2004-01-11 00:29:08

by Andries E. Brouwer

[permalink] [raw]
Subject:

: with 2.6.x:
: Loading /etc/console/boottime.kmap.gz
: KDSKBENT: Invalid argument
: failed to bind key 265 to value 638

You must be using a non-standard loadkeys, or an old loadkeys
compiled against new kernel headers.

The 2.6 situation is broken in several respects and one of
the bugs is a NR_KEYS that is 512, which the keys involved
are unsigned characters and cannot be larger than 255.

My opinion is that NR_KEYS must be decreased to 256, and maybe
I have seen patches by Vojtech somewhere that already did this,
but looking at current bk source it seems that they have not been
applied yet.

Anyway, I released kbd-1.10 last week or so, and it ignores the
kernel NR_KEYS but tries to adapt dynamically to the kernel.
It would not come with this error message, I suppose.

: with 2.4.20 & 2.4.22-ck2 the value is increasing from 128 to 511

: failed to bind key 128 to value 512

2.4 has NR_KEYS equal to 128, so key 128 and higher do not exist.

So, so far all is well understood.

: atkbd.c: Unknown key pressed (translated set 2, code 0x91 on isa0060/serio0)
: atkbd.c: Unknown key released (translated set 2, code 0x91 on isa0060/serio0)
: atkbd.c: Unknown key pressed (translated set 2, code 0x92 on isa0060/serio0)
: atkbd.c: Unknown key released (translated set 2, code 0x92 on isa0060/serio0)

This is something different, a key without associated keycode.
That is normal. If it really has high bit set it is a bit unusual.
(What does showkey -s show?)

Maybe you can make this addressable using setkeycodes.

Andries


2004-01-11 01:50:31

by sven kissner

[permalink] [raw]
Subject: Re: logitech cordless desktop deluxe optical keyboard issues

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

hi andries,

thanks for your reply.

On Sunday 11 January 2004 01:28, [email protected] wrote:
> I have seen patches by Vojtech somewhere that already did this,
> but looking at current bk source it seems that they have not been
> applied yet.

where can i find those patches? kml?

> Anyway, I released kbd-1.10 last week or so, and it ignores the
> kernel NR_KEYS but tries to adapt dynamically to the kernel.
> It would not come with this error message, I suppose.

the message doesn't appear anymore, but installing is giving me the following:
<-- snip -->
Setting up kbd (1.10-1) ...
Looking for keymap to install:
de-latin1-nodeadkeys
cannot open file de-latin1-nodeadkeys
Loading /etc/console/boottime.kmap.gz
<-- snap -->
i guess i just screwed up the paths ;)

> : atkbd.c: Unknown key pressed (translated set 2, code 0x91 on
> : isa0060/serio0) atkbd.c: Unknown key released (translated set 2, code
> : 0x91 on isa0060/serio0) atkbd.c: Unknown key pressed (translated set 2,
> : code 0x92 on isa0060/serio0) atkbd.c: Unknown key released (translated
> : set 2, code 0x92 on isa0060/serio0)
>
> This is something different, a key without associated keycode.
> That is normal. If it really has high bit set it is a bit unusual.
> (What does showkey -s show?)

showkey -s is giving me exactly the same:
<-- snip -->
atkbd.c: Unknown key pressed (translated set 2, code 0x92 on isa0060/serio0).
atkbd.c: Unknown key released (translated set 2, code 0x92 on isa0060/serio0).
<-- snap -->

sven
- --
..never argue with idiots. they drag you down to their level and beat you with
experience..
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQFAAKySPV/e7f4i4AERAm0FAJ97N/CT3PfUcJZK2+4gXy7e3lDFgwCgl/T2
qRVMrl5ik2NZ3dtGdxounJQ=
=8I8H
-----END PGP SIGNATURE-----