Common interface for all kernels implemented in C++. More...

#include <ICPPKernel.h>

Collaboration diagram for ICPPKernel:

Public Member Functions
virtual	~ICPPKernel ()=default
	Default destructor. More...

virtual void	run (const Window &window, const ThreadInfo &info)
	Execute the kernel on the passed window. More...

virtual void	run_nd (const Window &window, const ThreadInfo &info, const Window &thread_locator)
	legacy compatibility layer for implemantions which do not support thread_locator In these cases we simply narrow the interface down the legacy version More...

virtual void	run_op (ITensorPack &tensors, const Window &window, const ThreadInfo &info)
	Execute the kernel on the passed window. More...

virtual const char *	name () const =0
	Name of the kernel. More...

Public Member Functions inherited from IKernel
	IKernel ()
	Constructor. More...

virtual	~IKernel ()=default
	Destructor. More...

virtual bool	is_parallelisable () const
	Indicates whether or not the kernel is parallelisable. More...

virtual BorderSize	border_size () const
	The size of the border for that kernel. More...

const Window &	window () const
	The maximum window the kernel can be executed on. More...

bool	is_window_configured () const
	Function to check if the embedded window of this kernel has been configured. More...

Detailed Description

Common interface for all kernels implemented in C++.

Definition at line 38 of file ICPPKernel.h.

Constructor & Destructor Documentation

◆ ~ICPPKernel()

virtual ~ICPPKernel ( )

virtualdefault

Default destructor.

Member Function Documentation

◆ name()

virtual const char* name ( ) const

pure virtual

Name of the kernel.

Returns: Kernel name

Implemented in NEWinogradLayerTransformWeightsKernel< T, OutputTileRows, OutputTileCols, KernelRows, KernelCols >, NEWinogradLayerTransformOutputKernel< T, OutputTileRows, OutputTileCols, KernelRows, KernelCols >, NEWinogradLayerTransformInputKernel< T, OutputTileRows, OutputTileCols, KernelRows, KernelCols >, CpuComplexMulKernel, NEGEMMLowpMatrixBReductionKernel, CpuLogits1DSoftmaxKernel< IS_LOG >, CpuSubKernel, CpuMulKernel, NEGEMMLowpMatrixAReductionKernel, CpuDirectConvolutionOutputStageKernel, CpuAddKernel, CpuDirectConvolutionKernel, CpuScaleKernel, NEGEMMTranspose1xWKernel, CpuConcatenateDepthKernel, CpuConvertFullyConnectedWeightsKernel, CpuPoolingKernel, CpuFloorKernel, CpuConcatenateBatchKernel, CpuQuantizationKernel, CpuActivationKernel, CpuPermuteKernel, NEGEMMLowpOffsetContributionOutputStageKernel, NELogicalKernel, CpuConcatenateHeightKernel, CpuConcatenateWidthKernel, CpuCopyKernel, CpuDequantizationKernel, CpuReshapeKernel, CpuLogits1DMaxKernel, NEIm2ColKernel, CpuTransposeKernel, NEWeightsReshapeKernel, CpuPoolingAssemblyWrapperKernel, NEGatherKernel, NECol2ImKernel, NEGEMMInterleave4x4Kernel, CpuFillKernel, NEGEMMLowpQuantizeDownInt32ScaleKernel, NEGEMMLowpOffsetContributionKernel, NEGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel, NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel, NETileKernel, NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel, CpuElementwiseUnaryKernel, CpuDepthwiseConvolutionNativeKernel, CpuElementwiseKernel, NEGEMMLowpMatrixMultiplyKernel, NEGEMMMatrixAdditionKernel, NESelectKernel, NEReductionOperationKernel, NEFFTRadixStageKernel, NEGEMMMatrixMultiplyKernel, NERangeKernel, CPPNonMaximumSuppressionKernel, NEBitwiseAndKernel, NEBitwiseNotKernel, NEBitwiseOrKernel, NEBitwiseXorKernel, NEMeanStdDevNormalizationKernel, NEMinMaxLayerKernel, NEStridedSliceKernel, CPPPermuteKernel, NECropKernel, NEFFTScaleKernel, NEFillBorderKernel, CPPUpsampleKernel, NEBatchNormalizationLayerKernel, NEConvertQuantizedSignednessKernel, NEDepthConvertLayerKernel, NEFFTDigitReverseKernel, NESpaceToBatchLayerKernel, NEStackLayerKernel, CPPBoxWithNonMaximaSuppressionLimitKernel, NEChannelShuffleLayerKernel, NEDepthToSpaceLayerKernel, NEFuseBatchNormalizationKernel, NEInstanceNormalizationLayerKernel, NENormalizationLayerKernel, NEQLSTMLayerNormalizationKernel, NERemapKernel, NEReorgLayerKernel, NEROIAlignLayerKernel, NESpaceToDepthLayerKernel, CPPTopKVKernel, NEBatchToSpaceLayerKernel, NEBoundingBoxTransformKernel, NEL2NormalizeLayerKernel, NEMaxUnpoolingLayerKernel, NEPadLayerKernel, NEPriorBoxLayerKernel, NEReverseKernel, NEComputeAllAnchorsKernel, and NEROIPoolingLayerKernel.

