From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Marchand Subject: Re: [PATCH v6 3/3] ivshmem: add check on protocol version in QEMU Date: Thu, 25 Sep 2014 10:38:40 +0200 Message-ID: <5423D490.7050306@6wind.com> References: <1410167870-680-1-git-send-email-david.marchand@6wind.com> <1410167870-680-4-git-send-email-david.marchand@6wind.com> <20140908094948.GB457@redhat.com> <20140923155831.GB3150@stefanha-thinkpad.redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: qemu-devel@nongnu.org, kvm@vger.kernel.org, pbonzini@redhat.com, claudio.fontana@huawei.com, jani.kokkonen@huawei.com, eblake@redhat.com, cam@cs.ualberta.ca, armbru@redhat.com, arei.gonglei@huawei.com, mkletzan@redhat.com To: Stefan Hajnoczi , "Michael S. Tsirkin" Return-path: Received: from mail-wg0-f52.google.com ([74.125.82.52]:39729 "EHLO mail-wg0-f52.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751001AbaIYIio (ORCPT ); Thu, 25 Sep 2014 04:38:44 -0400 Received: by mail-wg0-f52.google.com with SMTP id n12so5821594wgh.35 for ; Thu, 25 Sep 2014 01:38:43 -0700 (PDT) In-Reply-To: <20140923155831.GB3150@stefanha-thinkpad.redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: On 09/23/2014 05:58 PM, Stefan Hajnoczi wrote: > I'm not sure a full-fledged feature negotiation system is needed. The > ivshmem protocol is local to the host and all participants are under > control of the administrator. > > I suggested a protocol version to protect against misconfiguration. For > example, building QEMU from source but talking to an outdated ivhsmem > server that is still running from before. > > Remember that ivshmem-server and QEMU are shipped together by the > distro. So in 99% of the cases they will have the same version anyway. > But we want to protect against rare misconfiguration that break things > (user mixing and matching incompatible software). Ok, so how about keeping with this simple mechanism at the moment ? If this versionning system is too limited in the future, I think we will need a global rework on the protocol, anyway. -- David Marchand From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:43067) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XX4hj-0001vr-JI for qemu-devel@nongnu.org; Thu, 25 Sep 2014 04:47:19 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1XX4hc-0005Mn-3l for qemu-devel@nongnu.org; Thu, 25 Sep 2014 04:47:11 -0400 Received: from mail-we0-f178.google.com ([74.125.82.178]:43738) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XX4hb-0005Jr-UF for qemu-devel@nongnu.org; Thu, 25 Sep 2014 04:47:04 -0400 Received: by mail-we0-f178.google.com with SMTP id t60so7573353wes.9 for ; Thu, 25 Sep 2014 01:46:57 -0700 (PDT) Message-ID: <5423D490.7050306@6wind.com> Date: Thu, 25 Sep 2014 10:38:40 +0200 From: David Marchand MIME-Version: 1.0 References: <1410167870-680-1-git-send-email-david.marchand@6wind.com> <1410167870-680-4-git-send-email-david.marchand@6wind.com> <20140908094948.GB457@redhat.com> <20140923155831.GB3150@stefanha-thinkpad.redhat.com> In-Reply-To: <20140923155831.GB3150@stefanha-thinkpad.redhat.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH v6 3/3] ivshmem: add check on protocol version in QEMU List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Stefan Hajnoczi , "Michael S. Tsirkin" Cc: kvm@vger.kernel.org, claudio.fontana@huawei.com, qemu-devel@nongnu.org, armbru@redhat.com, arei.gonglei@huawei.com, mkletzan@redhat.com, pbonzini@redhat.com, jani.kokkonen@huawei.com, cam@cs.ualberta.ca On 09/23/2014 05:58 PM, Stefan Hajnoczi wrote: > I'm not sure a full-fledged feature negotiation system is needed. The > ivshmem protocol is local to the host and all participants are under > control of the administrator. > > I suggested a protocol version to protect against misconfiguration. For > example, building QEMU from source but talking to an outdated ivhsmem > server that is still running from before. > > Remember that ivshmem-server and QEMU are shipped together by the > distro. So in 99% of the cases they will have the same version anyway. > But we want to protect against rare misconfiguration that break things > (user mixing and matching incompatible software). Ok, so how about keeping with this simple mechanism at the moment ? If this versionning system is too limited in the future, I think we will need a global rework on the protocol, anyway. -- David Marchand