support for Logitech Wave

support for Logitech Wave

[Timo Aaltonen]

 	Hi

   I've got a Logitech Wave keyboard, that has a number of extra keys on 
it, and most of them fail to work with evdev. I get no output with xev or 
while trying to cat the input device. I've been told that it's a kernel 
issue if the device shows no events, hence this email.

What should I do to get full support for all the keys on it? It's a nice 
keyboard, so I'd like to get the most of it :)

t

Re: support for Logitech Wave

[Dmitry Torokhov]

Hi Timo,

On Tue, Jan 15, 2008 at 03:49:14PM +0200, Timo Aaltonen wrote:
>
> 	Hi
>
>   I've got a Logitech Wave keyboard, that has a number of extra keys on it, 
> and most of them fail to work with evdev. I get no output with xev or while 
> trying to cat the input device. I've been told that it's a kernel issue if 
> the device shows no events, hence this email.
>
> What should I do to get full support for all the keys on it? It's a nice 
> keyboard, so I'd like to get the most of it :)
>

Please try booting with the latest -rc kernel (currently 2.6.26-rc8) and run
evtest utility. You should see it report MSC_SCAN events when you press keys
on your keyboard. Compose list of keys and corresponding codes and send it to
Jiri Kosina so he can add proper mappings to the HID driver.

Thanks!

-- 
Dmitry

Re: support for Logitech Wave

[Taneli Vähäkangas]

Hi! Chiming in, as I have a similar issue ...

On Wed, Jan 16, 2008 at 11:07:17AM -0500, Dmitry Torokhov wrote:
> Hi Timo,
> 
> On Tue, Jan 15, 2008 at 03:49:14PM +0200, Timo Aaltonen wrote:
> >
> > 	Hi
> >
> >   I've got a Logitech Wave keyboard, that has a number of extra keys on it, 
> > and most of them fail to work with evdev. I get no output with xev or while 
> > trying to cat the input device. I've been told that it's a kernel issue if 
> > the device shows no events, hence this email.
> >
> > What should I do to get full support for all the keys on it? It's a nice 
> > keyboard, so I'd like to get the most of it :)
> >
> 
> Please try booting with the latest -rc kernel (currently 2.6.26-rc8) and run
> evtest utility. You should see it report MSC_SCAN events when you press keys
> on your keyboard. Compose list of keys and corresponding codes and send it to
> Jiri Kosina so he can add proper mappings to the HID driver.

I tried the same on a Creative Prodikeys keyboard, but can't see any
events for some keys (most notably none of the "piano" keys work).

Is there anything that can be done to make it work?

Cheers,

	Taneli

Re: support for Logitech Wave

[Jiri Kosina]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 611 bytes --]

On Wed, 16 Jan 2008, Taneli Vähäkangas wrote:

> I tried the same on a Creative Prodikeys keyboard, but can't see any 
> events for some keys (most notably none of the "piano" keys work). Is 
> there anything that can be done to make it work?

Hi Taneli,

yes. Could you please grab reasonably recent kernel (2.6.24-rc<anything> 
would be ok), compile it with CONFIG_HID_DEBUG, modprobe the 'hid' module 
with 'debug=1' parameter, and send me the output from the time you connect 
the keyboard and from pressing the non-working keys? It should be trivial 
to fix then.

Thanks,

-- 
Jiri Kosina

Re: support for Logitech Wave

[Timo Aaltonen]

On Wed, 16 Jan 2008, Jiri Kosina wrote:

> On Wed, 16 Jan 2008, Taneli Vähäkangas wrote:
>
>> I tried the same on a Creative Prodikeys keyboard, but can't see any
>> events for some keys (most notably none of the "piano" keys work). Is
>> there anything that can be done to make it work?
>
> Hi Taneli,
>
> yes. Could you please grab reasonably recent kernel (2.6.24-rc<anything>
> would be ok), compile it with CONFIG_HID_DEBUG, modprobe the 'hid' module
> with 'debug=1' parameter, and send me the output from the time you connect
> the keyboard and from pressing the non-working keys? It should be trivial
> to fix then.

