From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:47979) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1c6NEm-0003Ju-3f for qemu-devel@nongnu.org; Mon, 14 Nov 2016 14:48:16 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1c6NEg-00006S-Nn for qemu-devel@nongnu.org; Mon, 14 Nov 2016 14:48:16 -0500 Received: from mx1.redhat.com ([209.132.183.28]:49730) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1c6NEg-000063-Hz for qemu-devel@nongnu.org; Mon, 14 Nov 2016 14:48:10 -0500 Date: Mon, 14 Nov 2016 12:48:08 -0700 From: Alex Williamson Message-ID: <20161114124808.2b37c0d9@t450s.home> In-Reply-To: <635f3da4-ce9f-5fe8-bbc7-e2fcf97bed63@redhat.com> References: <1477082896-26553-1-git-send-email-pbonzini@redhat.com> <1477082896-26553-8-git-send-email-pbonzini@redhat.com> <20161110210907.41ebcb50@t450s.home> <20161111140342.7825fd1b@t450s.home> <4e7ae711-d7c0-35d3-4823-5856922c98b4@redhat.com> <20161114100932.46a5a2a6@t450s.home> <20161114114929.0a45aacf@t450s.home> <635f3da4-ce9f-5fe8-bbc7-e2fcf97bed63@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH 07/13] virtio-scsi: always use dataplane path if ioeventfd is active List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Paolo Bonzini Cc: cornelia.huck@de.ibm.com, famz@redhat.com, qemu-devel@nongnu.org, stefanha@redhat.com, mst@redhat.com On Mon, 14 Nov 2016 20:33:32 +0100 Paolo Bonzini wrote: > On 14/11/2016 19:49, Alex Williamson wrote: > > On Mon, 14 Nov 2016 19:10:54 +0100 > > Paolo Bonzini wrote: > > > >> On 14/11/2016 18:09, Alex Williamson wrote: > >>> Hmm, fixed yet not fixed. I get a nice shutdown and it even eliminates > >>> a cpu spike shown in virt-manager at the end of shutdown that was > >>> typical previously, but then I noticed dmesg showing me segfaults, so I > >>> hooked up gdb and: > >> > >> Well, that I don't get the segfault already says something... Though > >> it's not even clear from the backtrace what is causing the segfault. > >> Let's try the same printf without the change. > > > > Log starting from virsh shutdown. The final line occurs just as the VM > > starts spinning on all vcpus. Thanks, > > Ok, this could be a start. Just for information what is your virtio-win > version? Device manager shows: Driver Date: 9/22/2015 Driver Version: 62.72.104.11000