Class Members

Here is a list of all class members with links to the classes they belong to:

- x -

• X : ck_tile::tile_distribution_encoding_pattern_aq_transposed_c< BlockGemmShape, WarpGemm, BlockSize, YPerTile, XPerTile, VecSize >
• x : ck_tile::naive_attention_fwd_kernel< QType, KType, VType, OType, AccType, KVScaleType, QLayout, KLayout, VLayout, OLayout, KScaleLayout, VScaleLayout, Traits >::page_addresser< T, Layout >, internal::Schema< SchemaDocumentType >
• X0 : ck_tile::tile_distribution_encoding_pattern_2d< BlockSize, YPerTile, XPerTile, VecSize, tile_distribution_pattern::block_raked, NumWaveGroups >, ck_tile::tile_distribution_encoding_pattern_2d< BlockSize, YPerTile, XPerTile, VecSize, tile_distribution_pattern::thread_raked, NumWaveGroups >, ck_tile::tile_distribution_encoding_pattern_2d< BlockSize, YPerTile, XPerTile, VecSize, tile_distribution_pattern::warp_raked, NumWaveGroups >
• X1 : ck_tile::tile_distribution_encoding_pattern_2d< BlockSize, YPerTile, XPerTile, VecSize, tile_distribution_pattern::block_raked, NumWaveGroups >, ck_tile::tile_distribution_encoding_pattern_2d< BlockSize, YPerTile, XPerTile, VecSize, tile_distribution_pattern::thread_raked, NumWaveGroups >, ck_tile::tile_distribution_encoding_pattern_2d< BlockSize, YPerTile, XPerTile, VecSize, tile_distribution_pattern::warp_raked, NumWaveGroups >
• X_ : ck::tensor_operation::device::DeviceColumnToImageImpl< NDimSpatial, ImageLayout, InputDataType, OutputDataType, BlockSize, MPerBlock, KPerBlock, ThreadClusterLengths, ScalarPerVector,, bool, type >::Argument, ck::tensor_operation::device::DeviceImageToColumnImpl< NDimSpatial, ImageLayout, InputDataType, OutputDataType, BlockSize, MPerBlock, KPerBlock, ThreadClusterLengths, ScalarPerVector,, bool, type >::Argument, ck::tensor_operation::TransformConvBwdDataToGemm_v1< NDimSpatial, ConvBwdDataSpecialization, AK1, BK1, GemmMPerBlock, GemmNPerBlock, GemmKPerBlock, DoPadGemmM, DoPadGemmN, ALayout, BLayout, CLayout, SplitN, ADataType, CDataType, NumGroupsToMerge, IndexType, CTranspose >, ck::tensor_operation::TransformConvFwdToGemm< NDimSpatial, ConvForwardSpecialization, SplitN, ADataType, CDataType, NumGroupsToMerge, IndexType, CTranspose >, ck_tile::TransformConvBwdDataToGemm< NDimSpatial, ConvolutionSpecialization, VectorSizeA, VectorSizeB, VectorSizeC, SplitN, ADataType, CDataType, NumGroupsToMerge, IndexType >, ck_tile::TransformConvBwdWeightToGemm< NDimSpatial, ConvolutionSpecialization, VectorSizeA, VectorSizeB, VectorSizeC, NumGroupsToMerge, SplitN, ADataType, CDataType, IndexType >
• x_elementwise_op_ : ck::tensor_operation::device::DeviceElementwiseNormalizationImpl< InDataTypeTuple, GammaDataType, BetaDataType, AccDataType, YDataType, XElementwiseOperation, YElementwiseOperation, Rank, NumReduceDim, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XYSrcVectorDim, XSrcVectorSize, GammaSrcVectorDim, GammaSrcVectorSize, BetaSrcVectorDim, BetaSrcVectorSize, YDstVectorSize >::Argument
• x_grid_desc_m_k : ck::tensor_operation::device::DeviceBatchNormBwdImpl< XDataType, DxDataType, DyDataType, AccDataType, ScaleDataType, DscaleDbiasDataType, MeanVarDataType, DyElementwiseOp, Rank, NumBatchNormReduceDim, UseMultiblockInK, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XDyDxVectorDim, XSrcVectorSize, DySrcVectorSize, DxDstVectorSize, ScaleSrcVectorSize, DscaleDbiasDstVectorSize, MeanVarSrcVectorSize >::Argument
