COMPMID-2754: Add support for QASYMM8_SIGNED in NE kernels/functions.

Kernels/Functions extended support:
- NEBatchToSpaceLayerKernel/NEBatchToSpaceLayer
- NEChannelShuffleLayerKernel/NEChannelShuffleLayer
- NECol2ImKernel/NECol2Im
- NEConvertFullyConnectedWeightsKernel/NEConvertFullyConnectedWeights
- NECopyKernel/NECopy
- NEConvolutionLayerReshapeWeights
- NEDepthToSpaceLayerKernel/NEDepthToSpaceLayer
- NEFlattenLayerKernel/NEFlattenLayer
- NEFillBorderKernel
- NEFullyConnectedLayerReshapeWeights
- NEGatherKernel/NEGather
- NEGEMMInterleave4x4Kernel
- NEGEMMTranspose1xWKernel
- NEIm2ColKernel/NEIm2Col
- NEMemsetKernel
- NEPadLayerKernel/NEPadLayer
- NEPermuteKernel/NEPermute
- NEReverseKernel/NEReverse
- NEReorgLayerKernel/NEReorgLayer
- NEReshapeLayerKernel/NEReshapeLayer
- NESplit
- NESlice
- NEStridedSliceKernel/NEStridedSlice
- NESpaceToBatchLayerKernel/NESpaceToBatchLayer
- NESpaceToDepthLayerKernel/NESpaceToDepthLayerKernel
- NEStackLayerKernel/NEStackLayer
- NETileKernel/NETile
- NETransposeKernel/NETranspose
- NEWidthConcatenateLayerKernel/NEHeightConcatenateLayer
- NEHeightConcatenateLayerKernel/NEHeightConcatenateLayer
- NEDepthConcatenateLayerKernel/NEDepthConcatenateLayer
- NEBathConcatenateLayerKernel/NEBatchConcatenateLayer

Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Change-Id: Ia070332ad4c4dbced2541dc46f7f2f3a86833b65
Reviewed-on: https://review.mlplatform.org/c/2442
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>

28 files changed, 74 insertions, 61 deletions

diff --git a/arm_compute/core/NEON/kernels/NEBatchConcatenateLayerKernel.h b/arm_compute/core/NEON/kernels/NEBatchConcatenateLayerKernel.h
index 69223eea6a..e3a7847025 100644
--- a/arm_compute/core/NEON/kernels/NEBatchConcatenateLayerKernel.h
+++ b/arm_compute/core/NEON/kernels/NEBatchConcatenateLayerKernel.h
@@ -29,6 +29,7 @@
 
 namespace arm_compute
 {
+// Forward declarations
 class ITensor;
 
 /** Interface for the batch concatenate kernel.
@@ -55,7 +56,7 @@ public:
     ~NEBatchConcatenateLayerKernel() = default;
     /** Initialise the kernel's inputs and output
      *
-     * @param[in]     input        Input tensor. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in]     input        Input tensor. Data types supported: All.
      * @param[in]     batch_offset The offset on axis # 3.
      * @param[in,out] output       Output tensor. Data types supported: Same as @p input.
      *
@@ -66,7 +67,7 @@ public:
     void configure(const ITensor *input, unsigned int batch_offset, ITensor *output);
     /**  Static function to check if given info will lead to a valid configuration of @ref NEBatchConcatenateLayerKernel
      *
-     * @param[in] input        Input tensor info. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in] input        Input tensor info. Data types supported: All.
      * @param[in] batch_offset The offset on axis # 3.
      * @param[in] output       Output tensor info. Data types supported: Same as @p input.
      *
diff --git a/arm_compute/core/NEON/kernels/NEBatchToSpaceLayerKernel.h b/arm_compute/core/NEON/kernels/NEBatchToSpaceLayerKernel.h
index cd1fa7e443..e9bbf4eb03 100644
--- a/arm_compute/core/NEON/kernels/NEBatchToSpaceLayerKernel.h
+++ b/arm_compute/core/NEON/kernels/NEBatchToSpaceLayerKernel.h
@@ -52,14 +52,14 @@ public:
     ~NEBatchToSpaceLayerKernel() = default;
     /** Initialise the kernel's inputs and output.
      *
-     * @param[in]  input       Tensor input. Supported tensor rank: 4. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in]  input       Tensor input. Supported tensor rank: 4. Data types supported: All.
      * @param[in]  block_shape 1-D tensor with shape [M]. Data types supported: S32
      * @param[out] output      Tensor output. Data types supported: same as @p input
      */
     void configure(const ITensor *input, const ITensor *block_shape, ITensor *output);
     /** Initialise the kernel's inputs and output (Static block shape).
      *
-     * @param[in]  input         Tensor input. Supported tensor rank: 4. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in]  input         Tensor input. Supported tensor rank: 4. Data types supported: All.
      * @param[in]  block_shape_x Block shape x value.
      * @param[in]  block_shape_y Block shape y value.
      * @param[out] output        Tensor output. Data types supported: same as @p input
@@ -67,7 +67,7 @@ public:
     void configure(const ITensor *input, const int32_t block_shape_x, const int32_t block_shape_y, ITensor *output);
     /** Static function to check if given info will lead to a valid configuration of @ref NEBatchToSpaceLayerKernel
      *
-     * @param[in] input       Tensor input. Supported tensor rank: 4. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in] input       Tensor input. Supported tensor rank: 4. Data types supported: All.
      * @param[in] block_shape 1-D tensor with shape [M]. Data types supported: S32
      * @param[in] output      Tensor output. Data types supported: same as @p input
      *
@@ -76,7 +76,7 @@ public:
     static Status validate(const ITensorInfo *input, const ITensorInfo *block_shape, const ITensorInfo *output);
     /** Static function to check if given info will lead to a valid configuration of @ref NEBatchToSpaceLayerKernel (Static block shape).
      *
-     * @param[in] input         Tensor input. Supported tensor rank: 4. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in] input         Tensor input. Supported tensor rank: 4. Data types supported: All.
      * @param[in] block_shape_x Block shape x value.
      * @param[in] block_shape_y Block shape y value.
      * @param[in] output        Tensor output. Data types supported: same as @p input
diff --git a/arm_compute/core/NEON/kernels/NEChannelShuffleLayerKernel.h b/arm_compute/core/NEON/kernels/NEChannelShuffleLayerKernel.h
index b6ec60030f..71659c4fcb 100644
--- a/arm_compute/core/NEON/kernels/NEChannelShuffleLayerKernel.h
+++ b/arm_compute/core/NEON/kernels/NEChannelShuffleLayerKernel.h
@@ -53,14 +53,14 @@ public:
     ~NEChannelShuffleLayerKernel() = default;
     /** Configure function's inputs and outputs.
      *
-     * @param[in]  input      Input tensor. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     * @param[in]  input      Input tensor. Data types supported: All
      * @param[out] output     Output tensor. Data type supported: Same as @p input
      * @param[in]  num_groups Number of groups. Must be greater than 1 and the number of channels of the tensors must be a multiple of the number of groups.
      */
     void configure(const ITensor *input, ITensor *output, unsigned int num_groups);
     /** Static function to check if given info will lead to a valid configuration of @ref NEChannelShuffleLayerKernel
      *
-     * @param[in]  input      Input tensor. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     * @param[in]  input      Input tensor. Data types supported: All
      * @param[out] output     Output tensor. Data type supported: Same as @p input
      * @param[in]  num_groups Number of groups. Must be greater than 1 and the number of channels of the tensors must be a multiple of the number of groups.
      *
diff --git a/arm_compute/core/NEON/kernels/NECol2ImKernel.h b/arm_compute/core/NEON/kernels/NECol2ImKernel.h
index 7c19f08b3b..9aa1062622 100644
--- a/arm_compute/core/NEON/kernels/NECol2ImKernel.h
+++ b/arm_compute/core/NEON/kernels/NECol2ImKernel.h
@@ -72,7 +72,7 @@ public:
 
