00:00:00  * ircretaryquit (Remote host closed the connection)
00:00:07  * sblomquit (Read error: Connection reset by peer)
00:00:08  * ircretaryjoined
00:01:06  * quijotequit (Ping timeout: 276 seconds)
00:31:20  * zz_karupachanged nick to karupa
00:44:17  * daviddiasjoined
00:45:53  * indexzerojoined
00:56:36  * quijotejoined
00:58:55  * janjongboomjoined
00:59:20  * mrvisserjoined
01:01:04  * quijotequit (Ping timeout: 265 seconds)
01:07:24  * mrvisserquit (Ping timeout: 276 seconds)
01:12:52  * kellabyte_joined
01:13:26  * yunongquit (Ping timeout: 255 seconds)
01:25:46  * cosnisjoined
01:28:55  * kazuponjoined
01:30:33  * cosnisquit (Ping timeout: 265 seconds)
01:31:16  * c4milojoined
01:34:10  * avalanche123quit (Remote host closed the connection)
01:34:37  * avalanche123joined
01:35:56  * c4miloquit (Ping timeout: 255 seconds)
01:39:05  * avalanche123quit (Ping timeout: 255 seconds)
01:49:20  * c4milojoined
01:50:55  * vtjnashjoined
01:54:15  * daviddiasquit (Remote host closed the connection)
01:54:47  * daviddiasjoined
01:57:11  * quijotejoined
01:58:06  <MI6>joyent/node: cjihrig master * c7b0203 : fs: close file if fstat() fails in readFile() - http://git.io/QtbL2g
01:58:42  <MI6>joyent/node: cjihrig v0.10 * 72cc66e : fs: close file if fstat() fails in readFile() - http://git.io/OdqQ_w
01:59:02  * daviddiasquit (Ping timeout: 245 seconds)
01:59:24  <MI6>joyent/node: cjihrig v0.8 * b93a51e : fs: close file if fstat() fails in readFile() - http://git.io/OZL9Ew
02:01:58  * quijotequit (Ping timeout: 265 seconds)
02:05:50  * daviddiasjoined
02:05:57  * daviddiasquit (Remote host closed the connection)
02:06:24  * daviddiasjoined
02:08:25  * avalanche123joined
02:10:43  * TooTallNatequit (Quit: Computer has gone to sleep.)
02:10:59  * daviddiasquit (Ping timeout: 252 seconds)
02:12:53  * avalanche123quit (Ping timeout: 264 seconds)
02:24:15  * rossk_quit
02:29:06  * AvianFluquit (Read error: Connection reset by peer)
02:33:49  * AvianFlujoined
02:34:01  * daviddiasjoined
02:34:16  * kazuponquit (Remote host closed the connection)
02:35:52  * kazupon_joined
02:38:51  * daviddiasquit (Remote host closed the connection)
02:38:57  * daviddia_joined
02:53:36  * TooTallNatejoined
02:56:45  * avalanche123joined
02:58:02  * quijotejoined
02:58:19  * parshap_changed nick to parshap
02:59:00  * daviddia_quit (Remote host closed the connection)
02:59:31  * daviddiasjoined
03:01:08  * daviddiasquit (Remote host closed the connection)
03:01:34  * daviddiasjoined
03:03:06  * quijotequit (Ping timeout: 276 seconds)
03:04:00  * daviddiasquit (Read error: Connection reset by peer)
03:04:09  * daviddiasjoined
03:05:07  * daviddiasquit (Remote host closed the connection)
03:05:35  * daviddiasjoined
03:07:18  * avalanche123quit (Remote host closed the connection)
03:07:44  * avalanche123joined
03:10:01  * daviddiasquit (Ping timeout: 252 seconds)
03:11:58  * avalanche123quit (Ping timeout: 240 seconds)
03:47:37  * brsonquit (Quit: leaving)
03:48:16  * hueniversequit (Ping timeout: 258 seconds)
03:50:52  * kazupon_quit (Remote host closed the connection)
03:54:42  * c4miloquit (Remote host closed the connection)
03:58:55  * quijotejoined
03:58:58  * avalanche123joined
04:00:51  * avalanche123quit (Remote host closed the connection)
04:00:58  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
04:01:19  * avalanche123joined
04:03:11  * quijotequit (Ping timeout: 252 seconds)
04:05:15  * daviddiasjoined
04:05:39  * avalanche123quit (Ping timeout: 252 seconds)
04:06:15  * kellabyte_quit (Quit: Connection closed for inactivity)
04:06:20  * daviddiasquit (Remote host closed the connection)
04:06:46  * daviddiasjoined
04:11:16  * daviddiasquit (Ping timeout: 258 seconds)
04:15:59  * sinclair|workquit (Remote host closed the connection)
04:19:10  * daviddiasjoined
04:21:26  * cosnisjoined
04:29:03  * m76joined
04:29:15  * kazuponjoined
04:29:18  * avalanche123joined
04:34:18  * kazuponquit (Ping timeout: 240 seconds)
04:40:19  * cosnis_joined
04:40:30  * cosnisquit (Ping timeout: 265 seconds)
04:43:24  * avalanche123quit (Remote host closed the connection)
04:43:51  * avalanche123joined
04:45:02  * piscisaureus___joined
04:45:09  * piscisaureus___changed nick to piscisaureus
04:47:58  * avalanche123quit (Ping timeout: 240 seconds)
04:48:17  * piscisaureus__quit (Ping timeout: 264 seconds)
04:50:01  * kazuponjoined
04:59:30  * quijotejoined
05:01:15  * cosnis_quit (Quit: Textual IRC Client: www.textualapp.com)
05:04:10  * quijotequit (Ping timeout: 258 seconds)
05:11:36  * inolenquit (Quit: Leaving.)
05:18:53  * piscisaureusquit (Ping timeout: 264 seconds)
05:20:33  * inolenjoined
05:38:04  * c4milojoined
05:42:53  * c4miloquit (Ping timeout: 258 seconds)
05:47:41  * TooTallNatequit (Quit: Computer has gone to sleep.)
05:50:29  * c4milojoined
05:53:27  * indexzeroquit (Quit: indexzero)
06:00:16  * quijotejoined
06:02:13  * sinclair|workjoined
06:04:55  * quijotequit (Ping timeout: 252 seconds)
06:08:41  * daviddiasquit (Remote host closed the connection)
06:09:16  * daviddiasjoined
06:11:41  * c4miloquit (Read error: Connection reset by peer)
06:11:56  * c4milojoined
06:13:24  * daviddiasquit (Ping timeout: 240 seconds)
06:27:27  * bajtosjoined
06:33:45  * avalanche123joined
06:36:14  * vtjnashquit (Remote host closed the connection)
06:39:46  * avalanche123quit (Remote host closed the connection)
06:49:06  * vtjnashjoined
06:50:19  * vtjnashquit (Remote host closed the connection)
06:50:51  * vtjnashjoined
06:51:35  * vtjnashquit (Read error: Connection reset by peer)
06:51:41  * vtjnash_joined
06:52:00  * vtjnash_quit (Remote host closed the connection)
06:52:36  * vtjnashjoined
06:55:11  * AvianFluquit (Read error: Connection reset by peer)
06:55:49  * quijotejoined
06:55:54  * AvianFlujoined
06:56:43  * c4milo_joined
06:56:53  * vtjnashquit (Ping timeout: 252 seconds)
06:57:36  * c4miloquit (Read error: Connection reset by peer)
06:58:26  * rendarjoined
07:05:06  * AvianFlu_joined
07:07:29  * AvianFluquit (Ping timeout: 264 seconds)
07:14:09  * zoinkjoined
07:15:31  <zoink>hi. does libuv do deferred signal handling?
07:19:29  <zoink>like, if you register a signal handler with uv_signal_start, is it actually called from the event loop, or from the "real" signal handler called by the kernel?
07:25:03  * c4milo_quit (Read error: Connection reset by peer)
07:25:54  * AvianFlu_quit (Read error: Connection reset by peer)
07:26:23  * c4milojoined
07:26:53  * AvianFlujoined
07:37:24  * quijotequit (Ping timeout: 265 seconds)
07:39:52  <roxlu>hey indutny
07:40:32  <roxlu>buf.base is what is being send, I meant a "data" member like other types have. (I do have a work around, but I wass curious why it was removed)
08:07:21  * AvianFlu_joined
08:08:33  * AvianFluquit (Ping timeout: 258 seconds)
08:14:19  * c4milo_joined
08:14:46  * c4miloquit (Read error: Connection reset by peer)
08:18:16  * AvianFlujoined
08:19:38  * AvianFlu_quit (Ping timeout: 240 seconds)
08:19:49  * quijotejoined
08:27:05  * quijotequit (Ping timeout: 252 seconds)
08:34:32  * Left_Turnjoined
08:35:41  * quijotejoined
08:42:33  * SMD1joined
08:42:35  * weoxjoined
08:43:02  * SMD1part
08:43:21  * SMD2joined
08:44:05  * c4milo_quit (Remote host closed the connection)
08:44:37  * c4milojoined
08:47:49  <SMD2>Hello.
