|
Parrot 6.1.0 "Black-collared Lovebird" | parrot.org/ | Log: irclog.perlgeek.de/parrot | #parrotsketch meeting Tuesday 19:30 UTC Set by moderator on 18 February 2014. |
|||
|
00:18
rurban1 joined
01:06
rurban1 joined
02:45
rurban1 joined
03:18
rurban1 joined
04:19
rurban1 joined
05:20
rurban1 joined
05:23
rurban2 joined
06:24
rurban1 joined
06:58
davidfetter joined
07:24
rurban1 joined
08:25
rurban1 joined
09:17
vincent21 joined
09:26
rurban1 joined
10:27
rurban1 joined
11:09
denis_boyun joined
11:23
PerlJam joined,
Util joined
11:24
Coke joined
11:29
rurban1 joined
12:30
rurban1 joined
13:01
webart joined
13:30
rurban1 joined
14:07
rurban1 joined
|
|||
| rurban | seems like windows builds are pretty broken: github.com/parrot/parrot/issues/1063 | 14:18 | |
|
14:36
travis-ci joined
|
|||
| travis-ci | [travis-ci] parrot/parrot#1006 (master - d22c064 : Paul Cochrane): The build passed. | 14:36 | |
| [travis-ci] Change view : github.com/parrot/parrot/compare/e...2c064a3540 | |||
| [travis-ci] Build details : travis-ci.org/parrot/parrot/builds/20939522 | |||
|
14:36
travis-ci left
|
|||
| rurban | dalek seems to be gone | 14:49 | |
| Didn't we have a Bender also? | 14:56 | ||
| I only see 5 admins | |||
|
14:59
travis-ci joined
|
|||
| travis-ci | [travis-ci] parrot/parrot#1007 (master - 479c806 : Reini Urban): The build passed. | 14:59 | |
| [travis-ci] Change view : github.com/parrot/parrot/compare/d...9c8066a55c | |||
| [travis-ci] Build details : travis-ci.org/parrot/parrot/builds/20939771 | |||
|
14:59
travis-ci left
15:01
webart left
15:03
dalek joined
|
|||
| rurban | thanks! | 15:06 | |
|
15:13
pmichaud joined
16:53
Chirag joined
|
|||
| rurban | Something like this: github.com/parrot/parrot/issues/1065 (parrot openpip env design) | 17:22 | |
| dalek | rrot/rurban/openpipe-env-gh1065: 30ba70a | rurban++ | / (2 files): [pmc] filehandle['env'] = env GH #1065 Support writable env attribute for a filehandle, needed to set the child env for pipes. The env for the child pipe cannot be set currently, we can only set the env for spawn and shell. |
17:26 | |
| rrot/rurban/openpipe-env-gh1065: fda1d06 | rurban++ | / (2 files): [pmc] filehandle['env'] = env GH #1065 Support writable env attribute for a filehandle, needed to set the child env for pipes. The env for the child pipe cannot be set currently, we can only set the env for spawn and shell. |
17:43 | ||
|
17:49
denisboyun joined
|
|||
| rurban | We need to revert -Werror=missing-prototypes | 18:24 | |
| rakudo is not clean enough yet, and they are blaming us ("parrrot change broke rakudo", ...) | 18:25 | ||
| esp. ops2c C --dynamic perl6.ops doesn't generate a Parrot_lib_perl6_ops_init prototype | 18:27 | ||
| hmm, but PARROT_DYNOP_PERL6_INIT is properly declared | 18:31 | ||
| Coke | I think in an ideal world, rakudo wouldn't care about the compiler warnings used to build one of the vms it's targetting. I wonder if decoupling that would be the right move. | ||
| rurban | Found it, rakudo's fault. I'll fix it | 18:33 | |
| I rather prefer to lead rakudo with our warnings. They have more such bugs than we. | 18:34 | ||
| They seem to ignore compiler warnings in general, and making it fatal seems to be the right move. | 18:35 | ||
| Coke | not going to touch that with a 10 meter cattle prod. | 18:40 | |
| as long as rakudo/nqp can update to the latest version of parrot when there's a new feature, seems fine. | |||
| rurban | currently rakudo is broken with latest parrot because they did not merged latest nqp (which fixed those prototypes warnings) | 18:42 | |
| merge => demand | 18:44 | ||
| trying now rakudo with parrot -O2 and latest fixes | 18:47 | ||
| Coke | rakudo doesn't depend on parrot, it depends on nqp. | 19:04 | |
| so if there's an nqp that fixes interaction with parrot, to use that parrot, the version of nqp required by rakudo needs to be bumped as well. | |||
| Chirag | Hey! a quick question - which part of the compiler is rakudo facing a problem with? | 19:22 | |
