COMPMID-2991: Add support for QASYMM8_SIGNED in CL kernels/functions - part 2

Adding support for QASYMM8_SIGNED to the following CL kernels/functions:

- CLActivationLayerKernel/CLActivationLayer
- CLComparisonKernel/CLComparison
- CLConvertFullyConnectedWeightsKernel/CLConvertFullyConnectedWeights
- CLDeconvolutionLayerUpsampleKernel/CLDeconvolutionLayerUpsample
- CLDepthToSpaceLayerKernel/CLDepthToSpaceLayer
- CLDequantizationLayerKernel/CLDequantizationLayer
- CLGEMMMatrixVectorMultiplyKernel
- CLNormalizePlanarYUVLayerKernel
- CLPReluLayer
- CLPixelWiseMultiplicationKernel/CLPixelWiseMultiplication
- CLPoolingLayerKernel/CLPoolingLayer

Change-Id: I874bbb7c2b08baa9c5ff4c9e6bc8778b42a6bec5
Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-on: https://review.mlplatform.org/c/2539
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>

20 files changed, 70 insertions, 88 deletions

diff --git a/arm_compute/core/CL/kernels/CLActivationLayerKernel.h b/arm_compute/core/CL/kernels/CLActivationLayerKernel.h
index cb2fda2be8..5b65a54824 100644
--- a/arm_compute/core/CL/kernels/CLActivationLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLActivationLayerKernel.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 ARM Limited.
+ * Copyright (c) 2016-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -51,7 +51,7 @@ public:
      * @note If the output tensor is a nullptr, the activation function will be performed in-place
      *
      * @param[in, out] input    Source tensor. In case of @p output tensor = nullptr, this tensor will store the result
-     *                          of the activation function. Data types supported: QASYMM8/QSYMM16/F16/F32.
+     *                          of the activation function. Data types supported: QASYMM8/QASYMM8_SIGNED/QSYMM16/F16/F32.
      * @param[out]     output   Destination tensor. Data type supported: same as @p input
      * @param[in]      act_info Activation layer information.
      */
@@ -59,7 +59,7 @@ public:
     /** Static function to check if given info will lead to a valid configuration of @ref CLActivationLayerKernel
      *
      * @param[in] input    Source tensor info. In case of @p output tensor info = nullptr, this tensor will store the result
-     *                     of the activation function. Data types supported: QASYMM8/QSYMM16/F16/F32.
+     *                     of the activation function. Data types supported: QASYMM8/QASYMM8_SIGNED/QSYMM16/F16/F32.
      * @param[in] output   Destination tensor info. Data type supported: same as @p input
      * @param[in] act_info Activation layer information.
      *
diff --git a/arm_compute/core/CL/kernels/CLComparisonKernel.h b/arm_compute/core/CL/kernels/CLComparisonKernel.h
index 21c6aeb064..a9c463901d 100644
--- a/arm_compute/core/CL/kernels/CLComparisonKernel.h
+++ b/arm_compute/core/CL/kernels/CLComparisonKernel.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 ARM Limited.
+ * Copyright (c) 2018-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -50,7 +50,7 @@ public:
     ~CLComparisonKernel() = default;
     /** Set the inputs and output tensors
      *
-     * @param[in]  input1    Source tensor. Data types supported: U8/S8/QASYMM8/U16/S16/U32/S32/F16/F32.
+     * @param[in]  input1    Source tensor. Data types supported: All.
      * @param[in]  input2    Source tensor. Data types supported: Same as @p input1.
      * @param[out] output    Destination tensor. Data types supported: U8.
      * @param[in]  operation Comparison operation to use.
@@ -58,7 +58,7 @@ public:
     void configure(const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output, ComparisonOperation operation);
     /** Static function to check if given info will lead to a valid configuration of @ref CLComparisonKernel
      *
-     * @param[in] input1    Source tensor. Data types supported: U8/S8/QASYMM8/U16/S16/U32/S32/F16/F32.
+     * @param[in] input1    Source tensor. Data types supported: All.
      * @param[in] input2    Source tensor. Data types supported: Same as @p input1.
      * @param[in] output    Destination tensor. Data types supported: U8.
