[libcxx-commits] [libcxx] [libc++] Complete `<charconv>` for 64-bit `long double` platforms (PR #117125)

[Nikolas Klauser via libcxx-commits]

philnik777 wrote:

> > But that's exactly the same with _and_ without this patch?
> 
> I think we're talking past each other. I think that this patch is necessary regardless (except we wouldn't need the `_LIBCPP_LONG_DOUBLE_IS_DOUBLE` macro), and then
> 
> ```
> template __from_chars_result<long double> __from_chars_floating_point(
>     _LIBCPP_NOESCAPE const char* __first, _LIBCPP_NOESCAPE const char* __last, chars_format __fmt);
> ```
> 
> is also necessary on top of that. Either way, I think we both agree that this patch doesn't buy us _much_.

But the main thing this patch does is introducing the `_LIBCPP_LONG_DOUBLE_IS_DOUBLE` macro to forward to the double version sometimes? There isn't much except adding the macro and some FTM configuration and a bit of testing. What am I missing here?

https://github.com/llvm/llvm-project/pull/117125