[for-next,07/13] RDMA/rxe: Extend srq verbs to support xrcd

Message ID	20220917031104.21222-8-rpearsonhpe@gmail.com (mailing list archive)
State	Superseded
Headers	show Return-Path: <linux-rdma-owner@kernel.org> From: Bob Pearson <rpearsonhpe@gmail.com> To: jgg@nvidia.com, zyjzyj2000@gmail.com, lizhijian@fujitsu.com, linux-rdma@vger.kernel.org Cc: Bob Pearson <rpearsonhpe@gmail.com> Subject: [PATCH for-next 07/13] RDMA/rxe: Extend srq verbs to support xrcd Date: Fri, 16 Sep 2022 22:10:58 -0500 Message-Id: <20220917031104.21222-8-rpearsonhpe@gmail.com> In-Reply-To: <20220917031104.21222-1-rpearsonhpe@gmail.com> References: <20220917031104.21222-1-rpearsonhpe@gmail.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Precedence: bulk
Series	Implement the xrc transport \| expand [for-next,00/13] Implement the xrc transport [for-next,01/13] RDMA/rxe: Replace START->FIRST, END->LAST [for-next,02/13] RDMA/rxe: Move next_opcode() to rxe_opcode.c [for-next,03/13] RDMA: Add xrc opcodes to ib_pack.h [for-next,04/13] RDMA/rxe: Extend opcodes and headers to support xrc [for-next,05/13] RDMA/rxe: Add xrc opcodes to next_opcode() [for-next,06/13] RDMA/rxe: Implement open_xrcd and close_xrcd [for-next,07/13] RDMA/rxe: Extend srq verbs to support xrcd [for-next,08/13] RDMA/rxe: Extend rxe_qp.c to support xrc qps [for-next,09/13] RDMA/rxe: Extend rxe_recv.c to support xrc [for-next,10/13] RDMA/rxe: Extend rxe_comp.c to support xrc qps [for-next,11/13] RDMA/rxe: Extend rxe_req.c to support xrc qps [for-next,12/13] RDMA/rxe: Extend rxe_net.c to support xrc qps [for-next,13/13] RDMA/rxe: Extend rxe_resp.c to support xrc qps

