[libhid-discuss] Some help?

Charles Lepple clepple at ghz.cc
Thu Aug 2 20:04:34 UTC 2007


Justin Williams wrote:
> Charles Lepple wrote:
>> I would guess that the Linux kernel HID driver wouldn't expose this as
>> a keyboard (it uses a non-keyboard usage page: 0xff00), but you might
>> also be able to read events from the evdev interface
>> (/dev/input/event* or something).
>>
>> This appears to be the catch-all for non-standard usage pages:
>>
>>    http://lxr.linux.no/source/drivers/hid/hid-input.c#L610
> I did this... And none of the event* devices show anything with the
> swiper... They are all either the touchscreen, the keyboard, or the
> extra mouse.

Worth a try, I guess.

>> As far as the usage path to use, what do you get when you modify the
>> sample code to connect to your device? It should dump a list of usage
>> paths for the report descriptor.
>>
> The output from test_libhid, when modified to connect to this device is:
> WARNING: hid_get_input_report(): failed to retrieve report from USB
> device 003/002[0]:error sending control message: Broken pipe.
>   TRACE: hid_get_input_report(): looking up report ID...
>   TRACE: hid_prepare_parse_path(): preparing search path of depth 3 for
> parse tree of USB device 003/002[0]...
>   TRACE: hid_prepare_parse_path(): search path prepared for parse tree
> of USB device 003/002[0].
>  NOTICE: hid_find_object(): found requested item.
>   TRACE: hid_get_input_report(): retrieving report ID 0x00 (length: 255)
> from USB device 003/002[0]...

There should be a call in there (maybe it's commented out?) to dump the
HID report descriptor.

> Also, what exactly do you mean regarding the page on the usb site?

It's documentation about HID, and it would be good to read that class
document.

-- 
Charles Lepple





More information about the libhid-discuss mailing list