      * @param[in] operation Comparison operation to use.
diff --git a/arm_compute/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.h b/arm_compute/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.h
index 6518dfc84c..b204eaa2ac 100644
--- a/arm_compute/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.h
+++ b/arm_compute/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 ARM Limited.
+ * Copyright (c) 2018-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -55,7 +55,7 @@ public:
     ~CLConvertFullyConnectedWeightsKernel() = 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.
@@ -63,7 +63,7 @@ public:
     void configure(const ICLTensor *input, ICLTensor *output, const TensorShape &original_input_shape, DataLayout data_layout);
     /** Static function to check if given info will lead to a valid configuration of @ref CLConvertFullyConnectedWeightsKernel
      *
-     * @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/CL/kernels/CLDeconvolutionLayerUpsampleKernel.h b/arm_compute/core/CL/kernels/CLDeconvolutionLayerUpsampleKernel.h
index 8382f3b4d7..a1c6bbdafe 100644
--- a/arm_compute/core/CL/kernels/CLDeconvolutionLayerUpsampleKernel.h
+++ b/arm_compute/core/CL/kernels/CLDeconvolutionLayerUpsampleKernel.h
@@ -50,14 +50,14 @@ public:
 
     /** Initialise the kernel's input and output.
      *
-     * @param[in]  input  Source tensor. Data types supported: QASYMM8/F16/F32.
+     * @param[in]  input  Source tensor. Data types supported: All.
      * @param[out] output Destination tensor. Data types supported: same as @p input. All but the lowest two dimensions must be the same size as in the input tensor, i.e. scaling is only performed within the XY-plane.
      * @param[in]  info   Contains padding and stride information described in @ref PadStrideInfo.
      */
     void configure(const ICLTensor *input, ICLTensor *output, const PadStrideInfo &info);
     /** Static function to check if given info will lead to a valid configuration of @ref CLDeconvolutionLayerUpsample
      *
-     * @param[in] input  Source tensor info. Data types supported: QASYMM8/F16/F32.
+     * @param[in] input  Source tensor info. Data types supported: All.
      * @param[in] output Destination tensor info. Data types supported: same as @p input. All but the lowest two dimensions must be the same size as in the input tensor, i.e. scaling is only performed within the XY-plane.
      * @param[in] info   Contains padding and stride information described in @ref PadStrideInfo.
      *
diff --git a/arm_compute/core/CL/kernels/CLDepthToSpaceLayerKernel.h b/arm_compute/core/CL/kernels/CLDepthToSpaceLayerKernel.h
index 6ae991c8e7..637e5fa960 100644
--- a/arm_compute/core/CL/kernels/CLDepthToSpaceLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLDepthToSpaceLayerKernel.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 ARM Limited.
+ * Copyright (c) 2019-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -49,14 +49,14 @@ public:
     ~CLDepthToSpaceLayerKernel() = 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 ICLTensor *input, ICLTensor *output, int32_t block_shape);
     /** Static function to check if given info will lead to a valid configuration of @ref CLDepthToSpaceLayerKernel.
      *
-     * @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/CL/kernels/CLDequantizationLayerKernel.h b/arm_compute/core/CL/kernels/CLDequantizationLayerKernel.h
index d79cd89883..78b5c14128 100644
--- a/arm_compute/core/CL/kernels/CLDequantizationLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLDequantizationLayerKernel.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 ARM Limited.
+ * Copyright (c) 2017-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -48,13 +48,13 @@ public:
     ~CLDequantizationLayerKernel() = default;
     /** Set the input, output, min and max.
      *
-     * @param[in]  input  Source tensor. Data types supported: QASYMM8/QSYMM8_PER_CHANNEL/QSYMM8/QSYMM16.
+     * @param[in]  input  Source tensor. Data types supported: QASYMM8/QASYMM8_SIGNED/QSYMM8_PER_CHANNEL/QSYMM8/QSYMM16.
      * @param[out] output Destination tensor. Data types supported: F16/F32.
      */
     void configure(const ICLTensor *input, ICLTensor *output);
     /** Static function to check if given info will lead to a valid configuration of @ref CLDequantizationLayerKernel
      *
-     * @param[in] input  Input tensor info. Data types supported: QASYMM8/QSYMM8_PER_CHANNEL/QSYMM8/QSYMM16.
