COMPMID-3280: Make all ML primitives for CL use the new interface - Part1 - Fix1

- const fix in the CLKernels part 1

Change-Id: I17340cb6ff26afd52b14b46645efedbe07aef1b6
Signed-off-by: Manuel Bottini <manuel.bottini@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3067
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>

50 files changed, 70 insertions, 71 deletions

diff --git a/arm_compute/core/CL/CLHelpers.h b/arm_compute/core/CL/CLHelpers.h
index 77c17c7d9c..c5db66c664 100644
--- a/arm_compute/core/CL/CLHelpers.h
+++ b/arm_compute/core/CL/CLHelpers.h
@@ -206,7 +206,7 @@ cl::Kernel create_opencl_kernel(CLCoreRuntimeContext *ctx, const std::string &ke
  *
  * @return An opencl kernel
  */
-cl::Kernel create_kernel(CLCompileContext &ctx, const std::string &kernel_name, const std::set<std::string> &build_opts = std::set<std::string>());
+cl::Kernel create_kernel(const CLCompileContext &ctx, const std::string &kernel_name, const std::set<std::string> &build_opts = std::set<std::string>());
 
 /** Creates a suitable LWS hint object for parallel implementations. Sets the number of WG based on the input size.
  *  If input width is smaller than 128 we can use fewer threads than 8.
diff --git a/arm_compute/core/CL/kernels/CLAbsoluteDifferenceKernel.h b/arm_compute/core/CL/kernels/CLAbsoluteDifferenceKernel.h
index 18896725e2..58dea3bdae 100644
--- a/arm_compute/core/CL/kernels/CLAbsoluteDifferenceKernel.h
+++ b/arm_compute/core/CL/kernels/CLAbsoluteDifferenceKernel.h
@@ -65,7 +65,7 @@ public:
      * @param[in]  input2          Source tensor. Data types supported: U8/S16.
      * @param[out] output          Destination tensor. Data types supported: U8/S16.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output);
 
     // Inherited methods overridden:
     void run(const Window &window, cl::CommandQueue &queue) override;
diff --git a/arm_compute/core/CL/kernels/CLAccumulateKernel.h b/arm_compute/core/CL/kernels/CLAccumulateKernel.h
index d7cb09fdd3..f639148e25 100644
--- a/arm_compute/core/CL/kernels/CLAccumulateKernel.h
+++ b/arm_compute/core/CL/kernels/CLAccumulateKernel.h
@@ -52,7 +52,7 @@ public:
      * @param[in]  input           Source tensor. Data types supported: U8.
      * @param[out] accum           Destination tensor. Data types supported: S16.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *accum);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *accum);
 };
 
 /** Interface for the accumulate weighted kernel.
@@ -81,7 +81,7 @@ public:
      * @param[in]     alpha           Scalar value in the range [0, 1.0]. Data types supported: F32.
      * @param[in,out] accum           Accumulated tensor. Data types supported: U8.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, float alpha, ICLTensor *accum);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, float alpha, ICLTensor *accum);
 };
 
 /** Interface for the accumulate squared kernel.
@@ -108,7 +108,7 @@ public:
      * @param[in]     shift           Shift value in the range of [0, 15]. Data types supported: U32.
      * @param[in,out] accum           Accumulated tensor. Data types supported: S16.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, uint32_t shift, ICLTensor *accum);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, uint32_t shift, ICLTensor *accum);
 };
 } // namespace arm_compute
 #endif /*ARM_COMPUTE_CLACCUMULATEKERNEL_H */
diff --git a/arm_compute/core/CL/kernels/CLActivationLayerKernel.h b/arm_compute/core/CL/kernels/CLActivationLayerKernel.h
index d25480cd60..1e83a689cd 100644
--- a/arm_compute/core/CL/kernels/CLActivationLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLActivationLayerKernel.h
@@ -65,7 +65,7 @@ public:
      * @param[out]     output          Destination tensor. Data type supported: same as @p input
      * @param[in]      act_info        Activation layer information.
      */
-    void configure(CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, ActivationLayerInfo act_info);
+    void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, ActivationLayerInfo act_info);
     /** 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
@@ -81,9 +81,9 @@ public:
     void run(const Window &window, cl::CommandQueue &queue) override;
 
 private:
-    ICLTensor            *_input;
-    ICLTensor            *_output;
-    bool                  _run_in_place;
+    ICLTensor *_input;
+    ICLTensor *_output;
+    bool       _run_in_place;
 };
 } // namespace arm_compute
 #endif /*ARM_COMPUTE_CLACTIVATIONLAYERKERNEL_H */
diff --git a/arm_compute/core/CL/kernels/CLArgMinMaxLayerKernel.h b/arm_compute/core/CL/kernels/CLArgMinMaxLayerKernel.h
index 831cee5e58..94e8baed13 100644
--- a/arm_compute/core/CL/kernels/CLArgMinMaxLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLArgMinMaxLayerKernel.h
@@ -76,7 +76,7 @@ public:
      * @param[in]  axis            Axis along which to reduce. Supported reduction axis : 0,1,2,3
      * @param[in]  op              Reduction operation to perform. Only ArgMin and ArgMax are supported.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *prev_output, ICLTensor *output, unsigned int axis, ReductionOperation op);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *prev_output, ICLTensor *output, unsigned int axis, ReductionOperation op);
 
