CudaExecItemT
Namespace: SymTensor.Compiler.Cuda
Parent Module: CudaExecUnitTypes
a CUDA operation that will be assigned to and executed in a CUDA stream
Union Cases
| Union Case | Description |
BlasGemm(...)
Signature: BlasTransposeOpT * BlasTransposeOpT * single * BlasTransposedMatrixTmpl * BlasTransposedMatrixTmpl * single * BlasTransposedMatrixTmpl
|
|
BlasGemmBatched(...)
Signature: BlasTransposeOpT * BlasTransposeOpT * single * BlasTransposedMatrixBatchTmpl * BlasTransposedMatrixBatchTmpl * single * BlasTransposedMatrixBatchTmpl
|
|
BlasGetrfBatched(...)
Signature: BlasTransposedMatrixBatchTmpl * BlasIntArrayTmpl * BlasIntArrayTmpl
|
|
BlasGetriBatched(...)
Signature: BlasTransposedMatrixBatchTmpl * BlasIntArrayTmpl * BlasTransposedMatrixBatchTmpl * BlasIntArrayTmpl
|
|
BlasInitPointerArray(...)
Signature: BlasTransposedMatrixBatchTmpl
|
|
CallCFunc(...)
Signature: TmplInstT * Type * ICudaArgTmpl list
|
|
CheckNonFiniteCounter(...)
Signature: string * ArrayNDManikinT
|
|
DumpValue(string,ArrayNDManikinT)
Signature: string * ArrayNDManikinT
|
|
ExecLoop(ExecLoopInfoT)
Signature: ExecLoopInfoT
|
|
ExtensionExecItem(ICudaExecItem)
Signature: ICudaExecItem
|
|
LaunchKernel(...)
Signature: TmplInstT * WorkDim * ICudaArgTmpl list
|
|
MemcpyDtoD(...)
Signature: IDevMemRngTmpl * IDevMemRngTmpl
|
|
MemcpyDtoH(...)
Signature: IDevMemRngTmpl * IHostMemRngTmpl
|
|
MemcpyHtoD(...)
Signature: IHostMemRngTmpl * IDevMemRngTmpl
|
|
MemsetSingle(single,IDevMemRngTmpl)
Signature: single * IDevMemRngTmpl
|
|
MemsetUInt32(uint32,IDevMemRngTmpl)
Signature: uint32 * IDevMemRngTmpl
|
|
PrintWithMsg(string,ArrayNDManikinT)
Signature: string * ArrayNDManikinT
|
|
Trace(UExprT,ArrayNDManikinT)
Signature: UExprT * ArrayNDManikinT
|
|
TraceEnteringLoop(UExprT)
Signature: UExprT
|
|
TraceLeavingLoop(UExprT)
Signature: UExprT
|
|
