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
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
|
// 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
// - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
// factory implementation
// - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
template <>
inline Simd4iFactory<const int&>::operator Scalar4i() const
{
return Scalar4i(v, v, v, v);
}
inline Simd4iFactory<detail::FourTuple>::operator Scalar4i() const
{
return reinterpret_cast<const Scalar4i&>(v);
}
template <int i>
inline Simd4iFactory<detail::IntType<i> >::operator Scalar4i() const
{
return Scalar4i(i, i, i, i);
}
template <>
inline Simd4iFactory<const int*>::operator Scalar4i() const
{
return Scalar4i(v[0], v[1], v[2], v[3]);
}
template <>
inline Simd4iFactory<detail::AlignedPointer<int> >::operator Scalar4i() const
{
return Scalar4i(v.ptr[0], v.ptr[1], v.ptr[2], v.ptr[3]);
}
template <>
inline Simd4iFactory<detail::OffsetPointer<int> >::operator Scalar4i() const
{
const int* ptr = reinterpret_cast<const int*>(reinterpret_cast<const char*>(v.ptr) + v.offset);
return Scalar4i(ptr[0], ptr[1], ptr[2], ptr[3]);
}
// - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
// operator implementations
// - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
namespace simdi
{
inline Scalar4i operator==(const Scalar4i& v0, const Scalar4i& v1)
{
return Scalar4i(v0.i4[0] == v1.i4[0], v0.i4[1] == v1.i4[1], v0.i4[2] == v1.i4[2], v0.i4[3] == v1.i4[3]);
}
inline Scalar4i operator<(const Scalar4i& v0, const Scalar4i& v1)
{
return Scalar4i(v0.i4[0] < v1.i4[0], v0.i4[1] < v1.i4[1], v0.i4[2] < v1.i4[2], v0.i4[3] < v1.i4[3]);
}
inline Scalar4i operator>(const Scalar4i& v0, const Scalar4i& v1)
{
return Scalar4i(v0.i4[0] > v1.i4[0], v0.i4[1] > v1.i4[1], v0.i4[2] > v1.i4[2], v0.i4[3] > v1.i4[3]);
}
inline Scalar4i operator+(const Scalar4i& v0, const Scalar4i& v1)
{
return Scalar4i(v0.i4[0] + v1.i4[0], v0.i4[1] + v1.i4[1], v0.i4[2] + v1.i4[2], v0.i4[3] + v1.i4[3]);
}
inline Scalar4i operator-(const Scalar4i& v)
{
return Scalar4i(-v.i4[0], -v.i4[1], -v.i4[2], -v.i4[3]);
}
inline Scalar4i operator-(const Scalar4i& v0, const Scalar4i& v1)
{
return Scalar4i(v0.i4[0] - v1.i4[0], v0.i4[1] - v1.i4[1], v0.i4[2] - v1.i4[2], v0.i4[3] - v1.i4[3]);
}
} // namespace simd
// - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
// function implementations
// - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
inline Scalar4i simd4i(const Scalar4f& v)
{
return v;
}
inline Scalar4i truncate(const Scalar4f& v)
{
return Scalar4i(int(v.f4[0]), int(v.f4[1]), int(v.f4[2]), int(v.f4[3]));
}
namespace simdi
{
inline int (&array(Scalar4i& v))[4]
{
return v.i4;
}
inline const int (&array(const Scalar4i& v))[4]
{
return v.i4;
}
} // namespace simdi
inline void store(int* ptr, const Scalar4i& v)
{
ptr[0] = v.i4[0];
ptr[1] = v.i4[1];
ptr[2] = v.i4[2];
ptr[3] = v.i4[3];
}
inline void storeAligned(int* ptr, const Scalar4i& v)
{
store(ptr, v);
}
inline void storeAligned(int* ptr, unsigned int offset, const Scalar4i& v)
{
store(reinterpret_cast<int*>(reinterpret_cast<char*>(ptr) + offset), v);
}
namespace simdi
{
inline int allEqual(const Scalar4i& v0, const Scalar4i& v1)
{
return v0.i4[0] == v1.i4[0] && v0.i4[1] == v1.i4[1] && v0.i4[2] == v1.i4[2] && v0.i4[3] == v1.i4[3];
}
inline int allEqual(const Scalar4i& v0, const Scalar4i& v1, Scalar4i& outMask)
{
bool b0 = v0.i4[0] == v1.i4[0], b1 = v0.i4[1] == v1.i4[1], b2 = v0.i4[2] == v1.i4[2], b3 = v0.i4[3] == v1.i4[3];
outMask = Scalar4f(b0, b1, b2, b3);
return b0 && b1 && b2 && b3;
}
inline int anyEqual(const Scalar4i& v0, const Scalar4i& v1)
{
return v0.i4[0] == v1.i4[0] || v0.i4[1] == v1.i4[1] || v0.i4[2] == v1.i4[2] || v0.i4[3] == v1.i4[3];
}
inline int anyEqual(const Scalar4i& v0, const Scalar4i& v1, Scalar4i& outMask)
{
bool b0 = v0.i4[0] == v1.i4[0], b1 = v0.i4[1] == v1.i4[1], b2 = v0.i4[2] == v1.i4[2], b3 = v0.i4[3] == v1.i4[3];
outMask = Scalar4f(b0, b1, b2, b3);
return b0 || b1 || b2 || b3;
}
inline int allGreater(const Scalar4i& v0, const Scalar4i& v1)
{
return v0.i4[0] > v1.i4[0] && v0.i4[1] > v1.i4[1] && v0.i4[2] > v1.i4[2] && v0.i4[3] > v1.i4[3];
}
inline int allGreater(const Scalar4i& v0, const Scalar4i& v1, Scalar4i& outMask)
{
bool b0 = v0.i4[0] > v1.i4[0], b1 = v0.i4[1] > v1.i4[1], b2 = v0.i4[2] > v1.i4[2], b3 = v0.i4[3] > v1.i4[3];
outMask = Scalar4f(b0, b1, b2, b3);
return b0 && b1 && b2 && b3;
}
inline int anyGreater(const Scalar4i& v0, const Scalar4i& v1)
{
return v0.i4[0] > v1.i4[0] || v0.i4[1] > v1.i4[1] || v0.i4[2] > v1.i4[2] || v0.i4[3] > v1.i4[3];
}
inline int anyGreater(const Scalar4i& v0, const Scalar4i& v1, Scalar4i& outMask)
{
bool b0 = v0.i4[0] > v1.i4[0], b1 = v0.i4[1] > v1.i4[1], b2 = v0.i4[2] > v1.i4[2], b3 = v0.i4[3] > v1.i4[3];
outMask = Scalar4f(b0, b1, b2, b3);
return b0 || b1 || b2 || b3;
}
} // namespace simd
|