Repository navigation
[LSRA] Further tuning of preferencing is needed #11959
Description
LSRA preferencing was improved with dotnet/coreclr#19429, but additional tuning is needed.
On x86, the Span.IndexerBench benchmark, the changes caused additional spill in some key methods due to different allocation choices. Specifically:
Span.IndexerBench:TestSameIndex1(struct,int,int):ubyte- V02 gets spilledSpan.IndexerBench:TestCoveredIndex2(struct,int,int):ubyte- V02 gets spilledSpan.IndexerBench:TestIndexer6(struct):ubyte- V02 gets spilledSpan.IndexerBench:TestCoveredIndex3(struct,int,int):ubyte- V04 gets spilled
Another area to investigate (probably independent from the above), is that it seems logical that RefPosition::getRangeEndRef() should only return 'nextRefPosition' if it is a lastUse and otherwise return lastRefPosition, but that tends to excessively lengthen the range for heuristic purposes. Another option might be to iterate to the next lastUse, but that might have excess throughput impact.
category:cq
theme:register-allocator
skill-level:expert
cost:large
Activity
dotnet-policy-service commentedon Apr 2, 2024
Due to lack of recent activity, this issue has been marked as a candidate for backlog cleanup. It will be closed if no further activity occurs within 14 more days. Any new comment (by anyone, not necessarily the author) will undo this process.
This process is part of our issue cleanup automation.
dotnet-policy-service commentedon Apr 17, 2024
This issue will now be closed since it had been marked
no-recent-activitybut received no further activity in the past 14 days. It is still possible to reopen or comment on the issue, but please note that the issue will be locked if it remains inactive for another 30 days.