| /external/webrtc/webrtc/common_audio/resampler/ |
| 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)));
44 m_sums1 = _mm_mul_ps(m_sums1, _mm_set_ps1(
46 m_sums2 = _mm_mul_ps(m_sums2, _mm_set_ps1(
|
| /external/webrtc/webrtc/modules/audio_processing/aec/ |
| aec_core_sse2.c | 57 const __m128 a = _mm_mul_ps(x_fft_buf_re, h_fft_buf_re);
58 const __m128 b = _mm_mul_ps(x_fft_buf_im, h_fft_buf_im);
59 const __m128 c = _mm_mul_ps(x_fft_buf_re, h_fft_buf_im);
60 const __m128 d = _mm_mul_ps(x_fft_buf_im, h_fft_buf_re);
104 const __m128 ef_re2 = _mm_mul_ps(ef_re, ef_re);
105 const __m128 ef_im2 = _mm_mul_ps(ef_im, ef_im);
111 __m128 ef_re_if = _mm_mul_ps(ef_re, absEfInv);
112 __m128 ef_im_if = _mm_mul_ps(ef_im, absEfInv);
119 ef_re = _mm_mul_ps(ef_re, kMu);
120 ef_im = _mm_mul_ps(ef_im, kMu)
[all...] |
| aec_rdft_sse2.c | 47 const __m128 a45_0v = _mm_mul_ps(wk2rv, x0v);
48 const __m128 a45_1v = _mm_mul_ps(wk2iv, x0w);
54 const __m128 x3s = _mm_mul_ps(mm_swap_sign, x3w);
57 a23_0v = _mm_mul_ps(wk1rv, x0v);
58 a23_1v = _mm_mul_ps(wk1iv, x0w);
65 const __m128 a67_0v = _mm_mul_ps(wk3rv, x0v);
66 const __m128 a67_1v = _mm_mul_ps(wk3iv, x0w);
119 const __m128 x3_swapped = _mm_mul_ps(mm_swap_sign, x3i0_3r0_3i1_x3r1);
127 const __m128 yy2 = _mm_mul_ps(mm_swap_sign, yy1);
129 const __m128 yy4 = _mm_mul_ps(wk1rv, yy3)
[all...] |
| /external/speex/libspeex/ |
| 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)));
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)));
67 sum = _mm_mul_ps(f, sum);
86 t = _mm_mul_ps(_mm_loadu_ps(a+i), _mm_loadu_ps(b+i));
90 t = _mm_mul_ps(_mm_loadu_ps(a+i+4), _mm_loadu_ps(b+i+4));
112 t = _mm_mul_ps(_mm_load1_ps(a+i), _mm_loadu_ps(b+i*oversample));
116 t = _mm_mul_ps(_mm_load1_ps(a+i+1), _mm_loadu_ps(b+(i+1)*oversample));
|
| 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)));
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]));
|
| filters_sse.h | 68 mem[0] = _mm_add_ps(mem[0], _mm_mul_ps(xx, num[0]));
69 mem[0] = _mm_sub_ps(mem[0], _mm_mul_ps(yy, den[0]));
74 mem[1] = _mm_add_ps(mem[1], _mm_mul_ps(xx, num[1]));
75 mem[1] = _mm_sub_ps(mem[1], _mm_mul_ps(yy, den[1]));
79 mem[2] = _mm_add_ps(mem[2], _mm_mul_ps(xx, num[2]));
80 mem[2] = _mm_sub_ps(mem[2], _mm_mul_ps(yy, den[2]));
118 mem[0] = _mm_add_ps(mem[0], _mm_mul_ps(xx, num[0]));
119 mem[0] = _mm_sub_ps(mem[0], _mm_mul_ps(yy, den[0]));
