This is the mail archive of the
mailing list for the GCC project.
Re: -m32 only ia32 compiler?
- To: Jakub Jelinek <jakub at redhat dot com>
- Subject: Re: -m32 only ia32 compiler?
- From: Jan Hubicka <jh at suse dot cz>
- Date: Fri, 4 May 2001 13:04:46 +0200
- Cc: rth at redhat dot com, aoliva at redhat dot com, jh at suse dot cz, gcc at gcc dot gnu dot org
- References: <20010503194906.I1863@sunsite.ms.mff.cuni.cz>
> With the recent 32->64bit cross patch suddenly HOST_WIDE_INT is long long on
I've noticed too.
> ia32 native, eventhough 99.999% people don't need -m64 support on ia32.
> Shouldn't there be separate configuration for i386-*-* (which would create a
> -m32 only compiler) and x86_64-*-* (which would create a -m32/-m64
> compiler)? On the SPARC, the compiler can be configured both to support only
> -m32 and -m32/-m64 (and actually in the distribution we use a driver hack,
> so that for -m32 compilations it looks into a special -B path where it finds a
> -m32 only cc1/cc1plus).
Yes, I plan to create ia32 only configuration too. Everything is handled by TARGET_64BIT
macro, so making 32bit only compiler should be easy.
I will give it priority now, since performance penalty due to HOST_WIDE_INT is high
now, so I will try to send patch till end of weekend.