int i; - hr_cmd->context = kmalloc(hr_cmd->max_cmds * - sizeof(struct hns_roce_cmd_context), - GFP_KERNEL); + hr_cmd->context = kmalloc_array(hr_cmd->max_cmds, + sizeof(*hr_cmd->context), + GFP_KERNEL); if (!hr_cmd->context) return -ENOMEM; diff --git a/drivers/infiniband/hw/hns/hns_roce_cq.c b/drivers/infiniband/hw/hns/hns_roce_cq.c index 589496c8fb9e..b89fd711019e 100644 --- a/drivers/infiniband/hw/hns/hns_roce_cq.c +++ b/drivers/infiniband/hw/hns/hns_roce_cq.c @@ -219,8 +219,7 @@ static int hns_roce_ib_get_cq_umem(struct hns_roce_dev *hr_dev, return PTR_ERR(*umem); ret = hns_roce_mtt_init(hr_dev, ib_umem_page_count(*umem), - ilog2((unsigned int)(*umem)->page_size), - &buf->hr_mtt); + (*umem)->page_shift, &buf->hr_mtt); if (ret) goto err_buf; diff --git a/drivers/infiniband/hw/hns/hns_roce_device.h b/drivers/infiniband/hw/hns/hns_roce_device.h index 1a6cb5d7a0dd..e493a61e14e1 100644 --- a/drivers/infiniband/hw/hns/hns_roce_device.h +++ b/drivers/infiniband/hw/hns/hns_roce_device.h @@ -687,9 +687,10 @@ void hns_roce_bitmap_free_range(struct hns_roce_bitmap *bitmap, unsigned long obj, int cnt, int rr); -struct ib_ah *hns_roce_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, +struct ib_ah *hns_roce_create_ah(struct ib_pd *pd, + struct rdma_ah_attr *ah_attr, struct ib_udata *udata); -int hns_roce_query_ah(struct ib_ah *ibah, struct ib_ah_attr *ah_attr); +int hns_roce_query_ah(struct ib_ah *ibah, struct rdma_ah_attr *ah_attr); int hns_roce_destroy_ah(struct ib_ah *ah); struct ib_pd *hns_roce_alloc_pd(struct ib_device *ib_dev, diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c index b8111b0c8877..014c8262bfff 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c @@ -33,6 +33,7 @@ #include <linux/platform_device.h> #include <linux/acpi.h> #include <linux/etherdevice.h> +#include <linux/of.h> #include <rdma/ib_umem.h> #include "hns_roce_common.h" #include "hns_roce_device.h" @@ -657,6 +658,7 @@ static int hns_roce_v1_rsv_lp_qp(struct hns_roce_dev *hr_dev) struct hns_roce_qp *hr_qp; struct ib_cq *cq; struct ib_pd *pd; + union ib_gid dgid; u64 subnet_prefix; int attr_mask = 0; int i; @@ -707,12 +709,8 @@ static int hns_roce_v1_rsv_lp_qp(struct hns_roce_dev *hr_dev) attr.rnr_retry = 7; attr.timeout = 0x12; attr.path_mtu = IB_MTU_256; - attr.ah_attr.ah_flags = 1; - attr.ah_attr.static_rate = 3; - attr.ah_attr.grh.sgid_index = 0; - attr.ah_attr.grh.hop_limit = 1; - attr.ah_attr.grh.flow_label = 0; - attr.ah_attr.grh.traffic_class = 0; + rdma_ah_set_grh(&attr.ah_attr, NULL, 0, 0, 1, 0); + rdma_ah_set_static_rate(&attr.ah_attr, 3); subnet_prefix = cpu_to_be64(0xfe80000000000000LL); for (i = 0; i < HNS_ROCE_V1_RESV_QP; i++) { @@ -741,24 +739,22 @@ static int hns_roce_v1_rsv_lp_qp(struct hns_roce_dev *hr_dev) hr_qp->ibqp.recv_cq = cq; hr_qp->ibqp.send_cq = cq; - attr.ah_attr.port_num = phy_port + 1; - attr.ah_attr.sl = sl; + rdma_ah_set_port_num(&attr.ah_attr, phy_port + 1); + rdma_ah_set_sl(&attr.ah_attr, phy_port + 1); attr.port_num = phy_port + 1; attr.dest_qp_num = hr_qp->qpn; - memcpy(attr.ah_attr.dmac, hr_dev->dev_addr[phy_port], + memcpy(rdma_ah_retrieve_dmac(&attr.ah_attr), + hr_dev->dev_addr[phy_port], MAC_ADDR_OCTET_NUM); - memcpy(attr.ah_attr.grh.dgid.raw, - &subnet_prefix, sizeof(u64)); - memcpy(&attr.ah_attr.grh.dgid.raw[8], - hr_dev->dev_addr[phy_port], 3); - memcpy(&attr.ah_attr.grh.dgid.raw[13], - hr_dev->dev_addr[phy_port] + 3, 3); - attr.ah_attr.grh.dgid.raw[11] = 0xff; - attr.ah_attr.grh.dgid.raw[12] = 0xfe; - attr.ah_attr.grh.dgid.raw[8] ^= 2; - + memcpy(&dgid.raw, &subnet_prefix, sizeof(u64)); + memcpy(&dgid.raw[8], hr_dev->dev_addr[phy_port], 3); + memcpy(&dgid.raw[13], hr_dev->dev_addr[phy_port] + 3, 3); + dgid.raw[11] = 0xff; + dgid.raw[12] = 0xfe; + dgid.raw[8] ^= 2; + rdma_ah_set_dgid_raw(&attr.ah_attr, dgid.raw); attr_mask |= IB_QP_PORT; ret = hr_dev->hw->modify_qp(&hr_qp->ibqp, &attr, attr_mask, @@ -1851,6 +1847,7 @@ void hns_roce_v1_cq_set_ci(struct hns_roce_cq *hr_cq, u32 cons_index) u32 doorbell[2]; doorbell[0] = cons_index & ((hr_cq->cq_depth << 1) - 1); + doorbell[1] = 0; roce_set_bit(doorbell[1], ROCEE_DB_OTHERS_H_ROCEE_DB_OTH_HW_SYNS_S, 1); roce_set_field(doorbell[1], ROCEE_DB_OTHERS_H_ROCEE_DB_OTH_CMD_M, ROCEE_DB_OTHERS_H_ROCEE_DB_OTH_CMD_S, 3); @@ -2565,6 +2562,7 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr, struct hns_roce_qp *hr_qp = to_hr_qp(ibqp); struct device *dev = &hr_dev->pdev->dev; struct hns_roce_qp_context *context; + const struct ib_global_route *grh = rdma_ah_read_grh(&attr->ah_attr); dma_addr_t dma_handle_2 = 0; dma_addr_t dma_handle = 0; uint32_t doorbell[2] = {0}; @@ -2573,6 +2571,7 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr, int ret = -EINVAL; u64 *mtts = NULL; int port; + u8 port_num; u8 *dmac; u8 *smac; @@ -2739,7 +2738,7 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr, goto out; } - dmac = (u8 *)attr->ah_attr.dmac; + dmac = (u8 *)attr->ah_attr.roce.dmac; context->sq_rq_bt_l = (u32)(dma_handle); roce_set_field(context->qpc_bytes_24, @@ -2780,7 +2779,7 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr, roce_set_bit(context->qpc_bytes_32, QP_CONTEXT_QPC_BYTE_32_GLOBAL_HEADER_S, - attr->ah_attr.ah_flags); + rdma_ah_get_ah_flags(&attr->ah_attr)); roce_set_field(context->qpc_bytes_32, QP_CONTEXT_QPC_BYTES_32_RESPONDER_RESOURCES_M, QP_CONTEXT_QPC_BYTES_32_RESPONDER_RESOURCES_S, @@ -2792,12 +2791,13 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr, attr->dest_qp_num); /* Configure GID index */ + port_num = rdma_ah_get_port_num(&attr->ah_attr); roce_set_field(context->qpc_bytes_36, QP_CONTEXT_QPC_BYTES_36_SGID_INDEX_M, QP_CONTEXT_QPC_BYTES_36_SGID_INDEX_S, - hns_get_gid_index(hr_dev, - attr->ah_attr.port_num - 1, - attr->ah_attr.grh.sgid_index)); + hns_get_gid_index(hr_dev, + port_num - 1, + grh->sgid_index)); memcpy(&(context->dmac_l), dmac, 4); @@ -2808,26 +2808,26 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr, roce_set_field(context->qpc_bytes_44, QP_CONTEXT_QPC_BYTES_44_MAXIMUM_STATIC_RATE_M, QP_CONTEXT_QPC_BYTES_44_MAXIMUM_STATIC_RATE_S, - attr->ah_attr.static_rate); + rdma_ah_get_static_rate(&attr->ah_attr)); roce_set_field(context->qpc_bytes_44, QP_CONTEXT_QPC_BYTES_44_HOPLMT_M, QP_CONTEXT_QPC_BYTES_44_HOPLMT_S, - attr->ah_attr.grh.hop_limit); + grh->hop_limit); roce_set_field(context->qpc_bytes_48, QP_CONTEXT_QPC_BYTES_48_FLOWLABEL_M, QP_CONTEXT_QPC_BYTES_48_FLOWLABEL_S, - attr->ah_attr.grh.flow_label); + grh->flow_label); roce_set_field(context->qpc_bytes_48, QP_CONTEXT_QPC_BYTES_48_TCLASS_M, QP_CONTEXT_QPC_BYTES_48_TCLASS_S, - attr->ah_attr.grh.traffic_class); + grh->traffic_class); roce_set_field(context->qpc_bytes_48, QP_CONTEXT_QPC_BYTES_48_MTU_M, QP_CONTEXT_QPC_BYTES_48_MTU_S, attr->path_mtu); - memcpy(context->dgid, attr->ah_attr.grh.dgid.raw, - sizeof(attr->ah_attr.grh.dgid.raw)); + memcpy(context->dgid, grh->dgid.raw, + sizeof(grh->dgid.raw)); dev_dbg(dev, "dmac:%x :%lx\n", context->dmac_l, roce_get_field(context->qpc_bytes_44, @@ -2907,8 +2907,9 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr, hr_qp->phy_port); roce_set_field(context->qpc_bytes_156, QP_CONTEXT_QPC_BYTES_156_SL_M, - QP_CONTEXT_QPC_BYTES_156_SL_S, attr->ah_attr.sl); - hr_qp->sl = attr->ah_attr.sl; + QP_CONTEXT_QPC_BYTES_156_SL_S, + rdma_ah_get_sl(&attr->ah_attr)); + hr_qp->sl = rdma_ah_get_sl(&attr->ah_attr); } else if (cur_state == IB_QPS_RTR && new_state == IB_QPS_RTS) { /* If exist optional param, return error */ @@ -3019,8 +3020,9 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr, hr_qp->phy_port); roce_set_field(context->qpc_bytes_156, QP_CONTEXT_QPC_BYTES_156_SL_M, - QP_CONTEXT_QPC_BYTES_156_SL_S, attr->ah_attr.sl); - hr_qp->sl = attr->ah_attr.sl; + QP_CONTEXT_QPC_BYTES_156_SL_S, + rdma_ah_get_sl(&attr->ah_attr)); + hr_qp->sl = rdma_ah_get_sl(&attr->ah_attr); roce_set_field(context->qpc_bytes_156, QP_CONTEXT_QPC_BYTES_156_INITIATOR_DEPTH_M, QP_CONTEXT_QPC_BYTES_156_INITIATOR_DEPTH_S, @@ -3355,28 +3357,33 @@ static int hns_roce_v1_q_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, if (hr_qp->ibqp.qp_type == IB_QPT_RC || hr_qp->ibqp.qp_type == IB_QPT_UC) { - qp_attr->ah_attr.sl = roce_get_field(context->qpc_bytes_156, - QP_CONTEXT_QPC_BYTES_156_SL_M, - QP_CONTEXT_QPC_BYTES_156_SL_S); - qp_attr->ah_attr.grh.flow_label = roce_get_field( - context->qpc_bytes_48, - QP_CONTEXT_QPC_BYTES_48_FLOWLABEL_M, - QP_CONTEXT_QPC_BYTES_48_FLOWLABEL_S); - qp_attr->ah_attr.grh.sgid_index = roce_get_field( - context->qpc_bytes_36, - QP_CONTEXT_QPC_BYTES_36_SGID_INDEX_M, - QP_CONTEXT_QPC_BYTES_36_SGID_INDEX_S); - qp_attr->ah_attr.grh.hop_limit = roce_get_field( - context->qpc_bytes_44, - QP_CONTEXT_QPC_BYTES_44_HOPLMT_M, - QP_CONTEXT_QPC_BYTES_44_HOPLMT_S); - qp_attr->ah_attr.grh.traffic_class = roce_get_field( - context->qpc_bytes_48, - QP_CONTEXT_QPC_BYTES_48_TCLASS_M, - QP_CONTEXT_QPC_BYTES_48_TCLASS_S); - - memcpy(qp_attr->ah_attr.grh.dgid.raw, context->dgid, - sizeof(qp_attr->ah_attr.grh.dgid.raw)); + struct ib_global_route *grh = + rdma_ah_retrieve_grh(&qp_attr->ah_attr); + + rdma_ah_set_sl(&qp_attr->ah_attr, + roce_get_field(context->qpc_bytes_156, + QP_CONTEXT_QPC_BYTES_156_SL_M, + QP_CONTEXT_QPC_BYTES_156_SL_S)); + rdma_ah_set_ah_flags(&qp_attr->ah_attr, IB_AH_GRH); + grh->flow_label = + roce_get_field(context->qpc_bytes_48, + QP_CONTEXT_QPC_BYTES_48_FLOWLABEL_M, + QP_CONTEXT_QPC_BYTES_48_FLOWLABEL_S); + grh->sgid_index = + roce_get_field(context->qpc_bytes_36, + QP_CONTEXT_QPC_BYTES_36_SGID_INDEX_M, + QP_CONTEXT_QPC_BYTES_36_SGID_INDEX_S); + grh->hop_limit = + roce_get_field(context->qpc_bytes_44, + QP_CONTEXT_QPC_BYTES_44_HOPLMT_M, + QP_CONTEXT_QPC_BYTES_44_HOPLMT_S); + grh->traffic_class = + roce_get_field(context->qpc_bytes_48, + QP_CONTEXT_QPC_BYTES_48_TCLASS_M, + QP_CONTEXT_QPC_BYTES_48_TCLASS_S); + + memcpy(grh->dgid.raw, context->dgid, + sizeof(grh->dgid.raw)); } qp_attr->pkey_index = roce_get_field(context->qpc_bytes_12, diff --git a/drivers/infiniband/hw/hns/hns_roce_mr.c b/drivers/infiniband/hw/hns/hns_roce_mr.c index 4139abee3b54..dc5c97c8f070 100644 --- a/drivers/infiniband/hw/hns/hns_roce_mr.c +++ b/drivers/infiniband/hw/hns/hns_roce_mr.c @@ -127,11 +127,12 @@ static int hns_roce_buddy_init(struct hns_roce_buddy *buddy, int max_order) buddy->max_order = max_order; spin_lock_init(&buddy->lock); - - buddy->bits = kzalloc((buddy->max_order + 1) * sizeof(long *), - GFP_KERNEL); - buddy->num_free = kzalloc((buddy->max_order + 1) * sizeof(int *), - GFP_KERNEL); + buddy->bits = kcalloc(buddy->max_order + 1, + sizeof(*buddy->bits), + GFP_KERNEL); + buddy->num_free = kcalloc(buddy->max_order + 1, + sizeof(*buddy->num_free), + GFP_KERNEL); if (!buddy->bits || !buddy->num_free) goto err_out; @@ -503,7 +504,8 @@ int hns_roce_ib_umem_write_mtt(struct hns_roce_dev *hr_dev, for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) { len = sg_dma_len(sg) >> mtt->page_shift; for (k = 0; k < len; ++k) { - pages[i++] = sg_dma_address(sg) + umem->page_size * k; + pages[i++] = sg_dma_address(sg) + + (k << umem->page_shift); if (i == PAGE_SIZE / sizeof(u64)) { ret = hns_roce_write_mtt(hr_dev, mtt, n, i, pages); @@ -563,9 +565,9 @@ struct ib_mr *hns_roce_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, } n = ib_umem_page_count(mr->umem); - if (mr->umem->page_size != HNS_ROCE_HEM_PAGE_SIZE) { - dev_err(dev, "Just support 4K page size but is 0x%x now!\n", - mr->umem->page_size); + if (mr->umem->page_shift != HNS_ROCE_HEM_PAGE_SHIFT) { + dev_err(dev, "Just support 4K page size but is 0x%lx now!\n", + BIT(mr->umem->page_shift)); ret = -EINVAL; goto err_umem; } diff --git a/drivers/infiniband/hw/hns/hns_roce_qp.c b/drivers/infiniband/hw/hns/hns_roce_qp.c index 3f44f2f91f03..054c52699090 100644 --- a/drivers/infiniband/hw/hns/hns_roce_qp.c +++ b/drivers/infiniband/hw/hns/hns_roce_qp.c @@ -437,8 +437,7 @@ static int hns_roce_create_qp_common(struct hns_roce_dev *hr_dev, } ret = hns_roce_mtt_init(hr_dev, ib_umem_page_count(hr_qp->umem), - ilog2((unsigned int)hr_qp->umem->page_size), - &hr_qp->mtt); + hr_qp->umem->page_shift, &hr_qp->mtt); if (ret) { dev_err(dev, "hns_roce_mtt_init error for create qp\n"); goto err_buf; diff --git a/drivers/infiniband/hw/i40iw/i40iw_cm.c b/drivers/infiniband/hw/i40iw/i40iw_cm.c index 95a0586a4da8..f3bc01bce483 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_cm.c +++ b/drivers/infiniband/hw/i40iw/i40iw_cm.c @@ -3184,9 +3184,8 @@ void i40iw_setup_cm_core(struct i40iw_device *iwdev) INIT_LIST_HEAD(&cm_core->connected_nodes); INIT_LIST_HEAD(&cm_core->listen_nodes); - init_timer(&cm_core->tcp_timer); - cm_core->tcp_timer.function = i40iw_cm_timer_tick; - cm_core->tcp_timer.data = (unsigned long)cm_core; + setup_timer(&cm_core->tcp_timer, i40iw_cm_timer_tick, + (unsigned long)cm_core); spin_lock_init(&cm_core->ht_lock); spin_lock_init(&cm_core->listen_list_lock); diff --git a/drivers/infiniband/hw/i40iw/i40iw_utils.c b/drivers/infiniband/hw/i40iw/i40iw_utils.c index 70c3e9e79508..409a3781e735 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_utils.c +++ b/drivers/infiniband/hw/i40iw/i40iw_utils.c @@ -844,10 +844,9 @@ void i40iw_terminate_start_timer(struct i40iw_sc_qp *qp) iwqp = (struct i40iw_qp *)qp->back_qp; i40iw_add_ref(&iwqp->ibqp); - init_timer(&iwqp->terminate_timer); - iwqp->terminate_timer.function = i40iw_terminate_timeout; + setup_timer(&iwqp->terminate_timer, i40iw_terminate_timeout, + (unsigned long)iwqp); iwqp->terminate_timer.expires = jiffies + HZ; - iwqp->terminate_timer.data = (unsigned long)iwqp; add_timer(&iwqp->terminate_timer); } @@ -1436,9 +1435,8 @@ void i40iw_hw_stats_start_timer(struct i40iw_sc_vsi *vsi) { struct i40iw_vsi_pestat *devstat = vsi->pestat; - init_timer(&devstat->stats_timer); - devstat->stats_timer.function = i40iw_hw_stats_timeout; - devstat->stats_timer.data = (unsigned long)vsi; + setup_timer(&devstat->stats_timer, i40iw_hw_stats_timeout, + (unsigned long)vsi); mod_timer(&devstat->stats_timer, jiffies + msecs_to_jiffies(STATS_TIMER_DELAY)); } diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.c b/drivers/infiniband/hw/i40iw/i40iw_verbs.c index 9b2849979756..4dbe61ec7a77 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_verbs.c +++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.c @@ -1345,7 +1345,7 @@ static void i40iw_copy_user_pgaddrs(struct i40iw_mr *iwmr, { struct ib_umem *region = iwmr->region; struct i40iw_pbl *iwpbl = &iwmr->iwpbl; - int chunk_pages, entry, pg_shift, i; + int chunk_pages, entry, i; struct i40iw_pble_alloc *palloc = &iwpbl->pble_alloc; struct i40iw_pble_info *pinfo; struct scatterlist *sg; @@ -1354,14 +1354,14 @@ static void i40iw_copy_user_pgaddrs(struct i40iw_mr *iwmr, pinfo = (level == I40IW_LEVEL_1) ? NULL : palloc->level2.leaf; - pg_shift = ffs(region->page_size) - 1; for_each_sg(region->sg_head.sgl, sg, region->nmap, entry) { - chunk_pages = sg_dma_len(sg) >> pg_shift; + chunk_pages = sg_dma_len(sg) >> region->page_shift; if ((iwmr->type == IW_MEMREG_TYPE_QP) && !iwpbl->qp_mr.sq_page) iwpbl->qp_mr.sq_page = sg_page(sg); for (i = 0; i < chunk_pages; i++) { - pg_addr = sg_dma_address(sg) + region->page_size * i; + pg_addr = sg_dma_address(sg) + + (i << region->page_shift); if ((entry + i) == 0) *pbl = cpu_to_le64(pg_addr & iwmr->page_msk); @@ -1847,7 +1847,7 @@ static struct ib_mr *i40iw_reg_user_mr(struct ib_pd *pd, iwmr->ibmr.device = pd->device; ucontext = to_ucontext(pd->uobject->context); - iwmr->page_size = region->page_size; + iwmr->page_size = PAGE_SIZE; iwmr->page_msk = PAGE_MASK; if (region->hugetlb && (req.reg_type == IW_MEMREG_TYPE_MEM)) @@ -2696,7 +2696,7 @@ static int i40iw_query_pkey(struct ib_device *ibdev, * @ah_attr: address handle attributes */ static struct ib_ah *i40iw_create_ah(struct ib_pd *ibpd, - struct ib_ah_attr *attr, + struct rdma_ah_attr *attr, struct ib_udata *udata) { diff --git a/drivers/infiniband/hw/mlx4/ah.c b/drivers/infiniband/hw/mlx4/ah.c index 077c33d2dc75..538c46a73248 100644 --- a/drivers/infiniband/hw/mlx4/ah.c +++ b/drivers/infiniband/hw/mlx4/ah.c @@ -40,36 +40,45 @@ #include "mlx4_ib.h" -static struct ib_ah *create_ib_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, +static struct ib_ah *create_ib_ah(struct ib_pd *pd, + struct rdma_ah_attr *ah_attr, struct mlx4_ib_ah *ah) { struct mlx4_dev *dev = to_mdev(pd->device)->dev; - ah->av.ib.port_pd = cpu_to_be32(to_mpd(pd)->pdn | (ah_attr->port_num << 24)); - ah->av.ib.g_slid = ah_attr->src_path_bits; - ah->av.ib.sl_tclass_flowlabel = cpu_to_be32(ah_attr->sl << 28); - if (ah_attr->ah_flags & IB_AH_GRH) { + ah->av.ib.port_pd = cpu_to_be32(to_mpd(pd)->pdn | + (rdma_ah_get_port_num(ah_attr) << 24)); + ah->av.ib.g_slid = rdma_ah_get_path_bits(ah_attr); + ah->av.ib.sl_tclass_flowlabel = + cpu_to_be32(rdma_ah_get_sl(ah_attr) << 28); + if (rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH) { + const struct ib_global_route *grh = rdma_ah_read_grh(ah_attr); + ah->av.ib.g_slid |= 0x80; - ah->av.ib.gid_index = ah_attr->grh.sgid_index; - ah->av.ib.hop_limit = ah_attr->grh.hop_limit; + ah->av.ib.gid_index = grh->sgid_index; + ah->av.ib.hop_limit = grh->hop_limit; ah->av.ib.sl_tclass_flowlabel |= - cpu_to_be32((ah_attr->grh.traffic_class << 20) | - ah_attr->grh.flow_label); - memcpy(ah->av.ib.dgid, ah_attr->grh.dgid.raw, 16); + cpu_to_be32((grh->traffic_class << 20) | + grh->flow_label); + memcpy(ah->av.ib.dgid, grh->dgid.raw, 16); } - ah->av.ib.dlid = cpu_to_be16(ah_attr->dlid); - if (ah_attr->static_rate) { - ah->av.ib.stat_rate = ah_attr->static_rate + MLX4_STAT_RATE_OFFSET; - while (ah->av.ib.stat_rate > IB_RATE_2_5_GBPS + MLX4_STAT_RATE_OFFSET && - !(1 << ah->av.ib.stat_rate & dev->caps.stat_rate_support)) - --ah->av.ib.stat_rate; + ah->av.ib.dlid = cpu_to_be16(rdma_ah_get_dlid(ah_attr)); + if (rdma_ah_get_static_rate(ah_attr)) { + u8 static_rate = rdma_ah_get_static_rate(ah_attr) + + MLX4_STAT_RATE_OFFSET; + + while (static_rate > IB_RATE_2_5_GBPS + MLX4_STAT_RATE_OFFSET && + !(1 << static_rate & dev->caps.stat_rate_support)) + --static_rate; + ah->av.ib.stat_rate = static_rate; } return &ah->ibah; } -static struct ib_ah *create_iboe_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, +static struct ib_ah *create_iboe_ah(struct ib_pd *pd, + struct rdma_ah_attr *ah_attr, struct mlx4_ib_ah *ah) { struct mlx4_ib_dev *ibdev = to_mdev(pd->device); @@ -79,17 +88,18 @@ static struct ib_ah *create_iboe_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr u16 vlan_tag = 0xffff; union ib_gid sgid; struct ib_gid_attr gid_attr; + const struct ib_global_route *grh = rdma_ah_read_grh(ah_attr); int ret; - memcpy(&in6, ah_attr->grh.dgid.raw, sizeof(in6)); + memcpy(&in6, grh->dgid.raw, sizeof(in6)); if (rdma_is_multicast_addr(&in6)) { is_mcast = 1; rdma_get_mcast_mac(&in6, ah->av.eth.mac); } else { - memcpy(ah->av.eth.mac, ah_attr->dmac, ETH_ALEN); + memcpy(ah->av.eth.mac, ah_attr->roce.dmac, ETH_ALEN); } - ret = ib_get_cached_gid(pd->device, ah_attr->port_num, - ah_attr->grh.sgid_index, &sgid, &gid_attr); + ret = ib_get_cached_gid(pd->device, rdma_ah_get_port_num(ah_attr), + grh->sgid_index, &sgid, &gid_attr); if (ret) return ERR_PTR(ret); eth_zero_addr(ah->av.eth.s_mac); @@ -100,36 +110,40 @@ static struct ib_ah *create_iboe_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr dev_put(gid_attr.ndev); } if (vlan_tag < 0x1000) - vlan_tag |= (ah_attr->sl & 7) << 13; - ah->av.eth.port_pd = cpu_to_be32(to_mpd(pd)->pdn | (ah_attr->port_num << 24)); - ret = mlx4_ib_gid_index_to_real_index(ibdev, ah_attr->port_num, ah_attr->grh.sgid_index); + vlan_tag |= (rdma_ah_get_sl(ah_attr) & 7) << 13; + ah->av.eth.port_pd = cpu_to_be32(to_mpd(pd)->pdn | + (rdma_ah_get_port_num(ah_attr) << 24)); + ret = mlx4_ib_gid_index_to_real_index(ibdev, + rdma_ah_get_port_num(ah_attr), + grh->sgid_index); if (ret < 0) return ERR_PTR(ret); ah->av.eth.gid_index = ret; ah->av.eth.vlan = cpu_to_be16(vlan_tag); - ah->av.eth.hop_limit = ah_attr->grh.hop_limit; - if (ah_attr->static_rate) { - ah->av.eth.stat_rate = ah_attr->static_rate + MLX4_STAT_RATE_OFFSET; + ah->av.eth.hop_limit = grh->hop_limit; + if (rdma_ah_get_static_rate(ah_attr)) { + ah->av.eth.stat_rate = rdma_ah_get_static_rate(ah_attr) + + MLX4_STAT_RATE_OFFSET; while (ah->av.eth.stat_rate > IB_RATE_2_5_GBPS + MLX4_STAT_RATE_OFFSET && !(1 << ah->av.eth.stat_rate & dev->caps.stat_rate_support)) --ah->av.eth.stat_rate; } ah->av.eth.sl_tclass_flowlabel |= - cpu_to_be32((ah_attr->grh.traffic_class << 20) | - ah_attr->grh.flow_label); + cpu_to_be32((grh->traffic_class << 20) | + grh->flow_label); /* * HW requires multicast LID so we just choose one. */ if (is_mcast) ah->av.ib.dlid = cpu_to_be16(0xc000); - memcpy(ah->av.eth.dgid, ah_attr->grh.dgid.raw, 16); - ah->av.eth.sl_tclass_flowlabel |= cpu_to_be32(ah_attr->sl << 29); - + memcpy(ah->av.eth.dgid, grh->dgid.raw, 16); + ah->av.eth.sl_tclass_flowlabel |= cpu_to_be32(rdma_ah_get_sl(ah_attr) + << 29); return &ah->ibah; } -struct ib_ah *mlx4_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, +struct ib_ah *mlx4_ib_create_ah(struct ib_pd *pd, struct rdma_ah_attr *ah_attr, struct ib_udata *udata) { @@ -140,8 +154,8 @@ struct ib_ah *mlx4_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, if (!ah) return ERR_PTR(-ENOMEM); - if (rdma_port_get_link_layer(pd->device, ah_attr->port_num) == IB_LINK_LAYER_ETHERNET) { - if (!(ah_attr->ah_flags & IB_AH_GRH)) { + if (ah_attr->type == RDMA_AH_ATTR_TYPE_ROCE) { + if (!(rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH)) { ret = ERR_PTR(-EINVAL); } else { /* @@ -163,34 +177,40 @@ struct ib_ah *mlx4_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, return create_ib_ah(pd, ah_attr, ah); /* never fails */ } -int mlx4_ib_query_ah(struct ib_ah *ibah, struct ib_ah_attr *ah_attr) +int mlx4_ib_query_ah(struct ib_ah *ibah, struct rdma_ah_attr *ah_attr) { struct mlx4_ib_ah *ah = to_mah(ibah); - enum rdma_link_layer ll; + int port_num = be32_to_cpu(ah->av.ib.port_pd) >> 24; memset(ah_attr, 0, sizeof *ah_attr); - ah_attr->port_num = be32_to_cpu(ah->av.ib.port_pd) >> 24; - ll = rdma_port_get_link_layer(ibah->device, ah_attr->port_num); - if (ll == IB_LINK_LAYER_ETHERNET) - ah_attr->sl = be32_to_cpu(ah->av.eth.sl_tclass_flowlabel) >> 29; - else - ah_attr->sl = be32_to_cpu(ah->av.ib.sl_tclass_flowlabel) >> 28; - - ah_attr->dlid = ll == IB_LINK_LAYER_INFINIBAND ? be16_to_cpu(ah->av.ib.dlid) : 0; - if (ah->av.ib.stat_rate) - ah_attr->static_rate = ah->av.ib.stat_rate - MLX4_STAT_RATE_OFFSET; - ah_attr->src_path_bits = ah->av.ib.g_slid & 0x7F; + ah_attr->type = ibah->type; + if (ah_attr->type == RDMA_AH_ATTR_TYPE_ROCE) { + rdma_ah_set_dlid(ah_attr, 0); + rdma_ah_set_sl(ah_attr, + be32_to_cpu(ah->av.eth.sl_tclass_flowlabel) + >> 29); + } else { + rdma_ah_set_dlid(ah_attr, be16_to_cpu(ah->av.ib.dlid)); + rdma_ah_set_sl(ah_attr, + be32_to_cpu(ah->av.ib.sl_tclass_flowlabel) + >> 28); + } + + rdma_ah_set_port_num(ah_attr, port_num); + if (ah->av.ib.stat_rate) + rdma_ah_set_static_rate(ah_attr, + ah->av.ib.stat_rate - + MLX4_STAT_RATE_OFFSET); + rdma_ah_set_path_bits(ah_attr, ah->av.ib.g_slid & 0x7F); if (mlx4_ib_ah_grh_present(ah)) { - ah_attr->ah_flags = IB_AH_GRH; - - ah_attr->grh.traffic_class = - be32_to_cpu(ah->av.ib.sl_tclass_flowlabel) >> 20; - ah_attr->grh.flow_label = - be32_to_cpu(ah->av.ib.sl_tclass_flowlabel) & 0xfffff; - ah_attr->grh.hop_limit = ah->av.ib.hop_limit; - ah_attr->grh.sgid_index = ah->av.ib.gid_index; - memcpy(ah_attr->grh.dgid.raw, ah->av.ib.dgid, 16); + u32 tc_fl = be32_to_cpu(ah->av.ib.sl_tclass_flowlabel); + + rdma_ah_set_grh(ah_attr, NULL, + tc_fl & 0xfffff, ah->av.ib.gid_index, + ah->av.ib.hop_limit, + tc_fl >> 20); + rdma_ah_set_dgid_raw(ah_attr, ah->av.ib.dgid); } return 0; diff --git a/drivers/infiniband/hw/mlx4/cm.c b/drivers/infiniband/hw/mlx4/cm.c index d64845335e87..1e6c526450d9 100644 --- a/drivers/infiniband/hw/mlx4/cm.c +++ b/drivers/infiniband/hw/mlx4/cm.c @@ -279,14 +279,14 @@ id_map_alloc(struct ib_device *ibdev, int slave_id, u32 sl_cm_id) } static struct id_map_entry * -id_map_get(struct ib_device *ibdev, int *pv_cm_id, int sl_cm_id, int slave_id) +id_map_get(struct ib_device *ibdev, int *pv_cm_id, int slave_id, int sl_cm_id) { struct id_map_entry *ent; struct mlx4_ib_sriov *sriov = &to_mdev(ibdev)->sriov; spin_lock(&sriov->id_map_lock); if (*pv_cm_id == -1) { - ent = id_map_find_by_sl_id(ibdev, sl_cm_id, slave_id); + ent = id_map_find_by_sl_id(ibdev, slave_id, sl_cm_id); if (ent) *pv_cm_id = (int) ent->pv_cm_id; } else @@ -414,7 +414,7 @@ void mlx4_ib_cm_paravirt_clean(struct mlx4_ib_dev *dev, int slave) struct rb_root *sl_id_map = &sriov->sl_id_map; struct list_head lh; struct rb_node *nd; - int need_flush = 1; + int need_flush = 0; struct id_map_entry *map, *tmp_map; /* cancel all delayed work queue entries */ INIT_LIST_HEAD(&lh); @@ -422,13 +422,13 @@ void mlx4_ib_cm_paravirt_clean(struct mlx4_ib_dev *dev, int slave) list_for_each_entry_safe(map, tmp_map, &dev->sriov.cm_list, list) { if (slave < 0 || slave == map->slave_id) { if (map->scheduled_delete) - need_flush &= !!cancel_delayed_work(&map->timeout); + need_flush |= !cancel_delayed_work(&map->timeout); } } spin_unlock(&sriov->id_map_lock); - if (!need_flush) + if (need_flush) flush_scheduled_work(); /* make sure all timers were flushed */ /* now, remove all leftover entries from databases*/ diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c index 6a0fec357dae..4f5a143fc0a7 100644 --- a/drivers/infiniband/hw/mlx4/cq.c +++ b/drivers/infiniband/hw/mlx4/cq.c @@ -147,7 +147,7 @@ static int mlx4_ib_get_cq_umem(struct mlx4_ib_dev *dev, struct ib_ucontext *cont return PTR_ERR(*umem); err = mlx4_mtt_init(dev->dev, ib_umem_page_count(*umem), - ilog2((*umem)->page_size), &buf->mtt); + (*umem)->page_shift, &buf->mtt); if (err) goto err_buf; diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c index db564ccc0f92..b4694717f6f3 100644 --- a/drivers/infiniband/hw/mlx4/mad.c +++ b/drivers/infiniband/hw/mlx4/mad.c @@ -189,25 +189,26 @@ int mlx4_MAD_IFC(struct mlx4_ib_dev *dev, int mad_ifc_flags, static void update_sm_ah(struct mlx4_ib_dev *dev, u8 port_num, u16 lid, u8 sl) { struct ib_ah *new_ah; - struct ib_ah_attr ah_attr; + struct rdma_ah_attr ah_attr; unsigned long flags; if (!dev->send_agent[port_num - 1][0]) return; memset(&ah_attr, 0, sizeof ah_attr); - ah_attr.dlid = lid; - ah_attr.sl = sl; - ah_attr.port_num = port_num; + ah_attr.type = rdma_ah_find_type(&dev->ib_dev, port_num); + rdma_ah_set_dlid(&ah_attr, lid); + rdma_ah_set_sl(&ah_attr, sl); + rdma_ah_set_port_num(&ah_attr, port_num); - new_ah = ib_create_ah(dev->send_agent[port_num - 1][0]->qp->pd, - &ah_attr); + new_ah = rdma_create_ah(dev->send_agent[port_num - 1][0]->qp->pd, + &ah_attr); if (IS_ERR(new_ah)) return; spin_lock_irqsave(&dev->sm_lock, flags); if (dev->sm_ah[port_num - 1]) - ib_destroy_ah(dev->sm_ah[port_num - 1]); + rdma_destroy_ah(dev->sm_ah[port_num - 1]); dev->sm_ah[port_num - 1] = new_ah; spin_unlock_irqrestore(&dev->sm_lock, flags); } @@ -509,7 +510,7 @@ int mlx4_ib_send_to_slave(struct mlx4_ib_dev *dev, int slave, u8 port, struct mlx4_ib_demux_pv_ctx *tun_ctx; struct mlx4_ib_demux_pv_qp *tun_qp; struct mlx4_rcv_tunnel_mad *tun_mad; - struct ib_ah_attr attr; + struct rdma_ah_attr attr; struct ib_ah *ah; struct ib_qp *src_qp = NULL; unsigned tun_tx_ix = 0; @@ -555,15 +556,18 @@ int mlx4_ib_send_to_slave(struct mlx4_ib_dev *dev, int slave, u8 port, /* create ah. Just need an empty one with the port num for the post send. * The driver will set the force loopback bit in post_send */ memset(&attr, 0, sizeof attr); - attr.port_num = port; + attr.type = rdma_ah_find_type(&dev->ib_dev, port); + + rdma_ah_set_port_num(&attr, port); if (is_eth) { union ib_gid sgid; + union ib_gid dgid; - if (get_gids_from_l3_hdr(grh, &sgid, &attr.grh.dgid)) + if (get_gids_from_l3_hdr(grh, &sgid, &dgid)) return -EINVAL; - attr.ah_flags = IB_AH_GRH; + rdma_ah_set_grh(&attr, &dgid, 0, 0, 0, 0); } - ah = ib_create_ah(tun_ctx->pd, &attr); + ah = rdma_create_ah(tun_ctx->pd, &attr); if (IS_ERR(ah)) return -ENOMEM; @@ -580,7 +584,7 @@ int mlx4_ib_send_to_slave(struct mlx4_ib_dev *dev, int slave, u8 port, tun_mad = (struct mlx4_rcv_tunnel_mad *) (tun_qp->tx_ring[tun_tx_ix].buf.addr); if (tun_qp->tx_ring[tun_tx_ix].ah) - ib_destroy_ah(tun_qp->tx_ring[tun_tx_ix].ah); + rdma_destroy_ah(tun_qp->tx_ring[tun_tx_ix].ah); tun_qp->tx_ring[tun_tx_ix].ah = ah; ib_dma_sync_single_for_cpu(&dev->ib_dev, tun_qp->tx_ring[tun_tx_ix].buf.map, @@ -653,7 +657,7 @@ int mlx4_ib_send_to_slave(struct mlx4_ib_dev *dev, int slave, u8 port, spin_unlock(&tun_qp->tx_lock); tun_qp->tx_ring[tun_tx_ix].ah = NULL; end: - ib_destroy_ah(ah); + rdma_destroy_ah(ah); return ret; } @@ -1018,7 +1022,7 @@ static void send_handler(struct ib_mad_agent *agent, struct ib_mad_send_wc *mad_send_wc) { if (mad_send_wc->send_buf->context[0]) - ib_destroy_ah(mad_send_wc->send_buf->context[0]); + rdma_destroy_ah(mad_send_wc->send_buf->context[0]); ib_free_send_mad(mad_send_wc->send_buf); } @@ -1073,7 +1077,7 @@ void mlx4_ib_mad_cleanup(struct mlx4_ib_dev *dev) } if (dev->sm_ah[p]) - ib_destroy_ah(dev->sm_ah[p]); + rdma_destroy_ah(dev->sm_ah[p]); } } @@ -1352,7 +1356,7 @@ static int is_proxy_qp0(struct mlx4_ib_dev *dev, int qpn, int slave) int mlx4_ib_send_to_wire(struct mlx4_ib_dev *dev, int slave, u8 port, enum ib_qp_type dest_qpt, u16 pkey_index, - u32 remote_qpn, u32 qkey, struct ib_ah_attr *attr, + u32 remote_qpn, u32 qkey, struct rdma_ah_attr *attr, u8 *s_mac, u16 vlan_id, struct ib_mad *mad) { struct ib_sge list; @@ -1363,6 +1367,7 @@ int mlx4_ib_send_to_wire(struct mlx4_ib_dev *dev, int slave, u8 port, struct mlx4_mad_snd_buf *sqp_mad; struct ib_ah *ah; struct ib_qp *send_qp = NULL; + struct ib_global_route *grh; unsigned wire_tx_ix = 0; int ret = 0; u16 wire_pkey_ix; @@ -1389,12 +1394,13 @@ int mlx4_ib_send_to_wire(struct mlx4_ib_dev *dev, int slave, u8 port, send_qp = sqp->qp; /* create ah */ - sgid_index = attr->grh.sgid_index; - attr->grh.sgid_index = 0; - ah = ib_create_ah(sqp_ctx->pd, attr); + grh = rdma_ah_retrieve_grh(attr); + sgid_index = grh->sgid_index; + grh->sgid_index = 0; + ah = rdma_create_ah(sqp_ctx->pd, attr); if (IS_ERR(ah)) return -ENOMEM; - attr->grh.sgid_index = sgid_index; + grh->sgid_index = sgid_index; to_mah(ah)->av.ib.gid_index = sgid_index; /* get rid of force-loopback bit */ to_mah(ah)->av.ib.port_pd &= cpu_to_be32(0x7FFFFFFF); @@ -1410,7 +1416,7 @@ int mlx4_ib_send_to_wire(struct mlx4_ib_dev *dev, int slave, u8 port, sqp_mad = (struct mlx4_mad_snd_buf *) (sqp->tx_ring[wire_tx_ix].buf.addr); if (sqp->tx_ring[wire_tx_ix].ah) - ib_destroy_ah(sqp->tx_ring[wire_tx_ix].ah); + rdma_destroy_ah(sqp->tx_ring[wire_tx_ix].ah); sqp->tx_ring[wire_tx_ix].ah = ah; ib_dma_sync_single_for_cpu(&dev->ib_dev, sqp->tx_ring[wire_tx_ix].buf.map, @@ -1442,7 +1448,7 @@ int mlx4_ib_send_to_wire(struct mlx4_ib_dev *dev, int slave, u8 port, if (s_mac) memcpy(to_mah(ah)->av.eth.s_mac, s_mac, 6); if (vlan_id < 0x1000) - vlan_id |= (attr->sl & 7) << 13; + vlan_id |= (rdma_ah_get_sl(attr) & 7) << 13; to_mah(ah)->av.eth.vlan = cpu_to_be16(vlan_id); @@ -1455,7 +1461,7 @@ int mlx4_ib_send_to_wire(struct mlx4_ib_dev *dev, int slave, u8 port, spin_unlock(&sqp->tx_lock); sqp->tx_ring[wire_tx_ix].ah = NULL; out: - ib_destroy_ah(ah); + rdma_destroy_ah(ah); return ret; } @@ -1467,12 +1473,13 @@ static int get_slave_base_gid_ix(struct mlx4_ib_dev *dev, int slave, int port) } static void fill_in_real_sgid_index(struct mlx4_ib_dev *dev, int slave, int port, - struct ib_ah_attr *ah_attr) + struct rdma_ah_attr *ah_attr) { + struct ib_global_route *grh = rdma_ah_retrieve_grh(ah_attr); if (rdma_port_get_link_layer(&dev->ib_dev, port) == IB_LINK_LAYER_INFINIBAND) - ah_attr->grh.sgid_index = slave; + grh->sgid_index = slave; else - ah_attr->grh.sgid_index += get_slave_base_gid_ix(dev, slave, port); + grh->sgid_index += get_slave_base_gid_ix(dev, slave, port); } static void mlx4_ib_multiplex_mad(struct mlx4_ib_demux_pv_ctx *ctx, struct ib_wc *wc) @@ -1482,11 +1489,13 @@ static void mlx4_ib_multiplex_mad(struct mlx4_ib_demux_pv_ctx *ctx, struct ib_wc int wr_ix = wc->wr_id & (MLX4_NUM_TUNNEL_BUFS - 1); struct mlx4_tunnel_mad *tunnel = tun_qp->ring[wr_ix].addr; struct mlx4_ib_ah ah; - struct ib_ah_attr ah_attr; + struct rdma_ah_attr ah_attr; u8 *slave_id; int slave; int port; u16 vlan_id; + u8 qos; + u8 *dmac; /* Get slave that sent this packet */ if (wc->src_qp < dev->dev->phys_caps.base_proxy_sqpn || @@ -1571,14 +1580,16 @@ static void mlx4_ib_multiplex_mad(struct mlx4_ib_demux_pv_ctx *ctx, struct ib_wc ah.av.ib.port_pd = cpu_to_be32(port << 24 | (be32_to_cpu(ah.av.ib.port_pd) & 0xffffff)); mlx4_ib_query_ah(&ah.ibah, &ah_attr); - if (ah_attr.ah_flags & IB_AH_GRH) + if (rdma_ah_get_ah_flags(&ah_attr) & IB_AH_GRH) fill_in_real_sgid_index(dev, slave, ctx->port, &ah_attr); - - memcpy(ah_attr.dmac, tunnel->hdr.mac, 6); + dmac = rdma_ah_retrieve_dmac(&ah_attr); + if (dmac) + memcpy(dmac, tunnel->hdr.mac, ETH_ALEN); vlan_id = be16_to_cpu(tunnel->hdr.vlan); /* if slave have default vlan use it */ - mlx4_get_slave_default_vlan(dev->dev, ctx->port, slave, - &vlan_id, &ah_attr.sl); + if (mlx4_get_slave_default_vlan(dev->dev, ctx->port, slave, + &vlan_id, &qos)) + rdma_ah_set_sl(&ah_attr, qos); mlx4_ib_send_to_wire(dev, slave, ctx->port, is_proxy_qp0(dev, wc->src_qp, slave) ? @@ -1714,7 +1725,7 @@ static void mlx4_ib_free_pv_qp_bufs(struct mlx4_ib_demux_pv_ctx *ctx, tx_buf_size, DMA_TO_DEVICE); kfree(tun_qp->tx_ring[i].buf.addr); if (tun_qp->tx_ring[i].ah) - ib_destroy_ah(tun_qp->tx_ring[i].ah); + rdma_destroy_ah(tun_qp->tx_ring[i].ah); } kfree(tun_qp->tx_ring); kfree(tun_qp->ring); @@ -1746,7 +1757,7 @@ static void mlx4_ib_tunnel_comp_worker(struct work_struct *work) pr_debug("received tunnel send completion:" "wrid=0x%llx, status=0x%x\n", wc.wr_id, wc.status); - ib_destroy_ah(tun_qp->tx_ring[wc.wr_id & + rdma_destroy_ah(tun_qp->tx_ring[wc.wr_id & (MLX4_NUM_TUNNEL_BUFS - 1)].ah); tun_qp->tx_ring[wc.wr_id & (MLX4_NUM_TUNNEL_BUFS - 1)].ah = NULL; @@ -1763,7 +1774,7 @@ static void mlx4_ib_tunnel_comp_worker(struct work_struct *work) " status = %d, wrid = 0x%llx\n", ctx->slave, wc.status, wc.wr_id); if (!MLX4_TUN_IS_RECV(wc.wr_id)) { - ib_destroy_ah(tun_qp->tx_ring[wc.wr_id & + rdma_destroy_ah(tun_qp->tx_ring[wc.wr_id & (MLX4_NUM_TUNNEL_BUFS - 1)].ah); tun_qp->tx_ring[wc.wr_id & (MLX4_NUM_TUNNEL_BUFS - 1)].ah = NULL; @@ -1900,7 +1911,7 @@ static void mlx4_ib_sqp_comp_worker(struct work_struct *work) if (wc.status == IB_WC_SUCCESS) { switch (wc.opcode) { case IB_WC_SEND: - ib_destroy_ah(sqp->tx_ring[wc.wr_id & + rdma_destroy_ah(sqp->tx_ring[wc.wr_id & (MLX4_NUM_TUNNEL_BUFS - 1)].ah); sqp->tx_ring[wc.wr_id & (MLX4_NUM_TUNNEL_BUFS - 1)].ah = NULL; @@ -1930,7 +1941,7 @@ static void mlx4_ib_sqp_comp_worker(struct work_struct *work) " status = %d, wrid = 0x%llx\n", ctx->slave, wc.status, wc.wr_id); if (!MLX4_TUN_IS_RECV(wc.wr_id)) { - ib_destroy_ah(sqp->tx_ring[wc.wr_id & + rdma_destroy_ah(sqp->tx_ring[wc.wr_id & (MLX4_NUM_TUNNEL_BUFS - 1)].ah); sqp->tx_ring[wc.wr_id & (MLX4_NUM_TUNNEL_BUFS - 1)].ah = NULL; diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index fba94df28cf1..521d0def2d9e 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -1173,7 +1173,7 @@ static void mlx4_ib_disassociate_ucontext(struct ib_ucontext *ibcontext) /* need to protect from a race on closing the vma as part of * mlx4_ib_vma_close(). */ - down_read(&owning_mm->mmap_sem); + down_write(&owning_mm->mmap_sem); for (i = 0; i < HW_BAR_COUNT; i++) { vma = context->hw_bar_info[i].vma; if (!vma) @@ -1187,11 +1187,13 @@ static void mlx4_ib_disassociate_ucontext(struct ib_ucontext *ibcontext) BUG_ON(1); } + context->hw_bar_info[i].vma->vm_flags &= + ~(VM_SHARED | VM_MAYSHARE); /* context going to be destroyed, should not access ops any more */ context->hw_bar_info[i].vma->vm_ops = NULL; } - up_read(&owning_mm->mmap_sem); + up_write(&owning_mm->mmap_sem); mmput(owning_mm); put_task_struct(owning_process); } @@ -2867,23 +2869,19 @@ static void *mlx4_ib_add(struct mlx4_dev *dev) if (mlx4_ib_init_sriov(ibdev)) goto err_mad; - if (dev->caps.flags & MLX4_DEV_CAP_FLAG_IBOE || - dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ROCE_V1_V2) { - if (!iboe->nb.notifier_call) { - iboe->nb.notifier_call = mlx4_ib_netdev_event; - err = register_netdevice_notifier(&iboe->nb); - if (err) { - iboe->nb.notifier_call = NULL; - goto err_notif; - } - } - if (dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ROCE_V1_V2) { - err = mlx4_config_roce_v2_port(dev, ROCE_V2_UDP_DPORT); - if (err) { - goto err_notif; - } + if (!iboe->nb.notifier_call) { + iboe->nb.notifier_call = mlx4_ib_netdev_event; + err = register_netdevice_notifier(&iboe->nb); + if (err) { + iboe->nb.notifier_call = NULL; + goto err_notif; } } + if (dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ROCE_V1_V2) { + err = mlx4_config_roce_v2_port(dev, ROCE_V2_UDP_DPORT); + if (err) + goto err_notif; + } for (j = 0; j < ARRAY_SIZE(mlx4_class_attributes); ++j) { if (device_create_file(&ibdev->ib_dev.dev, @@ -2941,6 +2939,7 @@ err_counter: mlx4_ib_delete_counters_table(ibdev, &ibdev->counters_table[i]); err_map: + mlx4_ib_free_eqs(dev, ibdev); iounmap(ibdev->uar_map); err_uar: diff --git a/drivers/infiniband/hw/mlx4/mcg.c b/drivers/infiniband/hw/mlx4/mcg.c index e010fe459e67..3405e947dc1e 100644 --- a/drivers/infiniband/hw/mlx4/mcg.c +++ b/drivers/infiniband/hw/mlx4/mcg.c @@ -209,7 +209,7 @@ static struct mcast_group *mcast_insert(struct mlx4_ib_demux_ctx *ctx, static int send_mad_to_wire(struct mlx4_ib_demux_ctx *ctx, struct ib_mad *mad) { struct mlx4_ib_dev *dev = ctx->dev; - struct ib_ah_attr ah_attr; + struct rdma_ah_attr ah_attr; unsigned long flags; spin_lock_irqsave(&dev->sm_lock, flags); @@ -231,20 +231,20 @@ static int send_mad_to_slave(int slave, struct mlx4_ib_demux_ctx *ctx, struct mlx4_ib_dev *dev = ctx->dev; struct ib_mad_agent *agent = dev->send_agent[ctx->port - 1][1]; struct ib_wc wc; - struct ib_ah_attr ah_attr; + struct rdma_ah_attr ah_attr; /* Our agent might not yet be registered when mads start to arrive */ if (!agent) return -EAGAIN; - ib_query_ah(dev->sm_ah[ctx->port - 1], &ah_attr); + rdma_query_ah(dev->sm_ah[ctx->port - 1], &ah_attr); if (ib_find_cached_pkey(&dev->ib_dev, ctx->port, IB_DEFAULT_PKEY_FULL, &wc.pkey_index)) return -EINVAL; wc.sl = 0; wc.dlid_path_bits = 0; wc.port_num = ctx->port; - wc.slid = ah_attr.dlid; /* opensm lid */ + wc.slid = rdma_ah_get_dlid(&ah_attr); /* opensm lid */ wc.src_qp = 1; return mlx4_ib_send_to_slave(dev, slave, ctx->port, IB_QPT_GSI, &wc, NULL, mad); } @@ -1102,7 +1102,8 @@ static void _mlx4_ib_mcg_port_cleanup(struct mlx4_ib_demux_ctx *ctx, int destroy while ((p = rb_first(&ctx->mcg_table)) != NULL) { group = rb_entry(p, struct mcast_group, node); if (atomic_read(&group->refcount)) - mcg_warn_group(group, "group refcount %d!!! (pointer %p)\n", atomic_read(&group->refcount), group); + mcg_debug_group(group, "group refcount %d!!! (pointer %p)\n", + atomic_read(&group->refcount), group); force_clean_group(group); } diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h index 64fed44b43a6..c2b9cbf4da05 100644 --- a/drivers/infiniband/hw/mlx4/mlx4_ib.h +++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h @@ -742,9 +742,9 @@ int mlx4_ib_arm_cq(struct ib_cq *cq, enum ib_cq_notify_flags flags); void __mlx4_ib_cq_clean(struct mlx4_ib_cq *cq, u32 qpn, struct mlx4_ib_srq *srq); void mlx4_ib_cq_clean(struct mlx4_ib_cq *cq, u32 qpn, struct mlx4_ib_srq *srq); -struct ib_ah *mlx4_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, +struct ib_ah *mlx4_ib_create_ah(struct ib_pd *pd, struct rdma_ah_attr *ah_attr, struct ib_udata *udata); -int mlx4_ib_query_ah(struct ib_ah *ibah, struct ib_ah_attr *ah_attr); +int mlx4_ib_query_ah(struct ib_ah *ibah, struct rdma_ah_attr *ah_attr); int mlx4_ib_destroy_ah(struct ib_ah *ah); struct ib_srq *mlx4_ib_create_srq(struct ib_pd *pd, @@ -833,7 +833,7 @@ int mlx4_ib_send_to_slave(struct mlx4_ib_dev *dev, int slave, u8 port, int mlx4_ib_send_to_wire(struct mlx4_ib_dev *dev, int slave, u8 port, enum ib_qp_type dest_qpt, u16 pkey_index, u32 remote_qpn, - u32 qkey, struct ib_ah_attr *attr, u8 *s_mac, + u32 qkey, struct rdma_ah_attr *attr, u8 *s_mac, u16 vlan_id, struct ib_mad *mad); __be64 mlx4_ib_get_new_demux_tid(struct mlx4_ib_demux_ctx *ctx); diff --git a/drivers/infiniband/hw/mlx4/mr.c b/drivers/infiniband/hw/mlx4/mr.c index 433bcdbdd680..e6f77f63da75 100644 --- a/drivers/infiniband/hw/mlx4/mr.c +++ b/drivers/infiniband/hw/mlx4/mr.c @@ -107,7 +107,7 @@ int mlx4_ib_umem_write_mtt(struct mlx4_ib_dev *dev, struct mlx4_mtt *mtt, len = sg_dma_len(sg) >> mtt->page_shift; for (k = 0; k < len; ++k) { pages[i++] = sg_dma_address(sg) + - umem->page_size * k; + (k << umem->page_shift); /* * Be friendly to mlx4_write_mtt() and * pass it chunks of appropriate size. @@ -155,7 +155,7 @@ struct ib_mr *mlx4_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, } n = ib_umem_page_count(mr->umem); - shift = ilog2(mr->umem->page_size); + shift = mr->umem->page_shift; err = mlx4_mr_alloc(dev->dev, to_mpd(pd)->pdn, virt_addr, length, convert_access(access_flags), n, shift, &mr->mmr); @@ -239,7 +239,7 @@ int mlx4_ib_rereg_user_mr(struct ib_mr *mr, int flags, goto release_mpt_entry; } n = ib_umem_page_count(mmr->umem); - shift = ilog2(mmr->umem->page_size); + shift = mmr->umem->page_shift; err = mlx4_mr_rereg_mem_write(dev->dev, &mmr->mmr, virt_addr, length, n, shift, diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c index c34eebc7db65..996e9058e515 100644 --- a/drivers/infiniband/hw/mlx4/qp.c +++ b/drivers/infiniband/hw/mlx4/qp.c @@ -745,7 +745,7 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, } err = mlx4_mtt_init(dev->dev, ib_umem_page_count(qp->umem), - ilog2(qp->umem->page_size), &qp->mtt); + qp->umem->page_shift, &qp->mtt); if (err) goto err_buf; @@ -1383,31 +1383,31 @@ static void mlx4_set_sched(struct mlx4_qp_path *path, u8 port) path->sched_queue = (path->sched_queue & 0xbf) | ((port - 1) << 6); } -static int _mlx4_set_path(struct mlx4_ib_dev *dev, const struct ib_ah_attr *ah, +static int _mlx4_set_path(struct mlx4_ib_dev *dev, + const struct rdma_ah_attr *ah, u64 smac, u16 vlan_tag, struct mlx4_qp_path *path, struct mlx4_roce_smac_vlan_info *smac_info, u8 port) { - int is_eth = rdma_port_get_link_layer(&dev->ib_dev, port) == - IB_LINK_LAYER_ETHERNET; int vidx; int smac_index; int err; - - path->grh_mylmc = ah->src_path_bits & 0x7f; - path->rlid = cpu_to_be16(ah->dlid); - if (ah->static_rate) { - path->static_rate = ah->static_rate + MLX4_STAT_RATE_OFFSET; + path->grh_mylmc = rdma_ah_get_path_bits(ah) & 0x7f; + path->rlid = cpu_to_be16(rdma_ah_get_dlid(ah)); + if (rdma_ah_get_static_rate(ah)) { + path->static_rate = rdma_ah_get_static_rate(ah) + + MLX4_STAT_RATE_OFFSET; while (path->static_rate > IB_RATE_2_5_GBPS + MLX4_STAT_RATE_OFFSET && !(1 << path->static_rate & dev->dev->caps.stat_rate_support)) --path->static_rate; } else path->static_rate = 0; - if (ah->ah_flags & IB_AH_GRH) { - int real_sgid_index = mlx4_ib_gid_index_to_real_index(dev, - port, - ah->grh.sgid_index); + if (rdma_ah_get_ah_flags(ah) & IB_AH_GRH) { + const struct ib_global_route *grh = rdma_ah_read_grh(ah); + int real_sgid_index = + mlx4_ib_gid_index_to_real_index(dev, port, + grh->sgid_index); if (real_sgid_index >= dev->dev->caps.gid_table_len[port]) { pr_err("sgid_index (%u) too large. max is %d\n", @@ -1417,19 +1417,19 @@ static int _mlx4_set_path(struct mlx4_ib_dev *dev, const struct ib_ah_attr *ah, path->grh_mylmc |= 1 << 7; path->mgid_index = real_sgid_index; - path->hop_limit = ah->grh.hop_limit; + path->hop_limit = grh->hop_limit; path->tclass_flowlabel = - cpu_to_be32((ah->grh.traffic_class << 20) | - (ah->grh.flow_label)); - memcpy(path->rgid, ah->grh.dgid.raw, 16); + cpu_to_be32((grh->traffic_class << 20) | + (grh->flow_label)); + memcpy(path->rgid, grh->dgid.raw, 16); } - if (is_eth) { - if (!(ah->ah_flags & IB_AH_GRH)) + if (ah->type == RDMA_AH_ATTR_TYPE_ROCE) { + if (!(rdma_ah_get_ah_flags(ah) & IB_AH_GRH)) return -1; path->sched_queue = MLX4_IB_DEFAULT_SCHED_QUEUE | - ((port - 1) << 6) | ((ah->sl & 7) << 3); + ((port - 1) << 6) | ((rdma_ah_get_sl(ah) & 7) << 3); path->feup |= MLX4_FEUP_FORCE_ETH_UP; if (vlan_tag < 0x1000) { @@ -1488,14 +1488,13 @@ static int _mlx4_set_path(struct mlx4_ib_dev *dev, const struct ib_ah_attr *ah, } else { smac_index = smac_info->smac_index; } - - memcpy(path->dmac, ah->dmac, 6); + memcpy(path->dmac, ah->roce.dmac, 6); path->ackto = MLX4_IB_LINK_TYPE_ETH; /* put MAC table smac index for IBoE */ path->grh_mylmc = (u8) (smac_index) | 0x80; } else { path->sched_queue = MLX4_IB_DEFAULT_SCHED_QUEUE | - ((port - 1) << 6) | ((ah->sl & 0xf) << 2); + ((port - 1) << 6) | ((rdma_ah_get_sl(ah) & 0xf) << 2); } return 0; @@ -1767,11 +1766,13 @@ static int __mlx4_ib_modify_qp(struct ib_qp *ibqp, u16 vlan = 0xffff; u8 smac[ETH_ALEN]; int status = 0; - int is_eth = rdma_cap_eth_ah(&dev->ib_dev, port_num) && - attr->ah_attr.ah_flags & IB_AH_GRH; + int is_eth = + rdma_cap_eth_ah(&dev->ib_dev, port_num) && + rdma_ah_get_ah_flags(&attr->ah_attr) & IB_AH_GRH; - if (is_eth && attr->ah_attr.ah_flags & IB_AH_GRH) { - int index = attr->ah_attr.grh.sgid_index; + if (is_eth) { + int index = + rdma_ah_read_grh(&attr->ah_attr)->sgid_index; status = ib_get_cached_gid(ibqp->device, port_num, index, &gid, &gid_attr); @@ -3394,39 +3395,37 @@ static int to_ib_qp_access_flags(int mlx4_flags) return ib_flags; } -static void to_ib_ah_attr(struct mlx4_ib_dev *ibdev, struct ib_ah_attr *ib_ah_attr, - struct mlx4_qp_path *path) +static void to_rdma_ah_attr(struct mlx4_ib_dev *ibdev, + struct rdma_ah_attr *ah_attr, + struct mlx4_qp_path *path) { struct mlx4_dev *dev = ibdev->dev; - int is_eth; - - memset(ib_ah_attr, 0, sizeof *ib_ah_attr); - ib_ah_attr->port_num = path->sched_queue & 0x40 ? 2 : 1; + u8 port_num = path->sched_queue & 0x40 ? 