/external/libaom/libaom/av1/common/x86/ |
D | reconinter_avx2.c | 57 const __m256i m8 = _mm256_packus_epi16(m16, _mm256_setzero_si256()); in av1_build_compound_diffwtd_mask_avx2() 82 const __m256i m8 = _mm256_packus_epi16(m16AC, m16BD); 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() 195 const __m256i m8 = _mm256_packus_epi16(m16, _mm256_setzero_si256()); in build_compound_diffwtd_mask_d16_avx2() 215 const __m256i m8 = _mm256_packus_epi16(m16AB, m16CD); in build_compound_diffwtd_mask_d16_avx2() 232 const __m256i m8 = _mm256_packus_epi16(m16A, m16B); in build_compound_diffwtd_mask_d16_avx2() 249 const __m256i m8 = _mm256_packus_epi16(m16A, m16B); in build_compound_diffwtd_mask_d16_avx2() 274 const __m256i m8AB = _mm256_packus_epi16(m16A, m16B); in build_compound_diffwtd_mask_d16_avx2() 275 const __m256i m8CD = _mm256_packus_epi16(m16C, m16D); in build_compound_diffwtd_mask_d16_avx2() [all …]
|
D | convolve_avx2.c | 100 __m256i res_8b_lo = _mm256_packus_epi16(res_16b_lo, res_16b_lo); in av1_convolve_y_sr_avx2() 110 __m256i res_8b_hi = _mm256_packus_epi16(res_16b_hi, res_16b_hi); in av1_convolve_y_sr_avx2() 213 __m256i res_8b_lo = _mm256_packus_epi16(res_16b_lo, res_16b_lo); in av1_convolve_y_sr_avx2() 223 __m256i res_8b_hi = _mm256_packus_epi16(res_16b_hi, res_16b_hi); in av1_convolve_y_sr_avx2() 318 __m256i res_8b = _mm256_packus_epi16(res_16b, res_16b); in av1_convolve_x_sr_avx2() 356 __m256i res_8b = _mm256_packus_epi16(res_16b, res_16b); in av1_convolve_x_sr_avx2() 391 __m256i res_8b = _mm256_packus_epi16(res_16b, res_16b); in av1_convolve_x_sr_avx2() 428 __m256i res_8b = _mm256_packus_epi16(res_16b, res_16b); in av1_convolve_x_sr_avx2()
|
D | jnt_convolve_avx2.c | 108 const __m256i res_8 = _mm256_packus_epi16(round_result, round_result); in av1_dist_wtd_convolve_x_avx2() 163 const __m256i res_8 = _mm256_packus_epi16(round_result, round_result); in av1_dist_wtd_convolve_x_avx2() 306 _mm256_packus_epi16(round_result, round_result); in av1_dist_wtd_convolve_y_avx2() 376 _mm256_packus_epi16(round_result_lo, round_result_hi); in av1_dist_wtd_convolve_y_avx2() 485 _mm256_packus_epi16(round_result, round_result); in av1_dist_wtd_convolve_y_avx2() 555 _mm256_packus_epi16(round_result_lo, round_result_hi); in av1_dist_wtd_convolve_y_avx2() 731 _mm256_packus_epi16(round_result, round_result); in av1_dist_wtd_convolve_2d_avx2() 762 _mm256_packus_epi16(round_result, round_result); in av1_dist_wtd_convolve_2d_avx2() 854 const __m256i res_8 = _mm256_packus_epi16(round_result, round_result); in av1_dist_wtd_convolve_2d_copy_avx2() 892 const __m256i res_8 = _mm256_packus_epi16(round_result, round_result); in av1_dist_wtd_convolve_2d_copy_avx2()
|
D | av1_inv_txfm_avx2.h | 50 _mm256_permute4x64_epi64(_mm256_packus_epi16(u, u), 168)); in write_recon_w16_avx2()
|
D | wiener_convolve_avx2.c | 191 const __m256i res_8b = _mm256_packus_epi16(res_16bit, res_16bit); in av1_wiener_convolve_add_src_avx2()
|
D | convolve_2d_avx2.c | 158 const __m256i res_8b = _mm256_packus_epi16(res_16bit, res_16bit); in av1_convolve_2d_sr_avx2()
|
D | cfl_avx2.c | 267 res = _mm256_packus_epi16(res, next); in cfl_predict_lbd_avx2()
|
D | warp_plane_avx2.c | 699 const __m256i res_8_lo = _mm256_packus_epi16(res_lo_16, res_lo_16); in store_vertical_filter_output_avx2() 742 __m256i res_8_hi = _mm256_packus_epi16(res_hi_16, res_hi_16); in store_vertical_filter_output_avx2() 761 const __m256i res_8bit = _mm256_packus_epi16(res_16bit, res_16bit); in store_vertical_filter_output_avx2()
|
D | selfguided_avx2.c | 717 _mm256_packus_epi16(tmp2, tmp2 /* "don't care" value */); in av1_apply_selfguided_restoration_avx2()
|
/external/libaom/libaom/aom_dsp/x86/ |
D | variance_impl_avx2.c | 301 src_pack = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in aom_sub_pixel_variance32xh_avx2() 307 src_reg = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in aom_sub_pixel_variance32xh_avx2() 332 src_pack = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in aom_sub_pixel_variance32xh_avx2() 338 src_reg = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in aom_sub_pixel_variance32xh_avx2() 514 src_pack = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in aom_sub_pixel_variance16xh_avx2() 521 src_reg = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in aom_sub_pixel_variance16xh_avx2() 557 src_pack = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in aom_sub_pixel_variance16xh_avx2() 564 src_reg = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in aom_sub_pixel_variance16xh_avx2() 645 src_reg = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in aom_sub_pixel_avg_variance32xh_avx2() 713 src_avg = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in aom_sub_pixel_avg_variance32xh_avx2() [all …]
|
D | aom_subpixel_8t_intrin_avx2.c | 121 _mm256_packus_epi16(srcRegFilt32b1_1, _mm256_setzero_si256()); in aom_filter_block1d4_h4_avx2() 217 _mm256_packus_epi16(srcRegFilt32b1_1, _mm256_setzero_si256()); in aom_filter_block1d4_h8_avx2() 315 srcRegFilt32b1_1 = _mm256_packus_epi16(srcRegFilt32b1_1, srcRegFilt32b1_1); in aom_filter_block1d8_h4_avx2() 431 _mm256_packus_epi16(srcRegFilt32b1_1, _mm256_setzero_si256()); in aom_filter_block1d8_h8_avx2() 560 srcRegFilt32b1_1 = _mm256_packus_epi16(srcRegFilt32b1_1, srcRegFilt32b2_1); in aom_filter_block1d16_h4_avx2() 595 srcRegFilt1_1 = _mm256_packus_epi16(srcRegFilt1_1, srcRegFilt1_1); in aom_filter_block1d16_h4_avx2() 707 srcRegFilt32b1_1 = _mm256_packus_epi16(srcRegFilt32b1_1, srcRegFilt32b2_1); in aom_filter_block1d16_h8_avx2() 872 resReg = _mm256_packus_epi16(resReglo, resReglo); in aom_filter_block1d8_v4_avx2() 980 srcReg32b1 = _mm256_packus_epi16(srcReg32b10, _mm256_setzero_si256()); in aom_filter_block1d8_v8_avx2() 1121 resReg = _mm256_packus_epi16(resReglo, resReghi); in aom_filter_block1d16_v4_avx2() [all …]
|
D | convolve_avx2.h | 127 const __m256i res_8b = _mm256_packus_epi16(res_16bit, res_16bit); \ 215 const __m256i res_8 = _mm256_packus_epi16(round_result, round_result); \ 244 const __m256i res_8 = _mm256_packus_epi16(round_result, round_result); \
|
D | masked_sad_intrin_avx2.c | 52 const __m256i pred = _mm256_packus_epi16(pred_l, pred_r); in masked_sad32xh_avx2() 106 const __m256i pred = _mm256_packus_epi16(pred_l, pred_r); in masked_sad16xh_avx2()
|
D | blend_a64_mask_avx2.c | 44 __m256i res = _mm256_packus_epi16(res0, res0); in blend_a64_d16_mask_w16_avx2() 77 __m256i res = _mm256_packus_epi16(res0, res1); in blend_a64_d16_mask_w32_avx2() 418 const __m256i v_res_b = _mm256_packus_epi16(v_res0_w, v_res0_w); in blend_16_u8_avx2() 439 const __m256i v_res = _mm256_packus_epi16(v_res0_w, v_res1_w); in blend_32_u8_avx2() 459 const __m256i v_m0_b = _mm256_packus_epi16(v_m0_w, v_m0_w); in blend_a64_mask_sx_sy_w16_avx2() 500 _mm256_permute4x64_epi64(_mm256_packus_epi16(v_m0l_w, v_m0h_w), 0xd8); in blend_a64_mask_sx_sy_w32n_avx2() 592 const __m256i v_m0_b = _mm256_packus_epi16(v_m0_w, _mm256_setzero_si256()); in blend_a64_mask_sx_w16_avx2()
