Gitiles
Code Review Sign In
review.mlplatform.org / ml / ComputeLibrary / 33f41fabd30fb444aaa0cf3e65b61794d498d151^! / . / src / core / NEON / kernels
commit33f41fabd30fb444aaa0cf3e65b61794d498d151[log] [tgz]
authorMichele Di Giorgio <michele.digiorgio@arm.com>Tue Mar 09 14:09:08 2021 +0000
committerMichele Di Giorgio <michele.digiorgio@arm.com>Wed Mar 31 17:08:51 2021 +0000
treea381cff3096a3b05198b0cd311fee28e40fd5a4f
parent5f91b5d7063462854b62d342f9d4e04ae647e9a6 [diff]
Fix trademarks throughout the codebase

Resolves: COMPMID-4299

Change-Id: Ie6a52c1371b9a2a7b5bb4f019ecd5e70a2008567
Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5338
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>

diff --git a/src/core/NEON/kernels/NEBatchNormalizationLayerKernel.cpp b/src/core/NEON/kernels/NEBatchNormalizationLayerKernel.cpp
index 9d47d7d..1691943 100644
--- a/src/core/NEON/kernels/NEBatchNormalizationLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEBatchNormalizationLayerKernel.cpp

@@ -148,7 +148,7 @@
 template <typename T, bool fused_activation, typename F>
 void NEBatchNormalizationLayerKernel::batch_normalization_nchw(const Window &window)
 {
-    /** Neon vector tag type. */
+    /** SIMD vector tag type. */
     using ExactTagType = typename wrapper::traits::neon_bitvector_tag_t<T, wrapper::traits::BitWidth::W128>;
 
     const int  window_step_x  = 16 / sizeof(T);
@@ -164,7 +164,7 @@
     F activation_functor(_act_info);
 
     // Hold information about the current feature map we are iterating.
-    // Only compute denominator and Neon vectors once per feature map.
+    // Only compute denominator and constants once per feature map.
     int slice = -1;
 
     const auto input_mean  = reinterpret_cast<const T *>(_mean->ptr_to_element(Coordinates(0, 0)));

diff --git a/src/core/NEON/kernels/NEChannelShuffleLayerKernel.cpp b/src/core/NEON/kernels/NEChannelShuffleLayerKernel.cpp
index c906657..008ad7c 100644
--- a/src/core/NEON/kernels/NEChannelShuffleLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEChannelShuffleLayerKernel.cpp

@@ -40,7 +40,7 @@
 {
 Status validate_arguments(const ITensorInfo *input, const ITensorInfo *output, unsigned int num_groups)
 {
-    // Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use Neon FP16 instructions.
+    // Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use CPU FP16 instructions.
     ARM_COMPUTE_RETURN_ERROR_ON(input->data_type() == DataType::UNKNOWN);
     ARM_COMPUTE_RETURN_ERROR_ON_DATA_LAYOUT_NOT_IN(input, DataLayout::NCHW, DataLayout::NHWC);
 

diff --git a/src/core/NEON/kernels/NECol2ImKernel.cpp b/src/core/NEON/kernels/NECol2ImKernel.cpp
index 26e4455..4ba02f1 100644
--- a/src/core/NEON/kernels/NECol2ImKernel.cpp
+++ b/src/core/NEON/kernels/NECol2ImKernel.cpp

@@ -44,7 +44,7 @@
 {
 Status validate_arguments(const ITensorInfo *input, const ITensorInfo *output, const Size2D &convolved_dims)
 {
-    //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use Neon FP16 instructions.
+    //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use CPU FP16 instructions.
     ARM_COMPUTE_RETURN_ERROR_ON(input->data_type() == DataType::UNKNOWN);
 
     // Validate configured output

diff --git a/src/core/NEON/kernels/NECol2ImKernel.h b/src/core/NEON/kernels/NECol2ImKernel.h
index 00a519d..397bf5a 100644
--- a/src/core/NEON/kernels/NECol2ImKernel.h
+++ b/src/core/NEON/kernels/NECol2ImKernel.h

@@ -32,7 +32,7 @@
 {
 class ITensor;
 
-/** Neon kernel to perform col2im reshaping.
+/** Kernel to perform col2im reshaping.
  *
  * Rearranges each matrix column into image blocks. It's the inverse operation of @ref NEIm2ColKernel.
  *

diff --git a/src/core/NEON/kernels/NEConvertFullyConnectedWeightsKernel.cpp b/src/core/NEON/kernels/NEConvertFullyConnectedWeightsKernel.cpp
index 101d138..4cd1bc7 100644
--- a/src/core/NEON/kernels/NEConvertFullyConnectedWeightsKernel.cpp
+++ b/src/core/NEON/kernels/NEConvertFullyConnectedWeightsKernel.cpp

@@ -69,7 +69,7 @@
                                                       DataLayout data_layout)
 {
     ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(input);
-    //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use Neon FP16 instructions.
+    //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use CPU FP16 instructions.
     ARM_COMPUTE_RETURN_ERROR_ON(input->data_type() == DataType::UNKNOWN);
     ARM_COMPUTE_RETURN_ERROR_ON(input->num_dimensions() != 2);
     ARM_COMPUTE_RETURN_ERROR_ON(input->dimension(1) != original_input_shape.total_size_lower(3));

diff --git a/src/core/NEON/kernels/NEConvertQuantizedSignednessKernel.h b/src/core/NEON/kernels/NEConvertQuantizedSignednessKernel.h
index 0be1fbe..67d5ca2 100644
--- a/src/core/NEON/kernels/NEConvertQuantizedSignednessKernel.h
+++ b/src/core/NEON/kernels/NEConvertQuantizedSignednessKernel.h

@@ -32,7 +32,7 @@
 // Forward declarations
 class ITensor;
 
-/** Neon kernel to convert asymmetric signed to asymmetric signed and vice-versa */
+/** Kernel to convert asymmetric signed to asymmetric signed and vice-versa */
 class NEConvertQuantizedSignednessKernel : public INEKernel
 {
 public:

diff --git a/src/core/NEON/kernels/NEDirectConvolutionLayerKernel.cpp b/src/core/NEON/kernels/NEDirectConvolutionLayerKernel.cpp
index e81d50f..09f9974 100644
--- a/src/core/NEON/kernels/NEDirectConvolutionLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEDirectConvolutionLayerKernel.cpp

@@ -149,7 +149,7 @@
 
 // Optimized convolver for 1x1 kernels used only where input width and height are both <= 8
 // For big Z as in Input=7x7x832, this implementation is faster than the general code becuase it doesn't need to
-// store intermidiate results in memory. Temporary results are stored in Neon registers directly and then written to the output buffer.
+// store intermidiate results in memory. Temporary results are stored in SIMD registers directly and then written to the output buffer.
 template <unsigned int stridex>
 class convolver_w1x1_i8x8_f32
 {

diff --git a/src/core/NEON/kernels/NEDirectConvolutionLayerKernel.h b/src/core/NEON/kernels/NEDirectConvolutionLayerKernel.h
index 58d385a..258def7 100644
--- a/src/core/NEON/kernels/NEDirectConvolutionLayerKernel.h
+++ b/src/core/NEON/kernels/NEDirectConvolutionLayerKernel.h

@@ -30,7 +30,7 @@
 {
 class ITensor;
 
-/** Neon interface for Direct Convolution Layer kernel */
+/** Interface for the kernel to perform Direct Convolution Layer. */
 class NEDirectConvolutionLayerKernel : public INEKernel
 {
 public:

diff --git a/src/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.cpp b/src/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.cpp
index 8dcbd00..3597045 100644
--- a/src/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.cpp
+++ b/src/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.cpp

@@ -93,7 +93,7 @@
 output_stage_nchw(ITensor *input, const ITensor *bias, const Window &window, ITensor *output,
                   int result_fixedpoint_multiplier, int result_shift, int result_offset_after_shift, bool has_bias)
 {
-    /** Neon vector tag type. */
+    /** SIMD vector tag type. */
     using ExactTagType = typename wrapper::traits::neon_bitvector_tag_t<T, wrapper::traits::BitWidth::W128>;
 
     ARM_COMPUTE_ERROR_ON(input->info()->data_layout() == DataLayout::UNKNOWN);

diff --git a/src/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.h b/src/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.h
index cd0710d..8f7eeb0 100644
--- a/src/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.h
+++ b/src/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.h

@@ -30,7 +30,7 @@
 namespace arm_compute
 {
 class ITensor;
-/** Neon kernel to accumulate the biases, if provided, or downscale in case of quantized input.
+/** Kernel to accumulate the biases, if provided, or downscale in case of quantized input.
  *
  * @note We assume bias to be shared
  * @note For quantized computations (i.e. @p input of S32 type) the output data type for auto-initialization must be passed as part

diff --git a/src/core/NEON/kernels/NEFillBorderKernel.cpp b/src/core/NEON/kernels/NEFillBorderKernel.cpp
index 70178df..1c7c1f9 100644
--- a/src/core/NEON/kernels/NEFillBorderKernel.cpp
+++ b/src/core/NEON/kernels/NEFillBorderKernel.cpp

@@ -103,7 +103,7 @@
 void NEFillBorderKernel::configure(ITensorInfo *tensor, BorderSize border_size, BorderMode border_mode, const PixelValue &constant_border_value)
 {
     ARM_COMPUTE_ERROR_ON_NULLPTR(tensor);
-    //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use Neon FP16 instructions.
+    //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use CPU FP16 instructions.
     ARM_COMPUTE_ERROR_ON(tensor->data_type() == DataType::UNKNOWN);
 
     _border_size           = border_size;

diff --git a/src/core/NEON/kernels/NEGEMMInterleave4x4Kernel.cpp b/src/core/NEON/kernels/NEGEMMInterleave4x4Kernel.cpp
index 96ed810..9011680 100644
--- a/src/core/NEON/kernels/NEGEMMInterleave4x4Kernel.cpp
+++ b/src/core/NEON/kernels/NEGEMMInterleave4x4Kernel.cpp

@@ -47,7 +47,7 @@
 Status validate_arguments(const ITensorInfo *input, const ITensorInfo *output)
 {
     ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(input);
-    //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use Neon FP16 instructions.
+    //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use CPU FP16 instructions.
     ARM_COMPUTE_RETURN_ERROR_ON(input->data_type() == DataType::UNKNOWN);
     ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(input, output);
 

diff --git a/src/core/NEON/kernels/NEGEMMInterleave4x4Kernel.h b/src/core/NEON/kernels/NEGEMMInterleave4x4Kernel.h
index 92fbd12..e592d5e 100644
--- a/src/core/NEON/kernels/NEGEMMInterleave4x4Kernel.h
+++ b/src/core/NEON/kernels/NEGEMMInterleave4x4Kernel.h

@@ -30,7 +30,7 @@
 {
 class ITensor;
 
-/** Neon kernel to interleave the elements of a matrix
+/** Kernel to interleave the elements of a matrix
  *
  * This function puts the values in a 4x4 block of Matrix A on the same row (Interleaved values)
  *

diff --git a/src/core/NEON/kernels/NEGEMMLowpMatrixMultiplyKernel.h b/src/core/NEON/kernels/NEGEMMLowpMatrixMultiplyKernel.h
index dfdb7b3..acfb79e 100644
--- a/src/core/NEON/kernels/NEGEMMLowpMatrixMultiplyKernel.h
+++ b/src/core/NEON/kernels/NEGEMMLowpMatrixMultiplyKernel.h

@@ -30,7 +30,7 @@
 {
 class ITensor;
 
-/** Neon kernel to multiply matrices
+/** Kernel to multiply matrices
  *
  * @note @ref NEGEMMLowpMatrixMultiplyKernel low precision matrix product kernel
  *  This kernel performs the following computation:

diff --git a/src/core/NEON/kernels/NEGEMMLowpOffsetContributionKernel.h b/src/core/NEON/kernels/NEGEMMLowpOffsetContributionKernel.h
index 9911ffc..f71929f 100644
--- a/src/core/NEON/kernels/NEGEMMLowpOffsetContributionKernel.h
+++ b/src/core/NEON/kernels/NEGEMMLowpOffsetContributionKernel.h

@@ -30,7 +30,7 @@
 {
 class ITensor;
 
