MemPool#

class torch.xpu.memory.MemPool(*args, **kwargs)[source]#

MemPool represents a pool of memory in a caching allocator. Currently, it’s just the ID of the pool object maintained in the XPUCachingAllocator.

Parameters:

allocator (torch._C._xpu_XPUAllocator, optional) – a torch._C._xpu_XPUAllocator object that can be used to define how memory gets allocated in the pool. If allocator is None (default), memory allocation follows the default/ current configuration of the XPUCachingAllocator.
use_on_oom (bool, optional) – a bool that indicates if this pool may be used as a last resort when an allocation outside the pool fails with OOM. Defaults to False.
no_split (bool, optional) – if True, cached segments in this pool are never split to serve smaller allocations. Defaults to False.

property id: tuple[int, int]#: Returns the ID of this pool as a tuple of two ints.

snapshot()[source]#

Return a snapshot of the XPU memory allocator pool state across all devices.

Interpreting the output of this function requires familiarity with the memory allocator internals.

use_count()[source]#

Returns the reference count of this pool.

Docs