00:00:02  <creationix>aww, my code doesn't compile on the v0.8 branch
00:00:04  <piscisaureus_>creationix: flags = UV_CLOSING | UV_CLOSED | UV_STREAM_READING | UV_STREAM_READABLE
00:00:12  <piscisaureus_>creationix: so apparently you already uv_close'd the handle
00:00:37  <piscisaureus_>I wonder how unix manages flags, given that UV_STREAM_READING kinda contradicts UV_CLOSED
00:01:16  <piscisaureus_>ah snap
00:01:18  <creationix>yeah, I never understood that
00:01:24  <piscisaureus_>I am weong tho :-)
00:01:58  * dedis4quit (K-Lined)
00:01:59  * brsonquit (Ping timeout: 252 seconds)
00:02:11  <piscisaureus_>it is UV_CLOSING | UV_STREAM_READABLE | UV_STREAM_WRITABLE | UV__HANDLE_REF
00:02:13  <piscisaureus_>that makes sense
00:02:32  <piscisaureus_>creationix: so probably you already called uv_close
00:02:40  <creationix>I'll trace it
00:02:46  <piscisaureus_>creationix: somehow uv-unix decides to call your read cb nonetheless
00:02:58  <piscisaureus_>which is legal btw
00:03:15  <piscisaureus_>but I would not expect UV_EOF in that case
00:03:42  * brsonjoined
00:03:58  <creationix>ahh, found it
00:04:05  <creationix>someone put a call to uv_close in my on_read handler
00:04:33  <creationix>right before looking for the UV_EOF
00:04:36  <piscisaureus_>ghe hahaha
00:04:37  <creationix>is that right?
00:04:40  <piscisaureus_>someone, riiiight :-)
00:04:45  <piscisaureus_>:-p
00:04:58  <piscisaureus_>creationix: it is fine to respond to UV_EOF with uv_close
00:05:06  <piscisaureus_>actually, you are sorta supposed to do that
00:05:06  <creationix>https://github.com/creationix/luv/blob/master/luv_stream.c#L26
00:05:18  <creationix>right, but shouldn't that be up to the client code?
00:05:22  <piscisaureus_>creationix: but uv_shutdown may be more practical because you want to drain the write queue first
00:05:28  <piscisaureus_>otherwise libuv will cancel pending writes
00:05:39  <creationix>right, I'm calling shutdown in the lua script
00:06:48  <creationix>what happens if I call uv_shutdown, but never call uv_close
00:06:54  <creationix>it will leak resources right?
00:06:57  <piscisaureus_>yes
00:07:02  <creationix>alright, thanks
00:07:23  <piscisaureus_>it would be nice if libuv had a "deferred close"
00:07:27  <piscisaureus_>that does not truncate the write queue
00:07:32  <piscisaureus_>I wonder how node handles this actually
00:07:54  <creationix>probably calls close in the on_shutdown callback
00:07:58  <creationix>that's what I usually do
00:08:12  <piscisaureus_>I don't think so
00:08:18  <piscisaureus_>it's less efficient :-)
00:10:18  <piscisaureus_>creationix: yeah, node does not close in the binding layer
00:10:59  <piscisaureus_>creationix: it might close in the js layer, depending on the state of the write queue and whether AllowHalfOpen is set
00:12:53  * sgallagh_afkquit (Remote host closed the connection)
00:38:44  * jmar777quit (Remote host closed the connection)
00:38:45  * loladiroquit (Quit: loladiro)
00:39:17  * jmar777joined
00:43:43  * jmar777_joined
00:43:52  * loladirojoined
00:44:51  * lohkeyquit (Quit: lohkey)
00:47:17  * jmar777quit (Ping timeout: 244 seconds)
00:51:46  * brsonquit (Ping timeout: 250 seconds)
01:03:00  * brsonjoined
01:20:24  * loladiroquit (Quit: loladiro)
01:23:52  * loladirojoined
01:27:52  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
01:28:52  * AvianFlujoined
01:32:06  * jmar777_quit (Remote host closed the connection)
01:32:41  * jmar777joined
01:36:53  * jmar777quit (Ping timeout: 244 seconds)
01:37:06  * abraxasjoined
01:40:04  * mmaleckichanged nick to mmalecki[zzz]
01:40:05  * loladiroquit (Quit: loladiro)
01:44:41  * loladirojoined
01:48:32  * dapquit (Quit: Leaving.)
