/external/webrtc/webrtc/common_audio/resampler/ |
H A D | sinc_resampler_sse.cc | 32 m_sums1 = _mm_add_ps(m_sums1, _mm_mul_ps(m_input, _mm_load_ps(k1 + i))); 33 m_sums2 = _mm_add_ps(m_sums2, _mm_mul_ps(m_input, _mm_load_ps(k2 + i))); 38 m_sums1 = _mm_add_ps(m_sums1, _mm_mul_ps(m_input, _mm_load_ps(k1 + i))); 39 m_sums2 = _mm_add_ps(m_sums2, _mm_mul_ps(m_input, _mm_load_ps(k2 + i))); 48 m_sums1 = _mm_add_ps(m_sums1, m_sums2); 52 m_sums2 = _mm_add_ps(_mm_movehl_ps(m_sums1, m_sums1), m_sums1);
|
/external/libopus/celt/x86/ |
H A D | pitch_sse.h | 52 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(_mm_shuffle_ps(x0,x0,0x00),yj)); 53 xsum2 = _mm_add_ps(xsum2,_mm_mul_ps(_mm_shuffle_ps(x0,x0,0x55), 55 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(_mm_shuffle_ps(x0,x0,0xaa), 57 xsum2 = _mm_add_ps(xsum2,_mm_mul_ps(_mm_shuffle_ps(x0,x0,0xff),y3)); 61 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(_mm_load1_ps(x+j),_mm_loadu_ps(y+j))); 64 xsum2 = _mm_add_ps(xsum2,_mm_mul_ps(_mm_load1_ps(x+j),_mm_loadu_ps(y+j))); 67 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(_mm_load1_ps(x+j),_mm_loadu_ps(y+j))); 71 _mm_storeu_ps(sum,_mm_add_ps(xsum1,xsum2)); 87 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(xi, y1i)); 88 xsum2 = _mm_add_ps(xsum [all...] |
/external/speex/libspeex/ |
H A D | ltp_sse.h | 45 sum = _mm_add_ps(sum, _mm_mul_ps(_mm_loadu_ps(a+0), _mm_loadu_ps(b+0))); 46 sum = _mm_add_ps(sum, _mm_mul_ps(_mm_loadu_ps(a+4), _mm_loadu_ps(b+4))); 50 sum = _mm_add_ps(sum, _mm_movehl_ps(sum, sum)); 82 sum = _mm_add_ps(sum, _mm_mul_ps(xx[0], yy[0])); 83 sum = _mm_add_ps(sum, _mm_mul_ps(xx[1], yy[1])); 87 sum = _mm_add_ps(sum, _mm_movehl_ps(sum, sum));
|
H A D | resample_sse.h | 47 sum = _mm_add_ps(sum, _mm_mul_ps(_mm_loadu_ps(a+i), _mm_loadu_ps(b+i))); 48 sum = _mm_add_ps(sum, _mm_mul_ps(_mm_loadu_ps(a+i+4), _mm_loadu_ps(b+i+4))); 50 sum = _mm_add_ps(sum, _mm_movehl_ps(sum, sum)); 64 sum = _mm_add_ps(sum, _mm_mul_ps(_mm_load1_ps(a+i), _mm_loadu_ps(b+i*oversample))); 65 sum = _mm_add_ps(sum, _mm_mul_ps(_mm_load1_ps(a+i+1), _mm_loadu_ps(b+(i+1)*oversample))); 68 sum = _mm_add_ps(sum, _mm_movehl_ps(sum, sum));
|
H A D | cb_search_sse.h | 77 resj = _mm_add_ps(resj, _mm_mul_ps(shape[k],r[j-k])); 80 EE = _mm_add_ps(EE, _mm_mul_ps(resj, resj));
|
