Rational Developer for System z
PL/I for Windows, Version 8.0, Programming Guide

FLOATINMATH

The FLOATINMATH option specifies that the precision that the compiler should use when invoking the mathematical built-in functions.

Read syntax diagramSkip visual syntax diagram                   .-ASIS-----.
>>-FLOATINMATH--(--+-LONG-----+--)-----------------------------><
                   '-EXTENDED-'
 
ASIS
Arguments to the mathematical built-in functions will not be forced to have long or extended floating-point precision.
LONG
Any argument to a mathematical built-in function with short floating-point precision will be converted to the maximum long floating-point precision to yield a result with the same maximum long floating-point precision.
EXTENDED
Any argument to a mathematical built-in function with short or long floating-point precision will be converted to the maximum extended floating-point precision to yield a result with the same maximum extended floating-point precision.

A FLOAT DEC expression with precision p has short floating-point precision if p <= 6, long floating-point precision if 6 < p <= 16 and extended floating-point precision if p > 16.

A FLOAT BIN expression with precision p has short floating-point precision if p <= 21, long floating-point precision if 21 <p <= 53 and extended floating-point precision if p > 53.


Terms of use | Feedback

This information center is powered by Eclipse technology. (http://www.eclipse.org)