• x_grid_desc_m_k_ : ck::tensor_operation::device::DeviceBatchNormFwdImpl< XDataType, YDataType, AccDataType, ScaleDataType, BiasDataType, MeanVarDataType, YElementwiseOp, Rank, NumBatchNormReduceDim, UseMultiblockInK, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XSrcYDstVectorDim, XSrcVectorSize, YDstVectorSize, ScaleSrcVectorSize, BiasSrcVectorSize, MeanVarSrcDstVectorSize >::Argument, ck::tensor_operation::device::DeviceElementwiseNormalizationImpl< InDataTypeTuple, GammaDataType, BetaDataType, AccDataType, YDataType, XElementwiseOperation, YElementwiseOperation, Rank, NumReduceDim, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XYSrcVectorDim, XSrcVectorSize, GammaSrcVectorDim, GammaSrcVectorSize, BetaSrcVectorDim, BetaSrcVectorSize, YDstVectorSize >::Argument, ck::tensor_operation::device::DeviceNormalizationBwdDataImpl< DYDataType, XDataType, GammaDataType, MeanInvStdDataType, ComputeDataType, DXDataType, Rank, NumReduceDim, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, IsDYFastestDimReduced, DYSrcVectorSize, IsXFastestDimReduced, XSrcVectorSize, IsGammaFastestDimReduced, GammaSrcVectorSize, IsMeanInvStdFastestDimReduced, MeanInvStdSrcVectorSize, IsDxFastestDimReduced, DXDstVectorSize >::Argument, ck::tensor_operation::device::DeviceNormalizationBwdGammaBetaImpl< DYDataType, XDataType, MeanInvStdDataType, ComputeDataType, DGammaDataType, DBetaDataType, Rank, NumReduceDim, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, IsDYFastestDimReduced, DYSrcVectorSize, IsXFastestDimReduced, XSrcVectorSize, IsMeanInvStdFastestDimReduced, MeanInvStdSrcVectorSize, DGammaDstVectorSize, DBetaDstVectorSize >::Argument, ck::tensor_operation::device::DeviceNormalizationFwdImpl< XDataType, GammaDataType, BetaDataType, ComputeDataType, YDataType, SaveMeanInvStdDataType, YElementwiseOperation, Rank, NumReduceDim, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XYSrcVectorDim, XSrcVectorSize, GammaSrcVectorDim, GammaSrcVectorSize, BetaSrcVectorDim, BetaSrcVectorSize, YDstVectorSize, SaveMeanInvStdDstVectorSize, UseWelford >::Argument, ck::tensor_operation::device::DeviceNormalizationFwdSplitKImpl< XDataType, GammaDataType, BetaDataType, ComputeDataType, YDataType, SaveMeanInvStdDataType, YElementwiseOperation, Rank, NumReduceDim, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XYVectorDim, XSrcVectorSize, GammaSrcVectorDim, GammaSrcVectorSize, BetaSrcVectorDim, BetaSrcVectorSize, YDstVectorSize, SaveMeanInvStdDstVectorSize >::Argument
• x_lds_size_ : ck::tensor_operation::device::DeviceElementwiseNormalizationImpl< InDataTypeTuple, GammaDataType, BetaDataType, AccDataType, YDataType, XElementwiseOperation, YElementwiseOperation, Rank, NumReduceDim, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XYSrcVectorDim, XSrcVectorSize, GammaSrcVectorDim, GammaSrcVectorSize, BetaSrcVectorDim, BetaSrcVectorSize, YDstVectorSize >::Argument
• x_stride : ck_tile::Layernorm2dFwd< Pipeline_, Epilogue_ >::Kargs, ck_tile::Layernorm2dFwdHostArgs, ck_tile::MoeSmoothquant< Pipeline_ >::Kargs, ck_tile::MoeSmoothquantHostArgs, ck_tile::Rmsnorm2dFwd< Pipeline_, Epilogue_ >::Kargs, ck_tile::Rmsnorm2dFwdHostArgs, ck_tile::Smoothquant< Pipeline_ >::Kargs, ck_tile::SmoothquantHostArgs
• x_thread_desc_m_k : ck::ThreadwiseWelford< T, XThreadDesc_M_K, MeanVarThreadDesc_M >
• XBiasDataType : ck_tile::Layernorm2dFwd< Pipeline_, Epilogue_ >, ck_tile::Layernorm2dFwdPipelineOnePass< Problem_, Policy_ >, ck_tile::Layernorm2dFwdPipelineProblem< XDataType_, XBiasDataType_, GammaDataType_, BetaDataType_, ComputeDataType_, YDataType_, MeanDataType_, InvStdDataType_, SmoothScaleDataType_, YScaleDataType_, BlockShape_, Traits_ >, ck_tile::Layernorm2dFwdPipelineTwoPass< Problem_, Policy_ >
