| /external/libaom/libaom/aom_dsp/x86/ |
| common_avx2.h | 129 out[0] = _mm256_permute2x128_si256(tr0_0, tr0_8, 0x20); // 0010 0000
130 out[8] = _mm256_permute2x128_si256(tr0_0, tr0_8, 0x31); // 0011 0001
131 out[1] = _mm256_permute2x128_si256(tr0_1, tr0_9, 0x20);
132 out[9] = _mm256_permute2x128_si256(tr0_1, tr0_9, 0x31);
133 out[2] = _mm256_permute2x128_si256(tr0_2, tr0_a, 0x20);
134 out[10] = _mm256_permute2x128_si256(tr0_2, tr0_a, 0x31);
135 out[3] = _mm256_permute2x128_si256(tr0_3, tr0_b, 0x20);
136 out[11] = _mm256_permute2x128_si256(tr0_3, tr0_b, 0x31);
138 out[4] = _mm256_permute2x128_si256(tr0_4, tr0_c, 0x20);
139 out[12] = _mm256_permute2x128_si256(tr0_4, tr0_c, 0x31)
[all...] |
| txfm_common_avx2.h | 150 out[0 + 0] = _mm256_permute2x128_si256(c[0], c[1], 0x20);
151 out[1 + 0] = _mm256_permute2x128_si256(c[8], c[9], 0x20);
152 out[2 + 0] = _mm256_permute2x128_si256(c[4], c[5], 0x20);
153 out[3 + 0] = _mm256_permute2x128_si256(c[12], c[13], 0x20);
155 out[0 + 8] = _mm256_permute2x128_si256(c[0], c[1], 0x31);
156 out[1 + 8] = _mm256_permute2x128_si256(c[8], c[9], 0x31);
157 out[2 + 8] = _mm256_permute2x128_si256(c[4], c[5], 0x31);
158 out[3 + 8] = _mm256_permute2x128_si256(c[12], c[13], 0x31);
160 out[4 + 0] = _mm256_permute2x128_si256(c[0 + 2], c[1 + 2], 0x20);
161 out[5 + 0] = _mm256_permute2x128_si256(c[8 + 2], c[9 + 2], 0x20)
[all...] |
| highbd_convolve_avx2.c | 148 __m256i s01 = _mm256_permute2x128_si256(
154 __m256i s12 = _mm256_permute2x128_si256(
160 __m256i s23 = _mm256_permute2x128_si256(
166 __m256i s34 = _mm256_permute2x128_si256(
172 __m256i s45 = _mm256_permute2x128_si256(
180 __m256i s56 = _mm256_permute2x128_si256(
196 const __m256i s67 = _mm256_permute2x128_si256(
205 const __m256i s78 = _mm256_permute2x128_si256(
306 const __m256i r0 = _mm256_permute2x128_si256(row0, row1, 0x20);
307 const __m256i r1 = _mm256_permute2x128_si256(row0, row1, 0x31)
[all...] |
| avg_intrin_avx2.c | 111 _mm256_permute2x128_si256(src[0], src[1], 0x20));
114 _mm256_permute2x128_si256(src[2], src[3], 0x20));
117 _mm256_permute2x128_si256(src[4], src[5], 0x20));
120 _mm256_permute2x128_si256(src[6], src[7], 0x20));
123 _mm256_permute2x128_si256(src[0], src[1], 0x31));
126 _mm256_permute2x128_si256(src[2], src[3], 0x31));
129 _mm256_permute2x128_si256(src[4], src[5], 0x31));
132 _mm256_permute2x128_si256(src[6], src[7], 0x31));
|
| aom_subpixel_8t_intrin_avx2.c | [all...] |
| highbd_quantize_intrin_avx2.c | 28 qp[i] = _mm256_permute2x128_si256(qp[i], qp[i], 0x11);
|
| intrapred_avx2.c | 24 __m256i u0 = _mm256_permute2x128_si256(y0, y0, 1);
34 __m256i u = _mm256_permute2x128_si256(y, y, 1);
318 __m256i r0 = _mm256_permute2x128_si256(t[i], t[i], 0);
319 __m256i r1 = _mm256_permute2x128_si256(t[i], t[i], 0x11);
811 return _mm256_permute2x128_si256(x0, x1, 0x20);
[all...] |
| /external/libaom/libaom/av1/common/x86/ |
| convolve_avx2.c | 63 const __m256i src_01a = _mm256_permute2x128_si256(
66 const __m256i src_12a = _mm256_permute2x128_si256(
