COMPMID-3280: Make all ML primitives for CL use the new interface - Part 1
authorManuel Bottini <manuel.bottini@arm.com>
Wed, 8 Apr 2020 09:15:51 +0000 (10:15 +0100)
committerManuel Bottini <manuel.bottini@arm.com>
Fri, 17 Apr 2020 14:10:38 +0000 (14:10 +0000)
commit4c6bd514a8d424a29b776754f1b3426fa3a8c339
treeed20ea238ae069ab138b62ea6879e0c1b955b606
parentd2f6d96cdc2ad2169c4abd0a8c4884f61ed6d186
COMPMID-3280: Make all ML primitives for CL use the new interface - Part 1

- Only CLKernels have been updated

Change-Id: Ife55b847c2e39e712a186eb6ca452503d5b66937
Signed-off-by: Manuel Bottini <manuel.bottini@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3001
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
271 files changed:
arm_compute/core/CL/CLHelpers.h
arm_compute/core/CL/kernels/CLAbsoluteDifferenceKernel.h
arm_compute/core/CL/kernels/CLAccumulateKernel.h
arm_compute/core/CL/kernels/CLActivationLayerKernel.h
arm_compute/core/CL/kernels/CLArgMinMaxLayerKernel.h
arm_compute/core/CL/kernels/CLBatchConcatenateLayerKernel.h
arm_compute/core/CL/kernels/CLBatchNormalizationLayerKernel.h
arm_compute/core/CL/kernels/CLBatchToSpaceLayerKernel.h
arm_compute/core/CL/kernels/CLBitwiseAndKernel.h
arm_compute/core/CL/kernels/CLBitwiseNotKernel.h
arm_compute/core/CL/kernels/CLBitwiseOrKernel.h
arm_compute/core/CL/kernels/CLBitwiseXorKernel.h
arm_compute/core/CL/kernels/CLBoundingBoxTransformKernel.h
arm_compute/core/CL/kernels/CLBox3x3Kernel.h
arm_compute/core/CL/kernels/CLCannyEdgeKernel.h
arm_compute/core/CL/kernels/CLChannelCombineKernel.h
arm_compute/core/CL/kernels/CLChannelExtractKernel.h
arm_compute/core/CL/kernels/CLChannelShuffleLayerKernel.h
arm_compute/core/CL/kernels/CLCol2ImKernel.h
arm_compute/core/CL/kernels/CLColorConvertKernel.h
arm_compute/core/CL/kernels/CLComparisonKernel.h
arm_compute/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.h
arm_compute/core/CL/kernels/CLConvolutionKernel.h
arm_compute/core/CL/kernels/CLCopyKernel.h
arm_compute/core/CL/kernels/CLCropKernel.h
arm_compute/core/CL/kernels/CLDeconvolutionLayerUpsampleKernel.h
arm_compute/core/CL/kernels/CLDeconvolutionReshapeOutputKernel.h
arm_compute/core/CL/kernels/CLDepthConcatenateLayerKernel.h
arm_compute/core/CL/kernels/CLDepthConvertLayerKernel.h
arm_compute/core/CL/kernels/CLDepthToSpaceLayerKernel.h
arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NCHWKernel.h
arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.h
arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayerNativeKernel.h
arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayerReshapeWeightsKernel.h
arm_compute/core/CL/kernels/CLDequantizationLayerKernel.h
arm_compute/core/CL/kernels/CLDerivativeKernel.h
arm_compute/core/CL/kernels/CLDilateKernel.h
arm_compute/core/CL/kernels/CLDirectConvolutionLayerKernel.h
arm_compute/core/CL/kernels/CLElementWiseUnaryLayerKernel.h
arm_compute/core/CL/kernels/CLElementwiseOperationKernel.h
arm_compute/core/CL/kernels/CLErodeKernel.h
arm_compute/core/CL/kernels/CLFFTDigitReverseKernel.h
arm_compute/core/CL/kernels/CLFFTRadixStageKernel.h
arm_compute/core/CL/kernels/CLFFTScaleKernel.h
arm_compute/core/CL/kernels/CLFastCornersKernel.h
arm_compute/core/CL/kernels/CLFillBorderKernel.h
arm_compute/core/CL/kernels/CLFlattenLayerKernel.h
arm_compute/core/CL/kernels/CLFuseBatchNormalizationKernel.h
arm_compute/core/CL/kernels/CLGEMMLowpMatrixMultiplyKernel.h
arm_compute/core/CL/kernels/CLGEMMLowpMatrixMultiplyNativeKernel.h
arm_compute/core/CL/kernels/CLGEMMLowpMatrixMultiplyReshapedKernel.h
arm_compute/core/CL/kernels/CLGEMMLowpMatrixMultiplyReshapedOnlyRHSKernel.h
arm_compute/core/CL/kernels/CLGEMMLowpOffsetContributionKernel.h
arm_compute/core/CL/kernels/CLGEMMLowpOffsetContributionOutputStageKernel.h
arm_compute/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel.h
arm_compute/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleKernel.h
arm_compute/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.h
arm_compute/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.h
arm_compute/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h
arm_compute/core/CL/kernels/CLGEMMLowpReductionKernel.h
arm_compute/core/CL/kernels/CLGEMMMatrixAccumulateBiasesKernel.h
arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyKernel.h
arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyNativeKernel.h
arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyReshapedKernel.h
arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyReshapedOnlyRHSKernel.h
arm_compute/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.h
arm_compute/core/CL/kernels/CLGEMMReshapeLHSMatrixKernel.h
arm_compute/core/CL/kernels/CLGEMMReshapeRHSMatrixKernel.h
arm_compute/core/CL/kernels/CLGatherKernel.h
arm_compute/core/CL/kernels/CLGaussian3x3Kernel.h
arm_compute/core/CL/kernels/CLGaussian5x5Kernel.h
arm_compute/core/CL/kernels/CLGaussianPyramidKernel.h
arm_compute/core/CL/kernels/CLGenerateProposalsLayerKernel.h
arm_compute/core/CL/kernels/CLHOGDescriptorKernel.h
arm_compute/core/CL/kernels/CLHOGDetectorKernel.h
arm_compute/core/CL/kernels/CLHarrisCornersKernel.h
arm_compute/core/CL/kernels/CLHeightConcatenateLayerKernel.h
arm_compute/core/CL/kernels/CLHistogramKernel.h
arm_compute/core/CL/kernels/CLIm2ColKernel.h
arm_compute/core/CL/kernels/CLInstanceNormalizationLayerKernel.h
arm_compute/core/CL/kernels/CLIntegralImageKernel.h
arm_compute/core/CL/kernels/CLL2NormalizeLayerKernel.h
arm_compute/core/CL/kernels/CLLKTrackerKernel.h
arm_compute/core/CL/kernels/CLLocallyConnectedMatrixMultiplyKernel.h
arm_compute/core/CL/kernels/CLMagnitudePhaseKernel.h
arm_compute/core/CL/kernels/CLMeanStdDevKernel.h
arm_compute/core/CL/kernels/CLMeanStdDevNormalizationKernel.h
arm_compute/core/CL/kernels/CLMedian3x3Kernel.h
arm_compute/core/CL/kernels/CLMemsetKernel.h
arm_compute/core/CL/kernels/CLMinMaxLayerKernel.h
arm_compute/core/CL/kernels/CLMinMaxLocationKernel.h
arm_compute/core/CL/kernels/CLNonLinearFilterKernel.h