2 : 1; - if (ib_ah_attr->port_num == 0 || ib_ah_attr->port_num > dev->caps.num_ports) + memset(ah_attr, 0, sizeof(*ah_attr)); + ah_attr->type = rdma_ah_find_type(&ibdev->ib_dev, port_num); + if (port_num == 0 || port_num > dev->caps.num_ports) return; - is_eth = rdma_port_get_link_layer(&ibdev->ib_dev, ib_ah_attr->port_num) == - IB_LINK_LAYER_ETHERNET; - if (is_eth) - ib_ah_attr->sl = ((path->sched_queue >> 3) & 0x7) | - ((path->sched_queue & 4) << 1); + if (ah_attr->type == RDMA_AH_ATTR_TYPE_ROCE) + rdma_ah_set_sl(ah_attr, ((path->sched_queue >> 3) & 0x7) | + ((path->sched_queue & 4) << 1)); else - ib_ah_attr->sl = (path->sched_queue >> 2) & 0xf; + rdma_ah_set_sl(ah_attr, (path->sched_queue >> 2) & 0xf); + rdma_ah_set_port_num(ah_attr, port_num); - ib_ah_attr->dlid = be16_to_cpu(path->rlid); - ib_ah_attr->src_path_bits = path->grh_mylmc & 0x7f; - ib_ah_attr->static_rate = path->static_rate ? path->static_rate - 5 : 0; - ib_ah_attr->ah_flags = (path->grh_mylmc & (1 << 7)) ? IB_AH_GRH : 0; - if (ib_ah_attr->ah_flags) { - ib_ah_attr->grh.sgid_index = path->mgid_index; - ib_ah_attr->grh.hop_limit = path->hop_limit; - ib_ah_attr->grh.traffic_class = - (be32_to_cpu(path->tclass_flowlabel) >> 20) & 0xff; - ib_ah_attr->grh.flow_label = - be32_to_cpu(path->tclass_flowlabel) & 0xfffff; - memcpy(ib_ah_attr->grh.dgid.raw, - path->rgid, sizeof ib_ah_attr->grh.dgid.raw); + rdma_ah_set_dlid(ah_attr, be16_to_cpu(path->rlid)); + rdma_ah_set_path_bits(ah_attr, path->grh_mylmc & 0x7f); + rdma_ah_set_static_rate(ah_attr, + path->static_rate ? path->static_rate - 5 : 0); + if (path->grh_mylmc & (1 << 7)) { + rdma_ah_set_grh(ah_attr, NULL, + be32_to_cpu(path->tclass_flowlabel) & 0xfffff, + path->mgid_index, + path->hop_limit, + (be32_to_cpu(path->tclass_flowlabel) + >> 20) & 0xff); + rdma_ah_set_dgid_raw(ah_attr, path->rgid); } } @@ -3467,10 +3466,11 @@ int mlx4_ib_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, int qp_attr to_ib_qp_access_flags(be32_to_cpu(context.params2)); if (qp->ibqp.qp_type == IB_QPT_RC || qp->ibqp.qp_type == IB_QPT_UC) { - to_ib_ah_attr(dev, &qp_attr->ah_attr, &context.pri_path); - to_ib_ah_attr(dev, &qp_attr->alt_ah_attr, &context.alt_path); + to_rdma_ah_attr(dev, &qp_attr->ah_attr, &context.pri_path); + to_rdma_ah_attr(dev, &qp_attr->alt_ah_attr, &context.alt_path); qp_attr->alt_pkey_index = context.alt_path.pkey_index & 0x7f; - qp_attr->alt_port_num = qp_attr->alt_ah_attr.port_num; + qp_attr->alt_port_num = + rdma_ah_get_port_num(&qp_attr->alt_ah_attr); } qp_attr->pkey_index = context.pri_path.pkey_index & 0x7f; diff --git a/drivers/infiniband/hw/mlx4/srq.c b/drivers/infiniband/hw/mlx4/srq.c index 7dd3f267f06b..e32dd58937a8 100644 --- a/drivers/infiniband/hw/mlx4/srq.c +++ b/drivers/infiniband/hw/mlx4/srq.c @@ -122,7 +122,7 @@ struct ib_srq *mlx4_ib_create_srq(struct ib_pd *pd, } err = mlx4_mtt_init(dev->dev, ib_umem_page_count(srq->umem), - ilog2(srq->umem->page_size), &srq->mtt); + srq->umem->page_shift, &srq->mtt); if (err) goto err_buf; diff --git a/drivers/infiniband/hw/mlx5/ah.c b/drivers/infiniband/hw/mlx5/ah.c index d090e96f6f01..3363e29157f6 100644 --- a/drivers/infiniband/hw/mlx5/ah.c +++ b/drivers/infiniband/hw/mlx5/ah.c @@ -34,50 +34,51 @@ static struct ib_ah *create_ib_ah(struct mlx5_ib_dev *dev, struct mlx5_ib_ah *ah, - struct ib_ah_attr *ah_attr, - enum rdma_link_layer ll) + struct rdma_ah_attr *ah_attr) { - if (ah_attr->ah_flags & IB_AH_GRH) { - memcpy(ah->av.rgid, &ah_attr->grh.dgid, 16); - ah->av.grh_gid_fl = cpu_to_be32(ah_attr->grh.flow_label | + if (rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH) { + const struct ib_global_route *grh = rdma_ah_read_grh(ah_attr); + + memcpy(ah->av.rgid, &grh->dgid, 16); + ah->av.grh_gid_fl = cpu_to_be32(grh->flow_label | (1 << 30) | - ah_attr->grh.sgid_index << 20); - ah->av.hop_limit = ah_attr->grh.hop_limit; - ah->av.tclass = ah_attr->grh.traffic_class; + grh->sgid_index << 20); + ah->av.hop_limit = grh->hop_limit; + ah->av.tclass = grh->traffic_class; } - ah->av.stat_rate_sl = (ah_attr->static_rate << 4); + ah->av.stat_rate_sl = (rdma_ah_get_static_rate(ah_attr) << 4); - if (ll == IB_LINK_LAYER_ETHERNET) { - memcpy(ah->av.rmac, ah_attr->dmac, sizeof(ah_attr->dmac)); + if (ah_attr->type == RDMA_AH_ATTR_TYPE_ROCE) { + memcpy(ah->av.rmac, ah_attr->roce.dmac, + sizeof(ah_attr->roce.dmac)); ah->av.udp_sport = - mlx5_get_roce_udp_sport(dev, - ah_attr->port_num, - ah_attr->grh.sgid_index); - ah->av.stat_rate_sl |= (ah_attr->sl & 0x7) << 1; + mlx5_get_roce_udp_sport(dev, + rdma_ah_get_port_num(ah_attr), + rdma_ah_read_grh(ah_attr)->sgid_index); + ah->av.stat_rate_sl |= (rdma_ah_get_sl(ah_attr) & 0x7) << 1; } else { - ah->av.rlid = cpu_to_be16(ah_attr->dlid); - ah->av.fl_mlid = ah_attr->src_path_bits & 0x7f; - ah->av.stat_rate_sl |= (ah_attr->sl & 0xf); + ah->av.rlid = cpu_to_be16(rdma_ah_get_dlid(ah_attr)); + ah->av.fl_mlid = rdma_ah_get_path_bits(ah_attr) & 0x7f; + ah->av.stat_rate_sl |= (rdma_ah_get_sl(ah_attr) & 0xf); } return &ah->ibah; } -struct ib_ah *mlx5_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, +struct ib_ah *mlx5_ib_create_ah(struct ib_pd *pd, struct rdma_ah_attr *ah_attr, struct ib_udata *udata) { struct mlx5_ib_ah *ah; struct mlx5_ib_dev *dev = to_mdev(pd->device); - enum rdma_link_layer ll; - - ll = pd->device->get_link_layer(pd->device, ah_attr->port_num); + enum rdma_ah_attr_type ah_type = ah_attr->type; - if (ll == IB_LINK_LAYER_ETHERNET && !(ah_attr->ah_flags & IB_AH_GRH)) + if ((ah_type == RDMA_AH_ATTR_TYPE_ROCE) && + !(rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH)) return ERR_PTR(-EINVAL); - if (ll == IB_LINK_LAYER_ETHERNET && udata) { + if (ah_type == RDMA_AH_ATTR_TYPE_ROCE && udata) { int err; struct mlx5_ib_create_ah_resp resp = {}; u32 min_resp_len = offsetof(typeof(resp), dmac) + @@ -92,7 +93,7 @@ struct ib_ah *mlx5_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, if (err) return ERR_PTR(err); - memcpy(resp.dmac, ah_attr->dmac, ETH_ALEN); + memcpy(resp.dmac, ah_attr->roce.dmac, ETH_ALEN); err = ib_copy_to_udata(udata, &resp, resp.response_length); if (err) return ERR_PTR(err); @@ -102,28 +103,29 @@ struct ib_ah *mlx5_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, if (!ah) return ERR_PTR(-ENOMEM); - return create_ib_ah(dev, ah, ah_attr, ll); /* never fails */ + return create_ib_ah(dev, ah, ah_attr); /* never fails */ } -int mlx5_ib_query_ah(struct ib_ah *ibah, struct ib_ah_attr *ah_attr) +int mlx5_ib_query_ah(struct ib_ah *ibah, struct rdma_ah_attr *ah_attr) { struct mlx5_ib_ah *ah = to_mah(ibah); u32 tmp; memset(ah_attr, 0, sizeof(*ah_attr)); + ah_attr->type = ibah->type; tmp = be32_to_cpu(ah->av.grh_gid_fl); if (tmp & (1 << 30)) { - ah_attr->ah_flags = IB_AH_GRH; - ah_attr->grh.sgid_index = (tmp >> 20) & 0xff; - ah_attr->grh.flow_label = tmp & 0xfffff; - memcpy(&ah_attr->grh.dgid, ah->av.rgid, 16); - ah_attr->grh.hop_limit = ah->av.hop_limit; - ah_attr->grh.traffic_class = ah->av.tclass; + rdma_ah_set_grh(ah_attr, NULL, + tmp & 0xfffff, + (tmp >> 20) & 0xff, + ah->av.hop_limit, + ah->av.tclass); + rdma_ah_set_dgid_raw(ah_attr, ah->av.rgid); } - ah_attr->dlid = be16_to_cpu(ah->av.rlid); - ah_attr->static_rate = ah->av.stat_rate_sl >> 4; - ah_attr->sl = ah->av.stat_rate_sl & 0xf; + rdma_ah_set_dlid(ah_attr, be16_to_cpu(ah->av.rlid)); + rdma_ah_set_static_rate(ah_attr, ah->av.stat_rate_sl >> 4); + rdma_ah_set_sl(ah_attr, ah->av.stat_rate_sl & 0xf); return 0; } diff --git a/drivers/infiniband/hw/mlx5/cmd.c b/drivers/infiniband/hw/mlx5/cmd.c index cdc2d3017da7..18d5e1db93ed 100644 --- a/drivers/infiniband/hw/mlx5/cmd.c +++ b/drivers/infiniband/hw/mlx5/cmd.c @@ -46,3 +46,14 @@ int mlx5_cmd_null_mkey(struct mlx5_core_dev *dev, u32 *null_mkey) null_mkey); return err; } + +int mlx5_cmd_query_cong_counter(struct mlx5_core_dev *dev, + bool reset, void *out, int out_size) +{ + u32 in[MLX5_ST_SZ_DW(query_cong_statistics_in)] = { }; + + MLX5_SET(query_cong_statistics_in, in, opcode, + MLX5_CMD_OP_QUERY_CONG_STATISTICS); + MLX5_SET(query_cong_statistics_in, in, clear, reset); + return mlx5_cmd_exec(dev, in, sizeof(in), out, out_size); +} diff --git a/drivers/infiniband/hw/mlx5/cmd.h b/drivers/infiniband/hw/mlx5/cmd.h index 7ca8a7b6434d..fa09228193a6 100644 --- a/drivers/infiniband/hw/mlx5/cmd.h +++ b/drivers/infiniband/hw/mlx5/cmd.h @@ -37,4 +37,6 @@ #include <linux/mlx5/driver.h> int mlx5_cmd_null_mkey(struct mlx5_core_dev *dev, u32 *null_mkey); +int mlx5_cmd_query_cong_counter(struct mlx5_core_dev *dev, + bool reset, void *out, int out_size); #endif /* MLX5_IB_CMD_H */ diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c index 31803b367104..94c049b62c2f 100644 --- a/drivers/infiniband/hw/mlx5/cq.c +++ b/drivers/infiniband/hw/mlx5/cq.c @@ -172,6 +172,8 @@ static void handle_responder(struct ib_wc *wc, struct mlx5_cqe64 *cqe, struct mlx5_ib_srq *srq; struct mlx5_ib_wq *wq; u16 wqe_ctr; + u8 roce_packet_type; + bool vlan_present; u8 g; if (qp->ibqp.srq || qp->ibqp.xrcd) { @@ -223,7 +225,6 @@ static void handle_responder(struct ib_wc *wc, struct mlx5_cqe64 *cqe, break; } wc->slid = be16_to_cpu(cqe->slid); - wc->sl = (be32_to_cpu(cqe->flags_rqpn) >> 24) & 0xf; wc->src_qp = be32_to_cpu(cqe->flags_rqpn) & 0xffffff; wc->dlid_path_bits = cqe->ml_path; g = (be32_to_cpu(cqe->flags_rqpn) >> 28) & 3; @@ -237,10 +238,22 @@ static void handle_responder(struct ib_wc *wc, struct mlx5_cqe64 *cqe, wc->pkey_index = 0; } - if (ll != IB_LINK_LAYER_ETHERNET) + if (ll != IB_LINK_LAYER_ETHERNET) { + wc->sl = (be32_to_cpu(cqe->flags_rqpn) >> 24) & 0xf; return; + } + + vlan_present = cqe->l4_l3_hdr_type & 0x1; + roce_packet_type = (be32_to_cpu(cqe->flags_rqpn) >> 24) & 0x3; + if (vlan_present) { + wc->vlan_id = (be16_to_cpu(cqe->vlan_info)) & 0xfff; + wc->sl = (be16_to_cpu(cqe->vlan_info) >> 13) & 0x7; + wc->wc_flags |= IB_WC_WITH_VLAN; + } else { + wc->sl = 0; + } - switch (wc->sl & 0x3) { + switch (roce_packet_type) { case MLX5_CQE_ROCE_L3_HEADER_TYPE_GRH: wc->network_hdr_type = RDMA_NETWORK_IB; break; @@ -818,7 +831,7 @@ static int create_cq_user(struct mlx5_ib_dev *dev, struct ib_udata *udata, return 0; err_cqb: - kfree(cqb); + kfree(*cqb); err_db: mlx5_ib_db_unmap_user(to_mucontext(context), &cq->db); diff --git a/drivers/infiniband/hw/mlx5/mad.c b/drivers/infiniband/hw/mlx5/mad.c index 8dacb49eabd9..f1b56de64871 100644 --- a/drivers/infiniband/hw/mlx5/mad.c +++ b/drivers/infiniband/hw/mlx5/mad.c @@ -187,6 +187,8 @@ static void pma_cnt_assign(struct ib_pma_portcounters *pma_cnt, port_xmit_discards); MLX5_ASSIGN_PMA_CNTR(pma_cnt->port_xmit_constraint_errors, port_xmit_constraint_errors); + MLX5_ASSIGN_PMA_CNTR(pma_cnt->port_xmit_wait, + port_xmit_wait); MLX5_ASSIGN_PMA_CNTR(pma_cnt->port_rcv_constraint_errors, port_rcv_constraint_errors); MLX5_ASSIGN_PMA_CNTR(pma_cnt->link_overrun_errors, diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 4dc0a8785fe0..9f3ba320ce70 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -57,6 +57,7 @@ #include <linux/mlx5/fs.h> #include <linux/mlx5/vport.h> #include "mlx5_ib.h" +#include "cmd.h" #define DRIVER_NAME "mlx5_ib" #define DRIVER_VERSION "2.2-1" @@ -165,15 +166,82 @@ static struct net_device *mlx5_ib_get_netdev(struct ib_device *device, return ndev; } -static int mlx5_query_port_roce(struct ib_device *device, u8 port_num, - struct ib_port_attr *props) +static int translate_eth_proto_oper(u32 eth_proto_oper, u8 *active_speed, + u8 *active_width) +{ + switch (eth_proto_oper) { + case MLX5E_PROT_MASK(MLX5E_1000BASE_CX_SGMII): + case MLX5E_PROT_MASK(MLX5E_1000BASE_KX): + case MLX5E_PROT_MASK(MLX5E_100BASE_TX): + case MLX5E_PROT_MASK(MLX5E_1000BASE_T): + *active_width = IB_WIDTH_1X; + *active_speed = IB_SPEED_SDR; + break; + case MLX5E_PROT_MASK(MLX5E_10GBASE_T): + case MLX5E_PROT_MASK(MLX5E_10GBASE_CX4): + case MLX5E_PROT_MASK(MLX5E_10GBASE_KX4): + case MLX5E_PROT_MASK(MLX5E_10GBASE_KR): + case MLX5E_PROT_MASK(MLX5E_10GBASE_CR): + case MLX5E_PROT_MASK(MLX5E_10GBASE_SR): + case MLX5E_PROT_MASK(MLX5E_10GBASE_ER): + *active_width = IB_WIDTH_1X; + *active_speed = IB_SPEED_QDR; + break; + case MLX5E_PROT_MASK(MLX5E_25GBASE_CR): + case MLX5E_PROT_MASK(MLX5E_25GBASE_KR): + case MLX5E_PROT_MASK(MLX5E_25GBASE_SR): + *active_width = IB_WIDTH_1X; + *active_speed = IB_SPEED_EDR; + break; + case MLX5E_PROT_MASK(MLX5E_40GBASE_CR4): + case MLX5E_PROT_MASK(MLX5E_40GBASE_KR4): + case MLX5E_PROT_MASK(MLX5E_40GBASE_SR4): + case MLX5E_PROT_MASK(MLX5E_40GBASE_LR4): + *active_width = IB_WIDTH_4X; + *active_speed = IB_SPEED_QDR; + break; + case MLX5E_PROT_MASK(MLX5E_50GBASE_CR2): + case MLX5E_PROT_MASK(MLX5E_50GBASE_KR2): + case MLX5E_PROT_MASK(MLX5E_50GBASE_SR2): + *active_width = IB_WIDTH_1X; + *active_speed = IB_SPEED_HDR; + break; + case MLX5E_PROT_MASK(MLX5E_56GBASE_R4): + *active_width = IB_WIDTH_4X; + *active_speed = IB_SPEED_FDR; + break; + case MLX5E_PROT_MASK(MLX5E_100GBASE_CR4): + case MLX5E_PROT_MASK(MLX5E_100GBASE_SR4): + case MLX5E_PROT_MASK(MLX5E_100GBASE_KR4): + case MLX5E_PROT_MASK(MLX5E_100GBASE_LR4): + *active_width = IB_WIDTH_4X; + *active_speed = IB_SPEED_EDR; + break; + default: + return -EINVAL; + } + + return 0; +} + +static void mlx5_query_port_roce(struct ib_device *device, u8 port_num, + struct ib_port_attr *props) { struct mlx5_ib_dev *dev = to_mdev(device); + struct mlx5_core_dev *mdev = dev->mdev; struct net_device *ndev, *upper; enum ib_mtu ndev_ib_mtu; u16 qkey_viol_cntr; + u32 eth_prot_oper; - /* props being zeroed by the caller, avoid zeroing it here */ + /* Possible bad flows are checked before filling out props so in case + * of an error it will still be zeroed out. + */ + if (mlx5_query_port_eth_proto_oper(mdev, ð_prot_oper, port_num)) + return; + + translate_eth_proto_oper(eth_prot_oper, &props->active_speed, + &props->active_width); props->port_cap_flags |= IB_PORT_CM_SUP; props->port_cap_flags |= IB_PORT_IP_BASED_GIDS; @@ -191,7 +259,7 @@ static int mlx5_query_port_roce(struct ib_device *device, u8 port_num, ndev = mlx5_ib_get_netdev(device, port_num); if (!ndev) - return 0; + return; if (mlx5_lag_is_active(dev->mdev)) { rcu_read_lock(); @@ -214,11 +282,6 @@ static int mlx5_query_port_roce(struct ib_device *device, u8 port_num, dev_put(ndev); props->active_mtu = min(props->max_mtu, ndev_ib_mtu); - - props->active_width = IB_WIDTH_4X; /* TODO */ - props->active_speed = IB_SPEED_QDR; /* TODO */ - - return 0; } static void ib_gid_to_mlx5_roce_addr(const union ib_gid *gid, @@ -924,7 +987,8 @@ int mlx5_ib_query_port(struct ib_device *ibdev, u8 port, return mlx5_query_hca_port(ibdev, port, props); case MLX5_VPORT_ACCESS_METHOD_NIC: - return mlx5_query_port_roce(ibdev, port, props); + mlx5_query_port_roce(ibdev, port, props); + return 0; default: return -EINVAL; @@ -1478,7 +1542,7 @@ static void mlx5_ib_disassociate_ucontext(struct ib_ucontext *ibcontext) /* need to protect from a race on closing the vma as part of * mlx5_ib_vma_close. */ - down_read(&owning_mm->mmap_sem); + down_write(&owning_mm->mmap_sem); list_for_each_entry_safe(vma_private, n, &context->vma_private_list, list) { vma = vma_private->vma; @@ -1488,11 +1552,12 @@ static void mlx5_ib_disassociate_ucontext(struct ib_ucontext *ibcontext) /* context going to be destroyed, should * not access ops any more. */ + vma->vm_flags &= ~(VM_SHARED | VM_MAYSHARE); vma->vm_ops = NULL; list_del(&vma_private->list); kfree(vma_private); } - up_read(&owning_mm->mmap_sem); + up_write(&owning_mm->mmap_sem); mmput(owning_mm); put_task_struct(owning_process); } @@ -1726,6 +1791,7 @@ static void set_tos(void *outer_c, void *outer_v, u8 mask, u8 val) #define LAST_TCP_UDP_FIELD src_port #define LAST_TUNNEL_FIELD tunnel_id #define LAST_FLOW_TAG_FIELD tag_id +#define LAST_DROP_FIELD size /* Field is the last supported field */ #define FIELDS_NOT_SUPPORTED(filter, field)\ @@ -1735,8 +1801,11 @@ static void set_tos(void *outer_c, void *outer_v, u8 mask, u8 val) offsetof(typeof(filter), field) -\ sizeof(filter.field)) -static int parse_flow_attr(u32 *match_c, u32 *match_v, - const union ib_flow_spec *ib_spec, u32 *tag_id) +#define IPV4_VERSION 4 +#define IPV6_VERSION 6 +static int parse_flow_attr(struct mlx5_core_dev *mdev, u32 *match_c, + u32 *match_v, const union ib_flow_spec *ib_spec, + u32 *tag_id, bool *is_drop) { void *misc_params_c = MLX5_ADDR_OF(fte_match_param, match_c, misc_parameters); @@ -1744,17 +1813,22 @@ static int parse_flow_attr(u32 *match_c, u32 *match_v, misc_parameters); void *headers_c; void *headers_v; + int match_ipv; if (ib_spec->type & IB_FLOW_SPEC_INNER) { headers_c = MLX5_ADDR_OF(fte_match_param, match_c, inner_headers); headers_v = MLX5_ADDR_OF(fte_match_param, match_v, inner_headers); + match_ipv = MLX5_CAP_FLOWTABLE_NIC_RX(mdev, + ft_field_support.inner_ip_version); } else { headers_c = MLX5_ADDR_OF(fte_match_param, match_c, outer_headers); headers_v = MLX5_ADDR_OF(fte_match_param, match_v, outer_headers); + match_ipv = MLX5_CAP_FLOWTABLE_NIC_RX(mdev, + ft_field_support.outer_ip_version); } switch (ib_spec->type & ~IB_FLOW_SPEC_INNER) { @@ -1810,10 +1884,17 @@ static int parse_flow_attr(u32 *match_c, u32 *match_v, if (FIELDS_NOT_SUPPORTED(ib_spec->ipv4.mask, LAST_IPV4_FIELD)) return -EOPNOTSUPP; - MLX5_SET(fte_match_set_lyr_2_4, headers_c, - ethertype, 0xffff); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, - ethertype, ETH_P_IP); + if (match_ipv) { + MLX5_SET(fte_match_set_lyr_2_4, headers_c, + ip_version, 0xf); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, + ip_version, IPV4_VERSION); + } else { + MLX5_SET(fte_match_set_lyr_2_4, headers_c, + ethertype, 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, + ethertype, ETH_P_IP); + } memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, src_ipv4_src_ipv6.ipv4_layout.ipv4), @@ -1842,10 +1923,17 @@ static int parse_flow_attr(u32 *match_c, u32 *match_v, if (FIELDS_NOT_SUPPORTED(ib_spec->ipv6.mask, LAST_IPV6_FIELD)) return -EOPNOTSUPP; - MLX5_SET(fte_match_set_lyr_2_4, headers_c, - ethertype, 0xffff); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, - ethertype, ETH_P_IPV6); + if (match_ipv) { + MLX5_SET(fte_match_set_lyr_2_4, headers_c, + ip_version, 0xf); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, + ip_version, IPV6_VERSION); + } else { + MLX5_SET(fte_match_set_lyr_2_4, headers_c, + ethertype, 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, + ethertype, ETH_P_IPV6); + } memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, src_ipv4_src_ipv6.ipv6_layout.ipv6), @@ -1937,6 +2025,12 @@ static int parse_flow_attr(u32 *match_c, u32 *match_v, *tag_id = ib_spec->flow_tag.tag_id; break; + case IB_FLOW_SPEC_ACTION_DROP: + if (FIELDS_NOT_SUPPORTED(ib_spec->drop, + LAST_DROP_FIELD)) + return -EOPNOTSUPP; + *is_drop = true; + break; default: return -EINVAL; } @@ -1967,26 +2061,60 @@ static bool flow_is_multicast_only(struct ib_flow_attr *ib_attr) is_multicast_ether_addr(eth_spec->val.dst_mac); } -static bool is_valid_attr(const struct ib_flow_attr *flow_attr) +static bool is_valid_ethertype(struct mlx5_core_dev *mdev, + const struct ib_flow_attr *flow_attr, + bool check_inner) { union ib_flow_spec *ib_spec = (union ib_flow_spec *)(flow_attr + 1); - bool has_ipv4_spec = false; - bool eth_type_ipv4 = true; + int match_ipv = check_inner ? + MLX5_CAP_FLOWTABLE_NIC_RX(mdev, + ft_field_support.inner_ip_version) : + MLX5_CAP_FLOWTABLE_NIC_RX(mdev, + ft_field_support.outer_ip_version); + int inner_bit = check_inner ? IB_FLOW_SPEC_INNER : 0; + bool ipv4_spec_valid, ipv6_spec_valid; + unsigned int ip_spec_type = 0; + bool has_ethertype = false; unsigned int spec_index; + bool mask_valid = true; + u16 eth_type = 0; + bool type_valid; /* Validate that ethertype is correct */ for (spec_index = 0; spec_index < flow_attr->num_of_specs; spec_index++) { - if (ib_spec->type == IB_FLOW_SPEC_ETH && + if ((ib_spec->type == (IB_FLOW_SPEC_ETH | inner_bit)) && ib_spec->eth.mask.ether_type) { - if (!((ib_spec->eth.mask.ether_type == htons(0xffff)) && - ib_spec->eth.val.ether_type == htons(ETH_P_IP))) - eth_type_ipv4 = false; - } else if (ib_spec->type == IB_FLOW_SPEC_IPV4) { - has_ipv4_spec = true; + mask_valid = (ib_spec->eth.mask.ether_type == + htons(0xffff)); + has_ethertype = true; + eth_type = ntohs(ib_spec->eth.val.ether_type); + } else if ((ib_spec->type == (IB_FLOW_SPEC_IPV4 | inner_bit)) || + (ib_spec->type == (IB_FLOW_SPEC_IPV6 | inner_bit))) { + ip_spec_type = ib_spec->type; } ib_spec = (void *)ib_spec + ib_spec->size; } - return !has_ipv4_spec || eth_type_ipv4; + + type_valid = (!has_ethertype) || (!ip_spec_type); + if (!type_valid && mask_valid) { + ipv4_spec_valid = (eth_type == ETH_P_IP) && + (ip_spec_type == (IB_FLOW_SPEC_IPV4 | inner_bit)); + ipv6_spec_valid = (eth_type == ETH_P_IPV6) && + (ip_spec_type == (IB_FLOW_SPEC_IPV6 | inner_bit)); + + type_valid = (ipv4_spec_valid) || (ipv6_spec_valid) || + (((eth_type == ETH_P_MPLS_UC) || + (eth_type == ETH_P_MPLS_MC)) && match_ipv); + } + + return type_valid; +} + +static bool is_valid_attr(struct mlx5_core_dev *mdev, + const struct ib_flow_attr *flow_attr) +{ + return is_valid_ethertype(mdev, flow_attr, false) && + is_valid_ethertype(mdev, flow_attr, true); } static void put_flow_table(struct mlx5_ib_dev *dev, @@ -2038,8 +2166,8 @@ enum flow_table_type { MLX5_IB_FT_TX }; -#define MLX5_FS_MAX_TYPES 10 -#define MLX5_FS_MAX_ENTRIES 32000UL +#define MLX5_FS_MAX_TYPES 6 +#define MLX5_FS_MAX_ENTRIES BIT(16) static struct mlx5_ib_flow_prio *get_flow_table(struct mlx5_ib_dev *dev, struct ib_flow_attr *flow_attr, enum flow_table_type ft_type) @@ -2048,11 +2176,14 @@ static struct mlx5_ib_flow_prio *get_flow_table(struct mlx5_ib_dev *dev, struct mlx5_flow_namespace *ns = NULL; struct mlx5_ib_flow_prio *prio; struct mlx5_flow_table *ft; + int max_table_size; int num_entries; int num_groups; int priority; int err = 0; + max_table_size = BIT(MLX5_CAP_FLOWTABLE_NIC_RX(dev->mdev, + log_max_ft_size)); if (flow_attr->type == IB_FLOW_ATTR_NORMAL) { if (flow_is_multicast_only(flow_attr) && !dont_trap) @@ -2091,6 +2222,9 @@ static struct mlx5_ib_flow_prio *get_flow_table(struct mlx5_ib_dev *dev, if (!ns) return ERR_PTR(-ENOTSUPP); + if (num_entries > max_table_size) + return ERR_PTR(-ENOMEM); + ft = prio->flow_table; if (!ft) { ft = mlx5_create_auto_grouped_flow_table(ns, priority, @@ -2118,12 +2252,15 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, struct mlx5_ib_flow_handler *handler; struct mlx5_flow_act flow_act = {0}; struct mlx5_flow_spec *spec; + struct mlx5_flow_destination *rule_dst = dst; const void *ib_flow = (const void *)flow_attr + sizeof(*flow_attr); unsigned int spec_index; u32 flow_tag = MLX5_FS_DEFAULT_FLOW_TAG; + bool is_drop = false; int err = 0; + int dest_num = 1; - if (!is_valid_attr(flow_attr)) + if (!is_valid_attr(dev->mdev, flow_attr)) return ERR_PTR(-EINVAL); spec = mlx5_vzalloc(sizeof(*spec)); @@ -2136,8 +2273,9 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, INIT_LIST_HEAD(&handler->list); for (spec_index = 0; spec_index < flow_attr->num_of_specs; spec_index++) { - err = parse_flow_attr(spec->match_criteria, - spec->match_value, ib_flow, &flow_tag); + err = parse_flow_attr(dev->mdev, spec->match_criteria, + spec->match_value, + ib_flow, &flow_tag, &is_drop); if (err < 0) goto free; @@ -2145,8 +2283,14 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, } spec->match_criteria_enable = get_match_criteria_enable(spec->match_criteria); - flow_act.action = dst ? MLX5_FLOW_CONTEXT_ACTION_FWD_DEST : - MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_PRIO; + if (is_drop) { + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP; + rule_dst = NULL; + dest_num = 0; + } else { + flow_act.action = dst ? MLX5_FLOW_CONTEXT_ACTION_FWD_DEST : + MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_PRIO; + } if (flow_tag != MLX5_FS_DEFAULT_FLOW_TAG && (flow_attr->type == IB_FLOW_ATTR_ALL_DEFAULT || @@ -2159,7 +2303,7 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, flow_act.flow_tag = flow_tag; handler->rule = mlx5_add_flow_rules(ft, spec, &flow_act, - dst, 1); + rule_dst, dest_num); if (IS_ERR(handler->rule)) { err = PTR_ERR(handler->rule); @@ -2315,7 +2459,7 @@ static struct ib_flow *mlx5_ib_create_flow(struct ib_qp *qp, int err; if (flow_attr->priority > MLX5_IB_FLOW_LAST_PRIO) - return ERR_PTR(-ENOSPC); + return ERR_PTR(-ENOMEM); if (domain != IB_FLOW_DOMAIN_USER || flow_attr->port > MLX5_CAP_GEN(dev->mdev, num_ports) || @@ -3133,7 +3277,7 @@ static void mlx5_disable_eth(struct mlx5_ib_dev *dev) mlx5_nic_vport_disable_roce(dev->mdev); } -struct mlx5_ib_q_counter { +struct mlx5_ib_counter { const char *name; size_t offset; }; @@ -3141,18 +3285,18 @@ struct mlx5_ib_q_counter { #define INIT_Q_COUNTER(_name) \ { .name = #_name, .offset = MLX5_BYTE_OFF(query_q_counter_out, _name)} -static const struct mlx5_ib_q_counter basic_q_cnts[] = { +static const struct mlx5_ib_counter basic_q_cnts[] = { INIT_Q_COUNTER(rx_write_requests), INIT_Q_COUNTER(rx_read_requests), INIT_Q_COUNTER(rx_atomic_requests), INIT_Q_COUNTER(out_of_buffer), }; -static const struct mlx5_ib_q_counter out_of_seq_q_cnts[] = { +static const struct mlx5_ib_counter out_of_seq_q_cnts[] = { INIT_Q_COUNTER(out_of_sequence), }; -static const struct mlx5_ib_q_counter retrans_q_cnts[] = { +static const struct mlx5_ib_counter retrans_q_cnts[] = { INIT_Q_COUNTER(duplicate_request), INIT_Q_COUNTER(rnr_nak_retry_err), INIT_Q_COUNTER(packet_seq_err), @@ -3160,22 +3304,31 @@ static const struct mlx5_ib_q_counter retrans_q_cnts[] = { INIT_Q_COUNTER(local_ack_timeout_err), }; -static void mlx5_ib_dealloc_q_counters(struct mlx5_ib_dev *dev) +#define INIT_CONG_COUNTER(_name) \ + { .name = #_name, .offset = \ + MLX5_BYTE_OFF(query_cong_statistics_out, _name ## _high)} + +static const struct mlx5_ib_counter cong_cnts[] = { + INIT_CONG_COUNTER(rp_cnp_ignored), + INIT_CONG_COUNTER(rp_cnp_handled), + INIT_CONG_COUNTER(np_ecn_marked_roce_packets), + INIT_CONG_COUNTER(np_cnp_sent), +}; + +static void mlx5_ib_dealloc_counters(struct mlx5_ib_dev *dev) { unsigned int i; for (i = 0; i < dev->num_ports; i++) { mlx5_core_dealloc_q_counter(dev->mdev, - dev->port[i].q_cnts.set_id); - kfree(dev->port[i].q_cnts.names); - kfree(dev->port[i].q_cnts.offsets); + dev->port[i].cnts.set_id); + kfree(dev->port[i].cnts.names); + kfree(dev->port[i].cnts.offsets); } } -static int __mlx5_ib_alloc_q_counters(struct mlx5_ib_dev *dev, - const char ***names, - size_t **offsets, - u32 *num) +static int __mlx5_ib_alloc_counters(struct mlx5_ib_dev *dev, + struct mlx5_ib_counters *cnts) { u32 num_counters; @@ -3186,27 +3339,32 @@ static int __mlx5_ib_alloc_q_counters(struct mlx5_ib_dev *dev, if (MLX5_CAP_GEN(dev->mdev, retransmission_q_counters)) num_counters += ARRAY_SIZE(retrans_q_cnts); + cnts->num_q_counters = num_counters; - *names = kcalloc(num_counters, sizeof(**names), GFP_KERNEL); - if (!