Re: [avrdude-dev] Support for Futurlec AVR development boards

[Brian Dean]

On Tue, Jan 27, 2004 at 06:56:02PM +0100, Jan-Hinnerk Reichert wrote:

> IMHO this is a good occasion to do some cleanup in the init, exit 
> code. I think there are a bit too much functions in the 
> PROGRAMMER-struct.

Sounds good.  The functions that are there are pretty much a first
pass as what I came up with back when I converted over from a purely
PPI programmer, to one that supports multiple programmer types.  I
don't doubt that I didn't quite get it right that first go around.

But the main thing to be concerned about with the PPI programmer is to
honor the -E exit specs on AVRDUDE exit.  This may or may not be
important depending on the type of programmer attached, but in
general, it is sometimes desireable to leave the port not as we found
it, but to instead ensure that /RESET is not pulled low on AVRDUDE
exit since that will keep the target from running.

Thus, the close routine needs to somehow be able to determine what
exit specs were specified on the command line so that it can do the
right thing in this case.  Perhaps a representation of the exit specs
should be passed in on the open() call, which can then be remembered
and applied appropriatedly by close(), or something similar.

Note that we lost this capability at some point and we don't currently
honor the exit specs, so this is not something that is working at the
moment.

Also, if we push all byte writing / reading, etc, down into each
programmer's functions, we can then hide the explicity LED setting as
well.  I think the only programmer that that applies to is the 'alf'
programmer which is a PPI programmer, so those functions could become
static within the PPI module.

-Brian