Fix XProf GCS upload bug in max_utils.Profiler#409
Open
eltsai wants to merge 1 commit into
Open
Conversation
e4b9135 to
d6c9c2a
Compare
Collaborator
Author
|
Also disabled profiler for the first two runs. The xprof functions is as expected now: |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Creating this PR based on CL:
This change fixes a bug where XProf traces were not uploaded to GCS when the profiler was used manually via
start()andstop()calls, which is the case in most MaxDiffusion trainers.The GCS upload logic was previously located in
Profiler.__exit__(used by the context manager), but not inProfiler.stop(). This change moves the upload logic tostop()and guards it with_jax_profiler_enabled(self.config)to ensure it only runs when JAX profiling was active.Also:
profiler_test.pyto verify GCS upload for both manual and context manager usage.profiler_testtarget toBUILD.