00:09:22  * isaacsquit (Remote host closed the connection)
00:18:15  * mikealquit (Quit: Leaving.)
00:28:58  * isaacsjoined
00:29:56  * mikealjoined
00:36:01  * isaacsquit (Ping timeout: 244 seconds)
00:55:07  * ericktquit (Ping timeout: 244 seconds)
01:05:25  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
01:25:24  * c4miloquit (Remote host closed the connection)
01:26:07  * pieternquit (Quit: pietern)
01:26:11  * c4milojoined
01:29:25  * dshaw_quit (Quit: Leaving.)
01:33:06  * ericktjoined
01:37:32  * ryahquit (Quit: leaving)
01:40:36  * AvianFlupart ("Leaving")
01:41:38  * ericktquit (Ping timeout: 260 seconds)
01:54:40  * theColejoined
01:56:32  * abraxasjoined
02:11:11  * TooTallNatequit (Ping timeout: 244 seconds)
02:18:29  * skmpyjoined
02:18:38  * brsonquit (Ping timeout: 256 seconds)
02:18:42  * perezdquit (Quit: perezd)
02:22:31  * brsonjoined
02:25:18  * ericktjoined
02:28:08  * skmpypart
02:29:03  * Ariajoined
02:33:44  * brsonquit (Ping timeout: 246 seconds)
02:35:57  * brsonjoined
02:40:10  * Ariaquit (Quit: Leaving...)
02:41:19  * Ariajoined
02:41:21  * loladiroquit (Remote host closed the connection)
02:55:43  * iraquit (Quit: Leaving...)
02:57:48  * brsonquit (Quit: leaving)
02:59:23  * bnoordhuisquit (Ping timeout: 240 seconds)
03:13:36  * AvianFlujoined
03:20:57  * Ariaquit (Remote host closed the connection)
03:35:51  * brsonjoined
03:36:22  * TooTallNatejoined
03:36:33  * brson_joined
03:38:03  * brson_quit (Client Quit)
03:38:12  * brson_joined
03:40:03  * brsonquit (Ping timeout: 244 seconds)
03:51:30  * bulatshakirzyanojoined
04:03:51  * c4miloquit (Remote host closed the connection)
04:44:23  * ericktquit (Quit: erickt)
04:53:51  * mikealquit (Quit: Leaving.)
05:11:07  * felixgejoined
05:11:07  * felixgequit (Changing host)
05:11:08  * felixgejoined
05:12:19  * felixge_joined
05:12:19  * felixge_quit (Changing host)
05:12:20  * felixge_joined
05:14:21  * benviejoined
05:15:22  * felixgequit (Ping timeout: 245 seconds)
05:15:23  * felixge_changed nick to felixge
05:16:56  * TooTallNatequit (Quit: Textual IRC Client: www.textualapp.com)
05:21:51  * isaacsjoined
05:23:16  * benviequit (Ping timeout: 252 seconds)
05:23:40  * mikealjoined
05:29:06  * benviejoined
05:41:06  * mikealquit (Quit: Leaving.)
05:56:07  * isaacsquit (Remote host closed the connection)
06:11:19  * brson_quit (Quit: leaving)
06:26:33  * mikealjoined
06:39:32  * brsonjoined
06:44:58  * brsonquit (Quit: leaving)
06:51:09  * stephankquit (Quit: *Poof!*)
06:59:33  * paddybyersjoined
07:16:06  * \toothrotjoined
07:16:50  * toothrquit (Ping timeout: 248 seconds)
07:26:57  * rendarjoined
07:50:24  * mmaleckijoined
07:56:12  * \toothrotquit (Ping timeout: 244 seconds)
07:56:34  * brsonjoined
07:56:36  * toothrjoined
08:14:07  * bulatshakirzyanoquit (Quit: Computer has gone to sleep.)
08:18:37  * hij1nxjoined
08:20:31  * hij1nxquit (Client Quit)
08:25:44  * felixgequit (Quit: felixge)
08:40:52  * mmalecki_joined
08:43:48  * dshaw_joined
08:43:58  * mmaleckiquit (Ping timeout: 256 seconds)
08:47:22  * brsonquit (Quit: leaving)
08:50:52  * theColequit (Read error: Connection reset by peer)
08:51:32  * theColejoined
08:57:59  * dshaw_quit (Quit: Leaving.)
