[dtensor] to_local backward grad placement passthrough #121474
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
to_local accepts a `grad_placements` if user choose to pass, previously we enforce the grad_out to be the "same" placement as the current DTensor for safety. But I realized that we DO NOT need to enforce this constraint. Why? backward placement does not need to be the same as fwd tensor placement, this is already the case for param vs param.grad (i.e. param can be replicate and grad can be partial), so we should not restrict this to activation vs activation grad too [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/121474
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (1 Unrelated Failure)As of commit fa58f6b with merge base 3d089de ( BROKEN TRUNK - The following job 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. |
wanchaol
added a commit
that referenced
this pull request
Mar 8, 2024
to_local accepts a `grad_placements` if user choose to pass, previously we enforce the grad_out to be the "same" placement as the current DTensor for safety. But I realized that we DO NOT need to enforce this constraint. Why? backward placement does not need to be the same as fwd tensor placement, this is already the case for param vs param.grad (i.e. param can be replicate and grad can be partial), so we should not restrict this to activation vs activation grad too ghstack-source-id: bdc6344 Pull Request resolved: #121474
github-actions
bot
added
oncall: distributed
|
@yoyoyocmu @xw285cornell This PR would directly produce reduce_scatter instead of allreduce + chunk, so that we don't need to add fusion pass to fuse this pattern in #120051 |
wanchaol
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 |
pianpwk
pushed a commit
that referenced
this pull request
Mar 11, 2024
to_local accepts a `grad_placements` if user choose to pass, previously we enforce the grad_out to be the "same" placement as the current DTensor for safety. But I realized that we DO NOT need to enforce this constraint. Why? backward placement does not need to be the same as fwd tensor placement, this is already the case for param vs param.grad (i.e. param can be replicate and grad can be partial), so we should not restrict this to activation vs activation grad too Pull Request resolved: #121474 Approved by: https://github.com/awgu, https://github.com/yoyoyocmu, https://github.com/yifuwang
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 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.
Stack from ghstack (oldest at bottom):
to_local accepts a
grad_placementsif user choose to pass, previouslywe enforce the grad_out to be the "same" placement as the current
DTensor for safety.
But I realized that we DO NOT need to enforce this constraint. Why?
backward placement does not need to be the same as fwd tensor placement, this
is already the case for param vs param.grad (i.e. param can be replicate
and grad can be partial), so we should not restrict this to activation
vs activation grad too
cc @mrshenli @pritamdamania87 @zhaojuanmao @satgera @rohan-varma @gqchen @aazzolini @osalpekar @jiayisuse @H-Huang @kwen2501 @awgu @penguinwu @fegin @XilunWu @fduwjj @wz337 @tianyu-l @wconstab @yf225 @chauhang