PAPI Pi5 Support #260
Open
PAPI Pi5 Support #260
Conversation
This file contains 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
The previous load/store validation tests were being optimized by the compiler in a way that caused the tests to mispredict the amount of memory instructions that are generated. This made it appear like the counters were incorrect, when it was really the test being inaccurate. To fix this, add assembly testcode for ARM to eliminate the problem of compiler optimizations. When load/store testcode is unavailable for the current platform, default back to the original matrix multiplication test.
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Outdated suggestions cannot be applied.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.
Pull Request Description
Add preset events for the Arm Cortex-A76, which is the processor used in the Raspberry Pi 5.
Additionally, add raw assembly versions of the load/store validation tests for ARM as the previous approach of estimating the amount of load/store instructions the compiler will generate for a matrix multiplication operation was providing false errors. This PR only has testcode for ARM, and in cases when there is no assembly testcode available the load/store validation tests still default to using the matrix multiplication method.
Author Checklist
Why this PR exists. Reference all relevant information, including background, issues, test failures, etc
Commits are self contained and only do one thing
Commits have a header of the form:
module: short descriptionCommits have a body (whenever relevant) containing a detailed description of the addressed problem and its solution
The PR needs to pass all the tests