Consider upgrading the module `mach.float`
That module should be better documented and should provide function on double-precision floats, not only on single-precision
That module should be better documented and should provide function on double-precision floats, not only on single-precision