|
Parrot 6.5.0 "Black-winged Lovebird" | parrot.org/ | Log: irclog.perlgeek.de/parrot | #parrotsketch meeting Tuesday 19:30 UTC Set by moderator on 17 June 2014. |
|||
| dalek | rrot: 7675f4b | rurban++ | / (2 files): [config] fix auto::libffi to expand -L$(libdir) pkg-config may store makefile macros. expand them if possible. fixes GH #1082 with -L$(libdir) in the linkflags of libffi-3.1.pc which broke auto::expext. Also fix stylistic issues and document TEST_PKGCONFIG. |
00:34 | |
|
00:35
travis-ci joined
|
|||
| travis-ci | [travis-ci] parrot/parrot#1055 (smoke-me/libffi-gh1082 - 7675f4b : Reini Urban): The build passed. | 00:35 | |
| [travis-ci] Change view : github.com/parrot/parrot/compare/s...ffi-gh1082 | |||
| [travis-ci] Build details : travis-ci.org/parrot/parrot/builds/29280653 | |||
|
00:35
travis-ci left
00:45
travis-ci joined
|
|||
| travis-ci | [travis-ci] parrot/parrot#1056 (master - 7675f4b : Reini Urban): The build passed. | 00:45 | |
| [travis-ci] Change view : github.com/parrot/parrot/compare/d...75f4bcf31d | |||
| [travis-ci] Build details : travis-ci.org/parrot/parrot/builds/29280922 | |||
|
00:45
travis-ci left
|
|||
| dalek | rrot/smoke-me/cygwin-tempfile-gh1077: bb757e0 | rurban++ | / (2 files): [test] fix wrong filehandle test cygwin 5.18 cannot do Cwd::abs_path on a non-existent file. Workaround that. The underlying problem is that we do not need to UNLINK a not-created file. Fixes GH #1077 on cygwin. |
01:13 | |
| rrot/smoke-me/cygwin-tempfile-gh1077: b1b4b87 | rurban++ | / (3 files): [test] fix wrong filehandle test cygwin 5.18 cannot do Cwd::abs_path on a non-existent file. Workaround that. The underlying problem is that we do not need to UNLINK a not-created file. Fixes GH #1077 on cygwin. |
01:20 | ||
|
01:32
travis-ci joined
|
|||
| travis-ci | [travis-ci] parrot/parrot#1058 (smoke-me/cygwin-tempfile-gh1077 - b1b4b87 : Reini Urban): The build passed. | 01:32 | |
| [travis-ci] Change view : github.com/parrot/parrot/commit/b1b4b875888b | |||
| [travis-ci] Build details : travis-ci.org/parrot/parrot/builds/29282524 | |||
|
01:32
travis-ci left
01:46
FROGGS_ joined
|
|||
| dalek | rrot: b1b4b87 | rurban++ | / (3 files): [test] fix wrong filehandle test cygwin 5.18 cannot do Cwd::abs_path on a non-existent file. Workaround that. The underlying problem is that we do not need to UNLINK a not-created file. Fixes GH #1077 on cygwin. |
01:54 | |
|
02:07
travis-ci joined
|
|||
| travis-ci | [travis-ci] parrot/parrot#1059 (master - b1b4b87 : Reini Urban): The build passed. | 02:07 | |
| [travis-ci] Change view : github.com/parrot/parrot/compare/7...b4b875888b | |||
| [travis-ci] Build details : travis-ci.org/parrot/parrot/builds/29284009 | |||
|
02:07
travis-ci left
06:15
bighugedog joined
06:57
FROGGS joined
08:49
basiliscos joined
13:32
rurban1 joined
13:53
Chirag joined
|
|||
| Chirag | rurban: Hey! Did you figure out the commit? I am stuck at my end... | 14:04 | |
|
14:07
bluescreen joined
|
|||
| rurban | Hi. Sorry over the weekend I had no access to irc.perl.org from my laptop | 14:16 | |
| what exactly is failing in Configure RELEASE 2.7- 2.8 for you? | 14:17 | ||
| it could be icu | |||
| I've benched through all single commits from 2.8 => 3.0, but haven't found culprits yet. will upload the log soon | 14:18 | ||
