Run benchmarks on CI, test for regressions
bitjson opened this issue · comments
Currently benchmarks can be manually run with yarn bench. It would be great if this was part of CI builds and tested for notable regressions from previous results.
Running benchmarks on CI as of bitjson@2910e19
It would be better to somehow automatically check for big regressions (by comparing results with previous CI builds), but I don't think that's worth working on/maintaining yet. For now, we can manually review the benchmarks for regression when pulling in changes that touch benchmarked functionality.
(If anyone wants to pick this up in the future, I'd love to see it happen.)