Date: Sun, 12 Jul 1998 19:22:48 +0300 (IDT)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
To: Dave Love <d DOT love AT dl DOT ac DOT uk>
cc: Andris Pavenis <pavenis AT lanet DOT lv>,
        "Gurunandan R. Bhat" <grbhat AT unigoa DOT ernet DOT in>,
        Alexey Yakovlev <jack AT catalysis DOT nsk DOT su>, djgpp-workers AT delorie DOT com,
        fortran AT gnu DOT org
Subject: Re: Inconsistencies between g77 v0.5.23 and v0.5.19
In-Reply-To: <rzqg1g84moo.fsf@djlvig.dl.ac.uk>
Message-ID: <Pine.SUN.3.91.980712192223.27114O-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Precedence: bulk


On 11 Jul 1998, Dave Love wrote:

>  AP> I think it should be treated as bug in f/runtime/configure.in
>  AP> as DJGPP defines both unix and __MSDOS__ 
> 
> I'm not really sure `bug' is fair :-), but perhaps it needs working
> round in the same way as the Windows versions (although cygwin32
> doesn't need that for binary mounts AFAIK).

IMHO, it *is* a bug.  It is not nice to rely on an assumption that
only compilers which produce native Unix code define `unix'.  DJGPP
shows that `unix' can be defined by a compiler which targets other
systems.

In general, it is best to test for a feature rather than for a name of
an operating system.  Perhaps that ifdef could be rewritten so that it
tests for the actual functionality?

> Does `port' mean that there are other changes to the source necessary
> that we (fortran) don't know about, or just mean the build on DJGPP?
> If other changes are needed, I expect we'd be happy to have them.  For
> instance, does the build now work with the normal configure under
> DJGPP bash or are (fixes to) the .bat files still needed?

Since the patch is for configure.in, the DJGPP version *must* be built
by running the configure script.

The DJGPP port of Autoconf usually takes care of other DOS-isms (such
as looking for executables with "test -x" instead of "test -f").