| /external/libgav1/libgav1/src/dsp/x86/ |
| D | super_res_sse4.cc | 82 const __m128i res0 = _mm_hadd_epi32(weighted_src[0], weighted_src[1]); in ComputeSuperRes4()
83 const __m128i res1 = _mm_hadd_epi32(weighted_src[2], weighted_src[3]); in ComputeSuperRes4()
84 const __m128i result0 = _mm_hadd_epi32(res0, res1); in ComputeSuperRes4()
103 const __m128i res0 = _mm_hadd_epi32(weighted_src[0], weighted_src[1]); in ComputeSuperRes8()
104 const __m128i res1 = _mm_hadd_epi32(weighted_src[2], weighted_src[3]); in ComputeSuperRes8()
105 const __m128i res2 = _mm_hadd_epi32(weighted_src[4], weighted_src[5]); in ComputeSuperRes8()
106 const __m128i res3 = _mm_hadd_epi32(weighted_src[6], weighted_src[7]); in ComputeSuperRes8()
107 const __m128i result0 = _mm_hadd_epi32(res0, res1); in ComputeSuperRes8()
108 const __m128i result1 = _mm_hadd_epi32(res2, res3); in ComputeSuperRes8()
|
| D | cdef_sse4.cc | 338 a = _mm_hadd_epi32(a, a); in SumVector_S32()
|
| /external/libaom/libaom/av1/common/x86/ |
| D | av1_convolve_horiz_rs_sse4.c | 105 const __m128i conv01_32 = _mm_hadd_epi32(conv0_32, conv1_32); in av1_convolve_horiz_rs_sse4_1()
106 const __m128i conv23_32 = _mm_hadd_epi32(conv2_32, conv3_32); in av1_convolve_horiz_rs_sse4_1()
108 const __m128i conv0123_32 = _mm_hadd_epi32(conv01_32, conv23_32); in av1_convolve_horiz_rs_sse4_1()
208 const __m128i conv01_32 = _mm_hadd_epi32(conv0_32, conv1_32); in av1_highbd_convolve_horiz_rs_sse4_1()
209 const __m128i conv23_32 = _mm_hadd_epi32(conv2_32, conv3_32); in av1_highbd_convolve_horiz_rs_sse4_1()
211 const __m128i conv0123_32 = _mm_hadd_epi32(conv01_32, conv23_32); in av1_highbd_convolve_horiz_rs_sse4_1()
|
| D | av1_convolve_scale_sse4.c | 75 const __m128i conv01lo = _mm_hadd_epi32(conv0lo, conv1lo); in hfilter8()
76 const __m128i conv23lo = _mm_hadd_epi32(conv2lo, conv3lo); in hfilter8()
77 const __m128i conv = _mm_hadd_epi32(conv01lo, conv23lo); in hfilter8()
161 const __m128i conv01 = _mm_hadd_epi32(conv0, conv1); in vfilter8()
162 const __m128i conv23 = _mm_hadd_epi32(conv2, conv3); in vfilter8()
163 __m128i conv = _mm_hadd_epi32(conv01, conv23); in vfilter8()
305 const __m128i conv01lo = _mm_hadd_epi32(conv0lo, conv1lo); in highbd_hfilter8()
306 const __m128i conv23lo = _mm_hadd_epi32(conv2lo, conv3lo); in highbd_hfilter8()
307 const __m128i conv = _mm_hadd_epi32(conv01lo, conv23lo); in highbd_hfilter8()
392 const __m128i conv01 = _mm_hadd_epi32(conv0, conv1); in highbd_vfilter8()
[all …]
|
| /external/libaom/libaom/aom_dsp/x86/ |
| D | highbd_variance_sse4.c | 54 s2 = _mm_hadd_epi32(t0, t1); in variance4x4_64_sse4_1()
55 s3 = _mm_hadd_epi32(s2, s2); in variance4x4_64_sse4_1()
56 y0 = _mm_hadd_epi32(s3, s3); in variance4x4_64_sse4_1()
61 s2 = _mm_hadd_epi32(t0, t1); in variance4x4_64_sse4_1()
62 s3 = _mm_hadd_epi32(s2, s2); in variance4x4_64_sse4_1()
63 x0 = _mm_hadd_epi32(s3, s3); in variance4x4_64_sse4_1()
|
| D | obmc_intrinsic_ssse3.h | 20 v_d = _mm_hadd_epi32(v_d, v_d); in xx_hsum_epi32_si32()
21 v_d = _mm_hadd_epi32(v_d, v_d); in xx_hsum_epi32_si32()
|
| D | masked_sad_intrin_ssse3.c | 338 res = _mm_hadd_epi32(res, res); in highbd_masked_sad_ssse3()
339 res = _mm_hadd_epi32(res, res); in highbd_masked_sad_ssse3()
398 res = _mm_hadd_epi32(res, res); in aom_highbd_masked_sad4xh_ssse3()
399 res = _mm_hadd_epi32(res, res); in aom_highbd_masked_sad4xh_ssse3()
|
| D | obmc_variance_avx2.c | 77 v_d = _mm_hadd_epi32(v_sum_d, v_sse_d); in obmc_variance_w8n()
78 v_d = _mm_hadd_epi32(v_d, v_d); in obmc_variance_w8n()
|
| D | masked_variance_intrin_ssse3.c | 452 sum = _mm_hadd_epi32(sum, sum_sq); in masked_variance()
453 sum = _mm_hadd_epi32(sum, sum); in masked_variance()
482 sum = _mm_hadd_epi32(sum, sum_sq); in masked_variance8xh()
483 sum = _mm_hadd_epi32(sum, sum); in masked_variance8xh()
514 sum = _mm_hadd_epi32(sum, sum_sq); in masked_variance4xh()
515 sum = _mm_hadd_epi32(sum, sum); in masked_variance4xh()
958 sum = _mm_hadd_epi32(sum, zero); in highbd_masked_variance()
959 sum = _mm_hadd_epi32(sum, zero); in highbd_masked_variance()
1024 sum = _mm_hadd_epi32(sum, sum_sq); in highbd_masked_variance4xh()
1025 sum = _mm_hadd_epi32(sum, zero); in highbd_masked_variance4xh()
|
| /external/flac/libFLAC/ |
| D | stream_encoder_intrin_ssse3.c | 89 mm_sum = _mm_hadd_epi32(mm_sum, mm_sum); in FLAC__precompute_partition_info_sums_intrin_ssse3()
90 mm_sum = _mm_hadd_epi32(mm_sum, mm_sum); in FLAC__precompute_partition_info_sums_intrin_ssse3()
|
| D | stream_encoder_intrin_avx2.c | 86 sum128 = _mm_hadd_epi32(sum128, sum128); in FLAC__precompute_partition_info_sums_intrin_avx2()
87 sum128 = _mm_hadd_epi32(sum128, sum128); in FLAC__precompute_partition_info_sums_intrin_avx2()
|
| /external/libaom/libaom/av1/encoder/x86/ |
| D | rdopt_sse4.c | 50 const __m128i xy32 = _mm_hadd_epi32(xy_madd_b, xy_madd_a); in horver_correlation_4x4()
62 const __m128i xz32 = _mm_hadd_epi32(xz_madd_b, xz_madd_a); in horver_correlation_4x4()
82 const __m128i sum2 = _mm_hadd_epi32(slli_a_2, swap_b_2); in horver_correlation_4x4()
|
| D | temporal_filter_sse4.c | 1500 y_reg_fst = _mm_hadd_epi32(y_fst, y_snd); in highbd_add_luma_dist_to_8_chroma_mod()