-/** Neon kernel used to add the offset contribution after @ref NEGEMMLowpMatrixMultiplyKernel. The computation is performed in-place
+/** Kernel used to add the offset contribution after @ref NEGEMMLowpMatrixMultiplyKernel. The computation is performed in-place
  *
  * This kernel takes a final int32 accumulator value (the output of @ref NEGEMMLowpMatrixMultiplyKernel),
  * and adds to it the offset contribution of matrix A and matrix B in-place.

diff --git a/src/core/NEON/kernels/NEGEMMLowpOffsetContributionOutputStageKernel.h b/src/core/NEON/kernels/NEGEMMLowpOffsetContributionOutputStageKernel.h
index 39fbd8e..6908f37 100644
--- a/src/core/NEON/kernels/NEGEMMLowpOffsetContributionOutputStageKernel.h
+++ b/src/core/NEON/kernels/NEGEMMLowpOffsetContributionOutputStageKernel.h

@@ -30,7 +30,7 @@
 {
 class ITensor;
 
-/** Neon kernel used to add the offset contribution and perform the output stage after @ref NEGEMMLowpMatrixMultiplyKernel.
+/** Kernel used to add the offset contribution and perform the output stage after @ref NEGEMMLowpMatrixMultiplyKernel.
  *
  * The computation is performed in-place
  *

diff --git a/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ScaleKernel.h b/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ScaleKernel.h
index 63d80aa..021ff8e 100644
--- a/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ScaleKernel.h
+++ b/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ScaleKernel.h

@@ -30,7 +30,7 @@
 {
 class ITensor;
 
-/** Neon kernel used to quantize down the int32 accumulator values of GEMMLowp to QASYMM8/QASYMM8_SIGNED
+/** Kernel used to quantize down the int32 accumulator values of GEMMLowp to QASYMM8/QASYMM8_SIGNED
  *
  * This kernel takes a final int32 accumulator value (the output of @ref NEGEMMLowpMatrixMultiplyKernel), and processes it to obtain the final QASYMM8/QASYMM8_SIGNED value.
  * The following computations will be performed by the kernel:

diff --git a/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.h b/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.h
index 8e92ba6..b01b204 100644
--- a/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.h
+++ b/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.h

@@ -30,7 +30,7 @@
 {
 class ITensor;
 
-/** Neon kernel used to quantize down the int32 accumulator values of GEMMLowp to QSYMM16
+/** Kernel used to quantize down the int32 accumulator values of GEMMLowp to QSYMM16
  *
  * This kernel takes a final int32 accumulator value (the output of @ref NEGEMMLowpMatrixMultiplyKernel), and processes it to obtain the final QSYMM16 value.
  * The following computations will be performed by the kernel:

diff --git a/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.h b/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.h
index 9b51a3b..9e7dc2f 100644
--- a/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.h
+++ b/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.h

@@ -30,7 +30,7 @@
 {
 class ITensor;
 
-/** Neon kernel used to quantize down the int32 accumulator values of GEMMLowp to QASYMM8_SIGNED
+/** Kernel used to quantize down the int32 accumulator values of GEMMLowp to QASYMM8_SIGNED
  *
  * This kernel takes a final int32 accumulator value (the output of @ref NEGEMMLowpMatrixMultiplyKernel), and processes it to obtain the final QASYMM8_SIGNED value.
  * The following computations will be performed by the kernel:

diff --git a/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h b/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h
index 4d43afa..def0573 100644
--- a/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h
+++ b/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h

@@ -30,7 +30,7 @@
 {
 class ITensor;
 
-/** Neon kernel used to quantize down the int32 accumulator values of GEMMLowp to QASYMM8
+/** Kernel used to quantize down the int32 accumulator values of GEMMLowp to QASYMM8
  *
  * This kernel takes a final int32 accumulator value (the output of @ref NEGEMMLowpMatrixMultiplyKernel), and processes it to obtain the final QASYMM8 value.
  * The following computations will be performed by the kernel:

diff --git a/src/core/NEON/kernels/NEGEMMLowpReductionKernel.h b/src/core/NEON/kernels/NEGEMMLowpReductionKernel.h
index 521adbf..9be618d 100644
--- a/src/core/NEON/kernels/NEGEMMLowpReductionKernel.h
+++ b/src/core/NEON/kernels/NEGEMMLowpReductionKernel.h

@@ -32,7 +32,7 @@
 class ITensor;
 struct GEMMLowpReductionKernelInfo;
 
-/** Common interface for all Neon reduction kernels */
+/** Common interface for all reduction kernels */
 class INEGEMMLowpReductionKernel : public INEKernel
 {
 public:
@@ -69,7 +69,7 @@
     bool           _mul_by_scalar;
 };
 
-/** Neon kernel used to compute the row-vectors of sums of all the entries in each row of Matrix A.
+/** Kernel used to compute the row-vectors of sums of all the entries in each row of Matrix A.
  *
  * @note This stage is needed to handle the offset of matrix product
  *       https://github.com/google/gemmlowp/blob/master/doc/low-precision.md
@@ -130,7 +130,7 @@
     void run_internal(const Window &window);
 };
 
-/** Neon kernel used to compute the row-vectors of sums of all the entries in each column of Matrix B.
+/** Kernel used to compute the row-vectors of sums of all the entries in each column of Matrix B.
  *
  * @note This stage is needed to handle the offset of matrix product
  *       https://github.com/google/gemmlowp/blob/master/doc/low-precision.md

diff --git a/src/core/NEON/kernels/NEGEMMMatrixAdditionKernel.h b/src/core/NEON/kernels/NEGEMMMatrixAdditionKernel.h
index f9ff143..c896cab 100644
--- a/src/core/NEON/kernels/NEGEMMMatrixAdditionKernel.h
+++ b/src/core/NEON/kernels/NEGEMMMatrixAdditionKernel.h

@@ -30,7 +30,7 @@
 {
 class ITensor;
 
