Return-Path: Received: from mx1.redhat.com ([209.132.183.28]:44647 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754259Ab1I1NpL (ORCPT ); Wed, 28 Sep 2011 09:45:11 -0400 Date: Wed, 28 Sep 2011 09:44:49 -0400 From: "J. Bruce Fields" To: Bryan Schumaker Cc: linux-nfs@vger.kernel.org Subject: Re: [PATCH 03/25] nfsd4: extend state lock over seqid replay logic Message-ID: <20110928134442.GD19435@pad.fieldses.org> References: <1316000721-3289-1-git-send-email-bfields@redhat.com> <1316000721-3289-4-git-send-email-bfields@redhat.com> <4E82001C.5030005@netapp.com> <20110928014930.GA18571@pad.fieldses.org> <4E831EB8.90800@netapp.com> Content-Type: text/plain; charset=us-ascii In-Reply-To: <4E831EB8.90800@netapp.com> Sender: linux-nfs-owner@vger.kernel.org List-ID: MIME-Version: 1.0 On Wed, Sep 28, 2011 at 09:18:48AM -0400, Bryan Schumaker wrote: > On 09/27/2011 09:49 PM, J. Bruce Fields wrote: > > On Tue, Sep 27, 2011 at 12:55:56PM -0400, Bryan Schumaker wrote: > >> I'm getting the following warning that I was able to bisect to this patch: > > > > I suspect this may fix it. (Untested.) > > Yeah, that seems to fix it. Thanks! Thanks for the testing; applying for 3.2.--b. > > > > --b. > > > > commit 8da87285a245c82d23183414cec7069f18721afd > > Author: J. Bruce Fields > > Date: Tue Sep 27 21:42:29 2011 -0400 > > > > nfsd4: fix state lock usage in LOCKU > > > > In commit 5ec094c1096ab3bb795651855d53f18daa26afde "nfsd4: extend state > > lock over seqid replay logic" I modified the exit logic of all the > > seqid-based procedures except nfsd4_locku(). Fix the oversight. > > > > The result of the bug was a double-unlock while handling the LOCKU > > procedure, and a warning like: > > > > [ 142.150014] WARNING: at kernel/mutex-debug.c:78 debug_mutex_unlock+0xda/0xe0() > > ... > > [ 142.152927] Pid: 742, comm: nfsd Not tainted 3.1.0-rc1-SLIM+ #9 > > [ 142.152927] Call Trace: > > [ 142.152927] [] warn_slowpath_common+0x7f/0xc0 > > [ 142.152927] [] warn_slowpath_null+0x1a/0x20 > > [ 142.152927] [] debug_mutex_unlock+0xda/0xe0 > > [ 142.152927] [] __mutex_unlock_slowpath+0x80/0x140 > > [ 142.152927] [] mutex_unlock+0xe/0x10 > > [ 142.152927] [] nfs4_lock_state+0x35/0x40 [nfsd] > > [ 142.152927] [] nfsd4_proc_compound+0x2a1/0x690 > > [nfsd] > > [ 142.152927] [] nfsd_dispatch+0xeb/0x230 [nfsd] > > [ 142.152927] [] svc_process_common+0x345/0x690 > > [sunrpc] > > [ 142.152927] [] ? try_to_wake_up+0x280/0x280 > > [ 142.152927] [] svc_process+0x102/0x150 [sunrpc] > > [ 142.152927] [] nfsd+0xbd/0x160 [nfsd] > > [ 142.152927] [] ? 0xffffffffa039efff > > [ 142.152927] [] kthread+0x8c/0xa0 > > [ 142.152927] [] kernel_thread_helper+0x4/0x10 > > [ 142.152927] [] ? kthread_worker_fn+0x190/0x190 > > [ 142.152927] [] ? gs_change+0x13/0x13 > > > > Reported-by: Bryan Schumaker > > Signed-off-by: J. Bruce Fields > > > > diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c > > index 752c0f8..9daf254 100644 > > --- a/fs/nfsd/nfs4state.c > > +++ b/fs/nfsd/nfs4state.c > > @@ -4199,7 +4199,8 @@ nfsd4_locku(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, > > memcpy(&locku->lu_stateid, &stp->st_stid.sc_stateid, sizeof(stateid_t)); > > > > out: > > - nfs4_unlock_state(); > > + if (!cstate->replay_owner) > > + nfs4_unlock_state(); > > return status; > > > > out_nfserr: > > -- > > To unsubscribe from this list: send the line "unsubscribe linux-nfs" in > > the body of a message to majordomo@vger.kernel.org > > More majordomo info at http://vger.kernel.org/majordomo-info.html >