Perl 6 language and compiler development | Logs at colabti.org/irclogger/irclogger_log/perl6-dev | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by Zoffix on 27 July 2018.
00:00 lizmat joined, r00t22 joined, p6bannerbot sets mode: +v lizmat, r00t22 left 00:09 MasterDuke left, lizmat left
jdv79 AlexDaniel: can you bench that again on head? 00:43
AlexDaniel jdv79: sorry, bench what? 00:45
00:47 timotimo left 00:48 Zoffix joined, p6bannerbot sets mode: +v Zoffix
Zoffix .tell AlexDaniel I believe this is the ticket for the JVM issue: github.com/perl6/nqp/issues/494 00:48
yoleaux Zoffix: I'll pass your message to AlexDaniel.
AlexDaniel . 00:49
yoleaux 00:48Z <Zoffix> AlexDaniel: I believe this is the ticket for the JVM issue: github.com/perl6/nqp/issues/494
AlexDaniel Thanks
00:49 timo joined, p6bannerbot sets mode: +v timo, timo is now known as timotimo, Zoffix left 01:06 x[LGWs4x4i]uG2N0 left 01:10 x[LGWs4x4i]uG2N0 joined, p6bannerbot sets mode: +v x[LGWs4x4i]uG2N0 01:52 MasterDuke joined, p6bannerbot sets mode: +v MasterDuke, MasterDuke left, MasterDuke joined, herbert.freenode.net sets mode: +v MasterDuke, p6bannerbot sets mode: +v MasterDuke 01:53 Ven` joined 01:54 p6bannerbot sets mode: +v Ven` 01:57 Ven` left
MasterDuke .tell jnthn i also saw some weird failures/segvs in rakudo's `make test` that i couldn't repro. i even ran one of the files that failed in a loop overnight and it was still going fine in the morning 01:58
yoleaux MasterDuke: I'll pass your message to jnthn.
02:06 Kaiepi left, Kaiepi joined 02:07 p6bannerbot sets mode: +v Kaiepi
jdv79 AlexDaniel: github.com/rakudo/rakudo/issues/21...-409032994 02:26
AlexDaniel c: 5cf0afc^,5cf0afc,HEAD gist.github.com/AlexDaniel/7102bb2...2a41ea121f 02:27
committable6 AlexDaniel, gist.github.com/c6b9ce68ecd2c28ea7...f4e7b92975
AlexDaniel c: 5cf0afc^,5cf0afc,HEAD gist.github.com/AlexDaniel/7102bb2...2a41ea121f
committable6 AlexDaniel, gist.github.com/692719aee7dd3ced42...deb5714f23 02:28
AlexDaniel jdv79: seems to be much better but still not as fast as before 5cf0afc
left a comment 02:29
jdv79 cool 02:32
releasable6 Next release in ≈6 days and ≈15 hours. 6 blockers. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 03:00
03:54 Ven` joined, p6bannerbot sets mode: +v Ven` 03:58 Ven` left 05:54 Ven` joined 05:55 p6bannerbot sets mode: +v Ven` 05:59 Ven` left 06:51 linear25 joined 06:52 p6bannerbot sets mode: +v linear25, linear25 left 07:04 steev10 joined, p6bannerbot sets mode: +v steev10 07:05 steev10 left 07:25 robertle joined 07:26 p6bannerbot sets mode: +v robertle 07:54 lizmat joined, p6bannerbot sets mode: +v lizmat, Ven` joined 07:55 p6bannerbot sets mode: +v Ven` 07:59 Ven` left
lizmat m: class A {}; my $a = A.new; dd << foo $a >> # underlying issue of #2203 08:32
camelia No such method 'words' for invocant of type 'A'
in block <unit> at <tmp> line 1
Geth rakudo: e5568aa502 | (Elizabeth Mattijsen)++ | src/Perl6/Actions.nqp
Make sure we call .words on stringified var

By taking out the call to Stringy in the infix:<~> handling, any variable would be directly exposed to having ".words" called upon it in << >> post-processing. This works for IO::Paths, but doesn't work for many other classes:
... (5 more lines)		 08:37
lizmat Fixes R#2203
synopsebot R#2203 [open]: github.com/rakudo/rakudo/issues/2203 [⚠ blocker ⚠] Failure in t/spec/S17-procasync/stress.t
08:56 travis-ci joined, p6bannerbot sets mode: +v travis-ci
travis-ci Rakudo build passed. Elizabeth Mattijsen 'Make sure we call .words on stringified var 08:56
travis-ci.org/rakudo/rakudo/builds/415054989 github.com/rakudo/rakudo/compare/2...568aa50238 08:57
08:57 travis-ci left 09:55 Ven` joined 09:56 p6bannerbot sets mode: +v Ven` 09:59 Ven` left 10:01 Tux__ joined, |Tux| left 10:02 p6bannerbot sets mode: +v Tux__ 10:24 astj joined 10:25 p6bannerbot sets mode: +v astj 10:29 Ven` joined 10:30 p6bannerbot sets mode: +v Ven` 10:41 Ven` left 11:02 Ven` joined, p6bannerbot sets mode: +v Ven`
nine Fixing the JVM bootstrap issue does not exactly become easier when the JVM segfaults... 11:07
[Tux] Rakudo version 2018.06-428-ge5568aa50 - MoarVM version 2018.06-417-g1e4b2c84d
csv-ip5xs0.961 - 0.981
csv-ip5xs-207.652 - 8.226
csv-parser23.586 - 24.964
csv-test-xs-200.431 - 0.463
test8.270 - 8.942
test-t2.210 - 2.213
test-t --race0.940 - 1.030
test-t-2038.749 - 39.714
test-t-20 --race12.914 - 12.940
11:21 astj left
Geth nqp: c908d61d01 | (Stefan Seifert)++ | 11 files
Fix bootstrap on JVM

Needed the same changes as in commit 256cd26919c9e49a6ed866e2221a12a5a387489e		 11:25
MasterDuke nine++ 11:32
11:34 astj joined 11:35 p6bannerbot sets mode: +v astj 11:36 lizmat left 11:42 Ven` left
Geth nqp: 70c17426ed | (Stefan Seifert)++ | 10 files
Fix JVM bootstrap after rename of getstrfromname to strfromname

Needs to have both ops available and rebootstrap with that before one can remove the old op.		 11:56
12:07 pmurias joined 12:08 p6bannerbot sets mode: +v pmurias 12:11 travis-ci joined, p6bannerbot sets mode: +v travis-ci
travis-ci NQP build passed. Stefan Seifert 'Fix JVM bootstrap after rename of getstrfromname to strfromname 12:11
travis-ci.org/perl6/nqp/builds/415087850 github.com/perl6/nqp/compare/c908d...c17426ed7c
12:11 travis-ci left 12:12 astj left 12:13 astj joined 12:14 p6bannerbot sets mode: +v astj
MasterDuke nine++ 12:14
MasterDuke wonders what he's up to now, nine must have evolved into at least a twenty 12:16
12:16 Ven` joined 12:17 p6bannerbot sets mode: +v Ven` 12:18 astj left
jnthn Hm, after clean builds of everything at HEAD, the SEGVs I saw in make test yesterday are gone. 12:27
yoleaux 01:58Z <MasterDuke> jnthn: i also saw some weird failures/segvs in rakudo's `make test` that i couldn't repro. i even ran one of the files that failed in a loop overnight and it was still going fine in the morning
12:29 astj joined 12:30 p6bannerbot sets mode: +v astj 12:32 astj_ joined 12:33 p6bannerbot sets mode: +v astj_ 12:36 astj left 12:38 astj_ left 12:42 lizmat joined 12:43 p6bannerbot sets mode: +v lizmat
dogbert17 looks good on my system as well as long as I don't run with any flags set 12:43
jnthn My spectest run had a bunch of segv :/
dogbert17 now?
there are really strange things going on but I have to set some flags to make them show up 12:44
jnthn But I can't reproduce anything when running them alone :/ 12:45
dogbert17 and you don't have MVM_SPESH_NODELAY and MVM_SPESH_BLOCKING set?
