/external/flac/libFLAC/ |
D | lpc_intrin_sse2.c | 71 … q0 = _mm_cvtsi32_si128(0xffff & qlp_coeff[0]); q0 = _mm_shuffle_epi32(q0, _MM_SHUFFLE(0,0,0,0)); in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_sse2() 72 … q1 = _mm_cvtsi32_si128(0xffff & qlp_coeff[1]); q1 = _mm_shuffle_epi32(q1, _MM_SHUFFLE(0,0,0,0)); in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_sse2() 73 … q2 = _mm_cvtsi32_si128(0xffff & qlp_coeff[2]); q2 = _mm_shuffle_epi32(q2, _MM_SHUFFLE(0,0,0,0)); in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_sse2() 74 … q3 = _mm_cvtsi32_si128(0xffff & qlp_coeff[3]); q3 = _mm_shuffle_epi32(q3, _MM_SHUFFLE(0,0,0,0)); in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_sse2() 75 … q4 = _mm_cvtsi32_si128(0xffff & qlp_coeff[4]); q4 = _mm_shuffle_epi32(q4, _MM_SHUFFLE(0,0,0,0)); in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_sse2() 76 … q5 = _mm_cvtsi32_si128(0xffff & qlp_coeff[5]); q5 = _mm_shuffle_epi32(q5, _MM_SHUFFLE(0,0,0,0)); in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_sse2() 77 … q6 = _mm_cvtsi32_si128(0xffff & qlp_coeff[6]); q6 = _mm_shuffle_epi32(q6, _MM_SHUFFLE(0,0,0,0)); in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_sse2() 78 … q7 = _mm_cvtsi32_si128(0xffff & qlp_coeff[7]); q7 = _mm_shuffle_epi32(q7, _MM_SHUFFLE(0,0,0,0)); in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_sse2() 79 … q8 = _mm_cvtsi32_si128(0xffff & qlp_coeff[8]); q8 = _mm_shuffle_epi32(q8, _MM_SHUFFLE(0,0,0,0)); in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_sse2() 80 … q9 = _mm_cvtsi32_si128(0xffff & qlp_coeff[9]); q9 = _mm_shuffle_epi32(q9, _MM_SHUFFLE(0,0,0,0)); in FLAC__lpc_compute_residual_from_qlp_coefficients_16_intrin_sse2() [all …]
|
D | lpc_intrin_sse41.c | 77 xmm0 = _mm_shuffle_epi32(xmm0, _MM_SHUFFLE(3,1,2,0)); // 0 q[1] 0 q[0] in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 78 xmm1 = _mm_shuffle_epi32(xmm1, _MM_SHUFFLE(3,1,2,0)); // 0 q[3] 0 q[2] in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 79 xmm2 = _mm_shuffle_epi32(xmm2, _MM_SHUFFLE(3,1,2,0)); // 0 q[5] 0 q[4] in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 80 xmm3 = _mm_shuffle_epi32(xmm3, _MM_SHUFFLE(3,1,2,0)); // 0 q[7] 0 q[6] in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 81 xmm4 = _mm_shuffle_epi32(xmm4, _MM_SHUFFLE(3,1,2,0)); // 0 q[9] 0 q[8] in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 82 xmm5 = _mm_shuffle_epi32(xmm5, _MM_SHUFFLE(3,1,2,0)); // 0 q[11] 0 