From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:58047) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XJZUB-0006wB-LJ for qemu-devel@nongnu.org; Mon, 18 Aug 2014 22:49:28 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1XJZU6-0003N2-Iz for qemu-devel@nongnu.org; Mon, 18 Aug 2014 22:49:23 -0400 Received: from mx1.redhat.com ([209.132.183.28]:51858) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XJZU6-0003Mq-Bw for qemu-devel@nongnu.org; Mon, 18 Aug 2014 22:49:18 -0400 Message-ID: <53F2BB23.9010506@redhat.com> Date: Tue, 19 Aug 2014 10:49:07 +0800 From: Jason Wang MIME-Version: 1.0 References: <1408355491-42089-1-git-send-email-jasowang@redhat.com> <20140818195328.GA30209@redhat.com> In-Reply-To: <20140818195328.GA30209@redhat.com> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH] vhost_net: start/stop guest notifiers properly List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Michael S. Tsirkin" Cc: zhangjie14@huawei.com, qemu-devel@nongnu.org On 08/19/2014 03:53 AM, Michael S. Tsirkin wrote: > On Mon, Aug 18, 2014 at 05:51:31PM +0800, Jason Wang wrote: >> > commit a9f98bb5ebe6fb1869321dcc58e72041ae626ad8 vhost: multiqueue >> > support changed the order of stopping the device. Previously >> > vhost_dev_stop would disable backend and only afterwards, unset guest >> > notifiers. We now unset guest notifiers while vhost is still >> > active. This can lose interrupts causing guest networking to fail. >> > >> > Additionally, remove the hdev->started assert in vhost.c since we may >> > want to start the guest notifiers before vhost starts and stop the >> > guest notifiers after vhost is stopped. >> > >> > In particular, this has been observed during migration. >> > >> > Reported-by: "Zhangjie (HZ)" >> > Signed-off-by: Michael S. Tsirkin >> > Signed-off-by: Jason Wang > This doesn't seem to apply to master. > Can you rebase please? Yes, will send a new version.