/external/libaom/libaom/aom_dsp/x86/ |
D | common_avx2.h | 57 __m256i tr1_0 = _mm256_unpacklo_epi32(tr0_0, tr0_2); in mm256_transpose_16x16() 59 __m256i tr1_2 = _mm256_unpacklo_epi32(tr0_1, tr0_3); in mm256_transpose_16x16() 61 __m256i tr1_4 = _mm256_unpacklo_epi32(tr0_4, tr0_6); in mm256_transpose_16x16() 63 __m256i tr1_6 = _mm256_unpacklo_epi32(tr0_5, tr0_7); in mm256_transpose_16x16() 66 __m256i tr1_8 = _mm256_unpacklo_epi32(tr0_8, tr0_a); in mm256_transpose_16x16() 68 __m256i tr1_a = _mm256_unpacklo_epi32(tr0_9, tr0_b); in mm256_transpose_16x16() 70 __m256i tr1_c = _mm256_unpacklo_epi32(tr0_c, tr0_e); in mm256_transpose_16x16() 72 __m256i tr1_e = _mm256_unpacklo_epi32(tr0_d, tr0_f); in mm256_transpose_16x16()
|
D | txfm_common_avx2.h | 136 u[i] = _mm256_unpacklo_epi32(t[i], t[i + 2]); in transpose2_8x8_avx2() 139 u[i + 4] = _mm256_unpacklo_epi32(t[i + 4], t[i + 6]); in transpose2_8x8_avx2() 209 const __m256i b0 = _mm256_unpacklo_epi32(a0, a1); in transpose_16bit_16x8_avx2() 210 const __m256i b1 = _mm256_unpacklo_epi32(a2, a3); in transpose_16bit_16x8_avx2() 211 const __m256i b2 = _mm256_unpacklo_epi32(a4, a5); in transpose_16bit_16x8_avx2() 212 const __m256i b3 = _mm256_unpacklo_epi32(a6, a7); in transpose_16bit_16x8_avx2()
|
D | avg_intrin_avx2.c | 66 b0 = _mm256_unpacklo_epi32(a0, a1); in hadamard_col8x2_avx2() 67 b1 = _mm256_unpacklo_epi32(a4, a5); in hadamard_col8x2_avx2() 70 b4 = _mm256_unpacklo_epi32(a2, a3); in hadamard_col8x2_avx2() 71 b5 = _mm256_unpacklo_epi32(a6, a7); in hadamard_col8x2_avx2() 298 a0 = _mm256_unpacklo_epi32(b0, b1); in highbd_hadamard_col8_avx2() 299 a1 = _mm256_unpacklo_epi32(b2, b3); in highbd_hadamard_col8_avx2() 302 a4 = _mm256_unpacklo_epi32(b4, b5); in highbd_hadamard_col8_avx2() 303 a5 = _mm256_unpacklo_epi32(b6, b7); in highbd_hadamard_col8_avx2()
|
D | blk_sse_sum_avx2.c | 76 temp_buffer1 = _mm256_unpacklo_epi32(sse_buffer, _mm256_setzero_si256()); in sse_sum_wd4_avx2() 111 temp_buffer1 = _mm256_unpacklo_epi32(sse_buffer, _mm256_setzero_si256()); in sse_sum_wd8_avx2() 145 temp_buffer1 = _mm256_unpacklo_epi32(sse_buffer, _mm256_setzero_si256()); in sse_sum_wd16_avx2()
|
D | highbd_variance_avx2.c | 45 __m256i v_d_l = _mm256_unpacklo_epi32(v_sum0, v_sse_d); in aom_highbd_calc8x8var_avx2() 73 __m256i v_d_l = _mm256_unpacklo_epi32(v_sum0, v_sse_d); in aom_highbd_calc16x16var_avx2()
|
