Temp fix for bug in scale-out profiling by adding env variable to disable profiler around allreduce #1697
Conversation
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Outdated suggestions cannot be applied.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.
Currently, a bug in PTI causes indefinite hang in all_reduce call when attempting to profile code on multiple nodes. If the upstream Kineto enables the toggle option for XPU PTI (pytorch/kineto#1088), then this PR would allow users to profile distributed scale-out code.
This PR uses an env variable to disable profiling around the call to XPU allreduce. An additional change in pytorch would be necessary to enable
toggleCollectionDynamicfor XPU.Ideally, a fix can be implemented in a new PTI version 0.12.3 or later before 2.8 deadline. If not, this fix would be required by customers.
@ashokei