Add logging for when inbuilt_inline_nn_modules will help with ID_MATCH guard triggered recompiles #160592

ppanchalia · 2025-08-14T00:01:57Z

We add a logging around when an ID_MATCH guard is added at a place where inbuilt_inline_nn_modules would inline it. This is done with the aim of tagging recompiles that could be avoided by setting inbuilt_inline_nn_modules flag.
It will help us log and track the flag's adoption and potentially quantify saving in the the number of recompiles.

Differential Revision: D80075975

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

pytorch-bot · 2025-08-14T00:02:00Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/160592

📄 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 or our office hours

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 338e2f3 with merge base 3650989 ():

BROKEN TRUNK - The following job failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

inductor / cuda12.8-py3.10-gcc9-sm86 / test (inductor_torchbench, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh) (trunk failure)
vision_maskrcnn

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

facebook-github-bot · 2025-08-14T00:02:16Z

This pull request was exported from Phabricator. Differential Revision: D80075975

torch/_dynamo/convert_frame.py

…H guard triggered recompiles (pytorch#160592) Summary: We add a logging around when an ID_MATCH guard is added at a place where inbuilt_inline_nn_modules would inline it. This is done with the aim of tagging recompiles that could be avoided by setting inbuilt_inline_nn_modules flag. It will help us log and track the flag's adoption and potentially quantify saving in the the number of recompiles. Test Plan: rm -rf /tmp/r && TORCH_TRACE=/tmp/r buck2 run //scripts/ppanchalia/pt2_workspace:inline_inbuilt_nn_modules_test and CI Tests Rollback Plan: Differential Revision: D80075975

facebook-github-bot · 2025-08-14T02:26:31Z

This pull request was exported from Phabricator. Differential Revision: D80075975

…H guard triggered recompiles (pytorch#160592) Summary: We add a logging around when an ID_MATCH guard is added at a place where inbuilt_inline_nn_modules would inline it. This is done with the aim of tagging recompiles that could be avoided by setting inbuilt_inline_nn_modules flag. It will help us log and track the flag's adoption and potentially quantify saving in the the number of recompiles. Test Plan: rm -rf /tmp/r && TORCH_TRACE=/tmp/r buck2 run //scripts/ppanchalia/pt2_workspace:inline_inbuilt_nn_modules_test and CI Tests Rollback Plan: Differential Revision: D80075975

facebook-github-bot · 2025-08-14T05:05:48Z

This pull request was exported from Phabricator. Differential Revision: D80075975

torch/_dynamo/convert_frame.py

torch/_dynamo/guards.py

…H guard triggered recompiles (pytorch#160592) Summary: We add a logging around when an ID_MATCH guard is added at a place where inbuilt_inline_nn_modules would inline it. This is done with the aim of tagging recompiles that could be avoided by setting inbuilt_inline_nn_modules flag. It will help us log and track the flag's adoption and potentially quantify saving in the the number of recompiles. Test Plan: rm -rf /tmp/r && TORCH_TRACE=/tmp/r buck2 run //scripts/ppanchalia/pt2_workspace:inline_inbuilt_nn_modules_test and CI Tests Rollback Plan: Differential Revision: D80075975

torch/_dynamo/guards.py

…H guard triggered recompiles (pytorch#160592) Summary: We add a logging around when an ID_MATCH guard is added at a place where inbuilt_inline_nn_modules would inline it. This is done with the aim of tagging recompiles that could be avoided by setting inbuilt_inline_nn_modules flag. It will help us log and track the flag's adoption and potentially quantify saving in the the number of recompiles. Test Plan: rm -rf /tmp/r && TORCH_TRACE=/tmp/r buck2 run //scripts/ppanchalia/pt2_workspace:inline_inbuilt_nn_modules_test and CI Tests Rollback Plan: Differential Revision: D80075975

facebook-github-bot · 2025-08-14T05:16:47Z

This pull request was exported from Phabricator. Differential Revision: D80075975

