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

[MiNT] uname reply - a final proposition



I've just posted this, but since some people might not have followed the
thread on Apache and how we got there, here's the short version:

Basically, it got started when I complained that Apache would not compile
correctly, because the GuessOS script could not recognize the output sent
by uname (different ports/versions of uname within the MiNT community may
give different responses for the same system - this is the main problem).

Anyhow, 3 different persons on this list are trying to agree on standard
replies for uname, so we could produce a white paper on porting stuff for
MiNT and also generate a better identification of the platform/OS for the
authors of Apache, GNU utils, Perl, SSH, etc. in the future.

The approach below is as best of a standard I could think of and why.

---------- Forwarded message ----------
Date: Wed, 3 Feb 1999 13:48:08 +0200 (EET)
From: Martin-Eric Racine <q-funk@pp.fishpool.com>
Subject: Re: [MiNT] uname - was sending diffs to Apache ...

> I'd vote for using m68k-atari-mint (perhaps followed by the release
> number) as the canonical system name.  How that is extracted from uname
> output is only a minor point.

Sounds good.

One approach that would work for MiNT:

-m  m68k
-p  68000 or 68020 or 68030 or 68040 or .....
-r  MiNT version
-s  MiNT
-v  TOS version

This way, -machine and -processor return more logical values, which
could be used as switches to compile proper ST, TT/Falcon or 020/040
binaries, in Makefile and Configure scripts.

Does this one make sense for everyone, finally?

PS:  Atari Corp. is gone, so is the copyright on MiNT still an issue?
     Can't we just finally use the more generic "MiNT" instead of
     "Fresh/Free/Friggin/FryingMiNT" to identify our OS??

----------------------------------------------------------------
  Martin-Eric Racine * http://www.pp.fishpool.com/~q-funk/M-E/
  The Atari TT030 Homepage * http://members.tripod.com/~TT030/
----------------------------------------------------------------