Who knows what the story is there ... according to the
USB spec, iso endpoints can't halt, so it's pointless
to even try.
So why does that driver try? Is that product breaking
the spec, or is that logic not necessary?
You probably refer to this:
usb_clear_halt(uvd->dev, usb_rcvisocpipe(uvd->dev, uvd->video_endp));
And its siblings, yes.
This request was observed being issued by the original Windows driver. I
noticed that this very request, illegal and pointless it might be, is
issued by the Windows driver each time the camera is reinitialized. I do
not know why. Maybe the device recognizes it somehow. Without it I had
troubles initializing the Model 1 camera. Don't know if other models
work better.
Hmm, curious. That's good to know, and it might be part of the
reason why that usb_clear_halt() check was commented out: if not
commented out, it'd fail unless the endpoint was halted. And no
HCD has a mechanism to flag iso endpoints as halted.
- Dave
-------------------------------------------------------
This sf.net email is sponsored by:
Access Your PC Securely with GoToMyPC. Try Free Now
https://www.gotomypc.com/s/OSND/DD
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
