Re: asus_wmi: Unknown key pressed

["Konstantin V. Gavrilenko" <k.gavrilenko@arhont.com>]

Yours sincerely,
Konstantin V. Gavrilenko

Arhont Information Security Ltd

web:    http://www.arhont.com
e-mail: k.gavrilenko@arhont.com

tel: +44 (0) 870 44 31337
fax: +44 (0) 208 429 3111

PGP: Key ID - 0xE81824F4
PGP: Server - keyserver.pgp.com


----- Original Message -----
From: "Corentin Chary" <corentincj@iksaif.net>
To: "Dmitry Torokhov" <dmitry.torokhov@gmail.com>
Cc: "Konstantin V. Gavrilenko" <k.gavrilenko@arhont.com>, "Yong Wang" <yong.y.wang@intel.com>, linux-input@vger.kernel.org, "Seth Forshee" <seth.forshee@canonical.com>
Sent: Sunday, 6 January, 2013 6:30:40 PM
Subject: Re: asus_wmi: Unknown key pressed

On Sat, Jan 5, 2013 at 11:09 PM, Dmitry Torokhov
<dmitry.torokhov@gmail.com> wrote:
> On Sat, Jan 05, 2013 at 08:06:32PM +0000, Corentin Chary wrote:
>> On Sat, Jan 5, 2013 at 7:19 PM, Konstantin V. Gavrilenko
>> <k.gavrilenko@arhont.com> wrote:
>> > Thanks for a prompt response.
>> >
>> > for all three events the showkey generates the same data
>> >
>> > keycode 240 press
>> > keycode 240 release
>> >
>> > ----- Original Message -----
>> > From: "Corentin Chary" <corentincj@iksaif.net>
>> > To: "Konstantin V. Gavrilenko" <k.gavrilenko@arhont.com>
>> > Cc: "Yong Wang" <yong.y.wang@intel.com>, "Dmitry Torokhov" <dtor@mail.ru>
>> > Sent: Saturday, 5 January, 2013 6:58:11 PM
>> > Subject: Re: asus_wmi: Unknown key pressed
>> >
>> > Hi,
>> >
>> > On Sat, Jan 5, 2013 at 3:17 PM, Konstantin V. Gavrilenko
>> > <k.gavrilenko@arhont.com> wrote:
>> >> Hi Guys,
>> >>
>> >> sorry for writing out of the blue, but your email addresses where provided in the kernel modules. I hope you will direct me to a better place where I can report the problem or someone that can help solve the problem with asus_wmi module.
>> >>
>> >> In my new laptop UX51VZA v203 my syslog is constantly filled with the following lines, even though I do not press any buttons on the keyboard.
>> >>
>> >> Dec 30 08:28:52 dynomob kernel: [22731.108645] asus_wmi: Unknown key c6 pressed
>> >> Dec 30 08:28:53 dynomob kernel: [22732.110834] asus_wmi: Unknown key c6 pressed
>> >> Dec 30 08:28:58 dynomob kernel: [22737.079312] asus_wmi: Unknown key c6 pressed
>> >> Dec 30 08:28:59 dynomob kernel: [22738.075352] asus_wmi: Unknown key c6 pressed
>> >> Dec 30 08:29:00 dynomob kernel: [22739.076342] asus_wmi: Unknown key c6 pressed
>> >> Dec 30 08:29:01 dynomob kernel: [22740.083493] asus_wmi: Unknown key c6 pressed
>> >>
>> >>
>> >> It took me a while to work out that this was information from the ambient light sensor )) and since there is no way to operate it or disable it, I had to cover it with a piece of black paper. Such default behavior is very disruptive as it generates an actual unseen key press, that is extremely annoying if you trying to type a password.
>> >
>> > This should not generate any key if the key is unknown. But maybe
>> > another key is generated (and not logged). Can you run showkey in a
>> > terminal and try to see what keycode you are getting ?
>> > I'll make a patch to ignore this key.
>> >
>> >> There are couple of other unknown signals tat I managed to figure out, namely
>> >>
>> >> Fn+A (ambient light button) generates
>> >> Jan  5 15:12:08 dynomob kernel: [46786.001579] asus_wmi: Unknown key 7a pressed
>> >
>> > Will map this key to some keycode.
>> >
>> >> unplugging of the powercord generates
>> >> Jan  5 15:13:08 dynomob kernel: [46845.606336] asus_wmi: Unknown key cf pressed
>> >
>> > Don't know what is means, so will probably just be ignored with a comment.
>> >
>> > Thanks,
>>
>> Oh right, Seth added some code to send KEY_UNKNOWN keys in such cases.
>> Seth, maybe keys present with KEY_IGNORE type should not send
>> KEY_UNKNOWN to avoid this kind of issues ?
>
> KEY_UNKNOWN should not really be mapped to any key so it should not
> interfere with normal typing/other operations while still being able to
> detect the events. The idea is for users to be able to notice the new
> events and then developers can decide if they should be ignored or
> mapped to some particular keycodes (or maybe routed through some other
> subsystem).

> The issue here is that the scancode is already present in the keymap
> with the type "KE_IGNORE" instead of "KE_KEY", but sparse_keymap seems
> to send KEY_UNKNOWN anyway.
> And Konstantin seems to say that it's bad in some cases (passwords?).



Depending on how frequently the light conditions change, the number of events generated can be quite large, as often as every second.
The main annoyances I noticed:
- the logs get clogged up very fast with "asus_wmi: Unknown key c6 pressed"
- those "hidden" characters are silently inserted, making it impossible to type the password correctly, unless the ambient light does not change.
- the automatic layout switcher (xneur) is impossible to use
- general experience of typing is degraded, as you are frequently interrupted.

yours,
Konstantin

--
Corentin Chary
http://xf.iksaif.net