H A D | filters_sse.h | 68 mem[0] = _mm_add_ps(mem[0], _mm_mul_ps(xx, num[0])); 74 mem[1] = _mm_add_ps(mem[1], _mm_mul_ps(xx, num[1])); 79 mem[2] = _mm_add_ps(mem[2], _mm_mul_ps(xx, num[2])); 118 mem[0] = _mm_add_ps(mem[0], _mm_mul_ps(xx, num[0])); 124 mem[1] = _mm_add_ps(mem[1], _mm_mul_ps(xx, num[1])); 271 mem[0] = _mm_add_ps(mem[0], _mm_mul_ps(xx, num[0])); 276 mem[1] = _mm_add_ps(mem[1], _mm_mul_ps(xx, num[1])); 280 mem[2] = _mm_add_ps(mem[2], _mm_mul_ps(xx, num[2])); 317 mem[0] = _mm_add_ps(mem[0], _mm_mul_ps(xx, num[0])); 322 mem[1] = _mm_add_ps(me [all...] |
H A D | vq_sse.h | 91 d = _mm_add_ps(d, _mm_mul_ps(in[j], *codebook++));
|
/external/webrtc/webrtc/modules/audio_processing/aec/ |
H A D | aec_core_sse2.c | 62 const __m128 f = _mm_add_ps(c, d); 63 const __m128 g = _mm_add_ps(y_fft_re, e); 64 const __m128 h = _mm_add_ps(y_fft_im, f); 101 const __m128 xPowPlus = _mm_add_ps(x_pow_local, k1e_10f); 106 const __m128 ef_sum2 = _mm_add_ps(ef_re2, ef_im2); 109 __m128 absEfPlus = _mm_add_ps(absEf, k1e_10f); 181 const __m128 e = _mm_add_ps(a, b); 223 wtBuf_re = _mm_add_ps(wtBuf_re, fft_re); 224 wtBuf_im = _mm_add_ps(wtBuf_im, fft_im); 295 const __m128 pol5_y_1 = _mm_add_ps(pol5_y_ [all...] |
H A D | aec_rdft_sse2.c | 38 __m128 x0v = _mm_add_ps(a01v, a23v); 40 const __m128 x2v = _mm_add_ps(a45v, a67v); 43 a01v = _mm_add_ps(x0v, x2v); 49 a45v = _mm_add_ps(a45_0v, a45_1v); 55 x0v = _mm_add_ps(x1v, x3s); 59 a23v = _mm_add_ps(a23_0v, a23_1v); 67 a67v = _mm_add_ps(a67_0v, a67_1v); 98 __m128 x0r0_0i0_0r1_x0i1 = _mm_add_ps(a_00_32, a_08_40); 111 const __m128 x2r0_2i0_2r1_x2i1 = _mm_add_ps(a_16_48, a_24_56); 114 const __m128 xx0 = _mm_add_ps(x0r0_0i0_0r1_x0i [all...] |
/external/webrtc/webrtc/common_audio/ |
H A D | fir_filter_sse.cc | 64 m_sum = _mm_add_ps(m_sum, _mm_mul_ps(m_in, _mm_load_ps(coef_ptr + j))); 69 m_sum = _mm_add_ps(m_sum, _mm_mul_ps(m_in, _mm_load_ps(coef_ptr + j))); 72 m_sum = _mm_add_ps(_mm_movehl_ps(m_sum, m_sum), m_sum);
|
/external/opencv3/modules/imgproc/src/ |
H A D | filter.cpp | 961 s0 = _mm_add_ps(_mm_mul_ps(s0, f), d4); 962 s1 = _mm_add_ps(_mm_mul_ps(s1, f), d4); 965 s2 = _mm_add_ps(_mm_mul_ps(s2, f), d4); 966 s3 = _mm_add_ps(_mm_mul_ps(s3, f), d4); 976 s0 = _mm_add_ps(s0, _mm_mul_ps(_mm_cvtepi32_ps(x0), f)); 977 s1 = _mm_add_ps(s1, _mm_mul_ps(_mm_cvtepi32_ps(x1), f)); 980 s2 = _mm_add_ps(s2, _mm_mul_ps(_mm_cvtepi32_ps(x0), f)); 981 s3 = _mm_add_ps(s3, _mm_mul_ps(_mm_cvtepi32_ps(x1), f)); 996 s0 = _mm_add_ps(_mm_mul_ps(s0, f), d4); 1005 s0 = _mm_add_ps(s [all...] |
