github.com/moarvm/moarvm | IRC logs at colabti.org/irclogger/irclogger_logs/moarvm
Set by AlexDaniel on 12 June 2018.
nwc10 jnthn: it's pretty calm here. The fun is forecast to arrive in 4.5 hours, and it's surprising how little evidence there is of it yet 06:29
jnthn It was very windy/wet here overnight. Then bizzarely I just walked to work in the sun. 09:52
moritz_ we quarter-hour stretches of sun here, and then it's all grey and windy and rainy again 10:08
nwc10 The fun started here about 10 minutes ago 11:18
jnthn I'm not sure if we had our share of fun overnight, or if it was just the appetizer... 11:19
timotimo we certainly had rattling and whistling 11:25
Guest4508 I wonder if I should report the "Corrupt multi dispatch cache" bug 13:00
at least we have a 100% reproducible case even though one might have to wait for it to trigger 13:01
here's the code in case anyone is interested: 13:02
while ./perl6-m -Ilib t/spec/S17-promise/lock-async-stress2.t; do :; done 13:03
sena_kun Guest4508: is thee any reason _not_ to report it? 13:07
I mean, if there is a fatal bug with steps to reproduce related to our basic test suite, what can be reasons not to report it? :) 13:08
timotimo implementorture 13:09
Guest4508 sena_kun: I didn't want to stress out jnthn :) 13:11
but of course I'll report it
nine -win 10 13:51
nwc10 nine: US vs AT keyboard layout? (your fingers thought US, but the OS thought AT) 13:59
lizmat Guest4508 jnthn confirmed on MacOS on HEAD in the 95th run 14:09
MoarVM panic: Corrupt multi dispatch cache: cur_node != 0, re-check == 0x0
jnthn 95th /o) 14:13
timotimo i recommend rr record if you can reproduce it on linux 14:19
repeat until it crashes, make sure to delete recordings in between runs, though
and not to delete the recording after the crash :D
oh, i think you can pass a directory that it'll use, with an env var maybe? 14:20
lizmat jnthn: 2nd time I tried, it died at the 53rd try 14:35
Guest4508 lizmat: rr doesn't work on Ryzen cpu's if you happen to have one 14:37
lizmat is on MacOS
Guest4508 jnthn: if you're utterly bored, this is what I managed to dig out wrt the 'Corrupt multi dispatch' error 14:39
gist.github.com/dogbert17/b74f1c97...3668979b26
jnthn Not that bored, alas ;) 14:45
I think I might well have investigated this one before to no avail
And I'm quite sure today isn't one of my smarter-than-average days :)
Guest4508 :) 14:48
moritz today is my "the answer, my friend, is blowing in the wind" day :-) 14:49
jnthn Only trouble is spotting it among all the *other* things blowing in the wind... 14:52
lizmat fwiw, upping the number of threads to 12, seems to make it crash sooner (29th attempt) 15:01
nine nwc10: yes. I use application specific keyboard layouts. US for terminal (where I do all my coding and IRC) and AT for the rest (i.e. writing emails in German). Sometimes after a reboot KDE forgets the application specific setting 15:16
lizmat fwiw: upping the number of threads to 64 makes it a lot slower, and I haven't seen it crash yet 15:43
(running for ~ hour now)
lizmat and running for another hour without fail, stopped now 16:52
it's that time of the week again: another Rakudo Weekly News hits the Net: rakudoweekly.blog/2020/02/10/2020-...r-brewing/ 17:09
jnthn lizmat++ 17:23