Da
Show enters and exits. Hide enters and exits.
| 00:04:31 | rubuildius_ppc | Dirkjan Bussink: 26ba94f12; bin/ci failed! http://pastie.caboo.se/paste/162588 |
| 00:04:52 | dbussink | brixen: there it is, the failure you wanted ;) |
| 00:04:55 | zenspider | argh!!! can we ban assignment in conditionals? |
| 00:05:22 | brixen | dbussink: ugh, goddamn it |
| 00:05:39 | brixen | dbussink: have to run to meet gf for a soc event :/ |
| 00:05:55 | dbussink | hehe, that happens :P |
| 00:06:05 | dbussink | i need to get some sleep, so now really bye |
| 00:06:23 | brixen | dbussink: later |
| 00:06:56 | rue | zenspider: Legalize it! |
| 00:16:23 | boyscout | 1 commit by Dirkjan Bussink |
| 00:16:24 | boyscout | * On BSD systems, getaddrinfo needs explicit NULL pointers; 85e87ed |
| 00:16:50 | dbussink | ok, that's really it for today :) |
| 00:16:53 | dbussink | night! |
| 00:17:12 | boyscout | 1 commit by Caleb Tennis |
| 00:17:12 | boyscout | * Add rb_define_global_function to subtend, with tests; d2d3750 |
| 00:20:42 | wmoxam leaves the room. | |
| 00:21:07 | panpot leaves the room. | |
| 00:22:52 | vintrepid leaves the room. | |
| 00:23:31 | lopex leaves the room. | |
| 00:26:03 | dewd_ leaves the room. | |
| 00:26:24 | loincloth_ leaves the room. | |
| 00:28:35 | twbray__ leaves the room. | |
| 00:29:22 | rubuildius_ppc | Caleb Tennis: d2d3750c4; bin/ci failed! http://pastie.caboo.se/paste/162595 |
| 00:29:55 | jtoy enters the room. | |
| 00:30:58 | zenspider | anyone here familiar with the stringio specs? is putc("AA") supposed to be something special? |
| 00:31:29 | alloy leaves the room. | |
| 00:31:33 | headius | hey zenspider, I almost have a builder done...there wasn't much more code needed than what I showed you |
| 00:31:50 | headius | we already have all the type-mapping logic in our java integration layer |
| 00:32:13 | headius | I just need to wire it in so it's managing output files the same way and it should be pretty good |
| 00:32:23 | headius | it's crapping them out in cwd right now |
| 00:32:45 | headius | pastie |
| 00:32:47 | therealadam leaves the room. | |
| 00:32:56 | pastie | http://pastie.org/162596 by headius. |
| 00:35:00 | headius | actually all that's really needed is line 40 down |
| 00:41:14 | crafterm enters the room. | |
| 00:50:58 | jero5 leaves the room. | |
| 00:50:59 | TheVoice leaves the room. | |
| 00:51:02 | radarek enters the room. | |
| 00:52:21 | TheVoice enters the room. | |
| 00:53:39 | MenTaLguY | ugh, I really hate Monitor::ConditionVariable#count_waiters |
| 00:54:31 | headius | why's that? |
| 00:54:51 | MenTaLguY | ideally, it should be protected by the monitor's mutex |
| 00:54:59 | MenTaLguY | but that seriously uglifies the design no matter how I slice it |
| 00:56:49 | TheVoic1 enters the room. | |
| 00:59:24 | MenTaLguY | well, found a solution that isn't as icky as it could have been |
| 01:02:03 | twbray enters the room. | |
| 01:02:22 | zenspider | argh |
| 01:02:36 | zenspider | fatal: unable to create '/Users/ryan/Work/git/rubinius/.git/index.lock': File exists ??? |
| 01:03:29 | TheVoice leaves the room. | |
| 01:03:47 | halorgium | lsof -nP |grep rubinius ? |
| 01:05:12 | TheVoice enters the room. | |
| 01:05:41 | octopod leaves the room. | |
| 01:06:52 | jayWHY enters the room. | |
| 01:09:39 | zenspider | hrm... anyone else seeing a Module#autoload error? |
| 01:10:02 | zenspider | it isn't in the failure above |
| 01:10:10 | mernen enters the room. | |
| 01:10:14 | zenspider | and I'm playing nowhere near it |
| 01:10:21 | zenspider | screw it... pushing |
| 01:11:32 | boyscout | 7 commits by Ryan Davis |
| 01:11:34 | boyscout | * excluded; cd0b896 |
| 01:11:35 | boyscout | * More specs to test out const scoping with eval; e40f2bb |
| 01:11:35 | boyscout | * Further clarified StringIO#getc specs.; 3926add |
| 01:11:36 | boyscout | * Added changes to make autotest work with lib files... prolly doesn't map for everythi ...; f05f58f |
| 01:11:37 | boyscout | * Removed bad test for stringio, we cover a lot more; 3bd10f9 |
| 01:11:38 | boyscout | ... |
| 01:12:40 | headius | zenspider: http://pastie.org/162611 |
| 01:12:46 | headius | that's basically the whole thing |
| 01:13:12 | headius | I haven't started monkeying with your tests to adapt them for this, since there's a lot of extra tests this won't need |
| 01:13:22 | zenspider | no tests? |
| 01:13:25 | headius | I'd also like the batch build/packaging working |
| 01:13:43 | cored enters the room. | |
| 01:13:57 | zenspider | would our packaging even work for jruby? I have no clue |
| 01:14:11 | headius | well, it would be a lot simpler |
| 01:14:21 | headius | I haven't looked at what all yours does right now |
| 01:15:11 | zenspider | a lot simpler? but you haven't looked at ours? |
| 01:15:18 | headius | I've skimmed it |
| 01:17:17 | evan | zenspider: i'm getting some random CI failures about bad file descriptors |
| 01:17:32 | headius | evan: lots of people getting those the past couple days apparently |
| 01:18:04 | zenspider | yeah. I saw that in the failure above... I don't think that's related to the autoload failure I just saw (tho... it might be. I have no clue) |
| 01:18:31 | rue | I think dbussink mentioned it yesterday |
| 01:18:42 | evan | i'll see if I can track it |
| 01:18:44 | rue | So probably not zenspider breaking stuff :) |
| 01:18:54 | evan | it looks a bit like someone is doing something bad with a descriptor |
| 01:19:04 | evan | closing something they don't control |
| 01:19:05 | zenspider | rue: THIS time. :P |
| 01:19:18 | rue | evan: How has your trip been? Caught the jet lag after all? I am getting worse at fighting it off as I age |
| 01:19:35 | rue | zenspider: Heh, all progress requires sacrifice |
| 01:19:37 | zenspider | oh. eric was talking about that wrt minitar |
| 01:20:26 | zenspider | rue: as you AGE?!?! how old are you again? (granted, the only photo I've seen of you was like 50x50) |
| 01:21:41 | rue | 130 |
| 01:21:50 | rue | It is pretty young for an elf though |
| 01:21:58 | evan | rue: jet lag has actually been no problem |
| 01:22:23 | evan | a little more tired in the evenings, but thats it. |
| 01:22:32 | rue | evan: Very cool! At least it is almost a 24-h difference which may help, I suppose |
| 01:22:37 | evan | last night I was totally over it (the alcohol helped, we were out until 4am) |
| 01:22:42 | rue | Heh |
| 01:22:46 | crafterm | always helps :) |
| 01:22:47 | zenspider | rue: :P |
| 01:22:50 | dysinger leaves the room. | |
| 01:23:33 | zenspider | how long do the build bots take these days? |
| 01:24:54 | rubuildius_ppc | Ryan Davis: cd0b89694; 1736 files, 5862 examples, 19916 expectations, 0 failures, 1 error; http://pastie.caboo.se/paste/162619 |
| 01:25:28 | evan | just long enough. |
| 01:25:36 | agardiner | 4am! wow... i feel like such a wuss, leaving before midnight! :-D |
| 01:25:44 | zenspider | huh... that was the failure I saw on my end |
| 01:25:54 | zenspider | but I didn't commit anything near that (that I know of) |
| 01:26:23 | zenspider | I added some failing specs to eval, and fixed StringIO#getc |
| 01:26:30 | zenspider | anyone have a clue on that one? |
| 01:27:33 | evan | interesting.... |
| 01:27:34 | evan | http://java.dzone.com/news/farewell-j-jvm |
| 01:27:35 | enebo leaves the room. | |
| 01:27:47 | evan | zenspider: i'm taking a look now. |
| 01:27:54 | evan | getting the EBADF crap now. |
| 01:29:06 | zenspider | I didn't do it! :P |
| 01:29:18 | evan | :D |
| 01:30:34 | evan | hah |
| 01:32:37 | evan | zenspider: oh btw, i'm not aware of any issues installing rubinius anymore |
| 01:32:46 | evan | thats to everyone actually |
| 01:36:05 | evan | zenspider: actually, the autoload failure is your fault |
| 01:36:07 | twbray leaves the room. | |
| 01:36:16 | evan | because your eval specs made a constant named A |
| 01:36:27 | evan | which overrides a constant named A that the autoload specs use |
| 01:37:49 | evan | thats why all constants need to be scoped in fixtures |
| 01:39:42 | agardiner | evan: this iseq clamping problem is causing spec failures on 64-bit... |
| 01:40:27 | agardiner | am i right in understanding that a bytearray doesn't store a size, but calculates it from the number of "fields" it has? |
| 01:40:44 | agardiner | in other words, it will always be a multiple of sizeof(OBJECT)? |
| 01:42:09 | evan | yep |
| 01:42:19 | agardiner | hmm... ok |
| 01:42:38 | agardiner | i need to work around the fact that an iseq may have noops at the end then |
| 01:43:00 | evan | these autoload specs leave something to be desired. |
| 01:43:01 | KirinDav enters the room. | |
| 01:43:27 | evan | like, why does it use global variables? |
| 01:45:22 | evan | oh, i see. |
| 01:47:23 | crossblaim_ leaves the room. | |
| 01:49:31 | agardiner | hmmm... |
| 01:49:56 | agardiner | evan: is there a special reason why iseq_new takes the number of fields to use, rather than a number of bytes? |
| 01:50:39 | agardiner | 'cause it means i need to know the size of the platform's OBJECT |
| 01:51:05 | rue | Not necessarily scoped in fixtures but scoped |
| 01:51:21 | rue | I could do without having to open a fixture file separately myself |
| 01:52:15 | vintrepid enters the room. | |
| 01:56:15 | zenspider | evan: my spec is excluded |
| 01:56:24 | zenspider | it shouldn't be loading it at all |
| 01:56:37 | zenspider | oh. wait. no... n/m. I'm a tard. the file is still loaded |
| 01:57:09 | zenspider | and given that my spec is about eval not working for constants... putting in fixtures seemed brokey |
| 01:57:15 | zenspider | sorry |
| 01:59:43 | rubymaverick enters the room. | |
| 02:01:58 | KirinDav leaves the room. | |
| 02:13:34 | nicksieger leaves the room. | |
| 02:13:41 | chop3 enters the room. | |
| 02:15:07 | nicksieger enters the room. | |
| 02:21:52 | TheVoice leaves the room. | |
| 02:24:14 | jinjing enters the room. | |
| 02:24:43 | ezmobius leaves the room. | |
| 02:25:18 | rue | Blah, apache is being a pain |
| 02:26:29 | jayWHY leaves the room. | |
| 02:26:50 | TheVoice enters the room. | |
| 02:27:44 | cored leaves the room. | |
| 02:29:02 | obiejuan leaves the room. | |
| 02:29:57 | jtoy leaves the room. | |
| 02:32:31 | imajes enters the room. | |
| 02:32:39 | jtoy enters the room. | |
| 02:34:58 | twbray enters the room. | |
| 02:40:55 | imajes leaves the room. | |
| 02:43:13 | boyscout | 1 commit by Adam Gardiner |
| 02:43:14 | boyscout | * InstructionSequence#decode now strips trailing noops; 8f3084f |
| 02:45:32 | VVSiz_ enters the room. | |
| 02:52:48 | _mutle enters the room. | |
| 02:54:48 | rubuildius_ppc | Adam Gardiner: 8f3084fb1; 1736 files, 5862 examples, 19915 expectations, 0 failures, 1 error; http://pastie.caboo.se/paste/162646 |
| 03:02:56 | cavalle_ enters the room. | |
| 03:03:13 | hoopy leaves the room. | |
| 03:03:39 | VVSiz leaves the room. | |
| 03:09:00 | mutle leaves the room. | |
| 03:12:16 | srbaker_ enters the room. | |
| 03:12:26 | srbaker_ leaves the room. | |
| 03:12:29 | srbaker leaves the room. | |
| 03:12:43 | srbaker enters the room. | |
| 03:19:42 | cavalle leaves the room. | |
| 03:20:11 | jrun leaves the room. | |
| 03:21:37 | tarcieri | hmmm, it'd be a fun exercise to port Tasks/Channels to MRI |
| 03:22:48 | tarcieri | on 1.8 you could use Threads and on 1.9 you could use Fibers |
| 03:23:05 | MenTaLguY | hm, I think you would need to use Continuations, actually |
| 03:23:15 | MenTaLguY | if you want to capture the full semantics of tasks |
| 03:23:37 | tarcieri | hmmm |
| 03:23:53 | radarek leaves the room. | |
| 03:23:55 | GMFlash leaves the room. | |
| 03:24:15 | tarcieri | I was thinking it could be done with Thread.stop |
| 03:24:21 | MenTaLguY | Tasks don't really map cleanly to anything in MRI though, I'm not sure even reified continuations are general enough (since you can't take continuations across threads) |
| 03:26:18 | agardiner | hmmm... i know that's something that can be done with tasks in Rubinius... |
| 03:26:29 | agardiner | is there anywhere this is used? |
| 03:26:33 | tarcieri | Does Omnibus allow for preemption between Actors? |
| 03:28:03 | MenTaLguY | sure, the current implementation binds them to threads |
| 03:28:21 | MenTaLguY | eventually I'd like to get back to it to work on pluggable schedulers for everything though |
| 03:28:34 | tarcieri | Yeah |
| 03:29:16 | tarcieri | So what semantics of Threads/Channels wouldn't be preserved on MRI if you had all Task threads stopped, then the scheduler starts one and joins it, then waits until that one calls Thread.stop, then schedules the next one? |
| 03:29:19 | tarcieri | Err |
| 03:29:22 | tarcieri | semantics of Tasks/Channels |
| 03:30:08 | MenTaLguY | tasks can be migrated across threads, is the big thing |
| 03:30:30 | tarcieri | Hmmm, how does that work? |
| 03:32:28 | agardiner | Tasks hold the reference to the operand stack and current method context |
| 03:32:32 | agardiner | not threads |