◆ run()

virtual void run	(	const Window &	window,
		const ThreadInfo &	info
	)

inlinevirtual

Execute the kernel on the passed window.

Warning: If is_parallelisable() returns false then the passed window must be equal to window()

Note: The window has to be a region within the window returned by the window() method; The width of the window has to be a multiple of num_elems_processed_per_iteration().

Parameters

[in]	window	Region on which to execute the kernel. (Must be a region of the window returned by window())
[in]	info	Info about executing thread and CPU.

Reimplemented in NEWinogradLayerTransformWeightsKernel< T, OutputTileRows, OutputTileCols, KernelRows, KernelCols >, NEWinogradLayerTransformOutputKernel< T, OutputTileRows, OutputTileCols, KernelRows, KernelCols >, NEWinogradLayerTransformInputKernel< T, OutputTileRows, OutputTileCols, KernelRows, KernelCols >, NEGEMMLowpMatrixBReductionKernel, NEGEMMLowpMatrixAReductionKernel, NEGEMMLowpOffsetContributionOutputStageKernel, NEIm2ColKernel, NEGEMMTranspose1xWKernel, NEWeightsReshapeKernel, NESpaceToBatchLayerKernel, NEFuseBatchNormalizationKernel, NEGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel, NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel, NEBatchNormalizationLayerKernel, NECol2ImKernel, NEGEMMLowpOffsetContributionKernel, NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel, NECropKernel, NEBatchToSpaceLayerKernel, NEGEMMLowpQuantizeDownInt32ScaleKernel, NEROIAlignLayerKernel, NEDepthConvertLayerKernel, NEGEMMInterleave4x4Kernel, NEGEMMMatrixMultiplyKernel, CPPNonMaximumSuppressionKernel, NEStackLayerKernel, NEBoundingBoxTransformKernel, NEFFTRadixStageKernel, NEGEMMLowpMatrixMultiplyKernel, NEReductionOperationKernel, NEGEMMMatrixAdditionKernel, NESelectKernel, NEFillBorderKernel, NEGatherKernel, NEMinMaxLayerKernel, NEL2NormalizeLayerKernel, NEPadLayerKernel, NEMeanStdDevNormalizationKernel, NEMaxUnpoolingLayerKernel, NENormalizationLayerKernel, NERangeKernel, NEFFTDigitReverseKernel, CPPPermuteKernel, CPPTopKVKernel, NEReorgLayerKernel, NEFFTScaleKernel, NEComputeAllAnchorsKernel, NEInstanceNormalizationLayerKernel, NEQLSTMLayerNormalizationKernel, NEPriorBoxLayerKernel, NEChannelShuffleLayerKernel, NEDepthToSpaceLayerKernel, NEReverseKernel, NESpaceToDepthLayerKernel, NEConvertQuantizedSignednessKernel, NETileKernel, CPPBoxWithNonMaximaSuppressionLimitKernel, NEROIPoolingLayerKernel, NEBitwiseAndKernel, NEBitwiseOrKernel, NEBitwiseXorKernel, NERemapKernel, CPPUpsampleKernel, and NEBitwiseNotKernel.

