/external/libaom/libaom/aom_dsp/x86/ |
D | common_avx2.h | 94 tr0_1 = _mm256_unpackhi_epi64(tr1_0, tr1_4); in mm256_transpose_16x16() 96 tr0_3 = _mm256_unpackhi_epi64(tr1_1, tr1_5); in mm256_transpose_16x16() 98 tr0_5 = _mm256_unpackhi_epi64(tr1_2, tr1_6); in mm256_transpose_16x16() 100 tr0_7 = _mm256_unpackhi_epi64(tr1_3, tr1_7); in mm256_transpose_16x16() 103 tr0_9 = _mm256_unpackhi_epi64(tr1_8, tr1_c); in mm256_transpose_16x16() 105 tr0_b = _mm256_unpackhi_epi64(tr1_9, tr1_d); in mm256_transpose_16x16() 107 tr0_d = _mm256_unpackhi_epi64(tr1_a, tr1_e); in mm256_transpose_16x16() 109 tr0_f = _mm256_unpackhi_epi64(tr1_b, tr1_f); in mm256_transpose_16x16()
|
D | adaptive_quantize_avx2.c | 137 zbin = _mm256_unpackhi_epi64(zbin, zbin); in aom_quantize_b_adaptive_avx2() 146 round = _mm256_unpackhi_epi64(round, round); in aom_quantize_b_adaptive_avx2() 147 quant = _mm256_unpackhi_epi64(quant, quant); in aom_quantize_b_adaptive_avx2() 148 shift = _mm256_unpackhi_epi64(shift, shift); in aom_quantize_b_adaptive_avx2() 149 dequant = _mm256_unpackhi_epi64(dequant, dequant); in aom_quantize_b_adaptive_avx2() 152 round = _mm256_unpackhi_epi64(round, round); in aom_quantize_b_adaptive_avx2() 153 quant = _mm256_unpackhi_epi64(quant, quant); in aom_quantize_b_adaptive_avx2() 154 shift = _mm256_unpackhi_epi64(shift, shift); in aom_quantize_b_adaptive_avx2() 161 dequant = _mm256_unpackhi_epi64(dequant, dequant); in aom_quantize_b_adaptive_avx2()
|
D | highbd_adaptive_quantize_avx2.c | 152 zbin = _mm256_unpackhi_epi64(zbin, zbin); in aom_highbd_quantize_b_adaptive_avx2() 162 round = _mm256_unpackhi_epi64(round, round); in aom_highbd_quantize_b_adaptive_avx2() 163 quant = _mm256_unpackhi_epi64(quant, quant); in aom_highbd_quantize_b_adaptive_avx2() 164 shift = _mm256_unpackhi_epi64(shift, shift); in aom_highbd_quantize_b_adaptive_avx2() 165 dequant = _mm256_unpackhi_epi64(dequant, dequant); in aom_highbd_quantize_b_adaptive_avx2() 168 round = _mm256_unpackhi_epi64(round, round); in aom_highbd_quantize_b_adaptive_avx2() 169 quant = _mm256_unpackhi_epi64(quant, quant); in aom_highbd_quantize_b_adaptive_avx2() 170 shift = _mm256_unpackhi_epi64(shift, shift); in aom_highbd_quantize_b_adaptive_avx2() 180 dequant = _mm256_unpackhi_epi64(dequant, dequant); in aom_highbd_quantize_b_adaptive_avx2()
|
