Fix FSDP+DP Transformer Engine error#344
Merged
copybara-service[bot] merged 1 commit intoAI-Hypercomputer:mainfrom Mar 3, 2026
Merged
Fix FSDP+DP Transformer Engine error#344copybara-service[bot] merged 1 commit intoAI-Hypercomputer:mainfrom
copybara-service[bot] merged 1 commit intoAI-Hypercomputer:mainfrom
Conversation
entrpn
approved these changes
Mar 3, 2026
copybara-service
bot
merged commit 68e0696
into
AI-Hypercomputer:main
26 of 27 checks passed
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR fixes a bug that prevents the use of FSDP and DP simultaneously. This is done by not passing the DP parallelism axis to the transformer engine mesh resource. I believe this restriction is in place for the transformer engine MultiHeadAttention class and does not apply to the DotProductAttention class used in this repo. The code has been tested on flux and wan training/inference.
This PR also applies the transformer_engine_context to most of the training/generate scripts and cleans up some previous code.