[FSDP] enable autograd in forward prefetching #116792
Closed
Conversation
This file contains 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
**problem** when prefetching for next forward, current forward may be annotated by `@torch.no_grad`. `param.grad_fn` keeps being None during prefetching. `_post_backward_hook` never gets triggered repro ```pytest test/distributed/fsdp/test_fsdp_freezing_weights.py``` **solution** this PR enabled autograd during prefetching (`_use_unsharded_views`), so `param.grad_fn` are properly assigned for next forward a longer-term fix would be moving `_use_unsharded_views` out of `_prefetch_handle` and put it in `_pre_forward_unshard`
Summary: Test Plan: Reviewers: Subscribers: Tasks: Tags:
[FSDP] enable autograd in forward prefetching
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/116792
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit e67952f with merge base 43fb1b6 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
pytorch-bot
bot
added
the
release notes: distributed (fsdp)
github-actions
bot
added
oncall: distributed
awgu
approved these changes
Jan 4, 2024
|
@weifengpy has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
@pytorchmergebot 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 |
Skylion007
pushed a commit
to Skylion007/pytorch
that referenced
this pull request
Feb 12, 2024
**problem** when prefetching for next forward, current forward may be annotated by `@torch.no_grad`. `param.grad_fn` keeps being None during prefetching. `_post_backward_hook` never gets triggered repro ```pytest test/distributed/fsdp/test_fsdp_freezing_weights.py``` **solution** this PR enabled autograd during prefetching (`_use_unsharded_views`), so `param.grad_fn` are properly assigned for next forward a longer-term fix would be moving `_use_unsharded_views` out of `_prefetch_handle` and put it in `_pre_forward_unshard` Pull Request resolved: pytorch#116792 Approved by: https://github.com/awgu
atalman
pushed a commit
that referenced
this pull request
Feb 14, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
problem
when prefetching for next forward, current forward may be annotated by
@torch.no_grad.param.grad_fnkeeps being None during prefetching._post_backward_hooknever gets triggeredrepro
pytest test/distributed/fsdp/test_fsdp_freezing_weights.pysolution
this PR enabled autograd during prefetching (
_use_unsharded_views), soparam.grad_fnare properly assigned for next forwarda longer-term fix would be moving
_use_unsharded_viewsout of_prefetch_handleand put it in_pre_forward_unshardcc @mrshenli @pritamdamania87 @zhaojuanmao @satgera @rohan-varma @gqchen @aazzolini @osalpekar @jiayisuse @H-Huang @kwen2501 @awgu @penguinwu @fegin @XilunWu @wanchaol @fduwjj @wz337 @tianyu-l @wconstab @yf225