| /external/libopus/celt/x86/ |
| pitch_sse.h | 43 xsum1 = _mm_loadu_ps(sum);
48 __m128 x0 = _mm_loadu_ps(x+j);
49 __m128 yj = _mm_loadu_ps(y+j);
50 __m128 y3 = _mm_loadu_ps(y+j+3);
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)));
84 __m128 xi = _mm_loadu_ps(x+i);
85 __m128 y1i = _mm_loadu_ps(y01+i);
86 __m128 y2i = _mm_loadu_ps(y02+i)
[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)));
61 __m128 f = _mm_loadu_ps(frac);
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)));
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))
[all...] |
| 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)));
68 x[i] = _mm_loadu_ps(_x+(i<<2));
72 y[i] = _mm_loadu_ps(_y+(i<<2)+offset);
|
| filters_sse.h | 46 mem[i] = _mm_loadu_ps(_mem+4*i);
47 num[i] = _mm_loadu_ps(_num+4*i);
48 den[i] = _mm_loadu_ps(_den+4*i);
99 mem[i] = _mm_loadu_ps(_mem+4*i);
100 num[i] = _mm_loadu_ps(_num+4*i);
101 den[i] = _mm_loadu_ps(_den+4*i);
153 mem[i] = _mm_loadu_ps(_mem+4*i);
154 den[i] = _mm_loadu_ps(_den+4*i);
202 mem[i] = _mm_loadu_ps(_mem+4*i);
203 den[i] = _mm_loadu_ps(_den+4*i)
[all...] |
| /external/webrtc/webrtc/modules/audio_processing/aec/ |
| aec_core_sse2.c | 51 const __m128 x_fft_buf_re = _mm_loadu_ps(&x_fft_buf[0][xPos + j]);
52 const __m128 x_fft_buf_im = _mm_loadu_ps(&x_fft_buf[1][xPos + j]);
53 const __m128 h_fft_buf_re = _mm_loadu_ps(&h_fft_buf[0][pos + j]);
54 const __m128 h_fft_buf_im = _mm_loadu_ps(&h_fft_buf[1][pos + j]);
55 const __m128 y_fft_re = _mm_loadu_ps(&y_fft[0][j]);
56 const __m128 y_fft_im = _mm_loadu_ps(&y_fft[1][j]);
97 const __m128 x_pow_local = _mm_loadu_ps(&x_pow[i]);
98 const __m128 ef_re_base = _mm_loadu_ps(&ef[0][i]);
99 const __m128 ef_im_base = _mm_loadu_ps(&ef[1][i]);
170 const __m128 x_fft_buf_re = _mm_loadu_ps(&x_fft_buf[0][xPos + j])
[all...] |
| aec_rdft_sse2.c | 23 __m128 a00v = _mm_loadu_ps(&a[j + 0]);
24 __m128 a04v = _mm_loadu_ps(&a[j + 4]);
25 __m128 a08v = _mm_loadu_ps(&a[j + 8]);
26 __m128 a12v = _mm_loadu_ps(&a[j + 12]);
253 const __m128 c_j1 = _mm_loadu_ps(&c[j1]); // 1, 2, 3, 4,
254 const __m128 c_k1 = _mm_loadu_ps(&c[29 - j1]); // 28, 29, 30, 31,
260 const __m128 a_j2_0 = _mm_loadu_ps(&a[0 + j2]); // 2, 3, 4, 5,
261 const __m128 a_j2_4 = _mm_loadu_ps(&a[4 + j2]); // 6, 7, 8, 9,
262 const __m128 a_k2_0 = _mm_loadu_ps(&a[122 - j2]); // 120, 121, 122, 123,
263 const __m128 a_k2_4 = _mm_loadu_ps(&a[126 - j2]); // 124, 125, 126, 127
