From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261664AbTJ1T75 (ORCPT ); Tue, 28 Oct 2003 14:59:57 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261670AbTJ1T75 (ORCPT ); Tue, 28 Oct 2003 14:59:57 -0500 Received: from hera.kernel.org ([63.209.29.2]:36745 "EHLO hera.kernel.org") by vger.kernel.org with ESMTP id S261664AbTJ1T74 (ORCPT ); Tue, 28 Oct 2003 14:59:56 -0500 To: linux-kernel@vger.kernel.org From: Stephen Hemminger Subject: Re: gettimeofday resolution seriously degraded in test9 Date: Tue, 28 Oct 2003 11:59:17 -0800 Organization: Open Source Development Lab Message-ID: <20031028115917.124a76bf.shemminger@osdl.org> References: Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Trace: build.pdx.osdl.net 1067371143 27928 172.20.1.60 (28 Oct 2003 19:59:03 GMT) X-Complaints-To: abuse@osdl.org NNTP-Posting-Date: Tue, 28 Oct 2003 19:59:03 +0000 (UTC) X-Newsreader: Sylpheed version 0.9.6claws (GTK+ 1.2.10; i686-pc-linux-gnu) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On 28 Oct 2003 11:19:21 -0800 David Mosberger-Tang wrote: > >>>>> On Tue, 28 Oct 2003 19:30:13 +0100, Stephen Hemminger said: > > Stephen> This should work better. Patch against 2.6.0-test9 > > Why not use the time-interpolator interface defined in timex.h? It > should handle such things without any special hacks. > > --david Because it has not been used yet outside of ia64. It would be worth investigating post 2.6.0 if it could be shown to be as fast and more correct. Several people have talked about redoing the existing mess, but now is not the time to attack this dragon...