124 mem[1] = _mm_add_ps(mem[1], _mm_mul_ps(xx, num[1]));
125 mem[1] = _mm_sub_ps(mem[1], _mm_mul_ps(yy, den[1]))
[all...] |
| 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));
|
| vq_sse.h | 50 __m128 d = _mm_mul_ps(E[i], half);
52 d = _mm_sub_ps(d, _mm_mul_ps(in[j], *codebook++));
91 d = _mm_add_ps(d, _mm_mul_ps(in[j], *codebook++));
|
| /external/libopus/celt/x86/ |
| 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)));
87 xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(xi, y1i));
88 xsum2 = _mm_add_ps(xsum2,_mm_mul_ps(xi, y2i));
132 yi = _mm_add_ps(yi, _mm_mul_ps(g10v,x2v))
[all...] |
| /external/eigen/Eigen/src/LU/arch/ |
| Inverse_SSE.h | 82 AB = _mm_mul_ps(_mm_shuffle_ps(A,A,0x0F), B);
83 AB = _mm_sub_ps(AB,_mm_mul_ps(_mm_shuffle_ps(A,A,0xA5), _mm_shuffle_ps(B,B,0x4E)));
85 DC = _mm_mul_ps(_mm_shuffle_ps(D,D,0x0F), C);
86 DC = _mm_sub_ps(DC,_mm_mul_ps(_mm_shuffle_ps(D,D,0xA5), _mm_shuffle_ps(C,C,0x4E)));
89 dA = _mm_mul_ps(_mm_shuffle_ps(A, A, 0x5F),A);
92 dB = _mm_mul_ps(_mm_shuffle_ps(B, B, 0x5F),B);
96 dC = _mm_mul_ps(_mm_shuffle_ps(C, C, 0x5F),C);
99 dD = _mm_mul_ps(_mm_shuffle_ps(D, D, 0x5F),D);
103 d = _mm_mul_ps(_mm_shuffle_ps(DC,DC,0xD8),AB);
106 iD = _mm_mul_ps(_mm_shuffle_ps(C,C,0xA0), _mm_movelh_ps(AB,AB))
[all...] |
| /external/eigen/Eigen/src/Geometry/arch/ |
| Geometry_SSE.h | 27 __m128 flip1 = _mm_xor_ps(_mm_mul_ps(vec4f_swizzle1(a,1,2,0,2),
29 __m128 flip2 = _mm_xor_ps(_mm_mul_ps(vec4f_swizzle1(a,3,3,3,1),
32 _mm_add_ps(_mm_sub_ps(_mm_mul_ps(a,vec4f_swizzle1(b,3,3,3,3)),
33 _mm_mul_ps(vec4f_swizzle1(a,2,0,1,0),
48 __m128 mul1=_mm_mul_ps(vec4f_swizzle1(a,1,2,0,3),vec4f_swizzle1(b,2,0,1,3));
49 __m128 mul2=_mm_mul_ps(vec4f_swizzle1(a,2,0,1,3),vec4f_swizzle1(b,1,2,0,3));
|
| /external/webrtc/webrtc/common_audio/ |
| 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)));
|
| /external/libgdx/extensions/gdx-bullet/jni/src/bullet/LinearMath/ |
| btMatrix3x3.h | 788 rv00 = _mm_mul_ps(rv00, mv0);
789 rv01 = _mm_mul_ps(rv01, mv1);
790 rv02 = _mm_mul_ps(rv02, mv2);
797 rv10 = _mm_mul_ps(rv10, mv0);
798 rv11 = _mm_mul_ps(rv11, mv1);
799 rv12 = _mm_mul_ps(rv12, mv2);
806 rv20 = _mm_mul_ps(rv20, mv0);
807 rv21 = _mm_mul_ps(rv21, mv1);
808 rv22 = _mm_mul_ps(rv22, mv2);
884 _mm_mul_ps(m[0].mVec128, vk),
[all...] |
| btVector3.h | 202 mVec128 = _mm_mul_ps(mVec128, vs);
