|
Parrot 3.3.0 released | parrot.org | Log: irclog.perlgeek.de/parrot/today” | Accepted GSoC Students announced! | GSoC student information emails coming out soon Set by moderator on 26 April 2011. |
|||
|
00:15
darbelo left
00:17
darbelo joined
00:19
darbelo left
|
|||
| dukeleto | can someone verify for me that this is a public link ? www.google-melange.com/gsoc/proposa...kpolulak/1 | 00:23 | |
| tcurtis | It is. | ||
| Assuming it is intended to link to the project proposal. | 00:24 | ||
|
00:26
ShaneC joined
|
|||
| kid51 | Well, I wasn't asked to login to view it. | 00:27 | |
|
00:27
kid51 is now known as kid51_at_dinner
00:28
theory left
|
|||
| dukeleto | tcurtis++ | 00:30 | |
| msg rohit_nsit08 what is your github username? | 00:41 | ||
| aloha | OK. I'll deliver the message. | ||
| dukeleto | msg lucian do you have a profile or something on the interwebs that you would like me to link to from a gsoc blog post? | 00:43 | |
| aloha | OK. I'll deliver the message. | ||
|
00:52
Coke left,
Coke joined,
mtk left
|
|||
| dalek | website: soh_cah_toa++ | It's the Most Wonderful Time of the Year! | 00:54 | |
| website: www.parrot.org/content/its-most-won...-time-year | |||
| whiteknight | xmas? | 00:55 | |
| I'm sweating too much for xmas | |||
| soh_cah_toa | yeah man, the heat is killing me | 00:56 | |
|
00:58
mtk joined
00:59
theory joined
|
|||
| cotto | kid51++ | 01:01 | |
| whiteknight | soh_cah_toa: you get any of those crazy thunderstorms today? | ||
| soh_cah_toa | whiteknight: almost. it got close | 01:03 | |
| whiteknight | we got hit by a few of them today. short but powerful | ||
| dukeleto | soh_cah_toa++ # bloggin' | ||
| soh_cah_toa | i never thought i'd see the day that i'd have a blog | 01:04 | |
| i've been meaning to ask this for a while...if parrot is register-based, how are functions called? | 01:13 | ||
| davidfetter | by their registered names | ||
| *rimshot* | |||
| thanks, i'm here all week. try the meatloaf. | 01:14 | ||
| whiteknight | :) | ||
| dukeleto | soh_cah_toa: continuation passing style | ||
| whiteknight | soh_cah_toa: dark magic, mostly | 01:15 | |
| soh_cah_toa: we push values into a CallContext object | |||
| davidfetter hopes he can convince calvin broadus to pen some music involving CP style | |||
| whiteknight | then the callee unpacks the CallContext into registers | ||
| soh_cah_toa | like the registers' state, return address, arguments? | 01:16 | |
| whiteknight | no, just the values | ||
| soh_cah_toa | hmm...well, how is the state before the function preservered? where is the return address stored? | 01:17 | |
| whiteknight | every executing function context has a CallContext object that stores its state | ||
| CallContexts form a linked list of sorts | |||
| soh_cah_toa | okay | ||
| whiteknight | it's like a stack in a C program, but more linked-listy | ||
| also, it's not really a list, because we can have closures and stuff, so it's an acyclic graph | 01:18 | ||
| soh_cah_toa | are the CallContext's stored in the default namespace? | ||
| whiteknight | no | ||
| soh_cah_toa | like, can i actually access this pmc? | ||
| whiteknight | interp->ctx contains a pointer to the current CallContext | ||
| soh_cah_toa | hmm... | 01:19 | |
| whiteknight | actually, there's a macro to use | ||
| CURRENT_CONTEXT(interp), I think | |||
| soh_cah_toa | that's another thing... | ||
|
01:19
kid51_at_dinner is now known as kid51
|
|||
| soh_cah_toa | where is "interp" defined? | 01:20 | |
| whiteknight | it's defined in most functions | ||
| soh_cah_toa | i see you can use it if you have PARROT_INTERP as the first arg | ||
| whiteknight | PARROT_INTERP | ||
| soh_cah_toa | then what's PARROT_INTERP? a macro? | ||
| whiteknight | PARROT_INTERP is #defined as "Interp *interp" | ||
| soh_cah_toa | ah, okay | ||
| whiteknight | I don't know where. You'll have to grep for it in include/parrot/*.h | 01:21 | |
| soh_cah_toa | i did | ||
| what a mess | |||
| whiteknight | :) | 01:22 | |
| dealing with that mess is why they pay you the big bucks | |||
| soh_cah_toa | how would i access CallConext in nqp? inside a Q:PIR {} block? say "Q:PIR { $P0 = new CallContext }"? | 01:23 | |
| oh, not the say function | 01:24 | ||
| whiteknight | if you want to create a new one, that's how you would do it | ||
| if you want the current one, you can get it from the interp | |||
| pir::getinterp()["context"] I think | |||
| you'll have to look up src/pmc/parrotinterpreter.pmc:get_attr_str to find out what it's called | |||
| soh_cah_toa | alright | 01:25 | |
| whiteknight | (it's not that I can't find out for you, I'm trying to help you solve it for yourself :) | ||
| I'm also trying to give you the grand tour of the codebase | 01:26 | ||
| soh_cah_toa | do i have to tip my tour guide? | ||
| coke_ | (grep for) or you could use "vi -t PARROT_INTERP" | 01:28 | |
| kid51 | seen plobsing? | ||
| aloha | plobsing was last seen in #perl6 1 days 9 hours ago joining the channel. | ||
| soh_cah_toa | well i did a grep a while ago. the thing was that i didn't know what type it was so i didn't know in what type of files to search so i could only do a find . -name "*.*" which was horrible | 01:29 | |
|
01:30
davidfetter left
|
|||
| kid51 | Can anyone advise me as to how I would "provide a C backtrace for [the] assertion violations" I'm reporting on parrot-dev re plobsing's tt1931-nci-parameters-deprecation branch? | 01:30 | |
| whiteknight | kid51: do you have gdb? | ||
| kid51 | Yes. | 01:31 | |
| whiteknight | kid51: gdb --args parrot foo.pir | ||
| coke_ | soh_cah_toa: see also ack. | ||
| whiteknight | (whatever your program is called) | ||
| kid51 | AAMOF, I've been learning more about it as part of my C++ study. | ||
| dukeleto .nom.nom.nom() | |||
