00:05:46  * c4milojoined
00:07:48  * c4miloquit (Remote host closed the connection)
00:20:36  * bcantrillquit (Quit: Leaving.)
00:22:08  * xaqquit (Remote host closed the connection)
00:33:51  * ericktquit (Quit: erickt)
00:37:53  * piscisaureus_joined
00:38:18  <piscisaureus_>bnoordhuis: re libeio blow up -> that sucks
00:38:24  <bnoordhuis>piscisaureus_: yes
00:38:30  <piscisaureus_>bnoordhuis: also, c-ares 1.9.0 landed so that's why it is closed :_)
00:38:37  <bnoordhuis>ah okay :)
00:39:34  * c4milojoined
00:40:38  <piscisaureus_>bnoordhuis: I just promised the audience that we will do EPOLLET so now there is no way back :-p
00:41:00  <bnoordhuis>piscisaureus_: good. i opened a tracking issue today
00:41:14  <piscisaureus_>wasn't that for ev removal:
00:41:24  <piscisaureus_>I suppose that has to happen first anyway so
00:41:30  <piscisaureus_>it doesn't really matter
00:42:11  * c4miloquit (Remote host closed the connection)
00:43:56  <bnoordhuis>yay, i'm down to < 40 unread github mails
00:46:34  * piscisaureus_quit (Ping timeout: 246 seconds)
01:06:12  <indutny>bnoordhuis: sorry, I was speaking at conf
01:06:48  <mmalecki>indutny: ++
01:06:48  <kohai>indutny has 26 beers
01:06:58  <mmalecki>indutny: asm slides are pretty badass
01:07:13  <bnoordhuis>indutny: what was your talk about?
01:07:18  <indutny>bnoordhuis: Candor
01:07:26  <indutny>bnoordhuis: unfortunatelly creationix hasn't came up
01:07:40  <indutny>so I missed the most of the bert's talk
01:07:41  <indutny>:(
01:07:48  <bnoordhuis>ah, too bad
01:07:56  <bnoordhuis>how was your talk received?
01:08:06  <indutny>well, they were laughing
01:09:23  <indutny>but I think some people even understand what I was talking about
01:10:17  <bnoordhuis>nice
01:11:43  <indutny>really sad that you ain't here
01:14:36  * abraxasjoined
01:39:41  * Ariajoined
02:09:34  * brsonquit (Ping timeout: 246 seconds)
02:15:41  <bnoordhuis>i will never understand the logic behind threads and signal delivery...
02:15:49  <bnoordhuis>maybe there is none
02:18:39  <indutny>bnoordhuis: no
02:18:56  <indutny>:D
02:18:59  <Aria>Heh. Yes.
02:19:00  <Aria>Oy.
02:19:27  <indutny>bnoordhuis: I'm going to eat something at the restoraunt here downstairs and start profiling SSL stuff
02:19:34  <indutny>bnoordhuis: what's the time in your place now?
02:19:43  <bnoordhuis>indutny: 4.15 am
02:19:47  <indutny>wow
02:19:49  <indutny>early bird
02:19:52  <indutny>ain't you?
02:19:52  <mmalecki>lol
02:19:59  <indutny>are you going to sleep soon?
02:20:03  <bnoordhuis>rather the other way around :)
02:20:10  <bnoordhuis>but yes, i think i'll hit the sack in abit
02:20:15  <bnoordhuis>enjoy your dinner, fedor
02:20:16  <indutny>ok
02:20:20  <indutny>sleep tight then
02:20:23  <indutny>thank you :)
02:20:23  <bnoordhuis>thanks
02:21:58  <mmalecki>bnoordhuis: https://github.com/joyent/node/pull/3621, review, would you? one line change :)
02:47:14  * brsonjoined
02:47:52  * c4milojoined
02:55:31  <bnoordhuis>mmalecki: add a test and i'll merge it
02:56:50  * c4miloquit (Remote host closed the connection)
02:57:05  <mmalecki>bnoordhuis: k, I'll ping you when it's done, doing something else now
03:00:47  * hij1nxjoined
03:16:54  * dapjoined
03:19:14  <indutny>bnoordhuis: still not sleeping, I suppose?
03:19:15  <indutny>:)
03:21:44  <bnoordhuis>indutny: hey
03:21:56  <bnoordhuis>well, i was about to turn off my pc :)
03:21:59  <indutny>haha
03:22:01  <indutny>ok,
03:22:05  <indutny>do it then :P
03:22:05  * tjfontainepulls the power adapter
03:22:17  <mmalecki>what do you mean by "turn off"?
03:22:37  <tjfontaine>mmalecki: shutdown -h now
03:22:58  <mmalecki>tjfontaine: wait, you mean you actually *can* turn this thing off?
03:23:39  <tjfontaine>mmalecki: according to my clients though they just pull the plug out of the wall and then put it back in, they got tired of waiting on the power button
03:23:52  <tjfontaine>(I cry myself to sleep most nights)
03:24:03  <mmalecki>well, doesn't work for me
03:24:04  <bnoordhuis>well, it worked in 1993, why shouldn't that work now
03:24:07  <mmalecki>at least on this macbook
03:24:20  <bnoordhuis>okay, off to bed. sleep tight, people
03:24:20  <mmalecki>no idea why, when I unplug it, it still works
03:24:23  <mmalecki>bnoordhuis: night
03:25:01  <tjfontaine>mmalecki: waiting on that battery to die really puts a damper on the turning it on/off 3 times thing
03:26:23  * bnoordhuisquit (Read error: Operation timed out)
03:29:58  <indutny>creationix: ping
03:30:05  <indutny>creationix: where are you man?
03:33:14  * dapquit (Quit: Leaving.)
03:34:50  * Ariaquit (Remote host closed the connection)
04:28:28  <deoxxa>ok, so i'm using uv_process_spawn
04:28:44  <deoxxa>sometimes i don't get any output data at all from the spawned process
04:29:08  <deoxxa>have there been any significant changes to the process stuff in the last couple of months?
05:36:58  * paddybyersjoined
05:48:37  <deoxxa>the answer was "yes", if anyone's wondering
05:59:32  * bcantrilljoined
06:26:23  * hzjoined
06:33:33  * hzquit (Read error: Connection reset by peer)
06:33:45  * hzjoined
06:41:20  * stephankquit (Quit: *Poof!*)
06:45:21  * hzquit (Ping timeout: 272 seconds)
06:49:38  * hzjoined
07:15:22  * AlbireoXjoined
07:22:47  * rendarjoined
07:26:04  * ^3rdEdenjoined
07:26:29  * paddybyers_joined
07:29:04  * paddybyersquit (Ping timeout: 246 seconds)
07:29:04  * paddybyers_changed nick to paddybyers
07:40:38  * brsonquit (Quit: leaving)
08:25:56  * mmaleckiquit (Quit: leaving)
08:49:58  * loladirojoined
08:51:00  * loladiroquit (Read error: Connection reset by peer)
08:51:06  * loladiro_joined
08:58:10  * bcantrillquit (Quit: Leaving.)
