/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/webrtc/webrtc/modules/audio_processing/aec/ |
D | aec_core_sse2.c | 62 const __m128 f = _mm_add_ps(c, d); in FilterFarSSE2() 63 const __m128 g = _mm_add_ps(y_fft_re, e); in FilterFarSSE2() 64 const __m128 h = _mm_add_ps(y_fft_im, f); in FilterFarSSE2() 101 const __m128 xPowPlus = _mm_add_ps(x_pow_local, k1e_10f); in ScaleErrorSignalSSE2() 106 const __m128 ef_sum2 = _mm_add_ps(ef_re2, ef_im2); in ScaleErrorSignalSSE2() 109 __m128 absEfPlus = _mm_add_ps(absEf, k1e_10f); in ScaleErrorSignalSSE2() 181 const __m128 e = _mm_add_ps(a, b); in FilterAdaptationSSE2() 223 wtBuf_re = _mm_add_ps(wtBuf_re, fft_re); in FilterAdaptationSSE2() 224 wtBuf_im = _mm_add_ps(wtBuf_im, fft_im); in FilterAdaptationSSE2() 295 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/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/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/opencv3/modules/imgproc/src/ |
D | filter.cpp | 961 s0 = _mm_add_ps(_mm_mul_ps(s0, f), d4); in operator ()() 962 s1 = _mm_add_ps(_mm_mul_ps(s1, f), d4); in operator ()() 965 s2 = _mm_add_ps(_mm_mul_ps(s2, f), d4); in operator ()() 966 s3 = _mm_add_ps(_mm_mul_ps(s3, f), d4); in operator ()() 976 s0 = _mm_add_ps(s0, _mm_mul_ps(_mm_cvtepi32_ps(x0), f)); in operator ()() 977 s1 = _mm_add_ps(s1, _mm_mul_ps(_mm_cvtepi32_ps(x1), f)); in operator ()() 980 s2 = _mm_add_ps(s2, _mm_mul_ps(_mm_cvtepi32_ps(x0), f)); in operator ()() 981 s3 = _mm_add_ps(s3, _mm_mul_ps(_mm_cvtepi32_ps(x1), f)); in operator ()() 996 s0 = _mm_add_ps(_mm_mul_ps(s0, f), d4); in operator ()() 1005 s0 = _mm_add_ps(s0, _mm_mul_ps(_mm_cvtepi32_ps(x0), f)); in operator ()() [all …]
|
D | imgwarp.cpp | 568 x0 = _mm_add_ps(_mm_mul_ps(x0, b0), _mm_mul_ps(y0, b1)); in operator ()() 569 x1 = _mm_add_ps(_mm_mul_ps(x1, b0), _mm_mul_ps(y1, b1)); in operator ()() 579 x0 = _mm_add_ps(_mm_mul_ps(x0, b0), _mm_mul_ps(y0, b1)); in operator ()() 580 x1 = _mm_add_ps(_mm_mul_ps(x1, b0), _mm_mul_ps(y1, b1)); in operator ()() 598 x0 = _mm_add_ps(_mm_mul_ps(x0, b0), _mm_mul_ps(y0, b1)); in operator ()() 599 x1 = _mm_add_ps(_mm_mul_ps(x1, b0), _mm_mul_ps(y1, b1)); in operator ()() 609 x0 = _mm_add_ps(_mm_mul_ps(x0, b0), _mm_mul_ps(y0, b1)); in operator ()() 610 x1 = _mm_add_ps(_mm_mul_ps(x1, b0), _mm_mul_ps(y1, b1)); in operator ()() 626 x0 = _mm_add_ps(_mm_mul_ps(x0, b0), _mm_mul_ps(y0, b1)); in operator ()() 663 x0 = _mm_add_ps(_mm_mul_ps(x0, b0), _mm_mul_ps(y0, b1)); in operator ()() [all …]
|
D | pyramids.cpp | 161 r0 = _mm_add_ps(r0, r4); in operator ()() 162 r1 = _mm_add_ps(_mm_add_ps(r1, r3), r2); in operator ()() 163 r0 = _mm_add_ps(r0, _mm_add_ps(r2, r2)); in operator ()() 164 t0 = _mm_add_ps(r0, _mm_mul_ps(r1, _4)); in operator ()() 171 r0 = _mm_add_ps(r0, r4); in operator ()() 172 r1 = _mm_add_ps(_mm_add_ps(r1, r3), r2); in operator ()() 173 r0 = _mm_add_ps(r0, _mm_add_ps(r2, r2)); in operator ()() 174 t1 = _mm_add_ps(r0, _mm_mul_ps(r1, _4)); in operator ()() 503 _mm_storeu_ps(dst1 + x, _mm_mul_ps(v_scale4, _mm_add_ps(v_r1, v_r2))); in operator ()() 504 …_mm_storeu_ps(dst0 + x, _mm_mul_ps(v_scale, _mm_add_ps(_mm_add_ps(v_r0, _mm_mul_ps(v_6, v_r1)), v_… in operator ()() [all …]
|
/external/vulkan-validation-layers/libs/glm/detail/ |
