Language Reference

Compiling and Exception Handling

XL Fortran provides a number of options for strict compliance with the IEEE standard.

• Use -qfloat=nomaf to ensure compatibility with the IEEE standard for floating point arithmetic (IEEE 754-1985).
• When compiling programs that change the rounding mode, use -qfloat=rrm.
• Use -qfloat=nans to detect signaling NaN values. Signaling NaN values can only occur if specified in a program.
• Use the -qstrict compiler option for strict conformance to the IEEE standard for floating-point arithmetic on programs compiled with an optimization level of -O3 or higher, -qhot, -qipa, -qpdf , or -qsmp.
• Compile programs containing a call to a convert-to-integer operation through assignment or the INT intrinsic with the -qarch=ppc compiler option, or the appropriate -qarch suboption for your PowerPC hardware. Specifying an appropriate -qarch compiler suboption allows the IEEE_INVALID exception to signal on invalid convert-to-integer operations on the PowerPC platform.
• To allow the IEEE_INVALID exception to signal on invalid SQRT operations on AIX 4.3 and above, specify the following command:

  export SQRT_EXCEPTION=3.1
  

• Invalid convert-to-integer and square root operations will not set the IEEE_INVALID exception flag on POWER platforms.

Related Information

For more information on IEEE floating-point and specific explanations of the compiler options listed above, see XL Fortran Floating-Point Processing in the User's Guide.