/external/libopus/celt/x86/ |
D | pitch_sse.h | 52 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(_mm_shuffle_ps(x0,x0,0x00),yj)); in xcorr_kernel() 53 xsum2 = _mm_add_ps(xsum2,_mm_mul_ps(_mm_shuffle_ps(x0,x0,0x55), in xcorr_kernel() 55 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(_mm_shuffle_ps(x0,x0,0xaa), in xcorr_kernel() 57 xsum2 = _mm_add_ps(xsum2,_mm_mul_ps(_mm_shuffle_ps(x0,x0,0xff),y3)); in xcorr_kernel() 61 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(_mm_load1_ps(x+j),_mm_loadu_ps(y+j))); in xcorr_kernel() 64 xsum2 = _mm_add_ps(xsum2,_mm_mul_ps(_mm_load1_ps(x+j),_mm_loadu_ps(y+j))); in xcorr_kernel() 67 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(_mm_load1_ps(x+j),_mm_loadu_ps(y+j))); in xcorr_kernel() 71 _mm_storeu_ps(sum,_mm_add_ps(xsum1,xsum2)); in xcorr_kernel() 87 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(xi, y1i)); in dual_inner_prod() 88 xsum2 = _mm_add_ps(xsum2,_mm_mul_ps(xi, y2i)); in dual_inner_prod() [all …]
|
/external/chromium_org/third_party/opus/src/celt/x86/ |
D | pitch_sse.h | 52 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(_mm_shuffle_ps(x0,x0,0x00),yj)); in xcorr_kernel() 53 xsum2 = _mm_add_ps(xsum2,_mm_mul_ps(_mm_shuffle_ps(x0,x0,0x55), in xcorr_kernel() 55 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(_mm_shuffle_ps(x0,x0,0xaa), in xcorr_kernel() 57 xsum2 = _mm_add_ps(xsum2,_mm_mul_ps(_mm_shuffle_ps(x0,x0,0xff),y3)); in xcorr_kernel() 61 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(_mm_load1_ps(x+j),_mm_loadu_ps(y+j))); in xcorr_kernel() 64 xsum2 = _mm_add_ps(xsum2,_mm_mul_ps(_mm_load1_ps(x+j),_mm_loadu_ps(y+j))); in xcorr_kernel() 67 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(_mm_load1_ps(x+j),_mm_loadu_ps(y+j))); in xcorr_kernel() 71 _mm_storeu_ps(sum,_mm_add_ps(xsum1,xsum2)); in xcorr_kernel() 87 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(xi, y1i)); in dual_inner_prod() 88 xsum2 = _mm_add_ps(xsum2,_mm_mul_ps(xi, y2i)); in dual_inner_prod() [all …]
|
/external/chromium_org/third_party/webrtc/common_audio/resampler/ |
D | sinc_resampler_sse.cc | 32 m_sums1 = _mm_add_ps(m_sums1, _mm_mul_ps(m_input, _mm_load_ps(k1 + i))); in Convolve_SSE() 33 m_sums2 = _mm_add_ps(m_sums2, _mm_mul_ps(m_input, _mm_load_ps(k2 + i))); in Convolve_SSE() 38 m_sums1 = _mm_add_ps(m_sums1, _mm_mul_ps(m_input, _mm_load_ps(k1 + i))); in Convolve_SSE() 39 m_sums2 = _mm_add_ps(m_sums2, _mm_mul_ps(m_input, _mm_load_ps(k2 + i))); in Convolve_SSE() 48 m_sums1 = _mm_add_ps(m_sums1, m_sums2); in Convolve_SSE() 52 m_sums2 = _mm_add_ps(_mm_movehl_ps(m_sums1, m_sums1), m_sums1); in Convolve_SSE()
|
/external/chromium_org/third_party/webrtc/modules/audio_processing/aec/ |
