/external/libvpx/libvpx/vpx_dsp/arm/ |
D | highbd_idct32x32_34_add_neon.c | 242 vst1q_s32(output, s3[0].val[0]); in vpx_highbd_idct32_6_neon() 244 vst1q_s32(output, s3[0].val[1]); in vpx_highbd_idct32_6_neon() 246 vst1q_s32(output, s3[1].val[0]); in vpx_highbd_idct32_6_neon() 248 vst1q_s32(output, s3[1].val[1]); in vpx_highbd_idct32_6_neon() 250 vst1q_s32(output, s3[2].val[0]); in vpx_highbd_idct32_6_neon() 252 vst1q_s32(output, s3[2].val[1]); in vpx_highbd_idct32_6_neon() 254 vst1q_s32(output, s3[3].val[0]); in vpx_highbd_idct32_6_neon() 256 vst1q_s32(output, s3[3].val[1]); in vpx_highbd_idct32_6_neon() 258 vst1q_s32(output, s3[4].val[0]); in vpx_highbd_idct32_6_neon() 260 vst1q_s32(output, s3[4].val[1]); in vpx_highbd_idct32_6_neon() [all …]
|
D | highbd_idct32x32_135_add_neon.c | 351 vst1q_s32(output + 0, s8[0].val[0]); in vpx_highbd_idct32_12_neon() 352 vst1q_s32(output + 4, s8[0].val[1]); in vpx_highbd_idct32_12_neon() 354 vst1q_s32(output + 0, s8[1].val[0]); in vpx_highbd_idct32_12_neon() 355 vst1q_s32(output + 4, s8[1].val[1]); in vpx_highbd_idct32_12_neon() 357 vst1q_s32(output + 0, s8[2].val[0]); in vpx_highbd_idct32_12_neon() 358 vst1q_s32(output + 4, s8[2].val[1]); in vpx_highbd_idct32_12_neon() 360 vst1q_s32(output + 0, s8[3].val[0]); in vpx_highbd_idct32_12_neon() 361 vst1q_s32(output + 4, s8[3].val[1]); in vpx_highbd_idct32_12_neon() 363 vst1q_s32(output + 0, s8[4].val[0]); in vpx_highbd_idct32_12_neon() 364 vst1q_s32(output + 4, s8[4].val[1]); in vpx_highbd_idct32_12_neon() [all …]
|
D | highbd_idct_neon.h | 365 vst1q_s32(output + 0, out[0].val[0]); in highbd_idct16x16_store_pass1() 366 vst1q_s32(output + 4, out[0].val[1]); in highbd_idct16x16_store_pass1() 368 vst1q_s32(output + 0, out[1].val[0]); in highbd_idct16x16_store_pass1() 369 vst1q_s32(output + 4, out[1].val[1]); in highbd_idct16x16_store_pass1() 371 vst1q_s32(output + 0, out[2].val[0]); in highbd_idct16x16_store_pass1() 372 vst1q_s32(output + 4, out[2].val[1]); in highbd_idct16x16_store_pass1() 374 vst1q_s32(output + 0, out[3].val[0]); in highbd_idct16x16_store_pass1() 375 vst1q_s32(output + 4, out[3].val[1]); in highbd_idct16x16_store_pass1() 377 vst1q_s32(output + 0, out[4].val[0]); in highbd_idct16x16_store_pass1() 378 vst1q_s32(output + 4, out[4].val[1]); in highbd_idct16x16_store_pass1() [all …]
|