| Chirag | my hdd had crashed a few days back.. took a while to reset things.. | 14:20 | |
| On my system, bench.sh does not work with RELEASE 2.7- 2.8 .. Actually, Configure.pl (from the initial commits) itself fails.. so make fails.. | 14:21 | ||
| rurban | yes, I read it at irclog.perlgeek.de/parrot/2014-07-05 | ||
| what exactly fails? | |||
| Chirag | init::manifest - Check MANIFEST.....................................done."my" variable $lib masks earlier declaration in same statement at config/init/defaults.pm line 291. | 14:22 | |
| syntax error at config/init/defaults.pm line 285, near "$cc qw(cc cxx link ld)" | |||
| and associated errors | |||
| rurban | oh, you need to patch that manually then | ||
| Chirag | but if I loop through the commits.. won't I lose it after every clean? | 14:23 | |
| rurban | if you got a patch, just apply it automatically in bench.sh | 14:24 | |
|
14:24
Chirag_ joined
|
|||
| Chirag_ | hmm.. I will do that.. | 14:25 | |
| rurban | but I don't have that problem, and I don't see any. which commit exactly? RELEASE_2_8_0? | 14:26 | |
| Chirag_ | for eg. f61b1d2fb32a43d11442fabbfa33fb00cea87fd0 | 14:27 | |
| rurban | ah, maybe you are using a too new perl. my $cc qw() is now illegal | ||
| I use a perl 5.14 | |||
| Chirag_ | oh | 14:28 | |
| rurban | yes, with 5.18 I get the same error. can you get a 5.14 perl? | ||
| Chirag_ | sure.. | ||
| rurban | And there's no need for hurry. Getting a 3rd task done would be nice, but I doubt it. | 14:35 | |
| It will be not that easy, I guess | |||
| Chirag_ | I did not find myself contributing much through task#2 :( | 14:38 | |
| rurban | Well, you did fine. I just rewrote it :) | 14:41 | |
| Chirag_ | :) | 14:43 | |
| Also, I wanted to run this by you.. Instead of a thesis, I am taking up an internship for the upcoming semester (part of my curriculum)... The dates will clash.. But since it's a day job, I will be able to give the same amount of time as now. So I hope it won't be a problem, right? | 14:47 | ||
| rurban | when does it start? | 14:50 | |
| end of gsoc work is august 11 | 14:51 | ||
| Chirag_ | They want me to start from 14 July but I am requesting them to postpone.. | 14:52 | |
| But I can assure that I will be able to give in the same amount of time anyhow... | 14:54 | ||
| rurban | We will see. you can benchmark single commits in the bad ranges to find out more culprits | ||
| this could run unattended. for the analysis we'll have to do it together anyway | |||
| Chirag_ | hmm.. We correspond in the evenings (IST), so it wont be disrupted, for sure.. | 14:56 | |
| rurban | one really bad commit is 2_11_0-614-gfdb6fa5\t8.486329896\t +- 0.49% plobsing: re-implement implicit :main | 15:14 | |
| 42% slowdown | 15:15 | ||
| Chirag_ | ooh.. too much.. that must be the one for the ~13% build | 15:16 | |
| rurban | but this could be because the previous commits were not correct. some didn't run at all | ||
| pushed my logs to github.com/parrot/parrot-bench | 15:20 | ||
| log.bench-3.0.slowdown.numbers | |||
| Chirag_ | u covered all from 2.7-2.8? | 15:24 | |
| rurban | nope | ||
| just until 2.9 | 15:25 | ||
| only 2.9, 2.10, 2.11, 3.0 | |||
| Chirag_ | oh.. the ones like 2_8 gave a false impression | 15:26 | |
| installing perl 5.14.4 .. taking a lot of time.. | 15:27 | ||
