From patchwork Mon May 11 18:03:04 2015 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Chuck Lever III X-Patchwork-Id: 6379901 Return-Path: X-Original-To: patchwork-linux-rdma@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork1.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.136]) by patchwork1.web.kernel.org (Postfix) with ESMTP id BA90A9F32E for ; Mon, 11 May 2015 18:03:21 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id 4BDDC20954 for ; Mon, 11 May 2015 18:03:20 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id DE5192094E for ; Mon, 11 May 2015 18:03:18 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754559AbbEKSDL (ORCPT ); Mon, 11 May 2015 14:03:11 -0400 Received: from mail-qc0-f176.google.com ([209.85.216.176]:33503 "EHLO mail-qc0-f176.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754550AbbEKSDH (ORCPT ); Mon, 11 May 2015 14:03:07 -0400 Received: by qcvo8 with SMTP id o8so49354388qcv.0; Mon, 11 May 2015 11:03:06 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=sender:subject:from:to:date:message-id:in-reply-to:references :user-agent:mime-version:content-type:content-transfer-encoding; bh=HdaLgc6cln3OyfTuugmQNq2EZV7g2whXlrdC9IQRu0o=; b=iyhqVfeV+uDZN9Nbp3Agy2KXGYpVG+pE/Gg14HbK/5Z5zsxinQikW00pKu0fh1tpWh SHLivCldCFbTFPEXpT7GmiuU5933hsR2rXpGr6CFcAp+n6R7XMCIb1UpcxjW60hKhahs 6rzTDRSrYiPe5RcrjtLw9hx1PJPI8w/euE/nCMJFX/8t0tJ758dIk1Ea4NJ/hqLmQryR DYOwJ62wB1r5PUSeRosfcgp2xnNV2f8YvygmFsHtBZxooTzFhdZ5q+nGvMa4GFYThjve f0y3qZuA37rbjBZF8i9OvocNjngXP3pnZYayE0PLoSyUmXX3s3nJlkxlvOr5pEe5qnD6 mlgA== X-Received: by 10.140.35.178 with SMTP id n47mr14815635qgn.24.1431367386911; Mon, 11 May 2015 11:03:06 -0700 (PDT) Received: from manet.1015granger.net ([2604:8800:100:81fc:82ee:73ff:fe43:d64f]) by mx.google.com with ESMTPSA id j40sm11066551qkh.46.2015.05.11.11.03.05 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Mon, 11 May 2015 11:03:05 -0700 (PDT) Subject: [PATCH v2 05/16] xprtrdma: Use ib_device pointer safely From: Chuck Lever To: linux-rdma@vger.kernel.org, linux-nfs@vger.kernel.org Date: Mon, 11 May 2015 14:03:04 -0400 Message-ID: <20150511180304.31263.36694.stgit@manet.1015granger.net> In-Reply-To: <20150511174401.31263.79596.stgit@manet.1015granger.net> References: <20150511174401.31263.79596.stgit@manet.1015granger.net> User-Agent: StGit/0.17.1-3-g7d0f MIME-Version: 1.0 Sender: linux-rdma-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-rdma@vger.kernel.org X-Spam-Status: No, score=-6.8 required=5.0 tests=BAYES_00,DKIM_SIGNED, RCVD_IN_DNSWL_HI,T_DKIM_INVALID,T_RP_MATCHES_RCVD,UNPARSEABLE_RELAY autolearn=ham version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP The connect worker can replace ri_id, but prevents ri_id->device from changing during the lifetime of a transport instance. The old ID is kept around until a new ID is created and the ->device is confirmed to be the same. Cache a copy of ri_id->device in rpcrdma_ia and in rpcrdma_rep. The cached copy can be used safely in code that does not serialize with the connect worker. Other code can use it to save an extra address generation (one pointer dereference instead of two). Signed-off-by: Chuck Lever Reviewed-by: Steve Wise --- net/sunrpc/xprtrdma/fmr_ops.c | 8 +---- net/sunrpc/xprtrdma/frwr_ops.c | 12 +++---- net/sunrpc/xprtrdma/physical_ops.c | 8 +---- net/sunrpc/xprtrdma/verbs.c | 61 +++++++++++++++++++----------------- net/sunrpc/xprtrdma/xprt_rdma.h | 2 + 5 files changed, 43 insertions(+), 48 