-
Notifications
You must be signed in to change notification settings - Fork 716
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
feat(TopDown): add TopDown PMU Events #4122
base: master
Are you sure you want to change the base?
Conversation
857989a
to
5d2533d
Compare
5d2533d
to
a861a9c
Compare
remember to merge the changes in submodules into their masters first a PR in XiangShan must have all its submodules checked-out to their masters (otherwise CI would fail) |
I don't understand. This PR does not distinguish frontend latency from bandwidth. Frontend latency is a counter that increases by 1 every time the decode bandwidth is not full. This does not make sense to me. |
* commit: 90aaf5935206ff322e461c3d021436c20dd0ac85 Including: * fix(ClockGate): avoid unused BlackBox modules (#184) * feat(TopDown): add l3Miss IO for Top-Down
* commmit: 0f9f93515853e66b8b1480fbe3f73508b1e270cc Including: * fix(MSHR): always clear meta.dirty on CMO Release (#313) * feat(MSHR): support WriteEvictOrEvict on eviction (#311) * style(Message): more robust issue isolation (#315) * feat(TopDown): add l2Miss IO for Top-Down * submodule(Huancun): bump Huancun
a861a9c
to
638179f
Compare
[Generated by IPC robot]
master branch:
|
This PR adds hardware synthesizable three-level categorized TopDown performance counters.
Level-1: Retiring, Frontend Bound, Bad Speculation, Backend Bound.
Level-2: Fetch Latency Bound, Fetch Bandwidth Bound, Branch Missprediction, machine clears, Core Bound, Memory Bound.
Leval-3: L1 Bound, L2 Bound, L3 Bound, Mem Bound, Store Bound.