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.
00:05 vrurg joined 00:10 lucasb left
Geth rakudo: Kaiepi++ created pull request #2890:
Don't assign to $*IN/$*OUT/$*ERR, bind instead		 01:00
04:23 skids joined 04:35 skids left 04:46 skids joined
Geth nqp/master: 5 commits pushed by (Nick Logan)++, (Christian Bartolomäus)++ 05:18
bartolin let's see what travis thinks about this. maybe we have to restore the 'travis_wait 30' and maybe the jdk-settings need tweaking. I've never fiddled with travis before, but it should be doable 05:21
05:26 skids left 05:31 robertle left
bartolin looks good \o/ 05:35
Geth rakudo/java_9: 090fb572ed | usev6++ | 2 files
Require Java 9 for JVM backend
 06:14
rakudo: usev6++ created pull request #2891:
Require Java 9 for JVM backend		 06:16
nqp: usev6++ created pull request #536:
[JVM] Don't warn about illegal reflective access		 06:17
07:14 patrickb joined
lizmat Files=1262, Tests=107941, 410 wallclock secs (29.84 usr 7.15 sys + 2910.71 cusr 255.15 csys = 3202.85 CPU) 09:32
|Tux| Rakudo version 2019.03.1-317-gb2e72d6d7 - MoarVM version 2019.05-4-g807b92f92
csv-ip5xs0.702 - 0.710
csv-ip5xs-205.701 - 5.794
csv-parser21.565 - 22.256
csv-test-xs-200.424 - 0.445
test6.618 - 6.721
test-t1.660 - 1.667
test-t --race0.797 - 0.870
test-t-2027.574 - 28.936
test-t-20 --race9.068 - 9.435
 09:54
11:05 patrickb left
lizmat looks like modules.perl6.org hasn't been updated since Mon May 6 12:12:44 2019 11:15
could someone give it a kick?
11:18 |Tux| left
timotimo sure 11:19
11:23 |Tux| joined 11:28 vrurg left
lizmat timotimo++ 11:34
timotimo modules page should be updated now 12:19
12:20 patrickb joined
lizmat timotimo: thanks! 12:22
13:00 sivoais left 13:01 sivoais joined 13:02 Kaiepi left, Kaiepi joined
lizmat notable6: weekly 13:24
13:25 vrurg joined
notable6 lizmat, 5 notes: gist.github.com/b9dcdf043e9311e643...8f1148ab7c 13:25
13:42 skids joined
lizmat reportable6: 2019-05-05T00:00:00Z 2019-05-13T00:00:00Z 13:44
reportable6 lizmat, Couldn't find a snapshot for 2019-05-05T00:00:00Z (try ｢list｣ command to see what's available).
lizmat reportable6: 2019-05-06T00:00:00Z 2019-05-13T00:00:00Z
reportable6 lizmat, Couldn't find a snapshot for 2019-05-06T00:00:00Z (try ｢list｣ command to see what's available).
lizmat reportable6: list
reportable6 lizmat, gist.github.com/6ab1d7cd555b4163ee...3ebadcbc47
lizmat reportable6: 2019-05-06T12:48:00Z 2019-05-13T00:00:00Z
reportable6 lizmat, OK, working on it! This may take up to 40 seconds
lizmat, gist.github.com/a8202bab4c6d889b85...dba91bd2d5 13:45
lizmat notable6: weekly reset 13:47
notable6 lizmat, Moved existing notes to “weekly_2019-05-13T13:47:00Z”
14:02 llfourn left 14:07 llfourn joined
lizmat And another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2019/05/13/...pumpkings/ 14:23
japhb lizmat++ 14:24
lizmat afk for a few hours& 14:25
15:38 patrickb left 16:22 patrickb joined
vrurg What is the current plan for release? 16:49
17:00 robertle joined
timotimo this weekend, i think? 17:15
or was it supposed to be the weekend that just passed? 17:16
vrurg timotimo: the latter. 17:20
That's why I'm asking. It's not the release itself that I need, but wanna get unblocked on my other work. 17:21
AlexDaniel kawaii: ↑ so what's the situation? 17:30
I think it may be reasonable to reschedule it to June 20th or whatever the regular date is 17:31
dogbert17 AlexDaniel: are you extremely busy? 17:32
AlexDaniel dogbert17: yeah
dogbert17 ok
Geth nqp: c88c64add0 | usev6++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java
[JVM] Don't warn about illegal reflective access

  ... on Java 9 and newer.
nqp: b38a28bcfd | (Christian Bartolomäus)++ (committed using GitHub Web editor) | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java
Merge pull request #536 from perl6/jvm_beyond_java_8

  [JVM] Don't warn about illegal reflective access
kawaii AlexDaniel: I was going to take your suggestion and make it a June release
AlexDaniel kawaii: just change it in the release guide then, it'll unblock people from doing great things :)
kawaii okay, will do :)
AlexDaniel the bot will pick it up automatically after a minute or two
vrurg: ↑ 17:33
Geth rakudo: 292235ce80 | (Kane Valentine)++ (committed using GitHub Web editor) | docs/release_guide.pod
Update release_guide.pod
kawaii done!
AlexDaniel I'll be able to give full attention from June 20th till July 1st :) 17:34
vrurg AlexDaniel: thanks for the note!
Can we proceed with R#2816 followed by R#2852? 17:35
synopsebot R#2816 [open]: github.com/rakudo/rakudo/pull/2816 Revamp of Configure.pl
R#2852 [open]: github.com/rakudo/rakudo/pull/2852 Add revision 6.e and improve multi-revision support
patrickb vrurg: I guess we can! 17:37
vrurg: Are you in for some post merge fallout fixups?
AlexDaniel releasable6: status 17:38
releasable6 AlexDaniel, Next release in ≈38 days and ≈1 hour. 1 blocker. 140 out of 318 commits logged (⚠ 2 warnings)
AlexDaniel, Details: gist.github.com/d31342f428f62b9743...4f57807de4
vrurg patrickb: I need to meet a screen repair guy and will be available after that.
patrickb :-P 17:39
AlexDaniel I guess the question was if you're ready to be fixing things during the next 38 days? :) 17:40
patrickb It was. I'll merge.
AlexDaniel ♥
Geth rakudo/master: 67 commits pushed by (Vadim Belman)++, (Patrick Böker)++
review: github.com/rakudo/rakudo/compare/2...43b8be148b
patrickb r#2852 is conflicting now :-/ 17:41
synopsebot R#2852 [open]: github.com/rakudo/rakudo/pull/2852 Add revision 6.e and improve multi-revision support
vrurg patrickb: Yep, I didn't have time to tell to postpone 2852 until I merge 17:57
patrickb vrurg: I didn't do anything about 2852. So no harm done. :-) 17:58
vrurg AlexDaniel: you're right, but the first problems are right here right now. ;)
I need 10 more mins and will prepare 2852 then. 17:59
MasterDuke gfldex: re github.com/rakudo/rakudo/issues/27...-491905807 what's your rakudo version? istr a fix for something like that landing recently 18:05
18:09 vrurg left 18:10 vrurg joined
vrurg AlexDaniel: could you please review 2852 when have time? Especially github.com/rakudo/rakudo/pull/2852...9cc935e02b – I'm not sure if I understood the logic of ModuleLoader correctly with respect to loading settings. 18:14
AlexDaniel vrurg: maybe ask jnthn or lizmat
I'd be glad to help but not now or any time soon 18:15
vrurg Ah, right. Sorry, forgot about it.
18:19 pyrimidine joined
vrurg patrickb: Due to the messy day, I forgot to tell that github.com/perl6/nqp/pull/531 should accompany 2816. 18:22
patrickb vrurg: Did you see the travis error on that PR? 18:24
vrurg: Also I don't have the priviledges to merge that PR... 18:25
gfldex MasterDuke: HEAD
MasterDuke gfldex: oh well, thought maybe there was a quick fix 18:26
vrurg patrickb: jvm... I may need some help with it. 18:28
vrurg never used Java.
patrickb I think the first two build jobs are not java... 18:29
Don't be fooled by the title. 18:30
vrurg I'm not worried about those two. 18:31
But java is problematic. 18:33
bartolin the failure for the seconde (and fourth) jvm build is expected. NQP master requires java 9 since earlier today. 18:43
and 'make test' didn't work before. so it looks as before the merge for the JVM backend, doesn't it? 18:45
vrurg bartolin: I'm fixing it for java 9 right now. Otherwise I had successfull builds of jvm earlier. 18:47
bartolin well, maybe I looked at the wrong place. what goes wrong for the jvm build at the moment? 18:48
MasterDuke gfldex: does --optimize=0 change anything?
bartolin vrurg: are you aware that there is a branch that should be merged after the next bump for nqp -- to make travis use java 9 to build rakudo? 18:52
vrurg bartolin: No, I'm not. 18:53
Ok, I'm into fixing a smaller problem for now. And would leave java 8 as the default for now. Would I need to pre-merge that branch to make travis happy when I finally switch to 9? 18:55
bartolin no problem, we (I) can do that later. 18:56
vrurg bartolin: thanks! May I ping you later today if help with jvm needed? 18:58
bartolin vrurg: yes, please do. I'll try to help -- unless I already went to bed :) in that case I could take a look tomorrow. 19:00
from my perspective we could fix arising problems for the jvm backend after a merge. 19:01
patrickb bartolin: That's great! You're a better fit to help with JVM stuff than I am. Thanks! 19:06
vrurg bartolin: that's the best gift I could dream of! ;) I'll try to get it compiling for me anyway. 19:07
Will step back only if it takes too long. 19:08
bartolin vrurg++ ;) 19:09
vrurg patrickb: one fix would involve a couple of changes into templates. Shall I make another PR for that? 19:10
vrurg fells like joining the committers club... 19:11
s/fells/feels/
patrickb vrurg: Please do! 19:19
Geth rakudo: vrurg++ created pull request #2893:
Fix for Configure.pl options when only --gen-moar is used		 19:51
nine So still more than a month till a reproducible release? :( 19:58
patrickb And a relocatable one. 20:02
Geth rakudo: 1fceeada7e | (Vadim Belman)++ | 3 files
Fix for Configure.pl options when only --gen-moar is used
 20:04
rakudo: af4e55a05d | (Patrick Böker)++ (committed using GitHub Web editor) | 3 files
Merge pull request #2893 from vrurg/build-revamp

Fix for Configure.pl options when only --gen-moar is used
vrurg Where do I get java 9? Oracle doesn't doesn't seem provide it anymore.
lizmat .seen kawaii 20:05
yoleaux I saw kawaii 17:33Z in #perl6-dev: <kawaii> done!
vrurg Ah, ok, java 10 seems to be working well.
lizmat hmmm.... HEAD configuring fails with: 20:06
===SORRY!===
Unknown macro configure_opts
for me
bartolin vrurg: I switched to java 11, since java 10 has reached end of support already. you can get one from adoptopenjdk.net/
lizmat vrurg patrickb ^^
vrurg lizmat: You need to update submodules. 20:07
bartolin vrurg: I'll look into telling travis to use java 11, too
lizmat and how does one do that ?
vrurg It was a tradeoff: either Configure.pl starts slow but updates them automatically, or we do it manually. 20:08
lizmat: git submodule update
20:08 ufobat_ joined
vrurg I'll try to figure out something. But nqp-configure isn't going to be updated often. 20:09
nine Considering stackoverflow.com/questions/560115...r-56013889 wouldn't the way for post Perl 6.c versions to add new methods to Setting classes be the $obj.&foo way of using subs as methods?
20:12 ufobat__ left 21:05 skids left
Geth nqp: 9a9b849d58 | usev6++ | examples/CallFromJava.java
[JVM] Fix order of arguments in example

There was a variant of findmethod that expected the ThreadContext as its first argument, but that was renamed with 49d6beb34a.
Thanks to @fabrice-ducos for reporting.		 21:17
rakudo: lizmat self-assigned dd fails on lazy lists github.com/rakudo/rakudo/issues/2892
0c8f269f81 | (Patrick Böker)++ | 2 files

Relocatability is disabled entirely on OpenBSD, because of technical difficulties. So no need to put hacks in to make it work in relocatable mode. Also simplify logic a bit more. Should still do the same as before.		 21:35
rakudo: 96e8d4f421 | (Patrick Böker)++ | 2 files
Fix OpenBSD

nqp::execname() is not '' in OpenBSD thus we still need some special casing. Kaiepi++ for the hint.
rakudo: bd89d51ed0 | (Patrick Böker)++ (committed using GitHub Web editor) | 2 files
Merge pull request #2857 from patzim/openbsd-cleanup

Don't special case OpenBSD in relocatability handling
patrickb timotimo: The two oldest PRs for rakudo are from you. Are they still relevant in the sense of it's worth looking at them? If not, can you close them? 21:53
timotimo i bet they're rubbish :) :)
you mean 184 and 236, right? 21:54
patrickb 2013 and 2014
yes
timotimo left margin from raw blocks and disable perl5 warnings
i think perl5 warnings has been superceded by "use isms" 21:55
vrurg Does it take some time for travis to update build status? nqp now builds with jvm but the summary still displays a problem. 21:57
Ah, nevermind, a test failed. 21:58
bartolin I wondered about the failing test, too. It seems to miss com.sun.jna.Callback. That doesn't happen on HEAD, so it might be related to a different travis setup. (it has 'language "java" on HEAD) 22:01
timotimo patrickb: i closed one of the two 22:02
vrurg bartolin: it just have failed on my mac too. So, it's not about travis.
bartolin vrurg: the test passes locally for me (FreeBSD with JDK 11). which java do you use? 22:04
vrurg Just have switched to 11.
patrickb one more PR and we are down to three pages :-D 22:05
timotimo aye!
bartolin well, I wasn't on your branch when I tested. but anyway, I think that could be fixed after a merge. It's much more important that the stuff works for MoarVM
vrurg bartolin: Ok, I'd leave it at this point. Need to make sure rakudo jvm builds too. 22:10
vrurg wonders who would merge NQP PR. 22:11
patrickb Just so I understand it. Is the rakudo build broken without that PR?
MasterDuke vrurg: there's an NQP PR that needs to be merged? 22:14
vrurg MasterDuke: github.com/perl6/nqp/pull/531 22:15
MasterDuke vrurg: btw, you don't have NQP commit ability?
vrurg patrickb: in certain cases.
MasterDuke: neither nqp nor rakudo. 22:16
MasterDuke vrurg: for rakudo you need to sign the CLA, but NQP is easier
vrurg patrickb: There some new command line options like --github-user which are aimed at easing synchronous development of rakudo-nqp-moar in forked repositories. 22:17
MasterDuke: no problem.
MasterDuke don't think i can give you that though, ping jnthn moritz timotimo lizmat 22:18
vrurg Where do I read it?
timotimo: what would you say?
Another thing I consider reasonable to do is to move vrurg/nqp-configure into perl6/. 22:19
MasterDuke hm, buggable or huggable used to have the instructions, but they aren't around anymore
vrurg patrickb: you were doing this recently. Would you share your experience? 22:20
Geth nqp/master: 58 commits pushed by (Vadim Belman)++, MasterDuke17++
review: github.com/perl6/nqp/compare/9a9b8...dae88d73d7		 22:22
patrickb let me search the logs
MasterDuke vrurg: merged the PR, looks like i don't have permissions to add you to the perl6 org (or at least i couldn't figure out how) 22:23
