[Bug c++/118038] constexpr std::bit_cast with long double/float80_t from some integer values

[gero.peterhoff at gmx dot net]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118038

--- Comment #5 from Gero <gero.peterhoff at gmx dot net> ---
Hello Andrew,
Thanks for your answers. However, I don't understand exactly what this has to
do with PR 104522.

bit_cast should only do the bit conversion and not worry about other
sensitivities. Whether the results are useful or not should not be of interest
at this point. This is in the hands of the caller.

In addition, bit_cast FP64<->I64 also works, for example, where I can also do
numerical nonsense.

thx
Gero