00:00:00  * ircretaryquit (Remote host closed the connection)
00:00:09  * ircretaryjoined
00:01:25  * quijotejoined
00:01:46  * octetcloudquit (Ping timeout: 264 seconds)
00:05:58  * quijotequit (Ping timeout: 255 seconds)
00:23:21  * rmgquit (Remote host closed the connection)
00:27:37  * zz_karupachanged nick to karupa
00:36:23  * petka_quit (Quit: Connection closed for inactivity)
00:40:16  * seldoquit (Remote host closed the connection)
00:51:25  * sh1mmerquit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
00:53:49  * rmgjoined
00:55:10  * inolenjoined
00:55:10  * inolenquit (Client Quit)
00:55:46  * brsonquit (Ping timeout: 264 seconds)
00:59:52  * rmgquit (Ping timeout: 245 seconds)
01:00:01  * rosskquit (Remote host closed the connection)
01:00:27  * rosskjoined
01:00:52  * mikealjoined
01:01:45  * sh1mmerjoined
01:02:09  * quijotejoined
01:04:52  * rosskquit (Ping timeout: 240 seconds)
01:06:52  * quijotequit (Ping timeout: 245 seconds)
01:14:20  * rosskjoined
01:17:05  * EhevuTovquit (Quit: This computer has gone to sleep)
01:17:40  * dshaw_joined
01:17:50  * mikealquit (Quit: Leaving.)
01:18:05  * Left_Turnquit (Remote host closed the connection)
01:18:34  * sh1mmerquit (Ping timeout: 264 seconds)
01:18:49  * jgiquit (Quit: jgi)
01:19:36  * abraxasjoined
01:20:53  * dshaw_quit (Client Quit)
01:23:31  * rosskquit (Remote host closed the connection)
01:23:57  * abraxasquit (Ping timeout: 240 seconds)
01:28:08  * dshaw_joined
01:28:51  * mikealjoined
01:29:01  * mikealquit (Client Quit)
01:37:57  * dshaw_quit (Quit: Leaving.)
01:42:49  * abraxasjoined
01:44:18  * kazuponjoined
01:53:27  * dshaw_joined
01:55:59  * kazuponquit (Remote host closed the connection)
01:58:17  * kazuponjoined
01:58:52  * janjongboomjoined
02:01:19  * AvianFluquit (Quit: Leaving)
02:02:57  * quijotejoined
02:07:47  * quijotequit (Ping timeout: 255 seconds)
02:13:08  * inolenjoined
02:13:12  * inolenquit (Client Quit)
02:13:45  * AvianFlujoined
02:14:30  * inolenjoined
02:14:31  * inolenquit (Client Quit)
02:17:02  * inolenjoined
02:17:02  * inolenquit (Client Quit)
02:19:22  * inolenjoined
02:19:55  * mikealjoined
02:22:21  * mikealquit (Client Quit)
02:24:38  * rosskjoined
02:31:30  * rosskquit (Remote host closed the connection)
02:32:41  * rosskjoined
02:33:42  * dshaw_quit (Quit: Leaving.)
02:38:07  * EhevuTovjoined
02:44:49  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
02:45:55  * bradleymeckquit (Quit: bradleymeck)
02:48:02  * c4miloquit (Remote host closed the connection)
02:55:39  * rosskquit (Remote host closed the connection)
03:00:02  * rosskjoined
03:03:49  * quijotejoined
03:05:35  * sh1mmerjoined
03:08:21  * quijotequit (Ping timeout: 240 seconds)
03:10:33  * EhevuTovquit (Quit: This computer has gone to sleep)
03:14:57  * sh1mmerquit (Quit: sh1mmer)
03:15:20  <creationix>trevnorris: is there a known bug where libuv leaks environment variables when creating child processes (meaning you pass in a custom env and random other props from parent end up in child)
03:15:42  <creationix>I'm seeing this in node 0.10, node 0.11 (both on osx) and in luv and luvit on linux and osx
03:15:45  <creationix>but not node on linux
03:20:36  <creationix>though the output is different on each machine, so it might be a false negative on linux node
03:32:01  <creationix>ok, I think the issue is OSX inserting new stuff into fresh processes
03:36:05  * cjihrigjoined
03:48:35  * sh1mmerjoined
03:49:14  * c4milojoined
03:53:37  * c4miloquit (Ping timeout: 245 seconds)
03:55:39  * rosskquit (Remote host closed the connection)
03:59:54  * AvianFluquit (Quit: Leaving)
04:01:52  * rosskjoined
04:04:25  * kazuponquit (Remote host closed the connection)
04:04:44  * quijotejoined
04:06:49  * rosskquit (Remote host closed the connection)
04:07:23  * rosskjoined
04:09:20  * quijotequit (Ping timeout: 250 seconds)
04:11:33  * rosskquit (Ping timeout: 240 seconds)
04:12:33  * dsantiagoquit (Ping timeout: 272 seconds)
04:15:31  * dsantiagojoined
04:20:57  * dsantiagoquit (Read error: Connection reset by peer)
04:23:32  * mikealjoined
04:27:38  * dsantiagojoined
04:35:04  * mikealquit (Quit: Leaving.)
