Refactor benchmarking infra #170
Labels
effort: medium
Default level of effort.
priority: 0
Nice-to-have. Willing to ship without this.
type: ref
A code update that doesn't meaningfully change functionality.
uphill
Some research/design is needed before this can be implemented.
Currently, our benchmarking suite is quite rough and has a bunch of code dependent on
e2e
, we should decouple them.Benches are also coupled to displaying the results, which should not be the case. Once this coupling is gone, it should make adding the results as a comment on a PR much easier, as mentioned in #169
The text was updated successfully, but these errors were encountered: