Re: Logitech M705 thumb button

[Robert Schedel <r.schedel@yahoo.de>]

Hello,

On 02.04.2010 13:27, Octoploid wrote:
[...]
>>> Could you please verify that there is indeed no USB event being triggered
>>> by the button press at all, with usbmon [1]?
>>
>> I already verified this, but forgot to mention it in my last mail.
>> There are indeed no USB events being triggered when I run
>> "cat /sys/kernel/debug/usb/usbmon/ou".
>>
>>> If it is the case, I am afraid the only option would be inspecting what
>>> initialization sequence does the driver for Another OS perform, so that it
>>> makes the button work.
> 
> OK, I have downloaded a trial version of a usb-monitor and
> recorded the following initialization sequence.
> Let me know it this helps.
> 
> http://pastebin.org/131184

You did not add info at which time you pressed which button, e.g. thumb button.

But, OK, init sequence is more important: Looking for "(DOWN)", there are several device-specific control blocks which resemble those sent by "lomoco" or other mouse tools for configuring Logitech features.

Example, request 0x9, value 0x210, sent downstream:

# 002448: Control Transfer (DOWN), 02.04.2010 11:14:55.341 +0.0
# Pipe Handle: 0x0
# Recipient: Interface
# Request Type: Class
# Direction: Host->Device
# Request: 0x9 (Unknown)
# Value: 0x210
# Index: 0x2
# Length: 0x7

My guess is, some new init trigger is required from a userspace tool (only) to enable thumb button, no kernel issue. This is difficult to analyze though w/o the device available and w/o spec. You might
a) fiddle with lomoco source code yourself, e.g. making the existing features work first + adding feature bits found in your protocol
b) get in touch with the developers
c) wait for someone more experienced to get the mouse and analyze

Regards,
Robert Schedel