From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756127AbXLBUfx (ORCPT ); Sun, 2 Dec 2007 15:35:53 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754749AbXLBUfn (ORCPT ); Sun, 2 Dec 2007 15:35:43 -0500 Received: from lazybastard.de ([212.112.238.170]:49316 "EHLO longford.lazybastard.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754668AbXLBUfn convert rfc822-to-8bit (ORCPT ); Sun, 2 Dec 2007 15:35:43 -0500 Date: Sun, 2 Dec 2007 21:30:16 +0100 From: =?utf-8?B?SsO2cm4=?= Engel To: Ingo Molnar Cc: =?utf-8?B?SsO2cm4=?= Engel , Mark Lord , Pavel Machek , Mark Lord , Thomas Gleixner , len.brown@intel.com, Andrew Morton , linux-kernel@vger.kernel.org, linux-pm@lists.linux-foundation.org, rjw@sisk.pl Subject: Re: [BUG] Strange 1-second pauses during Resume-from-RAM Message-ID: <20071202203016.GC10657@lazybastard.org> References: <20071201204747.GA6458@lazybastard.org> <20071201205456.GA8231@elte.hu> <20071201234112.GA7209@lazybastard.org> <20071202085607.GA28966@elte.hu> <20071202113143.GA9378@lazybastard.org> <20071202135711.GA20434@elte.hu> <20071202141117.GB9516@lazybastard.org> <20071202154746.GA27069@elte.hu> <20071202195501.GB10657@lazybastard.org> <20071202200722.GA12101@elte.hu> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8BIT In-Reply-To: <20071202200722.GA12101@elte.hu> User-Agent: Mutt/1.5.9i Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sun, 2 December 2007 21:07:22 +0100, Ingo Molnar wrote: > * Jörn Engel wrote: > > > Result looked like a livelock and finally convinced me to abandon the > > latency tracer. Sorry, but it appears to be the right tool for the > > wrong job. > > hm, we routinely use it in -rt to capture "what on earth is happening" > incidents. The snippet below is a random snipped from a trace that i've > just captured, with mcount enabled. It seems to work fine here, with and > without mcount. (pit clocksource is almost never used, that's why you > had those early problems.) > > oprofile helps if you can reliably reproduce the slowdown in a loop or > for a long amount of time, with lots of CPU utilization - and then it's > also lower overhead. The tracer can be used to capture rare or complex > events, and gives the full flow control and what is happening within the > kernel. Such a trace would be useful indeed. But so far the patch has only given me grief and nothing remotely like useful output. Maybe I should simply use the complete -rt patch instead of debugging the broken-out latency-tracer patch. Jörn -- Mundie uses a textbook tactic of manipulation: start with some reasonable talk, and lead the audience to an unreasonable conclusion. -- Bruce Perens