-/** Neon kernel to perform the in-place matrix addition between 2 matrices taking into account that the second matrix might be weighted by a scalar value beta:
+/** Kernel to perform the in-place matrix addition between 2 matrices taking into account that the second matrix might be weighted by a scalar value beta:
  *
  * @note [ MTX_OUT = MTX_0 + beta * MTX_1 ] with MTX_0 and MTX_1 of the same size
  *

diff --git a/src/core/NEON/kernels/NEGEMMMatrixMultiplyKernel.h b/src/core/NEON/kernels/NEGEMMMatrixMultiplyKernel.h
index e2945ee..3bc162a 100644
--- a/src/core/NEON/kernels/NEGEMMMatrixMultiplyKernel.h
+++ b/src/core/NEON/kernels/NEGEMMMatrixMultiplyKernel.h

@@ -30,7 +30,7 @@
 {
 class ITensor;
 
-/** Neon kernel to multiply two input matrices "A" and "B". All elements of the output matrix/vector will be multiplied by alpha after the matrix multiplication
+/** Kernel to multiply two input matrices "A" and "B". All elements of the output matrix/vector will be multiplied by alpha after the matrix multiplication
  *
  * @note If the output tensor is a matrix, the implementation assumes that the input tensors @p input0 and @p input1 are both matrices and reshaped respectively with @ref NEGEMMInterleave4x4Kernel" and @ref NEGEMMTranspose1xWKernel
  * @note If the output tensor is a vector and the data type is F32, the implementation assumes that the first input tensor @p input0 is a vector and the second input tensor @p input1 a matrix. The implementation also assumes that both tensors have not been reshaped

diff --git a/src/core/NEON/kernels/NEGEMMTranspose1xWKernel.cpp b/src/core/NEON/kernels/NEGEMMTranspose1xWKernel.cpp
index ad7ae50..f6a453c 100644
--- a/src/core/NEON/kernels/NEGEMMTranspose1xWKernel.cpp
+++ b/src/core/NEON/kernels/NEGEMMTranspose1xWKernel.cpp

@@ -51,7 +51,7 @@
 {
     ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(input);
     ARM_COMPUTE_RETURN_ERROR_ON(input->data_type() == DataType::UNKNOWN);
-    //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use Neon FP16 instructions.
+    //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use CPU FP16 instructions.
 
     if(output->total_size() != 0)
     {

diff --git a/src/core/NEON/kernels/NEGEMMTranspose1xWKernel.h b/src/core/NEON/kernels/NEGEMMTranspose1xWKernel.h
index 583588a..7ca71cf41 100644
--- a/src/core/NEON/kernels/NEGEMMTranspose1xWKernel.h
+++ b/src/core/NEON/kernels/NEGEMMTranspose1xWKernel.h

@@ -31,7 +31,7 @@
 // Forward declarations
 class ITensor;
 
-/** Neon kernel which transposes the elements of a matrix in chunks of 1xW, where W is equal to (16 / element size of the tensor)
+/** Kernel which transposes the elements of a matrix in chunks of 1xW, where W is equal to (16 / element size of the tensor)
  *
  * Following an example of how the transposition1xW works when the input data is F32
  *

diff --git a/src/core/NEON/kernels/NEGatherKernel.h b/src/core/NEON/kernels/NEGatherKernel.h
index 46b41b2..0711f81 100644
--- a/src/core/NEON/kernels/NEGatherKernel.h
+++ b/src/core/NEON/kernels/NEGatherKernel.h

@@ -33,7 +33,7 @@
 // Forward declarations
 class ITensor;
 
-/** Kernel to perform other operation on Neon */
+/** Kernel to perform gather operation. */
 class NEGatherKernel : public INEKernel
 {
 public:

diff --git a/src/core/NEON/kernels/NEInstanceNormalizationLayerKernel.cpp b/src/core/NEON/kernels/NEInstanceNormalizationLayerKernel.cpp
index fa1f7a6..d33431a 100644
--- a/src/core/NEON/kernels/NEInstanceNormalizationLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEInstanceNormalizationLayerKernel.cpp

@@ -82,7 +82,7 @@
 template <typename T, typename AccType = T>
 void instance_normalization_nchw(ITensor *input, ITensor *output, float gamma, float beta, float epsilon, const Window &window)
 {
-    /** Neon vector tag type. */
+    /** SIMD vector tag type. */
     using ExactTagType = typename wrapper::traits::neon_bitvector_tag_t<T, wrapper::traits::BitWidth::W128>;
 
     // Clear X/Y dimensions on execution window as we handle the planes manually

diff --git a/src/core/NEON/kernels/NENormalizationLayerKernel.cpp b/src/core/NEON/kernels/NENormalizationLayerKernel.cpp
index 21a70ae..f1c5d3f 100644
--- a/src/core/NEON/kernels/NENormalizationLayerKernel.cpp
+++ b/src/core/NEON/kernels/NENormalizationLayerKernel.cpp

@@ -170,7 +170,7 @@
 template <typename T, unsigned int S, unsigned int dim, bool do_2D_norm>
 void NENormalizationLayerKernel::normalize_float(const Window &window)
 {
-    /** Neon vector tag type. */
+    /** SIMD vector tag type. */
     using ExactTagType = typename wrapper::traits::neon_vector<T, S>::tag_type;
 
     Window win(window);

diff --git a/src/core/NEON/kernels/NEPadLayerKernel.h b/src/core/NEON/kernels/NEPadLayerKernel.h
index af0dbfd..00cda7d 100644
--- a/src/core/NEON/kernels/NEPadLayerKernel.h
+++ b/src/core/NEON/kernels/NEPadLayerKernel.h

@@ -30,10 +30,7 @@
 {
 class ITensor;
 
-/** Neon kernel to add padding to a tensor
- *
- * Add padding given padding information
- */
+/** Basic kernel to pad the input tensor given padding information. */
 class NEPadLayerKernel : public INEKernel
 {
 public:

diff --git a/src/core/NEON/kernels/NEQLSTMLayerNormalizationKernel.h b/src/core/NEON/kernels/NEQLSTMLayerNormalizationKernel.h
index 5522ae8..a3ff6e9 100644
--- a/src/core/NEON/kernels/NEQLSTMLayerNormalizationKernel.h
+++ b/src/core/NEON/kernels/NEQLSTMLayerNormalizationKernel.h

@@ -31,7 +31,7 @@
 {
 class ITensor;
 
