Thomas Munro

(Sorry for the force pushes, not too used to github workflow and didn't know that'd show up like that. :-)) Thanks for the feedback! I was writing up the docs...

Sorry for the delay: I will post a new patch this weekend coming, with an option as requested.

Done. Seems quite nice this way. BTW I plan to propose a couple more values for posixaio_wait, in later PRs (sigwaitinfo and kevent).

Thanks! They look great. I pushed an experimental change to use them, run with new non-advertised "--unicode" option. Seems like some more work is needed to space them out better,...

On my FreeBSD system it looks like this: ` _ _ _ _ __ _ ___ ___(_|_) (_)_ ____ ____ _ __| | ___ _ __ ___ / _` /...

Oh yeah I see the same corruption on Gnome Terminal on a Debian system. Funnily enough if I log into FreeBSD remotely through the Gnome Terminal on the same Debian...

Oh yeah, that fixes it! Now, I think we need to sort out the bombs (they kind of zigzag on the real game), the shields, the explosions, and the spacing...

Tests fixed as requested. I had another go at computing the length; I didn't like the way dmu_prefetch loops past the end limited only by the dmu_prefetch_max setting, so I...

Hmm. zloop failed with "ASSERT at cmd/ztest.c:5933:ztest_dmu_snapshot_hold()/lib/x86_64-linux-gnu/libasan.so.6(+0x45c0e)[0x7efebc67fc0e]", but when I try it locally I get various other random failures from an asan/ubsan build (on FreeBSD it fails fast, on Linux...