Searched refs:_mm256_loadu_si256 (Results 1 – 16 of 16) sorted by relevance
/external/flac/libFLAC/ |
D | lpc_intrin_avx2.c | 80 summ = _mm256_madd_epi16(q11, _mm256_loadu_si256((const __m256i*)(data+i-12))); in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_avx2() 81 …mull = _mm256_madd_epi16(q10, _mm256_loadu_si256((const __m256i*)(data+i-11))); summ = _mm256_add_… in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_avx2() 82 …mull = _mm256_madd_epi16(q9, _mm256_loadu_si256((const __m256i*)(data+i-10))); summ = _mm256_add_… in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_avx2() 83 …mull = _mm256_madd_epi16(q8, _mm256_loadu_si256((const __m256i*)(data+i-9 ))); summ = _mm256_add_… in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_avx2() 84 …mull = _mm256_madd_epi16(q7, _mm256_loadu_si256((const __m256i*)(data+i-8 ))); summ = _mm256_add_… in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_avx2() 85 …mull = _mm256_madd_epi16(q6, _mm256_loadu_si256((const __m256i*)(data+i-7 ))); summ = _mm256_add_… in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_avx2() 86 …mull = _mm256_madd_epi16(q5, _mm256_loadu_si256((const __m256i*)(data+i-6 ))); summ = _mm256_add_… in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_avx2() 87 …mull = _mm256_madd_epi16(q4, _mm256_loadu_si256((const __m256i*)(data+i-5 ))); summ = _mm256_add_… in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_avx2() 88 …mull = _mm256_madd_epi16(q3, _mm256_loadu_si256((const __m256i*)(data+i-4 ))); summ = _mm256_add_… in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_avx2() 89 …mull = _mm256_madd_epi16(q2, _mm256_loadu_si256((const __m256i*)(data+i-3 ))); summ = _mm256_add_… in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_avx2() [all …]
|
D | stream_encoder_intrin_avx2.c | 70 … __m256i res256 = _mm256_abs_epi32(_mm256_loadu_si256((const __m256i*)(residual+residual_sample))); in FLAC__precompute_partition_info_sums_intrin_avx2()
|
/external/libvpx/libvpx/vpx_dsp/x86/ |
D | sad_avx2.c | 23 ref1_reg = _mm256_loadu_si256((__m256i const *)ref_ptr); \ 24 ref2_reg = _mm256_loadu_si256((__m256i const *)(ref_ptr + 32)); \ 26 ref1_reg, _mm256_loadu_si256((__m256i const *)src_ptr)); \ 28 ref2_reg, _mm256_loadu_si256((__m256i const *)(src_ptr + 32))); \ 54 ref1_reg = _mm256_loadu_si256((__m256i const *)ref_ptr); \ 55 ref2_reg = _mm256_loadu_si256((__m256i const *)(ref_ptr + ref_stride)); \ 57 ref1_reg, _mm256_loadu_si256((__m256i const *)src_ptr)); \ 60 _mm256_loadu_si256((__m256i const *)(src_ptr + src_stride))); \ 101 ref1_reg = _mm256_loadu_si256((__m256i const *)ref_ptr); \ 102 ref2_reg = _mm256_loadu_si256((__m256i const *)(ref_ptr + 32)); \ [all …]
|
D | sad4d_avx2.c | 33 src_reg = _mm256_loadu_si256((const __m256i *)src); in vpx_sad32x32x4d_avx2() 34 ref0_reg = _mm256_loadu_si256((const __m256i *)ref0); in vpx_sad32x32x4d_avx2() 35 ref1_reg = _mm256_loadu_si256((const __m256i *)ref1); in vpx_sad32x32x4d_avx2() 36 ref2_reg = _mm256_loadu_si256((const __m256i *)ref2); in vpx_sad32x32x4d_avx2() 37 ref3_reg = _mm256_loadu_si256((const __m256i *)ref3); in vpx_sad32x32x4d_avx2() 103 src_reg = _mm256_loadu_si256((const __m256i *)src); in vpx_sad64x64x4d_avx2() 104 srcnext_reg = _mm256_loadu_si256((const __m256i *)(src + 32)); in vpx_sad64x64x4d_avx2() 105 ref0_reg = _mm256_loadu_si256((const __m256i *)ref0); in vpx_sad64x64x4d_avx2() 106 ref0next_reg = _mm256_loadu_si256((const __m256i *)(ref0 + 32)); in vpx_sad64x64x4d_avx2() 107 ref1_reg = _mm256_loadu_si256((const __m256i *)ref1); in vpx_sad64x64x4d_avx2() [all …]
|
D | variance_avx2.c | 108 const __m256i src0 = _mm256_loadu_si256((__m256i const *)(src_ptr)); in get32x16var_avx2() 110 _mm256_loadu_si256((__m256i const *)(src_ptr + source_stride)); in get32x16var_avx2() 111 const __m256i ref0 = _mm256_loadu_si256((__m256i const *)(ref_ptr)); in get32x16var_avx2() 113 _mm256_loadu_si256((__m256i const *)(ref_ptr + recon_stride)); in get32x16var_avx2() 223 const __m256i dst_reg = _mm256_loadu_si256((__m256i const *)dst); in spv32_x0_y0() 224 const __m256i src_reg = _mm256_loadu_si256((__m256i const *)src); in spv32_x0_y0() 226 const __m256i sec_reg = _mm256_loadu_si256((__m256i const *)sec); in spv32_x0_y0() 251 const __m256i dst_reg = _mm256_loadu_si256((__m256i const *)dst); in spv32_half_zero() 252 const __m256i src_0 = _mm256_loadu_si256((__m256i const *)src); in spv32_half_zero() 253 const __m256i src_1 = _mm256_loadu_si256((__m256i const *)(src + sstep)); in spv32_half_zero() [all …]
|
D | highbd_convolve_avx2.c | 34 const __m256i p0 = _mm256_loadu_si256((const __m256i *)src); in vpx_highbd_convolve_copy_avx2() 35 const __m256i p1 = _mm256_loadu_si256((const __m256i *)(src + 16)); in vpx_highbd_convolve_copy_avx2() 36 const __m256i p2 = _mm256_loadu_si256((const __m256i *)(src + 32)); in vpx_highbd_convolve_copy_avx2() 37 const __m256i p3 = _mm256_loadu_si256((const __m256i *)(src + 48)); in vpx_highbd_convolve_copy_avx2() 48 const __m256i p0 = _mm256_loadu_si256((const __m256i *)src); in vpx_highbd_convolve_copy_avx2() 49 const __m256i p1 = _mm256_loadu_si256((const __m256i *)(src + 16)); in vpx_highbd_convolve_copy_avx2() 59 p0 = _mm256_loadu_si256((const __m256i *)src); in vpx_highbd_convolve_copy_avx2() 61 p1 = _mm256_loadu_si256((const __m256i *)src); in vpx_highbd_convolve_copy_avx2() 117 p0 = _mm256_loadu_si256((const __m256i *)src); in vpx_highbd_convolve_avg_avx2() 118 p1 = _mm256_loadu_si256((const __m256i *)(src + 16)); in vpx_highbd_convolve_avg_avx2() [all …]
|
