Home
last modified time | relevance | path

Searched refs:_mm256_add_ps (Results 1 – 25 of 156) sorted by relevance

1234567

/external/XNNPACK/src/f32-dwconv/gen/
Dup16x25-avx-acc2.c100 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up16x25__avx_acc2()
101 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi0x89ABCDEF, vk0x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x25__avx_acc2()
118 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi2x01234567, vk2x01234567)); in xnn_f32_dwconv_ukernel_up16x25__avx_acc2()
119 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi2x89ABCDEF, vk2x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x25__avx_acc2()
127 vacc01234567p1 = _mm256_add_ps(vacc01234567p1, _mm256_mul_ps(vi3x01234567, vk3x01234567)); in xnn_f32_dwconv_ukernel_up16x25__avx_acc2()
128 vacc89ABCDEFp1 = _mm256_add_ps(vacc89ABCDEFp1, _mm256_mul_ps(vi3x89ABCDEF, vk3x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x25__avx_acc2()
136 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi4x01234567, vk4x01234567)); in xnn_f32_dwconv_ukernel_up16x25__avx_acc2()
137 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi4x89ABCDEF, vk4x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x25__avx_acc2()
145 vacc01234567p1 = _mm256_add_ps(vacc01234567p1, _mm256_mul_ps(vi5x01234567, vk5x01234567)); in xnn_f32_dwconv_ukernel_up16x25__avx_acc2()
146 vacc89ABCDEFp1 = _mm256_add_ps(vacc89ABCDEFp1, _mm256_mul_ps(vi5x89ABCDEF, vk5x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x25__avx_acc2()
[all …]
Dup16x25-avx.c100 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up16x25__avx()
101 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi0x89ABCDEF, vk0x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x25__avx()
109 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi1x01234567, vk1x01234567)); in xnn_f32_dwconv_ukernel_up16x25__avx()
110 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi1x89ABCDEF, vk1x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x25__avx()
118 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi2x01234567, vk2x01234567)); in xnn_f32_dwconv_ukernel_up16x25__avx()
119 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi2x89ABCDEF, vk2x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x25__avx()
127 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi3x01234567, vk3x01234567)); in xnn_f32_dwconv_ukernel_up16x25__avx()
128 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi3x89ABCDEF, vk3x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x25__avx()
136 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi4x01234567, vk4x01234567)); in xnn_f32_dwconv_ukernel_up16x25__avx()
137 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi4x89ABCDEF, vk4x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x25__avx()
[all …]
Dup8x25-avx.c97 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx()
103 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi1x01234567, vk1x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx()
109 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi2x01234567, vk2x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx()
115 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi3x01234567, vk3x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx()
121 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi4x01234567, vk4x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx()
127 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi5x01234567, vk5x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx()
133 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi6x01234567, vk6x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx()
139 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi7x01234567, vk7x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx()
145 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi8x01234567, vk8x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx()
151 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi9x01234567, vk9x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx()
[all …]
Dup8x25-avx-acc2.c97 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx_acc2()
109 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi2x01234567, vk2x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx_acc2()
115 vacc01234567p1 = _mm256_add_ps(vacc01234567p1, _mm256_mul_ps(vi3x01234567, vk3x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx_acc2()
121 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi4x01234567, vk4x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx_acc2()
127 vacc01234567p1 = _mm256_add_ps(vacc01234567p1, _mm256_mul_ps(vi5x01234567, vk5x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx_acc2()
133 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi6x01234567, vk6x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx_acc2()
139 vacc01234567p1 = _mm256_add_ps(vacc01234567p1, _mm256_mul_ps(vi7x01234567, vk7x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx_acc2()
145 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi8x01234567, vk8x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx_acc2()
151 vacc01234567p1 = _mm256_add_ps(vacc01234567p1, _mm256_mul_ps(vi9x01234567, vk9x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx_acc2()
157 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi10x01234567, vk10x01234567)); in xnn_f32_dwconv_ukernel_up8x25__avx_acc2()
[all …]