q[10] in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 89 xmm7 = _mm_shuffle_epi32(xmm7, _MM_SHUFFLE(2,0,3,1)); // 0 d[i-12] 0 d[i-11] in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 95 xmm6 = _mm_shuffle_epi32(xmm6, _MM_SHUFFLE(2,0,3,1)); in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 102 xmm6 = _mm_shuffle_epi32(xmm6, _MM_SHUFFLE(2,0,3,1)); in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 109 xmm6 = _mm_shuffle_epi32(xmm6, _MM_SHUFFLE(2,0,3,1)); in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() [all …]
|
D | lpc_intrin_sse.c | 83 d0 = _mm_shuffle_ps(d0, d0, _MM_SHUFFLE(2,1,0,3)); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_4_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() 171 d2 = _mm_shuffle_ps(d2, d2, _MM_SHUFFLE(2,1,0,3)); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_12_new() 172 d1 = _mm_shuffle_ps(d1, d1, _MM_SHUFFLE(2,1,0,3)); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_12_new() 173 d0 = _mm_shuffle_ps(d0, d0, _MM_SHUFFLE(2,1,0,3)); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_12_new() 227 d3 = _mm_shuffle_ps(d3, d3, _MM_SHUFFLE(2,1,0,3)); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_16_new() 228 d2 = _mm_shuffle_ps(d2, d2, _MM_SHUFFLE(2,1,0,3)); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_16_new() 229 d1 = _mm_shuffle_ps(d1, d1, _MM_SHUFFLE(2,1,0,3)); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_16_new() 230 d0 = _mm_shuffle_ps(d0, d0, _MM_SHUFFLE(2,1,0,3)); in FLAC__lpc_compute_autocorrelation_intrin_sse_lag_16_new() [all …]
|
D | fixed_intrin_ssse3.c | 71 last_error = _mm_shuffle_epi32(last_error, _MM_SHUFFLE(2,1,0,0)); in FLAC__fixed_compute_best_predictor_intrin_ssse3() 74 last_error = _mm_shuffle_epi32(last_error, _MM_SHUFFLE(2,1,0,0)); in FLAC__fixed_compute_best_predictor_intrin_ssse3() 77 last_error = _mm_shuffle_epi32(last_error, _MM_SHUFFLE(2,1,0,0)); in FLAC__fixed_compute_best_predictor_intrin_ssse3() 84 err1 = _mm_shuffle_epi32(err0, _MM_SHUFFLE(0,0,0,0)); // e0 e0 e0 e0 in FLAC__fixed_compute_best_predictor_intrin_ssse3() 162 last_error = _mm_shuffle_epi32(last_error, _MM_SHUFFLE(2,1,0,0)); in FLAC__fixed_compute_best_predictor_wide_intrin_ssse3() 165 last_error = _mm_shuffle_epi32(last_error, _MM_SHUFFLE(2,1,0,0)); in FLAC__fixed_compute_best_predictor_wide_intrin_ssse3() 168 last_error = _mm_shuffle_epi32(last_error, _MM_SHUFFLE(2,1,0,0)); in FLAC__fixed_compute_best_predictor_wide_intrin_ssse3() 175 err1 = _mm_shuffle_epi32(err0, _MM_SHUFFLE(0,0,0,0)); // e0 e0 e0 e0 in FLAC__fixed_compute_best_predictor_wide_intrin_ssse3()
