From patchwork Tue Oct 24 14:16:28 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Haakon Bugge X-Patchwork-Id: 829913 X-Patchwork-Delegate: davem@davemloft.net Return-Path: X-Original-To: patchwork-incoming@ozlabs.org Delivered-To: patchwork-incoming@ozlabs.org Authentication-Results: ozlabs.org; spf=none (mailfrom) smtp.mailfrom=vger.kernel.org (client-ip=209.132.180.67; helo=vger.kernel.org; envelope-from=netdev-owner@vger.kernel.org; receiver=) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id 3yLwMD47Z3z9t4P for ; Wed, 25 Oct 2017 01:17:12 +1100 (AEDT) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751645AbdJXOQt (ORCPT ); Tue, 24 Oct 2017 10:16:49 -0400 Received: from aserp1040.oracle.com ([141.146.126.69]:39836 "EHLO aserp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751399AbdJXOQq (ORCPT ); Tue, 24 Oct 2017 10:16:46 -0400 Received: from userv0022.oracle.com (userv0022.oracle.com [156.151.31.74]) by aserp1040.oracle.com (Sentrion-MTA-4.3.2/Sentrion-MTA-4.3.2) with ESMTP id v9OEGfTC031631 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK); Tue, 24 Oct 2017 14:16:42 GMT Received: from userv0121.oracle.com (userv0121.oracle.com [156.151.31.72]) by userv0022.oracle.com (8.14.4/8.14.4) with ESMTP id v9OEGelC002197 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK); Tue, 24 Oct 2017 14:16:40 GMT Received: from abhmp0006.oracle.com (abhmp0006.oracle.com [141.146.116.12]) by userv0121.oracle.com (8.14.4/8.13.8) with ESMTP id v9OEGeEs031026; Tue, 24 Oct 2017 14:16:40 GMT Received: from lab02.no.oracle.com (/10.172.144.56) by default (Oracle Beehive Gateway v4.0) with ESMTP ; Tue, 24 Oct 2017 07:16:39 -0700 From: =?utf-8?q?H=C3=A5kon_Bugge?= To: Santosh Shilimkar , "David S . Miller" Cc: netdev@vger.kernel.org, linux-rdma@vger.kernel.org, rds-devel@oss.oracle.com, linux-kernel@vger.kernel.org Subject: [PATCH] rds: Fix inaccurate accounting of unsignaled wrs Date: Tue, 24 Oct 2017 16:16:28 +0200 Message-Id: <20171024141628.14637-1-Haakon.Bugge@oracle.com> X-Mailer: git-send-email 2.9.3 MIME-Version: 1.0 X-Source-IP: userv0022.oracle.com [156.151.31.74] Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org The number of unsignaled work-requests posted to the IB send queue is tracked by a counter in the rds_ib_connection struct. When it reaches zero, or the caller explicitly asks for it, the send-signaled bit is set in send_flags and the counter is reset. This is performed by the rds_ib_set_wr_signal_state() function. However, this function is not always used which yields inaccurate accounting. This commit fixes this, re-factors a code bloat related to the matter, and makes the actual parameter type to the function consistent. Signed-off-by: Håkon Bugge --- net/rds/ib_send.c | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) diff --git a/net/rds/ib_send.c b/net/rds/ib_send.c index 6ab39db..d07ecb0 100644 --- a/net/rds/ib_send.c +++ b/net/rds/ib_send.c @@ -661,13 +661,15 @@ int rds_ib_xmit(struct rds_connection *conn, struct rds_message *rm, } } - rds_ib_set_wr_signal_state(ic, send, 0); + rds_ib_set_wr_signal_state(ic, send, false); /* * Always signal the last one if we're stopping due to flow control. */ - if (ic->i_flowctl && flow_controlled && i == (work_alloc-1)) - send->s_wr.send_flags |= IB_SEND_SIGNALED | IB_SEND_SOLICITED; + if (ic->i_flowctl && flow_controlled && i == (work_alloc - 1)) { + rds_ib_set_wr_signal_state(ic, send, true); + send->s_wr.send_flags |= IB_SEND_SOLICITED; + } if (send->s_wr.send_flags & IB_SEND_SIGNALED) nr_sig++; @@ -705,11 +707,8 @@ int rds_ib_xmit(struct rds_connection *conn, struct rds_message *rm, if (scat == &rm->data.op_sg[rm->data.op_count]) { prev->s_op = ic->i_data_op; prev->s_wr.send_flags |= IB_SEND_SOLICITED; - if (!(prev->s_wr.send_flags & IB_SEND_SIGNALED)) { - ic->i_unsignaled_wrs = rds_ib_sysctl_max_unsig_wrs; - prev->s_wr.send_flags |= IB_SEND_SIGNALED; - nr_sig++; - } + if (!(prev->s_wr.send_flags & IB_SEND_SIGNALED)) + nr_sig += rds_ib_set_wr_signal_state(ic, prev, true); ic->i_data_op = NULL; }