Welcome to the main channel on the development of MoarVM, a virtual machine for NQP and Rakudo (moarvm.org). This channel is being logged for historical purposes.
Set by lizmat on 24 May 2021.
nine So, I ran p6-jupyter-kernel/t/05-autocomplete.t in a loop for 16 hours and no more crashes after my spesh stats memory corruption fix :) 06:22
Nicholas and as it's clearly stuck at 0-0, now it goes to penalities? 06:23
nine However, it did run into a deadlock once (out of 10K runs). It got stuck waiting for a semaphore in await while all worker threads were waiting on the task queue: gist.github.com/niner/346dbbd64710...1f23197afd
I've seen this deadlock before in other programs
Anyway, time to create some PRs with my fixes 06:24
MasterDuke Nicholas: afk until this evening, but did you see my reply about benchmarking your branch? 06:36
Nicholas no, I hadn't 06:37
OK, odd. Interesting. If that thing can spin on the mutex there in the new code path, does that mean that it spins on a mutex in the existing places that the existing code calls. 06:38
I don't have time to investigate currently.
could you add that one or two line finding as a comment to the PR, so that we don't forget it
(maybe "this evening") 06:39
MasterDuke sure
Nicholas whatever this "evening" thing is. :-)
Nicholas I might not get a chance to get back to that stuff for (literally) a couple of months. 08:27
I have a backlog of "things that I should have been doing but PSC got in the way"
lizmat ah, not the RSC :-) 08:29
Nicholas indeed not. I wonder if there's a QSC
lizmat Questionable :-) 08:32
brrt good * #moarvm 11:16
jnthnwrthngtn o/
brrt congrats to the czechs for beating the dutch :-) 11:18
Nicholas hurrah! Now we no longer need to drink Heineken 11:19
brrt I certainly hope so 11:22
Nicholas or Super Bock 11:24

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!