Dup16x9-avx-acc2.c68 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up16x9__avx_acc2()
69 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi0x89ABCDEF, vk0x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x9__avx_acc2()
86 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi2x01234567, vk2x01234567)); in xnn_f32_dwconv_ukernel_up16x9__avx_acc2()
87 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi2x89ABCDEF, vk2x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x9__avx_acc2()
95 vacc01234567p1 = _mm256_add_ps(vacc01234567p1, _mm256_mul_ps(vi3x01234567, vk3x01234567)); in xnn_f32_dwconv_ukernel_up16x9__avx_acc2()
96 vacc89ABCDEFp1 = _mm256_add_ps(vacc89ABCDEFp1, _mm256_mul_ps(vi3x89ABCDEF, vk3x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x9__avx_acc2()
104 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi4x01234567, vk4x01234567)); in xnn_f32_dwconv_ukernel_up16x9__avx_acc2()
105 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi4x89ABCDEF, vk4x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x9__avx_acc2()
113 vacc01234567p1 = _mm256_add_ps(vacc01234567p1, _mm256_mul_ps(vi5x01234567, vk5x01234567)); in xnn_f32_dwconv_ukernel_up16x9__avx_acc2()
114 vacc89ABCDEFp1 = _mm256_add_ps(vacc89ABCDEFp1, _mm256_mul_ps(vi5x89ABCDEF, vk5x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x9__avx_acc2()
[all …]
Dup16x9-avx.c68 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up16x9__avx()
69 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi0x89ABCDEF, vk0x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x9__avx()
77 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi1x01234567, vk1x01234567)); in xnn_f32_dwconv_ukernel_up16x9__avx()
78 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi1x89ABCDEF, vk1x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x9__avx()
86 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi2x01234567, vk2x01234567)); in xnn_f32_dwconv_ukernel_up16x9__avx()
87 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi2x89ABCDEF, vk2x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x9__avx()
95 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi3x01234567, vk3x01234567)); in xnn_f32_dwconv_ukernel_up16x9__avx()
96 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi3x89ABCDEF, vk3x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x9__avx()
104 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi4x01234567, vk4x01234567)); in xnn_f32_dwconv_ukernel_up16x9__avx()
105 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi4x89ABCDEF, vk4x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x9__avx()
[all …]
Dup8x9-avx.c65 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx()
71 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi1x01234567, vk1x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx()
77 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi2x01234567, vk2x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx()
83 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi3x01234567, vk3x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx()
89 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi4x01234567, vk4x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx()
95 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi5x01234567, vk5x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx()
101 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi6x01234567, vk6x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx()
107 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi7x01234567, vk7x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx()
113 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi8x01234567, vk8x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx()
133 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx()
[all …]
Dup8x9-avx-acc2.c65 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx_acc2()
77 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi2x01234567, vk2x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx_acc2()
83 vacc01234567p1 = _mm256_add_ps(vacc01234567p1, _mm256_mul_ps(vi3x01234567, vk3x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx_acc2()
89 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi4x01234567, vk4x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx_acc2()
95 vacc01234567p1 = _mm256_add_ps(vacc01234567p1, _mm256_mul_ps(vi5x01234567, vk5x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx_acc2()
101 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi6x01234567, vk6x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx_acc2()
107 vacc01234567p1 = _mm256_add_ps(vacc01234567p1, _mm256_mul_ps(vi7x01234567, vk7x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx_acc2()
113 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi8x01234567, vk8x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx_acc2()
118 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, vacc01234567p1); in xnn_f32_dwconv_ukernel_up8x9__avx_acc2()
135 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up8x9__avx_acc2()
[all …]
Dup16x4-avx-acc2.c58 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up16x4__avx_acc2()
59 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi0x89ABCDEF, vk0x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x4__avx_acc2()
76 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi2x01234567, vk2x01234567)); in xnn_f32_dwconv_ukernel_up16x4__avx_acc2()
77 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi2x89ABCDEF, vk2x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x4__avx_acc2()
