From mboxrd@z Thu Jan 1 00:00:00 1970 From: santosh shilimkar Subject: Re: [PATCH net-next] RDS: fix rds-ping deadlock over TCP transport Date: Fri, 16 Oct 2015 18:57:10 -0700 Message-ID: <5621AAF6.9090805@oracle.com> References: <1445041610-30133-1-git-send-email-santosh.shilimkar@oracle.com> <20151017014534.GH5464@oracle.com> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, davem@davemloft.net, linux-kernel@vger.kernel.org, ssantosh@kernel.org To: Sowmini Varadhan Return-path: Received: from userp1040.oracle.com ([156.151.31.81]:31912 "EHLO userp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751602AbbJQB5Q (ORCPT ); Fri, 16 Oct 2015 21:57:16 -0400 In-Reply-To: <20151017014534.GH5464@oracle.com> Sender: netdev-owner@vger.kernel.org List-ID: On 10/16/2015 6:45 PM, Sowmini Varadhan wrote: > On (10/16/15 20:26), Santosh Shilimkar wrote: >> >> diff --git a/net/rds/send.c b/net/rds/send.c >> + if (!test_bit(RDS_LL_SEND_FULL, &conn->c_flags)) >> + queue_delayed_work(rds_wq, &conn->c_send_w, 0); > > A minor note- it would help to add some comments here explaining > that the pong has already been added to the sendq earlier.. > in the case of IB, if RDS_LL_SEND_FULL has been set, it takes some > head-scratching to figure out how the pong gets sent, and a few > comments could help clarify that. > Right. The check confused you. I will update the patch and drop the check all together and add a jiffies for the queuing which should take care of it. Will post v2 with that update. > but other than that, the contents look good to me, thus > > Acked-by: Sowmini Varadhan > Thanks !! regards, Santosh