|
Perl 6 language and compiler development | Logs at colabti.org/irclogger/irclogger_log/perl6-dev | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm Set by Zoffix on 27 July 2018. |
|||
| AlexDaniel | lizmat: I think we broke something… | 00:09 | |
| lizmat | Oops, wasn't it supposed to be published yet > | 00:10 | |
| ? | |||
| AlexDaniel | lizmat: it was published 10 minutes ago, but I think you were editing an older version for some reason | ||
| which I updated like 15 minutes ago maybe, was fixing < > | |||
| lizmat: ok don't touch it for now let me fix it… | |||
| lizmat takes her hands off | |||
| AlexDaniel | it's unrecoverable… ok let me fix that manually then | 00:15 | |
| lizmat | AlexDaniel: s/it creates it’s own/it creates its own/ | ||
| AlexDaniel | “recover” feature just doesn't work, it's so frustrating :s | 00:16 | |
| lizmat | :-( | ||
| AlexDaniel | fucking crap now it *removed* some text… | 00:18 | |
| lizmat: fixed that s/// | 00:20 | ||
| lizmat | AlexDaniel++ # also cool article about your work on the bots! | ||
| lizmat has done enough damage on this new day so is off to bed | 00:21 | ||
| AlexDaniel | lizmat: it's fine, I'm not sure what happened there | 00:22 | |
| but wordpress just likes to eat < > | |||
| lizmat | yeah, it does :-( | ||
| AlexDaniel | all edits are there, it just ate all the IRC nicknames and stuff | 00:23 | |
| lizmat: thanks! | 00:33 | ||
| jnthn | AlexDaniel: Great post, and thanks for your excellent work on bisectable and Blin. | 00:42 | |
| AlexDaniel | thank you thank you thank you! :) | ||
| timotimo | AlexDaniel: sorry, i was AFK :) | 02:34 | |
|
03:49
evalable6 left,
evalable6 joined,
ChanServ sets mode: +v evalable6
03:50
p6bannerbot sets mode: +v evalable6
04:45
Kaiepi left,
Kaiepi joined
04:46
p6bannerbot sets mode: +v Kaiepi
|
|||
| AlexDaniel | timotimo: it's ok! liz did some nice corrections, then I realized that now all < and > are gone, then I was in panic for 10 minutes trying to roll back to a previous version using the broken history feature in wordpress | 05:52 | |
| hm, interesting | 05:55 | ||
| I dared to touch it again, and it < > are gone again | |||
| pfft alright, no more touching :) | 06:00 | ||
| fixed it, of course | |||
| robertle | AlexDaniel: just read your blog post, great stuff. are all these bots running on a machine in your basement, or is this some sort of hosted/cloud setup? | 08:22 | |
|
08:35
ufobat___ joined,
p6bannerbot sets mode: +v ufobat___
08:57
pmurias left
09:00
pmurias joined,
p6bannerbot sets mode: +v pmurias
10:08
pmurias left,
[Tux] left
10:12
pmurias joined,
p6bannerbot sets mode: +v pmurias
|
|||
| lizmat | Files=1261, Tests=86370, 378 wallclock secs (18.75 usr 5.94 sys + 2704.84 cusr 223.17 csys = 2952.70 CPU) | 10:16 | |
| ^^ not a fluke :-( | |||
|
10:16
pmurias left
10:18
pmurias joined,
p6bannerbot sets mode: +v pmurias
|
|||
| nine | lizmat: yeah, it's probably the slowdown that can be fixed with gist.github.com/niner/92729569c472...be9cea3d33 | 10:26 | |
| lizmat has just removed 3.6G of .precomp | |||
| we'll see tomorrow morning if that made a difference | 10:27 | ||
| (having removed the files) | |||
| and then I'll try the patch after that, ok ? | |||
| nine | sure | 10:41 | |
|
10:58
ufobat___ left
11:04
[Tux] joined
11:05
p6bannerbot sets mode: +v [Tux]
11:29
pmurias left
11:31
pmurias joined,
p6bannerbot sets mode: +v pmurias
|
|||
