[comp.os.os2.misc] OS/2 port of gcc/g++?

seg@ingres.com (scott e garfinkle) (06/21/91)

In article <1991Jun20.113229.27552@brolga.cc.uq.oz.au> ant@brolga.cc.uq.oz.au (Anthony Murdoch) writes:
>strasser@psych.psy.uq.oz.au (Michael Strasser) writes:
>More importantly - is someone going to look at porting gcc/g++ to OS/2 v2
>when it is released properly ??  It would be soooo much better if it
>produced 32bit code

There are other issues that would need to be addressed, too.  The most important
one is the treatment of guard page and FP exceptions and automatic generation
of stack probes for stack frames >= 4k bytes (see new .INF doc on
DosCreateThread).  There's also the question of where you get a runtime
library for the beastie --especially one that is fullr reentrant.  On the whole,
I think the port would be pretty straightforward; however, the runtime
issues would be killer.  On the whole, I think I'll just go ahead and use
the new IBM OS/2 compiler when it's released pretty soon.
	-scott e. garfinkle

james@bigtex.cactus.org (James Van Artsdalen) (06/23/91)

In <1991Jun20.113229.27552@brolga.cc.uq.oz.au>, ant@brolga.cc.uq.oz.au
	(Anthony Murdoch) wrote:

> More importantly - is someone going to look at porting gcc/g++ to
> OS/2 v2 when it is released properly??  It would be soooo much better
> if it produced 32bit code

I ported gcc1 to use Intel's 386 syntax.  Unfortunately, at that time
I could find no Intel-syntax assembler that worked in `use32' mode,
including the OS/2 2.0 assembler.

The port of gcc to produce Intel-syntax is easy.  The problem is that
there isn't much use for the output yet.
-- 
James R. Van Artsdalen          james@bigtex.cactus.org   "Live Free or Die"
Dell Computer Co    9505 Arboretum Blvd Austin TX 78759         512-338-8789