From patchwork Wed Feb 28 23:40:02 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Logan Gunthorpe X-Patchwork-Id: 10250157 X-Patchwork-Delegate: bhelgaas@google.com Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id AA4FE60211 for ; Wed, 28 Feb 2018 23:42:08 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 9C97E28E57 for ; Wed, 28 Feb 2018 23:42:08 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 8BD0E28E8D; Wed, 28 Feb 2018 23:42:08 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-6.9 required=2.0 tests=BAYES_00,RCVD_IN_DNSWL_HI autolearn=unavailable version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id BB70F28E57 for ; Wed, 28 Feb 2018 23:42:07 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965222AbeB1Xl0 (ORCPT ); Wed, 28 Feb 2018 18:41:26 -0500 Received: from ale.deltatee.com ([207.54.116.67]:60180 "EHLO ale.deltatee.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S965039AbeB1XkZ (ORCPT ); Wed, 28 Feb 2018 18:40:25 -0500 Received: from cgy1-donard.priv.deltatee.com ([172.16.1.31]) by ale.deltatee.com with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.89) (envelope-from ) id 1erBKc-0003TV-OJ; Wed, 28 Feb 2018 16:40:20 -0700 Received: from gunthorp by cgy1-donard.priv.deltatee.com with local (Exim 4.89) (envelope-from ) id 1erBKZ-0005VL-RN; Wed, 28 Feb 2018 16:40:15 -0700 From: Logan Gunthorpe To: linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org, linux-nvme@lists.infradead.org, linux-rdma@vger.kernel.org, linux-nvdimm@lists.01.org, linux-block@vger.kernel.org Cc: Stephen Bates , Christoph Hellwig , Jens Axboe , Keith Busch , Sagi Grimberg , Bjorn Helgaas , Jason Gunthorpe , Max Gurtovoy , Dan Williams , =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= , Benjamin Herrenschmidt , Alex Williamson , Logan Gunthorpe Date: Wed, 28 Feb 2018 16:40:02 -0700 Message-Id: <20180228234006.21093-7-logang@deltatee.com> X-Mailer: git-send-email 2.11.0 In-Reply-To: <20180228234006.21093-1-logang@deltatee.com> References: <20180228234006.21093-1-logang@deltatee.com> X-SA-Exim-Connect-IP: 172.16.1.31 X-SA-Exim-Rcpt-To: linux-nvme@lists.infradead.org, linux-nvdimm@lists.01.org, linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org, linux-rdma@vger.kernel.org, linux-block@vger.kernel.org, sbates@raithlin.com, hch@lst.de, axboe@kernel.dk, sagi@grimberg.me, bhelgaas@google.com, jgg@mellanox.com, maxg@mellanox.com, keith.busch@intel.com, dan.j.williams@intel.com, benh@kernel.crashing.org, jglisse@redhat.com, alex.williamson@redhat.com, logang@deltatee.com X-SA-Exim-Mail-From: gunthorp@deltatee.com Subject: [PATCH v2 06/10] IB/core: Add optional PCI P2P flag to rdma_rw_ctx_[init|destroy]() X-SA-Exim-Version: 4.2.1 (built Tue, 02 Aug 2016 21:08:31 +0000) X-SA-Exim-Scanned: Yes (on ale.deltatee.com) Sender: linux-pci-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-pci@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP In order to use PCI P2P memory pci_p2pmem_[un]map_sg() functions must be called to map the correct DMA address. To do this, we add a flags variable and the RDMA_RW_CTX_FLAG_PCI_P2P flag. When the flag is specified use the appropriate map function. Signed-off-by: Logan Gunthorpe --- drivers/infiniband/core/rw.c | 21 +++++++++++++++++---- drivers/infiniband/ulp/isert/ib_isert.c | 5 +++-- drivers/infiniband/ulp/srpt/ib_srpt.c | 7 ++++--- drivers/nvme/target/rdma.c | 6 +++--- include/rdma/rw.h | 7 +++++-- net/sunrpc/xprtrdma/svc_rdma_rw.c | 6 +++--- 6 files changed, 35 insertions(+), 17 deletions(-) diff --git a/drivers/infiniband/core/rw.c