[Bug middle-end/123826] folding of `pow(a, 2.0)` into a*a causes errno not to be set

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123826

--- Comment #11 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Hongtao Liu from comment #10)
> 
> Yes, original code is std::pow (a, 2).
> 
> And I notice std::pow (a, -1) is still optimized to divsd, should it also be
> prevented?

Yes unless you know that a is nonzero.
```
If base is zero and exponent is negative, a domain error or a pole error may
occur.
```