04:43:47  * txdvquit (Read error: Connection reset by peer)
04:44:02  * sh1mmerquit (Quit: sh1mmer)
04:44:13  * txdvjoined
04:44:48  * kazuponjoined
04:45:57  * sh1mmerjoined
04:47:00  <creationix>I see uv_read2_start is not in libuv master? Is there something to replace it?
04:49:56  * kazuponquit (Ping timeout: 260 seconds)
04:52:27  * sh1mmerquit (Quit: sh1mmer)
04:58:53  <creationix>does node still pass fd's internally for cluster?
04:59:26  * kazuponjoined
05:05:20  * quijotejoined
05:10:14  * quijotequit (Ping timeout: 260 seconds)
05:18:32  * dap_quit (Quit: Leaving.)
05:21:21  * mikealjoined
05:28:25  * wolfeidauquit (Remote host closed the connection)
05:28:39  * wolfeidaujoined
05:30:11  * sh1mmerjoined
05:37:33  * c4milojoined
05:42:26  * c4miloquit (Ping timeout: 260 seconds)
05:51:18  * rosskjoined
05:55:21  * bajtosjoined
06:03:10  * cjihrigquit (Quit: Leaving.)
06:06:17  * quijotejoined
06:09:03  * jreyno40joined
06:10:40  * quijotequit (Ping timeout: 250 seconds)
06:20:33  * rosskquit (Remote host closed the connection)
06:31:45  * rosskjoined
06:34:53  * daviddiasjoined
06:46:44  * janjongboomjoined
06:47:30  * rosskquit (Remote host closed the connection)
06:48:05  * rosskjoined
06:52:21  * rosskquit (Ping timeout: 240 seconds)
07:01:06  <saghul|afk>creationix: use uv_pipe_pending_count and uv_pipe_pending_type on the read callback
07:01:16  <saghul|afk>creationix: API changed, but the mechanism is the same
07:06:58  * quijotejoined
07:08:18  * rendarjoined
07:11:10  * quijote_joined
07:11:32  * quijotequit (Ping timeout: 245 seconds)
07:25:51  * c4milojoined
07:30:42  * c4miloquit (Ping timeout: 260 seconds)
07:32:44  * Soarez|afkchanged nick to Soarez
08:01:55  * daviddiasquit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
08:03:18  * Soarezchanged nick to Soarez|afk
08:03:57  * Soarez|afkchanged nick to Soarez
08:10:30  * Left_Turnjoined
08:10:52  * quijote_quit (Ping timeout: 240 seconds)
08:11:49  * WalrusPony1joined
08:13:08  * Soarezchanged nick to Soarez|afk
08:13:11  * WalrusPonyquit (Ping timeout: 255 seconds)
08:13:16  * WalrusPony2joined
08:13:24  * jreyno40quit (Quit: jreyno40)
08:15:07  * Soarez|afkchanged nick to Soarez
08:16:02  * WalrusPony1quit (Ping timeout: 245 seconds)
08:16:12  * saghul|afkchanged nick to saghul
08:16:35  * WalrusPonyjoined
08:17:57  * daviddiasjoined
08:17:58  * WalrusPony2quit (Ping timeout: 264 seconds)
08:20:38  * daviddiasquit (Read error: Connection reset by peer)
08:20:54  * WalrusPonyquit (Ping timeout: 246 seconds)
08:25:01  * WalrusPonyjoined
08:26:57  * Soarezchanged nick to Soarez|afk
08:27:42  * WalrusPony1joined
08:28:27  * abraxasquit (Remote host closed the connection)
08:29:09  * WalrusPonyquit (Ping timeout: 240 seconds)
08:29:59  * abraxasjoined
08:32:51  * Soarez|afkchanged nick to Soarez
08:32:58  * WalrusPonyjoined
08:34:32  * WalrusPony1quit (Ping timeout: 250 seconds)
08:34:38  * abraxasquit (Ping timeout: 260 seconds)
08:36:44  * quijotejoined
08:38:06  * abraxasjoined
08:58:31  * txdvquit (Read error: Connection reset by peer)
08:59:18  * inolenquit (Quit: Leaving.)