Hi! I managed to get the event codes with evtest. Here's the output, 
followed by mappings to what the buttons really are. The ones that are 
mapped below 255 are working on X:

Input driver version is 1.0.0
Input device ID: bus 0x3 vendor 0x46d product 0xc317 version 0x110
Input device name: "Logitech USB Multimedia Keyboard"
Supported events:
   Event type 0 (Reset)
     Event code 0 (Reset)
     Event code 1 (Key)
     Event code 4 (?)
   Event type 1 (Key)
     Event code 113 (Mute)
     Event code 114 (VolumeDown)
     Event code 115 (VolumeUp)
     Event code 140 (Calc)
     Event code 142 (Sleep)
     Event code 155 (Mail)
     Event code 163 (NextSong)
     Event code 164 (PlayPause)
     Event code 165 (PreviousSong)
     Event code 166 (StopCD)
     Event code 171 (Config)
     Event code 172 (HomePage)
     Event code 256 (Btn0)
     Event code 257 (Btn1)
     Event code 258 (Btn2)
     Event code 259 (Btn3)
     Event code 260 (Btn4)
     Event code 261 (Btn5)
     Event code 262 (Btn6)
     Event code 263 (Btn7)
     Event code 264 (Btn8)
     Event code 266 (?)
     Event code 267 (?)
     Event code 268 (?)
     Event code 269 (?)
     Event code 270 (?)
     Event code 271 (?)
     Event code 272 (LeftBtn)
     Event code 273 (RightBtn)


Mappings:

256 (Btn0)	"Word"
257 (Btn1)	"Spreadsheet"
258 (Btn2)	"Calendar"
259 (Btn3)	"custom key a"
260 (Btn4)	"Photos"
261 (Btn5)	"custom key b"
262 (Btn6)	"custom key c"
263 (Btn7)	"Media center"
264 (Btn8)	"Messenger"
266 (?)		"local search"
267 (?)		"Vista Flip 3D" (go through windows, "alt-tab")
268 (?)		"Vista Gadget" (
269 (?)		"internet search"
270 (?)		"zoom -"
271 (?)		"zoom +"
272 (LeftBtn)	"eject"
273 (RightBtn)	"Function key, used for most of the above"

Hope this helps!

t
-
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: support for Logitech Wave

[Timo Aaltonen]

On Thu, 17 Jan 2008, Timo Aaltonen wrote:

> 268 (?)		"Vista Gadget" (

hmm, this was not complete.. that key is apparently used for launching the 
vista gadget library.. never seen one but that's the name of the key on 
the "manual".

btw, the kernel was 2.6.24-rc7 from Ubuntu Hardy.

t

Re: support for Logitech Wave

[Timo Aaltonen]

On Thu, 17 Jan 2008, Timo Aaltonen wrote:

> On Wed, 16 Jan 2008, Jiri Kosina wrote:
>
>> On Wed, 16 Jan 2008, Taneli Vähäkangas wrote:
>> 
>>> I tried the same on a Creative Prodikeys keyboard, but can't see any
>>> events for some keys (most notably none of the "piano" keys work). Is
>>> there anything that can be done to make it work?
>> 
>> Hi Taneli,
>> 
>> yes. Could you please grab reasonably recent kernel (2.6.24-rc<anything>
>> would be ok), compile it with CONFIG_HID_DEBUG, modprobe the 'hid' module
>> with 'debug=1' parameter, and send me the output from the time you connect
>> the keyboard and from pressing the non-working keys? It should be trivial
>> to fix then.
>
> Hi! I managed to get the event codes with evtest. Here's the output, followed 
> by mappings to what the buttons really are. The ones that are mapped below 
> 255 are working on X:
>
> Input driver version is 1.0.0
> Input device ID: bus 0x3 vendor 0x46d product 0xc317 version 0x110
> Input device name: "Logitech USB Multimedia Keyboard"
> Supported events:
>  Event type 0 (Reset)
>    Event code 0 (Reset)
>    Event code 1 (Key)
>    Event code 4 (?)
>  Event type 1 (Key)
>    Event code 113 (Mute)
>    Event code 114 (VolumeDown)
>    Event code 115 (VolumeUp)
>    Event code 140 (Calc)
>    Event code 142 (Sleep)
>    Event code 155 (Mail)
>    Event code 163 (NextSong)
>    Event code 164 (PlayPause)
>    Event code 165 (PreviousSong)
>    Event code 166 (StopCD)
>    Event code 171 (Config)
>    Event code 172 (HomePage)
>    Event code 256 (Btn0)
>    Event code 257 (Btn1)
>    Event code 258 (Btn2)
>    Event code 259 (Btn3)
>    Event code 260 (Btn4)
>    Event code 261 (Btn5)
>    Event code 262 (Btn6)
>    Event code 263 (Btn7)
>    Event code 264 (Btn8)
>    Event code 266 (?)
>    Event code 267 (?)
>    Event code 268 (?)
>    Event code 269 (?)
>    Event code 270 (?)
>    Event code 271 (?)
>    Event code 272 (LeftBtn)
>    Event code 273 (RightBtn)
>
>
> Mappings:
>
> 256 (Btn0)	"Word"
> 257 (Btn1)	"Spreadsheet"
> 258 (Btn2)	"Calendar"
> 259 (Btn3)	"custom key a"
> 260 (Btn4)	"Photos"
> 261 (Btn5)	"custom key b"
> 262 (Btn6)	"custom key c"
> 263 (Btn7)	"Media center"
> 264 (Btn8)	"Messenger"
> 266 (?)		"local search"
> 267 (?)		"Vista Flip 3D" (go through windows, "alt-tab")
> 268 (?)		"Vista Gadget" (Vista "desktop applets"?)
> 269 (?)		"internet search"
> 270 (?)		"zoom -"
> 271 (?)		"zoom +"
> 272 (LeftBtn)	"eject"
> 273 (RightBtn)	"Function key, used for most of the above"
>
> Hope this helps!

Jiri, did you notice this post? Is there enough information for you to 
patch the driver?

thanks,

t
-
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: support for Logitech Wave

[Jiri Kosina]

On Tue, 12 Feb 2008, Timo Aaltonen wrote:

> > > yes. Could you please grab reasonably recent kernel 
> > > (2.6.24-rc<anything> would be ok), compile it with CONFIG_HID_DEBUG, 
> > > modprobe the 'hid' module with 'debug=1' parameter, and send me the 
> > > output from the time you connect the keyboard and from pressing the 
> > > non-working keys? It should be trivial to fix then.
> > Hi! I managed to get the event codes with evtest. Here's the output, 
> > followed by mappings to what the buttons really are. The ones that are 
> > mapped below 255 are working on X:
> Jiri, did you notice this post? Is there enough information for you to patch
> the driver?

No, sorry, I missed it, please don't drop me from To:/Cc:, otherwise I am 
likely to miss posts :)

If all the events are seen by evtest, then everything is fine on the 
kernel side and you should just tweak your X configuration.

-- 
Jiri Kosina

Re: support for Logitech Wave

[Timo Aaltonen]

On Tue, 12 Feb 2008, Jiri Kosina wrote:

> On Tue, 12 Feb 2008, Timo Aaltonen wrote:
>
>>>> yes. Could you please grab reasonably recent kernel
>>>> (2.6.24-rc<anything> would be ok), compile it with CONFIG_HID_DEBUG,
>>>> modprobe the 'hid' module with 'debug=1' parameter, and send me the
>>>> output from the time you connect the keyboard and from pressing the
>>>> non-working keys? It should be trivial to fix then.
>>> Hi! I managed to get the event codes with evtest. Here's the output,
>>> followed by mappings to what the buttons really are. The ones that are
>>> mapped below 255 are working on X:
>> Jiri, did you notice this post? Is there enough information for you to patch
>> the driver?
>
> No, sorry, I missed it, please don't drop me from To:/Cc:, otherwise I am
> likely to miss posts :)

Ah ok, sorry about that :)

> If all the events are seen by evtest, then everything is fine on the
> kernel side and you should just tweak your X configuration.

hmm, but X doesn't allow key-id's >255, so I can only wait until they 
support more? (maybe in X.org R7.5..)

t

Re: support for Logitech Wave

[Jiri Kosina]

On Tue, 12 Feb 2008, Timo Aaltonen wrote:

> > If all the events are seen by evtest, then everything is fine on the
> > kernel side and you should just tweak your X configuration.
> hmm, but X doesn't allow key-id's >255, so I can only wait until they support
> more? (maybe in X.org R7.5..)

You need to specify "evdev" as the keyboard driver in the xorg.conf input 
device section and the correct input device for your keyboard. Then you 
should be able to get all the keys working correctly in xorg.

-- 
Jiri Kosina

Re: support for Logitech Wave

[Timo Aaltonen]

On Tue, 12 Feb 2008, Jiri Kosina wrote:

> On Tue, 12 Feb 2008, Timo Aaltonen wrote:
>
>>> If all the events are seen by evtest, then everything is fine on the
>>> kernel side and you should just tweak your X configuration.
>> hmm, but X doesn't allow key-id's >255, so I can only wait until they support
>> more? (maybe in X.org R7.5..)
>
> You need to specify "evdev" as the keyboard driver in the xorg.conf input
> device section and the correct input device for your keyboard. Then you
> should be able to get all the keys working correctly in xorg.

Yes, evdev is what I'm using, but I still don't get any events with xev. 
I'll ask the Xorg dev's again about it. IIRC it was a limitation of XKB..

t

Re: support for Logitech Wave

[Taneli Vähäkangas]

On Wed, Jan 16, 2008 at 11:00:48PM +0100, Jiri Kosina wrote:
> On Wed, 16 Jan 2008, Taneli Vähäkangas wrote:
> 
> > I tried the same on a Creative Prodikeys keyboard, but can't see any 
> > events for some keys (most notably none of the "piano" keys work). Is 
> > there anything that can be done to make it work?
> 
> Hi Taneli,
> 
> yes. Could you please grab reasonably recent kernel (2.6.24-rc<anything> 
> would be ok), compile it with CONFIG_HID_DEBUG, modprobe the 'hid' module 
> with 'debug=1' parameter, and send me the output from the time you connect 
> the keyboard and from pressing the non-working keys? It should be trivial 
> to fix then.

Thanks for prompt reply Jiri. Now I get some information for extra
function keys at the top row, but still nothing for the "piano" keys.
Here's the output for two previously non-functioning keys:


Jan 17 00:27:13 newer kernel: drivers/hid/hid-core.c: report (size 4)
(numbered)
Jan 17 00:27:13 newer kernel: drivers/hid/hid-core.c: report 4 (size 3) =  02 00 00
Jan 17 00:27:13 newer kernel: hid-debug: input ff00.0001 = 0
Jan 17 00:27:13 newer kernel: hid-debug: input ff00.0001 = 1
Jan 17 00:27:13 newer kernel: hid-debug: input ff00.0001 = 0
Jan 17 00:27:13 newer last message repeated 13 times
Jan 17 00:27:13 newer kernel: drivers/hid/hid-core.c: report (size 4) (numbered)
Jan 17 00:27:13 newer kernel: drivers/hid/hid-core.c: report 4 (size 3) =  00 00 00
Jan 17 00:27:13 newer kernel: hid-debug: input ff00.0001 = 0
Jan 17 00:27:13 newer last message repeated 15 times


Is this the information you asked? I can make you the full mapping of
non-working keys to log messages. However, I don't think that's very
useful if the "piano" keys don't work, as they're the main reason for
having such a keyboard in the first place.

