Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6-dev, 2017-02-24

| Channels | #perl6-dev index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:05 llfourn joined #perl6-dev
00:09 pyrimidine joined #perl6-dev
00:23 ZzZombo joined #perl6-dev
00:25 cognominal joined #perl6-dev
00:26 ZzZombo_ joined #perl6-dev
00:33 ZzZombo_ joined #perl6-dev
00:43 ZzZombo_ joined #perl6-dev
00:52 ZzZombo_ joined #perl6-dev
01:04 geekosaur joined #perl6-dev
01:06 ZzZombo_ joined #perl6-dev
01:06 ZzZombo_ joined #perl6-dev
01:25 pyrimidine joined #perl6-dev
01:47 llfourn joined #perl6-dev
02:05 pyrimidine joined #perl6-dev
02:07 stmuk joined #perl6-dev
02:23 user__ joined #perl6-dev
02:25 stmuk joined #perl6-dev
02:49 ilbot3 joined #perl6-dev
02:49 Topic for #perl6-dev is now Perl 6 language and compiler development 2.0 | Logs at http://irclog.perlgeek.de/perl6-dev/today
03:14 pyrimidine joined #perl6-dev
03:37 MasterDuke_ joined #perl6-dev
03:38 MasterDuke_ i spoke to soon. trying to build rakudo with that method removed nearly killed my box and the oom killer took out firefox
03:38 MasterDuke_ *too
03:40 MasterDuke_ huh, and did i misstype my password? it won't let me /nick to MasterDuke, it says it's already in use
03:47 llfourn joined #perl6-dev
04:34 llfourn joined #perl6-dev
06:16 pyrimidine joined #perl6-dev
06:25 ggoebel joined #perl6-dev
06:51 RabidGravy joined #perl6-dev
07:13 [Tux] joined #perl6-dev
07:29 [Tux] This is Rakudo version 2017.02-55-g320c2fb46 built on MoarVM version 2017.02-9-gc5379702
07:29 [Tux] csv-ip5xs        2.890
07:29 [Tux] test            12.664
07:29 [Tux] test-t           4.823 - 5.004
07:29 [Tux] csv-parser      13.261
07:52 stmuk joined #perl6-dev
08:13 pyrimidine joined #perl6-dev
08:25 [TuxCM] joined #perl6-dev
08:50 lizmat Files=1177, Tests=55938, 188 wallclock secs (11.31 usr  4.49 sys + 1123.82 cusr 105.84 csys = 1245.46 CPU)
08:56 lizmat http://marc.info/?l=git&m=148787047422954  # probably also interesting wrt our use of sha1 in compunit repos
09:14 pyrimidine joined #perl6-dev
09:16 samcv thanks lizmat
09:18 samcv lizmat, this isn't as much an issue as like perl 5 having ./ in @INC right?
09:22 samcv since you'd basically need write access to these locations anyway to exploit it
09:28 ZzZombo joined #perl6-dev
09:29 ZzZombo_ joined #perl6-dev
09:39 Geth ¦ star: cschwenz++ created pull request #86: Revert copyright year in LICENSE to original.
09:39 Geth ¦ star: review: https://github.com/rakudo/star/pull/86
10:03 ZzZombo_ joined #perl6-dev
10:16 ZzZombo__ joined #perl6-dev
10:16 pyrimidine joined #perl6-dev
10:16 ZzZombo joined #perl6-dev
11:24 lizmat samcv: indeed
11:25 Geth ¦ star: f78a2c06c1 | (Calvin Schwenzfeier)++ | LICENSE
11:25 Geth ¦ star: Revert copyright year in LICENSE to original.
11:25 Geth ¦ star:
11:25 Geth ¦ star: Per the commit message on the Rakudo LICENSE file[^1] and the source
11:25 Geth ¦ star: text for the original license[^2][^3], "Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed."
11:25 Geth ¦ star:
11:25 Geth ¦ star: [1] https://github.com/rakudo/rakudo/commit/47f5300f41cbce765471f81b52bda87bd4600be7
11:25 Geth ¦ star: [2] http://www.perlfoundation.org/artistic_license_2_0
11:25 Geth ¦ star: [3] https://opensource.org/licenses/Artistic-2.0
11:25 Geth ¦ star: review: https://github.com/rakudo/star/commit/f78a2c06c1
11:25 Geth ¦ star: 70da757d7c | lizmat++ | LICENSE
11:25 Geth ¦ star: Merge pull request #86 from cschwenz/master
11:25 Geth ¦ star:
11:25 Geth ¦ star: Revert copyright year in LICENSE to original.
