Enum CudaCooperativeLaunchMultiDeviceFlags
CudaCooperativeLaunchMultiDeviceFlags
Namespace: ManagedCuda.BasicTypes
Assembly: ManagedCuda.dll
Syntax
[Flags]
public enum CudaCooperativeLaunchMultiDeviceFlagsFields
| Name | Description |
|---|---|
| None | No flags |
| NoPostLaunchSync | If set, any subsequent work pushed in a stream that participated in a call to ::cuLaunchCooperativeKernelMultiDevice will only wait for the kernel launched on the GPU corresponding to that stream to complete before it begins execution. |
| NoPreLaunchSync | If set, each kernel launched as part of ::cuLaunchCooperativeKernelMultiDevice only waits for prior work in the stream corresponding to that GPU to complete before the kernel begins execution. |