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.
00:39 leont left
patrickz I created PRs in the MoarVM, nqp and rakudo repos for a relocatable perl6 and C based runners. I'd be grateful for feedback. 00:47
00:48 patrickb joined, patrickb left, patrickb joined, patrickb left 00:52 patrickz left 01:01 MasterDuke joined 01:02 MasterDuke left, MasterDuke joined
vrurg R#2694 02:07
synopsebot R#2694 [open]: github.com/rakudo/rakudo/issues/2694 Role consumption by multiple roles
02:38 Guest32385 is now known as masak 03:14 pyrimidine left, pyrimidine joined 03:26 travis-ci joined
travis-ci Rakudo build failed. Elizabeth Mattijsen 'Hacky fix for R#2686 03:26
synopsebot R#2686 [open]: github.com/rakudo/rakudo/issues/2686 [regression][testneeded] Some issue exposed by IO::Glob
travis-ci travis-ci.org/vrurg/rakudo/builds/492503740 github.com/vrurg/rakudo/compare/84...6e508d0d4b
03:26 travis-ci left 04:50 AlexDani` joined 04:55 AlexDaniel left 05:56 robertle joined 06:07 robertle left 08:47 ufobat joined
lizmat /vrurg/rakudo ??? 09:19
Files=1267, Tests=88061, 408 wallclock secs (21.04 usr 6.29 sys + 2950.60 cusr 228.55 csys = 3206.48 CPU) 09:26
am I wrong? It looks like travis is also stalled? at 30fc5146ca875a28977 ? 09:28
looks like the silentness of Geth coincides with the lack of Travis activity 09:30
jnthn Hm, I see nothing on github status about notification problems... 10:00
Though on the hooks page it shows appveyor and travis as getting hooks OK and some problem sending them to hack.p6c.org 10:01
"We couldnā€™t deliver this payload: Failure when receiving data from the peer" 10:02
Maybe somebody with access to restart Geth could try that 10:03
Geth: help
Geth jnthn, Source at github.com/perl6/geth To add repo, add an 'application/json' webhook on GitHub pointing it to hack.p6c.org:8888/?chan=#perl6 and choose 'Send me everything' for events to send | use `ver URL to commit` to fetch version bump changes
jnthn Geth: restart
10:21 AlexDani` is now known as AlexDaniel 10:22 Geth left
AlexDaniel jnthn: just did 10:22
10:22 Geth joined, ChanServ sets mode: +v Geth
Geth rakudo: 84ed8548f9 | (Elizabeth Mattijsen)++ | src/core/Buf.pm6
Add Buf/Blob.STORE candidates for non-iterables

