From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1O9aMA-0006VK-6H for qemu-devel@nongnu.org; Wed, 05 May 2010 04:53:26 -0400 Received: from [140.186.70.92] (port=41210 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1O9aM2-0006UX-Uw for qemu-devel@nongnu.org; Wed, 05 May 2010 04:53:23 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69) (envelope-from ) id 1O9aLz-0006hs-R3 for qemu-devel@nongnu.org; Wed, 05 May 2010 04:53:18 -0400 Received: from mx1.redhat.com ([209.132.183.28]:46886) by eggs.gnu.org with esmtp (Exim 4.69) (envelope-from ) id 1O9aLz-0006hf-HP for qemu-devel@nongnu.org; Wed, 05 May 2010 04:53:15 -0400 Message-ID: <4BE131EC.2030704@redhat.com> Date: Wed, 05 May 2010 10:53:00 +0200 From: Gerd Hoffmann MIME-Version: 1.0 Subject: Re: [Qemu-devel] Patch to improve handling of server sockets References: <4BE049F5.6050105@codemonkey.ws> <4BE0960E.5000407@redhat.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Reinhard Max Cc: qemu-devel@nongnu.org Hi, >> When going for multiple listening sockets in qemu we have to figure >> how we'll handle this in a number of places as there is no single >> listening address any more. > > Well, that's what my patch is about. Sure. > Did you take a look at it? Briefly, yes. Overall it looks sensible to me. Devil is in the details though, see below. Noticed that it probably should get a few helper functions to handle FdLists to avoid the quite simliar open-coded loop-over-all-fds loops all over the place. >> Reporting the vnc server address in QMP is one. > > Not sure what QMP is (this was the first time I looked at QEMU's > internals), You'll run into qmp for sure when forward-porting the patches to the latest qemu bits. It is the machine-readable version of the monitor protocol (in qemu 0.12+). > but I think my patch only leaves one place TODO where I > chose to report only the first address for now, but it shouldn't be too > hard to fix that as well. Yea. I've noticed that TODO ;) > BTW, in some places I circumvented the need for reporting multiple > addresses by simply reporting the name that was passed to QEMU instead. This is one of the issues which needs to be addressed somehow. First I think qemu should be self-consistent here, i.e. either report the (single) name or the list of addressed everythere. Second we have to care about the current users (especially libvirt). Today qemu usually reports the address I think. Thus I tend to stick to addresses to keep them happy. We'll have a externally visible change in any case though. Either the switch from the address to the name or the switch from a single address to a list of addresses. Both changes might break existing users. cheers, Gerd