08:47:49  <SMD2>Need some help with my c++ module to nodejs using libuv: I want to use node's event loop to call my callbacks.
08:47:50  <SMD2>sort of: uv_fs_write(uv_default_loop(), &write_req, open_req.result, cmd_buffer, cmd_count, -1, My_Callback);
08:47:50  <SMD2>The problem is that My_Callback is not called despite that uv_fs_write was.
08:48:50  * weoxquit (Ping timeout: 255 seconds)
08:48:58  * c4miloquit (Ping timeout: 240 seconds)
08:49:13  * weoxjoined
08:49:13  * weoxquit (Changing host)
08:49:13  * weoxjoined
08:51:46  * AvianFluquit (Read error: Connection reset by peer)
08:52:42  * AvianFlujoined
08:56:34  * weoxpart ("ERC Version 5.3 (IRC client for Emacs)")
08:56:40  * thlorenzjoined
09:01:00  * weoxjoined
09:09:28  <indutny>roxlu: ah, ok
09:09:33  <indutny>roxlu: hello
09:09:47  <indutny>roxlu: so this was done to simplify our internal code
09:09:55  <indutny>making it easy converting uv_buf_to to iov
09:10:20  <indutny>roxlu: you should really consider it as just a data
09:10:21  * thlorenzquit (Ping timeout: 252 seconds)
09:10:31  <indutny>not as a regular libuv thing that could have ->data field
09:11:01  <indutny>SMD2: Could you paste the rest of the code somewhere?
09:14:44  <SMD2>indutny: http://pastebin.com/SesP3mAh
09:14:44  <SMD2>This is part of my .so lib, which instance is createn in my nodejs addon.
09:14:44  <SMD2>http://pastebin.com/cnSErHKD - addon's code
09:15:23  <indutny>hm...
09:15:49  <SMD2>indutny: i'm interested in function Commit: it should write to file, then read, then call callback.
09:15:49  <SMD2>I'm wondering should I call uv_run(uv_default_loop()) or not
09:16:07  <indutny>you shouldn't
09:16:13  <indutny>I mean it is running in one place
09:16:34  <indutny>and it is in node internals
09:17:02  <SMD2>Yup, I thought that way but without uv_run() callbacks were not called.
09:19:06  <indutny>how is it getting called from node?
09:21:54  <SMD2>https://github.com/spacemonkeydelivers/libgt2d/blob/master/nodejs/myobject.cc
09:21:55  <SMD2>>tpl->PrototypeTemplate()->Set(v8::String::NewSymbol("commit"), v8::FunctionTemplate::New(Commit)->GetFunction..
09:21:55  <SMD2>and then in node's console:
09:21:55  <SMD2>var addon = require('./gt2d'); var obj = new addon.MyObject(); obj.commit(function(msg){})
09:22:13  <SMD2>i'll upload to github the latest sources now
09:22:43  <indutny>one second
09:22:46  <indutny>let me take a look
09:23:10  <indutny>ok
09:24:55  <SMD2>I've just uploaded the newest version
09:29:20  <indutny>SMD2: is there any possibility that the destructor is called on your class instance
09:29:24  <indutny>before calling the callback?
09:33:12  <SMD2>I dont' think so
09:34:52  * quijotequit (Ping timeout: 245 seconds)
09:41:25  <SMD2>indutny: even if i call only start_lib() no callback called
09:47:13  <indutny>hm...
09:53:23  <SMD2>oh, seems like I've found something:
09:53:23  <SMD2>If I run all the commands in the node's shell, everything works, but if I start nodejs with a file which contains i.e. "start_lib() stop_lib()" I get segfault. Seems like in the second case stop_lib() runs as a first one.
09:55:24  <SMD2>Or I may be wrong.
09:55:27  <indutny>huh?
09:55:49  <indutny>where is a Start_Lib declaration?
09:56:00  <SMD2>https://github.com/spacemonkeydelivers/libgt2d/blob/master/libgt2d.cpp
09:56:22  <SMD2>Sorry, wrong file
09:56:28  <SMD2>https://github.com/spacemonkeydelivers/libgt2d/blob/master/gt2d_init.cpp
09:57:24  <indutny>well
09:57:30  <indutny>open_req does not containt result
09:57:32  <indutny>contain*
09:57:34  <indutny>at this time
09:59:02  * janjongboomjoined
10:00:19  <SMD2>seems like it contains.
10:00:19  <SMD2>Start_Lib is called, Stop_Lib is called and it shows, that open_req.result is valid, but none of callbacks(On_Start, On_Stop) called.
10:00:19  <SMD2>If I write all commands in the shell, everything works fine %)
10:00:19  <SMD2>It also works if start node with a file containing "start_lib()", in that case callback is called. But if file would contain "start_lib(), stop_lib()" - segfault
10:05:43  * bajtosquit (Quit: bajtos)
10:10:38  <indutny>do you have a backtrace?
10:12:15  <SMD2>I'm looking how can I debug it with gdb
10:12:33  <SMD2>The target system is armv6
10:17:32  <indutny>you could get core dumps with it
10:17:34  <indutny>I think
10:17:38  <indutny>ulimit -c unlimited
10:17:45  <indutny>sorry
10:17:48  <indutny>gtg
10:21:19  <SMD2>Thank you a lot
10:21:41  * Kakerajoined
10:32:56  * quijotejoined
10:34:49  * quijote_joined
10:34:53  * quijotequit (Read error: Connection reset by peer)
10:37:57  * mrvisserjoined
10:38:15  * hzjoined
10:39:35  * quijote_quit (Ping timeout: 258 seconds)
10:47:24  * quijotejoined
11:13:14  * weoxpart ("ERC Version 5.3 (IRC client for Emacs)")
11:31:32  * quijotequit (Ping timeout: 245 seconds)
11:40:59  * thlorenzjoined
11:44:14  * mrvisserquit (Remote host closed the connection)
11:44:45  * thlorenzquit (Remote host closed the connection)
11:53:45  * quijotejoined
12:01:06  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
12:02:21  * kellabyte_joined
12:08:07  * mrvisserjoined
12:22:11  * kazuponquit (Remote host closed the connection)
12:25:28  * quijotequit (Ping timeout: 265 seconds)
12:27:41  * SMD2quit (Ping timeout: 258 seconds)
12:30:33  * Left_Turnquit (Ping timeout: 252 seconds)
12:33:54  * karupachanged nick to zz_karupa
12:52:58  <roxlu>thanks indutny
12:59:49  * bajtosjoined
13:03:24  * quijotejoined
13:22:19  * Left_Turnjoined
13:29:45  * quijotequit (Read error: Connection reset by peer)
13:30:18  * quijotejoined
13:31:29  * Kakeraquit (Ping timeout: 264 seconds)
13:45:41  * Kakerajoined
13:47:14  <kellabyte_>anyone know why on windows when I do this part of my code (used this as an example) https://github.com/joyent/libuv/blob/master/test/benchmark-multi-accept.c#L269 I would run into this assert? https://github.com/joyent/libuv/blob/master/src/win/stream.c#L42. it's not clear to me why this works on linux/osx but not windows
13:47:56  <kellabyte_>I also don't understand where stream->type gets set in the multi-accept example, so not sure where to hunt
13:59:33  * stephankquit (Quit: *Poof!*)
14:03:18  <rendar>kellabyte_: apparently uv_listen() returns an integer different than 0, what code does it return?
