/external/flac/libFLAC/ |
D | lpc_intrin_sse.c | 72 d = d0; d = _mm_shuffle_ps(d, d, 0); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_4_new() 82 d = _mm_load_ss(data+i); d = _mm_shuffle_ps(d, d, 0); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_4_new() 83 d0 = _mm_shuffle_ps(d0, d0, _MM_SHUFFLE(2,1,0,3)); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_4_new() 110 d = d0; d = _mm_shuffle_ps(d, d, 0); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_8_new() 122 d = _mm_load_ss(data+i); d = _mm_shuffle_ps(d, d, 0); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_8_new() 123 d1 = _mm_shuffle_ps(d1, d1, _MM_SHUFFLE(2,1,0,3)); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_8_new() 124 d0 = _mm_shuffle_ps(d0, d0, _MM_SHUFFLE(2,1,0,3)); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_8_new() 156 d = d0; d = _mm_shuffle_ps(d, d, 0); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_12_new() 170 d = _mm_load_ss(data+i); d = _mm_shuffle_ps(d, d, 0); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_12_new() 171 d2 = _mm_shuffle_ps(d2, d2, _MM_SHUFFLE(2,1,0,3)); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_12_new() [all …]
|
/external/webrtc/webrtc/modules/audio_processing/aec/ |
D | aec_rdft_sse2.c | 27 __m128 a01v = _mm_shuffle_ps(a00v, a08v, _MM_SHUFFLE(1, 0, 1, 0)); in cft1st_128_SSE2() 28 __m128 a23v = _mm_shuffle_ps(a00v, a08v, _MM_SHUFFLE(3, 2, 3, 2)); in cft1st_128_SSE2() 29 __m128 a45v = _mm_shuffle_ps(a04v, a12v, _MM_SHUFFLE(1, 0, 1, 0)); in cft1st_128_SSE2() 30 __m128 a67v = _mm_shuffle_ps(a04v, a12v, _MM_SHUFFLE(3, 2, 3, 2)); in cft1st_128_SSE2() 45 x0w = _mm_shuffle_ps(x0v, x0v, _MM_SHUFFLE(2, 3, 0, 1)); in cft1st_128_SSE2() 53 const __m128 x3w = _mm_shuffle_ps(x3v, x3v, _MM_SHUFFLE(2, 3, 0, 1)); in cft1st_128_SSE2() 56 x0w = _mm_shuffle_ps(x0v, x0v, _MM_SHUFFLE(2, 3, 0, 1)); in cft1st_128_SSE2() 62 x0w = _mm_shuffle_ps(x0v, x0v, _MM_SHUFFLE(2, 3, 0, 1)); in cft1st_128_SSE2() 70 a00v = _mm_shuffle_ps(a01v, a23v, _MM_SHUFFLE(1, 0, 1, 0)); in cft1st_128_SSE2() 71 a04v = _mm_shuffle_ps(a45v, a67v, _MM_SHUFFLE(1, 0, 1, 0)); in cft1st_128_SSE2() [all …]
|
D | aec_core_sse2.c | 220 _mm_shuffle_ps(fft0, fft4, _MM_SHUFFLE(2, 0, 2, 0)); in FilterAdaptationSSE2() 222 _mm_shuffle_ps(fft0, fft4, _MM_SHUFFLE(3, 1, 3, 1)); in FilterAdaptationSSE2() 437 sum = _mm_add_ps(sum, _mm_shuffle_ps(sum, sum, _MM_SHUFFLE(0, 0, 3, 2))); in _mm_add_ps_4x1() 439 sum = _mm_add_ps(sum, _mm_shuffle_ps(sum, sum, _MM_SHUFFLE(1, 1, 1, 1))); in _mm_add_ps_4x1() 533 __m128 vec_a = _mm_shuffle_ps(vec_3210, vec_7654, in SmoothedPSD() 535 __m128 vec_b = _mm_shuffle_ps(vec_3210, vec_7654, in SmoothedPSD() 554 __m128 vec_a = _mm_shuffle_ps(vec_3210, vec_7654, in SmoothedPSD() 556 __m128 vec_b = _mm_shuffle_ps(vec_3210, vec_7654, in SmoothedPSD() 632 _mm_shuffle_ps(vec_sqrtHanning_rev, vec_sqrtHanning_rev, in WindowDataSSE2() 647 const __m128 vec_a = _mm_shuffle_ps(vec_fft0, vec_fft4, in StoreAsComplexSSE2() [all …]
|
/external/eigen/Eigen/src/LU/arch/ |
D | Inverse_SSE.h | 84 AB = _mm_mul_ps(_mm_shuffle_ps(A,A,0x0F), B); 85 AB = _mm_sub_ps(AB,_mm_mul_ps(_mm_shuffle_ps(A,A,0xA5), _mm_shuffle_ps(B,B,0x4E))); 87 DC = _mm_mul_ps(_mm_shuffle_ps(D,D,0x0F), C); 88 DC = _mm_sub_ps(DC,_mm_mul_ps(_mm_shuffle_ps(D,D,0xA5), _mm_shuffle_ps(C,C,0x4E))); 91 dA = _mm_mul_ps(_mm_shuffle_ps(A, A, 0x5F),A); 94 dB = _mm_mul_ps(_mm_shuffle_ps(B, B, 0x5F),B); 98 dC = _mm_mul_ps(_mm_shuffle_ps(C, C, 0x5F),C); 101 dD = _mm_mul_ps(_mm_shuffle_ps(D, D, 0x5F),D); 105 d = _mm_mul_ps(_mm_shuffle_ps(DC,DC,0xD8),AB); 108 iD = _mm_mul_ps(_mm_shuffle_ps(C,C,0xA0), _mm_movelh_ps(AB,AB)); [all …]
|
/external/libopus/celt/x86/ |
D | pitch_sse.c | 56 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(_mm_shuffle_ps(x0,x0,0x00),yj)); in xcorr_kernel_sse() 57 xsum2 = _mm_add_ps(xsum2,_mm_mul_ps(_mm_shuffle_ps(x0,x0,0x55), in xcorr_kernel_sse() 58 _mm_shuffle_ps(yj,y3,0x49))); in xcorr_kernel_sse() 59 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(_mm_shuffle_ps(x0,x0,0xaa), in xcorr_kernel_sse() 60 _mm_shuffle_ps(yj,y3,0x9e))); in xcorr_kernel_sse() 61 xsum2 = _mm_add_ps(xsum2,_mm_mul_ps(_mm_shuffle_ps(x0,x0,0xff),y3)); in xcorr_kernel_sse() 96 xsum1 = _mm_add_ss(xsum1, _mm_shuffle_ps(xsum1, xsum1, 0x55)); in dual_inner_prod_sse() 99 xsum2 = _mm_add_ss(xsum2, _mm_shuffle_ps(xsum2, xsum2, 0x55)); in dual_inner_prod_sse() 124 sum = _mm_add_ss(sum, _mm_shuffle_ps(sum, sum, 0x55)); in celt_inner_prod_sse() 155 x2v = _mm_shuffle_ps(x0v, x4v, 0x4e); in comb_filter_const_sse() [all …]
|
D | vq_sse2.c | 82 sums = _mm_add_ps(sums, _mm_shuffle_ps(sums, sums, _MM_SHUFFLE(1, 0, 3, 2))); in op_pvq_search_sse2() 83 sums = _mm_add_ps(sums, _mm_shuffle_ps(sums, sums, _MM_SHUFFLE(2, 3, 0, 1))); in op_pvq_search_sse2() 129 xy4 = _mm_add_ps(xy4, _mm_shuffle_ps(xy4, xy4, _MM_SHUFFLE(1, 0, 3, 2))); in op_pvq_search_sse2() 130 xy4 = _mm_add_ps(xy4, _mm_shuffle_ps(xy4, xy4, _MM_SHUFFLE(2, 