This is the mail archive of the
mailing list for the GCC project.
Re: PowerPC -- support single-precision FPU
- From: "Andrew Pinski" <pinskia at gmail dot com>
- To: "Michael Eager" <eager at eagercon dot com>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Fri, 19 Sep 2008 16:35:20 -0700
- Subject: Re: PowerPC -- support single-precision FPU
- References: <48D3E488.firstname.lastname@example.org>
On Fri, Sep 19, 2008 at 10:42 AM, Michael Eager <email@example.com> wrote:
> The attached patch adds generic support for single-precision
> FPU on PowerPC. A follow-on patch will use this patch to add
> support for the single-precision FPU on the Xilinx PPC 405/440.
I noticed that "TARGET_SINGLE_FLOAT && (MODE) == SFmode" and
"TARGET_DOUBLE_FLOAT && (MODE) == DFmode" are used in more than once
place, could there be a macro that is defined to those?
Likewise for "((TARGET_HARD_FLOAT && TARGET_DOUBLE_FLOAT) ? DFmode : SFmode)"
Should floatsisf2_internal have the same constraints as floatdidf2?
These three small issues are only the issues I have with the patch.
Everything else looks good (though I cannot approve it).