• XDataType : ck::tensor_operation::device::DeviceElementwiseNormalizationImpl< InDataTypeTuple, GammaDataType, BetaDataType, AccDataType, YDataType, XElementwiseOperation, YElementwiseOperation, Rank, NumReduceDim, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XYSrcVectorDim, XSrcVectorSize, GammaSrcVectorDim, GammaSrcVectorSize, BetaSrcVectorDim, BetaSrcVectorSize, YDstVectorSize >, ck_tile::AddRmsnorm2dRdquantFwd< Pipeline_ >, ck_tile::AddRmsnorm2dRdquantFwdPipelineOnePass< Problem_, Policy_ >, ck_tile::AddRmsnorm2dRdquantFwdPipelineProblem< ADataType_, BDataType_, GammaDataType_, ComputeDataType_, XDataType_, YScaleDataType_, QYDataType_, BlockShape_, kPadN_, kSaveX_, kThreePass_ >, ck_tile::AddRmsnorm2dRdquantFwdPipelineThreePass< Problem_, Policy_ >, ck_tile::BlockNormReduce< Problem_, Policy_ >, ck_tile::BlockNormReduceProblem< XDataType_, ComputeDataType_, BlockShape_, kFastFDiv_, kWelford_ >, ck_tile::BlockReduce2d< Problem_, Policy_ >, ck_tile::BlockReduce2dProblem< XDataType_, ComputeDataType_, BlockShape_, OutputIndex_ >, ck_tile::ElementWiseKernel< Problem_, Policy_ >, ck_tile::ElementWisePipelineProblem< XDataType_, ComputeDataType_, YDataType_, BlockShape_, ElementWiseOperation_, kPad_ >, ck_tile::Layernorm2dFwd< Pipeline_, Epilogue_ >, ck_tile::Layernorm2dFwdPipelineOnePass< Problem_, Policy_ >, ck_tile::Layernorm2dFwdPipelineProblem< XDataType_, XBiasDataType_, GammaDataType_, BetaDataType_, ComputeDataType_, YDataType_, MeanDataType_, InvStdDataType_, SmoothScaleDataType_, YScaleDataType_, BlockShape_, Traits_ >, ck_tile::Layernorm2dFwdPipelineTwoPass< Problem_, Policy_ >, ck_tile::MoeSmoothquant< Pipeline_ >, ck_tile::Reduce2dProblem< XDataType_, ComputeDataType_, YDataType_, BlockShape_, ReduceOp_, OutputIndex_ >, ck_tile::Reduce< Problem_, Policy_ >, ck_tile::Rmsnorm2dFwd< Pipeline_, Epilogue_ >, ck_tile::Rmsnorm2dFwdPipelineModelSensitiveT5Pass< Problem_, Policy_ >, ck_tile::Rmsnorm2dFwdPipelineOnePass< Problem_, Policy_ >, ck_tile::Rmsnorm2dFwdPipelineProblem< XDataType_, GammaDataType_, ComputeDataType_, YDataType_, InvRmsDataType_, UnquantYDataType_, SmoothScaleDataType_, YScaleDataType_, BlockShape_, Traits_ >, ck_tile::Rmsnorm2dFwdPipelineTwoPass< Problem_, Policy_ >, ck_tile::Smoothquant< Pipeline_ >, ck_tile::SmoothquantPipelineOnePass< Problem_, Policy_ >, ck_tile::SmoothquantPipelineProblem< XDataType_, SmoothScaleDataType_, ComputeDataType_, YScaleDataType_, QYDataType_, BlockShape_, kPadN_, kTwoPass_ >, ck_tile::SmoothquantPipelineTwoPass< Problem_, Policy_ >
• XDL_PerScaleK : ck_tile::F16xMXF4FlatmmPipelineAGmemBGmemCRegV1< Problem, PipelinePolicy >
• XDL_PerScaleN : ck_tile::F16xMXF4FlatmmPipelineAGmemBGmemCRegV1< Problem, PipelinePolicy >
• XDL_PerWeightK : ck_tile::F16xMXF4FlatmmPipelineAGmemBGmemCRegV1< Problem, PipelinePolicy >
• XDLK_PerThread : ck_tile::F16xMXF4FlatmmPipelineAGmemBGmemCRegV1< Problem, PipelinePolicy >
• xdlops_gemm : ck::BlockwiseGemmXdlops_k0mk1_k0nk1_m0n0m1n1m2m3m4n2_v1< BlockSize, FloatA, FloatB, FloatAcc, AK0MK1BlockDesc, BK0NK1BlockDesc, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, ComputeTypeA, ComputeTypeB >, ck::BlockwiseGemmXdlops_k0mk1_k0nk1_m0n0m1n1m2m3m4n2_v1r1< BlockSize, FloatAB, FloatAcc, AK0MK1BlockDesc, BK0K0BN0N1N2N3K1BlockDesc, MPerBlock, NPerBlock, K0PerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_mx_pipeline_base< BlockSize, ADataType, BDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC >, ck::BlockwiseGemmXdlops_pipeline_base< BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC >, ck::BlockwiseGemmXdlops_pipeline_blockscale_bpreshuffle_v1< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MScaleBlock, NScaleBlock, KScaleBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_blockscale_bpreshuffle_v3< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MScaleBlock, NScaleBlock, KScaleBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_bpreshuffle_bdequant_v1< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_bpreshuffle_bdequant_v3< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_bpreshuffle_gufusion_bdequant_v1< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_bpreshuffle_gufusion_v1< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_bpreshuffle_gufusion_v3< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_bpreshuffle_mx_moe_gufusion_v3< BlockGemmPipelineScheduler::Intrawave, ThreadBlockSize, ScaleBlockSize, ADataType, AScaleDataType, BDataType, BScaleDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_bpreshuffle_mx_moe_v1< BlockGemmPipelineScheduler::Intrawave, ThreadBlockSize, ScaleBlockSize, ADataType, AScaleDataType, BDataType, BScaleDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_bpreshuffle_mx_moe_v3< BlockGemmPipelineScheduler::Intrawave, ThreadBlockSize, ScaleBlockSize, ADataType, AScaleDataType, BDataType, BScaleDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_bpreshuffle_v1< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_bpreshuffle_v2< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_bpreshuffle_v3< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_moe_blockscale_bpreshuffle_gufusion_v1< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MScaleBlock, NScaleBlock, KScaleBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_moe_blockscale_bpreshuffle_gufusion_v3< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MScaleBlock, NScaleBlock, KScaleBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_moe_blockscale_bpreshuffle_v1< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MScaleBlock, NScaleBlock, KScaleBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_moe_blockscale_bpreshuffle_v3< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MScaleBlock, NScaleBlock, KScaleBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_mx_moe_bns_gufusion_v3< BlockGemmPipelineScheduler::Intrawave, ThreadBlockSize, ScaleBlockSize, ADataType, AScaleDataType, BDataType, BScaleDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_mx_moe_nbs_v1< BlockGemmPipelineScheduler::Intrawave, ThreadBlockSize, ScaleBlockSize, ADataType, AScaleDataType, BDataType, BScaleDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_mx_moe_nbs_v3< BlockGemmPipelineScheduler::Intrawave, ThreadBlockSize, ScaleBlockSize, ADataType, AScaleDataType, BDataType, BScaleDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v1< BlockGemmPipelineScheduler::Interwave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v1< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v1_ab_scale< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v1_b_scale< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v1_mx< BlockGemmPipelineScheduler::Intrawave, ThreadBlockSize, ScaleBlockSize, ADataType, AScaleDataType, BDataType, BScaleDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v2< BlockGemmPipelineScheduler::Interwave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v2< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v2_ab_scale< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v2_b_scale< BlockGemmPipelineScheduler::Interwave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v2_b_scale< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v3< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v3_ab_scale< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v3_b_scale< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v3_mx< BlockGemmPipelineScheduler::Intrawave, ThreadBlockSize, ScaleBlockSize, ADataType, AScaleDataType, BDataType, BScaleDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v3_mx_bprehuffle< BlockGemmPipelineScheduler::Intrawave, ThreadBlockSize, ScaleBlockSize, ADataType, AScaleDataType, BDataType, BScaleDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v4< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride >, ck::BlockwiseGemmXdlops_pipeline_v4< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v4_b_scale< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_pipeline_v5< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride >, ck::BlockwiseGemmXdlopsDirectLoad_pipeline_v1< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::BlockwiseGemmXdlopsDirectLoad_pipeline_v4< BlockGemmPipelineScheduler::Intrawave, BlockSize, ADataType, BDataType, ComputeDataType, AccDataType, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, ABlockTransferSrcScalarPerVector, BBlockTransferSrcScalarPerVector, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack >, ck::GridwiseGemm_k0mk1_k0nk1_mn_xdlops_skip_b_lds_v1< BlockSize, FloatAB, FloatAcc, FloatC, CGlobalMemoryDataOperation, AGridDesc_K0_M_K1, BGridDesc_K0_N_K1, CGridDesc_M_N, AElementwiseOperation, BElementwiseOperation, CElementwiseOperation, MPerBlock, NPerBlock, K0PerBlock, MPerXdl, NPerXdl, K1Value, MXdlPerWave, NXdlPerWave, ABlockTransferThreadClusterLengths_K0_M_K1, ABlockTransferThreadClusterArrangeOrder, ABlockTransferSrcAccessOrder, ABlockTransferSrcVectorDim, ABlockTransferSrcScalarPerVector, ABlockTransferDstScalarPerVector_K1, AThreadTransferSrcResetCoordinateAfterRun, ABlockLdsExtraM, BBlockTransferSrcScalarPerVector, BThreadTransferSrcResetCoordinateAfterRun, BBlockBufferSize, CThreadTransferSrcDstAccessOrder, CThreadTransferSrcDstVectorDim, CThreadTransferDstScalarPerVector >
• XdlopsGemm() : ck::XdlopsGemm< base_type, MPerXdlops, NPerXdlops, KPack, additional_type, TransposeC, is_scale_mfma >
• XDot_ : ck::ConvBwdDataImplicitGemmOutTransform, ck::tensor_operation::TransformConvBwdDataToGemm_v1< NDimSpatial, ConvBwdDataSpecialization, AK1, BK1, GemmMPerBlock, GemmNPerBlock, GemmKPerBlock, DoPadGemmM, DoPadGemmN, ALayout, BLayout, CLayout, SplitN, ADataType, CDataType, NumGroupsToMerge, IndexType, CTranspose >, ck_tile::TransformConvBwdDataToGemm< NDimSpatial, ConvolutionSpecialization, VectorSizeA, VectorSizeB, VectorSizeC, SplitN, ADataType, CDataType, NumGroupsToMerge, IndexType >
• XDotSlice_K_ : ck::ConvBwdDataImplicitGemmOutTransform
• XIdx : ck::tensor_operation::device::DeviceColumnToImageImpl< NDimSpatial, ImageLayout, InputDataType, OutputDataType, BlockSize, MPerBlock, KPerBlock, ThreadClusterLengths, ScalarPerVector,, bool, type >
• Xor() : ck::Xor< LowLengths, ApplyModulo >
• xor_t() : ck_tile::xor_t< LowLengths >
• XR : ck_tile::tile_distribution_encoding_pattern_aq_transposed_c< BlockGemmShape, WarpGemm, BlockSize, YPerTile, XPerTile, VecSize >
• xr_stride : ck_tile::Layernorm2dFwd< Pipeline_, Epilogue_ >::Kargs, ck_tile::Layernorm2dFwdHostArgs, ck_tile::Rmsnorm2dFwd< Pipeline_, Epilogue_ >::Kargs, ck_tile::Rmsnorm2dFwdHostArgs
• XResidualDataType : ck_tile::Layernorm2dFwd< Pipeline_, Epilogue_ >, ck_tile::Layernorm2dFwdPipelineOnePass< Problem_, Policy_ >, ck_tile::Layernorm2dFwdPipelineTwoPass< Problem_, Policy_ >, ck_tile::Rmsnorm2dFwd< Pipeline_, Epilogue_ >, ck_tile::Rmsnorm2dFwdPipelineModelSensitiveT5Pass< Problem_, Policy_ >, ck_tile::Rmsnorm2dFwdPipelineOnePass< Problem_, Policy_ >, ck_tile::Rmsnorm2dFwdPipelineTwoPass< Problem_, Policy_ >
