diff options
Diffstat (limited to 'drivers/infiniband/core/verbs.c')
| -rw-r--r-- | drivers/infiniband/core/verbs.c | 198 |
1 files changed, 126 insertions, 72 deletions
diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c index 71580cc28c9e..fb98ed67d5bc 100644 --- a/drivers/infiniband/core/verbs.c +++ b/drivers/infiniband/core/verbs.c @@ -44,6 +44,7 @@ #include <linux/in.h> #include <linux/in6.h> #include <net/addrconf.h> +#include <linux/security.h> #include <rdma/ib_verbs.h> #include <rdma/ib_cache.h> @@ -311,7 +312,7 @@ EXPORT_SYMBOL(ib_dealloc_pd); /* Address handles */ -struct ib_ah *ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr) +struct ib_ah *rdma_create_ah(struct ib_pd *pd, struct rdma_ah_attr *ah_attr) { struct ib_ah *ah; @@ -321,12 +322,13 @@ struct ib_ah *ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr) ah->device = pd->device; ah->pd = pd; ah->uobject = NULL; + ah->type = ah_attr->type; atomic_inc(&pd->usecnt); } return ah; } -EXPORT_SYMBOL(ib_create_ah); +EXPORT_SYMBOL(rdma_create_ah); int ib_get_rdma_header_version(const union rdma_network_hdr *hdr) { @@ -450,9 +452,22 @@ int ib_get_gids_from_rdma_hdr(const union rdma_network_hdr *hdr, } EXPORT_SYMBOL(ib_get_gids_from_rdma_hdr); +/* + * This function creates ah from the incoming packet. + * Incoming packet has dgid of the receiver node on which this code is + * getting executed and, sgid contains the GID of the sender. + * + * When resolving mac address of destination, the arrived dgid is used + * as sgid and, sgid is used as dgid because sgid contains destinations + * GID whom to respond to. + * + * This is why when calling rdma_addr_find_l2_eth_by_grh() function, the + * position of arguments dgid and sgid do not match the order of the + * parameters. + */ int ib_init_ah_from_wc(struct ib_device *device, u8 port_num, const struct ib_wc *wc, const struct ib_grh *grh, - struct ib_ah_attr *ah_attr) + struct rdma_ah_attr *ah_attr) { u32 flow_class; u16 gid_index; @@ -464,6 +479,7 @@ int ib_init_ah_from_wc(struct ib_device *device, u8 port_num, union ib_gid sgid; memset(ah_attr, 0, sizeof *ah_attr); + ah_attr->type = rdma_ah_find_type(device, port_num); if (rdma_cap_eth_ah(device, port_num)) { if (wc->wc_flags & IB_WC_WITH_NETWORK_HDR_TYPE) net_type = wc->network_hdr_type; @@ -494,7 +510,7 @@ int ib_init_ah_from_wc(struct ib_device *device, u8 port_num, return -ENODEV; ret = rdma_addr_find_l2_eth_by_grh(&dgid, &sgid, - ah_attr->dmac, + ah_attr->roce.dmac, wc->wc_flags & IB_WC_WITH_VLAN ? NULL : &vlan_id, &if_index, &hoplimit); @@ -504,11 +520,6 @@ int ib_init_ah_from_wc(struct ib_device *device, u8 port_num, } resolved_dev = dev_get_by_index(&init_net, if_index); - if (resolved_dev->flags & IFF_LOOPBACK) { - dev_put(resolved_dev); - resolved_dev = idev; - dev_hold(resolved_dev); - } rcu_read_lock(); if (resolved_dev != idev && !rdma_is_upper_dev_rcu(idev, resolved_dev)) @@ -525,15 +536,12 @@ int ib_init_ah_from_wc(struct ib_device *device, u8 port_num, return ret; } - ah_attr->dlid = wc->slid; - ah_attr->sl = wc->sl; - ah_attr->src_path_bits = wc->dlid_path_bits; - ah_attr->port_num = port_num; + rdma_ah_set_dlid(ah_attr, wc->slid); + rdma_ah_set_sl(ah_attr, wc->sl); + rdma_ah_set_path_bits(ah_attr, wc->dlid_path_bits); + rdma_ah_set_port_num(ah_attr, port_num); if (wc->wc_flags & IB_WC_GRH) { - ah_attr->ah_flags = IB_AH_GRH; - ah_attr->grh.dgid = sgid; - if (!rdma_cap_eth_ah(device, port_num)) { if (dgid.global.interface_id != cpu_to_be64(IB_SA_WELL_KNOWN_GUID)) { ret = ib_find_cached_gid_by_port(device, &dgid, @@ -547,11 +555,12 @@ int ib_init_ah_from_wc(struct ib_device *device, u8 port_num, } } - ah_attr->grh.sgid_index = (u8) gid_index; flow_class = be32_to_cpu(grh->version_tclass_flow); - ah_attr->grh.flow_label = flow_class & 0xFFFFF; - ah_attr->grh.hop_limit = hoplimit; - ah_attr->grh.traffic_class = (flow_class >> 20) & 0xFF; + rdma_ah_set_grh(ah_attr, &sgid, + flow_class & 0xFFFFF, + (u8)gid_index, hoplimit, + (flow_class >> 20) & 0xFF); + } return 0; } @@ -560,34 +569,37 @@ EXPORT_SYMBOL(ib_init_ah_from_wc); struct ib_ah *ib_create_ah_from_wc(struct ib_pd *pd, const struct ib_wc *wc, const struct ib_grh *grh, u8 port_num) { - struct ib_ah_attr ah_attr; + struct rdma_ah_attr ah_attr; int ret; ret = ib_init_ah_from_wc(pd->device, port_num, wc, grh, &ah_attr); if (ret) return ERR_PTR(ret); - return ib_create_ah(pd, &ah_attr); + return rdma_create_ah(pd, &ah_attr); } EXPORT_SYMBOL(ib_create_ah_from_wc); -int ib_modify_ah(struct ib_ah *ah, struct ib_ah_attr *ah_attr) +int rdma_modify_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr) { + if (ah->type != ah_attr->type) + return -EINVAL; + return ah->device->modify_ah ? ah->device->modify_ah(ah, ah_attr) : -ENOSYS; } -EXPORT_SYMBOL(ib_modify_ah); +EXPORT_SYMBOL(rdma_modify_ah); -int ib_query_ah(struct ib_ah *ah, struct ib_ah_attr *ah_attr) +int rdma_query_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr) { return ah->device->query_ah ? ah->device->query_ah(ah, ah_attr) : -ENOSYS; } -EXPORT_SYMBOL(ib_query_ah); +EXPORT_SYMBOL(rdma_query_ah); -int ib_destroy_ah(struct ib_ah *ah) +int rdma_destroy_ah(struct ib_ah *ah) { struct ib_pd *pd; int ret; @@ -599,7 +611,7 @@ int ib_destroy_ah(struct ib_ah *ah) return ret; } -EXPORT_SYMBOL(ib_destroy_ah); +EXPORT_SYMBOL(rdma_destroy_ah); /* Shared receive queues */ @@ -710,12 +722,20 @@ static struct ib_qp *__ib_open_qp(struct ib_qp *real_qp, { struct ib_qp *qp; unsigned long flags; + int err; qp = kzalloc(sizeof *qp, GFP_KERNEL); if (!qp) return ERR_PTR(-ENOMEM); qp->real_qp = real_qp; + err = ib_open_shared_qp_security(qp, real_qp->device); + if (err) { + kfree(qp); + return ERR_PTR(err); + } + + qp->real_qp = real_qp; atomic_inc(&real_qp->usecnt); qp->device = real_qp->device; qp->event_handler = event_handler; @@ -801,6 +821,12 @@ struct ib_qp *ib_create_qp(struct ib_pd *pd, if (IS_ERR(qp)) return qp; + ret = ib_create_qp_security(qp, device); + if (ret) { + ib_destroy_qp(qp); + return ERR_PTR(ret); + } + qp->device = device; qp->real_qp = qp; qp->uobject = NULL; @@ -869,6 +895,7 @@ static const struct { } qp_state_table[IB_QPS_ERR + 1][IB_QPS_ERR + 1] = { [IB_QPS_RESET] = { [IB_QPS_RESET] = { .valid = 1 }, + [IB_QPS_ERR] = { .valid = 1 }, [IB_QPS_INIT] = { .valid = 1, .req_param = { @@ -1201,20 +1228,22 @@ int ib_modify_qp_is_ok(enum ib_qp_state cur_state, enum ib_qp_state next_state, EXPORT_SYMBOL(ib_modify_qp_is_ok); int ib_resolve_eth_dmac(struct ib_device *device, - struct ib_ah_attr *ah_attr) + struct rdma_ah_attr *ah_attr) { int ret = 0; + struct ib_global_route *grh; - if (ah_attr->port_num < rdma_start_port(device) || - ah_attr->port_num > rdma_end_port(device)) + if (!rdma_is_port_valid(device, rdma_ah_get_port_num(ah_attr))) return -EINVAL; - if (!rdma_cap_eth_ah(device, ah_attr->port_num)) + if (ah_attr->type != RDMA_AH_ATTR_TYPE_ROCE) return 0; - if (rdma_link_local_addr((struct in6_addr *)ah_attr->grh.dgid.raw)) { - rdma_get_ll_mac((struct