github.com/moarvm/moarvm | IRC logs at colabti.org/irclogger/irclogger_logs/moarvm
Set by AlexDaniel on 12 June 2018.
brrt moarning 06:09
nwc10 \o 06:35
samcv Morning On A Runtime 07:13
brrt ohai samcv 07:22
samcv heye brrt 07:43
brrt btw, anybody planning to go to FOSDEM this year? 10:43
nwc10 No. But I am planning to go to act.yapc.eu/gpw2020/ 10:47
brrt hmm 10:56
interesting
mildly inconvenient but interesting
jnthn I'll get at gpw2020 also, fwiw
nine brrt: I am 11:13
planning on both 11:14
brrt I wonder if it'd be reasonable to try and drive to Brussels... 11:27
it's much closer than I imagined from the train distance
nwc10 yes, but, is it reasonable to drive *back* from Brussels 11:28
brrt hehe, there is belgian traffic to account for as well
nwc10 Belgian beer good, Belgian traffic bad? 11:29
brrt from the PoV of the Netherlands, I'd say that's accurate 11:30
and driving is a strict choice for the one in favor of the other
nwc10 that was why I was suggesting that the "drive from" was a more important question than the "drive to" :-)
brrt the thing is, if/when I've taken the train, it's been... suboptimal 11:31
wait for the bus, no obvious way to pay for a ticket
wait for the shuttle bus back 11:32
miss the train
nwc10 convince someone else that they like driving more than they like beer?
brrt :-D 11:33
nine I'll probably rent a bike again in Brussels 12:34
rypervenche I noticed that MoarVM 2020.01.1 is not considered the "Latest release" on the github releases page. Is that intentional because there isn't yet a rakudo release? 14:26
sena_kun rypervenche: as far as I know, 2020.01.1 point release is not yet done 14:27
there are indeed preparations for it done
rypervenche Ahh, so it's just a tag then? I didn't know it would show up under Releases. 14:28
sena_kun awww, there is a tag already?
rypervenche samcv said "release is done" yesterday. 14:30
sena_kun rypervenche: ok, then it's the different beast. I think the releases page just wasn't updated.
because 2020.01 isn't there too
damn, I've missed that. :S 14:31
rypervenche: then rakudo will have a release tonight too. :)
rypervenche Glad I said something :)
sena_kun yup, thank you very much. :) 14:32
brrt nwc10: on the other hand, I didn't have any beer last FOSDEMs either 15:52
timotimo gist.github.com/timo/bd9661f276f68...8b9900c741 - counts of vmarray sizes, usages, and slot types at the end of perl6 -e 'use nqp; nqp::force_gc()' 17:00
one potential optimization is in-situ storage, i.e. instead of a pointer to an allocated blob we store the values there 17:02
important to not make the regular use case of "make array a given size, then resize it back to 0" any worse 17:05
japhb timotimo: Why are there thousands of "sz 0 use 0"? Why do those arrays even still exist at that point? 18:08
lizmat grammar stuff ? 18:09
japhb I'm also intrigued about why "sz 3 use 3" is far and away the most common, followed by 6 and 9. Wondering what 3-element arrays are being used so much for.
lizmat: Hmmm, that's an interesting thought (and possible LHF for optimization?) 18:10
lizmat possibly, but I could also see it being an optimization so that you wouldn't have to check for nullness 18:11
japhb lizmat: Right, but you could have all of those point to the *same* vmarray. 18:12
lizmat true
samcv sena_kun, do you mean the moarvm github releases page? 19:58
sena_kun samcv: apparently 20:04
I don't see latest moarvm release at github.com/MoarVM/MoarVM/releases at the very least