Introduce missing collectives and small fixes to support local tensor mode in AutoParallel#168110
Closed
dzmitry-huba wants to merge 2 commits intogh/dzmitry-huba/13/basefrom
Closed
Introduce missing collectives and small fixes to support local tensor mode in AutoParallel#168110dzmitry-huba wants to merge 2 commits intogh/dzmitry-huba/13/basefrom
dzmitry-huba wants to merge 2 commits intogh/dzmitry-huba/13/basefrom
Conversation
… mode in AutoParallel [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/168110
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (1 Unrelated Failure)As of commit c27928b with merge base fb6af11 ( BROKEN TRUNK - The following job failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
ezyang
reviewed
Nov 19, 2025
|
|
||
|
|
||
| _LOCAL_TENSOR_MODE: list["LocalTensorMode"] = [] | ||
| _GLOBAL_LOCAL_TENSOR_MODE: list["LocalTensorMode"] = [] |
Contributor
There was a problem hiding this comment.
err, so what's the global mode lol
ezyang
approved these changes
Nov 19, 2025
…ocal tensor mode in AutoParallel" This PR introduces support for additional functional collectives used in AutoParallel. Another change is in the semantic of the tolist() on the LocalTensor. Previously LocalTensor would reconcile first and then return a single tensor that is same on all ranks. AutoParallel uses tolist() to compute all-to-all splits during token dispatch and combine. [ghstack-poisoned]
Contributor
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 |
Contributor
Author
|
@pytorchbot help |
|
❌ 🤖 pytorchbot command failed: Try |
Contributor
Author
|
@pytorchbot --help |
PyTorchBot HelpMergeRevertRebaseLabelDr CIcherry-pick |
Collaborator
XueningXu
pushed a commit
to XueningXu/pytorch
that referenced
this pull request
Nov 19, 2025
… mode in AutoParallel (pytorch#168110) This PR introduces support for additional functional collectives used in AutoParallel. Another change is in the semantic of the tolist() on the LocalTensor. Previously LocalTensor would reconcile first and then return a single tensor that is same on all ranks. AutoParallel uses tolist() to compute all-to-all splits during token dispatch and combine. Pull Request resolved: pytorch#168110 Approved by: https://github.com/ezyang
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.
This PR introduces support for additional functional collectives used in AutoParallel.
Another change is in the semantic of the tolist() on the LocalTensor. Previously LocalTensor would reconcile first and then return a single tensor that is same on all ranks. AutoParallel uses tolist() to compute all-to-all splits during token dispatch and combine.
Stack from ghstack (oldest at bottom):