D | intrapred_avx2.c | 188 ww0 = _mm256_unpacklo_epi32(w0, w1); // 00 10 20 30 01 11 21 31 in highbd_transpose4x16_avx2() 189 ww1 = _mm256_unpacklo_epi32(w2, w3); // 40 50 60 70 41 51 61 71 in highbd_transpose4x16_avx2() 209 ww0 = _mm256_unpacklo_epi32(w0, w1); // 00 10 20 30 01 11 21 31 in highbd_transpose8x16_16x8_avx2() 210 ww1 = _mm256_unpacklo_epi32(w2, w3); // 40 50 60 70 41 51 61 71 in highbd_transpose8x16_16x8_avx2() 226 ww0 = _mm256_unpacklo_epi32(w0, w1); // 04 14 24 34 05 15 25 35 in highbd_transpose8x16_16x8_avx2() 227 ww1 = _mm256_unpacklo_epi32(w2, w3); // 44 54 64 74 45 55 65 75 in highbd_transpose8x16_16x8_avx2() 247 ww0 = _mm256_unpacklo_epi32(w0, w1); // in highbd_transpose16x16_avx2() 248 ww1 = _mm256_unpacklo_epi32(w2, w3); // in highbd_transpose16x16_avx2() 264 ww0 = _mm256_unpacklo_epi32(w0, w1); // in highbd_transpose16x16_avx2() 265 ww1 = _mm256_unpacklo_epi32(w2, w3); // in highbd_transpose16x16_avx2() [all …]
|
D | sse_avx2.c | 38 const __m256i sum0_4x64 = _mm256_unpacklo_epi32(*sum_all, zero); in summary_all_avx2()
|
/external/libvpx/libvpx/vp9/encoder/x86/ |
D | vp9_error_avx2.c | 47 sse_256 = _mm256_unpacklo_epi32(dqcoeff_256, zero); in vp9_block_error_avx2() 48 ssz_256 = _mm256_unpacklo_epi32(coeff_256, zero); in vp9_block_error_avx2() 76 exp_dqcoeff_lo = _mm256_unpacklo_epi32(dqcoeff_0, zero); in vp9_block_error_avx2() 79 exp_coeff_lo = _mm256_unpacklo_epi32(coeff_0, zero); in vp9_block_error_avx2() 131 sse_256 = _mm256_unpacklo_epi32(error, zero); in vp9_block_error_fp_avx2() 140 const __m256i exp_error_lo = _mm256_unpacklo_epi32(error, zero); in vp9_block_error_fp_avx2()
|
/external/libaom/libaom/av1/encoder/x86/ |
D | error_intrin_avx2.c | 53 sse_256 = _mm256_unpacklo_epi32(error, zero); in av1_block_error_lp_avx2() 62 const __m256i exp_error_lo = _mm256_unpacklo_epi32(error, zero); in av1_block_error_lp_avx2() 110 exp_dqcoeff_lo = _mm256_unpacklo_epi32(dqcoeff_reg, zero_reg); in av1_block_error_avx2() 113 exp_coeff_lo = _mm256_unpacklo_epi32(coeff_reg, zero_reg); in av1_block_error_avx2()
|
D | wedge_utils_avx2.c | 135 v_acc0_d = _mm256_add_epi64(_mm256_unpacklo_epi32(v_acc0_d, v_sign_d), in av1_wedge_sign_from_residuals_avx2()
|
D | av1_fwd_txfm2d_avx2.c | 1435 __m256i temp0 = _mm256_unpacklo_epi32(inputA[0], inputA[2]); in transpose_32_8x8_avx2() 1437 __m256i temp2 = _mm256_unpacklo_epi32(inputA[1], inputA[3]); in transpose_32_8x8_avx2() 1439 __m256i temp4 = _mm256_unpacklo_epi32(inputA[4], inputA[6]); in transpose_32_8x8_avx2() 1441 __m256i temp6 = _mm256_unpacklo_epi32(inputA[5], inputA[7]); in transpose_32_8x8_avx2() 1444 __m256i t0 = _mm256_unpacklo_epi32(temp0, temp2); in transpose_32_8x8_avx2() 1446 __m256i t2 = _mm256_unpacklo_epi32(temp1, temp3); in transpose_32_8x8_avx2() 1448 __m256i t4 = _mm256_unpacklo_epi32(temp4, temp6); in transpose_32_8x8_avx2() 1450 __m256i t6 = _mm256_unpacklo_epi32(temp5, temp7); in transpose_32_8x8_avx2()
|
/external/ruy/ruy/ |
