/external/libaom/libaom/aom_dsp/x86/ |
D | common_avx2.h | 58 __m256i tr1_1 = _mm256_unpackhi_epi32(tr0_0, tr0_2); in mm256_transpose_16x16() 60 __m256i tr1_3 = _mm256_unpackhi_epi32(tr0_1, tr0_3); in mm256_transpose_16x16() 62 __m256i tr1_5 = _mm256_unpackhi_epi32(tr0_4, tr0_6); in mm256_transpose_16x16() 64 __m256i tr1_7 = _mm256_unpackhi_epi32(tr0_5, tr0_7); in mm256_transpose_16x16() 67 __m256i tr1_9 = _mm256_unpackhi_epi32(tr0_8, tr0_a); in mm256_transpose_16x16() 69 __m256i tr1_b = _mm256_unpackhi_epi32(tr0_9, tr0_b); in mm256_transpose_16x16() 71 __m256i tr1_d = _mm256_unpackhi_epi32(tr0_c, tr0_e); in mm256_transpose_16x16() 73 __m256i tr1_f = _mm256_unpackhi_epi32(tr0_d, tr0_f); in mm256_transpose_16x16()
|
D | txfm_common_avx2.h | 137 u[i + 2] = _mm256_unpackhi_epi32(t[i], t[i + 2]); in transpose2_8x8_avx2() 140 u[i + 6] = _mm256_unpackhi_epi32(t[i + 4], t[i + 6]); in transpose2_8x8_avx2() 213 const __m256i b4 = _mm256_unpackhi_epi32(a0, a1); in transpose_16bit_16x8_avx2() 214 const __m256i b5 = _mm256_unpackhi_epi32(a2, a3); in transpose_16bit_16x8_avx2() 215 const __m256i b6 = _mm256_unpackhi_epi32(a4, a5); in transpose_16bit_16x8_avx2() 216 const __m256i b7 = _mm256_unpackhi_epi32(a6, a7); in transpose_16bit_16x8_avx2()
|
D | avg_intrin_avx2.c | 68 b2 = _mm256_unpackhi_epi32(a0, a1); in hadamard_col8x2_avx2() 69 b3 = _mm256_unpackhi_epi32(a4, a5); in hadamard_col8x2_avx2() 72 b6 = _mm256_unpackhi_epi32(a2, a3); in hadamard_col8x2_avx2() 73 b7 = _mm256_unpackhi_epi32(a6, a7); in hadamard_col8x2_avx2() 300 a2 = _mm256_unpackhi_epi32(b0, b1); in highbd_hadamard_col8_avx2() 301 a3 = _mm256_unpackhi_epi32(b2, b3); in highbd_hadamard_col8_avx2() 304 a6 = _mm256_unpackhi_epi32(b4, b5); in highbd_hadamard_col8_avx2() 305 a7 = _mm256_unpackhi_epi32(b6, b7); in highbd_hadamard_col8_avx2()
|
D | blk_sse_sum_avx2.c | 77 temp_buffer2 = _mm256_unpackhi_epi32(sse_buffer, _mm256_setzero_si256()); in sse_sum_wd4_avx2() 112 temp_buffer2 = _mm256_unpackhi_epi32(sse_buffer, _mm256_setzero_si256()); in sse_sum_wd8_avx2() 146 temp_buffer2 = _mm256_unpackhi_epi32(sse_buffer, _mm256_setzero_si256()); in sse_sum_wd16_avx2()
|
D | highbd_variance_avx2.c | 46 __m256i v_d_h = _mm256_unpackhi_epi32(v_sum0, v_sse_d); in aom_highbd_calc8x8var_avx2() 74 __m256i v_d_h = _mm256_unpackhi_epi32(v_sum0, v_sse_d); in aom_highbd_calc16x16var_avx2()
|
