Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757260AbbGGOUM (ORCPT ); Tue, 7 Jul 2015 10:20:12 -0400 Received: from mail-wg0-f43.google.com ([74.125.82.43]:33803 "EHLO mail-wg0-f43.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754787AbbGGOUF (ORCPT ); Tue, 7 Jul 2015 10:20:05 -0400 Date: Tue, 7 Jul 2015 16:20:01 +0200 From: Frederic Weisbecker To: Preeti U Murthy Cc: LKML , Peter Zijlstra , Thomas Gleixner , Christoph Lameter , Ingo Molnar , Viresh Kumar , Rik van Riel Subject: Re: [PATCH 5/8] nohz: Restart the tick from irq exit Message-ID: <20150707142000.GC20498@lerouge> References: <1434044168-23173-1-git-send-email-fweisbec@gmail.com> <1434044168-23173-6-git-send-email-fweisbec@gmail.com> <557D4714.1090600@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <557D4714.1090600@linux.vnet.ibm.com> User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1251 Lines: 24 On Sun, Jun 14, 2015 at 02:49:16PM +0530, Preeti U Murthy wrote: > On 06/11/2015 11:06 PM, Frederic Weisbecker wrote: > > Restart the tick when necessary from the irq exit path. It makes nohz > > full more flexible and allow it to piggyback the tick restart on the > > scheduler IPI in the future instead of sending a dedicated IPI that > > often doubles the scheduler IPI on task wakeup. This will require > > You can piggy back on the scheduler ipi when you add a timer/hrtimer and > add a new task to the runqueue of the nohz_full cpus, since we call > resched_curr() in these code paths. But what about the calls to kick > nohz_full cpus by perf events and posix cpu timers ? These call sites > seem to be concerned about specifically waking up nohz_full cpus as far > as I can see. IOW there is no scheduling ipi that we can fall back on in > these paths. Sure, those will need to keep the current IPI. They are less of a worry because they should be rare events compared to the scheduler. Thanks. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/