11:25 Geth ¦ star: review: https://github.com/rakudo/star/commit/70da757d7c
11:32 ZzZombo_ joined #perl6-dev
11:32 ZzZombo_ joined #perl6-dev
11:51 AlexDaniel joined #perl6-dev
12:17 pyrimidine joined #perl6-dev
13:19 pyrimidine joined #perl6-dev
13:21 MasterDuke_ if there aren't any further thoughts/comments, would anybody mind merging https://github.com/rakudo/rakudo/pull/1021 and https://github.com/rakudo/rakudo/pull/1018 (and maybe squashing 1018 so there aren't commits that break the jvm build)?
13:22 Geth ¦ rakudo/nom: 25a3cc54b1 | (Daniel Green)++ | src/Perl6/Actions.nqp
13:22 Geth ¦ rakudo/nom: Optimize Actions.(dec|hex|oct|bin)int()
13:22 Geth ¦ rakudo/nom:
13:22 Geth ¦ rakudo/nom: Create a string_to_int that uses nqp::radix instead of nqp::radix_I.
13:22 Geth ¦ rakudo/nom: Then call that from Actions.(dec|hex|oct|bin)int() instead of
13:22 Geth ¦ rakudo/nom: string_to_bigint when the number of chars in the string to be converted
13:22 Geth ¦ rakudo/nom: is small enough that it couldn't cause overflow.
13:22 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/25a3cc54b1
13:22 Geth ¦ rakudo/nom: b61b3c7a7f | (Daniel Green)++ | src/Perl6/Actions.nqp
13:22 Geth ¦ rakudo/nom: Use string_to_int for longer strings when 64-bit
13:22 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/b61b3c7a7f
13:22 Geth ¦ rakudo/nom: d41b68e16c | lizmat++ | src/Perl6/Actions.nqp
13:22 Geth ¦ rakudo/nom: Merge pull request #1021 from MasterDuke17/optimize_string_to_bigint
13:22 Geth ¦ rakudo/nom:
13:22 Geth ¦ rakudo/nom: Optimize Actions.(dec|hex|oct|bin)int()
13:22 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/d41b68e16c
13:23 lizmat MasterDuke_: re 1018, what would be the effect on the JVM build?
13:32 MasterDuke_ lizmat++ the first commit breaks it, the next unbusts building, but has an error so it won't really run correctly, the last commit fixes running also
13:42 IOninja s: CORE::.keys.grep(* ne "IterationEnd").map({CORE::{$_}}).grep({ !.DEFINITE and try .can("say") }), 'unique', \()
13:42 SourceBaby IOninja, Sauce is at https://github.com/rakudo/rakudo/blob/320c2fb/src/core/Any-iterable-methods.pm#L1487
13:58 IOninja bisectable6: Scalar.unique
13:58 bisectable6 IOninja, On both starting points (old=2015.12 new=d41b68e) the exit code is 0, exit signal is 11 (SIGSEGV) and the output is identical as well
13:58 bisectable6 IOninja, Output on both points:
13:58 IOninja c: 2014.01 Scalar.unique
13:58 committable6 IOninja, ¦«2014.01»: Cannot invoke null object␤  in block  at /tmp/vrw4a_2gQY:1␤ «exit code = 1»
13:58 IOninja c: 2015.01 Scalar.unique
13:58 committable6 IOninja, ¦«2015.01»:
13:58 IOninja c: 2015.07 Scalar.unique
13:58 committable6 IOninja, ¦«2015.07»:
13:59 IOninja c: all Scalar.unique
13:59 committable6 IOninja, https://gist.github.com/769828254057438050cb84a953f9041b
15:22 ggoebel joined #perl6-dev
15:45 llfourn joined #perl6-dev
16:03 llfourn joined #perl6-dev
16:10 pyrimidine joined #perl6-dev
16:11 jsimonet joined #perl6-dev
16:40 Geth ¦ rakudo: MasterDuke17++ created pull request #1025: Check that the argument to p6recont_ro IS_CONCRETE
16:40 Geth ¦ rakudo: review: https://github.com/rakudo/rakudo/pull/1025
16:42 IOninja Doesn't that make p6recont_ro a NOP for all type objects?
