[dynamo] rename set_fullgraph to error_on_graph_break by williamwen42 · Pull Request #161739 · pytorch/pytorch

williamwen42 · 2025-08-28T21:24:21Z

Stack from ghstack (oldest at bottom):

Renaming set_fullgraph to error_on_graph_break for now. There are no semantic differences yet. In a followup PR, we will introduce a new torch.compile option error_on_graph_break that has lower priority than fullgraph so that fullgraph really returns 1 graph.

I could keep set_fullgraph as a deprecated alias for error_on_graph_break for now, but I'm hoping that won't be necessary since it's still private API (there are no internal callsites yet, and there are no significant OSS callsites yet).

cc @albanD @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @chenyang78 @kadeng @chauhang @amjames @Lucaskabela @mlazos @guilhermeleobas @xmfan as primary users for set_fullgraph

cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @chenyang78 @kadeng @chauhang @amjames @Lucaskabela

[ghstack-poisoned]

pytorch-bot · 2025-08-28T21:24:26Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/161739

📄 Preview Python docs built from this PR
📄 Preview C++ docs built from this PR
❓ Need help or want to give feedback on the CI? Visit the bot commands wiki

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit 4130014 with merge base 734ce8e ():
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

ghstack-source-id: 8949b34 Pull Request resolved: #161739

williamwen42 · 2025-08-29T18:19:16Z

@albanD skipping PR sanity check since we're just doing a rename here

Lucaskabela · 2025-08-29T18:29:58Z

-    Context manager/decorator to toggle fullgraph setting.
+    Context manager/decorator to toggle error_on_graph_break (i.e. torch.compile's fullgraph) setting.

    More precisely, when encountering a graph break, we will decide to resume (fullgraph=False)


Should we update this to reference the new error graph values?

Next PR - this PR doesn't have semantic changes yet.

Renaming `set_fullgraph` to `error_on_graph_break` for now. There are no semantic differences yet. In a followup PR, we will introduce a new `torch.compile` option `error_on_graph_break` that has lower priority than `fullgraph` so that `fullgraph` really returns 1 graph. I could keep `set_fullgraph` as a deprecated alias for `error_on_graph_break` for now, but I'm hoping that won't be necessary since it's still private API (there are no internal callsites yet, and there are no significant OSS callsites yet). cc guilhermeleobas xmfan as primary users for `set_fullgraph` cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 kadeng chauhang amjames Lucaskabela [ghstack-poisoned]

williamwen42 · 2025-09-03T16:10:01Z

@pytorchbot merge

pytorchmergebot · 2025-09-03T16:12:01Z

Merge started

Your 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

Advanced Debugging

Check the merge workflow status
here

pytorchmergebot · 2025-09-03T18:58:15Z

Merge failed

Reason: Command git -C /home/runner/work/pytorch/pytorch rebase origin/main returned non-zero exit code 1

Rebasing (1/1)
Auto-merging test/dynamo/cpython/3_13/test_itertools.diff
CONFLICT (content): Merge conflict in test/dynamo/cpython/3_13/test_itertools.diff
Auto-merging test/dynamo/cpython/3_13/test_itertools.py
CONFLICT (content): Merge conflict in test/dynamo/cpython/3_13/test_itertools.py
error: could not apply 503f1469f40... [dynamo] rename set_fullgraph to error_on_graph_break (#161739)
hint: Resolve all conflicts manually, mark them as resolved with
hint: "git add/rm <conflicted_files>", then run "git rebase --continue".
hint: You can instead skip this commit: run "git rebase --skip".
hint: To abort and get back to the state before "git rebase", run "git rebase --abort".
hint: Disable this message with "git config set advice.mergeConflict false"
Could not apply 503f1469f40... # [dynamo] rename set_fullgraph to error_on_graph_break (#161739)