09:18:10  * ibcjoined
09:18:46  * ibcquit (Remote host closed the connection)
09:21:34  * hij1nxquit (Quit: hij1nx)
09:34:54  * loladiro_quit (Read error: Connection reset by peer)
09:35:00  * loladirojoined
09:45:21  * piscisaureus_joined
09:56:48  * piscisaureus_quit (Ping timeout: 245 seconds)
10:03:41  * mmaleckijoined
10:03:55  * loladiroquit (Read error: Connection reset by peer)
10:04:01  * loladiro_joined
10:33:52  * mmaleckiquit (Ping timeout: 246 seconds)
11:15:17  * hzquit
11:31:19  * loladiro_quit (Quit: loladiro_)
11:41:50  * loladirojoined
12:27:19  * chobi_e_changed nick to chobi_e
12:42:36  * mmaleckijoined
12:54:21  * bnoordhuisjoined
13:03:27  * mmaleckiquit (Ping timeout: 240 seconds)
13:32:06  * benviejoined
13:33:15  * benviequit
13:33:52  * bweaverjoined
13:42:25  <CIA-108>node: Ben Noordhuis v0.8 * rb1cce04 / (common.gypi configure): build: rename strict_aliasing to node_no_strict_aliasing - http://git.io/itKpog
13:42:27  <CIA-108>node: Ben Noordhuis v0.8 * ra0add91 / configure : build: detect cc version with -dumpversion - http://git.io/7JUMmA
13:42:28  <CIA-108>node: Ben Noordhuis v0.8 * rb731c96 / configure : build: disable -fstrict-aliasing for any gcc < 4.6.0 - http://git.io/Mw_p-Q
13:52:11  <indutny>bnoordhuis: good morning
13:52:21  <bnoordhuis>indutny: hey
13:56:25  * mmaleckijoined
14:14:10  <indutny>bnoordhuis: so I'm building node with -pg flag atm
14:14:33  <indutny>and lets see how it'll show up at "ab https://localhost:.../" :)
14:20:01  * hzjoined
14:24:01  <indutny>bnoordhuis: is 'cflags' working for darwin?
14:24:08  <indutny>bnoordhuis: I don't think they have affected final build
14:28:02  <mmalecki>indutny: hey, how's nodeconf?
14:30:15  <indutny>mmalecki: it was great
14:31:16  * c4milojoined
14:31:20  <mmalecki>wait, it's over already?
14:31:35  <indutny>mmalecki: only day one is over
14:32:04  <mmalecki>ah, so I'm not getting any sleep today too >.<
14:33:19  <mmalecki>indutny: btw, not sure if I told you, haibu-carapace@master works on 0.8
14:33:29  <indutny>yeah, already seen that
14:33:33  <indutny>and you told me
14:33:33  <indutny>:)
14:33:50  <mmalecki>right. sorry, I had like an hour of sleep today
14:35:39  <mmalecki>indutny: when are you coming back to Russia btw?
14:36:38  <indutny>July 9th
14:36:40  <indutny>or 10th
14:37:34  <mmalecki>indutny: !!! I'm visiting you then!
14:37:54  <indutny>ok, nice
14:38:06  <mmalecki>indutny: I need to change the plane in Moscow anyway
14:38:21  <indutny>are so you've connected flights here?
14:38:30  <mmalecki>indutny: I'll bring vodka, make sure you have glasses
14:38:36  <mmalecki>indutny: no, changing airlines too
14:38:46  <mmalecki>indutny: so they are not strictly connected
14:38:54  <mmalecki>getting to Nepal is kind of a pain
14:39:24  <indutny>ah, ok
14:39:30  <indutny>not sure if I'll be free at those days
14:39:35  <indutny>probably I'll continue traveling
14:39:49  <mmalecki>oh, no, it'll be sometime later
14:40:12  <mmalecki>but yeah, I don't know when I'll be travelling neither
14:53:49  * ericktjoined
14:55:03  <indutny>bnoordhuis: yt?
14:57:49  * einarosquit (Ping timeout: 260 seconds)
14:58:07  * einarosjoined
15:11:03  * hij1nx_joined
15:11:12  * ^3rdEdenquit (Quit: Linkinus - http://linkinus.com)
15:14:38  * theColejoined
15:14:47  * theColequit (Client Quit)
15:16:03  * theColejoined
15:25:40  <bnoordhuis>indutny: ih
15:26:35  <bnoordhuis>indutny: gyp on os x uses the xcode_settings, not cflags (mostly)
15:26:44  <bnoordhuis>*the xcode_settings field
15:27:58  * theColequit (Quit: theCole)
15:39:18  * ibcjoined
15:39:38  * bnoordhuisis afk again
15:39:55  <ibc>hi, can I rely in the existence of cares.c integration? or will it be removed in the future from libuv?
15:41:20  <tjfontaine>there was some talk that c-ares won't always be in libuv, but that it's removal is not imminent
15:42:29  <ibc>thanks, so I will take a copy of src/cares.c :)
15:47:17  * stephankjoined
15:48:13  <tjfontaine>it should be said that I am not a developer on libuv, but merely an active observer, and I am only repeating what I've seen said here
15:48:40  * piscisaureus_joined
15:49:21  <tjfontaine>ibc: you really want to direct that question at piscisaureus_
15:49:42  <ibc>thanks
15:56:31  * bcantrilljoined
16:00:36  * hij1nx_quit (Quit: hij1nx_)
16:01:25  <indutny>bnoordhuis: yeah, had already figured that out
16:02:09  <indutny>bnoordhuis: so there're a couple of things that we're doing wrong in tls.js
16:02:25  <indutny>bnoordhuis: like creating many buffers for one connection in place where it isn't really needed
16:02:41  <indutny>bnoordhuis: but most interestingly is that overhead of such stuff still much lower that SSL timing :P
16:03:01  * brsonjoined
16:03:05  <indutny>piscisaureus_: heya
16:03:26  <indutny>piscisaureus_: sadly, I have missed the most of your talk :(
16:03:38  <piscisaureus_>indutny: booo :-p
16:03:50  <indutny>piscisaureus_: I was hardly distinguishing from backstage what you're talking about
16:03:59  <indutny>piscisaureus_: heh
16:04:02  <piscisaureus_>indutny: no, that's normal. I missed part of your talk to :-)
16:04:05  <indutny>hahaha
16:04:07  <indutny>piscisaureus_: where was ya?
