/external/libaom/libaom/av1/encoder/x86/ |
D | encodetxb_avx2.c | 45 const __m256i abs01 = _mm256_abs_epi16(_mm256_packs_epi32(c0, c1)); in av1_txb_init_levels_avx2() 60 const __m256i coeffAB = _mm256_packs_epi32(coeffA, coeffB); in av1_txb_init_levels_avx2() 61 const __m256i coeffCD = _mm256_packs_epi32(coeffC, coeffD); in av1_txb_init_levels_avx2() 87 const __m256i coeffAB = _mm256_packs_epi32(coeffA, coeffB); in av1_txb_init_levels_avx2() 88 const __m256i coeffCD = _mm256_packs_epi32(coeffC, coeffD); in av1_txb_init_levels_avx2() 108 const __m256i coeffAB = _mm256_packs_epi32(coeffA, coeffB); in av1_txb_init_levels_avx2() 109 const __m256i coeffCD = _mm256_packs_epi32(coeffC, coeffD); in av1_txb_init_levels_avx2()
|
D | wedge_utils_avx2.c | 59 const __m256i v_t0_w = _mm256_packs_epi32(v_t0l_d, v_t0h_d); in av1_wedge_sse_from_residuals_avx2() 200 const __m256i v_r0_w = _mm256_packs_epi32(v_r0l_w, v_r0h_w); in av1_wedge_compute_delta_squares_avx2() 201 const __m256i v_r1_w = _mm256_packs_epi32(v_r1l_w, v_r1h_w); in av1_wedge_compute_delta_squares_avx2() 202 const __m256i v_r2_w = _mm256_packs_epi32(v_r2l_w, v_r2h_w); in av1_wedge_compute_delta_squares_avx2() 203 const __m256i v_r3_w = _mm256_packs_epi32(v_r3l_w, v_r3h_w); in av1_wedge_compute_delta_squares_avx2()
|
D | pickrst_avx2.c | 512 _mm256_packs_epi32(yy_loadu_256(flt0 + j), in av1_lowbd_pixel_proj_error_avx2() 516 _mm256_packs_epi32(yy_loadu_256(flt1 + j), in av1_lowbd_pixel_proj_error_avx2() 531 _mm256_add_epi16(_mm256_packs_epi32(vr0, vr1), d0), s0); in av1_lowbd_pixel_proj_error_avx2() 564 _mm256_packs_epi32(yy_loadu_256(flt + j), in av1_lowbd_pixel_proj_error_avx2() 576 _mm256_add_epi16(_mm256_packs_epi32(vr0, vr1), d0), s0); 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 | error_intrin_avx2.c | 25 const __m256i y = _mm256_packs_epi32(x0, x1); in read_coeff()
|
D | av1_quantize_avx2.c | 23 *c = _mm256_packs_epi32(x0, x1); in read_coeff()
|
/external/libvpx/libvpx/vpx_dsp/x86/ |
D | fwd_dct32x32_impl_avx2.h | 431 step2[20] = _mm256_packs_epi32(s2_20_6, s2_20_7); in FDCT32x32_2D_AVX2() 432 step2[21] = _mm256_packs_epi32(s2_21_6, s2_21_7); in FDCT32x32_2D_AVX2() 433 step2[22] = _mm256_packs_epi32(s2_22_6, s2_22_7); in FDCT32x32_2D_AVX2() 434 step2[23] = _mm256_packs_epi32(s2_23_6, s2_23_7); in FDCT32x32_2D_AVX2() 435 step2[24] = _mm256_packs_epi32(s2_24_6, s2_24_7); in FDCT32x32_2D_AVX2() 436 step2[25] = _mm256_packs_epi32(s2_25_6, s2_25_7); in FDCT32x32_2D_AVX2() 437 step2[26] = _mm256_packs_epi32(s2_26_6, s2_26_7); in FDCT32x32_2D_AVX2() 438 step2[27] = _mm256_packs_epi32(s2_27_6, s2_27_7); in FDCT32x32_2D_AVX2() 632 step3[10] = _mm256_packs_epi32(s3_10_6, s3_10_7); in FDCT32x32_2D_AVX2() 633 step3[11] = _mm256_packs_epi32(s3_11_6, s3_11_7); in FDCT32x32_2D_AVX2() [all …]
|
D | bitdepth_conversion_avx2.h | 25 return _mm256_packs_epi32(a_low, a_high); in load_tran_low()