14:07:16  * m76quit (Read error: Connection reset by peer)
14:09:08  * euoiajoined
14:10:51  * daviddiasjoined
14:13:08  * stephankjoined
14:14:09  * kazuponjoined
14:19:38  * bajtosquit (Quit: bajtos)
14:21:23  * daviddiasquit (Ping timeout: 255 seconds)
14:23:31  * daviddiasjoined
14:23:50  * bajtosjoined
14:30:33  * daviddiasquit (Remote host closed the connection)
14:31:07  * daviddiasjoined
14:35:09  * zz_karupaquit (*.net *.split)
14:35:35  * daviddiasquit (Ping timeout: 252 seconds)
14:37:07  * zz_karupajoined
14:38:15  * zz_karupachanged nick to karupa
14:39:21  * kazuponquit (Remote host closed the connection)
14:43:05  <kellabyte_>in debug mode it dies on the assert on line 42 that I linked, in release mode rc = -4071
14:43:34  <kellabyte_>which is defined as UV__EINVAL (-4071) in libuv
14:44:31  <kellabyte_>rendar: I'm not sure how in the multi-accept example that stream type gets set, so it does work on linux/osx but not quite sure how. I'm probably misunderstanding something
14:54:17  * quijotequit (Ping timeout: 252 seconds)
14:56:11  * avalanche123joined
15:05:19  * quijotejoined
15:06:36  * mrvisserquit (Remote host closed the connection)
15:10:10  <rendar>kellabyte_: you're right, UV__EINVAL means something is set wrongly
15:10:20  <rendar>i guess
15:11:58  <kellabyte_>Yeah. My code is based on the multi-accept and not much is changed and it works on Unix and OSX but throws that assert on windows. Not sure why yet lol
15:12:15  <kellabyte_>But I'm not sure where the stream type is set in the first place
15:13:18  * euoiaquit (Ping timeout: 240 seconds)
15:13:44  * kazuponjoined
15:14:34  <kellabyte_>The stream on windows seems to be set to unknown
15:27:59  * euoiajoined
15:33:31  <rendar>kellabyte_: hmm, sorry dunno that :(
15:39:59  <roxlu>kellabyte_: when I saw your post on performance and malloc I was wondering why you'r using malloc at all? Why not use a preallocated memory buffer?
15:42:06  * mrvisserjoined
15:43:24  <tjfontaine>yawn.
15:45:16  <kellabyte_>roxlu: yeah, that's my next step, I was keeping malloc() around so I could learn some tricks
15:45:35  <kellabyte_>morning tjfontaine :)
15:45:39  <roxlu>ah got it! that will probably improve performance a lot
15:46:37  <kellabyte_>hard to tell, there are very few mallocs happening now, I don't think it'll make a huge dent in the cpu usage
15:46:54  <kellabyte_>but probably will help latency like the last change did
15:47:53  <roxlu>I'm curious to the benchmarks, it's one of the things I check first when optimizing
15:48:02  <kellabyte_>yeah same
15:48:20  <roxlu>though I work with video/audio data which is a bit different in the way it's used
15:53:41  <tjfontaine>kellabyte_: can it be evening again?
15:53:45  <tjfontaine>I'm sure it is somewhere
15:55:39  <kellabyte_>lol
15:55:48  <kellabyte_>I'm good with that, I declare it evening!
15:56:20  <tjfontaine>sold.
16:02:07  <tjfontaine>indutny: you travelling?
16:05:17  * kazuponquit
16:05:36  <kellabyte_>I find this multi-accept stuff hard to follow and debug
16:06:10  <tjfontaine>which multi accept stuff
16:06:33  * kazuponjoined
16:07:06  <kellabyte_>do you have scroll buffer from a little before you yawned? :P I was talking about it just a bit ago, if not I can relink!
16:07:51  <tjfontaine>oh the benchmark
16:08:44  <kellabyte_>yeah, bnorris pointed me to it over a year ago as an example how to load balance across cores
16:08:56  <tjfontaine>right
16:08:57  <kellabyte_>so I've updated libuv but I'm getting that assert on windows (not on linux or osx)
16:09:51  <kellabyte_>it would be nice if it were easier to pass connections to new event loops though rather than going through all this IPC stuff
16:10:25  <tjfontaine>ya, we've talked about it before, we really need the ability to pass a uv_handle_t to a new event loop
16:10:31  <kellabyte_>yeah
16:10:40  <kellabyte_>can I log a feature request?
16:10:50  <tjfontaine>absolutely. issues are always ok :)
16:11:18  <tjfontaine>from an api perspective I don't think this is actually a problem I think it's just a difference between unix and windows and this benchmark probably isn't run often
16:11:31  <tjfontaine>oh wait, this is listen not accept
16:11:37  * rosskjoined
16:11:42  <kellabyte_>the error might be in my code somewhere
16:11:56  <tjfontaine>kellabyte_: you definitely want to cache the return value and inspect it
16:12:01  <tjfontaine>it will be quite helpful here
16:12:08  <kellabyte_>return value from what?