D | avg_intrin_avx2.c | 97 src[0] = _mm256_loadu_si256((const __m256i *)src_diff); in hadamard_8x8x2_avx2() 98 src[1] = _mm256_loadu_si256((const __m256i *)(src_diff += src_stride)); in hadamard_8x8x2_avx2() 99 src[2] = _mm256_loadu_si256((const __m256i *)(src_diff += src_stride)); in hadamard_8x8x2_avx2() 100 src[3] = _mm256_loadu_si256((const __m256i *)(src_diff += src_stride)); in hadamard_8x8x2_avx2() 101 src[4] = _mm256_loadu_si256((const __m256i *)(src_diff += src_stride)); in hadamard_8x8x2_avx2() 102 src[5] = _mm256_loadu_si256((const __m256i *)(src_diff += src_stride)); in hadamard_8x8x2_avx2() 103 src[6] = _mm256_loadu_si256((const __m256i *)(src_diff += src_stride)); in hadamard_8x8x2_avx2() 104 src[7] = _mm256_loadu_si256((const __m256i *)(src_diff += src_stride)); in hadamard_8x8x2_avx2() 150 const __m256i coeff0 = _mm256_loadu_si256((const __m256i *)t_coeff); in vpx_hadamard_16x16_avx2() 151 const __m256i coeff1 = _mm256_loadu_si256((const __m256i *)(t_coeff + 64)); in vpx_hadamard_16x16_avx2() [all …]
|
D | bitdepth_conversion_avx2.h | 23 const __m256i a_low = _mm256_loadu_si256((const __m256i *)a); in load_tran_low() 24 const __m256i a_high = _mm256_loadu_si256((const __m256i *)(a + 8)); in load_tran_low() 27 return _mm256_loadu_si256((const __m256i *)a); in load_tran_low()
|
D | fwd_dct32x32_impl_avx2.h | 117 const __m256i ina0 = _mm256_loadu_si256((const __m256i *)(ina)); in FDCT32x32_2D_AVX2() 119 _mm256_loadu_si256((const __m256i *)(ina + str1)); in FDCT32x32_2D_AVX2() 121 _mm256_loadu_si256((const __m256i *)(ina + str2)); in FDCT32x32_2D_AVX2() 123 _mm256_loadu_si256((const __m256i *)(ina + str3)); in FDCT32x32_2D_AVX2() 125 _mm256_loadu_si256((const __m256i *)(inb - str3)); in FDCT32x32_2D_AVX2() 127 _mm256_loadu_si256((const __m256i *)(inb - str2)); in FDCT32x32_2D_AVX2() 129 _mm256_loadu_si256((const __m256i *)(inb - str1)); in FDCT32x32_2D_AVX2() 130 const __m256i inb0 = _mm256_loadu_si256((const __m256i *)(inb)); in FDCT32x32_2D_AVX2() 153 const __m256i ina0 = _mm256_loadu_si256((const __m256i *)(ina)); in FDCT32x32_2D_AVX2() 155 _mm256_loadu_si256((const __m256i *)(ina + str1)); in FDCT32x32_2D_AVX2() [all …]
|
/external/mesa3d/src/gallium/drivers/swr/rasterizer/common/ |
D | simd16intrin.h | 242 result.lo = _mm256_loadu_si256(&m[0].lo); in _simd16_loadu_si() 243 result.hi = _mm256_loadu_si256(&m[0].hi); in _simd16_loadu_si()
|
D | simdintrin.h | 618 #define _simd_loadu_si _mm256_loadu_si256
|
/external/clang/test/CodeGen/ |
D | avx-builtins.c | 489 return _mm256_loadu_si256(A); in test_mm256_loadu_si256()
|
/external/eigen/Eigen/src/Core/arch/AVX/ |
D | PacketMath.h | 219 …adu<Packet8i>(const int* from) { EIGEN_DEBUG_UNALIGNED_LOAD return _mm256_loadu_si256(reinterpret_…
|
/external/clang/lib/Headers/ |
D | avxintrin.h | 2387 _mm256_loadu_si256(__m256i const *__p) in _mm256_loadu_si256() function
|
/external/eigen/Eigen/src/Core/arch/CUDA/ |
D | PacketMathHalf.h | 395 result.x = _mm256_loadu_si256(reinterpret_cast<const __m256i*>(from));
|
/external/pdfium/third_party/libopenjpeg20/ |
D | dwt.c | 607 #define LOADU(x) _mm256_loadu_si256((const VREG*)(x))
|