There was a great deal of log messages at the time of plugging in the
keyboard. I didn't feel comfortable spamming the list with it, so it is
available at:
http://www.cs.helsinki.fi/u/vahakang/hid-dbg.txt

(I think all the log entries after 00:23:59 are key presses, but
couldn't tell for sure.)

BTW, here's a picture of the keyboard in question ... just in case my
descriptions don't seem to make sense:
http://images.tw.creative.com/images/corporate/artwork/hires_prodikeyspcmidiusb.jpg

Cheers,

	Taneli
-
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: support for Logitech Wave

[Jiri Kosina]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 2082 bytes --]

On Thu, 17 Jan 2008, Taneli Vähäkangas wrote:

> Thanks for prompt reply Jiri. Now I get some information for extra 
> function keys at the top row, but still nothing for the "piano" keys. 

That's strange. Could you please try to snoop the traffic on USB, to see 
if the keys that don't produce any HID debugging output at all at least 
produce some USB traffic?

See Documentation/usb/usbmon.txt for details.

> Here's the output for two previously non-functioning keys:
> Jan 17 00:27:13 newer kernel: drivers/hid/hid-core.c: report (size 4)
> (numbered)
> Jan 17 00:27:13 newer kernel: drivers/hid/hid-core.c: report 4 (size 3) =  02 00 00
> Jan 17 00:27:13 newer kernel: hid-debug: input ff00.0001 = 0
> Jan 17 00:27:13 newer kernel: hid-debug: input ff00.0001 = 1
> Jan 17 00:27:13 newer kernel: hid-debug: input ff00.0001 = 0
> Jan 17 00:27:13 newer last message repeated 13 times
> Jan 17 00:27:13 newer kernel: drivers/hid/hid-core.c: report (size 4) (numbered)
> Jan 17 00:27:13 newer kernel: drivers/hid/hid-core.c: report 4 (size 3) =  00 00 00
> Jan 17 00:27:13 newer kernel: hid-debug: input ff00.0001 = 0
> Jan 17 00:27:13 newer last message repeated 15 times

Hmm, this looks only like output for on key, is it really for two?

> Is this the information you asked? I can make you the full mapping of
> non-working keys to log messages. However, I don't think that's very
> useful if the "piano" keys don't work, as they're the main reason for
> having such a keyboard in the first place.

Yes, we'll need to figure out where the data from these keys are getting 
lost.

> (I think all the log entries after 00:23:59 are key presses, but
> couldn't tell for sure.)

Yes, they are.

> BTW, here's a picture of the keyboard in question ... just in case my
> descriptions don't seem to make sense:
> http://images.tw.creative.com/images/corporate/artwork/hires_prodikeyspcmidiusb.jpg

Could it be that the keyboard is a totally separate device? Does alsa 
report anything, when you connect it, for example?

-- 
Jiri Kosina

Re: support for Logitech Wave

[Taneli Vähäkangas]

Hello!

On Thu, Jan 17, 2008 at 02:07:03PM +0100, Jiri Kosina wrote:
> On Thu, 17 Jan 2008, Taneli Vähäkangas wrote:
> 
> > Thanks for prompt reply Jiri. Now I get some information for extra 
> > function keys at the top row, but still nothing for the "piano" keys. 
> 
> That's strange. Could you please try to snoop the traffic on USB, to see 
> if the keys that don't produce any HID debugging output at all at least 
> produce some USB traffic?
> 
> See Documentation/usb/usbmon.txt for details.

I did that and there are no usbmon messages at all for the "piano" keys,
so this is most likely not linux-input or HID issue.