09:25:42  * mikealquit (Ping timeout: 244 seconds)
09:29:14  * mmalecki_quit (Ping timeout: 260 seconds)
09:32:59  * mikealjoined
09:45:27  * mmaleckijoined
09:49:09  * hij1nxjoined
09:53:40  * hij1nxquit (Client Quit)
10:47:50  <indutny>v8 profiler is a bit magic
10:48:02  * hij1nxjoined
10:58:06  * mmalecki_joined
11:00:14  * mmaleckiquit (Ping timeout: 252 seconds)
11:06:58  * mmalecki_quit (Ping timeout: 244 seconds)
11:07:48  * irajoined
11:08:24  * mmaleckijoined
11:08:49  * abraxasquit (Remote host closed the connection)
11:40:28  * felixgejoined
11:40:28  * felixgequit (Changing host)
11:40:28  * felixgejoined
12:25:13  * TheJHjoined
12:27:53  * mmaleckiquit (Ping timeout: 240 seconds)
12:57:49  * mmaleckijoined
13:25:25  * bnoordhuisjoined
13:26:51  * piscisaureus_joined
13:35:13  * bitprobejoined
13:36:12  * bitprobequit (Client Quit)
13:36:21  <bnoordhuis>piscisaureus_: does msvc understand 42ULL?
13:36:38  <piscisaureus_>bnoordhuis: yep
13:36:48  <bnoordhuis>okay, cool
13:37:12  <piscisaureus_>bnoordhuis: you should realize though that you might as well write `43`
13:37:17  <piscisaureus_>er, 42
13:37:36  <bnoordhuis>of course :)
13:38:02  <indutny>bnoordhuis: hope that's not related to that quiz
13:38:08  <indutny>:D
13:38:11  <piscisaureus_>bnoordhuis: so you are going to assume that unsigned long long ~ uint64_t on all platforms
13:38:19  <piscisaureus_>?
13:38:31  <bnoordhuis>that it's at least 64 bits
13:38:41  <bnoordhuis>which it should be, that's the whole point of unsigned long long
13:39:15  <piscisaureus_>I wonder if, once we switch to 128 bits platforms, there will be unsigned long long long
13:39:22  <bnoordhuis>probably :)
13:39:45  <bnoordhuis>i kid, i hope compiler vendors and programmers have learned their lesson by then
13:41:00  <piscisaureus_>yeah hopefully it will just just be uint_least128_t then
13:42:05  <indutny>is there any "real" reasons for switching to 128bits?
13:42:22  <bnoordhuis>only for some very particular workloads
13:42:28  <piscisaureus_>indutny: what could be an unreal reason?
13:42:39  <bnoordhuis>some crypto algorithms can benefit from 128 bits wide registers
13:42:44  <piscisaureus_>I mean, 64 bit address space is kinda crampy innit ?
13:43:17  <indutny>piscisaureus_: well 64bit is enough for addressing
13:43:31  <piscisaureus_>that's what the guys that invented 32 bit said :-p
13:43:35  <indutny>hahahaha
13:43:53  <indutny>well, it's exponential thing
13:44:00  <piscisaureus_>although, assuming that memory keeps groing exponentially it will take twice as long
13:44:07  <indutny>so I presume we'll be old and retarted once it'll arrive
13:44:13  <indutny>piscisaureus_: that's it
13:44:19  <indutny>twice?
13:44:23  <piscisaureus_>no, 4 actually
13:44:34  <piscisaureus_>from 16 -> 32 was +16 bits
13:44:40  <piscisaureus_>oh yeah
13:44:43  <piscisaureus_>twice as logn
13:44:57  <indutny>ah, yes
13:44:59  <indutny>you're right
13:45:04  <indutny>how much did that took?
13:45:12  <indutny>8-10 years?
13:45:19  <bnoordhuis>doesn't it depend on the exponent?