16:13:03  <kellabyte_>it's landing on the assert on line 42 of stream.c
16:13:18  <kellabyte_>basically stream->type is unknown, so it asserts
16:13:28  * SMD2joined
16:13:40  <tjfontaine>oh I thought you were saying the assert from uv_listen in the multi-accept was a problem
16:14:18  <kellabyte_>I haven't run that benchmark on windows yet to see if that code suffers from the same problem as mine
16:14:50  * piscisaureusjoined
16:14:51  <kellabyte_>I'm blaming my code first, maybe I'm losing a pointer somewhere, but I'm not sure where stream->type should be set in that benchmark
16:16:32  <tjfontaine>should probablye be a UV_PIPE or UV_TCP
16:16:45  <kellabyte_>yeah, it's not, its unknown
16:17:09  <kellabyte_>but where in that multi-accept benchmark would it be set to UV_PIPE or UV_TCP?
16:17:26  <tjfontaine>uv_*_init
16:18:13  <tjfontaine>looks like server_handle is init'd in ipc_read_cb, which maybe hasn't been called yet?
16:19:39  * yunongjoined
16:19:44  <kellabyte_>why would it do it on a read callback?
16:20:03  <tjfontaine>because that's when it's being sent the server it should be calling listen on
16:20:35  <tjfontaine>the master creates the handles, and sends via ipc the server handle to the child so the child can call accept on it
16:22:37  <kellabyte_>but from how I read it, server_cb() happens first from the spawned threads
16:22:47  * quijotequit (Ping timeout: 265 seconds)
16:24:19  <tjfontaine>oh I missed get_listen_handle
16:25:01  <tjfontaine>so it does do server_cb but that calls get_listen_handle which sets up the ipc and tries to prime the ipc between them
16:25:12  <tjfontaine>which hopefully results in ipc_cb happening
16:26:11  <kellabyte_>oh right, good catch. is this a potential race condition?
16:26:59  <kellabyte_>or is uv_pipe_connect() and the ipc_cb() going to be synchronous before uv_listen() gets called?
16:28:17  <tjfontaine>it's definitely a race
16:28:29  <tjfontaine>well, not an unexpected race
16:28:41  <tjfontaine>I bet this benchmark can be flakey as is on windows for us as well
16:28:49  * SMD3joined
16:30:43  * SMD2quit (Ping timeout: 258 seconds)
16:30:46  <kellabyte_>ok, so I'm not insane then :P
16:31:24  <tjfontaine>no, presumably you already have access to a windows machine to test this benchmark as is?
16:32:25  * m76joined
16:33:22  <kellabyte_>yeah
16:33:57  <kellabyte_>will that assert surface if I run the executable?
16:34:59  * quijotejoined
16:37:39  <kellabyte_>I'm running run-benchmarks.exe now
16:38:27  * c4milo_joined
16:40:57  <kellabyte_>logged the feature request: https://github.com/joyent/libuv/issues/1301
16:43:05  <tjfontaine>thanks
16:43:24  <kellabyte_>oh yeah, benchmark fails with asserts, but different asserts than my code
16:43:28  <kellabyte_>but I guess timing could be different
16:43:36  <SMD3>Hey guys, need some advice.
16:43:36  <SMD3>I write c++ lib using libuv. So I need for example open file and then close it asyncronycally. So i call uv_fs_open(..., On_Open) and then uv_fs_close(..., On_Close) and then uv_run(loop).
16:43:36  <SMD3>So the problem is that On_Start callback isn't called before uv_fs_close() function. Is there any possibility to add some sleep or wait to get On_Open callback to be called.
16:43:36  <SMD3>If I do sort of uv_fs_open(On_start(uv_fs_close(on_stop))), it works fine.
16:43:59  <kellabyte_>also pipe_pump100_client fails
16:44:05  * c4milo_quit (Ping timeout: 264 seconds)
16:44:24  * avalanche123quit (Remote host closed the connection)
16:44:51  * avalanche123joined
16:46:24  * mcavagejoined
16:49:09  * avalanche123quit (Ping timeout: 252 seconds)
16:49:56  <zoink>if you register a signal handler with uv_signal_start, is it called from the "real" signal handler that the kernel knows about, or is it actually called from the event loop?
