00:00:20  * paddybyersquit (Ping timeout: 245 seconds)
00:00:31  <ryah>can i have 20,000 uv_poll_t guys going?
00:00:48  <ryah>tcp fds
00:04:08  * c4milojoined
00:08:00  * c4miloquit (Remote host closed the connection)
00:14:28  * TooTallNatequit (Quit: Computer has gone to sleep.)
00:16:19  <bnoordhuis>ryah: sure, why not?
00:17:40  * hzquit
00:23:51  * TooTallNatejoined
00:28:00  * joshthecoderquit (Quit: Leaving...)
00:32:05  * toothrchanged nick to toothrot
00:34:54  * c4milojoined
00:35:07  * mmalecki_quit (Ping timeout: 255 seconds)
00:44:53  * bnoordhuisquit (Ping timeout: 272 seconds)
00:45:33  * ericktquit (Quit: erickt)
00:50:32  * ericktjoined
01:31:13  * joshthecoderjoined
01:51:24  * c4miloquit (Remote host closed the connection)
01:54:25  * ericktquit (Quit: erickt)
02:17:46  * wankdankerquit (Ping timeout: 246 seconds)
02:26:28  * brsonjoined
02:39:17  * toothrotquit (Quit: here we are)
02:44:28  <ryah>it's sad but im not going to use uv_tcp_t in my app
02:44:37  <ryah>it doesn't have the right semantics
02:44:45  <ryah>i need non-blocking writes and reads
02:49:08  * TooTallNatequit (Quit: ["Textual IRC Client: www.textualapp.com"])
02:56:50  * tomshredsquit (Quit: Leaving...)
03:06:01  * tomshredsjoined
03:06:33  * wankdankerjoined
03:32:47  * ericktjoined
03:38:19  * Ariaquit (Remote host closed the connection)
03:38:47  * Ariajoined
03:50:34  * joshthecoderquit (Quit: Leaving...)
03:52:14  <deoxxa>ryah: is uv_write blocking?
03:52:53  <deoxxa>i've been using uv_tcp_t successfully for about 6 months now, i've not found it to be blocking
04:32:45  * toothrjoined
04:51:43  * toothrchanged nick to toothrot
05:01:35  * ericktquit (Quit: erickt)
05:13:19  * wankdankerquit (Read error: Connection reset by peer)
05:13:30  * brsonquit (Ping timeout: 240 seconds)
05:13:39  * wankdankerjoined
05:26:21  <indutny>deoxxa: yes, it's blocking
05:26:36  <indutny>deoxxa: i.e. write() and read() calls are happening in event loop's thread
05:26:53  <indutny>but it should not matter much, since uv does it only at right time
05:27:02  <indutny>ryah: so?
05:35:26  * brsonjoined
05:55:31  <ryah>well it's not blocking .. but it's async
05:55:36  <ryah>rather than non-blocking
05:55:45  <ryah>non-blocking actually means it doesn't happen
05:55:50  <ryah>async is guarenteed to happen
05:56:01  <ryah>s/guarenteed/likely/
05:58:21  * paddybyersjoined
06:09:08  <deoxxa>ryah: ah, yes
07:04:53  * tomshredsquit (Quit: Leaving...)
07:16:36  * rendarjoined
07:27:06  * benoitcjoined
07:41:45  * brsonquit (Quit: leaving)
07:56:49  * paddybyersquit (Ping timeout: 272 seconds)
08:02:28  * paddybyersjoined
08:10:45  * AvianFluquit (Remote host closed the connection)
08:18:40  * paddybyersquit (Ping timeout: 245 seconds)
08:30:29  * EhevuTovjoined
08:38:01  * mikealquit (Quit: Leaving.)
08:38:27  * mikealjoined
08:56:46  * mikealquit (Quit: Leaving.)
