[ale] Q: RTFM pointer to GCC calling conventions on x86 targe ts

[russell.enderby at arris-i.com (Russell Enderby)]

Speaking of gcc.  Does anyone know how todefine the stack size for the application
instead of creating your own internal stack?  Similar to how borland does it under
dos/windows.

Thanks,
Russell

Dunlap, Randy wrote:

> pointer:  http://www.delorie.com/gnu/docs/
>
> This is the best that I've been able to find,
> although gcc doc is not up to date here.
>
> Good luck.
>
> ~Randy
>
> > From: John M. Mills [mailto:jmills at tga.com]
> >
> > Where can I check argument passing by gcc-2.95.2 in function calls and
> > returns for Intel Pentium uP's? I need to work on C and C++
> > sources with
> > in-line assembly for M$VC++ and try to come up with definitions or a
> > strategy so they can alternately be compiled by 'gcc' or
> > VC++, while being
> > maintained as a single set of sources.
> >
> > Executables will then run in LynxOS, for which we will use
> > Lynx's version of cross-gcc (MSWin to x86-lynxOS).
> >
> > Also, can 'gcc' produce assembly code for 'as86' (and by
> > implication allow
> > inclusion of Intel-syntax assembly code)?
> >
> > What does this imply about linkage? (Will I need to use ld86?) Can I
> > produce libs from Intel-syntax source which then link
> > properly to objects
> > from "generic" 'gcc' compilation of C/C++ source?
> >
> > I expect this is a guaranteed 'bucket of worms', but I also
> > expect some
> > folks have had to fight it through. Fortunately we're talking about a
> > small number of assembly lines (but buried in _lot_ of C/C++).
>
> --
> To unsubscribe: mail majordomo at ale.org with "unsubscribe ale" in message body.

--
To unsubscribe: mail majordomo at ale.org with "unsubscribe ale" in message body.