From patchwork Fri Aug 26 09:51:33 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "D. Wythe" X-Patchwork-Id: 12955801 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 91A38ECAAA3 for ; Fri, 26 Aug 2022 09:52:36 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1344142AbiHZJwe (ORCPT ); Fri, 26 Aug 2022 05:52:34 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:56962 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1344117AbiHZJwG (ORCPT ); Fri, 26 Aug 2022 05:52:06 -0400 Received: from out199-1.us.a.mail.aliyun.com (out199-1.us.a.mail.aliyun.com [47.90.199.1]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7F45CD87DC; Fri, 26 Aug 2022 02:52:01 -0700 (PDT) X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R281e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=ay29a033018045170;MF=alibuda@linux.alibaba.com;NM=1;PH=DS;RN=8;SR=0;TI=SMTPD_---0VNIjhT1_1661507516; Received: from localhost(mailfrom:alibuda@linux.alibaba.com fp:SMTPD_---0VNIjhT1_1661507516) by smtp.aliyun-inc.com; Fri, 26 Aug 2022 17:51:57 +0800 From: "D. Wythe" To: kgraul@linux.ibm.com, wenjia@linux.ibm.com Cc: kuba@kernel.org, davem@davemloft.net, netdev@vger.kernel.org, linux-s390@vger.kernel.org, linux-rdma@vger.kernel.org, "D. Wythe" Subject: [PATCH net-next v2 06/10] net/smc: use read semaphores to reduce unnecessary blocking in smc_buf_create() & smcr_buf_unuse() Date: Fri, 26 Aug 2022 17:51:33 +0800 Message-Id: <184516129e6d80241baa79cd2754b647504a3163.1661407821.git.alibuda@linux.alibaba.com> X-Mailer: git-send-email 1.8.3.1 In-Reply-To: References: Precedence: bulk List-ID: X-Mailing-List: linux-rdma@vger.kernel.org From: "D. Wythe" Following is part of Off-CPU graph during frequent SMC-R short-lived processing: process_one_work (51.19%) smc_close_passive_work (28.36%) smcr_buf_unuse (28.34%) rwsem_down_write_slowpath (28.22%) smc_listen_work (22.83%) smc_clc_wait_msg (1.84%) smc_buf_create (20.45%) smcr_buf_map_usable_links rwsem_down_write_slowpath (20.43%) smcr_lgr_reg_rmbs (0.53%) rwsem_down_write_slowpath (0.43%) smc_llc_do_confirm_rkey (0.08%) We can clearly see that during the connection establishment time, waiting time of connections is not on IO, but on llc_conf_mutex. What is more important, the core critical area (smcr_buf_unuse() & smc_buf_create()) only perfroms read semantics on links, we can easily replace it with read semaphore. Signed-off-by: D. Wythe --- net/smc/smc_core.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c index 95d8438..1f66f14 100644 --- a/net/smc/smc_core.c +++ b/net/smc/smc_core.c @@ -1362,10 +1362,10 @@ static void smcr_buf_unuse(struct smc_buf_desc *buf_desc, bool is_rmb, rc = smc_llc_flow_initiate(lgr, SMC_LLC_FLOW_RKEY); if (!rc) { /* protect against smc_llc_cli_rkey_exchange() */ - down_write(&lgr->llc_conf_mutex); + down_read(&lgr->llc_conf_mutex); smc_llc_do_delete_rkey(lgr, buf_desc); buf_desc->is_conf_rkey = false; - up_write(&lgr->llc_conf_mutex); + up_read(&lgr->llc_conf_mutex); smc_llc_flow_stop(lgr, &lgr->llc_flow_lcl); } } @@ -2567,7 +2567,7 @@ static int smcr_buf_map_usable_links(struct smc_link_group *lgr, int i, rc = 0; /* protect against parallel link reconfiguration */ - down_write(&lgr->llc_conf_mutex); + down_read(&lgr->llc_conf_mutex); for (i = 0; i < SMC_LINKS_PER_LGR_MAX; i++) { struct smc_link *lnk = &lgr->lnk[i]; @@ -2579,7 +2579,7 @@ static int smcr_buf_map_usable_links(struct smc_link_group *lgr, } } out: - up_write(&lgr->llc_conf_mutex); + up_read(&lgr->llc_conf_mutex); return rc; }