     /** Set the input and output of the kernel.
      *
-     * @param[in]  input          The input tensor to convert. Data types supported: Any
+     * @param[in]  input          The input tensor to convert. Data types supported: All
      * @param[out] output         The output tensor. 3 lower dimensions represent a single output [width, height, OFM],
      *                            while the rest represent batch of outputs. Data types supported: Same as @p input
      * @param[in]  convolved_dims Output convolved dimensions.
@@ -80,7 +80,7 @@ public:
     void configure(const ITensor *input, ITensor *output, const Size2D &convolved_dims);
     /** Static function to check if given info will lead to a valid configuration of @ref NECol2ImKernel
      *
-     * @param[in] input          The input tensor to convert. Data types supported: Any
+     * @param[in] input          The input tensor to convert. Data types supported: All
      * @param[in] output         The output tensor. 3 lower dimensions represent a single output [width, height, OFM],
      *                           while the rest represent batch of outputs. Data types supported: Same as @p input
      * @param[in] convolved_dims Output convolved dimensions.
diff --git a/arm_compute/core/NEON/kernels/NEConvertFullyConnectedWeightsKernel.h b/arm_compute/core/NEON/kernels/NEConvertFullyConnectedWeightsKernel.h
index c54339da72..d45191949a 100644
--- a/arm_compute/core/NEON/kernels/NEConvertFullyConnectedWeightsKernel.h
+++ b/arm_compute/core/NEON/kernels/NEConvertFullyConnectedWeightsKernel.h
@@ -28,6 +28,7 @@
 
 namespace arm_compute
 {
+// Forward declarations
 class ITensor;
 
 /** Interface to convert the 2D Fully Connected weights from NCHW to NHWC or vice versa.
@@ -59,7 +60,7 @@ public:
     ~NEConvertFullyConnectedWeightsKernel() = default;
     /** Set the input and output tensor.
      *
-     * @param[in]  input                Source weights tensor to convert. Must be 2 dimensional. Data types supported: U8/S8/QASYMM8/U16/S16/U32/S32/F16/F32.
+     * @param[in]  input                Source weights tensor to convert. Must be 2 dimensional. Data types supported: All.
      * @param[out] output               The converted weights tensor. Shape and Data Type: Same as @p input.
      * @param[in]  original_input_shape Shape of the original input tensor (the one entering fully connected layer).
      * @param[in]  data_layout          The data layout the weights have been trained in.
@@ -67,7 +68,7 @@ public:
     void configure(const ITensor *input, ITensor *output, const TensorShape &original_input_shape, DataLayout data_layout);
     /** Static function to check if given info will lead to a valid configuration of @ref NEConvertFullyConnectedWeightsKernel
      *
-     * @param[in] input                Source weights tensor info to convert. Must be 2 dimensional. Data types supported: U8/S8/QASYMM8/U16/S16/U32/S32/F16/F32.
+     * @param[in] input                Source weights tensor info to convert. Must be 2 dimensional. Data types supported: All.
      * @param[in] output               The converted weights tensor info. Shape and Data Type: Same as @p input.
      * @param[in] original_input_shape Shape of the original input tensor (the one entering fully connected layer).
      * @param[in] data_layout          The data layout the weights have been trained in.
diff --git a/arm_compute/core/NEON/kernels/NECopyKernel.h b/arm_compute/core/NEON/kernels/NECopyKernel.h
index db3f6a8ae8..d2dbbaef98 100644
--- a/arm_compute/core/NEON/kernels/NECopyKernel.h
+++ b/arm_compute/core/NEON/kernels/NECopyKernel.h
@@ -51,14 +51,14 @@ public:
     NECopyKernel &operator=(NECopyKernel &&) = default;
     /** Initialize the kernel's input, output.
      *
-     * @param[in]  input   Source tensor. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in]  input   Source tensor. Data types supported: All
      * @param[out] output  Destination tensor. Data types supported: same as @p input.
      * @param[in]  padding (Optional) Padding to be applied to the input tensor
      */
     void configure(const ITensor *input, ITensor *output, const PaddingList &padding = PaddingList());
     /** Static function to check if given info will lead to a valid configuration of @ref NECopyKernel
      *
-     * @param[in] input   Source tensor. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in] input   Source tensor. Data types supported: All
      * @param[in] output  Destination tensor. Data types supported: same as @p input.
      * @param[in] padding (Optional) Padding to be applied to the input tensor
      *
diff --git a/arm_compute/core/NEON/kernels/NEDepthConcatenateLayerKernel.h b/arm_compute/core/NEON/kernels/NEDepthConcatenateLayerKernel.h
index 4fcaa24707..ddbd0983e4 100644
--- a/arm_compute/core/NEON/kernels/NEDepthConcatenateLayerKernel.h
+++ b/arm_compute/core/NEON/kernels/NEDepthConcatenateLayerKernel.h
@@ -29,6 +29,7 @@
 
