Re: [avrdude-dev] Yet another config parameter

[Joerg Wunsch]

As Brian Dean wrote:

> If you want a config file parm to do some alternate behaviour, let me
> know and I'll add it (or someone else can - just look at the changes I
> recently made to add several new ones for an example).

Just to make sure we match what the datasheets say, i'd rather add the
option to either pulse /RESET or SCK to retry the programming enable.
I think i can do that myself.

> Are you saying that with the debugging code in place that it works,
> but without the debugging code it doesn't work?

No (but i can't deny that option either).  It worked when Eric changed
his code again to pulse SCK, as it is mandatory for the m163.  All the
other chips didn't have problems either way, they obviously always
succeed in the first attempt.

> I don't know how
> fast one can pulse a parallel port line using in/out instructions, ...

In theory, the parallel port is on the ISA bus, which has a "classic"
timing requirement of something like 1.25 µs for any 8-bit IO cycle.
But i'm not sure whether this is really still true for
state-of-the-art all-in-one Super-IO chips, even though the address
range covered is still within the ISA hole.

> > reliably at all with programmer-supplied power...  Remember, that's
> > all on the same notebook, with the same programmer hardware.
> 
> I think your laptop does not supply enough power.

The strange thing though is, under Windows, it provides enough power
to drive that m163, and there, using the port-supplied power is the
only way to make that m163 enter programming mode at the first try,
while with external power, it only enters programming mode after ~ 30
attempts.

> Didn't you have
> trouble with that before?

That's been with the Toshiba Libretto i normally use for my AVR
experiments.  This time, since Windows was required, i used the
(newer) Thinkpad 600.  I can see from the LED on my '1200 board that
the Thinkpad is driving a lot more power out of its parallel port.

> The other option would be to twiddle the I/O registers
> directly like Eric is doing for the Windows port, but that would
> require installing avrdude setuid root in order to gain access to
> them.

I would start with that to see how fast it could be that way.  If it
sounds like a big win, i'd start thinking more seriously about a
driver.  I probably wouldn't want to ship an official (Unix) version
that does direct port IO, for security concerns.

-- 
J"org Wunsch                                           Unix support engineer
address@hidden        http://www.interface-systems.de/~j/