Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262543AbTIUTeO (ORCPT ); Sun, 21 Sep 2003 15:34:14 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262544AbTIUTeO (ORCPT ); Sun, 21 Sep 2003 15:34:14 -0400 Received: from twilight.ucw.cz ([81.30.235.3]:28835 "EHLO twilight.ucw.cz") by vger.kernel.org with ESMTP id S262543AbTIUTeN (ORCPT ); Sun, 21 Sep 2003 15:34:13 -0400 Date: Sun, 21 Sep 2003 21:34:00 +0200 From: Vojtech Pavlik To: Peter Osterlund Cc: Vojtech Pavlik , Linus Torvalds , linux-kernel@vger.kernel.org, Dmitry Torokhov Subject: Re: Broken synaptics mouse.. Message-ID: <20030921193400.GA22743@ucw.cz> References: <20030919114806.GD784@ucw.cz> <20030921172758.GA21014@ucw.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.4i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1269 Lines: 30 On Sun, Sep 21, 2003 at 09:29:10PM +0200, Peter Osterlund wrote: > OK, below is a new patch that splits the W value as has been suggested > before. The synaptics driver now reports BTN_TOOL_FINGER, > BTN_TOOL_DOUBLETAP and BTN_TOOL_TRIPLETAP for one, two and three > fingers respectively, and it reports ABS_TOOL_WIDTH for the finger > width value. These event types are also used by mousedev.c to decide > if it is dealing with a touchpad. > > It should also gets the direction of the Y axis right. At first glance, patch looks OK. > One thing that it doesn't get right is the handling of invalid ABS_* > values. How is this supposed to be handled? The driver doesn't know > the exact limits for the X/Y values, and discarding values outside > some guessed limits will only have the effect that some parts of the > touchpad area becomes dead. I think something like 'if the finger is lifted so much above surface that X and Y are unreliable, don't report X and Y'. Is that doable? -- Vojtech Pavlik SuSE Labs, SuSE CR - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/