Member List

• ck
• tensor_operation
• device
• DeviceElementwiseImpl

ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq > Member List

This is the complete list of members for ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >, including all inherited members.

ck::BaseOperator()=default	ck::tensor_operation::device::BaseOperator
ck::BaseOperator(const BaseOperator &)=default	ck::tensor_operation::device::BaseOperator
DeviceElementwise< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, UnaryOperation, Scale, NumDim >::BaseOperator()=default	ck::tensor_operation::device::BaseOperator
DeviceElementwise< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, UnaryOperation, Scale, NumDim >::BaseOperator(const BaseOperator &)=default	ck::tensor_operation::device::BaseOperator
Block2TileMap typedef	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >
GenerateBatchDimsLenghtsTuple(const std::array< index_t, NumDim > &lengths, const index_t M0_dim, const index_t M1_dim)	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
GenerateInDataTypePointerTuple()	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
GenerateInDataTypePointerTuple()	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
GenerateInOutGrid1dDescTuple(Number< TupleSize >)	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
GenerateInOutGridDescTuple()	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
GenerateOutDataTypePointerTuple()	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
GenerateOutDataTypePointerTuple()	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
GetLowestStrideDim(const std::array< index_t, NumDim > &strides)	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
GetObjectName() const	ck::tensor_operation::device::BaseOperator	inlinevirtual
GetTemplateInfo() const	ck::tensor_operation::device::BaseOperator	inlinevirtual
GetTypeIdHashCode() const	ck::tensor_operation::device::BaseOperator	inlinevirtual
GetTypeIdName() const	ck::tensor_operation::device::BaseOperator	inlinevirtual
GetTypeString() const override	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinevirtual
GetTypeString() const override	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinevirtual
GetWorkSpaceSize(const BaseArgument *) const	ck::tensor_operation::device::BaseOperator	inlinevirtual
GridwiseElementwise typedef	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >
GridwiseElementwiseOp typedef	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >
GridwiseElementwiseOpSameInOutVectorDim typedef	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >
I0	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	static
I1	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	static
InDataTypePointerTuple typedef	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >
InDataTypePointerTuple typedef	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >
InGrid1dDescTuple typedef	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >
InGridDescTuple typedef	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >
IsSupportedArgument(const Argument &arg)	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
IsSupportedArgument(const BaseArgument *p_arg) override	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinevirtual
IsSupportedArgument(const Argument &arg)	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
IsSupportedArgument(const BaseArgument *p_arg) override	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinevirtual
MakeArgument(const std::array< index_t, NumDim > lengths, const std::array< std::array< index_t, NumDim >, NumInput > inStridesArray, const std::array< std::array< index_t, NumDim >, NumOutput > outStridesArray, const std::array< const void *, NumInput > in_dev_buffers, const std::array< void *, NumOutput > out_dev_buffers, ElementwiseOperation elementwise_op)	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
MakeArgument(const std::array< index_t, NumDim > lengths, const std::array< std::array< index_t, NumDim >, NumInput > inStridesArray, const std::array< std::array< index_t, NumDim >, NumOutput > outStridesArray, const std::array< const void *, NumInput > in_dev_buffers, const std::array< void *, NumOutput > out_dev_buffers, ElementwiseOperation elementwise_op, UnaryOperation unary_op, Scale scale_op)	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
MakeArgumentPointer(const std::array< index_t, NumDim > lengths, const std::array< std::array< index_t, NumDim >, NumInput > inStridesArray, const std::array< std::array< index_t, NumDim >, NumOutput > outStridesArray, const std::array< const void *, NumInput > in_dev_buffers, const std::array< void *, NumOutput > out_dev_buffers, ElementwiseOperation elementwise_op) override	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinevirtual
MakeArgumentPointer(const std::array< index_t, NumDim > lengths, const std::array< std::array< index_t, NumDim >, NumInput > inStridesArray, const std::array< std::array< index_t, NumDim >, NumOutput > outStridesArray, const std::array< const void *, NumInput > in_dev_buffers, const std::array< void *, NumOutput > out_dev_buffers, ElementwiseOperation elementwise_op, UnaryOperation unary_op, Scale scale_op) override	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinevirtual
MakeDescriptor(const std::array< index_t, NumDim > &lengths, const std::array< index_t, NumDim > &in_strides, const std::array< index_t, NumDim > &out_strides, const std::array< index_t, NumDim > &desc_strides)	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
MakeDescriptor_M(const std::array< index_t, NumDim > &lengths, const std::array< index_t, NumDim > &stride, index_t gridSize, index_t blockSize)	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
MakeInvoker()	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
MakeInvoker()	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
MakeInvokerPointer() override	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinevirtual
MakeInvokerPointer() override	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinevirtual
NumInput	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	static
NumOutput	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	static
ck::operator=(const BaseOperator &)=default	ck::tensor_operation::device::BaseOperator
DeviceElementwise< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, UnaryOperation, Scale, NumDim >::operator=(const BaseOperator &)=default	ck::tensor_operation::device::BaseOperator
OutDataTypePointerTuple typedef	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >
OutDataTypePointerTuple typedef	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >
OutGrid1dDescTuple typedef	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >
OutGridDescTuple typedef	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >
PadDescriptor_M_1d(Desc_M desc_m, index_t gridSize, index_t blockSize)	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
PadInputOutputDescriptor(const InOutDescriptor &desc)	ck::tensor_operation::device::DeviceElementwiseImpl< InDataTypeTuple, OutDataTypeTuple, ElementwiseOperation, NumDim, BlockSize, M0PerBlock, M1PerBlock, M0PerThread, M1PerThread, ThreadClusterArrangeOrder, InScalarPerVectorSeq, OutScalarPerVectorSeq >	inlinestatic
SetWorkSpacePointer(BaseArgument *p_arg, void *p_workspace, const StreamConfig &=StreamConfig{}) const	ck::tensor_operation::device::BaseOperator	inlinevirtual
~BaseOperator()	ck::tensor_operation::device::BaseOperator	inlinevirtual