[FSDP] Relax post-backward assert #89791
Closed
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
[ghstack-poisoned]
awgu
requested review from
H-Huang,
kwen2501,
mrshenli,
pritamdamania87,
rohan-varma and
zhaojuanmao
as code owners
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/89791
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit a510d8d: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
pytorch-bot
bot
added
the
release notes: distributed (fsdp)
zhaojuanmao
approved these changes
Nov 28, 2022
| "FullyShardedDataParallel._post_backward_hook" | ||
| ): | ||
| _assert_in_training_states(state, [TrainingState.FORWARD_BACKWARD]) | ||
| # For reentrant AC, the post-backward hook may run multiple times in |
Contributor
There was a problem hiding this comment.
nit: For reentrant AC multiple times
This assert was accidentally made stricter when transitioning from per-FSDP-instance training state to per-handle training state. This PR relaxes it again, which should restore compatibility for some reentrant AC plus FSDP cases. [ghstack-poisoned]
This assert was accidentally made stricter when transitioning from per-FSDP-instance training state to per-handle training state. This PR relaxes it again, which should restore compatibility for some reentrant AC plus FSDP cases. [ghstack-poisoned]
Collaborator
Author
|
@pytorchbot rebase -s |
Collaborator
|
@pytorchbot successfully started a rebase job. Check the current status here |
This assert was accidentally made stricter when transitioning from per-FSDP-instance training state to per-handle training state. This PR relaxes it again, which should restore compatibility for some reentrant AC plus FSDP cases. [ghstack-poisoned]
Collaborator
|
Successfully rebased |
Collaborator
Author
|
@pytorchbot merge |
Collaborator
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 |
mrshenli
added a commit
that referenced
this pull request
Nov 29, 2022
With combining FSDP with reentrant checkpointing, the post backward hook might run twice, and then hit [this error](https://github.com/pytorch/pytorch/blob/e20ec44544c17d6d3d411f88b870e05043bda731/torch/distributed/fsdp/_runtime_utils.py#L487). This is because reentrant backward uses nested autograd GraphTasks. The inner GraphTask is not aware of the outer one and therefore will flush pending `AccumulateGrad` invocations on exit, which in turn triggers the post backward hooks registered by FSDP. Later, the outer GraphTask will trigger that again, leading to the above error. PR #89791 relaxes the FSDP training state check, but we still run into grad value check failures occasionally. Therefore, this PR only lands the test for non-reentrant test, and we can enable the reentrant test when the accuracy issues are addressed. [ghstack-poisoned]
mrshenli
added a commit
that referenced
this pull request
Nov 29, 2022
With combining FSDP with reentrant checkpointing, the post backward hook might run twice, and then hit [this error](https://github.com/pytorch/pytorch/blob/e20ec44544c17d6d3d411f88b870e05043bda731/torch/distributed/fsdp/_runtime_utils.py#L487). This is because reentrant backward uses nested autograd GraphTasks. The inner GraphTask is not aware of the outer one and therefore will flush pending `AccumulateGrad` invocations on exit, which in turn triggers the post backward hooks registered by FSDP. Later, the outer GraphTask will trigger that again, leading to the above error. PR #89791 relaxes the FSDP training state check, but we still run into grad value check failures occasionally. Therefore, this PR only lands the test for non-reentrant test, and we can enable the reentrant test when the accuracy issues are addressed. ghstack-source-id: 8848c4c Pull Request resolved: #89781
Collaborator
Merge failedReason: 1 additional jobs have failed, first few of them are: trunk Details for Dev Infra teamRaised by workflow job |
pytorchmergebot
pushed a commit
that referenced
this pull request
Nov 29, 2022
With combining FSDP with reentrant checkpointing, the post backward hook might run twice, and then hit [this error](https://github.com/pytorch/pytorch/blob/e20ec44544c17d6d3d411f88b870e05043bda731/torch/distributed/fsdp/_runtime_utils.py#L487). This is because reentrant backward uses nested autograd GraphTasks. The inner GraphTask is not aware of the outer one and therefore will flush pending `AccumulateGrad` invocations on exit, which in turn triggers the post backward hooks registered by FSDP. Later, the outer GraphTask will trigger that again, leading to the above error. PR #89791 relaxes the FSDP training state check, but we still run into grad value check failures occasionally. Therefore, this PR only lands the test for non-reentrant test, and we can enable the reentrant test when the accuracy issues are addressed. Pull Request resolved: #89781 Approved by: https://github.com/rohan-varma
Collaborator
Author
|
@pytorchbot merge |
Collaborator
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 |
kulinseth
pushed a commit
to kulinseth/pytorch
that referenced
this pull request
Dec 10, 2022
With combining FSDP with reentrant checkpointing, the post backward hook might run twice, and then hit [this error](https://github.com/pytorch/pytorch/blob/e20ec44544c17d6d3d411f88b870e05043bda731/torch/distributed/fsdp/_runtime_utils.py#L487). This is because reentrant backward uses nested autograd GraphTasks. The inner GraphTask is not aware of the outer one and therefore will flush pending `AccumulateGrad` invocations on exit, which in turn triggers the post backward hooks registered by FSDP. Later, the outer GraphTask will trigger that again, leading to the above error. PR pytorch#89791 relaxes the FSDP training state check, but we still run into grad value check failures occasionally. Therefore, this PR only lands the test for non-reentrant test, and we can enable the reentrant test when the accuracy issues are addressed. Pull Request resolved: pytorch#89781 Approved by: https://github.com/rohan-varma
kulinseth
pushed a commit
to kulinseth/pytorch
that referenced
this pull request
Dec 10, 2022
This assert was accidentally made stricter when transitioning from per-FSDP-instance training state to per-handle training state. This PR relaxes it again, which should restore compatibility for some reentrant AC plus FSDP cases. Pull Request resolved: pytorch#89791 Approved by: https://github.com/zhaojuanmao
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.
Stack from ghstack (oldest at bottom):
This assert was accidentally made stricter when transitioning from per-FSDP-instance training state to per-handle training state. This PR relaxes it again, which should restore compatibility for some reentrant AC plus FSDP cases.