From: Hans-Bernhard Broeker <broeker AT physik DOT rwth-aachen DOT de>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Prolog
Date: 1 Aug 2002 10:24:53 GMT
Organization: Aachen University of Technology (RWTH)
Lines: 16
Message-ID: <aib29l$kjd$1@nets3.rz.RWTH-Aachen.DE>
References: <zdZ19.3130$6j5 DOT 101432 AT newsfep2-win DOT server DOT ntli DOT net> <3d48a1d8$0$1421$272ea4a1 AT news DOT execpc DOT com> <b6729.2752$U44 DOT 166519 AT newsfep2-gui>
NNTP-Posting-Host: acp3bf.physik.rwth-aachen.de
X-Trace: nets3.rz.RWTH-Aachen.DE 1028197493 21101 137.226.32.75 (1 Aug 2002 10:24:53 GMT)
X-Complaints-To: abuse AT rwth-aachen DOT de
NNTP-Posting-Date: 1 Aug 2002 10:24:53 GMT
Originator: broeker@
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

Lawrence Rust <lvr AT nospam DOT softsystem DOT co DOT uk> wrote:

> The problem is that the source is part of a library that will be compiled by
> others and I can't guarantee that the final makefile will specify
> '-fomit-frame-pointer'.  The function modifies the sp register so it's
> important that the calling stack frame format is known.

Hmm... this feels like your best bet should be to stop having this
assembly-only function sit in the middle of a C source file in the
first place.  Move it to a separate assembly source file and GCC will
no longer be an issue.  Or, if it really is so short that even the C
default prologue and epilogue are an inacceptable overhead, consider
making it an inline.
-- 
Hans-Bernhard Broeker (broeker AT physik DOT rwth-aachen DOT de)
Even if all the snow were burnt, ashes would remain.