H A D | imgwarp.cpp | 568 x0 = _mm_add_ps(_mm_mul_ps(x0, b0), _mm_mul_ps(y0, b1)); 569 x1 = _mm_add_ps(_mm_mul_ps(x1, b0), _mm_mul_ps(y1, b1)); 579 x0 = _mm_add_ps(_mm_mul_ps(x0, b0), _mm_mul_ps(y0, b1)); 580 x1 = _mm_add_ps(_mm_mul_ps(x1, b0), _mm_mul_ps(y1, b1)); 598 x0 = _mm_add_ps(_mm_mul_ps(x0, b0), _mm_mul_ps(y0, b1)); 599 x1 = _mm_add_ps(_mm_mul_ps(x1, b0), _mm_mul_ps(y1, b1)); 609 x0 = _mm_add_ps(_mm_mul_ps(x0, b0), _mm_mul_ps(y0, b1)); 610 x1 = _mm_add_ps(_mm_mul_ps(x1, b0), _mm_mul_ps(y1, b1)); 626 x0 = _mm_add_ps(_mm_mul_ps(x0, b0), _mm_mul_ps(y0, b1)); 663 x0 = _mm_add_ps(_mm_mul_p [all...] |
H A D | pyramids.cpp | 161 r0 = _mm_add_ps(r0, r4); 162 r1 = _mm_add_ps(_mm_add_ps(r1, r3), r2); 163 r0 = _mm_add_ps(r0, _mm_add_ps(r2, r2)); 164 t0 = _mm_add_ps(r0, _mm_mul_ps(r1, _4)); 171 r0 = _mm_add_ps(r0, r4); 172 r1 = _mm_add_ps(_mm_add_ps(r1, r3), r2); 173 r0 = _mm_add_ps(r [all...] |
H A D | corner.cpp | 89 t = _mm_add_ps(_mm_mul_ps(t, t), _mm_mul_ps(b,b)); 90 a = _mm_sub_ps(_mm_add_ps(a, c), _mm_sqrt_ps(t)); 156 t = _mm_add_ps(a, c); 726 v_s1 = _mm_mul_ps(v_factor, _mm_add_ps(v_s1, _mm_add_ps(v_s2, _mm_mul_ps(v_s3, v_m2))));
|
/external/opencv3/modules/hal/src/ |
H A D | stat.cpp | 195 d0 = _mm_add_ps(d0, _mm_mul_ps(t0, t0)); 196 d1 = _mm_add_ps(d1, _mm_mul_ps(t1, t1)); 198 _mm_store_ps(buf, _mm_add_ps(d0, d1)); 231 d0 = _mm_add_ps(d0, _mm_and_ps(t0, absmask)); 232 d1 = _mm_add_ps(d1, _mm_and_ps(t1, absmask)); 234 _mm_store_ps(buf, _mm_add_ps(d0, d1));
|
H A D | mathfuncs.cpp | 89 __m128 c = _mm_div_ps(tmin, _mm_add_ps(tmax, eps)); 92 a = _mm_mul_ps(_mm_add_ps(a, p5), c2); 93 a = _mm_mul_ps(_mm_add_ps(a, p3), c2); 94 a = _mm_mul_ps(_mm_add_ps(a, p1), c); 536 __m128 zf0 = _mm_add_ps(xf0, mA1); 537 __m128 zf1 = _mm_add_ps(xf1, mA1); 539 zf0 = _mm_add_ps(_mm_mul_ps(zf0, xf0), mA2); 540 zf1 = _mm_add_ps(_mm_mul_ps(zf1, xf1), mA2); 542 zf0 = _mm_add_ps(_mm_mul_ps(zf0, xf0), mA3); 543 zf1 = _mm_add_ps(_mm_mul_p [all...] |
/external/eigen/Eigen/src/Geometry/arch/ |
H A 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/opencv3/modules/core/src/ |
H A 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_scale), v_shift); 1485 __m128 v_dst2 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps(_mm_unpackhi_epi16(v_src12, v_zero_i)), v_scale), v_shift); 1487 __m128 v_dst3 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps(_mm_unpacklo_epi16(v_src_34, v_zero_i)), v_scale), v_shift); 1489 __m128 v_dst4 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps(_mm_unpackhi_epi16(v_src_34, v_zero_i)), v_scale), v_shift); 1521 __m128 v_dst1 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps( 