_status87, _statusfp, _statusfp2

Gets the floating-point status word.

Syntax

unsigned int _status87( void );
unsigned int _statusfp( void );
void _statusfp2(unsigned int *px86, unsigned int *pSSE2)

Parameters

px86
This address is filled with the status word for the x87 floating-point unit.

pSSE2
This address is filled with the status word for the SSE2 floating-point unit.

Return value

For _status87 and _statusfp, the bits in the value that's returned indicate the floating-point status. See the FLOAT.H include file for a definition of the bits that are returned by _statusfp. Many math library functions modify the floating-point status word, with unpredictable results. Optimization can reorder, combine, and eliminate floating-point operations around calls to _status87, _statusfp, and related functions. Use the /Od (Disable (Debug)) compiler option or the fenv_access pragma directive to prevent optimizations that reorder floating-point operations. Return values from _clearfp and _statusfp, and also the return parameters of _statusfp2, are more reliable if fewer floating-point operations are performed between known states of the floating-point status word.

Remarks

The _statusfp function gets the floating-point status word. The status word is a combination of the floating-point processor status and other conditions detected by the floating-point exception handler—for example, floating-point stack overflow and underflow. Unmasked exceptions are checked for before the contents of the status word are returned. In other words, the caller is informed of pending exceptions. On x86 platforms, _statusfp returns a combination of the x87 and SSE2 floating-point status. On x64 platforms, the status that's returned is based on the SSE's MXCSR status. On ARM64 platforms, _statusfp returns status from the FPSCR register.

_statusfp is a platform-independent, portable version of _status87. It's identical to _status87 on Intel (x86) platforms and is also supported by the x64 and ARM64 platforms. To ensure that your floating-point code is portable to all architectures, use _statusfp. If you're only targeting x86 platforms, you can use either _status87 or _statusfp.

We recommend _statusfp2 for chips (such as the Pentium IV) that have both an x87 and an SSE2 floating-point processor. For _statusfp2, the addresses are filled by using the floating-point status word for both the x87 or the SSE2 floating-point processor. For a chip that supports x87 and SSE2 floating-point processors, EM_AMBIGUOUS is set to 1 if _statusfp or _controlfp is used and the action was ambiguous because it could refer to the x87 or the SSE2 floating-point status word. The _statusfp2 function is only supported on x86 platforms.

These functions aren't useful for /clr (Common Language Runtime Compilation) because the common language runtime (CLR) only supports the default floating-point precision.

Requirements

For more compatibility information, see Compatibility.

Example

// crt_statusfp.c
// Build by using: cl /W4 /Ox /nologo crt_statusfp.c
// This program creates various floating-point errors and
// then uses _statusfp to display messages that indicate these problems.

#include <stdio.h>
#include <float.h>
#pragma fenv_access(on)

double test( void )
{
   double a = 1e-40;
   float b;
   double c;

   printf("Status = 0x%.8x - clear\n", _statusfp());

   // Assignment into b is inexact & underflows:
   b = (float)(a + 1e-40);
   printf("Status = 0x%.8x - inexact, underflow\n", _statusfp());

   // c is denormal:
   c = b / 2.0;
   printf("Status = 0x%.8x - inexact, underflow, denormal\n",
            _statusfp());

   // Clear floating point status:
   _clearfp();
   return c;
}

int main(void)
{
   return (int)test();
}

Status = 0x00000000 - clear
Status = 0x00000003 - inexact, underflow
Status = 0x00080003 - inexact, underflow, denormal

See also

Math and floating-point support
_clear87, _clearfp
_control87, _controlfp, __control87_2