Show enters and exits. Hide enters and exits.
| 01:04:13 | boyscout | Put melbourne ext for rbx in a subdir. - 02c813d - Brian Ford |
| 01:04:13 | boyscout | Build external libs for parser universal on darwin. - 47cfec0 - Brian Ford |
| 01:04:13 | boyscout | Build melbourne for MRI. - ef46d5d - Brian Ford |
| 01:06:46 | boyscout | CI: Build ef46d5d failed. http://ci.rubini.us/rubinius/builds/ef46d5d2db9019884a240d3d7df066c8a16e4026 |
| 01:08:30 | brixen | that's about as ugly a bt as you could get |
| 01:26:10 | evan | but hey! |
| 01:26:12 | evan | a backtrace! |
| 01:26:17 | evan | better than before |
| 01:28:03 | brixen | yep! |
| 01:28:09 | slava | evan: did you sort out the issue where brixen's new compiler was slowing down in GC? |
| 01:28:33 | evan | no. |
| 01:28:48 | evan | fixed one thing that might help |
| 01:28:58 | evan | but haven't focused strictly on that issue |
| 01:33:48 | imajes | stesla: ping |
| 01:42:52 | brixen | hrm, c-ng is partially working with melbourne under MRI |
| 01:43:09 | brixen | except that I'm getting Fixnums instead of Symbols |
| 01:44:01 | slava | are you porting it to MRI for bootstrap purposes? |
| 01:44:11 | brixen | yeah |
| 01:44:16 | slava | cool |
| 01:44:21 | slava | so its almost ready then |
| 01:44:37 | brixen | although, you know what I'm thinking |
| 01:44:47 | brixen | rbc files are platform independent |
| 01:45:09 | brixen | so, we should add a configure option for prebuilts just like we do for llvm |
| 01:45:57 | evan | brixen: we've been blindly calling unshare everytime we use as_rstring() |
| 01:46:05 | evan | which blindly duplicates data_ |
| 01:46:09 | evan | even if it's not shared. |
| 01:46:09 | evan | :/ |
| 01:46:49 | brixen | :( |
| 01:47:02 | evan | which is where the crash was (still not sure about it) |
| 01:47:11 | evan | but it's SUCKING down performance too. |
| 02:01:23 | brixen | hm, why would I get a Fixnum back from rb_intern on MRI... |
| 02:04:54 | evan | you shouldn't... |
| 02:04:57 | evan | you get an ID back |
| 02:04:59 | evan | not a Symbol |
| 02:05:04 | evan | you need to use SYM2ID() on it |
| 02:05:07 | evan | to get a Symbol |
| 02:05:12 | brixen | ID2SYM |
| 02:05:15 | evan | ID != Symbol on MRI |
| 02:05:18 | evan | er. |
| 02:05:19 | evan | yes. |
| 02:05:21 | evan | ID2SYM |
| 02:05:30 | brixen | I just realized that |
| 02:05:35 | brixen | MRI is so confusing |
| 02:05:37 | evan | yeah, thats silly |
| 02:05:43 | evan | it's a legacy thing |
| 02:05:49 | evan | because the parser uses ID's internally |
| 02:05:51 | evan | as you know. |
| 02:06:01 | brixen | yeah |
| 02:06:03 | Zoxc | why is the ID mess there? |
| 02:06:12 | evan | Zoxc: purely legacy |
| 02:06:15 | evan | it's never been cleaned up |
| 02:06:20 | Zoxc | in the first place? =P |
| 02:06:26 | evan | oh, in the first place |
| 02:06:40 | evan | it was purely the token the symbol table gave out |
| 02:06:46 | evan | ruby didn't have symbols in the beginning |
| 02:07:00 | evan | at one point, matz added a quick parser hack |
| 02:07:06 | evan | which created |
| 02:07:11 | evan | :foo.class # => Fixnum |
| 02:07:18 | evan | in time, he changed that to be Symbol |
| 02:07:26 | evan | but didn't go back and clean up the ID usage |
| 02:07:43 | Zoxc | ah |
| 02:07:47 | slava | matz has inflicted much pain on the world |
| 02:08:11 | evan | and much joy. |
| 02:09:29 | brixen | yeah, I forgive the pain |
| 02:09:42 | brixen | just wish we'd started rewriting MRI in like 2000 |
| 02:10:57 | imajes | that would have been interesting |
| 02:11:03 | imajes | but |
| 02:11:05 | imajes | it's OK |
| 02:11:09 | imajes | i'm just glad when we'll be done :) |
| 02:11:30 | brixen | heh |
| 02:11:37 | boyscout | Unshare intellegently, try and fix a strange bug - 10e0af7 - Evan Phoenix |
| 02:11:37 | boyscout | Fix Regexp management bug - ee4d78c - Evan Phoenix |
| 02:11:37 | boyscout | Fix operand name so instruction decode is nicer - e8187c5 - Evan Phoenix |
| 02:11:37 | boyscout | Fix describe showing things multiple times - ab3f496 - Evan Phoenix |
