Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932458Ab2JBVnl (ORCPT ); Tue, 2 Oct 2012 17:43:41 -0400 Received: from e3.ny.us.ibm.com ([32.97.182.143]:33196 "EHLO e3.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932356Ab2JBVnj (ORCPT ); Tue, 2 Oct 2012 17:43:39 -0400 Date: Tue, 2 Oct 2012 14:43:27 -0700 From: Nishanth Aravamudan To: Benjamin Herrenschmidt Cc: Alexander Graf , linux-nfs@vger.kernel.org, Jan Kara , Linus Torvalds , LKML List , "J. Bruce Fields" , anton@samba.org, skinsbursky@parallels.com, bfields@redhat.com, linuxppc-dev Subject: Re: [REGRESSION] nfsd crashing with 3.6.0-rc7 on PowerPC Message-ID: <20121002214327.GA29218@linux.vnet.ibm.com> References: <3BDA9E62-7031-42D6-8CA9-5327B61700F5@suse.de> <20120928151043.GA19102@fieldses.org> <2A52FC96-148C-4F7A-9950-E152E0C6698D@suse.de> <1349139509.3847.2.camel@pasglop> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1349139509.3847.2.camel@pasglop> X-Operating-System: Linux 3.2.0-31-generic (x86_64) User-Agent: Mutt/1.5.21 (2010-09-15) x-cbid: 12100221-8974-0000-0000-00000ED56779 X-IBM-ISS-SpamDetectors: X-IBM-ISS-DetailInfo: BY=3.00000294; HX=3.00000196; KW=3.00000007; PH=3.00000001; SC=3.00000008; SDB=6.00179273; UDB=6.00040600; UTC=2012-10-02 21:43:37 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1220 Lines: 31 Hi Ben, On 02.10.2012 [10:58:29 +1000], Benjamin Herrenschmidt wrote: > On Mon, 2012-10-01 at 16:03 +0200, Alexander Graf wrote: > > Phew. Here we go :). It looks to be more of a PPC specific problem > > than it appeared as at first: > > Ok, so I suspect the problem is the pushing down of the locks which > breaks with iommu backends that have a separate flush callback. In > that case, the flush moves out of the allocator lock. > > Now we do call flush before we return, still, but it becomes racy > I suspect, but somebody needs to give it a closer look. I'm hoping > Anton or Nish will later today. Started looking into this. If your suspicion were accurate, wouldn't the bisection have stopped at 0e4bc95d87394364f408627067238453830bdbf3 ("powerpc/iommu: Reduce spinlock coverage in iommu_alloc and iommu_free")? Alex, the error is reproducible, right? Does it go away by reverting that commit against mainline? Just trying to narrow down my focus. Thanks, Nish -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/