/external/libvpx/libvpx/vpx_dsp/arm/ |
highbd_idct32x32_34_add_neon.c | 241 vst1q_s32(output, s3[0].val[0]); 243 vst1q_s32(output, s3[0].val[1]); 245 vst1q_s32(output, s3[1].val[0]); 247 vst1q_s32(output, s3[1].val[1]); 249 vst1q_s32(output, s3[2].val[0]); 251 vst1q_s32(output, s3[2].val[1]); 253 vst1q_s32(output, s3[3].val[0]); 255 vst1q_s32(output, s3[3].val[1]); 257 vst1q_s32(output, s3[4].val[0]); 259 vst1q_s32(output, s3[4].val[1]) [all...] |
highbd_idct32x32_135_add_neon.c | 350 vst1q_s32(output + 0, s8[0].val[0]); 351 vst1q_s32(output + 4, s8[0].val[1]); 353 vst1q_s32(output + 0, s8[1].val[0]); 354 vst1q_s32(output + 4, s8[1].val[1]); 356 vst1q_s32(output + 0, s8[2].val[0]); 357 vst1q_s32(output + 4, s8[2].val[1]); 359 vst1q_s32(output + 0, s8[3].val[0]); 360 vst1q_s32(output + 4, s8[3].val[1]); 362 vst1q_s32(output + 0, s8[4].val[0]); 363 vst1q_s32(output + 4, s8[4].val[1]) [all...] |
highbd_idct16x16_add_neon.c | 546 vst1q_s32(output + 0, out[0].val[0]); 547 vst1q_s32(output + 4, out[0].val[1]); 549 vst1q_s32(output + 0, out[1].val[0]); 550 vst1q_s32(output + 4, out[1].val[1]); 552 vst1q_s32(output + 0, out[2].val[0]); 553 vst1q_s32(output + 4, out[2].val[1]); 555 vst1q_s32(output + 0, out[3].val[0]); 556 vst1q_s32(output + 4, out[3].val[1]); 558 vst1q_s32(output + 0, out[4].val[0]); 559 vst1q_s32(output + 4, out[4].val[1]) [all...] |
highbd_idct32x32_1024_add_neon.c | 41 vst1q_s32(out + first * 32, q0.val[0]); 42 vst1q_s32(out + first * 32 + 4, q0.val[1]); 43 vst1q_s32(out + second * 32, q1.val[0]); 44 vst1q_s32(out + second * 32 + 4, q1.val[1]); 163 vst1q_s32(*out, a0.val[0]); 165 vst1q_s32(*out, a0.val[1]); 167 vst1q_s32(*out, a1.val[0]); 169 vst1q_s32(*out, a1.val[1]); 171 vst1q_s32(*out, a2.val[0]); 173 vst1q_s32(*out, a2.val[1]) [all...] |
mem_neon.h | 65 vst1q_s32(buf, v0); 66 vst1q_s32(buf + 4, v1);
|
/external/libopus/silk/fixed/arm/ |
warped_autocorrelation_FIX_neon_intr.c | 91 vst1q_s32( input_QS, vdupq_n_s32( 0 ) ); 93 vst1q_s32( input_QS, vdupq_n_s32( 0 ) ); 95 vst1q_s32( input_QS, vdupq_n_s32( 0 ) ); 97 vst1q_s32( input_QS, vdupq_n_s32( 0 ) ); 99 vst1q_s32( input_QS, vdupq_n_s32( 0 ) ); 101 vst1q_s32( input_QS, vdupq_n_s32( 0 ) ); 107 vst1q_s32( input_QS + 0, vshll_n_s16( vget_low_s16( t0_s16x4 ), QS ) ); 108 vst1q_s32( input_QS + 4, vshll_n_s16( vget_high_s16( t0_s16x4 ), QS ) ); 113 vst1q_s32( input_QS, vdupq_n_s32( 0 ) ); 115 vst1q_s32( input_QS, vdupq_n_s32( 0 ) ) [all...] |
/external/libopus/silk/arm/ |
NSQ_del_dec_neon_intr.c | 170 vst1q_s32( &NSQ->sLTP_shp_Q14[ NSQ->sLTP_shp_buf_idx + offset + 0 ], t0_s32x4 ); 171 vst1q_s32( &NSQ->sLTP_shp_Q14[ NSQ->sLTP_shp_buf_idx + offset + 4 ], t1_s32x4 ); 284 vst1q_s32( psDelDec->RD_Q10, vdupq_n_s32( 0 ) ); 289 vst1q_s32( psDelDec->LF_AR_Q14, vld1q_dup_s32( &NSQ->sLF_AR_shp_Q14 ) ); 290 vst1q_s32( psDelDec->Diff_Q14, vld1q_dup_s32( &NSQ->sDiff_shp_Q14 ) ); 291 vst1q_s32( psDelDec->Shape_Q14[ 0 ], vld1q_dup_s32( &NSQ->sLTP_shp_Q14[ psEncC->ltp_mem_length - 1 ] ) ); 293 vst1q_s32( psDelDec->sLPC_Q14[ i ], vld1q_dup_s32( &NSQ->sLPC_Q14[ i ] ) ); 296 vst1q_s32( psDelDec->sAR2_Q14[ i ], vld1q_dup_s32( &NSQ->sAR2_Q14[ i ] ) ); 364 vst1q_s32( psDelDec->RD_Q10, RD_Q10_s32x4 ); 418 vst1q_s32( &NSQ->sLPC_Q14[ i ], t_s32x4 ) [all...] |