 namespace arm_compute
 {
+// Forward declarations
 class ITensor;
 
 /** Interface for the depth concatenate kernel.
@@ -55,7 +56,7 @@ public:
     ~NEDepthConcatenateLayerKernel() = default;
     /** Initialise the kernel's inputs and output
      *
-     * @param[in]     input        Input tensor. Data types supported: QASYMM8/F16/F32.
+     * @param[in]     input        Input tensor. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32.
      * @param[in]     depth_offset The offset on the Z axis.
      * @param[in,out] output       Output tensor. Data types supported: Same as @p input.
      *
@@ -66,7 +67,7 @@ public:
     void configure(const ITensor *input, unsigned int depth_offset, ITensor *output);
     /**  Static function to check if given info will lead to a valid configuration of @ref NEDepthConcatenateLayerKernel
      *
-     * @param[in] input        Input tensor info. Data types supported: QASYMM8/F16/F32.
+     * @param[in] input        Input tensor info. Data types supported:  QASYMM8/QASYMM8_SIGNED/F16/F32.
      * @param[in] depth_offset The offset on the Z axis.
      * @param[in] output       Output tensor info. Data types supported: Same as @p input.
      *
diff --git a/arm_compute/core/NEON/kernels/NEDepthToSpaceLayerKernel.h b/arm_compute/core/NEON/kernels/NEDepthToSpaceLayerKernel.h
index f27a56c779..3129ea9fb9 100644
--- a/arm_compute/core/NEON/kernels/NEDepthToSpaceLayerKernel.h
+++ b/arm_compute/core/NEON/kernels/NEDepthToSpaceLayerKernel.h
@@ -28,6 +28,7 @@
 
 namespace arm_compute
 {
+// Forward declarations
 class ITensor;
 
 /** Interface for the depth to space kernel */
@@ -52,14 +53,14 @@ public:
     ~NEDepthToSpaceLayerKernel() = default;
     /** Initialise the kernel's inputs and output.
      *
-     * @param[in]  input       Tensor input. Supported tensor rank: 4. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in]  input       Tensor input. Supported tensor rank: 4. Data types supported: All
      * @param[out] output      Tensor output. Data types supported: same as @p input
      * @param[in]  block_shape Block shape x value.
      */
     void configure(const ITensor *input, ITensor *output, int32_t block_shape);
     /** Static function to check if given info will lead to a valid configuration of @ref NEDepthToSpaceLayerKernel.
      *
-     * @param[in] input       Tensor input info. Supported tensor rank: 4. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in] input       Tensor input info. Supported tensor rank: 4. Data types supported: All
      * @param[in] output      Tensor output info. Data types supported: same as @p input
      * @param[in] block_shape Block shape value.
      *
diff --git a/arm_compute/core/NEON/kernels/NEFillBorderKernel.h b/arm_compute/core/NEON/kernels/NEFillBorderKernel.h
index b345ec4cc9..0c852e8232 100644
--- a/arm_compute/core/NEON/kernels/NEFillBorderKernel.h
+++ b/arm_compute/core/NEON/kernels/NEFillBorderKernel.h
@@ -30,6 +30,7 @@
 
 namespace arm_compute
 {
+// Forward declarations
 class ITensor;
 
 /** Interface for the kernel to fill borders */
@@ -57,7 +58,7 @@ public:
      *
      * @note This kernel fills the borders within the XY-planes.
      *
-     * @param[in,out] tensor                Tensor to process. Data types supported: U8/S8/QASYMM8/S16/S32/F32.
+     * @param[in,out] tensor                Tensor to process. Data types supported: All.
      * @param[in]     border_size           Size of the border to fill in elements.
      * @param[in]     border_mode           Border mode to use for the convolution.
      * @param[in]     constant_border_value (Optional) Constant value to use for borders if border_mode is set to CONSTANT.
diff --git a/arm_compute/core/NEON/kernels/NEFlattenLayerKernel.h b/arm_compute/core/NEON/kernels/NEFlattenLayerKernel.h
index 8a7597a929..ba2f99857f 100644
--- a/arm_compute/core/NEON/kernels/NEFlattenLayerKernel.h
+++ b/arm_compute/core/NEON/kernels/NEFlattenLayerKernel.h
@@ -54,7 +54,7 @@ public:
     /** Set the input and output of the kernel.
      *
      * @param[in]  input  First input tensor to flatten with at least 3 dimensions.
-     *                    The dimensions above the third will be interpreted as batches. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     *                    The dimensions above the third will be interpreted as batches. Data types supported: All
      * @param[out] output Output tensor with shape [w*h*d, input_batches] where:
      *                    w = width input tensor, h = height input tensor and d = depth input tensor. Data type supported: same as @p input
      */
@@ -62,7 +62,7 @@ public:
     /** Static function to check if given info will lead to a valid configuration of @ref NEFlattenLayerKernel
      *
      * @param[in]  input  First input tensor to flatten with at least 3 dimensions.
-     *                    The dimensions above the third will be interpreted as batches. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     *                    The dimensions above the third will be interpreted as batches. Data types supported: All
      * @param[out] output Output tensor with shape [w*h*d, input_batches] where:
      *                    w = width input tensor, h = height input tensor and d = depth input tensor. Data type supported: same as @p input
      *
diff --git a/arm_compute/core/NEON/kernels/NEGEMMTranspose1xWKernel.h b/arm_compute/core/NEON/kernels/NEGEMMTranspose1xWKernel.h
index 0e2cf0e82d..967a1b73dc 100644
--- a/arm_compute/core/NEON/kernels/NEGEMMTranspose1xWKernel.h
+++ b/arm_compute/core/NEON/kernels/NEGEMMTranspose1xWKernel.h
@@ -28,6 +28,7 @@
 