|
D | fixed_intrin_sse2.c | 71 last_error = _mm_shuffle_epi32(last_error, _MM_SHUFFLE(2,1,0,0)); in FLAC__fixed_compute_best_predictor_intrin_sse2() 74 last_error = _mm_shuffle_epi32(last_error, _MM_SHUFFLE(2,1,0,0)); in FLAC__fixed_compute_best_predictor_intrin_sse2() 77 last_error = _mm_shuffle_epi32(last_error, _MM_SHUFFLE(2,1,0,0)); in FLAC__fixed_compute_best_predictor_intrin_sse2() 84 err1 = _mm_shuffle_epi32(err0, _MM_SHUFFLE(0,0,0,0)); // e0 e0 e0 e0 in FLAC__fixed_compute_best_predictor_intrin_sse2() 168 last_error = _mm_shuffle_epi32(last_error, _MM_SHUFFLE(2,1,0,0)); in FLAC__fixed_compute_best_predictor_wide_intrin_sse2() 171 last_error = _mm_shuffle_epi32(last_error, _MM_SHUFFLE(2,1,0,0)); in FLAC__fixed_compute_best_predictor_wide_intrin_sse2() 174 last_error = _mm_shuffle_epi32(last_error, _MM_SHUFFLE(2,1,0,0)); in FLAC__fixed_compute_best_predictor_wide_intrin_sse2() 181 err1 = _mm_shuffle_epi32(err0, _MM_SHUFFLE(0,0,0,0)); // e0 e0 e0 e0 in FLAC__fixed_compute_best_predictor_wide_intrin_sse2()
|
/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))); 35 buf = _mm_shuffle_epi32(t2, _MM_SHUFFLE(2,1,0,3)); 38 t0 = _mm_shuffle_epi32(m2,_MM_SHUFFLE(0,0,2,0)); \ 41 buf = _mm_shuffle_epi32(t2, _MM_SHUFFLE(2,3,0,1)); 47 buf = _mm_shuffle_epi32(t2, _MM_SHUFFLE(2,3,0,1)); 53 buf = _mm_shuffle_epi32(t2, _MM_SHUFFLE(2,3,0,1)); 59 buf = _mm_shuffle_epi32(t2, _MM_SHUFFLE(3,1,0,2)); [all …]
|
D | blake2s-round.h | 60 row4 = _mm_shuffle_epi32( row4, _MM_SHUFFLE(2,1,0,3) ); \ 61 row3 = _mm_shuffle_epi32( row3, _MM_SHUFFLE(1,0,3,2) ); \ 62 row2 = _mm_shuffle_epi32( row2, _MM_SHUFFLE(0,3,2,1) ); 65 row4 = _mm_shuffle_epi32( row4, _MM_SHUFFLE(0,3,2,1) ); \ 66 row3 = _mm_shuffle_epi32( row3, _MM_SHUFFLE(1,0,3,2) ); \ 67 row2 = _mm_shuffle_epi32( row2, _MM_SHUFFLE(2,1,0,3) );
|
/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() 534 _MM_SHUFFLE(2, 0, 2, 0)); in SmoothedPSD() 536 _MM_SHUFFLE(3, 1, 3, 1)); in SmoothedPSD() 555 _MM_SHUFFLE(2, 0, 2, 0)); in SmoothedPSD() 557 _MM_SHUFFLE(3, 1, 3, 1)); in SmoothedPSD() 633 _MM_SHUFFLE(0, 1, 2, 3)); in WindowDataSSE2() 648 _MM_SHUFFLE(2, 0, 2, 0)); in StoreAsComplexSSE2() [all …]
|
/external/libaom/libaom/aom_dsp/simd/ |
