linuxconsole-dev

Detecting a keyboard

[Alastair M. <mck...@de...>]

Hello,

I am trying to detect, under Linux (2.4 or 2.5) if a keyboard is
_present_ (attached to a PS/2 or other connector). (Similarly, same ges
for any non-PC architecture. I know about USB, but PPC? Sparc?
Acorn/Amiga/Atari?)

This is the nearest thing I find to a console-experts list, so does
anyone know how to do it ?

(The point being, if I'm at, eg. a PC, and there is are USB & PS2
connectors, and I can detect a keyboard on USB, is that the only
keyboard attached ?)

- Alastair

--=20
Alastair McKinstry <mck...@de...>
GPG Key fingerprint =3D 9E64 E714 8E08 81F9 F3DC  1020 FA8E 3790 9051 38F4

He that would make his own liberty secure must guard even his enemy from
oppression; for if he violates this duty he establishes a precedent that
will reach to himself.

- --Thomas Paine

Re: Detecting a keyboard

[Brad H. <bh...@bi...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Sat, 7 Dec 2002 06:20, Alastair McKinstry wrote:
> Hello,
>
> I am trying to detect, under Linux (2.4 or 2.5) if a keyboard is
> _present_ (attached to a PS/2 or other connector). (Similarly, same ges
> for any non-PC architecture. I know about USB, but PPC? Sparc?
> Acorn/Amiga/Atari?)
Easy on 2.5, you could just parse the /proc/bus/input/devices list. Or do an 
EVIOCGBIT to test for a partcular event interface.

What are you actually trying to accomplish?

Brad

- -- 
http://linux.conf.au. 22-25Jan2003. Perth, Aust. I'm registered. Are you?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE98XSfW6pHgIdAuOMRAqiMAJ4qw+Cn1FxmUfZWhEdLbszVHhicUQCcDMRy
uB37rMfI4+4XCp/iRLRFEhE=
=AuHg
-----END PGP SIGNATURE-----

Re: Detecting a keyboard

[Alastair M. <mck...@de...>]

Thanks.
I'm working on the new debian-installer. I'm looking at the case of
having  multiple potential keyboards in a computer, eg. USB & PS/2.
I want to be able to tell, if possible, what keyboards are attached
(possible in the USB case), and failing that (eg in the PS2 case)
if there is a keyboard attached; so that if, eg we spot a keyboard on
the USB port, we configure that, but is there also a keyboard on the
PS/2 port ?

We may need different keymaps on the different keyboards (indeed, this
might be essential ; one of the few reasons a normal user might have
multiple keyboards is to have different types)


Alastair


On Sat, 2002-12-07 at 04:10, Brad Hards wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>=20
> On Sat, 7 Dec 2002 06:20, Alastair McKinstry wrote:
> > Hello,
> >
> > I am trying to detect, under Linux (2.4 or 2.5) if a keyboard is
> > _present_ (attached to a PS/2 or other connector). (Similarly, same ges
> > for any non-PC architecture. I know about USB, but PPC? Sparc?
> > Acorn/Amiga/Atari?)
> Easy on 2.5, you could just parse the /proc/bus/input/devices list. Or do=
 an=20
> EVIOCGBIT to test for a partcular event interface.
>=20
> What are you actually trying to accomplish?
>=20
> Brad
>=20
> - --=20
> http://linux.conf.au. 22-25Jan2003. Perth, Aust. I'm registered. Are you?
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.0.6 (GNU/Linux)
> Comment: For info see http://www.gnupg.org
>=20
> iD8DBQE98XSfW6pHgIdAuOMRAqiMAJ4qw+Cn1FxmUfZWhEdLbszVHhicUQCcDMRy
> uB37rMfI4+4XCp/iRLRFEhE=3D
> =3DAuHg
> -----END PGP SIGNATURE-----
>=20
>=20
--=20
Alastair McKinstry <mck...@de...>
GPG Key fingerprint =3D 9E64 E714 8E08 81F9 F3DC  1020 FA8E 3790 9051 38F4

He that would make his own liberty secure must guard even his enemy from
oppression; for if he violates this duty he establishes a precedent that
will reach to himself.