 namespace arm_compute
 {
+// 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)
@@ -74,13 +75,13 @@ public:
     }
     /** Initialise the kernel's input and output.
      *
-     * @param[in]  input  Input tensor. Data types supported: U8/S8/QASYMM8/QSYMM8_PER_CHANNEL/U16/S16/F16/U32/S32/F32
+     * @param[in]  input  Input tensor. Data types supported: All
      * @param[out] output Output tensor. Data type supported: same as @p input.
      */
     void configure(const ITensor *input, ITensor *output);
     /** Static function to check if given info will lead to a valid configuration of @ref NEGEMMTranspose1xWKernel
      *
-     * @param[in] input  Input tensor info. Data types supported: U8/S8/QASYMM8/QSYMM8_PER_CHANNEL/U16/S16/F16/U32/S32/F32
+     * @param[in] input  Input tensor info. Data types supported: All
      * @param[in] output Output tensor info. Data type supported: same as @p input.
      *
      * @return a status
diff --git a/arm_compute/core/NEON/kernels/NEGatherKernel.h b/arm_compute/core/NEON/kernels/NEGatherKernel.h
index e619ee3574..bfef40b53b 100644
--- a/arm_compute/core/NEON/kernels/NEGatherKernel.h
+++ b/arm_compute/core/NEON/kernels/NEGatherKernel.h
@@ -30,6 +30,7 @@
 
 namespace arm_compute
 {
+// Forward declarations
 class ITensor;
 
 /** Kernel to perform other operation on NEON */
@@ -59,7 +60,7 @@ public:
     }
     /** Initialise the kernel's inputs and outputs
      *
-     * @param[in]  input   Source tensor. Supported tensor rank: up to 4. Data type supported: U8/S8/QASYMM8/U16/S16/U32/S32/F16/F32
+     * @param[in]  input   Source tensor. Supported tensor rank: up to 4. Data type supported: All
      * @param[in]  indices Indices tensor. Supported tensor rank: up to 1. Must be one of the following type: U32/S32. Each value Must be in range [0, input.shape[@p axis])
      * @param[out] output  Destination tensor. Data type supported: Same as @p input
      * @param[in]  axis    (Optional) The axis in @p input to gather @p indices from. Negative values wrap around. Defaults to 0
@@ -67,7 +68,7 @@ public:
     void configure(const ITensor *input, const ITensor *indices, ITensor *output, int axis = 0);
     /** Static function to check if given info will lead to a valid configuration of @ref NEGatherKernel
      *
-     * @param[in] input   Source tensor info. Supported tensor rank: up to 4. Data type supported: U8/S8/QASYMM8/U16/S16/U32/S32/F16/F32
+     * @param[in] input   Source tensor info. Supported tensor rank: up to 4. Data type supported: All
      * @param[in] indices Indices tensor info. Supported tensor rank: up to 1. Must be one of the following type: U32/S32. Each value Must be in range [0, input.shape[@p axis])
      * @param[in] output  Destination tensor info. Data type supported: Same as @p input
      * @param[in] axis    (Optional) The axis in @p input to gather @p indices from. Negative values wrap around. Defaults to 0
diff --git a/arm_compute/core/NEON/kernels/NEHeightConcatenateLayerKernel.h b/arm_compute/core/NEON/kernels/NEHeightConcatenateLayerKernel.h
index 91370bca41..be81f2e963 100644
--- a/arm_compute/core/NEON/kernels/NEHeightConcatenateLayerKernel.h
+++ b/arm_compute/core/NEON/kernels/NEHeightConcatenateLayerKernel.h
@@ -30,6 +30,7 @@
 
 namespace arm_compute
 {
+// Forward declarations
 class ITensor;
 
