From owner-freebsd-stable@FreeBSD.ORG Thu Feb 26 16:31:42 2009 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 7FDA61065672 for ; Thu, 26 Feb 2009 16:31:42 +0000 (UTC) (envelope-from avg@icyb.net.ua) Received: from citadel.icyb.net.ua (citadel.icyb.net.ua [212.40.38.140]) by mx1.freebsd.org (Postfix) with ESMTP id B070F8FC0C for ; Thu, 26 Feb 2009 16:31:41 +0000 (UTC) (envelope-from avg@icyb.net.ua) Received: from odyssey.starpoint.kiev.ua (alpha-e.starpoint.kiev.ua [212.40.38.101]) by citadel.icyb.net.ua (8.8.8p3/ICyb-2.3exp) with ESMTP id SAA07408; Thu, 26 Feb 2009 18:31:38 +0200 (EET) (envelope-from avg@icyb.net.ua) Message-ID: <49A6C3EA.6060605@icyb.net.ua> Date: Thu, 26 Feb 2009 18:31:38 +0200 From: Andriy Gapon User-Agent: Thunderbird 2.0.0.19 (X11/20090110) MIME-Version: 1.0 To: "Paul B. Mahol" References: <49A6BFD1.6090601@icyb.net.ua> <3a142e750902260827w722f110eve23a7bdb07cfbe42@mail.gmail.com> In-Reply-To: <3a142e750902260827w722f110eve23a7bdb07cfbe42@mail.gmail.com> X-Enigmail-Version: 0.95.7 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: freebsd-stable@freebsd.org Subject: Re: qemu+aio vs C2 X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 26 Feb 2009 16:31:43 -0000 on 26/02/2009 18:27 Paul B. Mahol said the following: > On 2/26/09, Andriy Gapon wrote: >> I see something unusual and surprising for me: if I kldload aio for qemu's >> sake >> and then actually start qemu, I see that share of C2 in cx_usage is >> constantly >> dropping and then finally there is "too many short sleeps, backing off to >> C1". >> >> This is on i386 with stable/7 as of r188116. >> >> I am out of ideas. > > And qemu guest is? > > Perhaps you should try changing guest kern.hz sysctl setting. > I dont think related comitt got MFC-ed to STABLE. > For what I've been doing the guest was our boot code and loader. Nothing beyond that. And, pardon me, how guest software could affect host's hardware in this way? As i understand only a hardware even of some sort could interrupt C2. -- Andriy Gapon