Tamir Duberstein
Tamir Duberstein
> > Why is that better? If we know they can't be zero it's better to communicate that in the type. > > Because API ergonomics matter. > > `NonZero`...
@tyrone-wu can you rebase this please?
@tyrone-wu I rebased this for you.
Alrighty, we've agreed to take this and follow up with older kernel tests in follow up work.
@tyrone-wu interested in getting this over the finish line?
Weird, I pushed this branch and I can see that in https://github.com/tyrone-wu/aya/tree/aya/perf-event-type-config but the PR hasn't updated.
@alessandrod does the API look reasonable to you?
Hey this is a pretty big show stopper - my tests run fine locally but hang in Github Actions. Can we make mocking the monotonic clock optional?
@fhilgers as of #1148 our integration tests are running on ARM. Would you like to work on adding a test that exercises this?
I haven't forgotten about this; I think we should first remove the tokio dependency from aya-log so that it can be more-or-less agnostic to the user's chosen executor. I sent...