[cfe-dev] Floating-point performance question

[Renato Golin]

On 6 September 2013 21:55, Chris Lattner <clattner at apple.com> wrote:

> IMO, if glibc's implementation of exp2 is so bad, LibCall info should get
> a bit to indicate this.  This could disable pow(2 -> exp2, and could enable
> transformations from exp2(x) -> pow(2,x).
>

Maybe a cost table / cost info in LC, so that optimizers could work out if
it's worth exchanging N pows for M exps, or something like that, on a
target specific basis.

cheers,
--renato
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20130908/baa0d3d4/attachment.html>