Fast forward Orig/release/1.0 to viable strict #14341
Merged
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
Differential Revision: D81706654 Pull Request resolved: pytorch#13957
Differential Revision: D82280438 Pull Request resolved: pytorch#14247
As titled, rely on CI jobs
Swaps Llava export arg order, such that forward takes `embeddings, cache_position`, instead of `cache_position, embeddings`.
Update according to pytorch#14275
Summary: As title, moving to torchao Reviewed By: jerryzh168 Differential Revision: D82341472
Co-authored-by: Digant Desai <[email protected]>
Co-authored-by: Digant Desai <[email protected]>
Co-authored-by: Digant Desai <[email protected]>
Differential Revision: D79381533 Pull Request resolved: pytorch#14229
Differential Revision: D82265586 Pull Request resolved: pytorch#14241
Differential Revision: D82345075 Pull Request resolved: pytorch#14287
Differential Revision: D82160616 Pull Request resolved: pytorch#14189
### Summary Updated README and remove NeuronAdapter.h
Differential Revision: D81950579 Pull Request resolved: pytorch#14093
Differential Revision: D81951110 Pull Request resolved: pytorch#14094
Differential Revision: D81973532 Pull Request resolved: pytorch#14095
Tag name for MLSDK fetching was incorrect. cc @digantdesai @freddan80 @per @zingo @oscarandersson8218
…uilder (pytorch#14163) ### Summary Move optimization in keep_one_empty_buffer.py to IR Model builder. This functionality is not an optimization, but a required housekeeping step to build a valid IR model for Neutron Converter. ### Test plan All tests where a subgraph is delegated to Neutron.
This PR enables the visualisation of a tosa flatbuffer file in the arm `run.sh` script. The `model_explorer` command line flag is added which launches model explorer, with the tosa adapter plugin - https://github.com/Arm/tosa-adapter-model-explorer. cc @digantdesai @freddan80 @per @zingo @oscarandersson8218 Co-authored-by: Robert Taylor <[email protected]>
Causes the script to print commands it's executing which should make debugging easier
Summary: use expand decorator in test type dispatches Differential Revision: D82239681
Differential Revision: D82174198 Pull Request resolved: pytorch#14273
### Summary Remove NeuronAdapter.h Please don't update this branch if possible to prevent it from getting added back again, thanks.
Differential Revision: D80513725 Pull Request resolved: pytorch#13802 pytorch#13802 (comment) failed to cp to main Co-authored-by: Nitin Jain <[email protected]>
Fix failures in [D82358803](https://www.internalfb.com/diff/D82358803)
Differential Revision: D82109615 Pull Request resolved: pytorch#14140
…h#14307) …sing only performance cores (pytorch#14090)" This reverts commit 72d50b2. ### Summary Seeing crashes in macos unittest job.
Differential Revision: D82454608 Pull Request resolved: pytorch#14143
Temporarily disable the OpenVINO CI job due to consistent timeouts. See pytorch#14314 for the tracking task, which has more details. It is intended that this job is fixed and re-enabled quickly.
…ytorch#14225) For voxtral and phi-3, we construct the cache_position_tensor like before; for llava, it will construct underneath so we pass in size 1.
…ytorch#14070) Summary: update the codebase to allow a system whereby: - the test inventory in https://www.internalfb.com/code/fbsource/fbcode/executorch/backends/arm/test/ops/, which until now runs on open source CI can all be exercised internally on our internal diff-time sandcastle and CI runs - those tests can **transparently** switch between running an internal version of FVP using Meta's FVP runtimes, build configs and artifacts when run in fbsrouce, but then use the open-source FVp runtime when executed in github opens-ource CI Differential Revision: D81159036
…t in ET (pytorch#13989) These tests are failing for all backends due to lack of defunctionalization support for in-place activations. This PR disables them to reduce noise. They can be re-enabled in the future when supported. Tracking in pytorch#11700.
abhinaykukkadapu
requested review from
GregoryComer,
JacobSzwejbka,
cccclai,
digantdesai,
jackzhxng,
kimishpatel,
kirklandsign,
larryliu0820,
lucylq,
manuelcandales,
mergennachin and
swolchok
as code owners
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/14341
Note: Links to docs will display an error until the docs builds have been completed. ⏳ No Failures, 53 PendingAs of commit e0dda90 with merge base e0dda90 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
bot
added
the
CLA Signed
GregoryComer
approved these changes
Sep 16, 2025
abhinaykukkadapu
merged commit e0dda90
into
pytorch:orig/release/1.0
875 of 885 checks passed
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
20 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.
Fast forward Orig/release/1.0 to viable strict as of Tuesday Sept 16 11 AM PST