aboutsummaryrefslogtreecommitdiffstats
path: root/src/include/glm/detail/func_common_simd.inl
blob: c04e6e5402414e42d4d430006cc9688d12fa9d85 (plain)
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
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
/// @ref core
/// @file glm/detail/func_common_simd.inl

#if GLM_ARCH & GLM_ARCH_SSE2_BIT

#include "../simd/common.h"

#include <immintrin.h>

namespace glm{
namespace detail
{
	template<qualifier Q>
	struct compute_abs_vector<4, float, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, float, Q> call(vec<4, float, Q> const& v)
		{
			vec<4, float, Q> result;
			result.data = glm_vec4_abs(v.data);
			return result;
		}
	};

	template<qualifier Q>
	struct compute_abs_vector<4, int, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, int, Q> call(vec<4, int, Q> const& v)
		{
			vec<4, int, Q> result;
			result.data = glm_ivec4_abs(v.data);
			return result;
		}
	};

	template<qualifier Q>
	struct compute_floor<4, float, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, float, Q> call(vec<4, float, Q> const& v)
		{
			vec<4, float, Q> result;
			result.data = glm_vec4_floor(v.data);
			return result;
		}
	};

	template<qualifier Q>
	struct compute_ceil<4, float, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, float, Q> call(vec<4, float, Q> const& v)
		{
			vec<4, float, Q> result;
			result.data = glm_vec4_ceil(v.data);
			return result;
		}
	};

	template<qualifier Q>
	struct compute_fract<4, float, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, float, Q> call(vec<4, float, Q> const& v)
		{
			vec<4, float, Q> result;
			result.data = glm_vec4_fract(v.data);
			return result;
		}
	};

	template<qualifier Q>
	struct compute_round<4, float, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, float, Q> call(vec<4, float, Q> const& v)
		{
			vec<4, float, Q> result;
			result.data = glm_vec4_round(v.data);
			return result;
		}
	};

	template<qualifier Q>
	struct compute_mod<4, float, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, float, Q> call(vec<4, float, Q> const& x, vec<4, float, Q> const& y)
		{
			vec<4, float, Q> result;
			result.data = glm_vec4_mod(x.data, y.data);
			return result;
		}
	};

	template<qualifier Q>
	struct compute_min_vector<4, float, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, float, Q> call(vec<4, float, Q> const& v1, vec<4, float, Q> const& v2)
		{
			vec<4, float, Q> result;
			result.data = _mm_min_ps(v1.data, v2.data);
			return result;
		}
	};

	template<qualifier Q>
	struct compute_min_vector<4, int, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, int, Q> call(vec<4, int, Q> const& v1, vec<4, int, Q> const& v2)
		{
			vec<4, int, Q> result;
			result.data = _mm_min_epi32(v1.data, v2.data);
			return result;
		}
	};

	template<qualifier Q>
	struct compute_min_vector<4, uint, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, uint, Q> call(vec<4, uint, Q> const& v1, vec<4, uint, Q> const& v2)
		{
			vec<4, uint, Q> result;
			result.data = _mm_min_epu32(v1.data, v2.data);
			return result;
		}
	};

	template<qualifier Q>
	struct compute_max_vector<4, float, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, float, Q> call(vec<4, float, Q> const& v1, vec<4, float, Q> const& v2)
		{
			vec<4, float, Q> result;
			result.data = _mm_max_ps(v1.data, v2.data);
			return result;
		}
	};

	template<qualifier Q>
	struct compute_max_vector<4, int, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, int, Q> call(vec<4, int, Q> const& v1, vec<4, int, Q> const& v2)
		{
			vec<4, int, Q> result;
			result.data = _mm_max_epi32(v1.data, v2.data);
			return result;
		}
	};

	template<qualifier Q>
	struct compute_max_vector<4, uint, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, uint, Q> call(vec<4, uint, Q> const& v1, vec<4, uint, Q> const& v2)
		{
			vec<4, uint, Q> result;
			result.data = _mm_max_epu32(v1.data, v2.data);
			return result;
		}
	};

	template<qualifier Q>
	struct compute_clamp_vector<4, float, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, float, Q> call(vec<4, float, Q> const& x, vec<4, float, Q> const& minVal, vec<4, float, Q> const& maxVal)
		{
			vec<4, float, Q> result;
			result.data = _mm_min_ps(_mm_max_ps(x.data, minVal.data), maxVal.data);
			return result;
		}
	};

	template<qualifier Q>
	struct compute_clamp_vector<4, int, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, int, Q> call(vec<4, int, Q> const& x, vec<4, int, Q> const& minVal, vec<4, int, Q> const& maxVal)
		{
			vec<4, int, Q> result;
			result.data = _mm_min_epi32(_mm_max_epi32(x.data, minVal.data), maxVal.data);
			return result;
		}
	};

	template<qualifier Q>
	struct compute_clamp_vector<4, uint, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, uint, Q> call(vec<4, uint, Q> const& x, vec<4, uint, Q> const& minVal, vec<4, uint, Q> const& maxVal)
		{
			vec<4, uint, Q> result;
			result.data = _mm_min_epu32(_mm_max_epu32(x.data, minVal.data), maxVal.data);
			return result;
		}
	};

	template<qualifier Q>
	struct compute_mix_vector<4, float, bool, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, float, Q> call(vec<4, float, Q> const& x, vec<4, float, Q> const& y, vec<4, bool, Q> const& a)
		{
			__m128i const Load = _mm_set_epi32(-static_cast<int>(a.w), -static_cast<int>(a.z), -static_cast<int>(a.y), -static_cast<int>(a.x));
			__m128 const Mask = _mm_castsi128_ps(Load);

			vec<4, float, Q> Result;
#			if 0 && GLM_ARCH & GLM_ARCH_AVX
				Result.data = _mm_blendv_ps(x.data, y.data, Mask);
#			else
				Result.data = _mm_or_ps(_mm_and_ps(Mask, y.data), _mm_andnot_ps(Mask, x.data));
#			endif
			return Result;
		}
	};
/* FIXME
	template<qualifier Q>
	struct compute_step_vector<float, Q, tvec4>
	{
		GLM_FUNC_QUALIFIER static vec<4, float, Q> call(vec<4, float, Q> const& edge, vec<4, float, Q> const& x)
		{
			vec<4, float, Q> Result;
			result.data = glm_vec4_step(edge.data, x.data);
			return result;
		}
	};
*/
	template<qualifier Q>
	struct compute_smoothstep_vector<4, float, Q, true>
	{
		GLM_FUNC_QUALIFIER static vec<4, float, Q> call(vec<4, float, Q> const& edge0, vec<4, float, Q> const& edge1, vec<4, float, Q> const& x)
		{
			vec<4, float, Q> Result;
			Result.data = glm_vec4_smoothstep(edge0.data, edge1.data, x.data);
			return Result;
		}
	};
}//namespace detail
}//namespace glm

#endif//GLM_ARCH & GLM_ARCH_SSE2_BIT