     /** Static function to check if given info will lead to a valid configuration of @ref CLArgMinMaxLayerKernel.
      *
diff --git a/arm_compute/core/CL/kernels/CLBatchConcatenateLayerKernel.h b/arm_compute/core/CL/kernels/CLBatchConcatenateLayerKernel.h
index 06764302f4..163666853c 100644
--- a/arm_compute/core/CL/kernels/CLBatchConcatenateLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLBatchConcatenateLayerKernel.h
@@ -72,7 +72,7 @@ public:
      * @note: The gaps between the two lowest dimensions of input and output need to be divisible by 2.
      *
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, unsigned int batch_offset, ICLTensor *output);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, unsigned int batch_offset, ICLTensor *output);
     /**  Static function to check if given info will lead to a valid configuration of @ref CLBatchConcatenateLayerKernel
      *
      * @param[in] input        Input tensor info. Data types supported: All.
diff --git a/arm_compute/core/CL/kernels/CLBatchNormalizationLayerKernel.h b/arm_compute/core/CL/kernels/CLBatchNormalizationLayerKernel.h
index 564b21680b..8eaaca845a 100644
--- a/arm_compute/core/CL/kernels/CLBatchNormalizationLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLBatchNormalizationLayerKernel.h
@@ -81,9 +81,8 @@ public:
      * @param[in]      epsilon         (Optional) Small value to avoid division with zero. Default value is 0.001f.
      * @param[in]      act_info        (Optional) Activation layer information in case of a fused activation. Only RELU, BOUNDED_RELU and LU_BOUNDED_RELU supported.
      */
-    void configure(CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, const ICLTensor *mean, const ICLTensor *var, const ICLTensor *beta = nullptr, const ICLTensor *gamma = nullptr,
-                   float               epsilon  = 0.001f,
-                   ActivationLayerInfo act_info = ActivationLayerInfo());
+    void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, const ICLTensor *mean, const ICLTensor *var, const ICLTensor *beta = nullptr,
+                   const ICLTensor *gamma = nullptr, float epsilon = 0.001f, ActivationLayerInfo act_info = ActivationLayerInfo());
     /** Static function to check if given info will lead to a valid configuration of @ref CLBatchNormalizationLayerKernel
      *
      * @param[in] input    Source tensor info. In case of @p output tensor info = nullptr, this tensor will store the result.
diff --git a/arm_compute/core/CL/kernels/CLBatchToSpaceLayerKernel.h b/arm_compute/core/CL/kernels/CLBatchToSpaceLayerKernel.h
index f9289eab73..2b12ad094a 100644
--- a/arm_compute/core/CL/kernels/CLBatchToSpaceLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLBatchToSpaceLayerKernel.h
@@ -61,7 +61,7 @@ public:
      * @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(CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *block_shape, ICLTensor *output);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *block_shape, ICLTensor *output);
     /** Initialise the kernel's inputs and output (Static block shape).
      *
      * @param[in]  input         Tensor input. Supported tensor rank: 4. Data types supported: All.
@@ -78,7 +78,7 @@ public:
      * @param[in]  block_shape_y   Block shape y value.
      * @param[out] output          Tensor output. Data types supported: same as @p input
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, const int32_t block_shape_x, const int32_t block_shape_y, ICLTensor *output);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, const int32_t block_shape_x, const int32_t block_shape_y, ICLTensor *output);
     /** Static function to check if given info will lead to a valid configuration of @ref CLBatchToSpaceLayerKernel
      *
      * @param[in] input       Tensor input. Supported tensor rank: 4. Data types supported: All.
diff --git a/arm_compute/core/CL/kernels/CLBitwiseAndKernel.h b/arm_compute/core/CL/kernels/CLBitwiseAndKernel.h
index 6c60bc0f33..8defe32862 100644
--- a/arm_compute/core/CL/kernels/CLBitwiseAndKernel.h
+++ b/arm_compute/core/CL/kernels/CLBitwiseAndKernel.h
@@ -62,7 +62,7 @@ public:
      * @param[in]  input2          Source tensor. Data types supported: U8.
      * @param[out] output          Destination tensor. Data types supported: U8.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output);
 
     // Inherited methods overridden:
     void run(const Window &window, cl::CommandQueue &queue) override;
diff --git a/arm_compute/core/CL/kernels/CLBitwiseNotKernel.h b/arm_compute/core/CL/kernels/CLBitwiseNotKernel.h
index 0522841e73..b86ce7f173 100644
--- a/arm_compute/core/CL/kernels/CLBitwiseNotKernel.h
+++ b/arm_compute/core/CL/kernels/CLBitwiseNotKernel.h
@@ -50,7 +50,7 @@ public:
      * @param[in]  input           Source tensor. Data types supported: U8.
      * @param[out] output          Destination tensor. Data types supported: U8.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output);
 };
 } // namespace arm_compute
 #endif /* ARM_COMPUTE_CLBITWISENOTKERNEL_H */
diff --git a/arm_compute/core/CL/kernels/CLBitwiseOrKernel.h b/arm_compute/core/CL/kernels/CLBitwiseOrKernel.h
index 151f19d374..65eb50f0fd 100644
--- a/arm_compute/core/CL/kernels/CLBitwiseOrKernel.h
+++ b/arm_compute/core/CL/kernels/CLBitwiseOrKernel.h
@@ -62,7 +62,7 @@ public:
      * @param[in]  input2          Source tensor. Data types supported: U8.
      * @param[out] output          Destination tensor. Data types supported: U8.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output);
 
     // Inherited methods overridden:
     void run(const Window &window, cl::CommandQueue &queue) override;
diff --git a/arm_compute/core/CL/kernels/CLBitwiseXorKernel.h b/arm_compute/core/CL/kernels/CLBitwiseXorKernel.h
index 03c1e05da4..5c63a7f22c 100644
--- a/arm_compute/core/CL/kernels/CLBitwiseXorKernel.h
+++ b/arm_compute/core/CL/kernels/CLBitwiseXorKernel.h
@@ -62,7 +62,7 @@ public:
      * @param[in]  input2          Source tensor. Data types supported: U8.
      * @param[out] output          Destination tensor. Data types supported: U8.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output);
 
     // Inherited methods overridden:
     void run(const Window &window, cl::CommandQueue &queue) override;
diff --git a/arm_compute/core/CL/kernels/CLBoundingBoxTransformKernel.h b/arm_compute/core/CL/kernels/CLBoundingBoxTransformKernel.h
index ffa63bd5a4..bbe11562ed 100644
--- a/arm_compute/core/CL/kernels/CLBoundingBoxTransformKernel.h
+++ b/arm_compute/core/CL/kernels/CLBoundingBoxTransformKernel.h
@@ -71,7 +71,7 @@ public:
      * @note Only single image prediction is supported. Height and Width (and scale) of the image will be contained in the BoundingBoxTransformInfo struct.
      *
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *boxes, ICLTensor *pred_boxes, const ICLTensor *deltas, const BoundingBoxTransformInfo &info);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *boxes, ICLTensor *pred_boxes, const ICLTensor *deltas, const BoundingBoxTransformInfo &info);
 