| 03:33:12 | agardiner | so a task can be associated with any thread, which is what actually executes the instructions |
| 03:34:17 | tarcieri | yeah, there'd certainly be massive incompatibilities with them being overlying primitives instead of underlying ones |
| 03:34:26 | geekounet leaves the room. | |
| 03:37:08 | b3b0p leaves the room. | |
| 03:37:11 | agardiner | i can see the value of channels on MRI... not so sure about tasks... |
| 03:37:20 | b3b0p enters the room. | |
| 03:37:31 | agardiner | what would you use them for? something like fibers? |
| 03:37:40 | tarcieri | yeah |
| 03:38:04 | tarcieri | coroutines that can communicate/cooperatively multitask with message passing |
| 03:38:32 | agardiner | hmm... hard to make that fit when it isn't supported from the available primitives |
| 03:38:44 | agardiner | would be nice, though :-) |
| 03:39:06 | agardiner | can you use tasks like that on Rubinius currently? |
| 03:39:18 | agardiner | i've been meaning to ask evan about that... |
| 03:39:31 | tarcieri | well, see |
| 03:39:38 | agardiner | we don't have a way to yield at present, do we? |
| 03:39:43 | tarcieri | my main motivation for this is Revactor TCP |
| 03:39:51 | tarcieri | well, you yield by receiving a message, correct? |
| 03:40:23 | agardiner | hmmm... yeah, but that is the thread that is suspended, not just the task |
| 03:40:23 | tarcieri | I need some lower level communication channel than an Actor mailbox for things like Actor::TCP |
| 03:40:46 | tarcieri | for implementing synchronous "blocking" calls |
| 03:41:02 | agardiner | i don't think we (currently) have a way for a task to hand off to another task on the same thread... but i could be wrong?! |
| 03:41:29 | tarcieri | well, in this case, it'd be implemented in a higher level primitive (Actors) |
| 03:42:39 | headius | evening |
| 03:43:26 | agardiner | ok. Rubinius Actors are currently implemented using threads though, no? |
| 03:43:35 | agardiner | headius: evening |
| 03:43:41 | tarcieri | I was under the impression they were implemented on top of Tasks/Channels directly |
| 03:43:45 | tarcieri | but I may be mistaken |
| 03:44:11 | agardiner | i know they're implemented on channels... not sure about the task/thread bit... perhaps i should look! :-) |
| 03:44:50 | agardiner | yeah, its Thread that's used |
| 03:44:56 | tarcieri | aah |
| 03:45:31 | agardiner | Tasks are funny creatures right now... they don't seem to have a lot of utility when removed from threads |
| 03:45:49 | agardiner | (or as distinct from Thread) |
| 03:45:57 | tarcieri | well, seems like I need to better understand Tasks |
| 03:46:27 | agardiner | hehe... me too! |
| 03:46:40 | GMFlash enters the room. | |
| 03:47:21 | pth_ enters the room. | |
| 03:47:34 | agardiner | from what i've seen, threads in rubinius provide you with something that can be scheduled and used to store Thread locals |
| 03:48:01 | agardiner | pretty much everything else is on Task, often with convenience accessors from Thread to its current Task |
| 04:01:26 | cavalle enters the room. | |
| 04:02:29 | pth_ leaves the room. | |
| 04:03:54 | pth leaves the room. | |
| 04:06:49 | eventualbuddha leaves the room. | |
| 04:16:55 | cavalle_ leaves the room. | |
| 04:19:19 | hoopy enters the room. | |
| 04:23:25 | RyanTM leaves the room. | |
| 04:23:51 | RyanTM enters the room. | |
| 04:27:25 | srbaker leaves the room. | |
| 04:30:52 | wycats leaves the room. | |
| 04:34:46 | wifelette leaves the room. | |
| 04:35:06 | crafterm leaves the room. | |
| 04:37:31 | crafterm enters the room. | |
| 04:49:20 | lstoll_ enters the room. | |
| 04:57:07 | rue | agardiner: There is no value to channels in MRI :) |
| 04:58:03 | agardiner | well, i agree it would be a bit of a band-aid...! |
| 04:58:10 | rue | tarcieri: I think I will probably have a scratch that could use properly communicating coroutines |
| 04:58:22 | rue | agardiner: I mean for US there is no value in that ;) |
| 04:58:35 | agardiner | hehe... too true! |
| 04:58:43 | rue | tarcieri: So we shall see if I get anywhere with that |
| 04:59:05 | agardiner | rue: i'll second coroutines |
| 04:59:17 | agardiner | seems like it shouldn't be too hard to add to tasks |
| 05:02:19 | dysinger enters the room. | |
| 05:02:32 | evan_s leaves the room. | |
| 05:02:43 | rue | Hm, I thought we already have pre-emptive coroutines |
| 05:03:51 | agardiner | well, we have a way for a thread to associate a task, but no simple way for a task to yield to another on the same thread |
| 05:05:55 | rue | Interesting, I need to look at it again |
| 05:06:54 | digitalfallout enters the room. | |
| 05:07:29 | lstoll leaves the room. | |
| 05:10:27 | evan | digitalfallout: oh! another evan! |
| 05:10:59 | agardiner | ohoh... could be trouble! ;-) |
| 05:11:24 | agardiner | is the digitalfallout *because* there is another evan? :-P |
| 05:12:08 | lstoll_ leaves the room. | |
| 05:12:29 | tarcieri | hmm, what I think I'll do for my stuff is just add something with Channel-like semantics... I think MenTaLguY may have already done something like that for Omnibus |
| 05:13:35 | tarcieri | I mean, it wouldn't be that hard to implement the Fiber API on top of Tasks, would it? |
| 05:13:58 | srbaker enters the room. | |
| 05:14:58 | agardiner | probably not... |
| 05:16:38 | evan | tarcieri: it's already 1-to-1 |
| 05:16:53 | evan | could mostly be |
| 05:16:55 | evan | class Fiber < Task |
| 05:18:36 | agardiner | most of the functionality would be implemented on Task, though wouldn't it? |
| 05:18:44 | agardiner | otherwise the VM would need to get involved... |
| 05:18:49 | evan | yeah |
| 05:18:53 | agardiner | and no another class |
| 05:18:59 | agardiner | s/no/know/ |
| 05:18:59 | evan | yep. |
| 05:21:03 | lstoll enters the room. | |
| 05:25:04 | MenTaLguY | I don't think inheritance would really be appropriate for fibers-based-on-tasks |
| 05:25:27 | evan | oh? |
| 05:25:31 | MenTaLguY | in terms of their features, fibers aren't a proper subtype of task |
| 05:25:49 | MenTaLguY | you don't have liskov substitutability |
| 05:26:07 | evan | hm, ok. |
| 05:26:38 | brixen | google and wikipedia, kill combo :) |
| 05:26:43 | brixen | killer* |
| 05:27:20 | brixen | ohh, this is in the FOOL book |
| 05:27:26 | wmoxam enters the room. | |
| 05:29:43 | agardiner | could tasks be changed so that this wasn't violated? |
| 05:30:12 | MenTaLguY | I suppose you could take away features until they were the same as fibers, and then call them fibers |
