00:00:01  * ircretaryquit (Remote host closed the connection)
00:00:08  * ircretaryjoined
00:01:05  * TooTallNatequit (Quit: ["Textual IRC Client: www.textualapp.com"])
00:04:42  * piscisaureus_quit (Ping timeout: 250 seconds)
00:14:50  * piscisaureus_joined
00:25:36  * zot1joined
00:33:44  * dapquit (Quit: Leaving.)
00:35:59  * qmx|awaychanged nick to qmx
00:40:33  * piscisaureus_quit (Ping timeout: 276 seconds)
00:43:31  * kazuponjoined
00:45:17  * zot1quit (Quit: Leaving.)
00:51:57  * qmxchanged nick to qmx|away
00:57:08  * kazuponquit (Remote host closed the connection)
01:04:11  * bnoordhuisjoined
01:05:35  * CoverSlidequit (Ping timeout: 260 seconds)
01:07:13  * CoverSlidejoined
01:08:36  * bnoordhuisquit (Ping timeout: 264 seconds)
01:15:51  * stagasquit (Ping timeout: 260 seconds)
01:39:55  * indexzerojoined
01:43:57  * abraxasjoined
01:54:34  * loladiroquit (Quit: loladiro)
02:01:21  <MI6>joyent/node: Nathan Rajlich v0.10 * 598b5e4 : blog: fix small typo in v0.10.0 release article - http://git.io/_z-5dg
02:07:39  * kazuponjoined
02:13:24  * kazuponquit (Ping timeout: 264 seconds)
02:29:14  <creationix>mikeal: are you around?
02:30:58  <mikeal>yes
02:31:44  <creationix>mikeal: sent you a dm link to the kickstarter draft to see what you think
02:42:12  * hzquit (Ping timeout: 264 seconds)
02:49:06  * hzjoined
03:04:08  * brsonquit (Quit: leaving)
03:05:52  * hzquit
03:25:26  <MI6>nodejs-v0.10: #14 UNSTABLE windows-ia32 (5/559) osx-ia32 (1/559) windows-x64 (5/559) http://jenkins.nodejs.org/job/nodejs-v0.10/14/
03:26:19  <MI6>nodejs-master: #94 UNSTABLE linux-x64 (2/555) windows-ia32 (6/555) smartos-ia32 (28/555) linux-ia32 (2/555) windows-x64 (5/555) http://jenkins.nodejs.org/job/nodejs-master/94/
03:33:01  * mralephquit (Ping timeout: 240 seconds)
03:37:50  * AvianFlujoined
03:38:44  <tjfontaine>man that osx buildslave needs replaced :/
03:46:01  * kazuponjoined
03:46:16  * mralephjoined
03:49:16  * jguerreroquit (Quit: jguerrero)
03:50:00  * nsmjoined
03:54:21  * brsonjoined
03:58:34  * jguerrerojoined
04:15:25  * kazuponquit (Remote host closed the connection)
04:16:41  * perezdquit (Quit: perezd)
04:17:59  * perezdjoined
04:17:59  * kazuponjoined
04:29:18  * loladirojoined
04:42:47  * AvianFluquit (Remote host closed the connection)
04:42:55  * mralephquit (Read error: Operation timed out)
04:58:35  * mikealquit (Quit: Leaving.)
04:59:08  * mikealjoined
05:21:20  * c4miloquit (Remote host closed the connection)
05:47:09  * mralephjoined
06:07:10  * wolfeidauquit (Remote host closed the connection)
06:12:03  * nsmquit (Quit: nsm)
06:15:07  * brsonquit (Quit: leaving)
06:25:49  * AvianFlujoined
06:37:30  * perezdquit (Quit: perezd)
06:52:51  * defunctzombiechanged nick to defunctzombie_zz
07:33:27  * `3rdEdenjoined
07:34:40  * jguerreroquit (Quit: jguerrero)
07:35:17  * rendarjoined
07:41:09  * zotjoined
07:49:22  * stagasjoined
07:52:51  * AvianFluquit (Remote host closed the connection)
08:11:39  * loladiropart
08:28:39  * bnoordhuisjoined
08:33:53  * csaohjoined
08:39:33  * stagasquit (Read error: Connection reset by peer)
08:39:58  * stagasjoined
08:41:50  <bnoordhuis>morning all
08:53:00  * dominictarrquit (Quit: dominictarr)
08:55:15  * mmaleckichanged nick to mmalecki[out[
08:55:17  * mmalecki[out[changed nick to mmalecki[out]
09:01:03  * bnoordhuisquit (Ping timeout: 276 seconds)
09:01:55  * benoitcquit (Excess Flood)
09:05:08  * benoitcjoined
09:08:04  * zotquit (Quit: Leaving.)