08:59:30  * txdvjoined
09:13:15  * inolenjoined
09:14:05  * c4milojoined
09:18:44  * c4miloquit (Ping timeout: 260 seconds)
09:27:11  * WalrusPony1joined
09:28:46  * WalrusPonyquit (Ping timeout: 264 seconds)
09:33:57  * inolenquit (Quit: Leaving.)
09:36:07  * petka_joined
09:52:58  * quijotequit (Ping timeout: 250 seconds)
09:57:22  * quijotejoined
09:58:35  * abraxasquit (Remote host closed the connection)
10:00:05  * mitsuhikoquit (Quit: ZNC - http://znc.sourceforge.net)
10:00:38  * bajtosquit (Quit: bajtos)
10:00:53  * abraxasjoined
10:03:24  * abraxasquit (Remote host closed the connection)
10:12:00  * mitsuhikojoined
10:12:34  * quijotequit (Ping timeout: 264 seconds)
10:13:17  * seishunjoined
10:39:41  * quijotejoined
10:39:42  <_3rdEden>trevnorris: would you accept a patch that greatly improves the performance os EE.once but might make .emit a tiny bit slower?
10:43:57  * quijotequit (Ping timeout: 240 seconds)
10:44:39  <txdv>what is EE.once and what is .emit
10:45:27  <txdv>o you are talking about node.js and not libuv
10:53:27  * karupachanged nick to zz_karupa
10:58:45  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
11:02:22  * c4milojoined
11:05:00  * EhevuTovjoined
11:07:10  * c4miloquit (Ping timeout: 264 seconds)
11:12:27  * bajtosjoined
11:40:31  * quijotejoined
11:44:58  * quijotequit (Ping timeout: 264 seconds)
11:52:39  * Soarezchanged nick to Soarez|afk
11:56:17  * kazuponquit (Remote host closed the connection)
12:09:54  * Soarez|afkchanged nick to Soarez
12:41:27  * quijotejoined
12:45:33  * quijotequit (Ping timeout: 240 seconds)
12:50:44  * c4milojoined
12:55:24  * c4miloquit (Ping timeout: 250 seconds)
13:04:23  * quijotejoined
13:08:22  * AvianFlujoined
13:31:22  * c4milojoined
13:36:56  * kazuponjoined
13:44:50  * cjihrigjoined
13:47:16  * c4miloquit (Remote host closed the connection)
13:53:42  * AvianFluquit (Quit: Leaving)
13:55:35  * EhevuTovquit (Quit: Leaving)
13:56:09  * c4milojoined
14:00:01  * Soarezchanged nick to Soarez|afk
14:00:26  * AvianFlujoined
14:00:43  * c4miloquit (Ping timeout: 255 seconds)
14:11:32  * Soarez|afkchanged nick to Soarez
14:19:01  <txdv>saghul: i dont understand jenkins
14:19:10  <txdv>are these failures new?