| 05:30:24 | evan | there isn't much to take away |
| 05:30:32 | MenTaLguY | however, the right answer in this case is simple composition rather than inheritance |
| 05:30:32 | evan | the API is pretty sparse |
| 05:30:39 | MenTaLguY | fiber has-a task |
| 05:30:46 | MenTaLguY | rather than fiber is-a task |
| 05:30:52 | evan | sure |
| 05:30:54 | agardiner | yeah, that makes sense i guess |
| 05:32:50 | brixen | evan: if I: p _inspect(obj) => <MethodTable>, but p obj->obj_type => ObjectType, what might I have done? |
| 05:33:16 | agardiner | riddle me this... :-) |
| 05:33:21 | brixen | heh |
| 05:34:06 | MenTaLguY | The answer is chickens! |
| 05:34:30 | MenTaLguY | *Professor Layton tugs down his hat brim in shame* |
| 05:34:31 | evan | brixen: the class perhaps isn't setup to tag it's instances as the proper type |
| 05:34:57 | rue | Actually, the answer is "the egg" unless one wants to specify "chicken egg" |
| 05:35:04 | brixen | evan: ok, that's what I was thinking |
| 05:35:04 | rue | In which case it was, obviously, the chicken. |
| 05:35:20 | agardiner | does it need an lookup_table_init method? |
| 05:35:28 | agardiner | like sendsite? |
| 05:35:31 | evan | nah |
| 05:35:38 | evan | just needs to be done in bootstrap.c |
| 05:35:53 | brixen | yeah, I'll figure it out |
| 05:35:56 | evan | at some point, the bootstrap code has to be refactored and cleaned up |
| 05:36:32 | agardiner | ah, yeah, makes sense i think - lookup table is an object, not a struct as object... |
| 05:37:03 | brixen | well, I can do this: class F < LookupTable, and that works fine |
| 05:37:26 | brixen | and I've replaced BC(hash) for MethodTable with BC(lookuptable) |
| 05:37:46 | brixen | and fixed up methtbl_new, so I must be missing something else |
| 05:38:40 | evan | brixen: type's aren't inherited like that |
| 05:38:55 | evan | so be sure that class_set_object_type is called on BC(methtbl) |
| 05:39:03 | evan | i see that it is actualy |
| 05:39:19 | brixen | evan: where are you looking? |
| 05:39:25 | evan | brixen: oh... you konw what.. |
| 05:39:29 | evan | bootstrap.c |
| 05:39:31 | evan | line 44 |
| 05:39:39 | evan | er, line 77 |
| 05:39:48 | evan | line 77 needs to be moved I think |
| 05:40:05 | evan | i'll bet there are MethodTable instances being created BEFORE line 77 runs |
| 05:40:16 | brixen | ahh, could be |
| 05:40:52 | evan | i'm trying to track down the EBADF problem |
| 05:40:58 | evan | it seems to be something in the socket specs |
| 05:41:28 | brixen | yeah, I've only seen that on socket specs |
| 05:41:53 | evan | i think socket.rb is closing descriptors it shouldn't |
| 05:44:40 | GMFlash leaves the room. | |
| 05:46:31 | brixen | evan: that was it exactly, _methtbl_basic_class sets the class to ObjectType, I moved class_set_object_type(BC(methtbl), I2N(MTType)) right under that and it fixed it |
| 05:46:45 | evan | huzzah. |
| 05:46:56 | brixen | so, I'm to here: Unable to send [] on #<MethodTable> (NoMethodError) |
| 05:47:10 | brixen | on replacing MethodTable super with LookupTable |
| 05:47:12 | brixen | :) |
| 05:47:20 | evan | heh |
| 05:47:33 | ezmobius enters the room. | |
| 05:47:44 | evan | brixen: sounds like an order problem |
| 05:47:59 | brixen | evan: this should actually be a bit of a boost, because #hash was called over 600,000 times, just loading IRB |
| 05:48:00 | evan | brixen: you need LookupTable#[] to be available |
| 05:48:04 | mediogre enters the room. | |
| 05:48:10 | evan | awesome |
| 05:48:52 | brixen | evan: does this mean ordering in bootstrap? or I need to fix the depends on in core? |
| 05:49:12 | evan | ordering in bootstrap likely |
| 05:49:20 | brixen | hmm, ok |
| 05:49:21 | evan | since stuff like alias_method in bootstrap uses MethodTable#[] |
| 05:49:27 | brixen | right |
| 05:50:06 | brixen | did we not want to use depends on in bootstrap? |
| 05:51:42 | evan | so, wtf is with the TCPServer specs that test "can be interrupted by Thread#kill / Thread#raise" |
| 05:51:58 | evan | brixen: go ahead and use depends on in bootstrap |
| 05:52:18 | wycats enters the room. | |
| 05:52:31 | brixen | evan: ok |
| 05:52:57 | headius | evan: I added those |
| 05:53:28 | evan | ok |
| 05:53:29 | headius | TCPServer#accept doesn't prevent kill or raise from happening |
| 05:53:40 | evan | why would it? |
| 05:53:41 | headius | mongrel uses raise to shut down the server |
| 05:55:33 | RyanTM leaves the room. | |
| 05:55:45 | evan | ok, i see the idea |
| 05:55:49 | evan | seems like it should be more generic |
| 05:55:59 | headius | probably |
| 05:56:05 | evan | since it's really any 'blocking' call |
| 05:56:35 | headius | well, that's not entirely true, since calls out to native libraries can't be kill/raise'ed |
| 05:56:43 | headius | blocking IO, sure |
| 05:56:54 | evan | ah, true. |
| 05:58:03 | headius | in Ruby it's all done with selectors, I don't think EINTR enters into it |
| 05:58:20 | ezmobius leaves the room. | |
| 05:58:20 | aotearoa leaves the room. | |
| 05:58:20 | rphillips leaves the room. | |
| 05:58:20 | Chrononaut leaves the room. | |
| 05:58:20 | silex leaves the room. | |
| 05:58:20 | mojombo leaves the room. | |
| 05:58:33 | headius | er, in MRI |
| 05:58:47 | ezmobius enters the room. | |
| 05:58:47 | aotearoa enters the room. | |
| 05:58:47 | mojombo enters the room. | |
| 05:58:47 | rphillips enters the room. | |
| 05:58:47 | Chrononaut enters the room. | |
| 05:58:47 | silex enters the room. | |
| 05:59:00 | headius | MRI implements blocking IO in terms of non-blocking operations so it can deschedule that thread |
| 05:59:28 | headius | we're doing the same in JRuby (partly) to allow kill/raise to work |
| 06:00:10 | evan | yes |
| 06:00:15 | evan | so does rubinius |
| 06:00:28 | MenTaLguY | it's blocking polling, so it's not evil |
| 06:00:52 | headius | jruby currently has half and half |
| 06:00:52 | lstoll leaves the room. | |
| 06:01:07 | headius | since we don't need to worry about blocking to have other threads run |
| 06:01:16 | headius | but green threading operations like kill/raise are still an issue |
| 06:01:54 | MenTaLguY | Andys^: no. |
| 06:02:47 | MenTaLguY | Andys^: it requests notification when any outstanding nonblocking operations will be possible, and then blocks indefinitely until a pending operation is ready to perform |
| 06:04:26 | headius | I'm not sure green threaded impls can guarantee what thread will be running when the signal is received though |
