CUDACachingAllocator,c10d: fixes for IPC release performance #148805
Closed
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/148805
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 13aca6d with merge base b958890 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Skylion007
approved these changes
Mar 8, 2025
fegin
approved these changes
Mar 10, 2025
zdevito
approved these changes
Mar 10, 2025
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
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Outdated suggestions cannot be applied.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.
This has two fixes to improve IPC tensor release performance when using torchft's BabyProcessGroupNCCL.
ExpandableSegementsobject to avoid synchronizing under the lockTest plan:
Run with torchft + torchtitan
Check py-spy to verify no bottleneck on IPC lock when creating new shared tensors
cc @H-Huang @awgu @kwen2501 @wanchaol @fegin @fduwjj @wz337 @wconstab @c-p-i-o