+     * @param[in] input  Input tensor info. Data types supported: QASYMM8/QASYMM8_SIGNED/QSYMM8_PER_CHANNEL/QSYMM8/QSYMM16.
      * @param[in] output Output tensor info. Data types supported: F16/F32.
      *
      * @return a status
diff --git a/arm_compute/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.h b/arm_compute/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.h
index 821c6ae5bb..8ee911dc0e 100644
--- a/arm_compute/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.h
+++ b/arm_compute/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 ARM Limited.
+ * Copyright (c) 2017-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -46,16 +46,16 @@ public:
     CLGEMMMatrixVectorMultiplyKernel &operator=(CLGEMMMatrixVectorMultiplyKernel &&) = default;
     /** Set the input and output of the kernel.
      *
-     * @param[in]  input0 The reshaped input tensor. Data types supported: QASYMM8/F16/F32
-     * @param[in]  input1 The 2D reshaped weights tensor. Data type supported: Same as @p input, S32 for QASYMM8 input.
-     * @param[out] output The output 2D tensor. Data types supported: Same as @p input
+     * @param[in]  input0 The reshaped input tensor. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32
+     * @param[in]  input1 The 2D reshaped weights tensor. Data type supported: Same as @p input.
+     * @param[out] output The output 2D tensor. Data types supported: Same as @p input, S32 for QASYMM8/QASYMM8_SIGNED.
      */
     void configure(const ICLTensor *input0, const ICLTensor *input1, ICLTensor *output);
     /** Static function to check if given info will lead to a valid configuration of @ref CLGEMMMatrixVectorMultiplyKernel
      *
-     * @param[in] input0 The reshaped input tensor. Data types supported: QASYMM8/F16/F32
-     * @param[in] input1 The 2D reshaped weights tensor. Data type supported: Same as @p input, S32 for QASYMM8 input.
-     * @param[in] output The output 2D tensor. Data types supported: Same as @p input
+     * @param[in] input0 The reshaped input tensor info. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32
+     * @param[in] input1 The 2D reshaped weights tensor info. Data type supported: Same as @p input.
+     * @param[in] output The output 2D tensor info. Data types supported: Same as @p input, S32 for QASYMM8/QASYMM8_SIGNED.
      *
      * @return a status
      */
diff --git a/arm_compute/core/CL/kernels/CLNormalizePlanarYUVLayerKernel.h b/arm_compute/core/CL/kernels/CLNormalizePlanarYUVLayerKernel.h
index d54aae31c5..4334882fd8 100644
--- a/arm_compute/core/CL/kernels/CLNormalizePlanarYUVLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLNormalizePlanarYUVLayerKernel.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 ARM Limited.
+ * Copyright (c) 2018-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -50,7 +50,7 @@ public:
     /** Set the input and output tensors.
      *
      * @param[in]  input  Source tensor. 3 lower dimensions represent a single input with dimensions [width, height, channels].
-     *                    Data types supported: QASYMM8/F16/F32.
+     *                    Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32.
      * @param[out] output Destination tensor. Data type supported: same as @p input
      * @param[in]  mean   Mean values tensor. 1 dimension with size equal to the number of input channels. Data types supported: same as @p input
      * @param[in]  std    Standard deviation values tensor. 1 dimension with size equal to the number of input channels.
@@ -60,7 +60,7 @@ public:
     /** Static function to check if given info will lead to a valid configuration of @ref CLNormalizePlanarYUVLayerKernel
      *
      * @param[in]  input  Source tensor info. 3 lower dimensions represent a single input with dimensions [width, height, channels].
-     *                    Data types supported: QASYMM8/F16/F32.
+     *                    Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32.