| 06:05:13 | headius | plus the semantics of some operation require you to know the event has been received, like kill |
| 06:05:16 | TheVoic1 enters the room. | |
| 06:05:47 | TheVoic1 leaves the room. | |
| 06:06:45 | agardiner | later folks! o/ |
| 06:07:00 | agardiner leaves the room. | |
| 06:08:10 | headius | that's basically what jruby does |
| 06:08:19 | headius | though all threads are always awake |
| 06:08:33 | headius | so they periodically have to check "am I dead?" |
| 06:09:04 | headius | that's also basically what Ruby does...for raise, for example, it sets an exception for the target thread which gets raised when it wakes up |
| 06:09:09 | headius | but I think it also immediately wakes it up |
| 06:09:20 | headius | it's a pretty primitive scheduler |
| 06:09:37 | hoopy leaves the room. | |
| 06:10:03 | hoopy enters the room. | |
| 06:10:23 | headius | I'm going to reimplement JRuby's cross-thread events explicitly that way though |
| 06:10:36 | headius | so polling for an event is nearly zero cost until there's an event to be had |
| 06:12:18 | headius | sleep |
| 06:13:58 | headius leaves the room. | |
| 06:14:07 | b3b0p leaves the room. | |
| 06:15:32 | rubymaverick leaves the room. | |
| 06:16:00 | dkubb enters the room. | |
| 06:17:23 | wmoxam leaves the room. | |
| 06:18:00 | evan | OH |
| 06:18:03 | evan | I know whats wrong. |
| 06:21:36 | TheVoice leaves the room. | |
| 06:27:56 | hoopy leaves the room. | |
| 06:33:30 | hoopy enters the room. | |
| 06:34:20 | TheVoice enters the room. | |
| 06:49:00 | w1rele55 leaves the room. | |
| 06:51:18 | ttmrichter enters the room. | |
| 06:51:24 | w1rele55 enters the room. | |
| 06:57:10 | wifelette enters the room. | |
| 06:58:05 | srbaker leaves the room. | |
| 07:01:58 | brixen | evan: hah, I figured it out. it was failing in kernel/platform/env.rb calling #private, which of course manipulates the MT, and stables knew nothing about LookupTable, so looking up #[] on MT wouldn't find anything |
| 07:02:31 | brixen | so, no ordering issue in bootstrap still |
| 07:02:46 | evan | oh good. |
| 07:03:16 | ezmobius leaves the room. | |
| 07:03:22 | brixen | rbx has a vertable samurai armory of double-edged swords :) |
| 07:08:31 | hoopy | is trying to build rbx on a slug |
| 07:09:57 | ttmrichter leaves the room. | |
| 07:10:30 | ttmrichter enters the room. | |
| 07:10:35 | d2dchat enters the room. | |
| 07:11:49 | d2dchat leaves the room. | |
| 07:12:06 | d2dchat enters the room. | |
| 07:12:42 | benburkert enters the room. | |
| 07:14:52 | ttmrichter leaves the room. | |
| 07:25:31 | rue | brixen: Actually, did they not mostly use single-edged weapons? |
| 07:25:56 | perplexes leaves the room. | |
| 07:26:00 | _mutle leaves the room. | |
| 07:26:23 | brixen | rue: hmm, well yes, but the tip often had both edges sharpened |
| 07:27:20 | brixen | rue: just imagine a lot of swords, if they are single-edged, imagine a couple taped together :) |
| 07:31:49 | jinjing enters the room. | |
| 07:32:58 | dbussink | morning |
| 07:33:19 | ttmrichter enters the room. | |
| 07:34:07 | radarek enters the room. | |
| 07:34:46 | GMFlash enters the room. | |
| 07:39:51 | gnufied enters the room. | |
| 07:40:40 | crafterm leaves the room. | |
| 07:42:12 | crafterm enters the room. | |
| 07:42:32 | crafterm leaves the room. | |
| 07:50:33 | rue | Morning |
| 07:50:34 | aotearoa leaves the room. | |
| 07:50:37 | rue | Short night there |
| 07:50:44 | aotearoa enters the room. | |
| 07:52:06 | MenTaLguY leaves the room. | |
| 07:52:06 | quik_ leaves the room. | |
| 07:52:33 | digitalfallout leaves the room. | |
| 08:00:42 | wifelette leaves the room. | |
| 08:02:10 | dbussink | rue: yeah, i have a meeting this morning |
| 08:11:55 | thehcdreamer enters the room. | |
| 08:13:55 | benburkert leaves the room. | |
| 08:16:54 | gnufied leaves the room. | |
| 08:17:34 | lstoll enters the room. | |
| 08:19:05 | twbray leaves the room. | |
| 08:19:20 | perdix enters the room. | |
| 08:22:10 | _VVSiz_ enters the room. | |
| 08:23:11 | w1rele55 leaves the room. | |
| 08:30:08 | lstoll leaves the room. | |
| 08:31:15 | w1rele55 enters the room. | |
| 08:32:35 | nkpart leaves the room. | |
| 08:37:28 | gnufied enters the room. | |
| 08:38:52 | ragnard enters the room. | |
| 08:38:52 | ragnard leaves the room. | |
| 08:39:35 | ragnard enters the room. | |
| 08:40:23 | VVSiz_ leaves the room. | |
| 08:40:42 | zimbatm enters the room. | |
| 08:42:26 | cavalle leaves the room. | |
| 08:48:09 | vintrepid leaves the room. | |
| 08:48:39 | crossblaim enters the room. | |
| 08:49:52 | imajes enters the room. | |
| 08:50:40 | alloy enters the room. | |
| 08:52:50 | imajes leaves the room. | |
| 09:01:08 | alloy leaves the room. | |
| 09:08:26 | alloy enters the room. | |
| 09:15:26 | cavalle enters the room. | |
| 09:21:25 | aotearoa leaves the room. | |
| 09:26:34 | aotearoa enters the room. | |
| 09:26:56 | octopod enters the room. | |
| 09:38:33 | Arjen_ enters the room. | |
| 09:50:37 | aotearoa_ enters the room. | |
| 09:50:37 | alloy leaves the room. | |
| 09:50:52 | alloy enters the room. | |
| 09:51:40 | aotearoa leaves the room. | |
| 09:53:14 | TheVoice leaves the room. | |
| 09:55:21 | imajes enters the room. | |
| 09:56:10 | imajes_office leaves the room. | |
| 10:01:41 | geekounet leaves the room. | |
| 10:03:27 | JimMc enters the room. | |
| 10:04:03 | JimMc | morning |
| 10:08:37 | comboy leaves the room. | |
| 10:08:46 | JimMc | well fine, be that way :P |
| 10:13:50 | perdiy enters the room. | |
| 10:13:50 | perdix leaves the room. | |
| 10:14:10 | cavalle_ enters the room. | |
| 10:29:46 | cavalle leaves the room. | |
| 10:32:16 | GMFlash leaves the room. | |
| 10:38:13 | mernen leaves the room. | |
| 10:48:42 | dewd enters the room. | |
| 10:50:45 | chris2 enters the room. | |
| 10:58:04 | lstoll enters the room. | |
| 11:12:39 | Fullmoon leaves the room. | |
| 11:27:49 | JimMc | hey guys, in the /bin of the build dir are any of these utilities needed for the build process? |
| 11:28:09 | JimMc | autotest bm ci completeness contributors.sh mspec |
| 11:28:09 | JimMc | autotestng.rb bm-automatic cmspec contributors.rb mkspec name_map.rb |
| 11:28:32 | JimMc | **porting to solaris |
| 11:31:01 | JimMc | ok figured it out |
| 11:58:28 | rue | JimMc: bin/mspec is more or less essential |
| 12:23:54 | cavalle_ leaves the room. | |
| 12:25:44 | crossblaim leaves the room. | |
