From mboxrd@z Thu Jan 1 00:00:00 1970 From: Steven Whitehouse Date: Mon, 26 Mar 2007 09:18:39 +0100 Subject: [Cluster-devel] Re: [PATCH] dlm: fix coverity-spotted stupidity In-Reply-To: <460777A4.7060301@redhat.com> References: <1174562653.32601.172.camel@quoit.chygwyn.com> <46026D5E.1000601@redhat.com> <1174565693.32601.184.camel@quoit.chygwyn.com> <46043FB2.9030904@redhat.com> <460777A4.7060301@redhat.com> Message-ID: <1174897119.32601.240.camel@quoit.chygwyn.com> List-Id: To: cluster-devel.redhat.com MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Hi, Since I'm just updating the git tree due to 2.6.21-rc5, I've pulled this patch for now, Steve. On Mon, 2007-03-26 at 08:35 +0100, Patrick Caulfield wrote: > Abhijith Das wrote: > > Steven Whitehouse wrote: > > > >> Hi, > >> > >> Now applied to the -nmw git tree. Thanks, > >> > >> Steve. > >> > >> On Thu, 2007-03-22 at 11:49 +0000, Patrick Caulfield wrote: > >> > >> > >>> This patch fixes the EAGAIN checking in lowcomms-tcp.c so it actually > >>> works. > >>> Thanks to Adrian Bunk & the coverity checker for spotting that > >>> > >>> Signed-Off-By: Patrick Caulfield > >>> > > Hi, > > I pulled this patch along with the latest stuff from the nmw tree. Looks > > like this patch makes dlm_recv/0 use 100% CPU. I'm not able to restart > > clvmd and such. I'm guessing the kernel_recvmsg() keeps returning > > -EAGAIN and we keep rescheduling. Any clue as to what is going on? > > > Bugger. You're right. >