D | txfm_common_avx2.h | 150 out[2 * i + 1] = _mm256_unpackhi_epi64(u[2 * i], u[2 * i + 4]); in transpose2_8x8_avx2() 153 out[2 * i + 5] = _mm256_unpackhi_epi64(u[2 * i + 1], u[2 * i + 5]); in transpose2_8x8_avx2() 219 out[1] = _mm256_unpackhi_epi64(b0, b1); in transpose_16bit_16x8_avx2() 221 out[3] = _mm256_unpackhi_epi64(b4, b5); in transpose_16bit_16x8_avx2() 223 out[5] = _mm256_unpackhi_epi64(b2, b3); in transpose_16bit_16x8_avx2() 225 out[7] = _mm256_unpackhi_epi64(b6, b7); in transpose_16bit_16x8_avx2()
|
D | avg_intrin_avx2.c | 76 in[1] = _mm256_unpackhi_epi64(b0, b1); in hadamard_col8x2_avx2() 78 in[3] = _mm256_unpackhi_epi64(b2, b3); in hadamard_col8x2_avx2() 80 in[5] = _mm256_unpackhi_epi64(b4, b5); in hadamard_col8x2_avx2() 82 in[7] = _mm256_unpackhi_epi64(b6, b7); in hadamard_col8x2_avx2() 309 b2 = _mm256_unpackhi_epi64(a0, a1); in highbd_hadamard_col8_avx2() 310 b3 = _mm256_unpackhi_epi64(a4, a5); in highbd_hadamard_col8_avx2() 313 b6 = _mm256_unpackhi_epi64(a2, a3); in highbd_hadamard_col8_avx2() 314 b7 = _mm256_unpackhi_epi64(a6, a7); in highbd_hadamard_col8_avx2()
|
D | sad4d_avx2.c | 75 sum_mhigh = _mm256_unpackhi_epi64(sum_ref0, sum_ref2); in aom_sadMxNx4d_avx2()
|
D | intrapred_avx2.c | 27 u0 = _mm256_unpackhi_epi64(y0, y0); in dc_sum_64() 37 u = _mm256_unpackhi_epi64(y, y); in dc_sum_32() 192 d[1] = _mm256_unpackhi_epi64(ww0, ww1); // 01 11 21 31 41 51 61 71 in highbd_transpose4x16_avx2() 198 d[3] = _mm256_unpackhi_epi64(ww0, ww1); // 03 13 23 33 43 53 63 73 in highbd_transpose4x16_avx2() 213 d[1] = _mm256_unpackhi_epi64(ww0, ww1); // 01 11 21 31 41 51 61 71 in highbd_transpose8x16_16x8_avx2() 219 d[3] = _mm256_unpackhi_epi64(ww0, ww1); // 03 13 23 33 43 53 63 73 in highbd_transpose8x16_16x8_avx2() 230 d[5] = _mm256_unpackhi_epi64(ww0, ww1); // 05 15 25 35 45 55 65 75 in highbd_transpose8x16_16x8_avx2() 236 d[7] = _mm256_unpackhi_epi64(ww0, ww1); // 07 17 27 37 47 57 67 77 in highbd_transpose8x16_16x8_avx2() 251 dd[1] = _mm256_unpackhi_epi64(ww0, ww1); in highbd_transpose16x16_avx2() 257 dd[3] = _mm256_unpackhi_epi64(ww0, ww1); in highbd_transpose16x16_avx2() [all …]
|
D | sad_highbd_avx2.c | 556 u3 = _mm256_unpackhi_epi64(u0, u2); in get_4d_sad_from_mm256_epi32()
|
/external/ruy/ruy/ |
D | pack_avx2_fma.cc | 181 t2 = _mm256_unpackhi_epi64(r0, r1); 182 t6 = _mm256_unpackhi_epi64(r4, r5); 185 t3 = _mm256_unpackhi_epi64(r2, r3); 186 t7 = _mm256_unpackhi_epi64(r6, r7); 313 t2 = _mm256_unpackhi_epi64(r0, r1); 314 t6 = _mm256_unpackhi_epi64(r4, r5); 317 t3 = _mm256_unpackhi_epi64(r2, r3); 318 t7 = _mm256_unpackhi_epi64(r6, r7); 399 t2 = _mm256_unpackhi_epi64(r0, r1); 400 t6 = _mm256_unpackhi_epi64(r4, r5); [all …]
|
/external/libvpx/libvpx/vpx_dsp/x86/ |
