From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:55985) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XOMpX-0005cP-4e for qemu-devel@nongnu.org; Mon, 01 Sep 2014 04:19:21 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1XOMpR-0005di-0N for qemu-devel@nongnu.org; Mon, 01 Sep 2014 04:19:15 -0400 Received: from mx1.redhat.com ([209.132.183.28]:34883) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XOMpQ-0005dQ-PE for qemu-devel@nongnu.org; Mon, 01 Sep 2014 04:19:08 -0400 Date: Mon, 1 Sep 2014 11:18:24 +0300 From: "Michael S. Tsirkin" Message-ID: <20140901081824.GA21249@redhat.com> References: <1408424189-10510-1-git-send-email-jasowang@redhat.com> <53F46909.2040101@huawei.com> <53F575AF.2010100@redhat.com> <53F59174.7010104@huawei.com> <53F59771.2050602@redhat.com> <53F5A2FD.2060405@huawei.com> <20140827125939.GA17757@redhat.com> <54005898.70309@huawei.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <54005898.70309@huawei.com> Subject: Re: [Qemu-devel] [PATCH V3] vhost_net: start/stop guest notifiers properly List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Zhangjie (HZ)" Cc: Jason Wang , qemu-devel@nongnu.org, William Dauchy On Fri, Aug 29, 2014 at 06:40:24PM +0800, Zhangjie (HZ) wrote: > > > On 2014/8/27 20:59, Michael S. Tsirkin wrote: > > On Thu, Aug 21, 2014 at 03:42:53PM +0800, Zhangjie (HZ) wrote: > >> On 2014/8/21 14:53, Jason Wang wrote: > >>> On 08/21/2014 02:28 PM, Zhangjie (HZ) wrote: > >>>> > >>>> After migration, vhost is not disabled, virtual nic became unreachable because vhost is not awakened. > >>>> By the logical of EVENT_IDX, virtio-net will not kick vhost again if the used idx is not updated. > >>>> So, if one interrupts is lost during migration, virtio_net will not kick vhost again. > >>>> Then, no skb from virtio-net can be sent to tap. > >>> > >>> Yes and I mean to test vhost=off to see if it was the issue of vhost. > >> That sounds reasonable, but the test case is to test vhost. > >>>> > >>>> Jason's patch reduced the probability of occurrence, from about 1/20 to 1/80. It is really effective. I think the patch should be acked. > >>>> May be we can try to solve the problem from another perspective. Do you have some methods to sense the migration? > >>>> We can make up a signal from virtio-net after the migration. > >>> > >>> You can make a patch like this to debug. If problem disappears, it means > >>> interrupt was really lost anyway. > >>>> > >>>>> Anyway, I will try to reproduce it by myself. > >>>>> > >>>> The test environment is really terrible, I build a environment myself, but it problem did not occur. > >>>> The environment I use now is from a colleague Responsible for test work. > >>>> Two hosts, every host has about 20 vms, they send packages(ipv4 and ipv6) between each other. > >>>> The VM to be migrated also sens packages itself, and there is a ping(-i 0.001) from another host to it. > >>>> The physical nic is 1GE, connected through a internal nework. > >>> > >>> Just want to confirm. For the problem did not occur, you mean with my > >>> patch on top? > >>> . > >>> > >> I mean, with your patch, I have to test 80 times before it occurs, the probability is reduced. > > > > Could you please try to apply the patch > > [PATCH V4] net: Forbid dealing with packets when VM is not running > > on top and see if this helps? > > > > Thanks! > > > >> -- > >> Best Wishes! > >> Zhang Jie > > . > > > Thanks! I will have a test. Great, once you have the result of the two patches applied together, please let us know on the list. > -- > Best Wishes! > Zhang Jie