D | intrinsic_matrix.inl | 52 out[0] = _mm_add_ps(in1[0], in2[0]); 53 out[1] = _mm_add_ps(in1[1], in2[1]); 54 out[2] = _mm_add_ps(in1[2], in2[2]); 55 out[3] = _mm_add_ps(in1[3], in2[3]); 81 __m128 a0 = _mm_add_ps(m0, m1); 82 __m128 a1 = _mm_add_ps(m2, m3); 83 __m128 a2 = _mm_add_ps(a0, a1); 102 __m128 a0 = _mm_add_ps(u0, u1); 106 __m128 a1 = _mm_add_ps(u2, u3); 110 __m128 f2 = _mm_add_ps(f0, f1); [all …]
|
D | intrinsic_geometric.inl | 56 __m128 add0 = _mm_add_ps(mul0, swp0); 58 __m128 add1 = _mm_add_ps(add0, swp1); 68 __m128 add0 = _mm_add_ps(mov0, mul0); 137 __m128 add0 = _mm_add_ps(mul3, sqt0);
|
/external/vulkan-validation-layers/libs/glm/gtx/ |
D | simd_quat.inl | 107 return fquatSIMD(_mm_add_ps(q1.Data, q2.Data)); 135 __m128 add0 = _mm_add_ps(mul0, _mm_movehl_ps(mul0, mul0)); 139 __m128 add1 = _mm_add_ps(mul1, _mm_movehl_ps(mul1, mul1)); 143 __m128 add2 = _mm_add_ps(mul2, _mm_movehl_ps(mul2, mul2)); 147 __m128 add3 = _mm_add_ps(mul3, _mm_movehl_ps(mul3, mul3)); 192 return _mm_add_ps(v.Data, _mm_add_ps(uv, uuv)); 336 __m128 _xyzw2 = _mm_add_ps(_xyzw, _xyzw); 344 _tmp1 = _mm_add_ps(_tmp1, _mm_mul_ps(_zxyw2, _wwww)); 428 return _mm_add_ps(x.Data, _mm_mul_ps(_mm_set1_ps(a), _mm_sub_ps(y.Data, x.Data))); 454 return _mm_add_ps(x.Data, _mm_mul_ps(_mm_set1_ps(a), _mm_sub_ps(y.Data, x.Data))); [all …]
|
D | simd_mat4.inl | 132 this->Data[0].Data = _mm_add_ps(this->Data[0].Data, m[0].Data); 133 this->Data[1].Data = _mm_add_ps(this->Data[1].Data, m[1].Data); 134 this->Data[2].Data = _mm_add_ps(this->Data[2].Data, m[2].Data); 135 this->Data[3].Data = _mm_add_ps(this->Data[3].Data, m[3].Data); 178 this->Data[0].Data = _mm_add_ps(this->Data[0].Data, Operand); 179 this->Data[1].Data = _mm_add_ps(this->Data[1].Data, Operand); 180 this->Data[2].Data = _mm_add_ps(this->Data[2].Data, Operand); 181 this->Data[3].Data = _mm_add_ps(this->Data[3].Data, Operand); 226 this->Data[0].Data = _mm_add_ps(this->Data[0].Data, one); 227 this->Data[1].Data = _mm_add_ps(this->Data[1].Data, one); [all …]
|
D | simd_vec4.inl | 103 this->Data = _mm_add_ps(Data, _mm_set_ps1(s)); 109 this->Data = _mm_add_ps(this->Data , v.Data); 151 this->Data = _mm_add_ps(this->Data , glm::detail::one); 185 return fvec4SIMD(_mm_add_ps(v.Data, _mm_set1_ps(s))); 190 return fvec4SIMD(_mm_add_ps(_mm_set1_ps(s), v.Data)); 195 return fvec4SIMD(_mm_add_ps(v1.Data, v2.Data)); 262 return fvec4SIMD(_mm_add_ps(v.Data, glm::detail::one)); 455 return _mm_add_ps(x.Data, Mul0); 531 return _mm_add_ps(_mm_mul_ps(a.Data, b.Data), c.Data);
|
/external/webrtc/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/opencv3/modules/hal/src/ |
D | stat.cpp | 195 d0 = _mm_add_ps(d0, _mm_mul_ps(t0, t0)); in normL2Sqr_() 196 d1 = _mm_add_ps(d1, _mm_mul_ps(t1, t1)); in normL2Sqr_() 198 _mm_store_ps(buf, _mm_add_ps(d0, d1)); in normL2Sqr_() 231 d0 = _mm_add_ps(d0, _mm_and_ps(t0, absmask)); in normL1_() 232 d1 = _mm_add_ps(d1, _mm_and_ps(t1, absmask)); in normL1_() 234 _mm_store_ps(buf, _mm_add_ps(d0, d1)); in normL1_()
|