-/** Neon kernel to perform layer normalization */
+/** Kernel to perform layer normalization for QLSTM. */
 class NEQLSTMLayerNormalizationKernel : public INEKernel
 {
 public:

diff --git a/src/core/NEON/kernels/NERangeKernel.cpp b/src/core/NEON/kernels/NERangeKernel.cpp
index 170f58f..d0c51f8 100644
--- a/src/core/NEON/kernels/NERangeKernel.cpp
+++ b/src/core/NEON/kernels/NERangeKernel.cpp

@@ -43,7 +43,7 @@
 template <typename T>
 void range_function(ITensor *output, float start, float step, const Window &window)
 {
-    /** Neon vector tag type. */
+    /** SIMD vector tag type. */
     using ExactTagType = typename wrapper::traits::neon_bitvector<T, wrapper::traits::BitWidth::W128>::tag_type;
 
     const auto step_vec  = wrapper::vdup_n(static_cast<T>(step), ExactTagType{});

diff --git a/src/core/NEON/kernels/NEReductionOperationKernel.cpp b/src/core/NEON/kernels/NEReductionOperationKernel.cpp
index 875d233..553048c 100644
--- a/src/core/NEON/kernels/NEReductionOperationKernel.cpp
+++ b/src/core/NEON/kernels/NEReductionOperationKernel.cpp

@@ -386,7 +386,7 @@
 template <typename T, int S>
 struct RedOpX
 {
-    /** Neon vector tag type. */
+    /** SIMD vector tag type. */
     using ExactTagType = typename wrapper::traits::neon_vector<T, S>::tag_type;
 
     inline void operator()(const Window &in_window, Window &out_window, const ITensor *in, ITensor *out, const ReductionOperation op)
@@ -908,7 +908,7 @@
 template <typename T, int S>
 struct RedOpYZW
 {
-    /** Neon vector tag type. */
+    /** SIMD vector tag type. */
     using ExactTagType = typename wrapper::traits::neon_vector<T, S>::tag_type;
     using neon_vector  = typename wrapper::traits::neon_vector<T, S>::type;
 
@@ -1127,7 +1127,7 @@
 template <typename T, int S, int axis, ReductionOperation op>
 struct RedOpYZW_complex
 {
-    /** Neon vector tag type. */
+    /** SIMD vector tag type. */
     using ExactTagType = typename wrapper::traits::neon_vector<T, S>::tag_type;
     using neon_vector  = typename wrapper::traits::neon_vector<T, S>::type;
 

diff --git a/src/core/NEON/kernels/NEReductionOperationKernel.h b/src/core/NEON/kernels/NEReductionOperationKernel.h
index c7ed007..667305b 100644
--- a/src/core/NEON/kernels/NEReductionOperationKernel.h
+++ b/src/core/NEON/kernels/NEReductionOperationKernel.h

@@ -30,7 +30,7 @@
 {
 class ITensor;
 
-/** Neon kernel to perform a reduction operation
+/** Kernel to perform a reduction operation
  *
  * @note For ARG_MIN/ARG_MAX reduction, the default data type for an uninitialized
  *       output tensor is signed 32-bit integer (S32). It is the user's responsibility

diff --git a/src/core/NEON/kernels/NERemapKernel.h b/src/core/NEON/kernels/NERemapKernel.h
index adc7f4b..21cb67e 100644
--- a/src/core/NEON/kernels/NERemapKernel.h
+++ b/src/core/NEON/kernels/NERemapKernel.h

@@ -31,7 +31,7 @@
 {
 class ITensor;
 
-/** Neon kernel to perform a remap on a tensor */
+/** Kernel to perform a remap on a tensor */
 class NERemapKernel : public INEKernel
 {
 public:

diff --git a/src/core/NEON/kernels/NEReorgLayerKernel.cpp b/src/core/NEON/kernels/NEReorgLayerKernel.cpp
index 75a58fc..a7b830c 100644
--- a/src/core/NEON/kernels/NEReorgLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEReorgLayerKernel.cpp

@@ -42,7 +42,7 @@
 {
 Status validate_arguments(const ITensorInfo *input, const ITensorInfo *output, int32_t stride)
 {
-    //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use Neon FP16 instructions.
+    //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use CPU FP16 instructions.
     ARM_COMPUTE_RETURN_ERROR_ON(input->data_type() == DataType::UNKNOWN);
     ARM_COMPUTE_RETURN_ERROR_ON(input->data_layout() == DataLayout::UNKNOWN);
 

diff --git a/src/core/NEON/kernels/NEReverseKernel.cpp b/src/core/NEON/kernels/NEReverseKernel.cpp
index b2fce0f..758433f 100644
--- a/src/core/NEON/kernels/NEReverseKernel.cpp
+++ b/src/core/NEON/kernels/NEReverseKernel.cpp

@@ -37,7 +37,7 @@
 Status validate_arguments(const ITensorInfo *input, const ITensorInfo *output, const ITensorInfo *axis)
 {
     ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(input, output, axis);
-    //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use Neon FP16 instructions.
+    //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use CPU FP16 instructions.
     ARM_COMPUTE_RETURN_ERROR_ON(input->data_type() == DataType::UNKNOWN);
     ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(axis, 1, DataType::U32);
     ARM_COMPUTE_RETURN_ERROR_ON_MSG(axis->num_dimensions() > 1, "Axis must be a 1D tensor");

diff --git a/src/core/NEON/kernels/NEStackLayerKernel.cpp b/src/core/NEON/kernels/NEStackLayerKernel.cpp
index 64310e7..07ce829 100644
--- a/src/core/NEON/kernels/NEStackLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEStackLayerKernel.cpp

@@ -44,7 +44,7 @@
 Status validate_arguments(const ITensorInfo *input, unsigned int axis, unsigned int idx_input, unsigned int num_tensors, const ITensorInfo *output)
 {
     ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(input, output);
-    // Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use Neon FP16 instructions.
+    // Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use CPU FP16 instructions.
     ARM_COMPUTE_RETURN_ERROR_ON(input->data_type() == DataType::UNKNOWN);
     ARM_COMPUTE_RETURN_ERROR_ON(idx_input >= num_tensors);
     ARM_COMPUTE_RETURN_ERROR_ON(axis > input->num_dimensions());

diff --git a/src/core/NEON/kernels/NEStackLayerKernel.h b/src/core/NEON/kernels/NEStackLayerKernel.h
index dc5c7d7..9b36518 100644
--- a/src/core/NEON/kernels/NEStackLayerKernel.h
+++ b/src/core/NEON/kernels/NEStackLayerKernel.h