      * @param[out] output Destination tensor info. Data type supported: same as @p input
      * @param[in]  mean   Mean values tensor info. 1 dimension with size equal to the number of input channels. Data types supported: same as @p input
      * @param[in]  std    Standard deviation values tensor info. 1 dimension with size equal to the number of input channels.
diff --git a/arm_compute/core/CL/kernels/CLPixelWiseMultiplicationKernel.h b/arm_compute/core/CL/kernels/CLPixelWiseMultiplicationKernel.h
index 0e5027b29a..58471ab299 100644
--- a/arm_compute/core/CL/kernels/CLPixelWiseMultiplicationKernel.h
+++ b/arm_compute/core/CL/kernels/CLPixelWiseMultiplicationKernel.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 ARM Limited.
+ * Copyright (c) 2016-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -48,7 +48,7 @@ public:
     CLPixelWiseMultiplicationKernel &operator=(CLPixelWiseMultiplicationKernel &&) = default;
     /** Initialise the kernel's input, output and border mode.
      *
-     * @param[in]  input1          An input tensor. Data types supported: U8/QASYMM8/S16/QSYMM16/F16/F32.
+     * @param[in]  input1          An input tensor. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/F16/F32.
      * @param[in]  input2          An input tensor. Data types supported: same as @p input1.
      * @param[out] output          The output tensor, Data types supported: same as @p input1. Note: U8 requires both inputs to be U8.
      * @param[in]  scale           Scale to apply after multiplication.
@@ -60,7 +60,7 @@ public:
                    ConvertPolicy overflow_policy, RoundingPolicy rounding_policy);
     /** Static function to check if given info will lead to a valid configuration of @ref CLPixelWiseMultiplicationKernel
      *
-     * @param[in] input1          An input tensor info. Data types supported: U8/QASYMM8/S16/QSYMM16/F16/F32.
+     * @param[in] input1          An input tensor info. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/F16/F32.
      * @param[in] input2          An input tensor info. Data types supported: same as @p input1.
      * @param[in] output          The output tensor info, Data types supported: same as @p input1. Note: U8 requires both inputs to be U8.
      * @param[in] scale           Scale to apply after multiplication.
diff --git a/arm_compute/core/CL/kernels/CLPoolingLayerKernel.h b/arm_compute/core/CL/kernels/CLPoolingLayerKernel.h
index 95acdf4b6c..4b3ee24333 100644
--- a/arm_compute/core/CL/kernels/CLPoolingLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLPoolingLayerKernel.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 ARM Limited.
+ * Copyright (c) 2017-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -52,14 +52,14 @@ public:
     /** Set the input and output tensors.
      *
      *
-     * @param[in]  input     Source tensor. Data types supported: QASYMM8/F16/F32.
+     * @param[in]  input     Source tensor. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32.
      * @param[out] output    Destination tensor. Data types supported: Same as @p input.
      * @param[in]  pool_info Contains pooling operation information described in @ref PoolingLayerInfo.
      */
     void configure(const ICLTensor *input, ICLTensor *output, const PoolingLayerInfo &pool_info);
     /** Static function to check if given info will lead to a valid configuration of @ref CLPoolingLayerKernel
      *
-     * @param[in] input     Source tensor info. Data types supported: QASYMM8/F16/F32.
+     * @param[in] input     Source tensor info. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32.
      * @param[in] output    Destination tensor info. Data types supported: Same as @p input.
      * @param[in] pool_info Contains pooling operation information described in @ref PoolingLayerInfo.
      *
diff --git a/arm_compute/runtime/CL/functions/CLActivationLayer.h b/arm_compute/runtime/CL/functions/CLActivationLayer.h
index f7781480bf..09f5d2bf58 100644
--- a/arm_compute/runtime/CL/functions/CLActivationLayer.h
+++ b/arm_compute/runtime/CL/functions/CLActivationLayer.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 ARM Limited.
+ * Copyright (c) 2016-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -57,7 +57,7 @@ public:
      * @note If the output tensor is a nullptr or is equal to the input, the activation function will be performed in-place
      *
      * @param[in, out] input    Source tensor. In case of @p output tensor = nullptr, this tensor will store the result
-     *                          of the activation function. Data types supported: QASYMM8/QSYMM16/F16/F32.
+     *                          of the activation function. Data types supported: QASYMM8/QASYMM8_SIGNED/QSYMM16/F16/F32.
      * @param[out]     output   Destination tensor. Data type supported: same as @p input
      * @param[in]      act_info Activation layer parameters.
