[KB] Default linalg-to-loops pipeline#223
Merged
Merged
Conversation
Move the default loops lowering to Lighthouse Descriptor/Driver so that any user of lighthouse can have a reasonable behaviour without having to build their own pipeline for basic lowering. Advanced/optimizing pipelines will need to be built anyway, but they can also already use the internal components anyway.
adam-smnk
approved these changes
Jul 10, 2026
- rename linalg-to-loops to scalar-lowering - add TOSA pipeline to make it more generic - rename llvm_lowering to llvm-lowering
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Move the default loops lowering to Lighthouse Descriptor/Driver so that any user of lighthouse can have a reasonable behavior without having to build their own basic pipeline for loops lowering.
Advanced/optimizing pipelines will need to be built anyway, but they can also already use the internal components already.
Later we'll have to decide if we move the kernel bench schedules inside the lighthouse search path, or publish them as a separate thing and inform the user of the search path. For now, external users can just call the
linalg-to-loops.yamlby name, which is the point now.