Skip to content
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

Split measurement of walltime and perf. counters #1465

Merged
merged 1 commit into from Oct 13, 2022

Conversation

Kobzol
Copy link
Contributor

@Kobzol Kobzol commented Oct 12, 2022

It's all on a best effort basis.. maybe once we have the benchmarks running on perf.rlo we can do some experiments to see if this actually helps reduce variance.

@Kobzol Kobzol force-pushed the runtime-measurement-precision branch 2 times, most recently from 902eb15 to 2d156a1 Compare October 13, 2022 19:36
@Kobzol
Copy link
Contributor Author

Kobzol commented Oct 13, 2022

You're right, we shouldn't add it without evidence that it helps. I have removed cache clearing from this PR and only kept the wall-time/perf. counter split.

@Kobzol Kobzol force-pushed the runtime-measurement-precision branch from 2d156a1 to f173632 Compare October 13, 2022 19:37
@Kobzol Kobzol changed the title Try to clear caches before executing a benchmark and split measurement of walltime and perf. counters Split measurement of walltime and perf. counters Oct 13, 2022
@Kobzol Kobzol force-pushed the runtime-measurement-precision branch from f173632 to c1bb6d3 Compare October 13, 2022 21:18
@Kobzol Kobzol enabled auto-merge October 13, 2022 21:18
@Kobzol Kobzol merged commit a8fb050 into master Oct 13, 2022
@Kobzol Kobzol deleted the runtime-measurement-precision branch October 13, 2022 21:28
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants