Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262933AbUKYCxK (ORCPT ); Wed, 24 Nov 2004 21:53:10 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262934AbUKYCxJ (ORCPT ); Wed, 24 Nov 2004 21:53:09 -0500 Received: from brown.brainfood.com ([146.82.138.61]:7315 "EHLO gradall.private.brainfood.com") by vger.kernel.org with ESMTP id S262933AbUKYCxC (ORCPT ); Wed, 24 Nov 2004 21:53:02 -0500 Date: Wed, 24 Nov 2004 03:00:37 -0600 (CST) From: Adam Heath X-X-Sender: adam@gradall.private.brainfood.com To: Ingo Molnar cc: linux-kernel@vger.kernel.org Subject: Re: [patch] Real-Time Preemption, -RT-2.6.10-rc2-mm2-V0.7.30-2 In-Reply-To: <20041124040604.GA13340@elte.hu> Message-ID: References: <20041123115201.GA26714@elte.hu> <20041124040604.GA13340@elte.hu> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2049 Lines: 47 On Wed, 24 Nov 2004, Ingo Molnar wrote: > > * Adam Heath wrote: > > > > > I'm seeing something very odd. It's against 29-0. I also seem to > > > > recall seeing something similiar reported earlier. > > > > > > > > I'm seeing pauses on my system. Not certain what is causing it. > > > > Hitting a key on the keyboard unsticks it. > > > > > > at first sight this looks like a scheduling/wakeup anomaly. Please > > > re-report this if it happens with the current (30-4) kernel too. Also, > > > could you test the vanilla -mm tree, it has a few scheduler updates too. > > > > 2.6.10-rc1-mm3 doesn't have the same problem. Didn't have a more > > recent mm kernel available last night. Will compile one, and always > > keep it available. > > -rc2-mm2 would be nice to test - there are a number of new interactivity > fixes from Con being test-driven in -mm right now. In particular, these > patches were added in -rc1-mm4. These are the patches in question: > > sched-adjust_timeslice_granularity.patch > requeue_granularity.patch > sched-remove_interactive_credit.patch > > you can download them individually from: > > http://kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.10-rc2/2.6.10-rc2-mm2/broken-out/ > > so if these symptoms still occur with vanilla -rc2-mm2, could you try to > unapply them, in reverse order? (there might be rejects when you try > that, due to patch dependencies - let me know if it doesnt work out and > i'll do an undo patch.) The symptoms still occur with 30-9. I'll be trying rc2-mm2 over the holiday. Came in this morning, and after hitting a key, my machine said it was 2:38am, when it was actually 11:10am. All internet connections had died(obviously). But the machine started working fine once I hit that key. No messages in dmesg. - 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/