This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
RE: [patch] h8300.*: Add -ms2600 switch.
- To: "'law at cygnus dot com'" <law at cygnus dot com>
- Subject: RE: [patch] h8300.*: Add -ms2600 switch.
- From: Kazu Hirata <kazu at hxi dot com>
- Date: Tue, 15 Aug 2000 10:00:51 -0400
- Cc: gcc-patches at gcc dot gnu dot org
Hi,
> ie, it seems to me that we're going to need changes in other tools to
> distinguish between generic H8/S and H8/S2600 code. Note
> this may have
> effects on the GCC changes to support the H8/S2600.
Is it easy to support upper compatibility? One time I had a problem that
rejected the plain H8/300's code in a library when I tried to use it on the
H8/300H, which was the reason for the diff between rev 1.3 and 1.4 for
gcc/config/h8300/lib1funcs.asm.
> > -#define TARGET_H8300S (target_flags & 1)
> > +#define TARGET_H8300S ((target_flags & 1) ||
> (target_flags & 2))
> ?!? I don't understand this, it seems like you should have
> left this alone.
Are you considering a scheme where you specify -ms if you are using *any*
H8/S, and you additionally specify -ms2600 if you want to enable the mac
stuff? I was thinking of it as specifying one of -ms and -ms2600. Actually
-ms sounds like any H8/S. Should I just have an additional switch, -ms2600,
use it to turn on the predicate, TARGET_H8S2600, and in turn use the
predicate to enable the mac register and relevant instructions? I hope this
idea is somewhat fair.
Kazu Hirata