| ops2c? | |||
| rurban | no, they had their own ops problem. see github.com/rakudo/rakudo/pull/252 | 19:26 | |
| Coke: Yes, I know. But I will not decide which nqp they want. I just fixed nqp master | 19:27 | ||
| So they can move forward if wanted | |||
| Chirag | oh.. so is it like they did not change rakudo to work with parrot 6.1 ? | 19:31 | |
| Coke | normal release process for rakudo is to -not- upgrade to latest parrot. We upgrade when there's some new feature or bugfix needed. | 19:35 | |
| theoretically helps packagers. | |||
| rurban | I know. I thought they'd like -O2 | 19:36 | |
| Chirag | Coke: understood.. | 19:37 | |
| Coke | mmm, speed good. | 19:50 | |
|
19:51
prammer joined
|
|||
| rurban | There should be a lot of constant propagation in rakudo, but I have no timings yet. | 20:08 | |
|
20:12
kid51 joined
|
|||
| rurban | Chirag: You need to see github.com/parrot/misc/blob/master/timeline.md on whiteknight's PCC plans and benchmarks | 20:20 | |
| Chirag | sure! I will go through.. | 20:23 | |
| rurban | whiteknight.github.io/2011/05/11/pc...ments.html is a good writeup | 20:24 | |
| Chirag | the timeline was posted 3 years ago .. so some of the taks must have been completed right? | 20:28 | |
| rurban | yes. see gist.github.com/Whiteknight/1198521 in the comments | 20:31 | |
| but pcc not | |||
| I'll try the invokecc and returncc improvements as new ops and in the imcc compiler | 20:32 | ||
| e.g. a new new_call_context | 20:33 | ||
| this will reduce the needed signature arrays | |||
| Chirag | ok.. so will this tackle the problem of the resizablePMC array as well? | 20:39 | |
| rurban | maybe | ||
| the resize problem is not so big, fill_params is | 20:40 | ||
| Chirag | hmm.. i have look into the imcc compiler.. its working | ||
| rurban | I can do the imcc part. it's a big harder | 20:41 | |
| bit harder | |||
| we need to work together anyway | 20:42 | ||
| Chirag | oh... alright | ||
| I am just going through the docs actually.. | |||
| btw I was just curious to know if my proposal can be seen by all the admins? | 20:49 | ||
| rurban | yes, you are the only one so far | ||
| Chirag | :) | ||
| you saw highway? | 20:51 | ||
| rurban | No, just Queen yesterday | 20:52 | |
| Chirag | ah .. hows it? | ||
| rurban | It has Lisa Haydon in int :) | 20:53 | |
| It's totally european. cheap but good camera, plays in paris and amsterdam | 20:54 | ||
| very realistic | |||
| everybody liked it. | |||
| Rating 9.1: www.imdb.com/title/tt3322420/ | |||
| Chirag | whoa.. | ||
| rurban | not kitschy at all | ||
| Chirag | but i dont go by imdb | ||
| rurban | it's pretty unreliable with hindi films, yes | 20:55 | |
| Chirag | i find it misleading | ||
| even with english | |||
| rurban | too many fans, and no critics | ||
| Chirag | yes | ||
| i saw now u see me | |||
| too far-fetched | |||
| rurban | oh god, the ending! | 20:56 | |
| Chirag | i know! | ||
| rurban | Mark Ruffalo blows it in the end. But two thirds are pretty good | ||
| Chirag | had they made it supernatural altogether .. i would have still accepted it :P | 20:57 | |
| rurban | oh my | ||
| but it would have been a bit better at least | |||
| Chirag | yes.. | ||
| i'll leave now.. | 20:59 | ||
| Have a nice day! | |||
| rurban | bye! | 21:00 | |
| m headerizer | 21:20 | ||
| rebased 6model, seems to work fine. 200 new ops, uh | 22:20 | ||
| dalek | Heuristic branch merge: pushed 24 commits to parrot/rurban/6model by rurban | 22:46 | |
|
22:46
MadHatter87 joined
|
|||
| dalek | rrot/rurban/6model: bb6bbb3 | rurban++ | MANIFEST (2 files): manifest: regenerate for 6model |
22:47 | |
|
22:47
MadHatter87 joined
22:50
MadHatter87 left,
MadHatter87 joined
23:16
MadHatter87 left
|
|||
| dalek | rrot/rurban/6model: d6c9ce0 | rurban++ | / (7 files): 6model: add set_invocation_spec and set_boolification_spec ops from nqp: How do we invoke this thing? (attribute or method to call) How do we turn something of this type into a boolean? |
23:22 | |