Implement C23 pown [committed]

[Vincent Lefevre]

On 2025-06-04 09:54:57 -0500, Segher Boessenkool wrote:
> On Wed, Jun 04, 2025 at 02:35:39PM +0200, Vincent Lefevre wrote:
> > On 2025-06-03 11:09:07 -0500, Segher Boessenkool wrote:
> > > The commit message clearly says this implementation isd correct for
> > > IEEE float only.  And indeed it is not for double-double, which is very
> > > much not IEEE.
> > > 
> > > Maybe these tests should require IEEE float?  The implementation does,
> > > after all!
> > 
> > Shouldn't this implementation rather be disabled with non-IEEE floats
> > (thus not just the tests), as it is not correct?
> 
> Maybe?  Some of the end results are terrible (+inf for -inf raised to
> some big odd number, NaN where it should be some finite result), but
> perhaps most results are somewhat close to an appropriate number.
> 
> But testing the edge cases (as this testcase does) simply does not work.

If some results are significantly worse than powl (is it implemented?),
this could be badly surprising.

-- 
Vincent Lefèvre <vincent@vinc17.net> - Web: <https://www.vinc17.net/>
100% accessible validated (X)HTML - Blog: <https://www.vinc17.net/blog/>
Work: CR INRIA - computer arithmetic / Pascaline project (LIP, ENS-Lyon)