Perl 6 - the future is here, just unevenly distributed

IRC log for #moarvm, 2017-01-13

| Channels | #moarvm index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:03 Ven joined #moarvm
00:23 Ven joined #moarvm
00:44 Ven joined #moarvm
00:48 pyrimidine joined #moarvm
01:23 Ven joined #moarvm
01:32 dalek joined #moarvm
01:35 pyrimidine joined #moarvm
01:44 Ven joined #moarvm
01:50 leego joined #moarvm
01:55 leego joined #moarvm
01:58 pyrimidine joined #moarvm
02:01 FROGGS_ joined #moarvm
02:10 pyrimidine joined #moarvm
02:23 Ven joined #moarvm
02:46 Ven joined #moarvm
02:49 ilbot3 joined #moarvm
02:49 Topic for #moarvm is now https://github.com/moarvm/moarvm | IRC logs at  http://irclog.perlgeek.de/moarvm/today
03:03 Ven joined #moarvm
03:18 ggoebel joined #moarvm
03:20 pyrimidine joined #moarvm
03:23 Ven joined #moarvm
03:42 Ven joined #moarvm
04:02 Ven joined #moarvm
04:22 Ven joined #moarvm
04:43 Ven joined #moarvm
05:03 Ven joined #moarvm
05:05 pyrimidine joined #moarvm
05:22 Ven joined #moarvm
05:42 Ven joined #moarvm
06:02 Ven joined #moarvm
06:14 pyrimidine joined #moarvm
06:22 Ven joined #moarvm
06:22 nwc10 jnthn: spectest6 eventually gave up with a non-ASAN failure after just over 24 hours of looping: http://paste.scsys.co.uk/540693
06:22 nwc10 so, I think that makes it a good moarning
06:22 nwc10 (not perfect, but very much better)
06:48 pyrimidine joined #moarvm
07:00 pyrimidine joined #moarvm
07:08 domidumont joined #moarvm
07:15 domidumont joined #moarvm
07:29 JimmyZ joined #moarvm
07:30 geekosaur joined #moarvm
07:32 ilmari joined #moarvm
07:35 leego joined #moarvm
07:39 geekosaur joined #moarvm
07:41 pyrimidine joined #moarvm
07:57 geekosaur joined #moarvm
07:57 samcv nwc10, do you usually build with clang or gcc?
08:04 leego joined #moarvm
08:06 geekosaur joined #moarvm
08:11 pyrimidine joined #moarvm
09:01 zakharyas joined #moarvm
09:08 pyrimidine joined #moarvm
09:49 pyrimidine joined #moarvm
10:07 pyrimidine joined #moarvm
10:36 pyrimidine joined #moarvm
10:51 pyrimidine joined #moarvm
11:03 pyrimidine joined #moarvm
11:07 pyrimidine joined #moarvm
11:37 geekosaur joined #moarvm
11:52 pyrimidine joined #moarvm
11:52 Ven joined #moarvm
12:11 Ven joined #moarvm
12:31 pyrimidine joined #moarvm
12:31 Ven joined #moarvm
12:50 Ven joined #moarvm
13:23 jnthn samcv: One comment but it's not very important...
13:23 samcv yeah i noticed that. it was already there I think before I changed things
13:24 jnthn Yeah, I noticed that too :)
13:24 samcv yeah so. gonna just leave it for now I think. just in case
13:24 jnthn Maybe I'll just merge it and clean up later? Or you want to do it now?
13:24 jnthn OK
13:24 samcv nah just leave it in now
13:24 jnthn Also noticed we call the struct NamedAlias but elsewhere it's name alias
13:25 jnthn Again, don't think I'm bothered enough about this to suggest we change it :)
13:29 Geth MoarVM: 816186484b | (Samantha McVey)++ | 4 files
13:29 Geth MoarVM: Remove Unicode 1 names and add Unicode Name Aliases
13:29 Geth MoarVM:
13:29 Geth MoarVM: This adds support in the codepointfromname op for all of the official
13:29 Geth MoarVM: Unicode name aliases. It removes the old Unicode 1 names in the name
13:29 Geth MoarVM: registry and replaces them with <control-xxxx> with the codepoint
13:29 Geth MoarVM: number.