[all...] |
| /external/clang/test/Sema/ |
| uninit-variables-vectors.c | 6 __m128 _mm_loadu_ps(const float *p);
14 X = _mm_loadu_ps(&input[0]);
|
| /external/opencv3/modules/imgproc/src/ |
| corner.cpp | 75 __m128 t0 = _mm_loadu_ps(cov + j*3); // a0 b0 c0 x
76 __m128 t1 = _mm_loadu_ps(cov + j*3 + 3); // a1 b1 c1 x
77 __m128 t2 = _mm_loadu_ps(cov + j*3 + 6); // a2 b2 c2 x
78 __m128 t3 = _mm_loadu_ps(cov + j*3 + 9); // a3 b3 c3 x
145 __m128 t0 = _mm_loadu_ps(cov + j*3); // a0 b0 c0 x
146 __m128 t1 = _mm_loadu_ps(cov + j*3 + 3); // a1 b1 c1 x
147 __m128 t2 = _mm_loadu_ps(cov + j*3 + 6); // a2 b2 c2 x
148 __m128 t3 = _mm_loadu_ps(cov + j*3 + 9); // a3 b3 c3 x
330 __m128 v_dx_0 = _mm_loadu_ps(dxdata + j);
331 __m128 v_dx_1 = _mm_loadu_ps(dxdata + j + 4)
[all...] |
| imgwarp.cpp | 593 x0 = _mm_loadu_ps(S0 + x);
594 x1 = _mm_loadu_ps(S0 + x + 4);
595 y0 = _mm_loadu_ps(S1 + x);
596 y1 = _mm_loadu_ps(S1 + x + 4);
604 x0 = _mm_loadu_ps(S0 + x + 8);
605 x1 = _mm_loadu_ps(S0 + x + 12);
606 y0 = _mm_loadu_ps(S1 + x + 8);
607 y1 = _mm_loadu_ps(S1 + x + 12);
623 x0 = _mm_loadu_ps(S0 + x);
624 y0 = _mm_loadu_ps(S1 + x)
[all...] |
| filter.cpp | [all...] |
| thresh.cpp | 730 v0 = _mm_loadu_ps( src + j );
731 v1 = _mm_loadu_ps( src + j + 4 );
768 v0 = _mm_loadu_ps( src + j );
769 v1 = _mm_loadu_ps( src + j + 4 );
806 v0 = _mm_loadu_ps( src + j );
807 v1 = _mm_loadu_ps( src + j + 4 );
837 v0 = _mm_loadu_ps( src + j );
838 v1 = _mm_loadu_ps( src + j + 4 );
876 v0 = _mm_loadu_ps( src + j );
877 v1 = _mm_loadu_ps( src + j + 4 )
[all...] |
| morph.cpp | 155 __m128 s = _mm_loadu_ps((const float*)src + i);
158 __m128 x = _mm_loadu_ps((const float*)src + i + k);
469 __m128 s0 = _mm_loadu_ps(sptr);
470 __m128 s1 = _mm_loadu_ps(sptr + 4);
471 __m128 s2 = _mm_loadu_ps(sptr + 8);
472 __m128 s3 = _mm_loadu_ps(sptr + 12);
478 x0 = _mm_loadu_ps(sptr);
479 x1 = _mm_loadu_ps(sptr + 4);
480 x2 = _mm_loadu_ps(sptr + 8);
481 x3 = _mm_loadu_ps(sptr + 12)
[all...] |
| color.cpp | [all...] |
| pyramids.cpp | 499 __m128 v_r0 = _mm_loadu_ps(row0 + x);
500 __m128 v_r1 = _mm_loadu_ps(row1 + x);
501 __m128 v_r2 = _mm_loadu_ps(row2 + x);
506 v_r0 = _mm_loadu_ps(row0 + x + 4);
507 v_r1 = _mm_loadu_ps(row1 + x + 4);
508 v_r2 = _mm_loadu_ps(row2 + x + 4);
[all...] |
| /external/opencv3/modules/hal/src/ |
| stat.cpp | 193 __m128 t0 = _mm_sub_ps(_mm_loadu_ps(a + j), _mm_loadu_ps(b + j));
194 __m128 t1 = _mm_sub_ps(_mm_loadu_ps(a + j + 4), _mm_loadu_ps(b + j + 4));
