|
01:48
ilbot3 joined
03:00
lizmat joined
03:05
lizmat_ joined
05:57
oetiker joined
06:02
FROGGS joined
06:03
oetiker joined
06:06
oetiker joined
06:24
oetiker joined
06:55
zakharyas joined
06:56
oetiker joined
07:34
Ven joined
07:41
brrt joined
|
|||
| brrt | \o | 07:42 | |
| nwc10 | o/ | ||
| brrt | today i start hacking! | 07:45 | |
| nwc10 | rah! | ||
| brrt | :-) | ||
| nwc10 | any firm idea on where you're going to start? | ||
| FROGGS | beer! /o/ | 07:46 | |
| :P | |||
| nwc10 | even *I* think that it's a bit early for beer | ||
| give it half an hour or so... | |||
| FROGGS | *g* | ||
| brrt | tea! | 07:49 | |
| well, i was going to start by creating a prototype outside the moar code branch | |||
| i was thinking of naming the branch: even-moar-jit | 07:50 | ||
| and i'm going to start hacking on dynasm sometime | |||
|
08:59
lizmat joined
11:00
TimToady joined
11:42
Ven joined
|
|||
| masak | brrt++ # today he starts hacking! | 11:58 | |
| tadzik | yay :) | 11:59 | |
| FROGGS throws in qa.debian.org/bls/packages/m/moarvm.html | 12:19 | ||
| masak | MoarVM is a Debian package? cool! | 12:23 | |
| could something be done about those rather unsightly "7 issues"? | 12:24 | ||
| FROGGS | masak: look at: tracker.debian.org/pkg/moarvm | ||
| :o) | |||
| masak: yes, of course | |||
| masak | \o/ | 12:26 | |
| nwc10 | masak: see also the \o/ at irclog.perlgeek.de/moarvm/2015-06-14#i_10749126 | 12:27 | |
| FROGGS | tracker.debian.org/pkg/nqp is still at 2014.07, but will also be updated thanks to nebuchadnezzar and dod | ||
| masak | heh, interp.c is huge, and it's basically just one big switch statement :P | 12:31 | |
| FROGGS | well, it is true :o) | ||
| masak | there are 81 lines before and 16 lines after the switch statement, which itself is 4891 lines ;) | 12:32 | |
| arnsholt | Any idea why the build is reproducible? Is it just the SC ids being based on time and such, or something more problematic? | 13:00 | |
| s/is/isn't/ | |||
| moritz | that's also the first thing that came to my mind | 13:02 | |
| and you can't artifically set the timestamps all to the same time | |||
| cause then the different stages will get identical SC IDs, and things blow up prettily | |||
| nwc10 | you're thinking about NQP, I think | 13:03 | |
| this is MoarVM - does it compile anything? | |||
| arnsholt | Yeah, that's why I wasn't sure | ||
| moritz | right, NQP | ||
| FROGGS | it is about guids and stuff | 13:17 | |
| ahh, that explains it: wiki.debian.org/ReproducibleBuilds/Howto | 13:18 | ||
| stuff like: "Files in data.tar contains hostname, uname output, username" | 13:19 | ||
|
14:04
brrt joined
14:09
Ven joined
|
|||
| nebuchadnezzar | hello | 14:13 | |
| I'm trying to understand github.com/MoarVM/MoarVM/commit/d2...9281700ca, since it makes lintian unhappy (lists.alioth.debian.org/pipermail/...0567.html) | 14:14 | ||
| If I understand correctly, this directory will be use by rakudo packages which install a NativeCall library? | 14:15 | ||
| FROGGS | nebuchadnezzar: correct | 14:19 | |
| nebuchadnezzar | ok, so we need the rpath | 14:20 | |
| FROGGS: so MoarVM needs to find thoses libraries | 14:21 | ||
| FROGGS | nebuchadnezzar: yes, I think so | 14:22 | |
| we might be able to work around that | 14:23 | ||
| but I'm not sure | |||
| nebuchadnezzar | Thanks, will see what dod will say about this | 14:26 | |
| FROGGS | thank you | ||
|
15:29
Ven joined
15:41
lizmat joined,
FROGGS joined
|
|||
| FROGGS | froggs@kfreebsd:~/dev/MoarVM$ make | 17:42 | |
| linking moar | |||
| ./libmoar.so: undefined reference to `pthread_getaffinity_np' | |||
| :o( | |||
| tadzik | :o | 17:56 | |
| masak | o( | 18:11 | |
| FROGGS | ha! | 18:28 | |
| fixed it | |||
| (I think) | |||
| nqp will tell | 18:30 | ||
|
18:33
japhb joined
|
|||
| FROGGS | All tests successful. | 18:34 | |
| \o/ | |||
| nebuchadnezzar: gnu/kfreebsd will be supported soon | 18:36 | ||
| japhb | FROGGS++ # Rapid turnaround for porting to a new arch | 18:37 | |
| FROGGS | only semi new :o) | 18:38 | |
| japhb | Given how long it took Debian to support kfreebsd, I'd say porting to it in a few hours counts as excellent work. :-) | 18:40 | |
| jnthn | Wow, FROGGS++ | 19:09 | |
| FROGGS | jnthn: it is just a tiny patch, really :o) | 19:11 | |
| dalek | arVM/kfreebsd: 47290b4 | FROGGS++ | / (2 files): support GNU/kFreeBSD This does not include a necessary libuv patch that I backported from: anonscm.debian.org/cgit/pkg-javascr..._port.diff I'm not sure yet what to do with that patch... |
19:14 | |
| arVM: 47290b4 | FROGGS++ | / (2 files): support GNU/kFreeBSD This does not include a necessary libuv patch that I backported from: anonscm.debian.org/cgit/pkg-javascr..._port.diff I'm not sure yet what to do with that patch... |
19:17 | ||
| jnthn | I guess since we maintain our own MoarVM/libuv repo it could go in there... | 19:18 | |
| FROGGS | jnthn: in fact we do not maintain our own libuv repo | 19:23 | |
| we use their's | |||
| jnthn | oh :) | ||
| OK, then I've no idea what to do :P | |||
| FROGGS | this is the adjusted patch: gist.github.com/FROGGS/e78f1c16f5e8277fddeb | 19:31 | |
| we could do the packager's approach and apply patches to our deps at configure time :S | |||
| nebuchadnezzar: with latest moarvm patch and the gist I just posted MoarVM will work nicely on GNU/kFreeBSD | 19:36 | ||
| nebuchadnezzar: it is up to you if you drop in these two patches into the debian package for 2015.04 | |||
|
20:32
Ven joined
20:39
colomon joined
|
|||