Date:	Tue, 30 Sep 1997 09:00:58 +0700 (GMT)
From:	Alexey Yakovlev <jack AT catalysis DOT nsk DOT su>
To:	"Jude T. DaShiell" <jdashiel AT eagle1 DOT eaglenet DOT com>
cc:	djgpp AT delorie DOT com
Subject:	Re: g77 and functions
In-Reply-To:	<199709260155.VAA27568@eagle1.eaglenet.com>
Message-ID:	<Pine.GSO.3.96.SK.970930081959.11915A-100000@quant.catalysis.nsk.su>
MIME-Version:	1.0

On Thu, 25 Sep 1997, Jude T. DaShiell wrote:

> One problem with g77 for me is
> none of the functions appear to have
> any documentation in info.

 g77-0.5.21 info (don't know how about earlier version) contains
documentation on intrinsic functions and subroutines. (go to *Language::
section and then to *Functions and Subroutines:: and then to *Table of
Intrinsic Functions::).

> So I'm interested to know if
> djgpp followed a standard for its functions
> which machine architecture did it emulate?

 It depends on the g77 version. Since version 0.5.20 libf2c.a includes
Dave Love's U77 library. All timing subroutines and many unix specific
ones are there. I'm not sure that they will all work under DJGPP but,
actually, most of them are just stubs for corresponding C functions. If C
function works then Fortran's should do as well. For more details download
the latest g77 sources from any gnu mirror and look in the
f/runtime/libU77 subdirectory.
 BTW, you can get g77-0.5.21 binaries for DJGPP from 
http://www.catalysis.nsk.su/~jack/g770521b.zip OR
ftp://quant.catalysis.nsk.su/pub/djgpp/g770521b.zip

Regards
Alexey

================================================================
Alexey Yakovlev, Quantum Chemistry Lab.
Boreskov Institute of Catalysis, Russian Academy of Sciences
pr. Lavrentieva 5, 630090 Novosibirsk, Russia
phone: +7-(3832)-350264, fax: +7-(3832)-355766 or 357687
e-mail: jack AT catalysis DOT nsk DOT su
================================================================

- Raw text -