85 vacc01234567p1 = _mm256_add_ps(vacc01234567p1, _mm256_mul_ps(vi3x01234567, vk3x01234567)); in xnn_f32_dwconv_ukernel_up16x4__avx_acc2()
86 vacc89ABCDEFp1 = _mm256_add_ps(vacc89ABCDEFp1, _mm256_mul_ps(vi3x89ABCDEF, vk3x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x4__avx_acc2()
91 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, vacc01234567p1); in xnn_f32_dwconv_ukernel_up16x4__avx_acc2()
92 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, vacc89ABCDEFp1); in xnn_f32_dwconv_ukernel_up16x4__avx_acc2()
110 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up16x4__avx_acc2()
122 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi2x01234567, vk2x01234567)); in xnn_f32_dwconv_ukernel_up16x4__avx_acc2()
[all …]
Dup16x4-avx.c58 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up16x4__avx()
59 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi0x89ABCDEF, vk0x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x4__avx()
67 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi1x01234567, vk1x01234567)); in xnn_f32_dwconv_ukernel_up16x4__avx()
68 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi1x89ABCDEF, vk1x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x4__avx()
76 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi2x01234567, vk2x01234567)); in xnn_f32_dwconv_ukernel_up16x4__avx()
77 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi2x89ABCDEF, vk2x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x4__avx()
85 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi3x01234567, vk3x01234567)); in xnn_f32_dwconv_ukernel_up16x4__avx()
86 vacc89ABCDEFp0 = _mm256_add_ps(vacc89ABCDEFp0, _mm256_mul_ps(vi3x89ABCDEF, vk3x89ABCDEF)); in xnn_f32_dwconv_ukernel_up16x4__avx()
107 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up16x4__avx()
113 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi1x01234567, vk1x01234567)); in xnn_f32_dwconv_ukernel_up16x4__avx()
[all …]
Dup8x4-avx.c55 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up8x4__avx()
61 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi1x01234567, vk1x01234567)); in xnn_f32_dwconv_ukernel_up8x4__avx()
67 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi2x01234567, vk2x01234567)); in xnn_f32_dwconv_ukernel_up8x4__avx()
73 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi3x01234567, vk3x01234567)); in xnn_f32_dwconv_ukernel_up8x4__avx()
93 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up8x4__avx()
97 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi1x01234567, vk1x01234567)); in xnn_f32_dwconv_ukernel_up8x4__avx()
101 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi2x01234567, vk2x01234567)); in xnn_f32_dwconv_ukernel_up8x4__avx()
105 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi3x01234567, vk3x01234567)); in xnn_f32_dwconv_ukernel_up8x4__avx()
Dup8x4-avx-acc2.c55 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up8x4__avx_acc2()
67 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi2x01234567, vk2x01234567)); in xnn_f32_dwconv_ukernel_up8x4__avx_acc2()
73 vacc01234567p1 = _mm256_add_ps(vacc01234567p1, _mm256_mul_ps(vi3x01234567, vk3x01234567)); in xnn_f32_dwconv_ukernel_up8x4__avx_acc2()
78 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, vacc01234567p1); in xnn_f32_dwconv_ukernel_up8x4__avx_acc2()
95 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi0x01234567, vk0x01234567)); in xnn_f32_dwconv_ukernel_up8x4__avx_acc2()
103 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, _mm256_mul_ps(vi2x01234567, vk2x01234567)); in xnn_f32_dwconv_ukernel_up8x4__avx_acc2()
107 vacc01234567p1 = _mm256_add_ps(vacc01234567p1, _mm256_mul_ps(vi3x01234567, vk3x01234567)); in xnn_f32_dwconv_ukernel_up8x4__avx_acc2()
110 vacc01234567p0 = _mm256_add_ps(vacc01234567p0, vacc01234567p1); in xnn_f32_dwconv_ukernel_up8x4__avx_acc2()
/external/XNNPACK/src/f32-vscaleextexp/gen/
Davx2-p5-x96.c191 __m256 ve0 = _mm256_add_ps(vn0, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x96()
192 __m256 ve1 = _mm256_add_ps(vn1, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x96()
193 __m256 ve2 = _mm256_add_ps(vn2, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x96()
194 __m256 ve3 = _mm256_add_ps(vn3, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x96()
195 __m256 ve4 = _mm256_add_ps(vn4, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x96()
196 __m256 ve5 = _mm256_add_ps(vn5, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x96()
197 __m256 ve6 = _mm256_add_ps(vn6, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x96()
198 __m256 ve7 = _mm256_add_ps(vn7, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x96()
199 __m256 ve8 = _mm256_add_ps(vn8, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x96()
200 __m256 ve9 = _mm256_add_ps(vn9, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x96()
[all …]
Davx2-p5-x88.c181 __m256 ve0 = _mm256_add_ps(vn0, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x88()
182 __m256 ve1 = _mm256_add_ps(vn1, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x88()
183 __m256 ve2 = _mm256_add_ps(vn2, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x88()
184 __m256 ve3 = _mm256_add_ps(vn3, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x88()
185 __m256 ve4 = _mm256_add_ps(vn4, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x88()
186 __m256 ve5 = _mm256_add_ps(vn5, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x88()