LPC_inv_pred_gain_neon_intr.c | 134 vst1q_s32( A_QA + n, t0_s32x4 ); 135 vst1q_s32( A_QA + k - n - 4, t1_s32x4 ); 224 vst1q_s32( Atmp_QA + 16, vshll_n_s16( vget_low_s16 ( t2_s16x8 ), QA - 12 ) ); 225 vst1q_s32( Atmp_QA + 20, vshll_n_s16( vget_high_s16( t2_s16x8 ), QA - 12 ) ); 233 vst1q_s32( Atmp_QA + 8, vshll_n_s16( vget_low_s16 ( t1_s16x8 ), QA - 12 ) ); 234 vst1q_s32( Atmp_QA + 12, vshll_n_s16( vget_high_s16( t1_s16x8 ), QA - 12 ) ); 269 vst1q_s32( Atmp_QA + 0, vshll_n_s16( vget_low_s16 ( t0_s16x8 ), QA - 12 ) ); 270 vst1q_s32( Atmp_QA + 4, vshll_n_s16( vget_high_s16( t0_s16x8 ), QA - 12 ) );
|
NSQ_neon.c | 107 vst1q_s32(data1, a0); 108 vst1q_s32(data1 + 4, a1);
|
/external/webrtc/webrtc/modules/audio_coding/codecs/isac/fix/source/ |
lattice_neon.c | 85 vst1q_s32(ptr2, ptr2va); 86 vst1q_s32(ptr2 + 4, ptr2vb); 97 vst1q_s32(ptr1, ptr1va); 98 vst1q_s32(ptr1 + 4, ptr1vb); 128 vst1q_s32(ptr2, ptr2va); 137 vst1q_s32(ptr1, ptr1va);
|
transform_neon.c | 70 vst1q_s32(outreQ16, outr_0); 72 vst1q_s32(outreQ16, outr_1); 74 vst1q_s32(outimQ16, outi_0); 76 vst1q_s32(outimQ16, outi_1); 323 vst1q_s32(outre1, outr0); 325 vst1q_s32(outim1, outi0); 327 vst1q_s32(outre2, outr_1); 329 vst1q_s32(outim2, outi_1); 444 vst1q_s32(p_outre1, outre1_0); 446 vst1q_s32(p_outre1, outre1_1) [all...] |
entropy_coding_neon.c | 69 vst1q_s32(&matrix_product[matrix_prod_index], sum_32x4); 110 vst1q_s32(&matrix_product[matrix_prod_index], sum_32x4);
|
pitch_estimator_c.c | 81 vst1q_s32(vbuff, int_32x4_sum);
|
/external/webrtc/webrtc/modules/audio_processing/aecm/ |
aecm_core_neon.c | 87 vst1q_s32(echo_est_p, vreinterpretq_s32_u32(echo_est_v_low)); 88 vst1q_s32(echo_est_p + 4, vreinterpretq_s32_u32(echo_est_v_high)); 163 vst1q_s32(echo_est_p, vreinterpretq_s32_u32(echo_est_v_low)); 164 vst1q_s32(echo_est_p + 4, vreinterpretq_s32_u32(echo_est_v_high)); 203 vst1q_s32(start_adapt32_p, adapt32_v_low); 204 vst1q_s32(start_adapt32_p + 4, adapt32_v_high);
|
/external/tensorflow/tensorflow/contrib/lite/kernels/internal/optimized/ |
depthwiseconv_uint8.h | 71 vst1q_s32(acc_buffer_ptr + 4 * i, acc[i].val[0]); 72 vst1q_s32(acc_buffer_ptr + 4 * i + 8, acc[i].val[1]); 120 vst1q_s32(acc_buffer_ptr + 4 * i, acc[i]); 140 vst1q_s32(acc_buffer_ptr, acc[0]); 141 vst1q_s32(acc_buffer_ptr + 4, acc[1]); 182 vst1q_s32(acc_buffer_ptr + 4 * i, acc[i]); 210 vst1q_s32(acc_buffer_ptr + 4 * i, acc[i]); 259 vst1q_s32(acc_buffer_ptr + 4 * i, acc[i]); 287 vst1q_s32(acc_buffer_ptr + 4 * i, acc[i]); 333 vst1q_s32(acc_buffer_ptr + 4 * i, acc[i]) [all...] |