| 12:32:13 | crossblaim enters the room. | |
| 12:32:37 | jinjing leaves the room. | |
| 12:33:18 | jinjing enters the room. | |
| 12:36:06 | lstoll_ enters the room. | |
| 12:41:49 | lstoll leaves the room. | |
| 12:41:56 | lstoll_ leaves the room. | |
| 12:47:27 | rue | And good morning |
| 13:04:37 | jinjing leaves the room. | |
| 13:05:52 | jinjing enters the room. | |
| 13:18:20 | chop3_ enters the room. | |
| 13:19:02 | chop3 leaves the room. | |
| 13:19:51 | chop3_ leaves the room. | |
| 13:20:50 | obiejuan enters the room. | |
| 13:22:26 | skaar enters the room. | |
| 13:22:36 | jinjing leaves the room. | |
| 13:27:55 | wmoxam enters the room. | |
| 13:28:32 | dysinger leaves the room. | |
| 13:31:03 | ubiquitous leaves the room. | |
| 13:34:34 | aotearoa_ leaves the room. | |
| 13:48:23 | RyanTM enters the room. | |
| 14:00:46 | digitalfallout enters the room. | |
| 14:04:22 | wmoxam leaves the room. | |
| 14:13:51 | b3b0p enters the room. | |
| 14:23:58 | headius enters the room. | |
| 14:25:11 | enebo enters the room. | |
| 14:26:07 | ragnard leaves the room. | |
| 14:28:10 | AndrewO enters the room. | |
| 14:30:58 | nicksieger leaves the room. | |
| 14:31:42 | headius leaves the room. | |
| 14:39:15 | obiejuan leaves the room. | |
| 14:46:57 | wmoxam enters the room. | |
| 14:51:09 | jbwiv_ enters the room. | |
| 14:55:03 | probablycorey enters the room. | |
| 15:02:46 | headius enters the room. | |
| 15:04:14 | probablycorey_ enters the room. | |
| 15:05:54 | dysinger enters the room. | |
| 15:07:38 | wmoxam leaves the room. | |
| 15:08:25 | headius_ enters the room. | |
| 15:10:04 | gnufied leaves the room. | |
| 15:11:39 | probablycorey leaves the room. | |
| 15:11:42 | headius__ enters the room. | |
| 15:11:53 | headius leaves the room. | |
| 15:13:26 | headius_ leaves the room. | |
| 15:13:55 | therealadam enters the room. | |
| 15:14:00 | benburkert enters the room. | |
| 15:15:18 | nicksieger enters the room. | |
| 15:18:48 | rue | Man, this is a totally epic battle |
| 15:21:17 | probablycorey_ leaves the room. | |
| 15:21:33 | enebo leaves the room. | |
| 15:22:36 | probablycorey enters the room. | |
| 15:26:30 | wmoxam enters the room. | |
| 15:26:57 | vintrepid enters the room. | |
| 15:27:33 | GMFlash enters the room. | |
| 15:31:39 | probablycorey_ enters the room. | |
| 15:31:55 | obvio enters the room. | |
| 15:35:20 | Rue_ enters the room. | |
| 15:36:12 | Rue_Mac61 enters the room. | |
| 15:36:24 | Rue_Mac61 | Neat |
| 15:36:38 | srbaker enters the room. | |
| 15:37:05 | headius | epic, truly |
| 15:37:52 | macournoyer enters the room. | |
| 15:38:38 | probablycorey leaves the room. | |
| 15:40:01 | twbray enters the room. | |
| 15:43:06 | zimbatm leaves the room. | |
| 15:48:25 | rue | headius: Took me the entire night! |
| 15:50:05 | probablycorey_ leaves the room. | |
| 15:51:29 | headius | so what were you working on/ |
| 15:51:42 | headius | implementing flip-flop? |
| 15:52:09 | probablycorey enters the room. | |
| 15:56:45 | rue | headius: Something far more productive, I now have a fully working ethernet-capable x61 with OS X |
| 15:57:46 | kamal_fariz | what's the legality of that i wonder |
| 15:57:53 | headius | SHHH |
| 15:58:14 | kamal_fariz | lol |
| 16:04:08 | ttmrichter leaves the room. | |
| 16:08:16 | probablycorey leaves the room. | |
| 16:10:20 | obiejuan enters the room. | |
| 16:10:40 | probablycorey enters the room. | |
| 16:11:34 | KirinDav enters the room. | |
| 16:11:58 | d2dchat leaves the room. | |
| 16:13:05 | rue | kamal_fariz: Did you not see the photo? Totally legit :) |
| 16:13:28 | kamal_fariz | the imac, yeah |
| 16:14:06 | rue | I have an Apple sticker and a creative way of reading the EULA |
| 16:15:19 | benburkert leaves the room. | |
| 16:15:39 | headius leaves the room. | |
| 16:15:51 | enebo enters the room. | |
| 16:15:56 | headius enters the room. | |
| 16:18:23 | jbwiv | rue: osx on a thinkpad? |
| 16:18:50 | enebo_ enters the room. | |
| 16:20:18 | rue | jbwiv_: Ya |
| 16:20:37 | jbwiv | wow...how'd you accomplish that one? |
| 16:21:06 | rue | m4d 5ki11z |
| 16:21:14 | jbwiv | :) |
| 16:21:45 | rue | Also, lots of helpful folks at the insanely mac forums |
| 16:22:11 | jbwiv | sweet...I'll have to do some digging...would love to get that working |
| 16:22:14 | nicksieger leaves the room. | |
| 16:24:02 | nicksieger enters the room. | |
| 16:24:12 | scoopr | rue, so you are gonna work on mod_rubinius? |
| 16:26:27 | hoopy leaves the room. | |
| 16:27:00 | enebo__ enters the room. | |
| 16:27:02 | rue | scoopr: Have been for the past, what, three weeks |
| 16:27:12 | scoopr | ah okay =) |
| 16:27:22 | enebo leaves the room. | |
| 16:27:23 | enebo_ leaves the room. | |
| 16:27:44 | scoopr | have you had some decision made already for the overall design? |
| 16:28:51 | rue | Yeah |
| 16:30:09 | rue | Doing thin modules for the various web servers. All they do is ensure the daemon side is up and running and push data over |
| 16:30:39 | rue | Some finessing there to allow different sites to use different daemons and so on |
| 16:30:45 | obiejuan leaves the room. | |
| 16:31:00 | scoopr | aight, sounds good |
| 16:31:12 | rue | I am still not fully decided on the daemon, I will try to build it into Rubinius itself probably as an extension |
| 16:31:57 | rue | We already have libev in plus some other arch that should make it a bit more streamlined than going through a separate app into rbx |
| 16:33:07 | macournoyer | I heard someone was porting EventMachine to rubinius using Channels and stuff? |
| 16:35:47 | hoopy enters the room. | |
| 16:36:08 | jtoy leaves the room. | |
| 16:41:51 | imajes leaves the room. | |
| 16:42:34 | jinjing enters the room. | |
| 16:44:54 | benburkert enters the room. | |
| 16:45:15 | perplexes enters the room. | |
| 16:48:28 | twbray leaves the room. | |
| 16:48:44 | imajes enters the room. | |
| 16:50:02 | lopex enters the room. | |
| 16:50:24 | jtoy enters the room. | |
| 16:55:12 | wifelette enters the room. | |
| 16:57:11 | GMFlash leaves the room. | |
| 17:02:15 | benburkert leaves the room. | |
| 17:03:41 | vintrepid leaves the room. | |
| 17:16:32 | vintrepid enters the room. | |
| 17:23:39 | jtoy leaves the room. | |
| 17:24:06 | hoopy leaves the room. | |
| 17:24:30 | KirinDave leaves the room. | |
| 17:29:13 | d2dchat enters the room. | |
| 17:34:41 | thehcdreamer leaves the room. | |
| 17:39:24 | dodecaphonic enters the room. | |
| 17:42:09 | wycats leaves the room. | |
