I'm not sure that patch has to deal with "force to memory".
FYI: when I port python trunk to mingw platform I encounter some
inconsistency for isinf.

My note about issue follow:
- configure.in:
...
dnl FIXME: For mingw "isinf" is a macro defined in <math.h> and can't be
dnl detected as function. Also PC/pyconfig.h define HAVE_ISINF but it is
dnl useless since header define Py_IS_INFINITY as (!_finite(X) &&
!_isnan(X))
dnl Also in pymath.h is commented too how PC/pyconfig.h define _isinf.
dnl NOTE: For mingw to keep compatibility with native build we define
dnl Py_IS_INFINITY in pyport.h.
...

- pyport.h
#ifdef __MINGW32__
...
/*NOTE: mingw has isinf as macro defined in math.h.
  Since PC/pyconfig.h define Py_IS_INFINITY(X) that cover HAVE_ISINF
  here for Py_IS_INFINITY we define same as for native build.
  This makes HAVE_ISINF needless.
  Also see comments in configure.in and pymath.h. */
#define Py_IS_INFINITY(X) (!_finite(X) && !_isnan(X))
....


next lest see Py_IS_INFINITY in pymath.h
....
#ifndef Py_IS_INFINITY
#ifdef HAVE_ISINF
#define Py_IS_INFINITY(X) isinf(X)
#else
#define Py_IS_INFINITY(X) ((X) && (X)*0.5 == (X))
#endif
#endif
....

Is the macro Py_IS_INFINITY correct if isinf is not detected by
configure script ?

Also I think too that problem is that we has to classify result after
conversion to double.