Share via


IDkmGPUComputeKernelOperation Interface

Definition

Provides the compute kernel hierarchy, i.e., the thread group, compute vector and compute thread for view by the user.

Implementations of this interface may restrict when they are called using a filter defined in their component configuration. The following properties may be used: BaseDebugMonitorId, EngineId, TransportKind.

public interface class IDkmGPUComputeKernelOperation
public interface class IDkmGPUComputeKernelOperation
__interface IDkmGPUComputeKernelOperation
public interface IDkmGPUComputeKernelOperation
type IDkmGPUComputeKernelOperation = interface
Public Interface IDkmGPUComputeKernelOperation

Methods

GetActiveThreadGroups(DkmGPUComputeKernel, Int64[], Int32)

Obtain the active thread groups from the compute kernel.

GetComputeKernelName(DkmGPUComputeKernel)

Get the name of compute kernel.

GetComputeKernelProperties(DkmGPUComputeKernel, DkmComputeProperty[], Int32)

Get properties of the compute kernel.

GetComputeVectorWidth(DkmGPUComputeKernel, Int32)

Obtain the warp size of the hardware or emulator.

GetCurrentGroupDimensions(DkmGPUComputeKernel, Int32[], Int32)

Get the dimension of the thread block.

GetCurrentThreadDimensions(DkmGPUComputeKernel, Int32[], Int32)

Get the dimension of the thread block.

GetFlatComputeKernelDimensions(DkmGPUComputeKernel, Int32[], Int32[], Int32, DkmComputeKernelModel)

Get the dimension of the thread block.

GetGroupId(DkmGPUComputeThread, Int32[], Int32)

Get the dimension of the thread block.

GetStoppedThreads(DkmGPUComputeKernel)

Get all threads that hit breakpoint.

GetThisGroupDimension(DkmGPUComputeThread, Int32[], Int32)

Get the dimension of the thread block.

GetThisThreadDimension(DkmGPUComputeThread, Int32[], Int32)

Get the dimension of the thread block.

GetThreadFromId(DkmGPUComputeKernel, UInt64, DkmGPUComputeThread)

Gets the DkmGPUComputeThread object for a given thread ID.

GetThreadId(DkmGPUComputeThread, Int32[], Int32)

Get the dimension of the thread block.

GroupBy(DkmGPUComputeKernel, DkmQueryComputeThreadInfoFlags, ReadOnlyCollection<UInt64>, DkmWhereClause)

Runs the group by query on thread info objects.

Select(DkmGPUComputeKernel, ReadOnlyCollection<UInt64>, DkmWhereClause)

Runs the select query on thread info objects.

UpdateFlaggedState(DkmGPUComputeKernel, DkmWhereClause, Boolean)

Update flagged state of compute threads.

UpdateFrozenState(DkmGPUComputeKernel, DkmWhereClause, Boolean)

Update frozen state of compute threads.

Applies to