Backend data separation test#10531
Merged
Merged
Conversation
Add backend data separation test with demo ExecutorBackend. Note: ExecutorBackend is a wrapper around a portable PTE file. It doesn't support delegated program-data separation in the sense of tagging tensors as external and placing them into the named data store as such. Rather, this test: 1. Creates a linear PTE file with data removed, using the portable flow. 2. Packages (1) into preprocessed blob as a delegate, using ExecutorBackend. Note: this discards the PTD portion. 3. Re-create the PTD portion via export_program (portable flow again). 4. Runs the delegated ExecutorBackend linear with portable linear.ptd file. Caveat: this means that LinearModule definition in export_program and export_delegated_program must stay in sync, as we get the PTE artifact from export_delegated_program (wrapped in ExecutorBackend), and the PTD artifact from export_program. Differential Revision: [D73679733](https://our.internmc.facebook.com/intern/diff/D73679733/) [ghstack-poisoned]
lucylq
requested review from
JacobSzwejbka,
kirklandsign,
larryliu0820,
swolchok and
tarun292
as code owners
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/10531
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: ✅ No FailuresAs of commit 61a7b79 with merge base de98d25 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
This was referenced Apr 28, 2025
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D73679733 |
Add backend data separation test with demo ExecutorBackend. Note: ExecutorBackend is a wrapper around a portable PTE file. It doesn't support delegated program-data separation in the sense of tagging tensors as external and placing them into the named data store as such. Rather, this test: 1. Creates a linear PTE file with data removed, using the portable flow. 2. Packages (1) into preprocessed blob as a delegate, using ExecutorBackend. Note: this discards the PTD portion. 3. Re-create the PTD portion via export_program (portable flow again). 4. Runs the delegated ExecutorBackend linear with portable linear.ptd file. Caveat: this means that LinearModule definition in export_program and export_delegated_program must stay in sync, as we get the PTE artifact from export_delegated_program (wrapped in ExecutorBackend), and the PTD artifact from export_program. Differential Revision: [D73679733](https://our.internmc.facebook.com/intern/diff/D73679733/) [ghstack-poisoned]
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D73679733 |
Add backend data separation test with demo ExecutorBackend. Note: ExecutorBackend is a wrapper around a portable PTE file. It doesn't support delegated program-data separation in the sense of tagging tensors as external and placing them into the named data store as such. Rather, this test: 1. Creates a linear PTE file with data removed, using the portable flow. 2. Packages (1) into preprocessed blob as a delegate, using ExecutorBackend. Note: this discards the PTD portion. 3. Re-create the PTD portion via export_program (portable flow again). 4. Runs the delegated ExecutorBackend linear with portable linear.ptd file. Caveat: this means that LinearModule definition in export_program and export_delegated_program must stay in sync, as we get the PTE artifact from export_delegated_program (wrapped in ExecutorBackend), and the PTD artifact from export_program. Differential Revision: [D73679733](https://our.internmc.facebook.com/intern/diff/D73679733/) [ghstack-poisoned]
lucylq
added a commit
that referenced
this pull request
May 5, 2025
Pull Request resolved: #10531 Add backend data separation test with demo ExecutorBackend. Note: ExecutorBackend is a wrapper around a portable PTE file. It doesn't support delegated program-data separation in the sense of tagging tensors as external and placing them into the named data store as such. Rather, this test: 1. Creates a linear PTE file with data removed, using the portable flow. 2. Packages (1) into preprocessed blob as a delegate, using ExecutorBackend. Note: this discards the PTD portion. 3. Re-create the PTD portion via export_program (portable flow again). 4. Runs the delegated ExecutorBackend linear with portable linear.ptd file. Caveat: this means that LinearModule definition in export_program and export_delegated_program must stay in sync, as we get the PTE artifact from export_delegated_program (wrapped in ExecutorBackend), and the PTD artifact from export_program. ghstack-source-id: 282158282 @exported-using-ghexport Differential Revision: [D73679733](https://our.internmc.facebook.com/intern/diff/D73679733/)
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D73679733 |
mcr229
approved these changes
May 6, 2025
Add backend data separation test with demo ExecutorBackend. Note: ExecutorBackend is a wrapper around a portable PTE file. It doesn't support delegated program-data separation in the sense of tagging tensors as external and placing them into the named data store as such. Rather, this test: 1. Creates a linear PTE file with data removed, using the portable flow. 2. Packages (1) into preprocessed blob as a delegate, using ExecutorBackend. Note: this discards the PTD portion. 3. Re-create the PTD portion via export_program (portable flow again). 4. Runs the delegated ExecutorBackend linear with portable linear.ptd file. Caveat: this means that LinearModule definition in export_program and export_delegated_program must stay in sync, as we get the PTE artifact from export_delegated_program (wrapped in ExecutorBackend), and the PTD artifact from export_program. Differential Revision: [D73679733](https://our.internmc.facebook.com/intern/diff/D73679733/) [ghstack-poisoned]
lucylq
added a commit
that referenced
this pull request
May 6, 2025
Pull Request resolved: #10531 Add backend data separation test with demo ExecutorBackend. Note: ExecutorBackend is a wrapper around a portable PTE file. It doesn't support delegated program-data separation in the sense of tagging tensors as external and placing them into the named data store as such. Rather, this test: 1. Creates a linear PTE file with data removed, using the portable flow. 2. Packages (1) into preprocessed blob as a delegate, using ExecutorBackend. Note: this discards the PTD portion. 3. Re-create the PTD portion via export_program (portable flow again). 4. Runs the delegated ExecutorBackend linear with portable linear.ptd file. Caveat: this means that LinearModule definition in export_program and export_delegated_program must stay in sync, as we get the PTE artifact from export_delegated_program (wrapped in ExecutorBackend), and the PTD artifact from export_program. ghstack-source-id: 282178788 @exported-using-ghexport Differential Revision: [D73679733](https://our.internmc.facebook.com/intern/diff/D73679733/)
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D73679733 |
Add backend data separation test with demo ExecutorBackend. Note: ExecutorBackend is a wrapper around a portable PTE file. It doesn't support delegated program-data separation in the sense of tagging tensors as external and placing them into the named data store as such. Rather, this test: 1. Creates a linear PTE file with data removed, using the portable flow. 2. Packages (1) into preprocessed blob as a delegate, using ExecutorBackend. Note: this discards the PTD portion. 3. Re-create the PTD portion via export_program (portable flow again). 4. Runs the delegated ExecutorBackend linear with portable linear.ptd file. Caveat: this means that LinearModule definition in export_program and export_delegated_program must stay in sync, as we get the PTE artifact from export_delegated_program (wrapped in ExecutorBackend), and the PTD artifact from export_program. Differential Revision: [D73679733](https://our.internmc.facebook.com/intern/diff/D73679733/) [ghstack-poisoned]
lucylq
added a commit
that referenced
this pull request
May 6, 2025
Pull Request resolved: #10531 Add backend data separation test with demo ExecutorBackend. Note: ExecutorBackend is a wrapper around a portable PTE file. It doesn't support delegated program-data separation in the sense of tagging tensors as external and placing them into the named data store as such. Rather, this test: 1. Creates a linear PTE file with data removed, using the portable flow. 2. Packages (1) into preprocessed blob as a delegate, using ExecutorBackend. Note: this discards the PTD portion. 3. Re-create the PTD portion via export_program (portable flow again). 4. Runs the delegated ExecutorBackend linear with portable linear.ptd file. Caveat: this means that LinearModule definition in export_program and export_delegated_program must stay in sync, as we get the PTE artifact from export_delegated_program (wrapped in ExecutorBackend), and the PTD artifact from export_program. ghstack-source-id: 282187684 @exported-using-ghexport Differential Revision: [D73679733](https://our.internmc.facebook.com/intern/diff/D73679733/)
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D73679733 |
facebook-github-bot
had a problem deploying
to
cherry-pick-bot
GitHub Actions
Failure
lucylq
added a commit
that referenced
this pull request
May 6, 2025
Pull Request resolved: #10531 Add backend data separation test with demo ExecutorBackend. Note: ExecutorBackend is a wrapper around a portable PTE file. It doesn't support delegated program-data separation in the sense of tagging tensors as external and placing them into the named data store as such. Rather, this test: 1. Creates a linear PTE file with data removed, using the portable flow. 2. Packages (1) into preprocessed blob as a delegate, using ExecutorBackend. Note: this discards the PTD portion. 3. Re-create the PTD portion via export_program (portable flow again). 4. Runs the delegated ExecutorBackend linear with portable linear.ptd file. Caveat: this means that LinearModule definition in export_program and export_delegated_program must stay in sync, as we get the PTE artifact from export_delegated_program (wrapped in ExecutorBackend), and the PTD artifact from export_program. ghstack-source-id: 282187684 @exported-using-ghexport Differential Revision: [D73679733](https://our.internmc.facebook.com/intern/diff/D73679733/)
