From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965530AbXAWWEu (ORCPT ); Tue, 23 Jan 2007 17:04:50 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S965525AbXAWWEu (ORCPT ); Tue, 23 Jan 2007 17:04:50 -0500 Received: from www.osadl.org ([213.239.205.134]:35934 "EHLO mail.tglx.de" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S965520AbXAWWBW (ORCPT ); Tue, 23 Jan 2007 17:01:22 -0500 Message-Id: <20070123211209.237368000@localhost.localdomain> References: <20070123211159.178138000@localhost.localdomain> Date: Tue, 23 Jan 2007 22:01:37 -0000 From: Thomas Gleixner To: Andrew Morton Cc: LKML , Ingo Molnar , John Stultz , Arjan van de Veen , Roman Zippel Subject: [patch 40/46] i386 prepare nmi watchdog for dynticks Content-Disposition: inline; filename=dynticks-i386-prepare-nmi-watchdog.patch Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org From: Thomas Gleixner The NMI watchdog implementation assumes that the local APIC timer interrupt is happening. This assumption is not longer true when high resolution timers and dynamic ticks come into play, as they may switch off the local APIC timer completely. Take the PIT/HPET interrupts into account too, to avoid false positives. Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar Cc: Andi Kleen Cc: Zachary Amsden Cc: "Eric W. Biederman" Cc: Rohit Seth Signed-off-by: Andrew Morton --- arch/i386/kernel/nmi.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) Index: linux-2.6.20-rc4-mm1-bo/arch/i386/kernel/nmi.c =================================================================== --- linux-2.6.20-rc4-mm1-bo.orig/arch/i386/kernel/nmi.c +++ linux-2.6.20-rc4-mm1-bo/arch/i386/kernel/nmi.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include @@ -971,9 +972,13 @@ __kprobes int nmi_watchdog_tick(struct p cpu_clear(cpu, backtrace_mask); } - sum = per_cpu(irq_stat, cpu).apic_timer_irqs; + /* + * Take the local apic timer and PIT/HPET into account. We don't + * know which one is active, when we have highres/dyntick on + */ + sum = per_cpu(irq_stat, cpu).apic_timer_irqs + kstat_irqs(0); - /* if the apic timer isn't firing, this cpu isn't doing much */ + /* if the none of the timers isn't firing, this cpu isn't doing much */ if (!touched && last_irq_sums[cpu] == sum) { /* * Ayiee, looks like this CPU is stuck ... --