Return-Path: Received: from mx2.netapp.com ([216.240.18.37]:45995 "EHLO mx2.netapp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752854Ab1G3BJe convert rfc822-to-8bit (ORCPT ); Fri, 29 Jul 2011 21:09:34 -0400 Subject: Re: [PATCH v4 00/27] add block layout driver to pnfs client From: Trond Myklebust To: Jim Rees Cc: Christoph Hellwig , linux-nfs@vger.kernel.org, peter honeyman Date: Fri, 29 Jul 2011 21:09:32 -0400 In-Reply-To: <20110729191341.GC23061@merit.edu> References: <1311874276-1386-1-git-send-email-rees@umich.edu> <20110729155136.GB28306@infradead.org> <20110729185415.GA23061@merit.edu> <20110729190133.GA10946@infradead.org> <20110729191341.GC23061@merit.edu> Content-Type: text/plain; charset="UTF-8" Message-ID: <1311988172.16078.15.camel@lade.trondhjem.org> Sender: linux-nfs-owner@vger.kernel.org List-ID: MIME-Version: 1.0 On Fri, 2011-07-29 at 15:13 -0400, Jim Rees wrote: > Christoph Hellwig wrote: > > On Fri, Jul 29, 2011 at 02:54:15PM -0400, Jim Rees wrote: > > xfstests does require a small patch to work with NFSv4, which I can supply > > if anyone is interested. > > Please send it to xfs@oss.sgi.com with a proper description and signoff. > > I don't have a proper patch, just one that works for me. But I'll send a > bug report. It has to do with the mismatch between nfs and nfs4 mount > types, so it's not really a xfstests bug. I think this will fix itself when > the nfsvers=4 changes fully propagate. > > It seems like you didn't set up the SCRATCH_DEV variable properly. > > I was just skipping those tests so I could get to the one that fails. I've > also tested with the SCRATCH_DEV tests in other runs. > > > Message from syslogd@rhcl1 at Jul 29 14:42:05 ... > > kernel:------------[ cut here ]------------ > > > > Message from syslogd@rhcl1 at Jul 29 14:42:05 ... > > kernel:invalid opcode: 0000 [#1] SMP > > > > Message from syslogd@rhcl1 at Jul 29 14:42:05 ... > > kernel:Stack: > > > > Message from syslogd@rhcl1 at Jul 29 14:42:05 ... > > kernel:Call Trace: > > > > Message from syslogd@rhcl1 at Jul 29 14:42:05 ... > > Looks like we did find a bug in NFS. > > It kind of looks that way. Is that reproducible on the upstream kernel, or is it something that is being introduced by the pNFS blocks code? -- Trond Myklebust Linux NFS client maintainer NetApp Trond.Myklebust@netapp.com www.netapp.com