 /** Interface for the height concatenate kernel.
@@ -56,7 +57,7 @@ public:
     ~NEHeightConcatenateLayerKernel() = default;
     /** Initialise the kernel's inputs and output
      *
-     * @param[in]     input         Input tensor. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     * @param[in]     input         Input tensor. Data types supported: All
      * @param[in]     height_offset The starting offset on the Y axis for the output tensor.
      * @param[in,out] output        Output tensor. Data types supported: Same as @p input.
      *
@@ -64,7 +65,7 @@ public:
     void configure(const ITensor *input, unsigned int height_offset, ITensor *output);
     /**  Static function to check if given info will lead to a valid configuration of @ref NEHeightConcatenateLayerKernel
      *
-     * @param[in] input         Input tensor info. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     * @param[in] input         Input tensor info. Data types supported: All
      * @param[in] height_offset The starting offset on the Y axis for the output tensor.
      * @param[in] output        Output tensor info. Data types supported: Same as @p input.
      *
diff --git a/arm_compute/core/NEON/kernels/NEIm2ColKernel.h b/arm_compute/core/NEON/kernels/NEIm2ColKernel.h
index c9d22da254..8ee9660b95 100644
--- a/arm_compute/core/NEON/kernels/NEIm2ColKernel.h
+++ b/arm_compute/core/NEON/kernels/NEIm2ColKernel.h
@@ -77,7 +77,7 @@ public:
     /** Set the input and output of the kernel.
      *
      * @param[in]  input       The input tensor to convert. 3 lower dimensions represent a single input [width, height, IFM],
-     *                         while every optional dimension from 4 and above represent a batch of inputs. Data types supported: QASYMM8/F16/F32
+     *                         while every optional dimension from 4 and above represent a batch of inputs. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32
      *                         Note: QASYMM8 works only for has_bias = false
      * @param[out] output      The output tensor. Data types supported: Same as @p input
      * @param[in]  kernel_dims The kernel dimensions (width and height).
@@ -91,7 +91,7 @@ public:
     /** Static function to check if given info will lead to a valid configuration of @ref NEIm2ColKernel
      *
      * @param[in] input       The input tensor to convert. 3 lower dimensions represent a single input [width, height, IFM],
-     *                        while every optional dimension from 4 and above represent a batch of inputs. Data types supported: QASYMM8/F16/F32
+     *                        while every optional dimension from 4 and above represent a batch of inputs. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32
      *                        Note: QASYMM8 works only for has_bias = false
      * @param[in] output      The output tensor. Data types supported: Same as @p input
      * @param[in] kernel_dims The kernel dimensions (width and height).
diff --git a/arm_compute/core/NEON/kernels/NEMemsetKernel.h b/arm_compute/core/NEON/kernels/NEMemsetKernel.h
index 1ff302d6ed..b4bcd11b82 100644
--- a/arm_compute/core/NEON/kernels/NEMemsetKernel.h
+++ b/arm_compute/core/NEON/kernels/NEMemsetKernel.h
@@ -55,7 +55,7 @@ public:
     NEMemsetKernel &operator=(NEMemsetKernel &&) = default;
     /** Initialise the kernel's tensor and filling value
      *
-     * @param[in,out] tensor         Input tensor to fill. Supported data types: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     * @param[in,out] tensor         Input tensor to fill. Supported data types: All
      * @param[in]     constant_value The value used to fill the planes of the tensor
      */
     void configure(ITensor *tensor, const PixelValue &constant_value);
diff --git a/arm_compute/core/NEON/kernels/NEPadLayerKernel.h b/arm_compute/core/NEON/kernels/NEPadLayerKernel.h
index 8d58fa9c65..4cbefbd1e3 100644
--- a/arm_compute/core/NEON/kernels/NEPadLayerKernel.h
+++ b/arm_compute/core/NEON/kernels/NEPadLayerKernel.h
@@ -56,7 +56,7 @@ public:
 
     /** Initialize the function
      *
-     * @param[in]  input          Source tensor. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in]  input          Source tensor. Data types supported: All.
      * @param[out] output         Output tensor. Data type supported: same as @p input
      * @param[in]  padding        The padding for each spatial dimension of the input tensor. The pair padding[i]
      *                            specifies the front and the end padding in the i-th dimension.
@@ -67,7 +67,7 @@ public:
     void configure(ITensor *input, ITensor *output, const PaddingList &padding, const PixelValue constant_value = PixelValue(), const PaddingMode mode = PaddingMode::CONSTANT);
     /**  Static function to check if given info will lead to a valid configuration of @ref NEPadLayer.
      *
-     * @param[in] input          Source tensor info. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in] input          Source tensor info. Data types supported: All.
      * @param[in] output         Output tensor info. Data type supported: same as @p input
      * @param[in] padding        The padding for each spatial dimension of the input tensor. The pair padding[i]
      *                           specifies the front and the end padding in the i-th dimension.
diff --git a/arm_compute/core/NEON/kernels/NEPermuteKernel.h b/arm_compute/core/NEON/kernels/NEPermuteKernel.h
index 794259c262..89dc4e6fc7 100644
--- a/arm_compute/core/NEON/kernels/NEPermuteKernel.h
+++ b/arm_compute/core/NEON/kernels/NEPermuteKernel.h
@@ -28,6 +28,7 @@
 
 namespace arm_compute
 {
+// Forward declarations
 class ITensor;
 
 /** NEON kernel to perform tensor permutation.
@@ -58,7 +59,7 @@ public:
      *
      * @note Arbitrary permutation vectors are supported with rank not greater than 4
      *
-     * @param[in]  input  The input tensor to permute. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     * @param[in]  input  The input tensor to permute. Data types supported: All
      * @param[out] output The output tensor. Data types supported: Same as @p input
      * @param[in]  perm   Permutation vector
      */
@@ -67,7 +68,7 @@ public:
      *
      * @note Arbitrary permutation vectors are supported with rank not greater than 4
      *
-     * @param[in] input  The input tensor to permute. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     * @param[in] input  The input tensor to permute. Data types supported: All
      * @param[in] output The output tensor. Data types supported: Same as @p input
      * @param[in] perm   Permutation vector
      *
diff --git a/arm_compute/core/NEON/kernels/NEReorgLayerKernel.h b/arm_compute/core/NEON/kernels/NEReorgLayerKernel.h
index 195f0c2336..9277ddbe47 100644
--- a/arm_compute/core/NEON/kernels/NEReorgLayerKernel.h
+++ b/arm_compute/core/NEON/kernels/NEReorgLayerKernel.h
@@ -53,7 +53,7 @@ public:
     ~NEReorgLayerKernel() = default;
     /** Set the input and output of the kernel
      *
-     * @param[in]  input  Source tensor. Data type supported: U8/S8/U16/S16/QASYMM8/U32/S32/F16/F32
+     * @param[in]  input  Source tensor. Data type supported: All
      * @param[out] output Destination tensor. Data type supported: Same as @p input
      * @param[in]  stride Stride to be used during data re-organization.
      *                    It defines the spatial distance between 2 consecutive pixels in the x and y direction
@@ -62,7 +62,7 @@ public:
 