D | avg_intrin_avx2.c | 68 b2 = _mm256_unpackhi_epi64(a0, a1); in highbd_hadamard_col8_avx2() 69 b3 = _mm256_unpackhi_epi64(a4, a5); in highbd_hadamard_col8_avx2() 72 b6 = _mm256_unpackhi_epi64(a2, a3); in highbd_hadamard_col8_avx2() 73 b7 = _mm256_unpackhi_epi64(a6, a7); in highbd_hadamard_col8_avx2() 278 in[1] = _mm256_unpackhi_epi64(b0, b1); in hadamard_col8x2_avx2() 280 in[3] = _mm256_unpackhi_epi64(b2, b3); in hadamard_col8x2_avx2() 282 in[5] = _mm256_unpackhi_epi64(b4, b5); in hadamard_col8x2_avx2() 284 in[7] = _mm256_unpackhi_epi64(b6, b7); in hadamard_col8x2_avx2()
|
D | fwd_dct32x32_impl_avx2.h | 2833 __m256i tr2_1 = _mm256_unpackhi_epi64(tr1_0, tr1_4); in FDCT32x32_2D_AVX2() 2835 __m256i tr2_3 = _mm256_unpackhi_epi64(tr1_2, tr1_6); in FDCT32x32_2D_AVX2() 2837 __m256i tr2_5 = _mm256_unpackhi_epi64(tr1_1, tr1_5); in FDCT32x32_2D_AVX2() 2839 __m256i tr2_7 = _mm256_unpackhi_epi64(tr1_3, tr1_7); in FDCT32x32_2D_AVX2()
|
/external/libaom/libaom/av1/common/x86/ |
D | warp_plane_avx2.c | 189 coeff[1] = _mm256_unpackhi_epi64(res_0, res_2); in prepare_horizontal_filter_coeff_avx2() 191 coeff[3] = _mm256_unpackhi_epi64(res_1, res_3); in prepare_horizontal_filter_coeff_avx2() 229 coeff[1] = _mm256_unpackhi_epi64(res_0, res_2); in prepare_horizontal_filter_coeff_beta0_avx2() 231 coeff[3] = _mm256_unpackhi_epi64(res_1, res_3); in prepare_horizontal_filter_coeff_beta0_avx2() 490 coeffs[1] = _mm256_unpackhi_epi64(res_0, res_1); in prepare_vertical_filter_coeffs_avx2() 492 coeffs[3] = _mm256_unpackhi_epi64(res_2, res_3); in prepare_vertical_filter_coeffs_avx2() 535 coeffs[5] = _mm256_unpackhi_epi64(res_0, res_1); in prepare_vertical_filter_coeffs_avx2() 537 coeffs[7] = _mm256_unpackhi_epi64(res_2, res_3); in prepare_vertical_filter_coeffs_avx2() 566 coeffs[1] = _mm256_unpackhi_epi64(res_0, res_1); in prepare_vertical_filter_coeffs_delta0_avx2() 568 coeffs[3] = _mm256_unpackhi_epi64(res_2, res_3); in prepare_vertical_filter_coeffs_delta0_avx2() [all …]
|
D | highbd_inv_txfm_avx2.c | 181 x0 = _mm256_unpackhi_epi64(u0, u2); in transpose_8x8_avx2() 182 x1 = _mm256_unpackhi_epi64(u4, u6); in transpose_8x8_avx2() 191 x0 = _mm256_unpackhi_epi64(u1, u3); in transpose_8x8_avx2() 192 x1 = _mm256_unpackhi_epi64(u5, u7); in transpose_8x8_avx2() 218 x0 = _mm256_unpackhi_epi64(u0, u2); in transpose_8x8_flip_avx2() 219 x1 = _mm256_unpackhi_epi64(u4, u6); in transpose_8x8_flip_avx2() 228 x0 = _mm256_unpackhi_epi64(u1, u3); in transpose_8x8_flip_avx2() 229 x1 = _mm256_unpackhi_epi64(u5, u7); in transpose_8x8_flip_avx2()
|
/external/libaom/libaom/av1/encoder/x86/ |
D | corner_match_avx2.c | 68 __m256i unp_hig = _mm256_unpackhi_epi64(sumsq2_vec, cross_vec); in av1_compute_cross_correlation_avx2()
|
D | pickrst_avx2.c | 693 const __m256i c_high = _mm256_unpackhi_epi64(c0, c1); in calc_proj_params_r0_r1_avx2() 699 const __m256i h0x_high = _mm256_unpackhi_epi64(h00, h01); in calc_proj_params_r0_r1_avx2() 707 const __m256i h1x_high = _mm256_unpackhi_epi64(zero, h11); in calc_proj_params_r0_r1_avx2()
|
D | highbd_fwd_txfm_avx2.c | 167 x0 = _mm256_unpackhi_epi64(u0, u2); in fwd_txfm_transpose_8x8_avx2() 168 x1 = _mm256_unpackhi_epi64(u4, u6); in fwd_txfm_transpose_8x8_avx2() 177 x0 = _mm256_unpackhi_epi64(u1, u3); in fwd_txfm_transpose_8x8_avx2() 178 x1 = _mm256_unpackhi_epi64(u5, u7); in fwd_txfm_transpose_8x8_avx2()
|
/external/eigen/Eigen/src/Core/arch/CUDA/ |
D | PacketMathHalf.h | 587 __m256i abcdefgh_23 = _mm256_unpackhi_epi64(abcd_03, efgh_03); 589 __m256i ijklmnop_23 = _mm256_unpackhi_epi64(ijkl_03, mnop_03); 591 __m256i abcdefgh_67 = _mm256_unpackhi_epi64(abcd_47, efgh_47); 593 __m256i ijklmnop_67 = _mm256_unpackhi_epi64(ijkl_47, mnop_47); 595 __m256i abcdefgh_ab = _mm256_unpackhi_epi64(abcd_8b, efgh_8b); 597 __m256i ijklmnop_ab = _mm256_unpackhi_epi64(ijkl_8b, mnop_8b); 599 __m256i abcdefgh_ef = _mm256_unpackhi_epi64(abcd_cf, efgh_cf); 601 __m256i ijklmnop_ef = _mm256_unpackhi_epi64(ijkl_cf, mnop_cf);
|
/external/libaom/libaom/aom_dsp/simd/ |
D | v256_intrinsics_x86.h | 184 return _mm256_unpackhi_epi64( in v256_ziphi_64() 455 v256 t = _mm256_add_epi32(s, _mm256_unpackhi_epi64(s, s)); in v256_sad_u8_sum() 485 v256 t = _mm256_add_epi32(s, _mm256_unpackhi_epi64(s, s)); in v256_ssd_u8_sum()
|
/external/clang/test/CodeGen/ |
D | avx2-builtins.c | 1195 return _mm256_unpackhi_epi64(a, b); in test_mm256_unpackhi_epi64()
|
/external/llvm-project/clang/test/CodeGen/X86/ |
D | avx2-builtins.c | 1243 return _mm256_unpackhi_epi64(a, b); in test_mm256_unpackhi_epi64()
|
/external/llvm-project/clang/lib/Headers/ |
D | avx2intrin.h | 671 _mm256_unpackhi_epi64(__m256i __a, __m256i __b) in _mm256_unpackhi_epi64() function
|
D | avx512vlintrin.h | 6299 (__v4di)_mm256_unpackhi_epi64(__A, __B), in _mm256_mask_unpackhi_epi64() 6307 (__v4di)_mm256_unpackhi_epi64(__A, __B), in _mm256_maskz_unpackhi_epi64()
|
/external/clang/lib/Headers/ |
D | avx2intrin.h | 797 _mm256_unpackhi_epi64(__m256i __a, __m256i __b) in _mm256_unpackhi_epi64() function
|
D | avx512vlintrin.h | 6911 (__v4di)_mm256_unpackhi_epi64(__A, __B), in _mm256_mask_unpackhi_epi64() 6919 (__v4di)_mm256_unpackhi_epi64(__A, __B), in _mm256_maskz_unpackhi_epi64()
|