Glauber Costa

Definitely a possibility, but should come from someone with expertise in that technology. I don't have either the expertise or the bandwidth unfortunately

Yes, and that is quite amazing, actually! That would be the preferred direction for us to go forward with for sure.

I have a few problems with this strategy: First, please validate my understanding: what you are proposing we do here is prioritize explicit wakes within the same task queue by...

I must be missing something. In terms of results themselves, while you claim "A gets better while nothing gets worse", a lot of benchmarks are actually worse on "After". The...

Thanks! Realistically it will take me a couple of weeks to get to this as I am currently traveling. sorry about that!

tl;dr: I am 100% onboard, but I don't think we should rush to a solution. Connection balancing was by far one of the biggest issues in seastar applications, so I...

Fully Agree. Usually the way to make it work is to have a parametrized clock implementation that defaults to wallclock, but for tests you can pass your own that you...

Although obviously familiar with tracing systems in general, I was not familiar with the tracing crate. I would have to spend more time on it. In particular from an admittedly...

The main blocker for this has been lack of time on my end, as I am focusing on other issues right now. If someone wants to move it forward, that'd...

It's concerning that such a simple test fails. On the other hand, it's great that it is a simple test that fails =p However your C test doesn't seem fully...