     /** Static function to check if given info will lead to a valid configuration of @ref NEReshapeLayerKernel
      *
-     * @param[in] input  Source tensor info. Data type supported: U8/S8/U16/S16/QASYMM8/U32/S32/F16/F32
+     * @param[in] input  Source tensor info. Data type supported: All
      * @param[in] output Destination tensor info. Data type supported: Same as @p input
      * @param[in] stride Stride to be used during data re-organization
      *                   It defines the spatial distance between 2 consecutive pixels in the x and y direction
diff --git a/arm_compute/core/NEON/kernels/NEReshapeLayerKernel.h b/arm_compute/core/NEON/kernels/NEReshapeLayerKernel.h
index 2871bc2fb2..fccf2685a8 100644
--- a/arm_compute/core/NEON/kernels/NEReshapeLayerKernel.h
+++ b/arm_compute/core/NEON/kernels/NEReshapeLayerKernel.h
@@ -28,6 +28,7 @@
 
 namespace arm_compute
 {
+// Forward declarations
 class ITensor;
 
 /** Interface for the kernel to perform tensor reshaping */
@@ -40,14 +41,14 @@ public:
     }
     /** Set the input and output of the kernel
      *
-     * @param[in]  input  Source tensor. Data type supported: U8/S8/U16/S16/QASYMM8/U32/S32/F16/F32
+     * @param[in]  input  Source tensor. Data type supported: All
      * @param[out] output Destination tensor. Data type supported: Same as @p input
      */
     void configure(const ITensor *input, ITensor *output);
 
     /** Static function to check if given info will lead to a valid configuration of @ref NEReshapeLayerKernel
      *
-     * @param[in] input  Source tensor info. Data type supported: U8/S8/U16/S16/QASYMM8/U32/S32/F16/F32
+     * @param[in] input  Source tensor info. Data type supported: All
      * @param[in] output Destination tensor info. Data type supported: Same as @p input
      *
      * @return a status
diff --git a/arm_compute/core/NEON/kernels/NEReverseKernel.h b/arm_compute/core/NEON/kernels/NEReverseKernel.h
index a8a1fdaac8..516653b70d 100644
--- a/arm_compute/core/NEON/kernels/NEReverseKernel.h
+++ b/arm_compute/core/NEON/kernels/NEReverseKernel.h
@@ -52,7 +52,7 @@ public:
     ~NEReverseKernel() = default;
     /** Initialise the kernel's inputs and output
      *
-     * @param[in]  input  Input tensor. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     * @param[in]  input  Input tensor. Data types supported: All
      * @param[out] output Output tensor. Data type supported: Same as @p input
      * @param[in]  axis   Axis tensor. Contains the indices of the dimensions to reverse. Data type supported: U32
      */
@@ -60,7 +60,7 @@ public:
 
     /** Static function to check if given info will lead to a valid configuration of @ref NEReverseKernel
      *
-     * @param[in] input  Input tensor info. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     * @param[in] input  Input tensor info. Data types supported: All
      * @param[in] output Output tensor info. Data type supported: Same as @p input
      * @param[in] axis   Axis tensor info. Contains the indices of the dimensions to reverse. Data type supported: U32
      *
diff --git a/arm_compute/core/NEON/kernels/NESelectKernel.h b/arm_compute/core/NEON/kernels/NESelectKernel.h
index 1711246b27..51c8543ddc 100644
--- a/arm_compute/core/NEON/kernels/NESelectKernel.h
+++ b/arm_compute/core/NEON/kernels/NESelectKernel.h
@@ -29,6 +29,7 @@
 
 namespace arm_compute
 {
+// Forward declarations
 class ITensor;
 
 /** Interface for the select kernel
@@ -60,7 +61,7 @@ public:
     /** Common signature for all the specialised elementwise functions
      *
      * @param[in]  c      Condition input tensor. Data types supported: U8.
-     * @param[in]  x      First input tensor. Data types supported: U8/S8/U16/S16/U32/S32/F16/F32.
+     * @param[in]  x      First input tensor. Data types supported: All.
      * @param[out] y      Second input tensor. Data types supported: Same as @p x
      * @param[in]  output Output tensor. Data types supported: Same as @p x
      */
@@ -69,7 +70,7 @@ public:
     /** Validate the argument passed to the kernel
      *
      * @param[in] c      Condition input tensor. Data types supported: U8.
-     * @param[in] x      First input tensor. Data types supported: U8/S8/U16/S16/U32/S32/F16/F32.
+     * @param[in] x      First input tensor. Data types supported: All.
      * @param[in] y      Second input tensor. Data types supported: Same as @p x
      * @param[in] output Output tensor. Data types supported: Same as @p x.
      *
@@ -84,7 +85,7 @@ private:
     /** Common signature for all the specialised select functions
      *
      * @param[in] c      Condition input tensor. Data types supported: U8.
-     * @param[in] x      First input tensor. Data types supported: U8/S8/U16/S16/U32/S32/F16/F32.
+     * @param[in] x      First input tensor. Data types supported: All.
      * @param[in] y      Second input tensor. Data types supported: Same as @p x
      * @param[in] output Output tensor. Data types supported: Same as @p x.
      */
diff --git a/arm_compute/core/NEON/kernels/NESpaceToBatchLayerKernel.h b/arm_compute/core/NEON/kernels/NESpaceToBatchLayerKernel.h
index 79c321f919..ec88f03df4 100644
--- a/arm_compute/core/NEON/kernels/NESpaceToBatchLayerKernel.h
+++ b/arm_compute/core/NEON/kernels/NESpaceToBatchLayerKernel.h
@@ -29,6 +29,7 @@
 
 namespace arm_compute
 {
+// Forward declaration
 class ITensor;
 