/external/tensorflow/tensorflow/core/kernels/ |
quantized_mul_op.cc | 100 vst1q_s32(output_ptr + 0, z_low_low_32x4); 101 vst1q_s32(output_ptr + 4, z_low_high_32x4); 102 vst1q_s32(output_ptr + 8, z_high_low_32x4); 103 vst1q_s32(output_ptr + 12, z_high_high_32x4); 177 vst1q_s32(output_ptr + 0, z_low_low_32x4); 178 vst1q_s32(output_ptr + 4, z_low_high_32x4); 179 vst1q_s32(output_ptr + 8, z_high_low_32x4); 180 vst1q_s32(output_ptr + 12, z_high_high_32x4); 265 vst1q_s32(output_ptr + 0, z_low_low_32x4); 266 vst1q_s32(output_ptr + 4, z_low_high_32x4) [all...] |
quantized_add_op.cc | 100 vst1q_s32(output_ptr + 0, result_low_32x4); 101 vst1q_s32(output_ptr + 4, result_high_32x4); 221 vst1q_s32(output_ptr + 0, result_low_32x4); 222 vst1q_s32(output_ptr + 4, result_high_32x4); 361 vst1q_s32(output_ptr + 0, result_low_32x4); 362 vst1q_s32(output_ptr + 4, result_high_32x4);
|
quantized_resize_bilinear_op.cc | 364 vst1q_s32(reinterpret_cast<int32*>(output_y_ptr + x_start), x0x1x2x3); 444 vst1q_s32(reinterpret_cast<int32*>(output_y_ptr + x_start * 3), 446 vst1q_s32(reinterpret_cast<int32*>(output_y_ptr + x_start * 3 + 4), 448 vst1q_s32(reinterpret_cast<int32*>(output_y_ptr + x_start * 3 + 8),
|
quantization_utils.h | 617 vst1q_s32(output_ptr + 0, output_value[0]); 618 vst1q_s32(output_ptr + 4, output_value[1]); [all...] |
/external/gemmlowp/internal/ |
output_neon.h | 267 vst1q_s32(dst_ptr + i * stride, block.buf.reg[i]); 309 vst1q_s32(dst_ptr + i * col_stride + 0, src.buf.reg[2 * i + 0]); 310 vst1q_s32(dst_ptr + i * col_stride + 4, src.buf.reg[2 * i + 1]); 321 vst1q_s32(dst_ptr + i * row_stride, transpose_top.buf.reg[i]); 330 vst1q_s32(dst_ptr + (i + 4) * row_stride, transpose_bottom.buf.reg[i]); 380 vst1q_s32(dst_ptr + i * col_stride, src.buf.reg[2 * i]); 381 vst1q_s32(dst_ptr + i * col_stride + 4, src.buf.reg[2 * i + 1]); 392 vst1q_s32(dst_ptr + i * row_stride, transpose_top_left.buf.reg[i]); 401 vst1q_s32(dst_ptr + (i + 4) * row_stride, 411 vst1q_s32(dst_ptr + i * row_stride + 4, transpose_top_right.buf.reg[i]) [all...] |
pack_neon.h | 122 vst1q_s32(sums_of_each_slice_ptr, 239 vst1q_s32(sums_of_each_slice_ptr, 305 vst1q_s32(sums_ptr, sum);
|
simd_wrappers_neon.h | 56 vst1q_s32(dst, value);
|
/external/arm-neon-tests/ |
compute_ref.gccarm-rvct | |
/external/libopus/celt/arm/ |
celt_neon_intr.c | 96 vst1q_s32(sum, a);
|
/external/eigen/Eigen/src/Core/arch/NEON/ |
PacketMath.h | 268 template<> EIGEN_STRONG_INLINE void pstore<int32_t>(int32_t* to, const Packet4i& from) { EIGEN_DEBUG_ALIGNED_STORE vst1q_s32(to, from); } 271 template<> EIGEN_STRONG_INLINE void pstoreu<int32_t>(int32_t* to, const Packet4i& from) { EIGEN_DEBUG_UNALIGNED_STORE vst1q_s32(to, from); } 312 template<> EIGEN_STRONG_INLINE int32_t pfirst<Packet4i>(const Packet4i& a) { int32_t EIGEN_ALIGN16 x[4]; vst1q_s32(x, a); return x[0]; }
|