Floating-Point Support (Windows Embedded CE 6.0)
1/5/2010
Many Microsoft run-time library functions require floating-point support from a math coprocessor or from the floating-point libraries that accompany the compiler. Floating-point support functions are loaded only if required.
When you use a floating-point type specifier in the format string of a call to a function in the printf or scanf family, you must specify a floating-point value or a pointer to a floating-point value in the argument list to tell the compiler that floating-point support is required.
The math functions in the Microsoft run-time library handle exceptions the same way that UNIX V math functions do.
The Microsoft run-time library sets the default internal precision of the math coprocessor (or emulator) to 64 bits. This default applies only to the internal precision at which all intermediate calculations are performed; it does not apply to the size of arguments, return values, or variables.
You can override this default and set the chip (or emulator) back to 80-bit precision by linking your program with LIB/FP10.OBJ.
On the linker command line, FP10.OBJ must appear before LIBC.LIB, LIBCMT.LIB, or MSVCRT.LIB.
For detailed reference information about the CRT functions, see Alphabetical Function Reference (CRT).
| Routine | Use |
|---|---|
abs |
Returns absolute value of int |
acos |
Calculates arccosine |
asin |
Calculates arcsine |
atan, atan2 |
Calculates arctangent |
atof |
Converts character string to double-precision floating-point value |
_cabs |
Finds absolute value of complex number |
ceil, ceilf |
Finds integer ceiling |
_chgsign |
Reverses sign of double-precision floating-point argument |
clearerr |
Clears error. This function has been deprecated because a more secure version, clearerr_s, is available. |
clearerr_s |
Resets the error indicator for a stream. Security-enhanced version of clearerr. |
_controlfp |
Gets and sets the floating-point control word. This function has been deprecated because a more secure version, _controlfp_s, is available. |
controlfp_s |
Gets and sets the floating-point control word. Security-enhanced version of deprecated function _controlfp. Returns error for bad parameters. |
_copysign |
Returns one value with sign of another |
cos |
Calculates cosine |
cosh |
Calculates hyperbolic cosine |
div |
Divides one integer by another, returning quotient and remainder |
_ecvt |
Converts double to character string of specified length. This function has been deprecated because a more secure version, _evct_s, is available. |
_ecvt_s |
Converts a double number to a string. Security-enhanced version of deprecated function _ecvt. Validates parameters and sets errno. Template overload available. |
exp |
Calculates exponential function |
fabs, fabsf |
Finds absolute value |
_fcvt |
Converts double to string with specified number of digits following decimal point. This function has been deprecated because a more secure version, _fvct_s, is available. |
_fcvt_s |
Converts a floating-point number to a string. Security-enhanced version of deprecated function _fcvt. Validates parameters and sets errno. Template overload available. |
_finite |
Determines whether given double-precision floating-point value is finite |
floor, floorf |
Finds largest integer less than or equal to argument |
fmod, fmodf |
Finds floating-point remainder |
_fpclass |
Returns status word containing information on floating-point class |
_fpieee_flt |
Invokes user-defined trap handler for IEEE floating-point exceptions |
_fpreset |
Reinitializes floating-point math package |
frexp |
Calculates exponential value |
_gcvt |
Converts floating-point value to character string. This function has been deprecated because a more secure version, _gvct_s, is available. |
_gcvt_s |
Converts a floating-point value to a string. Security-enhanced version of deprecated function _gcvt. Validates parameters and sets errno. Template overload available. |
_hypot |
Calculates hypotenuse of right triangle |
_isnan |
Checks given double-precision floating-point value for not a number (NaN) |
_j0, _j1, _jn |
Calculates Bessel functions _j0, _j1, and _jn |
labs |
Returns absolute value of long |
ldexp |
Calculates product of argument and 2 to specified power |
ldiv |
Divides one long integer by another, returning quotient and remainder |
log, log10 |
Calculates logarithm |
_logb |
Extracts exponential value of double-precision floating-point argument |
_lrotl, _lrotr |
Shifts unsigned long int left (_lrotl) or right (_lrotr) |
modf |
Splits argument into integer and fractional parts |
_nextafter |
Returns next representable neighbor |
pow |
Calculates value raised to a power |
rand |
Gets pseudorandom number |
rand_s |
Generates a pseudorandom number. Security-enhanced version of rand. |
_rotl, _rotr |
Shifts unsigned int left (_rotl) or right (_rotr) |
_scalb |
Scales argument by power of 2 |
sin |
Calculates sine |
sinh |
Calculates hyperbolic sine |
sqrt |
Finds square root |
srand |
Initializes pseudorandom series |
_statusfp |
Gets floating-point status word |
strtod |
Converts character string to double-precision value |
tan |
Calculates tangent |
tanh |
Calculates hyperbolic tangent |
_y0, _y1, _yn |
Calculates Bessel functions _yo, _y1, and _yn |
See Also
Concepts
Other Resources
C/C++ Libraries for Windows Embedded CE
What's New in the C Run-Time Library