| 17:46:02 | hoopy enters the room. | |
| 17:48:31 | Defiler | rue: So.. hackintosh.. |
| 17:48:37 | Defiler | rue: Did you get audio working? |
| 17:49:53 | rue | Defiler: Have not tried at all, actually |
| 17:49:59 | rue | I usually have no sounds |
| 17:50:14 | hoopy leaves the room. | |
| 17:50:20 | hoopy enters the room. | |
| 17:51:11 | jtoy enters the room. | |
| 17:53:47 | Defiler | rue: Let me know how it goes, if you eventually try |
| 17:54:13 | rue | Defiler: Looks like a couple guys are running the SoundBlaster USB thing and the new HDA patch might resolve native sound |
| 17:54:28 | MenTaLguY enters the room. | |
| 17:55:00 | Defiler | Cool |
| 17:55:11 | MenTaLguY | howdy |
| 17:55:17 | rue | Defiler: I have full graphics, trackpad, power management (though I did not grab the SpeedStep kernel), ethernet through USB so I am set nicely |
| 17:55:50 | rue | Might investigate the builtin wifi/wlan/ethernet later on. Apple's kernel api is pretty damn nice I have to say |
| 17:55:55 | rue | Hello, MenTaLguY |
| 18:00:32 | sudoer enters the room. | |
| 18:00:57 | sudoer leaves the room. | |
| 18:01:48 | sudoer enters the room. | |
| 18:02:03 | sudoer leaves the room. | |
| 18:02:59 | binary42_ enters the room. | |
| 18:03:19 | binary42 leaves the room. | |
| 18:04:03 | wifelette leaves the room. | |
| 18:08:01 | TheVoice enters the room. | |
| 18:11:38 | mediogre leaves the room. | |
| 18:11:42 | jtoy leaves the room. | |
| 18:17:52 | d2dchat_ enters the room. | |
| 18:20:35 | probablycorey_ enters the room. | |
| 18:20:36 | probablycorey leaves the room. | |
| 18:20:41 | KirinDave enters the room. | |
| 18:22:27 | Arjen_ leaves the room. | |
| 18:22:43 | binary42_ leaves the room. | |
| 18:22:54 | wifelette enters the room. | |
| 18:28:18 | probablycorey enters the room. | |
| 18:33:30 | d2dchat leaves the room. | |
| 18:37:13 | jinjing leaves the room. | |
| 18:37:41 | probablycorey_ leaves the room. | |
| 18:42:00 | wycats enters the room. | |
| 18:47:27 | probablycorey leaves the room. | |
| 18:47:32 | jero5 enters the room. | |
| 18:48:49 | probablycorey enters the room. | |
| 18:49:03 | headius_ enters the room. | |
| 18:50:05 | enebo_ enters the room. | |
| 18:54:07 | brixen | git is pure awesome: stash, reset, commit --amend, and easy branching make my day |
| 18:59:31 | dysinger leaves the room. | |
| 18:59:40 | rue leaves the room. | |
| 18:59:47 | nicksieger leaves the room. | |
| 18:59:58 | headius leaves the room. | |
| 19:00:36 | enebo leaves the room. | |
| 19:05:40 | w1rele55 leaves the room. | |
| 19:33:26 | tizianobis enters the room. | |
| 19:33:55 | Arjen_ enters the room. | |
| 19:34:07 | brixen | is it possible? yes! MethodTable now uses LookupTable instead of Hash |
| 19:34:30 | brixen | thank evan for machine_handle_fire :) |
| 19:34:37 | anonuser | what is that? |
| 19:34:46 | ctennis_ enters the room. | |
| 19:34:58 | TheVoice leaves the room. | |
| 19:35:19 | brixen | it catches segfaults etc in C land and gives you a nice BT for one |
| 19:35:26 | anonuser | awesome. |
| 19:35:28 | brixen | it also gives you a nice debug point |
| 19:35:33 | anonuser | A boon to debugging? |
| 19:35:37 | brixen | indeed |
| 19:36:21 | brixen | now, to figure out why the counts go wonky when the LT redistributes the entries |
| 19:48:21 | vintrepid leaves the room. | |
| 19:49:38 | vintrepid enters the room. | |
| 19:49:46 | ezmobius enters the room. | |
| 19:49:58 | TheVoice enters the room. | |
| 19:53:43 | enebo_ leaves the room. | |
| 19:53:50 | headius leaves the room. | |
| 20:00:03 | wycats leaves the room. | |
| 20:04:39 | tizianobis_ enters the room. | |
| 20:06:22 | nicksieger leaves the room. | |
| 20:12:07 | boyscout | 2 commits by Caleb Tennis |
| 20:12:08 | boyscout | * Make rb_include_module public via ruby.h; dec9446 |
| 20:12:09 | boyscout | * Use more of a lookup table approach for exceptions; e23f2e9 |
| 20:13:51 | wycats enters the room. | |
| 20:20:09 | wmoxam leaves the room. | |
| 20:22:42 | tizianobis leaves the room. | |
| 20:25:20 | TheVoice leaves the room. | |
| 20:25:24 | rubuildius_ppc | Caleb Tennis: dec9446a5; 1736 files, 5862 examples, 19915 expectations, 0 failures, 1 error; http://pastie.caboo.se/paste/162960 |
| 20:26:14 | wmoxam enters the room. | |
| 20:32:01 | TheVoice enters the room. | |
| 20:32:46 | imajes enters the room. | |
| 20:35:09 | wmoxam leaves the room. | |
| 20:35:35 | headius enters the room. | |
| 20:36:15 | enebo enters the room. | |
| 20:37:25 | Arjen_ leaves the room. | |
| 20:40:30 | wmoxam enters the room. | |
| 20:40:52 | eugenebrechko enters the room. | |
| 20:45:44 | probablycorey leaves the room. | |
| 20:47:07 | eugenebrechko enters the room. | |
| 20:47:26 | chop3 enters the room. | |
| 20:47:57 | chop3 leaves the room. | |
| 20:54:01 | tizianobis_ leaves the room. | |
| 20:55:13 | therealadam leaves the room. | |
| 20:56:58 | chop3 enters the room. | |
| 21:01:06 | nicksieger enters the room. | |
| 21:04:38 | binary42 enters the room. | |
| 21:07:58 | wycats leaves the room. | |
| 21:08:32 | wycats enters the room. | |
| 21:21:42 | binary42 leaves the room. | |
| 21:22:12 | boyscout | 2 commits by Caleb Tennis |
| 21:22:13 | boyscout | * Minor fix to rb_require to make it a little more rubyesque; 5b999a1 |
| 21:22:14 | boyscout | * Add rb_gv_get and rb_gv_set, plus specs.; aea5cc4 |
| 21:24:10 | ctennis_ leaves the room. | |
| 21:34:45 | rubuildius_ppc | Caleb Tennis: 5b999a141; 1738 files, 5866 examples, 19922 expectations, 0 failures, 1 error; http://pastie.caboo.se/paste/163009 |
| 21:37:03 | rue enters the room. | |
| 21:37:19 | brixen | wb rue |
| 21:41:06 | rue | 'Lo |
| 21:41:42 | rue | I need a router, I think. I cannot seem to get all machines on at the same time |
| 21:43:04 | w1rele55 enters the room. | |
| 21:45:37 | squeegy | rue: twc only gives you two real ips |
| 21:45:54 | squeegy | at least that is the way it used to be. |
| 21:49:45 | rby enters the room. | |
| 21:50:07 | scoopr | hmh.. http://www.nabble.com/ANN:-New-Objective-C-Runtime-td13648692.html |
| 21:53:34 | brixen | that looks cool |
| 21:55:08 | jbwiv_ leaves the room. | |
| 21:59:42 | headius | apple doesn't use stock gnu obj-c, does it? |
| 21:59:55 | MenTaLguY | nope |
| 22:00:55 | benburkert enters the room. | |
| 22:02:24 | dodecaphonic leaves the room. | |
| 22:09:24 | krisps enters the room. | |
| 22:11:28 | skaar leaves the room. | |
| 22:13:35 | enebo leaves the room. | |
| 22:13:35 | headius leaves the room. | |
| 22:14:00 | brixen | boy, if you inadvertently run bin/mspec instead of bin/mspec ci, you might think the sky was falling |
| 22:14:09 | enebo enters the room. | |
| 22:14:09 | headius enters the room. | |
| 22:19:23 | cavalle enters the room. | |
| 22:23:08 | krisps_ enters the room. | |
| 22:24:29 | krisps_ leaves the room. | |
| 22:24:35 | krisps_ enters the room. | |
| 22:26:36 | imajes leaves the room. | |
| 22:26:40 | krisps_ leaves the room. | |
| 22:26:52 | krisps_ enters the room. | |
| 22:28:27 | krisps leaves the room. | |
| 22:32:08 | octopod leaves the room. | |
| 22:37:16 | wycats leaves the room. | |
| 22:37:49 | wycats enters the room. | |
| 22:39:56 | cavalle leaves the room. | |
| 22:40:08 | cavalle enters the room. | |
| 22:42:27 | enebo leaves the room. | |
| 22:43:42 | dewd_ enters the room. | |
| 22:46:28 | chris2 leaves the room. | |
| 22:47:54 | eugenebrechko leaves the room. | |
| 22:48:28 | GMFlash enters the room. | |
| 22:54:16 | evan | at the sprint |
| 22:54:24 | evan | on sloooooow hotel internet |
| 22:54:33 | krisps_ leaves the room. | |
| 22:54:57 | krisps enters the room. | |
| 22:56:42 | anonuser leaves the room. | |
| 22:56:42 | dewd leaves the room. | |
| 22:56:44 | krisps leaves the room. | |
| 22:57:10 | krisps enters the room. | |
| 22:57:13 | djwhitt | many people there at the sprint? |
| 22:57:46 | drbrain | 15 so far |
| 22:59:48 | wifelette leaves the room. | |
| 23:01:47 | krisps_ enters the room. | |
| 23:02:07 | krisps leaves the room. | |
| 23:04:25 | boyscout | 1 commit by Brian Ford |
| 23:04:26 | boyscout | * Bandaid fix for failing #autoload specs.; d7d9bfd |
| 23:04:42 | brixen | soo close on the MT switch to using LookupTable |
| 23:04:51 | brixen | if we could keep HEAD green, that would be nice :) |
| 23:05:09 | evan | brixen: zenspider needs to fix his eval spec |
| 23:05:18 | brixen | yeah, I know |
| 23:05:26 | w1rele55 leaves the room. | |
| 23:05:57 | brixen | evan: there's quite a few places that need proper scoping |
| 23:06:08 | brixen | I guess we'll do a cleanup pass soon |
| 23:06:28 | brixen | ok, must eat, bbiab.. |
| 23:07:25 | headius | hey, do you guys have a way to do full profiling right now? |
| 23:07:43 | macournoyer leaves the room. | |
| 23:08:42 | antares enters the room. | |
| 23:09:49 | lstoll enters the room. | |
| 23:10:06 | antares | on shotgun VM instance setup there's a Ruby constant VM_INFERIOR being set, like this: machine_set_const_under(m, "VM_INFERIOR", m->sub ? Qtrue : Qfalse, mod); — what does this mean, what is inferior VM in this case? |
| 23:10:38 | headius | it means "any VM other than Rubinius" |
| 23:14:41 | rubuildius_ppc | Brian Ford: d7d9bfd01; 1738 files, 5866 examples, 19924 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/163072 |
| 23:15:21 | antares | headius, sorry I still cannot get it |
| 23:15:31 | headius | I was just joking |
| 23:15:54 | antares | headius, what this Ruby constant means then? |
| 23:16:01 | antares | headius, you got me :) |
| 23:19:13 | d2dchat enters the room. | |
| 23:19:21 | d2dchat_ leaves the room. | |
| 23:20:11 | AndrewO leaves the room. | |
| 23:21:29 | dewd_ leaves the room. | |
| 23:23:48 | benburkert leaves the room. | |
| 23:24:00 | benburkert enters the room. | |
| 23:30:44 | brixen | antares: that looks like the mvm stuff, e.g. git show 34300952 |
| 23:31:51 | antares | brixen, yup, got it with git log <commit> |
| 23:31:59 | antares | should have done it earlier |
| 23:32:18 | antares | log message says: Expose a VM's id and whether it's a sub VM |
| 23:32:26 | brixen | yep |
| 23:32:38 | antares | so inferior VM means "spawned off of another VM", "child instance" |
| 23:32:46 | brixen | afaik, yes |
| 23:32:47 | krisps_ leaves the room. | |
| 23:33:01 | krisps enters the room. | |
| 23:33:35 | lopex leaves the room. | |
| 23:33:35 | TheVoice leaves the room. | |
| 23:36:39 | TheVoice enters the room. | |
| 23:37:15 | obvio leaves the room. | |
| 23:42:06 | lopex enters the room. | |
| 23:43:10 | antares | brixen, I have just realized full power of git log -S option |
| 23:43:22 | antares | git log -S"int sub" -- shotgun/lib/machine.h gives the answer |
| 23:43:54 | d2dchat leaves the room. | |
| 23:44:37 | jrun enters the room. | |
| 23:45:12 | dbussink | evening |
| 23:45:17 | dbussink | just in for a minute though :) |
| 23:45:23 | brixen | antares: nice |
| 23:45:35 | dbussink | brixen: i read work is progressing on the mt refactor? |
| 23:45:38 | brixen | antares: that -S is tricky, I tried it -S "int sub" the first time and got complaints |
| 23:45:50 | brixen | dbussink: *just* got my first clean ci run! |
| 23:45:52 | brixen | woot! |
| 23:45:58 | dbussink | woot! |
| 23:45:59 | antares | brixen, you must leave no space between -S and opening double quote |
| 23:46:12 | jinjing enters the room. | |
| 23:46:13 | brixen | antares: yeah, kindof a gotcha there |
| 23:46:23 | brixen | antares: and the man page for git-log doesn't list -S |
| 23:46:26 | brixen | what is it? |
| 23:46:39 | antares | brixen, I found that in man |
| 23:46:57 | antares | I am using man from kernel.org ftp, git 1.5.4 |
| 23:47:26 | brixen | ahh, interesting, I have 1.5.3.7 installed, dunno which version of the man pages though |
| 23:47:30 | brixen | I installed them separately |
| 23:48:28 | brixen | today's trivia fact: calls to cpu_find_method in a full CI run: 15,774,118 |
| 23:48:48 | brixen | no wonder removing 100k calls barely blips on overall time |
| 23:49:25 | wifelette enters the room. | |
| 23:49:38 | dbussink | what removed those 100k calls? |
| 23:52:48 | brixen | replaceing ranges like 0...n with 0, n |
| 23:52:57 | brixen | some stuff I did last week |
| 23:53:10 | dbussink | ah |
| 23:53:18 | dbussink | any difference because of the new methodtable? |
| 23:53:54 | brixen | heh, haven't run numbers |
| 23:54:04 | brixen | I'm still basking in the warm glow of a clean CI run |
| 23:54:16 | brixen | some serious head banging last night :) |
| 23:54:30 | dbussink | i can imagine yeah |
| 23:54:42 | dbussink | well, like they say, friday is a good day to break the build ;) |
| 23:54:56 | brixen | hah |
| 23:55:03 | brixen | this won't be broke :P |
| 23:57:15 | brixen | evan: did you find the issue with EBADF? |
| 23:57:15 | benburkert leaves the room. | |
| 23:57:57 | benburkert enters the room. | |
| 23:58:56 | lstoll leaves the room. | |
| 23:59:03 | wifelette leaves the room. |