@@ -32,7 +32,7 @@
 {
 class ITensor;
 
-/** Neon kernel to stacks a rank-R tensor into one with rank-(R+1) along the axis dimension.*/
+/** Basic kernel to stack a rank-R tensor into one with rank-(R+1) along the axis dimension. */
 class NEStackLayerKernel : public INEKernel
 {
 public:

diff --git a/src/core/NEON/kernels/NETileKernel.h b/src/core/NEON/kernels/NETileKernel.h
index e6ce953..47f306a 100644
--- a/src/core/NEON/kernels/NETileKernel.h
+++ b/src/core/NEON/kernels/NETileKernel.h

@@ -30,7 +30,7 @@
 {
 class ITensor;
 
-/** Neon kernel to perform a tile operation */
+/** Basic kernel to perform a tile operation */
 class NETileKernel : public INEKernel
 {
 public:

diff --git a/src/core/NEON/kernels/NEWeightsReshapeKernel.cpp b/src/core/NEON/kernels/NEWeightsReshapeKernel.cpp
index 44d6009..9bef9c3 100644
--- a/src/core/NEON/kernels/NEWeightsReshapeKernel.cpp
+++ b/src/core/NEON/kernels/NEWeightsReshapeKernel.cpp

@@ -47,7 +47,7 @@
 Status validate_arguments(const ITensorInfo *input, const ITensorInfo *biases, const ITensorInfo *output)
 {
     ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(input, output);
-    //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use Neon FP16 instructions.
+    //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use CPU FP16 instructions.
     ARM_COMPUTE_RETURN_ERROR_ON(input->data_type() == DataType::UNKNOWN);
 
     if(biases != nullptr)

diff --git a/src/core/NEON/kernels/NEWeightsReshapeKernel.h b/src/core/NEON/kernels/NEWeightsReshapeKernel.h
index a4a9e28..76eca9f 100644
--- a/src/core/NEON/kernels/NEWeightsReshapeKernel.h
+++ b/src/core/NEON/kernels/NEWeightsReshapeKernel.h

@@ -30,7 +30,7 @@
 {
 class ITensor;
 
-/** Neon kernel to perform reshaping on the weights used by convolution and locally connected layer
+/** Kernel to perform reshaping on the weights used by convolution and locally connected layer
  *
  * Rearranges each 3-dimensional kernel to a single row leading to a matrix with linearized kernels.
  * In combination with the @ref NEIm2ColKernel can transform a convolution to a matrix multiplication.

diff --git a/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.h b/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.h
index 3583735..75d257d 100644
--- a/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.h
+++ b/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.h

@@ -35,7 +35,7 @@
 // Forward declarations
 class ITensor;
 
-/** Interface for the Neon kernel to perform Winograd input transform. */
+/** Interface for the kernel to perform Winograd input transform. */
 class INEWinogradLayerTransformInputKernel : public INEKernel
 {
 public:
@@ -96,7 +96,7 @@
     }
 };
 
-/** Neon kernel to perform Winograd input transform. */
+/** Kernel to perform Winograd input transform. */
 template <typename T, int OutputTileRows, int OutputTileCols, int KernelRows, int KernelCols>
 class NEWinogradLayerTransformInputKernel : public INEWinogradLayerTransformInputKernel
 {
@@ -227,7 +227,7 @@
     ITensor                        *_workspace;
 };
 
-/** Interface for the Neon kernel to perform Winograd output transform. */
+/** Interface for the kernel to perform Winograd output transform. */
 class INEWinogradLayerTransformOutputKernel : public INEKernel
 {
 public:
@@ -310,7 +310,7 @@
     }
 };
 
-/** Neon kernel to perform Winograd output transform. */
+/** Kernel to perform Winograd output transform. */
 template <typename T, int OutputTileRows, int OutputTileCols, int KernelRows, int KernelCols>
 class NEWinogradLayerTransformOutputKernel : public INEWinogradLayerTransformOutputKernel
 {
@@ -437,7 +437,7 @@
     int                              _num_channels;
 };
 
-/** Interface for the Neon kernel to perform Winograd weights transform. */
+/** Interface for the kernel to perform Winograd weights transform. */
 class INEWinogradLayerTransformWeightsKernel : public INEKernel
 {
 public:
@@ -495,7 +495,7 @@
     static Status validate(const ITensorInfo *input, const ITensorInfo *weights);
 };
 
-/** Neon kernel to perform Winograd weights transform. */
+/** Kernel to perform Winograd weights transform. */
 template <typename T, int OutputTileRows, int OutputTileCols, int KernelRows, int KernelCols>
 class NEWinogradLayerTransformWeightsKernel final : public INEWinogradLayerTransformWeightsKernel
 {
@@ -577,7 +577,7 @@
     int                               _num_input_channels;
 };
 
-/** Neon kernel to perform Winograd. */
+/** Kernel to perform Winograd. */
 template <typename TIn, typename TOut, int OutputTileRows, int OutputTileCols, int KernelRows, int KernelCols>
 class NEWinogradLayerConfiguration
 {

diff --git a/src/core/NEON/kernels/arm_gemm/gemm_fp32.cpp b/src/core/NEON/kernels/arm_gemm/gemm_fp32.cpp
index 5c894c0..6c3743d 100644
--- a/src/core/NEON/kernels/arm_gemm/gemm_fp32.cpp
+++ b/src/core/NEON/kernels/arm_gemm/gemm_fp32.cpp

@@ -111,7 +111,7 @@
 },
 #endif // __ARM_FEATURE_SVE
 
-// Neon hybrid methods
+// Arm® Neon™ hybrid methods
 {
     GemmMethod::GEMM_HYBRID,
     "a64_smallK_hybrid_fp32_mla_8x4",

diff --git a/src/core/NEON/kernels/arm_gemm/interleave_indirect.cpp b/src/core/NEON/kernels/arm_gemm/interleave_indirect.cpp
index bb86d9e..0d56b46 100644
--- a/src/core/NEON/kernels/arm_gemm/interleave_indirect.cpp
+++ b/src/core/NEON/kernels/arm_gemm/interleave_indirect.cpp