14:25:06  * quijotequit (Ping timeout: 250 seconds)
14:26:44  * Soarezchanged nick to Soarez|afk
14:27:21  <saghul>txdv: sorry, I missed your messages. Yes they were but everything check out now :-)
14:28:02  <txdv>jenkins always shows the total amount of failures
14:28:14  <txdv>so when you run it and something on smartos shows a failure im never sure if it is ok or not
14:28:16  <MI6>joyent/libuv: Andrius Bentkus master * d5e6f43 : unix: fix tcp write after bad connect freezing - http://git.io/NEs3hA
14:28:51  <saghul>txdv: the +- thing is with regards to the previous run
14:29:07  <saghul>you can click on each build and check which tests have failed
14:30:50  * bradleymeckjoined
14:31:39  * cjihrigquit (Quit: Leaving.)
14:37:32  <txdv>saghul: you said that windows doesn't support write queuing?
14:37:34  <txdv>before connect?
14:38:33  * quijotejoined
14:42:13  <saghul>txdv: yes, it doesn't
14:42:31  <saghul>txdv: btw, did you have a chance to test that patch for EAI constants?
14:43:07  <txdv>what patch
14:44:20  <saghul>txdv: ttps://github.com/joyent/node/blob/master/lib/buffer.js#L40-L45
14:44:25  <saghul>uh, not that
14:44:29  <saghul>txdv: https://github.com/saghul/libuv/commit/a7e8589342cc608edb0a5a3d5d4e5dd11101267c
14:44:36  <saghul>that ^
14:47:49  <txdv>yeah
14:47:51  <txdv>thats the fix
14:48:01  <txdv>i tried to define __GNU_something
14:48:12  <txdv>i dont think that we need netdb.h in uv-common.h anymore
14:48:19  <txdv>and we can move it also to that file
14:49:46  * Soarez|afkchanged nick to Soarez
14:50:29  <saghul>txdv: thanks for testing, somehow I don't get the problem myself
14:51:29  <MI6>joyent/libuv: Saúl Ibarra Corretgé master * 76cd676 : unix: fix exposing EAI_* glibc-isms - http://git.io/oZQCGg
14:51:57  <creationix>saghul: thanks!
14:53:03  <saghul>creationix: welcome!
14:53:17  <creationix>I was worried the feature was removed and it's a requirement for something I'm working on
14:54:19  * Soarezchanged nick to Soarez|afk
14:54:59  * kazuponquit (Remote host closed the connection)
14:55:27  * kazuponjoined
14:57:07  <txdv>saghul: https://github.com/joyent/libuv/blob/76cd67686c36e535660b449fb3bd4042084f12ce/src/uv-common.c#L35-L40
14:57:12  <txdv>are these at the right placE?
14:57:58  <txdv>i think they belong in the same commit
14:58:07  <txdv>but you pushed without a lgtm
15:00:06  * Soarez|afkchanged nick to Soarez
15:04:33  <saghul>ups :-/ it does work, but they should probably be on the right place.
15:04:37  <saghul>https://github.com/saghul/libuv/tree/eai-error2 should do
15:04:48  <saghul>I'll verify a bit later
15:05:18  <txdv>on windows?
15:06:59  <saghul>on both, i didn't even try to compile it, I put it up there so I remember
15:12:20  <txdv>I wonder though why we have to stop the IO thingy before the connection callback
15:12:37  <txdv>it is actually a good idea to check after the connection callback whether the queue is empty or not
15:12:41  <txdv>ill have to check it out
15:13:25  <txdv>why should we even stop
15:13:33  <txdv>it is edge triggered
15:15:53  * cjihrigjoined
15:21:44  * AvianFluquit (Quit: Leaving)
15:24:09  * bajtosquit (Quit: bajtos)
15:27:02  * Soarezchanged nick to Soarez|afk
15:34:26  * inolenjoined
15:37:23  * rosskjoined
15:38:50  * inolenquit (Ping timeout: 260 seconds)
15:39:49  * kazuponquit (Remote host closed the connection)
15:40:32  * octetcloudjoined
15:54:13  * rmgjoined
16:02:39  * Soarez|afkchanged nick to Soarez
16:02:43  * jgijoined
16:07:59  * bradleymeckquit (Quit: bradleymeck)
16:10:24  * bradleymeckjoined
16:12:17  * cjihrigquit (Quit: Leaving.)
