From patchwork Sun Nov 19 15:28:19 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: "D. Wythe" X-Patchwork-Id: 13460480 X-Patchwork-Delegate: kuba@kernel.org Authentication-Results: smtp.subspace.kernel.org; dkim=none Received: from out30-97.freemail.mail.aliyun.com (out30-97.freemail.mail.aliyun.com [115.124.30.97]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 9F18E126; Sun, 19 Nov 2023 07:28:26 -0800 (PST) X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R711e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=ay29a033018045176;MF=alibuda@linux.alibaba.com;NM=1;PH=DS;RN=13;SR=0;TI=SMTPD_---0VweoVTA_1700407699; Received: from j66a10360.sqa.eu95.tbsite.net(mailfrom:alibuda@linux.alibaba.com fp:SMTPD_---0VweoVTA_1700407699) by smtp.aliyun-inc.com; Sun, 19 Nov 2023 23:28:23 +0800 From: "D. Wythe" To: kgraul@linux.ibm.com, wenjia@linux.ibm.com, jaka@linux.ibm.com, wintera@linux.ibm.com, guwen@linux.alibaba.com Cc: kuba@kernel.org, davem@davemloft.net, netdev@vger.kernel.org, linux-s390@vger.kernel.org, linux-rdma@vger.kernel.org, tonylu@linux.alibaba.com, pabeni@redhat.com, edumazet@google.com Subject: [PATCH net v3] net/smc: avoid data corruption caused by decline Date: Sun, 19 Nov 2023 23:28:19 +0800 Message-Id: <1700407699-97350-1-git-send-email-alibuda@linux.alibaba.com> X-Mailer: git-send-email 1.8.3.1 Precedence: bulk X-Mailing-List: netdev@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-Patchwork-Delegate: kuba@kernel.org From: "D. Wythe" We found a data corruption issue during testing of SMC-R on Redis applications. The benchmark has a low probability of reporting a strange error as shown below. "Error: Protocol error, got "\xe2" as reply type byte" Finally, we found that the retrieved error data was as follows: 0xE2 0xD4 0xC3 0xD9 0x04 0x00 0x2C 0x20 0xA6 0x56 0x00 0x16 0x3E 0x0C 0xCB 0x04 0x02 0x01 0x00 0x00 0x20 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0xE2 It is quite obvious that this is a SMC DECLINE message, which means that the applications received SMC protocol message. We found that this was caused by the following situations: client server ¦ proposal -------------> ¦ accept <------------- ¦ confirm -------------> wait confirm ¦failed llc confirm ¦ x------ (after 2s)timeout wait rsp wait decline (after 1s) timeout (after 2s) timeout ¦ decline --------------> ¦ decline <-------------- As a result, a decline message was sent in the implementation, and this message was read from TCP by the already-fallback connection. This patch double the client timeout as 2x of the server value, With this simple change, the Decline messages should never cross or collide (during Confirm link timeout). This issue requires an immediate solution, since the protocol updates involve a more long-term solution. Fixes: 0fb0b02bd6fd ("net/smc: adapt SMC client code to use the LLC flow") Signed-off-by: D. Wythe --- net/smc/af_smc.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index abd2667..8615cc0 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -598,8 +598,12 @@ static int smcr_clnt_conf_first_link(struct smc_sock *smc) struct smc_llc_qentry *qentry; int rc; - /* receive CONFIRM LINK request from server over RoCE fabric */ - qentry = smc_llc_wait(link->lgr, NULL, SMC_LLC_WAIT_TIME, + /* Receive CONFIRM LINK request from server over RoCE fabric. + * Increasing the client's timeout by twice as much as the server's + * timeout by default can temporarily avoid decline messages of + * both sides crossing or colliding + */ + qentry = smc_llc_wait(link->lgr, NULL, 2 * SMC_LLC_WAIT_TIME, SMC_LLC_CONFIRM_LINK); if (!qentry) { struct smc_clc_msg_decline dclc;