09:26:52  * bnoordhuisjoined
09:27:17  * mraleph1joined
09:31:59  * stephank_joined
09:34:16  <indutny>morning
09:35:00  * bnoordhuisquit (Ping timeout: 240 seconds)
09:35:40  * niska`joined
09:35:42  * kazuponquit (Remote host closed the connection)
09:36:21  * mralephquit (*.net *.split)
09:36:21  * stephankquit (*.net *.split)
09:36:21  * philipsquit (*.net *.split)
09:36:21  * niskaquit (*.net *.split)
09:36:22  * stephank_changed nick to stephank
09:43:00  * stagas_joined
09:48:21  * philipsjoined
09:57:07  * kazuponjoined
10:02:15  * stagas_quit (Ping timeout: 260 seconds)
10:07:42  * indexzeroquit (Quit: indexzero)
10:09:54  * piscisaureus_joined
10:10:52  * piscisaureus_quit (Read error: Connection reset by peer)
10:11:10  * piscisaureus_joined
10:28:59  * Nicholas__joined
10:46:05  * benoitcquit (Excess Flood)
10:47:52  <dostoyevsky>src/unix/stream.c:1115: uv__stream_io: Assertion `!(stream->flags & UV_CLOSING)' failed.
10:48:04  <dostoyevsky>I am getting this message when I try to reconnect to a server...
10:48:25  <dostoyevsky>I try to connect again in the callback to uv_close
10:50:14  <indutny>and you're using old socket?
10:50:17  <indutny>uv_stream_t
10:50:20  <indutny>err
10:50:21  <indutny>uv_tcp_t
10:51:01  <dostoyevsky>Yes... can I reinitialize it with uv_tcp_init() ?
10:52:20  <indutny>not from uv_close
10:52:25  <indutny>I guess
10:52:30  <indutny>but let me check
10:53:10  * benoitcjoined
10:53:12  <roxlu>does libuv has a function to get the directory from a filepath ? (so just strips the filename)
10:55:35  <dostoyevsky>roxlu: I don't think so. At least in the test code I've found #ifdefs that work on path strings for WINDOWS/UNIX
10:55:57  <roxlu>ok thanks!
10:59:53  <dostoyevsky>roxlu: You could, however, use libuv in c++ and there boost has something...
11:00:36  <roxlu>hmm boost :)
11:00:49  <roxlu>I'll write my own tiny function i guess
11:02:50  <dostoyevsky>indutny: At least the assert went away when I used uv_tcp_init() ... not sure if I did really reconnect, though.
11:07:21  <dostoyevsky>indutny: Test cases work with uv_tcp_init() now :)
11:07:42  <indutny>ok
11:08:43  * `3rdEdenchanged nick to `3E|OUT
11:12:17  * wolfeidaujoined
11:20:30  * sgallaghjoined
11:29:04  * bnoordhuisjoined
11:30:08  <bnoordhuis>dostoyevsky: you can uv_tcp_init() the handle again in the close callback but not right after calling uv_close()
11:34:26  * kazuponquit (Remote host closed the connection)
11:37:10  * csaohquit (Quit: csaoh)
11:52:43  * benoitcquit (Excess Flood)
11:58:10  * benoitcjoined
12:00:01  * `3E|OUTquit (Remote host closed the connection)
12:01:10  * abraxasquit (Remote host closed the connection)
12:13:00  * rvaggquit (Quit: ta ta)
12:33:22  * kazuponjoined
12:41:04  * csaohjoined
12:45:23  * kazuponquit (Remote host closed the connection)
12:54:42  <indutny>bnoordhuis: https://github.com/joyent/node/pull/5001
12:54:44  <indutny>please review
12:56:53  <bnoordhuis>what do i get in return?
12:57:36  <indutny>bounty
12:58:31  * brianc1quit (Read error: Connection reset by peer)
13:03:23  * benoitcquit (Excess Flood)
13:05:02  * qmx|awaychanged nick to qmx
13:05:45  * mmalecki[out]changed nick to mmalecki
13:06:40  <bnoordhuis>i prefer booty actually
13:07:04  * zot1joined
13:08:40  <indutny>meh
13:09:11  * benoitcjoined
13:15:45  * kazuponjoined
13:25:10  * kazuponquit (Ping timeout: 252 seconds)
13:25:17  <MI6>joyent/node: Fedor Indutny master * 26dae9a : net: handle 'finish' event only after 'connect' - http://git.io/JOCJ-w
13:29:36  * zot1quit (Quit: Leaving.)
13:30:27  * mikealquit (Quit: Leaving.)
13:44:59  * mikealjoined
13:55:09  * bradleymeckquit (Quit: bradleymeck)
13:55:24  * `3E|OUTjoined
13:57:08  * `3E|OUTchanged nick to `3rdEden
13:57:42  * piscisaureus_quit (Ping timeout: 264 seconds)
14:01:24  * abraxasjoined
14:05:44  * abraxasquit (Ping timeout: 246 seconds)
14:08:45  * c4milojoined
14:10:04  * mikealquit (Quit: Leaving.)
14:10:45  <roxlu>indutny: maybe a bit off topic question, but i was wondering what your thoughts on this are:
14:11:39  <roxlu>I'm working on something where need to calculate lots of vertices (opengl) which I want to do once and then load them from a file
14:12:18  <roxlu>I can just write a complete array with binary data to a file and then reload it again, right?
14:14:32  <roxlu>ah nvm.. of course I can do that :#
14:14:48  <roxlu>was just wondering what the fastest solution would be
14:15:36  <bnoordhuis>roxlu: that's not very portable though
14:15:59  <roxlu>bnoordhuis: yeah, I'm focussing on Mac and Windows
14:16:18  <roxlu>I think I won't get any problems with that
14:16:28  <roxlu>... or do I?