16:14:04  * inolenjoined
16:15:36  * brsonjoined
16:18:17  * rosskquit (Remote host closed the connection)
16:18:53  * rosskjoined
16:23:09  * rosskquit (Ping timeout: 240 seconds)
16:25:09  * Ralithquit (Ping timeout: 240 seconds)
16:26:32  * cjihrigjoined
16:37:59  * cjihrigquit (Quit: Leaving.)
16:43:41  * dap_joined
16:46:15  * mikealquit (Quit: Leaving.)
16:48:06  * Ralithjoined
16:48:37  * daviddiasjoined
16:50:46  * daviddiasquit (Client Quit)
16:51:09  * daviddiasjoined
17:00:23  * dshaw_joined
17:13:06  * rosskjoined
17:17:47  * bradleymeckquit (Quit: bradleymeck)
17:22:32  * wavdedjoined
17:24:32  * mikealjoined
17:28:56  <rendar>saghul: windows doesn't support write queuing before connect? it means that you cannot enqueue a write, before ConnectEx() OVERLAPPED packet is completed?
17:29:24  * bradleymeckjoined
17:33:43  * Soarezchanged nick to Soarez|afk
17:37:14  * rosskquit (Remote host closed the connection)
17:44:59  * kazuponjoined
17:45:21  * mikealquit (Quit: Leaving.)
17:52:43  * sh1mmerquit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
17:56:19  * wavdedquit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
17:56:32  * quijotequit (Ping timeout: 255 seconds)
18:02:14  * kazuponquit (Remote host closed the connection)
18:02:33  <indutny>rendar: yes
18:02:40  <indutny>rendar: this is why it doesn't work on unixes too
18:02:51  <indutny>we should return EINPROGRESS
18:05:50  * sh1mmerjoined
18:06:23  * petka_quit (Quit: Connection closed for inactivity)
18:07:39  * quijotejoined
18:11:08  * rosskjoined
18:12:24  * wavdedjoined
18:16:49  * txdvquit (Read error: Connection reset by peer)
18:17:50  * txdvjoined
18:37:33  * quijotequit (Ping timeout: 240 seconds)
18:38:45  <creationix>saghul: do you know if I have to let libuv create the pipes when creating child processes or can I create my own and pass them in to spawn's stdio option?
18:39:09  <creationix>I'm getting "spawn: invalid argument" trying to pass them in
18:39:37  * a_lejoined
18:49:05  * brsonquit (Ping timeout: 272 seconds)
18:53:25  * jgiquit (Quit: jgi)
18:53:45  * mikealjoined
19:04:37  * rendar_joined
19:05:27  <indutny>tjfontaine: i'm going to figure out arm build problems
19:05:29  <indutny>on v0.10
19:05:33  * rendarquit (Ping timeout: 240 seconds)
19:05:33  <indutny>after that and that v8 fi
19:05:34  <indutny>fix
19:05:42  <indutny>it might be a good idea to do another release
19:08:08  * dshaw_quit (Quit: Leaving.)
19:08:18  * mikealquit (Quit: Leaving.)
19:08:44  * dsantiagoquit (Quit: Computer has gone to sleep.)
19:09:09  * jgijoined
19:10:05  * daviddiasquit (Quit: Textual IRC Client: www.textualapp.com)
19:11:08  * mikealjoined
19:12:14  * yunongjoined
19:12:59  * kazuponjoined
19:13:09  <rendar_>indutny: i see, so the thing is in unix you can call successfully write(2) after connect(2) immediately returns, then the data you write there, will be put in the kernel's socket buffer, instead on windows you can't do this, you can call WSASend() only *AFTER* it completes the connect request
19:15:36  <saghul>creationix: you can do both, but you need to specify how you want to pass them
19:15:59  <saghul>also, on Windows only passing TCP and UDP handles is supported, not pipes
19:16:16  <creationix>saghul: this is linux only. I'm trying to talk to the fuse interface in the linux kernel
19:16:59  <creationix>fusermount is a setuid process that opens /dev/fuse and gives me the fd over an ipc pipe
19:17:31  * kazuponquit (Ping timeout: 255 seconds)
19:17:35  <creationix>so I'm trying to get a fd passed up from a child
19:17:48  <saghul>creationix: aha
19:18:22  <saghul>then you could just create a pipe with UV_CREATE_PIPE | UV_READABLE_PIPE | UV_WRITABLE_PIPE
19:18:30  <creationix>is that the only way?
