Re: [PATCH] (Attempt to) Fix link compatibility check (was: Re: recent mips-elf linker "architecture ... incompatible" regressions)

[Thiemo Seufer <ica2_ts at csv dot ica dot uni-stuttgart dot de>]

Eric Christopher wrote:
> 
> > mips-foo-gcc -mips64         -Dnormal -o sffp_normal.o super-fast-fp-rtns.c
> > mips-foo-gcc -mips64 -mips3d -Dmips3d -o sffp_mips3d.o super-fast-fp-rtns.c
> > 
> > ?
> > 
> 
> Or how about this:
> 
> mips-foo-gcc -mips64 -mips3d foo.c
> mips-foo-gcc -mips64 -mips3d bar.c
> 
> mips-foo-gcc foo.o bar.o -o a.exe
> 
> OK, so perhaps the ASEs aren't a good example at the moment, but if you
> start having gcc vectorize or start using the 3d instructions when you
> pass -mips3d then you start wanting it in the headers...

AFAICS ASE's are designed to be used together with a base architecture,
so there is not much point in warning about ASE flags as long as the
ASE isn't incompatible to the base arch (Is there any of this sort?).
IMHO it's best to keep the union of all ASE flags in the final
executable.

Checking CPU type is a different thing because of subtle
incompatibilities. AFAICS we should either (try to) warn about any
incompatibility or we shouldn't make any attempt on it.


Thiemo