From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nivedita Singhvi Subject: Re: -O option causing test jitter? Date: Tue, 22 Apr 2008 09:05:44 -0700 Message-ID: <480E0CD8.2070400@us.ibm.com> References: <1208476794.6307.15.camel@localhost.localdomain> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: john stultz , linux-rt-users To: Steven Rostedt Return-path: Received: from e4.ny.us.ibm.com ([32.97.182.144]:39009 "EHLO e4.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751007AbYDVQF6 (ORCPT ); Tue, 22 Apr 2008 12:05:58 -0400 Received: from d01relay02.pok.ibm.com (d01relay02.pok.ibm.com [9.56.227.234]) by e4.ny.us.ibm.com (8.13.8/8.13.8) with ESMTP id m3MG5wSj002240 for ; Tue, 22 Apr 2008 12:05:58 -0400 Received: from d01av02.pok.ibm.com (d01av02.pok.ibm.com [9.56.224.216]) by d01relay02.pok.ibm.com (8.13.8/8.13.8/NCO v8.7) with ESMTP id m3MG5v7c217162 for ; Tue, 22 Apr 2008 12:05:57 -0400 Received: from d01av02.pok.ibm.com (loopback [127.0.0.1]) by d01av02.pok.ibm.com (8.12.11.20060308/8.13.3) with ESMTP id m3MG5rGh015875 for ; Tue, 22 Apr 2008 12:05:57 -0400 In-Reply-To: Sender: linux-rt-users-owner@vger.kernel.org List-ID: Steven Rostedt wrote: > > On Thu, 17 Apr 2008, john stultz wrote: >> Anyway, I wanted to hit the list with this so folks could look at it and >> let me know if such severe jitter was expected from just compile >> options? >> > > I think this may simply be a simple case of cache misses. > > Attached is two files: sched_jitter-test.c and do_work.S. I found that by > switching the positions of do_work1 and do_work2, which ever one is first > performs better. do_work1 is me copying the output of objdump of -O1 of > do_work, and do_work2 is copying the output from objdump of -O2. So jitter going down might simply be a case of the faster cases actually slowing down, right - and making things uniformly slow? Do we know if the whole kaboodle is getting faster? thanks, Nivedita