yuzu/src/common/math_util.h
2015-05-07 15:45:22 -03:00

205 lines
5.0 KiB
C++

// Copyright 2013 Dolphin Emulator Project / 2014 Citra Emulator Project
// Licensed under GPLv2 or any later version
// Refer to the license.txt file included.
#pragma once
#include "common/common_types.h"
#include <algorithm>
#include <type_traits>
#include <vector>
namespace MathUtil
{
template<typename T>
inline T Clamp(const T val, const T& min, const T& max)
{
return std::max(min, std::min(max, val));
}
static const u64 DOUBLE_SIGN = 0x8000000000000000ULL,
DOUBLE_EXP = 0x7FF0000000000000ULL,
DOUBLE_FRAC = 0x000FFFFFFFFFFFFFULL,
DOUBLE_ZERO = 0x0000000000000000ULL;
static const u32 FLOAT_SIGN = 0x80000000,
FLOAT_EXP = 0x7F800000,
FLOAT_FRAC = 0x007FFFFF,
FLOAT_ZERO = 0x00000000;
union IntDouble {
double d;
u64 i;
};
union IntFloat {
float f;
u32 i;
};
inline bool IsNAN(double d)
{
IntDouble x; x.d = d;
return ( ((x.i & DOUBLE_EXP) == DOUBLE_EXP) &&
((x.i & DOUBLE_FRAC) != DOUBLE_ZERO) );
}
inline bool IsQNAN(double d)
{
IntDouble x; x.d = d;
return ( ((x.i & DOUBLE_EXP) == DOUBLE_EXP) &&
((x.i & 0x0007fffffffffffULL) == 0x000000000000000ULL) &&
((x.i & 0x000800000000000ULL) == 0x000800000000000ULL) );
}
inline bool IsSNAN(double d)
{
IntDouble x; x.d = d;
return( ((x.i & DOUBLE_EXP) == DOUBLE_EXP) &&
((x.i & DOUBLE_FRAC) != DOUBLE_ZERO) &&
((x.i & 0x0008000000000000ULL) == DOUBLE_ZERO) );
}
inline float FlushToZero(float f)
{
IntFloat x; x.f = f;
if ((x.i & FLOAT_EXP) == 0)
x.i &= FLOAT_SIGN; // turn into signed zero
return x.f;
}
inline double FlushToZeroAsFloat(double d)
{
IntDouble x; x.d = d;
if ((x.i & DOUBLE_EXP) < 0x3800000000000000ULL)
x.i &= DOUBLE_SIGN; // turn into signed zero
return x.d;
}
enum PPCFpClass
{
PPC_FPCLASS_QNAN = 0x11,
PPC_FPCLASS_NINF = 0x9,
PPC_FPCLASS_NN = 0x8,
PPC_FPCLASS_ND = 0x18,
PPC_FPCLASS_NZ = 0x12,
PPC_FPCLASS_PZ = 0x2,
PPC_FPCLASS_PD = 0x14,
PPC_FPCLASS_PN = 0x4,
PPC_FPCLASS_PINF = 0x5,
};
// Uses PowerPC conventions for the return value, so it can be easily
// used directly in CPU emulation.
u32 ClassifyDouble(double dvalue);
// More efficient float version.
u32 ClassifyFloat(float fvalue);
template<class T>
struct Rectangle
{
T left;
T top;
T right;
T bottom;
Rectangle()
{ }
Rectangle(T theLeft, T theTop, T theRight, T theBottom)
: left(theLeft), top(theTop), right(theRight), bottom(theBottom)
{ }
bool operator==(const Rectangle& r) { return left==r.left && top==r.top && right==r.right && bottom==r.bottom; }
T GetWidth() const { return std::abs(static_cast<typename std::make_signed<T>::type>(right - left)); }
T GetHeight() const { return std::abs(static_cast<typename std::make_signed<T>::type>(bottom - top)); }
// If the rectangle is in a coordinate system with a lower-left origin, use
// this Clamp.
void ClampLL(T x1, T y1, T x2, T y2)
{
if (left < x1) left = x1;
if (right > x2) right = x2;
if (top > y1) top = y1;
if (bottom < y2) bottom = y2;
}
// If the rectangle is in a coordinate system with an upper-left origin,
// use this Clamp.
void ClampUL(T x1, T y1, T x2, T y2)
{
if (left < x1) left = x1;
if (right > x2) right = x2;
if (top < y1) top = y1;
if (bottom > y2) bottom = y2;
}
};
} // namespace MathUtil
inline float pow2f(float x) {return x * x;}
inline double pow2(double x) {return x * x;}
float MathFloatVectorSum(const std::vector<float>&);
#define ROUND_UP(x, a) (((x) + (a) - 1) & ~((a) - 1))
#define ROUND_DOWN(x, a) ((x) & ~((a) - 1))
// Rounds down. 0 -> undefined
inline u64 Log2(u64 val)
{
#if defined(__GNUC__)
return 63 - __builtin_clzll(val);
#elif defined(_MSC_VER) && defined(_M_X64)
unsigned long result = -1;
_BitScanReverse64(&result, val);
return result;
#else
u64 result = -1;
while (val != 0)
{
val >>= 1;
++result;
}
return result;
#endif
}
// Tiny matrix/vector library.
// Used for things like Free-Look in the gfx backend.
class Matrix33
{
public:
static void LoadIdentity(Matrix33 &mtx);
// set mtx to be a rotation matrix around the x axis
static void RotateX(Matrix33 &mtx, float rad);
// set mtx to be a rotation matrix around the y axis
static void RotateY(Matrix33 &mtx, float rad);
// set result = a x b
static void Multiply(const Matrix33 &a, const Matrix33 &b, Matrix33 &result);
static void Multiply(const Matrix33 &a, const float vec[3], float result[3]);
float data[9];
};
class Matrix44
{
public:
static void LoadIdentity(Matrix44 &mtx);
static void LoadMatrix33(Matrix44 &mtx, const Matrix33 &m33);
static void Set(Matrix44 &mtx, const float mtxArray[16]);
static void Translate(Matrix44 &mtx, const float vec[3]);
static void Multiply(const Matrix44 &a, const Matrix44 &b, Matrix44 &result);
float data[16];
};