69 const __m256i src_23a = _mm256_permute2x128_si256(
72 const __m256i src_34a = _mm256_permute2x128_si256(
84 const __m256i src_45a = _mm256_permute2x128_si256(
88 const __m256i src_56a = _mm256_permute2x128_si256(
160 const __m256i src_01a = _mm256_permute2x128_si256(
163 const __m256i src_12a = _mm256_permute2x128_si256(
166 const __m256i src_23a = _mm256_permute2x128_si256(
169 const __m256i src_34a = _mm256_permute2x128_si256(
[all...] |
| highbd_jnt_convolve_avx2.c | 132 _mm256_permute2x128_si256(data_0, data_1, 0x20);
176 _mm256_permute2x128_si256(data_0, data_1, 0x20);
294 const __m256i r0 = _mm256_permute2x128_si256(row0, row1, 0x20);
295 const __m256i r1 = _mm256_permute2x128_si256(row0, row1, 0x31);
368 _mm256_permute2x128_si256(data_0, data_1, 0x20);
413 _mm256_permute2x128_si256(data_0, data_1, 0x20);
517 const __m256i r0 = _mm256_permute2x128_si256(row0, row1, 0x20);
518 const __m256i r1 = _mm256_permute2x128_si256(row0, row1, 0x31);
554 _mm256_permute2x128_si256(data_0, data_1, 0x20);
594 _mm256_permute2x128_si256(data_0, data_1, 0x20)
[all...] |
| jnt_convolve_avx2.c | 33 return _mm256_permute2x128_si256(
249 _mm256_permute2x128_si256(src_a[kk], src_a[kk + 1], 0x20);
263 const __m256i src_45a = _mm256_permute2x128_si256(src4, src5, 0x20);
267 const __m256i src_56a = _mm256_permute2x128_si256(src5, src4, 0x20);
427 _mm256_permute2x128_si256(src_a[kk], src_a[kk + 1], 0x20);
442 const __m256i src_67a = _mm256_permute2x128_si256(src6, src7, 0x20);
446 const __m256i src_78a = _mm256_permute2x128_si256(src7, src6, 0x20);
[all...] |
| highbd_convolve_2d_avx2.c | 75 const __m256i r0 = _mm256_permute2x128_si256(row0, row1, 0x20);
76 const __m256i r1 = _mm256_permute2x128_si256(row0, row1, 0x31);
|
| /external/libvpx/libvpx/vp9/encoder/x86/ |
| vp9_quantize_avx2.c | 105 dequant256 = _mm256_permute2x128_si256(dequant256, dequant256, 0x31);
106 quant256 = _mm256_permute2x128_si256(quant256, quant256, 0x31);
107 round256 = _mm256_permute2x128_si256(round256, round256, 0x31);
|
| /external/libvpx/libvpx/vpx_dsp/x86/ |
| highbd_convolve_avx2.c | 236 x[0] = _mm256_permute2x128_si256(pp[0], pp[4], 0x20);
237 x[1] = _mm256_permute2x128_si256(pp[1], pp[5], 0x20);
238 x[2] = _mm256_permute2x128_si256(pp[2], pp[6], 0x20);
239 x[3] = _mm256_permute2x128_si256(pp[3], pp[7], 0x20);
242 x[6] = _mm256_permute2x128_si256(pp[0], pp[4], 0x31);
243 x[7] = _mm256_permute2x128_si256(pp[1], pp[5], 0x31);
251 x[0] = _mm256_permute2x128_si256(pp[0], pp[2], 0x30);
252 x[1] = _mm256_permute2x128_si256(pp[1], pp[3], 0x30);
253 x[2] = _mm256_permute2x128_si256(pp[2], pp[0], 0x30);
254 x[3] = _mm256_permute2x128_si256(pp[3], pp[1], 0x30)
[all...] |
| avg_intrin_avx2.c | 75 in[0] = _mm256_permute2x128_si256(b0, b1, 0x20);
76 in[1] = _mm256_permute2x128_si256(b0, b1, 0x31);
77 in[2] = _mm256_permute2x128_si256(b2, b3, 0x20);
78 in[3] = _mm256_permute2x128_si256(b2, b3, 0x31);
79 in[4] = _mm256_permute2x128_si256(b4, b5, 0x20);
80 in[5] = _mm256_permute2x128_si256(b4, b5, 0x31);
81 in[6] = _mm256_permute2x128_si256(b6, b7, 0x20);
82 in[7] = _mm256_permute2x128_si256(b6, b7, 0x31);
313 _mm256_permute2x128_si256(src[0], src[1], 0x20));
316 _mm256_permute2x128_si256(src[2], src[3], 0x20))
[all...] |
| vpx_subpixel_8t_intrin_avx2.c | 488 src_reg_01 = _mm256_permute2x128_si256(src_reg_m10, src_reg_1, 0x21);
687 src_reg_01 = _mm256_permute2x128_si256(src_reg_m10, src_reg_1, 0x21);
[all...] |
| /external/libaom/libaom/av1/encoder/x86/ |
| av1_highbd_quantize_avx2.c | 27 qp[0] = _mm256_permute2x128_si256(qp[0], qp[0], 0x11);
28 qp[1] = _mm256_permute2x128_si256(qp[1], qp[1], 0x11);
29 qp[2] = _mm256_permute2x128_si256(qp[2], qp[2], 0x11);
|
| av1_quantize_avx2.c | 70 qp[0] = _mm256_permute2x128_si256(qp[0], qp[0], 0x11);
71 qp[1] = _mm256_permute2x128_si256(qp[1], qp[1], 0x11);
72 qp[2] = _mm256_permute2x128_si256(qp[2], qp[2], 0x11);
81 __m256i x0 = _mm256_permute2x128_si256(y0, y1, 0x20); \
82 __m256i x1 = _mm256_permute2x128_si256(y0, y1, 0x31); \
|
| av1_fwd_txfm2d_avx2.c | [all...] |
| /external/eigen/Eigen/src/Core/arch/CUDA/ |
| PacketMathHalf.h | 604 __m256i a_p_0 = _mm256_permute2x128_si256(abcdefgh_01, ijklmnop_01, 0x20);
605 __m256i a_p_1 = _mm256_permute2x128_si256(abcdefgh_01, ijklmnop_01, 0x31);
606 __m256i a_p_2 = _mm256_permute2x128_si256(abcdefgh_23, ijklmnop_23, 0x20);
607 __m256i a_p_3 = _mm256_permute2x128_si256(abcdefgh_23, ijklmnop_23, 0x31);
608 __m256i a_p_4 = _mm256_permute2x128_si256(abcdefgh_45, ijklmnop_45, 0x20);
609 __m256i a_p_5 = _mm256_permute2x128_si256(abcdefgh_45, ijklmnop_45, 0x31);
610 __m256i a_p_6 = _mm256_permute2x128_si256(abcdefgh_67, ijklmnop_67, 0x20);
611 __m256i a_p_7 = _mm256_permute2x128_si256(abcdefgh_67, ijklmnop_67, 0x31);
612 __m256i a_p_8 = _mm256_permute2x128_si256(abcdefgh_89, ijklmnop_89, 0x20);
613 __m256i a_p_9 = _mm256_permute2x128_si256(abcdefgh_89, ijklmnop_89, 0x31)
[all...] |
| /external/libaom/libaom/aom_dsp/simd/ |
| v256_intrinsics_x86.h | 350 _mm256_permute2x128_si256(a, a, _MM_SHUFFLE(0, 1, 0, 1)), pattern),
352 _mm256_permute2x128_si256(a, a, _MM_SHUFFLE(0, 0, 0, 0)), pattern),
361 _mm256_permute2x128_si256(a, b, _MM_SHUFFLE(0, 1, 0, 1)), p32),
363 _mm256_permute2x128_si256(a, b, _MM_SHUFFLE(0, 0, 0, 0)), p32),
367 _mm256_permute2x128_si256(a, b, _MM_SHUFFLE(0, 3, 0, 3)), pattern),
369 _mm256_permute2x128_si256(a, b, _MM_SHUFFLE(0, 2, 0, 2)), pattern),
396 r, _mm256_permute2x128_si256(r, r, _MM_SHUFFLE(2, 0, 0, 1))));
419 r, _mm256_permute2x128_si256(r, r, _MM_SHUFFLE(2, 0, 0, 1))));
672 _mm256_permute2x128_si256(a, a, _MM_SHUFFLE(2, 0, 0, 1)), a, n) \
|
| /external/clang/test/CodeGen/ |
| avx2-builtins.c | [all...] |
| /external/mesa3d/src/gallium/drivers/swr/rasterizer/core/ |
| format_utils.h | 243 simdscalari dst23 = _mm256_permute2x128_si256(src, src, 0x01);
[all...] |
| /external/clang/lib/Headers/ |
| avx2intrin.h | 965 #define _mm256_permute2x128_si256 macro
[all...] |
| /external/skia/src/opts/ |
| SkRasterPipeline_opts.h | [all...] |
| /external/skqp/src/opts/ |
| SkRasterPipeline_opts.h | [all...] |