13:45:40  <bnoordhuis>if it's 1.0000000001, we still have some time to go
13:45:53  <piscisaureus_>bnoordhuis: the assumption is that the exponent stays the same
13:46:09  <bnoordhuis>assumption is the mother of all screwups
13:46:52  <piscisaureus_>bnoordhuis: the exponent is defined by moore's law
13:47:07  * c4milojoined
13:47:56  <indutny>piscisaureus_: moore law is crap
13:48:02  <indutny>he's not a hipster
13:48:39  <piscisaureus_>indutny: 80386: introduced in 1985. Itanium: 2001
13:48:43  <piscisaureus_>indutny: so that makes 16 years
13:48:44  <indutny>ok, 16 years
13:48:57  <indutny>lets wait for 2033 then :P
13:49:01  <piscisaureus_>er
13:49:01  <indutny>and see if piscisaureus_ was right
13:49:02  <piscisaureus_>*2
13:49:09  <piscisaureus_>oh sure
13:49:22  * mmalecki_joined
13:49:26  <piscisaureus_>see you here in 32 years
13:49:34  <indutny>piscisaureus_: in this room
13:49:39  <piscisaureus_>sure
13:49:47  <piscisaureus_>On #skynet I'm afraid
13:49:56  <indutny>when node.js will reach version 128.0.0
13:50:04  <indutny>127.0.0.1 hahahah
13:50:08  <piscisaureus_>let's verify with moore's law
13:50:34  <piscisaureus_>it doesn't really work with moore's law, actually
13:50:53  <indutny>yeah, our release dates are inpredictable
13:51:11  <piscisaureus_>that would predict that in 32 year's we'll only need +24 bits
13:51:30  <indutny>I hope in 2033 quantum computer will be able to work with gbs of qbits
13:51:52  <indutny>and we'll be able to build node on it
13:51:57  <indutny>and auto-fix tests
13:52:08  <indutny>by trying all variants of small changes to source files
13:52:18  * mmaleckiquit (Ping timeout: 244 seconds)
13:52:37  <indutny>piscisaureus_: I can believe in that (+24 bits)
13:52:57  <indutny>piscisaureus_: +64 is less realistic, and to my mind more useless
13:53:28  <piscisaureus_>indutny: sure. But if we grow +33 bits we're in trouble right
13:53:39  <piscisaureus_>since then we would need 65-bit pointers
13:56:04  <indutny>hahaha
13:56:24  <bnoordhuis>we can use the parity bit of my ECC memory for that
13:56:32  <piscisaureus_>you still have that?
13:56:45  <piscisaureus_>indutny: actually there may be a flaw in our computation
13:57:06  <bnoordhuis>many (most?) high-end servers have
13:58:25  * bitprobejoined
13:59:15  <piscisaureus_>indutny: since before the 386 there was no flat memory model - it was possible to use more than 16 bits
13:59:28  <indutny>24bit addressation
13:59:34  <indutny>oh, not really
13:59:35  <indutny>25bit
13:59:45  <piscisaureus_>so 1985...2001 should be more like +
13:59:48  <piscisaureus_>8
13:59:52  <indutny>7
13:59:53  <piscisaureus_>yeah
13:59:53  <indutny>;)
14:00:39  <piscisaureus_>so that makes it take until 2074 :-)
14:01:45  <CIA-112>libuv: Shigeki Ohtsu master * re2aa39a / test/test-hrtime.c : test: change uv_hrtime() test to iterate upto 100 - http://git.io/8v1Grg
14:01:46  <CIA-112>libuv: Ben Noordhuis master * rc639f34 / (6 files in 2 dirs): unix, test: make NANOSEC unsigned long long - http://git.io/3eI7_w
14:03:31  <bnoordhuis>so... should i bite the bullet and drop libev altogether before 0.8?
14:03:49  * travis-cijoined
14:03:49  <travis-ci>[travis-ci] joyent/libuv#390 (master - c639f34 : Ben Noordhuis): The build is still failing.
14:03:49  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/b47af98...c639f34
14:03:49  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/1557770
14:03:49  * travis-cipart
14:04:02  <piscisaureus_>bnoordhuis: could you make it happen?
14:04:08  <bnoordhuis>yes, quite easily
14:04:12  <piscisaureus_>bnoordhuis: that means uv_signal + uv_child, you realize that right?
14:04:38  <piscisaureus_>bnoordhuis: plus epoll + port + kqueue backend
14:04:51  <bnoordhuis>piscisaureus_: yes. that are the only things libev still does
14:05:03  <piscisaureus_>bnoordhuis: oh, you replaced ev_stat now ?
14:05:30  <bnoordhuis>mostly. i should finish that sometime this week
14:05:54  <piscisaureus_>bnoordhuis: cool. Is it cross-platform ?
14:05:57  <indutny>piscisaureus_: so what are we filling our 64 bit memory nowadays
14:06:03  <bnoordhuis>piscisaureus_: yes
14:06:12  <indutny>piscisaureus_: like 64gbs of RAM
14:06:23  <indutny>piscisaureus_: many people are filling them with JVM's garbage
14:06:24  <indutny>:D
14:06:29  <indutny>very useful upgrade
14:06:39  <bnoordhuis>indutny: only 48 address lines are actually used on today's hardware
14:06:52  <indutny>bnoordhuis: well, you may have different processes
14:07:05  <indutny>bnoordhuis: i.e. use different pages
14:07:21  <indutny>or no?