vrurg MasterDuke: no rush. Thanks for the merge! 22:24
patrickb vrurg: colabti.org/irclogger/irclogger_lo...-03-25#l48
vrurg patrickb: thanks! 22:25
patrickb vrurg++, MasterDuke++
MasterDuke ++s all around 22:32
bartolin well, now I see the failing test t/nativecall/01-basic.t on the jvm backend, too. but that's for later. 22:35
more importantly, the new configure doesn't recognize my make (on FreeBSD 11). 'make -v' doesn't reveal much. 22:36
vrurg bartolin: BSD make, of course... Can you use gmake for a while? 22:37
I'll have it fixed ASAP.
bartolin vrurg: yeah, I can work around it. just wanted to report it for now.
vrurg++
22:41 travis-ci joined
travis-ci NQP build failed. MasterDuke17 'Merge pull request #531 from vrurg/build-revamp 22:41
travis-ci.org/perl6/nqp/builds/532023997 github.com/perl6/nqp/compare/9a9b8...dae88d73d7
22:41 travis-ci left
patrickb I'm off to bed. o/ 22:47
22:47 patrickb left
vrurg o/ 22:47
bartolin it looks like the generated executables 'nqp' and 'nqp-j' don't have the jna-4.0.0.jar in their classpath. If I add that manually, t/nativecall/01-basic.t passes 22:51
vrurg rakudo builds and install with javac --source 1.9 22:53
bartolin: Those are parts I know nothing about. Could have lost something when was moving command lines from generating scripts into templates. 22:54
bartolin I'm testing a patch that re-adds those jars. (But I have to admit that I don't know much about those parts as well.) 23:03
vrurg make test in rakudo hangs. Though do to parallelizing it's not clear what exact test hangs. 23:05
But ok, enough jvm for today.
MasterDuke you can set the number of jobs to 1 23:06
vrurg was always a java hater, to the irony of it.
bartolin oh, 'make test' doesn't work for Rakudo on the jvm backend :/ There is a severe memory (whatever) leak with the EvalServer
(it used to run a few test files successfully, but not many) 23:07
vrurg Then I'm not worried anymore. For what was started as a way to simplify adding a new revision letter it's already good enough output. ;)
MasterDuke huh, i thought `make test` did work for the jvm backend 23:08
Geth nqp: aac9f52958 | usev6++ | 2 files
[JVM] Re-add 3rdparty jar jna-4.0.0.jar
bartolin let's see if that make travis happy again.
MasterDuke: it works for nqp, but not for rakudo
MasterDuke ah
bartolin I run a spectest from time to time -- but without the EvalServer 23:09
vrurg 990% CPU consumption – java breaks the record of my macbook! 23:11
bartolin (and yes, it would make sense to look at 'make test', too. it's just a lack of time.
java--
vrurg This is better than any ports event. 1000! It did it! On just 6 cores. :D 23:12
bartolin famous last words? I hope the machine didn't die the heat death *g* 23:25
23:26 travis-ci joined
travis-ci NQP build failed. usev6 '[JVM] Re-add 3rdparty jar jna-4.0.0.jar' 23:26
travis-ci.org/perl6/nqp/builds/532038706 github.com/perl6/nqp/compare/bcdae...c9f5295833
23:26 travis-ci left
bartolin hmm. I'll try look into that tomorrow. 23:27
bartolin drops to bed
vrurg bartolin: o/
Tahnks for all the help today!
Geth nqp: f404a20003 | (Nick Logan)++ (committed using GitHub Web editor) | docs/ops.markdown
Update getppid backends
 23:32
23:49 travis-ci joined
travis-ci NQP build failed. Nick Logan 'Update getppid backends' 23:49
travis-ci.org/perl6/nqp/builds/532046206 github.com/perl6/nqp/compare/aac9f...04a20003ae
23:49 travis-ci left

Powered by Cro and the Raku® Programming Language.

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

Thank you!