Run benchmarks directly

[angerman]

With #168, we skipped using cabal to run the tests, we can do the same for benchmarks as well. For the sake of consistency this seems to be the best approach.

[angerman]

@ruhatch are you going to get around to this as well, or should I?

[rvl]

I fixed the benchmarks build in #175. Unlike test suites, it builds the benchmarks, but does not execute them. I think this is good, because there would be a lot of variance when running benchmarks on hydra build slaves.

This is one example of how we execute benchmarks: input-output-hk/cardano-wallet#531

One thing we could fix is installing benchmarks in $out/bin rather than a path which is different for every component.

[angerman]

One thing we could fix is installing benchmarks in $out/bin rather than a path which is different for every component.

you mean all benchmarks for a package into the same $out?

[ruhatch]

@angerman I'm not going to be able to take this one, so removing myself as assignee, sorry!

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.