      */
@@ -65,7 +65,7 @@ public:
     /** Static function to check if given info will lead to a valid configuration of @ref CLActivationLayer
      *
      * @param[in] input    Source tensor info. In case of @p output tensor info = nullptr, this tensor will store the result
-     *                     of the activation function. Data types supported: QASYMM8/QSYMM16/F16/F32.
+     *                     of the activation function. Data types supported: QASYMM8/QASYMM8_SIGNED/QSYMM16/F16/F32.
      * @param[in] output   Destination tensor info. Data type supported: same as @p input
      * @param[in] act_info Activation layer information.
      *
diff --git a/arm_compute/runtime/CL/functions/CLComparison.h b/arm_compute/runtime/CL/functions/CLComparison.h
index 7f0b22341f..85dbe7129d 100644
--- a/arm_compute/runtime/CL/functions/CLComparison.h
+++ b/arm_compute/runtime/CL/functions/CLComparison.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 ARM Limited.
+ * Copyright (c) 2018-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -38,7 +38,7 @@ class CLComparison : public ICLSimpleFunction
 public:
     /** Initialise the kernel's inputs and outputs.
      *
-     * @param[in]  input1    Source tensor. Data types supported: U8/S8/QASYMM8/U16/S16/U32/S32/F16/F32.
+     * @param[in]  input1    Source tensor. Data types supported: All.
      *                       The input1 tensor is [in, out] because its TensorInfo might be modified inside the kernel in case of broadcasting of dimension 0.
      * @param[in]  input2    Source tensor. Data types supported: Same as @p input1.
      *                       The input2 tensor is [in, out] because its TensorInfo might be modified inside the kernel in case of broadcasting of dimension 0.
@@ -48,7 +48,7 @@ public:
     void configure(ICLTensor *input1, ICLTensor *input2, ICLTensor *output, ComparisonOperation operation);
     /** Static function to check if given info will lead to a valid configuration of @ref CLComparison
      *
-     * @param[in]  input1    Source tensor. Data types supported: U8/S8/QASYMM8/U16/S16/U32/S32/F16/F32.
+     * @param[in]  input1    Source tensor. Data types supported: All.
      * @param[in]  input2    Source tensor. Data types supported: Same as @p input1.
      * @param[in]  output    Destination tensor. Data types supported: U8.
      * @param[out] operation Comparison operation to be used.
diff --git a/arm_compute/runtime/CL/functions/CLConvertFullyConnectedWeights.h b/arm_compute/runtime/CL/functions/CLConvertFullyConnectedWeights.h
index f0359ecc75..76a28ed6fe 100644
--- a/arm_compute/runtime/CL/functions/CLConvertFullyConnectedWeights.h
+++ b/arm_compute/runtime/CL/functions/CLConvertFullyConnectedWeights.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 ARM Limited.
+ * Copyright (c) 2018-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -39,7 +39,7 @@ class CLConvertFullyConnectedWeights : public ICLSimpleFunction
 public:
     /** Initialize the function.
      *
-     * @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.
@@ -49,7 +49,7 @@ public:
     void configure(const ICLTensor *input, ICLTensor *output, const TensorShape &original_input_shape, DataLayout data_layout);
     /** Static function to check if given info will lead to a valid configuration of @ref CLConvertFullyConnectedWeights
      *
-     * @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.
@@ -90,7 +90,7 @@ public:
     }
     /** Configures the @ref CLConvertFullyConnectedWeights function
      *
-     * @param[in] input                Source weights tensor info to convert.  Data type supported: U8/S8/QASYMM8/U16/S16/U32/S32/F16/F32.
+     * @param[in] input                Source weights tensor info to convert.  Data type supported: All.
      * @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/runtime/CL/functions/CLDeconvolutionLayerUpsample.h b/arm_compute/runtime/CL/functions/CLDeconvolutionLayerUpsample.h
index 6f015744bf..5a1009c79f 100644
--- a/arm_compute/runtime/CL/functions/CLDeconvolutionLayerUpsample.h
+++ b/arm_compute/runtime/CL/functions/CLDeconvolutionLayerUpsample.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 ARM Limited.
+ * Copyright (c) 2017-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -59,14 +59,14 @@ public:
 
     /** Initialize the function's source, destination, interpolation type and border_mode.
      *
-     * @param[in, out] input  Source tensor. Data type supported: QASYMM8/F16/F32.
+     * @param[in, out] input  Source tensor. Data type supported: All.
      * @param[out]     output Destination tensor. Data type supported: same as @p input.
      * @param[in]      info   Contains padding and policies to be used in the deconvolution.
      */
     void configure(ICLTensor *input, ICLTensor *output, const PadStrideInfo &info);
     /** Static function to check if given info will lead to a valid configuration of @ref CLDeconvolutionLayerUpsample
      *
-     * @param[in] input  Source tensor info. Data type supported: QASYMM8/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] info   Contains padding and policies to be used in the deconvolution.
      *
diff --git a/arm_compute/runtime/CL/functions/CLDepthToSpaceLayer.h b/arm_compute/runtime/CL/functions/CLDepthToSpaceLayer.h
index ddee04a3dd..0c33ed34be 100644
--- a/arm_compute/runtime/CL/functions/CLDepthToSpaceLayer.h
+++ b/arm_compute/runtime/CL/functions/CLDepthToSpaceLayer.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 ARM Limited.
+ * Copyright (c) 2019-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -24,43 +24,33 @@
 #ifndef ARM_COMPUTE_CLDEPTHTOSPACELAYER_H
 #define ARM_COMPUTE_CLDEPTHTOSPACELAYER_H
 
-#include "arm_compute/runtime/IFunction.h"
-
-#include "arm_compute/core/CL/kernels/CLDepthToSpaceLayerKernel.h"
 #include "arm_compute/core/Types.h"
+#include "arm_compute/runtime/CL/ICLSimpleFunction.h"
 
 namespace arm_compute
 {
 class ICLTensor;
 
 /** Basic function to run @ref CLDepthToSpaceLayerKernel. */
-class CLDepthToSpaceLayer : public IFunction
+class CLDepthToSpaceLayer : public ICLSimpleFunction
 {
 public:
-    /** Default constructor */
-    CLDepthToSpaceLayer();
     /** Set the input and output tensors.
      *
-     * @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 ICLTensor *input, ICLTensor *output, int32_t block_shape);
     /** Static function to check if given info will lead to a valid configuration of @ref CLDepthToSpaceLayer.
      *
-     * @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.
      *
      * @return a status
      */
     static Status validate(const ITensorInfo *input, const ITensorInfo *output, int32_t block_shape);
-
-    // Inherited methods overridden:
-    void run() override;
-
-private:
-    CLDepthToSpaceLayerKernel _depth_to_space_kernel; /**< CLDepthToSpaceLayerKernel to run */
 };
 }
 #endif /* ARM_COMPUTE_CLDEPTHTOSPACELAYER_H */
diff --git a/arm_compute/runtime/CL/functions/CLDequantizationLayer.h b/arm_compute/runtime/CL/functions/CLDequantizationLayer.h
index 308349af02..48d6ba8435 100644
--- a/arm_compute/runtime/CL/functions/CLDequantizationLayer.h
+++ b/arm_compute/runtime/CL/functions/CLDequantizationLayer.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 ARM Limited.
+ * Copyright (c) 2017-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -40,13 +40,13 @@ public:
     /** Set the input and output tensors.
      *
      * @param[in]  input  Source tensor with at least 3 dimensions. The dimensions over the third will be interpreted as batches.
-     *                    Data types supported: QASYMM8/QSYMM8_PER_CHANNEL/QSYMM8/QSYMM16.
+     *                    Data types supported: QASYMM8/QASYMM8_SIGNED/QSYMM8_PER_CHANNEL/QSYMM8/QSYMM16.
      * @param[out] output Destination tensor with the same dimensions of input. Data type supported: F16/F32.
      */
     void configure(const ICLTensor *input, ICLTensor *output);
     /** Static function to check if given info will lead to a valid configuration of @ref CLDequantizationLayer
      *
-     * @param[in] input  Input tensor info. Data types supported: QASYMM8/QSYMM8_PER_CHANNEL/QSYMM8/QSYMM16.
