Fortran: Module Interface nanMod (Source File: nanMod.F90)

Set parameters for the floating point flags "inf" Infinity and "nan" not-a-number. As well as "bigint" the point at which integers start to overflow. These values are used to initialize arrays with as a way to detect if arrays are being used before being set. Note that bigint is the largest possible 32-bit integer.

USES:

   use shr_kind_mod, only: r8 => shr_kind_r8

PUBLIC TYPES:

   implicit none
   save
 #ifdef __PGI
   quiet nan for portland group compilers
   real(r8), parameter :: inf = O'0777600000000000000000'
   real(r8), parameter :: nan = O'0777700000000000000000'
   integer,  parameter :: bigint = O'17777777777'
 #else
   signaling nan otherwise
   real(r8), parameter :: inf = O'0777600000000000000000'
   real(r8), parameter :: nan = O'0777610000000000000000'
   integer,  parameter :: bigint = O'17777777777'
 #endif

REVISION HISTORY:

   Created by Mariana Vertenstein based on cam module created by
   CCM core group