arm_compute/core/CL/kernels/CLNonMaximaSuppression3x3Kernel.h
arm_compute/core/CL/kernels/CLNormalizationLayerKernel.h
arm_compute/core/CL/kernels/CLNormalizePlanarYUVLayerKernel.h
arm_compute/core/CL/kernels/CLPadLayerKernel.h
arm_compute/core/CL/kernels/CLPermuteKernel.h
arm_compute/core/CL/kernels/CLPixelWiseMultiplicationKernel.h
arm_compute/core/CL/kernels/CLPoolingLayerKernel.h
arm_compute/core/CL/kernels/CLPriorBoxLayerKernel.h
arm_compute/core/CL/kernels/CLQuantizationLayerKernel.h
arm_compute/core/CL/kernels/CLROIAlignLayerKernel.h
arm_compute/core/CL/kernels/CLROIPoolingLayerKernel.h
arm_compute/core/CL/kernels/CLRangeKernel.h
arm_compute/core/CL/kernels/CLReductionOperationKernel.h
arm_compute/core/CL/kernels/CLRemapKernel.h
arm_compute/core/CL/kernels/CLReorgLayerKernel.h
arm_compute/core/CL/kernels/CLReshapeLayerKernel.h
arm_compute/core/CL/kernels/CLReverseKernel.h
arm_compute/core/CL/kernels/CLScaleKernel.h
arm_compute/core/CL/kernels/CLScharr3x3Kernel.h
arm_compute/core/CL/kernels/CLSelectKernel.h
arm_compute/core/CL/kernels/CLSobel3x3Kernel.h
arm_compute/core/CL/kernels/CLSobel5x5Kernel.h
arm_compute/core/CL/kernels/CLSobel7x7Kernel.h
arm_compute/core/CL/kernels/CLSoftmaxLayerKernel.h
arm_compute/core/CL/kernels/CLSpaceToBatchLayerKernel.h
arm_compute/core/CL/kernels/CLSpaceToDepthLayerKernel.h
arm_compute/core/CL/kernels/CLStackLayerKernel.h
arm_compute/core/CL/kernels/CLStridedSliceKernel.h
arm_compute/core/CL/kernels/CLTableLookupKernel.h
arm_compute/core/CL/kernels/CLThresholdKernel.h
arm_compute/core/CL/kernels/CLTileKernel.h
arm_compute/core/CL/kernels/CLTransposeKernel.h
arm_compute/core/CL/kernels/CLUpsampleLayerKernel.h
arm_compute/core/CL/kernels/CLWarpAffineKernel.h
arm_compute/core/CL/kernels/CLWarpPerspectiveKernel.h
arm_compute/core/CL/kernels/CLWeightsReshapeKernel.h
arm_compute/core/CL/kernels/CLWidthConcatenate2TensorsKernel.h
arm_compute/core/CL/kernels/CLWidthConcatenate4TensorsKernel.h
arm_compute/core/CL/kernels/CLWidthConcatenateLayerKernel.h
arm_compute/core/CL/kernels/CLWinogradFilterTransformKernel.h
arm_compute/core/CL/kernels/CLWinogradInputTransformKernel.h
arm_compute/core/CL/kernels/CLWinogradOutputTransformKernel.h
arm_compute/core/CL/kernels/CLYOLOLayerKernel.h
arm_compute/core/CL/kernels/ICLDepthwiseConvolutionLayer3x3Kernel.h
src/core/CL/kernels/CLAbsoluteDifferenceKernel.cpp
src/core/CL/kernels/CLAccumulateKernel.cpp
src/core/CL/kernels/CLActivationLayerKernel.cpp
src/core/CL/kernels/CLArgMinMaxLayerKernel.cpp
src/core/CL/kernels/CLBatchConcatenateLayerKernel.cpp
src/core/CL/kernels/CLBatchNormalizationLayerKernel.cpp
src/core/CL/kernels/CLBatchToSpaceLayerKernel.cpp
src/core/CL/kernels/CLBitwiseAndKernel.cpp
src/core/CL/kernels/CLBitwiseNotKernel.cpp
src/core/CL/kernels/CLBitwiseOrKernel.cpp
src/core/CL/kernels/CLBitwiseXorKernel.cpp
src/core/CL/kernels/CLBoundingBoxTransformKernel.cpp
src/core/CL/kernels/CLBox3x3Kernel.cpp
src/core/CL/kernels/CLCannyEdgeKernel.cpp
src/core/CL/kernels/CLChannelCombineKernel.cpp