+     * @param[in] input  Input tensor info. Data types supported: QASYMM8/QASYMM8_SIGNED/QSYMM8_PER_CHANNEL/QSYMM8/QSYMM16.
      * @param[in] output Output tensor info. Data type supported: F16/F32.
      *
      * @return a status
diff --git a/arm_compute/runtime/CL/functions/CLNormalizePlanarYUVLayer.h b/arm_compute/runtime/CL/functions/CLNormalizePlanarYUVLayer.h
index 4fe5a111b6..5fbfdd18b7 100644
--- a/arm_compute/runtime/CL/functions/CLNormalizePlanarYUVLayer.h
+++ b/arm_compute/runtime/CL/functions/CLNormalizePlanarYUVLayer.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 ARM Limited.
+ * Copyright (c) 2018-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -24,10 +24,10 @@
 #ifndef ARM_COMPUTE_CLNORMALIZEPLANARYUVLAYER_H
 #define ARM_COMPUTE_CLNORMALIZEPLANARYUVLAYER_H
 
-#include "arm_compute/runtime/IFunction.h"
-
-#include "arm_compute/core/CL/kernels/CLNormalizePlanarYUVLayerKernel.h"
 #include "arm_compute/core/Types.h"
+#include "arm_compute/runtime/CL/ICLSimpleFunction.h"
+
+#include <cstdint>
 
 namespace arm_compute
 {
@@ -37,15 +37,13 @@ class ICLTensor;
  *
  *  @note The function simulates a NormalizePlanarYUV layer.
  */
-class CLNormalizePlanarYUVLayer : public IFunction
+class CLNormalizePlanarYUVLayer : public ICLSimpleFunction
 {
 public:
-    /** Default constructor */
-    CLNormalizePlanarYUVLayer();
     /** Set the input and output tensors.
      *
      * @param[in]  input  Source tensor. 3 lower dimensions represent a single input with dimensions [width, height, channels].
-     *                    Data types supported: QASYMM8/F16/F32.
+     *                    Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32.
      * @param[out] output Destinationfeature tensor. Data type supported: same as @p input
      * @param[in]  mean   Mean values tensor. 1 dimension with size equal to the number of input channels. Data types supported: Same as @p input
      * @param[in]  std    Standard deviation values tensor. 1 dimension with size equal to the number of input channels.
@@ -55,7 +53,7 @@ public:
     /** Static function to check if given info will lead to a valid configuration of @ref CLNormalizePlanarYUVLayer
      *
      * @param[in]  input  Source tensor info. 3 lower dimensions represent a single input with dimensions [width, height, channels].
-     *                    Data types supported: QASYMM8/F16/F32.
+     *                    Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32.
      * @param[out] output Destination tensor info. Data type supported: same as @p input
      * @param[in]  mean   Mean values tensor info. 1 dimension with size equal to the number of input channels. Data types supported: Same as @p input
      * @param[in]  std    Standard deviation values tensor info. 1 dimension with size equal to the number of input channels.
@@ -64,12 +62,6 @@ public:
      * @return a status
      */
     static Status validate(const ITensorInfo *input, const ITensorInfo *output, const ITensorInfo *mean, const ITensorInfo *std);
-
-    // Inherited methods overridden:
-    void run() override;
-
-private:
-    CLNormalizePlanarYUVLayerKernel _norm_kernel; /**< NormalizePlanarYUV layer kernel to run */
 };
 } // namespace arm_compute
 #endif /* ARM_COMPUTE_CLNORMALIZEPLANARYUVLAYER_H */
diff --git a/arm_compute/runtime/CL/functions/CLPReluLayer.h b/arm_compute/runtime/CL/functions/CLPReluLayer.h
index 42876cd714..7f8a41238c 100644
--- a/arm_compute/runtime/CL/functions/CLPReluLayer.h
+++ b/arm_compute/runtime/CL/functions/CLPReluLayer.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 ARM Limited.
+ * Copyright (c) 2019-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -42,14 +42,14 @@ public:
      *
      * @note If the output tensor is a nullptr or is equal to the input, the activation function will be performed in-place
      *
-     * @param[in]  input  Source tensor. Data types supported: QASYMM8/F16/F32.
