IceFPU.h File Reference

Go to the source code of this file.

Macros
#define	SIGN_BITMASK   0x80000000
#define	IR(x)   ((udword&)(x))
	Integer representation of a floating-point value. More...
#define	SIR(x)   ((sdword&)(x))
	Signed integer representation of a floating-point value. More...
#define	AIR(x)   (IR(x)&0x7fffffff)
	Absolute integer representation of a floating-point value. More...
#define	FR(x)   ((float&)(x))
	Floating-point representation of an integer value. More...
#define	CHECK_VALID_FLOAT(x)   OPASSERT(IsValidFloat(x));

Functions
inline_ bool	IsNAN (float value)
	Is the float valid ? More...
inline_ bool	IsIndeterminate (float value)
inline_ bool	IsPlusInf (float value)
inline_ bool	IsMinusInf (float value)
inline_ bool	IsValidFloat (float value)

Detailed Description

Contains FPU related code.

Author

Pierre Terdiman

Date

April, 4, 2000

Definition in file IceFPU.h.

Macro Definition Documentation

#define AIR

(

x)

(IR(x)&0x7fffffff)

Absolute integer representation of a floating-point value.

Definition at line 24 of file IceFPU.h.

Referenced by Point::ClosestAxis().

#define CHECK_VALID_FLOAT

(

x)

OPASSERT(IsValidFloat(x));

Definition at line 44 of file IceFPU.h.

#define FR

(

x)

((float&)(x))

Floating-point representation of an integer value.

Definition at line 27 of file IceFPU.h.

Referenced by Point::Tweak(), and Point::TweakSmaller().

#define IR

(

x)

((udword&)(x))

Integer representation of a floating-point value.

Definition at line 18 of file IceFPU.h.

Referenced by Matrix3x3::IsIdentity(), Matrix4x4::IsIdentity(), IsIndeterminate(), IsMinusInf(), IsNAN(), Opcode::Point::IsNotUsed(), IsPlusInf(), Point::IsZero(), Point::operator!=(), Point::operator==(), Opcode::Point::SetNotUsed(), Matrix4x4::Transpose(), Matrix3x3::Transpose(), Point::Tweak(), Point::TweakSmaller(), and Point::VectorCode().

#define SIGN_BITMASK   0x80000000

Definition at line 15 of file IceFPU.h.

Referenced by Point::VectorCode().

#define SIR

(

x)

((sdword&)(x))

Signed integer representation of a floating-point value.

Definition at line 21 of file IceFPU.h.

Function Documentation

inline_ bool IsIndeterminate

(

float

value)

Definition at line 31 of file IceFPU.h.

References IR.

Referenced by IsValidFloat().

{ return IR(value) == 0xffc00000;               }

inline_ bool IsMinusInf

(

float

value)

Definition at line 33 of file IceFPU.h.

References IR.

Referenced by IsValidFloat().

{ return IR(value) == 0xff800000;               }

inline_ bool IsNAN

(

float

value)

Is the float valid ?

Definition at line 30 of file IceFPU.h.

References IR.

Referenced by IsValidFloat().

{ return (IR(value)&0x7f800000) == 0x7f800000;  }

inline_ bool IsPlusInf

(

float

value)

Definition at line 32 of file IceFPU.h.

References IR.

Referenced by IsValidFloat().

{ return IR(value) == 0x7f800000;               }

inline_ bool IsValidFloat

(

float

value)

Definition at line 35 of file IceFPU.h.

References IsIndeterminate(), IsMinusInf(), IsNAN(), and IsPlusInf().

Referenced by Matrix3x3::IsValid(), Matrix4x4::IsValid(), and Point::IsValid().

    {
        if(IsNAN(value))            return false;
        if(IsIndeterminate(value))  return false;
        if(IsPlusInf(value))        return false;
        if(IsMinusInf(value))       return false;
        return true;
    }