19:18:42  <creationix>I'd rather pass in an existing pipe if possible
19:18:48  * mikealquit (Quit: Leaving.)
19:18:49  <creationix>but if not, I'll just rearrange my logic
19:18:57  <saghul>no, if you already have the uv_pipe_t object, you can use UV_INHERIT_STREAM
19:19:14  <saghul>or if you just have a fd: UV_INHERIT_FD
19:19:22  <creationix>hmm, I wonder why it's telling me "spawn: invalid argument"
19:19:29  <saghul>but this for passing a fd from the parent to the child
19:19:56  <creationix>fusermount says to create a socket pair and pass one end via env variable
19:20:18  <saghul>yeah, you can do that too
19:20:37  <saghul>there are only 2 code paths leading to EINVAL explicitly
19:20:49  <saghul>https://github.com/joyent/libuv/blob/master/src/unix/process.c#L208 and https://github.com/joyent/libuv/blob/master/src/unix/process.c#L220
19:21:04  <saghul>what flags are you using in the stdio container?
19:22:24  <rendar_>saghul: since in windows you can't fork(), if you use UV_INHERIT_FD, how actually the child process will have the same fd number somewhere? the memory won't be shared as happen with fork()
19:23:04  <creationix>I'm trying just UV_INHERIT_STREAM with the pipe as the data.stream
19:23:21  <creationix>the pipe is created with ipc true
19:23:44  <saghul>creationix: is it bound or opened?
19:23:44  <creationix>but passing any pipe causes the same error
19:23:50  <creationix>the pipe, no
19:24:10  <saghul>creationix: it needs to have an fd, otherwise you cannot pass it
19:24:28  <creationix>right, so new pipes don't have an fd on their own. That's what I didn't know
19:24:36  <creationix>I'm trying to port this code https://github.com/hanwen/go-fuse/blob/master/fuse/mount_linux.go#L18
19:24:44  <creationix>but I don't think libuv can create a fd pair like this?
19:25:03  * zz_karupaquit (Ping timeout: 252 seconds)
19:25:22  <saghul>creationix: no, but you can do it easily:
19:25:33  <saghul>create the socketpair yourself, put it non-blocking
19:25:42  <saghul>and pass one end to uv_pipe_open
19:25:56  <saghul>then your pipe will use that fd
19:26:45  <saghul>rendar_: TBH I'm not sure :-/
19:27:35  <creationix>ok, I can create a pair with `socketpair(AF_UNIX, SOCK_SEQPACKET, 0, fds)`, I'll look up how to set it non-blocking
19:27:37  <saghul>creationix: now, if you are going to be creating a SOCK_SEQPACKET type of socket, you might need to use uv_udp_t instead since it's more similar to the datagram interface
19:27:50  <indutny>creationix: setsockopt ftw
19:28:02  <rendar_>well, isn't creationix trying to create a child process which inherits a socketpair from the parent process?
19:28:04  <creationix>I don't think SEQPACKET is required, that's just what the go library did
19:28:05  * seldojoined
19:28:06  * zz_karupajoined
19:28:06  * zz_karupachanged nick to karupa
19:28:25  <creationix>rendar_: no, the seduid child passes up a protected fd to the parent
19:28:31  <creationix>it checks to see if I have permissions
19:28:32  * seldoquit (Remote host closed the connection)
19:28:36  <saghul>creationix: then what would you use, SOCK_STREAM or SOCK_DGRAM?
19:28:43  <indutny>ah way
19:28:44  <rendar_>creationix: oh ok
19:28:45  <indutny>wait
19:28:45  * seldojoined
19:28:48  <indutny>creationix: ioctl
19:28:53  <indutny>creationix: FIONBIO
19:29:02  <creationix>http://ptspts.blogspot.com/2009/11/fuse-protocol-tutorial-for-linux-26.html says to use SOCK_STREAM
19:29:06  <creationix>so I can try that
19:29:16  <rendar_>creationix: what you mean with "procted" fd?