01:55:41  * jmar777joined
02:01:11  * stagasquit (Ping timeout: 252 seconds)
02:12:57  * mikealquit (Quit: Leaving.)
02:21:33  * jmar777quit (Remote host closed the connection)
02:22:07  * jmar777joined
02:26:14  * jmar777quit (Ping timeout: 240 seconds)
02:30:33  * jmar777joined
02:35:39  * TooTallNatequit (Quit: ["Textual IRC Client: www.textualapp.com"])
03:07:49  * c4milojoined
03:12:37  * loladiroquit (Quit: loladiro)
03:13:31  * kristatejoined
03:17:04  * loladirojoined
03:33:33  * kazuponjoined
03:53:10  * c4miloquit (Remote host closed the connection)
03:53:59  * c4milojoined
04:07:55  * c4miloquit (Remote host closed the connection)
04:13:19  * brsonquit (Ping timeout: 255 seconds)
04:19:13  * perezdquit (Quit: perezd)
04:20:31  * perezdjoined
04:22:34  * dapjoined
04:51:48  * loladiroquit (Quit: loladiro)
04:57:51  * loladirojoined
05:03:20  * coderarityjoined
05:12:48  * kristatequit (Remote host closed the connection)
05:26:24  * kristatejoined
05:32:53  * jmar777quit (Remote host closed the connection)
05:33:27  * jmar777joined
05:37:39  * jmar777quit (Ping timeout: 244 seconds)
05:53:55  * loladiroquit (Quit: loladiro)
05:54:12  * dapquit (Quit: Leaving.)
06:21:56  * loladirojoined
07:06:06  * AvianFluquit (Remote host closed the connection)
07:40:39  * rendarjoined
07:58:30  * bradleymeckjoined
07:59:29  * bradleymeckquit (Client Quit)
08:16:00  * perezdquit (Quit: perezd)
08:29:43  * abraxasquit (Remote host closed the connection)
09:04:20  * janjongboomjoined
09:04:52  * loladiroquit (Quit: loladiro)
09:12:56  * `3rdEdenjoined
09:23:43  * brsonjoined
09:39:34  * lohkeyjoined
10:03:31  * maheshjoined
10:04:51  <mahesh>How can I modify the http.js & net.js & recompile it?
10:05:26  <mahesh>in lib folder?
10:06:00  * lohkeyquit (Quit: lohkey)
10:08:20  <mahesh>Thanks in advance
10:08:44  <mahesh>please if anyone can help me out in this
10:17:56  * benoitcquit (Ping timeout: 245 seconds)
10:28:51  * abraxas_joined
10:31:40  * stagasjoined
10:36:02  * benoitcjoined
10:36:11  * abraxas_quit (Remote host closed the connection)
10:39:24  * coderarityquit (Remote host closed the connection)
10:40:40  * kazuponquit (Remote host closed the connection)
10:45:37  <indutny>mahesh: well, just modify them
10:45:39  <indutny>mahesh: and run make
10:45:55  <indutny>mahesh: they'll be automatically recompiled
10:46:10  <mahesh>I did that but seems no effect
10:46:40  <mahesh>for tesing purpose I put some prints
10:46:55  <indutny>well, I do it everyday
10:47:08  <indutny>are you sure you're executing node that's result of build?
10:47:13  <indutny>what file are you executing?
10:48:06  <mahesh>I put some prints in net.js & http.js
10:48:18  <mahesh>I ran make
10:48:31  <mahesh>& did
10:48:44  <mahesh>node test.js
10:49:23  <mahesh>in that case what shuld I use
10:49:32  <mahesh>It will be great help
10:51:30  <indutny>node?
10:51:31  <indutny>./node
10:51:38  <indutny>^ that's how it should be
10:51:47  <indutny>you're using global installed node
10:51:52  <indutny>not what you've just compiled
10:52:52  <mahesh>got it
10:52:55  <mahesh>thanks
10:53:49  <mahesh>how can I receive tcp packet over http server
10:53:52  <mahesh>?
10:54:02  <indutny>oh gosh :)
10:54:09  <indutny>what do you mean by tcp packet?