09:17:17  * paddybyersjoined
09:18:17  * hzjoined
09:37:59  * paddybyersquit (Remote host closed the connection)
09:38:17  * paddybyersjoined
09:54:09  * paddybyersquit (Ping timeout: 244 seconds)
10:29:55  * wankdankerquit (Remote host closed the connection)
10:30:22  * wankdankerjoined
10:32:03  * paddybyersjoined
10:45:53  * Ariaquit (Remote host closed the connection)
10:46:02  * Ariajoined
10:50:17  * AndreasMadsenjoined
10:51:15  * benoitcquit (Excess Flood)
10:56:48  * benoitcjoined
11:03:49  * wankdankerquit (Ping timeout: 246 seconds)
11:06:18  * benoitcquit (Excess Flood)
11:08:18  * benoitcjoined
11:18:22  * paddybyersquit (Ping timeout: 244 seconds)
11:24:08  * AndreasM_joined
11:24:08  * AndreasMadsenquit (Read error: Connection reset by peer)
11:27:51  * EhevuTovquit (Quit: This computer has gone to sleep)
11:34:31  * mmaleckijoined
11:51:00  <mmalecki>ircretary: tell bnoordhuis for the sake of debugging and introspection. it's exposed in libuv and it's good to know what type of handle is keeping your app up. nothing wrong with exposing that, since that's a private API anyway, imo
11:51:00  <ircretary>mmalecki: I'll be sure to tell bnoordhuis
11:55:47  <indutny>ohai
11:57:51  <mmalecki>indutny: hai
11:58:09  <mmalecki>indutny: you're coming to nodedublin?
12:00:35  * paddybyersjoined
12:23:18  <txdv_>what are teh girls coding on
12:24:02  <txdv_>why is there no node moscow
12:27:39  * piscisaureus_joined
12:28:30  <piscisaureus_>ryah: you can have 20k uv_poll_t's in your app
12:28:32  <piscisaureus_>even on windows
12:28:42  <piscisaureus_>except in particular circumstances :-)
12:47:58  * mikealjoined
12:48:22  <MI6>joyent/node: Jan Lehnardt v0.8 * a7b5938 : docs: fix copy and paste error - http://git.io/LIRQxQ
12:52:57  * c4milojoined
12:57:12  * mikealquit (Quit: Leaving.)
13:15:41  <txdv_>piscisaureus_: what particular circumstance?
13:16:16  <piscisaureus_>txdv_: when the user has LSPs installed on windows
13:16:24  <piscisaureus_>er, on windows XP
13:16:37  <piscisaureus_>(and I recently found out, also sometimes on newer windowses, but I will fix that)
13:16:50  <piscisaureus_>txdv_: so, not that often
13:17:00  <piscisaureus_>but *if* then 20k sockets will definitely blow up
13:17:18  <piscisaureus_>because libuv will create between 20000 and 40000 threads
13:17:26  <txdv_>fork bomb
13:17:59  <txdv_>what is LSP?
13:18:33  * benoitcquit (Excess Flood)
13:22:52  * benoitcjoined
13:26:39  <piscisaureus_>txdv_: user-mode network filter
13:26:45  <piscisaureus_>google is your best friend :-)
13:27:48  <txdv_>yeah i know, i am that lonely
13:28:19  <piscisaureus_>actually it won't create so many threads
13:28:24  <piscisaureus_>it just won't work
13:29:37  <piscisaureus_>txdv: aah, I feel sad for you
13:34:06  <benoitc>mmm from time to time I have a process status becoming a zombie after a uv_process_kill(SIGTERM) but on_exit isn't triggered
13:34:51  <benoitc>i'm using the pyuv binding so not sure if it's a bug there or in libuv
13:35:25  <benoitc>it appears when all other handles are inactive or unrefed and i'm waiting that the loop ends
13:35:57  * AvianFlujoined
13:48:28  <txdv_>http://jrvis.com/red-dwarf/?user=joyent&repo=libuv
13:48:39  <txdv_>a lot of japanese watching lib uv
14:13:09  * AndreasM_quit (Remote host closed the connection)
14:15:49  <indutny>mmalecki: nope
14:40:09  * c4miloquit (Remote host closed the connection)
14:58:14  <indutny>piscisaureus_: ok, time to fix openssl in master :)
15:00:45  * paddybyersquit (Ping timeout: 245 seconds)
15:01:53  * AndreasMadsenjoined
15:01:55  * paddybyersjoined
15:02:43  <piscisaureus_>indutny: godspeed
15:03:07  <indutny>heh
15:03:13  <indutny>I was really busy :)
15:06:13  <indutny>piscisaureus_: so it's like 8 commits, right?
15:06:48  <piscisaureus_>indutny: they have my name on it. something like that yeah
15:06:58  <indutny>ok
15:10:48  <indutny>piscisaureus_: btw, how are you?
15:11:03  <piscisaureus_>yhanks
15:17:59  <indutny>what does sed -s on linux?
15:18:25  <indutny>actually whole -sri
15:18:29  <indutny>it's not present in osx
15:20:01  <txdv_>when is rm-ev going to hit master
15:27:21  * ericktjoined
15:30:38  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
15:36:49  * bnoordhuisjoined
15:39:58  <indutny>bnoordhuis: what the heck is 586?
15:40:02  <indutny>I'm not that old to new about it
15:40:24  <bnoordhuis>indutny: eh? you mean the intel architecture?
15:40:28  <indutny>idk
15:40:32  <bnoordhuis>context?
15:40:33  <indutny>it's a postfix in deps/openssl/asm stuff
15:40:41  <bnoordhuis>oh right
15:40:47  <indutny>why osx is using it?
15:40:50  <bnoordhuis>it means pentium 1
15:41:18  <indutny>oh
15:41:19  <indutny>crap
15:41:20  <indutny>ok
15:41:46  <indutny>can I use x86_64 instead?
15:42:04  * benoitcquit (Excess Flood)
15:42:19  <bnoordhuis>indutny: only if the machine supports that, of course :)
15:42:43  <indutny>bnoordhuis: well
15:42:50  <indutny>lets suppose that osx can do it
15:43:19  <bnoordhuis>in that case, yes
15:43:25  <bnoordhuis>but not if node is compiled as 32 bits
15:43:35  <bnoordhuis>what exactly is the problem, fedor?
15:43:53  <indutny>well, there's aesni asm file
15:44:17  <indutny>and I need to figure out
15:44:21  <indutny>either use inlined assembly here
15:44:28  <indutny>or fallback to something
15:45:48  <bnoordhuis>i'm not following, i think
15:47:07  <indutny>http://en.wikipedia.org/wiki/AES_instruction_set
15:47:11  <indutny>that's it
15:47:32  * wankdankerjoined
15:48:12  <indutny>bnoordhuis: so I don't think that every CPU support this instruction
15:48:31  <bnoordhuis>indutny: that's correct. only recent(ish) intel cpus
15:48:36  <indutny>indeed
15:48:45  <indutny>and amd
15:49:00  <bnoordhuis>yes, i believe so
15:49:18  <bnoordhuis>though i don't have any amd cpus in my household that are new enough :)
15:49:22  * benoitcjoined
15:49:58  <indutny>:)
15:50:05  <indutny>my personal opinion
15:50:07  <indutny>fuck amd
15:50:18  <indutny>I don't follow how are they better than Intel
15:50:50  <indutny>I suppose every x64 CPU should support aesni
15:51:03  <bnoordhuis>they don't
15:51:20  <bnoordhuis>i'm not sure when exactly it was added but it was years after the first x64 cpu
15:51:24  * c4milojoined
15:51:28  <bnoordhuis>wikipedia no doubt knows :)
15:52:52  <bnoordhuis>dinner, biab
15:56:17  <indutny>see ya
16:02:11  * hzquit
16:08:24  <mmalecki>bnoordhuis: did you get my note?
16:22:04  * tomshredsjoined
16:28:27  * jmar777joined
16:32:08  <indutny>oh
16:32:10  <indutny>fucking cow
16:42:10  * jmar777quit (Remote host closed the connection)
16:42:38  * jmar777joined
16:45:30  <txdv_>why u badmouth amd
16:46:49  * jmar777quit (Ping timeout: 246 seconds)
17:05:27  * joshthecoderjoined
17:17:12  * c4miloquit (Remote host closed the connection)
17:27:11  * piscisaureus_joined
17:28:30  <piscisaureus_>indutny: you should use the 586 stuff on x86, and something else on x64
17:29:18  <piscisaureus_>indutny: the aesni asm file is only for the intel engine, I don't think you can assume it works on all systems *if* you use it that should be used at runtime
17:29:27  <txdv_>BUILD STATUS IS FAILING
17:29:29  <txdv_>AAAAAAAAA
17:31:21  <indutny>piscisaureus_: hm..
17:31:24  <indutny>piscisaureus_: better disable it
17:31:46  <indutny>piscisaureus_: also, something doesn't work :)
17:32:04  <piscisaureus_>indutny: you can compile aesni in but it won't be used by default
17:32:06  <indutny>piscisaureus_: but it builds fine
17:32:06  <indutny>piscisaureus_: also, performance is not really good
17:32:09  <piscisaureus_>for that you need to patch it
17:32:23  <indutny>I already patched it
17:32:26  <indutny>and rebuild all asm stuff
17:32:41  <indutny>but I don't understand what's happening
17:32:41  <indutny>SSL_read returns -1
17:35:26  <indutny>it can't parse client hello
17:39:36  <piscisaureus_>hm odd
17:43:32  <indutny>yeah
17:56:52  * joshthecoderquit (Quit: Linkinus - http://linkinus.com)
18:10:41  * mikealjoined
18:20:30  * mraleph1joined
18:20:30  * mralephquit (Ping timeout: 245 seconds)
18:30:21  * AndreasMadsenquit (Remote host closed the connection)
18:48:49  * mikealquit (Quit: Leaving.)
18:51:18  * mikealjoined
19:07:28  * AndreasMadsenjoined
19:07:29  * piscisaureus_quit (Ping timeout: 260 seconds)
19:10:55  * felixgejoined
19:10:55  * felixgequit (Changing host)
19:10:55  * felixgejoined
19:11:10  * piscisaureus_joined
19:15:06  * brsonjoined
19:37:54  <indutny>piscisaureus_: I found the case
19:38:00  <indutny>piscisaureus_: it's really interesting!
19:38:10  * joshthecoderjoined
19:39:01  <indutny>I was compiling tlsnappy with stale uv.h header
19:41:13  * piscisaureus_quit (Read error: Connection reset by peer)
19:41:22  * piscisaureus_joined
19:42:38  <indutny>piscisaureus_: have we broke ABI in 0.9.x since 0.9.2?
19:42:59  <piscisaureus_>indutny: shrug, I don't know to be honest
19:43:04  <piscisaureus_>indutny: not unlikely I suppose
19:43:23  <piscisaureus_>indutny: I am pretty confident that there is *some* ABI that broke
19:43:26  <indutny>we broke it
19:43:49  <indutny>yay
19:43:50  <indutny>it works now!
19:43:58  <indutny>very good
19:44:17  <indutny>ttyl
19:44:33  * bulatshakirzyanojoined
19:44:51  <piscisaureus_>ohno
19:44:57  <piscisaureus_>oraakle: http://www.stumbleupon.com/su/1b86y7/www.toolsjournal.com/integrations-articles/item/917-oracle-previews-new-mysql-cluster-with-forigen-key-and-nodejs-support?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+ToolsJournal+%28ToolsJournal%29&utm_content=FeedBurner
19:45:12  <felixge>: p
19:45:19  <piscisaureus_>hey felixge
19:45:28  <felixge>piscisaureus_: hey
19:45:33  <piscisaureus_>how are you?
19:45:36  <txdv_>you know there are url shortnerers
19:45:45  <txdv_>i think this is one of the cases when you should use them ^^
19:46:06  <piscisaureus_>felixge: can you now burp on nodeup?
19:46:13  <felixge>piscisaureus_: :P
19:46:19  <felixge>piscisaureus_: I'll try
19:46:21  <piscisaureus_>good
19:46:25  <felixge>but I think that's your brand now
19:46:26  <piscisaureus_>felixge: do it
19:46:26  <felixge>:D
19:46:54  <felixge>:)
19:47:04  <felixge>I have wine right now, not beer
19:47:05  <txdv_>nodejs support for mysql?
19:47:06  <felixge>will be hard
19:47:17  <piscisaureus_>swallow some air
19:50:05  <txdv_>99,999%
19:51:01  <piscisaureus_>no i want coros in js
19:51:20  <txdv_>coros?
19:51:23  <felixge>:)
19:52:07  <piscisaureus_>felige++
19:52:07  <kohai>felige has 1 beer
19:54:07  <felixge>now my mis-pelled sibbling has a beer and I don't :(
19:54:10  <felixge>* misspelled
19:54:15  <piscisaureus_>felixge++
19:54:15  <kohai>felixge has 2 beers
19:54:18  <piscisaureus_>felige--
19:54:18  <kohai>felige has 0 beer
19:54:21  <felixge>sweet
19:54:31  <piscisaureus_>dude
20:05:12  <felixge>y?
20:06:54  * AndreasMadsenquit (Remote host closed the connection)
20:08:15  <piscisaureus_>felixge: see http://www.youtube.com/watch?v=YmNmb2EReG4
20:09:22  <felixge>:)
20:28:25  * mikealquit (Quit: Leaving.)
20:30:35  <indutny>piscisaureus_: wanna take a look?
20:30:40  <indutny>piscisaureus_: at WIP?
20:32:02  <indutny>piscisaureus_: https://github.com/joyent/node/pull/4139
20:32:03  <indutny>bnoordhuis: ^
20:34:21  * bulatshakirzyanochanged nick to avalanche123
20:34:28  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
20:50:03  * hzjoined
20:58:29  * felixgequit (Quit: felixge)
21:05:23  * paddybyersquit (Remote host closed the connection)
21:05:56  * paddybyersjoined
21:07:32  * rendarquit
21:20:03  * isufyquit
21:48:17  * perezdjoined
22:09:10  * mraleph1quit (Ping timeout: 246 seconds)
22:11:11  * xaqjoined
22:18:34  * hnakamurjoined
22:20:21  * brsonquit (Ping timeout: 245 seconds)
22:22:11  * paddybyersquit (Ping timeout: 260 seconds)
22:28:16  * mmaleckiquit (Quit: Lost terminal)
22:30:56  * mmaleckijoined
22:33:59  * xaq_joined
22:37:25  * xaqquit (Ping timeout: 245 seconds)
22:37:29  * avalanche123quit (Quit: Computer has gone to sleep.)
22:43:11  * brsonjoined
22:55:11  * xaq_quit (Ping timeout: 255 seconds)
22:55:53  * avalanche123joined
23:06:38  * xaqjoined
23:10:13  * avalanche123quit (Quit: Computer has gone to sleep.)
23:11:15  <MI6>joyent/libuv: Ben Noordhuis master * fd136da : unix: remove always_inline attribute Fixes the following gcc 4.7+ warnin (+1 more commits) - http://git.io/0Z2gLA
23:11:28  <bnoordhuis>^ stupid gcc
23:12:51  * travis-cijoined
23:12:51  <travis-ci>[travis-ci] joyent/libuv#793 (master - fd136da : Ben Noordhuis): The build was fixed.
23:12:51  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/47b2cd31c9b7...fd136da04ade
23:12:51  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/2790420
23:12:51  * travis-cipart
23:17:42  <mmalecki>bnoordhuis: ping?
23:17:48  <bnoordhuis>mmalecki: pong?
23:17:57  <mmalecki>bnoordhuis: did you get my note?
23:18:15  <bnoordhuis>mmalecki: about process._getActiveHandles()?
23:18:30  <mmalecki>bnoordhuis: yes, and types exposed. I explained why I want that
23:19:10  <bnoordhuis>mmalecki: can you give it to me again in a nutshell? :)
23:19:43  <mmalecki>bnoordhuis: sure. let me grep ~/irclogs
23:20:14  <mmalecki>"or the sake of debugging and introspection. it's exposed in libuv and it's good to know what type of handle is keeping your app up. nothing wrong with exposing that, since that's a private API anyway, imo"
23:20:25  <mmalecki>*for the sake
23:21:02  <bnoordhuis>mmalecki: process._getActiveHandles() is not the same thing as the handle list in libuv
23:21:21  <mmalecki>bnoordhuis: what's the difference?
23:21:27  <bnoordhuis>it returns a list of wrap objects
23:21:37  * xaqquit (Remote host closed the connection)
23:21:38  <mmalecki>yup
23:21:50  <mmalecki>so I want to make it { type: handleType, handle: wrapObject }
23:21:56  <mmalecki>if that's possible, in any way
23:22:09  <mmalecki>I can implement it.
23:22:45  <bnoordhuis>mmalecki: it's not that useful in its curremt implementation
23:22:56  <mmalecki>bnoordhuis: I can confirm it isn't :)
23:22:59  <bnoordhuis>the wrap object is the internal handle wrap
23:23:25  <bnoordhuis>what users likely want is the net.Socket or the timer object
23:23:40  <bnoordhuis>however, for net.Socket there is no back-link and timers don't have a 1-to-1 mapping
23:23:45  <mmalecki>yeah
23:23:47  <mmalecki>I know
23:24:02  <bnoordhuis>so i guess my question is: what's the point? :)
23:24:07  <mmalecki>bnoordhuis: the fact is, I'm implementing a module for gracefully exiting node process
23:24:27  <mmalecki>i.e., waiting for all network connections to close and such
23:25:16  <mmalecki>bnoordhuis: and I'd love to ignore certain types of handles
23:25:35  <mmalecki>but I can imagine this gets more useful when debugging
23:28:19  <bnoordhuis>it's only useful for debugging - it's way too expensive to use in production code
23:30:43  <mmalecki>bnoordhuis: when you're exiting and not expecting any more requests, that doesn't matter much. and in our case, exiting before you make some requests can cause some problems. anyway, I think of it as a nifty feature for debugging. I don't really need it for my module - I can hack it around
23:31:33  * hnakamurpart ("Leaving...")
23:31:46  <bnoordhuis>mmalecki: i'm afraid i don't really see the point
23:31:51  <bnoordhuis>but maybe you can convince the others :)
23:31:53  <mmalecki>right now you only see an object with no hint of what it actually is
23:32:14  <mmalecki>type: "tcp" would be a convenient hint
23:34:35  <mmalecki>I mean, at the first glance at objects, I sometimes fail to see what type they are
23:35:20  <mmalecki>also, damn, I have a flight in 3 hours
23:35:37  <bnoordhuis>mmalecki: that kind of works, e.g. handle instanceof process.binding('tcp_wrap').TCP
23:36:04  <bnoordhuis>if you feel strongly about it, open a PR and we'll take it from there
23:36:13  <mmalecki>bnoordhuis: I shall
23:36:35  <mmalecki>I SHALL
23:36:53  <mmalecki>bnoordhuis: when are you getting in to Dublin/
23:36:54  <mmalecki>?
23:37:06  <bnoordhuis>wednesday, early in the evening
23:37:22  <mmalecki>shame
23:37:51  <bnoordhuis>yeah, i could've flown in friday morning at four and back at twelve
23:37:59  <bnoordhuis>four? six
23:38:22  <bnoordhuis>i half-kid but conventions are always enormous time sinks
23:38:45  <mmalecki>well, I'm flying in today
23:38:58  <mmalecki>and I don't think it's a waste of time, Dublin seems like a cool place :)
23:40:31  <bnoordhuis>time sink != waste of time, or at least not quite
23:41:01  <bnoordhuis>but that's three days i could've been getting work done
23:43:17  <mmalecki>bnoordhuis: engineyard offered everyone a coworking space
23:43:20  <mmalecki>also, cafes
23:44:22  <bnoordhuis>i can't work in a cafe, too much distraction
23:45:12  <mmalecki>aww, that sucks
23:47:55  <mmalecki>bnoordhuis: at any rate, beer on wednesday?
23:48:03  <bnoordhuis>sure
23:48:31  <mmalecki>cool, you're staying at central?
23:49:39  <bnoordhuis>mmalecki: you mean the hotel? i think so. what the organization booked
23:50:34  <mmalecki>that'd be central, yeah
23:52:15  * ericktquit (Quit: erickt)