Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6-dev, 2017-09-02

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

All times shown according to UTC.

Time Nick Message
01:22 Geth ¦ nqp: a1efba5b10 | (Samantha McVey)++ | tools/build/MOAR_REVISION
01:22 Geth ¦ nqp: Bump MoarVM for 2x faster index and eqat operations
01:22 Geth ¦ nqp:
01:22 Geth ¦ nqp: Speedup is for eq or index for strings of differing storage
01:22 Geth ¦ nqp: types, or where one is a strand (most of our operations).
01:22 Geth ¦ nqp:
01:22 Geth ¦ nqp: For eq: 1.7x when one is a strand and one is flat, and 2.5x when
01:22 Geth ¦ nqp: both strings are flat.
01:22 Geth ¦ nqp: <…commit message has 10 more lines…>
01:22 Geth ¦ nqp: review: https://github.com/perl6/nqp/commit/a1efba5b10
01:22 Geth ¦ nqp: version bump brought these changes: https://github.com/MoarVM/MoarVM/compare/2017.08.1-46-g79654248... No newline at end of file
01:25 Geth ¦ rakudo/nom: 5ebbc5baf8 | (Samantha McVey)++ | tools/build/NQP_REVISION
01:25 Geth ¦ rakudo/nom: Bump Moar/NQP for 2x faster index and eqat operations
01:25 Geth ¦ rakudo/nom:
01:25 Geth ¦ rakudo/nom: Speedup is for eq or index for strings of differing storage
01:25 Geth ¦ rakudo/nom: types, or where one is a strand (most of our operations).
01:25 Geth ¦ rakudo/nom:
01:25 Geth ¦ rakudo/nom: For eq: 1.7x when one is a strand and one is flat, and 2.5x when
01:25 Geth ¦ rakudo/nom: both strings are flat.
01:25 Geth ¦ rakudo/nom: <…commit message has 16 more lines…>
01:25 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/5ebbc5baf8
01:25 Geth ¦ rakudo/nom: version bump brought these changes: https://github.com/perl6/nqp/compare/2017.08-25-gfc57c9a20... No newline at end of file
01:38 geekosaur joined #perl6-dev
01:53 ilbot3 joined #perl6-dev
01:53 Topic for #perl6-dev is now Perl 6 language and compiler development 2.0 | Logs at http://irclog.perlgeek.de/perl6-dev/today | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
01:58 Geth ¦ rakudo/nom: 0bc1c8774c | (Samantha McVey)++ | 2 files
01:58 Geth ¦ rakudo/nom: Add --ignore-errors option to Configure.pl
01:58 Geth ¦ rakudo/nom:
01:58 Geth ¦ rakudo/nom: Allows you to ignore errors such as having an incorrect version of NQP.
01:58 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/0bc1c8774c
02:00 lizmat joined #perl6-dev
03:47 Ven`` joined #perl6-dev
04:38 astj joined #perl6-dev
04:42 astj joined #perl6-dev
07:24 Skarsnik joined #perl6-dev
07:25 Skarsnik .
07:50 [Tux] This is Rakudo version 2017.08-59-g0bc1c8774 built on MoarVM version 2017.08.1-53-gad5a6464
07:50 [Tux] csv-ip5xs        1.379 - 1.400
07:50 [Tux] test            10.853
07:50 [Tux] test-t           3.588 - 3.639
07:50 [Tux] csv-parser      11.425
08:59 astj joined #perl6-dev
09:25 Ven joined #perl6-dev
09:46 dogbert2 joined #perl6-dev
09:48 Skarsnik_ joined #perl6-dev
09:54 eater joined #perl6-dev
10:05 Ven joined #perl6-dev
10:22 astj joined #perl6-dev
10:38 Skarsnik joined #perl6-dev
11:01 Geth ¦ rakudo/nom: 47439e69ff | MasterDuke17++ | 2 files
11:01 Geth ¦ rakudo/nom: Make INTERPOLATE's parameters non-optional
11:01 Geth ¦ rakudo/nom:
11:01 Geth ¦ rakudo/nom: Since the ops for optional params aren't currently JITted.
11:01 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/47439e69ff
11:01 Geth ¦ rakudo/nom: d85569d036 | MasterDuke17++ | src/Perl6/Actions.nqp
11:01 Geth ¦ rakudo/nom: Align two similar method bodies the same way
11:01 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/d85569d036
11:01 Geth ¦ rakudo/nom: 85a97ab3d9 | lizmat++ (committed using GitHub Web editor) | 2 files
11:01 Geth ¦ rakudo/nom: Merge pull request #1149 from MasterDuke17/make_INTERPOLATEs_parameters_non_optional
11:01 Geth ¦ rakudo/nom:
11:01 Geth ¦ rakudo/nom: Make INTERPOLATE's parameters non-optional
11:01 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/85a97ab3d9
11:25 vendethiel- joined #perl6-dev
11:38 vendethiel- joined #perl6-dev
12:26 Geth ¦ rakudo/jit_nativecall: 7162a23b71 | (Stefan Seifert)++ | lib/NativeCall.pm6
12:26 Geth ¦ rakudo/jit_nativecall: Properly unbox arguments to JIT compiled native functions
12:26 Geth ¦ rakudo/jit_nativecall:
12:26 Geth ¦ rakudo/jit_nativecall: We now create 2 locals for each argument: the param local and a properly
12:26 Geth ¦ rakudo/jit_nativecall: typed local for storing the unboxed value. As NativeCall permits passing
12:26 Geth ¦ rakudo/jit_nativecall: type objects to natively typed arguments, we need to handle those cases
12:26 Geth ¦ rakudo/jit_nativecall: and set default values manually.
12:26 Geth ¦ rakudo/jit_nativecall: review: https://github.com/rakudo/rakudo/commit/7162a23b71
12:52 Skarsnik_ joined #perl6-dev
12:54 Skarsnik__ joined #perl6-dev
13:01 skids joined #perl6-dev
13:02 lizmat samcv++  # grant work beyond the call of duty: https://cry.nu/perl6/grant-status-update-4/
13:03 astj joined #perl6-dev
14:27 Skarsnik_ joined #perl6-dev
15:16 samcv lizmat,  :-)
15:53 [Tux] samcv++
16:36 Geth ¦ rakudo/jit_nativecall: de489a1cd8 | (Stefan Seifert)++ | lib/NativeCall.pm6
16:36 Geth ¦ rakudo/jit_nativecall: Give generated native call function bodies a name
16:36 Geth ¦ rakudo/jit_nativecall:
16:36 Geth ¦ rakudo/jit_nativecall: nqp::setcodename only sets the name of the runtime code object. The static
16:36 Geth ¦ rakudo/jit_nativecall: frame takes its name from the block itself.
16:36 Geth ¦ rakudo/jit_nativecall: timotimo++ for giving the deciding hint!
16:36 Geth ¦ rakudo/jit_nativecall: review: https://github.com/rakudo/rakudo/commit/de489a1cd8
16:54 squashable6 joined #perl6-dev
17:08 astj joined #perl6-dev
17:30 AlexDaniel joined #perl6-dev
19:38 [Coke] (grant work)++ Thanks to samcv for posting updates.
19:40 [Coke] (and doing the work, of course!)
19:45 Skarsnik joined #perl6-dev
19:49 timotimo grand work, samcv++
19:57 moritz grand[dt] work indeed! :-)
19:57 moritz samcv++
20:12 committable6 joined #perl6-dev
20:25 AlexDaniel samcv?++
20:44 astj joined #perl6-dev
20:52 [Coke] so, how goes the SA?
22:14 astj joined #perl6-dev
22:55 Zoffix Is there any good reason why roast doesn't default to using, say, TEST_JOBS=6 instead of 1?
22:57 Zoffix Want to make it default to 6, so anyone isn't aware of needing to set jobs isn't waiting half hour for finish. Also, seems rakudo star's `make rakudos-spectest` doesn't set test jobs either, so instructions for install ask the user to run a half-hour command
22:59 Zoffix Oh, nevermind, looks like it does set something. Took just 4m to run: Files=1092, Tests=51258, 278 wallclock secs ( 7.61 usr  1.77 sys + 812.02 cusr 170.77 csys = 992.17 CPU)
22:59 Zoffix oh wait, I have TEST_JOBS set by default -_-
23:00 Zoffix :)
23:01 Zoffix yeah, it's a lot slower with TEST_JOBS unset :)
23:03 samcv thanks [Coke], timotimo , moritz, AlexDaniel :-)
23:26 Zoffix ZOFFLOP: t/spec/S17-lowlevel/atomic.t
23:28 Geth ¦ rakudo/nom: 8019c15bb8 | (Zoffix Znet)++ | 3 files
23:28 Geth ¦ rakudo/nom: Make t/harness* use 6 TEST_JOBS by default
23:28 Geth ¦ rakudo/nom:
23:28 Geth ¦ rakudo/nom: Many of installations instructions fail to specify the need to
23:28 Geth ¦ rakudo/nom: use TEST_JOBS (or users may miss them), which causes the spectest
23:28 Geth ¦ rakudo/nom: to take over half an hour to run.
23:28 Geth ¦ rakudo/nom:
23:28 Geth ¦ rakudo/nom: Default the jobs to 6, which is a more suitable default for many
23:28 Geth ¦ rakudo/nom: of the today's common boxes.
23:28 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/8019c15bb8

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