14:16:36  <bnoordhuis>depends
14:16:52  <bnoordhuis>sizeof(long) on os x is different depending on the cpu arch
14:17:01  <roxlu>ah ok
14:17:10  <roxlu>I'll generate the cached files per platform
14:17:18  <bnoordhuis>that should work
14:18:52  * piscisaureus_joined
14:19:04  <roxlu>thanks bnoordhuis !
14:19:37  * piscisaureus_quit (Read error: Connection reset by peer)
14:19:50  * piscisaureus_joined
14:20:01  * AvianFlujoined
14:20:03  <bnoordhuis>np :)
14:20:05  * wavdedjoined
14:20:50  <bnoordhuis>indutny: https://github.com/joyent/node/issues/4771#issuecomment-14765987 <- suggestions?
14:20:59  <wavded>can someone help me understand where in libuv are things buffered?
14:21:10  <bnoordhuis>wavded: buffered?
14:21:17  * kazuponjoined
14:21:36  <wavded>are we buffering data at all for any I/O before passing it along back to Node or elsewhere
14:21:41  <MI6>joyent/node: bnoordhuis created branch issue4771-wip - http://git.io/S7N3Xg
14:21:49  <bnoordhuis>indutny: ^ related
14:22:02  <bnoordhuis>wavded: libuv doesn't really buffer
14:22:18  <bnoordhuis>write reqs go into a queue when they can't be written immediately but that's about it
14:22:33  <bnoordhuis>reads are always passed on to the callback immediately
14:23:48  <wavded>ok thx, that is helpful
14:24:45  <wavded>under heavy load (which I know can mean a lot of things), where usually are the bottlenecks if any in libuv (the write queue?)
14:25:20  <bnoordhuis>wavded: well, it all depends on your definition of heavy load
14:25:34  <wavded>hehe, well lets say a LOT of incoming requests for an HTTP server
14:26:15  <bnoordhuis>could be accept(), could be read() or write()
14:26:26  <bnoordhuis>but it's usually the kernel that's the bottleneck, not libuv
14:26:32  * kazuponquit (Ping timeout: 252 seconds)
14:26:41  <bnoordhuis>libuv doesn't really do anything that's computationally expensive
14:27:42  <bnoordhuis>if you profile node with e.g. perf -g, the first uv function is usually at #50 or even further down
14:27:45  <wavded>right, it seems like its always trying to delegate elsewhere
14:27:48  <bnoordhuis>yep
14:31:26  * bradleymeckjoined
14:39:22  * hzjoined
14:40:29  <piscisaureus_>isaacs: hey. When I run npm dedupe and npm ls afterwards, I get unmet dependencies...
14:45:46  * TooTallNatejoined
14:46:24  * benoitcquit (Excess Flood)
14:52:17  * nsmjoined
14:52:17  <bnoordhuis>gdb on os x...
14:52:20  * bnoordhuissighs
14:55:41  * benoitcjoined
14:56:33  * kazuponjoined
14:58:48  <MI6>nodejs-master: #95 UNSTABLE osx-x64 (1/556) windows-ia32 (6/556) windows-x64 (5/556) http://jenkins.nodejs.org/job/nodejs-master/95/
15:03:08  * TooTallNatequit (Remote host closed the connection)
15:21:23  * Nicholas__quit (Ping timeout: 252 seconds)
15:33:03  * jguerrerojoined
15:33:59  * defunctzombie_zzchanged nick to defunctzombie
15:38:50  * bradleymeckquit (Quit: bradleymeck)
15:44:23  * bradleymeckjoined
15:44:58  <MI6>joyent/node: Ben Noordhuis v0.8 * 2c41a80 : crypto: check key type in GetPeerCertificate() Works around the followin - http://git.io/5jCqjw
15:45:04  * c4miloquit (Remote host closed the connection)
15:46:20  * c4milojoined
15:49:42  * mikealjoined
15:49:59  * stagas_joined
15:51:12  * stagasquit (Ping timeout: 276 seconds)
15:51:24  * stagas_changed nick to stagas
15:51:40  * felixgejoined
15:55:59  <tjfontaine>bnoordhuis: use lldb
15:56:44  <bnoordhuis>tjfontaine: i do from time to time but it doesn't have all the features gdb has
15:56:55  <tjfontaine>indeed
15:57:00  <bnoordhuis>i know, i know - it's open source, i should fix that myself
15:58:18  <tjfontaine>meh, I'm not saying that, apple and others are working on it -- the features will come
16:00:51  <bnoordhuis>indutny: https://github.com/joyent/node/issues/4926 <- ideas?
16:01:43  <bnoordhuis>piscisaureus_: uint64_t total = *(uint64_t*)filetime;
16:01:53  <bnoordhuis>^ it's potentially dangerous
16:02:09  <bnoordhuis>don't say i didn't warn you
16:02:55  <tjfontaine>I was going to mention that I just tried to follow what the other macro did :)
16:03:06  <piscisaureus_>bnoordhuis: why
16:03:17  <piscisaureus_>bnoordhuis: also - join the hangout
16:05:58  * dscapequit (Ping timeout: 245 seconds)
16:09:49  * dapjoined
16:11:09  <bnoordhuis>piscisaureus_: FILETIME is a struct, the DWORDs don't have to be consecutive in memory
16:11:26  <bnoordhuis>it's not an issue on x86 or x64 but it will be on e.g. 64 bits arm
16:13:31  * perezdjoined
16:16:22  * piscisaureus_quit (Read error: Connection reset by peer)
16:16:29  * bradleymeckquit (Quit: bradleymeck)
16:17:30  * piscisaureus_joined
16:19:14  * mikealquit (Quit: Leaving.)
