This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [patches] Re: SSE min/max instruction support
> On Sun, Feb 25, 2001 at 12:25:24AM +0100, Jan Hubicka wrote:
> > Whats also about explicitly documenting the i386 behaviour for smin/smax
> > FP, so we can use them in IEEE compilation?
>
> Not possible. Naturally the ia64 definition is the same
> as the i386 definition. But rs6000 uses
>
> tmp = a - b;
> result = (tmp >= 0 ? a : b);
>
> where the >= is hardcoded in "fsel %0,%1,%2,%3". We could
Hmm, thats what I was affraid of.
> perhaps add a target macro that describes which operand is
> selected in the case of equality or NaN.
Or fall back to my solution to detect it in the md file expander
in IEEE case.
What solution look cleaner to you? If we want macro, what should be
the name?
Honza
>
>
> r~