Fix PgNccl coalseced profiling#160680
Closed
H-Huang wants to merge 1 commit intogh/H-Huang/210/basefrom
Closed
Conversation
pytorch-bot
bot
added
oncall: distributed
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/160680
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 114cc2d with merge base 1c25871 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
fegin
approved these changes
Sep 26, 2025
kwen2501
approved these changes
Sep 28, 2025
Member
Author
|
@pytorchbot merge |
Collaborator
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 |
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
4 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Stack from ghstack (oldest at bottom):
Admittedly I'm a noob when looking at traces, but this looked pretty off to me:
Here is what is happening:
CPU side: In
endCoalescing, we create a work object with the profiling title "nccl:coalesced"GPU side: The CUDA kernels will inherit this profiling title
What is missing:
We forgot to call the record function callback. With this change we finishs immediately on the CPU side, but the ncclDevKernel_SendRecv still have the coalesced title. New trace looks like this:
cc @awgu @wanchaol @fegin @fduwjj @wz337 @wconstab @d4l3k @pragupta @msaroufim @dcci