From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:36082) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UFq7C-0006QX-Q6 for qemu-devel@nongnu.org; Wed, 13 Mar 2013 14:09:27 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1UFq78-0000Yr-OZ for qemu-devel@nongnu.org; Wed, 13 Mar 2013 14:09:26 -0400 Received: from mail-ye0-f171.google.com ([209.85.213.171]:42856) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UFq78-0000Ym-Ks for qemu-devel@nongnu.org; Wed, 13 Mar 2013 14:09:22 -0400 Received: by mail-ye0-f171.google.com with SMTP id m13so224797yen.16 for ; Wed, 13 Mar 2013 11:09:22 -0700 (PDT) From: Anthony Liguori In-Reply-To: <20130313175204.GA7894@redhat.com> References: <1363154346-14827-1-git-send-email-qemulist@gmail.com> <51405BEA.6080109@redhat.com> <87hakfiild.fsf@codemonkey.ws> <5140A76E.1080902@redhat.com> <20130313170606.GD4005@vm> <5140B80D.7000301@redhat.com> <20130313175204.GA7894@redhat.com> Date: Wed, 13 Mar 2013 13:09:19 -0500 Message-ID: <871ubj894g.fsf@codemonkey.ws> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Subject: Re: [Qemu-devel] [RFC PATCH 0/2] port network layer onto glib List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Michael S. Tsirkin" , Paolo Bonzini Cc: qemu-devel@nongnu.org, Stefan Hajnoczi , mdroth , Liu Ping Fan "Michael S. Tsirkin" writes: > On Wed, Mar 13, 2013 at 06:31:57PM +0100, Paolo Bonzini wrote: >> > We could do that purely >> > with AioContexts as well, but that rules out a large class of >> > backends that offloaded event loops can interact with, such as Chardevs, >> > so I think modelling how to handle both will provide a threading model >> > that scales better with other devices/subsystems. >> >> .. but I think the "no magic" argument applies here too. After all we >> only have a handful of subsystems. If chardevs are not performance >> critical, they can keep running in the main thread. >> >> If one day we find out that we need a real-time serial port, and glib >> just doesn't cut it, we shouldn't be ashamed of ripping GIOChannels out, >> hand-writing the same stuff, and using a dedicated AioContext. Of >> course by the time we get there we'll have unit tests/qtests to make >> sure we do not regress. Right??? :) >> >> Paolo > > Since you mention serial port, just wanted to say that while it's > bandwidth requirements are not high, we do need to improve its latency. > ATM whenever someone tries to use the emulated serial, guest experiences > stalls and worst case latency jumps, and it's a pain point for many > users. Can you be more specific? I'm not familiar with this issue. Regards, Anthony Liguori > > -- > MST