From mboxrd@z Thu Jan 1 00:00:00 1970 From: Steven Rostedt Subject: Re: [RFC][PATCH v2] sched/rt: Use IPI to trigger RT task push migration instead of pulling Date: Thu, 26 Feb 2015 07:46:09 -0500 Message-ID: <20150226074609.27ccd39d@grimm.local.home> References: <20150224133946.3948c4b7@gandalf.local.home> <20150225103535.GJ5029@twins.programming.kicks-ass.net> <20150225105116.7fa03cc9@gandalf.local.home> <20150225171110.GO21418@twins.programming.kicks-ass.net> <20150225125015.6c5110ca@gandalf.local.home> <20150226074907.GQ21418@twins.programming.kicks-ass.net> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: LKML , Ingo Molnar , Thomas Gleixner , Clark Williams , linux-rt-users , Mike Galbraith , "Paul E. McKenney" , =?UTF-8?B?SsO2cm4=?= Engel To: Peter Zijlstra Return-path: Received: from cdptpa-outbound-snat.email.rr.com ([107.14.166.229]:8943 "EHLO cdptpa-oedge-vip.email.rr.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S932272AbbBZMpR (ORCPT ); Thu, 26 Feb 2015 07:45:17 -0500 In-Reply-To: <20150226074907.GQ21418@twins.programming.kicks-ass.net> Sender: linux-rt-users-owner@vger.kernel.org List-ID: On Thu, 26 Feb 2015 08:49:07 +0100 Peter Zijlstra wrote: > Yes, notice that we don't start iterating at the beginning; this in on > purpose. If we start iterating at the beginning, _every_ cpu will again > pile up on the first one. > > By starting at the current cpu, each cpu will start iteration some place > else and hopefully, with a big enough system, different CPUs end up on a > different rto cpu. Note, v3 doesn't start at the current CPU, it starts at the original CPU again. As the start of every IPI comes from a different CPU, this still keeps a restart from clobbering the same CPU. -- Steve