You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Heaptrack has proven to be a useful tool so we should try to automate its usage. One easy way to do this is to add tracking of the peak memory usage in PRs, any significant changes should be reported with a warning comment.
We shouldn't need too many programs for this: a small, medium and large one should suffice as it'll allow us to see the fixed size memory costs vs worst case.
Ideally this should emit a workflow artifact of the profiling run so we can easily just download it and open in heaptrack.
The text was updated successfully, but these errors were encountered:
Heaptrack has proven to be a useful tool so we should try to automate its usage. One easy way to do this is to add tracking of the peak memory usage in PRs, any significant changes should be reported with a warning comment.
We shouldn't need too many programs for this: a small, medium and large one should suffice as it'll allow us to see the fixed size memory costs vs worst case.
Ideally this should emit a workflow artifact of the profiling run so we can easily just download it and open in heaptrack.
The text was updated successfully, but these errors were encountered: