| /external/libaom/libaom/aom_dsp/simd/ |
| D | v256_intrinsics_x86.h | 143 _mm256_permute4x64_epi64(b, _MM_SHUFFLE(3, 1, 2, 0)), in v256_ziplo_8()
144 _mm256_permute4x64_epi64(a, _MM_SHUFFLE(3, 1, 2, 0))); in v256_ziplo_8()
149 _mm256_permute4x64_epi64(b, _MM_SHUFFLE(3, 1, 2, 0)), in v256_ziphi_8()
150 _mm256_permute4x64_epi64(a, _MM_SHUFFLE(3, 1, 2, 0))); in v256_ziphi_8()
155 _mm256_permute4x64_epi64(b, _MM_SHUFFLE(3, 1, 2, 0)), in v256_ziplo_16()
156 _mm256_permute4x64_epi64(a, _MM_SHUFFLE(3, 1, 2, 0))); in v256_ziplo_16()
161 _mm256_permute4x64_epi64(b, _MM_SHUFFLE(3, 1, 2, 0)), in v256_ziphi_16()
162 _mm256_permute4x64_epi64(a, _MM_SHUFFLE(3, 1, 2, 0))); in v256_ziphi_16()
167 _mm256_permute4x64_epi64(b, _MM_SHUFFLE(3, 1, 2, 0)), in v256_ziplo_32()
168 _mm256_permute4x64_epi64(a, _MM_SHUFFLE(3, 1, 2, 0))); in v256_ziplo_32()
[all …]
|
| /external/libaom/libaom/av1/common/x86/ |
| D | reconinter_avx2.c | 59 _mm256_castsi256_si128(_mm256_permute4x64_epi64(m8, 0xd8)); in av1_build_compound_diffwtd_mask_avx2()
104 _mm256_permute4x64_epi64(_mm256_packus_epi16(m16AL, m16BL), 0xd8); in av1_build_compound_diffwtd_mask_avx2()
126 _mm256_permute4x64_epi64(_mm256_packus_epi16(m16L, m16H), 0xd8); in av1_build_compound_diffwtd_mask_avx2()
197 _mm256_castsi256_si128(_mm256_permute4x64_epi64(m8, 0xd8))); in build_compound_diffwtd_mask_d16_avx2()
216 yy_storeu_256(mask, _mm256_permute4x64_epi64(m8, 0xd8)); in build_compound_diffwtd_mask_d16_avx2()
233 yy_storeu_256(mask, _mm256_permute4x64_epi64(m8, 0xd8)); in build_compound_diffwtd_mask_d16_avx2()
250 yy_storeu_256(mask, _mm256_permute4x64_epi64(m8, 0xd8)); in build_compound_diffwtd_mask_d16_avx2()
276 yy_storeu_256(mask, _mm256_permute4x64_epi64(m8AB, 0xd8)); in build_compound_diffwtd_mask_d16_avx2()
277 yy_storeu_256(mask + 32, _mm256_permute4x64_epi64(m8CD, 0xd8)); in build_compound_diffwtd_mask_d16_avx2()
321 yy_storeu_256(mask, _mm256_permute4x64_epi64(m8AB, 0xd8)); in build_compound_diffwtd_mask_d16_avx2()
[all …]
|
| D | cfl_avx2.c | 134 top = _mm256_permute4x64_epi64(top, _MM_SHUFFLE(3, 1, 2, 0)); in cfl_luma_subsampling_444_lbd_avx2()
180 hsum = _mm256_permute4x64_epi64(hsum, _MM_SHUFFLE(3, 1, 2, 0)); in cfl_luma_subsampling_420_hbd_avx2()
213 hsum = _mm256_permute4x64_epi64(hsum, _MM_SHUFFLE(3, 1, 2, 0)); in cfl_luma_subsampling_422_hbd_avx2()
266 res = _mm256_permute4x64_epi64(res, _MM_SHUFFLE(3, 1, 2, 0)); in cfl_predict_lbd_avx2()
383 a = _mm256_permute4x64_epi64(a, _MM_SHUFFLE(3, 1, 2, 0)); in fill_sum_epi32()
|
| D | av1_inv_txfm_avx2.h | 50 _mm256_permute4x64_epi64(_mm256_packus_epi16(u, u), 168)); in write_recon_w16_avx2()
|
| D | selfguided_avx2.c | 706 const __m256i tmp2 = _mm256_permute4x64_epi64(tmp, 0xd8); in apply_selfguided_restoration_avx2()
715 const __m256i tmp2 = _mm256_permute4x64_epi64(tmp, 0xd8); in apply_selfguided_restoration_avx2()