jnthn I didn't when I ran make spectest 12:47
dogbert17 interesting, I have to have them set in order to see anything but some well known flapper 12:48
if we disregard github.com/MoarVM/MoarVM/issues/913 there's at least one other recent bug hiding 12:49
12:50 astj joined
dogbert17 for example, what do you see if you run MVM_SPESH_NODELAY=1 MVM_SPESH_BLOCKING=1 ./perl6 t/04-nativecall/06-struct.t 12:50
jnthn Really can't get any of them to fail :/
12:51 p6bannerbot sets mode: +v astj
dogbert17 try the one above :-) 12:51
lizmat jnthn: have you tried running a single flapper while having "make spectest" run at the same time ? 12:52
dogbert17 only 16 degrees and rain today 12:54
12:54 astj left
jnthn Yup 12:56
Yes, but I care about what's failing without any of the special flags
NODELAY isn't representative of normal operating conditons, so I'm less worried about it 12:58
dogbert17 agreed, I'm trying to get stuff to fail without any flags but haven't had much luck today 12:59
only the occasional flapper
jnthn Also, when I build with --debug=3 the make test failures seem to vanish 13:00
dogbert17 that's interesting
jnthn Oh no, I still see a failure, just less often 13:01
oh, but it was the test I'm running in a loop in gdb and it's because they raced over a test file 13:02
jnthn tries creating the memory/CPU pressure with spectest instead 13:03
dogbert17 heh, suddenly MoarVM doesn't build on 32 bit - ./libmoar.so: undefined reference to `MVM_jit_code_copy' 13:06
perhaps something for nine to ponder 13:07
MasterDuke i just got fails in t/spec/S17-supply/syntax.t (think that's a known flopper), t/spec/S17-procasync/nonexistent.t, t/spec/S32-io/chdir-process.t, t/spec/S29-os/system.rakudo.moar, t/spec/S32-list/roll.t, and t/spec/integration/advent2014-day16.t 13:08
AlexDaniel dogbert17: ticket pl 13:10
z
MasterDuke gist.github.com/MasterDuke17/b5726...3f303e1e29 backtrace from t/spec/S17-procasync/nonexistent.t 13:11
Geth ¦ rakudo: AlexDaniel self-unassigned 2018-08-10 Toast results / 29 burns github.com/rakudo/rakudo/issues/2196
13:16 Zoffix joined, p6bannerbot sets mode: +v Zoffix
dogbert17 AlexDaniel: will do 13:16
Zoffix Random thought: should we actually exclude NQP from the list of repos of the "perl6" team to which all the new members get assigned; oftentimes after just a couple of quality commits? The specific point of concern is that we bootstrap NQP. If a person had a malicious intent, it'd be pretty easy for them to get standard perl6 org membership and then just offer help to bootstrap NQP, while inserting crap 13:20
into it.
jnthn Zoffix: Yeah, I've sometimes worried a tad about that 13:25
AlexDaniel so the default permission in perl6 org is “write”, is it even possible to selectively have “read” for nqp repo without moving it to another org? 13:27
pmurias would we refuse to take a PR that rebootstrap?
* rebootstraps
AlexDaniel we can of course change the default permission, but that is going to affect all repos in perl6 org 13:28
13:28 pmurias left
Ulti give I just accidentally created and deleted a branch on NQP just because I have a load of perms I never wanted I can +1 the sentiment of altering the defaults 13:30
also its kind of crap that github branch search if it finds nothing defaults to create the branch
13:31 pmurias joined
Ulti why the F would anyone wishing to browse branches want to create them that way at origin 13:31
Zoffix AlexDaniel: what about adding "perl6" team to `nqp`'s collaborators and settting it to read perms?
Zoffix does it
13:32 p6bannerbot sets mode: +v pmurias
Zoffix pmurias: sure, why not. 13:32
timotimo Ulti: yes, that hit me once, too
Zoffix Why would PR refusal be taboo
AlexDaniel Zoffix: I don't think it's going to work, but try it
Ulti also let me know if I jsut deleted the actual JS branch ;____;
looks like its now known as nqp in the browser though 13:33
AlexDaniel Ulti: don't worry too much, if it was something important the author will push it again
Ulti it was set to be owned by me so fairly sure its fine
Zoffix Ulti: can you try doing more things? Does it block you now? 13:35
13:35 Ven` left
Ulti Zoffix: one sec will check 13:37
I can still create a branch
might just be I need to log in and out 13:38
Zoffix Actually, I see a message: "The Perl 6 organization has their default repository permission set to write. This means that every member of this organization has write access to this repository, regardless of the team and collaborator access specified below." 13:40
AlexDaniel ok I see what can be done
Ulti why do people need to be in the perl6 org though?
Zoffix AlexDaniel: what? 13:41
'cause I'm already in the process of changing it to read
pmurias are we even sure that we want to change the nqp repo to read?
AlexDaniel Zoffix: set default org permission to read, then go here github.com/orgs/perl6/teams/perl6/repositories then select all and set to “write”
Ulti yeah cant do it now 13:42
AlexDaniel oh?
Zoffix AlexDaniel: what does the permission on the right in that table signify?
13:43 Ven` joined
AlexDaniel Zoffix: that's for “perl6” team 13:43
13:43 p6bannerbot sets mode: +v Ven`
Zoffix OK. Done 13:44
looks like a couple of repos should really be in perl6-community-modules and not in perl6 13:46
org
AlexDaniel yeah I noticed that too
13:48 Ven` left
pmurias Zoffix: how are commiters now addable to nqp? 13:49
AlexDaniel there's team `core` and team `nqp` 13:50
Zoffix pmurias: add them to team nqp
pmurias so all the people from the perl6 have commits carried over? 13:51
AlexDaniel one thing that bothers me is that there are 307 people in the org but only 287 in `perl6` team
Zoffix pmurias: "the perl6"... what do you mean? The "perl6" team? No, they now have read perms only 13:52
pmurias Zoffix: I mean who is able to commit to nqp now? 13:53
AlexDaniel heh… I went to api.github.com/orgs/perl6/members in a hope that it will give me the full list, but no it does pagination by with only 30 elements per page :( 13:54
pmurias: anyone in this team github.com/orgs/perl6/teams/nqp/members or this team github.com/orgs/perl6/teams/core/members 13:55
although I'm not sure if these pages are visible to all
Zoffix pmurias: members of "core" team ( github.com/orgs/perl6/teams/core/members ), "nqp" team ( github.com/orgs/perl6/teams/nqp/members ), and "Org-Admins" team ( github.com/orgs/perl6/teams/org-admins/members )
pmurias Zoffix: how was the team created? 13:57
Zoffix pmurias: dunno, it already existed.
pmurias so why are we channing the set of commiters to list we don't know what it is? 13:59
Zoffix Because I doubt a group named "nqp" spontanously came into existence on its own. 14:00
14:01 Zoffix left
pmurias but are all the nqp committers in it? 14:02
AlexDaniel pmurias: good question. Now yes 14:13
I invited pretty much every contributor that had 2+ commits to the repo *and* was already part of perl6 org 14:15
most of them were already in nqp team
14:18 __idiot__ joined 14:19 p6bannerbot sets mode: +v __idiot__ 14:20 __idiot__ left
AlexDaniel another good question is how to get all 306 people in `perl6` org into `perl6` team 14:20
because getting the list of nicknames is… not that easy :)
I mean sure you can overcome pagination just with 10 clicks, but then your requests also need to be authed otherwise some members will be hidden 14:21
pmurias if we want to have more trust than "nobody can be bothered to subvert Perl 6" in our bootstrapped files we need to have reproducable bootstraps at some point 14:23
MasterDuke pmurias: somewhat related question, have you tried rebase the truffle branch onto master? if not, should that be done? 14:24
pmurias MasterDuke: I haven't done that. It should be done :) 14:26
(I have been focusing on porting over deserialization to nqp-truffle recently) 14:36
14:39 emilsp9 joined 14:40 p6bannerbot sets mode: +v emilsp9 14:41 emilsp9 left 14:45 Madcotto9 joined, p6bannerbot sets mode: +v Madcotto9 14:46 Madcotto9 left 15:01 Ven` joined 15:02 p6bannerbot sets mode: +v Ven` 15:34 Ven` left 15:45 this joined 15:46 p6bannerbot sets mode: +v this 15:47 this left
Geth star: 58e41cdeca | (Steve Mynott)++ | 2 files
apply rakudo patch c3e1ec0 to fix Windows module paths
 15:53
15:59 possiblyanowl joined, possiblyanowl left 16:08 Pidgeotto11 joined 16:09 Pidgeotto11 left
stmuk .tell Zoffix can you please fix rakudo.org to point to R* source/windows 2018.06.1 (note Mac DMG version unchanged). I don't seem to have access anymore 16:15
yoleaux stmuk: I'll pass your message to Zoffix.
jnthn Now I'm getting really suspicious something on my side is oddly busted. I've a couple of times had gcc segfault while compiling Moar 16:18
stmuk lizmat: for the p6weekly there is a fixed (for windows linenoise) rakudo star 2018.06.1
lizmat stmuk++ 16:19
weekly: there is a fixed (for windows linenoise) rakudo star 2018.06.1
notable6 lizmat, Noted!