D | v256_intrinsics_x86.h | 143 _mm256_permute4x64_epi64(b, _MM_SHUFFLE(3, 1, 2, 0)), in v256_ziplo_8() 144 _mm256_permute4x64_epi64(a, _MM_SHUFFLE(3, 1, 2, 0))); in v256_ziplo_8() 149 _mm256_permute4x64_epi64(b, _MM_SHUFFLE(3, 1, 2, 0)), in v256_ziphi_8() 150 _mm256_permute4x64_epi64(a, _MM_SHUFFLE(3, 1, 2, 0))); in v256_ziphi_8() 155 _mm256_permute4x64_epi64(b, _MM_SHUFFLE(3, 1, 2, 0)), in v256_ziplo_16() 156 _mm256_permute4x64_epi64(a, _MM_SHUFFLE(3, 1, 2, 0))); in v256_ziplo_16() 161 _mm256_permute4x64_epi64(b, _MM_SHUFFLE(3, 1, 2, 0)), in v256_ziphi_16() 162 _mm256_permute4x64_epi64(a, _MM_SHUFFLE(3, 1, 2, 0))); in v256_ziphi_16() 167 _mm256_permute4x64_epi64(b, _MM_SHUFFLE(3, 1, 2, 0)), in v256_ziplo_32() 168 _mm256_permute4x64_epi64(a, _MM_SHUFFLE(3, 1, 2, 0))); in v256_ziplo_32() [all …]
|
/external/libopus/silk/fixed/x86/ |
D | prefilter_FIX_sse.c | 73 coef_Q13_0123 = _mm_shuffle_epi32( coef_Q13_3210, _MM_SHUFFLE( 0, 1, 2, 3 ) ); in silk_warped_LPC_analysis_filter_FIX_sse4_1() 74 coef_Q13_4567 = _mm_shuffle_epi32( coef_Q13_7654, _MM_SHUFFLE( 0, 1, 2, 3 ) ); in silk_warped_LPC_analysis_filter_FIX_sse4_1() 82 state_0123 = _mm_shuffle_epi32( state_0123, _MM_SHUFFLE( 0, 1, 2, 3 ) ); in silk_warped_LPC_analysis_filter_FIX_sse4_1() 83 state_4567 = _mm_shuffle_epi32( state_4567, _MM_SHUFFLE( 0, 1, 2, 3 ) ); in silk_warped_LPC_analysis_filter_FIX_sse4_1() 94 xmm_tempa = _mm_shuffle_epi32( state_0123, _MM_SHUFFLE( 0, 1, 2, 3 ) ); in silk_warped_LPC_analysis_filter_FIX_sse4_1() 95 xmm_tempb = _mm_shuffle_epi32( state_4567, _MM_SHUFFLE( 0, 1, 2, 3 ) ); in silk_warped_LPC_analysis_filter_FIX_sse4_1() 113 … xmm_tempa = _mm_add_epi32( xmm_tempa, _mm_shuffle_epi32( xmm_tempa, _MM_SHUFFLE( 0, 0, 0, 2 ) ) ); in silk_warped_LPC_analysis_filter_FIX_sse4_1() 126 …oreu_si128( (__m128i *)( &state[ 0 ] ), _mm_shuffle_epi32( state_0123, _MM_SHUFFLE( 0, 1, 2, 3 ) )… in silk_warped_LPC_analysis_filter_FIX_sse4_1() 127 …oreu_si128( (__m128i *)( &state[ 4 ] ), _mm_shuffle_epi32( state_4567, _MM_SHUFFLE( 0, 1, 2, 3 ) )… in silk_warped_LPC_analysis_filter_FIX_sse4_1()
|
D | vector_ops_FIX_sse4_1.c | 68 inVec1_76543210 = _mm_shuffle_epi32( inVec1_76543210, _MM_SHUFFLE( 0, 0, 3, 2 ) ); in silk_inner_prod16_aligned_64_sse4_1() 78 acc2 = _mm_shuffle_epi32( acc1, _MM_SHUFFLE( 0, 0, 3, 2 ) ); in silk_inner_prod16_aligned_64_sse4_1()
|
/external/libopus/celt/x86/ |
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() 126 … pulses_sum = _mm_add_epi32(pulses_sum, _mm_shuffle_epi32(pulses_sum, _MM_SHUFFLE(1, 0, 3, 2))); in op_pvq_search_sse2() 127 … pulses_sum = _mm_add_epi32(pulses_sum, _mm_shuffle_epi32(pulses_sum, _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() [all …]
