A small comparison

[Kai Nacke]

On Friday, 22 November 2013 at 21:44:28 UTC, Kai Nacke wrote:
> I would expect that the LLVM intrinsic is replaced with fsin 
> (at least with -m32). I am unsure why this does not happen.

The answer is simple: it is yet not implemented. Just look at the 
slides from last LLVM developer meeting: 
http://www.llvm.org/devmtg/2013-11/slides/Rotem-Vectorization.pdf

Regards,
Kai