jnthn reboots all the thing... 16:20
timotimo tee hee 16:30
the whole thing
stmuk oh rakudo.org seems to work for 2018.06.1 anyway 16:33
16:34 MasterDuke left
stmuk although the .1 isn't visible 16:34
16:37 LordLionM joined, LordLionM is now known as Guest15269 16:38 p6bannerbot sets mode: +v Guest15269 16:40 Ven` joined 16:41 p6bannerbot sets mode: +v Ven` 16:42 Guest15269 left 16:44 Ven` left
jnthn Yes, the reboot helped. What a waste of time. 16:58
timotimo ;( 16:59
that's disheartening
jnthn Good news for your commits though :P
timotimo yeah
jnthn timotimo++
timotimo i'd've prefered the commits to have been faulty and having figured it out in five minutes
btw i've looked at a spesh and jit log of jdv's code, and spesh is, like, really really busy in it 17:00
but not in the way that it's doing bogus logging of frames that should really have been speshed
it's just that doing all those evals over and over makes every single piece of the compiler so much warmer
Latest statistics for 'set_name' (cuid: 19, file: gen/moar/Metamodel.nqp:102) 17:03
[...]
Type tuple 132
Latest statistics for 'compose' (cuid: 51, file: gen/moar/Metamodel.nqp:273)
Type tuple 65
pretty amazing 17:04
jnthn So megamorph :)
Though at some point it should give up and make a certain specialization 17:05
timotimo mega morphin' bytecode rangers 17:06
whee, the maximum stack depth climbs over 5k 17:08
using set_name as an example, it doesn't ever make a certain specialization, but it also only reaches 146 hits 17:09
at which point it has, unsurprisingly, 145 type tuples
jnthn 5k...that sounds...unlikely
timotimo shall i investigate by doing some random MVM_dump_backtrace calls in gdb? 17:10
jnthn Ah, to see if it's for real? Can try, but I suspect the stack simulation is going pear-shaped somehow
btw, did you see I fixed the real "super deep stack" thing some weeks back that made profiles of some parallel stuff look odd? 17:11
timotimo oh?
i must have missed that
i was convinced it came from throwing past the prof_exit op 17:12
17:12 MasterDuke joined, p6bannerbot sets mode: +v MasterDuke
jnthn I think that's a separate problem 17:12
timotimo OK
jnthn This one was with loads of frames in ThreadPoolScheduler
timotimo oh!
yes, i know what you mean 17:13
17:13 MasterDuke left, MasterDuke joined, herbert.freenode.net sets mode: +v MasterDuke, p6bannerbot sets mode: +v MasterDuke
timotimo i did see that 17:13
that was very good
should also help memory consumption
jnthn indeed!
dinner, bbl
timotimo have a good one!
today was not a sensible day, which means i've had breakfast ~1.5h ago :|
here's the biggest guard tree i've seen yet: optimizer's visit_children has 44 nodes in its guard tree 17:14
Geth rakudo: d8d51d0da2 | (Nick Logan)++ | src/core/CompUnit/RepositoryRegistry.pm6
Revert "Use $*SPEC.dir-sep for separators in CompUnit::RepositoryRegistry"

This reverts commit c3e1ec0bb86fad16ecca6ca5997d81cd6afc2ba5.
First, I do not understand why changing the directory separator should be needed as internally it doesn't matter anywhere else. Second, things like `"inst#$prefix/vendor` are not paths ( its a cur spec ) so changing them explicitly seems strange -- we don't use `$*SPEC.dir-sep;` for meta6 provides data for instance. Third, the line `@binaries = flat @installations.map: { .script("bin{$sep}$script", :$name) };` will not work because we
I'm not sure what problem this intended to solve, but this is likely not the spot to go about solving it.		 17:16
stmuk github.com/rakudo/rakudo/commit/d8...t-30047587 17:21
looks like a rakudo star 2018.06.2 may be needed :/ 17:34
or maybe just remove Windows binary until next release 17:35
timotimo this commit doesn't fix that problem? 17:38
hm, maybe the nqpbook repository can actually go %) 17:40
17:42 Zoffix joined, p6bannerbot sets mode: +v Zoffix
Zoffix Just ran a stresstest on a 160-core box... Looking at htop, it lasts about 40s, but then a few cores sit and wait for some test files. In fact, t/spec/S17-channel/stress.t just hung entirely 17:50
Oh, nm, it just seemed like forever
But total run time was Files=1304, Tests=152985, 140 wallclock secs (32.93 usr 4.44 sys + 3708.72 cusr 221.77 csys = 3967.86 CPU) vs 167s on 24-core box
ZOFFLOP: t/spec/S02-types/mixhash.t 17:51
timotimo not such a gigantic win, eh?
that's what 40% time saved?