|
/external/eigen/Eigen/src/Core/arch/AVX/ |
D | PacketMath.h | 232 …mp,_mm256_castps128_ps256(_mm_permute_ps( _mm256_castps256_ps128(tmp), _MM_SHUFFLE(1,0,1,0))), 15); 234 return _mm256_permute_ps(tmp, _MM_SHUFFLE(3,3,2,2)); 414 tmp = _mm256_mul_ps(tmp, _mm256_shuffle_ps(tmp,tmp,_MM_SHUFFLE(1,0,3,2))); 427 tmp = _mm256_min_ps(tmp, _mm256_shuffle_ps(tmp,tmp,_MM_SHUFFLE(1,0,3,2))); 439 tmp = _mm256_max_ps(tmp, _mm256_shuffle_ps(tmp,tmp,_MM_SHUFFLE(1,0,3,2))); 458 Packet8f tmp1 = _mm256_permute_ps (first, _MM_SHUFFLE(0,3,2,1)); 465 Packet8f tmp1 = _mm256_permute_ps (first, _MM_SHUFFLE(1,0,3,2)); 472 Packet8f tmp1 = _mm256_permute_ps (first, _MM_SHUFFLE(2,1,0,3)); 479 Packet8f tmp1 = _mm256_permute_ps (first, _MM_SHUFFLE(3,2,1,0)); 481 first = _mm256_permute_ps(tmp2, _MM_SHUFFLE(3,2,1,0)); [all …]
|
/external/mesa3d/src/gallium/drivers/swr/rasterizer/core/ |
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() 73 __m128i vYsub = _mm_shuffle_epi32(vY, _MM_SHUFFLE(3, 0, 2, 1)); in triangleSetupABInt() 76 __m128i vXsub = _mm_shuffle_epi32(vX, _MM_SHUFFLE(3, 0, 2, 1)); in triangleSetupABInt() 126 __m128i vAShuf = _mm_shuffle_epi32(vA, _MM_SHUFFLE(0, 2, 0, 1)); in calcDeterminantInt() 128 __m128i vBShuf = _mm_shuffle_epi32(vB, _MM_SHUFFLE(0, 1, 0, 2)); in calcDeterminantInt() 134 __m128i vMul2 = _mm_shuffle_epi32(vMul, _MM_SHUFFLE(3, 2, 3, 2)); in calcDeterminantInt() 322 __m128i vX1 = _mm_shuffle_epi32(vX, _MM_SHUFFLE(3, 2, 0, 1)); in calcBoundingBoxInt() 323 __m128i vX2 = _mm_shuffle_epi32(vX, _MM_SHUFFLE(3, 0, 1, 2)); in calcBoundingBoxInt() 325 __m128i vY1 = _mm_shuffle_epi32(vY, _MM_SHUFFLE(3, 2, 0, 1)); in calcBoundingBoxInt() [all …]
|
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()
|
/external/eigen/Eigen/src/Core/arch/AVX512/ |
D | PacketMath.h | 468 _mm256_castps256_ps128(lane0), _MM_SHUFFLE(1, 0, 1, 0))), 472 lane0 = _mm256_permute_ps(lane0, _MM_SHUFFLE(3, 3, 2, 2)); 478 _mm256_castps256_ps128(lane1), _MM_SHUFFLE(1, 0, 1, 0))), 482 lane1 = _mm256_permute_ps(lane1, _MM_SHUFFLE(3, 3, 2, 2)); 939 res = pmul(res, _mm_permute_ps(res, _MM_SHUFFLE(0, 0, 3, 2))); 940 return pfirst(pmul(res, _mm_permute_ps(res, _MM_SHUFFLE(0, 0, 0, 1)))); 947 res = pmul(res, _mm_permute_ps(res, _MM_SHUFFLE(0, 0, 3, 2))); 948 return pfirst(pmul(res, _mm_permute_ps(res, _MM_SHUFFLE(0, 0, 0, 1)))); 967 res = _mm_min_ps(res, _mm_permute_ps(res, _MM_SHUFFLE(0, 0, 3, 2))); 968 return pfirst(_mm_min_ps(res, _mm_permute_ps(res, _MM_SHUFFLE(0, 0, 0, 1)))); [all …]
