Conversation
Change-Id: I12e721be649c85ce8adfd0b21eedf9a9dced7108
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: f1101dcddb
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| MemDepUnit* memDepUnit; | ||
| LSQ* lsq; | ||
| const int intel_fewops = 4; | ||
| const int intel_fewops = 8; |
There was a problem hiding this comment.
Raising intel_fewops to a hardcoded 8 makes exec_stall_cycle count cycles as “few uops” even when narrower cores are fully utilized, which skews the Topdown breakdown (coreBound/memoryBound) that is derived from these counters in src/cpu/o3/cpu.cc (lines 503-521) and normalized by cpu->issueWidth (lines 479-490). This is a regression for any config with issue width below 8 (for example, the decodeWidth = 4 Xiangshan ECore configs), where full-issue cycles will now be misclassified as stalls.
Useful? React with 👍 / 👎.
🚀 Coremark Smoke Test Results
✅ Difftest smoke test passed! |
1 participant
Change-Id: I12e721be649c85ce8adfd0b21eedf9a9dced7108