Details for Dev Infra team

Raised by workflow job

Renaming `set_fullgraph` to `error_on_graph_break` for now. There are no semantic differences yet. In a followup PR, we will introduce a new `torch.compile` option `error_on_graph_break` that has lower priority than `fullgraph` so that `fullgraph` really returns 1 graph. I could keep `set_fullgraph` as a deprecated alias for `error_on_graph_break` for now, but I'm hoping that won't be necessary since it's still private API (there are no internal callsites yet, and there are no significant OSS callsites yet). cc albanD voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 kadeng chauhang amjames Lucaskabela mlazos guilhermeleobas xmfan as primary users for `set_fullgraph` cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 kadeng chauhang amjames Lucaskabela [ghstack-poisoned]

williamwen42 · 2025-09-03T23:27:29Z

@pytorchbot merge

pytorchmergebot · 2025-09-03T23:29:12Z

Merge started

Your 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

Advanced Debugging

Check the merge workflow status
here

This PR implements the semantics change to `torch._dynamo.error_on_graph_break`: - ~`torch.compile` now has a new `error_on_graph_break` kwarg that serves as a lower-priority toggle for erroring/continuing on graph breaks~ - `error_on_graph_break` is a new internal `torch.compile `setting that is lower-priority than `fullgraph`. It allows the user to toggle erroring/continuing on graph breaks. - `error_on_graph_break` does nothing when `fullgraph=True` - `error_on_graph_break` does NOT guarantee a single graph Followup [DONE]: need to change the programming model docs to reflect the 3 graph break modes for compilation: - `fullgraph=True`: enforce one graph, no graph breaks, cannot be toggled - `fullgraph=False, error_on_graph_break=True`: errors on graph breaks, latter can be toggled during compile time - `fullgraph=False, error_on_graph_break=False`: resumes tracing on graph breaks, latter can be toggled during compile time Pull Request resolved: #161747 Approved by: https://github.com/mlazos ghstack dependencies: #161739

@albanD

Renaming `set_fullgraph` to `error_on_graph_break` for now. There are no semantic differences yet. In a followup PR, we will introduce a new `torch.compile` option `error_on_graph_break` that has lower priority than `fullgraph` so that `fullgraph` really returns 1 graph. I could keep `set_fullgraph` as a deprecated alias for `error_on_graph_break` for now, but I'm hoping that won't be necessary since it's still private API (there are no internal callsites yet, and there are no significant OSS callsites yet). cc @albanD @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @chenyang78 @kadeng @chauhang @amjames @Lucaskabela @mlazos @guilhermeleobas @xmfan as primary users for `set_fullgraph` Pull Request resolved: pytorch#161739 Approved by: https://github.com/xmfan, https://github.com/Lucaskabela, https://github.com/anijain2305, https://github.com/mlazos

) This PR implements the semantics change to `torch._dynamo.error_on_graph_break`: - ~`torch.compile` now has a new `error_on_graph_break` kwarg that serves as a lower-priority toggle for erroring/continuing on graph breaks~ - `error_on_graph_break` is a new internal `torch.compile `setting that is lower-priority than `fullgraph`. It allows the user to toggle erroring/continuing on graph breaks. - `error_on_graph_break` does nothing when `fullgraph=True` - `error_on_graph_break` does NOT guarantee a single graph Followup [DONE]: need to change the programming model docs to reflect the 3 graph break modes for compilation: - `fullgraph=True`: enforce one graph, no graph breaks, cannot be toggled - `fullgraph=False, error_on_graph_break=True`: errors on graph breaks, latter can be toggled during compile time - `fullgraph=False, error_on_graph_break=False`: resumes tracing on graph breaks, latter can be toggled during compile time Pull Request resolved: pytorch#161747 Approved by: https://github.com/mlazos ghstack dependencies: pytorch#161739