|
/external/pdfium/third_party/libopenjpeg20/ |
D | mct.c | 235 _MM_SHUFFLE(1, 0, 1, 0)); in opj_mct_encode_real() 245 hi = _mm_shuffle_epi32(r, _MM_SHUFFLE(3, 3, 1, 1)); in opj_mct_encode_real() 255 hi = _mm_shuffle_epi32(g, _MM_SHUFFLE(3, 3, 1, 1)); in opj_mct_encode_real() 265 hi = _mm_shuffle_epi32(b, _MM_SHUFFLE(3, 3, 1, 1)); in opj_mct_encode_real() 279 lo = _mm_cvtepi32_epi64(_mm_shuffle_epi32(b, _MM_SHUFFLE(3, 2, 2, 0))); in opj_mct_encode_real() 280 hi = _mm_cvtepi32_epi64(_mm_shuffle_epi32(b, _MM_SHUFFLE(3, 2, 3, 1))); in opj_mct_encode_real() 290 hi = _mm_shuffle_epi32(r, _MM_SHUFFLE(3, 3, 1, 1)); in opj_mct_encode_real() 300 hi = _mm_shuffle_epi32(g, _MM_SHUFFLE(3, 3, 1, 1)); in opj_mct_encode_real() 314 lo = _mm_cvtepi32_epi64(_mm_shuffle_epi32(r, _MM_SHUFFLE(3, 2, 2, 0))); in opj_mct_encode_real() 315 hi = _mm_cvtepi32_epi64(_mm_shuffle_epi32(r, _MM_SHUFFLE(3, 2, 3, 1))); in opj_mct_encode_real() [all …]
|
/external/libopus/silk/x86/ |
D | NSQ_del_dec_sse4_1.c | 373 …b_sr_Q12_0123 = _mm_shuffle_epi32( b_Q12_0123, _MM_SHUFFLE( 0, 3, 2, 1 ) ); /* equal shift right 4… in silk_noise_shape_quantizer_del_dec_sse4_1() 390 …pred_lag_ptr_tmp = _mm_shuffle_epi32( pred_lag_ptr_tmp, _MM_SHUFFLE( 0, 3, 2, 1 ) );/* equal shift… in silk_noise_shape_quantizer_del_dec_sse4_1() 395 …tmpb = _mm_shuffle_epi32( pred_lag_ptr_tmp, _MM_SHUFFLE( 0, 0, 3, 2 ) );/* equal shift right 8 byt… in silk_noise_shape_quantizer_del_dec_sse4_1() 447 …psLPC_Q14_tmp = _mm_shuffle_epi32( psLPC_Q14_tmp, _MM_SHUFFLE( 0, 3, 2, 1 ) ); /* equal shift righ… in silk_noise_shape_quantizer_del_dec_sse4_1() 448 …a_Q12_tmp = _mm_shuffle_epi32( a_Q12_0123, _MM_SHUFFLE(0, 3, 2, 1 ) ); /* equal shift right 4 byte… in silk_noise_shape_quantizer_del_dec_sse4_1() 460 …psLPC_Q14_tmp = _mm_shuffle_epi32( psLPC_Q14_tmp, _MM_SHUFFLE( 0, 3, 2, 1 ) ); /* equal shift righ… in silk_noise_shape_quantizer_del_dec_sse4_1() 461 …a_Q12_tmp = _mm_shuffle_epi32( a_Q12_4567, _MM_SHUFFLE(0, 3, 2, 1 ) ); /* equal shift right 4 byte… in silk_noise_shape_quantizer_del_dec_sse4_1() 475 …psLPC_Q14_tmp = _mm_shuffle_epi32( psLPC_Q14_tmp, _MM_SHUFFLE( 0, 3, 2, 1 ) ); /* equal shift righ… in silk_noise_shape_quantizer_del_dec_sse4_1() 476 …a_Q12_tmp = _mm_shuffle_epi32( a_Q12_89AB, _MM_SHUFFLE(0, 3, 2, 1 ) );/* equal shift right 4 bytes… in silk_noise_shape_quantizer_del_dec_sse4_1() 488 …psLPC_Q14_tmp = _mm_shuffle_epi32( psLPC_Q14_tmp, _MM_SHUFFLE( 0, 3, 2, 1 ) ); /* equal shift righ… in silk_noise_shape_quantizer_del_dec_sse4_1() [all …]