- --Thomas Paine

Re: Detecting a keyboard

[Brad H. <bh...@bi...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Sat, 7 Dec 2002 19:26, Alastair McKinstry wrote:
> Thanks.
> I'm working on the new debian-installer. I'm looking at the case of
> having  multiple potential keyboards in a computer, eg. USB & PS/2.
> I want to be able to tell, if possible, what keyboards are attached
> (possible in the USB case), and failing that (eg in the PS2 case)
> if there is a keyboard attached; so that if, eg we spot a keyboard on
> the USB port, we configure that, but is there also a keyboard on the
> PS/2 port ?
You shouldn't care. They will always be mixed together.

> We may need different keymaps on the different keyboards (indeed, this
> might be essential ; one of the few reasons a normal user might have
> multiple keyboards is to have different types)
USB keyboards don't support variable mappings. 

- -- 
http://linux.conf.au. 22-25Jan2003. Perth, Aust. I'm registered. Are you?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE98bf8W6pHgIdAuOMRAmeKAJwIVb5ssikwRD5Kicqbmi3vpXZLHgCfcGhb
Q5NTXhWCZ2LY5bZ7JGpUsm0=
=b9MT
-----END PGP SIGNATURE-----

Re: Detecting a keyboard

[Vojtech P. <vo...@su...>]

On Sat, Dec 07, 2002 at 07:57:32PM +1100, Brad Hards wrote:

> On Sat, 7 Dec 2002 19:26, Alastair McKinstry wrote:
> > Thanks.
> > I'm working on the new debian-installer. I'm looking at the case of
> > having  multiple potential keyboards in a computer, eg. USB & PS/2.
> > I want to be able to tell, if possible, what keyboards are attached
> > (possible in the USB case), and failing that (eg in the PS2 case)
> > if there is a keyboard attached; so that if, eg we spot a keyboard on
> > the USB port, we configure that, but is there also a keyboard on the
> > PS/2 port ?
> You shouldn't care. They will always be mixed together.

Unless you use the event output, of course, or have a multihead
patched kernel.

> > We may need different keymaps on the different keyboards (indeed, this
> > might be essential ; one of the few reasons a normal user might have
> > multiple keyboards is to have different types)
> USB keyboards don't support variable mappings. 

It is planned to have per-keyboard keymaps (keycode->keysym), though no
scancode->keycode maps for USB.

-- 
Vojtech Pavlik
SuSE Labs

Re: Detecting a keyboard

[Arndt S. <ab...@sr...>]

On Mon, Dec 09, 2002 at 11:30:39AM +0100, Vojtech Pavlik wrote:
> 
> > > We may need different keymaps on the different keyboards (indeed, this
> > > might be essential ; one of the few reasons a normal user might have
> > > multiple keyboards is to have different types)
> > USB keyboards don't support variable mappings. 
> 
> It is planned to have per-keyboard keymaps (keycode->keysym), though no
> scancode->keycode maps for USB.

Just curious: For what reasons do you plan to introduce per-keyboard
keycode->keysym maps, but not scancode->keycode maps? Is it because
the keycode->keysym map is the more general case?

Will these per-keyboard keymaps go into 2.6, or is this a feature to
be implemented later?

Thank you,
Arndt

Re: Detecting a keyboard

[Vojtech P. <vo...@su...>]

On Tue, Dec 10, 2002 at 06:59:37PM +0100, Arndt Schoenewald wrote:
> On Mon, Dec 09, 2002 at 11:30:39AM +0100, Vojtech Pavlik wrote:
> > 
> > > > We may need different keymaps on the different keyboards (indeed, this
> > > > might be essential ; one of the few reasons a normal user might have
> > > > multiple keyboards is to have different types)
> > > USB keyboards don't support variable mappings. 
> > 
> > It is planned to have per-keyboard keymaps (keycode->keysym), though no
> > scancode->keycode maps for USB.
> 
> Just curious: For what reasons do you plan to introduce per-keyboard
> keycode->keysym maps, but not scancode->keycode maps?

Scancode->keycode are already per-keyboard for AT keyboards and other
keyboards that need them. But for example on USB, there is no
scancode->keycode map, because all the information about the keyboard is
retrieved from the keyboard's USB descriptor.

> Is it because the keycode->keysym map is the more general case?

Basically yes.

The plan is to have every keyboard supply the same events to the input
subsystem. Then to map them using a keymap to what the user wants.

Basically, the scancode->keycode map should reflect the hardware layout of
the keyboard, and keycode->keysym the preferences of the user (language
layout, etc).

> Will these per-keyboard keymaps go into 2.6, or is this a feature to
> be implemented later?

Probably for later. Ask James, these are already beyond my realm of
input core. ;)

-- 
Vojtech Pavlik
SuSE Labs

Re: Detecting a keyboard

[James S. <jsi...@in...>]

> > Will these per-keyboard keymaps go into 2.6, or is this a feature to
> > be implemented later?
> 
> Probably for later. Ask James, these are already beyond my realm of
> input core. ;)

It will have to wait until the next developement kernel to go in. The fact 
we managed to rework all the fbdev drivers to the ruby api and make the 
console system rely on the input subsystem is amazing. Unfortunely we have 
alot of drivers to work on. When the core console changes go in in the 
next developement cycle it will be easy as we will not have to mess with 
all the input and fbdev drivers next time around.

Re: Detecting a keyboard

[Vojtech P. <vo...@su...>]

On Sat, Dec 07, 2002 at 08:26:59AM +0000, Alastair McKinstry wrote:
> Thanks.
> I'm working on the new debian-installer. I'm looking at the case of
> having  multiple potential keyboards in a computer, eg. USB & PS/2.
> I want to be able to tell, if possible, what keyboards are attached
> (possible in the USB case), and failing that (eg in the PS2 case)
> if there is a keyboard attached; so that if, eg we spot a keyboard on
> the USB port, we configure that, but is there also a keyboard on the
> PS/2 port ?

On 2.5, there is a complete list of attached input devices in
/proc/bus/input/devices. The list is correct only after you load all the
needed modules, of course, which shouldn't pose a problem.

On 2.4, the only way to check for a PS/2 keyboard is to grep 'dmesg' for 'AT
keyboard not present' ...

> We may need different keymaps on the different keyboards (indeed, this
> might be essential ; one of the few reasons a normal user might have
> multiple keyboards is to have different types)
> 
> Alastair
> 
> 
> On Sat, 2002-12-07 at 04:10, Brad Hards wrote:
> > -----BEGIN PGP SIGNED MESSAGE-----
> > Hash: SHA1
> > 
> > On Sat, 7 Dec 2002 06:20, Alastair McKinstry wrote:
> > > Hello,
> > >
> > > I am trying to detect, under Linux (2.4 or 2.5) if a keyboard is
> > > _present_ (attached to a PS/2 or other connector). (Similarly, same ges
> > > for any non-PC architecture. I know about USB, but PPC? Sparc?
> > > Acorn/Amiga/Atari?)
> > Easy on 2.5, you could just parse the /proc/bus/input/devices list. Or do an 
> > EVIOCGBIT to test for a partcular event interface.
> > 
> > What are you actually trying to accomplish?
> > 
> > Brad
> > 
> > - -- 
> > http://linux.conf.au. 22-25Jan2003. Perth, Aust. I'm registered. Are you?
> > -----BEGIN PGP SIGNATURE-----
> > Version: GnuPG v1.0.6 (GNU/Linux)
> > Comment: For info see http://www.gnupg.org
> > 
> > iD8DBQE98XSfW6pHgIdAuOMRAqiMAJ4qw+Cn1FxmUfZWhEdLbszVHhicUQCcDMRy
> > uB37rMfI4+4XCp/iRLRFEhE=
> > =AuHg
> > -----END PGP SIGNATURE-----
> > 
> > 
> -- 
> Alastair McKinstry <mck...@de...>
> GPG Key fingerprint = 9E64 E714 8E08 81F9 F3DC  1020 FA8E 3790 9051 38F4
> 
> He that would make his own liberty secure must guard even his enemy from
> oppression; for if he violates this duty he establishes a precedent that
> will reach to himself.
> 
> - --Thomas Paine



-- 
Vojtech Pavlik
SuSE Labs