/external/webrtc/webrtc/common_audio/resampler/ |
H A D | sinc_resampler_neon.cc | 29 m_input = vld1q_f32(input_ptr); 31 m_sums1 = vmlaq_f32(m_sums1, m_input, vld1q_f32(k1)); 33 m_sums2 = vmlaq_f32(m_sums2, m_input, vld1q_f32(k2));
|
/external/clang/test/CodeGen/ |
H A D | arm-vector-align.c | 18 float32x4_t a = vld1q_f32(addr1);
|
H A D | aarch64-neon-intrinsics.c | 5877 return vld1q_f32(a);
|
H A D | arm_neon_intrinsics.c | 2757 return vld1q_f32(a);
|
/external/webrtc/webrtc/modules/audio_processing/aec/ |
H A D | aec_core_neon.c | 55 const float32x4_t x_fft_buf_re = vld1q_f32(&x_fft_buf[0][xPos + j]); 56 const float32x4_t x_fft_buf_im = vld1q_f32(&x_fft_buf[1][xPos + j]); 57 const float32x4_t h_fft_buf_re = vld1q_f32(&h_fft_buf[0][pos + j]); 58 const float32x4_t h_fft_buf_im = vld1q_f32(&h_fft_buf[1][pos + j]); 59 const float32x4_t y_fft_re = vld1q_f32(&y_fft[0][j]); 60 const float32x4_t y_fft_im = vld1q_f32(&y_fft[1][j]); 143 const float32x4_t x_pow_local = vld1q_f32(&x_pow[i]); 144 const float32x4_t ef_re_base = vld1q_f32(&ef[0][i]); 145 const float32x4_t ef_im_base = vld1q_f32(&ef[1][i]); 209 const float32x4_t x_fft_buf_re = vld1q_f32( [all...] |
H A D | aec_rdft_neon.c | 26 const float32x4_t vec_swap_sign = vld1q_f32((float32_t*)k_swap_sign); 30 float32x4_t a00v = vld1q_f32(&a[j + 0]); 31 float32x4_t a04v = vld1q_f32(&a[j + 4]); 32 float32x4_t a08v = vld1q_f32(&a[j + 8]); 33 float32x4_t a12v = vld1q_f32(&a[j + 12]); 38 const float32x4_t wk1rv = vld1q_f32(&rdft_wk1r[k2]); 39 const float32x4_t wk1iv = vld1q_f32(&rdft_wk1i[k2]); 40 const float32x4_t wk2rv = vld1q_f32(&rdft_wk2r[k2]); 41 const float32x4_t wk2iv = vld1q_f32(&rdft_wk2i[k2]); 42 const float32x4_t wk3rv = vld1q_f32( [all...] |
/external/opencv3/modules/imgproc/src/ |
H A D | accum.cpp | 103 vst1q_f32(dst + x, vaddq_f32(vld1q_f32(dst + x), vcvtq_f32_u32(vmovl_u16(vget_low_u16(v_src0))))); 104 vst1q_f32(dst + x + 4, vaddq_f32(vld1q_f32(dst + x + 4), vcvtq_f32_u32(vmovl_u16(vget_high_u16(v_src0))))); 105 vst1q_f32(dst + x + 8, vaddq_f32(vld1q_f32(dst + x + 8), vcvtq_f32_u32(vmovl_u16(vget_low_u16(v_src1))))); 106 vst1q_f32(dst + x + 12, vaddq_f32(vld1q_f32(dst + x + 12), vcvtq_f32_u32(vmovl_u16(vget_high_u16(v_src1))))); 118 vst1q_f32(dst + x, vaddq_f32(vld1q_f32(dst + x), vcvtq_f32_u32(vmovl_u16(vget_low_u16(v_src0))))); 119 vst1q_f32(dst + x + 4, vaddq_f32(vld1q_f32(dst + x + 4), vcvtq_f32_u32(vmovl_u16(vget_high_u16(v_src0))))); 120 vst1q_f32(dst + x + 8, vaddq_f32(vld1q_f32(dst + x + 8), vcvtq_f32_u32(vmovl_u16(vget_low_u16(v_src1))))); 121 vst1q_f32(dst + x + 12, vaddq_f32(vld1q_f32(dst + x + 12), vcvtq_f32_u32(vmovl_u16(vget_high_u16(v_src1))))); 144 vst1q_f32(dst + x, vaddq_f32(vld1q_f32(dst + x), vcvtq_f32_u32(v_src0))); 145 vst1q_f32(dst + x + 4, vaddq_f32(vld1q_f32(ds [all...] |
