cuda.core.VirtualMemoryResourceOptions#
- class cuda.core.VirtualMemoryResourceOptions(
- allocation_type: ~cuda.core.typing.VirtualMemoryAllocationType = VirtualMemoryAllocationType.PINNED,
- location_type: ~cuda.core.typing.VirtualMemoryLocationType = VirtualMemoryLocationType.DEVICE,
- handle_type: ~cuda.core.typing.VirtualMemoryHandleType = VirtualMemoryHandleType.POSIX_FD,
- granularity: ~cuda.core.typing.VirtualMemoryGranularityType = VirtualMemoryGranularityType.RECOMMENDED,
- gpu_direct_rdma: bool = False,
- addr_hint: int | None = 0,
- addr_align: int | None = None,
- peers: ~typing.Iterable[int] = <factory>,
- self_access: ~cuda.core.typing.VirtualMemoryAccessType = VirtualMemoryAccessType.READ_WRITE,
- peer_access: ~cuda.core.typing.VirtualMemoryAccessType = VirtualMemoryAccessType.READ_WRITE,
- A configuration object for the VirtualMemoryResource
Stores configuration information which tells the resource how to use the CUDA VMM APIs
- allocation_type#
Controls the type of allocation.
- Type:
VirtualMemoryAllocationType| str
- location_type#
Controls the location of the allocation.
- Type:
VirtualMemoryLocationType| str
- handle_type#
Export handle type for the physical allocation. Use
"posix_fd"on Linux if you plan to import/export the allocation (required for cuMemRetainAllocationHandle). Use None if you don’t need an exportable handle.- Type:
VirtualMemoryHandleType| str
- granularity#
Controls granularity query and size rounding.
- Type:
VirtualMemoryGranularityType| str
- addr_hint#
A (optional) virtual address hint to try to reserve at. Setting it to 0 lets the CUDA driver decide.
- Type:
- self_access#
Access flags for the owning device.
- Type:
VirtualMemoryAccessType| None | str
- peer_access#
Access flags for peers.
- Type:
VirtualMemoryAccessType| None | str