From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756923Ab1JRHOx (ORCPT ); Tue, 18 Oct 2011 03:14:53 -0400 Received: from mail-qy0-f181.google.com ([209.85.216.181]:53932 "EHLO mail-qy0-f181.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753889Ab1JRHOw convert rfc822-to-8bit (ORCPT ); Tue, 18 Oct 2011 03:14:52 -0400 MIME-Version: 1.0 In-Reply-To: <1318872925.4172.81.camel@twins> References: <1318872925.4172.81.camel@twins> Date: Tue, 18 Oct 2011 09:14:51 +0200 Message-ID: Subject: Re: Scheduler BOF @ LinuxCon.EU/ELCE From: Vincent Guittot To: Peter Zijlstra Cc: linux-kernel , Ingo Molnar , Thomas Gleixner , Mike Galbraith , Paul Turner , Suresh Siddha , rostedt , Srivatsa Vaddagiri , Oleg Nesterov , Venkatesh Pallipadi , Glauber Costa , Arjan van de Ven Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 17 October 2011 19:35, Peter Zijlstra wrote: > Hi, > > several people have expressed interest of talking to me about the > scheduler during the RTLWS/KS/LinuxCon.EU time-frame. Since I've just > about lost track of who all wanted to talk about what and who can and > can not make it out to Prague I need help ;-) > I will attend the ELCE and I'm interested by scheduler discussion > Also, since I'm too dis-organized (and mostly busy) to really arrange > something, can someone volunteer to arrange a BOF or cafe with beer to > discuss things? > > The topics I'm most interested in are: > >  - power optimized load-balancing; >    do packing based on cpu utilization >    fix the current interface of sched_{mc,ht}_power_savings I'm also interested > >  - integration of the cpufreq/idle governors and the scheduler >    use the idle guestimate for NO_HZ >    use $foo for $prev-topic to reduce double accounting etc. I'm also interested > >  - what can we do about the cgroup crap and how can we win some >   of the performance back; >    start with fixing the cgroup enabled but unused case >    continue from there.. > >    pjt was rewriting load tracking, do tell us about it >        also tell us what the status of ncrao's hackery is, >        or does your hackery replace that? > > Things I specifically do not want to talk about: > >  - systemd, its shit, go away. >