lucylq
added a commit
that referenced
this pull request
May 6, 2025
Pull Request resolved: #10531 Add backend data separation test with demo ExecutorBackend. Note: ExecutorBackend is a wrapper around a portable PTE file. It doesn't support delegated program-data separation in the sense of tagging tensors as external and placing them into the named data store as such. Rather, this test: 1. Creates a linear PTE file with data removed, using the portable flow. 2. Packages (1) into preprocessed blob as a delegate, using ExecutorBackend. Note: this discards the PTD portion. 3. Re-create the PTD portion via export_program (portable flow again). 4. Runs the delegated ExecutorBackend linear with portable linear.ptd file. Caveat: this means that LinearModule definition in export_program and export_delegated_program must stay in sync, as we get the PTE artifact from export_delegated_program (wrapped in ExecutorBackend), and the PTD artifact from export_program. ghstack-source-id: 282187684 @exported-using-ghexport Differential Revision: [D73679733](https://our.internmc.facebook.com/intern/diff/D73679733/) Backend data separation test Add backend data separation test with demo ExecutorBackend. Note: ExecutorBackend is a wrapper around a portable PTE file. It doesn't support delegated program-data separation in the sense of tagging tensors as external and placing them into the named data store as such. Rather, this test: 1. Creates a linear PTE file with data removed, using the portable flow. 2. Packages (1) into preprocessed blob as a delegate, using ExecutorBackend. Note: this discards the PTD portion. 3. Re-create the PTD portion via export_program (portable flow again). 4. Runs the delegated ExecutorBackend linear with portable linear.ptd file. Caveat: this means that LinearModule definition in export_program and export_delegated_program must stay in sync, as we get the PTE artifact from export_delegated_program (wrapped in ExecutorBackend), and the PTD artifact from export_program. Differential Revision: [D73679733](https://our.internmc.facebook.com/intern/diff/D73679733/)
phaiting
pushed a commit
that referenced
this pull request
May 6, 2025
Pull Request resolved: #10531 Add backend data separation test with demo ExecutorBackend. Note: ExecutorBackend is a wrapper around a portable PTE file. It doesn't support delegated program-data separation in the sense of tagging tensors as external and placing them into the named data store as such. Rather, this test: 1. Creates a linear PTE file with data removed, using the portable flow. 2. Packages (1) into preprocessed blob as a delegate, using ExecutorBackend. Note: this discards the PTD portion. 3. Re-create the PTD portion via export_program (portable flow again). 4. Runs the delegated ExecutorBackend linear with portable linear.ptd file. Caveat: this means that LinearModule definition in export_program and export_delegated_program must stay in sync, as we get the PTE artifact from export_delegated_program (wrapped in ExecutorBackend), and the PTD artifact from export_program. ghstack-source-id: 282187684 @exported-using-ghexport Differential Revision: [D73679733](https://our.internmc.facebook.com/intern/diff/D73679733/) Backend data separation test Add backend data separation test with demo ExecutorBackend. Note: ExecutorBackend is a wrapper around a portable PTE file. It doesn't support delegated program-data separation in the sense of tagging tensors as external and placing them into the named data store as such. Rather, this test: 1. Creates a linear PTE file with data removed, using the portable flow. 2. Packages (1) into preprocessed blob as a delegate, using ExecutorBackend. Note: this discards the PTD portion. 3. Re-create the PTD portion via export_program (portable flow again). 4. Runs the delegated ExecutorBackend linear with portable linear.ptd file. Caveat: this means that LinearModule definition in export_program and export_delegated_program must stay in sync, as we get the PTE artifact from export_delegated_program (wrapped in ExecutorBackend), and the PTD artifact from export_program. Differential Revision: [D73679733](https://our.internmc.facebook.com/intern/diff/D73679733/)
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.
Stack from ghstack (oldest at bottom):
Add backend data separation test with demo ExecutorBackend.
Note: ExecutorBackend is a wrapper around a portable PTE file. It doesn't support delegated program-data separation in the sense of tagging tensors as external and placing them into the named data store as such. Rather, this test:
Caveat: this means that LinearModule definition in export_program and export_delegated_program must stay in sync, as we get the PTE artifact from export_delegated_program (wrapped in ExecutorBackend), and the PTD artifact from export_program.
Differential Revision: D73679733