On Wed, May 03, 2006 at 01:22:00PM -0400, antiright@gmail.com wrote:
> On Thu, Apr 27, 2006 at 09:48:13AM -0700, Bill Studenmund wrote:
> > Wouldn't it be better to increase the timeout time?
> > 
> > Either there is a longer, more-correct timeout that will work, or we're 
> > doing something wrong in the probe which results in the z22 not 
> > responding.
> 
> I will look for a more effective solution following your suggestion.
> My patch works, but it is obviously a hack, in that it does not directly
> address the cause of the problem, but rather prevents the timeout error
> from being issued.
> 

I have tried adjusting the timeout period (as defined in usbdi.h).
The same TIMEOUT error occurrs, regardless of the time used.
So, I would suspect the latter option mentioned, of something being wrong
with the probe, to be the case.  I think that the best solution would
be to isolate the error-checking code causing the false positive for a
TIMEOUT, and simply scrap it.  (I've also had better results with the
timeout period *reduced*, since ucom is only attached after the period
has passed.)

Jeff