From mboxrd@z Thu Jan 1 00:00:00 1970 From: dturvene Subject: Question about psmouse alps driver patches Date: Fri, 25 Jan 2013 10:47:46 -0500 Message-ID: <5102A922.4090203@dahetral.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from dahetral.com ([128.177.27.153]:45488 "EHLO dahetral.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757527Ab3AYQWI (ORCPT ); Fri, 25 Jan 2013 11:22:08 -0500 Received: from [192.168.1.13] (pool-108-28-181-146.washdc.fios.verizon.net [108.28.181.146]) (authenticated bits=0) by dahetral.com (8.13.8/8.13.8) with ESMTP id r0PFlk4M025623 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Fri, 25 Jan 2013 10:47:46 -0500 Sender: linux-input-owner@vger.kernel.org List-Id: linux-input@vger.kernel.org To: linux-input@vger.kernel.org I submitted normalized patches to Canonical Ubuntu in October for the alps psmouse touchpad. The thread is long and confusing at https://bugs.launchpad.net/ubuntu/+source/linux/+bug/606238 Since then, I have maintained a psmouse dlkm that provides alps support for a number of new systems, primarily from Dell. Recently, I have been asked by several people to submit the patches to linux-input to merge into the upstream kernel. I am looking for advice on how to proceed. The big problems are: 1) The alps touchpads seem to be mutating relatively quickly with several unrecognized signatures appearing over the last few months after I built my patches. 2) A number of people have submitted a patch for a particular alps touchpad signature, which will need to be reconciled and rolled-up into a single driver. See the Jan 20 3-part submission by cernekee@gmail.com. His patches look good, and clean up the code a good bit, but target a touchpad signature also reverse-engineered by bgarami.foss@gmail.com. I integrated the bgarami fixes into my patches but the patches from cernekee@gmail.com are radically different. 3) I built the patches against the 3.2 kernel. My understanding is they do not even compile against the kernel head - something like 3.5.x. I feel bad submitting the patches I have. They are big and rough because several of us reverse-engineered the ALPS interfaces but did not try to figure them out. It will take a lot of merge+test work to reconcile the patch submissions for the various alps target platforms. What do you suggest I do (e.g. submit a patch for only the new protocol I can test?)