Description
Fortran programs can use multiple-precision arithmetic routines when normal floating-point precision is not enough.
This is a scientific development library. It is useful for numerical software that needs careful precision handling and should be tested against known results.