16:50:10  * quijotequit (Read error: Connection reset by peer)
16:50:20  * brsonjoined
16:50:47  * quijotejoined
16:52:47  * c4milojoined
16:56:42  <tjfontaine>zoink: they're proxied over to the main thread
17:04:31  * avalanche123joined
17:05:08  <kellabyte_>tjfontaine: should I log the bench assertion failures too?
17:07:44  * kazuponquit (Remote host closed the connection)
17:15:00  * AvianFluquit (Remote host closed the connection)
17:16:18  * AvianFlujoined
17:21:48  * quijotequit (Ping timeout: 240 seconds)
17:24:45  * TooTallNatejoined
17:25:16  <Ralith>zoink: it's called from the event loop; one of the reasons to use that sort of thing, after all, is to work around the difficult concurrency problem signals usually present
17:29:40  * quijotejoined
17:37:43  <kellabyte_>logged the multi-accept failures on windows: https://github.com/joyent/libuv/issues/1302
17:39:20  * sblomjoined
17:40:17  * Ralithquit (Ping timeout: 255 seconds)
17:46:18  * quijotequit (Ping timeout: 240 seconds)
17:50:11  * AvianFluquit (Remote host closed the connection)
17:54:08  * yunongquit (Remote host closed the connection)
17:54:26  * yunongjoined
17:54:30  * yunongquit (Remote host closed the connection)
17:55:05  * yunongjoined
18:00:06  * yunongquit (Ping timeout: 276 seconds)
18:12:09  * sh1mmerjoined
18:15:38  * sblomquit (Read error: Connection reset by peer)
18:17:26  * Ralithjoined
18:29:59  * quijotejoined
18:32:08  * avalanche123quit (Remote host closed the connection)
18:32:37  * avalanche123joined
18:33:38  * yunongjoined
18:36:47  * mrvisserquit (Remote host closed the connection)
18:36:59  * avalanche123quit (Ping timeout: 255 seconds)
18:42:44  * AvianFlujoined
18:43:22  * AvianFluquit (Client Quit)
18:43:52  * AvianFlujoined
18:45:17  * avalanche123joined
18:47:50  * thlorenzjoined
18:53:57  * sh1mmerquit (Quit: sh1mmer)
18:56:00  * rendarquit (Ping timeout: 258 seconds)
18:57:23  * rendarjoined
18:59:11  * bajtosquit (Quit: bajtos)
19:01:03  * thlorenzquit (Ping timeout: 252 seconds)
19:06:54  <tjfontaine>kellabyte_: yes and thank you
19:08:55  <kellabyte_>tjfontaine: thank you for the help :) I was pretty confused for a couple days
19:09:48  * quijotequit (Ping timeout: 258 seconds)
19:12:27  <tjfontaine>kellabyte_: yup, no problem, I have a pretty good idea what could be the change that made it more racey -- the arch there should be reasonably sound though, I just think you need to be a bit more crisp and tolerant of when things happen
19:13:01  <tjfontaine>kellabyte_: this is essentially how node's clustering works, but across processes not threads (fine fine the OS can represent that in any way it likes)
19:13:20  <kellabyte_>right
19:13:55  <tjfontaine>bbiab lunch
19:14:28  <kellabyte_>yeah, it just seems like a lot of work when everything is in process and hard to follow
19:14:38  <kellabyte_>enjoy lunch :)
19:17:22  <nathan7>cluster should be able to work within the same process, no?
19:17:25  <nathan7>in theory
19:17:35  <nathan7>now that we're passing isolates around everywhere and such
19:18:20  <nathan7>could do nice things like share a LevelDB across cluster threads (it's threadsafe but single-process)
19:22:31  * sh1mmerjoined
19:23:28  * avalanche123quit (Remote host closed the connection)
19:24:59  <kellabyte_>I'd rather hand off the handle in a more efficient way
19:28:10  * mrvisserjoined
19:28:13  * avalanch_joined
19:38:58  * TooTallNatequit (Quit: Computer has gone to sleep.)
19:50:50  * c4miloquit (Remote host closed the connection)
19:53:55  * quijotejoined
19:58:21  * quijotequit (Ping timeout: 252 seconds)
20:11:52  * TooTallNatejoined
20:16:07  * avalanch_quit (Remote host closed the connection)
20:18:27  * c4milojoined
20:20:37  <zoink>Ralith: ah, thanks :)
20:28:58  * c4miloquit
20:32:08  * c4milojoined
20:32:42  * avalanche123joined
20:33:34  * dap_quit (Quit: Leaving.)
