From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S272638AbTG1DoK (ORCPT ); Sun, 27 Jul 2003 23:44:10 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S272639AbTG1DoK (ORCPT ); Sun, 27 Jul 2003 23:44:10 -0400 Received: from kweetal.tue.nl ([131.155.3.6]:25861 "EHLO kweetal.tue.nl") by vger.kernel.org with ESMTP id S272638AbTG1DoH (ORCPT ); Sun, 27 Jul 2003 23:44:07 -0400 Date: Mon, 28 Jul 2003 05:59:20 +0200 From: Andries Brouwer To: dean gaudet Cc: linux-kernel@vger.kernel.org Subject: Re: 2.6.0-test2 has i8042 mux problems Message-ID: <20030728035920.GA1660@win.tue.nl> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.3.25i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sun, Jul 27, 2003 at 07:13:26PM -0700, dean gaudet wrote: > this was a bug i reported back a few versions as well -- and i don't think > i received any responses (nor from the maintainer). > > i've got a box on which 2.4.x works fine, but 2.6.0-test2 gets into a snit > when it's trying to initialize the i8042. i can get 2.6.0-test2 to boot > if i add "i8042_nomux". > > the mux initialization code seems kind of ... wonk -- it seems to write > values to the registers then read back and if the value is the same then > it assumes the mux is there. that seems way too likely to be broken in > situations when the mux isn't there... it'd be better to be looking for > some value which is different after writing. No, it writes f0, 56, a4 and if it gets f0, 56, and not a4, then it assumes there is a mux. [This is a Synaptics convention.] What hardware do you have? And what is the conversation?