Disable FlexAttention max-autotune when deterministic is used #1808
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
With max-autotune, FlexAttention is not deterministic even if torch.use_deterministic_algorithms is True. Add a flag to disable max-autotune. Since this this is a debug flag, we use environment variables to implement this flag. DISABLE_TORCHTITAN_FLEX_MAX_AUTOTUNE is the flag.
meta-cla
bot
added
the
CLA Signed
wwwjn
reviewed
Oct 7, 2025
fegin
changed the title
githubsgi
pushed a commit
to githubsgi/torchtitan
that referenced
this pull request
Oct 13, 2025
…h#1808) With max-autotune, FlexAttention is not deterministic even if torch.use_deterministic_algorithms is True. When deterministic mode is set, we should also remove the usage of `max-autotune`.
githubsgi
pushed a commit
to githubsgi/torchtitan
that referenced
this pull request
Oct 15, 2025
…h#1808) With max-autotune, FlexAttention is not deterministic even if torch.use_deterministic_algorithms is True. When deterministic mode is set, we should also remove the usage of `max-autotune`.
githubsgi
pushed a commit
to githubsgi/torchtitan
that referenced
this pull request
Oct 16, 2025
…h#1808) With max-autotune, FlexAttention is not deterministic even if torch.use_deterministic_algorithms is True. When deterministic mode is set, we should also remove the usage of `max-autotune`.
githubsgi
pushed a commit
to githubsgi/torchtitan
that referenced
this pull request
Oct 16, 2025
…h#1808) With max-autotune, FlexAttention is not deterministic even if torch.use_deterministic_algorithms is True. When deterministic mode is set, we should also remove the usage of `max-autotune`.
githubsgi
pushed a commit
to githubsgi/torchtitan
that referenced
this pull request
Oct 29, 2025
…h#1808) With max-autotune, FlexAttention is not deterministic even if torch.use_deterministic_algorithms is True. When deterministic mode is set, we should also remove the usage of `max-autotune`.
githubsgi
pushed a commit
to githubsgi/torchtitan
that referenced
this pull request
Oct 29, 2025
…h#1808) With max-autotune, FlexAttention is not deterministic even if torch.use_deterministic_algorithms is True. When deterministic mode is set, we should also remove the usage of `max-autotune`.
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.
With max-autotune, FlexAttention is not deterministic even if torch.use_deterministic_algorithms is True. When deterministic mode is set, we should also remove the usage of
max-autotune.