https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118841
Jakub Jelinek <jakub at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |jakub at gcc dot gnu.org
--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
(In reply to Jonathan Wakely from comment #2)
> Maybe like this:
>
> --- a/libstdc++-v3/include/bits/random.tcc
> +++ b/libstdc++-v3/include/bits/random.tcc
> @@ -3351,12 +3351,17 @@ namespace __detail
> static_assert(std::is_floating_point<_RealType>::value,
> "template argument must be a floating point type");
>
> + using _MaxRealType
> + = __conditional_t<(std::numeric_limits<_RealType>::digits
> + <= std::numeric_limits<double>::digits),
> + double, long double>;
> const size_t __b
> =
> std::min(static_cast<size_t>(std::numeric_limits<_RealType>::digits),
> __bits);
> - const long double __r = static_cast<long double>(__urng.max())
> - - static_cast<long double>(__urng.min()) + 1.0L;
> - const size_t __log2r = std::log(__r) / std::log(2.0L);
> + const _MaxRealType __r = static_cast<_MaxRealType>(__urng.max())
> + - static_cast<_MaxRealType>(__urng.min())
> + + 1.0L;
This one would need to be _MaxRealType(1.0) or so. Otherwise it is still
evaluated in long double.
