This is the mail archive of the
gcc-help@gcc.gnu.org
mailing list for the GCC project.
RE: AS -force-long-branchs for all processors
- From: Kristis Makris <kristis dot makris at asu dot edu>
- To: Bansidhar Arvind Deshpande "- CTD, Chennai." <bansidhara at ctd dot hcltech dot com>
- Cc: gcc-help at gcc dot gnu dot org
- Date: 03 Oct 2003 11:11:45 -0700
- Subject: RE: AS -force-long-branchs for all processors
- References: <8DFBF35EE709D346B80B28E517B9024254F937@HARITHA.ctd.hcltech.com>
Well, on x86 the "call" instruction for function calls supports absolute
addressing, but still "-force-long-branchs argument" won't work in
producing x86 machine code. In fact if you try to assemble something for
x86 with that argument with gcc 2.95.4 the source won't compile.
I thought that absolute call addressing is a rather common feature in
processors(maybe it's not). How come it only made it in gcc for the
m68HC11/12s ? Isn't that odd ?
On Thu, 2003-10-02 at 21:06, Bansidhar Arvind Deshpande - CTD, Chennai.
wrote:
> Yes, Actually this is processor related issue. If ur processor
> supports relative jumps then only this option makes sense as
> in case 68CH11, 68HC12 processors.
>
> Is there a way to tell GCC to generate code that includes relative
> references for jumps inside a function(e.g. in che case of loops), but
> absolute references for calls to other functions ?
>
> I've see the -force-long-branchs argument on AS, but that seems to be
> available only for the 68CH11, 68HC12 processors.
>
> Thanks,
> Kristis