This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: The perennial i386 floating-point rounding issue...
- From: Jan Hubicka <jh at suse dot cz>
- To: Richard Henderson <rth at redhat dot com>,Stephen L Moshier <steve at moshier dot net>,mike stump <mrs at windriver dot com>, gcc at gcc dot gnu dot org,zack at codesourcery dot com
- Date: Tue, 11 Jun 2002 12:21:00 +0200
- Subject: Re: The perennial i386 floating-point rounding issue...
- References: <Pine.LNX.4.33.0206101956380.15319-100000@moshier.net> <20020610175614.A641@redhat.com>
> On Mon, Jun 10, 2002 at 08:03:26PM -0400, Stephen L Moshier wrote:
> > Or maybe we should offer a way to set it ourselves.
>
> This is (in theory) very easy to do. See optimize_mode_switching
> and NORMAL_MODE. It should be relatively easy to modify the
> existing optimize_mode_switching macros to add bits to control
> the rounding mode.
Another problem with this is that glibc floks will be really unhappy, as
they were when I made the mathematics correct for SSE math.
Does the idea of setting the FLT_* macro (forgot exact name) at
compilation time and possibly making SSE fp the default when -march
allows sound sane?
Honza
>
> I would be very open to a patch that optionally enabled setting
> the rounding precision bits for the computation.
>
> And on a related point, while I may have once argued against
> spilling in 80-bit mode (when not controling computation precision),
> I've been convinced that it's a mistake. So if someone wants to
> work on that as well, please do so. It'll be a bit harder though...
>
>
> r~