…H guard triggered recompiles (pytorch#160592) Summary: Pull Request resolved: pytorch#160592 We add a logging around when an ID_MATCH guard is added at a place where inbuilt_inline_nn_modules would inline it. This is done with the aim of tagging recompiles that could be avoided by setting inbuilt_inline_nn_modules flag. It will help us log and track the flag's adoption and potentially quantify saving in the the number of recompiles. Test Plan: rm -rf /tmp/r && TORCH_TRACE=/tmp/r buck2 run //scripts/ppanchalia/pt2_workspace:inline_inbuilt_nn_modules_test and CI Tests Rollback Plan: Differential Revision: D80075975

facebook-github-bot · 2025-08-14T05:16:54Z

This pull request was exported from Phabricator. Differential Revision: D80075975

anijain2305

Very close to be done. Can you also add a test? For test, you can add a guard hook and check if the str [inline-inbuilt-nn-modules-candidate] is present in the str representation of guards. A pointer to do that is here -

pytorch/test/dynamo/test_guard_manager.py

Line 1055 in 2898d3f

with install_guard_manager_testing_hook(hook):

facebook-github-bot · 2025-08-14T05:24:17Z

This pull request was exported from Phabricator. Differential Revision: D80075975

…H guard triggered recompiles (pytorch#160592) Summary: Pull Request resolved: pytorch#160592 We add a logging around when an ID_MATCH guard is added at a place where inbuilt_inline_nn_modules would inline it. This is done with the aim of tagging recompiles that could be avoided by setting inbuilt_inline_nn_modules flag. It will help us log and track the flag's adoption and potentially quantify saving in the the number of recompiles. Test Plan: rm -rf /tmp/r && TORCH_TRACE=/tmp/r buck2 run //scripts/ppanchalia/pt2_workspace:inline_inbuilt_nn_modules_test and CI Tests Rollback Plan: Differential Revision: D80075975

facebook-github-bot · 2025-08-14T05:24:24Z

This pull request was exported from Phabricator. Differential Revision: D80075975

…H guard triggered recompiles (pytorch#160592) Summary: We add a logging around when an ID_MATCH guard is added at a place where inbuilt_inline_nn_modules would inline it. This is done with the aim of tagging recompiles that could be avoided by setting inbuilt_inline_nn_modules flag. It will help us log and track the flag's adoption and potentially quantify saving in the the number of recompiles. Test Plan: rm -rf /tmp/r && TORCH_TRACE=/tmp/r buck2 run //scripts/ppanchalia/pt2_workspace:inline_inbuilt_nn_modules_test and CI Tests Rollback Plan: Differential Revision: D80075975

facebook-github-bot · 2025-08-14T18:27:22Z

This pull request was exported from Phabricator. Differential Revision: D80075975

anijain2305

Test is failing. But the code looks good.

…H guard triggered recompiles (pytorch#160592) Summary: We add a logging around when an ID_MATCH guard is added at a place where inbuilt_inline_nn_modules would inline it. This is done with the aim of tagging recompiles that could be avoided by setting inbuilt_inline_nn_modules flag. It will help us log and track the flag's adoption and potentially quantify saving in the the number of recompiles. Test Plan: rm -rf /tmp/r && TORCH_TRACE=/tmp/r buck2 run //scripts/ppanchalia/pt2_workspace:inline_inbuilt_nn_modules_test and CI Tests And UT added Rollback Plan: Differential Revision: D80075975

facebook-github-bot · 2025-08-14T19:08:56Z

This pull request was exported from Phabricator. Differential Revision: D80075975

facebook-github-bot · 2025-08-14T19:12:05Z

This pull request was exported from Phabricator. Differential Revision: D80075975

