Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
[ci/cd] Split Test pipeline according to
runtime-benchmarks
feature (…
…#48) ## Summary Based on original [issue](paritytech/ci_cd#864 ) from @gilescope related to the monorepo: > Currently the tests have started to be run with the `runtime-benchmarks` feature enabled ( E.g. https://github.com/paritytech/polkadot-sdk/blob/aabed6757e26e21ce06dd59946a17b72f349185e/.gitlab/pipeline/test.yml#L37 ). This was done so that the autogenerated tests that the benchmark macros create get run (but it seems that we already have a short-benchmark CI job... - do the additional tests give us anything extra? EDIT: the short-benchmarks only test 3 relay chains so not comprehensive enough). > However, this is inherently dangerous as not everything will be in a production like state when it is being tested. > For safety's sake we should be running without the `runtime-benchmarks` feature enabled. Assuming the CI system can take it, we should set up another job that runs the tests with the runtime-benchmarks feature enabled. (if we did not do this we would catch any problems later when the weights are rerun for the runtime release builds). ## Proposed solution Split [test pipeline](https://github.com/polkadot-fellows/runtimes/blob/main/.github/workflows/test.yml#L44) to the two pipelines: - one without `runtime-benchmarks` feature - one with `runtime-benchmarks` feature Co-authored-by: Bastian Köcher <[email protected]>
- Loading branch information