14:07:52  <bnoordhuis>indutny: no, i mean that you can only address 48 bits
14:08:03  <bnoordhuis>the hardware doesn't support addressing anything beyond that
14:08:12  <bnoordhuis>48 bits is still a ton load of memory though
14:10:44  <indutny>indeed
14:10:58  <indutny>a lot of memory that is really wasted
14:11:11  <indutny>when we had only 25 bits of address space - memory usage was more clever
14:11:24  <piscisaureus_>a lot of virtual address space
14:11:26  <piscisaureus_>now who cares
14:11:39  <indutny>ok, I'm going to trash my macbook and buy some old computer
14:11:58  <piscisaureus_>I don't agree
14:12:11  <piscisaureus_>the TLB is an amazing piece of hardware - use it every way you can
14:12:25  <piscisaureus_>bnoordhuis kom er maar in met je grapje
14:14:01  <bnoordhuis>piscisaureus_: i never jest, you know that
14:20:40  * isaacsjoined
14:22:15  <piscisaureus_>More fun with windows paths
14:22:15  <piscisaureus_>process.chdir('\\\\.\\pipe');
14:22:23  <piscisaureus_>process.binding('fs').readdir('..\\..\\c:')
14:22:29  <piscisaureus_>^-- works
14:32:50  <isaacs>piscisaureus_: wtf?
14:32:55  <isaacs>you can cd into non-directories?
14:32:59  <isaacs>windows, you so crazy!!
14:33:22  <piscisaureus_>isaacs: well "pipe" is not really a directory but it is similar to a directory
14:33:28  <piscisaureus_>a directory that contains named pipes
14:33:45  <piscisaureus_>it's fun tho, I never realized you could list pipe names with that
14:34:07  <piscisaureus_>process.binding('fs').readdir('\\\\.\\pipe')
14:43:15  * mmalecki_quit (Ping timeout: 256 seconds)
14:48:06  * theColequit (Quit: theCole)
14:53:53  * isaacsquit (Remote host closed the connection)
15:01:15  * xaqjoined
15:01:59  <CIA-112>libuv: Ben Noordhuis master * r59cda86 / (7 files in 3 dirs): unix, test: make NANOSEC a 64 bits unsigned int - http://git.io/1OleNA
15:02:00  <CIA-112>libuv: Ben Noordhuis master * r053d3af / include/uv-private/ngx-queue.h : Make ngx_queue_foreach() impl more explicit. - http://git.io/6BR3eQ
15:07:13  * ericktjoined
15:09:34  * travis-cijoined
15:09:34  <travis-ci>[travis-ci] joyent/libuv#391 (master - 053d3af : Ben Noordhuis): The build is still failing.
15:09:34  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/c639f34...053d3af
15:09:34  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/1558531
15:09:34  * travis-cipart
15:11:29  * paddybyersquit (Ping timeout: 246 seconds)
15:12:08  * pieternjoined
15:32:25  <indutny>piscisaureus_: is a call in 30 minutes?
15:32:32  <piscisaureus_>indutny: yeah
15:49:25  <piscisaureus_>ok. We have standup now, join you guys after that
15:49:32  <piscisaureus_>isaacs: ^
15:50:27  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
15:51:46  * isaacsjoined
15:53:48  * hij1nxquit (Read error: Connection reset by peer)
15:54:45  * paddybyersjoined
15:55:02  * piscisaureus_joined
15:57:11  * TooTallNatejoined
15:58:23  <isaacs>piscisaureus_: TooTallNate, bnoordhuis, igorzi, indutny: skytp in 5 minutes?
15:58:28  <isaacs>s/skytp/skype
15:58:32  <indutny>isaacs: sounds good
15:58:37  <TooTallNate>yup yup
15:58:51  <indutny>piscisaureus_ and bnoordhuis are on their own standup
15:59:25  <piscisaureus_>bnoordhuis is slacking actually
15:59:51  <piscisaureus_>but yeah we can start within 5 minutes as far as I am concerned
15:59:56  * ibcjoined
16:01:03  <ibc>hi, I will use a own ngx_queue to store handles. I want to add them at the end of the queue, which function should I use?
16:01:04  <ibc>1) ngx_queue_insert_tail()
16:01:04  <ibc>2) ngx_queue_add()
16:03:20  <bnoordhuis>ibc: ngx_queue_insert_tail
16:03:37  <bnoordhuis>ngx_queue_add appends two queues
16:03:47  <ibc>clear, thanks a lot
16:04:50  <indutny>piscisaureus_: where are ya?