16:04:14  * hij1nx_joined
16:04:16  <indutny>piscisaureus_: removing microphone and all the stuff?
16:04:28  <piscisaureus_>yeah, and then chill out outside for a couple of minutes
16:04:31  <piscisaureus_>I saw the end of it
16:04:38  <indutny>bnoordhuis: 22111 41.9% 41.9% _malloc_create_zone
16:04:46  <piscisaureus_>oh wow
16:04:51  <indutny>hehehe
16:04:55  * hij1nx_changed nick to hij1nx
16:04:56  <indutny>yeah, indeed
16:05:00  <piscisaureus_>indutny: I have to get up, see you around 10
16:05:01  <indutny>but I already have a patch for that stuff
16:05:08  <indutny>well, you won't see me, I suppose
16:05:14  <indutny>going to walk around in portland :)
16:06:11  * hij1nxquit (Client Quit)
16:06:17  <indutny>piscisaureus_: are there any interesting talks today? :P
16:06:25  <piscisaureus_>indutny: I have no clue :-)
16:06:34  <piscisaureus_>indutny: you should hang around at the party and stuff :-)
16:06:53  <indutny>piscisaureus_: so today is a party, ok, nice
16:06:53  <piscisaureus_>indutny: will you be away the entire day?
16:06:57  <indutny>piscisaureus_: dunno
16:07:02  <piscisaureus_>indutny: yesterday was a part too
16:07:06  <piscisaureus_>*party
16:07:13  <indutny>piscisaureus_: yeah, I know
16:07:21  <indutny>I just was completely wasted by TZ difference
16:07:27  <piscisaureus_>haha
16:07:28  <piscisaureus_>ok
16:07:30  <piscisaureus_>anyway
16:07:30  <indutny>and I don't really enjoy partying
16:07:33  <indutny>:P
16:08:32  <piscisaureus_>indutny: anyway, enjoy. I am going to take a shower now etc.
16:08:42  * piscisaureus_away
16:08:50  <indutny>ok
16:08:54  <indutny>:)
16:09:20  <indutny>bnoordhuis: oh, I just got 300 req/s with my SSL patches :P
16:09:52  <tjfontaine>what was it before? did you just change it to use the slab allocator?
16:09:57  <indutny>em... it was 380
16:09:59  <indutny>ok
16:09:59  <indutny>nvm
16:10:01  <indutny>:D
16:10:20  <indutny>tjfontaine: I don't think slab allocator will work with our tls logic
16:10:31  <indutny>but who knows
16:10:34  <indutny>I should just try
16:11:11  <indutny>but I'm reusing buffers, right
16:11:30  <indutny>and now I'm wasting time here: 17514 57.4% 57.4% node::crypto::Connection::ClearOut
16:11:38  <indutny>and only 5431 17.8% 17.8% _malloc_create_zone
16:11:51  <tjfontaine>well that's better I guess :)
16:12:06  <indutny>ah, I know what's happening
16:12:13  <indutny>I've changed a logic a little accidentaly
16:15:53  <indutny>ah, ok that was node's build with profiling enabled
16:15:55  <indutny>very nice
16:17:29  <indutny>ok 401 req/sec
16:17:31  <indutny>very nice
16:17:38  * ibcquit (Remote host closed the connection)
16:17:48  <indutny>ok, I'm going to run tests with that patch
16:17:58  <indutny>and in meantime I'm going to eat something :P
16:17:58  <indutny>ttyl
16:31:17  * hij1nxjoined
16:32:41  * hij1nx_joined
16:35:46  * hij1nxquit (Ping timeout: 252 seconds)
16:35:52  * hij1nx_changed nick to hij1nx
16:38:02  * zz_jcejoined
16:38:05  * zz_jcechanged nick to jce
16:44:48  * ericktquit (Quit: erickt)
16:53:41  * hij1nxquit (Read error: Connection reset by peer)
16:54:17  * hij1nxjoined
17:01:54  * ericktjoined
17:23:30  * hzquit
17:29:27  * brsonquit (Ping timeout: 240 seconds)
17:32:37  * brsonjoined
17:39:11  <bnoordhuis>back
17:49:40  * hij1nxquit (Quit: hij1nx)
17:51:56  <indutny>bnoordhuis: me too
17:52:20  * benviejoined
17:52:47  <indutny>bnoordhuis: so the tests are passing, wanna review?
17:53:29  <bnoordhuis>indutny: i'm not really in a reviewing mood :-/
17:54:08  <indutny>bnoordhuis: ah, ok
17:54:15  * hij1nx_joined
17:54:22  <indutny>everyone else should be drunk away :P
17:54:27  <indutny>or will be such soon
17:54:30  <bnoordhuis>heh
18:00:38  <tjfontaine>indutny: I wanna see!
18:01:59  <indutny>oh noes
18:02:02  <indutny>benchmarks are so hard
18:02:16  <indutny>it's showing 408 req/s for original non-patched version now
18:02:25  <indutny>or has me taking lunch affected that?
18:26:26  * piscisaureus_quit (Ping timeout: 244 seconds)
18:28:39  * hij1nx_quit (Read error: Connection reset by peer)
18:33:05  * c4miloquit (Remote host closed the connection)
18:34:01  * c4milojoined
18:36:45  * hij1nx_joined
18:37:43  <indutny>bnoordhuis: http://lists.apple.com/archives/apple-cdsa/2008/Apr/msg00024.html
18:37:48  <indutny>SSL_OP_MICROSOFT_BIG_SSLV3_BUFFER
18:37:54  <indutny>really interesting stuff is in openssl's internals
18:38:21  <bnoordhuis>yeah, a lot of workarounds for buggy clients and servers
18:42:12  <indutny>indeed
18:49:15  * `3rdEdenjoined
18:49:15  <indutny>their comments just make me sad
18:49:16  <indutny> /* ssl3_read_bytes decided to call s->handshake_func, which
18:49:16  <indutny> * called ssl3_read_bytes to read handshake data.
18:49:16  <indutny> * However, ssl3_read_bytes actually found application data
18:49:16  <indutny> * and thinks that application data makes sense here; so disable
18:49:16  <indutny> * handshake processing and try to read application data again. */
18:55:05  <CIA-108>libuv: Ben Noordhuis v0.8 * r5031a5b / (4 files in 2 dirs): unix: rename linux/core.c to linux/linux-core.c - http://git.io/XKcEDA
18:56:07  * loladiroquit (Ping timeout: 248 seconds)
18:57:09  * travis-cijoined
18:57:10  <travis-ci>[travis-ci] joyent/libuv#483 (v0.8 - 5031a5b : Ben Noordhuis): The build is still failing.
18:57:10  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/cc1b3de24732...5031a5b85a3f
18:57:10  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/1769635
18:57:10  * travis-cipart
18:57:44  * loladirojoined
18:59:17  <CIA-108>node: Ben Noordhuis v0.8 * r5da7890 / (4 files in 2 dirs): deps: upgrade libuv to 5031a5b - http://git.io/T7f-kQ
18:59:17  <CIA-108>node: Ben Noordhuis v0.8 * rfc4e12b / (611 files in 209 dirs): tools: update gyp to r1426 - http://git.io/5-xFDg
19:00:25  <indutny>bnoordhuis: so, surprisingly, it looks like we're spending a lot of time trying to read from clear part of ssl stream
19:01:22  <indutny>bnoordhuis: I've an idea to manually parse packet headers in node_crypto.cc or tls.js, and using this information to check if we have any data to read here
19:01:31  <indutny>s/to manullay parse/of manually parsing
19:01:46  <indutny>and, of course, I meant SSL packets
19:02:23  <bnoordhuis>indutny: read from clear part of ssl stream <- can you expand on that?
19:02:38  <indutny>well, there're two ins and two outs here
19:02:56  <indutny>encrypted stream is directly piped to socket
19:03:04  <indutny>and socket is piped to the encrypted stream
19:03:27  <indutny>and on we're periodically checking if openssl is able to decrypt anything by trying reading from cleartext stream
19:03:49  <indutny>bnoordhuis: that's this part ^ :)
19:04:38  <bnoordhuis>okay
19:04:42  <bnoordhuis>and why is it so slow?
19:05:06  <indutny>bnoordhuis: that's really important question
19:05:09  <indutny>I've no idea
19:05:24  <indutny>investigating that in the openssl
19:05:32  <indutny>however I know only statistical information
19:05:36  <indutny>because I was using v8 tick processor
19:05:46  <indutny>I'll try using callgrind
19:06:02  <bnoordhuis>i was going to suggest that :)
19:06:14  <indutny>hehe
19:13:24  <bnoordhuis>[% 100|+ 149|- 6]: Done. <- out/Debug/run-tests on freebsd 9
19:13:33  <indutny>hm...
19:13:38  <indutny>not good, I suppose?
19:13:49  <bnoordhuis>it's really five failing tests, the last one fails because there's no ipv6 on that machine
19:13:59  <bnoordhuis>indutny: it's better than i expected actually :)
19:15:53  <indutny>:)
19:17:21  * hij1nx_quit (Quit: hij1nx_)
19:22:46  <indutny>bnoordhuis: oh, callgrind is slow as hell
19:23:00  <indutny>bnoordhuis: I just got a footprint from 300 requests, but it doesn't really contain any useful information
19:23:23  <bnoordhuis>indutny: use callgrind_annotate --auto=yes
19:23:48  <bnoordhuis>and yes, it's slow :)
19:24:52  <indutny>yeah, I'm going to do at least few thousands of requests
19:25:01  <indutny>btw, is it tracking all the calls
19:25:28  <indutny>I suppose it's wrapping calls with some runtime stuff, right?
19:26:18  <bnoordhuis>indutny: no, it's running your code in a kind of vm
19:26:28  <bnoordhuis>it's more like emulation than profiling, really
19:26:32  <indutny>oh crap
19:26:38  <indutny>why not just replacing calls?
19:26:48  <indutny>it would be much faster, I suppose
19:26:53  <bnoordhuis>because that doesn't give accurate profiling data
19:26:55  <bnoordhuis>emulation does
19:27:04  <indutny>well, I actually just got it
19:27:14  <indutny>I think that's because you may be using shared libraries
19:27:34  <indutny>and it's cheapier to emulate than to transpile everything
19:28:25  <bnoordhuis>indutny: it's something of a mix between emulation and virtualization
19:28:30  <indutny>ok
19:28:35  <bnoordhuis>read the valgrind source sometime if you have the chance, it's quite interesting
19:28:47  <bnoordhuis>also shows what a pain in the ass the x86 instruction set is
19:29:11  <indutny>I already know that
19:29:14  <indutny>haha
19:29:17  <indutny>don't tell me about x86
19:29:25  <indutny>but I'll look at source, definitely
19:29:37  <indutny>so, do you want to see a footprint?
19:30:09  <indutny>it's not really interesting btw
19:30:16  <indutny>call times doesn't give me anything
19:30:33  <indutny>is it tracking time spent in function?
19:31:00  <bnoordhuis>indutny: well, a top 10 of functions would be interesting
19:31:07  <indutny>they ain't
19:31:14  <indutny>it's basically v8's ones an openssl's ones
19:31:20  <indutny>3,069,644,256 out/../deps/openssl/openssl/crypto/bn/bn_asm.c:bn_mul_add_words [./node_g]
19:31:20  <indutny>2,226,878,645 out/../deps/v8/src/handles.cc:v8::internal::HandleScope::ZapRange(v8::internal::Object**, v8::internal::Object**) [./node_g]
19:31:20  <indutny>1,457,725,640 out/../deps/openssl/openssl/crypto/bn/bn_asm.c:bn_sqr_comba8 [./node_g]
19:31:20  <indutny> 626,517,245 out/../deps/openssl/openssl/crypto/bn/bn_asm.c:bn_mul_comba8 [./node_g]
19:31:20  <indutny> 595,936,143 out/../deps/openssl/openssl/crypto/bn/bn_mont.c:BN_from_montgomery_word [./node_g]
19:31:20  <indutny> 349,813,980 out/../deps/v8/src/isolate.h:v8::internal::Isolate::Current() [./node_g]
19:31:20  <indutny> 279,851,196 out/../deps/v8/src/platform.h:v8::internal::Thread::GetExistingThreadLocal(v8::internal::Thread::LocalStorageKey) [./node_g]
19:31:21  <indutny> 210,228,864 out/../deps/openssl/openssl/crypto/sha/sha_locl.h:sha1_block_data_order [./node_g]
19:31:23  <bnoordhuis>okay, optimize those :)
19:31:27  <indutny>haha
19:31:34  <bnoordhuis>ah, zaprange. it's a debug build?
19:31:38  <indutny>yes
19:31:45  <bnoordhuis>what do you see with a release build?
19:31:50  <indutny>one moment
19:31:52  <indutny>ok, I just got it
19:31:58  <indutny>v8 was just zapping all the memory
19:32:02  <indutny>most of the time :P
19:32:16  <bnoordhuis>yeah. debug builds are usually pretty much dominated by v8
19:32:20  <indutny>yeah
19:32:22  <indutny>thanks
19:34:56  <indutny>bnoordhuis: btw, have you seen candor-lang.org ?
19:35:09  <bnoordhuis>indutny: no. should i?
19:35:21  <indutny>bnoordhuis: well, you decide
19:35:38  <bnoordhuis>indutny: ooh, sexy
19:35:50  <bnoordhuis>now to find some 0-day exploits in the candor vm :)
19:35:57  <indutny>hahahaha
19:36:01  <indutny>well, that's not a VM
19:36:12  <indutny>it translates candor to the javascript
19:36:28  <bnoordhuis>you've been busy
19:36:56  <indutny>yeah
19:36:57  <indutny>sort of
19:37:07  * hij1nxjoined
19:40:00  * xaqjoined
19:40:06  <indutny>bnoordhuis: so node_crypto.cc is on like 60 place of footprint
19:40:15  <bnoordhuis>well, that's good
19:40:17  <indutny>and bignumbers are still dominating there
19:40:33  <bnoordhuis>are we using the asm versions now?
19:40:51  <bnoordhuis>i believe bn_asm.c doesn't actually contain assembly
19:41:02  <indutny>one sec
19:41:07  <indutny>I'll gist footprint
19:41:40  <indutny>bnoordhuis: https://gist.github.com/54a628b81d538f0b2ff1
19:42:06  * hij1nxquit (Quit: hij1nx)
19:42:32  <indutny>bnoordhuis: yeah, it's plain C
19:42:35  <bnoordhuis>indutny: speed up bn_mul_add_words and bn_sqr_comba8 by a factor of 10 and we can go home early :)
19:42:42  <tjfontaine>haha
19:42:45  <indutny>well
19:42:52  <indutny>the problem is not really in speed
19:42:56  <indutny>but in the amount of calls to them
19:43:09  <indutny>2 billions is not really good number :)
19:43:15  <indutny>(almost 3)
19:43:22  <bnoordhuis>indutny: that's not calls, that's instructions executed
19:43:28  <indutny>oh
19:43:34  <indutny>so it has spent time here
19:43:40  <bnoordhuis>that's what i meant when i said that callgrind is a precise profiler :)
19:43:46  <bnoordhuis>yes, exactly
19:43:47  <indutny>yeah
19:43:59  <indutny>would be good to see stack traces for it
19:44:06  <indutny>this information is not really useful
19:44:12  <bnoordhuis>indutny: you can
19:44:20  <bnoordhuis>i don't remember the switch off the top of my head
19:44:24  <bnoordhuis>but it's in the man page somewhere
19:45:19  <indutny>--tree=caller
19:45:23  <tjfontaine>indeed
19:45:37  <indutny>it doesn't display anything for the first one
19:46:47  * hij1nxjoined
19:47:11  <indutny>ok, --tree=both was much more helpful
19:47:12  <kohai>ok has -1 beer
19:48:06  <bnoordhuis>hah
19:48:28  <indutny>kohai: you are dumb
19:48:28  <kohai>I am Kohai, semi-useful communications-facilitating pseudointelligence!
19:48:43  <bnoordhuis>stress on 'semi-useful'
19:50:09  <mmalecki>:(
19:50:24  <mmalecki>bnoordhuis: you're cruel!
19:50:35  <tjfontaine>kohai: botsnack
19:50:35  <kohai>I am Kohai, semi-useful communications-facilitating pseudointelligence!
19:50:40  <tjfontaine>tsk tsk
19:50:50  <indutny>hm...
19:50:54  <indutny>still nothing interesting
19:51:41  <bnoordhuis>mmalecki: only when i can get away with it
19:51:56  <bnoordhuis>indutny: might be worth checking if openssl is using asm at all
19:52:16  <bnoordhuis>we may (explicitly or implicitly) compile with OPENSSL_NO_ASM
19:52:22  <mmalecki>kohai is a very good bot!
19:52:22  <kohai>'Bot' is a derogatory term, and I'm offended.
19:53:36  <bnoordhuis>indutny: there's OPENSSL_NO_INLINE_ASM too
19:54:00  <indutny>oook
19:55:03  <mmalecki>!insult bnoordhuis
19:55:04  <kohai>bnoordhuis smells of elderberry wine!
19:55:19  <mmalecki>bnoordhuis: you can no longer get away with it!
19:55:32  <indutny>!insult kohai
19:55:32  <kohai>I punt thee squarely in the privy-stones, kohai!
19:55:55  <tjfontaine>self mutilation!
19:56:25  <indutny>oh noes, it's rebuilding v8 again
19:56:46  <indutny>I hope one day GYP will be able to check define's uses to see what files should be recompiled
19:56:52  <indutny>it's not that hard, actually
19:57:16  <tjfontaine>neither is P=NP
19:57:31  <indutny>well
19:57:44  <indutny>that's not really related
19:57:50  <tjfontaine>heh
19:57:50  <indutny>because it's possible and really simple
19:57:59  <indutny>just save hashes of source files with inlined defines
19:58:06  <indutny>and check if something was changed
19:59:01  <bnoordhuis>indutny: it's open source, go fix it!
19:59:10  <indutny>bnoordhuis: em... I don't really like python
19:59:21  <indutny>I would be much happier it'll be written in C or C++
19:59:26  <bnoordhuis>you'll get used to it
19:59:29  <indutny>s/it'll//if it'll/
19:59:43  <indutny>I don't think so, I spent a lot of time with python years ago
20:00:40  <indutny>bnoordhuis: speed hasn't changed with OPENSSL_NO_ASM
20:00:56  <indutny>that's definitely not good
20:01:04  <indutny> 'Release': {
20:01:05  <indutny> 'defines': [ 'OPENSSL_NO_ASM' ],
20:01:07  <indutny>that's how I defined it
20:01:17  <bnoordhuis>indutny: OPENSSL_NO_ASM <- disables asm optimized functions
20:01:24  <indutny>yeah
20:01:26  <indutny>indeed
20:01:33  <bnoordhuis>so if it's just as fast, that means we're not using the optimized versions
20:01:39  <indutny>indeed
20:01:48  <indutny>obviously
20:02:04  <bnoordhuis>so that should be a conceptually easy fix and quick win
20:02:09  <bnoordhuis>yeah, i just said quick win
20:02:32  * `3rdEdenquit (Quit: Leaving...)
20:02:52  <indutny>bnoordhuis: http://cdn.memegenerator.net/instances/400x/22912356.jpg
20:03:11  <bnoordhuis>haha, very true
20:04:16  * hij1nxquit (Quit: hij1nx)
20:04:21  <indutny>you can just use it as autoreply
20:07:42  <indutny> # performance stuff
20:07:42  <indutny> 'OPENSSL_BN_ASM_PART_WORDS',
20:07:42  <indutny> 'OPENSSL_BN_ASM_MONT',
20:07:42  <indutny> 'MD5_ASM',
20:07:42  <indutny> 'SHA1_ASM',
20:07:43  <indutny> 'RMD160_ASM',
20:07:43  <indutny> 'SHA256_ASM',
20:07:44  <indutny> 'SHA512_ASM',
20:07:44  <indutny> 'AES_ASM'
20:07:51  <indutny>bnoordhuis: trying that ^
20:07:54  <bnoordhuis>good
20:08:30  * hij1nxjoined
20:08:35  <bnoordhuis>god, compiling v8 on a single core machine is a painful exercise
20:08:43  <indutny>haha
20:08:49  <indutny>indeed
20:08:55  <indutny>so why are you doing that?
20:09:05  <bnoordhuis>because it's an old single core server :)
20:09:38  <indutny>ok
20:09:42  <indutny>kinda "retro"
20:10:02  <bnoordhuis>yeah. it's my freebsd machine
20:10:49  <indutny>god, I'm building v8 kinda 10th time at this day
20:10:58  <indutny>and only 1 times yesterday
20:10:59  <indutny>:P
20:11:38  <indutny>bnoordhuis: I think openssl.gyp is simply not-full
20:11:47  <indutny>but let me try modified version first
20:11:49  <bnoordhuis>'not-full'?
20:12:25  <bnoordhuis>hah, it took nearly a minute to compile lithium-codegen-ia32.cc
20:13:32  <tjfontaine>is it clang freebsd or gcc freebsd?
20:14:59  <indutny>bnoordhuis: what are ".s" files for?
20:15:10  <bnoordhuis>tjfontaine: it's gcc 4.2
20:15:15  <tjfontaine>thems the assembly files?
20:15:18  <tjfontaine>bnoordhuis: k
20:15:23  <bnoordhuis>err, that's gcc freebsd
20:15:29  <indutny>tjfontaine: yeah, looks like
20:15:34  <indutny>but we ain't using them anywhere :(
20:15:38  <bnoordhuis>indutny: check for .s and .S
20:15:58  <tjfontaine>bnoordhuis: could be worse, could be gcc 2.95 I guess :)
20:17:15  * benviequit
20:17:24  * hij1nxquit (Quit: hij1nx)
20:17:56  <bnoordhuis>gcc 2.73 is still the best and that's a fact
20:18:27  <bnoordhuis>and by that i mean it was really, really buggy
20:18:47  <bnoordhuis>i remember it had this bug where it would sometimes for no reason optimize entire for loops away on x86
20:18:48  <indutny>so how can one build ".s" files?
20:18:52  <bnoordhuis>fun debugging, that
20:18:58  <bnoordhuis>but i digress
20:19:07  <bnoordhuis>indutny: you run them through gas
20:19:31  <bnoordhuis>gyp should deal with them automatically, i think
20:19:52  <tjfontaine>ya, just call them a source file and then it should do the needfull
20:20:41  <tjfontaine>http://code.google.com/p/gyp/source/browse/trunk/test/win/asm-files/asm-files.gyp
20:20:41  <indutny>ok
20:20:46  <tjfontaine>for verification
20:21:20  <indutny>../deps/openssl/openssl/crypto/sha/sha1-x86_64.s:3:Unknown pseudo-op: .type
20:21:20  <indutny>../deps/openssl/openssl/crypto/sha/sha1-x86_64.s:3:Rest of line ignored. 1st junk character valued 115 (s).
20:21:20  <indutny>../deps/openssl/openssl/crypto/sha/sha1-x86_64.s:4:Alignment too large: 15. assumed.
20:21:20  <indutny>../deps/openssl/openssl/crypto/sha/sha1-x86_64.s:1280:Unknown pseudo-op: .size
20:21:20  <indutny>../deps/openssl/openssl/crypto/sha/sha1-x86_64.s:1280:Rest of line ignored. 1st junk character valued 115 (s).
20:21:21  <indutny>../deps/openssl/openssl/crypto/sha/sha1-x86_64.s:1282:Alignment too large: 15. assumed.
20:24:04  <indutny>googling it
20:25:45  <tjfontaine>you're on osx?
20:26:27  <bnoordhuis>indutny: what source tree is that? i don't seem to have that file
20:28:09  <indutny>tjfontaine: yes
20:28:14  <indutny>bnoordhuis: haha, indeed
20:28:21  <indutny>bnoordhuis: that's because we ain't compiling assembly at all
20:28:29  <indutny>trying to fix that atm
20:28:33  <bnoordhuis>ah, they're plain missing?
20:28:54  <tjfontaine>indutny: this may be related to gyp using as directly and not gcc/clang
20:28:54  <indutny>indeed
20:29:09  <indutny>tjfontaine: looks like so
20:29:28  <indutny>bnoordhuis: can we consider running "make -C deps/openssl/openssl/crypto" ?
20:30:24  <indutny>ah that won't help
20:30:36  <indutny>we need to compile files in that makefile somehow
20:32:26  <bnoordhuis>[04:33|% 100|+ 416|- 11]: Done <- `make test` (node) on freebsd 9
20:32:32  <indutny>em...
20:32:33  <indutny>worse?
20:34:23  <CIA-108>libuv: Ben Noordhuis v0.8 * r68b0c85 / test/test-hrtime.c : test: allow 80 ms intervals in hrtime test - http://git.io/DCuj2w
20:35:35  <bnoordhuis>indutny: no, that's node. the first one was libuv
20:35:57  <indutny>ah
20:36:03  <indutny>well, it's expected then
20:36:14  <indutny>what's the latest version, 11?
20:36:17  <indutny>freebsd 11
20:36:25  <indutny>oh, 9
20:36:31  <indutny>and we're not really good on it
20:36:32  <indutny>nice
20:36:33  <bnoordhuis>9 is stable, 10 is the next stable
20:36:40  * travis-cijoined
20:36:41  <travis-ci>[travis-ci] joyent/libuv#484 (v0.8 - 68b0c85 : Ben Noordhuis): The build is still failing.
20:36:41  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/5031a5b85a3f...68b0c85c0958
20:36:41  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/1770580
20:36:41  * travis-cipart
20:36:55  <bnoordhuis>ffs, mmalecki go fix that ^
20:37:46  <bnoordhuis>[04:25|% 100|+ 419|- 8]: Done <- second run :-/
20:38:04  <tjfontaine>what's failing there, I can't really tell
20:38:18  <bnoordhuis>tjfontaine: you fail as a psychic
20:38:21  <bnoordhuis>but i'll gist the results
20:38:24  <tjfontaine>not that one
20:38:27  <tjfontaine>the travis failure
20:38:51  <tjfontaine>I mean, I don't understand why gethostbyname is failing
20:39:40  <bnoordhuis>oh, it's probably because udp traffic is somehow blocked
20:39:51  <bnoordhuis>anyway, https://gist.github.com/39e8a1251aa3bbe33b81 <- `make test` on freebsd
20:40:06  <tjfontaine>well, I do udp out to google servers in node-dns without issue
20:40:43  <bnoordhuis>and incoming udp traffic?
20:40:58  <tjfontaine>well, it gets results back so ya
20:41:33  <tjfontaine>unless C vm's are different than nodejs vm's
20:41:44  <bnoordhuis>mmalecki: ^
20:43:12  <indutny>11th time I'm building v8 today
20:44:35  <bnoordhuis>i know the feeling
20:44:43  <bnoordhuis>i've been thinking about investing in a 16 core system
20:44:45  <tjfontaine>that freebsd upgrade server error is weird
20:45:04  <bnoordhuis>yeah, it's kind of worrisome. i'm going to look into that
20:45:38  <indutny>is golang using openssl?
20:45:55  <bnoordhuis>it's some kind of race because it doesn't always fail
20:45:58  <indutny>I think they're using it's own implementation
20:46:24  <indutny>I really wish the Candor VM will be as fast as v8 is
20:46:35  <indutny>and it'll still building in like 30 seconds
20:46:55  <indutny>at that days I'll be using only candor.io
20:47:02  <bnoordhuis>ah, i know what it is - someone even posted a patch for it
20:49:10  <tjfontaine>indutny: http://code.google.com/p/go/source/browse/?repo=crypto doesn't seem so
20:50:52  <bnoordhuis>but does go actually support tls?
20:51:07  <tjfontaine>http://golang.org/pkg/crypto/tls/
20:51:09  * tjfontaineshrugs
20:51:14  <tjfontaine>partially anyway
20:51:19  <indutny>that's it
20:51:24  <indutny>I seen a spdy module for it
20:52:01  <indutny>I think word *partially* means that "we do not do hacks here"
20:53:03  <tjfontaine>ah here it is http://code.google.com/p/go/source/browse?repo=default#hg%2Fsrc%2Fpkg%2Fcrypto%2Ftls
20:53:11  <bnoordhuis>a.k.a. "it will sometimes not work for no reason and you'll never be able to figure out why" :)
20:53:34  <tjfontaine>bnoordhuis: that's different how? :)
20:53:48  <indutny>no difference :P
20:54:48  <bnoordhuis>haha, okay. let me rephrase that as "will not work even more often"
20:55:19  <bnoordhuis>go code is pretty terse, i like that
20:56:00  <tjfontaine>it should have used llvm, google teams really have no idea what other teams are doing
20:56:03  <indutny>woooow
20:56:07  <indutny>537 req/sec
20:56:56  <indutny>https://gist.github.com/4cdf1f81da1953508fa9
20:57:39  <indutny>yeaaah
20:58:18  <bnoordhuis>indutny: what performance difference do you see?
20:58:21  <indutny>1.5x
20:58:31  * piscisaureus_joined
20:58:34  <bnoordhuis>indutny: try median-of-3
20:58:36  <tjfontaine>ut oh
20:58:41  <bnoordhuis>hey, look who drops by
20:58:57  <indutny>hahahaha
20:59:00  <indutny>I drop
21:00:07  <indutny>I think that it's not a limit
21:00:21  <indutny>as I only improved a bn speed
21:00:28  * perezdquit (Quit: perezd)
21:01:37  <indutny>so
21:01:42  <indutny>I'm shocked
21:01:53  <indutny>really considering buying a beer today
21:02:56  * perezdjoined
21:04:23  <bnoordhuis>don't do it, fedor, you'll end up like me
21:04:39  <indutny>hahahaha
21:05:16  <indutny>no I need to figure out how to run those perl scripts in gyp
21:05:23  <indutny>to bring even more inlined assembly to node
21:05:30  <bnoordhuis>don't go there
21:05:37  <bnoordhuis>it needs to be plain gyp
21:05:49  <tjfontaine>it's ugly ugly ugly
21:06:00  <tjfontaine>just run them manually for all the common platforms :)
21:06:10  <indutny>bnoordhuis: well, I'm not really planning to run this from gyp
21:06:17  <indutny>bnoordhuis: I think we may precompile them
21:06:23  <bnoordhuis>that's a better idea
21:06:35  <bnoordhuis>that reminds me
21:06:36  <bnoordhuis>piscisaureus_: ping
21:06:46  <tjfontaine>I looked at the script it just looked like a win64 exception
21:07:03  <piscisaureus_>bnoordhuis: wut?
21:07:12  <bnoordhuis>piscisaureus_: re libuv releases
21:07:31  <piscisaureus_>bnoordhuis: wut?
21:07:32  <bnoordhuis>piscisaureus_: i realized we may have to switch the plain makefile for autoconf
21:07:43  <indutny>hahaha
21:07:45  <bnoordhuis>at least, if you want uptake from the distros
21:07:45  <indutny>finally
21:07:54  <bnoordhuis>well... i'm not a big fan of autotools
21:07:58  <piscisaureus_>bnoordhuis: I don't give a fuck
21:08:10  <bnoordhuis>language, young man
21:08:13  <tjfontaine>bnoordhuis: the library shouldn't be terrible to maintain there, just a second place to update
21:08:41  <tjfontaine>bnoordhuis: I would suggest automake as well, in for a penny
21:09:20  <piscisaureus_>bnoordhuis: why do these ppl want autoconf?
21:09:28  <bnoordhuis>piscisaureus_: because that's all they know
21:09:36  <bnoordhuis>you should read the debian maintainers list sometime
21:09:38  <tjfontaine>consistency mostly
21:10:05  <bnoordhuis>even cmake is cause for a raised eyebrow or two
21:10:10  <bnoordhuis>let alone gyp :-/
21:10:10  * paddybyersquit (Quit: paddybyers)
21:10:56  <tjfontaine>well, gyp is in squeeze that's promising
21:11:51  * perezdquit (Quit: perezd)
21:13:30  <piscisaureus_>bnoordhuis: we have a plain old makefile right? Why is that not good?
21:13:44  <bnoordhuis>piscisaureus_: well, for one it's quite lacking :)
21:14:13  <piscisaureus_>bnoordhuis: well, whatever, I'll take patches if you are serious
21:14:13  <indutny>piscisaureus_: are you at hotel?
21:14:20  <piscisaureus_>indutny: in the audience
21:14:22  <indutny>or just with a notebook
21:14:24  <bnoordhuis>ah, the OSS way
21:14:27  <indutny>oh, in the audience
21:14:30  <piscisaureus_>indutny: I am looking at dap at the same time
21:14:32  <indutny>like, I'm in the court
21:14:35  <piscisaureus_>cuz he is talking
21:14:39  <indutny>dap?
21:14:47  <piscisaureus_>Dave Pacheco
21:14:59  <indutny>ok
21:15:04  <piscisaureus_>indutny: I will come to the court again after the talks
21:15:06  <piscisaureus_>:-)
21:15:11  <indutny>hahaha
21:15:11  <piscisaureus_>or earlier
21:15:15  <piscisaureus_>when I get fed up with it
21:15:31  <bnoordhuis>piscisaureus_: ask him if he's fixed the dtrace probes yet :)
21:15:34  <piscisaureus_>bnoordhuis: I will take your word for it, but I know nothing about autoconf and I don't care either
21:15:41  <piscisaureus_>bnoordhuis: he's talking about it
21:15:46  <piscisaureus_>:-p
21:15:51  <bnoordhuis>piscisaureus_: yell "fix it, slacker!"
21:16:08  <piscisaureus_>bnoordhuis: so I am afraid that you will have to build the autostuff stuff
21:16:23  <bnoordhuis>yeah, i anticipated that - no problem
21:17:09  <indutny>bnoordhuis: what platforms are we supporting?
21:17:20  <indutny>bnoordhuis: darwin/win/sunos/openbsd, all?
21:17:31  <bnoordhuis>indutny: linux, solaris, os x, windows
21:17:48  <bnoordhuis>and i try to keep the freebsd and openbsd code compiling
21:18:39  <indutny>ok
21:18:55  <indutny>I'm afraid that we'll need to have separate precompiled files for each of that platform
21:19:04  <indutny>probably I may port it to gyp somehow
21:19:31  <indutny>I actually want to say that openssl's build system is dumb
21:19:34  <indutny>completely dumb
21:24:59  <piscisaureus_>I'm offline for a bit
21:25:08  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
21:27:38  <bnoordhuis>indutny: that's an often-heard complaint about openssl
21:28:38  <indutny>so they're storing cross-platform assembly (well, sort of) in .pl files
21:28:49  <indutny>and running them later with platform option to get final ".s" files
21:29:34  <indutny>what I'm thinking about is that we can just have #ifdef's blocks in ".s" files, which will represent assembly code for each platform
21:29:47  <indutny>so we will only add an DEFINE in GYP to activate that chunks
21:29:50  <indutny>bnoordhuis: what do you think?
21:30:23  <bnoordhuis>indutny: i think that's going to be a royal pain every time we upgrade openssl
21:32:13  <indutny>bnoordhuis: well, I'll write a node script for it
21:32:24  <indutny>bnoordhuis: already thought about upgrades
21:32:32  <indutny>bnoordhuis: I'm not going to run each script manually :)
21:32:40  <indutny>there're a lot of them overhere
21:35:25  * perezdjoined
21:37:15  * rendarquit
21:40:23  * c4miloquit (Remote host closed the connection)
21:40:59  <indutny>bnoordhuis: are we supporting arm?
21:41:14  <indutny>oh it may exist without assembly, I suppose
21:41:16  <bnoordhuis>indutny: yes for certain values of 'yes'
21:48:48  * AvianFlujoined
21:49:22  * ibcjoined
21:50:18  <ibc>hi, hope I can get a response from the developers now :) => can I rely in the existence of cares.c integration? or will it be removed in the future from libuv?
21:53:28  * txdvquit (Ping timeout: 245 seconds)
22:02:59  <creationix>so, I want to fix uv_set_process_title, but I don't know where to start
22:03:08  <creationix>it doesn't seem to even work in node v0.8.1
22:04:02  <AvianFlu>creationix, there's a platform where it's basically a no-op� might be osx
22:04:12  <creationix>right, I'm working on osx
22:04:21  <creationix>setting it throws an error
22:04:29  <AvianFlu>I saw it in the code last week, one platform threw ENOTSUP, the other was a no-op
22:04:30  <creationix> return uv__new_artificial_error(UV_ENOSYS);
22:04:33  <AvianFlu>yeah
22:04:44  <creationix>and linux looked like a no-op
22:04:53  <creationix>but I thought node supported this somewhere
22:04:57  <AvianFlu>I would have sworn it worked on linux
22:05:01  <AvianFlu>I thought it was sunOS
22:05:03  <creationix>but It doesn't seem to be working there either
22:05:08  <creationix>node doesn't use this libuv function yet
22:08:27  <bnoordhuis>ibc: c-ares will be around for the foreseeable future
22:08:38  <bnoordhuis>ibc: but we may remove it someday, yes
22:09:02  <bnoordhuis>creationix: uv_set_process_title should work on linux
22:09:24  <creationix>bnoordhuis: when does uv_setup_args get called?
22:09:27  <creationix>(in darwin.c)
22:09:37  <ibc>ok, thanks a lot (I've open an issue about this question in Github, so I will close it now by adding your response). Thanks a lot
22:09:41  <bnoordhuis>creationix: are you talking about libuv or node?
22:09:46  <creationix>libuv
22:09:58  <creationix>or do I need to call it manually
22:09:59  <creationix>from luvit
22:10:03  <bnoordhuis>creationix: the latter
22:10:21  <creationix>cool, that explains why uv_get_process_title is not working for me
22:10:46  <creationix>so I see it returns a new char**
22:10:56  <creationix>should I use that as my new argv then?
22:12:25  <bnoordhuis>creationix: yes
22:12:26  <creationix>wohoo, getProcessTitle works in luvit
22:16:48  * ibcquit (Remote host closed the connection)
22:17:05  * AvianFluquit (Quit: Leaving)
22:22:27  * bnoordhuisquit (Ping timeout: 240 seconds)
22:33:41  <indutny>ircretary: tell bnoordhuis about https://github.com/joyent/node/pull/3629
22:33:41  <ircretary>indutny: I'll be sure to tell bnoordhuis
22:35:55  * piscisaureus_joined
22:36:13  <piscisaureus_>indutny: hey
22:36:19  <piscisaureus_>indutny: where are you?
22:40:19  * piscisaureus_quit (Client Quit)
22:43:30  * bcantrillquit (Quit: Leaving.)
22:49:28  * piscisaureus_joined
22:55:35  * perezdquit (Quit: perezd)
23:10:14  <piscisaureus_>ircretary: tell indutny I am at the theatre again. When you are at the Jupiter come and grab me there
23:10:15  <ircretary>piscisaureus_: I'll be sure to tell indutny
23:15:41  * piscisaureus_quit (Ping timeout: 255 seconds)
23:21:46  * sh1mmerjoined
23:24:42  * perezdjoined
23:28:34  * bcantrilljoined
23:44:44  <mmalecki>ircretary: tell bnoordhuis that node is running on node machines, there are no c machines
23:44:45  <ircretary>mmalecki: I'll be sure to tell bnoordhuis
23:50:10  * sh1mmerquit (Quit: sh1mmer)
23:53:41  * xaqquit (Remote host closed the connection)
23:57:23  * bcantrillquit (Quit: Leaving.)