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.
lizmat Files=1256, Tests=81347, 345 wallclock secs (16.24 usr 5.68 sys + 2436.46 cusr 227.57 csys = 2685.95 CPU) 07:06
yoleaux 21 Nov 2018 23:02Z <SmokeMachine> lizmat: I am thinking on doing something like this, what do you think? gist.github.com/FCO/8badb8ffc1ff86...7963f37fd2
21 Nov 2018 23:04Z <SmokeMachine> lizmat: but Im not sure how META6.json would handle that... should Bla, Ble and Bli (on that example) be resources?
lizmat .tell SmokeMachine not sure, and no time to look at as commuting to London 07:07
yoleaux lizmat: I'll pass your message to SmokeMachine.
AlexDaniel DrForr: any thoughts on github.com/rakudo/rakudo/issues/2499 ? 12:37
one way to work around is to change `Foo` in tests to unique names
but that doesn't feel right
I'm actually not sure what that issue bisects to. Once github.com/drforr/perl6-Perl6-Parser/pull/17 is merged that'd be easy to figure out, but otherwise I have to do some gymnastics 12:41
lucasb everything's ok? :) is it just pending a moar release? 12:53
6.d will be the default lang in this release?
timotimo yes, 6.d will be the default 12:55
releasable6: status
releasable6 timotimo, Next release will happen when it's ready. 2 blockers. 82 out of 172 commits logged (⚠ 1 warnings)
timotimo, Details: gist.github.com/cc553d47ee05b89f03...3b3ecddb50
AlexDaniel lucasb: we need to figure out what to do with these two blockers 12:56
then it's easy
lucasb ok. sorry to be anxious :) 12:57
|Tux| Rakudo version 2018.10-172-ge7ce194a7 - MoarVM version 2018.10-91-g8c67e1697
csv-ip5xs0.885 - 0.906
csv-ip5xs-206.661 - 6.887
csv-parser21.369 - 21.543
csv-test-xs-200.420 - 0.441
test7.865 - 8.036
test-t1.764 - 1.791
test-t --race0.807 - 0.816
test-t-2029.098 - 30.004
test-t-20 --race10.155 - 10.488
13:21
AlexDaniel releasable6: status 14:01
releasable6 AlexDaniel, Next release will happen when it's ready. 1 blocker. 82 out of 172 commits logged (⚠ 1 warnings)
AlexDaniel, Details: gist.github.com/d3c95081c6e4070679...5d5bca543e 14:02
AlexDaniel alright, the last one is even weirder I think…
jnthn AlexDaniel: Bad news: running something close to HEAD on a big $dayjob codebase, I get spurious test failures now and then when pieces of it need to be re-compiled. Then run them again and all passes. Not quite sure when exactly it regressed, nor how on earth I can golf it. 14:24
There's the occasional SEGV, the occasional complaint about get_cache being called on a weird type, and even the odd global merge failed. 14:25
Seems to never happen if the modules are all compiled
AlexDaniel jnthn: I have not seen any big issues during the last run of Blin, but since then there were some commits in MoarVM, maybe something there? 14:29
jnthn AlexDaniel: Maybe, I'm not absolutely at HEAD either, that said... 14:31
I'll compile very latest when I get chance to see if it helps
dogbert21 Perhaps it's the duplicate label problem
jnthn Also, it's possible that this is not a regression at all, and I just have a large enough codebase combined with 12-degree parallel testing to have exposed an existing issue.
I'll try and find out a bit more. 14:32
tbrowder___ .seen DrForr 14:43
yoleaux I saw DrForr 5 Oct 2018 17:53Z in #perl6: <DrForr> Sounds like something for the docs :)
dogbert21 jnthn: do you run your $dayjob system with automatic core dump generation, i.e. ulimit -c unlimited? 14:54
perhaps there's a coredump you can take a look at
jnthn dogbert21: Actually it was a panic, thinking about it, not a core dump 15:09
But that's by far the rareest of the cases
Installing HEAD now before I continue working to see if it happens there or if we're lucky and some fix I didn't have helped it 15:10
ooh, good news, HEAD seems potentially better...at least, first couple of times changing something deep in the dependency chain and running tests didn't have any spurious failures 15:20
Maybe it was "[JIT] Fix trunc to cast upwards" that helped...I think we had some other test mis-compiles that were fixed by that 15:21
I thought I was running with that fix, but apparently not. Sorry for the noise. 15:22
AlexDaniel that makes sense 15:28
and I still don't understand R#2499 15:29
synopsebot_ R#2499 [open]: github.com/rakudo/rakudo/issues/2499 [⚠ blocker ⚠] Failures in Perl6::Parser module 15:30
Geth nqp: 25e518596f | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/build/MOAR_REVISION
[MoarVM Bump] 8c67e1697 Revert "Remove broken […]

MoarVM bump brought: github.com/MoarVM/MoarVM/compare/2...g8c67e1697
15:48
¦ nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...g8c67e1697
rakudo: 8e27d5f5b0 | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/build/NQP_REVISION
[NQP Bump] 25e518596 [MoarVM Bump] 8c67e1697 R […]

NQP bump brought: github.com/perl6/nqp/compare/2018....g25e518596
¦ rakudo: version bump brought these changes: github.com/perl6/nqp/compare/2018....g25e518596
AlexDaniel if anyone feels like helping, maybe pay attention to #perl6 :) I don't know what's going on but there are so many new people there 22:04