Camelia, the Perl 6 bug

IRC log for #moarvm, 2013-06-03

| Channels | #moarvm index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:35 BrowserUk joined #moarvm
00:36 BrowserUk left #moarvm
01:01 woosley joined #moarvm
01:15 woosley left #moarvm
01:15 woosley joined #moarvm
02:36 FROGGS joined #moarvm
03:37 woosley joined #moarvm
04:17 labster joined #moarvm
05:20 birdwindupbird joined #moarvm
06:07 lizmat joined #moarvm
07:18 Guest1337 joined #moarvm
08:00 Guest1337 joined #moarvm
08:42 Guest1337 joined #moarvm
10:55 Guest1337 joined #moarvm
12:48 Ulti joined #moarvm
12:48 Ulti ohai
12:49 moritz it's Ulti!
13:03 lizmat joined #moarvm
13:20 Ulti I wanted moar perl6 scrolling text
13:25 cognominal joined #moarvm
13:54 colomon joined #moarvm
13:58 cognominal joined #moarvm
14:16 nwc10 jnthn: are you testing MoarVM on 32 or 64 bit Windows?
14:17 jnthn 64 on desktop, I think my laptop here is 32
14:17 nwc10 OK. does t/qast/qast_arguments.t fail on your laptop?
14:17 jnthn Notably 'cus I remember fixing 32-bit bugs before...
14:17 jnthn um, currently rebuilding loads of stuff...will tell you in a moment :)
14:17 nwc10 OK.
14:17 nwc10 I would expect tests 10 and 12 to fail
14:17 nwc10 as they fail on 32 bit linux
14:18 nwc10 also, not sure if it's helpful, but if anyone is keep to help and doesn't know what to do, run MoarVM under valgrind and plug the malloc leaks
14:19 JimmyZ joined #moarvm
14:19 jnthn yes, that'd be good :)
14:20 JimmyZ C:\Parrot\bin\nqp.exe nqp-moar-cc.nqp --setting=NULL --target=mbc --output=nqp-mo.moarvm nqp-src\nqp-mo.pm
14:20 JimmyZ No registered operation handler for 'setstaticlex'
14:20 JimmyZ I got this
14:20 jnthn Then the nqp you build with is too old
14:21 JimmyZ I git pull last nqp master
14:22 JimmyZ yes, It tell me my nqp is too old, but I don't know where is the newer
14:23 colomon joined #moarvm
14:26 JimmyZ Am I missing something?
14:26 [Coke] where is the nqp you pulled?
14:27 JimmyZ perl6/nqp
14:27 JimmyZ [Coke]: Did you see my message to you?
14:27 [Coke] I ended up having to pull a fresh nqp into moarVM/nqp-cc/nqp , and then run (in nqp-cc) perl Configure.pl --gen-nqp --gen-parrot
14:28 [Coke] Yes, thanks.
14:29 JimmyZ well, I just git pull nqp and make install to parrot/bin
14:30 [Coke] ok. I never bothered to get moarvm working with the globally installed nqp, just the locally (to nqp-cc) installed nqp.
14:30 [Coke] also moarvm requires a branch.
14:30 [Coke] (of nqp. not master)
14:30 nwc10 not quite NQP enough for the job...
14:30 [Coke] ETOOMANYPERLS
14:36 * JimmyZ still deos not know why his nqp is old
14:47 [Coke] JimmyZ: wrong branch
14:48 [Coke] if you have a pre-built version off master, it's not going to work
14:48 [Coke] I would recommend going with --gen-nqp --gen-parrot for now, esp. since moarvm will eventually not need them.
15:05 JimmyZ [Coke]: thanks, I think I should switch to rak-jvm-support.
15:07 [Coke] aye. if you do the -gen steps, it will handle that for you automatically, and not interfere with any other things using your installed latest-master version.
15:21 bronco_creek joined #moarvm
15:24 colomon joined #moarvm
15:34 JimmyZ yay, got it build eventually, thanks [Coke]
15:40 lizmat joined #moarvm
15:40 colomon [Coke]++
15:45 labster joined #moarvm
15:49 mj41 joined #moarvm
15:57 FROGGS joined #moarvm
15:58 JimmyZ joined #moarvm
15:59 JimmyZ >nqp t\moar\math.t
15:59 JimmyZ Malformed UTF-8 string
15:59 JimmyZ current instr.: 'slurp' pc 657 (src\stage2\gen\NQPCORE.setting.pir:302) (src\stage2\NQPCORE.setting:647)
15:59 JimmyZ run any test I get 'Malformed UTF-8 string'
16:34 labster joined #moarvm
16:37 tgt joined #moarvm
16:39 jnthn nwc10: Only 12 fails here
16:42 nwc10 odd, and a bit surprising
16:42 nwc10 but it might be that the 562949953421337 I get is because of something else with a value 2 being nearby
16:42 nwc10 and gcc makes it nearby, but MSVC doesn't
16:44 jnthn *nod*
16:52 labster joined #moarvm
17:00 jnthn Will take a look into it later todya
