avrdude-dev
[Top][All Lists]
Advanced

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

Re: [avrdude-dev] avrdude on FreeBSD/sparc64


From: Brian Dean
Subject: Re: [avrdude-dev] avrdude on FreeBSD/sparc64
Date: Thu, 22 Apr 2004 13:06:55 -0400
User-agent: Mutt/1.5.6i

On Thu, Apr 22, 2004 at 05:03:30PM +0200, Joerg Wunsch wrote:

> The good news: after upgrading the compiler port to GCC 3.4.0 (*), the
> entire toolchain compiles on the UltraSPARC machine quite well, and so
> far it seems to work.  (As already noticed by Matthias Wei?er in the
> AVR-GCC list, GCC 3.3.x did not compile at all on a 64-bit host OS.)

Excellent!

> Now I might give avrdude a try here, on a serial device only
> (FreeBSD/sparc64 doesn't seem to support the parallel port hardware on
> that machine yet).  However, so far I get:
> 
> xcd1# avrdude -c stk500 -p t26 
> lcreat(): warning, LISTSZ[32] != sizeof(LIST)[56]
> lcreat(): warning, LISTSZ[32] != sizeof(LIST)[56]
> lcreat(): warning, LISTSZ[32] != sizeof(LIST)[56]
> lcreat(): warning, LISTSZ[32] != sizeof(LIST)[56]
> ...
> 
> That seems to be a 64-bit OS issue to me, doesn't it?  I didn't look
> into the source code though.

Actually, this is coming from my linked list code - this is probably
the first it's been compiled on a 64-bit platform.  Honestly, I cannot
remember the purpose of that check.  You can probably remove the code
that checks for that in lists.c, function 'lcreate()' and also remove
the #define LISTSZ from lists.h.  I don't see anywehre else that it is
referenced.

The only thing I can think of is that I must have had an application
that used the linked list routines and I needed the size of the LIST
elemement, probably for an array of them or something, but I did not
want to expose the actual data structure, so I used that hard-coded
size instead.  The size there looks like it might be the size of a
pointer on the platform it is being compiled on?

At any rate, I think you can safely remove the #define and the check
and it should still work fine.

-Brian




reply via email to

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