• XSrcVectorDim : ck::tensor_operation::device::DeviceNormalizationBwdDataImpl< DYDataType, XDataType, GammaDataType, MeanInvStdDataType, ComputeDataType, DXDataType, Rank, NumReduceDim, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, IsDYFastestDimReduced, DYSrcVectorSize, IsXFastestDimReduced, XSrcVectorSize, IsGammaFastestDimReduced, GammaSrcVectorSize, IsMeanInvStdFastestDimReduced, MeanInvStdSrcVectorSize, IsDxFastestDimReduced, DXDstVectorSize >, ck::tensor_operation::device::DeviceNormalizationBwdGammaBetaImpl< DYDataType, XDataType, MeanInvStdDataType, ComputeDataType, DGammaDataType, DBetaDataType, Rank, NumReduceDim, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, IsDYFastestDimReduced, DYSrcVectorSize, IsXFastestDimReduced, XSrcVectorSize, IsMeanInvStdFastestDimReduced, MeanInvStdSrcVectorSize, DGammaDstVectorSize, DBetaDstVectorSize >
• XStride_ : ck::tensor_operation::TransformConvFwdToGemm< NDimSpatial, ConvForwardSpecialization, SplitN, ADataType, CDataType, NumGroupsToMerge, IndexType, CTranspose >
• xStrides_ : ck::tensor_operation::device::DeviceBatchNormBwdImpl< XDataType, DxDataType, DyDataType, AccDataType, ScaleDataType, DscaleDbiasDataType, MeanVarDataType, DyElementwiseOp, Rank, NumBatchNormReduceDim, UseMultiblockInK, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XDyDxVectorDim, XSrcVectorSize, DySrcVectorSize, DxDstVectorSize, ScaleSrcVectorSize, DscaleDbiasDstVectorSize, MeanVarSrcVectorSize >::Argument, ck::tensor_operation::device::DeviceBatchNormFwdImpl< XDataType, YDataType, AccDataType, ScaleDataType, BiasDataType, MeanVarDataType, YElementwiseOp, Rank, NumBatchNormReduceDim, UseMultiblockInK, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XSrcYDstVectorDim, XSrcVectorSize, YDstVectorSize, ScaleSrcVectorSize, BiasSrcVectorSize, MeanVarSrcDstVectorSize >::Argument, ck::tensor_operation::device::DeviceElementwiseNormalizationImpl< InDataTypeTuple, GammaDataType, BetaDataType, AccDataType, YDataType, XElementwiseOperation, YElementwiseOperation, Rank, NumReduceDim, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XYSrcVectorDim, XSrcVectorSize, GammaSrcVectorDim, GammaSrcVectorSize, BetaSrcVectorDim, BetaSrcVectorSize, YDstVectorSize >::Argument, ck::tensor_operation::device::DeviceNormalizationBwdDataImpl< DYDataType, XDataType, GammaDataType, MeanInvStdDataType, ComputeDataType, DXDataType, Rank, NumReduceDim, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, IsDYFastestDimReduced, DYSrcVectorSize, IsXFastestDimReduced, XSrcVectorSize, IsGammaFastestDimReduced, GammaSrcVectorSize, IsMeanInvStdFastestDimReduced, MeanInvStdSrcVectorSize, IsDxFastestDimReduced, DXDstVectorSize >::Argument, ck::tensor_operation::device::DeviceNormalizationBwdGammaBetaImpl< DYDataType, XDataType, MeanInvStdDataType, ComputeDataType, DGammaDataType, DBetaDataType, Rank, NumReduceDim, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, IsDYFastestDimReduced, DYSrcVectorSize, IsXFastestDimReduced, XSrcVectorSize, IsMeanInvStdFastestDimReduced, MeanInvStdSrcVectorSize, DGammaDstVectorSize, DBetaDstVectorSize >::Argument, ck::tensor_operation::device::DeviceNormalizationFwdImpl< XDataType, GammaDataType, BetaDataType, ComputeDataType, YDataType, SaveMeanInvStdDataType, YElementwiseOperation, Rank, NumReduceDim, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XYSrcVectorDim, XSrcVectorSize, GammaSrcVectorDim, GammaSrcVectorSize, BetaSrcVectorDim, BetaSrcVectorSize, YDstVectorSize, SaveMeanInvStdDstVectorSize, UseWelford >::Argument, ck::tensor_operation::device::DeviceNormalizationFwdSplitKImpl< XDataType, GammaDataType, BetaDataType, ComputeDataType, YDataType, SaveMeanInvStdDataType, YElementwiseOperation, Rank, NumReduceDim, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XYVectorDim, XSrcVectorSize, GammaSrcVectorDim, GammaSrcVectorSize, BetaSrcVectorDim, BetaSrcVectorSize, YDstVectorSize, SaveMeanInvStdDstVectorSize >::Argument