16:43 IOninja oh, it's just the container's concretness that's tested...
16:46 * IOninja wonders what happened to IS_WOOD, IS_STEEL, and IS_PAPER
17:08 llfourn joined #perl6-dev
17:32 pyrimidine joined #perl6-dev
17:36 dogbert17 IOninja; have you tried the mapper program with MasterDuke's fix?
17:41 Geth ¦ rakudo/nom: cfe0e04622 | (Daniel Green)++ | src/vm/moar/ops/perl6_ops.c
17:41 Geth ¦ rakudo/nom: Check that the argument to p6recont_ro IS_CONCRETE
17:41 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/cfe0e04622
17:41 Geth ¦ rakudo/nom: ae2c1dadf0 | lizmat++ | src/vm/moar/ops/perl6_ops.c
17:41 Geth ¦ rakudo/nom: Merge pull request #1025 from MasterDuke17/fix_segv_in_p6recont_ro
17:41 Geth ¦ rakudo/nom:
17:41 Geth ¦ rakudo/nom: Fix SEGV in nqp::p6recont_ro
17:41 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/ae2c1dadf0
17:44 robertle joined #perl6-dev
17:45 llfourn joined #perl6-dev
17:45 IOninja dogbert17: no, but I imagine it fixes it.
17:52 MasterDuke lizmat++ also, need any other info for https://github.com/rakudo/rakudo/pull/1018?
17:53 Geth ¦ rakudo/nom: 4 commits pushed by (Daniel Green)++, lizmat++
17:53 Geth ¦ rakudo/nom: b2e0ac09ed | Make Parameter.[named_names, type_captures] list_s
17:53 Geth ¦ rakudo/nom: b1def95c1a | JVM fix for native named_names and type_captures
17:53 Geth ¦ rakudo/nom: 2f6d2c6f2f | Add braces to the now two-line body of an if
17:53 Geth ¦ rakudo/nom: e9b55e5d83 | Merge pull request #1018 from MasterDuke17/make_Parameters_named_names_and_type_captures_native_str_lists
17:53 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/compare/ae2c1dadf0...e9b55e5d83
17:53 lizmat no  :-)
17:55 MasterDuke lizmat++
18:07 pyrimidine joined #perl6-dev
18:21 Geth ¦ rakudo/nom: 45b3af8dc2 | (Elizabeth Mattijsen)++ | src/core/Rakudo/Internals.pm
18:21 Geth ¦ rakudo/nom: Make MERGESORT also take IterationBuffers directly
18:21 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/45b3af8dc2
18:21 Geth ¦ rakudo/nom: 7010ae9bbe | (Elizabeth Mattijsen)++ | src/core/Rakudo/Iterator.pm
18:21 Geth ¦ rakudo/nom: Make R:I.ReifiedList take IterationBuffers directly
18:21 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/7010ae9bbe
18:29 MasterDuke lizmat: is there an easy way to call/use MERGESORT directly?
18:30 lizmat in the setting, it's just Rakudo::Internals.MERGESORT...()
18:30 lizmat MasterDuke: but why ?