 /** Interface for the space to batch kernel */
@@ -53,7 +54,7 @@ public:
     ~NESpaceToBatchLayerKernel() = default;
     /** Initialise the kernel's inputs and output.
      *
-     * @param[in]  input       Tensor input. Supported tensor rank: 4. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in]  input       Tensor input. Supported tensor rank: 4. Data types supported: All.
      * @param[in]  block_shape 1-D tensor with shape [M]. Data types supported: S32
      * @param[in]  paddings    2-D tensor with shape [2, M]. Data types supported: S32
      * @param[out] output      Tensor output. Data types supported: same as @p input
@@ -61,7 +62,7 @@ public:
     void configure(const ITensor *input, const ITensor *block_shape, const ITensor *paddings, ITensor *output);
     /** Initialise the kernel's input and output. (Static block shape and paddings)
      *
-     * @param[in]  input         Tensor input. Supported tensor rank: 4. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in]  input         Tensor input. Supported tensor rank: 4. Data types supported: All.
      * @param[in]  block_shape_x Block shape x value.
      * @param[in]  block_shape_y Block shape y value.
      * @param[in]  padding_left  The left padding of the output tensor.
@@ -71,7 +72,7 @@ public:
     void configure(const ITensor *input, const int block_shape_x, const int block_shape_y, const Size2D &padding_left, const Size2D &padding_right, ITensor *output);
     /** Static function to check if given info will lead to a valid configuration of @ref NESpaceToBatchLayerKernel
      *
-     * @param[in] input       Tensor input. Supported tensor rank: 4. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in] input       Tensor input. Supported tensor rank: 4. Data types supported: All.
      * @param[in] block_shape 1-D tensor with shape [M]. Data types supported: S32
      * @param[in] paddings    2-D tensor with shape [2, M]. Data types supported: S32
      * @param[in] output      Tensor output. Data types supported: same as @p input
@@ -81,7 +82,7 @@ public:
     static Status validate(const ITensorInfo *input, const ITensorInfo *block_shape, const ITensorInfo *paddings, const ITensorInfo *output);
     /** Static function to check if given info will lead to a valid configuration of @ref NESpaceToBatchLayerKernel (Static block shape and paddings)
      *
-     * @param[in] input         Tensor input. Supported tensor rank: 4. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in] input         Tensor input. Supported tensor rank: 4. Data types supported: All.
      * @param[in] block_shape_x Block shape x value.
      * @param[in] block_shape_y Block shape y value.
      * @param[in] padding_left  The left padding of the output tensor.
diff --git a/arm_compute/core/NEON/kernels/NESpaceToDepthLayerKernel.h b/arm_compute/core/NEON/kernels/NESpaceToDepthLayerKernel.h
index fe783a7159..87f8c67635 100644
--- a/arm_compute/core/NEON/kernels/NESpaceToDepthLayerKernel.h
+++ b/arm_compute/core/NEON/kernels/NESpaceToDepthLayerKernel.h
@@ -53,14 +53,14 @@ public:
     ~NESpaceToDepthLayerKernel() = default;
     /** Initialise the kernel's inputs and output.
      *
-     * @param[in]  input       Tensor input. Supported tensor rank: 4. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in]  input       Tensor input. Supported tensor rank: 4. Data types supported: All.
      * @param[out] output      Tensor output. Data types supported: same as @p input
      * @param[in]  block_shape Block shape value
      */
     void configure(const ITensor *input, ITensor *output, int32_t block_shape);
     /** Static function to check if given info will lead to a valid configuration of @ref NESpaceToDepthLayerKernel
      *
-     * @param[in] input       Tensor input info. Supported tensor rank: 4. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in] input       Tensor input info. Supported tensor rank: 4. Data types supported: All.
      * @param[in] output      Tensor output info. Data types supported: same as @p input
      * @param[in] block_shape Block shape value
      *
diff --git a/arm_compute/core/NEON/kernels/NEStackLayerKernel.h b/arm_compute/core/NEON/kernels/NEStackLayerKernel.h
index 04ad5d10dc..c4dc53eac6 100644
--- a/arm_compute/core/NEON/kernels/NEStackLayerKernel.h
+++ b/arm_compute/core/NEON/kernels/NEStackLayerKernel.h
@@ -56,7 +56,7 @@ public:
      *
      * @note Supported input tensor rank: up to 4
      *
-     * @param[in]  input       Input tensor. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     * @param[in]  input       Input tensor. Data types supported: All
      * @param[in]  axis        The dimension to stack the tensors along. It must be smaller than the number of input dimensions.
      * @param[in]  idx_input   Index of the input tensor in the list of tensors to stack.
      *                         All tensors in the list must have the same shape
@@ -69,7 +69,7 @@ public:
      *
      * @note Supported input tensor rank: up to 4
      *
-     * @param[in] input       Input tensor info. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     * @param[in] input       Input tensor info. Data types supported: All
      * @param[in] axis        The dimension to stack the tensors along. It must be smaller than the number of input dimensions.
      * @param[in] idx_input   Index of the input tensor in the list of tensors to stack
      *                        All tensors in the list must have the same shape
diff --git a/arm_compute/core/NEON/kernels/NEStridedSliceKernel.h b/arm_compute/core/NEON/kernels/NEStridedSliceKernel.h
index fc624f047b..6709619a62 100644
--- a/arm_compute/core/NEON/kernels/NEStridedSliceKernel.h
+++ b/arm_compute/core/NEON/kernels/NEStridedSliceKernel.h
@@ -58,7 +58,7 @@ public:
      *
      * @note Supported tensor rank: up to 4
      *
-     * @param[in]  input            Source tensor. Data type supported: U8/S8/QASYMM8/U16/S16/QASYMM16/QSYMM16/U32/S32/F16/F32
+     * @param[in]  input            Source tensor. Data type supported: All
      * @param[out] output           Destination tensor. Data type supported: Same as @p input
      * @param[in]  starts           The starts of the dimensions of the input tensor to be sliced. The length must be of rank(input).
      * @param[in]  ends             The ends of the dimensions of the input tensor to be sliced. The length must be of rank(input).
@@ -72,11 +72,11 @@ public:
                    const Coordinates &starts, const Coordinates &ends, const BiStrides &strides,
                    int32_t begin_mask, int32_t end_mask, int32_t shrink_axis_mask);
 