D | intrapred_avx2.c | 194 ww0 = _mm256_unpackhi_epi32(w0, w1); // 02 12 22 32 03 13 23 33 in highbd_transpose4x16_avx2() 195 ww1 = _mm256_unpackhi_epi32(w2, w3); // 42 52 62 72 43 53 63 73 in highbd_transpose4x16_avx2() 215 ww0 = _mm256_unpackhi_epi32(w0, w1); // 02 12 22 32 03 13 23 33 in highbd_transpose8x16_16x8_avx2() 216 ww1 = _mm256_unpackhi_epi32(w2, w3); // 42 52 62 72 43 53 63 73 in highbd_transpose8x16_16x8_avx2() 232 ww0 = _mm256_unpackhi_epi32(w0, w1); // 06 16 26 36 07 17 27 37 in highbd_transpose8x16_16x8_avx2() 233 ww1 = _mm256_unpackhi_epi32(w2, w3); // 46 56 66 76 47 57 67 77 in highbd_transpose8x16_16x8_avx2() 253 ww0 = _mm256_unpackhi_epi32(w0, w1); // in highbd_transpose16x16_avx2() 254 ww1 = _mm256_unpackhi_epi32(w2, w3); // in highbd_transpose16x16_avx2() 270 ww0 = _mm256_unpackhi_epi32(w0, w1); // in highbd_transpose16x16_avx2() 271 ww1 = _mm256_unpackhi_epi32(w2, w3); // in highbd_transpose16x16_avx2() [all …]
|
D | sse_avx2.c | 39 const __m256i sum1_4x64 = _mm256_unpackhi_epi32(*sum_all, zero); in summary_all_avx2()
|
/external/ruy/ruy/ |
D | pack_avx2_fma.cc | 172 r2 = _mm256_unpackhi_epi32(t0, t1); 173 r6 = _mm256_unpackhi_epi32(t4, t5); 176 r3 = _mm256_unpackhi_epi32(t2, t3); 177 r7 = _mm256_unpackhi_epi32(t6, t7); 304 r2 = _mm256_unpackhi_epi32(t0, t1); 305 r6 = _mm256_unpackhi_epi32(t4, t5); 308 r3 = _mm256_unpackhi_epi32(t2, t3); 309 r7 = _mm256_unpackhi_epi32(t6, t7); 390 r2 = _mm256_unpackhi_epi32(t0, t1); 391 r6 = _mm256_unpackhi_epi32(t4, t5); [all …]
|
/external/libvpx/libvpx/vpx_dsp/x86/ |
D | fwd_dct32x32_impl_avx2.h | 1554 u[1] = _mm256_unpackhi_epi32(lstep3[12], lstep3[10]); in FDCT32x32_2D_AVX2() 1556 u[3] = _mm256_unpackhi_epi32(lstep3[13], lstep3[11]); in FDCT32x32_2D_AVX2() 1593 u[1] = _mm256_unpackhi_epi32(lstep3[36], lstep3[58]); in FDCT32x32_2D_AVX2() 1595 u[3] = _mm256_unpackhi_epi32(lstep3[37], lstep3[59]); in FDCT32x32_2D_AVX2() 1597 u[5] = _mm256_unpackhi_epi32(lstep3[38], lstep3[56]); in FDCT32x32_2D_AVX2() 1599 u[7] = _mm256_unpackhi_epi32(lstep3[39], lstep3[57]); in FDCT32x32_2D_AVX2() 1601 u[9] = _mm256_unpackhi_epi32(lstep3[40], lstep3[54]); in FDCT32x32_2D_AVX2() 1603 u[11] = _mm256_unpackhi_epi32(lstep3[41], lstep3[55]); in FDCT32x32_2D_AVX2() 1605 u[13] = _mm256_unpackhi_epi32(lstep3[42], lstep3[52]); in FDCT32x32_2D_AVX2() 1607 u[15] = _mm256_unpackhi_epi32(lstep3[43], lstep3[53]); in FDCT32x32_2D_AVX2() [all …]
|
D | avg_intrin_avx2.c | 59 a2 = _mm256_unpackhi_epi32(b0, b1); in highbd_hadamard_col8_avx2() 60 a3 = _mm256_unpackhi_epi32(b2, b3); in highbd_hadamard_col8_avx2() 63 a6 = _mm256_unpackhi_epi32(b4, b5); in highbd_hadamard_col8_avx2() 64 a7 = _mm256_unpackhi_epi32(b6, b7); in highbd_hadamard_col8_avx2() 270 b2 = _mm256_unpackhi_epi32(a0, a1); in hadamard_col8x2_avx2() 271 b3 = _mm256_unpackhi_epi32(a4, a5); in hadamard_col8x2_avx2() 274 b6 = _mm256_unpackhi_epi32(a2, a3); in hadamard_col8x2_avx2() 275 b7 = _mm256_unpackhi_epi32(a6, a7); in hadamard_col8x2_avx2()
|
/external/libaom/libaom/av1/encoder/x86/ |
D | error_intrin_avx2.c | 63 const __m256i exp_error_hi = _mm256_unpackhi_epi32(error, zero); in av1_block_error_lp_avx2() 111 exp_dqcoeff_hi = _mm256_unpackhi_epi32(dqcoeff_reg, zero_reg); in av1_block_error_avx2() 114 exp_coeff_hi = _mm256_unpackhi_epi32(coeff_reg, zero_reg); in av1_block_error_avx2()
|
D | wedge_utils_avx2.c | 136 _mm256_unpackhi_epi32(v_acc0_d, v_sign_d)); in av1_wedge_sign_from_residuals_avx2()
|
D | av1_fwd_txfm2d_avx2.c | 1436 __m256i temp1 = _mm256_unpackhi_epi32(inputA[0], inputA[2]); in transpose_32_8x8_avx2() 1438 __m256i temp3 = _mm256_unpackhi_epi32(inputA[1], inputA[3]); in transpose_32_8x8_avx2() 1440 __m256i temp5 = _mm256_unpackhi_epi32(inputA[4], inputA[6]); in transpose_32_8x8_avx2() 1442 __m256i temp7 = _mm256_unpackhi_epi32(inputA[5], inputA[7]); in transpose_32_8x8_avx2() 1445 __m256i t1 = _mm256_unpackhi_epi32(temp0, temp2); in transpose_32_8x8_avx2() 1447 __m256i t3 = _mm256_unpackhi_epi32(temp1, temp3); in transpose_32_8x8_avx2() 1449 __m256i t5 = _mm256_unpackhi_epi32(temp4, temp6); in transpose_32_8x8_avx2() 1451 __m256i t7 = _mm256_unpackhi_epi32(temp5, temp7); in transpose_32_8x8_avx2()
|
D | highbd_fwd_txfm_avx2.c | 151 u1 = _mm256_unpackhi_epi32(in[0 * instride], in[1 * instride]); in fwd_txfm_transpose_8x8_avx2() 154 u3 = _mm256_unpackhi_epi32(in[2 * instride], in[3 * instride]); in fwd_txfm_transpose_8x8_avx2() 157 u5 = _mm256_unpackhi_epi32(in[4 * instride], in[5 * instride]); in fwd_txfm_transpose_8x8_avx2() 160 u7 = _mm256_unpackhi_epi32(in[6 * instride], in[7 * instride]); in fwd_txfm_transpose_8x8_avx2()
|
/external/libvpx/libvpx/vp9/encoder/x86/ |
D | vp9_error_avx2.c | 77 exp_dqcoeff_hi = _mm256_unpackhi_epi32(dqcoeff_0, zero); in vp9_block_error_avx2() 80 exp_coeff_hi = _mm256_unpackhi_epi32(coeff_0, zero); in vp9_block_error_avx2() 141 const __m256i exp_error_hi = _mm256_unpackhi_epi32(error, zero); in vp9_block_error_fp_avx2()
|
/external/libaom/libaom/av1/common/x86/ |
D | warp_plane_avx2.c | 184 const __m256i res_1 = _mm256_unpackhi_epi32(tmp_12, tmp_14); in prepare_horizontal_filter_coeff_avx2() 186 const __m256i res_3 = _mm256_unpackhi_epi32(tmp_13, tmp_15); in prepare_horizontal_filter_coeff_avx2() 224 const __m256i res_1 = _mm256_unpackhi_epi32(tmp_12, tmp_14); in prepare_horizontal_filter_coeff_beta0_avx2() 226 const __m256i res_3 = _mm256_unpackhi_epi32(tmp_13, tmp_15); in prepare_horizontal_filter_coeff_beta0_avx2() 486 __m256i res_2 = _mm256_unpackhi_epi32(filt_0, filt_1); in prepare_vertical_filter_coeffs_avx2() 487 __m256i res_3 = _mm256_unpackhi_epi32(filt_2, filt_3); in prepare_vertical_filter_coeffs_avx2() 531 res_2 = _mm256_unpackhi_epi32(filt_0, filt_1); in prepare_vertical_filter_coeffs_avx2() 532 res_3 = _mm256_unpackhi_epi32(filt_2, filt_3); in prepare_vertical_filter_coeffs_avx2() 562 __m256i res_2 = _mm256_unpackhi_epi32(filt_0, filt_1); in prepare_vertical_filter_coeffs_delta0_avx2() 563 __m256i res_3 = _mm256_unpackhi_epi32(filt_2, filt_3); in prepare_vertical_filter_coeffs_delta0_avx2() [all …]