     /** Static function to check if given info will lead to a valid configuration of @ref CLBoundingBoxTransform
      *
diff --git a/arm_compute/core/CL/kernels/CLBox3x3Kernel.h b/arm_compute/core/CL/kernels/CLBox3x3Kernel.h
index 572ae87d9a..ea3c1c1f3e 100644
--- a/arm_compute/core/CL/kernels/CLBox3x3Kernel.h
+++ b/arm_compute/core/CL/kernels/CLBox3x3Kernel.h
@@ -50,7 +50,7 @@ public:
      * @param[out] output           The output tensor. Data types supported: U8.
      * @param[in]  border_undefined True if the border mode is undefined. False if it's replicate or constant.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, bool border_undefined);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, bool border_undefined);
 
     //Inherited methods overriden:
     BorderSize border_size() const override;
diff --git a/arm_compute/core/CL/kernels/CLCannyEdgeKernel.h b/arm_compute/core/CL/kernels/CLCannyEdgeKernel.h
index 67c23dd811..40ad4dcd84 100644
--- a/arm_compute/core/CL/kernels/CLCannyEdgeKernel.h
+++ b/arm_compute/core/CL/kernels/CLCannyEdgeKernel.h
@@ -65,7 +65,7 @@ public:
      * @param[out] phase           Destination tensor - Quantized phase. Data types supported: U8.
      * @param[in]  norm_type       Normalization type. if 1, L1-Norm otherwise L2-Norm.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *gx, const ICLTensor *gy, ICLTensor *magnitude, ICLTensor *phase, int32_t norm_type);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *gx, const ICLTensor *gy, ICLTensor *magnitude, ICLTensor *phase, int32_t norm_type);
 
     // Inherited methods overridden:
     void run(const Window &window, cl::CommandQueue &queue) override;
@@ -111,7 +111,7 @@ public:
      * @param[in]  lower_thr        Lower threshold.
      * @param[in]  border_undefined True if the border mode is undefined. False if it's replicate or constant.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *magnitude, const ICLTensor *phase, ICLTensor *output, int32_t lower_thr, bool border_undefined);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *magnitude, const ICLTensor *phase, ICLTensor *output, int32_t lower_thr, bool border_undefined);
 
     // Inherited methods overridden:
     void run(const Window &window, cl::CommandQueue &queue) override;
@@ -167,7 +167,7 @@ public:
      * @param[in,out] l1_stack_counter Tensor for counting the elements in the L1 stack of each pixel. Data types supported: U8.
      *                                              Expected to be initialized to 0 before each run.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, int32_t upper_thr, int32_t lower_thr,
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, int32_t upper_thr, int32_t lower_thr,
                    ICLTensor *visited, ICLTensor *recorded, ICLTensor *l1_stack, ICLTensor *l1_stack_counter);
 
     // Inherited methods overridden:
diff --git a/arm_compute/core/CL/kernels/CLChannelCombineKernel.h b/arm_compute/core/CL/kernels/CLChannelCombineKernel.h
index 60d0bd4a45..32ddf152c3 100644
--- a/arm_compute/core/CL/kernels/CLChannelCombineKernel.h
+++ b/arm_compute/core/CL/kernels/CLChannelCombineKernel.h
@@ -69,7 +69,7 @@ public:
      * @param[in]  plane3          The 2D plane that forms channel 3. Must be of U8 format.
      * @param[out] output          The single planar output tensor.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *plane0, const ICLTensor *plane1, const ICLTensor *plane2, const ICLTensor *plane3, ICLTensor *output);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *plane0, const ICLTensor *plane1, const ICLTensor *plane2, const ICLTensor *plane3, ICLTensor *output);
     /** Configure function's inputs and outputs.
      *
      * @param[in]  plane0 The 2D plane that forms channel 0. Must be of U8 format.
@@ -86,7 +86,7 @@ public:
      * @param[in]  plane2          The 2D plane that forms channel 2. Must be of U8 format.
      * @param[out] output          The multi planar output tensor.
      */
-    void configure(CLCompileContext &compile_context, const ICLImage *plane0, const ICLImage *plane1, const ICLImage *plane2, ICLMultiImage *output);
+    void configure(const CLCompileContext &compile_context, const ICLImage *plane0, const ICLImage *plane1, const ICLImage *plane2, ICLMultiImage *output);
 
     // Inherited methods overridden:
     void run(const Window &window, cl::CommandQueue &queue) override;
diff --git a/arm_compute/core/CL/kernels/CLChannelExtractKernel.h b/arm_compute/core/CL/kernels/CLChannelExtractKernel.h
index 1f2cc8900a..6a0c4bb94e 100644
--- a/arm_compute/core/CL/kernels/CLChannelExtractKernel.h
+++ b/arm_compute/core/CL/kernels/CLChannelExtractKernel.h
@@ -65,7 +65,7 @@ public:
      * @param[in]  channel         Channel to extract.
      * @param[out] output          Destination tensor. Must be of U8 format.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, Channel channel, ICLTensor *output);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, Channel channel, ICLTensor *output);
     /** Set the input and output of the kernel
      *
      * @param[in]  input   Multi-planar source image. Formats supported: NV12/NV21/IYUV/YUV444
@@ -80,7 +80,7 @@ public:
      * @param[in]  channel         Channel to extract.
      * @param[out] output          Single-planar 2D destination image. Must be of U8 format.
      */
-    void configure(CLCompileContext &compile_context, const ICLMultiImage *input, Channel channel, ICLImage *output);
+    void configure(const CLCompileContext &compile_context, const ICLMultiImage *input, Channel channel, ICLImage *output);
 
     // Inherited methods overridden:
     void run(const Window &window, cl::CommandQueue &queue) override;
diff --git a/arm_compute/core/CL/kernels/CLChannelShuffleLayerKernel.h b/arm_compute/core/CL/kernels/CLChannelShuffleLayerKernel.h
index 921c20df10..14b59d325f 100644
--- a/arm_compute/core/CL/kernels/CLChannelShuffleLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLChannelShuffleLayerKernel.h
@@ -60,7 +60,7 @@ public:
      * @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(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, unsigned int num_groups);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, unsigned int num_groups);
     /** Static function to check if given info will lead to a valid configuration of @ref CLChannelShuffleLayerKernel
      *
      * @param[in] input      Input tensor info. Data types supported: All.
diff --git a/arm_compute/core/CL/kernels/CLCol2ImKernel.h b/arm_compute/core/CL/kernels/CLCol2ImKernel.h
index 6ef424853e..d0528ed21a 100644
--- a/arm_compute/core/CL/kernels/CLCol2ImKernel.h
+++ b/arm_compute/core/CL/kernels/CLCol2ImKernel.h
@@ -81,7 +81,7 @@ public:
      * @param[in]  convolved_dims  Output convolved dimensions.
      * @param[in]  num_groups      (Optional) Number of groups when performing a grouped convolution
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const Size2D &convolved_dims, unsigned int num_groups = 1);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const Size2D &convolved_dims, unsigned int num_groups = 1);
     /** Static function to check if given info will lead to a valid configuration of @ref CLCol2ImKernel
      *
      * @param[in] input          The input tensor to convert. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32
diff --git a/arm_compute/core/CL/kernels/CLColorConvertKernel.h b/arm_compute/core/CL/kernels/CLColorConvertKernel.h
index 25b95eb42c..2bcd141863 100644
--- a/arm_compute/core/CL/kernels/CLColorConvertKernel.h
+++ b/arm_compute/core/CL/kernels/CLColorConvertKernel.h
@@ -67,7 +67,7 @@ public:
      *                                                          RGBA8888 (if the formats of @p input are UYVY422/YUYV422/RGB888/),
      *                                                          U8 (if the formats of @p input is RGB888)
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output);
     /** Set the input and output of the kernel
      *
      * @param[in]  input  Multi-planar source image. Formats supported: NV12/NV21/IYUV
@@ -80,7 +80,7 @@ public:
      * @param[in]  input           Multi-planar source image. Formats supported: NV12/NV21/IYUV
      * @param[out] output          Single-planar destination image. Formats supported: RGB888/RGBA8888
      */
-    void configure(CLCompileContext &compile_context, const ICLMultiImage *input, ICLImage *output);
+    void configure(const CLCompileContext &compile_context, const ICLMultiImage *input, ICLImage *output);
     /** Set the input and output of the kernel
      *
      * @param[in]  input  Single-planar source image. Formats supported: RGB888/RGBA8888/UYVY422/YUYV422
@@ -93,7 +93,7 @@ public:
      * @param[in]  input           Single-planar source image. Formats supported: RGB888/RGBA8888/UYVY422/YUYV422
      * @param[out] output          Multi-planar destination image. Formats supported: NV12/IYUV/YUV444 (if the formats of @p input are RGB888/RGB8888)
      */
-    void configure(CLCompileContext &compile_context, const ICLImage *input, ICLMultiImage *output);
+    void configure(const CLCompileContext &compile_context, const ICLImage *input, ICLMultiImage *output);
     /** Set the input and output of the kernel
      *
      * @param[in]  input  Multi-planar source image. Formats supported: NV12/NV21/IYUV
@@ -106,7 +106,7 @@ public:
      * @param[in]  input           Multi-planar source image. Formats supported: NV12/NV21/IYUV
      * @param[out] output          Multi-planar destination image. Formats supported: YUV444/IYUV (if the formats of @p input are NV12/NV21)/NV12 (if the format of  @p input is IYUV)
      */
-    void configure(CLCompileContext &compile_context, const ICLMultiImage *input, ICLMultiImage *output);
+    void configure(const CLCompileContext &compile_context, const ICLMultiImage *input, ICLMultiImage *output);
 
     // Inherited methods overridden:
     void run(const Window &window, cl::CommandQueue &queue) override;
diff --git a/arm_compute/core/CL/kernels/CLComparisonKernel.h b/arm_compute/core/CL/kernels/CLComparisonKernel.h
index 15779938b2..d5c5297c61 100644
--- a/arm_compute/core/CL/kernels/CLComparisonKernel.h
+++ b/arm_compute/core/CL/kernels/CLComparisonKernel.h
@@ -64,7 +64,7 @@ public:
      * @param[out] output          Destination tensor. Data types supported: U8.
      * @param[in]  operation       Comparison operation to use.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output, ComparisonOperation operation);
+    void configure(const CLCompileContext &compile_context, 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: All.
diff --git a/arm_compute/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.h b/arm_compute/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.h
index f7e212e1e4..d3e57a6738 100644
--- a/arm_compute/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.h
+++ b/arm_compute/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.h
@@ -69,7 +69,7 @@ public:
      * @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.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const TensorShape &original_input_shape, DataLayout data_layout);
+    void configure(const CLCompileContext &compile_context, 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: All.
diff --git a/arm_compute/core/CL/kernels/CLConvolutionKernel.h b/arm_compute/core/CL/kernels/CLConvolutionKernel.h
index e1cdc88007..b6fe51dbaa 100644
--- a/arm_compute/core/CL/kernels/CLConvolutionKernel.h
+++ b/arm_compute/core/CL/kernels/CLConvolutionKernel.h
@@ -70,7 +70,7 @@ public:
      * @param[in]  scale            Scale of the convolution matrix. If 0 is passed, it will be set to the sum of the coefficients of the convolution or 1 if they add up to 0.
      * @param[in]  border_undefined True if the border mode is undefined. False if it's replicate or constant.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const int16_t *conv, uint32_t scale, bool border_undefined);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const int16_t *conv, uint32_t scale, bool border_undefined);
 
     // Inherited methods overridden:
     BorderSize border_size() const override;
@@ -112,7 +112,7 @@ public:
      * @param[in]  conv             Convolution matrix to apply to the input tensor.
      * @param[in]  border_undefined True if the border mode is undefined. False if it's replicate or constant.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const int16_t *conv, bool border_undefined);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const int16_t *conv, bool border_undefined);
 
     // Inherited methods overridden:
     BorderSize border_size() const override;
@@ -153,7 +153,7 @@ public:
      * @param[in]  border_undefined True if the border mode is undefined. False if it's replicate or constant.
      * @param[in]  data_type        Data type to use for intermeidate result. @sa data_type_for_convolution
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const int16_t *conv, uint32_t scale, bool border_undefined, DataType data_type = DataType::S32);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const int16_t *conv, uint32_t scale, bool border_undefined, DataType data_type = DataType::S32);
 