| AlexDaniel | robertle: all bots run on a dedicated server, for running Blin I use google compute engine (because more cores) | 12:09 | |
| robertle: if you're wondering, security is somewhat an issue. There are some basic protections in place, but someone dedicated will get around easily. Bots don't answer private messages though, somehow people are not very eager to poke bots in a malicious way publicly… so that kinda works | 12:19 | ||
|
12:33
[Tux] left
12:40
[Tux] joined
12:41
p6bannerbot sets mode: +v [Tux]
|
|||
| robertle | understood, I was also wondering about resources. is it enough or is the box too small? | 13:05 | |
| timotimo | AlexDaniel: maybe a "panic shutdown" message by a few trusted users could be useful if someone does show up | 13:11 | |
|
14:31
pmurias left
|
|||
| Geth | nqp: b0d32b635b | (Stefan Seifert)++ | t/moar/13-writeint.t Test readnum and writenum |
14:54 | |
| nqp: 86dbca4eab | (Stefan Seifert)++ | 11 files Rebootstrap for fixing writenum calls |
|||
|
15:10
travis-ci joined,
p6bannerbot sets mode: +v travis-ci
|
|||
| travis-ci | NQP build failed. Stefan Seifert 'Rebootstrap for fixing writenum calls' | 15:10 | |
| travis-ci.org/perl6/nqp/builds/471604743 github.com/perl6/nqp/compare/87ee8...dbca4eab48 | |||
|
15:10
travis-ci left
15:53
epony left
16:07
epony joined,
p6bannerbot sets mode: +v epony
17:07
lizmat left
17:29
pmurias joined
17:30
p6bannerbot sets mode: +v pmurias
|
|||
| pmurias | nine: re nqp::const:: being slower then literals don't they emit indentical code? | 17:30 | |
| nine | pmurias: no, I've seen some hash lookup in spesh logs | 17:42 | |
|
18:19
lizmat joined,
p6bannerbot sets mode: +v lizmat
18:42
pmurias left
20:19
Tux__ joined,
|Tux| left
20:20
p6bannerbot sets mode: +v Tux__
|
|||
| AlexDaniel | robertle: 21G of free space currently, and the whole data/ dir takes just 49G | 21:37 | |
| robertle: that's the beauty of long range compression :) you can give it two or more different (but similar) rakudo builds, and it ends up taking space of roughly one | 21:38 | ||
| timotimo: I thought about that, yeah. But then, trusted users can also kick users :) | 21:40 | ||
| maybe not from all channels, but still | |||
| robertle: regarding R#2564, so it's just a flapper? | 21:53 | ||
| synopsebot | R#2564 [open]: github.com/rakudo/rakudo/issues/2564 "Cannot invoke this object (REPR: Null; VMNull)" in t/04-nativecall/06-struct.t on mipsel | ||
| AlexDaniel | ah, but it's in t/, so that's not cool | 21:57 | |
| robertle | certainly does not happen every time | 22:30 | |
| I am currently trying to reproduce the third one, ppc64el, but it is surprisingly stubborn | 22:31 | ||
| Geth | rakudo: 83f0fd6c99 | (Nick Logan)++ (committed using GitHub Web editor) | src/core/CompUnit/Repository/FileSystem.pm6 Add distribution cache to CURFS Cache the generated Distribution to avoid doing directory recursion, as discovered by @niner. It also clears the cache when it finds an existing file that was not discovered during the initial generation of the Distribution. It is debatable if clearing the cache is ideal, but it is the current behavior according to a roast test that fails if caching is added without the clearer. |
23:15 | |
|
23:15
epony left
23:23
SqrtNegInf left
|
|||
| lizmat | notable6: weekly | 23:50 | |
| notable6 | lizmat, 2 notes: gist.github.com/d155aed54ea46fb06f...4477a537b0 | ||
|
23:54
lizmat left
23:56
lizmat joined,
p6bannerbot sets mode: +v lizmat
|
|||