719 _mm256_castsi256_si128(_mm256_permute4x64_epi64(res, 0xd8)); in apply_selfguided_restoration_avx2()
|
| D | convolve_avx2.c | 360 res_8b = _mm256_permute4x64_epi64(res_8b, 216); in av1_convolve_x_sr_avx2()
432 res_8b = _mm256_permute4x64_epi64(res_8b, 216); in av1_convolve_x_sr_avx2()
|
| D | wiener_convolve_avx2.c | 253 const __m256i res_8bit2 = _mm256_permute4x64_epi64(res_8bit, 0xd8); in av1_wiener_convolve_add_src_avx2()
|
| D | jnt_convolve_avx2.c | 855 const __m256i res_0 = _mm256_permute4x64_epi64(res_8, 0xD8); in av1_dist_wtd_convolve_2d_copy_avx2()
|
| /external/libvpx/libvpx/vp9/encoder/x86/ |
| D | vp9_quantize_avx2.c | 38 const __m256i _coeff256 = _mm256_permute4x64_epi64(*coeff256, 0xd8); in scan_eob_256()
78 round256 = _mm256_permute4x64_epi64(round256, 0x54); in vp9_quantize_fp_avx2()
81 quant256 = _mm256_permute4x64_epi64(quant256, 0x54); in vp9_quantize_fp_avx2()
84 dequant256 = _mm256_permute4x64_epi64(dequant256, 0x54); in vp9_quantize_fp_avx2()
|
| /external/libaom/libaom/av1/encoder/x86/ |
| D | encodetxb_avx2.c | 48 const __m256i res = _mm256_permute4x64_epi64(res_, 0xd8); in av1_txb_init_levels_avx2()
65 const __m256i res_ = _mm256_permute4x64_epi64(absABCD, 0xd8); in av1_txb_init_levels_avx2()
92 const __m256i res_ = _mm256_permute4x64_epi64(absABCD, 0xd8); in av1_txb_init_levels_avx2()
113 const __m256i res_ = _mm256_permute4x64_epi64(absABCD, 0xd8); in av1_txb_init_levels_avx2()
|
| D | error_intrin_avx2.c | 26 *c = _mm256_permute4x64_epi64(y, 0xD8); in read_coeff()
|
| D | pickrst_avx2.c | 511 const __m256i flt0_16b = _mm256_permute4x64_epi64( in av1_lowbd_pixel_proj_error_avx2()
515 const __m256i flt1_16b = _mm256_permute4x64_epi64( in av1_lowbd_pixel_proj_error_avx2()
563 const __m256i flt_16b = _mm256_permute4x64_epi64( in av1_lowbd_pixel_proj_error_avx2()
689 _mm256_permute4x64_epi64(_mm256_packs_epi32(vrl, vrh), 0xd8); in av1_highbd_pixel_proj_error_avx2()
766 _mm256_permute4x64_epi64(_mm256_packs_epi32(vrl, vrh), 0xd8); in av1_highbd_pixel_proj_error_avx2()
|
| D | rdopt_avx2.c | 46 const __m256i perm = _mm256_permute4x64_epi64(slli, 0x90); in horver_correlation_4x4()
|
| D | av1_quantize_avx2.c | 24 *c = _mm256_permute4x64_epi64(*c, 0xD8); in read_coeff()
|
| /external/libvpx/libvpx/vpx_dsp/x86/ |
| D | loopfilter_avx2.c | 623 _mm256_permute4x64_epi64(_mm256_packus_epi16(res_p, res_p), 168)); in vpx_lpf_horizontal_16_dual_avx2()
629 _mm256_permute4x64_epi64(_mm256_packus_epi16(res_q, res_q), 168)); in vpx_lpf_horizontal_16_dual_avx2()
637 _mm256_permute4x64_epi64(_mm256_packus_epi16(res_p, res_p), 168)); in vpx_lpf_horizontal_16_dual_avx2()
645 _mm256_permute4x64_epi64(_mm256_packus_epi16(res_q, res_q), 168)); in vpx_lpf_horizontal_16_dual_avx2()
663 _mm256_permute4x64_epi64(_mm256_packus_epi16(res_p, res_p), 168)); in vpx_lpf_horizontal_16_dual_avx2()