> > Here's the output for two previously non-functioning keys:
> > Jan 17 00:27:13 newer kernel: drivers/hid/hid-core.c: report (size 4)
> > (numbered)
> > Jan 17 00:27:13 newer kernel: drivers/hid/hid-core.c: report 4 (size 3) =  02 00 00
> > Jan 17 00:27:13 newer kernel: hid-debug: input ff00.0001 = 0
> > Jan 17 00:27:13 newer kernel: hid-debug: input ff00.0001 = 1
> > Jan 17 00:27:13 newer kernel: hid-debug: input ff00.0001 = 0
> > Jan 17 00:27:13 newer last message repeated 13 times
> > Jan 17 00:27:13 newer kernel: drivers/hid/hid-core.c: report (size 4) (numbered)
> > Jan 17 00:27:13 newer kernel: drivers/hid/hid-core.c: report 4 (size 3) =  00 00 00
> > Jan 17 00:27:13 newer kernel: hid-debug: input ff00.0001 = 0
> > Jan 17 00:27:13 newer last message repeated 15 times
> 
> Hmm, this looks only like output for on key, is it really for two?

I'll check again later, I was quite confident last night that it was two
keypresses, but I was also quite tired ...

> > BTW, here's a picture of the keyboard in question ... just in case my
> > descriptions don't seem to make sense:
> > http://images.tw.creative.com/images/corporate/artwork/hires_prodikeyspcmidiusb.jpg
> 
> Could it be that the keyboard is a totally separate device? Does alsa 
> report anything, when you connect it, for example?

Unfortunately alsa notices nothing (that was also my first guess, so I
tried that even before reporting the problem here).

Could it be that hid steals the device and that's why alsa doesn't see
it?

Maybe the keyboard expects some kind of trigger from the host side to
enable the missing functionality. It apparently requires a separate
driver in windows and doesn't work in Mac OS X (that's why my sister
borrowed the keyboard to me, it didn't work with a Mac ...)

Thanks a lot for your time and responses Jiri.

Cheers,

	Taneli

-
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: support for Logitech Wave

[Kevin Milburn]

[-- Attachment #1: Type: text/plain, Size: 1076 bytes --]

Hi Jiri

I'm having problems with a Logitech Wave (cordless) keyboard and its extra 
keys,   and am hoping the following information will help in resolving the 
problem.

This testing has been done with the following configuration :

Distribution : openSuse 10.3
Kernel :  2.6.25-18-default  (built with source from openSuse build service)
Keyboard :  Logitech Cordless Desktop Wave,  Model Y-RCP140  
<http://www.logitech.com/index.cfm/keyboards/keyboard_mice_combos/devices/3072&cl=us,en>

I've tested the keyboard with evtest,  and none of the extra keys are 
recognized,  and no information reported when they are pressed

Testing with Xorg indicates that some of the keys work when the 'kbd' driver 
is used,  but not when 'evdev' is used  (probably due to the above).

Evbug does see most of these extra keys,  but not all of them. 

I've attached a copy of  information from a debug session of the HID,  and 
information from evbug.  It also indicates which keys work (depending on 
layout).

If there's anything else I can do to help,  please let me know.

TIA
Kevin.

[-- Attachment #2: logicdwave.txt.bz2 --]
[-- Type: application/x-bzip2, Size: 8824 bytes --]

Re: support for Logitech Wave

[Jiri Kosina]

On Fri, 25 Apr 2008, Kevin Milburn wrote:

> I'm having problems with a Logitech Wave (cordless) keyboard and its 
> extra keys, and am hoping the following information will help in 
> resolving the problem.

Hi Kevin,

is the device in question vendor ID 0x046d and product ID 0xc517? (you can 
find out using lsusb, for example).

If so, then that's the very same receiver that Logitech ships with S510 
keyboards, so we'll have to put this extra mappings into the 
quirk_logitech_wireless() function.

> I've tested the keyboard with evtest, and none of the extra keys are 
> recognized, and no information reported when they are pressed

Could you please be more specific here? Those keys for which there is some 
'evbug' output should definitely also appear in evtest output (if X driver 
will then receive the code or not is a different story).

> I've attached a copy of information from a debug session of the HID, and 
> information from evbug.  It also indicates which keys work (depending on 
> layout).

Thanks, this is all the information that will be needed for establishing 
proper mappings for non-working keys, but I'd like to understand first 
whether there are keys that appear in evbug output, but there is no event 
happening on evdev (as seen by evtest).

A few remarks from quick look at the output:

- what are the "Vista Flip" and "Vista Gadget" keys supposed to do?
- the keys that are labelled "works depending on layout" -- what exactly 
  does that mean? Is there still output from hid-debug, no matter the 
  layout?

Thanks,

-- 
Jiri Kosina
SUSE Labs

Re: support for Logitech Wave

[Jiri Kosina]

On Mon, 28 Apr 2008, Kevin Milburn wrote:

> > If so, then that's the very same receiver that Logitech ships with 
> > S510 keyboards, so we'll have to put this extra mappings into the 
> > quirk_logitech_wireless() function.
> Yes, the device is reported as vendor ID 0x046d and product ID 0xc517r

OK, so I will add the new mapping into quirk_logitech_wireless() and will 
send you a patch to test.

> > Could you please be more specific here? Those keys for which there is 
> > some 'evbug' output should definitely also appear in evtest output (if 
> > X driver will then receive the code or not is a different story).
> Evtest is connecting to the device correctly and reports keypess 
> information for the standard keys, but nothing is output when one of the 
> extra keys is pressed..
[ ... ]
> Apr 28 10:47:31 ganges kernel: evbug.c: Event. Dev: usb-0000:00:02.0-7/input0, 
> Type: 4, Code: 4, Value: 458796
> Apr 28 10:47:31 ganges kernel: evbug.c: Event. Dev: usb-0000:00:02.0-7/input0, 
> Type: 1, Code: 57, Value: 1
> Apr 28 10:47:31 ganges kernel: evbug.c: Event. Dev: usb-0000:00:02.0-7/input0, 
> Type: 0, Code: 0, Value: 0
> Apr 28 10:47:31 ganges kernel: evbug.c: Event. Dev: usb-0000:00:02.0-7/input0, 
> Type: 4, Code: 4, Value: 458796
> Apr 28 10:47:31 ganges kernel: evbug.c: Event. Dev: usb-0000:00:02.0-7/input0, 
> Type: 1, Code: 57, Value: 0
> Apr 28 10:47:31 ganges kernel: evbug.c: Event. Dev: usb-0000:00:02.0-7/input0, 
> Type: 0, Code: 0, Value: 0
> Apr 28 10:48:09 ganges kernel: evbug.c: Event. Dev: usb-0000:00:02.0-7/input1, 
> Type: 4, Code: 4, Value: 786613
> Apr 28 10:48:09 ganges kernel: evbug.c: Event. Dev: usb-0000:00:02.0-7/input1, 
> Type: 1, Code: 163, Value: 1
> Apr 28 10:48:09 ganges kernel: evbug.c: Event. Dev: usb-0000:00:02.0-7/input1, 
> Type: 0, Code: 0, Value: 0
> Apr 28 10:48:09 ganges kernel: evbug.c: Event. Dev: usb-0000:00:02.0-7/input1, 
> Type: 4, Code: 4, Value: 786613
> Apr 28 10:48:09 ganges kernel: evbug.c: Event. Dev: usb-0000:00:02.0-7/input1, 
> Type: 1, Code: 163, Value: 0
> Apr 28 10:48:09 ganges kernel: evbug.c: Event. Dev: usb-0000:00:02.0-7/input1, 
> Type: 0, Code: 0, Value: 0

The keyboard probably has two separate interfaces, with some multimedia 
keys being reported on a different one. That's why you see them in evbug 
output, but not in evtest output, because they appear under different 
event device.

-- 
Jiri Kosina
SUSE Labs

Re: support for Logitech Wave

[Kevin Milburn]

On Wednesday 30 April 2008 10:21:16 Jiri Kosina wrote:
> The keyboard probably has two separate interfaces, with some multimedia
> keys being reported on a different one. That's why you see them in evbug
> output, but not in evtest output, because they appear under different
> event device.