For some reason we cannot make things like CArray an Iterable without breaking a lot of things (like R#2687). This fixes R#2681 the hard way. If the native types are the same, this could be sped up quite a bit, but I don't see a quick way to introspect that and fast-path that.		 10:22
synopsebot R#2687 [closed]: github.com/rakudo/rakudo/issues/2687 [regression][testneeded] NativeCall issue exposed by Audio::Convert::Samplerate
R#2681 [closed]: github.com/rakudo/rakudo/issues/2681 [regression][testneeded] Crypt::TweetNacl test suite fails since 762c708fdcef 10:23
AlexDaniel this is probably an old commit
jnthn AlexDaniel++ 10:28
AlexDaniel I didn't redeliver all of the missed payloads, but at least it should work from now on
lizmat AlexDaniel++ 10:29
AlexDaniel fwiw instructions to restart non-whateverables are typically found like this
Geth: source
Geth AlexDaniel, Source at github.com/perl6/geth To add repo, add an 'application/json' webhook on GitHub pointing it to hack.p6c.org:8888/?chan=#perl6 and choose 'Send me everything' for events to send | use `ver URL to commit` to fetch version bump changes
AlexDaniel ā†‘ in the README 10:30
|Tux| Hmmm :( 10:40
Rakudo version 2018.12-301-g9d6e508d0 - MoarVM version 2018.12-105-gc0435e55a
csv-ip5xs0.746 - 0.751
csv-ip5xs-206.190 - 6.342
csv-parser22.035 - 22.582
csv-test-xs-200.428 - 0.431
test7.211 - 7.419
test-t2.007 - 2.068
test-t --race0.873 - 1.001
test-t-2033.307 - 33.347
test-t-20 --race10.433 - 11.005
lizmat [Tux] yeah, I saw similar slowdowns in spectest runnings :-(
releasable6 Next release in ā‰ˆ3 days and ā‰ˆ7 hours. 4 blockers. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 11:00
dogbert2_ |Tux|: aren't the results supposed to be improving :) 11:04
|Tux| that is the goal in the end yes 11:05
dogbert2_ if I remember correctly we have regressed from ~1.6 for test-t 11:13
lizmat dogbert2_: indeed 11:21
dogbert2_ perhaps something for one of AlexDaniels superbots 11:34
12:01 AlexDaniel left 12:14 lucasb joined 12:20 robertle joined
dogbert2_ lizmat: do you remember the url to |Tux| csv code? 13:25
ah, found it 13:33
14:29 dogbert2_ left
Geth rakudo: cygx++ created pull request #2696:
Make CUR uninstall .bat wrapper scripts		 14:42
rakudo: 121ca5fda7 | cygx++ (committed by Nick Logan) | src/core/CompUnit/Repository/Installation.pm6
Make CUR uninstall .bat wrapper scripts (#2696)
 14:54
15:20 AlexDaniel joined 16:00 tyil left 16:03 leont joined 16:41 ufobat_ joined 16:42 tyil joined 16:44 ufobat left 16:48 tyil left 16:50 tyil joined 16:54 lucasb left 16:59 tyil left 17:04 tyil joined 17:13 tyil left
Geth nqp/identify-mixin-types: 146ede2920 | (Jonathan Worthington)++ | 26 files
Start explicitly marking out mixin types

MoarVM will start requiring types that are the target type of a rebless
  (only used for mixins) to up-front declare they will be used that way.
This means it can assume anything that is not such a type - the vast majority of cases - will never have extra data attached to them, which can be used to generate simpler attribute access code. ... (6 more lines)		 17:23
rakudo/identify-mixin-types: f09156240a | (Jonathan Worthington)++ | 3 files
Explicitly identify types used for mixins

So that we'll be able to better optimize access to the rest of them.		 17:24
17:31 vrurg left
Geth nqp/identify-mixin-types: d2b2e47578 | (Jonathan Worthington)++ | 5 files
Fix tests after mixin type change

Provide a quite cheaty trait to allow fixing of them without having to resort to writing a bunch of MOP code.		 17:37
18:16 vrurg joined 18:20 vendethiel- joined 18:36 vendethiel- left 18:37 vendethiel- joined 18:56 leont left
japhb jnthn: Why does github.com/perl6/nqp/commit/146ede2920 (the nqp one) require `$metaclass.set_is_mixin($new_type);` on the $is_mixin path but github.com/rakudo/rakudo/commit/f09156240a (the rakudo one) doesn't? 19:45
timotimo because of nqp::newmixintype 19:48
japhb But both commits use that .... 19:51
19:52 vendethiel- left
timotimo hm, maybe it was required for the rebootstrap 19:53
oh 19:54
perhaps because Metamodel::Mixins unconditionally passes :is_mixin
and maybe only that does?
dunno %)
i've only seen the code in the diffs 19:55
vrurg It looks like I'm getting closer to my dream: BUILD and TWEAK are being executed in roles in my fork.... That's so exciting, damn... ;) 20:22
20:32 vendethiel- joined
timotimo oh my 21:06
21:35 ufobat_ left 22:17 vendethiel- left
jnthn japhb: I moved the call to that in one of them, but not in the other, which I guess matches my "should I move this or not" thoughts. ;) 22:24
Perhaps it's more robust to move it in NQP also 22:25
japhb Heh, indecision eigencommits 22:26
22:40 tyil joined 22:57 dogbert17 left 23:04 tyil left 23:22 tyil joined 23:27 tyil left 23:46 tyil joined

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!