1512 y_reg_snd = _mm_hadd_epi32(y_fst, y_snd); in highbd_add_luma_dist_to_8_chroma_mod()
|
| /external/tensorflow/tensorflow/lite/kernels/internal/optimized/ |
| D | sse_tensor_utils.cc | 48 acc = _mm_hadd_epi32(acc, acc); in ReduceInt32x4()
52 acc = _mm_hadd_epi32(acc, acc); in ReduceInt32x4()
|
| /external/XNNPACK/src/q8-gemm/ |
| D | 2x4c8-sse2.c | 20 const __m128i xxyy = _mm_hadd_epi32(x, y); in sse_reduce4_i32()
22 const __m128i zzww = _mm_hadd_epi32(z, w); in sse_reduce4_i32()
24 return _mm_hadd_epi32(xxyy, zzww); in sse_reduce4_i32()
|
| /external/libavc/common/x86/ |
| D | ih264_ihadamard_scaling_sse42.c | 221 temp0 = _mm_hadd_epi32(plane_0, plane_1); //a0+a1 a2+a3 b0+b1 b2+b3 in ih264_ihadamard_scaling_2x2_uv_sse42()
223 …plane_0 = _mm_hadd_epi32(temp0, temp1); //a0+a1+a2+a3 b0+b1+b2+b3 a0-a1+a2-a3 b0-b1+b2… in ih264_ihadamard_scaling_2x2_uv_sse42()
|
| D | ih264_chroma_intra_pred_filters_ssse3.c | 292 hv_val_4x32b = _mm_hadd_epi32(h_val_4x32b, v_val_4x32b); in ih264_intra_pred_chroma_8x8_mode_plane_ssse3()
|
| D | ih264_resi_trans_quant_sse42.c | 918 temp0 = _mm_hadd_epi32(plane_0, plane_1); //a0+a1 a2+a3 b0+b1 b2+b3 in ih264_hadamard_quant_2x2_uv_sse42()
921 …plane_0 = _mm_hadd_epi32(temp0, temp1); //a0+a1+a2+a3 b0+b1+b2+b3 a0-a1+a2-a3 b0-b1+b2… in ih264_hadamard_quant_2x2_uv_sse42()
|
| D | ih264_luma_intra_pred_filters_ssse3.c | 2055 hv_val_4x32b = _mm_hadd_epi32(h_val_4x32b, v_val_4x32b); in ih264_intra_pred_luma_16x16_mode_plane_ssse3()
2056 hv_val_4x32b = _mm_hadd_epi32(hv_val_4x32b, hv_val_4x32b); in ih264_intra_pred_luma_16x16_mode_plane_ssse3()
|
| /external/clang/test/CodeGen/ |
| D | ssse3-builtins.c | 49 return _mm_hadd_epi32(a, b); in test_mm_hadd_epi32()
|
| /external/eigen/Eigen/src/Core/arch/SSE/ |
| D | PacketMath.h | 561 return _mm_hadd_epi32(_mm_hadd_epi32(vecs[0], vecs[1]),_mm_hadd_epi32(vecs[2], vecs[3]));
565 Packet4i tmp0 = _mm_hadd_epi32(a,a);
566 return pfirst<Packet4i>(_mm_hadd_epi32(tmp0,tmp0));
|
| /external/clang/lib/Headers/ |
| D | tmmintrin.h | 227 _mm_hadd_epi32(__m128i __a, __m128i __b) in _mm_hadd_epi32() function
|
| /external/libvpx/libvpx/vp9/encoder/x86/ |
| D | highbd_temporal_filter_sse4.c | 468 y_reg_fst = _mm_hadd_epi32(y_fst, y_snd); in highbd_add_luma_dist_to_8_chroma_mod()
480 y_reg_snd = _mm_hadd_epi32(y_fst, y_snd); in highbd_add_luma_dist_to_8_chroma_mod()
|
| /external/neon_2_sse/ |
| D | NEON_2_SSE.h | 6289 hadd128 = _mm_hadd_epi32 (_pM128i(a), _pM128i(b)); in vpadd_s32()
6376 r32_1 = _mm_hadd_epi32(r32_1, r32_1); //use low 64 bits in vpaddl_s16()
6439 return _mm_hadd_epi32 (r32_1, r32_2); in vpaddlq_s16()
|