-
Notifications
You must be signed in to change notification settings - Fork 25.7k
[AOTInductor] Allow using ProxyExecutor for ATen fallbacks #112976
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/112976
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (3 Unrelated Failures)As of commit c3a6133 with merge base 24b61a4 ( 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. |
|
This pull request was exported from Phabricator. Differential Revision: D51005807 |
This PR needs a
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Fix the failing tests?
05da94e to
23d6544
Compare
|
This pull request was exported from Phabricator. Differential Revision: D51005807 |
23d6544 to
5b06d71
Compare
|
This pull request was exported from Phabricator. Differential Revision: D51005807 |
5b06d71 to
e144a3c
Compare
|
This pull request was exported from Phabricator. Differential Revision: D51005807 |
…12976) Summary: Use ProxyExecutor for aten._scaled_dot_product_efficient_attention in ABI-mode Test Plan: OSS CI buck2 run mode/dev-nosan deeplearning/aot_inductor/test:test_custom_ops Reviewed By: chenyang78, amitaga Differential Revision: D51005807
e144a3c to
c3a6133
Compare
|
This pull request was exported from Phabricator. Differential Revision: D51005807 |
| else f"at::_ops::{kernel.__name__.replace('.', '_')}::call" | ||
| ) | ||
| schema = kernel._schema | ||
| if config.is_fbcode() and kernel in fbcode_use_proxy_executor: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why do we need to diverge behavior between OSS and fbcode here? It would seem much better to do the same thing in both.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
proxy executor doesn't work in OSS yet.
In fbcode, we use thrift for serialization/deserialization. However, in OSS, we don't have a useable serialization/deserialization solution.
export uses json.dumps for serialization and json.loads for deserialization back to python. However, this format is not very friendly for deserialization back to C++.
|
@pytorchbot merge (Initiating merge automatically since Phabricator Diff has merged) |
Merge startedYour 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 |
…12976) Summary: Use ProxyExecutor for aten._scaled_dot_product_efficient_attention in ABI-mode Test Plan: OSS CI Differential Revision: D51005807 Pull Request resolved: pytorch#112976 Approved by: https://github.com/chenyang78, https://github.com/jansel
Summary: Use ProxyExecutor for aten._scaled_dot_product_efficient_attention in ABI-mode
Test Plan: OSS CI
Differential Revision: D51005807
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @peterbell10 @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @aakhundov @ColinPeppler