Jimmy Brush

> As a user, you only notice these if you are actually trying to read the reply from the kernel, which apparently does not happen that often (nobody noticed it...

> My understanding is that a `null cap` is not a real thing as such, and instead is simply a marker that indicates an empty cap slot. While the CNode's...

> A slightly more radical departure from the current API would be to say "everything is a syscall", not "everything is an IPC with the kernel". Then it would make...

FYI, I've seen this test fail when `LibUtilsDefaultZfLogLevel` is increased. I assume the extra prints take long enough to exceed the margin.

I've definitely seen that flag reliably trigger this failure on another platform. If that's not supposed to happen I'll repro it again and file an issue about it.

I also think some of the timer drivers in libplatsupport might introduce some rounding error when converting from/to timer ticks and nanoseconds because they use a pre-calculated number of nanoseconds...

> The printing happens after the measurement in each loop, so this wont affect it. Ah, the failure I saw was in a different test, SCHED0021.