+     * @param[in]  input  Source tensor. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32.
      * @param[in]  alpha  PRelu layer parameters. Data types supported: same of @p input.
      * @param[out] output Destination tensor. Data type supported: same as @p input
      */
     void configure(ICLTensor *input, ICLTensor *alpha, ICLTensor *output);
     /** Static function to check if given info will lead to a valid configuration of @ref CLPReluLayer
      *
-     * @param[in] input  Source tensor info. Data types supported: QASYMM8/F16/F32.
+     * @param[in] input  Source tensor info. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32.
      * @param[in] alpha  PRelu layer parameters. Data types supported: same of @p input.
      * @param[in] output Destination tensor info. Data type supported: same as @p input
      *
diff --git a/arm_compute/runtime/CL/functions/CLPixelWiseMultiplication.h b/arm_compute/runtime/CL/functions/CLPixelWiseMultiplication.h
index fd64d7b939..72b1587b02 100644
--- a/arm_compute/runtime/CL/functions/CLPixelWiseMultiplication.h
+++ b/arm_compute/runtime/CL/functions/CLPixelWiseMultiplication.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 ARM Limited.
+ * Copyright (c) 2016-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -38,7 +38,7 @@ class CLPixelWiseMultiplication : public ICLSimpleFunction
 public:
     /** Initialise the kernel's inputs, output and convertion policy.
      *
-     * @param[in, out] input1          An input tensor. Data types supported: U8/QASYMM8/S16/QSYMM16/F16/F32.
+     * @param[in, out] input1          An input tensor. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/F16/F32.
      *                                 The input tensor is [in, out] because its TensorInfo might be modified inside the kernel in case of broadcasting of dimension 0.
      * @param[in, out] input2          An input tensor. Data types supported: same as @p input1.
      *                                 The input tensor is [in, out] because its TensorInfo might be modified inside the kernel in case of broadcasting of dimension 0.
@@ -52,7 +52,7 @@ public:
                    ConvertPolicy overflow_policy, RoundingPolicy rounding_policy);
     /** Static function to check if given info will lead to a valid configuration of @ref CLPixelWiseMultiplication
      *
-     * @param[in] input1          An input tensor info. Data types supported: U8/QASYMM8/S16/QSYMM16/F16/F32.
+     * @param[in] input1          An input tensor info. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/F16/F32.
      * @param[in] input2          An input tensor info. Data types supported: same as @p input1.
      * @param[in] output          The output tensor info, Data types supported: same as @p input1. Note: U8 requires both inputs to be U8.
      * @param[in] scale           Scale to apply after multiplication.
diff --git a/arm_compute/runtime/CL/functions/CLPoolingLayer.h b/arm_compute/runtime/CL/functions/CLPoolingLayer.h
index 19acb7fb40..c78b558ac8 100644
--- a/arm_compute/runtime/CL/functions/CLPoolingLayer.h
+++ b/arm_compute/runtime/CL/functions/CLPoolingLayer.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 ARM Limited.
+ * Copyright (c) 2017-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -43,14 +43,14 @@ class CLPoolingLayer : public ICLSimpleFunction
 public:
     /** Set the input and output tensors.
      *
-     * @param[in,out] input     Source tensor. (Written to only when padding != 0) Data types supported: QASYMM8/F16/F32.
+     * @param[in,out] input     Source tensor. (Written to only when padding != 0) Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32.
      * @param[out]    output    Destination tensor. Data types supported: Same as @p input.
      * @param[in]     pool_info Contains pooling operation information described in @ref PoolingLayerInfo.
      */
     void configure(ICLTensor *input, ICLTensor *output, const PoolingLayerInfo &pool_info);
     /** Static function to check if given info will lead to a valid configuration of @ref CLPoolingLayer
      *
-     * @param[in] input     Source tensor info. Data types supported: QASYMM8/F16/F32.
+     * @param[in] input     Source tensor info. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32.
      * @param[in] output    Destination tensor info. Data types supported: Same as @p input.
      * @param[in] pool_info Contains pooling operation information described in @ref PoolingLayerInfo.
      *