Re: weird optimization in sin+cos, x86 backend

[James Courtier-Dutton <james dot dutton at gmail dot com>]

2012/2/9 Andrew Haley <aph@redhat.com>:
> On 02/09/2012 01:38 PM, Tim Prince wrote:
>> x87 built-ins should be a fair compromise between speed, code size, and
>> accuracy, for long double, on most CPUs. ÂAs Richard says, it's
>> certainly possible to do better in the context of SSE, but gcc doesn't
>> know anything about the quality of math libraries present; it doesn't
>> even take into account whether it's glibc or something else.
>
> Yeah. ÂOn x86_64 glibc, we should really get the sincos bg fixed. ÂAnd
> it really is a bug: cos() and sin() are correctly rounded but sincos()
> isn't.
>
> Andrew.

Given:

gcc -g -O0 -c -o sincos1.o sincos1.c
gcc -static -g -o sincos1 sincos1.o -lm

./sincos1
sin = -8.52200849767188795e-01    (uses xmm register intructions)
sinl = 0.46261304076460176          (uses fprem and fsin)
sincos = 4.62613040764601746e-01 (uses fprem and fsin)
sincosl = 0.46261304076460176       (uses fprem and fsin)

It looks to me that only sin() is evaluation correctly using xmm
register instructions.
sinl, sincos, and sincosl all use fprem and fsin and fsincos.

From my analysis, fprem is causing a lot of the problems.
Using a combination of xmm register instructions instead of fprem
might cure our problems with fsincos.
Alternatively, make both sin() and sinl() use xmm register intructions.