@@ -292,21 +292,21 @@
 /* AArch32 */
 #ifdef __arm__
 /* FP32 */
-/* Neon implementation (height 6) */
+/* Arm® Neon™ implementation (height 6) */
 template void IndirectInterleave<6, 1, VLType::None>(float *, const float * const * const *, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void ConvolutionInterleave<6, 1, VLType::None>(float *, const float *, size_t, const convolver<float> &, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void Interleave<6, 1, VLType::None>(float *, const float *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 
 /* FP16 */
 #if __ARM_FP16_ARGS
-/* Neon implementation using FP32 kernel (height 6) */
+/* Arm® Neon™ implementation using FP32 kernel (height 6) */
 template void IndirectInterleave<6, 1, VLType::None>(float *, const __fp16 * const * const *, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void ConvolutionInterleave<6, 1, VLType::None>(float *, const __fp16 *, size_t, const convolver<__fp16> &, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void Interleave<6, 1, VLType::None>(float *, const __fp16 *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 #endif /* __ARM_FP16_ARGS */
 
 /* BF16 */
-/* Neon implementation using FP32 kernel */
+/* Arm® Neon™ implementation using FP32 kernel */
 template void IndirectInterleave<6, 1, VLType::None>(float *, const bfloat16 * const * const *, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void ConvolutionInterleave<6, 1, VLType::None>(float *, const bfloat16 *, size_t, const convolver<bfloat16> &, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void Interleave<6, 1, VLType::None>(float *, const bfloat16 *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
@@ -315,7 +315,7 @@
 /* AArch64 */
 #ifdef __aarch64__
 /* FP32 */
-/* Neon/SVE implementation (height 8) */
+/* Arm® Neon™/SVE implementation (height 8) */
 template void IndirectInterleave<8, 1, VLType::None>(float *, const float * const * const *, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void ConvolutionInterleave<8, 1, VLType::None>(float *, const float *, size_t, const convolver<float> &, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void Interleave<8, 1, VLType::None>(float *, const float *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
@@ -339,7 +339,7 @@
 template void Interleave<8, 1, VLType::None>(float *, const __fp16 *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 
 /* BF16 */
-/* Neon/SVE BFDOT */
+/* Arm® Neon™/SVE BFDOT */
 #ifdef V8P6_BF
 template void IndirectInterleave<8, 2, VLType::None>(bfloat16 *, const bfloat16 * const * const *, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void ConvolutionInterleave<8, 2, VLType::None>(bfloat16 *, const bfloat16 *, size_t, const convolver<bfloat16> &, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
@@ -350,7 +350,7 @@
 template void Interleave<8, 4, VLType::None>(bfloat16 *, const bfloat16 *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 #endif // V8P6_BF
 
-/* Neon/SVE using FP32 kernel */
+/* Arm® Neon™/SVE using FP32 kernel */
 template void IndirectInterleave<8, 1, VLType::None>(float *, const bfloat16 * const * const *, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void ConvolutionInterleave<8, 1, VLType::None>(float *, const bfloat16 *, size_t, const convolver<bfloat16> &, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void Interleave<8, 1, VLType::None>(float *, const bfloat16 *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
@@ -365,12 +365,12 @@
 template void Interleave<8, 1, VLType::None>(uint16_t *, const uint16_t *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 
 /* INT8 */
-/* Neon SMLA/SMLAL (height 4, block 16) */
+/* Arm® Neon™ SMLA/SMLAL (height 4, block 16) */
 template void IndirectInterleave<4, 16, VLType::None>(int8_t *, const int8_t * const * const *, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void ConvolutionInterleave<4, 16, VLType::None>(int8_t *, const int8_t *, size_t, const convolver<int8_t> &, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void Interleave<4, 16, VLType::None>(int8_t *, const int8_t *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 
-/* Neon SDOT (height 8, block 4) */
+/* Arm® Neon™ SDOT (height 8, block 4) */
 template void IndirectInterleave<8, 4, VLType::None>(int8_t *, const int8_t * const * const *, unsigned int, unsigned int, unsigned int y0, unsigned int ymax, unsigned int k0, unsigned int kmax, bool, int32_t);
 template void ConvolutionInterleave<8, 4, VLType::None>(int8_t *, const int8_t *, size_t, const convolver<int8_t> &, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void Interleave<8, 4, VLType::None>(int8_t *, const int8_t *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
@@ -382,17 +382,17 @@
 template void Interleave<8, 8, VLType::None>(int8_t *, const int8_t *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 #endif // MMLA_INT8
 
-/* Neon SDOT (height 8, block 1) */
+/* Arm® Neon™ SDOT (height 8, block 1) */
 template void IndirectInterleave<8, 1, VLType::None>(int16_t *, const int8_t * const * const *, unsigned int, unsigned int, unsigned int y0, unsigned int ymax, unsigned int k0, unsigned int kmax, bool, int32_t);
 template void ConvolutionInterleave<8, 1, VLType::None>(int16_t *, const int8_t *, size_t, const convolver<int8_t> &, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void Interleave<8, 1, VLType::None>(int16_t *, const int8_t *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 
-/* Neon SMLA/SMLAL (height 4, block 16) */
+/* Arm® Neon™ SMLA/SMLAL (height 4, block 16) */
 template void IndirectInterleave<4, 16, VLType::None>(uint8_t *, const uint8_t * const * const *, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void ConvolutionInterleave<4, 16, VLType::None>(uint8_t *, const uint8_t *, size_t, const convolver<uint8_t> &, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void Interleave<4, 16, VLType::None>(uint8_t *, const uint8_t *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 
-/* Neon SDOT (height 8, block 4) */
+/* Arm® Neon™ SDOT (height 8, block 4) */
 template void IndirectInterleave<8, 4, VLType::None>(uint8_t *, const uint8_t * const * const *, unsigned int, unsigned int, unsigned int y0, unsigned int ymax, unsigned int k0, unsigned int kmax, bool, int32_t);
 template void ConvolutionInterleave<8, 4, VLType::None>(uint8_t *, const uint8_t *, size_t, const convolver<uint8_t> &, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void Interleave<8, 4, VLType::None>(uint8_t *, const uint8_t *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
@@ -404,7 +404,7 @@
 template void Interleave<8, 8, VLType::None>(uint8_t *, const uint8_t *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 #endif // MMLA_INT8
 