Definition at line 55 of file ICPPKernel.h.

     {
         ARM_COMPUTE_UNUSED(window, info);
         ARM_COMPUTE_ERROR("default implementation of legacy run() virtual member function invoked");
     }

References ARM_COMPUTE_ERROR, ARM_COMPUTE_UNUSED, arm_compute::test::validation::info, and IKernel::window().

Referenced by ICPPKernel::run_nd(), and SingleThreadScheduler::schedule().

◆ run_nd()

virtual void run_nd	(	const Window &	window,
		const ThreadInfo &	info,
		const Window &	thread_locator
	)

inlinevirtual

legacy compatibility layer for implemantions which do not support thread_locator In these cases we simply narrow the interface down the legacy version

Parameters

[in]	window	Region on which to execute the kernel. (Must be a region of the window returned by window())
[in]	info	Info about executing thread and CPU.
[in]	thread_locator	Specifies "where" the current thread is in the multi-dimensional space

Definition at line 68 of file ICPPKernel.h.

     {
         ARM_COMPUTE_UNUSED(thread_locator);
         run(window, info);
     }

References ARM_COMPUTE_UNUSED, arm_compute::test::validation::info, ICPPKernel::run(), and IKernel::window().

◆ run_op()

virtual void run_op	(	ITensorPack &	tensors,
		const Window &	window,
		const ThreadInfo &	info
	)

inlinevirtual

Execute the kernel on the passed window.

Warning: If is_parallelisable() returns false then the passed window must be equal to window()

Note: The window has to be a region within the window returned by the window() method; The width of the window has to be a multiple of num_elems_processed_per_iteration().

Parameters

[in]	tensors	A vector containing the tensors to operate on.
[in]	window	Region on which to execute the kernel. (Must be a region of the window returned by window())
[in]	info	Info about executing thread and CPU.

Reimplemented in CpuComplexMulKernel, CpuLogits1DSoftmaxKernel< IS_LOG >, NEStridedSliceKernel, CpuSubKernel, CpuDepthwiseConvolutionNativeKernel, CpuPoolingAssemblyWrapperKernel, NEFillBorderKernel, CpuMulKernel, CpuDirectConvolutionOutputStageKernel, CpuAddKernel, CpuDirectConvolutionKernel, CpuElementwiseUnaryKernel, CpuScaleKernel, CpuConcatenateDepthKernel, CpuConvertFullyConnectedWeightsKernel, CpuPoolingKernel, CpuFloorKernel, CpuConcatenateBatchKernel, CpuQuantizationKernel, CpuActivationKernel, CpuElementwiseKernel, CpuPermuteKernel, NELogicalKernel, CpuConcatenateHeightKernel, CpuConcatenateWidthKernel, CpuCopyKernel, CpuDequantizationKernel, CpuReshapeKernel, CpuLogits1DMaxKernel, CpuTransposeKernel, and CpuFillKernel.

Definition at line 86 of file ICPPKernel.h.

     {
         ARM_COMPUTE_UNUSED(tensors, window, info);
     }

References ARM_COMPUTE_UNUSED, arm_compute::test::validation::info, and IKernel::window().

Referenced by SingleThreadScheduler::schedule_op(), and OMPScheduler::schedule_op().

The documentation for this class was generated from the following file:

arm_compute/core/CPP/ICPPKernel.h

Public Member Functions

Detailed Description

Constructor & Destructor Documentation

◆ ~ICPPKernel()

Member Function Documentation

◆ name()

◆ run()

◆ run_nd()

◆ run_op()