@albanD

Renaming `set_fullgraph` to `error_on_graph_break` for now. There are no semantic differences yet. In a followup PR, we will introduce a new `torch.compile` option `error_on_graph_break` that has lower priority than `fullgraph` so that `fullgraph` really returns 1 graph. I could keep `set_fullgraph` as a deprecated alias for `error_on_graph_break` for now, but I'm hoping that won't be necessary since it's still private API (there are no internal callsites yet, and there are no significant OSS callsites yet). cc @albanD @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @chenyang78 @kadeng @chauhang @amjames @Lucaskabela @mlazos @guilhermeleobas @xmfan as primary users for `set_fullgraph` Pull Request resolved: pytorch#161739 Approved by: https://github.com/xmfan, https://github.com/Lucaskabela, https://github.com/anijain2305, https://github.com/mlazos

) This PR implements the semantics change to `torch._dynamo.error_on_graph_break`: - ~`torch.compile` now has a new `error_on_graph_break` kwarg that serves as a lower-priority toggle for erroring/continuing on graph breaks~ - `error_on_graph_break` is a new internal `torch.compile `setting that is lower-priority than `fullgraph`. It allows the user to toggle erroring/continuing on graph breaks. - `error_on_graph_break` does nothing when `fullgraph=True` - `error_on_graph_break` does NOT guarantee a single graph Followup [DONE]: need to change the programming model docs to reflect the 3 graph break modes for compilation: - `fullgraph=True`: enforce one graph, no graph breaks, cannot be toggled - `fullgraph=False, error_on_graph_break=True`: errors on graph breaks, latter can be toggled during compile time - `fullgraph=False, error_on_graph_break=False`: resumes tracing on graph breaks, latter can be toggled during compile time Pull Request resolved: pytorch#161747 Approved by: https://github.com/mlazos ghstack dependencies: pytorch#161739

@albanD

Renaming `set_fullgraph` to `error_on_graph_break` for now. There are no semantic differences yet. In a followup PR, we will introduce a new `torch.compile` option `error_on_graph_break` that has lower priority than `fullgraph` so that `fullgraph` really returns 1 graph. I could keep `set_fullgraph` as a deprecated alias for `error_on_graph_break` for now, but I'm hoping that won't be necessary since it's still private API (there are no internal callsites yet, and there are no significant OSS callsites yet). cc @albanD @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @chenyang78 @kadeng @chauhang @amjames @Lucaskabela @mlazos @guilhermeleobas @xmfan as primary users for `set_fullgraph` Pull Request resolved: pytorch#161739 Approved by: https://github.com/xmfan, https://github.com/Lucaskabela, https://github.com/anijain2305, https://github.com/mlazos

) This PR implements the semantics change to `torch._dynamo.error_on_graph_break`: - ~`torch.compile` now has a new `error_on_graph_break` kwarg that serves as a lower-priority toggle for erroring/continuing on graph breaks~ - `error_on_graph_break` is a new internal `torch.compile `setting that is lower-priority than `fullgraph`. It allows the user to toggle erroring/continuing on graph breaks. - `error_on_graph_break` does nothing when `fullgraph=True` - `error_on_graph_break` does NOT guarantee a single graph Followup [DONE]: need to change the programming model docs to reflect the 3 graph break modes for compilation: - `fullgraph=True`: enforce one graph, no graph breaks, cannot be toggled - `fullgraph=False, error_on_graph_break=True`: errors on graph breaks, latter can be toggled during compile time - `fullgraph=False, error_on_graph_break=False`: resumes tracing on graph breaks, latter can be toggled during compile time Pull Request resolved: pytorch#161747 Approved by: https://github.com/mlazos ghstack dependencies: pytorch#161739

ghstack-source-id: 90fe29d Pull Request resolved: pytorch/pytorch#161739

[dynamo] rename set_fullgraph to error_on_graph_break

a02df43

[ghstack-poisoned]

pytorch-bot Bot added ciflow/inductor module: dynamo labels Aug 28, 2025

williamwen42 added a commit that referenced this pull request Aug 28, 2025

[dynamo] rename set_fullgraph to error_on_graph_break

61b43a2

ghstack-source-id: 8949b34 Pull Request resolved: #161739

williamwen42 added release notes: dynamo module: compile ux UX issues related to torch.compile labels Aug 28, 2025

williamwen42 mentioned this pull request Aug 28, 2025

[dynamo] change error_on_graph_break/fullgraph semantics #161747

Closed

williamwen42 requested review from Lucaskabela, anijain2305 and zou3519 August 29, 2025 18:13

xmfan approved these changes Aug 29, 2025

View reviewed changes

williamwen42 added the skip-pr-sanity-checks label Aug 29, 2025

Lucaskabela approved these changes Aug 29, 2025

View reviewed changes

williamwen42 added 2 commits September 2, 2025 14:14

williamwen42 mentioned this pull request Sep 2, 2025

[dynamo] add error_on_graph_break() to compiler namespace #162023

Closed

anijain2305 approved these changes Sep 2, 2025

View reviewed changes

williamwen42 mentioned this pull request Sep 3, 2025

[dynamo] add error_on_graph_break kwarg to torch.compile #162027

Closed

pytorch-bot Bot added the ciflow/trunk Trigger trunk jobs on your pull request label Sep 3, 2025

pytorchmergebot added the merging label Sep 3, 2025

pytorchmergebot removed the merging label Sep 3, 2025

mlazos approved these changes Sep 3, 2025

View reviewed changes

pytorchmergebot added the merging label Sep 3, 2025

pytorchmergebot added the Merged label Sep 4, 2025

pytorchmergebot closed this in 8678d83 Sep 4, 2025

pytorchmergebot removed the merging label Sep 4, 2025

github-actions Bot deleted the gh/williamwen42/283/head branch October 4, 2025 02:06

Khanaksahu pushed a commit to Khanaksahu/pytorch-fork that referenced this pull request Nov 17, 2025

[dynamo] rename set_fullgraph to error_on_graph_break

85b72a2

ghstack-source-id: 90fe29d Pull Request resolved: pytorch/pytorch#161739

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[dynamo] rename set_fullgraph to error_on_graph_break#161739

[dynamo] rename set_fullgraph to error_on_graph_break#161739
williamwen42 wants to merge 4 commits intogh/williamwen42/283/basefrom
gh/williamwen42/283/head

williamwen42 commented Aug 28, 2025 •

edited by pytorch-bot Bot

Loading

Uh oh!

pytorch-bot Bot commented Aug 28, 2025 •

edited

Loading

Uh oh!

williamwen42 commented Aug 29, 2025

Uh oh!

Lucaskabela Aug 29, 2025

Uh oh!

williamwen42 Aug 29, 2025

Uh oh!

williamwen42 commented Sep 3, 2025

Uh oh!

pytorchmergebot commented Sep 3, 2025

Uh oh!

pytorchmergebot commented Sep 3, 2025

Uh oh!

williamwen42 commented Sep 3, 2025

Uh oh!

pytorchmergebot commented Sep 3, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

Conversation

williamwen42 commented Aug 28, 2025 • edited by pytorch-bot Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-bot Bot commented Aug 28, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/161739

✅ No Failures

Uh oh!

williamwen42 commented Aug 29, 2025

Uh oh!

Lucaskabela Aug 29, 2025

Choose a reason for hiding this comment

Uh oh!

williamwen42 Aug 29, 2025

Choose a reason for hiding this comment

Uh oh!

williamwen42 commented Sep 3, 2025

Uh oh!

pytorchmergebot commented Sep 3, 2025

Merge started

Uh oh!

pytorchmergebot commented Sep 3, 2025

Merge failed

Uh oh!

williamwen42 commented Sep 3, 2025

Uh oh!

pytorchmergebot commented Sep 3, 2025

Merge started

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

williamwen42 commented Aug 28, 2025 •

edited by pytorch-bot Bot

Loading

pytorch-bot Bot commented Aug 28, 2025 •

edited

Loading