     // Inherited methods overridden:
     BorderSize border_size() const override;
@@ -209,7 +209,7 @@ public:
      * @param[in]  scale            Scale of the convolution matrix. If 0 is passed, it will be set to the sum of the coefficients of the convolution or 1 if they add up to 0.
      * @param[in]  border_undefined True if the border mode is undefined. False if it's replicate or constant.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const int16_t *conv, uint32_t width, uint32_t height, uint32_t scale, bool border_undefined);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const int16_t *conv, uint32_t width, uint32_t height, uint32_t scale, bool border_undefined);
 
     // Inherited methods overridden:
     void run(const Window &window, cl::CommandQueue &queue) override;
diff --git a/arm_compute/core/CL/kernels/CLCopyKernel.h b/arm_compute/core/CL/kernels/CLCopyKernel.h
index 1774f8ccad..05dff8ed0c 100644
--- a/arm_compute/core/CL/kernels/CLCopyKernel.h
+++ b/arm_compute/core/CL/kernels/CLCopyKernel.h
@@ -61,7 +61,7 @@ public:
      * @param[in]  padding         (Optional) Padding to be applied to the input tensor
      * @param[in]  output_window   (Optional) Window to be used in case only copying into part of a tensor. Default is nullptr.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const PaddingList &padding = PaddingList(), Window *output_window = nullptr);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const PaddingList &padding = PaddingList(), Window *output_window = nullptr);
     /** Static function to check if given info will lead to a valid configuration of @ref CLCopyKernel
      *
      * @param[in] input         Source tensor info. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
diff --git a/arm_compute/core/CL/kernels/CLCropKernel.h b/arm_compute/core/CL/kernels/CLCropKernel.h
index 103986a5f8..a1c6f901eb 100644
--- a/arm_compute/core/CL/kernels/CLCropKernel.h
+++ b/arm_compute/core/CL/kernels/CLCropKernel.h
@@ -71,7 +71,7 @@ public:
      * @param[in]  extrapolation_value Value to be used for values outside of the image. Default is 0.
      * @param[in]  output_window       Output window to be used in case cropped image is being copied into a tensor. Default is nullptr.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, Coordinates2D start, Coordinates2D end, uint32_t batch_index, float extrapolation_value = 0,
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, Coordinates2D start, Coordinates2D end, uint32_t batch_index, float extrapolation_value = 0,
                    Window *output_window = nullptr);
 
     /** Static function to check if given info will lead to a valid configuration of @ref CLStridedSliceKernel
diff --git a/arm_compute/core/CL/kernels/CLDeconvolutionLayerUpsampleKernel.h b/arm_compute/core/CL/kernels/CLDeconvolutionLayerUpsampleKernel.h
index 7e8a45fd8f..0c65f519cc 100644
--- a/arm_compute/core/CL/kernels/CLDeconvolutionLayerUpsampleKernel.h
+++ b/arm_compute/core/CL/kernels/CLDeconvolutionLayerUpsampleKernel.h
@@ -62,7 +62,7 @@ public:
      * @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(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const PadStrideInfo &info);
+    void configure(const CLCompileContext &compile_context, 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: All.
diff --git a/arm_compute/core/CL/kernels/CLDeconvolutionReshapeOutputKernel.h b/arm_compute/core/CL/kernels/CLDeconvolutionReshapeOutputKernel.h
index daeb8c1f9c..292c561e46 100644
--- a/arm_compute/core/CL/kernels/CLDeconvolutionReshapeOutputKernel.h
+++ b/arm_compute/core/CL/kernels/CLDeconvolutionReshapeOutputKernel.h
@@ -79,7 +79,7 @@ public:
      * @param[in]  weights_info    Deconvolution weights tensor info. Supported data types: same as @p input.  Supported data layouts: same as @p input.
      * @param[in]  deconv_info     Contains padding and policies to be used in the deconvolution, this is described in @ref PadStrideInfo. This kernel supports only stride_x = weights.width && stride_y = weights.height. Moreover, padding is not supported.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *bias, ICLTensor *output, const ITensorInfo *input_info, const ITensorInfo *weights_info,
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *bias, ICLTensor *output, const ITensorInfo *input_info, const ITensorInfo *weights_info,
                    const PadStrideInfo &deconv_info);
 
