Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755195AbXEXVWd (ORCPT ); Thu, 24 May 2007 17:22:33 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753463AbXEXVWL (ORCPT ); Thu, 24 May 2007 17:22:11 -0400 Received: from smtp-outbound-1.vmware.com ([65.113.40.141]:40509 "EHLO smtp-outbound-1.vmware.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753237AbXEXVWK (ORCPT ); Thu, 24 May 2007 17:22:10 -0400 Message-ID: <465600A5.3040806@vmware.com> Date: Thu, 24 May 2007 14:16:21 -0700 From: Zachary Amsden User-Agent: Thunderbird 1.5.0.10 (X11/20070221) MIME-Version: 1.0 To: Andrew Morton CC: Ben Collins , linux-kernel@vger.kernel.org Subject: Re: [PATCH 4/6] RTC: Ratelimit "lost interrupts" message References: <98cd38b5.git.bcollins@ubuntu.com> <20070524134822.ed020127.akpm@linux-foundation.org> <1180040212.12570.279.camel@cunning> <20070524140956.12ff7c99.akpm@linux-foundation.org> In-Reply-To: <20070524140956.12ff7c99.akpm@linux-foundation.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1411 Lines: 36 Andrew Morton wrote: > >>> Presumably someone is getting a lot of these messages. >>> >>> Do we know why? Is there something which needs fixing here? >>> >> I got it a lot when using any sort of virtualization on my Core2Duo >> (kvm, vmware, qemu). Just a big spew. Honestly, I was getting this with >> 2.6.20, and haven't tested again with latest kernel. >> > > OK, a bit of googling finds things like this: > http://chxo.com/be2/20060821_3333.html, so I assume that it's some sort of > generic problem with the hypervisor's emulation of the RTC. > > Zach, do you know what the score is here? > We at least jack up the rtc for fast clock timeouts with better granularity - so it can be a host kernel issue, although a guest kernel trying to get high frequency RTC will also be inaccurate, and inevitably will have unhidable interrupt lateness. Ratelimiting seems the right thing to do, as once you have noted late RTC interrupts, it doesn't need to be noted again, and there is nothing you can do to fix it. It is useful as a warning about interrupt disabled handlers taking too long, but there are other ways of detecting that too. Zach - 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/