17:00 jnthn *today
17:05 timotimo is there any plans to make moarvm embeddable in custom programs?
17:06 jnthn That's some way down my list.
17:06 jnthn If it happens soon, it'll be because somebody else decides they really want it :)
17:07 timotimo just wanted to make sure it's in the back of people's minds, because that would be a nice thing to have
17:42 tomyan joined #moarvm
18:04 timotimo ty
18:04 timotimo (typed that into an assumed-dead ssh session)
18:07 timotimo btw: repeating questions is something that's very, very helpful to livestream viewers
18:07 timotimo (i managed to get the dvr thing going on windows)
18:09 tomyan joined #moarvm
18:50 colomon joined #moarvm
18:50 FROGGS joined #moarvm
18:52 Guest1337 joined #moarvm
18:55 lizmat joined #moarvm
18:58 cognominal joined #moarvm
19:05 labster joined #moarvm
19:12 nevernotlurk joined #moarvm
19:16 rurban joined #moarvm
19:32 jnthn nwc10: 32-bit qast_arguments.t issue you noticed should now be fixed; dalek didn't report the commit for some reason.
19:34 jnthn um, 'cus I pushed to the wrong place...
19:35 nwc10 there's a private repository?
19:36 nwc10 and has been since 2012/04/07?
19:38 jnthn right
19:39 dalek MoarVM: 73625d1 | jonathan++ | src/core/args.c:
19:39 dalek MoarVM: Fix thinkos in flatten_args.
19:39 dalek MoarVM:
19:39 dalek MoarVM: Showed up on 32-bit platforms as corruption/data loss. Reported by
19:39 dalek MoarVM: nwc10++.
19:39 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/73625d12e7
19:39 moritz that's where they develop EvenMoarVM
19:39 jnthn There we go.
19:40 nwc10 works on x86
19:41 nwc10 well, that test works
19:41 jnthn \o/
19:41 jnthn Any more fails in 32-bit?
19:41 nwc10 there will be a delay because I just did git clean -dxf
19:41 nwc10 and i'm trying a clean build
19:41 jnthn k
19:41 nwc10 that was the only one, IIRC
19:46 nwc10 works on ARM
19:46 FROGGS nwc10: what? moarvm works on arm?
19:48 nwc10 why does everyone seem to be surprised by this
19:48 nwc10 I'd expect it to work on all 11 or so architectures that Debian build on
19:48 jnthn I'd hope so
19:48 nwc10 on the assmuption that APR already works there
19:48 tomyan joined #moarvm
19:48 nwc10 (annoyingly APR doesn't seem to do VMS)
19:49 FROGGS nwc10: I was surprised, yes
19:49 FROGGS cant tell why
19:52 Guest1337 joined #moarvm
19:56 FROGGS joined #moarvm
19:56 * jnthn is about to watch a MoarVM talk :)
19:57 FROGGS :o)
19:59 masak ++diakopter!
20:07 donaldh joined #moarvm
20:12 nwc10 jnthn: All tests successful.
20:12 jnthn \o/
20:12 nwc10 (in green, but I think that Freenode won't let me do that)
20:14 masak it's the thought that counts.
20:14 * masak .oO( colorless greed ideas IRC furiously )
20:14 masak green*, g'ah
20:17 leont joined #moarvm
20:36 chipdude joined #moarvm
20:38 sorear joined #moarvm
20:45 prammer joined #moarvm
20:46 eternaleye nwc10: Freenode does permit color, the ircbots in other channels use that.
20:47 rurban joined #moarvm
20:47 eternaleye err, commitbots
20:47 TimToady colorless green threads sleep independently
20:47 TimToady or nefariously
20:49 sorear +c is either strip color or strip CTCPs
20:56 Guest133_ joined #moarvm
21:03 eternaleye Mm, it may well be color. The channel I see it used in doesn't have that flag.
21:03 eternaleye Maybe ask the freenode folks?
21:04 eternaleye (#exherbo's commitbot uses it, as does #paludis')
21:11 colomon joined #moarvm
21:12 timotimo i think you want +C
21:13 lizmat joined #moarvm
21:16 FROGGS joined #moarvm
21:18 lizmat joined #moarvm
21:26 eternaleye timotimo: Well, the channels I see it used in are 'Lns' and 'nst'
21:26 eternaleye (rather, their modes are)
21:26 timotimo oh, hm
21:32 Guest1337 joined #moarvm
22:04 colomon joined #moarvm
22:05 tomyan joined #moarvm
22:06 lizmat joined #moarvm
22:13 rurban joined #moarvm
22:21 Guest1337 joined #moarvm
22:24 lizmat joined #moarvm
23:32 colomon joined #moarvm
23:49 FROGGS joined #moarvm

| Channels | #moarvm index | Today | | Search | Google Search | Plain-Text | summary