[export] inline jit.scripted function in export #155180
Closed
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
[ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/155180
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 503740c with merge base 3f34d26 ( 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. |
Does this handle control flows of the scripted function? I thought it would be nice to directly inline the scripted function to leverage its ability to capture them (which is most of the time why a user would want to script the function in the first place) |
It handles the control flow the same way as we exporting the python code: we specialize the constants and shapes, errors out for data-dependent ones. |
When we exporting a scripted function, we inline the original callable stored in "_torchdynamo_inline", this is the same strategy as torch.compile path. cc EikanWang jgong5 wenzhe-nrv sanchitintel [ghstack-poisoned]
I think it can be more helpful: the scripted functions are usually data dependent with control flows (that’s often why they are used) |
|
Yeah, I agree. Maybe we could try to use the auto converter tool we're developing when we see a jit.script annotated function. It's also a good test bed for testing the coverage of the auto converter. |
When we exporting a scripted function, we inline the original callable stored in "_torchdynamo_inline", this is the same strategy as torch.compile path. cc EikanWang jgong5 wenzhe-nrv sanchitintel [ghstack-poisoned]
When we export a scripted function, we inline the original callable stored in "_torchdynamo_inline", this is the same strategy as torch.compile path. We do the same thing for script method, where a "\_\_wrapped\_\_" attribute points to the original callable in most cases. There are some corner cases we identified: top-level jit.scripted modules' method doesn't have a \_\_wrapped\_\_. In this case, we fall back to the original scripted approach. Maybe there're more such cases but need verification. cc EikanWang jgong5 wenzhe-nrv sanchitintel [ghstack-poisoned]
When we export a scripted function, we inline the original callable stored in "_torchdynamo_inline", this is the same strategy as torch.compile path. We do the same thing for script method, where a "\_\_wrapped\_\_" attribute points to the original callable in most cases. There are some corner cases we identified: top-level jit.scripted modules' method doesn't have a \_\_wrapped\_\_. In this case, we fall back to the original scripted approach. Maybe there're more such cases but need verification. cc EikanWang jgong5 wenzhe-nrv sanchitintel [ghstack-poisoned]
When we export a scripted function, we inline the original callable stored in "_torchdynamo_inline", this is the same strategy as torch.compile path. We do the same thing for script method, where a "\_\_wrapped\_\_" attribute points to the original callable in most cases. There are some corner cases we identified: top-level jit.scripted modules' method doesn't have a \_\_wrapped\_\_. In this case, we fall back to the original scripted approach. Maybe there're more such cases but need verification. cc EikanWang jgong5 wenzhe-nrv sanchitintel [ghstack-poisoned]
zou3519
reviewed
Jun 5, 2025
zou3519
approved these changes
Jun 5, 2025
When we export a scripted function, we inline the original callable stored in "_torchdynamo_inline", this is the same strategy as torch.compile path. We do the same thing for script method, where a "\_\_wrapped\_\_" attribute points to the original callable in most cases. There are some corner cases we identified: top-level jit.scripted modules' method doesn't have a \_\_wrapped\_\_. In this case, we fall back to the original scripted approach. Maybe there're more such cases but need verification. cc EikanWang jgong5 wenzhe-nrv sanchitintel [ghstack-poisoned]
|
@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 |
thatgeeman
pushed a commit
to thatgeeman/pytorch-docathon
that referenced
this pull request
Jun 15, 2025
When we export a scripted function, we inline the original callable stored in "_torchdynamo_inline", this is the same strategy as torch.compile path. We do the same thing for script method, where a "\_\_wrapped\_\_" attribute points to the original callable in most cases. There are some corner cases we identified: top-level jit.scripted modules' method doesn't have a \_\_wrapped\_\_. In this case, we fall back to the original scripted approach. Maybe there're more such cases but need verification. Pull Request resolved: pytorch#155180 Approved by: https://github.com/zou3519
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.
Stack from ghstack (oldest at bottom):
When we export a scripted function, we inline the original callable stored in "_torchdynamo_inline", this is the same strategy as torch.compile path.
We do the same thing for script method, where a "__wrapped__" attribute points to the original callable in most cases. There are some corner cases we identified: top-level jit.scripted modules' method doesn't have a __wrapped__. In this case, we fall back to the original scripted approach. Maybe there're more such cases but need verification.
cc @EikanWang @jgong5 @wenzhe-nrv @sanchitintel