     /** Static function to check if given info will lead to a valid configuration of @ref  CLDeconvolutionReshapeOutputKernel.
diff --git a/arm_compute/core/CL/kernels/CLDepthConcatenateLayerKernel.h b/arm_compute/core/CL/kernels/CLDepthConcatenateLayerKernel.h
index 7b594417d6..5fe826d090 100644
--- a/arm_compute/core/CL/kernels/CLDepthConcatenateLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLDepthConcatenateLayerKernel.h
@@ -72,7 +72,7 @@ public:
      * @note: The gaps between the two lowest dimensions of input and output need to be divisible by 2.
      *
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, unsigned int depth_offset, ICLTensor *output);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, unsigned int depth_offset, ICLTensor *output);
     /**  Static function to check if given info will lead to a valid configuration of @ref CLDepthConcatenateLayerKernel
      *
      * @param[in] input        Input tensor info. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32
diff --git a/arm_compute/core/CL/kernels/CLDepthConvertLayerKernel.h b/arm_compute/core/CL/kernels/CLDepthConvertLayerKernel.h
index 8bbf9b3dce..66eb6222b2 100644
--- a/arm_compute/core/CL/kernels/CLDepthConvertLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLDepthConvertLayerKernel.h
@@ -75,7 +75,7 @@ public:
      * @param[in]  policy          Conversion policy
      * @param[in]  shift           Value for down/up conversions. Must be 0 <= shift < 8.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, ConvertPolicy policy, uint32_t shift);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, ConvertPolicy policy, uint32_t shift);
     /** Static function to check if given info will lead to a valid configuration of @ref CLDepthConvertLayerKernel
      *
      * @param[in] input  Source tensor info. Data types supported: U8/S8/QSYMM8_PER_CHANNEL/U16/S16/U32/S32/F16/F32.
diff --git a/arm_compute/core/CL/kernels/CLDepthToSpaceLayerKernel.h b/arm_compute/core/CL/kernels/CLDepthToSpaceLayerKernel.h
index 541506b521..87ac3c1ec1 100644
--- a/arm_compute/core/CL/kernels/CLDepthToSpaceLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLDepthToSpaceLayerKernel.h
@@ -61,7 +61,7 @@ public:
      * @param[out] output          Tensor output. Data types supported: same as @p input
      * @param[in]  block_shape     Block shape value.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, int32_t block_shape);
+    void configure(const CLCompileContext &compile_context, 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: All.
diff --git a/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NCHWKernel.h b/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NCHWKernel.h
index f68fde4737..6cf0326467 100644
--- a/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NCHWKernel.h
+++ b/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NCHWKernel.h
@@ -75,7 +75,7 @@ public:
      * @param[in]  output_shifts      (Optional) Output shifts tensor for quantized computations. In case of per-channel quantization,
      *                                the number of multipliers must be equal to the number of filters (IFM). Supported data types: S32
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info,
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info,
                    unsigned int depth_multiplier = 1, ActivationLayerInfo act_info = ActivationLayerInfo(), const Size2D &dilation = Size2D(1U, 1U),
                    const ICLTensor *output_multipliers = nullptr, const ICLTensor *output_shifts = nullptr) override;
     /** Static function to check if given info will lead to a valid configuration of @ref CLDepthwiseConvolutionLayer3x3NCHWKernel
diff --git a/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.h b/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.h
index f9fda0a42c..e564cf6fe0 100644
--- a/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.h
+++ b/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.h
@@ -76,7 +76,7 @@ public:
      * @param[in]  output_shifts      (Optional) Output shifts tensor for quantized computations. In case of per-channel quantization,
      *                                the number of multipliers must be equal to the number of filters (IFM). Supported data types: S32
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info,
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info,
                    unsigned int depth_multiplier = 1, ActivationLayerInfo act_info = ActivationLayerInfo(), const Size2D &dilation = Size2D(1U, 1U),
                    const ICLTensor *output_multipliers = nullptr, const ICLTensor *output_shifts = nullptr) override;
     /** Static function to check if given info will lead to a valid configuration of @ref CLDepthwiseConvolutionLayer3x3NHWCKernel
diff --git a/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayerNativeKernel.h b/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayerNativeKernel.h
index db26b4a06f..8847cf9c46 100644
--- a/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayerNativeKernel.h
+++ b/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayerNativeKernel.h
@@ -87,7 +87,7 @@ public:
      * @param[in]  output_shifts      (Optional) Output shifts tensor for quantized computations. In case of per-channel quantization,
      *                                the number of multipliers must be equal to the number of filters (IFM). Supported data types: S32
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const DWCWeightsKernelInfo &dwc_weights_info,
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const DWCWeightsKernelInfo &dwc_weights_info,
                    const DWCKernelInfo &dwc_info, const PadStrideInfo &conv_info, unsigned int depth_multiplier = 1, const Size2D &dilation = Size2D(1U, 1U),
                    const ICLTensor *output_multipliers = nullptr, const ICLTensor *output_shifts = nullptr);
     /** Static function to check if given info will lead to a valid configuration of @ref CLDepthwiseConvolutionLayerNativeKernel
diff --git a/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayerReshapeWeightsKernel.h b/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayerReshapeWeightsKernel.h
index e7fc6f8d81..8dc5d32e4f 100644
--- a/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayerReshapeWeightsKernel.h
+++ b/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayerReshapeWeightsKernel.h
@@ -59,7 +59,7 @@ public:
      * @param[out] output          The output tensor of dimension [W*H*C0, ceil(IFM/C0)]. C0 is the number of channels read by each thread. Data types supported: same as @p weights.
      * @param[in]  info            Depthwise convolution information to reshape the input tensor.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const DepthwiseConvolutionReshapeInfo &info);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const DepthwiseConvolutionReshapeInfo &info);
 
     /** Static function to check if given info will lead to a valid configuration of @ref CLDepthwiseConvolutionLayer3x3NHWCKernel
      *
diff --git a/arm_compute/core/CL/kernels/CLDequantizationLayerKernel.h b/arm_compute/core/CL/kernels/CLDequantizationLayerKernel.h
index 4cb1339300..bb154f1a5b 100644
--- a/arm_compute/core/CL/kernels/CLDequantizationLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLDequantizationLayerKernel.h
@@ -58,7 +58,7 @@ public:
      * @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(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output);
+    void configure(const CLCompileContext &compile_context, 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/QASYMM8_SIGNED/QSYMM8_PER_CHANNEL/QSYMM8/QSYMM16.
diff --git a/arm_compute/core/CL/kernels/CLDerivativeKernel.h b/arm_compute/core/CL/kernels/CLDerivativeKernel.h
index 5d5ad860f3..cd8ae90c2d 100644
--- a/arm_compute/core/CL/kernels/CLDerivativeKernel.h
+++ b/arm_compute/core/CL/kernels/CLDerivativeKernel.h
@@ -66,7 +66,7 @@ public:
      * @param[out] output_y         (Optional) Destination tensor for the Y gradient, Data types supported: S16.
      * @param[in]  border_undefined True if the border mode is undefined. False if it's replicate or constant.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output_x, ICLTensor *output_y, bool border_undefined);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output_x, ICLTensor *output_y, bool border_undefined);
 
     // Inherited methods overridden:
     void run(const Window &window, cl::CommandQueue &queue) override;
diff --git a/arm_compute/core/CL/kernels/CLDilateKernel.h b/arm_compute/core/CL/kernels/CLDilateKernel.h
index 9c41a84b31..45f5fe0764 100644
--- a/arm_compute/core/CL/kernels/CLDilateKernel.h
+++ b/arm_compute/core/CL/kernels/CLDilateKernel.h
@@ -50,7 +50,7 @@ public:
      * @param[out] output           The output tensor. Data types supported: U8.
      * @param[in]  border_undefined True if the border mode is undefined. False if it's replicate or constant.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, bool border_undefined);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, bool border_undefined);
 
