From owner-freebsd-hackers  Fri Jun 21 21:05:24 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id VAA26113
          for hackers-outgoing; Fri, 21 Jun 1996 21:05:24 -0700 (PDT)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id VAA26101
          for <freebsd-hackers@FreeBSD.ORG>; Fri, 21 Jun 1996 21:05:20 -0700 (PDT)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.12/8.6.9) id OAA16873; Sat, 22 Jun 1996 14:00:32 +1000
Date: Sat, 22 Jun 1996 14:00:32 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199606220400.OAA16873@godzilla.zeta.org.au>
To: bde@zeta.org.au, kuku@gilberto.physik.rwth-aachen.de
Subject: Re: TARGET_NO_FANCY_MATH_387
Cc: freebsd-hackers@FreeBSD.ORG, mrm@MARMOT.Mole.ORG
Sender: owner-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

>Could you give an example showing that 

>-ffancy-math-386 and -ffast-math

>have an effect? I tried with a small example and I'm always getting
>calls instead of fpu-inline statements. (gcc-2.6.3)

>#include <math.h>
>main()
>{
>    double d=2500.;
>    printf("%lf\n",sqrt(sin(d)));
>}

>gcc  -ffast-math -m486 -mfancy-math-387 -S p.c

This too is an example if you compile with -O.

Bruce