- allowCrossDeviceAccess
Enables/disables P2P memory access for multi-gpu
- allowMultiGPU
This method allows you to enable or disable multi-GPU mode. PLEASE NOTE: This is
NOT magic method, t
- getAvailableDevices
- isDebug
- setMaximumDeviceCache
This method allows you to specify maximum memory cache per device
- <init>
- allowPreallocation
If set to true, each non-cached allocation request will cause few additional
allocations, Default va
- enableDebug
With debug enabled all CUDA launches will become synchronous, with forced stream
synchronizations af
- getAllocationModel
- getCommandLanesNumber
- getExecutionModel
- getFirstMemory