225 mVec128 = _mm_mul_ps(mVec128, vs);
238 __m128 vd = _mm_mul_ps(mVec128, v.mVec128);
304 __m128 vd = _mm_mul_ps(mVec128, mVec128);
314 mVec128 = _mm_mul_ps(mVec128, vd);
331 mVec128 = _mm_mul_ps(mVec128, y);
385 V = _mm_mul_ps(V, mVec128);
386 T = _mm_mul_ps(T, v.mVec128);
423 V = _mm_mul_ps(V, v1.mVec128);
424 T = _mm_mul_ps(T, v2.mVec128)
[all...] |
| /external/opencv3/modules/imgproc/src/ |
| corner.cpp | 86 a = _mm_mul_ps(a, half);
87 c = _mm_mul_ps(c, half);
89 t = _mm_add_ps(_mm_mul_ps(t, t), _mm_mul_ps(b,b));
157 a = _mm_sub_ps(_mm_mul_ps(a, c), _mm_mul_ps(b, b));
158 t = _mm_mul_ps(_mm_mul_ps(k4, t), t);
335 __m128 v_dx2_0 = _mm_mul_ps(v_dx_0, v_dx_0);
336 __m128 v_dxy_0 = _mm_mul_ps(v_dx_0, v_dy_0)
[all...] |
| filter.cpp | [all...] |
| 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))
[all...] |
| /external/libgdx/extensions/gdx-bullet/jni/src/bullet/BulletDynamics/ConstraintSolver/ |
| btSequentialImpulseConstraintSolver.cpp | 111 __m128 result = _mm_mul_ps( vec0, vec1);
145 #define FMADD(a, b, c) _mm_add_ps(c, _mm_mul_ps(a, b))
147 #define FMNADD(a, b, c) _mm_sub_ps(c, _mm_mul_ps(a, b))
157 btSimdScalar deltaImpulse = _mm_sub_ps(_mm_set1_ps(c.m_rhs), _mm_mul_ps(_mm_set1_ps(c.m_appliedImpulse), _mm_set1_ps(c.m_cfm)));
160 deltaImpulse = _mm_sub_ps(deltaImpulse, _mm_mul_ps(deltaVel1Dotn, _mm_set1_ps(c.m_jacDiagABInv)));
161 deltaImpulse = _mm_sub_ps(deltaImpulse, _mm_mul_ps(deltaVel2Dotn, _mm_set1_ps(c.m_jacDiagABInv)));
172 __m128 linearComponentA = _mm_mul_ps(c.m_contactNormal1.mVec128, body1.internalGetInvMass().mVec128);
173 __m128 linearComponentB = _mm_mul_ps((c.m_contactNormal2).mVec128, body2.internalGetInvMass().mVec128);
175 body1.internalGetDeltaLinearVelocity().mVec128 = _mm_add_ps(body1.internalGetDeltaLinearVelocity().mVec128, _mm_mul_ps(linearComponentA, impulseMagnitude));
176 body1.internalGetDeltaAngularVelocity().mVec128 = _mm_add_ps(body1.internalGetDeltaAngularVelocity().mVec128, _mm_mul_ps(c.m_angularComponentA.mVec128, impulseMagnitude))
[all...] |
| /external/pdfium/third_party/libopenjpeg20/ |
| mct.c | 393 vr = _mm_add_ps(vy, _mm_mul_ps(vv, vrv));
394 vg = _mm_sub_ps(_mm_sub_ps(vy, _mm_mul_ps(vu, vgu)), _mm_mul_ps(vv, vgv));
395 vb = _mm_add_ps(vy, _mm_mul_ps(vu, vbu));
406 vr = _mm_add_ps(vy, _mm_mul_ps(vv, vrv));
407 vg = _mm_sub_ps(_mm_sub_ps(vy, _mm_mul_ps(vu, vgu)), _mm_mul_ps(vv, vgv));
408 vb = _mm_add_ps(vy, _mm_mul_ps(vu, vbu));
|
| /external/opencv3/modules/hal/src/ |
| mathfuncs.cpp | 90 __m128 c2 = _mm_mul_ps(c, c);
91 __m128 a = _mm_mul_ps(c2, p7);
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);
107 a = _mm_mul_ps(a, scale4);
510 xf0 = _mm_mul_ps(xf0, postscale4);
511 xf1 = _mm_mul_ps(xf1, postscale4);
533 yf0 = _mm_mul_ps(yf0, _mm_castsi128_ps(_mm_slli_epi32(xi0, 23)));
534 yf1 = _mm_mul_ps(yf1, _mm_castsi128_ps(_mm_slli_epi32(xi1, 23)))
[all...] |
| /external/eigen/Eigen/src/Core/arch/SSE/ |
| Complex.h | 65 return Packet2cf(_mm_addsub_ps(_mm_mul_ps(_mm_moveldup_ps(a.v), b.v),
66 _mm_mul_ps(_mm_movehdup_ps(a.v),
68 // return Packet2cf(_mm_addsub_ps(_mm_mul_ps(vec4f_swizzle1(a.v, 0, 0, 2, 2), b.v),
69 // _mm_mul_ps(vec4f_swizzle1(a.v, 1, 1, 3, 3),
73 return Packet2cf(_mm_add_ps(_mm_mul_ps(vec4f_swizzle1(a.v, 0, 0, 2, 2), b.v),
74 _mm_xor_ps(_mm_mul_ps(vec4f_swizzle1(a.v, 1, 1, 3, 3),
168 return Packet2cf(_mm_add_ps(_mm_xor_ps(_mm_mul_ps(vec4f_swizzle1(a.v, 0, 0, 2, 2), b.v), mask),
169 _mm_mul_ps(vec4f_swizzle1(a.v, 1, 1, 3, 3),
186 return Packet2cf(_mm_add_ps(_mm_mul_ps(vec4f_swizzle1(a.v, 0, 0, 2, 2), b.v),
187 _mm_xor_ps(_mm_mul_ps(vec4f_swizzle1(a.v, 1, 1, 3, 3)
[all...] |
| /external/opencv3/modules/core/src/ |
| convert.cpp | [all...] |
| /external/skia/src/core/ |
| SkHalf.h | 108 __m128i h = _mm_srli_epi32(_mm_castps_si128(_mm_mul_ps(fs.fVec, rebias)), 13);
|
| /prebuilts/gcc/linux-x86/host/x86_64-w64-mingw32-4.8/x86_64-w64-mingw32/include/ |
| fvec.h | 48 friend F32vec4 operator *(const F32vec4 &a,const F32vec4 &b) { return _mm_mul_ps(a,b); }
54 F32vec4& operator *=(F32vec4 &a) { return *this = _mm_mul_ps(vec,a); }
68 return _mm_sub_ps(_mm_add_ps(Ra0,Ra0),_mm_mul_ps(_mm_mul_ps(Ra0,a),Ra0));
|
| /external/opencv3/modules/video/src/ |
| optflowgf.cpp | 468 s0 = _mm_mul_ps(_mm_loadu_ps(sptr0 + x), g4);
469 s1 = _mm_mul_ps(_mm_loadu_ps(sptr0 + x + 4), g4);
470 s2 = _mm_mul_ps(_mm_loadu_ps(sptr0 + x + 8), g4);
471 s3 = _mm_mul_ps(_mm_loadu_ps(sptr0 + x + 12), g4);
480 s0 = _mm_add_ps(s0, _mm_mul_ps(x0, g4));
481 s1 = _mm_add_ps(s1, _mm_mul_ps(x1, g4));
484 s2 = _mm_add_ps(s2, _mm_mul_ps(x0, g4));
485 s3 = _mm_add_ps(s3, _mm_mul_ps(x1, g4));
498 __m128 s0 = _mm_mul_ps(_mm_loadu_ps(sptr0 + x), g4);
505 s0 = _mm_add_ps(s0, _mm_mul_ps(x0, g4))
[all...] |