D | highbd_idct32x32_1024_add_neon.c | 41 vst1q_s32(out + first * 32, q0.val[0]); in store_in_output() 42 vst1q_s32(out + first * 32 + 4, q0.val[1]); in store_in_output() 43 vst1q_s32(out + second * 32, q1.val[0]); in store_in_output() 44 vst1q_s32(out + second * 32 + 4, q1.val[1]); in store_in_output() 157 vst1q_s32(*out, a[0].val[0]); in transpose_and_store_s32_8x8() 159 vst1q_s32(*out, a[0].val[1]); in transpose_and_store_s32_8x8() 161 vst1q_s32(*out, a[1].val[0]); in transpose_and_store_s32_8x8() 163 vst1q_s32(*out, a[1].val[1]); in transpose_and_store_s32_8x8() 165 vst1q_s32(*out, a[2].val[0]); in transpose_and_store_s32_8x8() 167 vst1q_s32(*out, a[2].val[1]); in transpose_and_store_s32_8x8() [all …]
|
D | quantize_neon.c | 27 vst1q_s32(dqcoeff, dqcoeff_0); in calculate_dqcoeff_and_store() 28 vst1q_s32(dqcoeff + 4, dqcoeff_1); in calculate_dqcoeff_and_store() 181 vst1q_s32(dqcoeff, dqcoeff_0); in calculate_dqcoeff_and_store_32x32() 182 vst1q_s32(dqcoeff + 4, dqcoeff_1); in calculate_dqcoeff_and_store_32x32()
|
D | highbd_idct16x16_add_neon.c | 1035 vst1q_s32(output, out[0]); in vpx_highbd_idct16x16_10_add_half1d_pass1() 1037 vst1q_s32(output, out[1]); in vpx_highbd_idct16x16_10_add_half1d_pass1() 1039 vst1q_s32(output, out[2]); in vpx_highbd_idct16x16_10_add_half1d_pass1() 1041 vst1q_s32(output, out[3]); in vpx_highbd_idct16x16_10_add_half1d_pass1() 1043 vst1q_s32(output, out[4]); in vpx_highbd_idct16x16_10_add_half1d_pass1() 1045 vst1q_s32(output, out[5]); in vpx_highbd_idct16x16_10_add_half1d_pass1() 1047 vst1q_s32(output, out[6]); in vpx_highbd_idct16x16_10_add_half1d_pass1() 1049 vst1q_s32(output, out[7]); in vpx_highbd_idct16x16_10_add_half1d_pass1() 1051 vst1q_s32(output, out[8]); in vpx_highbd_idct16x16_10_add_half1d_pass1() 1053 vst1q_s32(output, out[9]); in vpx_highbd_idct16x16_10_add_half1d_pass1() [all …]
|
D | mem_neon.h | 80 vst1q_s32(buf, v0); in store_s16q_to_tran_low() 81 vst1q_s32(buf + 4, v1); in store_s16q_to_tran_low()
|
/external/libopus/silk/fixed/arm/ |
D | warped_autocorrelation_FIX_neon_intr.c | 91 vst1q_s32( input_QS, vdupq_n_s32( 0 ) ); in silk_warped_autocorrelation_FIX_neon() 93 vst1q_s32( input_QS, vdupq_n_s32( 0 ) ); in silk_warped_autocorrelation_FIX_neon() 95 vst1q_s32( input_QS, vdupq_n_s32( 0 ) ); in silk_warped_autocorrelation_FIX_neon() 97 vst1q_s32( input_QS, vdupq_n_s32( 0 ) ); in silk_warped_autocorrelation_FIX_neon() 99 vst1q_s32( input_QS, vdupq_n_s32( 0 ) ); in silk_warped_autocorrelation_FIX_neon() 101 vst1q_s32( input_QS, vdupq_n_s32( 0 ) ); in silk_warped_autocorrelation_FIX_neon() 107 vst1q_s32( input_QS + 0, vshll_n_s16( vget_low_s16( t0_s16x4 ), QS ) ); in silk_warped_autocorrelation_FIX_neon() 108 vst1q_s32( input_QS + 4, vshll_n_s16( vget_high_s16( t0_s16x4 ), QS ) ); in silk_warped_autocorrelation_FIX_neon() 113 vst1q_s32( input_QS, vdupq_n_s32( 0 ) ); in silk_warped_autocorrelation_FIX_neon() 115 vst1q_s32( input_QS, vdupq_n_s32( 0 ) ); in silk_warped_autocorrelation_FIX_neon() [all …]