diff --git a/drivers/infiniband/sw/rxe/rxe_srq.c b/drivers/infiniband/sw/rxe/rxe_srq.c index 02b39498c370..fcd1a58c3900 100644 --- a/drivers/infiniband/sw/rxe/rxe_srq.c +++ b/drivers/infiniband/sw/rxe/rxe_srq.c @@ -11,61 +11,85 @@ int rxe_srq_chk_init(struct rxe_dev *rxe, struct ib_srq_init_attr *init) { struct ib_srq_attr *attr = &init->attr; + int err = -EINVAL; - if (attr->max_wr > rxe->attr.max_srq_wr) { - pr_warn("max_wr(%d) > max_srq_wr(%d)\n", - attr->max_wr, rxe->attr.max_srq_wr); - goto err1; + if (init->srq_type == IB_SRQT_TM) { + err = -EOPNOTSUPP; + goto err_out; } - if (attr->max_wr <= 0) { - pr_warn("max_wr(%d) <= 0\n", attr->max_wr); - goto err1; + if (init->srq_type == IB_SRQT_XRC) { + if (!init->ext.cq || !init->ext.xrc.xrcd) + goto err_out; } + if (attr->max_wr > rxe->attr.max_srq_wr) + goto err_out; + + if (attr->max_wr <= 0) + goto err_out; + if (attr->max_wr < RXE_MIN_SRQ_WR) attr->max_wr = RXE_MIN_SRQ_WR; - if (attr->max_sge > rxe->attr.max_srq_sge) { - pr_warn("max_sge(%d) > max_srq_sge(%d)\n", - attr->max_sge, rxe->attr.max_srq_sge); - goto err1; - } + if (attr->max_sge > rxe->attr.max_srq_sge) + goto err_out; if (attr->max_sge < RXE_MIN_SRQ_SGE) attr->max_sge = RXE_MIN_SRQ_SGE; return 0; -err1: - return -EINVAL; +err_out: + pr_debug("%s: failed err = %d\n", __func__, err); + return err; } int rxe_srq_from_init(struct rxe_dev *rxe, struct rxe_srq *srq, struct ib_srq_init_attr *init, struct ib_udata *udata, struct rxe_create_srq_resp __user *uresp) { - int err; - int srq_wqe_size; + struct rxe_pd *pd = to_rpd(srq->ibsrq.pd); + struct rxe_cq *cq; + struct rxe_xrcd *xrcd; struct rxe_queue *q; - enum queue_type type; + int srq_wqe_size; + int err; + + rxe_get(pd); + srq->pd = pd; srq->ibsrq.event_handler = init->event_handler; srq->ibsrq.srq_context = init->srq_context; srq->limit = init->attr.srq_limit; - srq->srq_num = srq->elem.index; srq->rq.max_wr = init->attr.max_wr; srq->rq.max_sge = init->attr.max_sge; - srq_wqe_size = rcv_wqe_size(srq->rq.max_sge); + if (init->srq_type == IB_SRQT_XRC) { + cq = to_rcq(init->ext.cq); + if (cq) { + rxe_get(cq); + srq->cq = to_rcq(init->ext.cq); + } else { + return -EINVAL; + } + xrcd = to_rxrcd(init->ext.xrc.xrcd); + if (xrcd) { + rxe_get(xrcd); + srq->xrcd = to_rxrcd(init->ext.xrc.xrcd); + } + srq->ibsrq.ext.xrc.srq_num = srq->elem.index; + } spin_lock_init(&srq->rq.producer_lock); spin_lock_init(&srq->rq.consumer_lock); - type = QUEUE_TYPE_FROM_CLIENT; - q = rxe_queue_init(rxe, &srq->rq.max_wr, srq_wqe_size, type); + srq_wqe_size = rcv_wqe_size(srq->rq.max_sge); + q = rxe_queue_init(rxe, &srq->rq.max_wr, srq_wqe_size, + QUEUE_TYPE_FROM_CLIENT); if (!q) { - pr_warn("unable to allocate queue for srq\n"); + pr_debug("%s: srq#%d: unable to allocate queue\n", + __func__, srq->elem.index); return -ENOMEM; } @@ -79,66 +103,45 @@ int rxe_srq_from_init(struct rxe_dev *rxe, struct rxe_srq *srq, return err; } - if (uresp) { - if (copy_to_user(&uresp->srq_num, &srq->srq_num, - sizeof(uresp->srq_num))) { - rxe_queue_cleanup(q); - return -EFAULT; - } - } - return 0; } int rxe_srq_chk_attr(struct rxe_dev *rxe, struct rxe_srq *srq, struct ib_srq_attr *attr, enum ib_srq_attr_mask mask) { - if (srq->error) { - pr_warn("srq in error state\n"); - goto err1; - } + int err = -EINVAL; + + if (srq->error) + goto err_out; if (mask & IB_SRQ_MAX_WR) { - if (attr->max_wr > rxe->attr.max_srq_wr) { - pr_warn("max_wr(%d) > max_srq_wr(%d)\n", - attr->max_wr, rxe->attr.max_srq_wr); - goto err1; - } + if (attr->max_wr > rxe->attr.max_srq_wr) + goto err_out; - if (attr->max_wr <= 0) { - pr_warn("max_wr(%d) <= 0\n", attr->max_wr); - goto err1; - } + if (attr->max_wr <= 0) + goto err_out; - if (srq->limit && (attr->max_wr < srq->limit)) { - pr_warn("max_wr (%d) < srq->limit (%d)\n", - attr->max_wr, srq->limit); - goto err1; - } + if (srq->limit && (attr->max_wr < srq->limit)) + goto err_out; if (attr->max_wr < RXE_MIN_SRQ_WR) attr->max_wr = RXE_MIN_SRQ_WR; } if (mask & IB_SRQ_LIMIT) { - if (attr->srq_limit > rxe->attr.max_srq_wr) { - pr_warn("srq_limit(%d) > max_srq_wr(%d)\n", - attr->srq_limit, rxe->attr.max_srq_wr); - goto err1; - } + if (attr->srq_limit > rxe->attr.max_srq_wr) + goto err_out; - if (attr->srq_limit > srq->rq.queue->buf->index_mask) { - pr_warn("srq_limit (%d) > cur limit(%d)\n", - attr->srq_limit, - srq->rq.queue->buf->index_mask); - goto err1; - } + if (attr->srq_limit > srq->rq.queue->buf->index_mask) + goto err_out; } return 0; -err1: - return -EINVAL; +err_out: + pr_debug("%s: srq#%d: failed err = %d\n", __func__, + srq->elem.index, err); + return err; } int rxe_srq_from_attr(struct rxe_dev *rxe, struct rxe_srq *srq, @@ -182,6 +185,12 @@ void rxe_srq_cleanup(struct rxe_pool_elem *elem) if (srq->pd) rxe_put(srq->pd); + if (srq->cq) + rxe_put(srq->cq); + + if (srq->xrcd) + rxe_put(srq->xrcd); + if (srq->rq.queue) rxe_queue_cleanup(srq->rq.queue); } diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.c b/drivers/infiniband/sw/rxe/rxe_verbs.c index 4a5da079bf11..ef86f0c5890e 100644 --- a/drivers/infiniband/sw/rxe/rxe_verbs.c +++ b/drivers/infiniband/sw/rxe/rxe_verbs.c @@ -306,7 +306,6 @@ static int rxe_create_srq(struct ib_srq *ibsrq, struct ib_srq_init_attr *init, { int err; struct rxe_dev *rxe = to_rdev(ibsrq->device); - struct rxe_pd *pd = to_rpd(ibsrq->pd); struct rxe_srq *srq = to_rsrq(ibsrq); struct rxe_create_srq_resp __user *uresp = NULL; @@ -316,9 +315,6 @@ static int rxe_create_srq(struct ib_srq *ibsrq, struct ib_srq_init_attr *init, uresp = udata->outbuf; } - if (init->srq_type != IB_SRQT_BASIC) - return -EOPNOTSUPP; - err = rxe_srq_chk_init(rxe, init); if (err) return err; @@ -327,13 +323,11 @@ static int rxe_create_srq(struct ib_srq *ibsrq, struct ib_srq_init_attr *init, if (err) return err; - rxe_get(pd); - srq->pd = pd; - err = rxe_srq_from_init(rxe, srq, init, udata, uresp); if (err) goto err_cleanup; + rxe_finalize(srq); return 0; err_cleanup: @@ -367,6 +361,7 @@ static int rxe_modify_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr, err = rxe_srq_from_attr(rxe, srq, attr, mask, &ucmd, udata); if (err) return err; + return 0; } @@ -380,6 +375,7 @@ static int rxe_query_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr) attr->max_wr = srq->rq.queue->buf->index_mask; attr->max_sge = srq->rq.max_sge; attr->srq_limit = srq->limit; + return 0; } @@ -546,7 +542,6 @@ static void init_send_wr(struct rxe_qp *qp, struct rxe_send_wr *wr, const struct ib_send_wr *ibwr) { wr->wr_id = ibwr->wr_id; - wr->num_sge = ibwr->num_sge; wr->opcode = ibwr->opcode; wr->send_flags = ibwr->send_flags; @@ -628,6 +623,8 @@ static void init_send_wqe(struct rxe_qp *qp, const struct ib_send_wr *ibwr, return; } + wqe->dma.num_sge = ibwr->num_sge; + if (unlikely(ibwr->send_flags & IB_SEND_INLINE)) copy_inline_data_to_wqe(wqe, ibwr); else diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.h b/drivers/infiniband/sw/rxe/rxe_verbs.h index 6c4cfb802dd4..7dab7fa3ba6c 100644 --- a/drivers/infiniband/sw/rxe/rxe_verbs.h +++ b/drivers/infiniband/sw/rxe/rxe_verbs.h @@ -102,13 +102,19 @@ struct rxe_srq { struct ib_srq ibsrq; struct rxe_pool_elem elem; struct rxe_pd *pd; + struct rxe_xrcd *xrcd; /* xrc only */ + struct rxe_cq *cq; /* xrc only */ struct rxe_rq rq; - u32 srq_num; int limit; int error; }; +static inline u32 srq_num(struct rxe_srq *srq) +{ + return srq->ibsrq.ext.xrc.srq_num; +} + enum rxe_qp_state { QP_STATE_RESET, QP_STATE_INIT, diff --git a/include/uapi/rdma/rdma_user_rxe.h b/include/uapi/rdma/rdma_user_rxe.h index f09c5c9e3dd5..514a1b6976fe 100644 --- a/include/uapi/rdma/rdma_user_rxe.h +++ b/include/uapi/rdma/rdma_user_rxe.h @@ -74,7 +74,7 @@ struct rxe_av { struct rxe_send_wr { __aligned_u64 wr_id; - __u32 num_sge; + __u32 srq_num; /* xrc only */ __u32 opcode; __u32 send_flags; union { @@ -191,8 +191,6 @@ struct rxe_create_qp_resp { struct rxe_create_srq_resp { struct mminfo mi; - __u32 srq_num; - __u32 reserved; }; struct rxe_modify_srq_cmd {

[for-next,07/13] RDMA/rxe: Extend srq verbs to support xrcd

Commit Message

Comments

Patch