18:31 MasterDuke lizmat: i thought i noticed some weirdness (seemed slower than i expected), but that was in a very large script and i wasn't sure what under the hood was using it
18:32 MasterDuke so i was just going to try it with some random input and time it
18:32 lizmat MasterDuke: by all means...  :-)
18:32 lizmat it now directly takes IterationBuffers  :-)
18:32 lizmat still returns a List though, baby steps  :-)
18:36 pyrimidine joined #perl6-dev
18:46 llfourn joined #perl6-dev
19:12 lizmat joined #perl6-dev
19:20 MasterDuke lizmat: turns out a graph of Rakudo::Internals.MERGESORT-REIFIED-LIST performance has the same shape as Perl 5's sort, it's just the individual values are 15-20x higher
19:22 lizmat yeah... not unsurprising, as that is same order of magnitude between Text::CSV_PP in Perl 5 and Text::CSV in Perl 6
19:23 Geth ¦ roast: e31c78c52b | (Zoffix Znet)++ | S32-list/unique.t
19:23 Geth ¦ roast: Test Scalar.unique does not SEGV
19:23 Geth ¦ roast:
19:23 Geth ¦ roast: RT#130852: https://rt.perl.org/Ticket/Display.html?id=130852
19:23 Geth ¦ roast: Rakudo fix: https://github.com/rakudo/rakudo/commit/ae2c1dadf0
19:23 synopsebot6 Link:  https://rt.perl.org/rt3//Public/Bug/Display.html?id=130852
19:23 Geth ¦ roast: review: https://github.com/perl6/roast/commit/e31c78c52b
19:23 IOninja ZOFVM: Files=1226, Tests=132875, 220 wallclock secs (25.02 usr  3.88 sys + 4621.42 cusr 310.74 csys = 4961.06 CPU)
19:25 MasterDuke IOninja: i don't understand eval-lives-ok. on a perl6 build before my patch, `perl6 -e 'use Test; eval-lives-ok 「Scalar.unique」'` segfaults
19:28 IOninja sure, 'cause the bug is still in there.
19:28 IOninja You'd need to use `is_run` to make it fail just that test and not crash the test file.
19:30 IOninja why didn't I use lives-ok { Scalar.unique } ? Ignorance and paranoia. I don't know what sort of optimizations we do or will do that won't turn that into some optimized version that doesn't hit the bug. Maybe the same would happen with the eval-lives-ok version, I've no idea.
19:31 IOninja ...also 「...」 looks cool
19:33 MasterDuke i find it mildly annoying that roast doesn't have a RAKUDO_REVISION. if i'm working on something in a branch and a bunch of roast tests go in there are likely to be failures, and i have to figure out whether they are because of my changes or not
19:35 pyrimidine joined #perl6-dev
19:36 Geth ¦ rakudo/nom: 556db9d7bc | (Elizabeth Mattijsen)++ | src/core/Array.pm
19:36 Geth ¦ rakudo/nom: Make Array.splice use IterationBuffer internally
19:36 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/556db9d7bc
19:38 Geth ¦ rakudo/nom: 9e9a4ad250 | (Elizabeth Mattijsen)++ | src/core/Baggy.pm
19:38 Geth ¦ rakudo/nom: Re-imagine Baggy!LISTIFY
19:38 Geth ¦ rakudo/nom:
19:38 Geth ¦ rakudo/nom: Make it return a ready made Str.  This allows using native strs
19:38 Geth ¦ rakudo/nom: internally, and thereby speed up things like Bag.perl|gist|Str.
19:38 Geth ¦ rakudo/nom:
19:38 Geth ¦ rakudo/nom: Actually part of the nqp::list -> IterationBuffer hunt.
19:38 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/9e9a4ad250
19:47 llfourn joined #perl6-dev
19:48 committable6 joined #perl6-dev
19:57 MasterDuke joined #perl6-dev
19:58 Geth ¦ rakudo/nom: 8e8cd14bf4 | (Elizabeth Mattijsen)++ | src/core/Block.pm
19:58 Geth ¦ rakudo/nom: Make sure Block.phasers returns List with IB
19:58 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/8e8cd14bf4
20:26 ggoebel joined #perl6-dev
20:36 Geth ¦ rakudo/nom: 7556498daa | (Elizabeth Mattijsen)++ | src/core/CompUnit/RepositoryRegistry.pm
20:36 Geth ¦ rakudo/nom: Move some code out of a loop
20:36 Geth ¦ rakudo/nom:
20:36 Geth ¦ rakudo/nom: Actually part of the nqp::list -> IB search
20:36 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/7556498daa
20:49 llfourn joined #perl6-dev
21:17 pyrimidine joined #perl6-dev
21:17 MasterDuke re https://github.com/rakudo/rakudo/pull/876, if i removed all the controversial/incorrect commits from that branch/PR, would anybody have a problem with merging it?
21:18 [Coke] .... by definition, probably not? :)
21:19 [Coke] MasterDuke: coming in late: what is the purpose of the commit?
21:19 [Coke] s/commit/PR/
21:19 MasterDuke --> Nil return constraints are good for performance
21:20 MasterDuke so i went through and added them everywhere i thought it made sense
21:21 timotimo OK
21:21 MasterDuke a couple were mistaken, but it's still a bunch of changes (though it did spectest clean at the time)
21:21 timotimo were you able to measure any change in performance for those?