|
/external/libaom/libaom/av1/common/x86/ |
D | highbd_convolve_2d_avx2.c | 98 __m256i res_even1 = _mm256_packs_epi32(res_even, res_even); in av1_highbd_convolve_2d_sr_avx2() 99 __m256i res_odd1 = _mm256_packs_epi32(res_odd, res_odd); in av1_highbd_convolve_2d_sr_avx2() 149 __m256i res_16bit = _mm256_packs_epi32(res_a_round, res_b_round); in av1_highbd_convolve_2d_sr_avx2() 158 res_a_round = _mm256_packs_epi32(res_a_round, res_a_round); in av1_highbd_convolve_2d_sr_avx2() 167 res_a_round = _mm256_packs_epi32(res_a_round, res_a_round); in av1_highbd_convolve_2d_sr_avx2()
|
D | jnt_convolve_avx2.c | 289 _mm256_packs_epi32(res_lo_0_round, res_lo_1_round); in av1_dist_wtd_convolve_y_avx2() 347 _mm256_packs_epi32(res_hi_0_round, res_hi_1_round); in av1_dist_wtd_convolve_y_avx2() 468 _mm256_packs_epi32(res_lo_0_round, res_lo_1_round); in av1_dist_wtd_convolve_y_avx2() 526 _mm256_packs_epi32(res_hi_0_round, res_hi_1_round); in av1_dist_wtd_convolve_y_avx2() 717 const __m256i res_16b = _mm256_packs_epi32(res_a_round, res_b_round); in av1_dist_wtd_convolve_2d_avx2() 747 const __m256i res_16b = _mm256_packs_epi32(res_a_round, res_a_round); in av1_dist_wtd_convolve_2d_avx2()
|
D | highbd_wiener_convolve_avx2.c | 130 const __m256i res = _mm256_packs_epi32(res_even, res_odd); in av1_highbd_wiener_convolve_add_src_avx2() 236 _mm256_packs_epi32(res_lo_round, res_hi_round); in av1_highbd_wiener_convolve_add_src_avx2()
|
D | wiener_convolve_avx2.c | 138 const __m256i res = _mm256_packs_epi32(res_even, res_odd); in av1_wiener_convolve_add_src_avx2() 242 _mm256_packs_epi32(res_lo_round, res_hi_round); in av1_wiener_convolve_add_src_avx2()
|
D | convolve_2d_avx2.c | 156 const __m256i res_16bit = _mm256_packs_epi32(res_a_round, res_b_round); in av1_convolve_2d_sr_avx2()
|
D | cfl_avx2.c | 438 avg_epi16 = _mm256_packs_epi32(avg_epi16, avg_epi16); in subtract_average_avx2()
|
D | highbd_jnt_convolve_avx2.c | 317 __m256i res_even1 = _mm256_packs_epi32(res_even, res_even); in av1_highbd_dist_wtd_convolve_2d_avx2() 318 __m256i res_odd1 = _mm256_packs_epi32(res_odd, res_odd); in av1_highbd_dist_wtd_convolve_2d_avx2()
|
/external/libaom/libaom/aom_dsp/x86/ |
D | highbd_convolve_avx2.c | 223 __m256i res_16bit = _mm256_packs_epi32(res_a_round, res_b_round); in av1_highbd_convolve_y_sr_avx2() 232 res_a_round = _mm256_packs_epi32(res_a_round, res_a_round); in av1_highbd_convolve_y_sr_avx2() 241 res_a_round = _mm256_packs_epi32(res_a_round, res_a_round); in av1_highbd_convolve_y_sr_avx2() 334 __m256i res_even1 = _mm256_packs_epi32(res_even, res_even); in av1_highbd_convolve_x_sr_avx2() 335 __m256i res_odd1 = _mm256_packs_epi32(res_odd, res_odd); in av1_highbd_convolve_x_sr_avx2() 602 res = _mm256_packs_epi32(res, res); in aom_highbd_filter_block1d4_h4_avx2() 628 res = _mm256_packs_epi32(res, res); in aom_highbd_filter_block1d4_h4_avx2() 687 __m256i res = _mm256_packs_epi32(res_even, res_odd); in aom_highbd_filter_block1d8_h4_avx2() 716 res = _mm256_packs_epi32(res, res); in aom_highbd_filter_block1d8_h4_avx2() 1114 res_16bit = _mm256_packs_epi32(res_16bit, res_16bit); in aom_highbd_filter_block1d4_v4_avx2() [all …]