| whiteknight | kid51: when you get in, do "start", "break Parrot_confess", "c" | 01:32 | |
| kid51: if it asks you any questions, just say "yes" | |||
| kid51 | For 'foo.pir', would I here substitute compilers/opsc/gen/Ops/File.pir? | ||
| whiteknight | kid51: if that's the program that has the assertion failure, yes | ||
| kid51 | (Well, actually, I can't, because that was where it failed) | ||
| whiteknight | basically, it's "gdb --args", followed by the exact commandline you want to debug | 01:33 | |
| soh_cah_toa | coke_: oh...this looks nice. yeah, this will definitely be handy | ||
| whiteknight | soh_cah_toa: yeah, ack is a godsend | ||
| Andy makes it | |||
| kid51 | gdb --args ./parrot-nqp --target=pir --output=compilers/opsc/gen/Ops/File.pir compilers/opsc/src/Ops/File.pm ? | ||
| whiteknight | kid51: looks good to me | ||
| soh_cah_toa | that's awesome | 01:34 | |
| whiteknight | kid51: on a side note, I figured out my --no-line-directives bug. I had a typo inside an alias somewhere that was dropping that argument | ||
| kid51: so, egg on my face | |||
| kid51 | Undefined command: "start". Try "help". | 01:35 | |
| Function "Parrot_confess" not defined. | |||
| Make breakpoint pending on future shared library load? (y or [n]) | |||
| Well, I immediately run into the problem we always have on Darwin: | 01:36 | ||
| dyld: Library not loaded: /usr/local/lib/libparrot.dylib | |||
| Referenced from: /Users/jimk/gitwork/parrot/parrot-nqp | |||
| Reason: image not found | |||
| whiteknight | ...weird | ||
| okay, it makes no sense that the "start" command isn't found. that seems like a pretty fundamental command | 01:37 | ||
| unless mac developers call it something weird | |||
| try "genius" | |||
| :) | |||
|
01:37
Drossel left
|
|||
| kid51 | gdb --version | 01:38 | |
| GNU gdb 6.1-20040303 (Apple version gdb-437) (Sun Dec 25 08:31:29 GMT 2005) | |||
|
01:38
Kulag joined
|
|||
| kid51 | For purpose of using 'prove' or 'perl t/harness', the workaround is to do a non-official installation of Parrot, and then: | 01:39 | |
| export DYLD_LIBRARY_PATH="$HOMEDIR/work/pseudoinstall/lib:$HOMEDIR/gitwork/parrot/blib/lib" | |||
| So perhaps we have to somehow make gdb aware of DYLD_LIBRARY_PATH ? | 01:40 | ||
| Oh, wait a minute ... or an hour.... I currently don't have anything installed. | 01:41 | ||
| Well, this will take me at least an hour, by which time it will be sleep time. | 01:43 | ||
| So I'll do a pseudoinstall of a gcc/g++/g++ build, then reconfigure and build with all g++ until I get to the failure point. | 01:46 | ||
| At which point maybe gdb will start to work. | |||
|
01:47
ShaneC left
01:49
benabik joined
|
|||
| benabik | ~~ | 01:50 | |
|
01:50
woosley joined
01:51
Kulag left
|
|||
| whiteknight | okay, I'm heading to bed | 01:52 | |
| goodnight | |||
| benabik | whiteknight: 'night | ||
|
01:52
whiteknight left
|
|||
| cotto | 'night | 01:52 | |
|
01:52
Kulag joined
|
|||
| benabik | bacek: ping | 01:55 | |
|
01:56
bubaflub left
|
|||
| kid51 | right about now, that's probably "bacek_at_work ping" ;-) | 01:56 | |
| PerlJam | benabik: how fortuitous ... I volunteered to be bacek's backup mentor for gsoc (he doesn't know yet either). And ... I have no idea where you're at or what you've done so far :) | 01:58 | |
| benabik | kid51: Don't bother me with details. | ||
| PerlJam: I've blogged, forked parrot on github, and just now tried to get a hold of bacek. | 01:59 | ||
| PerlJam | benabik: where's your blog? | ||
| benabik | parrot.org/blog/1240 | ||
| PerlJam | danke | ||
| benabik: just checking ... when you say NQP, you really mean nqp-rx, right? (not the new and improved nqp that jnthn is working on) | 02:02 | ||
| benabik | PerlJam: Most specifically I really mean parrot-nqp, but jnthn__ said there should be no syntax differences. | ||
| PerlJam | benabik: so ... what did you need from bacek? (I don't know if I can help, but I'll try.) | 02:11 | |
| benabik | PerlJam: Mostly just trying to sync up and see how badly mis-matched our schedules will be. I'm supposed to be talking to somebody once a week. :-) | 02:13 | |
| PerlJam | What time zone are you in? | 02:14 | |
| benabik | EDT (-4) | ||
| PerlJam | I'm CDT (well, EDT at the moment since I'm in Florida, but I'll be back to CDT tomorrow :) | 02:16 | |
| benabik | Which means I'm more likely to be able to get ahold of you than bacek. :-/ I'm still going to try to figure out when I can talk to him though. | 02:19 | |
|
02:20
ybit left
02:21
Caelum left
|
|||
| soh_cah_toa | msg cotto what hours are you at work? i'd like to just go over some things about gsoc w/ you but i know sometimes you're work | 02:22 | |
| aloha | OK. I'll deliver the message. | ||
| cotto | ~~ | 02:23 | |
| soh_cah_toa, usually 8-10 to 5-6:30 | |||
| it's flexible | |||
| PerlJam | benabik: are you working full time while getting your MS? | 02:24 | |
| soh_cah_toa | cotto: okay, that works | 02:25 | |
| benabik | PerlJam: At the moment all I have is a contracting gig a few hours a week. | ||
|
02:25
Caelum joined
|
|||
| soh_cah_toa | cotto: how about friday night? or do you usually go out then? | 02:26 | |
| cotto | soh_cah_toa, this Friday I'll most likely be hanging out with dukeleto at lfnw | 02:27 | |
| soh_cah_toa | cotto: oh yeah. forgot about that | ||
| cotto: when do you get back? | 02:28 | ||
| cotto | Sunday evening | ||
|
02:28
ybit joined
|
|||
| soh_cah_toa | cotto: that'll work | 02:29 | |
| PerlJam | benabik: so ... I haven't used NQP (of any flavor) in a while and I'm going to have to come up to speed on the nqp_pct branch and such. (just so you know) | 02:32 | |
| benabik | PerlJam: Fair enough. I'm not really going to get much work done prior to May 20th, so there's time for everyone to get up to speed. :-) | 02:33 | |
|
02:48
kid51 left
|
|||