D | aec_core_sse2.c | 57 const __m128 f = _mm_add_ps(c, d); in FilterFarSSE2() 58 const __m128 g = _mm_add_ps(yf_re, e); in FilterFarSSE2() 59 const __m128 h = _mm_add_ps(yf_im, f); in FilterFarSSE2() 92 const __m128 xPowPlus = _mm_add_ps(xPow, k1e_10f); in ScaleErrorSignalSSE2() 97 const __m128 ef_sum2 = _mm_add_ps(ef_re2, ef_im2); in ScaleErrorSignalSSE2() 100 __m128 absEfPlus = _mm_add_ps(absEf, k1e_10f); in ScaleErrorSignalSSE2() 169 const __m128 e = _mm_add_ps(a, b); in FilterAdaptationSSE2() 211 wtBuf_re = _mm_add_ps(wtBuf_re, fft_re); in FilterAdaptationSSE2() 212 wtBuf_im = _mm_add_ps(wtBuf_im, fft_im); in FilterAdaptationSSE2() 283 const __m128 pol5_y_1 = _mm_add_ps(pol5_y_0, *((__m128*)C4)); in mm_pow_ps() [all …]
|
D | aec_rdft_sse2.c | 38 __m128 x0v = _mm_add_ps(a01v, a23v); in cft1st_128_SSE2() 40 const __m128 x2v = _mm_add_ps(a45v, a67v); in cft1st_128_SSE2() 43 a01v = _mm_add_ps(x0v, x2v); in cft1st_128_SSE2() 49 a45v = _mm_add_ps(a45_0v, a45_1v); in cft1st_128_SSE2() 55 x0v = _mm_add_ps(x1v, x3s); in cft1st_128_SSE2() 59 a23v = _mm_add_ps(a23_0v, a23_1v); in cft1st_128_SSE2() 67 a67v = _mm_add_ps(a67_0v, a67_1v); in cft1st_128_SSE2() 98 __m128 x0r0_0i0_0r1_x0i1 = _mm_add_ps(a_00_32, a_08_40); in cftmdl_128_SSE2() 111 const __m128 x2r0_2i0_2r1_x2i1 = _mm_add_ps(a_16_48, a_24_56); in cftmdl_128_SSE2() 114 const __m128 xx0 = _mm_add_ps(x0r0_0i0_0r1_x0i1, x2r0_2i0_2r1_x2i1); in cftmdl_128_SSE2() [all …]
|
/external/webrtc/src/modules/audio_processing/aec/ |
D | aec_rdft_sse2.c | 41 __m128 x0v = _mm_add_ps(a01v, a23v); in cft1st_128_SSE2() 43 const __m128 x2v = _mm_add_ps(a45v, a67v); in cft1st_128_SSE2() 46 a01v = _mm_add_ps(x0v, x2v); in cft1st_128_SSE2() 52 a45v = _mm_add_ps(a45_0v, a45_1v); in cft1st_128_SSE2() 58 x0v = _mm_add_ps(x1v, x3s); in cft1st_128_SSE2() 62 a23v = _mm_add_ps(a23_0v, a23_1v); in cft1st_128_SSE2() 70 a67v = _mm_add_ps(a67_0v, a67_1v); in cft1st_128_SSE2() 101 __m128 x0r0_0i0_0r1_x0i1 = _mm_add_ps(a_00_32, a_08_40); in cftmdl_128_SSE2() 114 const __m128 x2r0_2i0_2r1_x2i1 = _mm_add_ps(a_16_48, a_24_56); in cftmdl_128_SSE2() 117 const __m128 xx0 = _mm_add_ps(x0r0_0i0_0r1_x0i1, x2r0_2i0_2r1_x2i1); in cftmdl_128_SSE2() [all …]
|
D | aec_core_sse2.c | 59 const __m128 f = _mm_add_ps(c, d); in FilterFarSSE2() 60 const __m128 g = _mm_add_ps(yf_re, e); in FilterFarSSE2() 61 const __m128 h = _mm_add_ps(yf_im, f); in FilterFarSSE2() 88 const __m128 xPowPlus = _mm_add_ps(xPow, k1e_10f); in ScaleErrorSignalSSE2() 93 const __m128 ef_sum2 = _mm_add_ps(ef_re2, ef_im2); in ScaleErrorSignalSSE2() 96 __m128 absEfPlus = _mm_add_ps(absEf, k1e_10f); in ScaleErrorSignalSSE2() 155 const __m128 e = _mm_add_ps(a, b); in FilterAdaptationSSE2() 194 wtBuf_re = _mm_add_ps(wtBuf_re, fft_re); in FilterAdaptationSSE2() 195 wtBuf_im = _mm_add_ps(wtBuf_im, fft_im); in FilterAdaptationSSE2() 267 const __m128 pol5_y_1 = _mm_add_ps(pol5_y_0, *((__m128 *)C4)); in mm_pow_ps() [all …]
|
/external/speex/libspeex/ |
D | ltp_sse.h | 45 sum = _mm_add_ps(sum, _mm_mul_ps(_mm_loadu_ps(a+0), _mm_loadu_ps(b+0))); in inner_prod() 46 sum = _mm_add_ps(sum, _mm_mul_ps(_mm_loadu_ps(a+4), _mm_loadu_ps(b+4))); in inner_prod() 50 sum = _mm_add_ps(sum, _mm_movehl_ps(sum, sum)); in inner_prod() 82 sum = _mm_add_ps(sum, _mm_mul_ps(xx[0], yy[0])); in pitch_xcorr() 83 sum = _mm_add_ps(sum, _mm_mul_ps(xx[1], yy[1])); in pitch_xcorr() 87 sum = _mm_add_ps(sum, _mm_movehl_ps(sum, sum)); in pitch_xcorr()
|
D | resample_sse.h | 47 sum = _mm_add_ps(sum, _mm_mul_ps(_mm_loadu_ps(a+i), _mm_loadu_ps(b+i))); in inner_product_single() 48 sum = _mm_add_ps(sum, _mm_mul_ps(_mm_loadu_ps(a+i+4), _mm_loadu_ps(b+i+4))); in inner_product_single() 50 sum = _mm_add_ps(sum, _mm_movehl_ps(sum, sum)); in inner_product_single() 64 sum = _mm_add_ps(sum, _mm_mul_ps(_mm_load1_ps(a+i), _mm_loadu_ps(b+i*oversample))); in interpolate_product_single() 65 sum = _mm_add_ps(sum, _mm_mul_ps(_mm_load1_ps(a+i+1), _mm_loadu_ps(b+(i+1)*oversample))); in interpolate_product_single() 68 sum = _mm_add_ps(sum, _mm_movehl_ps(sum, sum)); in interpolate_product_single()
|
D | filters_sse.h | 68 mem[0] = _mm_add_ps(mem[0], _mm_mul_ps(xx, num[0])); in filter_mem16_10() 74 mem[1] = _mm_add_ps(mem[1], _mm_mul_ps(xx, num[1])); in filter_mem16_10() 79 mem[2] = _mm_add_ps(mem[2], _mm_mul_ps(xx, num[2])); in filter_mem16_10() 118 mem[0] = _mm_add_ps(mem[0], _mm_mul_ps(xx, num[0])); in filter_mem16_8() 124 mem[1] = _mm_add_ps(mem[1], _mm_mul_ps(xx, num[1])); in filter_mem16_8() 271 mem[0] = _mm_add_ps(mem[0], _mm_mul_ps(xx, num[0])); in fir_mem16_10() 276 mem[1] = _mm_add_ps(mem[1], _mm_mul_ps(xx, num[1])); in fir_mem16_10() 280 mem[2] = _mm_add_ps(mem[2], _mm_mul_ps(xx, num[2])); in fir_mem16_10() 317 mem[0] = _mm_add_ps(mem[0], _mm_mul_ps(xx, num[0])); in fir_mem16_8() 322 mem[1] = _mm_add_ps(mem[1], _mm_mul_ps(xx, num[1])); in fir_mem16_8()
|
D | cb_search_sse.h | 77 resj = _mm_add_ps(resj, _mm_mul_ps(shape[k],r[j-k])); in compute_weighted_codebook() 80 EE = _mm_add_ps(EE, _mm_mul_ps(resj, resj)); in compute_weighted_codebook()
|
/external/chromium_org/third_party/openmax_dl/dl/sp/src/x86/ |
