[code-reflection] RFR: Model lifetimes of onnx session-related objects more explicitly [v2]
Adam Pocock
duke at openjdk.org
Mon Mar 3 13:13:08 UTC 2025
On Mon, 3 Mar 2025 12:21:38 GMT, Maurizio Cimadamore <mcimadamore at openjdk.org> wrote:
>> The class representing an onnx session is auto closeable. But, in the current code, a session is closed immediately after its `run` method is called. This is problematic because a session returns some ORTValues (tensors) which also need to be freed, but that cannot be freed immediately after calling `run` (as they need to be used by clients).
>>
>> To address this problem, I tweaked the session code to accept an external arena. All the allocation of session-related data structures now happens using that external arena. This means that the client can now be in charge of managing the lifetime of a session (see changes to MNIST demo).
>>
>> To test, I tweaked the MNIST code to do 10K iterations on each button pressed. Predictably, a single button pressed resulted in over 3g of memory being leaked. With these changes the memory arrives at ~400K (there is still some minor leak, but not sure worth pushing more).
>>
>> If the changes to the demo are not deemed good, I can withdraw this PR -- I mostly wanted to capture the result of my exploration somewhere.
>
> Maurizio Cimadamore has updated the pull request incrementally with one additional commit since the last revision:
>
> Rename local arena variables
Releasing the ONNX environment will cause it to crash the next time one is created when certain kinds of sessions are created. It needs to be created at most once per JVM instantiation.
-------------
PR Comment: https://git.openjdk.org/babylon/pull/332#issuecomment-2694347199
More information about the babylon-dev
mailing list