🦋 Welcome to the IRC channel of the core developers of the Raku Programming Language (raku.org #rakulang). This channel is logged for the purpose of history keeping about its development | evalbot usage: 'm: say 3;' or /msg camelia m: ... | log inspection situation still under development | For MoarVM see #moarvm
Set by lizmat on 22 May 2021.
Xliff MasterDuke: Seeing improvements over last week after the first project. More than made up the 30 seconds list. 00:10
s/list/lost/
MasterDuke nice
Xliff Will post whole summary when done. Takes about 4 hours total.
gfldex m: (^∞).grep(*.is-prime).is-lazy.say; (^∞).hyper.grep(*.is-prime).is-lazy.say; 00:11
camelia True
False
gfldex Is this intentional?
Voldenet you don't even need grep 00:20
m: (^∞).is-lazy.say; (^∞).hyper.is-lazy.say;
camelia True
False
gfldex method is-lazy(--> False) { } # can be found in ParallelSequence.pm6 00:21
And I don't get why it has to be.
Voldenet gfldex: related commit, when you look at the history it was added around hyperseq rework github.com/rakudo/rakudo/commit/83...1482406843 00:35
Xliff MasterDuke: This weeks compile times were 98.63% of last weeks 06:17
It is worth noting that last week were 102.16% above the week prior to that. 06:18
p6-GLib Projects Total Compile time: 13437,31s over 31 projects 06:19
Parallel compile times cancelled again due to #4655 06:20
nine Xliff: FWIW I consider all such bugs as urgent. Fixing them still depends on having time, being able to reproduce and having any idea at all what the hell is going wrong... 07:55
Xliff nine: Fair enough. However there has been no indication of progress on the bug report. 07:56
I figure after another few weeks I'll have enough data to figure out where the bug might be introduced by my code, given its intermittent nature. 07:57
Still it is nice to hear someone is still working on it.
nine: I did make a mention in the report about the possibility CompUnit::Util
CompUnit::Util's re-export() involed. Has anyone looked into that? 07:58
nine Not me. I've been pretty busy with the whole unsigned native ints fixing business. 07:59
Xliff Oh. Does this mean I will no longer get signed integers for unsigned values? 08:07
nine lizmat_: nqp repo looks quite ok to me 08:08
nine Xliff: At some point. I've gotten it far enought that fixing such issues will finally be possible. It took some 8k lines of diff to teach the whole stack that ints and uints are different things. With that in place we will be able to look at individual places that need to distinguish. 08:11
lizmat Files=1351, Tests=117094, 295 wallclock secs (35.53 usr 10.96 sys + 3971.34 cusr 340.15 csys = 4357.98 CPU)
nine Basically I'm fixing a design mistake in a well established code base. That's quite a lot of work 08:12
Xliff nine++ - Thanks!
lizmat that's the lowest spectest time I've seen since April 2021
I think MasterDuke's mimalloc work could be involved in that :-) 08:13
Xliff I just noticed that change, yesterday. I had to reinit the submodules to update rakudo. 08:15
lizmat notable6: weekly 10:39
notable6 lizmat, 2 notes: 2022-01-12T18:28:12Z <El_Che>: www.youtube.com/watch?v=dQw4w9WgXcQ ; 2022-01-12T18:30:04Z <lizmat>: rakuforprediction.wordpress.com/20...with-raku/
lizmat notable6: weekly reset
notable6 lizmat, Moved existing notes to “weekly_2022-01-17T10:39:55Z”
lizmat and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2022/01/17/2022-...kucon-how/ 13:01
sena_kun lizmat++ 13:02
releasable6 Next release in ≈4 days and ≈3 hours. 1 blocker. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 15:00
jdv MasterDuke: did you look at MessagePack blin failure? 16:00
i think it bisected to your ternary thing and it was a segv or something like that
can someone look at the Trait::Traced and SAT::Solver::MiniSAT blin failures? 16:03
MasterDuke jdv: yeah, my new code isn't actually to blame for the segv, it just uncovered some other bug. but i haven't figured out/fixed the real bug yet. and given github.com/rakudo/rakudo/issues/4719, there's a good chance we'll need to revert the ternary warning (at least) for the release 16:06
jdv ah, ok 16:07
MasterDuke jdv: that bisect for SAT::Solver::MiniSAT doesn't look right 16:10
jdv i got snowed in upstate and i don't have access to most of my stuff up here:( hopefully i can do a bit more tomorrow. 16:11
MasterDuke oh fun. hopefully you can get in some skiing/snowboarding?
jdv likely not this time. still not feeling great from the last bout with covid... 16:13
Xliff jdv: :(
jdv mostly better though 16:15
i think the Trait::Traced might be the easiest... it looks like maybe just some rejiggering. there's info if you look back through the bisected commits stuff from jnthn. 16:16
*commit's 16:17
[Coke] jdv: upstate sounds like you might be in a 6 hour drive from me. :) 21:10
patrickb fg 22:59
patrickb is off to bed 23:00
o/

Powered by Cro and the Raku® Programming Language.

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

Thank you!