This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Calculating cosinus/sinus
- From: Robert Dewar <dewar at adacore dot com>
- To: jacob at jacob dot remcomp dot fr
- Cc: gcc at gcc dot gnu dot org, Marc Glisse <marc dot glisse at inria dot fr>
- Date: Sat, 11 May 2013 09:34:37 -0400
- Subject: Re: Calculating cosinus/sinus
- References: <518E0C29 dot 6050204 at jacob dot remcomp dot fr> <alpine dot DEB dot 2 dot 02 dot 1305111128410 dot 3954 at laptop-mg dot saclay dot inria dot fr> <518E1299 dot 80609 at jacob dot remcomp dot fr>
On 5/11/2013 5:42 AM, jacob navia wrote:
1) The fsin instruction is ONE instruction! The sin routine is (at
least) thousand instructions!
Even if the fsin instruction itself is "slow" it should be thousand
times faster than the
complicated routine gcc calls.
2) The FPU is at 64 bits mantissa using gcc, i.e. fsin will calculate
with 64 bits mantissa and
NOT only 53 as SSE2. The fsin instruction is more precise!
You are making conclusions based on naive assumptions here.
I think that gcc has a problem here. I am pointing you to this problem,
but please keep in mind
I am no newbee...
Sure, but that does not mean you are familiar with the intracacies
of accurate computation of transcendental functions!
jacob