*names) + if (MLX5_CAP_GEN(dev->mdev, cc_query_allowed)) { + cnts->num_cong_counters = ARRAY_SIZE(cong_cnts); + num_counters += ARRAY_SIZE(cong_cnts); + } + + cnts->names = kcalloc(num_counters, sizeof(cnts->names), GFP_KERNEL); + if (!cnts->names) return -ENOMEM; - *offsets = kcalloc(num_counters, sizeof(**offsets), GFP_KERNEL); - if (!*offsets) + cnts->offsets = kcalloc(num_counters, + sizeof(cnts->offsets), GFP_KERNEL); + if (!cnts->offsets) goto err_names; - *num = num_counters; - return 0; err_names: - kfree(*names); + kfree(cnts->names); return -ENOMEM; } -static void mlx5_ib_fill_q_counters(struct mlx5_ib_dev *dev, - const char **names, - size_t *offsets) +static void mlx5_ib_fill_counters(struct mlx5_ib_dev *dev, + const char **names, + size_t *offsets) { int i; int j = 0; @@ -3229,9 +3387,16 @@ static void mlx5_ib_fill_q_counters(struct mlx5_ib_dev *dev, offsets[j] = retrans_q_cnts[i].offset; } } + + if (MLX5_CAP_GEN(dev->mdev, cc_query_allowed)) { + for (i = 0; i < ARRAY_SIZE(cong_cnts); i++, j++) { + names[j] = cong_cnts[i].name; + offsets[j] = cong_cnts[i].offset; + } + } } -static int mlx5_ib_alloc_q_counters(struct mlx5_ib_dev *dev) +static int mlx5_ib_alloc_counters(struct mlx5_ib_dev *dev) { int i; int ret; @@ -3240,7 +3405,7 @@ static int mlx5_ib_alloc_q_counters(struct mlx5_ib_dev *dev) struct mlx5_ib_port *port = &dev->port[i]; ret = mlx5_core_alloc_q_counter(dev->mdev, - &port->q_cnts.set_id); + &port->cnts.set_id); if (ret) { mlx5_ib_warn(dev, "couldn't allocate queue counter for port %d, err %d\n", @@ -3248,15 +3413,12 @@ static int mlx5_ib_alloc_q_counters(struct mlx5_ib_dev *dev) goto dealloc_counters; } - ret = __mlx5_ib_alloc_q_counters(dev, - &port->q_cnts.names, - &port->q_cnts.offsets, - &port->q_cnts.num_counters); + ret = __mlx5_ib_alloc_counters(dev, &port->cnts); if (ret) goto dealloc_counters; - mlx5_ib_fill_q_counters(dev, port->q_cnts.names, - port->q_cnts.offsets); + mlx5_ib_fill_counters(dev, port->cnts.names, + port->cnts.offsets); } return 0; @@ -3264,7 +3426,7 @@ static int mlx5_ib_alloc_q_counters(struct mlx5_ib_dev *dev) dealloc_counters: while (--i >= 0) mlx5_core_dealloc_q_counter(dev->mdev, - dev->port[i].q_cnts.set_id); + dev->port[i].cnts.set_id); return ret; } @@ -3279,44 +3441,93 @@ static struct rdma_hw_stats *mlx5_ib_alloc_hw_stats(struct ib_device *ibdev, if (port_num == 0) return NULL; - return rdma_alloc_hw_stats_struct(port->q_cnts.names, - port->q_cnts.num_counters, + return rdma_alloc_hw_stats_struct(port->cnts.names, + port->cnts.num_q_counters + + port->cnts.num_cong_counters, RDMA_HW_STATS_DEFAULT_LIFESPAN); } -static int mlx5_ib_get_hw_stats(struct ib_device *ibdev, - struct rdma_hw_stats *stats, - u8 port_num, int index) +static int mlx5_ib_query_q_counters(struct mlx5_ib_dev *dev, + struct mlx5_ib_port *port, + struct rdma_hw_stats *stats) { - struct mlx5_ib_dev *dev = to_mdev(ibdev); - struct mlx5_ib_port *port = &dev->port[port_num - 1]; int outlen = MLX5_ST_SZ_BYTES(query_q_counter_out); void *out; __be32 val; - int ret; - int i; - - if (!stats) - return -ENOSYS; + int ret, i; out = mlx5_vzalloc(outlen); if (!out) return -ENOMEM; ret = mlx5_core_query_q_counter(dev->mdev, - port->q_cnts.set_id, 0, + port->cnts.set_id, 0, out, outlen); if (ret) goto free; - for (i = 0; i < port->q_cnts.num_counters; i++) { - val = *(__be32 *)(out + port->q_cnts.offsets[i]); + for (i = 0; i < port->cnts.num_q_counters; i++) { + val = *(__be32 *)(out + port->cnts.offsets[i]); stats->value[i] = (u64)be32_to_cpu(val); } free: kvfree(out); - return port->q_cnts.num_counters; + return ret; +} + +static int mlx5_ib_query_cong_counters(struct mlx5_ib_dev *dev, + struct mlx5_ib_port *port, + struct rdma_hw_stats *stats) +{ + int outlen = MLX5_ST_SZ_BYTES(query_cong_statistics_out); + void *out; + int ret, i; + int offset = port->cnts.num_q_counters; + + out = mlx5_vzalloc(outlen); + if (!out) + return -ENOMEM; + + ret = mlx5_cmd_query_cong_counter(dev->mdev, false, out, outlen); + if (ret) + goto free; + + for (i = 0; i < port->cnts.num_cong_counters; i++) { + stats->value[i + offset] = + be64_to_cpup((__be64 *)(out + + port->cnts.offsets[i + offset])); + } + +free: + kvfree(out); + return ret; +} + +static int mlx5_ib_get_hw_stats(struct ib_device *ibdev, + struct rdma_hw_stats *stats, + u8 port_num, int index) +{ + struct mlx5_ib_dev *dev = to_mdev(ibdev); + struct mlx5_ib_port *port = &dev->port[port_num - 1]; + int ret, num_counters; + + if (!stats) + return -EINVAL; + + ret = mlx5_ib_query_q_counters(dev, port, stats); + if (ret) + return ret; + num_counters = port->cnts.num_q_counters; + + if (MLX5_CAP_GEN(dev->mdev, cc_query_allowed)) { + ret = mlx5_ib_query_cong_counters(dev, port, stats); + if (ret) + return ret; + num_counters += port->cnts.num_cong_counters; + } + + return num_counters; } static void *mlx5_ib_add(struct mlx5_core_dev *mdev) @@ -3523,14 +3734,14 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) goto err_rsrc; if (MLX5_CAP_GEN(dev->mdev, max_qp_cnt)) { - err = mlx5_ib_alloc_q_counters(dev); + err = mlx5_ib_alloc_counters(dev); if (err) goto err_odp; } dev->mdev->priv.uar = mlx5_get_uars_page(dev->mdev); if (!dev->mdev->priv.uar) - goto err_q_cnt; + goto err_cnt; err = mlx5_alloc_bfreg(dev->mdev, &dev->bfreg, false, false); if (err) @@ -3574,9 +3785,9 @@ err_bfreg: err_uar_page: mlx5_put_uars_page(dev->mdev, dev->mdev->priv.uar); -err_q_cnt: +err_cnt: if (MLX5_CAP_GEN(dev->mdev, max_qp_cnt)) - mlx5_ib_dealloc_q_counters(dev); + mlx5_ib_dealloc_counters(dev); err_odp: mlx5_ib_odp_remove_one(dev); @@ -3610,7 +3821,7 @@ static void mlx5_ib_remove(struct mlx5_core_dev *mdev, void *context) mlx5_free_bfreg(dev->mdev, &dev->bfreg); mlx5_put_uars_page(dev->mdev, mdev->priv.uar); if (MLX5_CAP_GEN(dev->mdev, max_qp_cnt)) - mlx5_ib_dealloc_q_counters(dev); + mlx5_ib_dealloc_counters(dev); destroy_umrc_res(dev); mlx5_ib_odp_remove_one(dev); destroy_dev_resources(&dev->devr); diff --git a/drivers/infiniband/hw/mlx5/mem.c b/drivers/infiniband/hw/mlx5/mem.c index 778d8a18925f..914f212e7ef6 100644 --- a/drivers/infiniband/hw/mlx5/mem.c +++ b/drivers/infiniband/hw/mlx5/mem.c @@ -59,15 +59,14 @@ void mlx5_ib_cont_pages(struct ib_umem *umem, u64 addr, u64 pfn; struct scatterlist *sg; int entry; - unsigned long page_shift = ilog2(umem->page_size); + unsigned long page_shift = umem->page_shift; - /* With ODP we must always match OS page size. */ if (umem->odp_data) { - *count = ib_umem_page_count(umem); - *shift = PAGE_SHIFT; - *ncont = *count; + *ncont = ib_umem_page_count(umem); + *count = *ncont << (page_shift - PAGE_SHIFT); + *shift = page_shift; if (order) - *order = ilog2(roundup_pow_of_two(*count)); + *order = ilog2(roundup_pow_of_two(*ncont)); return; } @@ -156,7 +155,7 @@ void __mlx5_ib_populate_pas(struct mlx5_ib_dev *dev, struct ib_umem *umem, int page_shift, size_t offset, size_t num_pages, __be64 *pas, int access_flags) { - unsigned long umem_page_shift = ilog2(umem->page_size); + unsigned long umem_page_shift = umem->page_shift; int shift = page_shift - umem_page_shift; int mask = (1 << shift) - 1; int i, k, idx; diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index ce8ba617d46e..38c877bc45e5 100644 --- a/drivers/infiniband/hw/mlx5/mlx5_ib.h +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -513,6 +513,7 @@ struct mlx5_ib_mr { struct mlx5_ib_mw { struct ib_mw ibmw; struct mlx5_core_mkey mmkey; + int ndescs; }; struct mlx5_ib_umr_context { @@ -595,15 +596,16 @@ struct mlx5_ib_resources { struct mutex mutex; }; -struct mlx5_ib_q_counters { +struct mlx5_ib_counters { const char **names; size_t *offsets; - u32 num_counters; + u32 num_q_counters; + u32 num_cong_counters; u16 set_id; }; struct mlx5_ib_port { - struct mlx5_ib_q_counters q_cnts; + struct mlx5_ib_counters cnts; }; struct mlx5_roce { @@ -738,9 +740,9 @@ void mlx5_ib_free_srq_wqe(struct mlx5_ib_srq *srq, int wqe_index); int mlx5_MAD_IFC(struct mlx5_ib_dev *dev, int ignore_mkey, int ignore_bkey, u8 port, const struct ib_wc *in_wc, const struct ib_grh *in_grh, const void *in_mad, void *response_mad); -struct ib_ah *mlx5_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, +struct ib_ah *mlx5_ib_create_ah(struct ib_pd *pd, struct rdma_ah_attr *ah_attr, struct ib_udata *udata); -int mlx5_ib_query_ah(struct ib_ah *ibah, struct ib_ah_attr *ah_attr); +int mlx5_ib_query_ah(struct ib_ah *ibah, struct rdma_ah_attr *ah_attr); int mlx5_ib_destroy_ah(struct ib_ah *ah); struct ib_srq *mlx5_ib_create_srq(struct ib_pd *pd, struct ib_srq_init_attr *init_attr, diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c index b8f9382a8b7d..366433f71b58 100644 --- a/drivers/infiniband/hw/mlx5/mr.c +++ b/drivers/infiniband/hw/mlx5/mr.c @@ -1009,7 +1009,7 @@ int mlx5_ib_update_xlt(struct mlx5_ib_mr *mr, u64 idx, int npages, } if (!xlt) { - uctx = to_mucontext(mr->ibmr.uobject->context); + uctx = to_mucontext(mr->ibmr.pd->uobject->context); mlx5_ib_warn(dev, "Using XLT emergency buffer\n"); size = PAGE_SIZE; xlt = (void *)uctx->upd_xlt_page; @@ -1045,8 +1045,9 @@ int mlx5_ib_update_xlt(struct mlx5_ib_mr *mr, u64 idx, int npages, for (pages_mapped = 0; pages_mapped < pages_to_map && !err; pages_mapped += pages_iter, idx += pages_iter) { + npages = min_t(int, pages_iter, pages_to_map - pages_mapped); dma_sync_single_for_cpu(ddev, dma, size, DMA_TO_DEVICE); - npages = populate_xlt(mr, idx, pages_iter, xlt, + npages = populate_xlt(mr, idx, npages, xlt, page_shift, size, flags); dma_sync_single_for_device(ddev, dma, size, DMA_TO_DEVICE); @@ -1687,6 +1688,7 @@ struct ib_mw *mlx5_ib_alloc_mw(struct ib_pd *pd, enum ib_mw_type type, mw->mmkey.type = MLX5_MKEY_MW; mw->ibmw.rkey = mw->mmkey.key; + mw->ndescs = ndescs; resp.response_length = min(offsetof(typeof(resp), response_length) + sizeof(resp.response_length), udata->outlen); @@ -1782,7 +1784,7 @@ mlx5_ib_sg_to_klms(struct mlx5_ib_mr *mr, klms[i].va = cpu_to_be64(sg_dma_address(sg) + sg_offset); klms[i].bcount = cpu_to_be32(sg_dma_len(sg) - sg_offset); klms[i].key = cpu_to_be32(lkey); - mr->ibmr.length += sg_dma_len(sg); + mr->ibmr.length += sg_dma_len(sg) - sg_offset; sg_offset = 0; } diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c index d7b12f0750e2..ae0746754008 100644 --- a/drivers/infiniband/hw/mlx5/odp.c +++ b/drivers/infiniband/hw/mlx5/odp.c @@ -57,7 +57,7 @@ static int check_parent(struct ib_umem_odp *odp, { struct mlx5_ib_mr *mr = odp->private; - return mr && mr->parent == parent; + return mr && mr->parent == parent && !odp->dying; } static struct ib_umem_odp *odp_next(struct ib_umem_odp *odp) @@ -158,13 +158,6 @@ static void mr_leaf_free_action(struct work_struct *work) mr->parent = NULL; synchronize_srcu(&mr->dev->mr_srcu); - if (!READ_ONCE(odp->dying)) { - mr->parent = imr; - if (atomic_dec_and_test(&imr->num_leaf_free)) - wake_up(&imr->q_leaf_free); - return; - } - ib_umem_release(odp->umem); if (imr->live) mlx5_ib_update_xlt(imr, idx, 1, 0, @@ -206,8 +199,8 @@ void mlx5_ib_invalidate_range(struct ib_umem *umem, unsigned long start, * but they will write 0s as well, so no difference in the end result. */ - for (addr = start; addr < end; addr += (u64)umem->page_size) { - idx = (addr - ib_umem_start(umem)) / PAGE_SIZE; + for (addr = start; addr < end; addr += BIT(umem->page_shift)) { + idx = (addr - ib_umem_start(umem)) >> umem->page_shift; /* * Strive to write the MTTs in chunks, but avoid overwriting * non-existing MTTs. The huristic here can be improved to @@ -225,8 +218,7 @@ void mlx5_ib_invalidate_range(struct ib_umem *umem, unsigned long start, if (in_block && umr_offset == 0) { mlx5_ib_update_xlt(mr, blk_start_idx, - idx - blk_start_idx, - PAGE_SHIFT, + idx - blk_start_idx, 0, MLX5_IB_UPD_XLT_ZAP | MLX5_IB_UPD_XLT_ATOMIC); in_block = 0; @@ -235,8 +227,7 @@ void mlx5_ib_invalidate_range(struct ib_umem *umem, unsigned long start, } if (in_block) mlx5_ib_update_xlt(mr, blk_start_idx, - idx - blk_start_idx + 1, - PAGE_SHIFT, + idx - blk_start_idx + 1, 0, MLX5_IB_UPD_XLT_ZAP | MLX5_IB_UPD_XLT_ATOMIC); /* @@ -297,24 +288,6 @@ void mlx5_ib_internal_fill_odp_caps(struct mlx5_ib_dev *dev) return; } -static struct mlx5_ib_mr *mlx5_ib_odp_find_mr_lkey(struct mlx5_ib_dev *dev, - u32 key) -{ - u32 base_key = mlx5_base_mkey(key); - struct mlx5_core_mkey *mmkey = __mlx5_mr_lookup(dev->mdev, base_key); - struct mlx5_ib_mr *mr; - - if (!mmkey || mmkey->key != key || mmkey->type != MLX5_MKEY_MR) - return NULL; - - mr = container_of(mmkey, struct mlx5_ib_mr, mmkey); - - if (!mr->live) - return NULL; - - return container_of(mmkey, struct mlx5_ib_mr, mmkey); -} - static void mlx5_ib_page_fault_resume(struct mlx5_ib_dev *dev, struct mlx5_pagefault *pfault, int error) @@ -436,8 +409,6 @@ next_mr: nentries++; } - odp->dying = 0; - /* Return first odp if region not covered by single one */ if (likely(!result)) result = odp; @@ -522,77 +493,38 @@ void mlx5_ib_free_implicit_mr(struct mlx5_ib_mr *imr) wait_event(imr->q_leaf_free, !atomic_read(&imr->num_leaf_free)); } -/* - * Handle a single data segment in a page-fault WQE or RDMA region. - * - * Returns number of pages retrieved on success. The caller may continue to - * the next data segment. - * Can return the following error codes: - * -EAGAIN to designate a temporary error. The caller will abort handling the - * page fault and resolve it. - * -EFAULT when there's an error mapping the requested pages. The caller will - * abort the page fault handling. - */ -static int pagefault_single_data_segment(struct mlx5_ib_dev *dev, - u32 key, u64 io_virt, size_t bcnt, - u32 *bytes_committed, - u32 *bytes_mapped) +static int pagefault_mr(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr, + u64 io_virt, size_t bcnt, u32 *bytes_mapped) { - int srcu_key; - unsigned int current_seq = 0; - u64 start_idx; - int npages = 0, ret = 0; - struct mlx5_ib_mr *mr; u64 access_mask = ODP_READ_ALLOWED_BIT; + int npages = 0, page_shift, np; + u64 start_idx, page_mask; struct ib_umem_odp *odp; - int implicit = 0; + int current_seq; size_t size; - - srcu_key = srcu_read_lock(&dev->mr_srcu); - mr = mlx5_ib_odp_find_mr_lkey(dev, key); - /* - * If we didn't find the MR, it means the MR was closed while we were - * handling the ODP event. In this case we return -EFAULT so that the - * QP will be closed. - */ - if (!mr || !mr->ibmr.pd) { - mlx5_ib_dbg(dev, "Failed to find relevant mr for lkey=0x%06x, probably the MR was destroyed\n", - key); - ret = -EFAULT; - goto srcu_unlock; - } - if (!mr->umem->odp_data) { - mlx5_ib_dbg(dev, "skipping non ODP MR (lkey=0x%06x) in page fault handler.\n", - key); - if (bytes_mapped) - *bytes_mapped += - (bcnt - *bytes_committed); - goto srcu_unlock; - } - - /* - * Avoid branches - this code will perform correctly - * in all iterations (in iteration 2 and above, - * bytes_committed == 0). - */ - io_virt += *bytes_committed; - bcnt -= *bytes_committed; + int ret; if (!mr->umem->odp_data->page_list) { odp = implicit_mr_get_data(mr, io_virt, bcnt); - if (IS_ERR(odp)) { - ret = PTR_ERR(odp); - goto srcu_unlock; - } + if (IS_ERR(odp)) + return PTR_ERR(odp); mr = odp->private; - implicit = 1; } else { odp = mr->umem->odp_data; } next_mr: + size = min_t(size_t, bcnt, ib_umem_end(odp->umem) - io_virt); + + page_shift = mr->umem->page_shift; + page_mask = ~(BIT(page_shift) - 1); + start_idx = (io_virt - (mr->mmkey.iova & page_mask)) >> page_shift; + + if (mr->umem->writable) + access_mask |= ODP_WRITE_ALLOWED_BIT; + current_seq = READ_ONCE(odp->notifiers_seq); /* * Ensure the sequence number is valid for some time before we call @@ -600,51 +532,43 @@ next_mr: */ smp_rmb(); - size = min_t(size_t, bcnt, ib_umem_end(odp->umem) - io_virt); - start_idx = (io_virt - (mr->mmkey.iova & PAGE_MASK)) >> PAGE_SHIFT; - - if (mr->umem->writable) - access_mask |= ODP_WRITE_ALLOWED_BIT; - ret = ib_umem_odp_map_dma_pages(mr->umem, io_virt, size, access_mask, current_seq); if (ret < 0) - goto srcu_unlock; + goto out; - if (ret > 0) { - int np = ret; - - mutex_lock(&odp->umem_mutex); - if (!ib_umem_mmu_notifier_retry(mr->umem, current_seq)) { - /* - * No need to check whether the MTTs really belong to - * this MR, since ib_umem_odp_map_dma_pages already - * checks this. - */ - ret = mlx5_ib_update_xlt(mr, start_idx, np, - PAGE_SHIFT, - MLX5_IB_UPD_XLT_ATOMIC); - } else { - ret = -EAGAIN; - } - mutex_unlock(&odp->umem_mutex); - if (ret < 0) { - if (ret != -EAGAIN) - mlx5_ib_err(dev, "Failed to update mkey page tables\n"); - goto srcu_unlock; - } + np = ret; - if (bytes_mapped) { - u32 new_mappings = np * PAGE_SIZE - - (io_virt - round_down(io_virt, PAGE_SIZE)); - *bytes_mapped += min_t(u32, new_mappings, size); - } + mutex_lock(&odp->umem_mutex); + if (!ib_umem_mmu_notifier_retry(mr->umem, current_seq)) { + /* + * No need to check whether the MTTs really belong to + * this MR, since ib_umem_odp_map_dma_pages already + * checks this. + */ + ret = mlx5_ib_update_xlt(mr, start_idx, np, + page_shift, MLX5_IB_UPD_XLT_ATOMIC); + } else { + ret = -EAGAIN; + } + mutex_unlock(&odp->umem_mutex); - npages += np; + if (ret < 0) { + if (ret != -EAGAIN) + mlx5_ib_err(dev, "Failed to update mkey page tables\n"); + goto out; + } + + if (bytes_mapped) { + u32 new_mappings = (np << page_shift) - + (io_virt - round_down(io_virt, 1 << page_shift)); + *bytes_mapped += min_t(u32, new_mappings, size); } + npages += np << (page_shift - PAGE_SHIFT); bcnt -= size; + if (unlikely(bcnt)) { struct ib_umem_odp *next; @@ -653,17 +577,18 @@ next_mr: if (unlikely(!next || next->umem->address != io_virt)) { mlx5_ib_dbg(dev, "next implicit leaf removed at 0x%llx. got %p\n", io_virt, next); - ret = -EAGAIN; - goto srcu_unlock_no_wait; + return -EAGAIN; } odp = next; mr = odp->private; goto next_mr; } -srcu_unlock: + return npages; + +out: if (ret == -EAGAIN) { - if (implicit || !odp->dying) { + if (mr->parent || !odp->dying) { unsigned long timeout = msecs_to_jiffies(MMU_NOTIFIER_TIMEOUT); @@ -679,7 +604,155 @@ srcu_unlock: } } -srcu_unlock_no_wait: + return ret; +} + +struct pf_frame { + struct pf_frame *next; + u32 key; + u64 io_virt; + size_t bcnt; + int depth; +}; + +/* + * Handle a single data segment in a page-fault WQE or RDMA region. + * + * Returns number of OS pages retrieved on success. The caller may continue to + * the next data segment. + * Can return the following error codes: + * -EAGAIN to designate a temporary error. The caller will abort handling the + * page fault and resolve it. + * -EFAULT when there's an error mapping the requested pages. The caller will + * abort the page fault handling. + */ +static int pagefault_single_data_segment(struct mlx5_ib_dev *dev, + u32 key, u64 io_virt, size_t bcnt, + u32 *bytes_committed, + u32 *bytes_mapped) +{ + int npages = 0, srcu_key, ret, i, outlen, cur_outlen = 0, depth = 0; + struct pf_frame *head = NULL, *frame; + struct mlx5_core_mkey *mmkey; + struct mlx5_ib_mw *mw; + struct mlx5_ib_mr *mr; + struct mlx5_klm *pklm; + u32 *out = NULL; + size_t offset; + + srcu_key = srcu_read_lock(&dev->mr_srcu); + + io_virt += *bytes_committed; + bcnt -= *bytes_committed; + +next_mr: + mmkey = __mlx5_mr_lookup(dev->mdev, mlx5_base_mkey(key)); + if (!mmkey || mmkey->key != key) { + mlx5_ib_dbg(dev, "failed to find mkey %x\n", key); + ret = -EFAULT; + goto srcu_unlock; + } + + switch (mmkey->type) { + case MLX5_MKEY_MR: + mr = container_of(mmkey, struct mlx5_ib_mr, mmkey); + if (!mr->live || !mr->ibmr.pd) { + mlx5_ib_dbg(dev, "got dead MR\n"); + ret = -EFAULT; + goto srcu_unlock; + } + + ret = pagefault_mr(dev, mr, io_virt, bcnt, bytes_mapped); + if (ret < 0) + goto srcu_unlock; + + npages += ret; + ret = 0; + break; + + case MLX5_MKEY_MW: + mw = container_of(mmkey, struct mlx5_ib_mw, mmkey); + + if (depth >= MLX5_CAP_GEN(dev->mdev, max_indirection)) { + mlx5_ib_dbg(dev, "indirection level exceeded\n"); + ret = -EFAULT; + goto srcu_unlock; + } + + outlen = MLX5_ST_SZ_BYTES(query_mkey_out) + + sizeof(*pklm) * (mw->ndescs - 2); + + if (outlen > cur_outlen) { + kfree(out); + out = kzalloc(outlen, GFP_KERNEL); + if (!out) { + ret = -ENOMEM; + goto srcu_unlock; + } + cur_outlen = outlen; + } + + pklm = (struct mlx5_klm *)MLX5_ADDR_OF(query_mkey_out, out, + bsf0_klm0_pas_mtt0_1); + + ret = mlx5_core_query_mkey(dev->mdev, &mw->mmkey, out, outlen); + if (ret) + goto srcu_unlock; + + offset = io_virt - MLX5_GET64(query_mkey_out, out, + memory_key_mkey_entry.start_addr); + + for (i = 0; bcnt && i < mw->ndescs; i++, pklm++) { + if (offset >= be32_to_cpu(pklm->bcount)) { + offset -= be32_to_cpu(pklm->bcount); + continue; + } + + frame = kzalloc(sizeof(*frame), GFP_KERNEL); + if (!frame) { + ret = -ENOMEM; + goto srcu_unlock; + } + + frame->key = be32_to_cpu(pklm->key); + frame->io_virt = be64_to_cpu(pklm->va) + offset; + frame->bcnt = min_t(size_t, bcnt, + be32_to_cpu(pklm->bcount) - offset); + frame->depth = depth + 1; + frame->next = head; + head = frame; + + bcnt -= frame->bcnt; + } + break; + + default: + mlx5_ib_dbg(dev, "wrong mkey type %d\n", mmkey->type); + ret = -EFAULT; + goto srcu_unlock; + } + + if (head) { + frame = head; + head = frame->next; + + key = frame->key; + io_virt = frame->io_virt; + bcnt = frame->bcnt; + depth = frame->depth; + kfree(frame); + + goto next_mr; + } + +srcu_unlock: + while (head) { + frame = head; + head = frame->next; + kfree(frame); + } + kfree(out); + srcu_read_unlock(&dev->mr_srcu, srcu_key); *bytes_committed = 0; return ret ? ret : npages; @@ -997,9 +1070,6 @@ static void mlx5_ib_mr_wqe_pfault_handler(struct mlx5_ib_dev *dev, resume_with_error = 0; goto resolve_page_fault; } else if (ret < 0 || total_wqe_bytes > bytes_mapped) { - if (ret != -ENOENT) - mlx5_ib_err(dev, "PAGE FAULT error: %d. QP 0x%x. type: 0x%x\n", - ret, pfault->wqe.wq_num, pfault->type); goto resolve_page_fault; } @@ -1059,8 +1129,8 @@ static void mlx5_ib_mr_rdma_pfault_handler(struct mlx5_ib_dev *dev, } else if (ret < 0 || pages_in_range(address, length) > ret) { mlx5_ib_page_fault_resume(dev, pfault, 1); if (ret != -ENOENT) - mlx5_ib_warn(dev, "PAGE FAULT error %d. QP 0x%x, type: 0x%x\n", - ret, pfault->token, pfault->type); + mlx5_ib_dbg(dev, "PAGE FAULT error %d. QP 0x%x, type: 0x%x\n", + ret, pfault->token, pfault->type); return; } @@ -1081,8 +1151,8 @@ static void mlx5_ib_mr_rdma_pfault_handler(struct mlx5_ib_dev *dev, prefetch_len, &bytes_committed, NULL); if (ret < 0 && ret != -EAGAIN) { - mlx5_ib_warn(dev, "Prefetch failed. ret: %d, QP 0x%x, address: 0x%.16llx, length = 0x%.16x\n", - ret, pfault->token, address, prefetch_len); + mlx5_ib_dbg(dev, "Prefetch failed. ret: %d, QP 0x%x, address: 0x%.16llx, length = 0x%.16x\n", + ret, pfault->token, address, prefetch_len); } } } diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index ed6320186f89..93959e1e43a3 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -2206,63 +2206,65 @@ static int modify_raw_packet_tx_affinity(struct mlx5_core_dev *dev, } static int mlx5_set_path(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp, - const struct ib_ah_attr *ah, + const struct rdma_ah_attr *ah, struct mlx5_qp_path *path, u8 port, int attr_mask, u32 path_flags, const struct ib_qp_attr *attr, bool alt) { - enum rdma_link_layer ll = rdma_port_get_link_layer(&dev->ib_dev, port); + const struct ib_global_route *grh = rdma_ah_read_grh(ah); int err; enum ib_gid_type gid_type; + u8 ah_flags = rdma_ah_get_ah_flags(ah); + u8 sl = rdma_ah_get_sl(ah); if (attr_mask & IB_QP_PKEY_INDEX) path->pkey_index = cpu_to_be16(alt ? attr->alt_pkey_index : attr->pkey_index); - if (ah->ah_flags & IB_AH_GRH) { - if (ah->grh.sgid_index >= + if (ah_flags & IB_AH_GRH) { + if (grh->sgid_index >= dev->mdev->port_caps[port - 1].gid_table_len) { pr_err("sgid_index (%u) too large. max is %d\n", - ah->grh.sgid_index, + grh->sgid_index, dev->mdev->port_caps[port - 1].gid_table_len); return -EINVAL; } } - if (ll == IB_LINK_LAYER_ETHERNET) { - if (!