3, 0, 1))); in op_pvq_search_sse2() 132 yy4 = _mm_add_ps(yy4, _mm_shuffle_ps(yy4, yy4, _MM_SHUFFLE(1, 0, 3, 2))); in op_pvq_search_sse2() 133 yy4 = _mm_add_ps(yy4, _mm_shuffle_ps(yy4, yy4, _MM_SHUFFLE(2, 3, 0, 1))); in op_pvq_search_sse2() 183 max2 = _mm_max_ps(max, _mm_shuffle_ps(max, max, _MM_SHUFFLE(1, 0, 3, 2))); in op_pvq_search_sse2() 184 max2 = _mm_max_ps(max2, _mm_shuffle_ps(max2, max2, _MM_SHUFFLE(2, 3, 0, 1))); in op_pvq_search_sse2()
|
/external/mesa3d/src/gallium/drivers/swr/rasterizer/core/ |
D | rasterizer.cpp | 92 __m128 vXa = _mm_shuffle_ps(vX, vX, _MM_SHUFFLE(1, 1, 0, 0)); in RasterizeLine() 93 __m128 vYa = _mm_shuffle_ps(vY, vY, _MM_SHUFFLE(1, 1, 0, 0)); in RasterizeLine() 94 __m128 vZa = _mm_shuffle_ps(vZ, vZ, _MM_SHUFFLE(1, 1, 0, 0)); in RasterizeLine() 95 __m128 vRecipWa = _mm_shuffle_ps(vRecipW, vRecipW, _MM_SHUFFLE(1, 1, 0, 0)); in RasterizeLine() 178 vXa = _mm_shuffle_ps(vX, vX, _MM_SHUFFLE(1, 0, 1, 1)); in RasterizeLine() 179 vYa = _mm_shuffle_ps(vY, vY, _MM_SHUFFLE(1, 0, 1, 1)); in RasterizeLine() 180 vZa = _mm_shuffle_ps(vZ, vZ, _MM_SHUFFLE(1, 0, 1, 1)); in RasterizeLine() 181 vRecipWa = _mm_shuffle_ps(vRecipW, vRecipW, _MM_SHUFFLE(1, 0, 1, 1)); in RasterizeLine()
|
D | frontend.h | 46 __m128 vYsub = _mm_shuffle_ps(vY, vY, _MM_SHUFFLE(3, 0, 2, 1)); in triangleSetupAB() 56 __m128 vXsub = _mm_shuffle_ps(vX, vX, _MM_SHUFFLE(3, 0, 2, 1)); in triangleSetupAB()
|
/external/tensorflow/tensorflow/core/kernels/ |
D | sparse_matmul_op.h | 235 return _mm_set1_ps(_mm_cvtss_f32(_mm_shuffle_ps(a, a, 1))); 241 return _mm_set1_ps(_mm_cvtss_f32(_mm_shuffle_ps(a, a, 2))); 247 return _mm_set1_ps(_mm_cvtss_f32(_mm_shuffle_ps(a, a, 3))); 263 return _mm512_broadcastss_ps(_mm_shuffle_ps(a, a, _MM_SHUFFLE(1, 1, 1, 1))); 269 return _mm512_broadcastss_ps(_mm_shuffle_ps(a, a, _MM_SHUFFLE(2, 2, 2, 2))); 275 return _mm512_broadcastss_ps(_mm_shuffle_ps(a, a, _MM_SHUFFLE(3, 3, 3, 3)));
|
/external/eigen/Eigen/src/Core/arch/SSE/ |
D | PacketMath.h | 74 (_mm_shuffle_ps( (a), (b), ((s)<<6|(r)<<4|(q)<<2|(p)))) 77 …(_mm_castps_si128( (_mm_shuffle_ps( _mm_castsi128_ps(a), _mm_castsi128_ps(b), ((s)<<6|(r)<<4|(q)<<… 382 to[stride*1] = _mm_cvtss_f32(_mm_shuffle_ps(from, from, 1)); 383 to[stride*2] = _mm_cvtss_f32(_mm_shuffle_ps(from, from, 2)); 384 to[stride*3] = _mm_cvtss_f32(_mm_shuffle_ps(from, from, 3)); 436 { return _mm_shuffle_ps(a,a,0x1B); } 543 return pfirst<Packet4f>(_mm_add_ss(tmp, _mm_shuffle_ps(tmp,tmp, 1))); 595 return pfirst<Packet4f>(_mm_mul_ss(tmp, _mm_shuffle_ps(tmp,tmp, 1))); 615 return pfirst<Packet4f>(_mm_min_ss(tmp, _mm_shuffle_ps(tmp,tmp, 1))); 641 return pfirst<Packet4f>(_mm_max_ss(tmp, _mm_shuffle_ps(tmp,tmp, 1))); [all …]