10:54:11  * stagasquit (Quit: ChatZilla 0.9.89-rdmsoft [XULRunner 1.9.0.17/2009122204])
10:54:35  <mahesh>I want to build a application
10:54:59  <mahesh>in which I have tcp client & http client
10:55:06  <mahesh>on same port
10:55:54  <indutny>well, it's pretty hacky
10:56:16  <indutny>you need to remove 'connection' listener and replace it with your own
10:56:17  <mahesh>so befor processing a http request I have to make sure wht client it is?
10:56:27  <indutny>I don't know your logic
10:56:30  <mahesh>in http .js?
10:56:38  <indutny>you don't need to patch it for this
10:58:42  <mahesh>wht if I use the same & chane the listener of tht perticular socket?
10:59:07  <indutny>well, you can just implement your own listener
10:59:08  <mahesh>What could be the good logic?
10:59:21  <indutny>and call old listener if you think it should be processed as a http protocol
10:59:25  <indutny>s/protocol/connection
10:59:51  <indutny>and btw, it's better ask this things on #node.js channel
11:00:07  <indutny>#libuv is only for core development
11:01:09  <mahesh>okk
11:01:34  <mahesh>one more devlopment am doing for tap device?
11:01:52  <mahesh>How can I read & write data over tap device?
11:02:06  <mahesh>using node js?
11:02:52  <indutny>idk, ask on #node.js, probably someone may know it there
11:04:30  <mahesh>ok thanks.
11:04:39  <mahesh>got really good help
11:05:29  <indutny>good
11:05:45  * janjongboomquit (Quit: janjongboom)
11:41:35  * kazuponjoined
11:47:17  * janjongboomjoined
11:50:53  * kazuponquit (Ping timeout: 255 seconds)
11:51:21  * indexzerojoined
12:13:58  * brsonquit (Ping timeout: 265 seconds)
12:16:59  * kazuponjoined
12:17:07  * rjequit (Ping timeout: 246 seconds)
12:18:43  * rjejoined
12:19:17  * mmalecki[zzz]changed nick to mmalecki
12:22:19  * bnoordhuisjoined
12:26:26  * kazuponquit (Ping timeout: 255 seconds)
12:31:01  <indutny>bnoordhuis: hoya
12:31:08  <indutny>bnoordhuis: wanna land osx select() stuff?
12:38:34  * kristatequit (Ping timeout: 240 seconds)
12:46:59  * piscisaureus_joined
12:48:46  * bnoordhuisquit (Ping timeout: 245 seconds)
12:49:37  <indutny>piscisaureus_: heya
12:49:40  <indutny>piscisaureus_: ben is ignoring me
12:49:58  <indutny>philips-: please, slap him in the face
12:50:01  <indutny>oops
12:50:03  <indutny>piscisaureus_: ^^
13:16:41  * bnoordhuisjoined
13:16:56  <indutny>bnoordhuis: ping?
13:18:23  <bnoordhuis>indutny: pong
13:19:28  <indutny>bnoordhuis: are you in good mood today?
13:19:29  <indutny>:)
13:23:21  <bnoordhuis>indutny: i always am
13:23:39  <indutny>bnoordhuis: ok, mind reviewing and accepting select() patch then?
13:24:17  <bnoordhuis>well, maybe. i have some other stuff to work on as well though
13:24:32  <bnoordhuis>indutny: btw, how are you adding value to our company today?
13:25:04  <indutny>bnoordhuis: do you have any concerns?
13:25:06  <txdv>does he work at cloud9?
13:25:12  <indutny>txdv: yes, I do
13:25:48  <txdv>bnoordhuis: you have to define value
13:31:05  <txdv>do you guys care about codestyle in libuv?
