linux-kernel - Re: sony-laptop driver: Volume keys on Sony Vaio TX3 don't work

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:	Thu, 20 Oct 2011 00:01:12 +0200
From:	John Hughes <john@...vaedi.com>
To:	John Hughes <john@...va.COM>
CC:	Mattia Dongili <malattia@...ux.it>, Valdis.Kletnieks@...edu,
	linux-kernel@...r.kernel.org, Stelian Pop <stelian@...ies.net>,
	platform-driver-x86 <platform-driver-x86@...r.kernel.org>
Subject: Re: sony-laptop driver: Volume keys on Sony Vaio TX3 don't work

On 10/18/2011 11:07 AM, John Hughes wrote:
> udev does have some per-model stuff for the Vaio's.
>
> Doesn't work right because the scancodes reported the sony-laptop 
> driver are not the right ones.
>
> (sony-laptop sends its "event" code as the scancode, but to change the 
> keymap you have to use the index in the sony_laptop_input_keycode_map 
> table.  I have posted one patch for this which makes the scancodes at 
> least consistent but I think I'll do another that makes them be the 
> same as what current udev thinks they should be.  I'll have that ready 
> this evening I thing - have to work now :-( )
>
Ok, here is a patch that makes the scancodes reported by sony-laptop be 
the same as the SONY_EVENT defines.

Unfortunately this doesn't work terribly well because the scancodes used 
by udev seem to be chosen randomly:

$ cat /lib/udev/keymaps/module-sony
0x06 mute # Fn+F2
0x07 volumedown # Fn+F3
0x08 volumeup # Fn+F4
0x09 brightnessdown # Fn+F5
0x0A brightnessup # Fn+F6
0x0B switchvideomode # Fn+F7
0x0E zoom # Fn+F10
0x10 suspend # Fn+F12

and

$ cat /lib/udev/keymaps/module-sony-vgn
0x00 brightnessdown # Fn+F5
0x10 brightnessup # Fn+F6
0x11 switchvideomode # Fn+F7
0x12 zoomout
0x14 zoomin
0x15 suspend # Fn+F12
0x17 prog1


So the different ideas are:

key      module-sony  module-sony-vgn   orig code(*)   event(*)
fn_f1                                   0x05           0x0c
fn_f2    0x06                           0x06           0x0d
fn_f3    0x07                           0x07           0x0e
fn_f4    0x08                           0x08           0x0f
fn_f5    0x09          0x00 (!)         0x09           0x10
fn_f6    0x0a          0x10             0x0a           0x11
fn_f7    0x0b          0x11             0x0b           0x12
fn_f8                                   0x0c           0x13
fn_f9                                   0x0d           0x14
fn_f10   0x0e                           0x0e           0x15
fn_f11                                  0x0f           0x16
fn_f12   0x10          0x15             0x10           0x17


(*) Orig code - the scancodes that map for the original code, and the 
scancodes returned by my first patch
(*) event - the scancodes from the patch included here.

So, it seems to me that:

1. my original patch is the smaller change, and works with the udev 
"module-sony" keymap file

2. this patch is prettier but doesn't work with existing userspace.

3. the poor sod who wrote the module-sony-vgn file was terribly confused 
by the scancodes returned by the original code.

My recommendation would be to use my first patch - it makes the driver 
return  scancodes that correspond to the scancodes that you need to map.

OK?

Want me to do the whole formal "blablabla-by" stuff now?


View attachment "sony-scancode-event.patch" of type "text/x-patch" (11186 bytes)