-
Notifications
You must be signed in to change notification settings - Fork 29.8k
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
benchmark: Improve event performance tests. #14052
Conversation
Currently most of the event tests only test a single event type, which might let those benchmark take fast-paths (in V8) that aren't taken in realistic use cases, and thus the benchmarks are not good proxies of real world uses.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM. For completeness, can you include the result before and after on master?
Not sure what you mean by that? |
The output of running those benchmarks before and after the change you are proposing. |
Oh, sure, here you go. Before:
After:
|
Thanks @bmeurer! |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM. Cf. #12657 (comment).
ACK. |
CI before landing: https://ci.nodejs.org/job/node-test-pull-request/8985/ |
Landed as 0011eb2 |
Currently most of the event tests only test a single event type, which might let those benchmark take fast-paths (in V8) that aren't taken in realistic use cases, and thus the benchmarks are not good proxies of real world uses. PR-URL: #14052 Reviewed-By: Matteo Collina <[email protected]> Reviewed-By: Timothy Gu <[email protected]> Reviewed-By: Michaël Zasso <[email protected]>
Currently most of the event tests only test a single event type, which might let those benchmark take fast-paths (in V8) that aren't taken in realistic use cases, and thus the benchmarks are not good proxies of real world uses. PR-URL: #14052 Reviewed-By: Matteo Collina <[email protected]> Reviewed-By: Timothy Gu <[email protected]> Reviewed-By: Michaël Zasso <[email protected]>
Currently most of the event tests only test a single event type, which might let those benchmark take fast-paths (in V8) that aren't taken in realistic use cases, and thus the benchmarks are not good proxies of real world uses. PR-URL: #14052 Reviewed-By: Matteo Collina <[email protected]> Reviewed-By: Timothy Gu <[email protected]> Reviewed-By: Michaël Zasso <[email protected]>
Currently most of the event tests only test a single event type, which might let those benchmark take fast-paths (in V8) that aren't taken in realistic use cases, and thus the benchmarks are not good proxies of real world uses. PR-URL: #14052 Reviewed-By: Matteo Collina <[email protected]> Reviewed-By: Timothy Gu <[email protected]> Reviewed-By: Michaël Zasso <[email protected]>
Should this be backported to |
Currently most of the event tests only test a single event type,
which might let those benchmark take fast-paths (in V8) that aren't
taken in realistic use cases, and thus the benchmarks are not good
proxies of real world uses.
Checklist
make -j4 test
(UNIX), orvcbuild test
(Windows) passesAffected core subsystem(s)
benchmark