fdimf16
fdimf128
Use the generic algorithms to provide implementations for these routines.
test-multiprecision
build-mpfr
f16_enabled
f128_enabled