13:31:23  <indutny>yes
13:31:33  <indutny>a lot
13:31:38  <indutny>except windows code
13:31:44  <indutny>it's hard to write it good looking
13:31:51  <indutny>all that HANDLE and PCHAR stuff
13:31:58  <txdv>xD
13:32:04  <txdv>it's so ugly that we don't care any more
13:32:51  <txdv>windows code is so horrible that all the coders go into a CAPS LOCK RAGE (HANDLE)
13:33:22  <indutny>well, it's just a matter of habbit
13:34:26  <txdv>I know
13:34:32  <bnoordhuis>indutny: no concerns, i just like being a high-level ideas kind of manager
13:34:33  <txdv>like hungarian variable notation
13:34:50  <txdv>it is like this _t thing in unix
13:34:57  <indutny>complex sentence, didn't understand anything :)
13:35:02  <bnoordhuis>we have to consolidate in order to maximize our synergy, that kind of thing
13:35:07  <indutny>aaah
13:35:09  <indutny>ok
13:35:24  <bnoordhuis>it doesn't have to mean anything, just sound nice
13:35:26  <indutny>well, you know I do a lot, even while I'm a little bit not-in-shape
13:36:02  <indutny>it's harder to guess what you're doing there ;)
13:36:21  <indutny>(not offense meant)
13:37:15  <bnoordhuis>oh, i'm just living from paycheck to paycheck
13:37:22  <bnoordhuis>working on my minesweeper skills in between
13:38:06  <txdv>no reddit?
13:38:14  <txdv>I thought your skillset at least includes reddit
13:38:38  <bnoordhuis>i'm more of a HN kind of guy
13:38:43  <bnoordhuis>though why HN i don't know either
13:38:50  <indutny>:)
13:39:18  <txdv>what is HN?
13:39:24  <indutny>hackernews?
13:39:27  <indutny>http://news.ycombinator.com/
13:39:38  <indutny>it's where we're sitting for a whole day
13:39:44  <indutny>when there're no question on this channel
13:40:05  <txdv>it's reddit just for hackers?
13:40:16  <bnoordhuis>and wannabe hackers
13:40:23  <bnoordhuis>plus a lot of annoying startup weenies
13:41:03  <txdv>good
13:41:13  <txdv>a place where I can whine about how important my bindings for libuv are
13:41:43  <bnoordhuis>tbh i like lwn best
13:41:59  <bnoordhuis>but lwn doesn't post nearly enough articles to quench my infothirst
13:42:02  * indutnygoes all 1337
13:42:08  <txdv>lwn?
13:42:14  <bnoordhuis>lwn.net
13:42:21  * kazuponjoined
13:42:23  <indutny>txdv: man, you're missing a lot of stuff
13:42:34  <indutny>bnoordhuis: wanna hack mail.ru?
13:42:37  * sergimjoined
13:42:42  <indutny>bnoordhuis: they're using /dev/urandom for SSL
13:42:57  <bnoordhuis>indutny: i'm afraid whole of russia will hack me back
13:43:04  <indutny>whale of russia
13:43:09  <bnoordhuis>that too
13:43:11  <indutny>will hit you
13:43:24  <bnoordhuis>i'll refrain from inserting a "your mom" joke here
13:43:36  * kazupon_joined
13:43:41  <indutny>I've few jokes that I'm often refraining when talking to you :)
13:43:42  <bnoordhuis>okay, enough merry banter. time for work
13:43:53  <txdv>I read the codeproject news letters, they always have articles from various sources
13:44:07  * kazuponquit (Read error: Connection reset by peer)
13:44:16  <txdv>so I have seen lwn and HN, but I never prescribed to them
13:44:38  <piscisaureus_>yeah bnoordhuis should really review indutny's patch now
13:44:44  <txdv>xbut HN Is like a proffesional version of it
13:49:55  * indexzeroquit (Quit: indexzero)
13:53:00  * stagasjoined
14:09:51  <txdv>so with no more ev
14:09:55  <txdv>what is next on libuvs roadmap?
14:10:52  <indutny>removing C code from it
14:10:56  <indutny>it should be plain headers
14:10:59  <indutny>and defines
14:11:03  <indutny>bnoordhuis: right?
14:22:40  <txdv>your sarcasm is appreciated, but the real deal would be better
14:24:24  <indutny>well, fixing bugs
14:24:31  <indutny>improving performance
14:24:38  <indutny>probably ben or bert has other plans that this
14:25:00  <indutny>actually, I like idea with headers and defines
14:25:13  <indutny>we could make g++ run eventloop at compile time
14:29:56  <saghul>indutny you could transform uv_wait into uv_backend_fd and uv_get_timeout ;-)
14:30:23  <indutny>oh yes
14:30:24  <indutny>and this
14:30:31  <indutny>well, actually, I can do it
14:30:40  <indutny>but looks like bnoordhuis is no longer accepting patches from me :(
14:30:42  * AvianFlujoined
14:31:59  <bnoordhuis>indutny: i'm on review strike, it's nothing personal
14:32:08  <indutny>kk :)
14:32:13  <bnoordhuis>but yes, uv_backend_*() would be nice
14:32:27  <indutny>how should timeout thingy be called?