|
D | TypeCasting.h | 55 return _mm_shuffle_ps(_mm_cvtpd_ps(a), _mm_cvtpd_ps(b), (1 << 2) | (1 << 6));
|
D | Complex.h | 125 to[stride*0] = std::complex<float>(_mm_cvtss_f32(_mm_shuffle_ps(from.v, from.v, 0)), 126 _mm_cvtss_f32(_mm_shuffle_ps(from.v, from.v, 1))); 127 to[stride*1] = std::complex<float>(_mm_cvtss_f32(_mm_shuffle_ps(from.v, from.v, 2)), 128 _mm_cvtss_f32(_mm_shuffle_ps(from.v, from.v, 3)));
|
/external/python/cpython3/Modules/_blake2/impl/ |
D | blake2s-load-sse41.h | 20 buf = TOI(_mm_shuffle_ps(TOF(m0), TOF(m1), _MM_SHUFFLE(2,0,2,0))); 23 buf = TOI(_mm_shuffle_ps(TOF(m0), TOF(m1), _MM_SHUFFLE(3,1,3,1))); 26 buf = TOI(_mm_shuffle_ps(TOF(m2), TOF(m3), _MM_SHUFFLE(2,0,2,0))); 29 buf = TOI(_mm_shuffle_ps(TOF(m2), TOF(m3), _MM_SHUFFLE(3,1,3,1)));
|
/external/eigen/Eigen/src/Core/arch/AVX/ |
D | Complex.h | 106 to[stride*0] = std::complex<float>(_mm_cvtss_f32(_mm_shuffle_ps(low, low, 0)), 107 _mm_cvtss_f32(_mm_shuffle_ps(low, low, 1))); 108 to[stride*1] = std::complex<float>(_mm_cvtss_f32(_mm_shuffle_ps(low, low, 2)), 109 _mm_cvtss_f32(_mm_shuffle_ps(low, low, 3))); 112 to[stride*2] = std::complex<float>(_mm_cvtss_f32(_mm_shuffle_ps(high, high, 0)), 113 _mm_cvtss_f32(_mm_shuffle_ps(high, high, 1))); 114 to[stride*3] = std::complex<float>(_mm_cvtss_f32(_mm_shuffle_ps(high, high, 2)), 115 _mm_cvtss_f32(_mm_shuffle_ps(high, high, 3)));
|
D | PacketMath.h | 274 to[stride*1] = _mm_cvtss_f32(_mm_shuffle_ps(low, low, 1)); 275 to[stride*2] = _mm_cvtss_f32(_mm_shuffle_ps(low, low, 2)); 276 to[stride*3] = _mm_cvtss_f32(_mm_shuffle_ps(low, low, 3)); 280 to[stride*5] = _mm_cvtss_f32(_mm_shuffle_ps(high, high, 1)); 281 to[stride*6] = _mm_cvtss_f32(_mm_shuffle_ps(high, high, 2)); 282 to[stride*7] = _mm_cvtss_f32(_mm_shuffle_ps(high, high, 3));
|
/external/speex/libspeexdsp/ |
D | resample_sse.h | 51 sum = _mm_add_ss(sum, _mm_shuffle_ps(sum, sum, 0x55)); in inner_product_single() 69 sum = _mm_add_ss(sum, _mm_shuffle_ps(sum, sum, 0x55)); in interpolate_product_single()
|
/external/webrtc/webrtc/common_audio/resampler/ |
D | sinc_resampler_sse.cc | 53 _mm_store_ss(&result, _mm_add_ss(m_sums2, _mm_shuffle_ps( in Convolve_SSE()
|
/external/webrtc/webrtc/common_audio/ |
D | fir_filter_sse.cc | 73 _mm_store_ss(out + i, _mm_add_ss(m_sum, _mm_shuffle_ps(m_sum, m_sum, 1))); in Filter()