     // Inherited methods overridden:
     BorderSize border_size() const override;
diff --git a/arm_compute/core/CL/kernels/CLDirectConvolutionLayerKernel.h b/arm_compute/core/CL/kernels/CLDirectConvolutionLayerKernel.h
index f1409b6339..489d7c27c5 100644
--- a/arm_compute/core/CL/kernels/CLDirectConvolutionLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLDirectConvolutionLayerKernel.h
@@ -88,7 +88,7 @@ public:
      *                             The 3rd dimensions must be equal to the 4th dimension of the @p kernels tensor. Data types supported: Same as @p input.
      * @param[in]  conv_info       Contains padding and stride information described in @ref PadStrideInfo.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info);
     /** Static function to check if given info will lead to a valid configuration of @ref CLDirectConvolutionLayerKernel
      *
      * @param[in] input     The input tensor to convolve. 3 lower dimensions represent a single input [width, height, IFM],
diff --git a/arm_compute/core/CL/kernels/CLElementWiseUnaryLayerKernel.h b/arm_compute/core/CL/kernels/CLElementWiseUnaryLayerKernel.h
index 1f76992b96..e190bdebbe 100644
--- a/arm_compute/core/CL/kernels/CLElementWiseUnaryLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLElementWiseUnaryLayerKernel.h
@@ -50,7 +50,7 @@ public:
      * @param[out] output          Output tensor. Data types supported: Same as @p input.
      * @param[in]  op              Element wise unary operation to perform.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const ElementWiseUnary &op);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const ElementWiseUnary &op);
     /** Static function to check if given info will lead to a valid configuration of @ref CLElementWiseUnaryLayerKernel
      *
      * @param[in] input  First tensor input info. Data types supported: F16/F32.
diff --git a/arm_compute/core/CL/kernels/CLElementwiseOperationKernel.h b/arm_compute/core/CL/kernels/CLElementwiseOperationKernel.h
index 2f1060126a..4d3d4bc834 100644
--- a/arm_compute/core/CL/kernels/CLElementwiseOperationKernel.h
+++ b/arm_compute/core/CL/kernels/CLElementwiseOperationKernel.h
@@ -99,7 +99,7 @@ protected:
     /** Commmon configure function for element-wise operators with no additional options (e.g., Div, Min, Max, SquaredDiff)
      *
      */
-    void configure_common(CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output);
+    void configure_common(const CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output);
 
     ActivationLayerInfo _act_info;
 
@@ -138,7 +138,7 @@ public:
      * @param[in] policy          Policy to use to handle overflow.
      * @param[in] act_info        (Optional) Activation layer information in case of a fused activation.
      */
-    void configure(CLCompileContext &compile_context, ArithmeticOperation op, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output, const ConvertPolicy &policy,
+    void configure(const CLCompileContext &compile_context, ArithmeticOperation op, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output, const ConvertPolicy &policy,
                    const ActivationLayerInfo &act_info = ActivationLayerInfo());
 
     /** Static function to check if given info will lead to a valid configuration of @ref CLSaturatedArithmeticOperationKernel
@@ -194,7 +194,7 @@ public:
      * @param[in] output          Output tensor. Data types supported: Same as @p input1.
      * @param[in] act_info        (Optional) Activation layer information in case of a fused activation.
      */
-    void configure(CLCompileContext &compile_context, ArithmeticOperation op, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output,
+    void configure(const CLCompileContext &compile_context, ArithmeticOperation op, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output,
                    const ActivationLayerInfo &act_info = ActivationLayerInfo());
 
     /** Static function to check if given info will lead to a valid configuration of @ref CLArithmeticOperationKernel
diff --git a/arm_compute/core/CL/kernels/CLErodeKernel.h b/arm_compute/core/CL/kernels/CLErodeKernel.h
index 8ba6ff8408..cbc748194c 100644
--- a/arm_compute/core/CL/kernels/CLErodeKernel.h
+++ b/arm_compute/core/CL/kernels/CLErodeKernel.h
@@ -50,7 +50,7 @@ public:
      * @param[out] output           The output tensor. Data types supported: U8.
      * @param[in]  border_undefined True if the border mode is undefined. False if it's replicate or constant.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, bool border_undefined);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, bool border_undefined);
 
     // Inherited methods overridden:
     BorderSize border_size() const override;
diff --git a/arm_compute/core/CL/kernels/CLFFTDigitReverseKernel.h b/arm_compute/core/CL/kernels/CLFFTDigitReverseKernel.h
index eac03ff868..a8da1246bb 100644
--- a/arm_compute/core/CL/kernels/CLFFTDigitReverseKernel.h
+++ b/arm_compute/core/CL/kernels/CLFFTDigitReverseKernel.h
@@ -65,7 +65,7 @@ public:
      * @param[in]  idx             Digit reverse index tensor. Data type supported: U32
      * @param[in]  config          Kernel configuration.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const ICLTensor *idx, const FFTDigitReverseKernelInfo &config);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const ICLTensor *idx, const FFTDigitReverseKernelInfo &config);
     /** Static function to check if given info will lead to a valid configuration of @ref CLFFTDigitReverseKernel
      *
      * @param[in] input  Source tensor info. Data types supported: F32.
diff --git a/arm_compute/core/CL/kernels/CLFFTRadixStageKernel.h b/arm_compute/core/CL/kernels/CLFFTRadixStageKernel.h
index 85bf4cce66..e3f53462d9 100644
--- a/arm_compute/core/CL/kernels/CLFFTRadixStageKernel.h
+++ b/arm_compute/core/CL/kernels/CLFFTRadixStageKernel.h
@@ -69,7 +69,7 @@ public:
      * @param[out]    output          Destination tensor. Can be nullptr. Data type supported: same as @p input
      * @param[in]     config          FFT descriptor metadata.
      */
-    void configure(CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, const FFTRadixStageKernelInfo &config);
+    void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, const FFTRadixStageKernelInfo &config);
     /** Static function to check if given info will lead to a valid configuration of @ref CLFFTRadixStageKernel
      *
      * @param[in] input  Source tensor info. Data types supported: F32.
diff --git a/arm_compute/core/CL/kernels/CLFFTScaleKernel.h b/arm_compute/core/CL/kernels/CLFFTScaleKernel.h
index cd4fe58b9c..d0d2b7613c 100644
--- a/arm_compute/core/CL/kernels/CLFFTScaleKernel.h
+++ b/arm_compute/core/CL/kernels/CLFFTScaleKernel.h
@@ -63,7 +63,7 @@ public:
      * @param[out]    output          Destination tensor. Data type supported: same as @p input
      * @param[in]     config          Kernel configuration
      */
-    void configure(CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, const FFTScaleKernelInfo &config);
+    void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, const FFTScaleKernelInfo &config);
     /** Static function to check if given info will lead to a valid configuration of @ref CLFFTScaleKernel
      *
      * @param[in] input  Source tensor info. Data types supported: F32.
diff --git a/arm_compute/core/CL/kernels/CLFastCornersKernel.h b/arm_compute/core/CL/kernels/CLFastCornersKernel.h
index 2a6102036f..1a0d4e36a5 100644
--- a/arm_compute/core/CL/kernels/CLFastCornersKernel.h
+++ b/arm_compute/core/CL/kernels/CLFastCornersKernel.h
@@ -75,7 +75,7 @@ public:
      * @param[in]  non_max_suppression True if non-maxima suppresion is applied, false otherwise.
      * @param[in]  border_mode         Strategy to use for borders.
      */
-    void configure(CLCompileContext &compile_context, const ICLImage *input, ICLImage *output, float threshold, bool non_max_suppression, BorderMode border_mode);
+    void configure(const CLCompileContext &compile_context, const ICLImage *input, ICLImage *output, float threshold, bool non_max_suppression, BorderMode border_mode);
 