14:32:34  <indutny>uv_backend_get_timeout?
14:32:41  <bnoordhuis>uv_backend_timeout?
14:32:53  <indutny>ok too
14:33:18  <bnoordhuis>i guess it should fail with UV_ENOSYS on the os that shall not be named
14:33:32  <indutny>:)
14:36:37  <txdv>what would be the purpose of uv_backend_* ?
14:38:05  <txdv>indutny: would that define rewrite work on gcc too?
14:38:24  <indutny>txdv: yeah, gcc needs to be patched to support it
14:38:37  <indutny>txdv: it's purpose is embedding uv's loop in another loops
14:38:56  <ArmyOfBruce>pasing that uv_buf_t by value makes wrapping in some languages a tiny bit harder (when the C-FFI doesn't support structs args by value).
14:39:30  <piscisaureus_>https://gist.github.com/63cc751daf20dad33df1
14:39:39  <piscisaureus_>^-- merged all teh config files
14:39:47  <piscisaureus_>most of it makes no sense to me tho
14:39:53  <txdv>passing uv_buf_t by value is a freaking bitch
14:39:55  <piscisaureus_>looks like just plain stupidness sometimes
14:40:17  <txdv>I have to do runtime checks if i'm on windows or on unix and pass the appropriate struct
14:40:59  <piscisaureus_>txdv: well, that sucks :-)
14:41:18  <indutny>piscisaureus_: you scary me
14:41:32  <txdv>piscisaureus_: luckily for you, it sucks only for me
14:41:37  <piscisaureus_>txdv: uv_buf_t looks like that for a reason, it has the same layout as the native iovec
14:41:50  <piscisaureus_>txdv: yes
14:41:56  <piscisaureus_>indutny: what do you mean?
14:42:02  <piscisaureus_>indutny: I am the scary version of you?
14:42:09  <indutny>yes
14:42:10  <piscisaureus_>indutny: you posted an ellipsis again
14:42:12  <txdv>well, passing it by value would ease the pain
14:42:16  <txdv>i mean by reference
14:42:18  <indutny>that's what I meant
14:42:23  <txdv>and sockaddr should be by reference too
14:42:30  <txdv>I already whined about this
14:42:36  <piscisaureus_>I have sympathy for your sockaddr ideas
14:42:36  * jmar777joined
14:42:43  <ArmyOfBruce>txdv: there's also uv_tcp_connect & uv_tcp_connect6 which could've just been a sockaddr_storage* and a single function.
14:42:46  <piscisaureus_>but I think we will keep uv_buf_t just the same
14:43:04  <piscisaureus_>infact uv_buf_t makes me extremely happy
14:43:07  <piscisaureus_>to demonstrate
14:43:11  <txdv>ArmyOfBruce: let's fork libuv and show them how much we hate 'by value'!
14:43:15  <piscisaureus_>HAHAHAH HAHA HAAHAHA <-- me happy
14:43:26  <ArmyOfBruce>txdv: what language do you use it from?
14:43:33  <txdv>c#
14:44:11  * jmar777quit (Remote host closed the connection)
14:44:18  <txdv>what about you ArmyOfBruce?
14:44:47  * jmar777joined
14:44:59  <ArmyOfBruce>txdv: I'm regularly considering whether I use it within the OpenDylan runtime or not. (http://opendylan.org/) … I'm currently 30/70% (30 in favor).
14:45:00  <txdv>Social protocol engagement text on #libuv sounds like: "what language are you writing libuv bindings for?"
