Jessica Clarke

Probably raced and sleeping in sigsuspend, never waking up again?

How are you ever expecting to wake up, anyway? What's delivering that signal?

Hm, I expect the idea is for the `close(pfd[1]);` to cause an EOF condition on the pending AIO read? Strange that it consistently works for cheritest but not cheriabitest. I...

We need to figure out what's going on with AIO...

Noticed this before but IIRC nothing passes non-NULL for the 9th argument so in practice it's dead code (other than the load from the stack, which works due to endianness)....

The struct probably should be register_t though to match the argument types until such time as PSCI gains a capability-aware interface (which changes almost nothing except for CPU_ON, IIRC)

Yeah in the GNU world there is no LD_32 etc it's just LD. The current FreeBSD approach is more in keeping with how it does ABI in the kernel, but...

> I think LD_PRELOAD in particular is a bit dangerous if all the linkers honor it. If i do `env LD_PRELOAD=foo.so ' where said program executes a purecap program (or...

This fails because forks don't have permission to add comments. #1432 shows it working (ish) from a branch in this repo, with the fixups squashed and the bogus permissions diff...

There are ways to make it work from forks too, but it's a bit clunky. You have to have the clang-format check perform its actions and upload the result as...