D | x86SP_SSE_Math.h | 42 out->imag = _mm_add_ps(_mm_mul_ps(a->real, b->imag), in VC_MUL() 48 out->real = _mm_add_ps(_mm_mul_ps(a->real, b->real), in VC_CONJ_MUL() 62 out->real = _mm_add_ps(a->real, b->real); in VC_ADD() 63 out->imag = _mm_add_ps(a->imag, b->imag); in VC_ADD() 71 out->real = _mm_add_ps(a->real, b->imag); in VC_ADD_X() 72 out->imag = _mm_add_ps(b->real, a->imag); in VC_ADD_X() 81 _mm_store_ps(out, _mm_add_ps(a->real, b->real)); in VC_ADD_STORE_SPLIT() 82 _mm_store_ps(out + offset, _mm_add_ps(a->imag, b->imag)); in VC_ADD_STORE_SPLIT() 115 out->real = _mm_add_ps(a->real, b->real); in VC_ADD_SUB() 124 out->real = _mm_add_ps(a->real, b->imag); in VC_ADD_SUB_X() [all …]
|
/external/chromium_org/third_party/qcms/src/ |
D | transform-sse2.c | 105 vec_r = _mm_add_ps(vec_r, _mm_add_ps(vec_g, vec_b)); in qcms_transform_data_rgb_out_lut_sse2() 136 vec_r = _mm_add_ps(vec_r, _mm_add_ps(vec_g, vec_b)); in qcms_transform_data_rgb_out_lut_sse2() 225 vec_r = _mm_add_ps(vec_r, _mm_add_ps(vec_g, vec_b)); in qcms_transform_data_rgba_out_lut_sse2() 258 vec_r = _mm_add_ps(vec_r, _mm_add_ps(vec_g, vec_b)); in qcms_transform_data_rgba_out_lut_sse2()
|
D | transform-sse1.c | 105 vec_r = _mm_add_ps(vec_r, _mm_add_ps(vec_g, vec_b)); in qcms_transform_data_rgb_out_lut_sse1() 138 vec_r = _mm_add_ps(vec_r, _mm_add_ps(vec_g, vec_b)); in qcms_transform_data_rgb_out_lut_sse1() 231 vec_r = _mm_add_ps(vec_r, _mm_add_ps(vec_g, vec_b)); in qcms_transform_data_rgba_out_lut_sse1() 266 vec_r = _mm_add_ps(vec_r, _mm_add_ps(vec_g, vec_b)); in qcms_transform_data_rgba_out_lut_sse1()
|
/external/chromium_org/third_party/webrtc/common_audio/ |
D | fir_filter_sse.cc | 64 m_sum = _mm_add_ps(m_sum, _mm_mul_ps(m_in, _mm_load_ps(coef_ptr + j))); in Filter() 69 m_sum = _mm_add_ps(m_sum, _mm_mul_ps(m_in, _mm_load_ps(coef_ptr + j))); in Filter() 72 m_sum = _mm_add_ps(_mm_movehl_ps(m_sum, m_sum), m_sum); in Filter()
|
/external/chromium_org/media/base/ |
D | sinc_resampler.cc | 343 m_sums1 = _mm_add_ps(m_sums1, _mm_mul_ps(m_input, _mm_load_ps(k1 + i))); in Convolve_SSE() 344 m_sums2 = _mm_add_ps(m_sums2, _mm_mul_ps(m_input, _mm_load_ps(k2 + i))); in Convolve_SSE() 349 m_sums1 = _mm_add_ps(m_sums1, _mm_mul_ps(m_input, _mm_load_ps(k1 + i))); in Convolve_SSE() 350 m_sums2 = _mm_add_ps(m_sums2, _mm_mul_ps(m_input, _mm_load_ps(k2 + i))); in Convolve_SSE() 357 m_sums1 = _mm_add_ps(m_sums1, m_sums2); in Convolve_SSE() 361 m_sums2 = _mm_add_ps(_mm_movehl_ps(m_sums1, m_sums1), m_sums1); in Convolve_SSE()
|
D | vector_math.cc | 103 _mm_store_ps(dest + i, _mm_add_ps(_mm_load_ps(dest + i), in FMAC_SSE() 157 ewma_x4 = _mm_add_ps(ewma_x4, in EWMAAndMaxPower_SSE()
|
/external/chromium_org/third_party/WebKit/Source/platform/audio/ |
D | VectorMath.cpp | 157 dest = _mm_add_ps(dest, temp); \ 308 *pDest = _mm_add_ps(*pSource1, *pSource2); 319 dest = _mm_add_ps(*pSource1, *pSource2); 332 *pDest = _mm_add_ps(*pSource1, source2); 342 dest = _mm_add_ps(*pSource1, source2); 486 imag = _mm_add_ps(imag, _mm_mul_ps(imag1, real2)); 544 mSum = _mm_add_ps(mSum, source);