-/* Neon 16-bit (height 8, block 1) */
+/* Arm® Neon™ 16-bit (height 8, block 1) */
 template void IndirectInterleave<8, 1, VLType::None>(uint16_t *, const uint8_t * const * const *, unsigned int, unsigned int, unsigned int y0, unsigned int ymax, unsigned int k0, unsigned int kmax, bool, int32_t);
 template void ConvolutionInterleave<8, 1, VLType::None>(uint16_t *, const uint8_t *, size_t, const convolver<uint8_t> &, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void Interleave<8, 1, VLType::None>(uint16_t *, const uint8_t *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);

diff --git a/src/core/NEON/kernels/batchnormalization/impl/NEON/fp16.cpp b/src/core/NEON/kernels/batchnormalization/impl/NEON/fp16.cpp
index c105ada..ed5254a 100644
--- a/src/core/NEON/kernels/batchnormalization/impl/NEON/fp16.cpp
+++ b/src/core/NEON/kernels/batchnormalization/impl/NEON/fp16.cpp

@@ -44,7 +44,7 @@
 void batch_normalization(ITensor *src, ITensor *dst, const ITensor *mean, const ITensor *var, const ITensor *beta, const ITensor *gamma,
                          float epsilon, ActivationLayerInfo &act_info, const Window &window)
 {
-    /** Neon vector tag type. */
+    /** SIMD vector tag type. */
     using ExactTagType = typename wrapper::traits::neon_bitvector_tag_t<float16_t, wrapper::traits::BitWidth::W128>;
 
     const int  window_step_x  = 8;

diff --git a/src/core/NEON/kernels/batchnormalization/impl/NEON/fp32.cpp b/src/core/NEON/kernels/batchnormalization/impl/NEON/fp32.cpp
index 4a90a21..d6e22e1 100644
--- a/src/core/NEON/kernels/batchnormalization/impl/NEON/fp32.cpp
+++ b/src/core/NEON/kernels/batchnormalization/impl/NEON/fp32.cpp

@@ -43,7 +43,7 @@
 void batch_normalization(ITensor *src, ITensor *dst, const ITensor *mean, const ITensor *var, const ITensor *beta, const ITensor *gamma,
                          float epsilon, ActivationLayerInfo &act_info, const Window &window)
 {
-    /** Neon vector tag type. */
+    /** SIMD vector tag type. */
     using ExactTagType = typename wrapper::traits::neon_bitvector_tag_t<float, wrapper::traits::BitWidth::W128>;
 
     const int  window_step_x  = 4;

diff --git a/src/core/NEON/kernels/detail/NEActivationFunctionDetail.h b/src/core/NEON/kernels/detail/NEActivationFunctionDetail.h
index e68f111..25d682d 100644
--- a/src/core/NEON/kernels/detail/NEActivationFunctionDetail.h
+++ b/src/core/NEON/kernels/detail/NEActivationFunctionDetail.h

@@ -34,7 +34,7 @@
 template <typename T, int S>
 struct dummy
 {
-    /** Neon vector type. */
+    /** SIMD vector type. */
     using ExactType = typename wrapper::traits::neon_vector<T, S>::type;
 
     /** Construct a dummy activation object.
@@ -68,9 +68,9 @@
 template <typename T, int S>
 struct linear
 {
-    /** Neon vector type. */
+    /** SIMD vector type. */
     using ExactType = typename wrapper::traits::neon_vector<T, S>::type;
-    /** Neon vector tag type. */
+    /** SIMD vector tag type. */
     using ExactTagType = typename wrapper::traits::neon_vector<T, S>::tag_type;
 
     /** Construct a Linear activation object.
@@ -112,9 +112,9 @@
 template <typename T, int S>
 struct square
 {
-    /** Neon vector type. */
+    /** SIMD vector type. */
     using ExactType = typename wrapper::traits::neon_vector<T, S>::type;
-    /** Neon vector tag type. */
+    /** SIMD vector tag type. */
     using ExactTagType = typename wrapper::traits::neon_vector<T, S>::tag_type;
 
     /** Construct a Square activation object.
@@ -148,9 +148,9 @@
 template <typename T, int S>
 struct logistic
 {
-    /** Neon vector type. */
+    /** SIMD vector type. */
     using ExactType = typename wrapper::traits::neon_vector<T, S>::type;
-    /** Neon vector tag type. */
+    /** SIMD vector tag type. */
     using ExactTagType = typename wrapper::traits::neon_vector<T, S>::tag_type;
 
     /** Construct a Logistic activation object.
@@ -188,9 +188,9 @@
 template <typename T, int S>
 struct relu
 {
-    /** Neon vector type. */
+    /** SIMD vector type. */
     using ExactType = typename wrapper::traits::neon_vector<T, S>::type;
-    /** Neon vector tag type. */
+    /** SIMD vector tag type. */
     using ExactTagType = typename wrapper::traits::neon_vector<T, S>::tag_type;
 
     /** Construct a RELU activation object.
@@ -228,9 +228,9 @@
 template <typename T, int S>
 struct brelu
 {
-    /** Neon vector type. */
+    /** SIMD vector type. */
     using ExactType = typename wrapper::traits::neon_vector<T, S>::type;
-    /** Neon vector tag type. */
+    /** SIMD vector tag type. */
     using ExactTagType = typename wrapper::traits::neon_vector<T, S>::tag_type;
 
     /** Construct a bounded RELU activation object.
@@ -270,9 +270,9 @@
 template <typename T, int S>
 struct lubrelu
 {
-    /** Neon vector type. */
+    /** SIMD vector type. */
     using ExactType = typename wrapper::traits::neon_vector<T, S>::type;
-    /** Neon vector tag type. */
+    /** SIMD vector tag type. */
     using ExactTagType = typename wrapper::traits::neon_vector<T, S>::tag_type;
 
     /** Construct a lower-upper bounded RELU activation object.