| atrodo | Ya, I have a laundry list in progress to fix/improve for isparrotfastyet, just need the time to do it | 03:05 | |
|
03:22
soh_cah_toa left
|
|||
| dukeleto | ~~ (for a bit) | 03:24 | |
|
03:35
hudnix left
|
|||
| dalek | rrot/leto/embed_grant: 504e86f | dukeleto++ | t/src/extend_vtable.t: [t] Parrot_PMC_set_pointer_keyed_int using a HashIteratorKey |
03:41 | |
|
04:23
plobsing_ left
|
|||
| dukeleto | Check it out, I have parrot.git doing the Jitterbug again: new.leto.net:3000/project/parrot | 04:36 | |
| dalek | rrot/tt2103_cmp_pmc: e71819b | dukeleto++ | t/src/extend_vtable.t: [t][TT#2103] Parrot_PMC_cmp_pmc This test gives this error: even though PP->P seems to be the correct signature. |
04:45 | |
| dukeleto | interesting. the actual error in the commit message got removed | 04:48 | |
| dalek | TT #2103 created by dukeleto++: cmp_pmc not working in extend_vtable | 04:57 | |
| TT #2103: trac.parrot.org/parrot/ticket/2103 | |||
| rrot: 2888490 | dukeleto++ | config/auto/ipv6/test.in: [TT #2101] Attempt to fix ipv6 detection on Gentoo, tadzik++ |
04:58 | ||
| dukeleto | wow, 16 consecutive leading integers in a SHA1! That is exceedingly rare. | 04:59 | |
| dalek | rrot: e5452a6 | dukeleto++ | t/tools/parrot_debugger.t: [TT #1848] Remove useless parrot_debugger tests A change in the parrot_debugger has rendered all these tests useless, and we will be getting a shiny, well-tested new debugger from soh_cah_toa++ soon, so these tests go to the bitbucket. |
05:04 | |
| TT #1848 closed by dukeleto++: delete placeholder t/tools/parrot_debugger.t | 05:13 | ||
| TT #1848: trac.parrot.org/parrot/ticket/1848 | |||
|
05:15
jrtayloriv joined
|
|||
| cotto | I'm comfortable in upgrading my talk to "probably not a trainwreck". | 05:23 | |
|
05:23
kuku joined
|
|||
| cotto | a couple more practice runs and it'll up to "boo-yah" | 05:26 | |
|
05:26
kuku is now known as birdwindupbird
|
|||
| dalek | TT #2100 closed by dukeleto++: Automate the release process | 05:28 | |
| TT #2100: trac.parrot.org/parrot/ticket/2100 | 05:29 | ||
| cotto | closed? | ||
| ah | 05:30 | ||
|
05:33
birdwindupbird left
05:35
ShaneC joined
05:36
theory left
05:37
kuku joined
05:59
fperrad joined
|
|||
| dukeleto | leto.net/dukeleto.pl/2011/04/parrot...bbles.html | 06:12 | |
| cotto: well, then it is about 5 levels above mine, currently | 06:13 | ||
| cotto | dukeleto, have you spoken much before? | 06:14 | |
|
06:15
he joined
06:39
UltraDM joined
06:47
rhebus joined
06:49
mj41 joined
07:01
rhebus_ joined
07:03
rhebus left,
rhebus_ is now known as rhebus
07:05
kuku left
07:13
rhebus_ joined
07:14
rhebus left,
rhebus_ is now known as rhebus
07:23
kuku joined
07:25
rhebus_ joined
|
|||
| dukeleto | cotto: yeah, I guess so, at various confs. | 07:25 | |
| cotto: me go bed now. | |||
| cotto: see ya on the flip side | |||
|
07:26
rhebus left,
rhebus_ is now known as rhebus
|
|||
| cotto | 'night | 07:27 | |
|
07:29
UltraDM left
07:38
rhebus_ joined
07:40
rhebus left,
rhebus_ is now known as rhebus
07:49
jrtayloriv left
07:55
ShaneC left
08:04
contingencyplan left
08:22
rhebus left
08:40
mtk left
08:46
mtk joined
08:48
krunen left
09:36
Kulag left
09:42
Kulag joined
10:08
ShaneC joined,
ShaneC left
10:11
woosley left,
jsut joined
|
|||
| dalek | p/ctmo: b3299e5 | moritz++ | build/Makefile.in: make nqp installable again Sadly this overwrites some files from parrots nqp-rx, which means that it breaks recompilation, because .ops file translation uses nqp-rx internally |
10:12 | |
| Patterner | "./include/parrot/config.h:73:44: error: conflicting types for 'Parrot_PackFile'" | 10:13 | |
|
10:16
jsut_ left
10:18
krunen joined
|
|||
| dalek | p/ctmo: 86fd707 | moritz++ | build/Makefile.in: [build] unbreak parallel build |
10:18 | |
|
10:21
dip joined
|
|||
| dalek | p/ctmo: f73152c | moritz++ | build/Makefile.in: [build] stop installing nqprx.pbc |
10:29 | |
|
10:29
UltraDM joined
|
|||
| bacek | benabik, pong | 10:32 | |
| benabik, I'm on biztrip to Singapore now. GMT+8 I think. | 10:33 | ||
|
10:41
lucian joined
|
|||
| bacek | msg benabik | 10:57 | |
| msg benabik I think you can just continue nqp_pct branch in original parrot repo (as soon as you'll have your commit bit). I'll try to attend next #ps and ask for commit bit for you. If I will not able to do it - just ask any metacommitter to give it to you. | 10:59 | ||
| aloha | OK. I'll deliver the message. | ||
| bacek | msg benabik It will simplify code review on my side. Just because I'm too lazy to set up new repo :) | 11:00 | |
| aloha | OK. I'll deliver the message. | ||
| bacek | msg benabik For any "urgent" queries I always available via email on bacek@bacek.com. Or xmpp:bacek@jabber.ru. Messages will arrive to my phone and I can answer them quickly (if I'm not totally drunk or burned in meetings at $dayjob :) | 11:03 | |
| aloha | OK. I'll deliver the message. | ||
| bacek | afk # dinner time | 11:04 | |
|
11:10
fperrad left
11:11
fperrad joined
11:20
kuku left
|
|||
| PerlJam | msg bacek Did you see that I volunteered to be your backup mentor for benabik? | 11:20 | |
| aloha | OK. I'll deliver the message. | ||
|
11:26
fperrad left
11:27
fperrad joined
11:31
kuku joined
11:37
Coke left,
kuku is now known as birdwindupbird,
Coke joined
11:42
lucian left
11:57
Patterner left,
Psyche^ joined,
Psyche^ is now known as Patterner
12:11
bubaflub joined
12:33
plobsing joined
12:44
whiteknight joined
|
|||
| whiteknight | good morning, #parrot | 12:45 | |
|
12:45
lucian joined
|
|||
| bubaflub | morning whiteknight | 12:46 | |
| whiteknight | hello bubaflub. how are you today? | 12:47 | |
| I'm going to try to update my VM to Ubuntu 11.04 today | |||
| bubaflub | whiteknight: not too bad. still stressed over finals but surviving. | ||
| whiteknight | so that promises to be a lot of fun | ||
| bubaflub: what classes are you taking this year? | |||
