benchmark-action / benchmark-action/github-action-benchmark

Summary table loses some of the extra information from go benchmarks when running with multiple CPU scenarios

Open
#242 1 comment 0 reactions 0 assignees View on GitHub
Dominant language
TypeScript
Stars
1.3k
Forks
184
PR merge metrics
No merged PRs in 30d

Description

https://github.com/google/mtail/actions/runs/8951254978?pr=858

The table includes lines like

BenchmarkStore/Add 818.6 ns/op 821.1 ns/op 1.00
BenchmarkStore/Add-1 778.3 ns/op 780.2 ns/op 1.00
BenchmarkStore/Add-1 779.9 ns/op 780.2 ns/op 1.00

which is confusng because these are three different scenarios -- the number of procs per the -cpu flag to `go test -bench`.

This is being recorded in the JSON file found in https://github.com/google/mtail/actions/runs/8951254978/job/24587370213?pr=858

```
{
name: 'BenchmarkStore/Add',
value: 818.6,
unit: 'ns/op',
extra: '1260484 times\n1 procs'
},
{
name: 'BenchmarkStore/Add-1',
value: 778.3,
unit: 'ns/op',
extra: '1538625 times\n2 procs'
},
{
name: 'BenchmarkStore/Add-1',
value: 779.9,
unit: 'ns/op',
extra: '1574536 times\n4 procs'
```

in this `extra` field. It would be useful if the name column included the "N procs" part as well to differentiate the names. Or it would be fine to leave the name alone as `BenchmarkStore/Add-1-4` and _not_ extract the procs count into a special "extra"

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.