     // Inherited methods overridden
     void run(const Window &window, cl::CommandQueue &queue) override;
@@ -119,7 +119,7 @@ public:
      * @param[out] corners         Array of keypoints to store the results.
      * @param[out] num_buffers     Number of keypoints to store the results.
      */
-    void configure(CLCompileContext &compile_context, const ICLImage *input, bool update_number, ICLKeyPointArray *corners, cl::Buffer *num_buffers);
+    void configure(const CLCompileContext &compile_context, const ICLImage *input, bool update_number, ICLKeyPointArray *corners, cl::Buffer *num_buffers);
 
     // Inherited methods overridden:
     void run(const Window &window, cl::CommandQueue &queue) override;
diff --git a/arm_compute/core/CL/kernels/CLFillBorderKernel.h b/arm_compute/core/CL/kernels/CLFillBorderKernel.h
index 226b611bcb..d00ea55a83 100644
--- a/arm_compute/core/CL/kernels/CLFillBorderKernel.h
+++ b/arm_compute/core/CL/kernels/CLFillBorderKernel.h
@@ -65,7 +65,7 @@ public:
      * @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.
      */
-    void configure(CLCompileContext &compile_context, ICLTensor *tensor, BorderSize border_size, BorderMode border_mode, const PixelValue &constant_border_value = PixelValue());
+    void configure(const CLCompileContext &compile_context, ICLTensor *tensor, BorderSize border_size, BorderMode border_mode, const PixelValue &constant_border_value = PixelValue());
 
     /** Function to set the constant value on fill border kernel depending on type.
      *
diff --git a/arm_compute/core/CL/kernels/CLFlattenLayerKernel.h b/arm_compute/core/CL/kernels/CLFlattenLayerKernel.h
index b795e03a34..ab009e1aa8 100644
--- a/arm_compute/core/CL/kernels/CLFlattenLayerKernel.h
+++ b/arm_compute/core/CL/kernels/CLFlattenLayerKernel.h
@@ -60,7 +60,7 @@ public:
      * @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
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output);
     /** Static function to check if given info will lead to a valid configuration of @ref CLFlattenLayerKernel
      *
      * @param[in]  input  First input tensor to flatten with at least 3 dimensions.
diff --git a/arm_compute/core/CL/kernels/CLFloorKernel.h b/arm_compute/core/CL/kernels/CLFloorKernel.h
index a3ccb96c61..4d1ed789db 100644
--- a/arm_compute/core/CL/kernels/CLFloorKernel.h
+++ b/arm_compute/core/CL/kernels/CLFloorKernel.h
@@ -59,7 +59,7 @@ public:
      * @param[in]  input           Source tensor. Data type supported: F16/F32.
      * @param[out] output          Destination tensor. Same as @p input
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output);
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output);
 
     /** Static function to check if given info will lead to a valid configuration of @ref CLFloorKernel
      *
diff --git a/arm_compute/core/CL/kernels/CLFuseBatchNormalizationKernel.h b/arm_compute/core/CL/kernels/CLFuseBatchNormalizationKernel.h
index 2d62a576bb..2fe6b223ca 100644
--- a/arm_compute/core/CL/kernels/CLFuseBatchNormalizationKernel.h
+++ b/arm_compute/core/CL/kernels/CLFuseBatchNormalizationKernel.h
@@ -81,7 +81,7 @@ public:
      * @param[in]  epsilon         (Optional) Batch normalization layer epsilon parameter. Defaults to 0.001f.
      * @param[in]  fbn_type        (Optional) Fused batch normalization type. Defaults to CONVOLUTION.
      */
-    void configure(CLCompileContext &compile_context, const ICLTensor *input_weights, const ICLTensor *bn_mean, const ICLTensor *bn_var, ICLTensor *fused_weights, ICLTensor *fused_bias,
+    void configure(const CLCompileContext &compile_context, const ICLTensor *input_weights, const ICLTensor *bn_mean, const ICLTensor *bn_var, ICLTensor *fused_weights, ICLTensor *fused_bias,
                    const ICLTensor *input_bias = nullptr, const ICLTensor *bn_beta = nullptr, const ICLTensor *bn_gamma = nullptr,
                    float epsilon = 0.001f, FuseBatchNormalizationType fbn_type = FuseBatchNormalizationType::CONVOLUTION);
     /** Static function to check if given info will lead to a valid configuration of @ref CLFuseBatchNormalizationKernel
diff --git a/arm_compute/core/CL/kernels/ICLDepthwiseConvolutionLayer3x3Kernel.h b/arm_compute/core/CL/kernels/ICLDepthwiseConvolutionLayer3x3Kernel.h
index 040ca157de..f0f7754960 100644
--- a/arm_compute/core/CL/kernels/ICLDepthwiseConvolutionLayer3x3Kernel.h
+++ b/arm_compute/core/CL/kernels/ICLDepthwiseConvolutionLayer3x3Kernel.h
@@ -86,7 +86,7 @@ public:
      * @param[in]  output_shifts      (Optional) Output shifts tensor for quantized computations. In case of per-channel quantization,
      *                                the number of multipliers must be equal to the number of filters (IFM). Supported data types: S32
      */
-    virtual void configure(CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info,
+    virtual void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info,
                            unsigned int depth_multiplier = 1, ActivationLayerInfo act_info = ActivationLayerInfo(), const Size2D &dilation = Size2D(1U, 1U),
                            const ICLTensor *output_multipliers = nullptr, const ICLTensor *output_shifts = nullptr) = 0;