Inline all scalars by default in export path#6803
Merged
Conversation
qihqi
approved these changes
Mar 22, 2024
JackCaoG
approved these changes
Mar 22, 2024
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Motivation of the change:
Currently only special scalars (0 or 1) will be inlined in the lowered HLO/StableHLO. In export path with HLFB, lifted scalars generated during tracing would result in additional arg in the
stablehlo.composite. This makes downstream compiler confusing when mapping the args in the model code andstablehlo.composite.Inlining scalars should work well in the export path. Because
torch.exportdoesn't trace non-tensor input at this moment, all non-tensor values are burned in the exported FX graph (with dangling placeholders).Change:
at::Scalar.Test:
Added new tests cases.