Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751359AbWELTQW (ORCPT ); Fri, 12 May 2006 15:16:22 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751360AbWELTQW (ORCPT ); Fri, 12 May 2006 15:16:22 -0400 Received: from 216-54-166-5.static.twtelecom.net ([216.54.166.5]:7564 "EHLO mx1.compro.net") by vger.kernel.org with ESMTP id S1751359AbWELTQV (ORCPT ); Fri, 12 May 2006 15:16:21 -0400 Message-ID: <4464DF00.1050501@compro.net> Date: Fri, 12 May 2006 15:16:16 -0400 From: Mark Hounschell Reply-To: markh@compro.net Organization: Compro Computer Svcs. User-Agent: Thunderbird 1.5 (X11/20060111) MIME-Version: 1.0 To: john stultz Cc: Ingo Molnar , lkml , Thomas Gleixner , Steven Rostedt Subject: Re: [RFC][PATCH -rt] irqd starvation on SMP by a single process? References: <1147401812.1907.14.camel@cog.beaverton.ibm.com> <20060512055025.GA25824@elte.hu> <1147457058.9343.22.camel@localhost.localdomain> In-Reply-To: <1147457058.9343.22.camel@localhost.localdomain> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1273 Lines: 48 john stultz wrote: > --- 2.6-rt/kernel/irq/manage.c 2006-05-11 18:37:36.000000000 -0500 > +++ dev-rt/kernel/irq/manage.c 2006-05-12 12:55:56.000000000 -0500 > @@ -724,6 +724,7 @@ > set_current_state(TASK_INTERRUPTIBLE); > do_hardirq(desc); > cond_resched_all(); > + local_irq_disable(); > __do_softirq(); > // do_softirq_from_hardirq(); > local_irq_enable(); > @@ -731,10 +732,8 @@ > /* > * Did IRQ affinities change? > */ > - if (!cpu_isset(smp_processor_id(), irq_affinity[irq])) { > - mask = cpumask_of_cpu(any_online_cpu(irq_affinity[irq])); > - set_cpus_allowed(current, mask); > - } > + if(!cpus_equal(current->cpus_allowed, irq_affinity[irq])) > + set_cpus_allowed(current, irq_affinity[irq]); > #endif > schedule(); > } > > > > Steve, This patch seems to have made a huge difference in my 'complete preemption' mode 'pauses' problem. >From many seconds, even minutes at a time, to 5 seconds and less when it even does happen??? It does still happen however. Mark - 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/