[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAKdAkRSYxM0_JOD-fR_n49PFxPUbZbVZ4nDqhgzm=HzCb8FQHg@mail.gmail.com>
Date: Thu, 21 Jul 2016 17:57:55 -0700
From: Dmitry Torokhov <dmitry.torokhov@...il.com>
To: Peter Hutterer <peter.hutterer@...-t.net>
Cc: Pavel Machek <pavel@....cz>, Michael Welling <mwelling@...e.org>,
kernel list <linux-kernel@...r.kernel.org>,
"linux-input@...r.kernel.org" <linux-input@...r.kernel.org>,
Pali Rohár <pali.rohar@...il.com>,
Sebastian Reichel <sre@...nel.org>,
"aaro.koskinen@....fi" <aaro.koskinen@....fi>,
Ivaylo Dimitrov <ivo.g.dimitrov.75@...il.com>,
Patrik Bachan <patrikbachan@...il.com>, serge@...lyn.com
Subject: Re: v4.1 to v4.7: regression in tsc2005 driver
On Thu, Jul 21, 2016 at 5:10 PM, Peter Hutterer
<peter.hutterer@...-t.net> wrote:
> On Thu, Jul 21, 2016 at 10:54:21AM +0200, Pavel Machek wrote:
>> Should we have calibration info in the device tree, with kernel
>> passing it to the x?
>>
>> Should kernel somehow do the calibration itself?
>
> if the kernel knows about the calibration it would make more sense to just
> do it in the kernel directly.
Historically we relied on userspace to perform
calibration/transformation. Lately we've been introducing trivial
transformations (inversion and axes swapping) and we allow plumbing
this though DT; the rest I think is still better done in userspace
(where you can do floating point, if desired).
> but for anything even remotely run-time or
> user-configured the xorg.conf snippet is the best solution.
Thanks.
--
Dmitry
Powered by blists - more mailing lists