16:09:02  <bnoordhuis>skype crashed...
16:09:38  <piscisaureus_>bnoordhuis: come baaaack
16:09:44  <bnoordhuis>i'm online again
16:14:01  * bulatshakirzyanojoined
16:16:55  * mmaleckijoined
16:26:24  <piscisaureus_>net.createConnection({fd: 4})
16:26:32  <piscisaureus_>net.createServer({fd: 4})
16:29:54  <isaacs>https://github.com/joyent/node/issues/3388
16:31:09  <bnoordhuis>unix people will be thrilled
16:31:30  <bnoordhuis>for a second anyway, then they'll find something else to bitch about
16:32:50  <piscisaureus_>isaacs: https://github.com/joyent/node/issues/3388#issuecomment-6180699
16:32:52  <piscisaureus_>yes
16:33:09  <piscisaureus_>probably the lack of proper PTYs
16:33:27  <piscisaureus_>(how hard would it to get that in btw, now that we have the stdio argument)
16:33:50  <piscisaureus_>After that, they will complain about the absence of an automake script
16:35:00  <bnoordhuis>hah
16:35:24  <bnoordhuis>i have Opinions about people that like autotools
16:35:51  <bnoordhuis>re ptys: someone that wants to use them is halfway there
16:36:40  <bnoordhuis>you still need to create the actual pty but we don't have to accommodate that, that's what add-ons are for
16:36:56  <bnoordhuis>okay, dinner time
16:38:40  * mmaleckiquit (Ping timeout: 252 seconds)
16:48:47  * TheJHquit (Ping timeout: 246 seconds)
16:48:50  <dap>it looks like the eio_custom() signature has changed between 0.6 and 0.8?
16:49:10  <dap>sorry if I'm late to a subject already discussed — just wondering if that should be on the API changes page.
16:51:30  * ibcquit (Remote host closed the connection)
16:53:54  * ibcjoined
16:55:19  * japjjoined
16:56:44  * dshaw_joined
16:57:21  <ibc>hi, is the write callback order guaranteed across the write reqs of the same handle?
16:57:29  <ibc>(in *nix and Windows)
16:57:35  <ibc>or will be?
16:57:40  <TooTallNate>dap: you should be using uv_queue_work() now
16:58:03  <dap>TooTallNate: Okay. Are there instructions for transitioning?
16:58:56  <TooTallNate>dap: mostly just examples in node and other add ons. i have a gist that does a backwards-compat kinda thing: https://gist.github.com/1368935
16:59:05  <igorzi>ibc: on windows write callback order is not guaranteed
16:59:49  <igorzi>ibc: (due to IOCP short-circuit optimisation that we do)
17:01:29  * dshaw_quit (Client Quit)
17:02:59  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
17:03:57  <ibc>ok
17:04:30  * ibcquit (Remote host closed the connection)
17:04:31  <isaacs>bnoordhuis: perhaps relevant to your interests: https://gist.github.com/1368935
17:08:14  * piscisaureus_joined
17:12:28  * dshaw_joined
17:15:31  * felixgequit (Quit: felixge)
17:17:49  * bnoordhuisquit (Ping timeout: 265 seconds)
17:24:56  <isaacs>piscisaureus_: Hey, is there a different CLA for libuv?
17:25:03  <isaacs>piscisaureus_: or do we just use the same node one?
17:27:20  <japj>isaacs: my bet is it's the same, since the libuv copyright notices contain "Copyright Joyent, Inc. and other Node contributors. All rights reserved." and "libuv is part of the Node project: http://nodejs.org/ libuv may be distributed alone under Node's license:"
17:27:42  <japj>but it would probably be good to mention it explicitly somewhere
17:27:56  <TooTallNate>but http_parser's a different for some reason :D
17:28:59  <isaacs>TooTallNate: i think because http_parser predates node by quite some time
17:29:17  <japj>http_parser.c is based on src/http/ngx_http_parse.c from NGINX copyright Igor Sysoev.
17:31:11  <japj>Have all the contributions to libuv be done by people that signed the Node CLA? (or phrased differently, are there libuv contributions done by people who didn't sign the Node CLA?)
17:32:06  <japj>I think for node it is explicitly checked (atleast that's what I noticed from contributing myself, and looking at pull requests in the node issue tracker)
17:32:19  <piscisaureus_>isaacs: it's the same as node
17:32:30  <piscisaureus_>isaacs: although, that's sort of lame, right?
