Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932206Ab0LRUou (ORCPT ); Sat, 18 Dec 2010 15:44:50 -0500 Received: from mail-yw0-f46.google.com ([209.85.213.46]:46307 "EHLO mail-yw0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932108Ab0LRUos convert rfc822-to-8bit (ORCPT ); Sat, 18 Dec 2010 15:44:48 -0500 MIME-Version: 1.0 In-Reply-To: <20101218175408.GA1709@polaris.bitmath.org> References: <1292683981-6908-1-git-send-email-rydberg@euromail.se> <1292683981-6908-2-git-send-email-rydberg@euromail.se> <20101218175408.GA1709@polaris.bitmath.org> Date: Sat, 18 Dec 2010 14:44:47 -0600 Message-ID: Subject: Re: [PATCH 1/4] Input: synaptics - add multitouch packet support From: Chris Bagwell To: Henrik Rydberg Cc: Dmitry Torokhov , Jiri Kosina , Chase Douglas , linux-input@vger.kernel.org, linux-kernel@vger.kernel.org, Takashi Iwai Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2175 Lines: 51 On Sat, Dec 18, 2010 at 11:54 AM, Henrik Rydberg wrote: > Hi Chris, > > Thanks for you comments. > >> > While in MT mode and two or three fingers are touching, two sets of >> > data are sent. ?The first is a new format buffer with lower resolution >> > reporting of stationary finger and the second is standard data format >> > reporting movement. >> >> We should probably update this comment and remove parts that mention >> stationary and movement. ?It appears to be doing finger tracking and >> first packet is always first touch... except we are getting reports >> that clickpads/integrated button pads may not be consistently like >> this. > > Yes, good point. Will update. > >> This tracking is partly why type B protocol is being used in later patches. > > The original thought was that the two points corresponded well to the > two fingers, only sometimes in the wrong order. The solution then was > to use the tracking module to order them appropriately. As it turned > out that the fingers are not followed that well after all, the whole > idea fell flat, so we resorted to only sending the bounding rectangle, > which requires no tracking. > > All in all, this patchset contains enough information to resolve > jumping and clickpad issues in user space, at the same time as it > keeps compatibility with mousedev and older hardware. We know there > are still issues, but I see no real reason to postpone these changes > any further. > Agree. I've found no issues in daily usage for a couple of weeks now and with an unmodified xf86-input-synaptics (newer then 1.3.0 btw). I can't speak for clickpads since I do not have access to one but I'm happy to help narrow down any problems. I really suspect any remaining issues are on user land side. The fact that this patch allows BTN_TOOL_DOUBLETAP to finally be sent will surely help user land filter out cursor jumps much easier as well. Chris -- 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/