-    /** Static function to check if given info will lead to a valid configuration of @ref CLStridedSliceKernel
+    /** Static function to check if given info will lead to a valid configuration of @ref NEStridedSliceKernel
      *
      * @note Supported tensor rank: up to 4
      *
-     * @param[in] input            Source tensor info. Data type supported: U8/S8/QASYMM8/U16/S16/QASYMM16/QSYMM16/U32/S32/F16/F32
+     * @param[in] input            Source tensor info. Data type supported: All
      * @param[in] output           Destination tensor info. Data type supported: Same as @p input
      * @param[in] starts           The starts of the dimensions of the input tensor to be sliced. The length must be of rank(input).
      * @param[in] ends             The ends of the dimensions of the input tensor to be sliced. The length must be of rank(input).
diff --git a/arm_compute/core/NEON/kernels/NETileKernel.h b/arm_compute/core/NEON/kernels/NETileKernel.h
index 93afc80651..a64470ffd0 100644
--- a/arm_compute/core/NEON/kernels/NETileKernel.h
+++ b/arm_compute/core/NEON/kernels/NETileKernel.h
@@ -50,14 +50,14 @@ public:
     }
     /** Set the source, destination of the kernel
      *
-     * @param[in]  input     Source tensor. Data type supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in]  input     Source tensor. Data type supported: All.
      * @param[out] output    Destination tensor. Same as @p input
      * @param[in]  multiples Contains the number of times the input tensor should be replicated on the given dimension.
      */
     void configure(const ITensor *input, ITensor *output, const Multiples &multiples);
     /** Static function to check if given info will lead to a valid configuration of @ref NETileKernel
      *
-     * @param[in] input     Source tensor info. Data type supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in] input     Source tensor info. Data type supported: All.
      * @param[in] output    Destination tensor info. Same as @p input
      * @param[in] multiples Contains the number of times the input tensor should be replicated on the given dimension.
      *
diff --git a/arm_compute/core/NEON/kernels/NETransposeKernel.h b/arm_compute/core/NEON/kernels/NETransposeKernel.h
index d73813c846..a14dece0d6 100644
--- a/arm_compute/core/NEON/kernels/NETransposeKernel.h
+++ b/arm_compute/core/NEON/kernels/NETransposeKernel.h
@@ -57,13 +57,13 @@ public:
 
     /** Initialise the kernel's input and output.
      *
-     * @param[in]  input  Input tensor. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     * @param[in]  input  Input tensor. Data types supported: All
      * @param[out] output Output tensor. Data type supported: Same as @p input
      */
     void configure(const ITensor *input, ITensor *output);
     /** Static function to check if given info will lead to a valid configuration of @ref NETransposeKernel
      *
-     * @param[in] input  Input tensor. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     * @param[in] input  Input tensor. Data types supported: All
      * @param[in] output Output tensor. Data type supported: Same as @p input
      *
      * @return a status
@@ -76,7 +76,7 @@ public:
 private:
     /** Common signature for all the transpose functions
      *
-     * @param[in]  input  An input tensor. Data types supported: U8/S8/U16/S16/F16/U32/S32/F32
+     * @param[in]  input  An input tensor. Data types supported: All
      * @param[out] output The output tensor. Data type supported: same as @p input
      * @param[in]  window Region on which to execute the kernel.
      */
diff --git a/arm_compute/core/NEON/kernels/NEWidthConcatenateLayerKernel.h b/arm_compute/core/NEON/kernels/NEWidthConcatenateLayerKernel.h
index ccd70a4ebd..f22f18f09f 100644
--- a/arm_compute/core/NEON/kernels/NEWidthConcatenateLayerKernel.h
+++ b/arm_compute/core/NEON/kernels/NEWidthConcatenateLayerKernel.h
@@ -30,6 +30,7 @@
 
 namespace arm_compute
 {
+// Forward declarations
 class ITensor;
 
 /** Interface for the width concatenate kernel.
@@ -56,7 +57,7 @@ public:
     ~NEWidthConcatenateLayerKernel() = default;
     /** Initialise the kernel's inputs and output
      *
-     * @param[in]     input        Input tensor. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     * @param[in]     input        Input tensor. Data types supported: All
      * @param[in]     width_offset The offset on the X axis.
      * @param[in,out] output       Output tensor. Data types supported: Same as @p input.
      *
@@ -64,7 +65,7 @@ public:
     void configure(const ITensor *input, unsigned int width_offset, ITensor *output);
     /**  Static function to check if given info will lead to a valid configuration of @ref NEWidthConcatenateLayerKernel
      *
-     * @param[in] input        Input tensor info. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32
+     * @param[in] input        Input tensor info. Data types supported: All
      * @param[in] width_offset The offset on the X axis.
      * @param[in] output       Output tensor info. Data types supported: Same as @p input.
      *