D | pack_avx2_fma.cc | 170 r0 = _mm256_unpacklo_epi32(t0, t1); 171 r4 = _mm256_unpacklo_epi32(t4, t5); 174 r1 = _mm256_unpacklo_epi32(t2, t3); 175 r5 = _mm256_unpacklo_epi32(t6, t7); 302 r0 = _mm256_unpacklo_epi32(t0, t1); 303 r4 = _mm256_unpacklo_epi32(t4, t5); 306 r1 = _mm256_unpacklo_epi32(t2, t3); 307 r5 = _mm256_unpacklo_epi32(t6, t7); 388 r0 = _mm256_unpacklo_epi32(t0, t1); 389 r4 = _mm256_unpacklo_epi32(t4, t5); [all …]
|
/external/libvpx/libvpx/vpx_dsp/x86/ |
D | fwd_dct32x32_impl_avx2.h | 1553 u[0] = _mm256_unpacklo_epi32(lstep3[12], lstep3[10]); in FDCT32x32_2D_AVX2() 1555 u[2] = _mm256_unpacklo_epi32(lstep3[13], lstep3[11]); in FDCT32x32_2D_AVX2() 1592 u[0] = _mm256_unpacklo_epi32(lstep3[36], lstep3[58]); in FDCT32x32_2D_AVX2() 1594 u[2] = _mm256_unpacklo_epi32(lstep3[37], lstep3[59]); in FDCT32x32_2D_AVX2() 1596 u[4] = _mm256_unpacklo_epi32(lstep3[38], lstep3[56]); in FDCT32x32_2D_AVX2() 1598 u[6] = _mm256_unpacklo_epi32(lstep3[39], lstep3[57]); in FDCT32x32_2D_AVX2() 1600 u[8] = _mm256_unpacklo_epi32(lstep3[40], lstep3[54]); in FDCT32x32_2D_AVX2() 1602 u[10] = _mm256_unpacklo_epi32(lstep3[41], lstep3[55]); in FDCT32x32_2D_AVX2() 1604 u[12] = _mm256_unpacklo_epi32(lstep3[42], lstep3[52]); in FDCT32x32_2D_AVX2() 1606 u[14] = _mm256_unpacklo_epi32(lstep3[43], lstep3[53]); in FDCT32x32_2D_AVX2() [all …]
|
D | avg_intrin_avx2.c | 57 a0 = _mm256_unpacklo_epi32(b0, b1); in highbd_hadamard_col8_avx2() 58 a1 = _mm256_unpacklo_epi32(b2, b3); in highbd_hadamard_col8_avx2() 61 a4 = _mm256_unpacklo_epi32(b4, b5); in highbd_hadamard_col8_avx2() 62 a5 = _mm256_unpacklo_epi32(b6, b7); in highbd_hadamard_col8_avx2() 268 b0 = _mm256_unpacklo_epi32(a0, a1); in hadamard_col8x2_avx2() 269 b1 = _mm256_unpacklo_epi32(a4, a5); in hadamard_col8x2_avx2() 272 b4 = _mm256_unpacklo_epi32(a2, a3); in hadamard_col8x2_avx2() 273 b5 = _mm256_unpacklo_epi32(a6, a7); in hadamard_col8x2_avx2()
|
/external/libaom/libaom/av1/common/x86/ |
D | warp_plane_avx2.c | 183 const __m256i res_0 = _mm256_unpacklo_epi32(tmp_12, tmp_14); in prepare_horizontal_filter_coeff_avx2() 185 const __m256i res_2 = _mm256_unpacklo_epi32(tmp_13, tmp_15); in prepare_horizontal_filter_coeff_avx2() 223 const __m256i res_0 = _mm256_unpacklo_epi32(tmp_12, tmp_14); in prepare_horizontal_filter_coeff_beta0_avx2() 225 const __m256i res_2 = _mm256_unpacklo_epi32(tmp_13, tmp_15); in prepare_horizontal_filter_coeff_beta0_avx2() 484 __m256i res_0 = _mm256_unpacklo_epi32(filt_0, filt_1); in prepare_vertical_filter_coeffs_avx2() 485 __m256i res_1 = _mm256_unpacklo_epi32(filt_2, filt_3); in prepare_vertical_filter_coeffs_avx2() 529 res_0 = _mm256_unpacklo_epi32(filt_0, filt_1); in prepare_vertical_filter_coeffs_avx2() 530 res_1 = _mm256_unpacklo_epi32(filt_2, filt_3); in prepare_vertical_filter_coeffs_avx2() 560 __m256i res_0 = _mm256_unpacklo_epi32(filt_0, filt_1); in prepare_vertical_filter_coeffs_delta0_avx2() 561 __m256i res_1 = _mm256_unpacklo_epi32(filt_2, filt_3); in prepare_vertical_filter_coeffs_delta0_avx2() [all …]
