1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
|
// This code contains NVIDIA Confidential Information and is disclosed to you
// under a form of NVIDIA software license agreement provided separately to you.
//
// Notice
// NVIDIA Corporation and its licensors retain all intellectual property and
// proprietary rights in and to this software and related documentation and
// any modifications thereto. Any use, reproduction, disclosure, or
// distribution of this software and related documentation without an express
// license agreement from NVIDIA Corporation is strictly prohibited.
//
// ALL NVIDIA DESIGN SPECIFICATIONS, CODE ARE PROVIDED "AS IS.". NVIDIA MAKES
// NO WARRANTIES, EXPRESSED, IMPLIED, STATUTORY, OR OTHERWISE WITH RESPECT TO
// THE MATERIALS, AND EXPRESSLY DISCLAIMS ALL IMPLIED WARRANTIES OF NONINFRINGEMENT,
// MERCHANTABILITY, AND FITNESS FOR A PARTICULAR PURPOSE.
//
// Information and code furnished is believed to be accurate and reliable.
// However, NVIDIA Corporation assumes no responsibility for the consequences of use of such
// information or for any infringement of patents or other rights of third parties that may
// result from its use. No license is granted by implication or otherwise under any patent
// or patent rights of NVIDIA Corporation. Details are subject to change without notice.
// This code supersedes and replaces all information previously supplied.
// NVIDIA Corporation products are not authorized for use as critical
// components in life support devices or systems without express written approval of
// NVIDIA Corporation.
//
// Copyright (c) 2008-2014 NVIDIA Corporation. All rights reserved.
// Copyright (c) 2004-2008 AGEIA Technologies, Inc. All rights reserved.
// Copyright (c) 2001-2004 NovodeX AG. All rights reserved.
#pragma once
#include <cmath>
// ps4 compiler defines _M_X64 without value
#if ((defined _M_IX86) || (defined _M_X64) || (defined __i386__) || (defined __x86_64__)) && !defined(__ANDROID__)
#define NVMATH_SSE2 1
#else
#define NVMATH_SSE2 0
#endif
#define NVMATH_VMX128 (defined _M_PPC)
#define NVMATH_ALTIVEC (defined __CELLOS_LV2__)
#define NVMATH_NEON (defined _M_ARM || defined __ARM_NEON__)
// which simd types are implemented (one or both are all valid options)
#define NVMATH_SIMD (NVMATH_SSE2 || NVMATH_VMX128 || NVMATH_ALTIVEC || NVMATH_NEON)
#define NVMATH_SCALAR !NVMATH_SIMD
// #define NVMATH_SCALAR 1
#ifdef _MSC_VER
#define NVMATH_ALIGN(alignment, decl) __declspec(align(alignment)) decl
#else
#define NVMATH_ALIGN(alignment, decl) decl __attribute__ ((aligned(alignment)))
#endif
#ifdef min
#undef min
#endif
#ifdef max
#undef max
#endif
// use template expression to fuse multiply-adds into a single instruction
#define NVMATH_FUSE_MULTIPLY_ADD (NVMATH_VMX128 || NVMATH_ALTIVEC || NVMATH_NEON)
// support shift by vector operarations
#define NVMATH_SHIFT_BY_VECTOR (NVMATH_VMX128 || NVMATH_ALTIVEC || NVMATH_NEON)
// Simd4f and Simd4i map to different types
#define NVMATH_DISTINCT_TYPES (NVMATH_SSE2 || NVMATH_ALTIVEC || NVMATH_NEON)
// support inline assembler
#define NVMATH_INLINE_ASSEMBLER !((defined _M_ARM) || (defined SN_TARGET_PSP2) || (defined __arm64__))
// - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
// expression template
// - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
/*! \brief Expression template to fuse and-not. */
template <typename T>
struct ComplementExpr
{
inline ComplementExpr(T const& v_) : v(v_)
{
}
inline operator T() const;
const T v;
private:
ComplementExpr& operator=(const ComplementExpr&); // not implemented
};
// - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
// helper functions
// - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
template <typename T>
T sqr(const T& x)
{
return x * x;
}
// - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
// details
// - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
namespace detail
{
template <typename T>
struct AlignedPointer
{
AlignedPointer(const T* p) : ptr(p)
{
}
const T* ptr;
};
template <typename T>
struct OffsetPointer
{
OffsetPointer(const T* p, unsigned int off) : ptr(p), offset(off)
{
}
const T* ptr;
unsigned int offset;
};
struct FourTuple
{
};
// zero and one literals
template <int i>
struct IntType
{
};
}
// Supress warnings
#if defined(__GNUC__) || defined(__SNC__)
#define NVMATH_UNUSED __attribute__((unused))
#else
#define NVMATH_UNUSED
#endif
static detail::IntType<0> _0 NVMATH_UNUSED;
static detail::IntType<1> _1 NVMATH_UNUSED;
static detail::IntType<int(0x80000000)> _sign NVMATH_UNUSED;
static detail::IntType<-1> _true NVMATH_UNUSED;
// - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
// platform specific includes
// - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
#if NVMATH_SSE2
#include "sse2/SimdTypes.h"
#elif NVMATH_VMX128
#include "xbox360/SimdTypes.h"
#elif NVMATH_ALTIVEC
#include "ps3/SimdTypes.h"
#elif NVMATH_NEON
#include "neon/SimdTypes.h"
#else
struct Simd4f;
struct Simd4i;
#endif
#if NVMATH_SCALAR
#include "scalar/SimdTypes.h"
#else
struct Scalar4f;
struct Scalar4i;
#endif
|