17:32:37  <isaacs>meh.
17:32:43  <isaacs>i guess.
17:32:50  <piscisaureus_>but nobody cares enough
17:40:59  <indutny>I have signed only node's one
17:41:10  <indutny>so I'm owner of a part of the libuv
17:41:11  <indutny>hahahaha
17:41:18  <indutny>I'll sue you guys
17:41:24  <indutny>once you'll be big and rich
17:41:40  <indutny>ok, gtg
17:41:41  <indutny>ttyl :D
17:41:41  <isaacs>piscisaureus_, indutny, bnoordhuis: any objection to https://github.com/joyent/libuv/pull/437?
17:41:58  <isaacs>i'm told by my networking friends at Joyent that it's safe for all versions of sunos that matter.
17:42:15  <isaacs>(that is, not just smartos, but the other ones, as well.)
17:42:27  <isaacs>ie, opensolaris and illumos
17:44:01  * dshaw_quit (Remote host closed the connection)
17:45:32  <CIA-112>node: Bert Belder v0.6 * rf482236 / src/platform_win32.cc : windows: don't print error when GetConsoleTitleW returns an empty string - http://git.io/LU-vlw
17:45:51  <piscisaureus_>isaacs: so why was it commented out?
17:46:10  <piscisaureus_>isaacs: not that I care that much - nobody uses solaris anyway (except you guys)
17:46:29  <isaacs>piscisaureus_: long ago, that flag was added when ifaddrs was added, and then became a custom? i don't really know.
17:46:41  <isaacs>waf set -DSUNOS_HAVE_IFADDRS for any sunos
17:46:44  <isaacs>gyp doesn't
17:47:17  <piscisaureus_>aha
17:47:22  <piscisaureus_>yeah just land that PR then
17:47:28  <isaacs>kewl, landing
17:47:36  <isaacs>hahah
17:47:40  <isaacs>from rmustacc: "It didn't exist in JPC 1 which is probably why that happened because Ryan had an old machine in JPC 1."
17:48:01  <isaacs>jpc1 = solaris nevada-121
17:48:07  <isaacs>ooollllldddd stuff
17:48:12  <isaacs>like, 2008 stuff.
17:48:16  <isaacs>no body uses taht.
17:50:00  <piscisaureus_>you should know, solaris is even more magic to me than windows is to you
17:50:11  <piscisaureus_>not magic. uncontrollable
17:50:33  <piscisaureus_>I tried to put smartos in a vm last week, but ehm, barely could get anything to work
17:50:43  <isaacs>yeah, smartos is pretty magic/weird to me as well.
17:50:48  <isaacs>in my heart of hearts, i <3 the bsd
17:51:24  <isaacs>sunos is bsd-ish in many ways, but linuxy in others, and just batshit weird in others. it's like opera, even when it's better, it's weird about it.
17:52:11  <isaacs>i've been slowly learning it and forgetting linux, though
17:52:30  <piscisaureus_>I tried to install virtualbox client extensions in smartos
17:52:34  <piscisaureus_>of course, didn't work
17:52:46  <piscisaureus_>because smartos uses a different package format
17:52:48  <CIA-112>libuv: Mark Cavage master * r9cb8bdc / src/unix/sunos.c : sunos: uv_interface_addresses needlessly #ifdef'd out - http://git.io/vnfXzw
17:52:51  <piscisaureus_>than anything else
17:53:25  <isaacs>piscisaureus_: yes, i am not in love with pkgin
17:53:26  <isaacs>pkgsrc
17:53:57  <isaacs>but really, every package manager makes me rageful.
17:54:08  <isaacs>*especially* npm, but even more especially, all the other ones.
17:57:40  * isaacsquit (Remote host closed the connection)
18:05:36  <piscisaureus_>I know my way around apt-get
18:05:45  <piscisaureus_>and dpkg
18:05:59  <piscisaureus_>but everything on mac os always drives me nuts
18:07:25  * brsonjoined
18:10:48  * felixgejoined
18:10:48  * felixgequit (Changing host)
18:10:48  * felixgejoined
18:13:59  <piscisaureus_>indutny: hey, I have a question about nodejitsu
18:14:10  <piscisaureus_>indutny: is it true that you guys don't allow child_process.spawn?
18:14:26  <piscisaureus_>AvianFlu: ^ ?