16:19:58  * wavdedquit (Quit: Textual IRC Client: www.textualapp.com)
16:39:19  * loladirojoined
16:40:57  * benoitcquit (Excess Flood)
16:45:22  * AvianFluquit (Remote host closed the connection)
16:46:17  <piscisaureus_>bnoordhuis: well the value that comes out of the kernel is actually int64_t that gets cast to DWORD[2]
16:46:34  <piscisaureus_>bnoordhuis: so if you're right windows would be totally broken on arm
16:46:55  <tjfontaine>anyway, I'm pretty sure my math there is wrong :)
16:47:11  * benoitcjoined
16:47:24  <tjfontaine>that value is in 100's of ns?
16:47:29  <piscisaureus_>yes
16:47:41  <piscisaureus_>with epoch being 1600 or something
16:47:56  <piscisaureus_>you should just look at the FILETIME_TO_TIME_T macro, it has the correct math
16:47:57  <tjfontaine>right that's what the magic number subtraction thing is for, as I googled
16:48:17  <piscisaureus_>It seems that you re-implemented what was already in there.
16:48:22  <tjfontaine>yes
16:58:32  <isaacs>good morning
16:58:52  <isaacs>bnoordhuis: what do you think about this? https://gist.github.com/5154078
16:59:06  <isaacs>bnoordhuis: it fixes a crash that i can't seem to figure out how to reproduce reliably
16:59:25  <isaacs>bnoordhuis: but, the wrap is checked for null right after uwrapping, which it can never be if you use UNWRAP, since there's an assertion in there.
16:59:35  * qmxchanged nick to qmx|lunch
17:10:21  * loladiroquit (Quit: loladiro)
17:10:55  <bnoordhuis>piscisaureus_: oh, okay.
17:11:09  * bradleymeckjoined
17:12:31  <bnoordhuis>isaacs: lgtm i guess. that wrap != NULL check was just a defense-in-depth thing
17:14:10  * piscisaureus_quit (Ping timeout: 252 seconds)
17:14:11  <isaacs>bnoordhuis: it can blow the assertion in some race condition where you try to console.log() the handle too early
17:14:26  <isaacs>bnoordhuis: this calls util.inspect, enumerates, boom
17:16:41  <MI6>joyent/node: isaacs v0.10 * 53f2381 : core: Unwrap without aborting in handle fd getter (+1 more commits) - http://git.io/inUJUg
17:20:34  * kazuponquit (Remote host closed the connection)
17:21:03  <isaacs>i think we need some kind of internal "error is about to be emitted" event
17:21:16  <isaacs>or some way to know, in an error event, if there are other handlers besides this one
17:22:14  <isaacs>in src.pipe(dest), if dest has an error, we want to stop writing to it.
17:22:28  <isaacs>however, we want to NOT suppress the throw in that case, which we will if we add a listener.
17:22:47  <isaacs>so, what the listener does, is it checks the listenerCount, and if it's the only listener, it throws.
17:23:05  <isaacs>but this is shit, because the user might have done dest.once('error',fn) which will have been removed by the time we get it.
17:25:24  <isaacs>2 possible ideas:
17:25:31  <isaacs>emit "_error" when we're about to emit "error"
17:26:00  <isaacs>or, add emitter.onButNotSideEffects(ev,fn)
17:26:14  <isaacs>the first is much lower impact. the second is kind of scary, i think, but it's also a bit ugly.
17:27:18  * AvianFlujoined
17:27:35  <MI6>joyent/node: Fedor Indutny v0.10 * 8135ac1 : net: handle 'finish' event only after 'connect' - http://git.io/eoqROg
17:28:01  <MI6>joyent/node: bnoordhuis created branch fix-path-join-docs - http://git.io/20TEPQ
17:28:07  <bnoordhuis>^ review anyone?
17:29:34  * piscisaureus_joined
17:31:39  * bajtosjoined
17:31:40  * hzquit
17:32:17  * dominictarrjoined
17:32:45  * mikealjoined
17:33:50  * isaacsreviewing
17:34:24  <isaacs>bnoordhuis: lgtm
17:35:04  * Nicholas__joined
17:35:15  <bajtos>names #libuv
17:36:19  <bradleymeck>did linkedin report that the security breach recently was due to node, some dude is claiming it was
17:38:20  * `3rdEdenquit (Remote host closed the connection)
17:41:46  * stagas_joined
17:42:11  <bnoordhuis>bradleymeck: i assume linkedin would've contacted us if that was the case
17:42:36  <isaacs>bradleymeck: most of their stack is rails.
17:42:41  <isaacs>node powers their mobile app.
17:42:47  <MI6>joyent/node: Ben Noordhuis v0.10 * 9af0085 : doc: path.join() arguments must be strings In v0.8, non-string arguments - http://git.io/ekECSg
17:42:51  <isaacs>but their core infrastructure and website is all rails.
17:43:29  <isaacs>and yes, someone would have surely emailed me. i've met a few people over there, they know they can bug me/joyent with issues.
17:43:54  * stagasquit (Ping timeout: 264 seconds)
17:43:55  * stagas_changed nick to stagas
17:45:24  <bradleymeck>notified him. thanks for confirmation
17:46:20  <isaacs>bradleymeck: of course, it MAY be that their node code wsa insecure.
17:46:30  <bradleymeck>isaacs: thats true of anything
17:46:36  <isaacs>bradleymeck: but that might be like saying that node is insecure because C
17:46:37  <bradleymeck>but if it was a CORE bug
17:46:38  <isaacs>:)
17:46:39  <isaacs>right
17:46:44  <isaacs>and we've had precious few of those.
17:46:57  <tjfontaine>there better be some real wood around for you to knock on
17:47:12  <bradleymeck>isaacs: remember, i argued with black water ops, and was told it was insecure because an addon could drop the CPU into 16bit mode and blow up v8
17:47:29  <isaacs>hahahah
17:47:41  <isaacs>"this platform is insecure because it provides access to the file system!"
17:47:45  <isaacs>i've heard that one, too
17:47:52  <bradleymeck>nice
17:48:10  <isaacs>and seen presentations arguing about how node is insecure because var at the top level makes a global (it doesn't, as you know)
17:48:14  <bradleymeck>as an aside: any arguments for why node_extensions.h should not be dynamically generated?
17:48:36  <isaacs>bnoordhuis: how do you feel about a "_error" event?
17:50:20  <isaacs>we could even go __error__ es5 style
17:50:25  <isaacs>wunderbar
17:50:38  * `3rdEdenjoined
17:50:58  * kazuponjoined
17:51:00  <bnoordhuis>isaacs: i'm in a conference call :/
17:51:32  <isaacs>bnoordhuis: i'm so sorry for your loss.
17:54:30  * stagasquit (Ping timeout: 252 seconds)
17:58:38  * Raltjoined
17:59:23  <isaacs>tjfontaine: progress bars!!
17:59:25  <isaacs>tjfontaine: nice!
17:59:48  <tjfontaine>oh god
17:59:57  <tjfontaine>it's all building :)
17:59:59  * kazuponquit (Ping timeout: 245 seconds)
18:00:09  * stagasjoined
18:00:19  * qmx|lunchchanged nick to qmx
18:01:01  <tjfontaine>I've decided to change how I iterate through the benchmarks, previously I was running each category multiple times, that takes to long to burn through and things sit in the build queue for too long
18:03:32  * dscapejoined
18:04:41  <tjfontaine>isaacs: btw you broke the build, I'm just letting you know now :)
18:04:57  * hzjoined
18:05:30  * brsonjoined
18:07:45  * perezdquit (Quit: perezd)
18:08:11  * perezdjoined
18:09:22  * dominictarrquit (Quit: dominictarr)
18:09:48  * nsmquit (Quit: nsm)
18:10:09  * Raltquit (Remote host closed the connection)
18:13:34  <MI6>nodejs-v0.8: #33 FAILURE linux-x64 (1/468) osx-x64 (1/468) smartos-x64 (2/468) linux-ia32 (1/468) osx-ia32 (1/468) smartos-ia32 (2/468) http://jenkins.nodejs.org/job/nodejs-v0.8/33/
18:14:11  * Raltjoined
18:14:22  * csaohquit (Quit: csaoh)
18:14:27  <MI6>nodejs-v0.10: #15 FAILURE windows-ia32 (4/560) windows-x64 (4/560) http://jenkins.nodejs.org/job/nodejs-v0.10/15/
18:15:22  <tjfontaine>I guess that stream unwrap isn't done on windows?
18:19:20  <isaacs>tjfontaine: well, that function isn't called on windows.
18:19:25  <isaacs>tjfontaine: er, wait.
18:19:28  * isaacslooking at the error
18:20:07  <isaacs>oh, glerp.
18:20:10  * isaacsis dumb
18:21:55  <isaacs>only landed half the commit.
18:22:23  <tjfontaine>I hadn't really looked at it, just saw a bunch of red
18:22:49  <isaacs>yeah
18:22:55  <isaacs>the windows red is old red
18:23:03  <isaacs>but the unix failure is all new
18:24:03  <tjfontaine>right
18:26:49  <MI6>joyent/node: isaacs v0.10 * 110cacd : core: Move UNWRAP_NO_ABORT to handle_wrap.h Otherwise it cannot be used - http://git.io/Bk8vIg
18:26:50  <isaacs>pushed fix ^
18:27:02  * perezdquit (Quit: perezd)
18:27:05  <isaacs>had to reset my dns, because comcast decided to get their wires crossed.
18:27:56  * wavded_quit (Ping timeout: 245 seconds)
18:28:22  * dscapequit (Ping timeout: 245 seconds)
18:31:25  * Raltquit (Remote host closed the connection)
18:36:57  * bnoordhuisquit (Ping timeout: 276 seconds)
18:38:33  * Raltjoined
18:39:19  * c4miloquit (Remote host closed the connection)
18:39:57  <MI6>nodejs-v0.10: #16 UNSTABLE windows-ia32 (5/560) smartos-ia32 (1/560) linux-x64 (1/560) windows-x64 (4/560) http://jenkins.nodejs.org/job/nodejs-v0.10/16/
18:40:04  <tjfontaine>much better
18:44:08  * isaacsoff to yoga class
18:44:36  * mikealquit (Quit: Leaving.)
18:44:37  * stagasquit (Ping timeout: 248 seconds)
18:51:35  * mikealjoined
18:55:57  * indexzerojoined
18:56:52  * kazuponjoined
18:57:57  * perezdjoined
19:01:15  * kazuponquit (Ping timeout: 245 seconds)
19:05:33  * c4milojoined
19:13:32  * indexzeroquit (Quit: indexzero)
19:14:18  * indexzerojoined
19:16:32  * mikealquit (Quit: Leaving.)
19:31:21  * bradleymeckquit (Quit: bradleymeck)
19:39:15  * indexzeroquit (Quit: indexzero)
19:42:58  * bnoordhuisjoined
19:47:55  * bnoordhuisquit (Ping timeout: 264 seconds)
20:01:26  * indexzerojoined
20:02:14  * benoitcquit (Excess Flood)
20:06:43  * benoitcjoined
20:08:29  * stagasjoined
20:12:06  * bajtosquit (Quit: Leaving)
20:15:53  * AvianFluquit (Remote host closed the connection)
20:29:20  * wavded_joined
20:29:54  * dscapejoined
20:31:52  * benoitcquit (Excess Flood)
20:34:43  * benoitcjoined
20:36:49  * bnoordhuisjoined
20:40:00  * mikealjoined
20:50:11  * dominictarrjoined
20:53:56  * `3rdEdenquit (Remote host closed the connection)
20:58:02  * kazuponjoined
20:59:47  * isaacsfg
21:02:32  * kazuponquit (Ping timeout: 252 seconds)
21:03:43  <indutny>^Z
21:03:49  <indutny>err
21:03:51  <indutny>bg
21:08:03  * qmxchanged nick to qmx|afk
21:14:01  * c4miloquit (Remote host closed the connection)
21:27:25  * mikealquit (Quit: Leaving.)
21:30:37  * mikealjoined
21:31:43  * stagasquit (Ping timeout: 256 seconds)
21:34:09  * c4milojoined
21:35:22  * qmx|afkchanged nick to qmx
21:41:55  * mikealquit (Quit: Leaving.)
21:42:10  <MI6>joyent/node: bnoordhuis created branch issue5005 - http://git.io/e1TO7A
21:42:16  <bnoordhuis>isaacs: ^ review?
21:42:56  <bnoordhuis>i'll update the documentation in a separate commit
21:43:41  * Raltquit (Remote host closed the connection)
21:43:43  <tjfontaine>bnoordhuis: how would you suggest I debug the freebsd i386 issue?
21:44:16  <tjfontaine>#4010 / #4412
21:46:37  * AvianFlujoined
21:47:03  <bnoordhuis>tjfontaine: with truss or dtrace? though dtrace is a pain to build on freebsd
21:47:16  <tjfontaine>lucky for you I've already got dtrace on fbsd
21:47:24  <bnoordhuis>i'm 99% sure it's because the mmap() call fails
21:47:30  <piscisaureus_>saghul: ha! what a great fix :)
21:47:31  <tjfontaine>dtruss would agree
21:47:43  <tjfontaine> 9784/100116: mprotect(0x28A71000, 0x1000, 0x5) = 0 0
21:47:43  <tjfontaine> 9784/100116: mmap(0x0, 0x1000, 0x7) = 682041344 0
21:47:43  <tjfontaine> 9784/100116: mprotect(0x28A72000, 0x1000, 0x5) = 0 0
21:47:43  <tjfontaine> 9784/100116: mmap(0x0, 0x1000, 0x7) = 682045440 0
21:47:43  <tjfontaine> 9784/100116: mprotect(0x0, 0x0, 0x0) = 0 0
21:48:10  <bnoordhuis>looks healthy. it's not returning MAP_FAILED
21:48:23  <bnoordhuis>which is (void*)-1 on fbsd, i think
21:48:39  <tjfontaine>I did core dump though
21:48:58  <tjfontaine># Fatal error in ../deps/v8/src/store-buffer.cc, line 98
21:48:59  <tjfontaine># CHECK(virtual_memory_->Commit(reinterpret_cast<Address>(start_), kStoreBufferSize, false)) failed
21:49:26  <saghul>piscisaureus_ yeah, I don't even know how to feel about it :-)
21:49:42  <bnoordhuis>tjfontaine: right, so it _is_ failing - are you sure you got all the mmap syscalls?
21:49:46  <piscisaureus_>saghul: I have no clue why you needed that but it seems to make sense :)
21:49:52  <bnoordhuis>anyway, i'd start with reverting that mmap patch
21:53:29  <MI6>joyent/node: Ben Noordhuis issue5005 * 770c2f6 : doc: implicit fs callbacks don't throw in v0.10 But they will in v0.12. - http://git.io/TMSFug
21:54:18  <bnoordhuis>tjfontaine: is your fbsd machine something i can log into?
21:55:08  <tjfontaine>bnoordhuis: at the moment no, it's a vm on my laptop and I'm not sure the protocol for firewall pinhole here, https://gist.github.com/tjfontaine/5156720 here's a dtruss from a debug build which reports more syscalls
21:55:49  <tjfontaine> 10002/100061: mmap(0xBFEF9000, 0x16B000, 0x0) = -1 Err#12
21:56:08  <bnoordhuis>ah, there's your MMAP_FAILED
21:56:12  <tjfontaine>yup
21:56:26  <tjfontaine>there seem to be other failures on fbsd i386
21:56:32  <tjfontaine>not just this
21:56:37  <bnoordhuis>you mean test failures?
21:57:08  <tjfontaine>no I mean if the other didn't show an mmap failure, yet still core'd
21:58:42  <tjfontaine>when node_g fails it seems to always be on the mmap
21:58:49  * kazuponjoined
21:59:23  * perezd_joined
21:59:37  <bnoordhuis>ah. that's odd
21:59:55  <bnoordhuis>is that the CHECK you pasted?
22:00:04  <tjfontaine>ya
22:00:12  * perezdquit (Ping timeout: 258 seconds)
22:00:14  * perezd_changed nick to perezd
22:00:27  <tjfontaine>when node_g fails the reported check (as seen in dtruss) is
22:00:28  <tjfontaine># Fatal error in ../deps/v8/src/platform.h, line 385
22:00:28  <tjfontaine># CHECK(IsReserved()) failed
22:01:39  <bnoordhuis>it has to be the mmap() call
22:01:57  <bnoordhuis>because Commit() is what calls mmap()
22:03:29  * kazuponquit (Ping timeout: 240 seconds)
22:07:04  * wolfeidauquit (Remote host closed the connection)
22:08:07  <tjfontaine>bnoordhuis: so I guess my excersize is to learn why the mmap is failing?
22:09:02  * rendarquit
22:11:21  <isaacs>bnoordhuis: ping
22:11:40  <isaacs>bnoordhuis: i'd just sent https://github.com/joyent/node/pull/5008 when you posted your comment :)
22:11:52  <isaacs>bnoordhuis: great minds and their thinking similarly :)
22:12:05  <bnoordhuis>tjfontaine: yes, though i can guess why - there's not that much address space on i386
22:12:08  <bnoordhuis>isaacs: indeed :)
22:12:48  <isaacs>bnoordhuis: i think my js patch is a bit simpler.
22:12:59  <isaacs>bnoordhuis: but i forgot the doc patch
22:13:25  <isaacs>bnoordhuis: what do you think?
22:13:53  <isaacs>bnoordhuis: i don't have strong feelings about it, as long as the functionality is the same. --throw-deprecation = crash, --trace-deprecation = console.trace, --no-deprecation = ignore
22:15:34  * benoitcquit (Excess Flood)
22:16:39  <bnoordhuis>isaacs: oh, the reason i did it like that is that it'll be easier to switch back to rethrow() again
22:16:49  <bnoordhuis>because the patch doesn't touch that function at all
22:17:48  <isaacs>bnoordhuis: we can use the magic of git to switch back to rethrow from either with one command :)
22:18:18  <isaacs>bnoordhuis: in the meantime, i'd rather have logic that's simpler to read.
22:18:34  <isaacs>bnoordhuis: what's the latest on the mmap thing?
22:19:06  <bnoordhuis>tjfontaine is working on it
22:19:32  <bnoordhuis>someone from joyent should work on getting that patch upstream though or i'll throw a hissy fit
22:19:36  <isaacs>bnoordhuis: oh, kewl.
22:19:42  <isaacs>bnoordhuis: no need for hissint
22:19:58  <bnoordhuis>ah, but i like that
22:20:31  <MI6>joyent/node: Gil Pedersen v0.10 * 615d809 : stream: Never call decoder.end() multiple times Fixes decoder.end() bein - http://git.io/NWHHOQ
22:20:57  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
22:21:14  <bnoordhuis>re #5005, i have no strong opinions or emotions. i like my patch a bit better for the aforementioned reason (and because it will also be easy to revert when `git revert` fails)
22:21:26  * mikealjoined
22:21:43  * benoitcjoined
22:22:53  * felixgequit (Quit: felixge)
22:24:50  * wolfeidaujoined
22:28:11  <MI6>joyent/node: Ben Noordhuis v0.10 * 7b7235a : doc: add note on process.title max length Fixes #5006. - http://git.io/T5vlUw
22:28:23  * loladirojoined
22:32:24  * bradleymeckjoined
22:34:27  <isaacs>bnoordhuis: i combined our forces.
22:34:28  <MI6>joyent/node: isaacs v0.10 * 6bd8b7e : fs: Missing cb errors are deprecated, not a throw Commit a804347 makes f (+1 more commits) - http://git.io/v5XxJQ
22:34:41  <isaacs>bnoordhuis: the good news is, we can merge v0.10 into master right now, and just revert the commit there while it's easy.