1524 __m128 v_dst2 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps( 1527 __m128 v_dst3 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps( 1530 __m128 v_dst4 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps( 1561 __m128 v_dst1 = _mm_add_ps(_mm_mul_ps(_mm_cvtepi32_ps(_mm_unpacklo_epi16(v_src, v_zero_i)), v_scale), v_shift); 1563 __m128 v_dst2 = _mm_add_ps(_mm_mul_p [all...] |
H A D | matmul.cpp | 1700 __m128 y0 = _mm_add_ps(_mm_add_ps(_mm_add_ps( 1704 __m128 y1 = _mm_add_ps(_mm_add_ps(_mm_add_ps( 1708 __m128 y2 = _mm_add_ps(_mm_add_ps(_mm_add_ps( 1712 __m128 y3 = _mm_add_ps(_mm_add_p [all...] |
/external/opencv3/modules/video/src/ |
H A D | optflowgf.cpp | 478 x0 = _mm_add_ps(_mm_loadu_ps(sptr0 + x), _mm_loadu_ps(sptr1 + x)); 479 x1 = _mm_add_ps(_mm_loadu_ps(sptr0 + x + 4), _mm_loadu_ps(sptr1 + x + 4)); 480 s0 = _mm_add_ps(s0, _mm_mul_ps(x0, g4)); 481 s1 = _mm_add_ps(s1, _mm_mul_ps(x1, g4)); 482 x0 = _mm_add_ps(_mm_loadu_ps(sptr0 + x + 8), _mm_loadu_ps(sptr1 + x + 8)); 483 x1 = _mm_add_ps(_mm_loadu_ps(sptr0 + x + 12), _mm_loadu_ps(sptr1 + x + 12)); 484 s2 = _mm_add_ps(s2, _mm_mul_ps(x0, g4)); 485 s3 = _mm_add_ps(s3, _mm_mul_ps(x1, g4)); 504 __m128 x0 = _mm_add_ps(_mm_loadu_ps(sptr0 + x), _mm_loadu_ps(sptr1 + x)); 505 s0 = _mm_add_ps(s [all...] |
H A D | lkpyramid.cpp | 357 qA22 = _mm_add_ps(qA22, _mm_mul_ps(fy, fy)); 358 qA12 = _mm_add_ps(qA12, _mm_mul_ps(fx, fy)); 359 qA11 = _mm_add_ps(qA11, _mm_mul_ps(fx, fx)); 574 qb0 = _mm_add_ps(qb0, _mm_cvtepi32_ps(v00)); 575 qb1 = _mm_add_ps(qb1, _mm_cvtepi32_ps(v10)); 580 qb0 = _mm_add_ps(qb0, _mm_cvtepi32_ps(v00)); 581 qb1 = _mm_add_ps(qb1, _mm_cvtepi32_ps(v10)); 669 _mm_store_ps(bbuf, _mm_add_ps(qb0, qb1));
|
/external/eigen/Eigen/src/Core/arch/SSE/ |
H A D | Complex.h | 48 template<> EIGEN_STRONG_INLINE Packet2cf 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(s), 0xb1)))));
|
H A D | PacketMath.h | 118 template<> EIGEN_STRONG_INLINE Packet4f plset<float>(const float& a) { return _mm_add_ps(pset1<Packet4f>(a), _mm_set_ps(3,2,1,0)); } 122 template<> EIGEN_STRONG_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/pdfium/third_party/libopenjpeg20/ |
H A D | mct.c | 393 vr = _mm_add_ps(vy, _mm_mul_ps(vv, vrv)); 395 vb = _mm_add_ps(vy, _mm_mul_ps(vu, vbu)); 406 vr = _mm_add_ps(vy, _mm_mul_ps(vv, vrv)); 408 vb = _mm_add_ps(vy, _mm_mul_ps(vu, vbu));
|
/external/eigen/Eigen/src/LU/arch/ |
H A 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));
|