deletions(-) -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html diff --git a/net/sunrpc/xprtrdma/fmr_ops.c b/net/sunrpc/xprtrdma/fmr_ops.c index 302d4eb..0a96155 100644 --- a/net/sunrpc/xprtrdma/fmr_ops.c +++ b/net/sunrpc/xprtrdma/fmr_ops.c @@ -85,7 +85,7 @@ fmr_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg, int nsegs, bool writing) { struct rpcrdma_ia *ia = &r_xprt->rx_ia; - struct ib_device *device = ia->ri_id->device; + struct ib_device *device = ia->ri_device; enum dma_data_direction direction = rpcrdma_data_dir(writing); struct rpcrdma_mr_seg *seg1 = seg; struct rpcrdma_mw *mw = seg1->rl_mw; @@ -137,17 +137,13 @@ fmr_op_unmap(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg) { struct rpcrdma_ia *ia = &r_xprt->rx_ia; struct rpcrdma_mr_seg *seg1 = seg; - struct ib_device *device; int rc, nsegs = seg->mr_nsegs; LIST_HEAD(l); list_add(&seg1->rl_mw->r.fmr->list, &l); rc = ib_unmap_fmr(&l); - read_lock(&ia->ri_qplock); - device = ia->ri_id->device; while (seg1->mr_nsegs--) - rpcrdma_unmap_one(device, seg++); - read_unlock(&ia->ri_qplock); + rpcrdma_unmap_one(ia->ri_device, seg++); if (rc) goto out_err; return nsegs; diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c index dff0481..66a85fa 100644 --- a/net/sunrpc/xprtrdma/frwr_ops.c +++ b/net/sunrpc/xprtrdma/frwr_ops.c @@ -137,7 +137,7 @@ static int frwr_op_init(struct rpcrdma_xprt *r_xprt) { struct rpcrdma_buffer *buf = &r_xprt->rx_buf; - struct ib_device *device = r_xprt->rx_ia.ri_id->device; + struct ib_device *device = r_xprt->rx_ia.ri_device; unsigned int depth = r_xprt->rx_ia.ri_max_frmr_depth; struct ib_pd *pd = r_xprt->rx_ia.ri_pd; int i; @@ -178,7 +178,7 @@ frwr_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg, int nsegs, bool writing) { struct rpcrdma_ia *ia = &r_xprt->rx_ia; - struct ib_device *device = ia->ri_id->device; + struct ib_device *device = ia->ri_device; enum dma_data_direction direction = rpcrdma_data_dir(writing); struct rpcrdma_mr_seg *seg1 = seg; struct rpcrdma_mw *mw = seg1->rl_mw; @@ -263,7 +263,6 @@ frwr_op_unmap(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg) struct rpcrdma_ia *ia = &r_xprt->rx_ia; struct ib_send_wr invalidate_wr, *bad_wr; int rc, nsegs = seg->mr_nsegs; - struct ib_device *device; seg1->rl_mw->r.frmr.fr_state = FRMR_IS_INVALID; @@ -273,10 +272,9 @@ frwr_op_unmap(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg) invalidate_wr.ex.invalidate_rkey = seg1->rl_mw->r.frmr.fr_mr->rkey; DECR_CQCOUNT(&r_xprt->rx_ep); - read_lock(&ia->ri_qplock); - device = ia->ri_id->device; while (seg1->mr_nsegs--) - rpcrdma_unmap_one(device, seg++); + rpcrdma_unmap_one(ia->ri_device, seg++); + read_lock(&ia->ri_qplock); rc = ib_post_send(ia->ri_id->qp, &invalidate_wr, &bad_wr); read_unlock(&ia->ri_qplock); if (rc) @@ -304,7 +302,7 @@ static void frwr_op_reset(struct rpcrdma_xprt *r_xprt) { struct rpcrdma_buffer *buf = &r_xprt->rx_buf; - struct ib_device *device = r_xprt->rx_ia.ri_id->device; + struct ib_device *device = r_xprt->rx_ia.ri_device; unsigned int depth = r_xprt->rx_ia.ri_max_frmr_depth; struct ib_pd *pd = r_xprt->rx_ia.ri_pd; struct rpcrdma_mw *r; diff --git a/net/sunrpc/xprtrdma/physical_ops.c b/net/sunrpc/xprtrdma/physical_ops.c index ba518af..da149e8 100644 --- a/net/sunrpc/xprtrdma/physical_ops.c +++ b/net/sunrpc/xprtrdma/physical_ops.c @@ -50,8 +50,7 @@ physical_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg, { struct rpcrdma_ia *ia = &r_xprt->rx_ia; - rpcrdma_map_one(ia->ri_id->device, seg, - rpcrdma_data_dir(writing)); + rpcrdma_map_one(ia->ri_device, seg, rpcrdma_data_dir(writing)); seg->mr_rkey = ia->ri_bind_mem->rkey; seg->mr_base = seg->mr_dma; seg->mr_nsegs = 1; @@ -65,10 +64,7 @@ physical_op_unmap(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg) { struct rpcrdma_ia *ia = &r_xprt->rx_ia; - read_lock(&ia->ri_qplock); - rpcrdma_unmap_one(ia->ri_id->device, seg); - read_unlock(&ia->ri_qplock); - + rpcrdma_unmap_one(ia->ri_device, seg); return 1; } diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c index 8e0bd84..ddd5b36 100644 --- a/net/sunrpc/xprtrdma/verbs.c +++ b/net/sunrpc/xprtrdma/verbs.c @@ -272,7 +272,6 @@ rpcrdma_recvcq_process_wc(struct ib_wc *wc, struct list_head *sched_list) { struct rpcrdma_rep *rep = (struct rpcrdma_rep *)(unsigned long)wc->wr_id; - struct rpcrdma_ia *ia; /* WARNING: Only wr_id and status are reliable at this point */ if (wc->status != IB_WC_SUCCESS) @@ -285,9 +284,8 @@ rpcrdma_recvcq_process_wc(struct ib_wc *wc, struct list_head *sched_list) dprintk("RPC: %s: rep %p opcode 'recv', length %u: success\n", __func__, rep, wc->byte_len); - ia = &rep->rr_rxprt->rx_ia; rep->rr_len = wc->byte_len; - ib_dma_sync_single_for_cpu(ia->ri_id->device, + ib_dma_sync_single_for_cpu(rep->rr_device, rdmab_addr(rep->rr_rdmabuf), rep->rr_len, DMA_FROM_DEVICE); prefetch(rdmab_to_msg(rep->rr_rdmabuf)); @@ -483,7 +481,7 @@ connected: pr_info("rpcrdma: connection to %pIS:%u on %s, memreg '%s', %d credits, %d responders%s\n", sap, rpc_get_port(sap), - ia->ri_id->device->name, + ia->ri_device->name, ia->ri_ops->ro_displayname, xprt->rx_buf.rb_max_requests, ird, ird < 4 && ird < tird / 2 ? " (low!)" : ""); @@ -584,8 +582,9 @@ rpcrdma_ia_open(struct rpcrdma_xprt *xprt, struct sockaddr *addr, int memreg) rc = PTR_ERR(ia->ri_id); goto out1; } + ia->ri_device = ia->ri_id->device; - ia->ri_pd = ib_alloc_pd(ia->ri_id->device); + ia->ri_pd = ib_alloc_pd(ia->ri_device); if (IS_ERR(ia->ri_pd)) { rc = PTR_ERR(ia->ri_pd); dprintk("RPC: %s: ib_alloc_pd() failed %i\n", @@ -593,7 +592,7 @@ rpcrdma_ia_open(struct rpcrdma_xprt *xprt, struct sockaddr *addr, int memreg) goto out2; } - rc = ib_query_device(ia->ri_id->device, devattr); + rc = ib_query_device(ia->ri_device, devattr); if (rc) { dprintk("RPC: %s: ib_query_device failed %d\n", __func__, rc); @@ -602,7 +601,7 @@ rpcrdma_ia_open(struct rpcrdma_xprt *xprt, struct sockaddr *addr, int memreg) if (devattr->device_cap_flags & IB_DEVICE_LOCAL_DMA_LKEY) { ia->ri_have_dma_lkey = 1; - ia->ri_dma_lkey = ia->ri_id->device->local_dma_lkey; + ia->ri_dma_lkey = ia->ri_device->local_dma_lkey; } if (memreg == RPCRDMA_FRMR) { @@ -617,7 +616,7 @@ rpcrdma_ia_open(struct rpcrdma_xprt *xprt, struct sockaddr *addr, int memreg) } } if (memreg == RPCRDMA_MTHCAFMR) { - if (!ia->ri_id->device->alloc_fmr) { + if (!ia->ri_device->alloc_fmr) { dprintk("RPC: %s: MTHCAFMR registration " "not supported by HCA\n", __func__); memreg = RPCRDMA_ALLPHYSICAL; @@ -767,9 +766,9 @@ rpcrdma_ep_create(struct rpcrdma_ep *ep, struct rpcrdma_ia *ia, init_waitqueue_head(&ep->rep_connect_wait); INIT_DELAYED_WORK(&ep->rep_connect_worker, rpcrdma_connect_worker); - sendcq = ib_create_cq(ia->ri_id->device, rpcrdma_sendcq_upcall, - rpcrdma_cq_async_error_upcall, ep, - ep->rep_attr.cap.max_send_wr + 1, 0); + sendcq = ib_create_cq(ia->ri_device, rpcrdma_sendcq_upcall, + rpcrdma_cq_async_error_upcall, ep, + ep->rep_attr.cap.max_send_wr + 1, 0); if (IS_ERR(sendcq)) { rc = PTR_ERR(sendcq); dprintk("RPC: %s: failed to create send CQ: %i\n", @@ -784,9 +783,9 @@ rpcrdma_ep_create(struct rpcrdma_ep *ep, struct rpcrdma_ia *ia, goto out2; } - recvcq = ib_create_cq(ia->ri_id->device, rpcrdma_recvcq_upcall, - rpcrdma_cq_async_error_upcall, ep, - ep->rep_attr.cap.max_recv_wr + 1, 0); + recvcq = ib_create_cq(ia->ri_device, rpcrdma_recvcq_upcall, + rpcrdma_cq_async_error_upcall, ep, + ep->rep_attr.cap.max_recv_wr + 1, 0); if (IS_ERR(recvcq)) { rc = PTR_ERR(recvcq); dprintk("RPC: %s: failed to create recv CQ: %i\n", @@ -907,7 +906,7 @@ retry: * More stuff I haven't thought of! * Rrrgh! */ - if (ia->ri_id->device != id->device) { + if (ia->ri_device != id->device) { printk("RPC: %s: can't reconnect on " "different device!\n", __func__); rdma_destroy_id(id); @@ -1049,6 +1048,7 @@ rpcrdma_create_rep(struct rpcrdma_xprt *r_xprt) goto out_free; } + rep->rr_device = ia->ri_device; rep->rr_rxprt = r_xprt; return rep; @@ -1449,9 +1449,9 @@ rpcrdma_register_internal(struct rpcrdma_ia *ia, void *va, int len, /* * All memory passed here was kmalloc'ed, therefore phys-contiguous. */ - iov->addr = ib_dma_map_single(ia->ri_id->device, + iov->addr = ib_dma_map_single(ia->ri_device, va, len, DMA_BIDIRECTIONAL); - if (ib_dma_mapping_error(ia->ri_id->device, iov->addr)) + if (ib_dma_mapping_error(ia->ri_device, iov->addr)) return -ENOMEM; iov->length = len; @@ -1495,8 +1495,8 @@ rpcrdma_deregister_internal(struct rpcrdma_ia *ia, { int rc; - ib_dma_unmap_single(ia->ri_id->device, - iov->addr, iov->length, DMA_BIDIRECTIONAL); + ib_dma_unmap_single(ia->ri_device, + iov->addr, iov->length, DMA_BIDIRECTIONAL); if (NULL == mr) return 0; @@ -1589,15 +1589,18 @@ rpcrdma_ep_post(struct rpcrdma_ia *ia, send_wr.num_sge = req->rl_niovs; send_wr.opcode = IB_WR_SEND; if (send_wr.num_sge == 4) /* no need to sync any pad (constant) */ - ib_dma_sync_single_for_device(ia->ri_id->device, - req->rl_send_iov[3].addr, req->rl_send_iov[3].length, - DMA_TO_DEVICE); - ib_dma_sync_single_for_device(ia->ri_id->device, - req->rl_send_iov[1].addr, req->rl_send_iov[1].length, - DMA_TO_DEVICE); - ib_dma_sync_single_for_device(ia->ri_id->device, - req->rl_send_iov[0].addr, req->rl_send_iov[0].length, - DMA_TO_DEVICE); + ib_dma_sync_single_for_device(ia->ri_device, + req->rl_send_iov[3].addr, + req->rl_send_iov[3].length, + DMA_TO_DEVICE); + ib_dma_sync_single_for_device(ia->ri_device, + req->rl_send_iov[1].addr, + req->rl_send_iov[1].length, + DMA_TO_DEVICE); + ib_dma_sync_single_for_device(ia->ri_device, + req->rl_send_iov[0].addr, + req->rl_send_iov[0].length, + DMA_TO_DEVICE); if (DECR_CQCOUNT(ep) > 0) send_wr.send_flags = 0; @@ -1630,7 +1633,7 @@ rpcrdma_ep_post_recv(struct rpcrdma_ia *ia, recv_wr.sg_list = &rep->rr_rdmabuf->rg_iov; recv_wr.num_sge = 1; - ib_dma_sync_single_for_cpu(ia->ri_id->device, + ib_dma_sync_single_for_cpu(ia->ri_device, rdmab_addr(rep->rr_rdmabuf), rdmab_length(rep->rr_rdmabuf), DMA_BIDIRECTIONAL); diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h index 230e7fe..300423d 100644 --- a/net/sunrpc/xprtrdma/xprt_rdma.h +++ b/net/sunrpc/xprtrdma/xprt_rdma.h @@ -62,6 +62,7 @@ struct rpcrdma_ia { const struct rpcrdma_memreg_ops *ri_ops; rwlock_t ri_qplock; + struct ib_device *ri_device; struct rdma_cm_id *ri_id; struct ib_pd *ri_pd; struct ib_mr *ri_bind_mem; @@ -173,6 +174,7 @@ struct rpcrdma_buffer; struct rpcrdma_rep { unsigned int rr_len; + struct ib_device *rr_device; struct rpcrdma_xprt *rr_rxprt; struct list_head rr_list; struct rpcrdma_regbuf *rr_rdmabuf;