avrdude-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [avrdude-dev] avrdude release?


From: Theodore A. Roth
Subject: Re: [avrdude-dev] avrdude release?
Date: Thu, 15 Jul 2004 12:47:54 -0700 (PDT)

On Thu, 15 Jul 2004, Bob Paddock wrote:

> >> CVS checkout on Jun/28/2004 worked.  See this thread:
> >> http://lists.gnu.org/archive/html/avrdude-dev/2004-06/msg00052.html
> >> For starters how do I get ppiwin.c-Jun/28 from CVS?
>
> > So the last change before mine was on June 24th by Alex. I'd say try
> > reverting my patch ('patch -p0 -r < foo.diff') and see it the problem
> > goes away.
>
> That should be a "-R" I assume?

Yes, brain fart on my part. Sorry.

>
> It did not go away.  Sometimes the first or the last byte of the signature
> would
> be correct, never the middle byte.
>
> As a shot in the dark I put in:
>
> us = (unsigned long) (us * 256);
>
> at the start of usleep() in ppiwin.c.  Now the signature always reads
> the wrong 0x000102, which is what you get if the part is locked (it is not
> locked).
> Smaller values gave more random results, but less random than not putting
> in something.

That doesn't make any sense to me. Why would slowing down the SPI clock
make it fail more reliably?

>
> > Although, looking over my change diffs just now, I don't see how I could
> > have broken it since I only changed debugging code that was #if'd out.
>
> That took out a long that was being put on the stack which would change
> the timing,
> but if it is that touchy it needs more serious help.

It's still there, it just got moved into the DEBUG_DisplayTimingInfo()
macro. It was also #if'd out before, so it shouldn't be affecting
anything here. But yeah, the timing should not be that critical.

>
> Anyone have suggestions?

Is there any way to can try using an external clock instead of the
internal one and see if that makes a difference? Kind of a "chicken and
egg" problem though if you have to use the dongle to change the fuses.
:-(

I wish I had one of those chips just to see if the problem is specific
to your setup or if it a problem in the silicon.

I assume you haven't had a chance to try programming it with a
jtagice-mkII and AvrStudio?

Is the pony dongle the only programmer you have?

---
Ted Roth
PGP Key ID: 0x18F846E9
Jabber ID: address@hidden




reply via email to

[Prev in Thread] Current Thread [Next in Thread]