18:14:41  <AvianFlu>it should work, I think
18:14:57  <AvianFlu>we tend to discourage it a little, 256mb vpses and all
18:15:17  <AvianFlu>but that's more for the people who want to like, spawn ffmpeg or something
18:15:40  <AvianFlu>in general it shouldn't cause problems
18:16:11  <piscisaureus_>AvianFlu: ok, so you don't like, block people from doing it?
18:16:18  <piscisaureus_>that's all I need to know :-)
18:16:21  <AvianFlu>yeah, no
18:16:27  <AvianFlu>people use it for the imagemagick tools, actually
18:16:30  <AvianFlu>so it should definitely work
18:16:42  <AvianFlu>it used to crash stuff, like, 8 months ago
18:16:53  <AvianFlu>but it should all play nice these days (tm)
18:16:58  <piscisaureus_>AvianFlu: right, ok. Thanks :-)
18:16:59  * felixgequit (Quit: felixge)
18:21:24  * brsonquit (Read error: Connection reset by peer)
18:21:59  * brsonjoined
18:24:52  * mikealquit (Quit: Leaving.)
18:28:06  * `3rdEdenjoined
18:32:42  * theColejoined
18:39:16  * isaacsjoined
18:46:19  * theColequit (Quit: theCole)
18:49:05  <piscisaureus_>ok, ehading out
18:49:13  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
18:51:15  <indutny>isaacs: heya
18:51:25  <indutny>isaacs: what do you think about giving me push access to libuv?
18:54:24  <CIA-112>node: isaacs reviewme * r84ffdb0 / (192 files in 15 dirs): Roll V8 back to 3.10.8.13 - http://git.io/s9cpBg
18:54:24  <CIA-112>node: isaacs reviewme * r355db80 / deps/v8/build/common.gypi : v8: Floating patches - http://git.io/hYcuZQ
18:54:24  <CIA-112>node: Ben Noordhuis reviewme * r476fe87 / deps/v8/src/debug-agent.cc : v8: debug: fix error handling in SendConnectMessage() - http://git.io/jI1ANQ
18:54:24  <CIA-112>node: Bert Belder reviewme * ra2b6aae / deps/v8/build/common.gypi : Unbreak the Windows build - http://git.io/RYWXXg
18:54:25  <CIA-112>node: isaacs reviewme * r09c3b73 / deps/v8/build/common.gypi : v8: Don't use 'echo -n' in configure script - http://git.io/ed3-JA
18:54:40  <isaacs>review, please^
18:54:50  <isaacs>rolling back v8 to 3.10
18:58:07  * isaacsquit (Remote host closed the connection)
18:58:43  * felixgejoined
18:58:43  * felixgequit (Changing host)
18:58:44  * felixgejoined
19:20:29  * `3rdEdenquit (Quit: Leaving...)
19:21:55  <CIA-112>node: Fedor Indutny master * r3116522 / (4 files in 2 dirs): child_process: spawn().ref() and spawn().unref() - http://git.io/Vd0cnA
19:21:59  <indutny>^^^
19:22:34  <indutny>AvianFlu: ^
19:25:24  <AvianFlu>excellent!
19:28:52  <tjfontaine>indutny: up next net.Sockets right? :)
19:30:08  <indutny>tjfontaine: what do you mean?
19:30:15  <indutny>tjfontaine: new net.Socket({ fd: ?? }) ?
19:30:23  <indutny>ah, ref/unref
19:30:39  <indutny>probably
19:31:20  <tjfontaine>indeed, ref/unref
19:33:32  * TheJHjoined
19:36:09  * kohaiquit (Remote host closed the connection)
19:43:30  * loladirojoined
20:11:15  * bulatshakirzyanoquit (Quit: Computer has gone to sleep.)
20:33:12  * japjquit (Remote host closed the connection)
20:58:14  * stephankjoined
21:18:41  * paddybyersquit (Ping timeout: 252 seconds)
21:36:52  * rendarquit
21:37:37  * piscisaureus_joined
21:38:14  * TooTallNatequit (Ping timeout: 252 seconds)
22:07:20  * c4miloquit (Remote host closed the connection)
22:09:27  <loladiro>piscisaureus_: Are you gonna be around for a little while? I'm just finishing up the windows part of the pipe/spawn change we talked about
22:09:41  <piscisaureus_>loladiro: for an hour or so, I think
22:10:02  <loladiro>ok, I'll see if I can get it done. Just adjusting the unit tests
22:10:07  <piscisaureus_>ok
22:15:04  <piscisaureus_>I wish I could use c++ for this
22:17:40  * felixgequit (Quit: felixge)
22:24:00  <piscisaureus_>hmm
22:24:32  <piscisaureus_>process.binding('fs').stat('c:\\windows\\write.exe\\.') <-- that way, node doesn't do any path normalization.... but I don't get an error anyway
22:43:17  * TheJHquit (Ping timeout: 245 seconds)
22:43:46  * avalanche123quit (Ping timeout: 248 seconds)
22:45:29  * avalanche123joined
22:51:50  * bitprobequit (Quit: Computer has gone to sleep.)