20:34:00  * dap_joined
20:37:49  * cjbjoined
20:38:55  * thlorenzjoined
20:46:04  * c4miloquit (Remote host closed the connection)
20:50:28  * roxluquit (Quit: leaving)
20:54:48  * quijotejoined
20:57:17  * thlorenzquit (Remote host closed the connection)
20:58:59  * quijotequit (Ping timeout: 240 seconds)
20:59:44  * thlorenzjoined
21:04:21  * thlorenzquit (Remote host closed the connection)
21:05:01  <tjfontaine>nathan7: it can, and should, the example as it exists today though isn't very resilient
21:08:24  <nathan7>tjfontaine: mhm
21:14:48  * sblomjoined
21:17:55  * sblomquit (Read error: Connection reset by peer)
21:31:47  * dap_quit (Quit: Leaving.)
21:37:04  * mrvisserquit (Remote host closed the connection)
21:37:04  * dap_joined
21:40:19  * jmar777joined
21:42:06  * sh1mmer_joined
21:44:17  * sh1mmerquit (Ping timeout: 258 seconds)
21:44:17  * sh1mmer_changed nick to sh1mmer
21:51:24  * thlorenzjoined
21:55:25  * quijotejoined
21:59:13  * thlorenzquit (Remote host closed the connection)
22:00:09  * quijotequit (Ping timeout: 265 seconds)
22:01:20  * thlorenzjoined
22:04:12  * thlorenzquit (Remote host closed the connection)
22:07:45  * rendarquit
22:07:57  * jwilmjoined
22:11:10  <indutny>hey hey hey
22:11:17  <indutny>tjfontaine: yep
22:11:25  <indutny>tjfontaine: currently in Edinburgh
22:14:18  * euoiaquit (Ping timeout: 240 seconds)
22:15:22  <qard>Anyone have a minute to help me figure out how to do something in V8? I'm trying to have a method of one C++ class return an instance of another in JS. See here; https://gist.github.com/Qard/2970e09ae96475c8a41a
22:16:14  <qard>I tried constructor->GetFunction()->NewInstance(0, argv), but the class takes non-v8 values as arguments.
22:17:48  <qard>I basically need a new args.This() pointing to the correct function instance. I'm not sure if you can do that without NewInstance() though.
22:19:37  <tjfontaine>indutny: what days are you in the bay area then?
22:19:48  <indutny>I think June 4th
22:19:55  <indutny>is the first day when I'll get there
22:19:58  <tjfontaine>indutny: ok
22:20:02  <indutny>and will be there until 18th
22:20:08  <tjfontaine>ok
22:26:08  * sh1mmerquit (Quit: sh1mmer)
22:31:02  * sblomjoined
22:31:02  * sh1mmerjoined
22:32:17  * c4milojoined
22:33:20  * AvianFluquit (Remote host closed the connection)
22:34:23  * Left_Turnquit (Read error: Connection reset by peer)
22:37:18  * Left_Turnjoined
22:39:13  * janjongboomjoined
22:39:30  <wrl>hey, i'm having an odd issue on windows
22:39:56  <wrl>i'm using libuv as the event loop in a widget toolkit, and right now my event handling loop is just libuv with a uv_timer_t firing at ~60fps (16ms, repeating)
22:40:32  <wrl>i'm seeing extremely high CPU usage (~50%), which i don't see when i use a SetTimer/WM_TIMER
22:40:41  <wrl>my animation is considerably less smooth with SetTimer however
22:40:53  * m76quit (Read error: Connection reset by peer)
22:43:03  * c4milo_joined
22:43:37  * c4miloquit (Ping timeout: 252 seconds)
22:56:27  * quijotejoined
22:58:41  * sblomquit (Read error: Connection reset by peer)
22:59:25  * rosskquit
23:00:51  * quijotequit (Ping timeout: 252 seconds)
23:13:24  * Kakeraquit (Ping timeout: 276 seconds)
23:34:49  * avalanche123quit (Remote host closed the connection)
23:35:15  * avalanche123joined
23:39:18  * avalanche123quit (Ping timeout: 240 seconds)
23:45:30  * mcavagequit (Remote host closed the connection)
23:52:15  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
23:56:55  * yunongquit (Ping timeout: 258 seconds)
23:56:58  * quijotejoined
23:58:40  * wwicksjoined
23:59:25  * jmar777_joined