src/core/CL/kernels/CLChannelExtractKernel.cpp
src/core/CL/kernels/CLChannelShuffleLayerKernel.cpp
src/core/CL/kernels/CLCol2ImKernel.cpp
src/core/CL/kernels/CLColorConvertKernel.cpp
src/core/CL/kernels/CLComparisonKernel.cpp
src/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.cpp
src/core/CL/kernels/CLConvolutionKernel.cpp
src/core/CL/kernels/CLCopyKernel.cpp
src/core/CL/kernels/CLCropKernel.cpp
src/core/CL/kernels/CLDeconvolutionLayerUpsampleKernel.cpp
src/core/CL/kernels/CLDeconvolutionReshapeOutputKernel.cpp
src/core/CL/kernels/CLDepthConcatenateLayerKernel.cpp
src/core/CL/kernels/CLDepthConvertLayerKernel.cpp
src/core/CL/kernels/CLDepthToSpaceLayerKernel.cpp
src/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NCHWKernel.cpp
src/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.cpp
src/core/CL/kernels/CLDepthwiseConvolutionLayerNativeKernel.cpp
src/core/CL/kernels/CLDepthwiseConvolutionLayerReshapeWeightsKernel.cpp
src/core/CL/kernels/CLDequantizationLayerKernel.cpp
src/core/CL/kernels/CLDerivativeKernel.cpp
src/core/CL/kernels/CLDilateKernel.cpp
src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp
src/core/CL/kernels/CLElementWiseUnaryLayerKernel.cpp
src/core/CL/kernels/CLElementwiseOperationKernel.cpp
src/core/CL/kernels/CLErodeKernel.cpp
src/core/CL/kernels/CLFFTDigitReverseKernel.cpp
src/core/CL/kernels/CLFFTRadixStageKernel.cpp
src/core/CL/kernels/CLFFTScaleKernel.cpp
src/core/CL/kernels/CLFastCornersKernel.cpp
src/core/CL/kernels/CLFillBorderKernel.cpp
src/core/CL/kernels/CLFlattenLayerKernel.cpp
src/core/CL/kernels/CLFuseBatchNormalizationKernel.cpp
src/core/CL/kernels/CLGEMMLowpMatrixMultiplyKernel.cpp
src/core/CL/kernels/CLGEMMLowpMatrixMultiplyNativeKernel.cpp
src/core/CL/kernels/CLGEMMLowpMatrixMultiplyReshapedKernel.cpp
src/core/CL/kernels/CLGEMMLowpMatrixMultiplyReshapedOnlyRHSKernel.cpp
src/core/CL/kernels/CLGEMMLowpOffsetContributionKernel.cpp
src/core/CL/kernels/CLGEMMLowpOffsetContributionOutputStageKernel.cpp
src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel.cpp
src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleKernel.cpp
src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.cpp
src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.cpp
src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.cpp
src/core/CL/kernels/CLGEMMLowpReductionKernel.cpp
src/core/CL/kernels/CLGEMMMatrixAccumulateBiasesKernel.cpp
src/core/CL/kernels/CLGEMMMatrixMultiplyKernel.cpp
src/core/CL/kernels/CLGEMMMatrixMultiplyNativeKernel.cpp
src/core/CL/kernels/CLGEMMMatrixMultiplyReshapedKernel.cpp
src/core/CL/kernels/CLGEMMMatrixMultiplyReshapedOnlyRHSKernel.cpp
src/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.cpp
src/core/CL/kernels/CLGEMMReshapeLHSMatrixKernel.cpp
src/core/CL/kernels/CLGEMMReshapeRHSMatrixKernel.cpp
src/core/CL/kernels/CLGatherKernel.cpp
src/core/CL/kernels/CLGaussian3x3Kernel.cpp
src/core/CL/kernels/CLGaussian5x5Kernel.cpp
src/core/CL/kernels/CLGaussianPyramidKernel.cpp