| bubaflub | whiteknight: what VM are you using? | ||
| whiteknight: i'm in seminary, so the list might appear weird | 12:48 | ||
| whiteknight | I'm using VirtualBox here. I've got Ubuntu 10.10 on there now and want to upgrade | ||
| bubaflub | whiteknight: History, Old Testament Prophets, Ministry, Theology and Contemporary Issues | ||
| whiteknight | I have absolutely no idea what kinds of classes are taught at seminary, so I'm even more interested to hear the list | ||
| bubaflub | whiteknight: yeah, i really like VirtualBox though at work we use VMWare | 12:49 | |
|
12:50
lucian_ joined
|
|||
| bubaflub | whiteknight: VMWare is nice in that I can pull up specific applications | 12:50 | |
| whiteknight | bubaflub: are you planning to be a priest or something like that when you graduate? | ||
|
12:50
hudnix joined
|
|||
| whiteknight | I really don't know a whole heck of a lot about how seminary works | 12:51 | |
| bubaflub | whiteknight: not exactly; "priest" usually refers to Catholic, Greek Orthodox, Anglican, Episcopalian, and probably some others. over here in Protestant land they are usually called pastors | 12:52 | |
| whiteknight | so you're planning to be a pastor? :) | 12:53 | |
| bubaflub | whiteknight: hahaha, sorry my typing is slow today. | ||
| whiteknight: maybe - the degree i am going to get (M. Div.) is a 3-year non-thesis degree | |||
| whiteknight: it's a very broad, non-specific degree that a lot of pastors will wind up getting | 12:54 | ||
|
12:54
lucian__ joined,
ambs joined,
lucian left,
bluescreen joined
|
|||
| bubaflub | whiteknight: covers Greek and Hebrew, History, Theology, Biblical Studies, Practical Ministry (like counseling and preaching) | 12:54 | |
| whiteknight | oh, okay. cool. So you just do software on the side? | 12:55 | |
| bubaflub | whiteknight: and a bunch of other stuff from the abstract like Theology and Culture to the very nitty gritty like how to handle a meeting | ||
| whiteknight: yep. my undergrad was Math and Computer Science and it funds my other activities | |||
| whiteknight | that's all very interesting. Good luck with it | ||
| bubaflub | whiteknight: i do perl programming for Grant Street Group right now; i've bounced around in webdev stuff | 12:56 | |
| whiteknight: thanks | |||
|
12:57
lucian_ left,
lucian joined
13:01
lucian__ left
|
|||
| whiteknight | blah, the upgrade to 11.04 is going to take 3 more hours | 13:03 | |
| but I want my daily dose of fancy *now* | |||
|
13:04
bluescreen left
|
|||
| atrodo | 3 hours? Good gravy, are they including a movie in the install? A DoD wipe? Harddrive encryption? | 13:14 | |
| moritz | atrodo: it's easy: install all packages related to tex :-) | ||
| whiteknight | it's down to less than 2 hours now. I suspect the network here is T3H SUX | ||
| moritz | atrodo: that would be > 200 packages, each of which needs to update some directory listing => lots of IO that doesn't cache very well | 13:15 | |
| atrodo | moritz++ # I could never understand the appeal of TeX | ||
| bubaflub | atrodo: i'm a big fan of LaTeX, if that counts... | 13:17 | |
| atrodo | bubaflub> I have no problems with it, just could never really understand it or understand the appeal of it. It's a me issue | 13:18 | |
| Which is also kind of strange since I spent 4 years in the print industry | |||
|
13:18
bluescreen joined
|
|||
| moritz | atrodo: without tex, I would have committed suizide many times over while trying to subdue some word processor | 13:19 | |
| bubaflub | atrodo: heh. yeah, also on Mac there are convenient bundles and pre-built binaries. there is also a build called XeTeX that handles unicode / multi-language stuff well. | ||
| moritz | atrodo: if you write a lot of text and mathematical formulas, there's nothing like it | ||
| whiteknight | I just don't do a heck of a lot of wordprocessing, but I rarely find an application that Markdown, POD, or some flavor of wikitext can't handle | 13:20 | |
| I've opened my copy of OpenOffice Write exactly once since I installed it: to update my resume which was stored as a .doc | |||
|
13:20
plobsing left
|
|||
| moritz | try to write a 500 pages report with lots of images in Word | 13:20 | |
| whiteknight | try to avoid writing 500 page reports with lots of images | ||
| problem solved | |||
| atrodo | moritz> Haven't done a lot of math formulas, I do understand it's second to none there | ||
| moritz | and then curse every programmer in the world when it says "Can't open file: too big" | 13:21 | |
| whiteknight | atrodo: yes. Wikipedia uses LaTeX for it's math typesetting engine, and I always appreciated it there | ||
| it uses a very limited subset of TeX, I think | 13:22 | ||
| atrodo | moritz> I also didn't know TeX existed until college | ||
| moritz | atrodo: that happens to a lot of people | 13:23 | |
| lucian | in my uni, .docx files are the norm apparently | 13:26 | |
| whiteknight | One day, when I write my first great american romance novella, maybe I'll want to use TeX | ||
| probably not | |||
| lucian | dukeleto: like a webpage? | 13:27 | |
| jnthn__ is writing course material for $dayjob in TeX today. :) | |||
| whiteknight | course material? you a teacher or professor or something? | 13:29 | |
| jnthn__ | whiteknight: My $dayjob is partly teaching, yes :) | 13:30 | |
| Util is working to fix broken Trac report {6}, as requested by dukeleto; ETA 30 minutes. | 13:31 | ||
| whiteknight | jnthn__: grad student? | ||
| jnthn__ | whiteknight: No, it's outside of academic world, for businesses. | 13:33 | |
|
13:33
darbelo joined
|
|||
| jnthn__ | whiteknight: I teach stuff like TDD, software architecture, etc. Currently putting together a web security course. | 13:33 | |
| whiteknight | oh, okay | ||
|
13:34
birdwindupbird left,
birdwindupbird joined
|
|||
| moritz | jnthn__: point them to autologout.org/all :-) | 13:34 | |
| lucian | moritz: pretty cool | 13:35 | |
| moritz | lucian: thanks | 13:36 | |
| next step: a submission form, and review of submitted links | |||
| lucian | did you make it? awesome | ||
| there's a typo in <title>, though | |||
| jnthn__ | moritz: oh, NICE! | ||
| moritz | lucian: me and a fellow perl programmer | 13:37 | |
| which typo? | |||
| moritz seems blind today | |||
| lucian | moritz: says AutLogin | 13:38 | |
| moritz | ah, thanks | ||
| jnthn__ | gah, that cost me my GitHub and Google logins :P | 13:41 | |
| moritz | jnthn__: I was quite surprised that those two were affected | 13:42 | |
| same for slashdot | |||
|
13:45
woosley joined
|
|||
| lucian | aloha: tell dukeleto if it's a website you need, you can put in my "blog", honeyweb.wordpress.com | 13:45 | |
|
13:45
darbelo left,
darbelo joined
13:46
darbelo_ joined,
darbelo left
13:53
lucian_ joined
13:57
lucian left
|
|||
| Util has finished fixing Trac report {6} | 13:57 | ||
|
13:59
darbelo joined
|
|||
| moritz | Util: and 4 minutes before your ETA. | 14:01 | |
|
14:03
darbelo_ left,
plobsing joined
14:04
darbelo left
14:05
he left
14:17
bluescreen left,
jsut left,
mtk left,
benabik left,
cosimo left,
mikehh left,
PacoLinux left,
coke_ left,
utsl left,
kthakore left,
preflex left,
marc left,
jnthn__ left,
NotFound left,
TimToady left,
pmichaud left,
PerlJam left,
tadzik left,
ingy left,
hudnix left,
whiteknight left,
mj41 left,
wagle left,
jjore left,
cotto left,
TonyC left,
perlite left,
jevin left,
tewk left,
AzureStone left,
Maddingue left,
ascent left,
pjcj left,
arnsholt left,
KaeseEs left,
slavorg left,
tcurtis left,
frodwith left,
Util left,
ttbot left,
knewt left,
woosley left,
ambs left,
Coke left,
dip left,
krunen left,
Caelum left,
particle1 left,
szbalint left,
sorear left,
dalek left,
sirmacik left,
eternaleye left,
sri left,
allison left,
cotto_work left,
moritz left
14:25
woosley joined,
bluescreen joined,
ambs joined,
hudnix joined,
whiteknight joined,
Coke joined,
dip joined,
krunen joined,
jsut joined,
mtk joined,
mj41 joined,
Caelum joined,
benabik joined,
cosimo joined,
particle1 joined,
wagle joined,
mikehh joined,
jjore joined,
cotto joined,
TonyC joined,
PacoLinux joined,
coke_ joined,
perlite joined,
jevin joined,
utsl joined,
tewk joined,
kthakore joined,
AzureStone joined,
preflex joined,
Maddingue joined,
szbalint joined,
ascent joined,
pjcj joined,
arnsholt joined,
KaeseEs joined,
slavorg joined,
tcurtis joined,
sorear joined,
marc joined,
jnthn__ joined,
dalek joined,
frodwith joined,
NotFound joined,
sirmacik joined,
TimToady joined,
eternaleye joined,
pmichaud joined,
PerlJam joined,
tadzik joined,
Util joined,
cotto_work joined,
sri joined,
allison joined,
ingy joined,
ttbot joined,
knewt joined,
moritz joined
14:35
lucian joined
14:36
bluescreen left
14:39
lucian_ left
|
|||
| whiteknight | blah, my VM is at 98% disk usage during the upgrade | 14:46 | |
| I'm going to have to purge a lot of crap, methinks | |||
|
14:46
bluescreen joined
|
|||
| whiteknight | maybe I'll just try to grow the disk | 14:47 | |
| no, can't automatically grow disks in virtualbox | 14:49 | ||
|
15:01
lucian_ joined
15:05
Andy_ joined
15:06
lucian left
15:12
woosley left
15:13
NotFound left,
Andy_ left,
PacoLinux left,
PacoLinux joined
15:14
NotFound joined
15:15
Andy_ joined
15:30
UltraDM left
15:32
davidfetter joined
|
|||
| whiteknight | why is config telling me that we are missing t/tools/parrot_debugger.t | 15:34 | |
| ? | |||
| coke_ | didn't dukeleto just remove it? | 15:35 | |
|
15:35
bluescreen left
|
|||
| whiteknight | oh, did he? | 15:37 | |
| then he didn't remove it enough | |||
|
15:42
theory joined
15:49
bluescreen joined
|
|||
| whiteknight | On ubuntu 11.04, it appears that asm/errno.h and asm/socket.h among others are not there | 15:53 | |
| instead, it looks like they've been moved to asm-generic/errno.h and asm-generic/socket.h | 15:54 | ||
| moritz | whiteknight: maybe you need to install some -dev packages first? | ||
| hm | |||
| whiteknight | I just upgraded from 10.10 | ||
| so at the least it should still have the packages that I had installed previously | |||
| moritz | "should" | ||
| whiteknight | I fixed the problem with a symlink and parrot is building now | 15:55 | |
| I don't know if that's a "bug" in ubuntu, or if we should update Parrot's configure system accordingly | |||
|
15:55
contingencyplan joined
|
|||
| whiteknight | I'm finding plenty of other bugs in the new ubuntu, so I'm inclined towards the former | 15:55 | |
| the fact that it sponaneously resets my desktop theme to the hideously ugly Gnome defaults is quite annoying. It's doing that randomly it seems | 16:01 | ||
| moritz | there have been lots of blog posts about how unreliable the ubuntu beta was | 16:02 | |
| whiteknight | yeah, I tried the beta, and it was just as bad as this | ||
|
16:02
rhebus joined
|
|||
| whiteknight | that's kind of disappointing | 16:03 | |
|
16:03
rohit_nsit08 joined
|
|||
| rohit_nsit08 | hello #parrot | 16:03 | |
| whiteknight | and sockaddr.t is segfaulting on the system now. It passed just fine when I ran the test this morning before the upgrade | ||
| coke_ | rohit_nsit08: hio. | ||
| whiteknight | good morning rohit_nsit08 | ||
| rohit_nsit08 | coke_: hello | 16:04 | |
| coke_ | rohit_nsit08: Are you on mumbai time? | ||
| atrodo | whiteknight> I'm glad they decided that the beta was good enough | ||
| rohit_nsit08 | coke_: yes, It us UTC+5:30 | ||
| coke_ | currently 21:34, yes? | ||
| rohit_nsit08 | coke_: yup | 16:05 | |
| coke_: this will be my usual time, fine? | |||
| coke_ | danyawaad. | ||
| rohit_nsit08 | coke_: vow, u know hindi? | 16:06 | |
| coke_ | Sure. it's lunch time here, which makes it amenable to talking at work. | ||
