00:17 kjp left
japhb m: my $hash = Hash::Ordered.new( (a => 1, c => 3, b => 2) ); $hash.pairs.say; $hash.raku.say 00:29
camelia (a => 1 c => 3 b => 2)
Hash::Ordered.new((:a(1), :b(2), :c(3)))
japhb lizmat: ^^ Hash::Ordered.raku is reordering the keys
03:36 Voldenet_ joined, vrurg joined, nativecallable6 left, lucs left 03:37 vrurg_ left, Voldenet left, JRaspass left, Voldenet_ is now known as Voldenet, lucs joined 03:39 nativecallable6 joined 03:41 Voldenet_ joined 03:42 Voldenet left, Voldenet_ is now known as Voldenet 03:43 nativecallable6 left, lucs left 03:44 nativecallable6 joined 03:45 lucs joined 03:49 nativecallable6 left, JRaspass joined, nativecallable6 joined 07:01 melezhik joined
melezhik . 08:49
patrickb melezhik is there any run in progress? I could boot up the machine now. 09:05
melezhik Cool , will run test round
Run 2000 modules for Rakudo 2025.10 09:06
patrickb: please pull the latest commit 09:09
Thanks
patrickb: let me know when you run 09:23
lizmat japhb: looks like something changed in rakudo that broke handling of multis in roles 09:26
patrickb melezhik: it's up 09:29
melezhik Yep, see it . heartbeat from loving-buck-34594827, version: 0.0.36, max threads: 19 09:31
ab5tract: would you like to join the party ? )) 09:42
patrickb: “time: 21m | tests total: 2000 | finished tests: 268” not bad at all 😊 09:55
What about CPU/RAM on agent so far ? Does it hold ? 09:56
patrickb Works so far. Not yet swapping. 09:57
melezhik Excellent
I have added my tiny agent with 1 thread as well , he he ) 09:58
Ok, it’s almost 400 modules under 30 minutes , incredible) 09:59
||| TESTS STAT: time: 25m | tests total: 2000 | finished tests: 382
Cc ab5tract
The good thing with sparky architecture is modules installations runs in isolated short living jobs ( Linux processes technically) , so there is no memory leaching , as a process goes and dies freeing all resources 10:01
Leaking
The only place where leak may happen is agent sparky web server and agent main job long running process 10:03
10:14 committable6 left 10:15 committable6 joined
melezhik ||| TESTS STAT: time: 53m | tests total: 2000 | finished tests: 903 10:27
tests total: 2000 | finished tests: 1757 11:28
In less then 2 hours
Finished . ✔️ 2000 modules in 2 hour 24 minutes . Cc patrickb [Coke] an 12:10
Thanks a lot patrickb 12:11
Now I can run comparison report
patrickb ram usage is still at 30gb 12:17
most of it taken by loads (guess 50) of `zef.raku install ParaSeq` processes and 8 `zef.raku install Proc::ZMQed` processes. 12:19
there are also many `rakudo t/11-reduce.rakutest` processes still there 12:26
melezhik Ok. From o10r point I f view round has finished , it’s no longer distribute load to agents . There might be some redundant jobs running on agents, but sooner or later they will finish 12:33
ping from agent: loving-buck-34594827, version: 0.0.36, jobs-run-cnt: 16, max-threads: 19
There 16 jobs now on agent
Even 15 12:34
jobs-run-cnt: 15, max-threads: 19
patrickb: you may terminate all running agent jobs via UI 12:41
14:07 melezhik_ joined
melezhik_ Started to post finding into gist - Diff between rakudo 2025.10 & rakudo 2025.11 - gist.github.com/melezhik/914cb7c84...5cd1d3f5bc 14:08
cc [Coke]:
sorry rakudo version mixed up in this report 14:10
seems Hashids issue was for Rakudo 2025.10 and has gone for Rakudo 2025.11 14:11
14:14 melezhik_ left 14:50 melezhik left
Geth rakudo/r5970-generic-finale: 8ede616be1 | ab5tract++ | 2 files
Fix for curried classes
15:29
japhb lizmat: Bah, that's no good. 16:52
16:56 Pixi` is now known as Pixi 17:39 melezhik_ joined
melezhik_ Ok. Modules stopped passing tests in 2025.11, but passed in 2025.10 17:40
brw.sparrowhub.io/file_view/brw-rep...orkers.log
brw.sparrowhub.io/file_view/brw-rep...st::XS.log 17:44
lizmat japhb ab5tract looks like 0a90592a5eebc3d19d23062 borked Hash::Ordered (doesn't pass tests anymore) 17:45
linkable6 (2025-11-18) github.com/rakudo/rakudo/commit/0a90592a5e Add an istype check on the role group
lizmat afk again&
melezhik_ brw.sparrowhub.io/file_view/brw-rep...ersion.log 17:51
brw.sparrowhub.io/file_view/brw-rep...33/Mux.log cc tony-o: 17:53
brw.sparrowhub.io/file_view/brw-rep...ephone.log 17:54
18:05 melezhik_ left, melezhik_ joined 18:10 melezhik_ left, melezhik_ joined 18:25 melezhik_ left, melezhik_ joined 18:31 melezhik_ left, melezhik_ joined 18:36 melezhik_ left, melezhik_ joined 18:53 melezhik_ left
librasteve_ rakudoweekly.blog/2025/12/01/2025-...t-is-here/ 19:42
21:44 librasteve_ left