b/drivers/infiniband/core/rw.c index c8963e91f92a..775a9f8b15a6 100644 --- a/drivers/infiniband/core/rw.c +++ b/drivers/infiniband/core/rw.c @@ -12,6 +12,7 @@ */ #include #include +#include #include #include @@ -269,18 +270,24 @@ static int rdma_rw_init_single_wr(struct rdma_rw_ctx *ctx, struct ib_qp *qp, * @remote_addr:remote address to read/write (relative to @rkey) * @rkey: remote key to operate on * @dir: %DMA_TO_DEVICE for RDMA WRITE, %DMA_FROM_DEVICE for RDMA READ + * @flags: any of the RDMA_RW_CTX_FLAG_* flags * * Returns the number of WQEs that will be needed on the workqueue if * successful, or a negative error code. */ int rdma_rw_ctx_init(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u8 port_num, struct scatterlist *sg, u32 sg_cnt, u32 sg_offset, - u64 remote_addr, u32 rkey, enum dma_data_direction dir) + u64 remote_addr, u32 rkey, enum dma_data_direction dir, + unsigned int flags) { struct ib_device *dev = qp->pd->device; int ret; - ret = ib_dma_map_sg(dev, sg, sg_cnt, dir); + if (flags & RDMA_RW_CTX_FLAG_PCI_P2PDMA) + ret = pci_p2pdma_map_sg(dev->dma_device, sg, sg_cnt, dir); + else + ret = ib_dma_map_sg(dev, sg, sg_cnt, dir); + if (!ret) return -ENOMEM; sg_cnt = ret; @@ -579,9 +586,11 @@ EXPORT_SYMBOL(rdma_rw_ctx_post); * @sg: scatterlist that was used for the READ/WRITE * @sg_cnt: number of entries in @sg * @dir: %DMA_TO_DEVICE for RDMA WRITE, %DMA_FROM_DEVICE for RDMA READ + * @flags: the same flags used to init the context */ void rdma_rw_ctx_destroy(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u8 port_num, - struct scatterlist *sg, u32 sg_cnt, enum dma_data_direction dir) + struct scatterlist *sg, u32 sg_cnt, enum dma_data_direction dir, + unsigned int flags) { int i; @@ -602,7 +611,11 @@ void rdma_rw_ctx_destroy(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u8 port_num, break; } - ib_dma_unmap_sg(qp->pd->device, sg, sg_cnt, dir); + if (flags & RDMA_RW_CTX_FLAG_PCI_P2PDMA) + pci_p2pdma_unmap_sg(qp->pd->device->dma_device, sg, + sg_cnt, dir); + else + ib_dma_unmap_sg(qp->pd->device, sg, sg_cnt, dir); } EXPORT_SYMBOL(rdma_rw_ctx_destroy); diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index fff40b097947..933200645262 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -1497,7 +1497,8 @@ isert_rdma_rw_ctx_destroy(struct isert_cmd *cmd, struct isert_conn *conn) se_cmd->t_prot_nents, dir); } else { rdma_rw_ctx_destroy(&cmd->rw, conn->qp, conn->cm_id->port_num, - se_cmd->t_data_sg, se_cmd->t_data_nents, dir); + se_cmd->t_data_sg, se_cmd->t_data_nents, + dir, 0); } cmd->rw.nr_ops = 0; @@ -2152,7 +2153,7 @@ isert_rdma_rw_ctx_post(struct isert_cmd *cmd, struct isert_conn *conn, } else { ret = rdma_rw_ctx_init(&cmd->rw, conn->qp, port_num, se_cmd->t_data_sg, se_cmd->t_data_nents, - offset, addr, rkey, dir); + offset, addr, rkey, dir, 0); } if (ret < 0) { diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.c b/drivers/infiniband/ulp/srpt/ib_srpt.c index 0373b7c40902..5dcbe0a18db8 100644 --- a/drivers/infiniband/ulp/srpt/ib_srpt.c +++ b/drivers/infiniband/ulp/srpt/ib_srpt.c @@ -899,7 +899,8 @@ static int srpt_alloc_rw_ctxs(struct srpt_send_ioctx *ioctx, goto unwind; ret = rdma_rw_ctx_init(&ctx->rw, ch->qp, ch->sport->port, - ctx->sg, ctx->nents, 0, remote_addr, rkey, dir); + ctx->sg, ctx->nents, 0, remote_addr, rkey, + dir, 0); if (ret < 0) { target_free_sgl(ctx->sg, ctx->nents); goto unwind; @@ -928,7 +929,7 @@ static int srpt_alloc_rw_ctxs(struct srpt_send_ioctx *ioctx, struct srpt_rw_ctx *ctx = &ioctx->rw_ctxs[i]; rdma_rw_ctx_destroy(&ctx->rw, ch->qp, ch->sport->port, - ctx->sg, ctx->nents, dir); + ctx->sg, ctx->nents, dir, 0); target_free_sgl(ctx->sg, ctx->nents); } if (ioctx->rw_ctxs != &ioctx->s_rw_ctx) @@ -946,7 +947,7 @@ static void srpt_free_rw_ctxs(struct srpt_rdma_ch *ch, struct srpt_rw_ctx *ctx = &ioctx->rw_ctxs[i]; rdma_rw_ctx_destroy(&ctx->rw, ch->qp, ch->sport->port, - ctx->sg, ctx->nents, dir); + ctx->sg, ctx->nents, dir, 0); target_free_sgl(ctx->sg, ctx->nents); } diff --git a/drivers/nvme/target/rdma.c b/drivers/nvme/target/rdma.c index 978e169c11bf..020354e11351 100644 --- a/drivers/nvme/target/rdma.c +++ b/drivers/nvme/target/rdma.c @@ -427,7 +427,7 @@ static void nvmet_rdma_release_rsp(struct nvmet_rdma_rsp *rsp) if (rsp->n_rdma) { rdma_rw_ctx_destroy(&rsp->rw, queue->cm_id->qp, queue->cm_id->port_num, rsp->req.sg, - rsp->req.sg_cnt, nvmet_data_dir(&rsp->req)); + rsp->req.sg_cnt, nvmet_data_dir(&rsp->req), 0); } if (rsp->req.sg != &rsp->cmd->inline_sg) @@ -510,7 +510,7 @@ static void nvmet_rdma_read_data_done(struct ib_cq *cq, struct ib_wc *wc) atomic_add(rsp->n_rdma, &queue->sq_wr_avail); rdma_rw_ctx_destroy(&rsp->rw, queue->cm_id->qp, queue->cm_id->port_num, rsp->req.sg, - rsp->req.sg_cnt, nvmet_data_dir(&rsp->req)); + rsp->req.sg_cnt, nvmet_data_dir(&rsp->req), 0); rsp->n_rdma = 0; if (unlikely(wc->status != IB_WC_SUCCESS)) { @@ -579,7 +579,7 @@ static u16 nvmet_rdma_map_sgl_keyed(struct nvmet_rdma_rsp *rsp, ret = rdma_rw_ctx_init(&rsp->rw, cm_id->qp, cm_id->port_num, rsp->req.sg, rsp->req.sg_cnt, 0, addr, key, - nvmet_data_dir(&rsp->req)); + nvmet_data_dir(&rsp->req), 0); if (ret < 0) return NVME_SC_INTERNAL; rsp->req.transfer_len += len; diff --git a/include/rdma/rw.h b/include/rdma/rw.h index a3cbbc7b6417..dbc342b53901 100644 --- a/include/rdma/rw.h +++ b/include/rdma/rw.h @@ -59,12 +59,15 @@ struct rdma_rw_ctx { }; }; +#define RDMA_RW_CTX_FLAG_PCI_P2PDMA (1 << 0) + int rdma_rw_ctx_init(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u8 port_num, struct scatterlist *sg, u32 sg_cnt, u32 sg_offset, - u64 remote_addr, u32 rkey, enum dma_data_direction dir); + u64 remote_addr, u32 rkey, enum dma_data_direction dir, + unsigned int flags); void rdma_rw_ctx_destroy(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u8 port_num, struct scatterlist *sg, u32 sg_cnt, - enum dma_data_direction dir); + enum dma_data_direction dir, unsigned int flags); int rdma_rw_ctx_signature_init(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u8 port_num, struct scatterlist *sg, u32 sg_cnt, diff --git a/net/sunrpc/xprtrdma/svc_rdma_rw.c b/net/sunrpc/xprtrdma/svc_rdma_rw.c index 12b9a7e0b6d2..4364a8e98470 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_rw.c +++ b/net/sunrpc/xprtrdma/svc_rdma_rw.c @@ -140,7 +140,7 @@ static void svc_rdma_cc_release(struct svc_rdma_chunk_ctxt *cc, rdma_rw_ctx_destroy(&ctxt->rw_ctx, rdma->sc_qp, rdma->sc_port_num, ctxt->rw_sg_table.sgl, - ctxt->rw_nents, dir); + ctxt->rw_nents, dir, 0); svc_rdma_put_rw_ctxt(rdma, ctxt); } svc_xprt_put(&rdma->sc_xprt); @@ -433,7 +433,7 @@ svc_rdma_build_writes(struct svc_rdma_write_info *info, ret = rdma_rw_ctx_init(&ctxt->rw_ctx, rdma->sc_qp, rdma->sc_port_num, ctxt->rw_sg_table.sgl, ctxt->rw_nents, 0, seg_offset, - seg_handle, DMA_TO_DEVICE); + seg_handle, DMA_TO_DEVICE, 0); if (ret < 0) goto out_initerr; @@ -639,7 +639,7 @@ static int svc_rdma_build_read_segment(struct svc_rdma_read_info *info, ret = rdma_rw_ctx_init(&ctxt->rw_ctx, cc->cc_rdma->sc_qp, cc->cc_rdma->sc_port_num, ctxt->rw_sg_table.sgl, ctxt->rw_nents, - 0, offset, rkey, DMA_FROM_DEVICE); + 0, offset, rkey, DMA_FROM_DEVICE, 0); if (ret < 0) goto out_initerr;