qyliss changed the topic of #spectrum to: A compartmentalized operating system | https://spectrum-os.org/ | Logs: https://logs.spectrum-os.org/spectrum/
<qyliss> A triumphant return to form :3
* cole-h pulls crosvm and checks the git log
<qyliss> cole-h: not much there yet -- I've been pushing to get this done, and need to go back and commit stuff now it works
<cole-h> I noticed :D Still, excited to read it when you do push it :^)
cole-h has quit [Quit: Goodbye]
tilpner has joined #spectrum
tilpner has quit [Remote host closed the connection]
tilpner has joined #spectrum
ilmu has joined #spectrum
<qyliss> Can somebody with a crosvm checkout try cargo test -p sys_util and let me know what revision you tried and if there are any failures?
<qyliss> I seem to have a failure even on upstream, and am wondering if that's something in my environment
<qyliss> I get the same failures with a fresh upstream clone in a fresh container -- could it be my kernel?
<qyliss> It's not the Rust version...
cole-h has joined #spectrum
<qyliss> Tried on another NixOS machine and that didn't help
<qyliss> Setting up a Debian VM now...
<cole-h> What's happening?
<qyliss> cole-h: some tests have started failing
<cole-h> Confirmed -- `descriptor::clone_equality`, right?
<qyliss> err, no
<cole-h> At least, that's what I get at crosvm HEAD
<cole-h> 9fef2380da26bf8ec8a2d14a33a4441a1553f001
<qyliss> Well, that's failing on HEAD
<qyliss> But it's seek_hole ones that are failing for me
<qyliss> failures:
<qyliss> seek_hole::tests::seek_data
<qyliss> seek_hole::tests::seek_hole
<qyliss> They're not failing for you?
<cole-h> I only got clone_equality.
<qyliss> Rust version?
<cole-h> 1.44.1
<qyliss> Kernel version?
<cole-h> 5.7.2-zen1
<qyliss> Hmm
<qyliss> cole-h: Nixpkgs rev?
<cole-h> a59af90880e47ead87e3854fcd71e4357dfa5665
<qyliss> Aha!
<qyliss> Yep, that fixes it
<cole-h> Time 2 bisect :P
<qyliss> yeah
<qyliss> Although it might also just be time for me to upgrade :)
<cole-h> Oh, I thought you were on a newer nixpkgs rev
<qyliss> Doubt it -- I've fallen a bit behind
<cole-h> Hehe, then no need to bisect
<qyliss> Well, would be interesting to find out why it's broken
<qyliss> Esp because it's also broken on debian AFAICT
<cole-h> Interesting..
<qyliss> Hmm bad news
<qyliss> Now it passes for me always
<cole-h> wot
<qyliss> yeah...
<qyliss> oh, wrong nixpkgs rev nm
<cole-h> lmao
<qyliss> Okay I don't care enough
<cole-h> :P
<qyliss> I just need to use a newer nixpkgs
<qyliss> Thanks!
* cole-h sees patches in inbox
<cole-h> 👀
<qyliss> wow that was quick
<qyliss> Pushed to interguest with all the WIP crosvm code
<cole-h> Got some nice reading materials for later, then :D
<cole-h> Heh, I just noticed I commented on something untouched by the doc patch... Oops. Feel free to disregard that one :^)
cole-h has quit [Quit: Goodbye]
jb55 has quit [Remote host closed the connection]
jb55 has joined #spectrum
Profpatsch has quit [Quit: WeeChat 2.8]
Profpatsch has joined #spectrum
Profpats1 has joined #spectrum
Profpatsch has quit [Ping timeout: 272 seconds]
Profpats1 has quit [Quit: WeeChat 2.8]
Profpatsch has joined #spectrum
<qyliss> Any meson experts in here?
cole-h has joined #spectrum
<tazjin> qyliss: not an expert, but I've used it before - what's up?
<tazjin> (I'm sure with the other people in here that have used it before we can combine into one whole expert)
<qyliss> Essentially, what I'd like is to conditionally compile some code depending on whether I'm building libwayland-server or libwayland-client
<qyliss> I'm wondering if there's already a way to do that, or if not how I could add one
<qyliss> Oh, hmm, actually, this should be part of wayland_private anyway
<tazjin> hmm, it would have to set a variable for the target being compiled
<qyliss> Ignore this question
<tazjin> but it sounds a bit odd
<tazjin> oh ok
<qyliss> I may have other meson questions in the near future though
<tazjin> cool
<qyliss> Hehe, I just tried to run host Firefox in the VM compositor, and it displayed a window, but then immediately froze
<qyliss> So I suppose I have a bit of a way still to go
<qyliss> Still awesome that Alacritty works though
<qyliss> Okay, think that's enough for today. :)
<qyliss> crosvm code pushed + a bit of it ready for master being reviewed; Wayland code cleaned up
<cole-h> Woot!
jb55 has quit [Remote host closed the connection]