…H guard triggered recompiles (pytorch#160592) Summary: Pull Request resolved: pytorch#160592 We add a logging around when an ID_MATCH guard is added at a place where inbuilt_inline_nn_modules would inline it. This is done with the aim of tagging recompiles that could be avoided by setting inbuilt_inline_nn_modules flag. It will help us log and track the flag's adoption and potentially quantify saving in the the number of recompiles. Test Plan: rm -rf /tmp/r && TORCH_TRACE=/tmp/r buck2 run //scripts/ppanchalia/pt2_workspace:inline_inbuilt_nn_modules_test and CI Tests And UT added Rollback Plan: Differential Revision: D80075975

facebook-github-bot · 2025-08-14T19:12:16Z

This pull request was exported from Phabricator. Differential Revision: D80075975

…H guard triggered recompiles (pytorch#160592) Summary: We add a logging around when an ID_MATCH guard is added at a place where inbuilt_inline_nn_modules would inline it. This is done with the aim of tagging recompiles that could be avoided by setting inbuilt_inline_nn_modules flag. It will help us log and track the flag's adoption and potentially quantify saving in the the number of recompiles. Test Plan: rm -rf /tmp/r && TORCH_TRACE=/tmp/r buck2 run //scripts/ppanchalia/pt2_workspace:inline_inbuilt_nn_modules_test and CI Tests And UT added Rollback Plan: Differential Revision: D80075975

facebook-github-bot · 2025-08-14T22:01:34Z

This pull request was exported from Phabricator. Differential Revision: D80075975

…H guard triggered recompiles (pytorch#160592) Summary: Pull Request resolved: pytorch#160592 We add a logging around when an ID_MATCH guard is added at a place where inbuilt_inline_nn_modules would inline it. This is done with the aim of tagging recompiles that could be avoided by setting inbuilt_inline_nn_modules flag. It will help us log and track the flag's adoption and potentially quantify saving in the the number of recompiles. Test Plan: rm -rf /tmp/r && TORCH_TRACE=/tmp/r buck2 run //scripts/ppanchalia/pt2_workspace:inline_inbuilt_nn_modules_test and CI Tests And UT added Rollback Plan: Differential Revision: D80075975

facebook-github-bot · 2025-08-14T22:01:48Z

This pull request was exported from Phabricator. Differential Revision: D80075975

ppanchalia · 2025-08-15T17:01:34Z

@pytorchbot merge

pytorchmergebot · 2025-08-15T17:03:56Z

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

…H guard triggered recompiles (#160592) Summary: We add a logging around when an ID_MATCH guard is added at a place where inbuilt_inline_nn_modules would inline it. This is done with the aim of tagging recompiles that could be avoided by setting inbuilt_inline_nn_modules flag. It will help us log and track the flag's adoption and potentially quantify saving in the the number of recompiles. X-link: pytorch/pytorch#160592 Approved by: https://github.com/anijain2305 Reviewed By: anijain2305, clee2000 Differential Revision: D80075975 fbshipit-source-id: 21c53f015c7aed9f6cf6e5da7e2b59fbc340c476

…H guard triggered recompiles (pytorch#160592) We add a logging around when an ID_MATCH guard is added at a place where inbuilt_inline_nn_modules would inline it. This is done with the aim of tagging recompiles that could be avoided by setting inbuilt_inline_nn_modules flag. It will help us log and track the flag's adoption and potentially quantify saving in the the number of recompiles. Differential Revision: D80075975 Pull Request resolved: pytorch#160592 Approved by: https://github.com/anijain2305

pytorch-bot bot added ciflow/inductor module: dynamo labels Aug 14, 2025

facebook-github-bot added the fb-exported label Aug 14, 2025

anijain2305 reviewed Aug 14, 2025

View reviewed changes

torch/_dynamo/convert_frame.py Outdated Show resolved Hide resolved

ppanchalia force-pushed the export-D80075975 branch from 62d5b2d to bd31dd2 Compare August 14, 2025 02:26

ppanchalia force-pushed the export-D80075975 branch from bd31dd2 to 19c06fa Compare August 14, 2025 05:05

