From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751601AbaL3TxF (ORCPT ); Tue, 30 Dec 2014 14:53:05 -0500 Received: from mx1.redhat.com ([209.132.183.28]:55877 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751321AbaL3TxC (ORCPT ); Tue, 30 Dec 2014 14:53:02 -0500 Date: Tue, 30 Dec 2014 17:52:28 -0200 From: Marcelo Tosatti To: Thomas Gleixner Cc: Steven Rostedt , Clark Williams , Luiz Capitulino , Rik van Riel , linux-kernel@vger.kernel.org Subject: [PATCH -rt] timer: upper bound on loops of __run_timers processing Message-ID: <20141230195228.GA3057@amt.cnet> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Commit "timers: do not raise softirq unconditionally", allows for timer wheel processing (__run_timers) to be delayed for long periods of time. The effect is that loops = jiffies - base->timer_jiffies Can grow to very large values resulting in __run_timers taking hundreds of milliseconds to execute. Fix by creating an upper bound on the number of loops to be processed. This allows a nohz=off kernel to achieve desired latencies. Signed-off-by: Marcelo Tosatti diff --git a/kernel/timer.c b/kernel/timer.c index f59e18c..c128416 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -1488,6 +1488,12 @@ void run_local_timers(void) } #endif + if (time_after_eq(jiffies, base->timer_jiffies)) { + unsigned long jiffies_delta = jiffies - base->timer_jiffies; + if (jiffies_delta > TVR_SIZE) + raise_softirq(TIMER_SOFTIRQ); + } + if (!base->active_timers) goto out;