|
D | VQ_WMat_EC_sse4_1.c | 90 C_tmp1 = _mm_shuffle_epi32( C_tmp1, _MM_SHUFFLE( 0, 3, 2, 1 ) ); /* shift right 4 bytes */ in silk_VQ_WMat_EC_sse4_1() 91 C_tmp3 = _mm_shuffle_epi32( C_tmp3, _MM_SHUFFLE( 0, 3, 2, 1 ) ); /* shift right 4 bytes */ in silk_VQ_WMat_EC_sse4_1() 99 C_tmp5 = _mm_add_epi32( C_tmp5, _mm_shuffle_epi32( C_tmp5, _MM_SHUFFLE( 0, 0, 0, 2 ) ) ); in silk_VQ_WMat_EC_sse4_1()
|
/external/tensorflow/tensorflow/core/kernels/ |
D | sparse_matmul_op.h | 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))); 308 return _mm512_broadcastd_epi32(_mm_shuffle_epi32(a, _MM_SHUFFLE(1, 1, 1, 1))); 314 return _mm512_broadcastd_epi32(_mm_shuffle_epi32(a, _MM_SHUFFLE(2, 2, 2, 2))); 320 return _mm512_broadcastd_epi32(_mm_shuffle_epi32(a, _MM_SHUFFLE(3, 3, 3, 3))); 330 _MM_SHUFFLE(3, 1, 2, 0)));
|
/external/mesa3d/src/gallium/auxiliary/util/ |
D | u_sse.h | 192 a13 = _mm_shuffle_epi32(a, _MM_SHUFFLE(2,3,0,1)); in mm_mullohi_epi32() 193 b13 = _mm_shuffle_epi32(b, _MM_SHUFFLE(2,3,0,1)); in mm_mullohi_epi32() 296 #define SCALAR_EPI32(m, i) _mm_shuffle_epi32((m), _MM_SHUFFLE(i,i,i,i))
|
/external/libvpx/libvpx/vpx_dsp/x86/ |
D | fwd_txfm_sse2.h | 31 __m128i buf0 = _mm_shuffle_epi32(a, _MM_SHUFFLE(0, 0, 2, 0)); in k_packs_epi64() 32 __m128i buf1 = _mm_shuffle_epi32(b, _MM_SHUFFLE(0, 0, 2, 0)); in k_packs_epi64() 152 _mm_shuffle_epi32(reg0_shifted, _MM_SHUFFLE(0, 0, 3, 1)); in k_check_epi32_overflow_4() 154 _mm_shuffle_epi32(reg1_shifted, _MM_SHUFFLE(0, 0, 3, 1)); in k_check_epi32_overflow_4() 156 _mm_shuffle_epi32(reg2_shifted, _MM_SHUFFLE(0, 0, 3, 1)); in k_check_epi32_overflow_4() 158 _mm_shuffle_epi32(reg3_shifted, _MM_SHUFFLE(0, 0, 3, 1)); in k_check_epi32_overflow_4()
|
/external/libaom/libaom/av1/common/x86/ |
D | cfl_avx2.c | 134 top = _mm256_permute4x64_epi64(top, _MM_SHUFFLE(3, 1, 2, 0)); in cfl_luma_subsampling_444_lbd_avx2() 180 hsum = _mm256_permute4x64_epi64(hsum, _MM_SHUFFLE(3, 1, 2, 0)); in cfl_luma_subsampling_420_hbd_avx2() 213 hsum = _mm256_permute4x64_epi64(hsum, _MM_SHUFFLE(3, 1, 2, 0)); in cfl_luma_subsampling_422_hbd_avx2() 266 res = _mm256_permute4x64_epi64(res, _MM_SHUFFLE(3, 1, 2, 0)); in cfl_predict_lbd_avx2() 383 a = _mm256_permute4x64_epi64(a, _MM_SHUFFLE(3, 1, 2, 0)); in fill_sum_epi32()
|
D | cfl_sse2.c | 18 l0 = _mm_add_epi32(l0, _mm_shuffle_epi32(l0, _MM_SHUFFLE(1, 0, 3, 2))); in fill_sum_epi32() 19 return _mm_add_epi32(l0, _mm_shuffle_epi32(l0, _MM_SHUFFLE(2, 3, 0, 1))); in fill_sum_epi32()
|