From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:50372) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1X2xaS-0005Ry-Ha for qemu-devel@nongnu.org; Fri, 04 Jul 2014 03:07:18 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1X2xaM-0002oH-Cu for qemu-devel@nongnu.org; Fri, 04 Jul 2014 03:07:12 -0400 Received: from mx1.redhat.com ([209.132.183.28]:57055) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1X2xaM-0002oD-4W for qemu-devel@nongnu.org; Fri, 04 Jul 2014 03:07:06 -0400 Message-ID: <53B65290.8070207@redhat.com> Date: Fri, 04 Jul 2014 09:06:56 +0200 From: Paolo Bonzini MIME-Version: 1.0 References: <1404395521-11158-1-git-send-email-stefanha@redhat.com> <53B64A5B.2020204@redhat.com> In-Reply-To: Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH 0/2] coroutine: dynamically scale pool size List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Ming Lei Cc: Kevin Wolf , qemu-devel , Stefan Hajnoczi Il 04/07/2014 09:02, Ming Lei ha scritto: >> > So the excessive writes (to eventfd and irqfd) are not a problem? That'd be >> > a relief. :) > I mean it is in a level, but your aio_notify() patch still can improve virtioblk > dataplane performance some, in my test, with 5~10K IOPS improvement, > which should belong to same level, because my test is a bit coarse, :-) > > IMO not only virtioblk can benefit from your patch, others can benefit too. Definitely, but it's not the kind of patch I enjoy including 3 weeks from release. I have already broken enough eggs during the 2.1 development period. :) Paolo