This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: PATCH: i386 target support for decimal float moves
- From: Kelley Cook <kcook at gcc dot gnu dot org>
- To: Ben Elliston <bje at au1 dot ibm dot com>, GCC Patches <gcc-patches at gcc dot gnu dot org>, Richard Henderson <rth at redhat dot com>
- Date: Tue, 20 Dec 2005 09:03:45 -0800 (PST)
- Subject: Re: PATCH: i386 target support for decimal float moves
- Reply-to: kcook at gcc dot gnu dot org
> One thing that bothers me is that it's necessary to explicitly
> pass -msse to get the target support enabled. Would it be
> reasonable to hardwire -msse if --enable-decimal-float is used?
If the rest of the patch is accepted, in my opinion, a better option
would be to also set --with-arch=pentium3 from its normal default of
i386 if --enable-decimal-float is used. For reference, config.gcc is
where this is usually set.
You would probably want to error out if the user attempts to override
this new default setting with a --with-arch or --with-tune to a lesser
processor without SSE support. Ditto for -mtune/-march in the compiled
processor.
KC