NAME
fegetflushtozero() — get floating-point underflow mode
SYNOPSIS
#include <fenv.h>
int fegetflushtozero(void);
DESCRIPTION
The
fegetflushtozero()
function retrieves the value representing the current underflow mode,
which is either IEEE-754-compliant (gradual) underflow mode or
flush-to-zero mode.
The default underflow mode is IEEE-754-compliant.
Flush-to-zero mode, also known as fast underflow mode, is supported on
most PA1.1 systems and on all PA2.0 systems. On HP 9000
systems, most underflow cases are supported by trapping into the
kernel, where the IEEE-mandated conversion of the result into a
denormalized value or zero is accomplished by software emulation.
Flush-to-zero mode causes the hardware to substitute a zero for the
result of an operation, with no fault occurring.
This may be a
significant performance optimization for applications that underflow
frequently.
Flush-to-zero mode also causes denormalized
floating-point operands to be treated as if they were true zero
operands.
To use this function, compile either with the default
-Ae
option or with the
-Aa
and
-D_HPUX_SOURCE
options. Make sure your program includes
<fenv.h>.
Link in the math library by specifying
-lm
on the compiler or linker command line.
For more information, see the
HP-UX Floating-Point Guide.
RETURN VALUE
The
fegetflushtozero()
function returns zero if the current underflow mode is
IEEE-754-compliant.
The function returns 1 if the current underflow
mode is flush-to-zero.
On systems that do not support flush-to-zero mode, this function
returns an undefined value.
ERRORS
No errors are defined.
EXAMPLE
Save the current underflow mode, set flush-to-zero mode, and restore
the previous mode.
#include <fenv.h>
/*...*/
int fm_saved;
fm_saved = fegetflushtozero();
fesetflushtozero(1);
/*...*/
fesetflushtozero(fm_saved);
AUTHOR
fegetflushtozero()
was developed by HP and is not required by any current standard.