configure check for MPFR?

[Steve Kargl]

gfortran currently uses GMP for its internal floating point
arithematic.  It has been suggested that gfortran should 
instead use MPFR.  I've started to look at the possibility of
doing some of the work to switch over from GMP to MPFR.
However, I don't know the configure magic I need to link
gfortran with the MPFR and GMP libraries.  Can someone
with configury knowledge provide a patch or enough
hand holding so I can come up with a patch?

-- 
Steve