D | mathfuncs.cpp | 89 __m128 c = _mm_div_ps(tmin, _mm_add_ps(tmax, eps)); in fastAtan2() 92 a = _mm_mul_ps(_mm_add_ps(a, p5), c2); in fastAtan2() 93 a = _mm_mul_ps(_mm_add_ps(a, p3), c2); in fastAtan2() 94 a = _mm_mul_ps(_mm_add_ps(a, p1), c); in fastAtan2() 536 __m128 zf0 = _mm_add_ps(xf0, mA1); in exp() 537 __m128 zf1 = _mm_add_ps(xf1, mA1); in exp() 539 zf0 = _mm_add_ps(_mm_mul_ps(zf0, xf0), mA2); in exp() 540 zf1 = _mm_add_ps(_mm_mul_ps(zf1, xf1), mA2); in exp() 542 zf0 = _mm_add_ps(_mm_mul_ps(zf0, xf0), mA3); in exp() 543 zf1 = _mm_add_ps(_mm_mul_ps(zf1, xf1), mA3); in exp() [all …]
|
/external/opencv3/modules/core/src/ |
D | convert.cpp | 1483 …__m128 v_dst1 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps(_mm_unpacklo_epi16(v_src12, v_zero_i)), v_sc… in operator ()() 1485 …__m128 v_dst2 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps(_mm_unpackhi_epi16(v_src12, v_zero_i)), v_sc… in operator ()() 1487 …__m128 v_dst3 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps(_mm_unpacklo_epi16(v_src_34, v_zero_i)), v_s… in operator ()() 1489 …__m128 v_dst4 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps(_mm_unpackhi_epi16(v_src_34, v_zero_i)), v_s… in operator ()() 1521 __m128 v_dst1 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps( in operator ()() 1524 __m128 v_dst2 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps( in operator ()() 1527 __m128 v_dst3 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps( in operator ()() 1530 __m128 v_dst4 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps( in operator ()() 1561 …__m128 v_dst1 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps(_mm_unpacklo_epi16(v_src, v_zero_i)), v_scal… in operator ()() 1563 …__m128 v_dst2 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps(_mm_unpackhi_epi16(v_src, v_zero_i)), v_scal… in operator ()() [all …]
|
/external/opencv3/modules/video/src/ |
D | optflowgf.cpp | 478 x0 = _mm_add_ps(_mm_loadu_ps(sptr0 + x), _mm_loadu_ps(sptr1 + x)); in FarnebackUpdateFlow_GaussianBlur() 479 x1 = _mm_add_ps(_mm_loadu_ps(sptr0 + x + 4), _mm_loadu_ps(sptr1 + x + 4)); in FarnebackUpdateFlow_GaussianBlur() 480 s0 = _mm_add_ps(s0, _mm_mul_ps(x0, g4)); in FarnebackUpdateFlow_GaussianBlur() 481 s1 = _mm_add_ps(s1, _mm_mul_ps(x1, g4)); in FarnebackUpdateFlow_GaussianBlur() 482 x0 = _mm_add_ps(_mm_loadu_ps(sptr0 + x + 8), _mm_loadu_ps(sptr1 + x + 8)); in FarnebackUpdateFlow_GaussianBlur() 483 x1 = _mm_add_ps(_mm_loadu_ps(sptr0 + x + 12), _mm_loadu_ps(sptr1 + x + 12)); in FarnebackUpdateFlow_GaussianBlur() 484 s2 = _mm_add_ps(s2, _mm_mul_ps(x0, g4)); in FarnebackUpdateFlow_GaussianBlur() 485 s3 = _mm_add_ps(s3, _mm_mul_ps(x1, g4)); in FarnebackUpdateFlow_GaussianBlur() 504 __m128 x0 = _mm_add_ps(_mm_loadu_ps(sptr0 + x), _mm_loadu_ps(sptr1 + x)); in FarnebackUpdateFlow_GaussianBlur() 505 s0 = _mm_add_ps(s0, _mm_mul_ps(x0, g4)); in FarnebackUpdateFlow_GaussianBlur() [all …]
|
/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); } 488 Packet4f tmp = _mm_add_ps(a, _mm_movehl_ps(a,a)); 502 tmp0 = _mm_add_ps(tmp0, tmp1); 504 tmp1 = _mm_add_ps(tmp1, tmp2); 507 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/pdfium/third_party/libopenjpeg20/ |
D | mct.c | 393 vr = _mm_add_ps(vy, _mm_mul_ps(vv, vrv)); in opj_mct_decode_real() 395 vb = _mm_add_ps(vy, _mm_mul_ps(vu, vbu)); in opj_mct_decode_real() 406 vr = _mm_add_ps(vy, _mm_mul_ps(vv, vrv)); in opj_mct_decode_real() 408 vb = _mm_add_ps(vy, _mm_mul_ps(vu, vbu)); in opj_mct_decode_real()
|