|
/external/libopus/silk/arm/ |
D | NSQ_del_dec_neon_intr.c | 170 vst1q_s32( &NSQ->sLTP_shp_Q14[ NSQ->sLTP_shp_buf_idx + offset + 0 ], t0_s32x4 ); in copy_winner_state_kernel() 171 vst1q_s32( &NSQ->sLTP_shp_Q14[ NSQ->sLTP_shp_buf_idx + offset + 4 ], t1_s32x4 ); in copy_winner_state_kernel() 284 vst1q_s32( psDelDec->RD_Q10, vdupq_n_s32( 0 ) ); in silk_NSQ_del_dec_neon() 289 vst1q_s32( psDelDec->LF_AR_Q14, vld1q_dup_s32( &NSQ->sLF_AR_shp_Q14 ) ); in silk_NSQ_del_dec_neon() 290 vst1q_s32( psDelDec->Diff_Q14, vld1q_dup_s32( &NSQ->sDiff_shp_Q14 ) ); in silk_NSQ_del_dec_neon() 291 …vst1q_s32( psDelDec->Shape_Q14[ 0 ], vld1q_dup_s32( &NSQ->sLTP_shp_Q14[ psEncC->ltp_mem_length - 1… in silk_NSQ_del_dec_neon() 293 vst1q_s32( psDelDec->sLPC_Q14[ i ], vld1q_dup_s32( &NSQ->sLPC_Q14[ i ] ) ); in silk_NSQ_del_dec_neon() 296 vst1q_s32( psDelDec->sAR2_Q14[ i ], vld1q_dup_s32( &NSQ->sAR2_Q14[ i ] ) ); in silk_NSQ_del_dec_neon() 364 vst1q_s32( psDelDec->RD_Q10, RD_Q10_s32x4 ); in silk_NSQ_del_dec_neon() 418 vst1q_s32( &NSQ->sLPC_Q14[ i ], t_s32x4 ); in silk_NSQ_del_dec_neon() [all …]
|
D | LPC_inv_pred_gain_neon_intr.c | 134 vst1q_s32( A_QA + n, t0_s32x4 ); in LPC_inverse_pred_gain_QA_neon() 135 vst1q_s32( A_QA + k - n - 4, t1_s32x4 ); in LPC_inverse_pred_gain_QA_neon() 224 vst1q_s32( Atmp_QA + 16, vshll_n_s16( vget_low_s16 ( t2_s16x8 ), QA - 12 ) ); in silk_LPC_inverse_pred_gain_neon() 225 vst1q_s32( Atmp_QA + 20, vshll_n_s16( vget_high_s16( t2_s16x8 ), QA - 12 ) ); in silk_LPC_inverse_pred_gain_neon() 233 vst1q_s32( Atmp_QA + 8, vshll_n_s16( vget_low_s16 ( t1_s16x8 ), QA - 12 ) ); in silk_LPC_inverse_pred_gain_neon() 234 vst1q_s32( Atmp_QA + 12, vshll_n_s16( vget_high_s16( t1_s16x8 ), QA - 12 ) ); in silk_LPC_inverse_pred_gain_neon() 269 vst1q_s32( Atmp_QA + 0, vshll_n_s16( vget_low_s16 ( t0_s16x8 ), QA - 12 ) ); in silk_LPC_inverse_pred_gain_neon() 270 vst1q_s32( Atmp_QA + 4, vshll_n_s16( vget_high_s16( t0_s16x8 ), QA - 12 ) ); in silk_LPC_inverse_pred_gain_neon()
|
D | NSQ_neon.c | 107 vst1q_s32(data1, a0); in silk_NSQ_noise_shape_feedback_loop_neon() 108 vst1q_s32(data1 + 4, a1); in silk_NSQ_noise_shape_feedback_loop_neon()
|