• XThreadBufferDimAccessOrder : ck::GridwiseNormalizationBwdData_mk_to_mk< DYDataType, XDataType, GammaDataType, MeanInvStdDataType, ComputeDataType, DXDataType, GridDesc_M_K, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, DYSrcVectorDim, DYSrcVectorSize, XSrcVectorDim, XSrcVectorSize, GammaSrcVectorDim, GammaSrcVectorSize, MeanInvStdSrcVectorDim, MeanInvStdSrcVectorSize, DXDstVectorDim, DXDstVectorSize, SweepOnce >, ck::GridwiseNormalizationBwdGammaBeta_mk_to_k< DYDataType, XDataType, MeanInvStdDataType, ComputeDataType, DGammaDataType, DBetaDataType, GridDesc_M_K, GridDesc_M, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, DYSrcVectorDim, DYSrcVectorSize, XSrcVectorDim, XSrcVectorSize, MeanInvStdSrcVectorDim, MeanInvStdSrcVectorSize, DGammaDstVectorSize, DBetaDstVectorSize >
• XThreadBufferNumber : ck::GridwiseElementwiseLayernormWelfordVariance_mk_to_mk< InDataTypePointerTuple, XDataType, GammaDataType, BetaDataType, YDataType, AccDataType, XElementwiseOperation, YElementwiseOperation, InGrid2dDescTuple, GridDesc_M_K, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XSrcVectorDim, XSrcVectorSize, GammaSrcVectorDim, GammaSrcVectorSize, BetaSrcVectorDim, BetaSrcVectorSize, YDstVectorDim, YDstVectorSize, SweepOnce >
• XTilde_ : ck::tensor_operation::TransformConvBwdDataToGemm_v1< NDimSpatial, ConvBwdDataSpecialization, AK1, BK1, GemmMPerBlock, GemmNPerBlock, GemmKPerBlock, DoPadGemmM, DoPadGemmN, ALayout, BLayout, CLayout, SplitN, ADataType, CDataType, NumGroupsToMerge, IndexType, CTranspose >, ck_tile::TransformConvBwdDataToGemm< NDimSpatial, ConvolutionSpecialization, VectorSizeA, VectorSizeB, VectorSizeC, SplitN, ADataType, CDataType, NumGroupsToMerge, IndexType >
• XYGridDesc_M_K : ck::tensor_operation::device::DeviceBatchNormBwdImpl< XDataType, DxDataType, DyDataType, AccDataType, ScaleDataType, DscaleDbiasDataType, MeanVarDataType, DyElementwiseOp, Rank, NumBatchNormReduceDim, UseMultiblockInK, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XDyDxVectorDim, XSrcVectorSize, DySrcVectorSize, DxDstVectorSize, ScaleSrcVectorSize, DscaleDbiasDstVectorSize, MeanVarSrcVectorSize >, ck::tensor_operation::device::DeviceBatchNormFwdImpl< XDataType, YDataType, AccDataType, ScaleDataType, BiasDataType, MeanVarDataType, YElementwiseOp, Rank, NumBatchNormReduceDim, UseMultiblockInK, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XSrcYDstVectorDim, XSrcVectorSize, YDstVectorSize, ScaleSrcVectorSize, BiasSrcVectorSize, MeanVarSrcDstVectorSize >
• xyLengths_ : ck::tensor_operation::device::DeviceBatchNormBwdImpl< XDataType, DxDataType, DyDataType, AccDataType, ScaleDataType, DscaleDbiasDataType, MeanVarDataType, DyElementwiseOp, Rank, NumBatchNormReduceDim, UseMultiblockInK, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XDyDxVectorDim, XSrcVectorSize, DySrcVectorSize, DxDstVectorSize, ScaleSrcVectorSize, DscaleDbiasDstVectorSize, MeanVarSrcVectorSize >::Argument, ck::tensor_operation::device::DeviceBatchNormFwdImpl< XDataType, YDataType, AccDataType, ScaleDataType, BiasDataType, MeanVarDataType, YElementwiseOp, Rank, NumBatchNormReduceDim, UseMultiblockInK, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, XSrcYDstVectorDim, XSrcVectorSize, YDstVectorSize, ScaleSrcVectorSize, BiasSrcVectorSize, MeanVarSrcDstVectorSize >::Argument