14:45:54  <txdv>the good thing about languages which are not widespread
14:45:59  <txdv>you have to write all the libs yourself
14:46:36  <indutny>huh
14:46:39  <ArmyOfBruce>txdv: which means they work well for your usecase … and you don't do things you don't need because you don't want to write yet another library. :)
14:46:48  * hzjoined
14:47:26  <txdv>Hell, C# is widespread but all the code written uses blocking functionality, which makes it impossible to integrate with my bindings for libuv
14:47:58  * Raltquit (Read error: Connection reset by peer)
14:48:09  <txdv>because all they use is native sockets
14:48:40  <ArmyOfBruce>txdv: that's where it is an advantage to largely control your compiler + runtime … I can decide to make it coroutine-based.
14:48:56  * jmar777quit (Ping timeout: 244 seconds)
14:49:22  <txdv>there is a lib for coroutines in c#
14:49:58  <txdv>https://github.com/txdv/LibuvSharp.Blocking
14:52:22  <txdv>I just realized that I prefixed all of the classes with Blocking just to make go away name clashes
14:52:24  <txdv>xD
14:55:16  * janjongboomquit (Read error: Connection reset by peer)
14:55:53  * janjongboomjoined
14:56:11  <txdv>hei
14:56:21  <txdv>a feature request
14:56:32  <bnoordhuis>a denial
14:56:45  <txdv>there is uv_*_start/stop for reading from sockets
14:57:07  <txdv>or like uv_*_read assumes the position of starting it
14:57:17  <txdv>but why is there no start/stop for the writing queue?
14:57:32  <indutny>like .pause()/.resume() in node?
14:57:36  <txdv>da
14:57:37  <bnoordhuis>because you control that yourself
14:57:43  <indutny>indeed
14:57:50  <txdv>Hm
14:57:53  <txdv>I guess I can do that
14:58:00  <indutny>you may want to coalesce buffers
14:58:07  <indutny>or discard some of them
14:58:12  <indutny>depending on your application
14:58:21  <indutny>it's never going to fit everyone's needs
14:59:25  <txdv>I guess it is quite reasonable to do it on my own
15:01:46  <ArmyOfBruce>You could do thread names - uv_thread_set_name().
15:02:07  <txdv>i always wanted to name a thread 'dick' with libuv
15:02:30  <txdv>but you heartless people always denied me of this happyness
15:04:42  <indutny>:)
15:15:53  * Raltjoined
15:17:13  <MI6>joyent/node: Ben Noordhuis master * 8d2753c : constants: fix wrapping of large constants Use Number::New() instead of - http://git.io/3ekvKg
15:19:51  <indutny>bnoordhuis: won't they wrap up anyway?
15:20:16  <indutny>or do you mean something like 2**33 converted to double and back to int64_t
15:22:25  <MI6>joyent/libuv: Ben Noordhuis master * dcd3b55 : unix: fix thread pool race condition Send the wakeup signal to the main - http://git.io/tRvyRA
15:24:06  * travis-cijoined
15:24:06  <travis-ci>[travis-ci] joyent/libuv#885 (master - dcd3b55 : Ben Noordhuis): The build passed.
15:24:06  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/fc5984fb45b2...dcd3b55c1990
15:24:06  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/3315895
15:24:06  * travis-cipart
15:28:22  <bnoordhuis>indutny: there's an example in the commit log
15:35:29  * bradleymeckjoined
15:37:58  * tomshredsjoined
15:46:13  <bnoordhuis>call in 15?
15:46:34  * tommybergeronjoined
15:46:53  * tommybergeronquit (Client Quit)
15:47:04  * tommybergeronjoined
15:47:17  * tommybergeronquit (Client Quit)
15:47:25  <piscisaureus_>I think it will be a little delayed
15:47:32  <piscisaureus_>by a couple of minutes
15:50:30  * tomshredsquit (Ping timeout: 264 seconds)
15:50:50  * bnoordhuisquit (Ping timeout: 260 seconds)
16:00:08  * hzquit
16:06:57  * `3rdEdenquit (Remote host closed the connection)
16:07:43  * bnoordhuisjoined
16:28:12  * kazupon_quit (Remote host closed the connection)
16:35:57  * hzjoined
16:39:03  * sergimquit (Quit: Computer has gone to sleep.)
16:42:03  * Raltquit (Remote host closed the connection)
16:47:17  * Raltjoined
16:51:18  <indutny>bnoordhuis: so that's what I was talkin about
16:51:19  <indutny>ok
16:52:21  * loladirojoined
16:53:09  * AvianFluquit (Remote host closed the connection)
16:54:46  <bnoordhuis>indutny: you're refering to the commit log?