|
D | highbd_wiener_convolve_avx2.c | 226 const __m256i res_hi = _mm256_unpackhi_epi32(res_even, res_odd); in av1_highbd_wiener_convolve_add_src_avx2()
|
D | highbd_inv_txfm_avx2.c | 165 u1 = _mm256_unpackhi_epi32(in[0], in[1]); in transpose_8x8_avx2() 168 u3 = _mm256_unpackhi_epi32(in[2], in[3]); in transpose_8x8_avx2() 171 u5 = _mm256_unpackhi_epi32(in[4], in[5]); in transpose_8x8_avx2() 174 u7 = _mm256_unpackhi_epi32(in[6], in[7]); in transpose_8x8_avx2() 202 u1 = _mm256_unpackhi_epi32(in[7], in[6]); in transpose_8x8_flip_avx2() 205 u3 = _mm256_unpackhi_epi32(in[5], in[4]); in transpose_8x8_flip_avx2() 208 u5 = _mm256_unpackhi_epi32(in[3], in[2]); in transpose_8x8_flip_avx2() 211 u7 = _mm256_unpackhi_epi32(in[1], in[0]); in transpose_8x8_flip_avx2()
|
D | highbd_jnt_convolve_avx2.c | 585 __m256i res2 = _mm256_unpackhi_epi32(res_even, res_odd); in av1_highbd_dist_wtd_convolve_x_avx2()
|
/external/gemmlowp/internal/ |
D | pack_avx.h | 94 __m256i ymm8 = _mm256_unpackhi_epi32(ymm5, ymm6); in Pack() 97 __m256i ymm14 = _mm256_unpackhi_epi32(ymm9, ymm10); in Pack()
|
/external/eigen/Eigen/src/Core/arch/CUDA/ |
D | PacketMathHalf.h | 569 __m256i abcd_47 = _mm256_unpackhi_epi32(ab_07, cd_07); 571 __m256i efgh_47 = _mm256_unpackhi_epi32(ef_07, gh_07); 573 __m256i ijkl_47 = _mm256_unpackhi_epi32(ij_07, kl_07); 575 __m256i mnop_47 = _mm256_unpackhi_epi32(mn_07, op_07); 578 __m256i abcd_cf = _mm256_unpackhi_epi32(ab_8f, cd_8f); 580 __m256i efgh_cf = _mm256_unpackhi_epi32(ef_8f, gh_8f); 582 __m256i ijkl_cf = _mm256_unpackhi_epi32(ij_8f, kl_8f); 584 __m256i mnop_cf = _mm256_unpackhi_epi32(mn_8f, op_8f);
|
/external/libaom/libaom/aom_dsp/simd/ |
D | v256_intrinsics_x86.h | 172 return _mm256_unpackhi_epi32( in v256_ziphi_32() 739 return v256_add_64(s, v256_add_64(_mm256_unpackhi_epi32(d, v256_zero()), in v256_ssd_s16()
|
/external/skia/src/opts/ |
D | SkSwizzler_opts.h | 651 ba = _mm256_unpackhi_epi32(*lo, *hi); // bbbbBBBB aaaaAAAA bbbbBBBB aaaaAAAA in premul_should_swapRB() 809 yk = _mm256_unpackhi_epi32(*lo, *hi); // yyyyYYYY kkkkKKKK yyyyYYYY kkkkKKKK in inverted_cmyk_to()
|
/external/clang/test/CodeGen/ |
D | avx2-builtins.c | 1189 return _mm256_unpackhi_epi32(a, b); in test_mm256_unpackhi_epi32()
|
/external/llvm-project/clang/test/CodeGen/X86/ |
D | avx2-builtins.c | 1237 return _mm256_unpackhi_epi32(a, b); in test_mm256_unpackhi_epi32()
|