HP-UX Floating-Point Guide

48 Chapter 2
Floating-Point Principles and the IEEE Standard for Binary Floating-Point Arithmetic
Floating-Point Formats
each component is a double-precision IEEE operand type. HP Fortran 90
and HP FORTRAN/9000 support both complex data types and a full
range of complex arithmetic operations.
NOTE HP Fortran 90 and HP FORTRAN/9000 also support the nonstandard
data type names DOUBLE COMPLEX and COMPLEX*16 (equivalent to
COMPLEX(KIND=8)) and COMPLEX*8 (equivalent to COMPLEX).
IEEE Representation Summary
Table 2-6, Table 2-7, Table 2-8, and Table 2-9 summarize how IEEE
values are represented in binary. To determine the class (normalized,
infinity, NaN, and so on) of a floating-point value at run time, you can
• Use one of the following macros:
• fpclassify to determine the value class
• signbit to determine the sign
• isnan to determine if the value is a NaN
• isinf to determine if the value is an infinity
• isfinite to determine if the value is finite (that is, neither
infinity nor NaN)
• isnormal to determine if the value is normalized
See “Floating-Point Classification Macro” on page 116 for information
about fpclassify. See the online man pages for information about
all of these macros.
• Provide code in your program that writes out the value in
hexadecimal (see “Displaying Floating-Point Values in Binary” on
page 77 for an example) and look up the value in Table 2-6 or Table
2-8
• Examine the value in hexadecimal using the debugger