[Bug libstdc++/117406] New: std::ilogb should return INT_MAX when argument is infinite also for 16-bit floats

[pkeir at outlook dot com via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=117406

            Bug ID: 117406
           Summary: std::ilogb should return INT_MAX when argument is
                    infinite also for 16-bit floats
           Product: gcc
           Version: 15.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: libstdc++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: pkeir at outlook dot com
  Target Milestone: ---

The std::ilogb overloads for std::float16_t and std::bfloat16_t should return
INT_MAX when given an infinite argument, but they don't. The second assertion
in the C++23 program below shows this:

#include <stdfloat>
#include <cmath>
#include <climits>
#include <cassert>

// g++ -std=c++23 gcc-ilogb.cpp

int main(int argc, char*argv[])
{
  auto inf    = std::numeric_limits<float>::infinity();
  auto inf16  = std::numeric_limits<std::float16_t>::infinity();
  auto infb16 = std::numeric_limits<std::bfloat16_t>::infinity();

  assert(INT_MAX == std::ilogb(inf)); // correct
  assert(INT_MAX == std::ilogb(inf16) || INT_MAX == std::ilogb(infb16));

  return 0;
}