janestreet / janestreet/magic-trace

Changing c-states breaks magic-trace / IPT

Open
#3 3 comments 0 reactions 0 assignees View on GitHub
question
Dominant language
OCaml
Stars
6.3k
Forks
209
PR merge metrics
No merged PRs in 30d

Description

I was experiencing issues where my traces were completely empty around the snapshot. After investigating the perf file it hinted at "instruction trace errors" which led me to

https://perf.wiki.kernel.org/index.php/Perf_tools_support_for_Intel%C2%AE_Processor_Trace

which mentions

> It is not uncommon to get overflows when transitioning to a C-state, so these errors are not significant.

I was testing this on a TGL laptop and after disabling turbo boost I got pretty stable traces again.

I am wondering whether other people share the same experience with switching c-states or whether there is maybe something else behind it?

If not it might be worth mentioning disabling c-states in the readme / tutorial? Turbo boost was enough for me but probably something lik e the max_cstate kernel flags work as well.

This is on 5.15.17.

Contributor guide

Open the contributing guide

Research direction

Start by reproducing the empty snapshots on the reported TGL laptop with c-states and turbo boost varied, then inspect the perf file for instruction trace errors. Compare the results with the linked Intel Processor Trace guidance and the max_cstate kernel flags. Done means establishing whether c-state transitions cause the failures and, if confirmed, identifying the README or tutorial location for the guidance.

Written by the indexing model from the issue text.

Assessment

Tech stack
linux, ocaml
Domain
operating-systems, performance
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.