(ah->ah_flags & IB_AH_GRH)) + if (ah->type == RDMA_AH_ATTR_TYPE_ROCE) { + if (!(ah_flags & IB_AH_GRH)) return -EINVAL; - err = mlx5_get_roce_gid_type(dev, port, ah->grh.sgid_index, + err = mlx5_get_roce_gid_type(dev, port, grh->sgid_index, &gid_type); if (err) return err; - memcpy(path->rmac, ah->dmac, sizeof(ah->dmac)); + memcpy(path->rmac, ah->roce.dmac, sizeof(ah->roce.dmac)); path->udp_sport = mlx5_get_roce_udp_sport(dev, port, - ah->grh.sgid_index); - path->dci_cfi_prio_sl = (ah->sl & 0x7) << 4; + grh->sgid_index); + path->dci_cfi_prio_sl = (sl & 0x7) << 4; if (gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP) - path->ecn_dscp = (ah->grh.traffic_class >> 2) & 0x3f; + path->ecn_dscp = (grh->traffic_class >> 2) & 0x3f; } else { path->fl_free_ar = (path_flags & MLX5_PATH_FLAG_FL) ? 0x80 : 0; path->fl_free_ar |= (path_flags & MLX5_PATH_FLAG_FREE_AR) ? 0x40 : 0; - path->rlid = cpu_to_be16(ah->dlid); - path->grh_mlid = ah->src_path_bits & 0x7f; - if (ah->ah_flags & IB_AH_GRH) + path->rlid = cpu_to_be16(rdma_ah_get_dlid(ah)); + path->grh_mlid = rdma_ah_get_path_bits(ah) & 0x7f; + if (ah_flags & IB_AH_GRH) path->grh_mlid |= 1 << 7; - path->dci_cfi_prio_sl = ah->sl & 0xf; + path->dci_cfi_prio_sl = sl & 0xf; } - if (ah->ah_flags & IB_AH_GRH) { - path->mgid_index = ah->grh.sgid_index; - path->hop_limit = ah->grh.hop_limit; + if (ah_flags & IB_AH_GRH) { + path->mgid_index = grh->sgid_index; + path->hop_limit = grh->hop_limit; path->tclass_flowlabel = - cpu_to_be32((ah->grh.traffic_class << 20) | - (ah->grh.flow_label)); - memcpy(path->rgid, ah->grh.dgid.raw, 16); + cpu_to_be32((grh->traffic_class << 20) | + (grh->flow_label)); + memcpy(path->rgid, grh->dgid.raw, 16); } - err = ib_rate_to_mlx5(dev, ah->static_rate); + err = ib_rate_to_mlx5(dev, rdma_ah_get_static_rate(ah)); if (err < 0) return err; path->static_rate = err; @@ -2274,7 +2276,7 @@ static int mlx5_set_path(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp, if ((qp->ibqp.qp_type == IB_QPT_RAW_PACKET) && qp->sq.wqe_cnt) return modify_raw_packet_eth_prio(dev->mdev, &qp->raw_packet_qp.sq, - ah->sl & 0xf); + sl & 0xf); return 0; } @@ -2799,7 +2801,7 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp, qp->port) - 1; mibport = &dev->port[port_num]; context->qp_counter_set_usr_page |= - cpu_to_be32((u32)(mibport->q_cnts.set_id) << 24); + cpu_to_be32((u32)(mibport->cnts.set_id) << 24); } if (!ibqp->uobject && cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) @@ -2827,7 +2829,7 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp, raw_qp_param.operation = op; if (cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) { - raw_qp_param.rq_q_ctr_id = mibport->q_cnts.set_id; + raw_qp_param.rq_q_ctr_id = mibport->cnts.set_id; raw_qp_param.set_mask |= MLX5_RAW_QP_MOD_SET_RQ_Q_CTR_ID; } @@ -4249,33 +4251,36 @@ static int to_ib_qp_access_flags(int mlx5_flags) return ib_flags; } -static void to_ib_ah_attr(struct mlx5_ib_dev *ibdev, struct ib_ah_attr *ib_ah_attr, - struct mlx5_qp_path *path) +static void to_rdma_ah_attr(struct mlx5_ib_dev *ibdev, + struct rdma_ah_attr *ah_attr, + struct mlx5_qp_path *path) { struct mlx5_core_dev *dev = ibdev->mdev; - memset(ib_ah_attr, 0, sizeof(*ib_ah_attr)); - ib_ah_attr->port_num = path->port; + memset(ah_attr, 0, sizeof(*ah_attr)); - if (ib_ah_attr->port_num == 0 || - ib_ah_attr->port_num > MLX5_CAP_GEN(dev, num_ports)) + ah_attr->type = rdma_ah_find_type(&ibdev->ib_dev, path->port); + rdma_ah_set_port_num(ah_attr, path->port); + if (rdma_ah_get_port_num(ah_attr) == 0 || + rdma_ah_get_port_num(ah_attr) > MLX5_CAP_GEN(dev, num_ports)) return; - ib_ah_attr->sl = path->dci_cfi_prio_sl & 0xf; + rdma_ah_set_port_num(ah_attr, path->port); + rdma_ah_set_sl(ah_attr, path->dci_cfi_prio_sl & 0xf); - ib_ah_attr->dlid = be16_to_cpu(path->rlid); - ib_ah_attr->src_path_bits = path->grh_mlid & 0x7f; - ib_ah_attr->static_rate = path->static_rate ? path->static_rate - 5 : 0; - ib_ah_attr->ah_flags = (path->grh_mlid & (1 << 7)) ? IB_AH_GRH : 0; - if (ib_ah_attr->ah_flags) { - ib_ah_attr->grh.sgid_index = path->mgid_index; - ib_ah_attr->grh.hop_limit = path->hop_limit; - ib_ah_attr->grh.traffic_class = - (be32_to_cpu(path->tclass_flowlabel) >> 20) & 0xff; - ib_ah_attr->grh.flow_label = - be32_to_cpu(path->tclass_flowlabel) & 0xfffff; - memcpy(ib_ah_attr->grh.dgid.raw, - path->rgid, sizeof(ib_ah_attr->grh.dgid.raw)); + rdma_ah_set_dlid(ah_attr, be16_to_cpu(path->rlid)); + rdma_ah_set_path_bits(ah_attr, path->grh_mlid & 0x7f); + rdma_ah_set_static_rate(ah_attr, + path->static_rate ? path->static_rate - 5 : 0); + if (path->grh_mlid & (1 << 7)) { + u32 tc_fl = be32_to_cpu(path->tclass_flowlabel); + + rdma_ah_set_grh(ah_attr, NULL, + tc_fl & 0xfffff, + path->mgid_index, + path->hop_limit, + (tc_fl >> 20) & 0xff); + rdma_ah_set_dgid_raw(ah_attr, path->rgid); } } @@ -4440,11 +4445,12 @@ static int query_qp_attr(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp, to_ib_qp_access_flags(be32_to_cpu(context->params2)); if (qp->ibqp.qp_type == IB_QPT_RC || qp->ibqp.qp_type == IB_QPT_UC) { - to_ib_ah_attr(dev, &qp_attr->ah_attr, &context->pri_path); - to_ib_ah_attr(dev, &qp_attr->alt_ah_attr, &context->alt_path); + to_rdma_ah_attr(dev, &qp_attr->ah_attr, &context->pri_path); + to_rdma_ah_attr(dev, &qp_attr->alt_ah_attr, &context->alt_path); qp_attr->alt_pkey_index = be16_to_cpu(context->alt_path.pkey_index); - qp_attr->alt_port_num = qp_attr->alt_ah_attr.port_num; + qp_attr->alt_port_num = + rdma_ah_get_port_num(&qp_attr->alt_ah_attr); } qp_attr->pkey_index = be16_to_cpu(context->pri_path.pkey_index); @@ -4965,7 +4971,8 @@ int mlx5_ib_modify_wq(struct ib_wq *wq, struct ib_wq_attr *wq_attr, if (MLX5_CAP_GEN(dev->mdev, modify_rq_counter_set_id)) { MLX5_SET64(modify_rq_in, in, modify_bitmask, MLX5_MODIFY_RQ_IN_MODIFY_BITMASK_RQ_COUNTER_SET_ID); - MLX5_SET(rqc, rqc, counter_set_id, dev->port->q_cnts.set_id); + MLX5_SET(rqc, rqc, counter_set_id, + dev->port->cnts.set_id); } else pr_info_once("%s: Receive WQ counters are not supported on current FW\n", dev->ib_dev.name); diff --git a/drivers/infiniband/hw/mthca/mthca_av.c b/drivers/infiniband/hw/mthca/mthca_av.c index c9f0f364f484..2aec9908c40a 100644 --- a/drivers/infiniband/hw/mthca/mthca_av.c +++ b/drivers/infiniband/hw/mthca/mthca_av.c @@ -152,7 +152,7 @@ u8 mthca_get_rate(struct mthca_dev *dev, int static_rate, u8 port) int mthca_create_ah(struct mthca_dev *dev, struct mthca_pd *pd, - struct ib_ah_attr *ah_attr, + struct rdma_ah_attr *ah_attr, struct mthca_ah *ah) { u32 index = -1; @@ -196,21 +196,26 @@ on_hca_fail: ah->key = pd->ntmr.ibmr.lkey; - av->port_pd = cpu_to_be32(pd->pd_num | (ah_attr->port_num << 24)); - av->g_slid = ah_attr->src_path_bits; - av->dlid = cpu_to_be16(ah_attr->dlid); + av->port_pd = cpu_to_be32(pd->pd_num | + (rdma_ah_get_port_num(ah_attr) << 24)); + av->g_slid = rdma_ah_get_path_bits(ah_attr); + av->dlid = cpu_to_be16(rdma_ah_get_dlid(ah_attr)); av->msg_sr = (3 << 4) | /* 2K message */ - mthca_get_rate(dev, ah_attr->static_rate, ah_attr->port_num); - av->sl_tclass_flowlabel = cpu_to_be32(ah_attr->sl << 28); - if (ah_attr->ah_flags & IB_AH_GRH) { + mthca_get_rate(dev, rdma_ah_get_static_rate(ah_attr), + rdma_ah_get_port_num(ah_attr)); + av->sl_tclass_flowlabel = cpu_to_be32(rdma_ah_get_sl(ah_attr) << 28); + if (rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH) { + const struct ib_global_route *grh = rdma_ah_read_grh(ah_attr); + av->g_slid |= 0x80; - av->gid_index = (ah_attr->port_num - 1) * dev->limits.gid_table_len + - ah_attr->grh.sgid_index; - av->hop_limit = ah_attr->grh.hop_limit; + av->gid_index = (rdma_ah_get_port_num(ah_attr) - 1) * + dev->limits.gid_table_len + + grh->sgid_index; + av->hop_limit = grh->hop_limit; av->sl_tclass_flowlabel |= - cpu_to_be32((ah_attr->grh.traffic_class << 20) | - ah_attr->grh.flow_label); - memcpy(av->dgid, ah_attr->grh.dgid.raw, 16); + cpu_to_be32((grh->traffic_class << 20) | + grh->flow_label); + memcpy(av->dgid, grh->dgid.raw, 16); } else { /* Arbel workaround -- low byte of GID must be 2 */ av->dgid[3] = cpu_to_be32(2); @@ -287,33 +292,35 @@ int mthca_read_ah(struct mthca_dev *dev, struct mthca_ah *ah, return 0; } -int mthca_ah_query(struct ib_ah *ibah, struct ib_ah_attr *attr) +int mthca_ah_query(struct ib_ah *ibah, struct rdma_ah_attr *attr) { struct mthca_ah *ah = to_mah(ibah); struct mthca_dev *dev = to_mdev(ibah->device); + u8 port_num = be32_to_cpu(ah->av->port_pd) >> 24; /* Only implement for MAD and memfree ah for now. */ if (ah->type == MTHCA_AH_ON_HCA) return -ENOSYS; memset(attr, 0, sizeof *attr); - attr->dlid = be16_to_cpu(ah->av->dlid); - attr->sl = be32_to_cpu(ah->av->sl_tclass_flowlabel) >> 28; - attr->port_num = be32_to_cpu(ah->av->port_pd) >> 24; - attr->static_rate = mthca_rate_to_ib(dev, ah->av->msg_sr & 0x7, - attr->port_num); - attr->src_path_bits = ah->av->g_slid & 0x7F; - attr->ah_flags = mthca_ah_grh_present(ah) ? IB_AH_GRH : 0; - - if (attr->ah_flags) { - attr->grh.traffic_class = - be32_to_cpu(ah->av->sl_tclass_flowlabel) >> 20; - attr->grh.flow_label = - be32_to_cpu(ah->av->sl_tclass_flowlabel) & 0xfffff; - attr->grh.hop_limit = ah->av->hop_limit; - attr->grh.sgid_index = ah->av->gid_index & - (dev->limits.gid_table_len - 1); - memcpy(attr->grh.dgid.raw, ah->av->dgid, 16); + attr->type = ibah->type; + rdma_ah_set_dlid(attr, be16_to_cpu(ah->av->dlid)); + rdma_ah_set_sl(attr, be32_to_cpu(ah->av->sl_tclass_flowlabel) >> 28); + rdma_ah_set_port_num(attr, port_num); + rdma_ah_set_static_rate(attr, + mthca_rate_to_ib(dev, ah->av->msg_sr & 0x7, + port_num)); + rdma_ah_set_path_bits(attr, ah->av->g_slid & 0x7F); + if (mthca_ah_grh_present(ah)) { + u32 tc_fl = be32_to_cpu(ah->av->sl_tclass_flowlabel); + + rdma_ah_set_grh(attr, NULL, + tc_fl & 0xfffff, + ah->av->gid_index & + (dev->limits.gid_table_len - 1), + ah->av->hop_limit, + (tc_fl >> 20) & 0xff); + rdma_ah_set_dgid_raw(attr, ah->av->dgid); } return 0; diff --git a/drivers/infiniband/hw/mthca/mthca_cmd.c b/drivers/infiniband/hw/mthca/mthca_cmd.c index c7f49bbb0c72..9d83a53c0c67 100644 --- a/drivers/infiniband/hw/mthca/mthca_cmd.c +++ b/drivers/infiniband/hw/mthca/mthca_cmd.c @@ -367,12 +367,16 @@ static int mthca_cmd_poll(struct mthca_dev *dev, goto out; } - if (out_is_imm) + if (out_is_imm && out_param) { *out_param = (u64) be32_to_cpu((__force __be32) __raw_readl(dev->hcr + HCR_OUT_PARAM_OFFSET)) << 32 | (u64) be32_to_cpu((__force __be32) __raw_readl(dev->hcr + HCR_OUT_PARAM_OFFSET + 4)); + } else if (out_is_imm) { + err = -EINVAL; + goto out; + } status = be32_to_cpu((__force __be32) __raw_readl(dev->hcr + HCR_STATUS_OFFSET)) >> 24; if (status) { @@ -450,8 +454,12 @@ static int mthca_cmd_wait(struct mthca_dev *dev, err = mthca_status_to_errno(context->status); } - if (out_is_imm) + if (out_is_imm && out_param) { *out_param = context->out_param; + } else if (out_is_imm) { + err = -EINVAL; + goto out; + } out: spin_lock(&dev->cmd.context_lock); diff --git a/drivers/infiniband/hw/mthca/mthca_dev.h b/drivers/infiniband/hw/mthca/mthca_dev.h index 4393a022867b..ec7da9a474cd 100644 --- a/drivers/infiniband/hw/mthca/mthca_dev.h +++ b/drivers/infiniband/hw/mthca/mthca_dev.h @@ -560,12 +560,12 @@ int mthca_alloc_sqp(struct mthca_dev *dev, void mthca_free_qp(struct mthca_dev *dev, struct mthca_qp *qp); int mthca_create_ah(struct mthca_dev *dev, struct mthca_pd *pd, - struct ib_ah_attr *ah_attr, + struct rdma_ah_attr *ah_attr, struct mthca_ah *ah); int mthca_destroy_ah(struct mthca_dev *dev, struct mthca_ah *ah); int mthca_read_ah(struct mthca_dev *dev, struct mthca_ah *ah, struct ib_ud_header *header); -int mthca_ah_query(struct ib_ah *ibah, struct ib_ah_attr *attr); +int mthca_ah_query(struct ib_ah *ibah, struct rdma_ah_attr *attr); int mthca_ah_grh_present(struct mthca_ah *ah); u8 mthca_get_rate(struct mthca_dev *dev, int static_rate, u8 port); enum ib_rate mthca_rate_to_ib(struct mthca_dev *dev, u8 mthca_rate, u8 port); diff --git a/drivers/infiniband/hw/mthca/mthca_mad.c b/drivers/infiniband/hw/mthca/mthca_mad.c index 9139405c4810..7df3db71777a 100644 --- a/drivers/infiniband/hw/mthca/mthca_mad.c +++ b/drivers/infiniband/hw/mthca/mthca_mad.c @@ -75,25 +75,26 @@ static void update_sm_ah(struct mthca_dev *dev, u8 port_num, u16 lid, u8 sl) { struct ib_ah *new_ah; - struct ib_ah_attr ah_attr; + struct rdma_ah_attr ah_attr; unsigned long flags; if (!dev->send_agent[port_num - 1][0]) return; memset(&ah_attr, 0, sizeof ah_attr); - ah_attr.dlid = lid; - ah_attr.sl = sl; - ah_attr.port_num = port_num; + ah_attr.type = rdma_ah_find_type(&dev->ib_dev, port_num); + rdma_ah_set_dlid(&ah_attr, lid); + rdma_ah_set_sl(&ah_attr, sl); + rdma_ah_set_port_num(&ah_attr, port_num); - new_ah = ib_create_ah(dev->send_agent[port_num - 1][0]->qp->pd, - &ah_attr); + new_ah = rdma_create_ah(dev->send_agent[port_num - 1][0]->qp->pd, + &ah_attr); if (IS_ERR(new_ah)) return; spin_lock_irqsave(&dev->sm_lock, flags); if (dev->sm_ah[port_num - 1]) - ib_destroy_ah(dev->sm_ah[port_num - 1]); + rdma_destroy_ah(dev->sm_ah[port_num - 1]); dev->sm_ah[port_num - 1] = new_ah; spin_unlock_irqrestore(&dev->sm_lock, flags); } @@ -345,6 +346,6 @@ void mthca_free_agents(struct mthca_dev *dev) } if (dev->sm_ah[p]) - ib_destroy_ah(dev->sm_ah[p]); + rdma_destroy_ah(dev->sm_ah[p]); } } diff --git a/drivers/infiniband/hw/mthca/mthca_provider.c b/drivers/infiniband/hw/mthca/mthca_provider.c index 22d0e6ee5af6..c197cd9b193f 100644 --- a/drivers/infiniband/hw/mthca/mthca_provider.c +++ b/drivers/infiniband/hw/mthca/mthca_provider.c @@ -410,7 +410,7 @@ static int mthca_dealloc_pd(struct ib_pd *pd) } static struct ib_ah *mthca_ah_create(struct ib_pd *pd, - struct ib_ah_attr *ah_attr, + struct rdma_ah_attr *ah_attr, struct ib_udata *udata) { @@ -937,7 +937,7 @@ static struct ib_mr *mthca_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, goto err; } - shift = ffs(mr->umem->page_size) - 1; + shift = mr->umem->page_shift; n = mr->umem->nmap; mr->mtt = mthca_alloc_mtt(dev, n); @@ -959,8 +959,7 @@ static struct ib_mr *mthca_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, for_each_sg(mr->umem->sg_head.sgl, sg, mr->umem->nmap, entry) { len = sg_dma_len(sg) >> shift; for (k = 0; k < len; ++k) { - pages[i++] = sg_dma_address(sg) + - mr->umem->page_size * k; + pages[i++] = sg_dma_address(sg) + (k << shift); /* * Be friendly to write_mtt and pass it chunks * of appropriate size. diff --git a/drivers/infiniband/hw/mthca/mthca_qp.c b/drivers/infiniband/hw/mthca/mthca_qp.c index 96e5fb91fb48..d21960cd9a49 100644 --- a/drivers/infiniband/hw/mthca/mthca_qp.c +++ b/drivers/infiniband/hw/mthca/mthca_qp.c @@ -393,31 +393,36 @@ static int to_ib_qp_access_flags(int mthca_flags) return ib_flags; } -static void to_ib_ah_attr(struct mthca_dev *dev, struct ib_ah_attr *ib_ah_attr, - struct mthca_qp_path *path) +static void to_rdma_ah_attr(struct mthca_dev *dev, + struct rdma_ah_attr *ah_attr, + struct mthca_qp_path *path) { - memset(ib_ah_attr, 0, sizeof *ib_ah_attr); - ib_ah_attr->port_num = (be32_to_cpu(path->port_pkey) >> 24) & 0x3; + u8 port_num = (be32_to_cpu(path->port_pkey) >> 24) & 0x3; - if (ib_ah_attr->port_num == 0 || ib_ah_attr->port_num > dev->limits.num_ports) - return; + memset(ah_attr, 0, sizeof(*ah_attr)); - ib_ah_attr->dlid = be16_to_cpu(path->rlid); - ib_ah_attr->sl = be32_to_cpu(path->sl_tclass_flowlabel) >> 28; - ib_ah_attr->src_path_bits = path->g_mylmc & 0x7f; - ib_ah_attr->static_rate = mthca_rate_to_ib(dev, - path->static_rate & 0xf, - ib_ah_attr->port_num); - ib_ah_attr->ah_flags = (path->g_mylmc & (1 << 7)) ? IB_AH_GRH : 0; - if (ib_ah_attr->ah_flags) { - ib_ah_attr->grh.sgid_index = path->mgid_index & (dev->limits.gid_table_len - 1); - ib_ah_attr->grh.hop_limit = path->hop_limit; - ib_ah_attr->grh.traffic_class = - (be32_to_cpu(path->sl_tclass_flowlabel) >> 20) & 0xff; - ib_ah_attr->grh.flow_label = - be32_to_cpu(path->sl_tclass_flowlabel) & 0xfffff; - memcpy(ib_ah_attr->grh.dgid.raw, - path->rgid, sizeof ib_ah_attr->grh.dgid.raw); + if (port_num == 0 || port_num > dev->limits.num_ports) + return; + ah_attr->type = rdma_ah_find_type(&dev->ib_dev, port_num); + rdma_ah_set_port_num(ah_attr, port_num); + + rdma_ah_set_dlid(ah_attr, be16_to_cpu(path->rlid)); + rdma_ah_set_sl(ah_attr, be32_to_cpu(path->sl_tclass_flowlabel) >> 28); + rdma_ah_set_path_bits(ah_attr, path->g_mylmc & 0x7f); + rdma_ah_set_static_rate(ah_attr, + mthca_rate_to_ib(dev, + path->static_rate & 0xf, + port_num)); + if (path->g_mylmc & (1 << 7)) { + u32 tc_fl = be32_to_cpu(path->sl_tclass_flowlabel); + + rdma_ah_set_grh(ah_attr, NULL, + tc_fl & 0xfffff, + path->mgid_index & + (dev->limits.gid_table_len - 1), + path->hop_limit, + (tc_fl >> 20) & 0xff); + rdma_ah_set_dgid_raw(ah_attr, path->rgid); } } @@ -468,11 +473,12 @@ int mthca_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, int qp_attr_m to_ib_qp_access_flags(be32_to_cpu(context->params2)); if (qp->transport == RC || qp->transport == UC) { - to_ib_ah_attr(dev, &qp_attr->ah_attr, &context->pri_path); - to_ib_ah_attr(dev, &qp_attr->alt_ah_attr, &context->alt_path); + to_rdma_ah_attr(dev, &qp_attr->ah_attr, &context->pri_path); + to_rdma_ah_attr(dev, &qp_attr->alt_ah_attr, &context->alt_path); qp_attr->alt_pkey_index = be32_to_cpu(context->alt_path.port_pkey) & 0x7f; - qp_attr->alt_port_num = qp_attr->alt_ah_attr.port_num; + qp_attr->alt_port_num = + rdma_ah_get_port_num(&qp_attr->alt_ah_attr); } qp_attr->pkey_index = be32_to_cpu(context->pri_path.port_pkey) & 0x7f; @@ -512,30 +518,36 @@ out: return err; } -static int mthca_path_set(struct mthca_dev *dev, const struct ib_ah_attr *ah, +static int mthca_path_set(struct mthca_dev *dev, const struct rdma_ah_attr *ah, struct mthca_qp_path *path, u8 port) { - path->g_mylmc = ah->src_path_bits & 0x7f; - path->rlid = cpu_to_be16(ah->dlid); - path->static_rate = mthca_get_rate(dev, ah->static_rate, port); + path->g_mylmc = rdma_ah_get_path_bits(ah) & 0x7f; + path->rlid = cpu_to_be16(rdma_ah_get_dlid(ah)); + path->static_rate = mthca_get_rate(dev, rdma_ah_get_static_rate(ah), + port); + + if (rdma_ah_get_ah_flags(ah) & IB_AH_GRH) { + const struct ib_global_route *grh = rdma_ah_read_grh(ah); - if (ah->ah_flags & IB_AH_GRH) { - if (ah->grh.sgid_index >= dev->limits.gid_table_len) { + if (grh->sgid_index >= dev->limits.gid_table_len) { mthca_dbg(dev, "sgid_index (%u) too large. max is %d\n", - ah->grh.sgid_index, dev->limits.gid_table_len-1); + grh->sgid_index, + dev->limits.gid_table_len - 1); return -1; } path->g_mylmc |= 1 << 7; - path->mgid_index = ah->grh.sgid_index; - path->hop_limit = ah->grh.hop_limit; + path->mgid_index = grh->sgid_index; + path->hop_limit = grh->hop_limit; path->sl_tclass_flowlabel = - cpu_to_be32((ah->sl << 28) | - (ah->grh.traffic_class << 20) | - (ah->grh.flow_label)); - memcpy(path->rgid, ah->grh.dgid.raw, 16); - } else - path->sl_tclass_flowlabel = cpu_to_be32(ah->sl << 28); + cpu_to_be32((rdma_ah_get_sl(ah) << 28) | + (grh->traffic_class << 20) | + (grh->flow_label)); + memcpy(path->rgid, grh->dgid.raw, 16); + } else { + path->sl_tclass_flowlabel = cpu_to_be32(rdma_ah_get_sl(ah) << + 28); + } return 0; } @@ -680,7 +692,7 @@ static int __mthca_modify_qp(struct ib_qp *ibqp, } if (mthca_path_set(dev, &attr->alt_ah_attr, &qp_context->alt_path, - attr->alt_ah_attr.port_num)) + rdma_ah_get_port_num(&attr->alt_ah_attr))) goto out_mailbox; qp_context->alt_path.port_pkey |= cpu_to_be32(attr->alt_pkey_index | diff --git a/drivers/infiniband/hw/nes/nes_hw.c b/drivers/infiniband/hw/nes/nes_hw.c index 19acd13c6cb1..8f9d8b4ad583 100644 --- a/drivers/infiniband/hw/nes/nes_hw.c +++ b/drivers/infiniband/hw/nes/nes_hw.c @@ -1849,9 +1849,8 @@ int nes_init_nic_qp(struct nes_device *nesdev, struct net_device *netdev) wqe_count -= counter; nes_write32(nesdev->regs+NES_WQE_ALLOC, (counter << 24) | nesvnic->nic.qp_id); } while (wqe_count); - init_timer(&nesvnic->rq_wqes_timer); - nesvnic->rq_wqes_timer.function = nes_rq_wqes_timeout; - nesvnic->rq_wqes_timer.data = (unsigned long)nesvnic; + setup_timer(&nesvnic->rq_wqes_timer, nes_rq_wqes_timeout, + (unsigned long)nesvnic); nes_debug(NES_DBG_INIT, "NAPI support Enabled\n"); if (nesdev->nesadapter->et_use_adaptive_rx_coalesce) { @@ -3055,7 +3054,7 @@ static void nes_cqp_ce_handler(struct nes_device *nesdev, struct nes_hw_cq *cq) memcpy(cqp_wqe, &cqp_request->cqp_wqe, sizeof(*cqp_wqe)); barrier(); - opcode = cqp_wqe->wqe_words[NES_CQP_WQE_OPCODE_IDX]; + opcode = le32_to_cpu(cqp_wqe->wqe_words[NES_CQP_WQE_OPCODE_IDX]); if ((opcode & NES_CQP_OPCODE_MASK) == NES_CQP_DOWNLOAD_SEGMENT) ctx_index = NES_CQP_WQE_DL_COMP_CTX_LOW_IDX; else diff --git a/drivers/infiniband/hw/nes/nes_mgt.c b/drivers/infiniband/hw/nes/nes_mgt.c index 33624f17c347..77226cf4ea02 100644 --- a/drivers/infiniband/hw/nes/nes_mgt.c +++ b/drivers/infiniband/hw/nes/nes_mgt.c @@ -1040,9 +1040,8 @@ int nes_init_mgt_qp(struct nes_device *nesdev, struct net_device *netdev, struct mgtvnic->mgt.rx_skb[counter] = skb; } - init_timer(&mgtvnic->rq_wqes_timer); - mgtvnic->rq_wqes_timer.function = nes_mgt_rq_wqes_timeout; - mgtvnic->rq_wqes_timer.data = (unsigned long)mgtvnic; + setup_timer(&mgtvnic->rq_wqes_timer, nes_mgt_rq_wqes_timeout, + (unsigned long)mgtvnic); wqe_count = NES_MGT_WQ_COUNT - 1; mgtvnic->mgt.rq_head = wqe_count; diff --git a/drivers/infiniband/hw/nes/nes_verbs.c b/drivers/infiniband/hw/nes/nes_verbs.c index ccf0a4cffe9c..25dcd7573df9 100644 --- a/drivers/infiniband/hw/nes/nes_verbs.c +++ b/drivers/infiniband/hw/nes/nes_verbs.c @@ -761,7 +761,8 @@ static int nes_dealloc_pd(struct ib_pd *ibpd) /** * nes_create_ah */ -static struct ib_ah *nes_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, +static struct ib_ah *nes_create_ah(struct ib_pd *pd, + struct rdma_ah_attr *ah_attr, struct ib_udata *udata) { return ERR_PTR(-ENOSYS); @@ -1308,9 +1309,8 @@ static struct ib_qp *nes_create_qp(struct ib_pd *ibpd, init_completion(&nesqp->rq_drained); nesqp->sig_all = (init_attr->sq_sig_type == IB_SIGNAL_ALL_WR); - init_timer(&nesqp->terminate_timer); - nesqp->terminate_timer.function = nes_terminate_timeout; - nesqp->terminate_timer.data = (unsigned long)nesqp; + setup_timer(&nesqp->terminate_timer, nes_terminate_timeout, + (unsigned long)nesqp); /* update the QP table */ nesdev->nesadapter->qp_table[nesqp->hwqp.qp_id-NES_FIRST_QPN] = nesqp; @@ -2165,9 +2165,9 @@ static struct ib_mr *nes_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, } nes_debug(NES_DBG_MR, "User base = 0x%lX, Virt base = 0x%lX, length = %u," - " offset = %u, page size = %u.\n", + " offset = %u, page size = %lu.\n", (unsigned long int)start, (unsigned long int)virt, (u32)length, - ib_umem_offset(region), region->page_size); + ib_umem_offset(region), BIT(region->page_shift)); skip_pages = ((u32)ib_umem_offset(region)) >> 12; diff --git a/drivers/infiniband/hw/ocrdma/ocrdma.h b/drivers/infiniband/hw/ocrdma/ocrdma.h index 45bdfa0e3b2b..7baedc74e39d 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma.h +++ b/drivers/infiniband/hw/ocrdma/ocrdma.h @@ -527,17 +527,17 @@ static inline int is_cqe_wr_imm(struct ocrdma_cqe *cqe) } static inline int ocrdma_resolve_dmac(struct ocrdma_dev *dev, - struct ib_ah_attr *ah_attr, u8 *mac_addr) + struct rdma_ah_attr *ah_attr, u8 *mac_addr) { struct in6_addr in6; - memcpy(&in6, ah_attr->grh.dgid.raw, sizeof(in6)); + memcpy(&in6, rdma_ah_read_grh(ah_attr)->dgid.raw, sizeof(in6)); if (rdma_is_multicast_addr(&in6)) rdma_get_mcast_mac(&in6, mac_addr); else if (rdma_link_local_addr(&in6)) rdma_get_ll_mac(&in6, mac_addr); else - memcpy(mac_addr, ah_attr->dmac, ETH_ALEN); + memcpy(mac_addr, ah_attr->roce.dmac, ETH_ALEN); return 0; } diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_ah.c b/drivers/infiniband/hw/ocrdma/ocrdma_ah.c index cd66e1e45dd7..d0249e463338 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_ah.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_ah.c @@ -71,7 +71,7 @@ static u16 ocrdma_hdr_type_to_proto_num(int devid, u8 hdr_type) } static inline int set_av_attr(struct ocrdma_dev *dev, struct ocrdma_ah *ah, - struct ib_ah_attr *attr, union ib_gid *sgid, + struct rdma_ah_attr *attr, union ib_gid *sgid, int pdid, bool *isvlan, u16 vlan_tag) { int status; @@ -81,6 +81,7 @@ static inline int set_av_attr(struct ocrdma_dev *dev, struct ocrdma_ah *ah, u16 proto_num = 0; u8 nxthdr = 0x11; struct iphdr ipv4; + const struct ib_global_route *ib_grh; union { struct sockaddr _sockaddr; struct sockaddr_in _sockaddr_in; @@ -120,32 +121,33 @@ static inline int set_av_attr(struct ocrdma_dev *dev, struct ocrdma_ah *ah, status = ocrdma_resolve_dmac(dev, attr, ð.dmac[0]); if (status) return status; - ah->sgid_index = attr->grh.sgid_index; + ib_grh = rdma_ah_read_grh(attr); + ah->sgid_index = ib_grh->sgid_index; /* Eth HDR */ memcpy(&ah->av->eth_hdr, ð, eth_sz); if (ah->hdr_type == RDMA_NETWORK_IPV4) { *((__be16 *)&ipv4) = htons((4 << 12) | (5 << 8) | - attr->grh.traffic_class); + ib_grh->traffic_class); ipv4.id = cpu_to_be16(pdid); ipv4.frag_off = htons(IP_DF); ipv4.tot_len = htons(0); - ipv4.ttl = attr->grh.hop_limit; + ipv4.ttl = ib_grh->hop_limit; ipv4.protocol = nxthdr; rdma_gid2ip(&sgid_addr._sockaddr, sgid); ipv4.saddr = sgid_addr._sockaddr_in.sin_addr.s_addr; - rdma_gid2ip(&dgid_addr._sockaddr, &attr->grh.dgid); + rdma_gid2ip(&dgid_addr._sockaddr, &ib_grh->dgid); ipv4.daddr = dgid_addr._sockaddr_in.sin_addr.s_addr; memcpy((u8 *)ah->av + eth_sz, &ipv4, sizeof(struct iphdr)); } else { memcpy(&grh.sgid[0], sgid->raw, sizeof(union ib_gid)); grh.tclass_flow = cpu_to_be32((6 << 28) | - (attr->grh.traffic_class << 24) | - attr->grh.flow_label); - memcpy(&grh.dgid[0], attr->grh.dgid.raw, - sizeof(attr->grh.dgid.raw)); + (ib_grh->traffic_class << 24) | + ib_grh->flow_label); + memcpy(&grh.dgid[0], ib_grh->dgid.raw, + sizeof(ib_grh->dgid.raw)); grh.pdid_hoplimit = cpu_to_be32((pdid << 16) | (nxthdr << 8) | - attr->grh.hop_limit); + ib_grh->hop_limit); memcpy((u8 *)ah->av + eth_sz, &grh, sizeof(struct ocrdma_grh)); } if (*isvlan) @@ -154,7 +156,7 @@ static inline int set_av_attr(struct ocrdma_dev *dev, struct ocrdma_ah *ah, return status; } -struct ib_ah *ocrdma_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr, +struct ib_ah *ocrdma_create_ah(struct ib_pd *ibpd, struct rdma_ah_attr *attr, struct ib_udata *udata) { u32 *ahid_addr; @@ -165,11 +167,14 @@ struct ib_ah *ocrdma_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr, struct ib_gid_attr sgid_attr; struct ocrdma_pd *pd = get_ocrdma_pd(ibpd); struct ocrdma_dev *dev = get_ocrdma_dev(ibpd->device); + const struct ib_global_route *grh; union ib_gid sgid; - if (!(attr->ah_flags & IB_AH_GRH)) + if ((attr->type != RDMA_AH_ATTR_TYPE_ROCE) || + !(rdma_ah_get_ah_flags(attr) & IB_AH_GRH)) return ERR_PTR(-EINVAL); + grh = rdma_ah_read_grh(attr); if (atomic_cmpxchg(&dev->update_sl, 1, 0)) ocrdma_init_service_level(dev); @@ -181,7 +186,7 @@ struct ib_ah *ocrdma_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr, if (status) goto av_err; - status = ib_get_cached_gid(&dev->ibdev, 1, attr->grh.sgid_index, &sgid, + status = ib_get_cached_gid(&dev->ibdev, 1, grh->sgid_index, &sgid, &sgid_attr); if (status) { pr_err("%s(): Failed to query sgid, status = %d\n", @@ -197,10 +202,11 @@ struct ib_ah *ocrdma_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr, ah->hdr_type = ib_gid_to_network_type(sgid_attr.gid_type, &sgid); if ((pd->uctx) && - (!rdma_is_multicast_addr((struct in6_addr *)attr->grh.dgid.raw)) && - (!rdma_link_local_addr((struct in6_addr *)attr->grh.dgid.raw))) { - status = rdma_addr_find_l2_eth_by_grh(&sgid, &attr->grh.dgid, - attr->dmac, &vlan_tag, + (!rdma_is_multicast_addr((struct in6_addr *)grh->dgid.raw)) && + (!rdma_link_local_addr((struct in6_addr *)grh->dgid.raw))) { + status = rdma_addr_find_l2_eth_by_grh(&sgid, &grh->dgid, + attr->roce.dmac, + &vlan_tag, &sgid_attr.ndev->ifindex, NULL); if (status) { @@ -216,7 +222,7 @@ struct ib_ah *ocrdma_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr, /* if pd is for the user process, pass the ah_id to user space */ if ((pd->uctx) && (pd->uctx->ah_tbl.va)) { - ahid_addr = pd->uctx->ah_tbl.va + attr->dlid; + ahid_addr = pd->uctx->ah_tbl.va + rdma_ah_get_dlid(attr); *ahid_addr = 0; *ahid_addr |= ah->id & OCRDMA_AH_ID_MASK; if (ocrdma_is_udp_encap_supported(dev)) { @@ -248,30 +254,32 @@ int ocrdma_destroy_ah(struct ib_ah *ibah) return 0; } -int ocrdma_query_ah(struct ib_ah *ibah, struct ib_ah_attr *attr) +int ocrdma_query_ah(struct ib_ah *ibah, struct rdma_ah_attr *attr) { struct ocrdma_ah *ah = get_ocrdma_ah(ibah); struct ocrdma_av *av = ah->av; struct ocrdma_grh *grh; - attr->ah_flags |= IB_AH_GRH; + + attr->type = ibah->type; if (ah->av->valid & OCRDMA_AV_VALID) { grh = (struct ocrdma_grh *)((u8 *)ah->av + sizeof(struct ocrdma_eth_vlan)); - attr->sl = be16_to_cpu(av->eth_hdr.vlan_tag) >> 13; + rdma_ah_set_sl(attr, be16_to_cpu(av->eth_hdr.vlan_tag) >> 13); } else { grh = (struct ocrdma_grh *)((u8 *)ah->av + sizeof(struct ocrdma_eth_basic)); - attr->sl = 0; + rdma_ah_set_sl(attr, 0); } - memcpy(&attr->grh.dgid.raw[0], &grh->dgid[0], sizeof(grh->dgid)); - attr->grh.sgid_index = ah->sgid_index; - attr->grh.hop_limit = be32_to_cpu(grh->pdid_hoplimit) & 0xff; - attr->grh.traffic_class = be32_to_cpu(grh->tclass_flow) >> 24; - attr->grh.flow_label = be32_to_cpu(grh->tclass_flow) & 0x00ffffffff; + rdma_ah_set_grh(attr, NULL, + be32_to_cpu(grh->tclass_flow) & 0xffffffff, + ah->sgid_index, + be32_to_cpu(grh->pdid_hoplimit) & 0xff, + be32_to_cpu(grh->tclass_flow) >> 24); + rdma_ah_set_dgid_raw(attr, &grh->dgid[0]); return 0; } -int ocrdma_modify_ah(struct ib_ah *ibah, struct ib_ah_attr *attr) +int ocrdma_modify_ah(struct ib_ah *ibah, struct rdma_ah_attr *attr) { /* modify_ah is unsupported */ return -ENOSYS; diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_ah.h b/drivers/infiniband/hw/ocrdma/ocrdma_ah.h index 0704a24b17c8..1a65c47945aa 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_ah.h +++ b/drivers/infiniband/hw/ocrdma/ocrdma_ah.h @@ -51,11 +51,11 @@ enum { OCRDMA_AH_L3_TYPE_SHIFT = 0x1D /* 29 bits */ }; -struct ib_ah *ocrdma_create_ah(struct ib_pd *, struct ib_ah_attr *, - struct ib_udata *); -int ocrdma_destroy_ah(struct ib_ah *); -int ocrdma_query_ah(struct ib_ah *, struct ib_ah_attr *); -int ocrdma_modify_ah(struct ib_ah *, struct ib_ah_attr *); +struct ib_ah *ocrdma_create_ah(struct ib_pd *pd, struct rdma_ah_attr *ah_attr, + struct ib_udata *udata); +int ocrdma_destroy_ah(struct ib_ah *ah); +int ocrdma_query_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr); +int ocrdma_modify_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr); int ocrdma_process_mad(struct ib_device *, int process_mad_flags, diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_hw.c b/drivers/infiniband/hw/ocrdma/ocrdma_hw.c index aa6967197620..dcb5942f9fb5 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_hw.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_hw.c @@ -2499,7 +2499,7 @@ static int ocrdma_set_av_params(struct ocrdma_qp *qp, int attr_mask) { int status; - struct ib_ah_attr *ah_attr = &attrs->ah_attr; + struct rdma_ah_attr *ah_attr = &attrs->ah_attr; union ib_gid sgid, zgid; struct ib_gid_attr sgid_attr; u32 vlan_id = 0xFFFF; @@ -2510,25 +2510,28 @@ static int ocrdma_set_av_params(struct ocrdma_qp *qp, struct sockaddr_in6 _sockaddr_in6; } sgid_addr, dgid_addr; struct ocrdma_dev *dev = get_ocrdma_dev(qp->ibqp.device); + const struct ib_global_route *grh; - if ((ah_attr->ah_flags & IB_AH_GRH) == 0) + if ((rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH) == 0) return -EINVAL; + grh = rdma_ah_read_grh(ah_attr); if (atomic_cmpxchg(&dev->update_sl, 1, 0)) ocrdma_init_service_level(dev); cmd->params.tclass_sq_psn |= - (ah_attr->grh.traffic_class << OCRDMA_QP_PARAMS_TCLASS_SHIFT); + (grh->traffic_class << OCRDMA_QP_PARAMS_TCLASS_SHIFT); cmd->params.rnt_rc_sl_fl |= - (ah_attr->grh.flow_label & OCRDMA_QP_PARAMS_FLOW_LABEL_MASK); - cmd->params.rnt_rc_sl_fl |= (ah_attr->sl << OCRDMA_QP_PARAMS_SL_SHIFT); + (grh->flow_label & OCRDMA_QP_PARAMS_FLOW_LABEL_MASK); + cmd->params.rnt_rc_sl_fl |= (rdma_ah_get_sl(ah_attr) << + OCRDMA_QP_PARAMS_SL_SHIFT); cmd->params.hop_lmt_rq_psn |= - (ah_attr->grh.hop_limit << OCRDMA_QP_PARAMS_HOP_LMT_SHIFT); + (grh->hop_limit << OCRDMA_QP_PARAMS_HOP_LMT_SHIFT); cmd->flags |= OCRDMA_QP_PARA_FLOW_LBL_VALID; /* GIDs */ - memcpy(&cmd->params.dgid[0], &ah_attr->grh.dgid.raw[0], + memcpy(&cmd->params.dgid[0], &grh->dgid.raw[0], sizeof(cmd->params.dgid)); - status = ib_get_cached_gid(&dev->ibdev, 1, ah_attr->grh.sgid_index, + status = ib_get_cached_gid(&dev->ibdev, 1, grh->sgid_index, &sgid, &sgid_attr); if (!status && sgid_attr.ndev) { vlan_id = rdma_vlan_dev_vlan_id(sgid_attr.ndev); @@ -2540,7 +2543,7 @@ static int ocrdma_set_av_params(struct ocrdma_qp *qp, if (!memcmp(&sgid, &zgid, sizeof(zgid))) return -EINVAL; - qp->sgid_idx = ah_attr->grh.sgid_index; + qp->sgid_idx = grh->sgid_index; memcpy(&cmd->params.sgid[0], &sgid.raw[0], sizeof(cmd->params.sgid)); status = ocrdma_resolve_dmac(dev, ah_attr, &mac_addr[0]); if (status) @@ -2551,7 +2554,7 @@ static int ocrdma_set_av_params(struct ocrdma_qp *qp, hdr_type = ib_gid_to_network_type(sgid_attr.gid_type, &sgid); if (hdr_type == RDMA_NETWORK_IPV4) { rdma_gid2ip(&sgid_addr._sockaddr, &sgid); - rdma_gid2ip(&dgid_addr._sockaddr, &ah_attr->grh.dgid); + rdma_gid2ip(&dgid_addr._sockaddr, &grh->dgid); memcpy(&cmd->params.dgid[0], &dgid_addr._sockaddr_in.sin_addr.s_addr, 4); memcpy(&cmd->params.sgid[0], diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_stats.c b/drivers/infiniband/hw/ocrdma/ocrdma_stats.c index f8e4b0a6486f..66056f9a9700 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_stats.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_stats.c @@ -643,7 +643,7 @@ static ssize_t ocrdma_dbgfs_ops_write(struct file *filp, struct ocrdma_stats *pstats = filp->private_data; struct ocrdma_dev *dev = pstats->dev; - if (count > 32) + if (*ppos != 0 || count == 0 || count > sizeof(tmp_str)) goto err; if (copy_from_user(tmp_str, buffer, count)) diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c index c52edeafd616..2f30bda8457a 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c @@ -914,21 +914,18 @@ static void build_user_pbes(struct ocrdma_dev *dev, struct ocrdma_mr *mr, pbe = (struct ocrdma_pbe *)pbl_tbl->va; pbe_cnt = 0; - shift = ilog2(umem->page_size); + shift = umem->page_shift; for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) { pages = sg_dma_len(sg) >> shift; for (pg_cnt = 0; pg_cnt < pages; pg_cnt++) { /* store the page address in pbe */ pbe->pa_lo = - cpu_to_le32(sg_dma_address - (sg) + - (umem->page_size * pg_cnt)); + cpu_to_le32(sg_dma_address(sg) + + (pg_cnt << shift)); pbe->pa_hi = - cpu_to_le32(upper_32_bits - ((sg_dma_address - (sg) + - umem->page_size * pg_cnt))); + cpu_to_le32(upper_32_bits(sg_dma_address(sg) + + (pg_cnt << shift))); pbe_cnt += 1; total_num_pbes += 1; pbe++; @@ -978,7 +975,7 @@ struct ib_mr *ocrdma_reg_user_mr(struct ib_pd *ibpd, u64 start, u64 len, if (status) goto umem_err; - mr->hwmr.pbe_size = mr->umem->page_size; + mr->hwmr.pbe_size = BIT(mr->umem->page_shift); mr->hwmr.fbo = ib_umem_offset(mr->umem); mr->hwmr.va = usr_addr; mr->hwmr.len = len; @@ -1601,23 +1598,24 @@ int ocrdma_query_qp(struct ib_qp *ibqp, qp_attr->cap.max_recv_sge = qp->rq.max_sges; qp_attr->cap.max_inline_data = qp->max_inline_data; qp_init_attr->cap = qp_attr->cap; - memcpy(&qp_attr->ah_attr.grh.dgid, ¶ms.dgid[0], - sizeof(params.dgid)); - qp_attr->ah_attr.grh.flow_label = params.rnt_rc_sl_fl & - OCRDMA_QP_PARAMS_FLOW_LABEL_MASK; - qp_attr->ah_attr.grh.sgid_index = qp->sgid_idx; - qp_attr->ah_attr.grh.hop_limit = (params.hop_lmt_rq_psn & - OCRDMA_QP_PARAMS_HOP_LMT_MASK) >> - OCRDMA_QP_PARAMS_HOP_LMT_SHIFT; - qp_attr->ah_attr.grh.traffic_class = (params.tclass_sq_psn & - OCRDMA_QP_PARAMS_TCLASS_MASK) >> - OCRDMA_QP_PARAMS_TCLASS_SHIFT; - - qp_attr->ah_attr.ah_flags = IB_AH_GRH; - qp_attr->ah_attr.port_num = 1; - qp_attr->ah_attr.sl = (params.rnt_rc_sl_fl & - OCRDMA_QP_PARAMS_SL_MASK) >> - OCRDMA_QP_PARAMS_SL_SHIFT; + qp_attr->ah_attr.type = RDMA_AH_ATTR_TYPE_ROCE; + + rdma_ah_set_grh(&qp_attr->ah_attr, NULL, + params.rnt_rc_sl_fl & + OCRDMA_QP_PARAMS_FLOW_LABEL_MASK, + qp->sgid_idx, + (params.hop_lmt_rq_psn & + OCRDMA_QP_PARAMS_HOP_LMT_MASK) >> + OCRDMA_QP_PARAMS_HOP_LMT_SHIFT, + (params.tclass_sq_psn & + OCRDMA_QP_PARAMS_TCLASS_MASK) >> + OCRDMA_QP_PARAMS_TCLASS_SHIFT); + rdma_ah_set_dgid_raw(&qp_attr->ah_attr, ¶ms.dgid[0]); + + rdma_ah_set_port_num(&qp_attr->ah_attr, 1); + rdma_ah_set_sl(&qp_attr->ah_attr, (params.rnt_rc_sl_fl & + OCRDMA_QP_PARAMS_SL_MASK) >> + OCRDMA_QP_PARAMS_SL_SHIFT); qp_attr->timeout = (params.ack_to_rnr_rtc_dest_qpn & OCRDMA_QP_PARAMS_ACK_TIMEOUT_MASK) >> OCRDMA_QP_PARAMS_ACK_TIMEOUT_SHIFT; @@ -1630,8 +1628,8 @@ int ocrdma_query_qp(struct ib_qp *ibqp, qp_attr->min_rnr_timer = 0; qp_attr->pkey_index = 0; qp_attr->port_num = 1; - qp_attr->ah_attr.src_path_bits = 0; - qp_attr->ah_attr.static_rate = 0; + rdma_ah_set_path_bits(&qp_attr->ah_attr, 0); + rdma_ah_set_static_rate(&qp_attr->ah_attr, 0); qp_attr->alt_pkey_index = 0; qp_attr->alt_port_num = 0; qp_attr->alt_timeout = 0; diff --git a/drivers/infiniband/hw/qedr/main.c b/drivers/infiniband/hw/qedr/main.c index ced0461d6e9f..ef11e770f822 100644 --- a/drivers/infiniband/hw/qedr/main.c +++ b/drivers/infiniband/hw/qedr/main.c @@ -340,43 +340,58 @@ static void qedr_remove_sysfiles(struct qedr_dev *dev) static void qedr_pci_set_atomic(struct qedr_dev *dev, struct pci_dev *pdev) { struct pci_dev *bridge; - u32 val; - - dev->atomic_cap = IB_ATOMIC_NONE; + u32 ctl2, cap2; + u16 flags; + int rc; bridge = pdev->bus->self; if (!bridge) - return; - - /* Check whether we are connected directly or via a switch */ - while (bridge && bridge->bus->parent) { - DP_DEBUG(dev, QEDR_MSG_INIT, - "Device is not connected directly to root. bridge->bus->number=%d primary=%d\n", - bridge->bus->number, bridge->bus->primary); - /* Need to check Atomic Op Routing Supported all the way to - * root complex. - */ - pcie_capability_read_dword(bridge, PCI_EXP_DEVCAP2, &val); - if (!(val & PCI_EXP_DEVCAP2_ATOMIC_ROUTE)) { - pcie_capability_clear_word(pdev, - PCI_EXP_DEVCTL2, - PCI_EXP_DEVCTL2_ATOMIC_REQ); - return; - } + goto disable; + + /* Check atomic routing support all the way to root complex */ + while (bridge->bus->parent) { + rc = pcie_capability_read_word(bridge, PCI_EXP_FLAGS, &flags); + if (rc || ((flags & PCI_EXP_FLAGS_VERS) < 2)) + goto disable; + + rc = pcie_capability_read_dword(bridge, PCI_EXP_DEVCAP2, &cap2); + if (rc) + goto disable; + + rc = pcie_capability_read_dword(bridge, PCI_EXP_DEVCTL2, &ctl2); + if (rc) + goto disable; + + if (!(cap2 & PCI_EXP_DEVCAP2_ATOMIC_ROUTE) || + (ctl2 & PCI_EXP_DEVCTL2_ATOMIC_EGRESS_BLOCK)) + goto disable; bridge = bridge->bus->parent->self; } - bridge = pdev->bus->self; - /* according to bridge capability */ - pcie_capability_read_dword(bridge, PCI_EXP_DEVCAP2, &val); - if (val & PCI_EXP_DEVCAP2_ATOMIC_COMP64) { - pcie_capability_set_word(pdev, PCI_EXP_DEVCTL2, - PCI_EXP_DEVCTL2_ATOMIC_REQ); - dev->atomic_cap = IB_ATOMIC_GLOB; - } else { - pcie_capability_clear_word(pdev, PCI_EXP_DEVCTL2, - PCI_EXP_DEVCTL2_ATOMIC_REQ); - } + rc = pcie_capability_read_word(bridge, PCI_EXP_FLAGS, &flags); + if (rc || ((flags & PCI_EXP_FLAGS_VERS) < 2)) + goto disable; + + rc = pcie_capability_read_dword(bridge, PCI_EXP_DEVCAP2, &cap2); + if (rc || !(cap2 & PCI_EXP_DEVCAP2_ATOMIC_COMP64)) + goto disable; + + /* Set atomic operations */ + pcie_capability_set_word(pdev, PCI_EXP_DEVCTL2, + PCI_EXP_DEVCTL2_ATOMIC_REQ); + dev->atomic_cap = IB_ATOMIC_GLOB; + + DP_DEBUG(dev, QEDR_MSG_INIT, "Atomic capability enabled\n"); + + return; + +disable: + pcie_capability_clear_word(pdev, PCI_EXP_DEVCTL2, + PCI_EXP_DEVCTL2_ATOMIC_REQ); + dev->atomic_cap = IB_ATOMIC_NONE; + + DP_DEBUG(dev, QEDR_MSG_INIT, "Atomic capability disabled\n"); + } static const struct qed_rdma_ops *qed_ops; @@ -423,14 +438,21 @@ static irqreturn_t qedr_irq_handler(int irq, void *handle) cq->arm_flags = 0; - if (cq->ibcq.comp_handler) + if (!cq->destroyed && cq->ibcq.comp_handler) (*cq->ibcq.comp_handler) (&cq->ibcq, cq->ibcq.cq_context); + /* The CQ's CNQ notification counter is checked before + * destroying the CQ in a busy-wait loop that waits for all of + * the CQ's CNQ interrupts to be processed. It is increased + * here, only after the completion handler, to ensure that the + * the handler is not running when the CQ is destroyed. + */ + cq->cnq_notif++; + sw_comp_cons = qed_chain_get_cons_idx(&cnq->pbl); cnq->n_comp++; - } qed_ops->rdma_cnq_prod_update(cnq->dev->rdma_ctx, cnq->index, diff --git a/drivers/infiniband/hw/qedr/qedr.h b/drivers/infiniband/hw/qedr/qedr.h index 5cb9195513bd..aa08c76a4245 100644 --- a/drivers/infiniband/hw/qedr/qedr.h +++ b/drivers/infiniband/hw/qedr/qedr.h @@ -272,6 +272,8 @@ struct qedr_cq { u32 cq_cons; struct qedr_userq q; + u8 destroyed; + u16 cnq_notif; }; struct qedr_pd { @@ -390,7 +392,7 @@ struct qedr_qp { struct qedr_ah { struct ib_ah ibah; - struct ib_ah_attr attr; + struct rdma_ah_attr attr; }; enum qedr_mr_type { @@ -429,7 +431,8 @@ struct qedr_mr { RDMA_CQE_RESPONDER_IMM_FLG_SHIFT) #define QEDR_RESP_RDMA (RDMA_CQE_RESPONDER_RDMA_FLG_MASK << \ RDMA_CQE_RESPONDER_RDMA_FLG_SHIFT) -#define QEDR_RESP_RDMA_IMM (QEDR_RESP_IMM | QEDR_RESP_RDMA) +#define QEDR_RESP_INV (RDMA_CQE_RESPONDER_INV_FLG_MASK << \ + RDMA_CQE_RESPONDER_INV_FLG_SHIFT) static inline void qedr_inc_sw_cons(struct qedr_qp_hwq_info *info) { @@ -443,19 +446,24 @@ static inline void qedr_inc_sw_prod(struct qedr_qp_hwq_info *info) } static inline int qedr_get_dmac(struct qedr_dev *dev, - struct ib_ah_attr *ah_attr, u8 *mac_addr) + struct rdma_ah_attr *ah_attr, u8 *mac_addr) { union ib_gid zero_sgid = { { 0 } }; struct in6_addr in6; + const struct ib_global_route *grh = rdma_ah_read_grh(ah_attr); + u8 *dmac; - if (!memcmp(&ah_attr->grh.dgid, &zero_sgid, sizeof(union ib_gid))) { + if (!memcmp(&grh->dgid, &zero_sgid, sizeof(union ib_gid))) { DP_ERR(dev, "Local port GID not supported\n"); eth_zero_addr(mac_addr); return -EINVAL; } - memcpy(&in6, ah_attr->grh.dgid.raw, sizeof(in6)); - ether_addr_copy(mac_addr, ah_attr->dmac); + memcpy(&in6, grh->dgid.raw, sizeof(in6)); + dmac = rdma_ah_retrieve_dmac(ah_attr); + if (!dmac) + return -EINVAL; + ether_addr_copy(mac_addr, dmac); return 0; } diff --git a/drivers/infiniband/hw/qedr/qedr_cm.c b/drivers/infiniband/hw/qedr/qedr_cm.c index a6280ce3e2a5..3d7705cec770 100644 --- a/drivers/infiniband/hw/qedr/qedr_cm.c +++ b/drivers/infiniband/hw/qedr/qedr_cm.c @@ -243,8 +243,8 @@ static inline int qedr_gsi_build_header(struct qedr_dev *dev, int *roce_mode) { bool has_vlan = false, has_grh_ipv6 = true; - struct ib_ah_attr *ah_attr = &get_qedr_ah(ud_wr(swr)->ah)->attr; - struct ib_global_route *grh = &ah_attr->grh; + struct rdma_ah_attr *ah_attr = &get_qedr_ah(ud_wr(swr)->ah)->attr; + const struct ib_global_route *grh = rdma_ah_read_grh(ah_attr); union ib_gid sgid; int send_size = 0; u16 vlan_id = 0; @@ -260,12 +260,13 @@ static inline int qedr_gsi_build_header(struct qedr_dev *dev, for (i = 0; i < swr->num_sge; ++i) send_size += swr->sg_list[i].length; - rc = ib_get_cached_gid(qp->ibqp.device, ah_attr->port_num, + rc = ib_get_cached_gid(qp->ibqp.device, rdma_ah_get_port_num(ah_attr), grh->sgid_index, &sgid, &sgid_attr); if (rc) { DP_ERR(dev, "gsi post send: failed to get cached GID (port=%d, ix=%d)\n", - ah_attr->port_num, grh->sgid_index); + rdma_ah_get_port_num(ah_attr), + grh->sgid_index); return rc; } @@ -277,7 +278,7 @@ static inline int qedr_gsi_build_header(struct qedr_dev *dev, if (!memcmp(&sgid, &zgid, sizeof(sgid))) { DP_ERR(dev, "gsi post send: GID not found GID index %d\n", - ah_attr->grh.sgid_index); + grh->sgid_index); return -ENOENT; } @@ -307,7 +308,7 @@ static inline int qedr_gsi_build_header(struct qedr_dev *dev, } /* ENET + VLAN headers */ - ether_addr_copy(udh->eth.dmac_h, ah_attr->dmac); + ether_addr_copy(udh->eth.dmac_h, ah_attr->roce.dmac); ether_addr_copy(udh->eth.smac_h, dev->ndev->dev_addr); if (has_vlan) { udh->eth.type = htons(ETH_P_8021Q); @@ -341,13 +342,13 @@ static inline int qedr_gsi_build_header(struct qedr_dev *dev, u32 ipv4_addr; udh->ip4.protocol = IPPROTO_UDP; - udh->ip4.tos = htonl(ah_attr->grh.flow_label); + udh->ip4.tos = htonl(grh->flow_label); udh->ip4.frag_off = htons(IP_DF); - udh->ip4.ttl = ah_attr->grh.hop_limit; + udh->ip4.ttl = grh->hop_limit; ipv4_addr = qedr_get_ipv4_from_gid(sgid.raw); udh->ip4.saddr = ipv4_addr; - ipv4_addr = qedr_get_ipv4_from_gid(ah_attr->grh.dgid.raw); + ipv4_addr = qedr_get_ipv4_from_gid(grh->dgid.raw); udh->ip4.daddr = ipv4_addr; /* note: checksum is calculated by the device */ } diff --git a/drivers/infiniband/hw/qedr/qedr_cm.h b/drivers/infiniband/hw/qedr/qedr_cm.h index 78efb1b056d1..a55916323ea9 100644 --- a/drivers/infiniband/hw/qedr/qedr_cm.h +++ b/drivers/infiniband/hw/qedr/qedr_cm.h @@ -39,7 +39,7 @@ #define QEDR_ROCE_V2_UDP_SPORT (0000) -static inline u32 qedr_get_ipv4_from_gid(u8 *gid) +static inline u32 qedr_get_ipv4_from_gid(const u8 *gid) { return *(u32 *)(void *)&gid[12]; } diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c index 2091902848e6..17685cfea6a2 100644 --- a/drivers/infiniband/hw/qedr/verbs.c +++ b/drivers/infiniband/hw/qedr/verbs.c @@ -681,16 +681,16 @@ static void qedr_populate_pbls(struct qedr_dev *dev, struct ib_umem *umem, pbe_cnt = 0; - shift = ilog2(umem->page_size); + shift = umem->page_shift; for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) { pages = sg_dma_len(sg) >> shift; for (pg_cnt = 0; pg_cnt < pages; pg_cnt++) { /* store the page address in pbe */ pbe->lo = cpu_to_le32(sg_dma_address(sg) + - umem->page_size * pg_cnt); + (pg_cnt << shift)); addr = upper_32_bits(sg_dma_address(sg) + - umem->page_size * pg_cnt); + (pg_cnt << shift)); pbe->hi = cpu_to_le32(addr); pbe_cnt++; total_num_pbes++; @@ -822,6 +822,17 @@ int qedr_arm_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags flags) { struct qedr_cq *cq = get_qedr_cq(ibcq); unsigned long sflags; + struct qedr_dev *dev; + + dev = get_qedr_dev(ibcq->device); + + if (cq->destroyed) { + DP_ERR(dev, + "warning: arm was invoked after destroy for cq %p (icid=%d)\n", + cq, cq->icid); + return -EINVAL; + } + if (cq->cq_type == QEDR_CQ_TYPE_GSI) return 0; @@ -987,35 +998,82 @@ int qedr_resize_cq(struct ib_cq *ibcq, int new_cnt, struct ib_udata *udata) return 0; } +#define QEDR_DESTROY_CQ_MAX_ITERATIONS (10) +#define QEDR_DESTROY_CQ_ITER_DURATION (10) + int qedr_destroy_cq(struct ib_cq *ibcq) { struct qedr_dev *dev = get_qedr_dev(ibcq->device); struct qed_rdma_destroy_cq_out_params oparams; struct qed_rdma_destroy_cq_in_params iparams; struct qedr_cq *cq = get_qedr_cq(ibcq); + int iter; + int rc; - DP_DEBUG(dev, QEDR_MSG_CQ, "destroy cq: cq_id %d", cq->icid); + DP_DEBUG(dev, QEDR_MSG_CQ, "destroy cq %p (icid=%d)\n", cq, cq->icid); + + cq->destroyed = 1; /* GSIs CQs are handled by driver, so they don't exist in the FW */ - if (cq->cq_type != QEDR_CQ_TYPE_GSI) { - int rc; + if (cq->cq_type == QEDR_CQ_TYPE_GSI) + goto done; - iparams.icid = cq->icid; - rc = dev->ops->rdma_destroy_cq(dev->rdma_ctx, &iparams, - &oparams); - if (rc) - return rc; - dev->ops->common->chain_free(dev->cdev, &cq->pbl); - } + iparams.icid = cq->icid; + rc = dev->ops->rdma_destroy_cq(dev->rdma_ctx, &iparams, &oparams); + if (rc) + return rc; + + dev->ops->common->chain_free(dev->cdev, &cq->pbl); if (ibcq->uobject && ibcq->uobject->context) { qedr_free_pbl(dev, &cq->q.pbl_info, cq->q.pbl_tbl); ib_umem_release(cq->q.umem); } + /* We don't want the IRQ handler to handle a non-existing CQ so we + * wait until all CNQ interrupts, if any, are received. This will always + * happen and will always happen very fast. If not, then a serious error + * has occured. That is why we can use a long delay. + * We spin for a short time so we don’t lose time on context switching + * in case all the completions are handled in that span. Otherwise + * we sleep for a while