|
D | intrapred_avx2.c | 740 const __m256i p = _mm256_packus_epi16(p0, p1); in paeth_16x1_pred() 852 const __m256i x0 = _mm256_packus_epi16(p0, p1); in paeth_32x1_pred() 856 const __m256i x1 = _mm256_packus_epi16(p0, p1); in paeth_32x1_pred() 3621 res = _mm256_packus_epi16( in dr_prediction_z1_HxW_internal_avx2() 3719 res16[jj] = _mm256_packus_epi16( in dr_prediction_z1_32xN_internal_avx2() 3803 res = _mm256_packus_epi16( in dr_prediction_z1_64xN_avx2() 4171 resx = _mm256_castsi256_si128(_mm256_packus_epi16( in dr_prediction_z2_HxW_avx2() 4244 resy = _mm256_castsi256_si128(_mm256_packus_epi16( in dr_prediction_z2_HxW_avx2()
|
D | variance_avx2.c | 361 const __m256i roundA = _mm256_packus_epi16(roundAL, roundAH); in comp_mask_pred_line_avx2()
|
/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 | 96 outReg32b1 = _mm256_packus_epi16(outReg32b1, outReg32b2); in vpx_filter_block1d16_h8_x_avx2() 255 s1[0] = _mm256_packus_epi16(s1[0], s2[0]); in vpx_filter_block1d16_v8_x_avx2() 406 dst_first = _mm256_packus_epi16(dst_first, dst_second); in vpx_filter_block1d16_h4_avx2() 429 dst_first = _mm256_packus_epi16(dst_first, dst_first); in vpx_filter_block1d16_h4_avx2() 522 res_reg = _mm256_packus_epi16(res_reg_lo, res_reg_hi); in vpx_filter_block1d16_v4_avx2() 599 dst_reg = _mm256_packus_epi16(dst_reg, dst_reg); in vpx_filter_block1d8_h4_avx2() 709 res_reg = _mm256_packus_epi16(res_reg, res_reg); in vpx_filter_block1d8_v4_avx2() 774 dst = _mm256_packus_epi16(dst, _mm256_setzero_si256()); in vpx_filter_block1d4_h4_avx2() 880 res_reg = _mm256_packus_epi16(res_reg, res_reg); in vpx_filter_block1d4_v4_avx2()
|
D | variance_avx2.c | 361 const __m256i exp_src = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in spv32_bilin_zero() 418 const __m256i exp_src_avg = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in spv32_x4_yb() 448 src_pack = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in spv32_xb_y4() 460 src_reg = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in spv32_xb_y4() 501 prev_src_pack = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in spv32_xb_yb() 512 src_pack = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in spv32_xb_yb() 521 const __m256i exp_src = _mm256_packus_epi16(exp_src_lo, exp_src_hi); in spv32_xb_yb()
|
/external/mesa3d/src/gallium/drivers/swr/rasterizer/common/ |
D | simdlib_256_avx512_core.inl | 94 SIMD_IWRAPPER_2_8(packus_epi16); // uint16 --> uint8 See documentation for _mm256_packus_epi16 and
|
D | simdlib_128_avx512_core.inl | 144 SIMD_IWRAPPER_2_8(packus_epi16); // uint16 --> uint8 See documentation for _mm256_packus_epi16 and
|
D | simdlib_256_avx2.inl | 181 SIMD_IWRAPPER_2(packus_epi16); // See documentation for _mm256_packus_epi16 and _mm512_packus_epi16
|
D | simdlib_256_avx512.inl | 225 // uint8 See documentation for _mm256_packus_epi16 and _mm512_packus_epi16
|
D | simdlib_128_avx512.inl | 225 // uint8 See documentation for _mm256_packus_epi16 and _mm512_packus_epi16
|
/external/clang/test/CodeGen/ |
D | avx2-builtins.c | 883 return _mm256_packus_epi16(a, b); in test_mm256_packs_epu16()
|