src/core/CL/kernels/CLGenerateProposalsLayerKernel.cpp
src/core/CL/kernels/CLHOGDescriptorKernel.cpp
src/core/CL/kernels/CLHOGDetectorKernel.cpp
src/core/CL/kernels/CLHarrisCornersKernel.cpp
src/core/CL/kernels/CLHeightConcatenateLayerKernel.cpp
src/core/CL/kernels/CLHistogramKernel.cpp
src/core/CL/kernels/CLIm2ColKernel.cpp
src/core/CL/kernels/CLInstanceNormalizationLayerKernel.cpp
src/core/CL/kernels/CLIntegralImageKernel.cpp
src/core/CL/kernels/CLL2NormalizeLayerKernel.cpp
src/core/CL/kernels/CLLKTrackerKernel.cpp
src/core/CL/kernels/CLLocallyConnectedMatrixMultiplyKernel.cpp
src/core/CL/kernels/CLMagnitudePhaseKernel.cpp
src/core/CL/kernels/CLMeanStdDevKernel.cpp
src/core/CL/kernels/CLMeanStdDevNormalizationKernel.cpp
src/core/CL/kernels/CLMedian3x3Kernel.cpp
src/core/CL/kernels/CLMemsetKernel.cpp
src/core/CL/kernels/CLMinMaxLayerKernel.cpp
src/core/CL/kernels/CLMinMaxLocationKernel.cpp
src/core/CL/kernels/CLNonLinearFilterKernel.cpp
src/core/CL/kernels/CLNonMaximaSuppression3x3Kernel.cpp
src/core/CL/kernels/CLNormalizationLayerKernel.cpp
src/core/CL/kernels/CLNormalizePlanarYUVLayerKernel.cpp
src/core/CL/kernels/CLPadLayerKernel.cpp
src/core/CL/kernels/CLPermuteKernel.cpp
src/core/CL/kernels/CLPixelWiseMultiplicationKernel.cpp
src/core/CL/kernels/CLPoolingLayerKernel.cpp
src/core/CL/kernels/CLPriorBoxLayerKernel.cpp
src/core/CL/kernels/CLQuantizationLayerKernel.cpp
src/core/CL/kernels/CLROIAlignLayerKernel.cpp
src/core/CL/kernels/CLROIPoolingLayerKernel.cpp
src/core/CL/kernels/CLRangeKernel.cpp
src/core/CL/kernels/CLReductionOperationKernel.cpp
src/core/CL/kernels/CLRemapKernel.cpp
src/core/CL/kernels/CLReorgLayerKernel.cpp
src/core/CL/kernels/CLReshapeLayerKernel.cpp
src/core/CL/kernels/CLReverseKernel.cpp
src/core/CL/kernels/CLScaleKernel.cpp
src/core/CL/kernels/CLScharr3x3Kernel.cpp
src/core/CL/kernels/CLSelectKernel.cpp
src/core/CL/kernels/CLSobel3x3Kernel.cpp
src/core/CL/kernels/CLSobel5x5Kernel.cpp
src/core/CL/kernels/CLSobel7x7Kernel.cpp
src/core/CL/kernels/CLSoftmaxLayerKernel.cpp
src/core/CL/kernels/CLSpaceToBatchLayerKernel.cpp
src/core/CL/kernels/CLSpaceToDepthLayerKernel.cpp
src/core/CL/kernels/CLStackLayerKernel.cpp
src/core/CL/kernels/CLStridedSliceKernel.cpp
src/core/CL/kernels/CLTableLookupKernel.cpp
src/core/CL/kernels/CLThresholdKernel.cpp
src/core/CL/kernels/CLTileKernel.cpp
src/core/CL/kernels/CLTransposeKernel.cpp
src/core/CL/kernels/CLUpsampleLayerKernel.cpp
src/core/CL/kernels/CLWarpAffineKernel.cpp
src/core/CL/kernels/CLWarpPerspectiveKernel.cpp
src/core/CL/kernels/CLWeightsReshapeKernel.cpp
src/core/CL/kernels/CLWidthConcatenate2TensorsKernel.cpp
src/core/CL/kernels/CLWidthConcatenate4TensorsKernel.cpp
src/core/CL/kernels/CLWidthConcatenateLayerKernel.cpp
src/core/CL/kernels/CLWinogradFilterTransformKernel.cpp
src/core/CL/kernels/CLWinogradInputTransformKernel.cpp
src/core/CL/kernels/CLWinogradOutputTransformKernel.cpp
src/core/CL/kernels/CLYOLOLayerKernel.cpp
src/runtime/CL/functions/CLActivationLayer.cpp