Re: [avrdude-dev] [bug #30498] Add support for new avr devices (atmega324pa)

[Joerg Wunsch]

(Not filed to the bug tracker, just to the list only.)

As jonathan nifenecker wrote:

> Follow-up Comment #1, bug #30498 (project avrdude):
> 
> part definition for :
> atmega164a
> atmega164pa
> atmega324a
> atmega644a
> atmega644pa
> atmega1284

I somewhat hesitate to duplicate entries for all the "A" devices.  The
entire avrdude.conf file already suffers from being about half a
megabyte in size, and the "A" devices are supposed to be
plug-compatible counterparts to the older non-"A" devices, so the new
entry will become a complete duplicate of an existing entry.

Right now, the search for a particular device match is done in
avrpart.c based on their "id" or "desc" field:

  for (ln1=lfirst(parts); ln1 && !found; ln1=lnext(ln1)) {
    p = ldata(ln1);
    if ((strcasecmp(partdesc, p->id) == 0) ||
        (strcasecmp(partdesc, p->desc) == 0))
      found = 1;
  }

My suggestion is to add an "alias" field to the config syntax, which
by itself is threated as a list of aliases, and is also searched.
That way, devices that are completely identical to other devices
(including their signature) can be mentioned that way, like:

part
    id               = "m88p";
    desc             = "ATMEGA88P";
    alias            = "ATMEGA88PA", "ATMEGA88PV";

(In case someone would want to be diligent enough, we could add all
the different suffix variants like "V" or "L" that way.)

I don't think there's a point in adding more abbreviations (the "id"
field) for the "A" devices: the abbreviation is only useful for humans
to save some typing work, so it makes sense to keep them at the
shortest string possible that will select a particular entry.

Opinions?
-- 
cheers, J"org               .-.-.   --... ...--   -.. .  DL8DTL

http://www.sax.de/~joerg/                        NIC: JW11-RIPE
Never trust an operating system you don't have sources for. ;-)