-
Notifications
You must be signed in to change notification settings - Fork 399
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
fix: remove prod perf markers on wc.ts #811
Conversation
Benchmark resultsBase commit: lwc-engine-benchmark
|
720ff20
to
75668ed
Compare
Benchmark resultsBase commit: lwc-engine-benchmark
|
avoid string concat, instead add it in the enum.
75668ed
to
c7e0d3e
Compare
Benchmark resultsBase commit: lwc-engine-benchmark
|
c7e0d3e
to
8a1d899
Compare
Benchmark resultsBase commit: lwc-engine-benchmark
|
@dbajric these are benchmarks explanation:
i still think that we should remove the thoughts? |
markers caused a bit of noise on wc-* perf tests, we can remove these cause it defeats the purpose of the performance markers in production: we dont want to measure individual components performance, wich is basically what they are doing when wc are used. summarizing… we are keeping the markers: init, hydrate for root components created with createElement, and processing rehydrate queue (all)
what is this? why only in wc.ts? what about the other marks? wc.ts is not really used in platform today, so it is definitely not the one affecting the perf. |
Benchmark resultsBase commit: lwc-engine-benchmark
|
@caridy i updated the details and desc on the pr. you are correct, they are not affecting the perf in platform, but was a red flag. When using wc, we are measuring individual components, that we agreed we dont want. The main issue on wc-table-10k is because the markers were not being cleared, that solved the perf regression on wc, cause again, the volume of marks and measures in this test is really high in comparison with the regular perf tests. I think we should remove these (wc prod marks) cause we are not using them on platform, and we can always bring them back. |
Benchmark resultsBase commit: lwc-engine-benchmark
|
@dbajric can you take a look make sure is fine by you now? |
Details
markers caused a bit of noise on wc-* perf tests, we can remove these cause it defeats the purpose of the performance markers in production: we dont want to measure individual components performance, which is basically what they are doing when wc is used.
summarizing… we are keeping the markers: init, hydrate for root components created with createElement, and processing rehydrate queue (all).
Does this PR introduce a breaking change?