/external/tensorflow/tensorflow/lite/kernels/internal/optimized/ |
D | depthwiseconv_uint8.h | 74 vst1q_s32(acc_buffer_ptr + 4 * i, acc[i].val[0]); 75 vst1q_s32(acc_buffer_ptr + 4 * i + 8, acc[i].val[1]); 123 vst1q_s32(acc_buffer_ptr + 4 * i, acc[i]); 143 vst1q_s32(acc_buffer_ptr, acc[0]); 144 vst1q_s32(acc_buffer_ptr + 4, acc[1]); 185 vst1q_s32(acc_buffer_ptr + 4 * i, acc[i]); 213 vst1q_s32(acc_buffer_ptr + 4 * i, acc[i]); 262 vst1q_s32(acc_buffer_ptr + 4 * i, acc[i]); 290 vst1q_s32(acc_buffer_ptr + 4 * i, acc[i]); 336 vst1q_s32(acc_buffer_ptr + 4 * i, acc[i]); [all …]
|
/external/tensorflow/tensorflow/core/kernels/ |
D | quantized_mul_op.cc | 99 vst1q_s32(output_ptr + 0, z_low_low_32x4); in ScalarMultiply() 100 vst1q_s32(output_ptr + 4, z_low_high_32x4); in ScalarMultiply() 101 vst1q_s32(output_ptr + 8, z_high_low_32x4); in ScalarMultiply() 102 vst1q_s32(output_ptr + 12, z_high_high_32x4); in ScalarMultiply() 176 vst1q_s32(output_ptr + 0, z_low_low_32x4); in VectorMultiply() 177 vst1q_s32(output_ptr + 4, z_low_high_32x4); in VectorMultiply() 178 vst1q_s32(output_ptr + 8, z_high_low_32x4); in VectorMultiply() 179 vst1q_s32(output_ptr + 12, z_high_high_32x4); in VectorMultiply() 264 vst1q_s32(output_ptr + 0, z_low_low_32x4); in VectorTensorMultiply() 265 vst1q_s32(output_ptr + 4, z_low_high_32x4); in VectorTensorMultiply() [all …]
|
D | quantized_add_op.cc | 99 vst1q_s32(output_ptr + 0, result_low_32x4); in ScalarAddition() 100 vst1q_s32(output_ptr + 4, result_high_32x4); in ScalarAddition() 220 vst1q_s32(output_ptr + 0, result_low_32x4); in VectorAddition() 221 vst1q_s32(output_ptr + 4, result_high_32x4); in VectorAddition() 360 vst1q_s32(output_ptr + 0, result_low_32x4); in VectorTensorAddition() 361 vst1q_s32(output_ptr + 4, result_high_32x4); in VectorTensorAddition()
|
D | quantized_resize_bilinear_op.cc | 370 vst1q_s32(reinterpret_cast<int32*>(output_y_ptr + x_start), x0x1x2x3); in OutputLerp32x4x1() 450 vst1q_s32(reinterpret_cast<int32*>(output_y_ptr + x_start * 3), in OutputLerp32x4x3() 452 vst1q_s32(reinterpret_cast<int32*>(output_y_ptr + x_start * 3 + 4), in OutputLerp32x4x3() 454 vst1q_s32(reinterpret_cast<int32*>(output_y_ptr + x_start * 3 + 8), in OutputLerp32x4x3()
|
/external/webrtc/webrtc/modules/audio_coding/codecs/isac/fix/source/ |
D | lattice_neon.c | 85 vst1q_s32(ptr2, ptr2va); in WebRtcIsacfix_FilterMaLoopNeon() 86 vst1q_s32(ptr2 + 4, ptr2vb); in WebRtcIsacfix_FilterMaLoopNeon() 97 vst1q_s32(ptr1, ptr1va); in WebRtcIsacfix_FilterMaLoopNeon() 98 vst1q_s32(ptr1 + 4, ptr1vb); in WebRtcIsacfix_FilterMaLoopNeon() 128 vst1q_s32(ptr2, ptr2va); in WebRtcIsacfix_FilterMaLoopNeon() 137 vst1q_s32(ptr1, ptr1va); in WebRtcIsacfix_FilterMaLoopNeon()