| rohit_nsit08: I know 3 or 4 words, 2 of which i cannot say in a public channel. | |||
|
16:06
rhebus_ joined
|
|||
| rohit_nsit08 | coke_: ya, okay that is fine :-) | 16:06 | |
| coke_ | (current company has a lot of Indians on staff, both locally and remotely.) | 16:07 | |
| So how goes? This is my first time as a primary mentor. Typically I'm the backup and never have to do anything. ;) | |||
| rohit_nsit08 | coke_: which company do u work for? | 16:08 | |
| coke_ | General Electric | ||
| rohit_nsit08 | coke_: great, ya i wrote yesterday a post on parrot regarding my so far status | ||
| coke_ | www.parrot.org/content/hello-parrot, ja? | ||
|
16:08
rhebus left,
rhebus_ is now known as rhebus
|
|||
| rohit_nsit08 | coke_: I was reading about PCT to get more acquaint with parrot | 16:09 | |
| coke_ | the "important points" is good. | ||
| rohit_nsit08 | ya that one | ||
| coke_ | (PCT) aye. I've done compilers on parrot both ways (raw pir, vs. PCT that generates pir), and doing it with PCT lets you avoid writing a lot of code. | 16:10 | |
| rohit_nsit08 | ya but that will require npq and perl i guess, and the javascript compiler will be written mostly in javascript, I'm currently experimenting on "cafe" on my system. | 16:11 | |
| coke_ | I have found that the best docs on that were (sadly) other people's code. I have a bunch of code using PCT classes from NQP (in partcl) | ||
| You don't need nqp to use PCT. it does give you a nicer front end, though. | 16:12 | ||
| whiteknight | It's not even possible to use PCT in stage 1. You need to already have the compiler running on Parrot for that | ||
| coke_ | whiteknight: you could use nqp in stage 1 and eventually rip it out. | 16:13 | |
| Generating raw PIR is fine. winxed and partcl-old both do that. | |||
| whiteknight | true. He can generate any kind of code he wants in the backend of stage 1 | ||
| coke_ | so, if you've got a comfort level with raw PIR, that's fine. | ||
| rohit_nsit08 | whiteknight: ya, I was just reading pct for fun, I have gone through the cafe's code generator and will start hacking on it now. | 16:14 | |
| coke_ | rohit_nsit08: what is "T representation" ? | ||
| whiteknight | it's a graphical way to represent what a compiler does and how it is written itself, though I'm fuzzy on those details myself | 16:15 | |
| coke_ | (looking at rohitnsit08.blogspot.com/2011/04/bo...iler.html) | ||
| also, you have a broken image in that post | |||
|
16:16
theory left
|
|||
| coke_ | but so far the plan looks good. | 16:16 | |
| rohit_nsit08 | coke_: T representation is simply to represent compiler , it represents source language, target language, and the implementation language in an easy to understand format. I read some research paper where they were used and helped me to grab the concept better | ||
| let me check for the broken image | |||
| coke_ | ok. in the image on the post, which parts of the T are which? | 16:17 | |
| (left, right, bottom) = (source, target, implementation) ? | |||
| rohit_nsit08 | coke_: yup | ||
| the bottom triangle represents the machine on which this compiler will run | |||
| coke_ | k. | 16:18 | |
| rohit_nsit08 | without that it is just a program only | ||
| didn't found any broken image, is it opening now? | |||
|
16:19
mj41 left
|
|||
| coke_ | "the This step can be visualized by this diagram. | 16:21 | |
| s/the// | |||
| 3.bp.blogspot.com/-8S3LOkheb5s/Tacg...1%2529.jpg | |||
| also: | 16:22 | ||
| 3.bp.blogspot.com/-_A0W5Qtfamw/Tach...1%2529.jpg | |||
| rohit_nsit08: are you blocking on anything from me at the moment? | |||
| my email is will@coleda.com - feel free to email me (or google chat) any questions there, if I'm not around on irc. | 16:23 | ||
| rohit_nsit08 | coke_: i guess not. | ||
| coke_: ya sure | |||
| thanks | |||
| i'll check the privacy settings if some problem is there | 16:24 | ||
| coke_ | ah, that could be, yah. | ||
| (could just be me.) | |||
| I have to go grab some lunch. :) | 16:25 | ||
|
16:26
varta left
|
|||
| rohit_nsit08 | coke_: okay, I just had dinner :-) | 16:26 | |
|
16:29
varta joined
16:34
birdwindupbird left
16:35
rohit_nsit08 left
|
|||
| dalek | rrot/tt1931-nci-parameters-deprecation: 398f171 | plobsing++ | src/nci/extra_thunks.nci: comment-ify garbage at EOL |
16:37 | |
| rrot/tt1931-nci-parameters-deprecation: 962e551 | plobsing++ | tools/dev/nci_thunk_gen.pir: rewrite thunk generator to work with new, more descriptive signatures * maintains backwards-compat with old-style sigs * works directly with sig FIAs * uses PCRE |
|||
| rrot/tt1931-nci-parameters-deprecation: 55bfc77 | plobsing++ | src/nci/extra_thunks.c: reboostrap thunks |
|||
| dukeleto | ~~ | 16:47 | |
|
16:47
rohit_nsit08 joined
16:58
rohit_nsit08 left
|
|||
| whiteknight | plobsing++ is kicking some ass in that branch | 17:00 | |
|
17:02
JimmyZ joined
17:11
rohit_nsit08 joined
|
|||
| dalek | p/ctmo: 4042390 | moritz++ | build/gen_parrot.pl: [Configure] git fetch in the right place |
17:13 | |
|
17:14
plobsing left
17:17
JimmyZ left
17:18
ShaneC joined
17:20
rhebus left
|
|||
| dukeleto | looks like Ubuntu 11.04 is out. Anybody tried it? | 17:31 | |
|
17:35
bluescreen left
|
|||
| coke_ | dukeleto: wonder if it fixes my power issues. :P | 17:35 | |
| coke_ wanders off to get his laptop out of the car. | 17:36 | ||
| moritz | dukeleto: see the backlog | ||
|
17:36
ShaneC left
|
|||
| bubaflub | dukeleto: i think whiteknight was installing/upgrading it on his VM | 17:38 | |
| dukeleto backlogs | |||
| bubaflub: how goes classes and such? | |||
| bubaflub | dukeleto: not bad! in the home stretch | 17:39 | |
| dukeleto: lots of time spent at the library this week; but knocked out almost all my reading. all i have left are final papers (and a few exams) | |||
