| /external/libaom/libaom/aom_dsp/x86/ |
| D | common_avx2.h | 93 tr0_0 = _mm256_unpacklo_epi64(tr1_0, tr1_4); in mm256_transpose_16x16()
95 tr0_2 = _mm256_unpacklo_epi64(tr1_1, tr1_5); in mm256_transpose_16x16()
97 tr0_4 = _mm256_unpacklo_epi64(tr1_2, tr1_6); in mm256_transpose_16x16()
99 tr0_6 = _mm256_unpacklo_epi64(tr1_3, tr1_7); in mm256_transpose_16x16()
102 tr0_8 = _mm256_unpacklo_epi64(tr1_8, tr1_c); in mm256_transpose_16x16()
104 tr0_a = _mm256_unpacklo_epi64(tr1_9, tr1_d); in mm256_transpose_16x16()
106 tr0_c = _mm256_unpacklo_epi64(tr1_a, tr1_e); in mm256_transpose_16x16()
108 tr0_e = _mm256_unpacklo_epi64(tr1_b, tr1_f); in mm256_transpose_16x16()
|
| D | txfm_common_avx2.h | 149 out[2 * i] = _mm256_unpacklo_epi64(u[2 * i], u[2 * i + 4]); in transpose2_8x8_avx2()
152 out[2 * i + 4] = _mm256_unpacklo_epi64(u[2 * i + 1], u[2 * i + 5]); in transpose2_8x8_avx2()
218 out[0] = _mm256_unpacklo_epi64(b0, b1); in transpose_16bit_16x8_avx2()
220 out[2] = _mm256_unpacklo_epi64(b4, b5); in transpose_16bit_16x8_avx2()
222 out[4] = _mm256_unpacklo_epi64(b2, b3); in transpose_16bit_16x8_avx2()
224 out[6] = _mm256_unpacklo_epi64(b6, b7); in transpose_16bit_16x8_avx2()
|
| D | avg_intrin_avx2.c | 75 in[0] = _mm256_unpacklo_epi64(b0, b1); in hadamard_col8x2_avx2()
77 in[2] = _mm256_unpacklo_epi64(b2, b3); in hadamard_col8x2_avx2()
79 in[4] = _mm256_unpacklo_epi64(b4, b5); in hadamard_col8x2_avx2()
81 in[6] = _mm256_unpacklo_epi64(b6, b7); in hadamard_col8x2_avx2()
307 b0 = _mm256_unpacklo_epi64(a0, a1); in highbd_hadamard_col8_avx2()
308 b1 = _mm256_unpacklo_epi64(a4, a5); in highbd_hadamard_col8_avx2()
311 b4 = _mm256_unpacklo_epi64(a2, a3); in highbd_hadamard_col8_avx2()
312 b5 = _mm256_unpacklo_epi64(a6, a7); in highbd_hadamard_col8_avx2()
|
| D | sad4d_avx2.c | 74 sum_mlow = _mm256_unpacklo_epi64(sum_ref0, sum_ref2); in aom_sadMxNx4d_avx2()
|
| D | intrapred_avx2.c | 191 d[0] = _mm256_unpacklo_epi64(ww0, ww1); // 00 10 20 30 40 50 60 70 in highbd_transpose4x16_avx2()
197 d[2] = _mm256_unpacklo_epi64(ww0, ww1); // 02 12 22 32 42 52 62 72 in highbd_transpose4x16_avx2()
212 d[0] = _mm256_unpacklo_epi64(ww0, ww1); // 00 10 20 30 40 50 60 70 in highbd_transpose8x16_16x8_avx2()
218 d[2] = _mm256_unpacklo_epi64(ww0, ww1); // 02 12 22 32 42 52 62 72 in highbd_transpose8x16_16x8_avx2()
229 d[4] = _mm256_unpacklo_epi64(ww0, ww1); // 04 14 24 34 44 54 64 74 in highbd_transpose8x16_16x8_avx2()
235 d[6] = _mm256_unpacklo_epi64(ww0, ww1); // 06 16 26 36 46 56 66 76 in highbd_transpose8x16_16x8_avx2()
250 dd[0] = _mm256_unpacklo_epi64(ww0, ww1); in highbd_transpose16x16_avx2()
256 dd[2] = _mm256_unpacklo_epi64(ww0, ww1); in highbd_transpose16x16_avx2()
267 dd[4] = _mm256_unpacklo_epi64(ww0, ww1); in highbd_transpose16x16_avx2()
273 dd[6] = _mm256_unpacklo_epi64(ww0, ww1); in highbd_transpose16x16_avx2()
[all …]
|
| D | sad_highbd_avx2.c | 555 u1 = _mm256_unpacklo_epi64(u0, u2); in get_4d_sad_from_mm256_epi32()
|
| D | blend_a64_mask_avx2.c | 625 _mm256_permute4x64_epi64(_mm256_unpacklo_epi64(v_al_b, v_ah_b), 0xd8); in blend_a64_mask_sx_w32n_avx2()
|
| /external/ruy/ruy/ |
| D | pack_avx2_fma.cc | 179 t0 = _mm256_unpacklo_epi64(r0, r1);
