Searched refs:vpadalq_s16 (Results 1 – 10 of 10) sorted by relevance
/external/libaom/libaom/aom_dsp/arm/ |
D | variance_neon.c | 188 q8s32 = vpadalq_s16(q8s32, vreinterpretq_s16_u16(q11u16)); in aom_variance16x8_neon() 194 q8s32 = vpadalq_s16(q8s32, vreinterpretq_s16_u16(q12u16)); in aom_variance16x8_neon() 200 q8s32 = vpadalq_s16(q8s32, vreinterpretq_s16_u16(q13u16)); in aom_variance16x8_neon() 206 q8s32 = vpadalq_s16(q8s32, vreinterpretq_s16_u16(q14u16)); in aom_variance16x8_neon() 262 q8s32 = vpadalq_s16(q8s32, vreinterpretq_s16_u16(q11u16)); in aom_variance8x16_neon() 268 q8s32 = vpadalq_s16(q8s32, vreinterpretq_s16_u16(q12u16)); in aom_variance8x16_neon()
|
/external/libopus/silk/arm/ |
D | LPC_inv_pred_gain_neon_intr.c | 221 t0_s32x4 = vpadalq_s16( t0_s32x4, t2_s16x8 ); in silk_LPC_inverse_pred_gain_neon() 227 t0_s32x4 = vpadalq_s16( t0_s32x4, t1_s16x8 ); in silk_LPC_inverse_pred_gain_neon()
|
/external/tensorflow/tensorflow/lite/kernels/internal/optimized/ |
D | neon_tensor_utils.cc | 789 dotprod_32x4 = vpadalq_s16(dotprod_32x4, prod_16x8); in NeonMatrixBatchVectorMultiplyImpl() 803 dotprod_32x4 = vpadalq_s16(dotprod_32x4, prod_16x8); in NeonMatrixBatchVectorMultiplyImpl() 1115 dotprod_32x4 = vpadalq_s16(dotprod_32x4, prod_16x8); in NeonMatrixBatchVectorMultiplyAccumulate() 1129 dotprod_32x4 = vpadalq_s16(dotprod_32x4, prod_16x8); in NeonMatrixBatchVectorMultiplyAccumulate() 1206 row_sum = vpadalq_s16(row_sum, temp); in NeonMatrixScalarMultiplyAccumulate() 1279 dotprod_32x4 = vpadalq_s16(dotprod_32x4, prod_16x8); in NeonMatrixBatchVectorMultiplyAccumulate() 1283 row_sum_32x4 = vpadalq_s16(row_sum_32x4, row_sum_16x8); in NeonMatrixBatchVectorMultiplyAccumulate() 1292 dotprod_32x4 = vpadalq_s16(dotprod_32x4, prod_16x8); in NeonMatrixBatchVectorMultiplyAccumulate() 1296 row_sum_32x4 = vpadalq_s16(row_sum_32x4, row_sum_16x8); in NeonMatrixBatchVectorMultiplyAccumulate() 1852 dotprod_32x4 = vpadalq_s16(dotprod_32x4, prod_16x8); in NeonSparseMatrixBatchVectorMultiplyAccumulate()
|
D | legacy_optimized_ops.h | 1499 row_accum00 = vpadalq_s16(row_accum00, local_accum00); in GEMVForLstmCellWithSymmetricRange() 1500 row_accum01 = vpadalq_s16(row_accum01, local_accum01); in GEMVForLstmCellWithSymmetricRange() 1501 row_accum10 = vpadalq_s16(row_accum10, local_accum10); in GEMVForLstmCellWithSymmetricRange() 1502 row_accum11 = vpadalq_s16(row_accum11, local_accum11); in GEMVForLstmCellWithSymmetricRange() 1520 row_accum00 = vpadalq_s16(row_accum00, local_accum00); in GEMVForLstmCellWithSymmetricRange() 1521 row_accum01 = vpadalq_s16(row_accum01, local_accum01); in GEMVForLstmCellWithSymmetricRange() 1522 row_accum10 = vpadalq_s16(row_accum10, local_accum10); in GEMVForLstmCellWithSymmetricRange() 1523 row_accum11 = vpadalq_s16(row_accum11, local_accum11); in GEMVForLstmCellWithSymmetricRange() 1570 row_accum20 = vpadalq_s16(row_accum20, local_accum00); in GEMVForLstmCellWithSymmetricRange() 1571 row_accum21 = vpadalq_s16(row_accum21, local_accum01); in GEMVForLstmCellWithSymmetricRange() [all …]
|
D | optimized_ops.h | 473 row_accum0 = vpadalq_s16(row_accum0, local_accum0); in ShuffledFullyConnectedWorkerImpl() 474 row_accum1 = vpadalq_s16(row_accum1, local_accum1); in ShuffledFullyConnectedWorkerImpl() 475 row_accum2 = vpadalq_s16(row_accum2, local_accum2); in ShuffledFullyConnectedWorkerImpl() 476 row_accum3 = vpadalq_s16(row_accum3, local_accum3); in ShuffledFullyConnectedWorkerImpl() 555 row_accum0##B = vpadalq_s16(row_accum0##B, local_accum0); \ in ShuffledFullyConnectedWorkerImpl() 556 row_accum1##B = vpadalq_s16(row_accum1##B, local_accum1); \ in ShuffledFullyConnectedWorkerImpl() 557 row_accum2##B = vpadalq_s16(row_accum2##B, local_accum2); \ in ShuffledFullyConnectedWorkerImpl() 558 row_accum3##B = vpadalq_s16(row_accum3##B, local_accum3); in ShuffledFullyConnectedWorkerImpl()
|
/external/gemmlowp/internal/ |
D | pack_neon.h | 307 sum = vpadalq_s16(sum, sums8); in Pack() 368 sum = vpadalq_s16(sum, sums8); in Pack()
|
/external/clang/test/CodeGen/ |
D | aarch64-neon-misc.c | 1160 return vpadalq_s16(a, b); in test_vpadalq_s16()
|
D | arm_neon_intrinsics.c | 11178 return vpadalq_s16(a, b); in test_vpadalq_s16()
|
/external/gemmlowp/standalone/ |
D | neon-gemm-kernel-benchmark.cc | 3712 acc[i][j] = vpadalq_s16(acc[i][j], local_acc); in Run()
|
/external/neon_2_sse/ |
D | NEON_2_SSE.h | 927 _NEON2SSESTORAGE int32x4_t vpadalq_s16(int32x4_t a, int16x8_t b); // VPADAL.S16 q0,q0 6505 return64(vpadalq_s16(_pM128i(a), _pM128i(b))); in vpadal_s16() 6548 _NEON2SSESTORAGE int32x4_t vpadalq_s16(int32x4_t a, int16x8_t b); // VPADAL.S16 q0,q0 6549 _NEON2SSE_INLINE int32x4_t vpadalq_s16(int32x4_t a, int16x8_t b) // VPADAL.S16 q0,q0 in vpadalq_s16() function
|