github.com/moarvm/moarvm | IRC logs at irclog.perlgeek.de/moarvm/today
Set by moderator on 31 May 2013.
colomon apr's configure, or apr's make? 00:00
diakopter make
it doesn't install it to the system
colomon ah, MoarVM's root Makefile works for me if I build apr by hand.
make worked this time! 00:11
make test reports All tests successful. 00:15
make nqptest reports All tests successful 00:20
pmichaud \\o/ 00:24
japhb_ Yay! 00:29
00:33 dukeleto joined
colomon still would be good to figure out why MoarVM/Configure.pl doesn't work for me. 00:42
[Coke] colomon: I basically just copied linux's config, and removed 2 options that caused issues on my machine. 00:53
colomon [Coke]: I meant, doesn't work for me in Linux. OS X seems like even more trouble. 00:58
[Coke] oh. 01:23
someone else reported trouble on ubuntu earlier also.
the moar announcement makes me even more meloncholy about not going this year. bummer. 01:24
diakopter [Coke]: I tried :P
[Coke] diakopter: taking someone else's money to go on vacation isn't really going to help. :) 01:29
diakopter that wasn't it :P you were gonna code awesomely
01:30 FROGGS_ joined 03:42 tokuhirom_ joined 04:23 Tene joined, woosley1 joined 04:31 woosley1 left 04:43 lizmat joined
lizmat you go sightseeing for a day, then this happens! Congratulations jnthn++, diakopter++, japhb++ & masak++ ! 04:46
Tene Just in time for the weekend. :D 04:55
06:33 guset joined 07:53 sa1 joined
sa1 I received a build error: pastie.org/7992338. There was no apr.h, but a apr.h.in 08:00
diakopter sa1: a problem building apr during the moar configure step 08:01
see if you can build apr manually in its dir there 08:02
sa1 diakopter: Yes, I was able to build it there. 08:03
diakopter try making moar now? 08:04
sa1 Yes, it builds.
diakopter what command line did you use to build apr?
sa1 ./configure, make 08:05
diakopter heh. ok
08:24 tgt joined 08:29 sa1 left
Tene Yeah, I've got the same problem. 08:52
The makefile is missing a dependency
Huh. It's linked against my system libapr. 08:56
Got it running and passing tests; time for sleep now, and I'll try to look into the build issues tomorrow. 09:01
09:27 patspam joined
nwc10 yes, native build of parrot and the cross compiler did work out on the Rasberry Pi. But it's faster to ship bytecode across. 09:30
diakopter :)
09:34 tomyan joined
colomon Tene: it's not the makefile, I don't think... the high level MoarVM Configure.pl script doesn't handle APR right on our systems, for some reason. Doesn't report it, either. Seems that if you manually build APR, everything is fine. 11:02
nwc10 would one of the "real" owners like to add MoarVM to ohloh? 11:42
seems to be this page: www.ohloh.net/p/new
12:17 tomyan joined
nwc10 partly I'm curious what Ohloh's analysis makes of which languages are used, in what proportions 12:20
12:52 tomyan_ joined 13:21 vmspb joined
pmichaud I'll add to ohloh 13:24
Added. 13:25
www.ohloh.net/p/moarvm 13:26
lizmat now all we need to do is to add "Written in the Perl Programming Language, Powered by Moar" on all pages on the internet :-) 13:27
to get TIOBI ratings up :-)
14:04 rurban joined
rurban tadzik: aren't you already here in Austin? 14:08
14:16 Guest133_ joined 14:17 lizmat joined
lizmat arrived at the hackathon 14:17
nwc10 pmichaud: thanks 14:39
pmichaud yes, looks like ohloh updated the stats 14:42
took an estimated 50 years of effort (COCOMO model)
starting with its first commit in April, 2012
bwahahahaha
ohhh 14:43
because it's scanning the 3rdparty directory
diakopter heh. 14:44
pmichaud okay, I've set ohloh to ignore "3rdparty"... we'll see how it does.
