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.
vrurg 2019.03 doesn't build on macOS with MacPorts. 00:30
Ok, found it's moar, found the README. Bailing out because of "wong" make?? Seriously??? 00:41
vrurg is furios
s/wong/wrong/ 00:48
timotimo vrurg, you mean ports vs xcode? 00:51
perhaps the detection of that pathological situation is a bit too sensitive 00:52
but bailing out early is much preferable to getting the other failure mode
vrurg aha. MacPorts installs gnu make as a dependency. This is normal. XCode uses GNU make too. What was the point of checking for make in Configure.pl?
timotimo which is a message that is just wrong and completely misleading
vrurg I'm thinking to patch it. Maybe with a hack to use /usr/bin/make if cc is found in that location. 00:54
timotimo github.com/MoarVM/MoarVM/pull/1055
check the pr
vrurg And anyway I totally agree to the need of validating ar/ld/cc for consistency.
timotimo: thank! I found the commit, it's self-explanatory. 00:55
timotimo ok, if checking make doesn't make sense, a patch would be welcome, perhaps even a short comment I. the file 00:56
for now I will go to bed and may not be able to do anything all day tomorrow
if I'm not reacting to your patch, it's not you, it's me ;)
vrurg timotimo: no problem. Thanks! :) 00:57
AlexDaniel vrurg: um, sorry, so is it a new issue? 01:42
or was it just a problem with your setup?
AlexDaniel is confused
vrurg AlexDaniel: it's a new one. Arises from a good intention to prevent a mixture of XCode and gnu tools. 01:43
AlexDaniel hmm
vrurg timotimo pointed to the pull request which introduced it. I'm trying to figure out a better way to achieve the goal.
Basically, it's either about ignoring make which from another toolchain or about enforcing the right one. 01:44
Though I would probably combine both and make rakudo and moar rely on /usr/bin/make if /usr/bin/cc or clang is used. 01:45
01:54 lgtaube left 01:55 lgtaube joined
Geth rakudo: vrurg++ created pull request #2750:
Follow up to PR MoarVM/MoarVM#1067
02:44
vrurg bisectable6: multi f(Str) { say "Str" }; multi f(Int) { say "Int" }; &f.candidates[1].wrap( sub (|) { note "Wrapper"; callsame } ); f(42); CATCH { when X::AdHoc { exit 0 } } 03:45
bisectable6 vrurg, On both starting points (old=2015.12 new=70d61b2) the exit code is 0 and the output is identical as well
vrurg, Output on both points: Ā«Ā»
vrurg bisectable6: help
bisectable6 vrurg, Like this: bisectable6: old=2015.12 new=HEAD exit 1 if (^āˆž).grep({ last })[5] // 0 == 4 # See wiki for more examples: github.com/perl6/whateverable/wiki/Bisectable
vrurg bisectable6: good=2018.12 bad=2019.03 multi f(Str) { say "Str" }; multi f(Int) { say "Int" }; &f.candidates[1].wrap( sub (|) { note "Wrapper"; callsame } ); f(42); CATCH { when X::AdHoc { say "ok"; exit 0 } } 03:50
bisectable6 vrurg, On both starting points (old=2018.12 new=2019.03) the exit code is 0 and the output is identical as well
vrurg, Output on both points: Ā«okā¤Ā»
vrurg bisectable6: good=2018.12 bad=70d61b21446fb5156d0b00a9187de7b04add6b8c multi f(Str) { say "Str" }; multi f(Int) { say "Int" }; &f.candidates[1].wrap( sub (|) { note "Wrapper"; callsame } ); f(42); CATCH { when X::AdHoc { say "ok"; exit 0 } } 03:51
bisectable6 vrurg, On both starting points (old=2018.12 new=70d61b2) the exit code is 0 and the output is identical as well
vrurg, Output on both points: Ā«okā¤Ā»
vrurg m: multi f(Str) { say "Str" }; multi f(Int) { say "Int" }; &f.candidates[1].wrap( sub (|) { note "Wrapper"; callsame } ); f(42); CATCH { when X::AdHoc { say "ok"; exit 0 } } 03:52
camelia ok
vrurg m: $*PERL.compiler.version.say 03:54
camelia v2019.03.22.g.70.d.61.b.214
MasterDuke .ask pmurias i'm trying to get the jvm backing working with my default-nqp-to-int branch, but am getting an error i'm not familiar with. i've got some backtrace from before any code change, backtrace from after code change, and the change diff here gist.github.com/MasterDuke17/b69bf...473842af1. any suggestions? 04:11
yoleaux MasterDuke: I'll pass your message to pmurias.
