[effects] Fix backwards handling #122346
[effects] Fix backwards handling #122346
Conversation
[ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/122346
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit c775fed with merge base 16935de ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
| def runtime_wrapper(*args): | ||
| # Pass in effect tokens (See Note [Side-Effectful Tokens in AOTAutograd]) | ||
| args = (*[torch.tensor([])] * num_tokens, *args) | ||
| if num_tokens > 0: |
There was a problem hiding this comment.
@zhuhaozhe addressing #121882 here
I didn't previously test the `.backwards()` call, and when testing on #122348 I realized we were missing some token handling in some places. [ghstack-poisoned]
|
@pytorchbot merge |
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 |
Merge failedReason: 3 jobs have failed, first few of them are: .github/workflows/trunk.yml, .github/workflows/generated-linux-binary-libtorch-cxx11-abi-main.yml, .github/workflows/generated-linux-binary-libtorch-pre-cxx11-main.yml Details for Dev Infra teamRaised by workflow job |
angelayi
added
ciflow/trunk
|
@pytorchbot merge |
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 |
I didn't previously test the `.backwards()` call, and when testing on #122348 I realized we were missing some token handling in some places. Pull Request resolved: #122346 Approved by: https://github.com/zou3519
3 participants
Stack from ghstack (oldest at bottom):
I didn't previously test the
.backwards()call, and when testing on #122348 I realized we were missing some token handling in some places.