|
D | transform_neon.c | 70 vst1q_s32(outreQ16, outr_0); in ComplexMulAndFindMaxNeon() 72 vst1q_s32(outreQ16, outr_1); in ComplexMulAndFindMaxNeon() 74 vst1q_s32(outimQ16, outi_0); in ComplexMulAndFindMaxNeon() 76 vst1q_s32(outimQ16, outi_1); in ComplexMulAndFindMaxNeon() 323 vst1q_s32(outre1, outr0); in TransformAndFindMaxNeon() 325 vst1q_s32(outim1, outi0); in TransformAndFindMaxNeon() 327 vst1q_s32(outre2, outr_1); in TransformAndFindMaxNeon() 329 vst1q_s32(outim2, outi_1); in TransformAndFindMaxNeon() 444 vst1q_s32(p_outre1, outre1_0); in PostShiftAndDivideAndDemodulateNeon() 446 vst1q_s32(p_outre1, outre1_1); in PostShiftAndDivideAndDemodulateNeon() [all …]
|
D | entropy_coding_neon.c | 69 vst1q_s32(&matrix_product[matrix_prod_index], sum_32x4); in WebRtcIsacfix_MatrixProduct1Neon() 110 vst1q_s32(&matrix_product[matrix_prod_index], sum_32x4); in WebRtcIsacfix_MatrixProduct1Neon()
|
D | pitch_estimator_c.c | 81 vst1q_s32(vbuff, int_32x4_sum); in WebRtcIsacfix_PCorr2Q32()
|
/external/webrtc/webrtc/modules/audio_processing/aecm/ |
D | aecm_core_neon.c | 87 vst1q_s32(echo_est_p, vreinterpretq_s32_u32(echo_est_v_low)); in WebRtcAecm_CalcLinearEnergiesNeon() 88 vst1q_s32(echo_est_p + 4, vreinterpretq_s32_u32(echo_est_v_high)); in WebRtcAecm_CalcLinearEnergiesNeon() 163 vst1q_s32(echo_est_p, vreinterpretq_s32_u32(echo_est_v_low)); in WebRtcAecm_StoreAdaptiveChannelNeon() 164 vst1q_s32(echo_est_p + 4, vreinterpretq_s32_u32(echo_est_v_high)); in WebRtcAecm_StoreAdaptiveChannelNeon() 203 vst1q_s32(start_adapt32_p, adapt32_v_low); in WebRtcAecm_ResetAdaptiveChannelNeon() 204 vst1q_s32(start_adapt32_p + 4, adapt32_v_high); in WebRtcAecm_ResetAdaptiveChannelNeon()
|
/external/gemmlowp/internal/ |
D | 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 …]
|
D | pack_neon.h | 122 vst1q_s32(sums_of_each_slice_ptr, in Pack() 239 vst1q_s32(sums_of_each_slice_ptr, in Pack() 305 vst1q_s32(sums_ptr, sum); in Pack()
|
D | simd_wrappers_neon.h | 56 vst1q_s32(dst, value);
|
/external/libaom/libaom/av1/common/arm/ |
D | av1_txfm_neon.c | 27 vst1q_s32(arr, tmp_q_s32); in av1_round_shift_array_neon()
|
D | mem_neon.h | 471 vst1q_s32(s, s1); in store_s32_4x4() 473 vst1q_s32(s, s2); in store_s32_4x4() 475 vst1q_s32(s, s3); in store_s32_4x4() 477 vst1q_s32(s, s4); in store_s32_4x4()
|