test: migrate slog benchmark from golang.org/x/exp/slog to log/slog

[chenyanchen]

Bechmark test log/slog.

[CLAassistant]

All committers have signed the CLA.

[chenyanchen]

Did you have uniform benchmark environment? If you do, please update benchmark result in README.

[abhinav]

Hey, @chenyanchen. Thanks for this PR. This makes sense, and .

However, I think this could benefit from a small change:
What you've done in b.Run("slog" is great.
I think we need a second b.Run("slog/LogAttrs" that is the same, except it uses Logger.LogAttrs and the old fakeSlogFields() list.
This will provide a comparison of both, slog with any, and slog with slog.Attr.

Would you be willing to make that change?
If not, one of us will try to get to it as soon as we have some time to do it.

[codecov]

Codecov Report

Merging #1363 (fb96ef2) into master (4ddf010) will increase coverage by 0.14%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master    #1363      +/-   ##
==========================================
+ Coverage   98.28%   98.42%   +0.14%     
==========================================
  Files          53       53              
  Lines        3493     3493              
==========================================
+ Hits         3433     3438       +5     
+ Misses         50       46       -4     
+ Partials       10        9       -1     

see 1 file with indirect coverage changes

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[chenyanchen]

It's very thoughtful of you.

The slog Logger.LogAttrs benchmark been added.

[chenyanchen]

Did you have uniform benchmark environment? If you do, please update benchmark result in README.

If you do not ask for specific hardware, I can do benchmarks on my M1 Mac.

[abhinav]

Did you have uniform benchmark environment? If you do, please update benchmark result in README.

If you do not ask for specific hardware, I can do benchmarks on my M1 Mac.

Hey, we do not have a uniform benchmark environment.
The benchmarks are updated with whatever machine someone was on at the time they ran the benchmarks.
It hasn't been a problem because the absolute numbers are less important than the relative numbers, which tend to follow the same trend regardless of where we're running the benchmarks.

[chenyanchen]

Benchmars has been updated. The results looks as expected.

[abhinav]

It's weird that there's no difference between slog with any and slog with LogAttrs, but that's something to investigate later.

Thanks!