Zoffix m: say 40/167
camelia 0.239521
Zoffix like 77% if you count the bulk. The rest, I'm guessing it's just single test files taking long to run 17:52
stmuk I'm deleting R* 2018.06.1
Zoffix ok
stmuk: FWIW, you should now have correct perms on rakudo.org repo
stmuk ok thanks I can test by reverting the last two changes! 17:55
works 17:56
I should have tested more :-\ "but just a little change what damage can it cause" :) 17:59
18:06 TriJetScud12 joined, p6bannerbot sets mode: +v TriJetScud12 18:07 TriJetScud12 left
stmuk It's not even Aug 13th yet 18:08
18:41 reportable6 left 18:49 wook_ joined 18:50 p6bannerbot sets mode: +v wook_ 18:51 wook_ left 18:52 Zoffix left 18:53 cronic joined, cronic left 19:03 RoyK20 joined, RoyK20 left 19:08 Ven` joined, p6bannerbot sets mode: +v Ven` 19:12 Ven` left 19:14 sirnaysayer26 joined 19:15 p6bannerbot sets mode: +v sirnaysayer26 19:20 sirnaysayer26 left 19:44 Ven` joined 19:45 p6bannerbot sets mode: +v Ven` 19:49 Ven` left 19:56 robertle left 20:20 Ven` joined 20:21 p6bannerbot sets mode: +v Ven` 20:26 Ven` left 20:50 bray90820_ joined, bray90820_ left 20:57 Ven` joined, p6bannerbot sets mode: +v Ven` 21:01 Ven` left 21:15 Kaiepi left, Kaiepi joined 21:16 p6bannerbot sets mode: +v Kaiepi 21:33 Ven` joined 21:34 p6bannerbot sets mode: +v Ven` 21:38 Ven` left 22:09 Ven` joined 22:10 p6bannerbot sets mode: +v Ven` 22:13 channels joined, channels left 22:14 Ven` left 22:18 rigel118 joined 22:19 p6bannerbot sets mode: +v rigel118
rigel118 8,8 0,0 8,8 22:19
8,8 0,0 1,1 0,0 8,8
8,8 0,0 1,1 0,0 1,1 0,0 8,8
8,8 0,0 12,12 8,8
8,8 12,12 0,0 12,12 8,8
8,8 12,12 0,0 12,12 8,8
8,8 12,12 0,0 12,12 8,8
8,8 12,12 1,1 12,12 0,0 12,12 8,8
8,8 12,12 1,1 12,12 0,0 12,12 8,8
8,8 12,12 0,0 12,12 8,8
8,8 12,12 11,11 12,12 1,1 12,12 8,8
8,8 12,12 1,1 12,12 11,11 0,0 11,11 12,12 1,1 12,12 8,8
8,8 12,12 1,1 12,12 11,11 12,12 1,1 12,12
8,8 12,12 1,1 12,12 11,11 0,0 11,11 12,12 1,1 12,12 22:20
8,8 12,12 1,1 12,12 11,11 12,12 1,1 12,12
1,0 kloeri: and I'm tired of you guys
1,0 thinking you can ban staff
1,0 when we complain about your spam
22:20 rigel118 left 22:46 Ven` joined, p6bannerbot sets mode: +v Ven` 22:50 Ven` left
lizmat sigh 22:52
22:57 lizmat_ joined 22:58 p6bannerbot sets mode: +v lizmat_
releasable6 Next release in ≈5 days and ≈19 hours. 4 blockers. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 23:00
23:01 lizmat left 23:22 Ven` joined 23:23 p6bannerbot sets mode: +v Ven` 23:27 Ven` left 23:33 Kaiepi left 23:38 Kaiepi joined, p6bannerbot sets mode: +v Kaiepi 23:58 Ven` joined 23:59 p6bannerbot sets mode: +v Ven`

Powered by Cro and the Raku® Programming Language.

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

Thank you!