Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[HAL] Add a pass to materialize homogeneous encodings. #14724

Closed
wants to merge 5 commits into from

Conversation

hanhanW
Copy link
Contributor

@hanhanW hanhanW commented Aug 17, 2023

  • Adds buildMaterializeEncodingsPassPipeline method to TargetBackend. In this revision, only llvm-cpu backend implements the method.

- Adds buildMaterializeEncodingsPassPipeline method to TargetBackend. In
  this revision, only llvm-cpu backend implements the method.
- Updates CPU upper_bound_tile_size materialization be only accepting
  single executable target.
Copy link
Collaborator

@benvanik benvanik left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't understand what this is for - you're running a pass on the host module that should have nothing you can change - no linalg ops, no tensors, no nothing - at the head of the HAL pipeline where this is running there should only be hal.executables, and the next pass to run is translate executables and that lets you run your backend-specific passes on your backend-specific IR.

@hanhanW hanhanW marked this pull request as draft August 22, 2023 23:47
@hanhanW
Copy link
Contributor Author

hanhanW commented Aug 23, 2023

I moved the pass to GlobalOptimization phase, so it no longer needs to update HAL things.

#14792

@hanhanW hanhanW closed this Aug 23, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants