[Issue 16026] std.math.frexp!float() wrong for very small subnormal values

[via Digitalmars-d-bugs]

https://issues.dlang.org/show_bug.cgi?id=16026

Iain Buclaw <ibuclaw at gdcproject.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |ibuclaw at gdcproject.org

--- Comment #1 from Iain Buclaw <ibuclaw at gdcproject.org> ---
This is actually a result of turning frexp into a template.

Before it did the operation at real precision, then down casted.  Now, it calls
frexp with the precision of the type T.

It's not a codegen bug, I can reproduce the same using GDC and Phobos 2.067.

--