Movatterモバイル変換

The characteristics of floating types are defined in terms of a modelthat describes a representation of floating-point numbers and values that provide informationabout an implementation's floating-point arithmetic.

The following parameters are used to define the model for each floating-pointtype:

s: sign (±1)
b: base or radix of exponent representation (an integer >1)
e: exponent (an integer between a minimum e_min and a maximum e_max)
p: precision (the number of base-b digits in the significand)
f_k: non-negative integers less thanb (the significand digits)

In addition to normalized floating-point numbers (f₁>0 ifx≠0), floating types mightbe able to contain other kinds of floating-point numbers, such as subnormalfloating-point numbers (x≠0, e=e_min, f₁=0) and unnormalized floating-point numbers (x≠0, e=e_min, f₁=0),and values that are not floating-point numbers, such as infinities and NaNs.ANaN is an encoding signifying Not-a-Number. Aquiet NaN propagates through almostevery arithmetic operation without raising a floating-point exception; asignaling NaN generally raisesa floating-point exception when occurring as an arithmetic operand.

The accuracy of the library functions inmath.h(3HEAD) andcomplex.h(3HEAD) that return floating-pointresults is defined on thelibm(3LIB) manual page.

All integer values in the <float.h> header, exceptFLT_ROUNDS, are constant expressionssuitable for use in#if preprocessing directives; all floating values are constantexpressions. All exceptDECIMAL_DIG,FLT_EVAL_METHOD,FLT_RADIX, andFLT_ROUNDS have separate names forall three floating-point types. The floating-point model representation is provided for allvalues exceptFLT_EVAL_METHOD andFLT_ROUNDS.

The rounding mode for floating-point addition is characterized by the value ofFLT_ROUNDS:

–1: Indeterminable.
0: Toward zero.
1: To nearest.
2: Toward positive infinity.
3: Toward negative infinity.

The values of operations with floating operands and values subject to theusual arithmetic conversions and of floating constants are evaluated to a formatwhose range and precision might be greater than required by the type.The use of evaluation formats is characterized by the architecture-dependent value ofFLT_EVAL_METHOD:

-1: Indeterminable.
0: Evaluate all operations and constants just to the range and precision of the type.
1: Evaluate operations and constants of type float and double to the range and precision of the double type; evaluate long double operations and constants to the range and precision of the long double type.
2: Evaluate all operations and constants to the range and precision of the long double type.

The values given in the following list are defined as constants.

Radix of exponent representation,b.
```
FLT_RADIX
```
Number of base-FLT_RADIX digits in the floating-point significand,p.
```
FLT_MANT_DIGDBL_MANT_DIGLDBL_MANT_DIG
```
Number of decimal digits,n, such that any floating-point number in the widest supported floating type withp_max radixb digits can be rounded to a floating-point number withn decimal digits and back again without change to the value.
```
DECIMAL_DIG
```
Number of decimal digits,q, such that any floating-point number withq decimal digits can be rounded into a floating-point number withp radixb digits and back again without change to theq decimal digits.
```
FLT_DIGDBL_DIGLDBL_DIG
```
Minimum negative integer such thatFLT_RADIX raised to that power minus 1 is a normalized floating-point number, e_min.
```
FLT_MIN_EXPDBL_MIN_EXPLDBL_MIN_EXP
```
Minimum negative integer such that 10 raised to that power is in the range of normalized floating-point numbers.
```
FLT_MIN_10_EXPDBL_MIN_10_EXPLDBL_MIN_10_EXP
```
Maximum integer such thatFLT_RADIX raised to that power minus 1 is a representable finite floating-point number, e_max.
```
FLT_MAX_EXPDBL_MAX_EXPLDBL_MAX_EXP
```
Maximum integer such that 10 raised to that power is in the range of representable finite floating-point numbers.
```
FLT_MAX_10_EXPDBL_MAX_10_EXPLDBL_MAX_10_EXP
```

The values given in the following list are defined as constant expressionswith values that are greater than or equal to those shown:

Maximum representable finite floating-point number.
```
FLT_MAXDBL_MAXLDBL_MAX
```

The values given in the following list are defined as constant expressionswith implementation-defined (positive) values that are less than or equal to thoseshown:

The difference between 1 and the least value greater than 1 that is representable in the given floating-point type,b^{1 - p}.
```
FLT_EPSILONDBL_EPSILONLDBL_EPSILON
```
Minimum normalized positive floating-point number,b^e_min^-1.
```
FLT_MINDBL_MINLDBL_MIN
```

Attributes

Seeattributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE	ATTRIBUTE VALUE
Interface Stability	Committed
Standard	See`standards`(5).

Skip Navigation Links
Exit Print View
	man pages section 3: Library Interfaces and Headers Oracle Solaris 11 Information Library

Movatterモバイル変換

float.h

, float

Synopsis

Description

Attributes

See Also