onnxruntime/include
RandySheriffH 009cd4ea2e
Allow cuda custom ops allocate deferred cpu mem (#17893)
Expose a new allocator from cuda stream.
The allocator manages deferred cpu memory which only get recycled before
stream destruction.

---------

Co-authored-by: Randy Shuai <rashuai@microsoft.com>
2023-10-20 16:12:21 -07:00
..
onnxruntime/core Allow cuda custom ops allocate deferred cpu mem (#17893) 2023-10-20 16:12:21 -07:00