|
D | highbd_wiener_convolve_avx2.c | 225 const __m256i res_lo = _mm256_unpacklo_epi32(res_even, res_odd); in av1_highbd_wiener_convolve_add_src_avx2()
|
D | highbd_inv_txfm_avx2.c | 164 u0 = _mm256_unpacklo_epi32(in[0], in[1]); in transpose_8x8_avx2() 167 u2 = _mm256_unpacklo_epi32(in[2], in[3]); in transpose_8x8_avx2() 170 u4 = _mm256_unpacklo_epi32(in[4], in[5]); in transpose_8x8_avx2() 173 u6 = _mm256_unpacklo_epi32(in[6], in[7]); in transpose_8x8_avx2() 201 u0 = _mm256_unpacklo_epi32(in[7], in[6]); in transpose_8x8_flip_avx2() 204 u2 = _mm256_unpacklo_epi32(in[5], in[4]); in transpose_8x8_flip_avx2() 207 u4 = _mm256_unpacklo_epi32(in[3], in[2]); in transpose_8x8_flip_avx2() 210 u6 = _mm256_unpacklo_epi32(in[1], in[0]); in transpose_8x8_flip_avx2()
|
D | highbd_jnt_convolve_avx2.c | 543 __m256i res1 = _mm256_unpacklo_epi32(res_even, res_odd); in av1_highbd_dist_wtd_convolve_x_avx2()
|
/external/gemmlowp/internal/ |
D | pack_avx.h | 93 __m256i ymm7 = _mm256_unpacklo_epi32(ymm5, ymm6); in Pack() 96 __m256i ymm13 = _mm256_unpacklo_epi32(ymm9, ymm10); in Pack()
|
/external/eigen/Eigen/src/Core/arch/CUDA/ |
D | PacketMathHalf.h | 568 __m256i abcd_03 = _mm256_unpacklo_epi32(ab_07, cd_07); 570 __m256i efgh_03 = _mm256_unpacklo_epi32(ef_07, gh_07); 572 __m256i ijkl_03 = _mm256_unpacklo_epi32(ij_07, kl_07); 574 __m256i mnop_03 = _mm256_unpacklo_epi32(mn_07, op_07); 577 __m256i abcd_8b = _mm256_unpacklo_epi32(ab_8f, cd_8f); 579 __m256i efgh_8b = _mm256_unpacklo_epi32(ef_8f, gh_8f); 581 __m256i ijkl_8b = _mm256_unpacklo_epi32(ij_8f, kl_8f); 583 __m256i mnop_8b = _mm256_unpacklo_epi32(mn_8f, op_8f);
|
/external/libgav1/libgav1/src/dsp/x86/ |
D | cdef_avx2.cc | 316 _mm256_unpacklo_epi64(_mm256_unpacklo_epi32(v_hsum_1_0, v_hsum_3_2), in AddPartial() 317 _mm256_unpacklo_epi32(v_hsum_5_4, v_hsum_7_6)); in AddPartial()
|
/external/libaom/libaom/aom_dsp/simd/ |
D | v256_intrinsics_x86.h | 166 return _mm256_unpacklo_epi32( in v256_ziplo_32() 740 _mm256_unpacklo_epi32(d, v256_zero()))); in v256_ssd_s16()
|
/external/skia/src/opts/ |
D | SkSwizzler_opts.h | 650 __m256i rg = _mm256_unpacklo_epi32(*lo, *hi), // rrrrRRRR ggggGGGG rrrrRRRR ggggGGGG in premul_should_swapRB() 808 __m256i cm = _mm256_unpacklo_epi32(*lo, *hi), // ccccCCCC mmmmMMMM ccccCCCC mmmmMMMM in inverted_cmyk_to()
|
/external/clang/test/CodeGen/ |
D | avx2-builtins.c | 1213 return _mm256_unpacklo_epi32(a, b); in test_mm256_unpacklo_epi32()
|
/external/llvm-project/clang/test/CodeGen/X86/ |
D | avx2-builtins.c | 1261 return _mm256_unpacklo_epi32(a, b); in test_mm256_unpacklo_epi32()
|