Re: Kernel 2.6.22 some Logitech webcams

On Friday 05 October 2007 16:18, Alan Stern wrote:
> On Fri, 5 Oct 2007, Laurent Pinchart wrote:
> > > One possibility is to have the uvcvideo probe routine check whether the
> > > audio interface is bound.  If it isn't, have the probe fail and start
> > > up a delayed workqueue routine to probe again 5 seconds later.  It's a
> > > nasty hack, but it ought to work.
> >
> > Thanks for the tip. I'll try it.
>
> Like I said, that's just a hack.

Obviously.

> A better solution would be to figure out a way for uvcvideo to work
> correctly even when usbaudio hasn't bound to the audio interface.  But
> without knowing the details of the problem, I can't suggest a way to
> accomplish that.

I've been working with Logitech on this and we haven't been able to find a 
solution. The devices are buggy and they sometimes miss an interrupt 
associated with USB control requests.

The Windows USB stack timings are such as this doesn't happen, but Linux has 
faster timings. Adding delays in the driver doesn't help, the problem is 
related to intra-request timings (time between packets). It might be possible 
to "fix" the Linux USB stack, but that would slow down all control requests 
and influence timings for all devices. Some of them might even break because 
of that.

For now the first and second generation Logitech UVC webcams are considered 
broken. The new generation has been fixed.

Best regards,

Laurent Pinchart

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
[email protected]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel