[OLPC-devel] Re: Build image 76.

[Jordan Crouse]

On 24/08/06 14:28 -0500, Richard Smith wrote:
> >> From a quick look, it seems that GPIO18 was already enabled and
> >> toggling it doesn't make the device come up.  I could be doing
> >> something wrong, or this might not be where our problem is.
> >
> >I just checked L21 and it had 3.3V on it so... Rats.  Thats not it.
> >
> >What about P4 USB is it enabled?
> 
> Another possibility is IRQ routing.  Can the southbridge have
> different IRQs for the USB ports and we don't have one routed
> correctly?

This is what I was trying to get at last night, but people kept confusing
the issue.  We have not verified that EHCI is correctly routed.  There are
two easy ways to check that.  Don't load EHCI at boot time, and see if the
device gets enumerated (per USB requirements, the device must talk at 1.1
speeds, even if it would prefer not to).  If it does, then we've found our 
culprit.  On the other hand, since we know that OHCI is good, if it doesn't
enumerate without EHCI, then obviously something else is wrong.

Like I just said, I doubt that Insyde went out of their way to make wireless
work - so we need to stop looking for secret sauce that only affects the 
wireless and start looking at issues on the fundamental busses at play here.
Step one is to get rid of EHCI.

Jordan

-- 
Jordan Crouse
Senior Linux Engineer
Advanced Micro Devices, Inc.
<www.amd.com/embeddedprocessors>