Hi,
> are you aware of the legousb project at sourceforge
> (http://sourceforge.net/projects/legousb)?
> It would be nice to pool efforts to get a well-working driver.
> The project is in need of a little pushing,
> which I unfortunately didn't have time to do recently.
Thanks, I'll have a look for this.
> As for your problem, I'm not much of a kernel guru,
> but it seems to me that your completion code deadlocks
> somehow and needs to be interupted.
> Without the code it's difficult to say more.
I don't think that it's deadlocking - the whole thing runs
quite happily without the printk() statements (ie I don't get lock ups
or anything), its just that the data sent gets lost/corrupted somehow
(or so I assume, since the high level comms protocol then fails).
It sort of looks as though some defect in the low-level usb
code (or the physical device ?) means that sending a second urb too
quickly after the first causes a problem - but if that's the case, why
would the printk() solve the problem when used in the completion code,
but not when used in the write() call ?
Cheers,
Richard
_______________________________________________________________
Don't miss the 2002 Sprint PCS Application Developer's Conference
August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