669 _mm256_permute4x64_epi64(_mm256_packus_epi16(res_q, res_q), 168)); in vpx_lpf_horizontal_16_dual_avx2()
681 _mm256_permute4x64_epi64(_mm256_packus_epi16(res_p, res_p), 168)); in vpx_lpf_horizontal_16_dual_avx2()
689 _mm256_permute4x64_epi64(_mm256_packus_epi16(res_q, res_q), 168)); in vpx_lpf_horizontal_16_dual_avx2()
707 _mm256_permute4x64_epi64(_mm256_packus_epi16(res_p, res_p), 168)); in vpx_lpf_horizontal_16_dual_avx2()
713 _mm256_permute4x64_epi64(_mm256_packus_epi16(res_q, res_q), 168)); in vpx_lpf_horizontal_16_dual_avx2()
[all …]
|
| D | vpx_subpixel_8t_intrin_avx2.c | 418 src_reg = _mm256_permute4x64_epi64(src_reg, 0x94); in vpx_filter_block1d16_h4_avx2()
430 dst_first = _mm256_permute4x64_epi64(dst_first, 0x8); in vpx_filter_block1d16_h4_avx2()
|
| D | highbd_convolve_avx2.c | 1094 src_reg = _mm256_permute4x64_epi64(src_reg, 0x94); in vpx_highbd_filter_block1d8_h4_avx2()
1105 res_reg = _mm256_permute4x64_epi64(res_reg, 0x8); in vpx_highbd_filter_block1d8_h4_avx2()
|
| /external/libaom/libaom/aom_dsp/x86/ |
| D | blend_a64_mask_avx2.c | 45 res = _mm256_permute4x64_epi64(res, 0xd8); in blend_a64_d16_mask_w16_avx2()
78 res = _mm256_permute4x64_epi64(res, 0xd8); in blend_a64_d16_mask_w32_avx2()
410 const __m256i v_s0_s_b = _mm256_permute4x64_epi64(v_s0_b, 0xd8); in blend_16_u8_avx2()
411 const __m256i v_s1_s_b = _mm256_permute4x64_epi64(v_s1_b, 0xd8); in blend_16_u8_avx2()
419 const __m256i v_res = _mm256_permute4x64_epi64(v_res_b, 0xd8); in blend_16_u8_avx2()
500 _mm256_permute4x64_epi64(_mm256_packus_epi16(v_m0l_w, v_m0h_w), 0xd8); in blend_a64_mask_sx_sy_w32n_avx2()
625 _mm256_permute4x64_epi64(_mm256_unpacklo_epi64(v_al_b, v_ah_b), 0xd8); in blend_a64_mask_sx_w32n_avx2()
|
| D | masked_sad_intrin_avx2.c | 63 res = _mm256_permute4x64_epi64(res, 0xd8); in masked_sad32xh_avx2()
116 res = _mm256_permute4x64_epi64(res, 0xd8); in masked_sad16xh_avx2()
|
| D | txfm_common_avx2.h | 106 return _mm256_permute4x64_epi64(b, 0xD8); in load_32bit_to_16bit_w16_avx2()
|
| D | aom_subpixel_8t_intrin_avx2.c | 575 srcReg12 = _mm256_permute4x64_epi64(srcReg1, 0x94); in aom_filter_block1d16_h4_avx2()
596 srcRegFilt1_1 = _mm256_permute4x64_epi64(srcRegFilt1_1, 0x8); in aom_filter_block1d16_h4_avx2()
|
| /external/tensorflow/tensorflow/core/kernels/ |
| D | sparse_matmul_op.h | 329 return _mm256_castsi256_ps(_mm256_permute4x64_epi64(_mm256_castps_si256(from),
|
| /external/clang/test/CodeGen/ |
| D | avx2-builtins.c | 901 return _mm256_permute4x64_epi64(a, 35); in test_mm256_permute4x64_epi64()
|
| /external/mesa3d/src/gallium/drivers/swr/rasterizer/memory/ |
| D | StoreTile.h | 787 final = _mm256_permute4x64_epi64(final, 0xD8);
937 final = _mm256_permute4x64_epi64(final, 0xD8);
|
| /external/clang/lib/Headers/ |
| D | avx2intrin.h | 957 #define _mm256_permute4x64_epi64(V, M) __extension__ ({ \ macro
|