| rurban | yes | 15:29 | |
| found no .deb for it? | |||
| Chirag_ | was using perlbrew... didnt realize it would take so long | 15:30 | |
| rurban | ~ 15 minutes | 15:31 | |
| and you need TAP::Harness::Archive also | 15:33 | ||
| for make smoke | |||
| Chirag_ | oh.. i will install that as well.. | 15:37 | |
| less RAM is making it slow... | 15:38 | ||
| rurban | and the make test step is slow | 15:39 | |
|
15:48
FROGGS joined
16:11
rurban1 joined
16:14
basiliscos joined
16:32
null0 left
|
|||
| Chirag_ | I am still getting that error.. definitely running 5.14.4... but maybe havent extracted TAP::Harness::Archive correctly | 17:05 | |
| rurban: should I fall back further from 5.14.4? | 17:09 | ||
| rurban | 5.14.4 is fine | 17:14 | |
| did you run your perl properly? | 17:15 | ||
| what is the name of your perl-5.14? perlbrew? | |||
| if perlbrew, do perlbrew switch 5.14.4 | 17:16 | ||
| Chirag_ | yes i did that | ||
| rurban | hmm, then you shouldn't get this error anymore. I avoid perlbrew because I hate it | 17:17 | |
| Chirag_ | it gives a subshell bash-4.3 | ||
| rurban | it seems to be highly environment specific | ||
| Chirag_ | i ll try finding a .deb now | 17:18 | |
| how do I handle TAP::Harness::Archive? extract and merge with Parrot.. these are part of a tar.gz | |||
| rurban | just $perl -S cpan TAP::Harness::Archive | 17:19 | |
| there is no extract and merge with Parrot | |||
| it is only needed for make smoke | |||
| Chirag_ | oh.. thanks! | 17:20 | |
|
21:22
basiliscos joined
23:43
kid51 joined
|
|||
| kid51 | $ perl tools/dev/resubmit_smolder.pl | 23:43 | |
| Could not upload report to Smolder at smolder.parrot.org | |||
| HTTP CODE: 500 (Internal Server Error) | |||
| Does anyone recall how we get that fixed? | 23:44 | ||
| Util | kid51: When it has been a problem in the past, smolder.parrot.org/ itself was not responding. | 23:45 | |
| kid51 | Yes | ||
| Util | Before I ask osuosl to reboot it, would you verify that smolder.parrot.org/ responds for you (it does for me, just now) in a web browser, | 23:46 | |
| kid51 | smolder.parrot.org/app/projects/smoke_reports/1 | ||
| Util | and also try resubmit_smolder.pl once more. | ||
| kid51 | The web site is there, but all the data for each report is '0' | ||
| or "Unknown" | |||
| Util | Ah! | 23:47 | |
| Never seen *that* before! | |||
| kid51 | I've just forgotten the address for osuosl | ||
| But that's what I usually see when the server is not accepting uploads. | |||
| Util | kid51: support@osuosl.org | 23:48 | |
| I was heading out to class; do you have time to submit a ticket? | |||
| kid51 | will do | 23:49 | |
| Util | Thanks! | 23:50 | |
| Also, if you want to reach them on IRC: /JOIN -freenode #osuosl | |||
|
23:50
janus joined
|
|||
| Util | I will check back in, in 3 hours. | 23:50 | |
| kid51 | Hmm, this may be my problem more than OSUOSL's. | 23:53 | |
| I just successfully uploaded a report from my Mac | |||
| But the failures are coming from my Linode | 23:54 | ||
| In the past, both worked. | |||
| For hosting support and inquiries, please email support@osuosl.org.. This IRC channel is a secondary means of communication meant to supplement the support tickets generated through that email address. | 23:56 | ||
| So I'll file a ticket first. | |||