13:29 Geth MoarVM:
13:29 Geth MoarVM: There are only 5 codepoints which have changed their functionality
13:29 Geth MoarVM: aside from the 400+ Name Aliases added:
13:29 Geth MoarVM:
13:29 Geth MoarVM: Correction:
13:29 Geth MoarVM:   * BELL
13:29 Geth MoarVM: <…commit message has 18 more lines…>
13:29 Geth MoarVM: review: https://github.com/MoarVM/MoarVM/commit/816186484b
13:29 Geth MoarVM: fc12a60ec3 | (Jonathan Worthington)++ | 4 files
13:29 Geth MoarVM: Merge pull request #497 from samcv/nou1names
13:29 Geth MoarVM:
13:29 Geth MoarVM: Remove Unicode 1 names and add Unicode Name Aliases
13:29 Geth MoarVM: review: https://github.com/MoarVM/MoarVM/commit/fc12a60ec3
13:50 pyrimidine joined #moarvm
14:05 brrt joined #moarvm
14:08 pyrimidine joined #moarvm
15:05 pyrimidine joined #moarvm
15:26 pyrimidine joined #moarvm
15:46 pyrimidine joined #moarvm
15:52 Geth MoarVM: 8442892451 | MasterDuke17++ | 8 files
15:52 Geth MoarVM: Convert MVM_malloc+memset to MVM_calloc
15:52 Geth MoarVM:
15:52 Geth MoarVM: In cases of `foo = MVM_malloc(...); memset(foo, 0, ...);`, convert them
15:52 Geth MoarVM: to `foo = MVM_calloc(...);
15:52 Geth MoarVM: review: https://github.com/MoarVM/MoarVM/commit/8442892451
15:52 Geth MoarVM: d9482b9ba3 | (Jonathan Worthington)++ | 8 files
15:52 Geth MoarVM: Merge pull request #498 from MasterDuke17/replace_malloc_plus_memset_zero_with_calloc
15:52 Geth MoarVM:
15:52 Geth MoarVM: Convert MVM_malloc+memset to MVM_calloc
15:52 Geth MoarVM: review: https://github.com/MoarVM/MoarVM/commit/d9482b9ba3
15:53 Geth MoarVM/master: 5 commits pushed by jnthn++
15:53 Geth MoarVM/master: 1f20dc77bb | Fix GC in spesh triggered by managed mutex use.…
15:53 Geth MoarVM/master: fc771134f2 | Fix data race in inlining extop fixup.…
15:53 Geth MoarVM/master: 5389d6080b | Fix data race in callsite fixup during inlining.…
15:53 Geth MoarVM/master: ef4d6a767a | Fix data race in string fixup during inlining.…
15:53 Geth MoarVM/master: d1da1bac41 | Merge pull request #495 from MoarVM/fix-inline-and-threads…
15:53 Geth MoarVM/master: review: https://github.com/MoarVM/MoarVM/compare/1f20dc77bb…d1da1bac41
15:54 vendethiel joined #moarvm
15:55 jnthn That's pretty :)
15:58 timotimo yup
17:36 domidumont joined #moarvm
17:49 nwc10 ASAN is very excited by generate_codepoints_by_name
17:49 nwc10 for example http://paste.scsys.co.uk/540698
17:50 nwc10 I hope that the causeis obvious to someone. If not, I'll bisect
17:55 timotimo i'm glad we have you :)
17:56 lizmat nwc10: pretty sure samcv will have a look at it when she's awake again :-)
18:11 vendethiel- joined #moarvm
18:35 vendethiel joined #moarvm
19:50 vendethiel joined #moarvm
20:07 timotimo samcv: does the backtrace in nwc10's paste ring a bell to you?
20:09 nwc10 I did bisect it. It's commit 816186484b5cc52f9ff1be6afa3b6f49264335bf
20:10 nwc10 Remove Unicode 1 names and add Unicode Name Aliases
20:17 vendethiel joined #moarvm
20:39 samcv yeah that struct didn't even exist before that commit
20:39 samcv i am mostly awake let me look at it one sec
20:50 vendethiel joined #moarvm
21:08 samcv how do I work with asan? I did ./Configure.pl --asan
21:08 samcv not getting any messages
21:09 samcv weird perl6 still works. but running moar i get that it's compiled with incompatible asan libs
21:09 samcv ==21995==Your application is linked against incompatible ASan runtimes.