19:29:20  <rendar_>protected* sorry
19:29:40  <creationix>rendar_: I'm trying to access /dev/fuse so I can talk to the kernel, but only root processes are allowed to access it
19:29:53  <saghul>creationix: if your linux is recent enough you can even create the socketpair non-blocking in a single go, or-ing SOCK_NONBLOCK with the type
19:29:56  <creationix>fusermount is setuid and is the gatekeeper
19:30:26  <saghul>linux >= 2.6.27 it seems
19:31:14  <creationix>thanks
19:32:02  <rendar_>creationix: got it
19:34:39  <rendar_>well, btw, in libuv isn't possible to create a socketpair, and make the child process to inherit that socketpair, becuase libuv dones't abstract fork(), of course..so, the process you run is started from scratch, it will be a fork()+exec(), right?
19:35:37  * quijotejoined
19:36:04  * mikealjoined
19:36:11  <saghul>rendar_: you can make the child process inherit one end of the socketpair, if you don't set CLOEXEC in it AFAICT
19:37:17  * quijotequit (Read error: Connection reset by peer)
19:37:21  * rosskquit (Remote host closed the connection)
19:37:27  * quijote_joined
19:37:56  * rosskjoined
19:38:08  <creationix>saghul: that's what I'm trying. I'm passing one half of the pair as stdio[3] and telling fusermount to use FD 3
19:38:33  <creationix>so now I just need to use uv_read2_start I think (using older libuv)
19:38:56  <creationix>so for the local fd I just wrap in a uv_pipe_t right?
19:38:58  <rendar_>saghul: yah, but you have to call fork() to do that, so it became a non-portable code
19:39:07  <rendar_>saghul: i meant portable code which will work also on Windows
19:39:34  * wavdedquit (Ping timeout: 264 seconds)
19:39:46  <saghul>rendar_: ah, windows, that guy... :-P
19:40:08  <creationix>rendar_: I'm fine with linux only since I'm writing code to talk to the linux kernel :P
19:40:23  <creationix>if there is a fuse for windows, I'm sure it works quite different
19:40:50  * quijotejoined
19:41:23  <rendar_>creationix: yeah right
19:41:26  <rendar_>saghul: lol
19:41:57  * rosskquit (Ping timeout: 240 seconds)
19:42:16  * quijote_quit (Ping timeout: 255 seconds)
19:43:01  <saghul>it would be cool to have some portable uv_pipepair(loop, pipe1, pipe2) thing
19:44:20  * wavdedjoined
19:45:07  * brsonjoined
19:45:09  * quijotequit (Ping timeout: 246 seconds)
19:45:17  <saghul>rendar_: https://github.com/ncm/selectable-socketpair
19:46:12  <creationix>saghul: so if I go the UV_CREATE_PIPE route, how do I set the IPC flag on the new pipe?
19:46:42  <creationix>I wonder what node's "ipc" does
19:50:34  <saghul>creationix: you need to initialize the pipe with uv_pipe_init
19:50:38  <saghul>so you set it there
20:00:17  * mikealquit (Quit: Leaving.)
20:00:46  * mikealjoined
20:24:22  * AvianFlujoined
20:40:08  * inolenquit (Quit: Leaving.)
20:41:36  * quijotejoined
20:46:46  * quijotequit (Ping timeout: 264 seconds)
20:48:33  * seldoquit (Remote host closed the connection)
20:48:48  * seldojoined
20:49:57  * dsantiagojoined
20:55:04  * seldoquit (Remote host closed the connection)
20:55:28  * seldojoined
20:57:24  * rosskjoined
21:00:16  * seldoquit (Remote host closed the connection)
21:01:11  * seldojoined
21:05:03  <jgi>octetcloud: Do you think creating a GitHub issue like “add test case for Worker.prototype.destroy when called within worker or remove method” is overkill? I think it’s worth it, otherwise we’ll be bitten by the same issue as I encountered (uncaught bug) in the future.