22:52:00  * c4milojoined
22:59:13  * bnoordhuisjoined
22:59:34  <bnoordhuis>yay, and we're back
22:59:43  <piscisaureus_>ben and ?
23:00:13  <bnoordhuis>my split personality
23:00:37  <bnoordhuis>had a network outage for most of the night
23:00:48  <piscisaureus_>that must be tele2 or ziggo
23:01:46  <bnoordhuis>yeah, tele2
23:03:42  * loladiroquit (Ping timeout: 245 seconds)
23:13:49  <piscisaureus_>bnoordhuis: I suppose you must've been very productive, then, today
23:15:16  <bnoordhuis>piscisaureus_: oh, not at all - i don't need an internet connection to waste time
23:15:34  * isaacsjoined
23:16:51  * TooTallNatejoined
23:19:20  <bnoordhuis>piscisaureus_: there's this thing called a book, it's like a paper ipad
23:19:39  <bnoordhuis>no touch screen though
23:22:57  <isaacs>bnoordhuis: it's got a swipe interface
23:24:13  <bnoordhuis>yeah, tactile feedback is pretty good
23:24:16  * loladirojoined
23:34:31  * theColejoined
23:38:54  <CIA-112>libuv: Ben Noordhuis master * rdea4594 / src/unix/core.c : unix: fix req cb / close cb invoke delay - http://git.io/xw5d6Q
23:40:54  * travis-cijoined
23:40:54  <travis-ci>[travis-ci] joyent/libuv#392 (master - dea4594 : Ben Noordhuis): The build was fixed.
23:40:55  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/9cb8bdc...dea4594
23:40:55  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/1562911
23:40:55  * travis-cipart
23:42:08  <bnoordhuis>hurray!
23:42:17  <piscisaureus_>what's to hurray?
23:42:27  <bnoordhuis>piscisaureus_: [travis-ci] joyent/libuv#392 (master - dea4594 : Ben Noordhuis): The build was fixed
23:43:20  <piscisaureus_>bnoordhuis: if you paid attention, you could have seen that coming by looking at win/core.c :-)
23:43:37  <bnoordhuis>piscisaureus_: you mean the commit itself?
23:43:38  <piscisaureus_>bnoordhuis: but, huzzay
23:43:42  <piscisaureus_>yeah
23:43:54  <bnoordhuis>oh, it's pretty old actually
23:44:00  <bnoordhuis>i tried some different approaches
23:44:12  <bnoordhuis>but this turned out to be the simplest
23:44:31  <piscisaureus_>also the only one that is correct ;-0
23:44:39  <bnoordhuis>in this implementation, yes
23:44:54  <bnoordhuis>i still feel there should be some way to make it all fall out automagically
23:45:11  <piscisaureus_>well, I suppose that if you have closing handles, you could skip the prepare .. poll .. check chant
23:45:40  <bnoordhuis>does it hurt to have those run?
23:45:43  <piscisaureus_>no
23:45:46  <piscisaureus_>I suppose not
23:45:53  <piscisaureus_>no
23:46:55  <bnoordhuis>i would think that if you don't poll if closing_handles > 0, you introduce a possible loop starvation issue
23:47:35  <loladiro>piscisaureus_: How do I debug a failing libuv unit test in visual studio (i.e. how do I attach to the test child process)?
23:49:42  * theColequit (Quit: theCole)
23:50:58  * mikealjoined
23:54:24  <piscisaureus_>loladiro: does the test need helpers?
23:54:34  <piscisaureus_>like echo-server or something?
23:54:56  <loladiro>It's the stdio-over-pipes one. There with my new pipes I need to debug
23:56:01  <piscisaureus_>no helpers
23:56:03  <piscisaureus_>ok
23:56:23  <piscisaureus_>loladiro: just set the command line arguments to `stdio_over_pipes stdio_over_pipes`
23:56:45  <piscisaureus_>(set run-tests as the startup project)
23:56:50  <piscisaureus_>loladiro: and off ya go
23:57:19  <loladiro>piscisaureus_: ok that works great. I tried just one stdio_over_pipes ;)
23:57:24  <loladiro>thanks