[Coke] crowd here might care to know: I've stepped down as the TPF Grants Committee Secretary. 00:26
jdv79 why? 01:25
Geth rakudo: 2b81f97819 | (Christian Bartolomäus)++ | src/core.c/Rakudo/Internals/JSON.pm6
[JVM] Don't use native string arrays in setting

As already pointed out in e22aeafb50 these blow up on the JVM backend.
This should fix github.com/rakudo/rakudo/issues/3695.
The problem only surfaced during installation on Windows, because there pathes contain backslashes that need to be escaped in JSON.		 05:44
Altai-man_ Awesome to have everything passing CI. \o/ 06:49
lizmat Files=1306, Tests=111304, 213 wallclock secs (28.76 usr 8.34 sys + 2992.84 cusr 274.13 csys = 3304.07 CPU) 07:08
[Tux] Rakudo version 2020.05.1-144-g2b81f9781 - MoarVM version 2020.05-13-g03c9154e8
csv-ip5xs0.957 - 0.968
csv-ip5xs-209.558 - 9.617
csv-parser30.887 - 34.534
csv-test-xs-200.459 - 0.532
test8.483 - 9.496
test-t2.374 - 2.471
test-t --race1.008 - 1.361
test-t-2035.848 - 37.974
test-t-20 --race11.544 - 12.280
 10:43
lizmat wot?
[Tux] Oops (timings done during teams meeting) or explainable?
I'll run it again after meeting 10:44
lizmat not explainable
looking at the commit log
Altai-man_ sigh 10:46
lizmat sigh? for now I'm pretty sure it's a fluke 10:47
Altai-man_ quite possibly, I'd say. just was thinking it was at around 1.8 for some days and I am unused already. :) 10:50
Or, rather, *not used to it anymore 10:51
Xliff .tell nine I think it was one of your bug fixes, that again required the use of a full directory lock. I think it was the .repo-id situation you fixed in f5ce80e1a6d60a6a51a898da85f3100bb088dfb4 12:29
linkable6 (2020-02-13) github.com/rakudo/rakudo/commit/f5ce80e1a6 Fix concurrency issue with re-checking a precomp file's dependencies
tellable6 Xliff, I'll pass your message to nine
[Coke] jdv79: just moving on. no drama. 13:30
[Tux] After the reboot … 14:43
Rakudo version 2020.05.1-144-g2b81f9781 - MoarVM version 2020.05-13-g03c9154e8
csv-ip5xs0.807 - 0.813
csv-ip5xs-207.807 - 7.827
csv-parser24.657 - 26.264
csv-test-xs-200.390 - 0.413
test7.317 - 7.726
test-t1.864 - 1.874
test-t --race0.811 - 0.879
test-t-2030.398 - 33.730
test-t-20 --race9.221 - 9.800
lizmat ah, much better :-) 14:47
Altai-man_ [Tux]++ 14:50
Geth nqp/new-disp: 090d67c189 | (Jonathan Worthington)++ | src/vm/moar/QAST/QASTOperationsMAST.nqp
Only re-use arg registers until after dispatch

The code here was nabbed from that for emitting invoke instructions, but since we now no longer copy the arguments into a buffer, but instead just have a list of registers to source them from, we musn't reuse those registers until after the dispatch is done.		 14:55
nqp/new-disp: ae04baf5e2 | (Jonathan Worthington)++ | t/moar/53-dispatch.t
Basic tests for boot-code-constant dispatcher
 14:58
rakudo: 2eec62feef | (Elizabeth Mattijsen)++ | src/core.c/Code.pm6
Reduce overhead of Code.ACCEPTS(value) by 20%

Not a lot, but it this code gets hot pretty easily for e.g. dir()		 14:59
[Coke] installed valgrind on osx to see if that helped with the abort trap I was seeing in cro. forgot how slow that is. :) 15:46
moritz with things being slow and starting witih v, I first read virtualbox, not valgrind :D 16:27
[Coke] yah, this feels like it froze and is not going to throw the abort trap 16:28
Geth rakudo/master: 4 commits pushed by (Vadim Belman)++ 17:30
Xliff m: my ($x, $y, $w, $h) = 5 xx 4; ($w, $h) »-=» ( 2 * $x, 2 * $y); ($w, $h).say 22:02
camelia (-5 -5)
Xliff m: my ($x, $y, $w, $h) = 5 xx 4; ($w, $h) »-» ( 2 * $x, 2 * $y); ($w, $h).say
camelia Potential difficulties:
Useless use of »-» in sink context
at <tmp>:1
------> 3my ($x, $y, $w, $h) = 5 xx 4; ($w, $h) 7⏏5»-» ( 2 * $x, 2 * $y); ($w, $h).say
(5 5)
Xliff m: my $r = 0..1; say 0.5 ~~ $r 22:18
camelia True
Xliff m: my $r = 0..1; say 0.5 ~~ $r; $r.min.say; $r.max.say
camelia True
0
1

Powered by Cro and the Raku® Programming Language.

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

Thank you!