187 __m256 ve6 = _mm256_add_ps(vn6, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x88()
188 __m256 ve7 = _mm256_add_ps(vn7, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x88()
189 __m256 ve8 = _mm256_add_ps(vn8, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x88()
190 __m256 ve9 = _mm256_add_ps(vn9, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x88()
[all …]
Davx2-p5-x80.c171 __m256 ve0 = _mm256_add_ps(vn0, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x80()
172 __m256 ve1 = _mm256_add_ps(vn1, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x80()
173 __m256 ve2 = _mm256_add_ps(vn2, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x80()
174 __m256 ve3 = _mm256_add_ps(vn3, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x80()
175 __m256 ve4 = _mm256_add_ps(vn4, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x80()
176 __m256 ve5 = _mm256_add_ps(vn5, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x80()
177 __m256 ve6 = _mm256_add_ps(vn6, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x80()
178 __m256 ve7 = _mm256_add_ps(vn7, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x80()
179 __m256 ve8 = _mm256_add_ps(vn8, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x80()
180 __m256 ve9 = _mm256_add_ps(vn9, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x80()
[all …]
Davx2-p5-x72.c161 __m256 ve0 = _mm256_add_ps(vn0, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x72()
162 __m256 ve1 = _mm256_add_ps(vn1, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x72()
163 __m256 ve2 = _mm256_add_ps(vn2, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x72()
164 __m256 ve3 = _mm256_add_ps(vn3, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x72()
165 __m256 ve4 = _mm256_add_ps(vn4, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x72()
166 __m256 ve5 = _mm256_add_ps(vn5, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x72()
167 __m256 ve6 = _mm256_add_ps(vn6, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x72()
168 __m256 ve7 = _mm256_add_ps(vn7, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x72()
169 __m256 ve8 = _mm256_add_ps(vn8, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x72()
188 …const __m256 vs0 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve0, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x72()
[all …]
Davx2-p5-x56.c141 __m256 ve0 = _mm256_add_ps(vn0, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x56()
142 __m256 ve1 = _mm256_add_ps(vn1, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x56()
143 __m256 ve2 = _mm256_add_ps(vn2, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x56()
144 __m256 ve3 = _mm256_add_ps(vn3, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x56()
145 __m256 ve4 = _mm256_add_ps(vn4, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x56()
146 __m256 ve5 = _mm256_add_ps(vn5, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x56()
147 __m256 ve6 = _mm256_add_ps(vn6, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x56()
164 …const __m256 vs0 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve0, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x56()
165 …const __m256 vs1 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve1, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x56()
166 …const __m256 vs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve2, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x56()
[all …]
Davx2-p5-x64.c151 __m256 ve0 = _mm256_add_ps(vn0, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x64()
152 __m256 ve1 = _mm256_add_ps(vn1, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x64()
153 __m256 ve2 = _mm256_add_ps(vn2, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x64()
154 __m256 ve3 = _mm256_add_ps(vn3, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x64()
155 __m256 ve4 = _mm256_add_ps(vn4, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x64()
156 __m256 ve5 = _mm256_add_ps(vn5, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x64()
157 __m256 ve6 = _mm256_add_ps(vn6, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x64()
158 __m256 ve7 = _mm256_add_ps(vn7, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x64()
176 …const __m256 vs0 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve0, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x64()
177 …const __m256 vs1 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve1, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x64()
[all …]
Davx2-p5-x48.c131 __m256 ve0 = _mm256_add_ps(vn0, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x48()
132 __m256 ve1 = _mm256_add_ps(vn1, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x48()
133 __m256 ve2 = _mm256_add_ps(vn2, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x48()
134 __m256 ve3 = _mm256_add_ps(vn3, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x48()
135 __m256 ve4 = _mm256_add_ps(vn4, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x48()
136 __m256 ve5 = _mm256_add_ps(vn5, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x48()
152 …const __m256 vs0 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve0, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x48()
153 …const __m256 vs1 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve1, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x48()
154 …const __m256 vs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve2, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x48()