04:54 dhyan_nataraj joined 07:07 lizmat joined 07:23 jdv79 left 07:37 lizmat left 07:39 sortiz joined 07:40 jdv79 joined
AlexDaniel 6c: multi f(Str) { say "Str" }; multi f(Int) { say "Int" }; &f.candidates[1].wrap( sub (|) { note "Wrapper"; callsame } ); f(42); CATCH { when X::AdHoc { exit 0 } } 08:38
committable6 AlexDaniel, Ā¦6c (37 commits): Ā«Ā»
AlexDaniel vrurg: ā†‘ this way it's a bit easier to see what was going on 08:39
so on all releases since 2015.12 the result is the same
Geth Ā¦ rakudo: AlexDaniel self-unassigned Issues with contributors generating script github.com/rakudo/rakudo/issues/2028 08:43
gfldex I got a script that can segfault moarvm quite fast. Needs linux and iostat. If a single line is commented out, it runs the whole night without a segfault: raw.githubusercontent.com/gfldex/b.../iostat.p6 09:08
Also it's quite resistent to golfing (or I would have). 09:09
AlexDaniel gfldex: so I'm running that in a loop and nothing is happening, what am I doing wrong? 09:24
gfldex AlexDaniel: ohh, please start it with 1 a parameter 09:28
it tends to segfault after 4-8 secs for me. Sometimes it takes a little longer.
I golfed it a little more. Update is on github. 09:33
AlexDaniel yeah, I see a segfault 09:58
gfldex: it'd help if you manage to include JSON::Stream in that golfed file
gfldex AlexDaniel: That module is not mine and fairly big. I should be able to simulate it and get rid of shelling out altogether. 10:04
10:04 pmurias joined
AlexDaniel gfldex: yeah but once it is all in the same script we'll have an actual golf 10:04
and we'll be able to throw that into committable 10:05
wellā€¦ we can also make it work with a module, too 10:06
I'm just not in the mood :D
pmurias MasterDuke: I don't have a direct solution, an obvious solution would be figure out what that type is 10:07
yoleaux 04:11Z <MasterDuke> pmurias: i'm trying to get the jvm backing working with my default-nqp-to-int branch, but am getting an error i'm not familiar with. i've got some backtrace from before any code change, backtrace from after code change, and the change diff here gist.github.com/MasterDuke17/b69bf...473842af1. any suggestions?
gfldex AlexDaniel: No luck. When I take the module out the segfault goes away. 10:25
|Tux| Rakudo version 2019.03-22-g70d61b214 - MoarVM version 2019.03-7-gabdddb018
csv-ip5xs0.739 - 0.761
csv-ip5xs-206.204 - 6.408
csv-parser22.317 - 22.650
csv-test-xs-200.431 - 0.432
test7.778 - 7.791
test-t1.870 - 1.935
test-t --race0.837 - 0.989
test-t-2032.163 - 32.491
test-t-20 --race10.194 - 10.894
10:38
10:45 lizmat joined 10:52 lizmat left 11:07 lizmat joined 11:47 AlexDaniel left 11:54 lizmat left 12:19 lizmat joined 12:36 lizmat left 12:37 lizmat joined, lizmat left 12:39 lizmat joined 12:41 pmurias left 12:57 lizmat left 13:02 AlexDaniel joined 13:06 lizmat joined, pmurias joined 13:13 lizmat left 13:30 lizmat joined 13:31 lizmat left 13:48 lizmat joined 13:49 lizmat left 13:52 lizmat joined, lizmat left 13:56 lucasb joined, lizmat joined 13:57 lizmat left 13:58 lizmat joined 14:03 lizmat left 14:07 pmurias left, lizmat joined 14:08 pmurias joined 14:09 lizmat left 14:11 lizmat joined, lizmat left 14:14 robertle joined 14:22 lizmat joined, lizmat left 14:23 lizmat joined, lizmat left 14:32 lizmat joined 14:33 lizmat left 14:34 sortiz left 14:49 lizmat joined 14:53 lizmat left 15:24 pmurias left 15:50 patrickb joined
patrickb o/ 15:50
Congratulations all around! Great to have a release again.
I don't have much time in the coming week for post merge fixups, but I'd like to get my relocatable branches merged soon after, given there is are no other objections to the merge. 15:51
15:57 patrickb left 16:24 AlexDaniel left 17:12 |Tux| left 17:18 |Tux| joined 18:03 dhyan_nataraj left 18:05 lucasb left 18:11 AlexDaniel joined
AlexDaniel gfldex: submit a ticket anyway, maybe someone else will have more luck 18:13
vrurg AlexDaniel: thanks for the advice. 18:16
vrurg was trying to locate a leftover of own experiments, as it turend out...
Kaiepi has 2019.03 been released yet? 18:20
robertle yes: github.com/rakudo/rakudo/releases 18:27
Kaiepi perfect
can someone review github.com/rakudo/rakudo/pull/2478 and the related pullreqs?