| 02:11:37 | boyscout | Fix Dir["*/foo"] from picking up incorrect files - b1ca90b - Evan Phoenix |
| 02:11:37 | boyscout | Update Rubygems to 1.3.5 - d44fed4 - Evan Phoenix |
| 02:11:42 | brixen | imajes: you'll be at RC? |
| 02:12:07 | evan | ok, bbiab. |
| 02:12:07 | brixen | woot! |
| 02:12:13 | brixen | evan: you updated rubygems! |
| 02:12:15 | brixen | yay |
| 02:12:33 | imajes | brixen: not this time :( |
| 02:12:38 | brixen | imajes: :( |
| 02:12:39 | imajes | brixen: stuck in argentina ;) |
| 02:12:46 | brixen | ohh so stuck :P |
| 02:12:49 | imajes | heh |
| 02:12:56 | brixen | I'd like to be stuck in argentina |
| 02:13:05 | imajes | yeah, more time for me in the US this year is .. problematic |
| 02:13:29 | brixen | did you use up all your time chips? |
| 02:14:58 | imajes | something like that |
| 02:15:13 | imajes | brixen: trips via US INS usually end up with lots of questions |
| 02:15:15 | imajes | and bag searches |
| 02:15:39 | brixen | bastards |
| 02:15:58 | brixen | imajes: if you weren't such a kingpin and all :) |
| 02:16:49 | imajes | brixen: riight |
| 02:19:12 | boyscout | CI: Build d44fed4 failed. http://ci.rubini.us/rubinius/builds/d44fed4c2d47ede57ffbe95436a1622ae1af5be2 |
| 07:03:08 | boyscout | Revert of 47cfec0d61, Snow Leopard builds fail once you specify the arch options for just 32-bit platforms. - e269a86 - David Waite |
| 07:03:35 | ddub | (brixen tricked me into testing my commit bit) |
| 07:04:42 | boyscout | CI: Build e269a86 failed. http://ci.rubini.us/rubinius/builds/e269a86f92e6bd5dddbb8517d0c17849633699a9 |
| 07:08:45 | ddub | oh no you didn't |
| 07:09:36 | ddub | :( couldn't be me |
| 07:09:43 | ddub | ahh ok, wasn't |
| 07:09:44 | ddub | whew |
| 07:09:51 | brianmario | haha |
| 07:09:51 | brixen | yeah, that's not you |
| 07:46:42 | ice799 | 8388608 |
| 07:46:44 | ice799 | Sigh. |
| 07:46:48 | ice799 | default stack size on my amd64 box. |
| 07:46:51 | ice799 | gonna be here for everrr |
| 07:56:05 | ddub | whats up ice799? |
| 07:56:20 | ddub | 8MB isn't that off for a default stack size |
| 07:56:43 | ddub | (the kernel may only allocate/map that into the process a page at a time) |
| 08:05:35 | ice799 | wrong channel, my bad. |
| 08:05:43 | ice799 | i'm talking about thread stack size, though. |
| 08:08:21 | ice799 | but, otherwise, ddub you are correct. the kernel maps memory into the process to extend the "main thread" stack when a page fault happens. |
| 08:18:22 | boyscout | Imported iseq and compiled_file to c-ng dir. - 2f02f9d - Brian Ford |
| 08:18:22 | boyscout | Fixed getting Symbols in melbourne under MRI. - 0cc2831 - Brian Ford |
| 08:18:22 | boyscout | More MRI bridging for c-ng. - 575dfad - Brian Ford |
| 08:18:22 | boyscout | Don't use uninitialize line variable (thanks ddub). - 430eaf7 - Brian Ford |
| 08:20:56 | boyscout | CI: Build 430eaf7 failed. http://ci.rubini.us/rubinius/builds/430eaf74d3dda168d0a941311a55bd43736bbb56 |
| 15:20:48 | ddub | morning |
| 16:59:00 | brixen | morning |
| 17:07:41 | stesla | imajes: pong (you pinged last night) |
| 17:08:19 | imajes | stesla: yeah, about hoptoad/toadhopper i think |
| 17:10:44 | evan | morning! |
| 17:10:50 | evan | tries to start out the day upbeat |
| 17:11:06 | imajes | GOOD MORNING BIG BEAR! |
| 17:13:22 | evan | hehehe |
| 17:15:32 | dbussink | evan: howdy how! |
| 17:19:29 | rue | Evening |
| 17:20:05 | evan | hi! |
| 17:23:56 | Zoxc | does rubinius still use a spaghetti stack? |
| 17:24:00 | evan | no |
| 17:27:20 | stesla | imajes: Ahh, still have questions? |
| 17:27:29 | boyscout | CI: 430eaf7 success. 3003 files, 11767 examples, 35776 expectations, 0 failures, 0 errors |
| 17:28:07 | rue | Waait, how did it succeed now? |
| 17:28:10 | imajes | stesla: ah, no. i was starting to make an adapter for ramaze, but managed to get rack_hoptoad working instead |
| 17:28:45 | stesla | Aha. Cool. |
| 17:31:31 | imajes | stesla: yeah, went with the simplest-possible-thing-that-works rule |
| 17:33:01 | stesla | imajes: Good way to do it. I ended up with Toadhopper becuase some of the other devs at EY said to check it out, and I was able to pretty quickly patch it to do what I wanted. |
