Received: from eggs.gnu.org ([2001:4830:134:3::10]:56137) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VfqqR-0005XU-RC for qemu-devel@nongnu.org; Mon, 11 Nov 2013 07:44:01 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1VfqqL-0005N0-RM for qemu-devel@nongnu.org; Mon, 11 Nov 2013 07:43:55 -0500 Received: from mx1.redhat.com ([209.132.183.28]:9107) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VfqqL-0005Mw-HK for qemu-devel@nongnu.org; Mon, 11 Nov 2013 07:43:49 -0500 Date: Mon, 11 Nov 2013 13:43:29 +0100 From: Stefan Hajnoczi Message-ID: <20131111124329.GA1036@stefanha-thinkpad.redhat.com> References: <1383560924-15788-1-git-send-email-matthias.bgg@gmail.com> <20131105132509.GC16457@stefanha-thinkpad.redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Subject: Re: [Qemu-devel] [PATCH v2 0/3] Make thread pool implementation modular List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Matthias Brugger Cc: Kevin Wolf , Liu Ping Fan , Anthony Liguori , Stefan Hajnoczi , Jeff Cody , Michael Tokarev , qemu-devel@nongnu.org, Markus Armbruster , malc , Alex Bligh , Stefan Weil , Paolo Bonzini , Asias He , Luiz Capitulino , Andreas =?iso-8859-1?Q?F=E4rber?= , Eduardo Habkost On Mon, Nov 11, 2013 at 11:00:45AM +0100, Matthias Brugger wrote: > 2013/11/5 Stefan Hajnoczi : > > I'd also like to see the thread pool implementation you wish to add > > before we add a layer of indirection which has no users yet. > > Fair enough, I will evaluate if it will make more sense to implement a > new AIO infrastructure instead to try reuse the thread-pool. > Actually my implementation will differ in the way, that we will have > several workerthreads with everyone of them having its own queue. The > requests will be distributed between them depending on an identifier. > The request function which the worker_thread call will be the same as > using aio=threads, so I'm not quite sure which will be the way to go. > Any opinions and hints, like the one you gave are highly appreciated. If I understand the slides you linked to correctly, the guest will pass an identifier with each request. The host has worker threads allowing each stream of requests to be serviced independently. The idea is to associate guest processes with unique identifiers. The guest I/O scheduler is supposed to submit requests in a way that meets certain policies (e.g. fairness between processes, deadlines, etc). Why is it necessary to push this task down into the host? I don't understand the advantage of this approach except that maybe it works around certain misconfigurations, I/O scheduler quirks, or plain old bugs - all of which should be investigated and fixed at the source instead of adding another layer of code to mask them. Stefan