propagate debug handle from edge dialect graph back to exported graph #12330
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
Using exported graph from torch.export as source of truth for aot intermediate output is our target. Once thing blocking us is the exported graph does not have debug handle, which will not show up in the export flow until DebugHandleGenerationPass, the last step of to_edge(). We need to equip the graph with same debug handle used in ExecuTorch flow. This diff creates a propagate_back_debug_handle function, which propagate debug handle from edge dialect program back to the exported program while maintain the correctness of operator tracing. Differential Revision: [D78051614](https://our.internmc.facebook.com/intern/diff/D78051614/) [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/12330
Note: Links to docs will display an error until the docs builds have been completed. ❌ 4 New Failures, 3 Unrelated FailuresAs of commit effafaa with merge base dd06b3b ( NEW FAILURES - The following jobs have failed:
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. |
Gasoonjia
added a commit
that referenced
this pull request
Jul 10, 2025
Using exported graph from torch.export as source of truth for aot intermediate output is our target. Once thing blocking us is the exported graph does not have debug handle, which will not show up in the export flow until DebugHandleGenerationPass, the last step of to_edge(). We need to equip the graph with same debug handle used in ExecuTorch flow. This diff creates a propagate_back_debug_handle function, which propagate debug handle from edge dialect program back to the exported program while maintain the correctness of operator tracing. Differential Revision: [D78051614](https://our.internmc.facebook.com/intern/diff/D78051614/) ghstack-source-id: 295264711 Pull Request resolved: #12330
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D78051614 |
larryliu0820
approved these changes
Jul 10, 2025
Gasoonjia
added
release notes: none
…orted graph" Using exported graph from torch.export as source of truth for aot intermediate output is our target. Once thing blocking us is the exported graph does not have debug handle, which will not show up in the export flow until DebugHandleGenerationPass, the last step of to_edge(). We need to equip the graph with same debug handle used in ExecuTorch flow. This diff creates a propagate_back_debug_handle function, which propagate debug handle from edge dialect program back to the exported program while maintain the correctness of operator tracing. Differential Revision: [D78051614](https://our.internmc.facebook.com/intern/diff/D78051614/) [ghstack-poisoned]
Gasoonjia
added a commit
that referenced
this pull request
Jul 10, 2025
Pull Request resolved: #12330 Using exported graph from torch.export as source of truth for aot intermediate output is our target. Once thing blocking us is the exported graph does not have debug handle, which will not show up in the export flow until DebugHandleGenerationPass, the last step of to_edge(). We need to equip the graph with same debug handle used in ExecuTorch flow. This diff creates a propagate_back_debug_handle function, which propagate debug handle from edge dialect program back to the exported program while maintain the correctness of operator tracing. ghstack-source-id: 295337600 @exported-using-ghexport Differential Revision: [D78051614](https://our.internmc.facebook.com/intern/diff/D78051614/)
|
This pull request was exported from Phabricator. Differential Revision: D78051614 |
facebook-github-bot
merged commit c18b49d
into
gh/gasoonjia/19/base
90 of 99 checks passed
facebook-github-bot
temporarily deployed
to
cherry-pick-bot
GitHub Actions
Inactive
Gasoonjia
added a commit
that referenced
this pull request
Jul 10, 2025
…#12337) This PR was created by the merge bot to help merge the original PR into the main branch. ghstack PR number: #12330 by @Gasoonjia ^ Please use this as the source of truth for the PR details, comments, and reviews ghstack PR base: https://github.com/pytorch/executorch/tree/gh/gasoonjia/19/base ghstack PR head: https://github.com/pytorch/executorch/tree/gh/gasoonjia/19/head Merge bot PR base: https://github.com/pytorch/executorch/tree/gh/gasoonjia/18/orig Merge bot PR head: https://github.com/pytorch/executorch/tree/gh/gasoonjia/19/orig @diff-train-skip-merge --------- Co-authored-by: gasoonjia <[email protected]>
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.
Stack from ghstack (oldest at bottom):
Using exported graph from torch.export as source of truth for aot intermediate output is our target. Once thing blocking us is the exported graph does not have debug handle, which will not show up in the export flow until DebugHandleGenerationPass, the last step of to_edge(). We need to equip the graph with same debug handle used in ExecuTorch flow.
This diff creates a propagate_back_debug_handle function, which propagate debug handle from edge dialect program back to the exported program while maintain the correctness of operator tracing.
Differential Revision: D78051614