21:14:21  * sblomjoined
21:14:31  * seldoquit (Remote host closed the connection)
21:15:25  * seldojoined
21:16:34  * seldoquit (Remote host closed the connection)
21:17:34  * seldojoined
21:21:24  * yunongquit (Ping timeout: 246 seconds)
21:22:45  * brsonquit (Quit: Lost terminal)
21:23:46  * daviddiasjoined
21:27:13  * yunongjoined
21:28:25  * seldoquit (Remote host closed the connection)
21:30:23  * seldojoined
21:31:11  * yunong_joined
21:33:21  <MI6>joyent/libuv: Saúl Ibarra Corretgé master * a87619c : unix, windows: move includes for EAI constants - http://git.io/qOqP6A
21:34:22  * dap_quit (Ping timeout: 240 seconds)
21:34:46  * yunongquit (Ping timeout: 255 seconds)
21:37:30  * dap_joined
21:40:57  <MI6>joyent/libuv: Saúl Ibarra Corretgé master * 2f83613 : test: fix watcher_cross_stop on Windows - http://git.io/KuYTow
21:42:22  * quijotejoined
21:42:59  * seldoquit (Remote host closed the connection)
21:43:20  * seldojoined
21:46:46  * quijotequit (Ping timeout: 250 seconds)
21:52:48  * geekintimejoined
21:53:07  <geekintime>What priority in the message loop are callbacks from C++ given? Is it possible to give them i/o priority?
21:53:15  <geekintime>re:nodejs
21:54:31  * a_lequit (Remote host closed the connection)
21:55:08  * a_lejoined
21:56:31  * mikealquit (Quit: Leaving.)
21:58:29  * seldoquit (Remote host closed the connection)
22:03:00  * daviddiasquit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
22:03:15  * brsonjoined
22:06:15  * wavdedquit (Quit: Textual IRC Client: www.textualapp.com)
22:08:10  * bradleymeckquit (Quit: bradleymeck)
22:08:16  * yunongjoined
22:11:32  * yunong_quit (Ping timeout: 260 seconds)
22:12:33  * bradleymeckjoined
22:13:08  * a_lequit (Remote host closed the connection)
22:16:57  * seldojoined
22:31:31  * seldoquit (Remote host closed the connection)
22:31:52  <bradleymeck>geekintime: nope
22:32:02  * seldojoined
22:32:06  <bradleymeck>spawn a diff thread and queue results
22:38:40  * seishunquit (Disconnected by services)
22:38:46  * seishunjoined
22:40:46  * bradleymeckquit (Quit: bradleymeck)
22:43:32  * quijotejoined
22:43:58  <geekintime>bradleymeck: seriously? they all just go in the loop at lowest priority?
22:47:06  * Soarez|afkchanged nick to Soarez
22:47:07  <geekintime>bradleymeck: I appreciate the feedback. I assume this means that node.js handles i/o code in a manner that modules cannot?
22:47:52  * quijotequit (Ping timeout: 240 seconds)
22:49:51  * seldoquit (Remote host closed the connection)
22:51:16  * seldojoined
22:54:50  <geekintime>bradleymeck: rather not spawn a thread. Operating in a resource limited device. Thanks for the thought though. Maybe I will go down that road.
22:55:52  * seldoquit (Ping timeout: 260 seconds)
22:57:02  * a_lejoined
23:02:17  * Soarezchanged nick to Soarez|afk
23:03:09  * isaacs_changed nick to isaacs
23:06:49  * rendar_quit
23:09:52  * seishunquit (Ping timeout: 250 seconds)
23:11:54  * a_lequit (Remote host closed the connection)
23:14:03  * seldojoined
23:16:33  * a_lejoined
23:19:01  <trevnorris>_3rdEden: depends. .emit() is already 3x slower in v0.11 than it was in v0.10. though figuring out what's going on will have to wait until core code stabilizes enough for the v0.12 release.
23:30:25  * a_lequit (Remote host closed the connection)
23:35:21  <MI6>joyent/node: Jackson Tian v0.12 * c0f30f6 : http: avoid create difference hidden class - http://git.io/8ZcB0A
23:36:55  * seldoquit (Remote host closed the connection)
23:38:13  * seldojoined
23:40:28  * seldoquit (Remote host closed the connection)
23:41:14  * dsantiagoquit (Quit: Computer has gone to sleep.)
23:44:00  * quijotejoined
23:48:32  * quijotequit (Ping timeout: 245 seconds)
23:49:13  * mikealjoined
23:50:03  * seldojoined
23:53:57  * mikealquit (Client Quit)
23:57:04  * mikealjoined