From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: Re: [PATCH cgroup/for-3.11 1/3] cgroup: mark "tasks" cgroup file as insane Date: Fri, 7 Jun 2013 13:05:19 -0700 Message-ID: <20130607200519.GB14781@mtj.dyndns.org> References: <20130604021302.GH29989@mtj.dyndns.org> <20130604111556.GA4963@redhat.com> <20130604201947.GE14916@htj.dyndns.org> <20130606092055.GF30217@redhat.com> <20130606211410.GF5045@htj.dyndns.org> <20130607051040.GA1364@tango.0pointer.de> <20130607093050.GA10742@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=sender:date:from:to:cc:subject:message-id:references:mime-version :content-type:content-disposition:in-reply-to:user-agent; bh=fb87XNiWsRgPhr/lT/2bmEEX7q5xN4QzY4PJ7hSkhNk=; b=U83mV/5royL8AQMCuX8AL65hGmZcge8r4ljOfQniujinXoyzdeRl/ap3KXqtHdtvQY g/6DtdwRcjEGuT/ZPoJZu9C615sySLstw5vka9r+XGor8YdicwgFzYi5HIzF8Sz5ywau 8pvRqOGlgmTIEIGGzzFt8TVZf1o4oQAjj/jVb78f2sSkiyjgy1gPjL2E0SS8tiKcEsd/ 5ZuKJhf9qMXAQahhG7rsQjeOE/24Vnns21tMVgl8xQg0T1+yc4WuuLXDjrdQ7FQzJSBC /mljKQokmDwOpMR6yuxV72u6BUewIzaDTrnV8rT6IrRIKJPL9+Kt5FuAU/bcIqsKrRjH mAIg== Content-Disposition: inline In-Reply-To: <20130607093050.GA10742-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: containers-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org Errors-To: containers-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org To: "Daniel P. Berrange" Cc: containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org, kay.sievers-tD+1rO4QERM@public.gmane.org, Michal Hocko , Lennart Poettering , Johannes Weiner , cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Vivek Goyal Hello, Daniel. On Fri, Jun 07, 2013 at 10:30:50AM +0100, Daniel P. Berrange wrote: > > Uhm. So I don't think we will support two ways to set this up for long > > in systemd (if at all). And even if we did, the distributions would pick > > one or the other as default, and if you are unlucky, then you couldn't > > run libvirt on it without reconfiguration and rebooting to get the other > > cgroup setup logic... > > Yep, that's exactly what concerns me. If systemd introduced the option I don't think you two are speaking about the same problem. Lennart is saying that backward compat option won't stay for long and you're worrying about the method to enable compatibility mode. > for a new style setup, then it would almost certainly be adopted by > default by Fedora (otherwise there's no point adding it). At which > point libvirt is doomed. We can't require people to reconfigure systemd > and reboot to use virt. Yeah, it's cumbersome, but if it's pressing we can make the release of controllers from unified hierarchy dynamic I suppose. I'll think more about it. Thanks. -- tejun