refactor(Prefetcher): split L2 prefetch request queue by slices#13
Merged
Conversation
ywlcode
requested review from
Ivyfeather,
Kumonda221-CrO3,
Maxpicca-Li and
linjuanZ
as code owners
Maxpicca-Li
approved these changes
Jun 4, 2026
linjuanZ
requested changes
Jun 8, 2026
linjuanZ
approved these changes
Jun 8, 2026
linjuanZ
pushed a commit
to OpenXiangShan/XiangShan
that referenced
this pull request
Jun 10, 2026
bump XScache (OpenXiangShan/XSCache#10) (OpenXiangShan/XSCache#13) Co-authored-by: ywlcode <yuwl2024@shanghaitech.edu.cn>
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
This PR refactors the L2 prefetch request path in XSCache by replacing the shared prefetch request queue with per-bank request queues and per-bank request ports.
The main goal is to reduce cross-bank arbitration blocking for L2 prefetch requests, especially when multiple prefetch sources target different L2 banks in the same cycle.
Performance
On SPEC06 gcc15 0.8cov:
SPEC2006/GHz:19.442 -> 19.500,+0.298%SPECint2006/GHz:+0.201%SPECfp2006/GHz:+0.367%Key counter changes over matched checkpoints:
70.525% -> 96.584%63.898% -> 96.131%57.384% -> 94.386%Extra work
Add performance counters for the stamping loss situation in train and resp.