Re: [BENCHMARK]-mfpmath=sse should disable x387 intrinsics

[Richard Guenther <richard dot guenther at gmail dot com>]

On Thu, 25 Nov 2004 14:45:20 +0100, Uros Bizjak <uros@kss-loka.si> wrote:
> Richard Guenther wrote:
> 
> >For me, specifying -mfpmath=sse,387 is 4% slower than -mfpmath=sse.  I gues
> >that 387 does not only enable 387 intrinsics but also all other
> >regular patterns?
> >
> >I would prefer the -mfpmath=sse behavior _not_ to be changed for ia32.
> >
> >
> Yes, Roger's patch is a good thing, and should stay in. Previously, you
> didn't have any switch available to turn i387 intrinsics off.

Err, but if I want the previous -mfpmath=sse behavior there is now no
way for it.
-mfpmath=sse,387 generates different code than -mfpmath=sse before.

> BTW: Did you add "-D __NO_MATH_INLINES" to compile parameters when
> compiling your application?

Yes, I did.

Also, out IP optimization could change the ABI for static functions to not have
the penalty of returning on the FP stack, testcase (compile with
-fno-inline -mfpmath=sse):

static double foo(double x)
{
  return x*x;
}
double bar(double y)
{
   y = 2*y+foo(y);
   return y;
}

Richard.