VMManager: Refactor Affinity Control to Thread Pinning #11403
Merged
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
Instead of having control over specific threads, thread pinning puts the EE/VU/GS threads on the most performant cores, then the software threads on the remaining cores, but only if they're in the same cluster. This way we don't end up pinning across clusters with different performance characteristics, which would harm instead of help software renderer performance. Also unpins on shutdown, that way we don't keep CPU cores awake.
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Outdated suggestions cannot be applied.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.
Description of Changes
Instead of having control over specific threads, thread pinning puts the EE/VU/GS threads on the most performant cores, then the software threads on the remaining cores, but only if they're in the same cluster.
This way we don't end up pinning across clusters with different performance characteristics, which would harm instead of help software renderer performance.
Also unpins on shutdown, that way we don't keep CPU cores awake.
And ignores SMT cores, because you don't want to be pinning two heavy threads on the same hyperthread...
Rationale behind Changes
The option in its current form is potentially detrimental to performance on heterogeneous CPUs.
Suggested Testing Steps
Check option behaves as expected. You can see what threads are pinned in the log.