229 __m128 t0 = _mm_sub_ps(_mm_loadu_ps(a + j), _mm_loadu_ps(b + j));
230 __m128 t1 = _mm_sub_ps(_mm_loadu_ps(a + j + 4), _mm_loadu_ps(b + j + 4));
|
| mathfuncs.cpp | 85 __m128 x = _mm_loadu_ps(X + i), y = _mm_loadu_ps(Y + i);
479 xf0 = _mm_loadu_ps(&x[i].f);
480 xf1 = _mm_loadu_ps(&x[i+4].f);
[all...] |
| /external/webrtc/webrtc/common_audio/resampler/ |
| sinc_resampler_sse.cc | 31 m_input = _mm_loadu_ps(input_ptr + i);
|
| /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);
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));
482 x0 = _mm_add_ps(_mm_loadu_ps(sptr0 + x + 8), _mm_loadu_ps(sptr1 + x + 8))
[all...] |
| /external/webrtc/webrtc/common_audio/ |
| fir_filter_sse.cc | 63 m_in = _mm_loadu_ps(in_ptr + j);
|
| /external/eigen/Eigen/src/Core/arch/SSE/ |
| PacketMath.h | 223 return _mm_loadu_ps(from);
251 // NOTE Some version of MSVC10 generates bad code when using _mm_loadu_ps
253 // TODO On most architectures this version should also be faster than a single _mm_loadu_ps
259 return _mm_loadu_ps(from);
286 return _mm_loadu_ps(from);
|
| /external/opencv3/modules/objdetect/src/ |
| hog.cpp | 421 __m128 _mag = _mm_loadu_ps(dbuf + x + (width << 1));
422 __m128 _angle = _mm_loadu_ps(dbuf + x + width * 3);
826 __m128 _w = _mm_mul_ps(_mm_set1_ps(pk.gradWeight), _mm_loadu_ps(pk.histWeights));
874 __m128 _w = _mm_mul_ps(_mm_set1_ps(pk.gradWeight), _mm_loadu_ps(pk.histWeights));
[all...] |
| /external/opencv3/modules/core/src/ |
| mathfuncs.cpp | 280 __m128 v_src = _mm_loadu_ps(buf[0] + k);
417 __m128 v_src = _mm_loadu_ps(buf[0] + k);
510 __m128 v_angle = _mm_loadu_ps(angle + i);
701 __m128 v_m = _mm_loadu_ps(mag + k);
702 _mm_storeu_ps(x + k, _mm_mul_ps(_mm_loadu_ps(x + k), v_m));
703 _mm_storeu_ps(y + k, _mm_mul_ps(_mm_loadu_ps(y + k), v_m));
[all...] |
| rand.cpp | 232 __m128 q0 = _mm_loadu_ps((const float*)(p + i));
233 __m128 q1 = _mm_loadu_ps((const float*)(p + i + 2));
241 _mm_storeu_ps(arr + i, _mm_add_ps(_mm_mul_ps(_mm_loadu_ps(f), p0), p1));
|
| /frameworks/rs/cpu_ref/ |
| rsCpuIntrinsics_x86.cpp | 387 pf = _mm_mul_ps(g0, _mm_loadu_ps(pi));
390 gx = _mm_loadu_ps((const float *)gptr + r);
391 p0 = _mm_loadu_ps(pi + r);
392 p1 = _mm_loadu_ps(pi + r + 4);
[all...] |
| /prebuilts/gcc/linux-x86/host/x86_64-w64-mingw32-4.8/x86_64-w64-mingw32/include/ |
| fvec.h | 118 inline void loadu(F32vec4 &a,float *p) { a = _mm_loadu_ps(p); }
|