22:34:48  <tjfontaine>ya this patched fbsd with the hint lower is still re-spawning, for 15 mins without any failure
22:35:06  <isaacs>bnoordhuis: the pain of merges will inspire us to not muck around with fs any more in v0.10 :)
22:36:03  <bnoordhuis>hah, indeed :)
22:39:03  <isaacs>running tests on master, then gonna push the merge+revert
22:41:12  * benoitcquit (Excess Flood)
22:42:06  <MI6>nodejs-v0.10: #17 UNSTABLE windows-ia32 (5/560) smartos-ia32 (3/560) osx-ia32 (2/560) linux-x64 (2/560) windows-x64 (6/560) smartos-x64 (2/560) osx-x64 (3/560) linux-ia32 (2/560) http://jenkins.nodejs.org/job/nodejs-v0.10/17/
22:42:42  <tjfontaine>hmm
22:43:42  <tjfontaine>I'll wait for this build to complete, but test-stream2-push.js and test-stream2-readable-empty-buffer-no-eof.js seem unhappy
22:44:44  <isaacs>ack, crap.
22:44:46  <isaacs>man.
22:44:48  <isaacs>yes.
22:44:54  * isaacssigh.
22:44:57  * isaacsis too trusting sometimes.
22:45:45  * benoitcjoined
22:49:11  <MI6>joyent/node: isaacs v0.10 * 6399839 : Revert "stream: Never call decoder.end() multiple times" This reverts co - http://git.io/K7vaAA
22:53:10  <tjfontaine>damn I should have just installed debian-kfreebsd, screw you ports
22:55:18  <bnoordhuis>haha
22:55:27  <bnoordhuis>what are you trying to do, tj?
22:55:38  <MI6>nodejs-v0.10: #18 UNSTABLE windows-ia32 (7/560) smartos-ia32 (3/560) osx-ia32 (3/560) linux-x64 (2/560) windows-x64 (6/560) smartos-x64 (2/560) osx-x64 (2/560) linux-ia32 (2/560) http://jenkins.nodejs.org/job/nodejs-v0.10/18/
22:55:40  <tjfontaine>I want my gnu environment :P
22:56:20  <tjfontaine>anyway I think I'm going to get a debug libc or just recompile it so I can see useful ustack from dtrace
22:56:43  <tjfontaine>but I'd like to know where the enomem is happening
22:57:01  <MI6>joyent/node: isaacs master * 5e140b3 : Revert "fs: Missing cb errors are deprecated, not a throw" This reverts (+29 more commits) - http://git.io/GOJegg
22:59:39  * kazuponjoined
23:01:38  * brson_joined
23:02:30  * brsonquit (Ping timeout: 252 seconds)
23:04:11  * c4miloquit (Remote host closed the connection)
23:04:36  * kazuponquit (Ping timeout: 272 seconds)
23:08:03  * indexzeroquit (Quit: indexzero)
23:09:37  * bradleymeckquit (Quit: bradleymeck)
23:13:22  <MI6>nodejs-master: #96 UNSTABLE osx-x64 (2/560) windows-ia32 (5/560) linux-ia32 (1/560) windows-x64 (4/560) osx-ia32 (1/560) http://jenkins.nodejs.org/job/nodejs-master/96/
23:17:12  * AvianFluquit (Remote host closed the connection)
23:21:15  <tjfontaine>oh it's so fucking noisey
23:21:30  <tjfontaine>DMAN YOU JENKINS
23:21:31  <LOUDBOT>OH WELL NEW YORK IS BETTER THAN ALL YOUR CRAPPY CITIES ANYWAY
23:21:54  * gblockjoined
23:22:00  <gblock>hello folks
23:22:26  <gblock>got an npm issue related to engines strict. where would be the best place to ask?
23:23:29  <gblock>we have a breaking issue on our modules related to node 0.10
23:23:44  <gblock>our modules won't work with node > 0.8.x
23:24:09  <gblock>so we…..tried to rev our modules adding a node constraint of < 9
23:24:38  <gblock>and adding enginesstrict but that did not have the desired effect
23:27:13  <tjfontaine>what does your engines clause look like?
23:30:51  * bradleymeckjoined
23:31:05  <MI6>nodejs-v0.10: #19 ABORTED windows-ia32 (4/560) smartos-ia32 (2/560) linux-x64 (2/560) http://jenkins.nodejs.org/job/nodejs-v0.10/19/
23:37:58  * hzquit
23:38:56  * bradleymeckquit (Quit: bradleymeck)
23:39:38  * qmxchanged nick to qmx|away
23:40:14  * mikealquit (Quit: Leaving.)
23:44:58  * c4milojoined
23:46:29  * Nicholas__quit (Ping timeout: 248 seconds)
23:48:02  <bnoordhuis>err, why do i have 80 new github emails all of a sudden?
23:48:43  <bnoordhuis>tjfontaine: "Can one of the admins verify this patch?" <- is that your work?
23:50:58  <MI6>nodejs-pullrequest: #1 UNSTABLE osx-ia32 (22/534) windows-x64 (88/534) linux-x64 (40/534) linux-ia32 (47/534) smartos-ia32 (10/534) windows-ia32 (52/534) smartos-x64 (96/534) osx-x64 (29/534) http://jenkins.nodejs.org/job/nodejs-pullrequest/1/
23:54:13  <tjfontaine>bnoordhuis: unfortunately :/
23:55:02  <tjfontaine>bnoordhuis: you are an admin, if you're interested in trigger thing https://wiki.jenkins-ci.org/display/JENKINS/GitHub+pull+request+builder+plugin