[dynamo] Make global state guards and torch function stack guards droppable.#167674
Closed
[dynamo] Make global state guards and torch function stack guards droppable.#167674
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/167674
Note: Links to docs will display an error until the docs builds have been completed. ⏳ 1 Pending, 4 Unrelated FailuresAs of commit ec46563 with merge base 4de24bc ( FLAKY - The following job failed but was likely due to flakiness present on trunk:
BROKEN TRUNK - The following jobs failed but was present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
UNSTABLE - The following job is marked as unstable, possibly due to flakiness on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
ee7e48e to
b023bb6
Compare
…ppable. Summary: Prior to this PR we will always build global and torch funciton guards in all cases. In this PR we did 2 changes to dynamo guards: 1. Created a new guard called "GLOBAL_STATE" which corresponds to the global state guard and can be filtered out using guard_filter_fn 2. Repurpose the existing "TORCH_FUNCTION_STATE" guard for checking torch function mode stack. Also added a new helper `torch.compiler.skip_all_guards_unsafe` which can be useful for use cases like vllm Test Plan: CI
b023bb6 to
ec46563
Compare
anijain2305
reviewed
Nov 13, 2025
| return [not entry.is_global for entry in guard_entries] | ||
|
|
||
|
|
||
| def skip_all_guards_unsafe(guard_entries): |
Contributor
There was a problem hiding this comment.
This is good.
Is it also possible to add a config option somewhere to skip building the guards itself? If user does not want guards, we can just short-circuit sooner.
Contributor
Author
There was a problem hiding this comment.
Yeah, sounds a good item to do. Not sure about the amount of changed code yet if we skip builder entirely. Can take a look further.
anijain2305
approved these changes
Nov 13, 2025
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 |
jsuarez5341
pushed a commit
to PufferAI/pytorch
that referenced
this pull request
Nov 15, 2025
…ppable. (pytorch#167674) Summary: Prior to this PR we will always build global and torch funciton guards in all cases. In this PR we did 2 changes to dynamo guards: 1. Created a new guard called "GLOBAL_STATE" which corresponds to the global state guard and can be filtered out using guard_filter_fn 2. Repurpose the existing "TORCH_FUNCTION_STATE" guard for checking torch function mode stack. Also added a new helper `torch.compiler.skip_all_guards_unsafe` which can be useful for use cases like vllm Test Plan: CI Fixes #ISSUE_NUMBER Pull Request resolved: pytorch#167674 Approved by: https://github.com/anijain2305
Silv3S
pushed a commit
to Silv3S/pytorch
that referenced
this pull request
Nov 18, 2025
…ppable. (pytorch#167674) Summary: Prior to this PR we will always build global and torch funciton guards in all cases. In this PR we did 2 changes to dynamo guards: 1. Created a new guard called "GLOBAL_STATE" which corresponds to the global state guard and can be filtered out using guard_filter_fn 2. Repurpose the existing "TORCH_FUNCTION_STATE" guard for checking torch function mode stack. Also added a new helper `torch.compiler.skip_all_guards_unsafe` which can be useful for use cases like vllm Test Plan: CI Fixes #ISSUE_NUMBER Pull Request resolved: pytorch#167674 Approved by: https://github.com/anijain2305
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.
Summary:
Prior to this PR we will always build global and torch funciton guards in all cases.
In this PR we did 2 changes to dynamo guards:
Also added a new helper
torch.compiler.skip_all_guards_unsafewhich can be useful for use cases like vllmTest Plan:
CI
Fixes #ISSUE_NUMBER
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @chenyang78 @kadeng @chauhang @amjames @Lucaskabela @mlazos