Re: [Libusb-win32-devel] Extra endpoints on HID mouse device

[<li...@ze...>]

* On 2006-03-27 Stephan Meyer <ste...@we...> wrote  :

> I'm no HID expert, but I think what you're trying to accomplish can
> be done completely without libusb-win32:
> 
> Solution 1): Add feature reports to your mouse. So that you can use
> Windows' HidD_SetFeature() and HidD_GetFeature() API functions
> to send vendor specific commands to your device (I haven't done this
> on my own, so I don't know if this solution actually works).

You are right; reading the HID spec more closefully, I think this is the
right thing to do. 

My only problem is that I am using a very inflexible USB firmware
library from Atmel in my device, which does not seem to support feature
request. Maybe rewriting (part of) this library is the best solution
after all, since there are more limitations to it that I run into every
now and then.

I will still need one bulk OUT enpoint though, for the purpose of
firmware uploading.

Thanks,

Ico

> 
> > 
> > I am developing an USB pointer device. The device acts as a 'normal'
> > UBS mouse, so it is recognized by every OS without using special
> > drivers. This all works fine.
> > 
> > I need to be able to change configuration settings on the device as
> > well. For this I added an extra OUT endpoint where I can send special
> > commands to; this is used for configuring the device and uploading new
> > firmware.
> > 
> > With linux, the HID mouse is claimed by the usbhid driver right away,
> > but using the usb_detach_kernel_driver_np() function I can still take
> > control over the device. libusb-win32 does not support this call,
> > however.
> > 
> > Summarizing: I want the normal windows driver to handle my mouse, but
> > still be able to talk to the other endpoints using libusb.  What would
> > be the best way to handle this with libusb-win32 ?
>  

-- 
:wq
^X^Cy^K^X^C^C^C^C