[DTensor] no-op redistribution shouldn't create _TransformInfo#172924
Closed
wconstab wants to merge 7 commits intogh/wconstab/505/basefrom
Closed
[DTensor] no-op redistribution shouldn't create _TransformInfo#172924wconstab wants to merge 7 commits intogh/wconstab/505/basefrom
wconstab wants to merge 7 commits intogh/wconstab/505/basefrom
Conversation
Previously, generate_greedy_transform_infos was happy to generate a TransformInfo with src placements == dst placements, which was pointless. Graph-based algo was safe, avoiding this. [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/172924
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 1b71100 with merge base 4ac0624 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
…Info" Previously, generate_greedy_transform_infos was happy to generate a TransformInfo with src placements == dst placements, which was pointless. Graph-based algo was safe, avoiding this. [ghstack-poisoned]
…Info" Previously, generate_greedy_transform_infos was happy to generate a TransformInfo with src placements == dst placements, which was pointless. Graph-based algo was safe, avoiding this. [ghstack-poisoned]
wconstab
commented
Jan 22, 2026
| return local_tensor | ||
|
|
||
| # Short-circuit if placements are already equal (no-op redistribution) | ||
| if current_spec.placements == target_spec.placements: |
Contributor
Author
There was a problem hiding this comment.
@zpcore I confirmed that if I comment out these 2 lines, the tests all pass. With this short circuit, I get a hang in python test/distributed/tensor/test_redistribute.py DistributeWithDeviceOrderTest.test_ordered_redistribute, and I didn't debug it yet but I wonder if you can think of any reason why this could happen.
Member
There was a problem hiding this comment.
This is because even though current_spec.placements == target_spec.placements, current_spec.shard_order and target_spec.shard_order can still be different, which requires a redistribution. We can update this line to:
if current_spec.placements == target_spec.placements and current_spec.shard_order == target_spec.shard_order:
Contributor
Author
There was a problem hiding this comment.
thanks, that makes sense!!
…Info" Previously, generate_greedy_transform_infos was happy to generate a TransformInfo with src placements == dst placements, which was pointless. Graph-based algo was safe, avoiding this. [ghstack-poisoned]
…Info" Previously, generate_greedy_transform_infos was happy to generate a TransformInfo with src placements == dst placements, which was pointless. Graph-based algo was safe, avoiding this. [ghstack-poisoned]
…Info" Previously, generate_greedy_transform_infos was happy to generate a TransformInfo with src placements == dst placements, which was pointless. Graph-based algo was safe, avoiding this. [ghstack-poisoned]
…Info" Previously, generate_greedy_transform_infos was happy to generate a TransformInfo with src placements == dst placements, which was pointless. Graph-based algo was safe, avoiding this. [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 |
suncapitalllc007-star
pushed a commit
to suncapitalllc007-star/pytorch
that referenced
this pull request
Jan 25, 2026
Previously, generate_greedy_transform_infos was happy to generate a TransformInfo with src placements == dst placements, which was pointless. Graph-based algo was safe, avoiding this. ghstack-source-id: b83774e Pull Request resolved: pytorch/pytorch#172924
pytorchmergebot
pushed a commit
that referenced
this pull request
Jan 26, 2026
Pull Request resolved: #172609 Approved by: https://github.com/zpcore, https://github.com/tianyu-l ghstack dependencies: #172924
riccardofelluga
pushed a commit
to riccardofelluga/pytorch
that referenced
this pull request
Jan 27, 2026
…ch#172924) Previously, generate_greedy_transform_infos was happy to generate a TransformInfo with src placements == dst placements, which was pointless. Graph-based algo was safe, avoiding this. Pull Request resolved: pytorch#172924 Approved by: https://github.com/zpcore
riccardofelluga
pushed a commit
to riccardofelluga/pytorch
that referenced
this pull request
Jan 27, 2026
) Pull Request resolved: pytorch#172609 Approved by: https://github.com/zpcore, https://github.com/tianyu-l ghstack dependencies: pytorch#172924
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
3 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):
Previously, generate_greedy_transform_infos was happy to generate a
TransformInfo with src placements == dst placements, which was
pointless. Graph-based algo was safe, avoiding this.