| 17:34:17 | imajes | yeah, i got close with ramaze but really, it was more effort than i wanted already |
| 18:04:28 | evan | wow |
| 18:04:30 | evan | for some reason |
| 18:04:41 | evan | the password reset for the rubinius twitter account finally come through |
| 18:04:42 | evan | like |
| 18:04:47 | evan | 4 weeks after I requested it. |
| 18:04:53 | brixen | heh, crazy |
| 18:04:56 | brixen | but yay! |
| 18:04:58 | brixen | finally |
| 18:05:24 | brixen | we all know twitter's got queue issues ;) |
| 18:05:53 | evan | heh |
| 18:06:13 | brixen | don't go posting all your tweets at once :) |
| 18:06:26 | evan | heh |
| 18:07:08 | imajes | ah twitter. |
| 18:07:15 | imajes | where not everything is eventually consistent |
| 18:09:29 | brixen | imajes: pioneering a new tech: eventually asymptotically consistent with a constant factor |
| 18:10:36 | brixen | k-EAC (in honor of k-CFA) |
| 18:12:49 | imajes | heh |
| 18:17:04 | evan | i'm rewriting most of our class variable support |
| 18:17:07 | evan | and i'm adding |
| 18:17:16 | evan | class_variable_get_or_set(sym, val) |
| 18:17:22 | evan | to use in the |
| 18:17:25 | evan | @@blah ||= 3 |
| 18:17:26 | evan | case. |
| 18:20:34 | evan | nicely, now that the JIT is humming along |
| 18:20:51 | evan | a monomorphic primitive is quite efficient |
| 18:26:25 | evan | wow, my C++ is getting... interesting. |
| 18:27:02 | evan | class thing : public Parent { virtual bool ok_p(...) { ... } } match; |
| 18:27:51 | evan | C++ could really have used anonymous subclasses |
| 18:30:41 | evan | something wierd I found out about yesterday |
| 18:30:44 | evan | haskell has no backtraces |
| 18:35:40 | brixen | it does try pretty hard to tell you exactly why you fucked up though |
| 19:31:19 | evan | I found a nice syntax hole |
| 19:31:22 | evan | that we can play with |
| 19:31:28 | evan | +@name = whatever |
| 19:32:48 | brixen | hmm |
| 19:33:18 | evan | I was actually thinking about |
| 19:33:23 | evan | +@@whatever |
| 19:33:29 | evan | for more sane class variables |
| 19:34:31 | brixen | sane how? |
| 19:34:50 | evan | not searching upwards |
| 19:35:03 | evan | so strictly shared between instances of that specific class |
| 19:35:40 | brixen | so like class instance vars with easy syntax from an instance? |
| 19:36:10 | evan | right |
| 19:36:20 | brixen | yeah, that'd be cool |
| 20:42:44 | boyscout | Rename rb_intern used in melbourne to avoid confusion. - ad69def - Brian Ford |
| 20:42:44 | boyscout | Always build melbourne with debug symbols. - c888a85 - Brian Ford |
| 20:42:44 | boyscout | Don't segfault MRI when quarks go wild. - 233fc9b - Brian Ford |
| 20:42:44 | boyscout | Melbourne must NOT use any MRI symbol machinery (any, ever, ever). - 2b0a4e9 - Brian Ford |
| 20:42:44 | boyscout | Name BlockPass body consistent with Iter, For. - 2a3ebd6 - Brian Ford |
| 20:43:04 | brixen | jj``: super is a keyword |
| 20:43:10 | brixen | it's handled in the compiler |
| 20:43:45 | brixen | heads to lunch |
| 20:43:53 | brixen | boyscout: be good while I'm away |
| 20:45:44 | boyscout | CI: Build 2a3ebd6 failed. http://ci.rubini.us/rubinius/builds/2a3ebd6d42f1eab83323bcab5f447ac34971e988 |
| 20:55:12 | dbussink | boyscout is good at generating failures lately |
| 20:55:31 | dbussink | evan: does it already help to identify issues? |
| 20:58:24 | evan | yep |
| 21:05:01 | dbussink | evan: seems like capi is most often to blame |
| 21:05:13 | evan | in those cases, yeah. |
| 21:05:24 | evan | i'm planning on trying to debug it more today |
| 22:25:34 | evan | ARG |
| 22:25:47 | evan | spent 20 minutes debugging a "bug" which is really just a bug in rubygems |
| 22:25:58 | evan | because it transports Marshal'd objects |
| 22:26:03 | evan | and they changed the name of an ivar |
| 22:31:21 | imajes | evan: i beat you. |
| 22:31:36 | imajes | evan: i just spent an hour debugging why capistrano wasn't working, because textmate had me editing the WRONG deploy.rb |
| 22:31:46 | evan | ha. |
| 22:32:08 | imajes | that app needs to be fixed, and badly |