21:11 samcv i have clang 3.9.1-2
21:13 timotimo i haven't been able to build moarvm with asan last time i tried
21:15 jnthn samcv: If you have valgrind installed, it may be easier (doesn't need a special build)
21:15 jnthn It's slower, but should give at least as good diagnostics (sometimes better)
21:15 jnthn If you're on Linux valgrind should be a really easy install
21:21 samcv ok
21:21 samcv i have valgrind
21:22 jnthn \o/
21:22 jnthn There's a perl6-valgrind-m script btw
21:22 jnthn Which makes things super convenient, except on the occasion it doesn't :)
21:23 samcv yeah i have used it. i made my own perl6-callgrind script as well
21:25 jnthn ooh
21:27 jnthn I wondered a bit about making the perl6-valgrind-m respect a VALGRIND_TOOL env var or something so I could do VALGRIND_TOOL=massif perl6-valgrind-m ...
21:27 jnthn Which may be neater than a proliferation of scripts
21:29 timotimo right
21:38 samcv jnthn, how do i do leakcheck full using the valgrind script?
21:45 jnthn samcv: Lsast time I needed that I just edited it and put the arg after the valgrind call
21:45 samcv kk
21:45 jnthn (and was careful not to commit it :))
21:45 samcv heh
21:46 jnthn Maybe a generic VALGRIND_ARGS would be more useful
21:47 samcv m: say "\c[keycap: #, HUNDRED POINTS SYMBOL]"
21:47 camelia rakudo-moar 232641: OUTPUT«===SORRY!=== Error while compiling <tmp>␤Unrecognized character name keycap:␤at <tmp>:1␤------> say "\c[keycap:⏏ #, HUNDRED POINTS SYMBOL]"␤»
21:47 samcv m: say "\c[keycap: #]"
21:47 camelia rakudo-moar 232641: OUTPUT«===SORRY!=== Error while compiling <tmp>␤Unrecognized character name keycap:␤at <tmp>:1␤------> say "\c[keycap:⏏ #]"␤»
21:47 samcv m: say "\c[keycap: 3]"
21:47 camelia rakudo-moar 232641: OUTPUT«3️⃣␤»
21:47 samcv doesn't like the hash symbol hmm
21:47 samcv m: say "\c[keycap: #]"
21:47 camelia rakudo-moar 232641: OUTPUT«===SORRY!=== Error while compiling <tmp>␤Unrecognized character name keycap:␤at <tmp>:1␤------> say "\c[keycap:⏏ #]"␤»
21:51 jnthn samcv: Might need to loosen stuff up here: https://github.com/perl6/nqp/blob/master/src/HLL/Grammar.nqp#L157
21:51 jnthn ohh
21:51 jnthn Wow, we allow comments inside of \c[...] ??
21:51 samcv why…
21:52 jnthn I have no clue!
21:52 samcv it's basically already commented :P
21:52 jnthn Just reading the grammar and guessing :P
21:53 jnthn (Guessing that's why # is special-cased, I mean. But it calls <.ws> which will parse a comment.
21:55 timotimo we're mad!
21:55 samcv ok i got asan working
22:02 jnthn Nice
22:02 jnthn games &
22:02 samcv ok I fixed asan. will PR in a sec
22:07 Geth MoarVM: samcv++ created pull request #499: Fix ASAN errors in generate_codepoints_by_name
22:07 Geth MoarVM: review: https://github.com/MoarVM/MoarVM/pull/499
22:08 lizmat ah, the infamous off-by-one plot!
22:08 samcv yeah.
22:08 samcv i had to change it to iterate down instead of up
22:09 samcv well I maybe shouldn't have. i guess I _could_ have done it from 0
22:09 samcv oh well it works now and still does every alias
22:10 lizmat I'll leave it to jnthn to merge, since it is not a blocker for you now, right ?
22:11 samcv no error results
22:11 samcv if it was off by the opposite side it would fault
23:58 * jnthn checks le PR

| Channels | #moarvm index | Today | | Search | Google Search | Plain-Text | summary