Overflow Conditions [ HP FORTRAN 77/iX Reference ] MPE/iX 5.0 Documentation
HP FORTRAN 77/iX Reference
Overflow Conditions
Each data type has its own format and range. An overflow condition
occurs when numbers outside the range of a particular type are assigned
to, or read into, the corresponding variables at run-time.
For read operations, an error message is issued and the program
terminates unless the ERR or IOSTAT specifier is present. For
assignments, truncation occurs. For integer variables, this truncation
is performed such that the high order bits are ignored. For REAL*4
variables assigned as REAL*8 values, the low order bits are ignored,
preserving the magnitude, but losing the precision. Truncation for
complex variables is the same as that for real variables.
This process is not to be confused with evaluation of an expression whose
result is too large or small for the data type involved, which causes
machine errors unless the corresponding ON statement trap has been set.
MPE/iX 5.0 Documentation