[dtensor] Add propagate_tensor_meta function that skips cache if _are_we_tracing #161334
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/161334
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (10 Unrelated Failures)As of commit c94282a with merge base 2f0de0f ( FLAKY - The following jobs failed but were likely due to flakiness present on trunk:
BROKEN TRUNK - The following jobs 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. |
pytorch-bot
bot
added
ciflow/inductor
oncall: distributed
xmfan
approved these changes
Aug 25, 2025
azahed98
force-pushed
the
fix/log_softmax_cache
branch
from
86cdb4b to
92c690e
Compare
|
@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 |
bdhirsh
reviewed
Aug 27, 2025
| """ | ||
| return self._propagate_tensor_meta_non_cached(op_schema) | ||
|
|
||
| def propagate_tensor_meta( |
There was a problem hiding this comment.
PR to rename: #161744
pytorchmergebot
pushed a commit
that referenced
this pull request
Sep 3, 2025
Rename the wrapper `propagate_tensor_meta` added in #161334 to make it clearly private, and rename the existing LRU function to accommodate. Pull Request resolved: #161744 Approved by: https://github.com/bdhirsh
markc-614
pushed a commit
to markc-614/pytorch
that referenced
this pull request
Sep 17, 2025
…_we_tracing (pytorch#161334) Fixes an issue where the log softmax handler checked the tensor metadata cache without checking for tracing or symints. Probably best to merge this after pytorch#160798, but not strictly blocking. Pull Request resolved: pytorch#161334 Approved by: https://github.com/xmfan
markc-614
pushed a commit
to markc-614/pytorch
that referenced
this pull request
Sep 17, 2025
Rename the wrapper `propagate_tensor_meta` added in pytorch#161334 to make it clearly private, and rename the existing LRU function to accommodate. Pull Request resolved: pytorch#161744 Approved by: https://github.com/bdhirsh
mansiag05
pushed a commit
to mansiag05/pytorch
that referenced
this pull request
Sep 22, 2025
Rename the wrapper `propagate_tensor_meta` added in pytorch#161334 to make it clearly private, and rename the existing LRU function to accommodate. Pull Request resolved: pytorch#161744 Approved by: https://github.com/bdhirsh
dsashidh
pushed a commit
to dsashidh/pytorch
that referenced
this pull request
Sep 26, 2025
Rename the wrapper `propagate_tensor_meta` added in pytorch#161334 to make it clearly private, and rename the existing LRU function to accommodate. Pull Request resolved: pytorch#161744 Approved by: https://github.com/bdhirsh
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.
Fixes an issue where the log softmax handler checked the tensor metadata cache without checking for tracing or symints.
Probably best to merge this after #160798, but not strictly blocking.
cc @H-Huang @awgu @wanchaol @fegin @fduwjj @wz337 @wconstab @d4l3k @pragupta