From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262973AbVGHXRQ (ORCPT ); Fri, 8 Jul 2005 19:17:16 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262964AbVGHXQ7 (ORCPT ); Fri, 8 Jul 2005 19:16:59 -0400 Received: from dvhart.com ([64.146.134.43]:36021 "EHLO localhost.localdomain") by vger.kernel.org with ESMTP id S262955AbVGHXPC (ORCPT ); Fri, 8 Jul 2005 19:15:02 -0400 Date: Fri, 08 Jul 2005 16:14:59 -0700 From: "Martin J. Bligh" Reply-To: "Martin J. Bligh" To: Chris Wedgwood Cc: Andrew Morton , linux-kernel@vger.kernel.org, torvalds@osdl.org Subject: Re: [PATCH] i386: Selectable Frequency of the Timer Interrupt Message-ID: <136640000.1120864499@flay> In-Reply-To: <20050708230303.GA19188@taniwha.stupidest.org> References: <20050708145953.0b2d8030.akpm@osdl.org> <133660000.1120863575@flay> <20050708230303.GA19188@taniwha.stupidest.org> X-Mailer: Mulberry/2.1.2 (Linux/x86) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org --On Friday, July 08, 2005 16:03:03 -0700 Chris Wedgwood wrote: > On Fri, Jul 08, 2005 at 03:59:35PM -0700, Martin J. Bligh wrote: > >> I think we're talking between 2.6.12-git5 and 2.6.12-git6 right? I >> can confirm more explicitly if really need be. 48s -> 45.5s elapsed. > > That's a huge difference (5%) --- what hardware is that on? 16 CPU x440 (about a 2-3 year old NUMA box). flat 4x seems about a 1% gain, still statistically significant, but definitely smaller. I'm not saying there isn't data supporting higher HZ ... I just haven't seen it published. I get the feeling what people really want is high-res timers anyway ... high HZ is just concealing the issue and making it slightly less crap, not actually fixing it. M.