REAL*16 not available in FORTRAN for RISC

[cqny at vax5.CIT.CORNELL.EDU]

I have an application which I am porting to the DECsystem 5400 which
uses REAL*16 (128 bit) floating point, currently in VAX FORTRAN.  The
largest floating point format in FORTRAN for RISC is REAL*8.
Does anyone know of extended precision libraries which might be usable
with FORTRAN for RISC?

Since the critical code isn't impossibly long, I could recode using
an extended precision integer formulation, if all I could find was an
extended integer package.

Thanks
Rich Marisa
Cornell Programmable Automation
marisa at cpa.tn.cornell.edu  or cqny at vax5.tn.cornell.edu