21:21 MasterDuke nothing very noticeable
21:22 timotimo ok :(
21:22 MasterDuke though i like being explicit, so i would argue that alone is a benefit
21:22 timotimo fair enough
21:23 [Coke] if the performance increase isn't noticable, why do it?
21:24 [Coke] so that leaves "making it explicit", which I would argue we don't care about in core. we want fast in core.
21:24 MasterDuke well, just because the difference isn't noticeable now doesn't mean it won't be more so in the future
21:36 lizmat right: afaik, --> Nil is not handled by the optimizer just yet
21:37 MasterDuke i know timotimo said it was faster, but maybe that was more in a theoretical way
21:40 MasterDuke benchable6: compare sub a() { my $a = rand; }; a() for ^5_000_000 ||| sub b(--> Nil) { my $a = rand; }; a() for ^5_000_000;
21:40 benchable6 MasterDuke, ¦«sub»:Cannot find this revision
21:41 MasterDuke benchable6: compare HEAD sub a() { my $a = rand; }; a() for ^5_000_000 ||| sub b(--> Nil) { my $a = rand; }; a() for ^5_000_000;
21:41 benchable6 MasterDuke, starting to benchmark the 1 given commit
21:41 benchable6 MasterDuke, ¦«HEAD»:===SORRY!=== Error while compiling -e␤Undeclared routine:␤    a used at line 1␤
21:41 MasterDuke benchable6: compare HEAD sub a() { my $a = rand; }; a() for ^5_000_000 ||| sub b(--> Nil) { my $a = rand; }; b() for ^5_000_000;
21:41 benchable6 MasterDuke, starting to benchmark the 1 given commit
21:41 benchable6 MasterDuke, ¦«HEAD»:Benchmark: ␤Timing 10 iterations of 0, 1...␤«timed out after 10 seconds, output»:
21:42 MasterDuke benchable6: compare HEAD sub a() { my $a = rand; }; a() for ^2_000_000 ||| sub b(--> Nil) { my $a = rand; }; b() for ^2_000_000;
21:42 benchable6 MasterDuke, starting to benchmark the 1 given commit
21:42 benchable6 MasterDuke, ¦«HEAD»:Benchmark: ␤Timing 10 iterations of 0, 1...␤         0: 8.2902 wallclock secs @ 1.2062/s (n=10)␤«timed out after 10 seconds, output»:
21:42 [Coke] gah
21:42 MasterDuke benchable6: compare HEAD sub a() { my $a = rand; }; a() for ^1_000_000 ||| sub b(--> Nil) { my $a = rand; }; b() for ^1_000_000;
21:42 benchable6 MasterDuke, starting to benchmark the 1 given commit
21:42 benchable6 MasterDuke, https://gist.github.com/aa7e16a86d05f1bf8115339931d5b9cf
21:42 lizmat MasterDuke: the https://github.com/rakudo/rakudo/pull/876 seems to have conflicts ?
21:44 MasterDuke lizmat: yeah, it's pretty old. i was wondering if i should go to the effort of fixing them and addressing the comments
21:44 lizmat I'll merge it if it's clean  :-)
21:45 lizmat unless people have objections ?
21:45 MasterDuke [Coke], timotimo, lizmat: if you look at the benchable6 results (and i confirmed the same thing on my machine), the --> Nil version is slightly faster
21:45 MasterDuke but it does take a lot of iterations to notice it
21:46 MasterDuke lizmat: cool, i'll see about revamping it then
21:46 * [Coke] defers to lizmat.
21:47 Geth ¦ rakudo/nom: adb6f87e88 | (Elizabeth Mattijsen)++ | src/core/control.pm
21:47 Geth ¦ rakudo/nom: Make CLONE-HASH-DECONTAINERIZED a bit smarter
21:47 Geth ¦ rakudo/nom:
21:47 Geth ¦ rakudo/nom: Also tried moving that to Rakudo::Internals, but as a method the result
21:47 Geth ¦ rakudo/nom: appears to always be HLLized, which sorta defeats the purpose.
21:47 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/adb6f87e88
21:48 * [Coke] really dislikes the opcode name "iterval". :)
21:48 IOninja why?
21:48 [Coke] because it's "interval" spelled wrong (except it isn't)
21:49 IOninja ah :)
21:51 llfourn joined #perl6-dev
22:02 sivoais joined #perl6-dev
22:12 sivoais joined #perl6-dev
22:20 lizmat m: my %h = a => 42, b => 666; say %h.DUMP  # this appears to be borked for a long time already
22:20 camelia rakudo-moar adb6f8: OUTPUT: «Hash<1>(␤    :$!descriptor(Perl6::Metamodel::ContainerDescriptor<2>(...)),␤    :$!storage(BOOTHash<3>(...))␤)␤»
22:22 lizmat ah, it only looks borked  :-)
22:22 * [Coke] has apparently forgotten there was a .DUMP
22:23 [Coke] is that the method form of dd?
22:24 lizmat no, this predates dd by a long shot: appears to be some debugging aid from way back
22:24 [Coke] kill it!
22:24 [Coke] :)
22:26 * lizmat is tempted
22:26 * lizmat has some experience
22:35 Geth ¦ rakudo/nom: 3de7b087b4 | (Elizabeth Mattijsen)++ | src/core/Hash.pm
22:35 Geth ¦ rakudo/nom: Streamline Hash.DUMP a bit
22:35 Geth ¦ rakudo/nom:
22:35 Geth ¦ rakudo/nom: But it should probably be removed, as it looks like a very old
22:35 Geth ¦ rakudo/nom: debugging aid.
22:35 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/3de7b087b4
22:44 lizmat good night, #perl6-dev!
22:44 IOninja night
22:51 timotimo MasterDuke: i'm not actually sure i said anything about --> Nil being faster actually? :S
22:51 IOninja you did
22:52 timotimo oh
22:52 IOninja timotimo: https://irclog.perlgeek.de/perl6-dev/2016-09-10#i_13184818
22:52 llfourn joined #perl6-dev
22:52 timotimo maybe we're clever about generating the right kind of return code when we have --> Nil
22:53 IOninja I think it bypasses the return type typecheck if it detects a Nil
22:53 timotimo cool
22:53 timotimo maybe i leaned a bit too far out of the window when i said "very good for performance"
22:54 IOninja m: sub foo () {$ = 42}; for ^2000000 { $ = foo }; say now - INIT now
22:54 camelia rakudo-moar 3de7b0: OUTPUT: «Cannot assign to an immutable value␤  in block <unit> at <tmp> line 1␤␤»
22:54 IOninja wat?
22:55 IOninja oh, optimizer messes it up
22:58 IOninja m: sub foo () {42}; for ^2000000 { $ = foo }; say now - INIT now
22:58 camelia rakudo-moar 3de7b0: OUTPUT: «1.036167␤»
22:58 IOninja m: sub foo (-->Nil) {42}; for ^2000000 { $ = foo }; say now - INIT now
22:58 camelia rakudo-moar 3de7b0: OUTPUT: «WARNINGS for <tmp>:␤Useless use of constant integer 42 in sink context (line 1)␤1.2373463␤»
22:58 IOninja m: sub foo (-->Nil) {$ = 42}; for ^2000000 { $ = foo }; say now - INIT now
22:58 camelia rakudo-moar 3de7b0: OUTPUT: «1.23535806␤»
22:58 IOninja m: sub foo (-->Nil) {$ = 42}; for ^2000000 { $ = foo }; say now - INIT now
22:58 camelia rakudo-moar 3de7b0: OUTPUT: «1.2348875␤»
22:58 IOninja m: sub foo (-->Nil) {$ = 42}; for ^2000000 { $ = foo }; say now - INIT now
22:58 camelia rakudo-moar 3de7b0: OUTPUT: «1.317078601␤»
22:59 IOninja m: sub foo () {42}; for ^2000000 { $ = foo }; say now - INIT now
22:59 camelia rakudo-moar 3de7b0: OUTPUT: «0.9883682␤»
22:59 IOninja m: sub foo () {42}; for ^2000000 { $ = foo }; say now - INIT now
22:59 camelia rakudo-moar 3de7b0: OUTPUT: «0.9765939␤»
22:59 * IOninja shrugs
22:59 timotimo hm.
23:53 llfourn joined #perl6-dev

| Channels | #perl6-dev index | Today | | Search | Google Search | Plain-Text | summary