and check again. Since the CNQ may be + * associated with (only) the current CPU we use msleep to allow the + * current CPU to be freed. + * The CNQ notification is increased in qedr_irq_handler(). + */ + iter = QEDR_DESTROY_CQ_MAX_ITERATIONS; + while (oparams.num_cq_notif != READ_ONCE(cq->cnq_notif) && iter) { + udelay(QEDR_DESTROY_CQ_ITER_DURATION); + iter--; + } + + iter = QEDR_DESTROY_CQ_MAX_ITERATIONS; + while (oparams.num_cq_notif != READ_ONCE(cq->cnq_notif) && iter) { + msleep(QEDR_DESTROY_CQ_ITER_DURATION); + iter--; + } + + if (oparams.num_cq_notif != cq->cnq_notif) + goto err; + + /* Note that we don't need to have explicit code to wait for the + * completion of the event handler because it is invoked from the EQ. + * Since the destroy CQ ramrod has also been received on the EQ we can + * be certain that there's no event handler in process. + */ +done: + cq->sig = ~cq->sig; + kfree(cq); return 0; + +err: + DP_ERR(dev, + "CQ %p (icid=%d) not freed, expecting %d ints but got %d ints\n", + cq, cq->icid, oparams.num_cq_notif, cq->cnq_notif); + + return -EINVAL; } static inline int get_gid_info_from_table(struct ib_qp *ibqp, @@ -1026,13 +1084,15 @@ static inline int get_gid_info_from_table(struct ib_qp *ibqp, { enum rdma_network_type nw_type; struct ib_gid_attr gid_attr; + const struct ib_global_route *grh = rdma_ah_read_grh(&attr->ah_attr); union ib_gid gid; u32 ipv4_addr; int rc = 0; int i; - rc = ib_get_cached_gid(ibqp->device, attr->ah_attr.port_num, - attr->ah_attr.grh.sgid_index, &gid, &gid_attr); + rc = ib_get_cached_gid(ibqp->device, + rdma_ah_get_port_num(&attr->ah_attr), + grh->sgid_index, &gid, &gid_attr); if (rc) return rc; @@ -1049,7 +1109,7 @@ static inline int get_gid_info_from_table(struct ib_qp *ibqp, memcpy(&qp_params->sgid.bytes[0], &gid.raw[0], sizeof(qp_params->sgid)); memcpy(&qp_params->dgid.bytes[0], - &attr->ah_attr.grh.dgid, + &grh->dgid, sizeof(qp_params->dgid)); qp_params->roce_mode = ROCE_V2_IPV6; SET_FIELD(qp_params->modify_flags, @@ -1059,7 +1119,7 @@ static inline int get_gid_info_from_table(struct ib_qp *ibqp, memcpy(&qp_params->sgid.bytes[0], &gid.raw[0], sizeof(qp_params->sgid)); memcpy(&qp_params->dgid.bytes[0], - &attr->ah_attr.grh.dgid, + &grh->dgid, sizeof(qp_params->dgid)); qp_params->roce_mode = ROCE_V1; break; @@ -1069,7 +1129,7 @@ static inline int get_gid_info_from_table(struct ib_qp *ibqp, ipv4_addr = qedr_get_ipv4_from_gid(gid.raw); qp_params->sgid.ipv4_addr = ipv4_addr; ipv4_addr = - qedr_get_ipv4_from_gid(attr->ah_attr.grh.dgid.raw); + qedr_get_ipv4_from_gid(grh->dgid.raw); qp_params->dgid.ipv4_addr = ipv4_addr; SET_FIELD(qp_params->modify_flags, QED_ROCE_MODIFY_QP_VALID_ROCE_MODE, 1); @@ -1691,6 +1751,7 @@ int qedr_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, struct qedr_qp *qp = get_qedr_qp(ibqp); struct qed_rdma_modify_qp_in_params qp_params = { 0 }; struct qedr_dev *dev = get_qedr_dev(&qp->dev->ibdev); + const struct ib_global_route *grh = rdma_ah_read_grh(&attr->ah_attr); enum ib_qp_state old_qp_state, new_qp_state; int rc = 0; @@ -1773,17 +1834,17 @@ int qedr_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, SET_FIELD(qp_params.modify_flags, QED_ROCE_MODIFY_QP_VALID_ADDRESS_VECTOR, 1); - qp_params.traffic_class_tos = attr->ah_attr.grh.traffic_class; - qp_params.flow_label = attr->ah_attr.grh.flow_label; - qp_params.hop_limit_ttl = attr->ah_attr.grh.hop_limit; + qp_params.traffic_class_tos = grh->traffic_class; + qp_params.flow_label = grh->flow_label; + qp_params.hop_limit_ttl = grh->hop_limit; - qp->sgid_idx = attr->ah_attr.grh.sgid_index; + qp->sgid_idx = grh->sgid_index; rc = get_gid_info_from_table(ibqp, attr, attr_mask, &qp_params); if (rc) { DP_ERR(dev, "modify qp: problems with GID index %d (rc=%d)\n", - attr->ah_attr.grh.sgid_index, rc); + grh->sgid_index, rc); return rc; } @@ -1968,25 +2029,21 @@ int qedr_query_qp(struct ib_qp *ibqp, qp_attr->cap.max_inline_data = ROCE_REQ_MAX_INLINE_DATA_SIZE; qp_init_attr->cap = qp_attr->cap; - memcpy(&qp_attr->ah_attr.grh.dgid.raw[0], ¶ms.dgid.bytes[0], - sizeof(qp_attr->ah_attr.grh.dgid.raw)); - - qp_attr->ah_attr.grh.flow_label = params.flow_label; - qp_attr->ah_attr.grh.sgid_index = qp->sgid_idx; - qp_attr->ah_attr.grh.hop_limit = params.hop_limit_ttl; - qp_attr->ah_attr.grh.traffic_class = params.traffic_class_tos; - - qp_attr->ah_attr.ah_flags = IB_AH_GRH; - qp_attr->ah_attr.port_num = 1; - qp_attr->ah_attr.sl = 0; + qp_attr->ah_attr.type = RDMA_AH_ATTR_TYPE_ROCE; + rdma_ah_set_grh(&qp_attr->ah_attr, NULL, + params.flow_label, qp->sgid_idx, + params.hop_limit_ttl, params.traffic_class_tos); + rdma_ah_set_dgid_raw(&qp_attr->ah_attr, ¶ms.dgid.bytes[0]); + rdma_ah_set_port_num(&qp_attr->ah_attr, 1); + rdma_ah_set_sl(&qp_attr->ah_attr, 0); qp_attr->timeout = params.timeout; qp_attr->rnr_retry = params.rnr_retry; qp_attr->retry_cnt = params.retry_cnt; qp_attr->min_rnr_timer = params.min_rnr_nak_timer; qp_attr->pkey_index = params.pkey_index; qp_attr->port_num = 1; - qp_attr->ah_attr.src_path_bits = 0; - qp_attr->ah_attr.static_rate = 0; + rdma_ah_set_path_bits(&qp_attr->ah_attr, 0); + rdma_ah_set_static_rate(&qp_attr->ah_attr, 0); qp_attr->alt_pkey_index = 0; qp_attr->alt_port_num = 0; qp_attr->alt_timeout = 0; @@ -2054,7 +2111,7 @@ int qedr_destroy_qp(struct ib_qp *ibqp) return rc; } -struct ib_ah *qedr_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr, +struct ib_ah *qedr_create_ah(struct ib_pd *ibpd, struct rdma_ah_attr *attr, struct ib_udata *udata) { struct qedr_ah *ah; @@ -2190,7 +2247,7 @@ struct ib_mr *qedr_reg_user_mr(struct ib_pd *ibpd, u64 start, u64 len, mr->hw_mr.pbl_ptr = mr->info.pbl_table[0].pa; mr->hw_mr.pbl_two_level = mr->info.pbl_info.two_layered; mr->hw_mr.pbl_page_size_log = ilog2(mr->info.pbl_info.pbl_size); - mr->hw_mr.page_size_log = ilog2(mr->umem->page_size); + mr->hw_mr.page_size_log = mr->umem->page_shift; mr->hw_mr.fbo = ib_umem_offset(mr->umem); mr->hw_mr.length = len; mr->hw_mr.vaddr = usr_addr; @@ -2625,6 +2682,8 @@ static int qedr_prepare_reg(struct qedr_qp *qp, fwqe1->addr.lo = lower_32_bits(mr->ibmr.iova); fwqe1->l_key = wr->key; + fwqe2->access_ctrl = 0; + SET_FIELD2(fwqe2->access_ctrl, RDMA_SQ_FMR_WQE_2ND_REMOTE_READ, !!(wr->access & IB_ACCESS_REMOTE_READ)); SET_FIELD2(fwqe2->access_ctrl, RDMA_SQ_FMR_WQE_2ND_REMOTE_WRITE, @@ -3271,57 +3330,81 @@ static int qedr_poll_cq_req(struct qedr_dev *dev, return cnt; } -static void __process_resp_one(struct qedr_dev *dev, struct qedr_qp *qp, - struct qedr_cq *cq, struct ib_wc *wc, - struct rdma_cqe_responder *resp, u64 wr_id) +static inline int qedr_cqe_resp_status_to_ib(u8 status) { - enum ib_wc_status wc_status = IB_WC_SUCCESS; - u8 flags; - - wc->opcode = IB_WC_RECV; - wc->wc_flags = 0; - - switch (resp->status) { + switch (status) { case RDMA_CQE_RESP_STS_LOCAL_ACCESS_ERR: - wc_status = IB_WC_LOC_ACCESS_ERR; - break; + return IB_WC_LOC_ACCESS_ERR; case RDMA_CQE_RESP_STS_LOCAL_LENGTH_ERR: - wc_status = IB_WC_LOC_LEN_ERR; - break; + return IB_WC_LOC_LEN_ERR; case RDMA_CQE_RESP_STS_LOCAL_QP_OPERATION_ERR: - wc_status = IB_WC_LOC_QP_OP_ERR; - break; + return IB_WC_LOC_QP_OP_ERR; case RDMA_CQE_RESP_STS_LOCAL_PROTECTION_ERR: - wc_status = IB_WC_LOC_PROT_ERR; - break; + return IB_WC_LOC_PROT_ERR; case RDMA_CQE_RESP_STS_MEMORY_MGT_OPERATION_ERR: - wc_status = IB_WC_MW_BIND_ERR; - break; + return IB_WC_MW_BIND_ERR; case RDMA_CQE_RESP_STS_REMOTE_INVALID_REQUEST_ERR: - wc_status = IB_WC_REM_INV_RD_REQ_ERR; - break; + return IB_WC_REM_INV_RD_REQ_ERR; case RDMA_CQE_RESP_STS_OK: - wc_status = IB_WC_SUCCESS; - wc->byte_len = le32_to_cpu(resp->length); + return IB_WC_SUCCESS; + default: + return IB_WC_GENERAL_ERR; + } +} - flags = resp->flags & QEDR_RESP_RDMA_IMM; +static inline int qedr_set_ok_cqe_resp_wc(struct rdma_cqe_responder *resp, + struct ib_wc *wc) +{ + wc->status = IB_WC_SUCCESS; + wc->byte_len = le32_to_cpu(resp->length); - if (flags == QEDR_RESP_RDMA_IMM) + if (resp->flags & QEDR_RESP_IMM) { + wc->ex.imm_data = le32_to_cpu(resp->imm_data_or_inv_r_Key); + wc->wc_flags |= IB_WC_WITH_IMM; + + if (resp->flags & QEDR_RESP_RDMA) wc->opcode = IB_WC_RECV_RDMA_WITH_IMM; - if (flags == QEDR_RESP_RDMA_IMM || flags == QEDR_RESP_IMM) { - wc->ex.imm_data = - le32_to_cpu(resp->imm_data_or_inv_r_Key); - wc->wc_flags |= IB_WC_WITH_IMM; - } - break; - default: - wc->status = IB_WC_GENERAL_ERR; - DP_ERR(dev, "Invalid CQE status detected\n"); + if (resp->flags & QEDR_RESP_INV) + return -EINVAL; + + } else if (resp->flags & QEDR_RESP_INV) { + wc->ex.imm_data = le32_to_cpu(resp->imm_data_or_inv_r_Key); + wc->wc_flags |= IB_WC_WITH_INVALIDATE; + + if (resp->flags & QEDR_RESP_RDMA) + return -EINVAL; + + } else if (resp->flags & QEDR_RESP_RDMA) { + return -EINVAL; } - /* fill WC */ - wc->status = wc_status; + return 0; +} + +static void __process_resp_one(struct qedr_dev *dev, struct qedr_qp *qp, + struct qedr_cq *cq, struct ib_wc *wc, + struct rdma_cqe_responder *resp, u64 wr_id) +{ + /* Must fill fields before qedr_set_ok_cqe_resp_wc() */ + wc->opcode = IB_WC_RECV; + wc->wc_flags = 0; + + if (likely(resp->status == RDMA_CQE_RESP_STS_OK)) { + if (qedr_set_ok_cqe_resp_wc(resp, wc)) + DP_ERR(dev, + "CQ %p (icid=%d) has invalid CQE responder flags=0x%x\n", + cq, cq->icid, resp->flags); + + } else { + wc->status = qedr_cqe_resp_status_to_ib(resp->status); + if (wc->status == IB_WC_GENERAL_ERR) + DP_ERR(dev, + "CQ %p (icid=%d) contains an invalid CQE status %d\n", + cq, cq->icid, resp->status); + } + + /* Fill the rest of the WC */ wc->vendor_err = 0; wc->src_qp = qp->id; wc->qp = &qp->ibqp; @@ -3416,6 +3499,13 @@ int qedr_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *wc) int update = 0; int done = 0; + if (cq->destroyed) { + DP_ERR(dev, + "warning: poll was invoked after destroy for cq %p (icid=%d)\n", + cq, cq->icid); + return 0; + } + if (cq->cq_type == QEDR_CQ_TYPE_GSI) return qedr_gsi_poll_cq(ibcq, num_entries, wc); diff --git a/drivers/infiniband/hw/qedr/verbs.h b/drivers/infiniband/hw/qedr/verbs.h index 070677ca4d19..0f8ab49d5a1a 100644 --- a/drivers/infiniband/hw/qedr/verbs.h +++ b/drivers/infiniband/hw/qedr/verbs.h @@ -70,7 +70,7 @@ int qedr_query_qp(struct ib_qp *, struct ib_qp_attr *qp_attr, int qp_attr_mask, struct ib_qp_init_attr *); int qedr_destroy_qp(struct ib_qp *ibqp); -struct ib_ah *qedr_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr, +struct ib_ah *qedr_create_ah(struct ib_pd *ibpd, struct rdma_ah_attr *attr, struct ib_udata *udata); int qedr_destroy_ah(struct ib_ah *ibah); diff --git a/drivers/infiniband/hw/qib/qib_iba6120.c b/drivers/infiniband/hw/qib/qib_iba6120.c index 06de1cbcf67d..e423b71e6ea0 100644 --- a/drivers/infiniband/hw/qib/qib_iba6120.c +++ b/drivers/infiniband/hw/qib/qib_iba6120.c @@ -3295,13 +3295,11 @@ static int init_6120_variables(struct qib_devdata *dd) dd->rhdrhead_intr_off = 1ULL << 32; /* setup the stats timer; the add_timer is done at end of init */ - init_timer(&dd->stats_timer); - dd->stats_timer.function = qib_get_6120_faststats; - dd->stats_timer.data = (unsigned long) dd; + setup_timer(&dd->stats_timer, qib_get_6120_faststats, + (unsigned long)dd); - init_timer(&dd->cspec->pma_timer); - dd->cspec->pma_timer.function = pma_6120_timer; - dd->cspec->pma_timer.data = (unsigned long) ppd; + setup_timer(&dd->cspec->pma_timer, pma_6120_timer, + (unsigned long)ppd); dd->ureg_align = qib_read_kreg32(dd, kr_palign); diff --git a/drivers/infiniband/hw/qib/qib_iba7220.c b/drivers/infiniband/hw/qib/qib_iba7220.c index 55a18384c22d..c3679c48e61c 100644 --- a/drivers/infiniband/hw/qib/qib_iba7220.c +++ b/drivers/infiniband/hw/qib/qib_iba7220.c @@ -4074,9 +4074,8 @@ static int qib_init_7220_variables(struct qib_devdata *dd) if (!qib_mini_init) qib_write_kreg(dd, kr_rcvbthqp, QIB_KD_QP); - init_timer(&ppd->cpspec->chase_timer); - ppd->cpspec->chase_timer.function = reenable_7220_chase; - ppd->cpspec->chase_timer.data = (unsigned long)ppd; + setup_timer(&ppd->cpspec->chase_timer, reenable_7220_chase, + (unsigned long)ppd); qib_num_cfg_vls = 1; /* if any 7220's, only one VL */ diff --git a/drivers/infiniband/hw/qib/qib_iba7322.c b/drivers/infiniband/hw/qib/qib_iba7322.c index af9f596bb68b..bb2439fff8fa 100644 --- a/drivers/infiniband/hw/qib/qib_iba7322.c +++ b/drivers/infiniband/hw/qib/qib_iba7322.c @@ -6611,9 +6611,8 @@ static int qib_init_7322_variables(struct qib_devdata *dd) if (!qib_mini_init) write_7322_init_portregs(ppd); - init_timer(&cp->chase_timer); - cp->chase_timer.function = reenable_chase; - cp->chase_timer.data = (unsigned long)ppd; + setup_timer(&cp->chase_timer, reenable_chase, + (unsigned long)ppd); ppd++; } @@ -6639,9 +6638,8 @@ static int qib_init_7322_variables(struct qib_devdata *dd) (u64) rcv_int_count << IBA7322_HDRHEAD_PKTINT_SHIFT; /* setup the stats timer; the add_timer is done at end of init */ - init_timer(&dd->stats_timer); - dd->stats_timer.function = qib_get_7322_faststats; - dd->stats_timer.data = (unsigned long) dd; + setup_timer(&dd->stats_timer, qib_get_7322_faststats, + (unsigned long)dd); dd->ureg_align = 0x10000; /* 64KB alignment */ diff --git a/drivers/infiniband/hw/qib/qib_init.c b/drivers/infiniband/hw/qib/qib_init.c index b50240b1d5a4..6c16ba1107ba 100644 --- a/drivers/infiniband/hw/qib/qib_init.c +++ b/drivers/infiniband/hw/qib/qib_init.c @@ -233,9 +233,8 @@ int qib_init_pportdata(struct qib_pportdata *ppd, struct qib_devdata *dd, spin_lock_init(&ppd->cc_shadow_lock); init_waitqueue_head(&ppd->state_wait); - init_timer(&ppd->symerr_clear_timer); - ppd->symerr_clear_timer.function = qib_clear_symerror_on_linkup; - ppd->symerr_clear_timer.data = (unsigned long)ppd; + setup_timer(&ppd->symerr_clear_timer, qib_clear_symerror_on_linkup, + (unsigned long)ppd); ppd->qib_wq = NULL; ppd->ibport_data.pmastats = @@ -429,9 +428,8 @@ static int loadtime_init(struct qib_devdata *dd) qib_get_eeprom_info(dd); /* setup time (don't start yet) to verify we got interrupt */ - init_timer(&dd->intrchk_timer); - dd->intrchk_timer.function = verify_interrupt; - dd->intrchk_timer.data = (unsigned long) dd; + setup_timer(&dd->intrchk_timer, verify_interrupt, + (unsigned long)dd); done: return ret; } @@ -755,9 +753,8 @@ done: continue; if (dd->flags & QIB_HAS_SEND_DMA) ret = qib_setup_sdma(ppd); - init_timer(&ppd->hol_timer); - ppd->hol_timer.function = qib_hol_event; - ppd->hol_timer.data = (unsigned long)ppd; + setup_timer(&ppd->hol_timer, qib_hol_event, + (unsigned long)ppd); ppd->hol_state = QIB_HOL_UP; } diff --git a/drivers/infiniband/hw/qib/qib_mad.c b/drivers/infiniband/hw/qib/qib_mad.c index d2ac29861af5..da295e0392ed 100644 --- a/drivers/infiniband/hw/qib/qib_mad.c +++ b/drivers/infiniband/hw/qib/qib_mad.c @@ -717,9 +717,10 @@ static int subn_set_portinfo(struct ib_smp *smp, struct ib_device *ibdev, spin_lock_irqsave(&ibp->rvp.lock, flags); if (ibp->rvp.sm_ah) { if (smlid != ibp->rvp.sm_lid) - ibp->rvp.sm_ah->attr.dlid = smlid; + rdma_ah_set_dlid(&ibp->rvp.sm_ah->attr, + smlid); if (msl != ibp->rvp.sm_sl) - ibp->rvp.sm_ah->attr.sl = msl; + rdma_ah_set_sl(&ibp->rvp.sm_ah->attr, msl); } spin_unlock_irqrestore(&ibp->rvp.lock, flags); if (smlid != ibp->rvp.sm_lid) @@ -2500,5 +2501,5 @@ void qib_notify_free_mad_agent(struct rvt_dev_info *rdi, int port_idx) del_timer_sync(&dd->pport[port_idx].cong_stats.timer); if (dd->pport[port_idx].ibport_data.smi_ah) - ib_destroy_ah(&dd->pport[port_idx].ibport_data.smi_ah->ibah); + rdma_destroy_ah(&dd->pport[port_idx].ibport_data.smi_ah->ibah); } diff --git a/drivers/infiniband/hw/qib/qib_qp.c b/drivers/infiniband/hw/qib/qib_qp.c index 2ac0c0f79e74..5984981e7dd4 100644 --- a/drivers/infiniband/hw/qib/qib_qp.c +++ b/drivers/infiniband/hw/qib/qib_qp.c @@ -489,7 +489,7 @@ void qib_qp_iter_print(struct seq_file *s, struct qib_qp_iter *iter) qp->s_last, qp->s_acked, qp->s_cur, qp->s_tail, qp->s_head, qp->s_size, qp->remote_qpn, - qp->remote_ah_attr.dlid); + rdma_ah_get_dlid(&qp->remote_ah_attr)); } #endif diff --git a/drivers/infiniband/hw/qib/qib_rc.c b/drivers/infiniband/hw/qib/qib_rc.c index 12658e3fe154..fc8b88514da5 100644 --- a/drivers/infiniband/hw/qib/qib_rc.c +++ b/drivers/infiniband/hw/qib/qib_rc.c @@ -234,7 +234,7 @@ int qib_make_rc_req(struct rvt_qp *qp, unsigned long *flags) int delta; ohdr = &priv->s_hdr->u.oth; - if (qp->remote_ah_attr.ah_flags & IB_AH_GRH) + if (rdma_ah_get_ah_flags(&qp->remote_ah_attr) & IB_AH_GRH) ohdr = &priv->s_hdr->u.l.oth; /* Sending responses has higher priority over sending requests. */ @@ -637,9 +637,11 @@ void qib_send_rc_ack(struct rvt_qp *qp) lrh0 = QIB_LRH_BTH; /* header size in 32-bit words LRH+BTH+AETH = (8+12+4)/4. */ hwords = 6; - if (unlikely(qp->remote_ah_attr.ah_flags & IB_AH_GRH)) { + if (unlikely(rdma_ah_get_ah_flags(&qp->remote_ah_attr) & + IB_AH_GRH)) { hwords += qib_make_grh(ibp, &hdr.u.l.grh, - &qp->remote_ah_attr.grh, hwords, 0); + rdma_ah_read_grh(&qp->remote_ah_attr), + hwords, 0); ohdr = &hdr.u.l.oth; lrh0 = QIB_LRH_GRH; } @@ -653,12 +655,13 @@ void qib_send_rc_ack(struct rvt_qp *qp) IB_AETH_CREDIT_SHIFT)); else ohdr->u.aeth = rvt_compute_aeth(qp); - lrh0 |= ibp->sl_to_vl[qp->remote_ah_attr.sl] << 12 | - qp->remote_ah_attr.sl << 4; + lrh0 |= ibp->sl_to_vl[rdma_ah_get_sl(&qp->remote_ah_attr)] << 12 | + rdma_ah_get_sl(&qp->remote_ah_attr) << 4; hdr.lrh[0] = cpu_to_be16(lrh0); - hdr.lrh[1] = cpu_to_be16(qp->remote_ah_attr.dlid); + hdr.lrh[1] = cpu_to_be16(rdma_ah_get_dlid(&qp->remote_ah_attr)); hdr.lrh[2] = cpu_to_be16(hwords + SIZE_OF_CRC); - hdr.lrh[3] = cpu_to_be16(ppd->lid | qp->remote_ah_attr.src_path_bits); + hdr.lrh[3] = cpu_to_be16(ppd->lid | + rdma_ah_get_path_bits(&qp->remote_ah_attr)); ohdr->bth[0] = cpu_to_be32(bth0); ohdr->bth[1] = cpu_to_be32(qp->remote_qpn); ohdr->bth[2] = cpu_to_be32(qp->r_ack_psn & QIB_PSN_MASK); @@ -938,7 +941,10 @@ void qib_rc_send_complete(struct rvt_qp *qp, struct ib_header *hdr) /* see post_send() */ barrier(); rvt_put_swqe(wqe); - rvt_qp_swqe_complete(qp, wqe, IB_WC_SUCCESS); + rvt_qp_swqe_complete(qp, + wqe, + ib_qib_wc_opcode[wqe->wr.opcode], + IB_WC_SUCCESS); } /* * If we were waiting for sends to complete before resending, @@ -983,7 +989,10 @@ static struct rvt_swqe *do_rc_completion(struct rvt_qp *qp, qp->s_last = s_last; /* see post_send() */ barrier(); - rvt_qp_swqe_complete(qp, wqe, IB_WC_SUCCESS); + rvt_qp_swqe_complete(qp, + wqe, + ib_qib_wc_opcode[wqe->wr.opcode], + IB_WC_SUCCESS); } else this_cpu_inc(*ibp->rvp.rc_delayed_comp); @@ -1898,8 +1907,8 @@ send_last: wc.opcode = IB_WC_RECV; wc.qp = &qp->ibqp; wc.src_qp = qp->remote_qpn; - wc.slid = qp->remote_ah_attr.dlid; - wc.sl = qp->remote_ah_attr.sl; + wc.slid = rdma_ah_get_dlid(&qp->remote_ah_attr); + wc.sl = rdma_ah_get_sl(&qp->remote_ah_attr); /* zero fields that are N/A */ wc.vendor_err = 0; wc.pkey_index = 0; diff --git a/drivers/infiniband/hw/qib/qib_ruc.c b/drivers/infiniband/hw/qib/qib_ruc.c index 17655cc3e6fe..bd09de7c6e56 100644 --- a/drivers/infiniband/hw/qib/qib_ruc.c +++ b/drivers/infiniband/hw/qib/qib_ruc.c @@ -195,7 +195,7 @@ void qib_migrate_qp(struct rvt_qp *qp) qp->s_mig_state = IB_MIG_MIGRATED; qp->remote_ah_attr = qp->alt_ah_attr; - qp->port_num = qp->alt_ah_attr.port_num; + qp->port_num = rdma_ah_get_port_num(&qp->alt_ah_attr); qp->s_pkey_index = qp->s_alt_pkey_index; ev.device = qp->ibqp.device; @@ -235,18 +235,23 @@ int qib_ruc_check_hdr(struct qib_ibport *ibp, struct ib_header *hdr, if (qp->s_mig_state == IB_MIG_ARMED && (bth0 & IB_BTH_MIG_REQ)) { if (!has_grh) { - if (qp->alt_ah_attr.ah_flags & IB_AH_GRH) + if (rdma_ah_get_ah_flags(&qp->alt_ah_attr) & + IB_AH_GRH) goto err; } else { - if (!(qp->alt_ah_attr.ah_flags & IB_AH_GRH)) + const struct ib_global_route *grh; + + if (!(rdma_ah_get_ah_flags(&qp->alt_ah_attr) & + IB_AH_GRH)) goto err; - guid = get_sguid(ibp, qp->alt_ah_attr.grh.sgid_index); + grh = rdma_ah_read_grh(&qp->alt_ah_attr); + guid = get_sguid(ibp, grh->sgid_index); if (!gid_ok(&hdr->u.l.grh.dgid, ibp->rvp.gid_prefix, guid)) goto err; if (!gid_ok(&hdr->u.l.grh.sgid, - qp->alt_ah_attr.grh.dgid.global.subnet_prefix, - qp->alt_ah_attr.grh.dgid.global.interface_id)) + grh->dgid.global.subnet_prefix, + grh->dgid.global.interface_id)) goto err; } if (!qib_pkey_ok((u16)bth0, @@ -259,27 +264,33 @@ int qib_ruc_check_hdr(struct qib_ibport *ibp, struct ib_header *hdr, goto err; } /* Validate the SLID. See Ch. 9.6.1.5 and 17.2.8 */ - if (be16_to_cpu(hdr->lrh[3]) != qp->alt_ah_attr.dlid || - ppd_from_ibp(ibp)->port != qp->alt_ah_attr.port_num) + if ((be16_to_cpu(hdr->lrh[3]) != + rdma_ah_get_dlid(&qp->alt_ah_attr)) || + ppd_from_ibp(ibp)->port != + rdma_ah_get_port_num(&qp->alt_ah_attr)) goto err; spin_lock_irqsave(&qp->s_lock, flags); qib_migrate_qp(qp); spin_unlock_irqrestore(&qp->s_lock, flags); } else { if (!has_grh) { - if (qp->remote_ah_attr.ah_flags & IB_AH_GRH) + if (rdma_ah_get_ah_flags(&qp->remote_ah_attr) & + IB_AH_GRH) goto err; } else { - if (!(qp->remote_ah_attr.ah_flags & IB_AH_GRH)) + const struct ib_global_route *grh; + + if (!