155 …const __m256 vs3 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve3, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x48()
[all …]
Davx2-p5-x40.c121 __m256 ve0 = _mm256_add_ps(vn0, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x40()
122 __m256 ve1 = _mm256_add_ps(vn1, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x40()
123 __m256 ve2 = _mm256_add_ps(vn2, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x40()
124 __m256 ve3 = _mm256_add_ps(vn3, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x40()
125 __m256 ve4 = _mm256_add_ps(vn4, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x40()
140 …const __m256 vs0 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve0, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x40()
141 …const __m256 vs1 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve1, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x40()
142 …const __m256 vs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve2, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x40()
143 …const __m256 vs3 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve3, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x40()
144 …const __m256 vs4 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve4, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x40()
[all …]
/external/XNNPACK/src/f32-raddextexp/gen/
Davx2-p5-x96-acc6.c231 …const __m256 vaccs0 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x96_acc6()
232 …const __m256 vaccs1 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x96_acc6()
233 …const __m256 vaccs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x96_acc6()
234 …const __m256 vaccs3 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x96_acc6()
235 …const __m256 vaccs4 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x96_acc6()
236 …const __m256 vaccs5 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x96_acc6()
237 …const __m256 vs0 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x96_acc6()
238 …const __m256 vs1 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x96_acc6()
239 …const __m256 vs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x96_acc6()
240 …const __m256 vs3 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x96_acc6()
[all …]
Davx2-p5-x64-acc4.c181 …const __m256 vaccs0 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x64_acc4()
182 …const __m256 vaccs1 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x64_acc4()
183 …const __m256 vaccs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x64_acc4()
184 …const __m256 vaccs3 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x64_acc4()
185 …const __m256 vs0 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x64_acc4()
186 …const __m256 vs1 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x64_acc4()
187 …const __m256 vs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x64_acc4()
188 …const __m256 vs3 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x64_acc4()
189 …const __m256 vs4 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x64_acc4()
190 …const __m256 vs5 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x64_acc4()
[all …]
Davx2-p5-x80-acc5.c206 …const __m256 vaccs0 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x80_acc5()
207 …const __m256 vaccs1 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x80_acc5()
208 …const __m256 vaccs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x80_acc5()
209 …const __m256 vaccs3 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x80_acc5()
210 …const __m256 vaccs4 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x80_acc5()
211 …const __m256 vs0 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x80_acc5()
212 …const __m256 vs1 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x80_acc5()
213 …const __m256 vs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x80_acc5()
214 …const __m256 vs3 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x80_acc5()
215 …const __m256 vs4 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x80_acc5()
[all …]
Davx2-p5-x72-acc3.c189 …const __m256 vaccs0 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x72_acc3()
190 …const __m256 vaccs1 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x72_acc3()
191 …const __m256 vaccs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdel… in xnn_f32_raddextexp_ukernel__avx2_p5_x72_acc3()
192 …const __m256 vs0 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x72_acc3()
193 …const __m256 vs1 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x72_acc3()
194 …const __m256 vs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x72_acc3()
195 …const __m256 vs3 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x72_acc3()
196 …const __m256 vs4 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x72_acc3()
197 …const __m256 vs5 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x72_acc3()
198 …const __m256 vs6 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_… in xnn_f32_raddextexp_ukernel__avx2_p5_x72_acc3()
[all …]
/external/libaom/libaom/aom_dsp/x86/
Dfft_avx2.c24 _mm256_store_ps, _mm256_set1_ps, _mm256_add_ps, _mm256_sub_ps,
27 _mm256_store_ps, _mm256_set1_ps, _mm256_add_ps, _mm256_sub_ps,
30 _mm256_store_ps, _mm256_set1_ps, _mm256_add_ps, _mm256_sub_ps,
50 _mm256_store_ps, _mm256_set1_ps, _mm256_add_ps, _mm256_sub_ps,
53 _mm256_store_ps, _mm256_set1_ps, _mm256_add_ps, _mm256_sub_ps,
56 _mm256_store_ps, _mm256_set1_ps, _mm256_add_ps, _mm256_sub_ps,

1234567