AlexDaniel robertle: the signature is going to change for the next release 18:28
robertle: I'll let you know as soon as it's 100% clear who is going to do the next release :) 18:29
robertle what do you mean with "signature"? I don't understand... 18:31
AlexDaniel robertle: gpg stuff, there's automation in debian that does some checks, right? 18:34
or am I making stuff up? :) 18:35
this stuff sources.debian.org/src/rakudo/2018.../upstream/ 18:38
18:44 cfa joined
cfa morning all 18:44
m: sub panic-stations { await start { die "ā˜ "; CATCH { default { return #`(oops!) } } } }; panic-stations
camelia MoarVM panic: Internal error: Unwound entire stack and missed handler
cfa this seems lta; encountered during a refactor
i think the panic's been there or a long time
bisectable6: help 18:45
bisectable6 cfa, Like this: bisectable6: old=2015.12 new=HEAD exit 1 if (^āˆž).grep({ last })[5] // 0 == 4 # See wiki for more examples: github.com/perl6/whateverable/wiki/Bisectable
cfa bisect: sub panic-stations { await start { die "ā˜ "; CATCH { default { return #`(oops!) } } } }; panic-stations
bisectable6 cfa, Bisecting by output (old=2015.12 new=70d61b2) because on both starting points the exit code is 1
robertle AlexDaniel: hm, but if noone signed the original tarball on github, where we get it from?
bisectable6 cfa, bisect log: gist.github.com/3a1f1d15dc73a63a1b...9f11b444ef 18:46
cfa, (2016-11-01) github.com/rakudo/rakudo/commit/b4...7b49ddf957
AlexDaniel robertle: from here? rakudo.org/files/rakudo
cfa oh, even simpler
m: sub { await start { return } }()
camelia MoarVM panic: Internal error: Unwound entire stack and missed handler
AlexDaniel robertle: also, all my commits and tags are signed
cfa vs. 18:47
m: await start { return }
camelia An operation first awaited:
in block <unit> at <tmp> line 1

Died with the exception:
Attempt to return outside of any Routine
in block at <tmp> line 1
AlexDaniel robertle: so you can also check it in git by looking at the tagged commit
actually, the tag itself is also signed!
everything is signed. Everything! 18:48
6c: sub { await start { return } }()
committable6 AlexDaniel, gist.github.com/99ac72bb06c33274df...b56117bbd3 18:49
AlexDaniel cfa: wow, this is really simple, but I haven't seen a bug report for that! Yeah, the code of course doesn't make much sense, but MoarVM panic is not the right answer :) 18:50
cfa right
i mean, the toplevel versoin gives a reasonable error
my refactor was closer to the first golf
where a return ended up inside an await start { }
version* 18:51
robertle right, I was confused. we do get the files from rakudo.perl6.org/downloads/rakudo/ where they do have signatures. who will be doing the releases in the future?
AlexDaniel actually, I don't understand why the top level version works but the same thing in a sub doesn't
cfa me neither!
but the error makes sense 18:52
the panic does not (from a user perspective ofc)
AlexDaniel robertle: there's a volunteer but it's not official yet
robertle awesome! viele haende schnelles ende 18:54
AlexDaniel cfa: soo, file a bug report? :) 19:02
cfa yep 19:10
(sorry, was afk)
m: sub { start { return } }(); sleep 1; 19:12
camelia Unhandled exception in code scheduled on thread 4
Attempt to return outside of immediately-enclosing Routine (i.e. `return` execution is outside the dynamic scope of the Routine where `return` was used)
in block at <tmp> line 1
cfa interesting contrast there too
19:15 dhyan_nataraj joined 19:18 dhyan_nataraj left, dhyan_nataraj joined
cfa github.com/rakudo/rakudo/issues/2753 19:18
issue title can probably use work :) 19:23
19:45 TreyHarris joined 20:28 dhyan_nataraj left 20:35 go|dfish left 20:48 go|dfish joined
MasterDuke there's a small chance i could go to either the swiss perl workshop or the europe perlcon. any recommendation for one over the other? 22:05
timotimo i know i will not be at riga, i don't know if i will make it to SPW 22:10
OTOH, riga has the potential to draw a much bigger crowd in general
MasterDuke any thought on the locations (non-conf stuff to see/do)? 22:11
timotimo somehow managed a segfault in readint 22:13
no thoughts
only dinner
22:31 epony left 22:55 squashable6 left 22:57 squashable6 joined
MasterDuke .tell pmurias it's a P6num (not surprisingly) 23:17
yoleaux MasterDuke: I'll pass your message to pmurias.
23:19 cfa left 23:43 SergiusUA joined 23:48 lizmat joined 23:58 lizmat left