[code-reflection] RFR: Model lifetimes of onnx session-related objects more explicitly
Maurizio Cimadamore
mcimadamore at openjdk.org
Fri Feb 28 12:46:49 UTC 2025
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.
-------------
Commit messages:
- Fix bug: replace confined arena with automatic arena in OnnxInterpreter
- Model lifetimes of session-related objects more explicitly
Changes: https://git.openjdk.org/babylon/pull/332/files
Webrev: https://webrevs.openjdk.org/?repo=babylon&pr=332&range=00
Stats: 90 lines in 4 files changed: 24 ins; 9 del; 57 mod
Patch: https://git.openjdk.org/babylon/pull/332.diff
Fetch: git fetch https://git.openjdk.org/babylon.git pull/332/head:pull/332
PR: https://git.openjdk.org/babylon/pull/332
More information about the babylon-dev
mailing list