|
D | convolve_avx2.h | 101 const __m256i res_16bit = _mm256_packs_epi32(res_a_round, res_b_round); \ 178 const __m256i res_16b = _mm256_packs_epi32(res_a_round, res_b_round); \ 206 const __m256i res_16b = _mm256_packs_epi32(res_a_round, res_a_round); \ 392 res = _mm256_packs_epi32(res_lo, res_hi); in comp_avg()
|
D | bitdepth_conversion_avx2.h | 21 return _mm256_packs_epi32(a_low, a_high); in load_tran_low()
|
D | txfm_common_avx2.h | 48 *in0 = _mm256_packs_epi32(c0, c1); in btf_16_w16_avx2() 49 *in1 = _mm256_packs_epi32(d0, d1); in btf_16_w16_avx2() 105 const __m256i b = _mm256_packs_epi32(a_low, *(const __m256i *)(a + 8)); in load_32bit_to_16bit_w16_avx2()
|
D | blend_a64_mask_avx2.c | 43 const __m256i res0 = _mm256_packs_epi32(res0_lo, res0_hi); in blend_a64_d16_mask_w16_avx2() 75 const __m256i res0 = _mm256_packs_epi32(res0_lo, res0_hi); in blend_a64_d16_mask_w32_avx2() 76 const __m256i res1 = _mm256_packs_epi32(res1_lo, res1_hi); in blend_a64_d16_mask_w32_avx2() 945 const __m256i pack = _mm256_packs_epi32(roundl, roundh); in highbd_blend_a64_d16_mask_w4_avx2() 1082 const __m256i packa = _mm256_packs_epi32(roundal, roundah); in highbd_blend_a64_d16_mask_w8_avx2() 1085 const __m256i packb = _mm256_packs_epi32(roundbl, roundbh); in highbd_blend_a64_d16_mask_w8_avx2() 1217 const __m256i packa = _mm256_packs_epi32(resal, resah); in highbd_blend_a64_d16_mask_w16_avx2() 1218 const __m256i packb = _mm256_packs_epi32(resbl, resbh); in highbd_blend_a64_d16_mask_w16_avx2()
|
D | masked_sad_intrin_avx2.c | 240 const __m256i pred = _mm256_packs_epi32(pred_l, pred_r); in highbd_masked_sad8xh_avx2() 297 const __m256i pred = _mm256_packs_epi32(pred_l, pred_r); in highbd_masked_sad16xh_avx2()
|
D | obmc_variance_avx2.c | 131 const __m256i v_rdiff01_w = _mm256_packs_epi32(v_rdiff0_d, v_rdiff2_d); in obmc_variance_w16n()
|
/external/mesa3d/src/gallium/drivers/swr/rasterizer/common/ |
D | simdlib_256_avx512_core.inl | 92 SIMD_IWRAPPER_2_16(packs_epi32); // int32 --> int16 See documentation for _mm256_packs_epi32 a…
|
D | simdlib_128_avx512_core.inl | 141 SIMD_IWRAPPER_2_16(packs_epi32); // int32 --> int16 See documentation for _mm256_packs_epi32 a…
|
D | simdlib_256_avx2.inl | 174 SIMD_IWRAPPER_2(packs_epi32); // See documentation for _mm256_packs_epi32 and _mm512_packs_epi32
|
/external/libaom/libaom/aom_dsp/simd/ |
D | v256_intrinsics_x86.h | 221 _mm256_packs_epi32(_mm256_srai_epi32(b, 16), _mm256_srai_epi32(a, 16)), in v256_unziphi_16() 294 return _mm256_permute4x64_epi64(_mm256_packs_epi32(b, a), in v256_pack_s32_s16()
|