(rdma_ah_get_ah_flags(&qp->remote_ah_attr) & + IB_AH_GRH)) goto err; - guid = get_sguid(ibp, - qp->remote_ah_attr.grh.sgid_index); + grh = rdma_ah_read_grh(&qp->remote_ah_attr); + guid = get_sguid(ibp, grh->sgid_index); if (!gid_ok(&hdr->u.l.grh.dgid, ibp->rvp.gid_prefix, guid)) goto err; if (!gid_ok(&hdr->u.l.grh.sgid, - qp->remote_ah_attr.grh.dgid.global.subnet_prefix, - qp->remote_ah_attr.grh.dgid.global.interface_id)) + grh->dgid.global.subnet_prefix, + grh->dgid.global.interface_id)) goto err; } if (!qib_pkey_ok((u16)bth0, @@ -292,7 +303,8 @@ int qib_ruc_check_hdr(struct qib_ibport *ibp, struct ib_header *hdr, goto err; } /* Validate the SLID. See Ch. 9.6.1.5 */ - if (be16_to_cpu(hdr->lrh[3]) != qp->remote_ah_attr.dlid || + if (be16_to_cpu(hdr->lrh[3]) != + rdma_ah_get_dlid(&qp->remote_ah_attr) || ppd_from_ibp(ibp)->port != qp->port_num) goto err; if (qp->s_mig_state == IB_MIG_REARM && @@ -528,8 +540,8 @@ again: wc.byte_len = wqe->length; wc.qp = &qp->ibqp; wc.src_qp = qp->remote_qpn; - wc.slid = qp->remote_ah_attr.dlid; - wc.sl = qp->remote_ah_attr.sl; + wc.slid = rdma_ah_get_dlid(&qp->remote_ah_attr); + wc.sl = rdma_ah_get_sl(&qp->remote_ah_attr); wc.port_num = 1; /* Signal completion event if the solicited bit is set. */ rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc, @@ -619,7 +631,7 @@ done: * Return the size of the header in 32 bit words. */ u32 qib_make_grh(struct qib_ibport *ibp, struct ib_grh *hdr, - struct ib_global_route *grh, u32 hwords, u32 nwords) + const struct ib_global_route *grh, u32 hwords, u32 nwords) { hdr->version_tclass_flow = cpu_to_be32((IB_GRH_VERSION << IB_GRH_VERSION_SHIFT) | @@ -652,20 +664,23 @@ void qib_make_ruc_header(struct rvt_qp *qp, struct ib_other_headers *ohdr, extra_bytes = -qp->s_cur_size & 3; nwords = (qp->s_cur_size + extra_bytes) >> 2; lrh0 = QIB_LRH_BTH; - if (unlikely(qp->remote_ah_attr.ah_flags & IB_AH_GRH)) { - qp->s_hdrwords += qib_make_grh(ibp, &priv->s_hdr->u.l.grh, - &qp->remote_ah_attr.grh, - qp->s_hdrwords, nwords); + if (unlikely(rdma_ah_get_ah_flags(&qp->remote_ah_attr) & IB_AH_GRH)) { + qp->s_hdrwords += + qib_make_grh(ibp, &priv->s_hdr->u.l.grh, + rdma_ah_read_grh(&qp->remote_ah_attr), + qp->s_hdrwords, nwords); lrh0 = QIB_LRH_GRH; } - lrh0 |= ibp->sl_to_vl[qp->remote_ah_attr.sl] << 12 | - qp->remote_ah_attr.sl << 4; + lrh0 |= ibp->sl_to_vl[rdma_ah_get_sl(&qp->remote_ah_attr)] << 12 | + rdma_ah_get_sl(&qp->remote_ah_attr) << 4; priv->s_hdr->lrh[0] = cpu_to_be16(lrh0); - priv->s_hdr->lrh[1] = cpu_to_be16(qp->remote_ah_attr.dlid); + priv->s_hdr->lrh[1] = + cpu_to_be16(rdma_ah_get_dlid(&qp->remote_ah_attr)); priv->s_hdr->lrh[2] = cpu_to_be16(qp->s_hdrwords + nwords + SIZE_OF_CRC); - priv->s_hdr->lrh[3] = cpu_to_be16(ppd_from_ibp(ibp)->lid | - qp->remote_ah_attr.src_path_bits); + priv->s_hdr->lrh[3] = + cpu_to_be16(ppd_from_ibp(ibp)->lid | + rdma_ah_get_path_bits(&qp->remote_ah_attr)); bth0 |= qib_get_pkey(ibp, qp->s_pkey_index); bth0 |= extra_bytes << 20; if (qp->s_mig_state == IB_MIG_MIGRATED) @@ -703,7 +718,8 @@ void qib_do_send(struct rvt_qp *qp) if ((qp->ibqp.qp_type == IB_QPT_RC || qp->ibqp.qp_type == IB_QPT_UC) && - (qp->remote_ah_attr.dlid & ~((1 << ppd->lmc) - 1)) == ppd->lid) { + (rdma_ah_get_dlid(&qp->remote_ah_attr) & + ~((1 << ppd->lmc) - 1)) == ppd->lid) { qib_ruc_loopback(qp); return; } @@ -769,7 +785,10 @@ void qib_send_complete(struct rvt_qp *qp, struct rvt_swqe *wqe, qp->ibqp.qp_type == IB_QPT_GSI) atomic_dec(&ibah_to_rvtah(wqe->ud_wr.ah)->refcount); - rvt_qp_swqe_complete(qp, wqe, status); + rvt_qp_swqe_complete(qp, + wqe, + ib_qib_wc_opcode[wqe->wr.opcode], + status); if (qp->s_acked == old_last) qp->s_acked = last; diff --git a/drivers/infiniband/hw/qib/qib_uc.c b/drivers/infiniband/hw/qib/qib_uc.c index b337b60fc40d..498e2202e72c 100644 --- a/drivers/infiniband/hw/qib/qib_uc.c +++ b/drivers/infiniband/hw/qib/qib_uc.c @@ -74,7 +74,7 @@ int qib_make_uc_req(struct rvt_qp *qp, unsigned long *flags) } ohdr = &priv->s_hdr->u.oth; - if (qp->remote_ah_attr.ah_flags & IB_AH_GRH) + if (rdma_ah_get_ah_flags(&qp->remote_ah_attr) & IB_AH_GRH) ohdr = &priv->s_hdr->u.l.oth; /* header size in 32-bit words LRH+BTH = (8+12)/4. */ @@ -394,8 +394,8 @@ last_imm: wc.status = IB_WC_SUCCESS; wc.qp = &qp->ibqp; wc.src_qp = qp->remote_qpn; - wc.slid = qp->remote_ah_attr.dlid; - wc.sl = qp->remote_ah_attr.sl; + wc.slid = rdma_ah_get_dlid(&qp->remote_ah_attr); + wc.sl = rdma_ah_get_sl(&qp->remote_ah_attr); /* zero fields that are N/A */ wc.vendor_err = 0; wc.pkey_index = 0; diff --git a/drivers/infiniband/hw/qib/qib_ud.c b/drivers/infiniband/hw/qib/qib_ud.c index ddd4e7458750..341a123ee95c 100644 --- a/drivers/infiniband/hw/qib/qib_ud.c +++ b/drivers/infiniband/hw/qib/qib_ud.c @@ -54,7 +54,7 @@ static void qib_ud_loopback(struct rvt_qp *sqp, struct rvt_swqe *swqe) struct qib_devdata *dd = ppd->dd; struct rvt_dev_info *rdi = &dd->verbs_dev.rdi; struct rvt_qp *qp; - struct ib_ah_attr *ah_attr; + struct rdma_ah_attr *ah_attr; unsigned long flags; struct rvt_sge_state ssge; struct rvt_sge *sge; @@ -92,13 +92,13 @@ static void qib_ud_loopback(struct rvt_qp *sqp, struct rvt_swqe *swqe) pkey1 = qib_get_pkey(ibp, sqp->s_pkey_index); pkey2 = qib_get_pkey(ibp, qp->s_pkey_index); if (unlikely(!qib_pkey_ok(pkey1, pkey2))) { - lid = ppd->lid | (ah_attr->src_path_bits & + lid = ppd->lid | (rdma_ah_get_path_bits(ah_attr) & ((1 << ppd->lmc) - 1)); qib_bad_pqkey(ibp, IB_NOTICE_TRAP_BAD_PKEY, pkey1, - ah_attr->sl, + rdma_ah_get_sl(ah_attr), sqp->ibqp.qp_num, qp->ibqp.qp_num, cpu_to_be16(lid), - cpu_to_be16(ah_attr->dlid)); + cpu_to_be16(rdma_ah_get_dlid(ah_attr))); goto drop; } } @@ -116,13 +116,13 @@ static void qib_ud_loopback(struct rvt_qp *sqp, struct rvt_swqe *swqe) if (unlikely(qkey != qp->qkey)) { u16 lid; - lid = ppd->lid | (ah_attr->src_path_bits & + lid = ppd->lid | (rdma_ah_get_path_bits(ah_attr) & ((1 << ppd->lmc) - 1)); qib_bad_pqkey(ibp, IB_NOTICE_TRAP_BAD_QKEY, qkey, - ah_attr->sl, + rdma_ah_get_sl(ah_attr), sqp->ibqp.qp_num, qp->ibqp.qp_num, cpu_to_be16(lid), - cpu_to_be16(ah_attr->dlid)); + cpu_to_be16(rdma_ah_get_dlid(ah_attr))); goto drop; } } @@ -168,11 +168,11 @@ static void qib_ud_loopback(struct rvt_qp *sqp, struct rvt_swqe *swqe) goto bail_unlock; } - if (ah_attr->ah_flags & IB_AH_GRH) { + if (rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH) { struct ib_grh grh; - struct ib_global_route grd = ah_attr->grh; + const struct ib_global_route *grd = rdma_ah_read_grh(ah_attr); - qib_make_grh(ibp, &grh, &grd, 0, 0); + qib_make_grh(ibp, &grh, grd, 0, 0); qib_copy_sge(&qp->r_sge, &grh, sizeof(grh), 1); wc.wc_flags |= IB_WC_GRH; @@ -220,9 +220,10 @@ static void qib_ud_loopback(struct rvt_qp *sqp, struct rvt_swqe *swqe) wc.src_qp = sqp->ibqp.qp_num; wc.pkey_index = qp->ibqp.qp_type == IB_QPT_GSI ? swqe->ud_wr.pkey_index : 0; - wc.slid = ppd->lid | (ah_attr->src_path_bits & ((1 << ppd->lmc) - 1)); - wc.sl = ah_attr->sl; - wc.dlid_path_bits = ah_attr->dlid & ((1 << ppd->lmc) - 1); + wc.slid = ppd->lid | (rdma_ah_get_path_bits(ah_attr) & + ((1 << ppd->lmc) - 1)); + wc.sl = rdma_ah_get_sl(ah_attr); + wc.dlid_path_bits = rdma_ah_get_dlid(ah_attr) & ((1 << ppd->lmc) - 1); wc.port_num = qp->port_num; /* Signal completion event if the solicited bit is set. */ rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc, @@ -246,7 +247,7 @@ int qib_make_ud_req(struct rvt_qp *qp, unsigned long *flags) { struct qib_qp_priv *priv = qp->priv; struct ib_other_headers *ohdr; - struct ib_ah_attr *ah_attr; + struct rdma_ah_attr *ah_attr; struct qib_pportdata *ppd; struct qib_ibport *ibp; struct rvt_swqe *wqe; @@ -289,14 +290,15 @@ int qib_make_ud_req(struct rvt_qp *qp, unsigned long *flags) ibp = to_iport(qp->ibqp.device, qp->port_num); ppd = ppd_from_ibp(ibp); ah_attr = &ibah_to_rvtah(wqe->ud_wr.ah)->attr; - if (ah_attr->dlid >= be16_to_cpu(IB_MULTICAST_LID_BASE)) { - if (ah_attr->dlid != be16_to_cpu(IB_LID_PERMISSIVE)) + if (rdma_ah_get_dlid(ah_attr) >= be16_to_cpu(IB_MULTICAST_LID_BASE)) { + if (rdma_ah_get_dlid(ah_attr) != + be16_to_cpu(IB_LID_PERMISSIVE)) this_cpu_inc(ibp->pmastats->n_multicast_xmit); else this_cpu_inc(ibp->pmastats->n_unicast_xmit); } else { this_cpu_inc(ibp->pmastats->n_unicast_xmit); - lid = ah_attr->dlid & ~((1 << ppd->lmc) - 1); + lid = rdma_ah_get_dlid(ah_attr) & ~((1 << ppd->lmc) - 1); if (unlikely(lid == ppd->lid)) { unsigned long tflags = *flags; /* @@ -328,17 +330,17 @@ int qib_make_ud_req(struct rvt_qp *qp, unsigned long *flags) qp->s_hdrwords = 7; qp->s_cur_size = wqe->length; qp->s_cur_sge = &qp->s_sge; - qp->s_srate = ah_attr->static_rate; + qp->s_srate = rdma_ah_get_static_rate(ah_attr); qp->s_wqe = wqe; qp->s_sge.sge = wqe->sg_list[0]; qp->s_sge.sg_list = wqe->sg_list + 1; qp->s_sge.num_sge = wqe->wr.num_sge; qp->s_sge.total_len = wqe->length; - if (ah_attr->ah_flags & IB_AH_GRH) { + if (rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH) { /* Header size in 32-bit words. */ qp->s_hdrwords += qib_make_grh(ibp, &priv->s_hdr->u.l.grh, - &ah_attr->grh, + rdma_ah_read_grh(ah_attr), qp->s_hdrwords, nwords); lrh0 = QIB_LRH_GRH; ohdr = &priv->s_hdr->u.l.oth; @@ -357,18 +359,20 @@ int qib_make_ud_req(struct rvt_qp *qp, unsigned long *flags) bth0 = IB_OPCODE_UD_SEND_ONLY_WITH_IMMEDIATE << 24; } else bth0 = IB_OPCODE_UD_SEND_ONLY << 24; - lrh0 |= ah_attr->sl << 4; + lrh0 |= rdma_ah_get_sl(ah_attr) << 4; if (qp->ibqp.qp_type == IB_QPT_SMI) lrh0 |= 0xF000; /* Set VL (see ch. 13.5.3.1) */ else - lrh0 |= ibp->sl_to_vl[ah_attr->sl] << 12; + lrh0 |= ibp->sl_to_vl[rdma_ah_get_sl(ah_attr)] << 12; priv->s_hdr->lrh[0] = cpu_to_be16(lrh0); - priv->s_hdr->lrh[1] = cpu_to_be16(ah_attr->dlid); /* DEST LID */ + priv->s_hdr->lrh[1] = + cpu_to_be16(rdma_ah_get_dlid(ah_attr)); /* DEST LID */ priv->s_hdr->lrh[2] = cpu_to_be16(qp->s_hdrwords + nwords + SIZE_OF_CRC); lid = ppd->lid; if (lid) { - lid |= ah_attr->src_path_bits & ((1 << ppd->lmc) - 1); + lid |= rdma_ah_get_path_bits(ah_attr) & + ((1 << ppd->lmc) - 1); priv->s_hdr->lrh[3] = cpu_to_be16(lid); } else priv->s_hdr->lrh[3] = IB_LID_PERMISSIVE; @@ -382,8 +386,9 @@ int qib_make_ud_req(struct rvt_qp *qp, unsigned long *flags) /* * Use the multicast QP if the destination LID is a multicast LID. */ - ohdr->bth[1] = ah_attr->dlid >= be16_to_cpu(IB_MULTICAST_LID_BASE) && - ah_attr->dlid != be16_to_cpu(IB_LID_PERMISSIVE) ? + ohdr->bth[1] = rdma_ah_get_dlid(ah_attr) >= + be16_to_cpu(IB_MULTICAST_LID_BASE) && + rdma_ah_get_dlid(ah_attr) != be16_to_cpu(IB_LID_PERMISSIVE) ? cpu_to_be32(QIB_MULTICAST_QPN) : cpu_to_be32(wqe->ud_wr.remote_qpn); ohdr->bth[2] = cpu_to_be32(wqe->psn & QIB_PSN_MASK); diff --git a/drivers/infiniband/hw/qib/qib_verbs.c b/drivers/infiniband/hw/qib/qib_verbs.c index 83f8b5f24381..ac42dce7e281 100644 --- a/drivers/infiniband/hw/qib/qib_verbs.c +++ b/drivers/infiniband/hw/qib/qib_verbs.c @@ -114,6 +114,19 @@ module_param_named(disable_sma, ib_qib_disable_sma, uint, S_IWUSR | S_IRUGO); MODULE_PARM_DESC(disable_sma, "Disable the SMA"); /* + * Translate ib_wr_opcode into ib_wc_opcode. + */ +const enum ib_wc_opcode ib_qib_wc_opcode[] = { + [IB_WR_RDMA_WRITE] = IB_WC_RDMA_WRITE, + [IB_WR_RDMA_WRITE_WITH_IMM] = IB_WC_RDMA_WRITE, + [IB_WR_SEND] = IB_WC_SEND, + [IB_WR_SEND_WITH_IMM] = IB_WC_SEND, + [IB_WR_RDMA_READ] = IB_WC_RDMA_READ, + [IB_WR_ATOMIC_CMP_AND_SWP] = IB_WC_COMP_SWAP, + [IB_WR_ATOMIC_FETCH_AND_ADD] = IB_WC_FETCH_ADD +}; + +/* * System image GUID. */ __be64 ib_qib_sys_image_guid; @@ -343,7 +356,7 @@ void qib_ib_rcv(struct qib_ctxtdata *rcd, void *rhdr, void *data, u32 tlen) if (lnh != QIB_LRH_GRH) goto drop; - mcast = rvt_mcast_find(&ibp->rvp, &hdr->u.l.grh.dgid); + mcast = rvt_mcast_find(&ibp->rvp, &hdr->u.l.grh.dgid, lid); if (mcast == NULL) goto drop; this_cpu_inc(ibp->pmastats->n_multicast_rcv); @@ -1323,16 +1336,16 @@ static int qib_get_guid_be(struct rvt_dev_info *rdi, struct rvt_ibport *rvp, return 0; } -int qib_check_ah(struct ib_device *ibdev, struct ib_ah_attr *ah_attr) +int qib_check_ah(struct ib_device *ibdev, struct rdma_ah_attr *ah_attr) { - if (ah_attr->sl > 15) + if (rdma_ah_get_sl(ah_attr) > 15) return -EINVAL; return 0; } static void qib_notify_new_ah(struct ib_device *ibdev, - struct ib_ah_attr *ah_attr, + struct rdma_ah_attr *ah_attr, struct rvt_ah *ah) { struct qib_ibport *ibp; @@ -1343,25 +1356,29 @@ static void qib_notify_new_ah(struct ib_device *ibdev, * done being setup. We can however modify things which we need to set. */ - ibp = to_iport(ibdev, ah_attr->port_num); + ibp = to_iport(ibdev, rdma_ah_get_port_num(ah_attr)); ppd = ppd_from_ibp(ibp); - ah->vl = ibp->sl_to_vl[ah->attr.sl]; + ah->vl = ibp->sl_to_vl[rdma_ah_get_sl(&ah->attr)]; ah->log_pmtu = ilog2(ppd->ibmtu); } struct ib_ah *qib_create_qp0_ah(struct qib_ibport *ibp, u16 dlid) { - struct ib_ah_attr attr; + struct rdma_ah_attr attr; struct ib_ah *ah = ERR_PTR(-EINVAL); struct rvt_qp *qp0; + struct qib_pportdata *ppd = ppd_from_ibp(ibp); + struct qib_devdata *dd = dd_from_ppd(ppd); + u8 port_num = ppd->port; memset(&attr, 0, sizeof(attr)); - attr.dlid = dlid; - attr.port_num = ppd_from_ibp(ibp)->port; + attr.type = rdma_ah_find_type(&dd->verbs_dev.rdi.ibdev, port_num); + rdma_ah_set_dlid(&attr, dlid); + rdma_ah_set_port_num(&attr, port_num); rcu_read_lock(); qp0 = rcu_dereference(ibp->rvp.qp[0]); if (qp0) - ah = ib_create_ah(qp0->ibqp.pd, &attr); + ah = rdma_create_ah(qp0->ibqp.pd, &attr); rcu_read_unlock(); return ah; } diff --git a/drivers/infiniband/hw/qib/qib_verbs.h b/drivers/infiniband/hw/qib/qib_verbs.h index 212e8ce71be8..da0db5485ddc 100644 --- a/drivers/infiniband/hw/qib/qib_verbs.h +++ b/drivers/infiniband/hw/qib/qib_verbs.h @@ -310,7 +310,7 @@ void qib_uc_rcv(struct qib_ibport *ibp, struct ib_header *hdr, void qib_rc_rcv(struct qib_ctxtdata *rcd, struct ib_header *hdr, int has_grh, void *data, u32 tlen, struct rvt_qp *qp); -int qib_check_ah(struct ib_device *ibdev, struct ib_ah_attr *ah_attr); +int qib_check_ah(struct ib_device *ibdev, struct rdma_ah_attr *ah_attr); int qib_check_send_wqe(struct rvt_qp *qp, struct rvt_swqe *wqe); @@ -335,7 +335,7 @@ int qib_ruc_check_hdr(struct qib_ibport *ibp, struct ib_header *hdr, int has_grh, struct rvt_qp *qp, u32 bth0); u32 qib_make_grh(struct qib_ibport *ibp, struct ib_grh *hdr, - struct ib_global_route *grh, u32 hwords, u32 nwords); + const struct ib_global_route *grh, u32 hwords, u32 nwords); void qib_make_ruc_header(struct rvt_qp *qp, struct ib_other_headers *ohdr, u32 bth0, u32 bth2); diff --git a/drivers/infiniband/hw/usnic/usnic_common_util.h b/drivers/infiniband/hw/usnic/usnic_common_util.h index b54986de5f0c..ddd81294fa46 100644 --- a/drivers/infiniband/hw/usnic/usnic_common_util.h +++ b/drivers/infiniband/hw/usnic/usnic_common_util.h @@ -34,21 +34,7 @@ #ifndef USNIC_CMN_UTIL_H #define USNIC_CMN_UTIL_H -static inline void -usnic_mac_to_gid(const char *const mac, char *raw_gid) -{ - raw_gid[0] = 0xfe; - raw_gid[1] = 0x80; - memset(&raw_gid[2], 0, 6); - raw_gid[8] = mac[0]^2; - raw_gid[9] = mac[1]; - raw_gid[10] = mac[2]; - raw_gid[11] = 0xff; - raw_gid[12] = 0xfe; - raw_gid[13] = mac[3]; - raw_gid[14] = mac[4]; - raw_gid[15] = mac[5]; -} +#include <net/addrconf.h> static inline void usnic_mac_ip_to_gid(const char *const mac, const __be32 inaddr, char *raw_gid) @@ -57,27 +43,7 @@ usnic_mac_ip_to_gid(const char *const mac, const __be32 inaddr, char *raw_gid) raw_gid[1] = 0x80; memset(&raw_gid[2], 0, 2); memcpy(&raw_gid[4], &inaddr, 4); - raw_gid[8] = mac[0]^2; - raw_gid[9] = mac[1]; - raw_gid[10] = mac[2]; - raw_gid[11] = 0xff; - raw_gid[12] = 0xfe; - raw_gid[13] = mac[3]; - raw_gid[14] = mac[4]; - raw_gid[15] = mac[5]; -} - -static inline void -usnic_write_gid_if_id_from_mac(char *mac, char *raw_gid) -{ - raw_gid[8] = mac[0]^2; - raw_gid[9] = mac[1]; - raw_gid[10] = mac[2]; - raw_gid[11] = 0xff; - raw_gid[12] = 0xfe; - raw_gid[13] = mac[3]; - raw_gid[14] = mac[4]; - raw_gid[15] = mac[5]; + addrconf_addr_eui48(&raw_gid[8], mac); } #endif /* USNIC_COMMON_UTIL_H */ diff --git a/drivers/infiniband/hw/usnic/usnic_ib_sysfs.c b/drivers/infiniband/hw/usnic/usnic_ib_sysfs.c index 04443242e258..32956f9f5715 100644 --- a/drivers/infiniband/hw/usnic/usnic_ib_sysfs.c +++ b/drivers/infiniband/hw/usnic/usnic_ib_sysfs.c @@ -44,6 +44,7 @@ #include "usnic_vnic.h" #include "usnic_ib_verbs.h" #include "usnic_log.h" +#include "usnic_ib_sysfs.h" static ssize_t usnic_ib_show_board(struct device *device, struct device_attribute *attr, diff --git a/drivers/infiniband/hw/usnic/usnic_ib_verbs.c b/drivers/infiniband/hw/usnic/usnic_ib_verbs.c index 3284730d3c09..4996984885c2 100644 --- a/drivers/infiniband/hw/usnic/usnic_ib_verbs.c +++ b/drivers/infiniband/hw/usnic/usnic_ib_verbs.c @@ -46,6 +46,7 @@ #include "usnic_log.h" #include "usnic_uiom.h" #include "usnic_transport.h" +#include "usnic_ib_verbs.h" #define USNIC_DEFAULT_TRANSPORT USNIC_TRANSPORT_ROCE_CUSTOM @@ -151,7 +152,7 @@ find_free_vf_and_create_qp_grp(struct usnic_ib_dev *us_ibdev, struct usnic_vnic *vnic; struct usnic_ib_qp_grp *qp_grp; struct device *dev, **dev_list; - int i, found = 0; + int i; BUG_ON(!mutex_is_locked(&us_ibdev->usdev_lock)); @@ -173,8 +174,13 @@ find_free_vf_and_create_qp_grp(struct usnic_ib_dev *us_ibdev, us_ibdev->ib_dev.name, pci_name(usnic_vnic_get_pdev( vnic))); - found = 1; - break; + qp_grp = usnic_ib_qp_grp_create(us_ibdev->ufdev, + vf, pd, + res_spec, + trans_spec); + + spin_unlock(&vf->lock); + goto qp_grp_check; } spin_unlock(&vf->lock); @@ -182,34 +188,30 @@ find_free_vf_and_create_qp_grp(struct usnic_ib_dev *us_ibdev, usnic_uiom_free_dev_list(dev_list); } - if (!found) { - /* Try to find resources on an unused vf */ - list_for_each_entry(vf, &us_ibdev->vf_dev_list, link) { - spin_lock(&vf->lock); - vnic = vf->vnic; - if (vf->qp_grp_ref_cnt == 0 && - usnic_vnic_check_room(vnic, res_spec) == 0) { - found = 1; - break; - } + /* Try to find resources on an unused vf */ + list_for_each_entry(vf, &us_ibdev->vf_dev_list, link) { + spin_lock(&vf->lock); + vnic = vf->vnic; + if (vf->qp_grp_ref_cnt == 0 && + usnic_vnic_check_room(vnic, res_spec) == 0) { + qp_grp = usnic_ib_qp_grp_create(us_ibdev->ufdev, vf, + pd, res_spec, + trans_spec); + spin_unlock(&vf->lock); + goto qp_grp_check; } + spin_unlock(&vf->lock); } - if (!found) { - usnic_info("No free qp grp found on %s\n", - us_ibdev->ib_dev.name); - return ERR_PTR(-ENOMEM); - } + usnic_info("No free qp grp found on %s\n", us_ibdev->ib_dev.name); + return ERR_PTR(-ENOMEM); - qp_grp = usnic_ib_qp_grp_create(us_ibdev->ufdev, vf, pd, res_spec, - trans_spec); - spin_unlock(&vf->lock); +qp_grp_check: if (IS_ERR_OR_NULL(qp_grp)) { usnic_err("Failed to allocate qp_grp\n"); return ERR_PTR(qp_grp ? PTR_ERR(qp_grp) : -ENOMEM); } - return qp_grp; } @@ -738,7 +740,7 @@ int usnic_ib_mmap(struct ib_ucontext *context, /* In ib callbacks section - Start of stub funcs */ struct ib_ah *usnic_ib_create_ah(struct ib_pd *pd, - struct ib_ah_attr *ah_attr, + struct rdma_ah_attr *ah_attr, struct ib_udata *udata) { diff --git a/drivers/infiniband/hw/usnic/usnic_ib_verbs.h b/drivers/infiniband/hw/usnic/usnic_ib_verbs.h index 0ed8e072329e..172e43b6fa95 100644 --- a/drivers/infiniband/hw/usnic/usnic_ib_verbs.h +++ b/drivers/infiniband/hw/usnic/usnic_ib_verbs.h @@ -75,7 +75,7 @@ int usnic_ib_dealloc_ucontext(struct ib_ucontext *ibcontext); int usnic_ib_mmap(struct ib_ucontext *context, struct vm_area_struct *vma); struct ib_ah *usnic_ib_create_ah(struct ib_pd *pd, - struct ib_ah_attr *ah_attr, + struct rdma_ah_attr *ah_attr, struct ib_udata *udata); int usnic_ib_destroy_ah(struct ib_ah *ah); diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma.h b/drivers/infiniband/hw/vmw_pvrdma/pvrdma.h index 9fbe22d3467b..8e2f0a11690f 100644 --- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma.h +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma.h @@ -440,10 +440,10 @@ void pvrdma_global_route_to_ib(struct ib_global_route *dst, const struct pvrdma_global_route *src); void ib_global_route_to_pvrdma(struct pvrdma_global_route *dst, const struct ib_global_route *src); -void pvrdma_ah_attr_to_ib(struct ib_ah_attr *dst, - const struct pvrdma_ah_attr *src); -void ib_ah_attr_to_pvrdma(struct pvrdma_ah_attr *dst, - const struct ib_ah_attr *src); +void pvrdma_ah_attr_to_rdma(struct rdma_ah_attr *dst, + const struct pvrdma_ah_attr *src); +void rdma_ah_attr_to_pvrdma(struct pvrdma_ah_attr *dst, + const struct rdma_ah_attr *src); int pvrdma_uar_table_init(struct pvrdma_dev *dev); void pvrdma_uar_table_cleanup(struct pvrdma_dev *dev); diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_misc.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_misc.c index 948b5ccd2a70..ec6a4ca1eeb7 100644 --- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_misc.c +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_misc.c @@ -194,7 +194,7 @@ int pvrdma_page_dir_insert_umem(struct pvrdma_page_dir *pdir, len = sg_dma_len(sg) >> PAGE_SHIFT; for (j = 0; j < len; j++) { dma_addr_t addr = sg_dma_address(sg) + - umem->page_size * j; + (j << umem->page_shift); ret = pvrdma_page_dir_insert_dma(pdir, i, addr); if (ret) @@ -277,28 +277,29 @@ void ib_global_route_to_pvrdma(struct pvrdma_global_route *dst, dst->traffic_class = src->traffic_class; } -void pvrdma_ah_attr_to_ib(struct ib_ah_attr *dst, - const struct pvrdma_ah_attr *src) +void pvrdma_ah_attr_to_rdma(struct rdma_ah_attr *dst, + const struct pvrdma_ah_attr *src) { - pvrdma_global_route_to_ib(&dst->grh, &src->grh); - dst->dlid = src->dlid; - dst->sl = src->sl; - dst->src_path_bits = src->src_path_bits; - dst->static_rate = src->static_rate; - dst->ah_flags = src->ah_flags; - dst->port_num = src->port_num; - memcpy(&dst->dmac, &src->dmac, sizeof(dst->dmac)); + dst->type = RDMA_AH_ATTR_TYPE_ROCE; + pvrdma_global_route_to_ib(rdma_ah_retrieve_grh(dst), &src->grh); + rdma_ah_set_dlid(dst, src->dlid); + rdma_ah_set_sl(dst, src->sl); + rdma_ah_set_path_bits(dst, src->src_path_bits); + rdma_ah_set_static_rate(dst, src->static_rate); + rdma_ah_set_ah_flags(dst, src->ah_flags); + rdma_ah_set_port_num(dst, src->port_num); + memcpy(dst->roce.dmac, &src->dmac, ETH_ALEN); } -void ib_ah_attr_to_pvrdma(struct pvrdma_ah_attr *dst, - const struct ib_ah_attr *src) +void rdma_ah_attr_to_pvrdma(struct pvrdma_ah_attr *dst, + const struct rdma_ah_attr *src) { - ib_global_route_to_pvrdma(&dst->grh, &src->grh); - dst->dlid = src->dlid; - dst->sl = src->sl; - dst->src_path_bits = src->src_path_bits; - dst->static_rate = src->static_rate; - dst->ah_flags = src->ah_flags; - dst->port_num = src->port_num; - memcpy(&dst->dmac, &src->dmac, sizeof(dst->dmac)); + ib_global_route_to_pvrdma(&dst->grh, rdma_ah_read_grh(src)); + dst->dlid = rdma_ah_get_dlid(src); + dst->sl = rdma_ah_get_sl(src); + dst->src_path_bits = rdma_ah_get_path_bits(src); + dst->static_rate = rdma_ah_get_static_rate(src); + dst->ah_flags = rdma_ah_get_ah_flags(src); + dst->port_num = rdma_ah_get_port_num(src); + memcpy(&dst->dmac, src->roce.dmac, sizeof(dst->dmac)); } diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_qp.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_qp.c index 30062aad3af1..ed34d5a581fa 100644 --- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_qp.c +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_qp.c @@ -533,8 +533,8 @@ int pvrdma_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, cmd->attrs.alt_port_num = attr->alt_port_num; cmd->attrs.alt_timeout = attr->alt_timeout; ib_qp_cap_to_pvrdma(&cmd->attrs.cap, &attr->cap); - ib_ah_attr_to_pvrdma(&cmd->attrs.ah_attr, &attr->ah_attr); - ib_ah_attr_to_pvrdma(&cmd->attrs.alt_ah_attr, &attr->alt_ah_attr); + rdma_ah_attr_to_pvrdma(&cmd->attrs.ah_attr, &attr->ah_attr); + rdma_ah_attr_to_pvrdma(&cmd->attrs.alt_ah_attr, &attr->alt_ah_attr); ret = pvrdma_cmd_post(dev, &req, &rsp, PVRDMA_CMD_MODIFY_QP_RESP); if (ret < 0) { @@ -938,8 +938,8 @@ int pvrdma_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, attr->alt_port_num = resp->attrs.alt_port_num; attr->alt_timeout = resp->attrs.alt_timeout; pvrdma_qp_cap_to_ib(&attr->cap, &resp->attrs.cap); - pvrdma_ah_attr_to_ib(&attr->ah_attr, &resp->attrs.ah_attr); - pvrdma_ah_attr_to_ib(&attr->alt_ah_attr, &resp->attrs.alt_ah_attr); + pvrdma_ah_attr_to_rdma(&attr->ah_attr, &resp->attrs.ah_attr); + pvrdma_ah_attr_to_rdma(&attr->alt_ah_attr, &resp->attrs.alt_ah_attr); qp->state = attr->qp_state; diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c index fec17c49103b..28517042011d 100644 --- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c @@ -520,20 +520,20 @@ int pvrdma_dealloc_pd(struct ib_pd *pd) * * @return: the ib_ah pointer on success, otherwise errno. */ -struct ib_ah *pvrdma_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, +struct ib_ah *pvrdma_create_ah(struct ib_pd *pd, struct rdma_ah_attr *ah_attr, struct ib_udata *udata) { struct pvrdma_dev *dev = to_vdev(pd->device); struct pvrdma_ah *ah; - enum rdma_link_layer ll; + const struct ib_global_route *grh; + u8 port_num = rdma_ah_get_port_num(ah_attr); - if (!(ah_attr->ah_flags & IB_AH_GRH)) + if (!(rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH)) return ERR_PTR(-EINVAL); - ll = rdma_port_get_link_layer(pd->device, ah_attr->port_num); - - if (ll != IB_LINK_LAYER_ETHERNET || - rdma_is_multicast_addr((struct in6_addr *)ah_attr->grh.dgid.raw)) + grh = rdma_ah_read_grh(ah_attr); + if ((ah_attr->type != RDMA_AH_ATTR_TYPE_ROCE) || + rdma_is_multicast_addr((struct in6_addr *)grh->dgid.raw)) return ERR_PTR(-EINVAL); if (!atomic_add_unless(&dev->num_ahs, 1, dev->dsr->caps.max_ah)) @@ -545,15 +545,15 @@ struct ib_ah *pvrdma_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, return ERR_PTR(-ENOMEM); } - ah->av.port_pd = to_vpd(pd)->pd_handle | (ah_attr->port_num << 24); - ah->av.src_path_bits = ah_attr->src_path_bits; + ah->av.port_pd = to_vpd(pd)->pd_handle | (port_num << 24); + ah->av.src_path_bits = rdma_ah_get_path_bits(ah_attr); ah->av.src_path_bits |= 0x80; - ah->av.gid_index = ah_attr->grh.sgid_index; - ah->av.hop_limit = ah_attr->grh.hop_limit; - ah->av.sl_tclass_flowlabel = (ah_attr->grh.traffic_class << 20) | - ah_attr->grh.flow_label; - memcpy(ah->av.dgid, ah_attr->grh.dgid.raw, 16); - memcpy(ah->av.dmac, ah_attr->dmac, 6); + ah->av.gid_index = grh->sgid_index; + ah->av.hop_limit = grh->hop_limit; + ah->av.sl_tclass_flowlabel = (grh->traffic_class << 20) | + grh->flow_label; + memcpy(ah->av.dgid, grh->dgid.raw, 16); + memcpy(ah->av.dmac, ah_attr->roce.dmac, ETH_ALEN); ah->ibah.device = pd->device; ah->ibah.pd = pd; diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.h b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.h index bfbe96b56255..002a9b066e70 100644 --- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.h +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.h @@ -417,7 +417,7 @@ int pvrdma_resize_cq(struct ib_cq *ibcq, int entries, int pvrdma_destroy_cq(struct ib_cq *cq); int pvrdma_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *wc); int pvrdma_req_notify_cq(struct ib_cq *cq, enum ib_cq_notify_flags flags); -struct ib_ah *pvrdma_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, +struct ib_ah *pvrdma_create_ah(struct ib_pd *pd, struct rdma_ah_attr *ah_attr, struct ib_udata *udata); int pvrdma_destroy_ah(struct ib_ah *ah); struct ib_qp *pvrdma_create_qp(struct ib_pd *pd, |