| seen plobsing? | 17:41 | ||
| aloha | plobsing was last seen in #perl6 3 hours 37 mins ago joining the channel. | ||
| bubaflub | msg plobsing does the ncidef2pir script need to be updated after the changes in your branch for TT1931 land? | ||
| aloha | OK. I'll deliver the message. | ||
|
17:46
bluescreen joined
|
|||
| dalek | rrot: 97c32c7 | dukeleto++ | MANIFEST: Remove parrot_debugger.t from the MANIFEST |
17:49 | |
| dukeleto | msg plobsing got a build failure on your recent commit new.leto.net:3000/api/build/parrot/...rl-v5.10.1 | 17:54 | |
| aloha | OK. I'll deliver the message. | ||
|
17:56
theory joined
|
|||
| dukeleto | new.leto.net:3000/project/parrot is already being very useful | 17:57 | |
|
17:58
Kulag left,
Kulag joined
|
|||
| bubaflub | dukeleto: is that your CI? | 18:02 | |
| dukeleto | bubaflub: yeps. it is Jitterbug : github.com/leto/jitterbug/tree/reuse_repo | 18:08 | |
|
18:08
davidfetter left
|
|||
| dukeleto | bubaflub: i got an instance for work at jitterbug.sgn.cornell.edu:8090/ | 18:09 | |
|
18:09
davidfetter joined
|
|||
| dukeleto | bubaflub: there is plenty to improve, but it is useful | 18:10 | |
| bubaflub | dukeleto: very nice. i've played a bit with Integrity before. but i like the idea of a Perl CI system | ||
| dukeleto | bubaflub: it is written in Perl, but knows how to run many kinds of tests | 18:12 | |
| bubaflub: it supports Build.PL, Makefile.PL, Makefile, Rakefile and maybe something else, now | 18:13 | ||
| bubaflub: I want to add support for running any kind of test suite | |||
| bubaflub: and i just added the ability to write a custom test script, if it can't autodetect | |||
| bubaflub | dukeleto: very nice. | ||
| dukeleto | bubaflub: oh yeah, it supports Configure.pl and setup.pir for Parrot and Perl 6 projects | 18:14 | |
| of course ;) | |||
| bubaflub | dukeleto: awesome. i'll be using disutils and some Rosella goodness for the GSoC project | 18:16 | |
|
18:17
lucian_ is now known as lucian
|
|||
| cotto_work | ~~ | 18:20 | |
|
18:23
mtk left
18:25
Kulag left,
Kulag joined
|
|||
| cotto_work | bubaflub: what happened with Parrot on RTEMS? | 18:25 | |
| bubaflub | cotto_work: i have a hints file, got stuck on the boot-strappy bits | 18:26 | |
| cotto_work: when i get time i'd like to revisit it | |||
| cotto_work | I'd like that too. | ||
| Was it you who was working on separate build dirs? | |||
| bubaflub | cotto_work: what i think would have to happen is you build a system parrot first, build up to miniparrot with the cross-compiler / RTEMS, steal some of the bootstrapped bits and finish to build libparrot for the target system | 18:27 | |
| cotto_work: yes; it was a major pain. i have a good amount of work on it but it was a bit inelegant; a lot of FindBin qw($bin) to figure out where i was | |||
| cotto_work: also, a lot of stuff is actually in a BEGIN {} block which make debugging *real* hard - i.e. it would error out before the debugger could even start | 18:28 | ||
|
18:28
mtk joined
|
|||
| cotto_work | bubaflub: do you have a branch for that? | 18:32 | |
| bubaflub | cotto_work: github.com/bubaflub/gsoc-rtems-par...ints-maker | 18:33 | |
| some of the diffs in there should do it | |||
| dukeleto | bubaflub: i think compiling parrot under qemu to cross-compile is the easier way to skin that cat | ||
| bubaflub: that is what the emacs bitbake recipe does to compile on arbitrary embedded machines | 18:34 | ||
| bubaflub | dukeleto: good to know | ||
| dukeleto | cotto_work: Parrot on RTEMS is a huge undertaking | ||
| bubaflub | dukeleto: anyway we can get a libparrot compiled on the target platform i'm game for | ||
| dukeleto | cotto_work: i don't think it will happen unless somebody funds it. | ||
| cotto_work | dukeleto: sure. There are a lot of prerequisites we don't have in place. | ||
| bubaflub | dukeleto and cotto_work: also because getting RTEMS up and running is a bit hairy itself; lots of automake and what not | ||
| cotto_work: i do think getting an out of directory build is definitely a first step | 18:35 | ||
|
18:35
dukeleto left,
dukeleto joined
|
|||
| dukeleto | blarg | 18:35 | |
|
18:44
jsut_ joined
18:50
jsut left
19:02
mj41 joined
19:06
Coke left,
Coke joined
19:15
rohit_nsit08 left
19:17
Coke left,
Coke joined
19:25
lucian left
19:26
rohit_nsit08 joined
19:29
Coke left,
Coke joined
19:41
theory left
19:45
Coke left,
Coke joined
19:50
plobsing joined
|
|||
| dalek | rrot/tt1931-nci-parameters-deprecation: 96c86cc | plobsing++ | tools/dev/nci_thunk_gen.pir: rearrange thunk gen to handle pass-by-ref |
19:51 | |
| rrot/tt1931-nci-parameters-deprecation: 4f81c7f | plobsing++ | config/gen/makefiles/root.in: new dependancies of thunk gen |
|||
| rrot/tt1931-nci-parameters-deprecation: 83bcbea | plobsing++ | src/nci/extra_thunks.nci: add new thunk for nci_thunk_gen and remaining PCRE thunks |
|||
| rrot/tt1931-nci-parameters-deprecation: edc12c7 | plobsing++ | src/nci/extra_thunks.c: rebootstrap thunks |
|||
| rrot/tt1931-nci-parameters-deprecation: 0bb3627 | plobsing++ | tools/dev/nci_thunk_gen.pir: [codingstd] trailing whitespace |
|||
| whiteknight | plobsing: so is pcre a requirement now to bootstrap ops? | 19:53 | |
| plobsing | no. it is required to build new static thunks and to bootstrap thunks. | ||
| bubaflub | plobsing: what's the best way to marshal C-strings in and out through NCI after these deprecations? | 19:55 | |
| plobsing | bubaflub ncidef2pir does not require updates to keep doing what it does. if we want to expose the new, shiny thunk descriptions, it will need to change. | ||
| whiteknight | er, thunks. that's what I meant | 19:56 | |
| plobsing | bubaflub: the same way as before - do it yourself and manage the object's lifetime appropriately | ||
| bubaflub | plobsing: okey dokey. | 19:57 | |
| plobsing | bubaflub: check out the PCRE bindings. it handles strings in this way. | ||
| bubaflub | plobsing: great, thanks. | ||
| plobsing | the PG bindings do too, and use automatic wrappers (to deal with the large amount of NCIs using strings) | 19:58 | |
|
19:58
bubaflub left
|
|||
| plobsing | whiteknight: I don't see the pcre dependency as particularly onerous and the operation is done rather infrequently. | 20:01 | |
| dukeleto | plobsing: are you seeing the failures at new.leto.net:3000/project/parrot ? | ||
| whiteknight | plobsing: yeah, I didn't say it was a problem. I don't have pcre on my machine so I guess I will have to isntall it | ||
| plobsing | whiteknight: how do you not have pcre installed? | 20:02 | |
| my machine has a zillion dependancies on it - glib2, grep, less, zsh, ... | 20:03 | ||
|
20:03
ttbot left
|
|||
| plobsing | dukeleto: hmmm... I was expecting the extra thunks to help out there | 20:03 | |
| whiteknight | I might have it. I certainly never manually installed it | 20:05 | |
| I've obviously never put much thought into it | |||
| dukeleto | many things bundle their own pcre, such as php | 20:06 | |
| dalek | rrot/tt1931-nci-parameters-deprecation: 0d8ab04 | plobsing++ | config/gen/makefiles/root.in: forgot to add pcre dep |
20:08 | |
|
20:12
bubaflub joined,
theory joined
20:15
perlite_ joined,
perlite left,
perlite_ is now known as perlite
20:19
Coke left,
Coke joined
|
|||
| dalek | rrot/tt1931-nci-parameters-deprecation: f2b7f12 | plobsing++ | config/gen/makefiles/root.in: remove extra_nci_thunks library this never got past the demo stage as interest never took off |
20:20 | |
|
20:33
particle1 left
20:34
whiteknight left
20:47
dmalcolm joined
20:52
particle joined
21:10
bluescreen left
21:11
rohit_nsit08 left
21:30
ambs left
21:32
bubaflub left
|
|||
| Andy_ | YOU GOT RID OF EXTRA_NCI_THUNKS? | 21:33 | |
|
21:33
donaldh joined
|
|||
| Andy_ | BLESS YOU | 21:33 | |
| dukeleto | And then there were fewer hunks, and the masses were pleased. | 21:39 | |
| Patterner | do you care about the masses? | 21:43 | |
|
21:49
ttbot joined
21:52
fperrad left
21:54
bubaflub joined
22:00
mj41 left
22:05
donaldh left
22:09
autark joined
22:10
PerlJam left,
PerlJam joined
22:17
darbelo joined
|
|||
| dukeleto forgot the difference between a hunk and a thunk | 22:19 | ||
|
22:19
bacek left
|
|||
| darbelo | dukeleto: One has a t at the start. | 22:19 | |
| ;) | |||
| dukeleto | darbelo: inteet it toes | 22:20 | |
| cotto_work: linuxfestnorthwest.org/forum/bof-se...ds-feather | 22:21 | ||
| cotto_work: parrot bof ? | |||
|
22:25
plobsing left
22:31
plobsing joined
|
|||
| dukeleto | msg cotto linuxfestnorthwest.org/forum/bof-se...ckathonbof | 22:39 | |
| aloha | OK. I'll deliver the message. | ||
| cotto_work | ~~ | ||
|
22:39
darbelo left
|
|||
| cotto_work | dukeleto: nice | 22:39 | |
|
22:39
darbelo joined
|
|||
| cotto_work | At some point I'd like to get a new logo for Parrot. I like that you mentioned that graphic artists are welcome too. | 22:41 | |
|
22:41
darbelo_ joined
|
|||
| dalek | tracwiki: v32 | dukeleto++ | NewParrotDeveloperGuide | 22:41 | |
| tracwiki: trac.parrot.org/parrot/wiki/NewParr...ction=diff | |||
| theory | tadzik: ping | 22:43 | |
| dukeleto | theory: tadzik is on europe time | 22:44 | |
| theory | the fuck you say | ||
| dukeleto | theory: poland, methinks | ||
| theory | Seems like half the time I talk to .eu folks on my time, they reply anyway. | ||
| I think half the people I know there don't sleep. | |||
| dukeleto: Thanks. | |||
| dukeleto | theory: you can message him | 22:45 | |
| msg tadzik theory++ is lookin' for ya | |||
| aloha | OK. I'll deliver the message. | ||
| theory | dukeleto: Yeah | ||
| oh, me tends to use /mst | |||
| /msg | |||
| dukeleto | theory: sure. but aloha will tell him you were looking for him, when he joins irc again. SOME people don't have persistent irc connections. | 22:46 | |
| probably healthier | |||
| theory | dukeleto: Yeah, I'm one of them. | ||
|
22:46
darbelo left
23:01
elmex left
23:03
darbelo_ left
23:04
darbelo joined
23:07
darbelo_ joined
23:08
bacek joined
23:09
darbelo__ joined,
darbelo left
23:12
darbelo joined,
darbelo__ left
23:14
darbelo_ left
23:15
dip left
23:19
whiteknight joined,
AzureSto_ joined
23:22
AzureStone left
23:23
bubaflub left
|
|||
| whiteknight | good evening, #parrot | 23:24 | |
| cotto_work | hi whiteknight | 23:27 | |
| whiteknight | hello cotto_work | ||
|
23:29
AzureSto_ left
23:32
AzureStone joined
23:34
AzureSto_ joined
|
|||
| whiteknight | I'm really starting to get excited about concurrency, but I figure we havea few other things to take care of first | 23:38 | |
| I don't want to start on it until at least we do some Lorito work. Context/Interp refactors are especially important in that regard | |||
| If we can start getting a plan together I would like to start putting in some deprecations by 3.6 | 23:39 | ||
|
23:39
darbelo left,
AzureStone left
23:48
davidfetter left,
theory left
23:51
bacek left
|
|||
| whiteknight | We really need to focus some effort on packfiles to help out Rakudo. Then I really want to focus on 6model | 23:52 | |
| after that, if Lorito has moved far enough forward, concurrency | |||
| I think we can get some good packfile work done, and 6model moved into Parrot before 4.0. We can probably get concurrency started by then, but not merged | 23:53 | ||
| jnthn__ | I need to look in more depth at the profile, but 6model uses the fixed size allocator in Parrot for pretty much all of its object body allocations now. The profile I glanced a few days ago showed stuff related to the fixed size allocator being a hot path now. | 23:58 | |
| In fact, I think fill_params, find_lex and the fixed size allocator were the three "hottest" things that showed up in the profile. | 23:59 | ||