anijain2305 reviewed Aug 14, 2025

View reviewed changes

torch/_dynamo/convert_frame.py Outdated Show resolved Hide resolved

anijain2305 reviewed Aug 14, 2025

View reviewed changes

torch/_dynamo/guards.py Outdated Show resolved Hide resolved

ppanchalia force-pushed the export-D80075975 branch from 19c06fa to 63ad578 Compare August 14, 2025 05:09

anijain2305 reviewed Aug 14, 2025

View reviewed changes

torch/_dynamo/guards.py Outdated Show resolved Hide resolved

ppanchalia force-pushed the export-D80075975 branch from 63ad578 to f8c455c Compare August 14, 2025 05:14

ppanchalia force-pushed the export-D80075975 branch from f8c455c to 69528a7 Compare August 14, 2025 05:16

anijain2305 reviewed Aug 14, 2025

View reviewed changes

ppanchalia force-pushed the export-D80075975 branch from 69528a7 to eae71d3 Compare August 14, 2025 05:24

ppanchalia force-pushed the export-D80075975 branch from eae71d3 to fa04744 Compare August 14, 2025 05:29

ppanchalia self-assigned this Aug 14, 2025

anijain2305 reviewed Aug 14, 2025

View reviewed changes

ppanchalia force-pushed the export-D80075975 branch from 2668eb3 to 73f3926 Compare August 14, 2025 19:07

ppanchalia force-pushed the export-D80075975 branch from 73f3926 to 5915d4e Compare August 14, 2025 19:08

ppanchalia force-pushed the export-D80075975 branch 2 times, most recently from 96d626f to 6e68921 Compare August 14, 2025 21:57

ppanchalia force-pushed the export-D80075975 branch from 6e68921 to 338e2f3 Compare August 14, 2025 22:01

anijain2305 approved these changes Aug 14, 2025

View reviewed changes

pytorchmergebot added the merging label Aug 15, 2025

pytorchmergebot closed this in 052c441 Aug 15, 2025

pytorchmergebot added Merged and removed merging labels Aug 15, 2025

ppanchalia deleted the export-D80075975 branch August 15, 2025 17:26

Add logging for when inbuilt_inline_nn_modules will help with ID_MATCH guard triggered recompiles #160592

Add logging for when inbuilt_inline_nn_modules will help with ID_MATCH guard triggered recompiles #160592

Uh oh!

Conversation

ppanchalia commented Aug 14, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-bot bot commented Aug 14, 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/160592

✅ You can merge normally! (1 Unrelated Failure)

Uh oh!

facebook-github-bot commented Aug 14, 2025

Uh oh!

Uh oh!

facebook-github-bot commented Aug 14, 2025

Uh oh!

facebook-github-bot commented Aug 14, 2025

Uh oh!

Uh oh!

Uh oh!

Uh oh!

facebook-github-bot commented Aug 14, 2025

Uh oh!

facebook-github-bot commented Aug 14, 2025

Uh oh!

anijain2305 left a comment

Choose a reason for hiding this comment

Uh oh!

facebook-github-bot commented Aug 14, 2025

Uh oh!

facebook-github-bot commented Aug 14, 2025

Uh oh!

facebook-github-bot commented Aug 14, 2025

Uh oh!

anijain2305 left a comment

Choose a reason for hiding this comment

Uh oh!

facebook-github-bot commented Aug 14, 2025

Uh oh!

facebook-github-bot commented Aug 14, 2025

Uh oh!

facebook-github-bot commented Aug 14, 2025

Uh oh!

facebook-github-bot commented Aug 14, 2025

Uh oh!

facebook-github-bot commented Aug 14, 2025

Uh oh!

ppanchalia commented Aug 15, 2025

Uh oh!

pytorchmergebot commented Aug 15, 2025

Merge started

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

ppanchalia commented Aug 14, 2025 •

edited

Loading

pytorch-bot bot commented Aug 14, 2025 •

edited

Loading