nwc10 that's somewhat cheating. It also messes up the stats for amount and proportion of code actually needs to be maintained
pmichaud it won't update until the next analysis
nwc10 oh, cool - I didn't know it let you do that
can it ignore the generated Unicode file? 14:45
pmichaud well, it has an "ignore files" option... I don't know if will ignore full directories. I guess we'll find out in a day or two.
which file is the generated Unicode one? 14:46
I can add that also
diakopter strings/unicode_gen.c
nwc10 src/strings/unicode_db.c
diakopter er that
pmichaud okay, added src/strings/unicode_db.c
nwc10 cool, thanks. That should make the stats more sane
diakopter :) 14:47
nwc10: I'm thinking of spending time talking about what generates that 14:48
do you think that would be interssting? 14:49
nwc10 possibly. I suspect I'd be more interested in the other stuff about how a VM design hangs together. But I'm not your typical audience 14:50
and I'm not there
diakopter heh, oh
nwc10: can you he more specific about "hangs together" 14:51
nwc10 yes
1) signals and threads don't really mix. So how does a VM cope with that? 14:52
2) how do events and concurrency hang together
3) how does the GC cope with concurrency
4) how does IO fit in with all of the above
5) portably :-)
diakopter k 14:53
nwc10 in that most of those are "easy" enough if you ignore the others. But it's how to do it all at once
also, am I right in reading the docs that there isn't even an "add" operation. At least, not in the main opcode set
diakopter 1. jnthn says he has some ideas, if I wers doing it, I'd simply study mono 14:54
there is
nwc10 oh, I missed it
diakopter add_i I think
nwc10 aha, it wasn't in anything I read in docs/
diakopter see file oplist
nwc10 OK
ah. 14:55
diakopter yes, the docs :)
definitely deprioritized
actually 14:56
perhaps most interesting to this crowd could be the p5 interop (planned/designed extremely thoroughly) 14:57
nwc10 yes, probably that would be most interesting
more interesting than the hitlist of questions I had 14:58
diakopter though probably the next 2kloc of that should be transferred to somewhere other than my brain first
pmichaud and, afaict, not yet truly revealed yet
(in the sense of not on the intertubes, yet)
nwc10 how many hours until your talk?
pmichaud 48, I think :) 14:59
nwc10 the beginning, not the end. :-(
er, :-)
diakopter namedrop: TimToady and I talked through it a number of evenings and weekends
pmichaud his talk starts in 48 hours, I think.
colomon afk # shutting down computer to pack it
15:08 Guest1337 joined 15:23 tgt joined
diakopter boarding a plane 15:26
ffs. 15:29
guy next to me hasn't bathed in some non-small number of weeks 15:31
it'll be a long 4 hours
nwc10 pmichaud: looks like Ohloh believed you when you asked it to ignore src/strings/unicode_db.c code.ohloh.net/project?pid=BOqaafYz...WOUENJT9Co 15:54
(it's not shown there)
but didn't believe you when you asked it to ignore the 3rdparty directory
15:56 colomon joined 16:16 lizmat joined
pmichaud ohloh might not allow me to ignore full directories 16:41
and I'm not sure I want to list all of the 3rd party files
nwc10 certainly not manually :-)
pmichaud oh! 16:42
checking
found it. 16:52
the instructions for specifying ignored files and directories was "below the fold" on the page, so I didn't see them. 16:53
next time ohloh updates, 3rdparty should not be included.
17:16 tomyan joined
nwc10 yay! 17:30
looks like i thas updated, as it's now saying Analyzed 29 minutes ago based on code collected about 4 hours ago. 17:31
and it's 35% Perl, 62% C
pmichaud estimated as 8 person-years of effort 17:41
might still be some more ignorage we need 17:42
17:42 cognominal joined
moritz jnthn's man year count 20-fold 18:08
18:21 rblackwe joined 18:23 lizmat joined 19:02 cognominal joined 19:13 labster joined 19:59 cognominal joined 20:22 tomyan joined 21:05 japhb_ joined 21:49 cognominal joined, tomyan joined 21:52 tomyan_ joined 21:59 tomyan joined 22:23 cognominal joined 22:37 BrowserUk joined

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!