180 t4 = _mm256_unpacklo_epi64(r4, r5);
183 t1 = _mm256_unpacklo_epi64(r2, r3);
184 t5 = _mm256_unpacklo_epi64(r6, r7);
311 t0 = _mm256_unpacklo_epi64(r0, r1);
312 t4 = _mm256_unpacklo_epi64(r4, r5);
315 t1 = _mm256_unpacklo_epi64(r2, r3);
316 t5 = _mm256_unpacklo_epi64(r6, r7);
397 t0 = _mm256_unpacklo_epi64(r0, r1);
398 t4 = _mm256_unpacklo_epi64(r4, r5);
[all …]
|
| /external/libvpx/libvpx/vpx_dsp/x86/ |
| D | avg_intrin_avx2.c | 66 b0 = _mm256_unpacklo_epi64(a0, a1); in highbd_hadamard_col8_avx2()
67 b1 = _mm256_unpacklo_epi64(a4, a5); in highbd_hadamard_col8_avx2()
70 b4 = _mm256_unpacklo_epi64(a2, a3); in highbd_hadamard_col8_avx2()
71 b5 = _mm256_unpacklo_epi64(a6, a7); in highbd_hadamard_col8_avx2()
277 in[0] = _mm256_unpacklo_epi64(b0, b1); in hadamard_col8x2_avx2()
279 in[2] = _mm256_unpacklo_epi64(b2, b3); in hadamard_col8x2_avx2()
281 in[4] = _mm256_unpacklo_epi64(b4, b5); in hadamard_col8x2_avx2()
283 in[6] = _mm256_unpacklo_epi64(b6, b7); in hadamard_col8x2_avx2()
|
| D | fwd_dct32x32_impl_avx2.h | 39 return _mm256_unpacklo_epi64(buf0, buf1); in k_packs_epi64_avx2()
2832 __m256i tr2_0 = _mm256_unpacklo_epi64(tr1_0, tr1_4); in FDCT32x32_2D_AVX2()
2834 __m256i tr2_2 = _mm256_unpacklo_epi64(tr1_2, tr1_6); in FDCT32x32_2D_AVX2()
2836 __m256i tr2_4 = _mm256_unpacklo_epi64(tr1_1, tr1_5); in FDCT32x32_2D_AVX2()
2838 __m256i tr2_6 = _mm256_unpacklo_epi64(tr1_3, tr1_7); in FDCT32x32_2D_AVX2()
|
| /external/libgav1/libgav1/src/dsp/x86/ |
| D | cdef_avx2.cc | 303 const __m256i v_src_4_0 = _mm256_unpacklo_epi64(v_src[0], v_src[4]); in AddPartial()
304 const __m256i v_src_5_1 = _mm256_unpacklo_epi64(v_src[1], v_src[5]); in AddPartial()
305 const __m256i v_src_6_2 = _mm256_unpacklo_epi64(v_src[2], v_src[6]); in AddPartial()
306 const __m256i v_src_7_3 = _mm256_unpacklo_epi64(v_src[3], v_src[7]); in AddPartial()
316 _mm256_unpacklo_epi64(_mm256_unpacklo_epi32(v_hsum_1_0, v_hsum_3_2), in AddPartial()
|
| D | convolve_avx2.cc | 237 StoreUnaligned32(&dest8[x], _mm256_unpacklo_epi64(result, result2)); in FilterHorizontal()
277 const __m256i packed_result = _mm256_unpacklo_epi64(result, result2); in FilterHorizontal()
|
| /external/libaom/libaom/av1/common/x86/ |
| D | warp_plane_avx2.c | 188 coeff[0] = _mm256_unpacklo_epi64(res_0, res_2); in prepare_horizontal_filter_coeff_avx2()
190 coeff[2] = _mm256_unpacklo_epi64(res_1, res_3); in prepare_horizontal_filter_coeff_avx2()
228 coeff[0] = _mm256_unpacklo_epi64(res_0, res_2); in prepare_horizontal_filter_coeff_beta0_avx2()
230 coeff[2] = _mm256_unpacklo_epi64(res_1, res_3); in prepare_horizontal_filter_coeff_beta0_avx2()
489 coeffs[0] = _mm256_unpacklo_epi64(res_0, res_1); in prepare_vertical_filter_coeffs_avx2()
491 coeffs[2] = _mm256_unpacklo_epi64(res_2, res_3); in prepare_vertical_filter_coeffs_avx2()
534 coeffs[4] = _mm256_unpacklo_epi64(res_0, res_1); in prepare_vertical_filter_coeffs_avx2()
536 coeffs[6] = _mm256_unpacklo_epi64(res_2, res_3); in prepare_vertical_filter_coeffs_avx2()
565 coeffs[0] = _mm256_unpacklo_epi64(res_0, res_1); in prepare_vertical_filter_coeffs_delta0_avx2()
567 coeffs[2] = _mm256_unpacklo_epi64(res_2, res_3); in prepare_vertical_filter_coeffs_delta0_avx2()
