From mboxrd@z Thu Jan 1 00:00:00 1970 From: Benjamin Tissoires Subject: Re: [PATCH 0/2] input: Add INPUT_PROP_POINTING_STICK property Date: Tue, 2 Sep 2014 10:34:13 -0400 Message-ID: <20140902143413.GA25423@mail.corp.redhat.com> References: <1409661804-10489-1-git-send-email-hdegoede@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Return-path: Received: from mx1.redhat.com ([209.132.183.28]:40775 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752121AbaIBOeU (ORCPT ); Tue, 2 Sep 2014 10:34:20 -0400 Content-Disposition: inline In-Reply-To: Sender: linux-input-owner@vger.kernel.org List-Id: linux-input@vger.kernel.org To: David Herrmann Cc: Hans de Goede , Dmitry Torokhov , Peter Hutterer , "open list:HID CORE LAYER" Hi David, On Sep 02 2014 or thereabouts, David Herrmann wrote: > Hi > > On Tue, Sep 2, 2014 at 2:43 PM, Hans de Goede wrote: > > Hi, > > > > It is useful for userspace to know that they're not dealing with a regular > > mouse but rather with a pointing stick (e.g. a trackpoint) so that userspace > > can e.g. automatically enable middle button scrollwheel emulation. > > > > It is impossible to tell the difference from the evdev info without resorting > > to putting a list of device / driver names in userspace, this is undesirable. > > ..so it is better to put that table into the kernel? > > I thought the plan was to avoid putting more hardware tables into the > kernel. All those tables are always loaded into RAM, whereas solutions > like udev hwdb can provide the exact same features in user-space but > avoid loading it into RAM, except if used. hwdb is pretty nice to > store information about hardware. It is not meant for configuration or > volatile data, but rather as read-only lookup table for hardware > information. I think you misread Hans in this case. Hans stated: - if this is not provided by the kernel, we have to create a user space table with all old and new future devices - if we put this in the kernel, then we only have to add this property to the few generic drivers which present this feature to the user space, and done. See patch 1/2. There is no table in the kernel. > > You could easily extend the hwdb with a trackpoint variable that marks > specific devices as trackpoints. I don't think this is reasonable. You will have to review all laptops with trackstick to add them into this (huge) db. > > Anyhow, if you guys want it in the kernel, go ahead. I just don't see > the advantage of having it in kernel-space. > Again, it's either 10 lines in the kernel, and we forget about it, or we have to carry a db which has to be updated with every single different laptop model available. I am definitively in favor of those 10 lines. Cheers, Benjamin