I''m sure I tried that before,  but from retesting today I guess I didn't....

You're right,   the missing keys reported by evbug are showing up in evtest 
when I connect to the second event interface,  a.k.a 
usb-Logitech_USB_Receiver-event-mouse.

so it is just the ones that evbug wasn't reporting that are missing...

Looks like I need to have some fun with getting xkbd and evdev in X working 
properly.....

TIA
Kev.

Re: support for Logitech Wave

[Kevin Milburn]

On Wednesday 30 April 2008 11:41:08 Kevin Milburn wrote:

> Looks like I need to have some fun with getting xkbd and evdev in X working
> properly.....
>

OK,   I've had some success with this,   but it's left me with a few more 
questions.

I can now use all the extra keys which result in a keycode of less than 255,  
which appears to be a limitation within the Xorg evdev driver.

while experimenting,  I've tried using showkey and setkeycodes commands but 
have encountered a few issues..

Firstly,  when showkey is set to keycode mode,  which sets the ioctl KDSKBMODE 
to K_MEDIUMRAW,  it show all the expected keycodes.  setting it to show 
scancodes, which sets the ioctl KDSKBMODE to K_RAW,  results in the scancodes 
being show for all the expected keys  EXCEPT those with a keycode of 392 or 
higher  (I don't know where the actual break is as this keyboard has no 
keycodes between 236 and 392). 

Is this a known issue where these scancodes don't appear,  or is it something 
that needs looking into?

Secondly,  the primary reason for trying the above was to obtain the scancodes 
so I could alter the keycodes to work with X.  Trying to alter one of the 
working scancodes (e071) to a new keycode using the setkeycodes command 
failed with an 'Invalid Argument'.  

Further research lead me to a comment indicating that setkeycodes does not 
work on USB keyboards.   If this is correct  (which it seems to be),  is 
there any other way of altering the keycodes generated by the kernel?

TIA
Kevin

Re: support for Logitech Wave

[Kevin Milburn]

Hi Jiri

Good news everyone, 

I now have all the keys working (that are recognised) thanks to the latest 
release of keytouch and it's editor,  which appeared this morning...

This also has the advantage of negating to need to alter keycodes as keytouch 
does that for us.  so my message asking about showkeys and setkeycodes can be 
ignored.

Once you're able to provide a patch for me to test I should now be able to do 
it with ease!

Kevin

Re: support for Logitech Wave

[Jiri Kosina]

On Wed, 30 Apr 2008, Kevin Milburn wrote:

> I can now use all the extra keys which result in a keycode of less than 
> 255, which appears to be a limitation within the Xorg evdev driver.

Yes, exactly. See for example 
http://bugs.freedesktop.org/show_bug.cgi?id=11227

> Secondly, the primary reason for trying the above was to obtain the 
> scancodes 

Just a sidenote -- USB keyboards have concept of "usage codes", rather 
than "scancodes".

> so I could alter the keycodes to work with X.  Trying to alter one of 
> the working scancodes (e071) to a new keycode using the setkeycodes 
> command failed with an 'Invalid Argument'.

Newer kernels have hidinput_setkeycode() callback, which should allow 
setting the usage code <-> keycode association using 'setkeycodes'. Could 
you please post hid debugging output?

-EINVAL should be returned only in the following cases:

	- (keycode < 0 || keycode > KEY_MAX).
	- the usage code you have specified is not supported by the 
          keyboard, according to its report descriptor

(see hidinput_setkeycode() code).

> Further research lead me to a comment indicating that setkeycodes does 
> not work on USB keyboards.  If this is correct (which it seems to be), 
> is there any other way of altering the keycodes generated by the kernel?

The support for setkeycodes working with USB keyboards has been added in 
commit fe7ba31fead.

-- 
Jiri Kosina
SUSE Labs