|
D | SincResampler.cpp | 292 sums1 = _mm_add_ps(sums1, mul1); \ in process() 293 sums2 = _mm_add_ps(sums2, mul2); \ in process()
|
/external/pdfium/core/src/fxcodec/fx_libopenjpeg/libopenjpeg20/ |
D | mct.c | 168 vr = _mm_add_ps(vy, _mm_mul_ps(vv, vrv)); in opj_mct_decode_real() 170 vb = _mm_add_ps(vy, _mm_mul_ps(vu, vbu)); in opj_mct_decode_real() 181 vr = _mm_add_ps(vy, _mm_mul_ps(vv, vrv)); in opj_mct_decode_real() 183 vb = _mm_add_ps(vy, _mm_mul_ps(vu, vbu)); in opj_mct_decode_real()
|
/external/eigen/Eigen/src/Core/arch/SSE/ |
D | Complex.h | 48 … padd<Packet2cf>(const Packet2cf& a, const Packet2cf& b) { return Packet2cf(_mm_add_ps(a.v,b.v)); } 73 return Packet2cf(_mm_add_ps(_mm_mul_ps(vec4f_swizzle1(a.v, 0, 0, 2, 2), b.v), 131 return pfirst(Packet2cf(_mm_add_ps(a.v, _mm_movehl_ps(a.v,a.v)))); 136 …return Packet2cf(_mm_add_ps(_mm_movelh_ps(vecs[0].v,vecs[1].v), _mm_movehl_ps(vecs[1].v,vecs[0].v)… 168 return Packet2cf(_mm_add_ps(_mm_xor_ps(_mm_mul_ps(vec4f_swizzle1(a.v, 0, 0, 2, 2), b.v), mask), 186 return Packet2cf(_mm_add_ps(_mm_mul_ps(vec4f_swizzle1(a.v, 0, 0, 2, 2), b.v), 234 …return Packet2cf(_mm_div_ps(res.v,_mm_add_ps(s,_mm_castsi128_ps(_mm_shuffle_epi32(_mm_castps_si128…
|
D | PacketMath.h | 118 template<> EIGEN_STRONG_INLINE Packet4f plset<float>(const float& a) { return _mm_add_ps(pset1<Pack… 122 …NG_INLINE Packet4f padd<Packet4f>(const Packet4f& a, const Packet4f& b) { return _mm_add_ps(a,b); } 425 Packet4f tmp = _mm_add_ps(a, _mm_movehl_ps(a,a)); 439 tmp0 = _mm_add_ps(tmp0, tmp1); 441 tmp1 = _mm_add_ps(tmp1, tmp2); 444 return _mm_add_ps(tmp0, tmp2);
|
/external/eigen/Eigen/src/Geometry/arch/ |
D | Geometry_SSE.h | 32 _mm_add_ps(_mm_sub_ps(_mm_mul_ps(a,vec4f_swizzle1(b,3,3,3,3)), 35 _mm_add_ps(flip1,flip2)));
|
/external/skia/src/opts/ |
D | SkBitmapFilter_opts_SSE2.cpp | 88 accum = _mm_add_ps(accum, _mm_mul_ps(cfloat, weightVector)); in highQualityFilter_SSE2() 89 weight = _mm_add_ps( weight, weightVector ); in highQualityFilter_SSE2() 94 accum = _mm_add_ps(accum, _mm_set1_ps(0.5f)); in highQualityFilter_SSE2() 154 accum = _mm_add_ps(accum, _mm_mul_ps(c, weightVector)); in highQualityFilter_ScaleOnly_SSE2() 161 accum = _mm_add_ps(accum, _mm_set1_ps(0.5f)); in highQualityFilter_ScaleOnly_SSE2()
|
/external/eigen/Eigen/src/LU/arch/ |
D | Inverse_SSE.h | 107 iD = _mm_add_ps(iD,_mm_mul_ps(_mm_shuffle_ps(C,C,0xF5), _mm_movehl_ps(AB,AB))); 110 iA = _mm_add_ps(iA,_mm_mul_ps(_mm_shuffle_ps(B,B,0xF5), _mm_movehl_ps(DC,DC))); 113 d = _mm_add_ps(d, _mm_movehl_ps(d, d));
|