H A D | imgwarp.cpp | 1088 float32x4_t v_src00 = vld1q_f32(S0 + x), v_src01 = vld1q_f32(S0 + x + 4); 1089 float32x4_t v_src10 = vld1q_f32(S1 + x), v_src11 = vld1q_f32(S1 + x + 4); 1116 float32x4_t v_src00 = vld1q_f32(S0 + x), v_src01 = vld1q_f32(S0 + x + 4); 1117 float32x4_t v_src10 = vld1q_f32(S1 + x), v_src11 = vld1q_f32(S1 + x + 4); 1144 float32x4_t v_src00 = vld1q_f32(S0 + x), v_src01 = vld1q_f32(S [all...] |
H A D | pyramids.cpp | 637 float32x4_t v_r0 = vld1q_f32(row0 + x); 638 float32x4_t v_r1 = vld1q_f32(row1 + x); 639 float32x4_t v_r2 = vld1q_f32(row2 + x); 640 float32x4_t v_r3 = vld1q_f32(row3 + x); 641 float32x4_t v_r4 = vld1q_f32(row4 + x); 647 v_r0 = vld1q_f32(row0 + x + 4); 648 v_r1 = vld1q_f32(row1 + x + 4); 649 v_r2 = vld1q_f32(row2 + x + 4); 650 v_r3 = vld1q_f32(row3 + x + 4); 651 v_r4 = vld1q_f32(row [all...] |
H A D | corner.cpp | 315 float32x4_t v_dx = vld1q_f32(dxdata + j); 316 float32x4_t v_dy = vld1q_f32(dydata + j); 733 float32x4_t v_dx = vld1q_f32(dxdata + j), v_dy = vld1q_f32(dydata + j); 734 float32x4_t v_s = vmulq_f32(v_dx, vmulq_f32(v_dx, vld1q_f32(d2ydata + j))); 735 v_s = vmlaq_f32(v_s, vld1q_f32(d2xdata + j), vmulq_f32(v_dy, v_dy)); 736 v_s = vmlaq_f32(v_s, vld1q_f32(dxydata + j), vmulq_n_f32(vmulq_f32(v_dy, v_dx), -2));
|
H A D | filter.cpp | 2825 x0l = vld1q_f32(S); 2826 x0h = vld1q_f32(S + 4); 2831 x1l = vld1q_f32(S); 2832 x1h = vld1q_f32(S + 4); 2833 x2l = vld1q_f32(S2); 2834 x2h = vld1q_f32(S2 + 4); 2848 x3l = vld1q_f32(S); 2849 x3h = vld1q_f32(S + 4); 2850 x4l = vld1q_f32(S2); 2851 x4h = vld1q_f32(S [all...] |
H A D | thresh.cpp | 746 float32x4_t v_src = vld1q_f32(src + j); 784 float32x4_t v_src = vld1q_f32(src + j); 818 vst1q_f32(dst + j, vminq_f32(vld1q_f32(src + j), v_thresh)); 850 float32x4_t v_src = vld1q_f32(src + j); 889 float32x4_t v_src = vld1q_f32(src + j);
|
H A D | histogram.cpp | 2371 v_result = vaddq_f32(v_result, vminq_f32(vld1q_f32(h1 + j), vld1q_f32(h2 + j)));
|
H A D | smooth.cpp | 2348 arg_type load(const float* ptr) { return vld1q_f32(ptr); }
|
/external/webrtc/webrtc/common_audio/ |
H A D | fir_filter_neon.cc | 60 m_in = vld1q_f32(in_ptr + j); 61 m_sum = vmlaq_f32(m_sum, m_in, vld1q_f32(coef_ptr + j));
|
/external/eigen/Eigen/src/Core/arch/NEON/ |
H A D | PacketMath.h | 95 EIGEN_STRONG_INLINE float32x4_t vld1q_f32(const float* x) { return ::vld1q_f32((const float32_t*)x); } function in namespace:Eigen::internal 194 template<> EIGEN_STRONG_INLINE Packet4f pload<Packet4f>(const float* from) { EIGEN_DEBUG_ALIGNED_LOAD return vld1q_f32(from); } 197 template<> EIGEN_STRONG_INLINE Packet4f ploadu<Packet4f>(const float* from) { EIGEN_DEBUG_UNALIGNED_LOAD return vld1q_f32(from); }
|
/external/opencv3/modules/hal/src/ |
H A D | stat.cpp | 239 v_sum = vaddq_f32(v_sum, vabdq_f32(vld1q_f32(a + j), vld1q_f32(b + j)));
|
H A D | mathfuncs.cpp | 119 float32x4_t x = vld1q_f32(X + i), y = vld1q_f32(Y + i);
|
/external/opencv3/modules/core/src/ |
H A D | arithm.cpp | 630 FUNCTOR_LOADSTORE(VLoadStore128, float, float32x4_t, vld1q_f32, vst1q_f32); 2531 float32x4_t v_dst1 = vmulq_f32(vld1q_f32(src1 + x), vld1q_f32(src2 + x)); 2532 float32x4_t v_dst2 = vmulq_f32(vld1q_f32(src1 + x + 4), vld1q_f32(src2 + x + 4)); 2541 float32x4_t v_dst1 = vmulq_f32(vld1q_f32(src1 + x), vld1q_f32(src2 + x)); 2544 float32x4_t v_dst2 = vmulq_f32(vld1q_f32(src1 + x + 4), vld1q_f32(src2 + x + 4)); 4326 uint32x4_t v_dst1 = vcgtq_f32(vld1q_f32(src [all...] |
H A D | convert.cpp | 1891 float32x4_t v_dst_0 = vmulq_n_f32(vld1q_f32(src + x), scale); 1895 float32x4_t v_dst_1 = vmulq_n_f32(vld1q_f32(src + x + 4), scale); 4122 float32x4_t v_dst1 = vaddq_f32(vmulq_f32(vld1q_f32(src + x), v_scale), v_shift); 4123 float32x4_t v_dst2 = vaddq_f32(vmulq_f32(vld1q_f32(src + x + 4), v_scale), v_shift); 4144 float32x4_t v_dst1 = vaddq_f32(vmulq_f32(vld1q_f32(src + x), v_scale), v_shift); 4145 float32x4_t v_dst2 = vaddq_f32(vmulq_f32(vld1q_f32(src + x + 4), v_scale), v_shift); 4166 float32x4_t v_dst1 = vaddq_f32(vmulq_f32(vld1q_f32(src + x), v_scale), v_shift); 4167 float32x4_t v_dst2 = vaddq_f32(vmulq_f32(vld1q_f32(src + x + 4), v_scale), v_shift); 4188 float32x4_t v_dst1 = vaddq_f32(vmulq_f32(vld1q_f32(src + x), v_scale), v_shift); 4189 float32x4_t v_dst2 = vaddq_f32(vmulq_f32(vld1q_f32(sr [all...] |
H A D | mathfuncs.cpp | 692 float32x4_t v_m = vld1q_f32(mag + k); 693 vst1q_f32(x + k, vmulq_f32(vld1q_f32(x + k), v_m)); 694 vst1q_f32(y + k, vmulq_f32(vld1q_f32(y + k), v_m));
|
H A D | matmul.cpp | 2204 float32x4_t v_src1 = vld1q_f32(src1 + i), v_src2 = vld1q_f32(src2 + i); 3193 v_sum = vmlaq_f32(v_sum, vld1q_f32(src1 + j), vld1q_f32(src2 + j));
|
H A D | stat.cpp | 700 v_pz = vaddq_u16(v_pz, vandq_u16(vcombine_u16(vmovn_u32(vceqq_f32(vld1q_f32(src + k), v_zero)), 701 vmovn_u32(vceqq_f32(vld1q_f32(src + k + 4), v_zero))), v_1));
|
/external/skia/src/opts/ |
H A D | SkNx_neon.h | 134 static SkNx Load(const void* ptr) { return vld1q_f32((const float*)ptr); }
|
/external/opencv3/modules/hal/include/opencv2/hal/ |
H A D | intrin_neon.hpp | 184 val = vld1q_f32(v);
|