16:54:50  <indutny>bnoordhuis: yes
16:54:55  <bnoordhuis>okay
16:55:07  <indutny>bnoordhuis: it's a hack, basically
16:55:16  <indutny>bnoordhuis: but it works well for PowerOfTwo numbers
16:55:57  <MI6>joyent/libuv: Ben Noordhuis master * 90271e1 : darwin, freebsd: set SO_NOSIGPIPE on sockets Suppresses delivery of SIGP - http://git.io/RDsu4A
16:57:46  * travis-cijoined
16:57:46  <travis-ci>[travis-ci] joyent/libuv#886 (master - 90271e1 : Ben Noordhuis): The build was broken.
16:57:46  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/dcd3b55c1990...90271e1a3cf6
16:57:46  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/3316993
16:57:46  * travis-cipart
17:07:55  * stagasquit (Ping timeout: 244 seconds)
17:12:01  * bnoordhu1sjoined
17:13:08  * bnoordhuisquit (Ping timeout: 265 seconds)
17:13:20  * bnoordhu1schanged nick to bnoordhuis
17:31:09  <MI6>joyent/node: Trevor Norris v0.8 * bb867c0 : doc: Add lines about additonal uses of Buffer That Buffers can be used w - http://git.io/06LycA
17:35:00  * loladiroquit (Quit: loladiro)
17:38:41  * kazuponjoined
17:43:17  * kazuponquit (Ping timeout: 260 seconds)
17:44:48  * bradleymeckquit (Quit: bradleymeck)
17:51:49  <txdv>O yeah
17:51:59  <txdv>I have thought write_start/stop
17:52:45  <txdv>if I write 2 buffers, and the tcp socket can handle them without queing
17:53:35  <txdv>is there a difference in speed compared to writing one buffer and the second only in the callback of the first one?
17:56:54  * loladirojoined
18:05:07  * brsonjoined
18:11:20  <bnoordhuis>txdv: yes, there could be
18:12:06  <txdv>in that case, we need start/stop write because I see no way how to implement it on top of libuv without sacrificing speed
18:13:21  * loladiroquit (Quit: loladiro)
18:13:47  * bradleymeckjoined
18:14:46  * bradleymeckquit (Client Quit)
18:30:42  * bnoordhuisquit (Ping timeout: 264 seconds)
18:38:11  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
18:43:07  * janjongboomquit (Quit: janjongboom)
18:51:17  * brsonquit (Quit: leaving)
18:57:05  <txdv>but if you want to know why I need it
18:57:19  <txdv>welll
19:17:58  * stagasjoined
19:29:18  * brsonjoined
19:46:57  * `3rdEdenjoined
19:55:22  * hzquit
19:57:23  * hzjoined
20:04:58  * TooTallNatejoined
20:09:10  * TheJHjoined
20:14:20  * `3rdEdenquit (Remote host closed the connection)
20:27:15  * Ralt_joined
20:41:51  * Ralt_quit (Read error: Connection reset by peer)
20:44:35  * Ralt_joined
20:59:53  * Ralt_quit (Remote host closed the connection)
21:02:15  * deoxxa_changed nick to deoxxa
21:08:45  * Ralt_joined
21:09:02  * jmar777joined
21:16:02  * c4milojoined
21:18:33  * c4miloquit (Remote host closed the connection)
21:27:00  * stagas_joined
21:29:47  * stagasquit (Ping timeout: 265 seconds)
21:29:55  * stagas_changed nick to stagas
22:02:18  * Ralt_quit (Remote host closed the connection)
22:07:58  * rendarquit
22:13:40  * jmar777quit (Remote host closed the connection)
22:14:12  * jmar777joined
22:18:26  * jmar777quit (Ping timeout: 244 seconds)
22:27:24  * bradleymeckjoined
22:53:07  * piscisaureus_joined
22:54:53  * philips-quit (Excess Flood)
22:56:54  * philips_joined
23:01:34  * piscisaureus_quit (Ping timeout: 260 seconds)
23:17:22  * bradleymeckquit (Quit: bradleymeck)
23:19:09  * bradleymeckjoined
23:54:21  * jmar777joined