in6_addr *)ah_attr->grh.dgid.raw, - ah_attr->dmac); + grh = rdma_ah_retrieve_grh(ah_attr); + + if (rdma_link_local_addr((struct in6_addr *)grh->dgid.raw)) { + rdma_get_ll_mac((struct in6_addr *)grh->dgid.raw, + ah_attr->roce.dmac); } else { union ib_gid sgid; struct ib_gid_attr sgid_attr; @@ -1222,8 +1251,8 @@ int ib_resolve_eth_dmac(struct ib_device *device, int hop_limit; ret = ib_query_gid(device, - ah_attr->port_num, - ah_attr->grh.sgid_index, + rdma_ah_get_port_num(ah_attr), + grh->sgid_index, &sgid, &sgid_attr); if (ret || !sgid_attr.ndev) { @@ -1234,34 +1263,50 @@ int ib_resolve_eth_dmac(struct ib_device *device, ifindex = sgid_attr.ndev->ifindex; - ret = rdma_addr_find_l2_eth_by_grh(&sgid, - &ah_attr->grh.dgid, - ah_attr->dmac, - NULL, &ifindex, &hop_limit); + ret = + rdma_addr_find_l2_eth_by_grh(&sgid, &grh->dgid, + ah_attr->roce.dmac, + NULL, &ifindex, &hop_limit); dev_put(sgid_attr.ndev); - ah_attr->grh.hop_limit = hop_limit; + grh->hop_limit = hop_limit; } out: return ret; } EXPORT_SYMBOL(ib_resolve_eth_dmac); -int ib_modify_qp(struct ib_qp *qp, - struct ib_qp_attr *qp_attr, - int qp_attr_mask) +/** + * ib_modify_qp_with_udata - Modifies the attributes for the specified QP. + * @qp: The QP to modify. + * @attr: On input, specifies the QP attributes to modify. On output, + * the current values of selected QP attributes are returned. + * @attr_mask: A bit-mask used to specify which attributes of the QP + * are being modified. + * @udata: pointer to user's input output buffer information + * are being modified. + * It returns 0 on success and returns appropriate error code on error. + */ +int ib_modify_qp_with_udata(struct ib_qp *qp, struct ib_qp_attr *attr, + int attr_mask, struct ib_udata *udata) { + int ret; - if (qp_attr_mask & IB_QP_AV) { - int ret; - - ret = ib_resolve_eth_dmac(qp->device, &qp_attr->ah_attr); + if (attr_mask & IB_QP_AV) { + ret = ib_resolve_eth_dmac(qp->device, &attr->ah_attr); if (ret) return ret; } + return ib_security_modify_qp(qp, attr, attr_mask, udata); +} +EXPORT_SYMBOL(ib_modify_qp_with_udata); - return qp->device->modify_qp(qp->real_qp, qp_attr, qp_attr_mask, NULL); +int ib_modify_qp(struct ib_qp *qp, + struct ib_qp_attr *qp_attr, + int qp_attr_mask) +{ + return ib_modify_qp_with_udata(qp, qp_attr, qp_attr_mask, NULL); } EXPORT_SYMBOL(ib_modify_qp); @@ -1290,6 +1335,7 @@ int ib_close_qp(struct ib_qp *qp) spin_unlock_irqrestore(&real_qp->device->event_handler_lock, flags); atomic_dec(&real_qp->usecnt); + ib_close_shared_qp_security(qp->qp_sec); kfree(qp); return 0; @@ -1330,6 +1376,7 @@ int ib_destroy_qp(struct ib_qp *qp) struct ib_cq *scq, *rcq; struct ib_srq *srq; struct ib_rwq_ind_table *ind_tbl; + struct ib_qp_security *sec; int ret; WARN_ON_ONCE(qp->mrs_used > 0); @@ -1345,6 +1392,9 @@ int ib_destroy_qp(struct ib_qp *qp) rcq = qp->recv_cq; srq = qp->srq; ind_tbl = qp->rwq_ind_tbl; + sec = qp->qp_sec; + if (sec) + ib_destroy_qp_security_begin(sec); if (!qp->uobject) rdma_rw_cleanup_mrs(qp); @@ -1361,6 +1411,11 @@ int ib_destroy_qp(struct ib_qp *qp) atomic_dec(&srq->usecnt); if (ind_tbl) atomic_dec(&ind_tbl->usecnt); + if (sec) + ib_destroy_qp_security_end(sec); + } else { + if (sec) + ib_destroy_qp_security_abort(sec); } return ret; @@ -1520,7 +1575,9 @@ int ib_attach_mcast(struct ib_qp *qp, union ib_gid *gid, u16 lid) if (!qp->device->attach_mcast) return -ENOSYS; - if (gid->raw[0] != 0xff || qp->qp_type != IB_QPT_UD) + if (gid->raw[0] != 0xff || qp->qp_type != IB_QPT_UD || + lid < be16_to_cpu(IB_MULTICAST_LID_BASE) || + lid == be16_to_cpu(IB_LID_PERMISSIVE)) return -EINVAL; ret = qp->device->attach_mcast(qp, gid, lid); @@ -1536,7 +1593,9 @@ int ib_detach_mcast(struct ib_qp *qp, union ib_gid *gid, u16 lid) if (!qp->device->detach_mcast) return -ENOSYS; - if (gid->raw[0] != 0xff || qp->qp_type != IB_QPT_UD) + if (gid->raw[0] != 0xff || qp->qp_type != IB_QPT_UD || + lid < be16_to_cpu(IB_MULTICAST_LID_BASE) || + lid == be16_to_cpu(IB_LID_PERMISSIVE)) return -EINVAL; ret = qp->device->detach_mcast(qp, gid, lid); @@ -1949,17 +2008,12 @@ static void ib_drain_qp_done(struct ib_cq *cq, struct ib_wc *wc) */ static void __ib_drain_sq(struct ib_qp *qp) { + struct ib_cq *cq = qp->send_cq; struct ib_qp_attr attr = { .qp_state = IB_QPS_ERR }; struct ib_drain_cqe sdrain; struct ib_send_wr swr = {}, *bad_swr; int ret; - if (qp->send_cq->poll_ctx == IB_POLL_DIRECT) { - WARN_ONCE(qp->send_cq->poll_ctx == IB_POLL_DIRECT, - "IB_POLL_DIRECT poll_ctx not supported for drain\n"); - return; - } - swr.wr_cqe = &sdrain.cqe; sdrain.cqe.done = ib_drain_qp_done; init_completion(&sdrain.done); @@ -1976,7 +2030,11 @@ static void __ib_drain_sq(struct ib_qp *qp) return; } - wait_for_completion(&sdrain.done); + if (cq->poll_ctx == IB_POLL_DIRECT) + while (wait_for_completion_timeout(&sdrain.done, HZ / 10) <= 0) + ib_process_cq_direct(cq, -1); + else + wait_for_completion(&sdrain.done); } /* @@ -1984,17 +2042,12 @@ static void __ib_drain_sq(struct ib_qp *qp) */ static void __ib_drain_rq(struct ib_qp *qp) { + struct ib_cq *cq = qp->recv_cq; struct ib_qp_attr attr = { .qp_state = IB_QPS_ERR }; struct ib_drain_cqe rdrain; struct ib_recv_wr rwr = {}, *bad_rwr; int ret; - if (qp->recv_cq->poll_ctx == IB_POLL_DIRECT) { - WARN_ONCE(qp->recv_cq->poll_ctx == IB_POLL_DIRECT, - "IB_POLL_DIRECT poll_ctx not supported for drain\n"); - return; - } - rwr.wr_cqe = &rdrain.cqe; rdrain.cqe.done = ib_drain_qp_done; init_completion(&rdrain.done); @@ -2011,7 +2064,11 @@ static void __ib_drain_rq(struct ib_qp *qp) return; } - wait_for_completion(&rdrain.done); + if (cq->poll_ctx == IB_POLL_DIRECT) + while (wait_for_completion_timeout(&rdrain.done, HZ / 10) <= 0) + ib_process_cq_direct(cq, -1); + else + wait_for_completion(&rdrain.done); } /** @@ -2028,8 +2085,7 @@ static void __ib_drain_rq(struct ib_qp *qp) * ensure there is room in the CQ and SQ for the drain work request and * completion. * - * allocate the CQ using ib_alloc_cq() and the CQ poll context cannot be - * IB_POLL_DIRECT. + * allocate the CQ using ib_alloc_cq(). * * ensure that there are no other contexts that are posting WRs concurrently. * Otherwise the drain is not guaranteed. @@ -2057,8 +2113,7 @@ EXPORT_SYMBOL(ib_drain_sq); * ensure there is room in the CQ and RQ for the drain work request and * completion. * - * allocate the CQ using ib_alloc_cq() and the CQ poll context cannot be - * IB_POLL_DIRECT. + * allocate the CQ using ib_alloc_cq(). * * ensure that there are no other contexts that are posting WRs concurrently. * Otherwise the drain is not guaranteed. @@ -2082,8 +2137,7 @@ EXPORT_SYMBOL(ib_drain_rq); * ensure there is room in the CQ(s), SQ, and RQ for drain work requests * and completions. * - * allocate the CQs using ib_alloc_cq() and the CQ poll context cannot be - * IB_POLL_DIRECT. + * allocate the CQs using ib_alloc_cq(). * * ensure that there are no other contexts that are posting WRs concurrently. * Otherwise the drain is not guaranteed. |