|
/external/skia/include/private/ |
D | SkNx_sse.h | 206 SkNx min = Min(*this, _mm_shuffle_ps(fVec, fVec, _MM_SHUFFLE(2,3,0,1))); in min() 207 min = Min(min, _mm_shuffle_ps(min.fVec, min.fVec, _MM_SHUFFLE(0,1,2,3))); in min() 212 SkNx max = Max(*this, _mm_shuffle_ps(fVec, fVec, _MM_SHUFFLE(2,3,0,1))); in max() 213 max = Max(max, _mm_shuffle_ps(max.fVec, max.fVec, _MM_SHUFFLE(0,1,2,3))); in max()
|
/external/skqp/include/private/ |
D | SkNx_sse.h | 206 SkNx min = Min(*this, _mm_shuffle_ps(fVec, fVec, _MM_SHUFFLE(2,3,0,1))); in min() 207 min = Min(min, _mm_shuffle_ps(min.fVec, min.fVec, _MM_SHUFFLE(0,1,2,3))); in min() 212 SkNx max = Max(*this, _mm_shuffle_ps(fVec, fVec, _MM_SHUFFLE(2,3,0,1))); in max() 213 max = Max(max, _mm_shuffle_ps(max.fVec, max.fVec, _MM_SHUFFLE(0,1,2,3))); in max()
|
/external/libaom/libaom/aom_dsp/simd/ |
D | v128_intrinsics_x86.h | 221 return _mm_castps_si128(_mm_shuffle_ps( in v128_unziphi_32() 226 return _mm_castps_si128(_mm_shuffle_ps( in v128_unziplo_32()
|
/external/clang/test/CodeGen/ |
D | sse-builtins.c | 658 return _mm_shuffle_ps(A, B, 0); in test_mm_shuffle_ps()
|
/external/compiler-rt/lib/tsan/rtl/ |
D | tsan_rtl.cc | 673 #define SHUF(v0, v1, i0, i1, i2, i3) _mm_castps_si128(_mm_shuffle_ps( \
|
/external/clang/lib/Headers/ |
D | xmmintrin.h | 2488 #define _mm_shuffle_ps(a, b, mask) __extension__ ({ \ macro
|
/external/neon_2_sse/ |
D | NEON_2_SSE.h | 6347 hadd128 = _mm_shuffle_ps (hadd128, hadd128, _MM_SHUFFLE(3,1, 2, 0)); //use low 64 bits in vpadd_f32() 11274 …v.val[1] = _mm_castps_si128(_mm_shuffle_ps(_mm_castsi128_ps(tmp2),_mm_castsi128_ps(tmp1), _MM_SHUF… 11278 …v.val[0] = _mm_castps_si128(_mm_shuffle_ps(_mm_castsi128_ps(tmp0),_mm_castsi128_ps(tmp1), _MM_SHUF… 11310 v.val[1] = _mm_shuffle_ps(tmp2,tmp1, _MM_SHUFFLE(1,0,3,2)); //b1,c1,a2,b2, 11312 v.val[2] = _mm_shuffle_ps(v.val[2],v.val[2], _MM_SHUFFLE(3,1,0,2)); //c2,a3,b3,c3 11314 v.val[0] = _mm_shuffle_ps(tmp0,tmp1, _MM_SHUFFLE(3,2,1,0)); //a0,b0,c0,a1, 14479 #define vrev64q_f32(vec) _mm_shuffle_ps (vec, vec, _MM_SHUFFLE(2,3, 0,1)) 15770 a_sh = _mm_shuffle_ps (a, a, _MM_SHUFFLE(3,1, 2, 0)); //a0, a2, a1, a3, need to check endiness in vtrnq_f32() 15771 b_sh = _mm_shuffle_ps (b, b, _MM_SHUFFLE(3,1, 2, 0)); //b0, b2, b1, b3, need to check endiness in vtrnq_f32() 15987 …v32x4.val[0] = _mm_shuffle_ps(a, b, _MM_SHUFFLE(2,0, 2, 0)); //a0, a2, b0, b2 , need to check endi… in vuzpq_f32() [all …]
|