[all …]
|
| D | convolve_avx2.c | 112 __m256i res_a = _mm256_unpacklo_epi64(res_8b_lo, res_8b_hi); in av1_convolve_y_sr_avx2()
225 __m256i res_a = _mm256_unpacklo_epi64(res_8b_lo, res_8b_hi); in av1_convolve_y_sr_avx2()
|
| D | highbd_inv_txfm_avx2.c | 176 x0 = _mm256_unpacklo_epi64(u0, u2); in transpose_8x8_avx2()
177 x1 = _mm256_unpacklo_epi64(u4, u6); in transpose_8x8_avx2()
186 x0 = _mm256_unpacklo_epi64(u1, u3); in transpose_8x8_avx2()
187 x1 = _mm256_unpacklo_epi64(u5, u7); in transpose_8x8_avx2()
213 x0 = _mm256_unpacklo_epi64(u0, u2); in transpose_8x8_flip_avx2()
214 x1 = _mm256_unpacklo_epi64(u4, u6); in transpose_8x8_flip_avx2()
223 x0 = _mm256_unpacklo_epi64(u1, u3); in transpose_8x8_flip_avx2()
224 x1 = _mm256_unpacklo_epi64(u5, u7); in transpose_8x8_flip_avx2()
|
| /external/libaom/libaom/av1/encoder/x86/ |
| D | corner_match_avx2.c | 67 __m256i unp_low = _mm256_unpacklo_epi64(sumsq2_vec, cross_vec); in av1_compute_cross_correlation_avx2()
|
| D | pickrst_avx2.c | 692 __m256i c_low = _mm256_unpacklo_epi64(c0, c1); in calc_proj_params_r0_r1_avx2()
698 __m256i h0x_low = _mm256_unpacklo_epi64(h00, h01); in calc_proj_params_r0_r1_avx2()
706 __m256i h1x_low = _mm256_unpacklo_epi64(zero, h11); in calc_proj_params_r0_r1_avx2()
|
| D | highbd_fwd_txfm_avx2.c | 162 x0 = _mm256_unpacklo_epi64(u0, u2); in fwd_txfm_transpose_8x8_avx2()
163 x1 = _mm256_unpacklo_epi64(u4, u6); in fwd_txfm_transpose_8x8_avx2()
172 x0 = _mm256_unpacklo_epi64(u1, u3); in fwd_txfm_transpose_8x8_avx2()
173 x1 = _mm256_unpacklo_epi64(u5, u7); in fwd_txfm_transpose_8x8_avx2()
|
| /external/eigen/Eigen/src/Core/arch/CUDA/ |
| D | PacketMathHalf.h | 586 __m256i abcdefgh_01 = _mm256_unpacklo_epi64(abcd_03, efgh_03);
588 __m256i ijklmnop_01 = _mm256_unpacklo_epi64(ijkl_03, mnop_03);
590 __m256i abcdefgh_45 = _mm256_unpacklo_epi64(abcd_47, efgh_47);
592 __m256i ijklmnop_45 = _mm256_unpacklo_epi64(ijkl_47, mnop_47);
594 __m256i abcdefgh_89 = _mm256_unpacklo_epi64(abcd_8b, efgh_8b);
596 __m256i ijklmnop_89 = _mm256_unpacklo_epi64(ijkl_8b, mnop_8b);
598 __m256i abcdefgh_cd = _mm256_unpacklo_epi64(abcd_cf, efgh_cf);
600 __m256i ijklmnop_cd = _mm256_unpacklo_epi64(ijkl_cf, mnop_cf);
|
| /external/libaom/libaom/aom_dsp/simd/ |
| D | v256_intrinsics_x86.h | 178 return _mm256_unpacklo_epi64( in v256_ziplo_64()
481 _mm256_srl_epi64(_mm256_sll_epi64(_mm256_unpacklo_epi64(rl, rh), c), c)); in v256_ssd_u8()
|
| /external/clang/test/CodeGen/ |
| D | avx2-builtins.c | 1219 return _mm256_unpacklo_epi64(a, b); in test_mm256_unpacklo_epi64()
|
| /external/llvm-project/clang/test/CodeGen/X86/ |
| D | avx2-builtins.c | 1267 return _mm256_unpacklo_epi64(a, b); in test_mm256_unpacklo_epi64()
|
| /external/llvm-project/clang/lib/Headers/ |
| D | avx2intrin.h | 695 _mm256_unpacklo_epi64(__m256i __a, __m256i __b) in _mm256_unpacklo_epi64() function
|
| /external/clang/lib/Headers/ |
| D | avx2intrin.h | 821 _mm256_unpacklo_epi64(__m256i __a, __m256i __b) in _mm256_unpacklo_epi64() function
|
| D | avx512vlintrin.h | 6975 (__v4di)_mm256_unpacklo_epi64(__A, __B), in _mm256_mask_unpacklo_epi64()
6983 (__v4di)_mm256_unpacklo_epi64(__A, __B), in _mm256_maskz_unpacklo_epi64()
|