From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:33282) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dsrd6-0005oC-9q for qemu-devel@nongnu.org; Fri, 15 Sep 2017 10:30:06 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dsrd2-0001HQ-8r for qemu-devel@nongnu.org; Fri, 15 Sep 2017 10:30:04 -0400 Received: from mx1.redhat.com ([209.132.183.28]:49770) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dsrd1-0001H6-VY for qemu-devel@nongnu.org; Fri, 15 Sep 2017 10:30:00 -0400 Date: Fri, 15 Sep 2017 15:29:46 +0100 From: "Dr. David Alan Gilbert" Message-ID: <20170915142945.GP2170@work-vm> References: <1505375436-28439-1-git-send-email-peterx@redhat.com> <20170914151911.GB7370@stefanha-x1.localdomain> <20170915035057.GM3617@pxdev.xzpeter.org> <20170915104926.GA14994@stefanha-x1.localdomain> <20170915113428.GF13610@redhat.com> <20170915120643.GN2170@work-vm> <20170915121433.GI13610@redhat.com> <20170915121956.GO2170@work-vm> <20170915122913.GJ13610@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline In-Reply-To: <20170915122913.GJ13610@redhat.com> Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [RFC 00/15] QMP: out-of-band (OOB) execution support List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Daniel P. Berrange" Cc: Stefan Hajnoczi , Peter Xu , =?iso-8859-1?Q?Marc-Andr=E9?= Lureau , QEMU , Paolo Bonzini , Stefan Hajnoczi , Fam Zheng , Juan Quintela , Michael Roth , Eric Blake , Laurent Vivier , Markus Armbruster * Daniel P. Berrange (berrange@redhat.com) wrote: > On Fri, Sep 15, 2017 at 01:19:56PM +0100, Dr. David Alan Gilbert wrote: > > * Daniel P. Berrange (berrange@redhat.com) wrote: > > > On Fri, Sep 15, 2017 at 01:06:44PM +0100, Dr. David Alan Gilbert wr= ote: > > > > * Daniel P. Berrange (berrange@redhat.com) wrote: > > > > > On Fri, Sep 15, 2017 at 11:49:26AM +0100, Stefan Hajnoczi wrote= : > > > > > > On Fri, Sep 15, 2017 at 11:50:57AM +0800, Peter Xu wrote: > > > > > > > On Thu, Sep 14, 2017 at 04:19:11PM +0100, Stefan Hajnoczi w= rote: > > > > > > > > On Thu, Sep 14, 2017 at 01:15:09PM +0200, Marc-Andr=E9 Lu= reau wrote: > > > > > > > > > There should be a limit in the number of requests the t= hread can > > > > > > > > > queue. Before the patch, the limit was enforced by syst= em socket > > > > > > > > > buffering I think. Now, should oob commands still be pr= ocessed even if > > > > > > > > > the queue is full? If so, the thread can't be suspended= . > > > > > > > >=20 > > > > > > > > I agree. > > > > > > > >=20 > > > > > > > > Memory usage must be bounded. The number of requests is = less important > > > > > > > > than the amount of memory consumed by them. > > > > > > > >=20 > > > > > > > > Existing QMP clients that send multiple QMP commands with= out waiting for > > > > > > > > replies need to rethink their strategy because OOB comman= ds cannot be > > > > > > > > processed if queued non-OOB commands consume too much mem= ory. > > > > > > >=20 > > > > > > > Thanks for pointing out this. Yes the memory usage problem= is valid, > > > > > > > as Markus pointed out as well in previous discussions (in "= Flow > > > > > > > Control" section of that long reply). Hopefully this serie= s basically > > > > > > > can work from design prospective, then I'll add this flow c= ontrol in > > > > > > > next version. > > > > > > >=20 > > > > > > > Regarding to what we should do if the limit is reached: Mar= kus > > > > > > > provided a few options, but the one I prefer most is that w= e don't > > > > > > > respond, but send an event showing that a command is droppe= d. > > > > > > > However, I would like it not queued, but a direct reply (af= ter all, > > > > > > > it's an event, and we should not need to care much on order= ing of it). > > > > > > > Then we can get rid of the babysitting of those "to be fail= ed" > > > > > > > requests asap, meanwhile we don't lose anything IMHO. > > > > > > >=20 > > > > > > > I think I also missed at least a unit test for this new int= erface. > > > > > > > Again, I'll add it after the whole idea is proved solid. T= hanks, > > > > > >=20 > > > > > > Another solution: the server reports available receive buffer= space to > > > > > > the client. The server only guarantees immediate OOB process= ing when > > > > > > the client stays within the receive buffer size. > > > > > >=20 > > > > > > Clients wishing to take advantage of OOB must query the recei= ve buffer > > > > > > size and make sure to leave enough room. > > > > >=20 > > > > > I don't think having to query it ahead of time is particularly = nice, > > > > > and of course it is inherantly racy. > > > > >=20 > > > > > I would just have QEMU emit an event when it pausing processing= of the > > > > > incoming commands due to a full queue. If the event includes t= he ID > > > > > of the last queued command, the client will know which (if any)= of > > > > > its outstanding commands are delayed. Another even can be sent = when > > > > > it restarts reading. > > > >=20 > > > > Hmm and now we're implementing flow control! > > > >=20 > > > > a) What exactly is the current semantics/buffer sizes? > > > > b) When do clients send multiple QMP commands on one channel with= out > > > > waiting for the response to the previous command? > > > > c) Would one queue entry for each class of commands/channel work > > > > (Where a class of commands is currently 'normal' and 'oob') > > >=20 > > > I do wonder if we need to worry about request limiting at all from = the > > > client side. For non-OOB commands clients will wait for a reply be= fore > > > sending a 2nd non-OOB command, so you'll never get a deep queue for= . > > >=20 > > > OOB commands are supposed to be things which can be handled quickly > > > without blocking, so even if a client sent several commands at once > > > without waiting for replies, they're going to be processed quickly, > > > so whether we temporarily block reading off the wire is a minor > > > detail. > >=20 > > Lets just define it so that it can't - you send an OOB command and wa= it > > for it's response before sending another on that channel. > >=20 > > > IOW, I think we could just have a fixed 10 command queue and apps j= ust > > > pretend that there's an infinite queue and nothing bad would happen= from > > > the app's POV. > >=20 > > Can you justify 10 as opposed to 1? >=20 > Semantically I don't think it makes a difference if the OOB commands ar= e > being processed sequentially by their thread. A >1 length queue would o= nly > matter for non-OOB commands if an app was filling the pipeline with non= -OOB > requests, as then that could block reading of OOB commands.=20 But can't we just tell the app not to? Dave >=20 > Regards, > Daniel > --=20 > |: https://berrange.com -o- https://www.flickr.com/photos/dberr= ange :| > |: https://libvirt.org -o- https://fstop138.berrange= .com :| > |: https://entangle-photo.org -o- https://www.instagram.com/dberr= ange :| -- Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK