00:02:06  <bnoordhuis>anyone seen this? http://armservers.com/2012/06/18/apache-benchmarks-for-calxedas-5-watt-web-server/
00:02:22  <bnoordhuis>like any benchmarks, take with a grain of salt
00:02:45  <bnoordhuis>but 5.26 watt at peak load is kind of nice :)
00:08:29  * dvv-androidquit (Remote host closed the connection)
00:08:46  * dvv-androidjoined
00:15:07  <dshaw_>Watts as a metric for server perf, trippy.
00:17:53  * isufyquit
00:19:19  * isufyjoined
00:25:38  * dvv-androidquit (Remote host closed the connection)
00:25:55  * dvv-androidjoined
00:34:50  <bnoordhuis>dshaw_: yeah, but not the worst one
00:35:02  <bnoordhuis>friends of mine operate one of the biggest DCs here in the netherlands
00:35:20  * hzjoined
00:35:24  <bnoordhuis>their utility bill is shocking
00:35:50  * hzquit (Client Quit)
00:36:19  <bnoordhuis>i half-kid but electricity is their biggest ongoing operational cost
00:38:26  * dvv-androidquit (Remote host closed the connection)
00:38:42  * dvv-androidjoined
00:41:24  <tjfontaine>oh the puns
00:41:25  * isufyquit
00:44:05  <bnoordhuis>yeah, i'm on a roll tonight
00:44:05  * dvv-androidquit (Remote host closed the connection)
00:44:07  <dshaw_>bnoordhuis: Just saw Ebay talk about their new cooling-less datacenters at Twitter ops. For big DC, power is definitely a driving concern.
00:44:22  * dvv-androidjoined
00:52:27  * dvv-androidquit (Remote host closed the connection)
00:52:43  * dvv-androidjoined
00:55:52  * dshaw_quit (Quit: Leaving.)
01:05:31  * dvv-androidquit (Remote host closed the connection)
01:05:47  * dvv-androidjoined
01:05:58  * isufyjoined
01:22:31  * toothrotquit (Quit: sjdafadgf)
01:25:13  * toothrjoined
01:25:56  * toothrquit (Changing host)
01:25:56  * toothrjoined
01:25:57  * dvv-androidquit (Remote host closed the connection)
01:26:13  * dvv-androidjoined
01:42:02  <bnoordhuis>anyone wants to buy a linux kernel 0-day information leak?
01:43:21  <bnoordhuis>that's admittedly not as hot as a root exploit
01:44:24  <tjfontaine>can you reproduce on android?
01:48:38  * dvv-androidquit (Remote host closed the connection)
01:48:54  * dvv-androidjoined
01:58:17  <CIA-108>libuv: Xidorn Quan master * raec320b / src/unix/darwin.c : darwin: get cpu model correctly on mac - http://git.io/seHO5g
01:59:19  <CIA-108>libuv: Ben Noordhuis master * r34fb759 / AUTHORS : Update AUTHORS - http://git.io/quKClA
02:00:02  <bnoordhuis>tjfontaine: send me an android phone and i'll tell you
02:00:05  <bnoordhuis>the answer is probably yes though
02:00:10  * travis-cijoined
02:00:10  <travis-ci>[travis-ci] joyent/libuv#456 (master - aec320b : Xidorn Quan): The build passed.
02:00:10  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/0005b52b7056...aec320b76ce1
02:00:10  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/1697702
02:00:10  * travis-cipart
02:00:20  <bnoordhuis>this is pretty core kernel networking code
02:00:41  <tjfontaine>bnoordhuis: well then you might be able to get some google monies this time
02:00:57  <bnoordhuis>tjfontaine: good idea!
02:01:23  * travis-cijoined
02:01:24  <travis-ci>[travis-ci] joyent/libuv#457 (master - 34fb759 : Ben Noordhuis): The build passed.
02:01:24  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/aec320b76ce1...34fb759508fe
02:01:24  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/1697736
02:01:24  * travis-cipart
02:05:44  * dvv-androidquit (Remote host closed the connection)
02:05:58  * dvv-androidjoined
02:15:51  * sh1mmerquit (Ping timeout: 250 seconds)
02:20:47  * dvv-androidquit (Remote host closed the connection)
02:21:04  * dvv-androidjoined
02:35:32  * dvv-androidquit (Remote host closed the connection)
02:35:48  * dvv-androidjoined
02:52:38  * dvv-androidquit (Remote host closed the connection)
02:52:55  * dvv-androidjoined
03:11:53  * dvv-androidquit (Remote host closed the connection)
03:12:10  * dvv-androidjoined
03:15:16  * bnoordhuisquit (Ping timeout: 245 seconds)
03:25:39  * brsonquit (Ping timeout: 252 seconds)
03:31:08  * dvv-androidquit (Remote host closed the connection)
03:31:25  * dvv-androidjoined
03:44:06  * dvv-androidquit (Remote host closed the connection)
03:44:23  * dvv-androidjoined
03:57:05  * dvv-androidquit (Remote host closed the connection)
03:57:21  * dvv-androidjoined
03:57:51  * isufyquit
03:59:50  * isufyjoined
04:04:40  * dvv-androidquit (Remote host closed the connection)
04:04:57  * dvv-androidjoined
04:17:32  * dvv-androidquit (Remote host closed the connection)
04:17:48  * dvv-androidjoined
04:30:40  * brsonjoined
04:31:39  * sh1mmerjoined
04:32:23  * dvv-androidquit (Remote host closed the connection)
04:32:40  * dvv-androidjoined
04:50:08  * dvv-androidquit (Remote host closed the connection)
04:50:25  * dvv-androidjoined
05:02:12  * isufyquit
05:04:46  * isufyjoined
05:10:58  * dvv-androidquit (Remote host closed the connection)
05:11:15  * dvv-androidjoined
05:24:12  * dshaw_joined
05:25:56  * dvv-androidquit (Remote host closed the connection)
05:26:13  * dvv-androidjoined
05:31:55  * mikealquit (Quit: Leaving.)
05:33:35  * mikealjoined
05:33:39  * mikealquit (Client Quit)
05:34:05  * mikealjoined
05:41:31  * dvv-androidquit (Remote host closed the connection)
05:41:48  * dvv-androidjoined
05:58:29  * dvv-androidquit (Remote host closed the connection)
05:58:46  * dvv-androidjoined
05:59:25  <CIA-108>node: isaacs v0.8 * r9183596 / (102 files in 8 dirs): npm: Upgrade to 1.1.31 - http://git.io/dbzgaQ
05:59:26  <CIA-108>node: isaacs v0.8 * r38250f6 / (98 files in 6 dirs): npm: Upgrade to 1.1.32 - http://git.io/GiKvhQ
05:59:26  <CIA-108>node: isaacs v0.8 * r20c5ca2 / (ChangeLog src/node_version.h): 2012.06.25, Version 0.8.0 (stable) - http://git.io/w9Ca3A
05:59:26  <CIA-108>node: isaacs v0.8 * re6cf29b / (105 files in 10 dirs): Merge branch 'v0.8.0-release' into v0.8 - http://git.io/I0O6ZA
05:59:27  <CIA-108>node: isaacs v0.8 * ref8d952 / src/node_version.h : Now working on 0.8.1 - http://git.io/W68hyg
06:03:02  <CIA-108>node: isaacs v0.8 * rbccadec / doc/blog/release/node-v0.8.0.md : blog post for v0.8.0 - http://git.io/kRrgdA
06:07:01  * dvv-androidquit (Remote host closed the connection)
06:07:17  * dvv-androidjoined
06:16:15  <CIA-108>node: isaacs v0.8 * re9a68ca / doc/blog/release/node-v0.8.0.md : blog: Add shasums to 0.8.0 blog post - http://git.io/CojdiQ
06:24:19  * dvv-androidquit (Remote host closed the connection)
06:24:35  * dvv-androidjoined
06:41:32  * dvv-androidquit (Remote host closed the connection)
06:41:49  * dvv-androidjoined
06:47:42  * mikealquit (Quit: Leaving.)
06:48:21  * mikealjoined
06:51:58  * dvv-androidquit (Remote host closed the connection)
06:52:15  * dvv-androidjoined
06:53:18  * stephankquit (Quit: *Poof!*)
06:59:07  * dvv-androidquit (Remote host closed the connection)
06:59:23  * dvv-androidjoined
07:08:00  * dvv-androidquit (Remote host closed the connection)
07:08:17  * dvv-androidjoined
07:18:09  * dvv-androidquit (Remote host closed the connection)
07:18:25  * dvv-androidjoined
07:24:36  * dvv-androidquit (Remote host closed the connection)
07:24:52  * dvv-androidjoined
07:30:13  * dvv-androidquit (Remote host closed the connection)
07:30:27  * dvv-androidjoined
07:38:56  * dvv-androidquit (Remote host closed the connection)
07:39:13  * dvv-androidjoined
07:43:19  * rendarjoined
07:49:47  * dvv-androidquit (Remote host closed the connection)
07:50:03  * dvv-androidjoined
07:53:02  * brsonquit (Quit: leaving)
07:59:49  * loladirojoined
08:00:02  * mmaleckijoined
08:02:29  * isufyquit
08:02:44  * dvv-androidquit (Remote host closed the connection)
08:03:01  * dvv-androidjoined
08:15:41  * dvv-androidquit (Remote host closed the connection)
08:15:59  * dvv-androidjoined
08:33:37  * dvv-androidquit (Remote host closed the connection)
08:33:54  * dvv-androidjoined
08:50:26  * toothrquit (Ping timeout: 240 seconds)
08:51:34  * hzjoined
08:52:06  * dvv-androidquit (Remote host closed the connection)
08:52:22  * dvv-androidjoined
08:54:12  * toothrjoined
09:06:24  * toothrquit (Ping timeout: 255 seconds)
09:06:24  * dvv-androidquit (Remote host closed the connection)
09:06:41  * dvv-androidjoined
09:09:20  * \toothrotjoined
09:20:18  * dvv-androidquit (Remote host closed the connection)
09:20:35  * dvv-androidjoined
09:22:58  <abraxas>isaacs: I guess you won't be adding the benchmark comparison with node 0.4 then? :)
09:23:18  <abraxas>isaacs: but good to see GYP got a mention
09:24:45  * theColejoined
09:28:59  * dvv-androidquit (Remote host closed the connection)
09:29:16  * dvv-androidjoined
09:47:48  * dvv-androidquit (Remote host closed the connection)
09:48:02  * dvv-androidjoined
10:06:59  * dvv-androidquit (Remote host closed the connection)
10:07:16  * dvv-androidjoined
10:24:08  * dvv-androidquit (Remote host closed the connection)
10:24:25  * dvv-androidjoined
10:42:14  * dvv-androidquit (Remote host closed the connection)
10:42:30  * dvv-androidjoined
11:05:26  * dvv-androidquit (Remote host closed the connection)
11:05:43  * dvv-androidjoined
11:23:26  * dvv-androidquit (Remote host closed the connection)
11:23:43  * dvv-androidjoined
11:39:40  * dvv-androidquit (Remote host closed the connection)
11:39:57  * dvv-androidjoined
11:52:41  * dvv-androidquit (Remote host closed the connection)
11:52:58  * dvv-androidjoined
11:59:43  * \toothrotquit (Ping timeout: 264 seconds)
12:06:17  * dshaw_quit (Quit: Leaving.)
12:11:56  * dvv-androidquit (Remote host closed the connection)
12:12:11  * dvv-androidjoined
12:12:42  * \toothrotjoined
12:16:30  * abraxasquit (Remote host closed the connection)
12:25:22  * \toothrotchanged nick to toothrot
12:25:27  * toothrotquit (Changing host)
12:25:27  * toothrotjoined
12:29:05  * dvv-androidquit (Remote host closed the connection)
12:29:22  * dvv-androidjoined
12:41:49  * theColequit (Quit: theCole)
12:46:23  * dvv-androidquit (Remote host closed the connection)
12:46:39  * dvv-androidjoined
13:04:10  * pthrasherquit (Quit: Leaving...)
13:04:10  * dvv-androidquit (Remote host closed the connection)
13:04:26  * dvv-androidjoined
13:06:14  * theColejoined
13:16:41  <indutny>good morning
13:19:08  * dvv-androidquit (Remote host closed the connection)
13:19:25  * dvv-androidjoined
13:34:11  * dvv-androidquit (Remote host closed the connection)
13:34:28  * dvv-androidjoined
13:41:47  * paddybyersjoined
13:50:41  * bnoordhuisjoined
13:51:20  * dvv-androidquit (Remote host closed the connection)
13:51:38  * dvv-androidjoined
13:55:51  * theColequit (Quit: theCole)
13:59:29  * theColejoined
14:06:12  * dvv-androidquit (Remote host closed the connection)
14:06:28  * dvv-androidjoined
14:07:33  * piscisaureus_joined
14:08:08  * pthrasherjoined
14:12:26  * toothrotquit (Ping timeout: 240 seconds)
14:14:12  * toothrjoined
14:14:43  * xaqjoined
14:14:47  * dvv-androidquit (Remote host closed the connection)
14:15:04  * dvv-androidjoined
14:17:46  * mmaleckiquit (Ping timeout: 244 seconds)
14:23:02  <isaacs>good morning
14:23:13  <isaacs>found a pretty gnarly bug last night, i guess
14:23:16  <isaacs>going to delay a little bit
14:25:47  <isaacs>https://github.com/joyent/node/issues/3528
14:28:27  * dvv-androidquit (Remote host closed the connection)
14:28:44  * dvv-androidjoined
14:29:18  * theColequit (Quit: theCole)
14:29:55  <isaacs>bnoordhuis: are you around? can you look at the diff bmc attached?
14:31:45  * theColejoined
14:33:42  <bnoordhuis>isaacs: looks plausible
14:34:06  * pthrasherquit (Quit: Linkinus - http://linkinus.com)
14:34:23  * thr4shjoined
14:37:41  <CIA-108>node: Ben Noordhuis master * r6003912 / test/simple/test-tls-set-ciphers.js : test: use RC4-MD5 cipher in tls test - http://git.io/Vyv4Zg
14:37:55  <creationix>that's so cool to see bcantrill's comments on the bug
14:38:10  <creationix>(re node #3528)
14:38:12  <CIA-108>libuv: saghul v0.8 * r431d61a / (src/win/error.c test/test-fs.c): win: map ERROR_PRIVILEGE_NOT_HELD to UV_EPERM - http://git.io/WYZj-A
14:38:20  <CIA-108>node: isaacs v0.8.0-release * r9a7158d / deps/uv/src/unix/sunos.c : Hotfix for #3528 Add strdup in uv_cpu_info - http://git.io/Ccja0g
14:38:21  <CIA-108>node: isaacs v0.8.0-release * r8b8a7a7 / (ChangeLog src/node_version.h): 2012.06.25, Version 0.8.0 (stable) - http://git.io/wApMSA
14:39:05  * dvv-androidquit (Remote host closed the connection)
14:39:18  * dvv-androidjoined
14:41:57  <bnoordhuis>hah, that code does indeed look copy/pasted
14:42:16  <bnoordhuis>that acpi comment doesn't make sense in that context
14:42:57  * travis-cijoined
14:42:57  <travis-ci>[travis-ci] joyent/libuv#458 (v0.8 - 431d61a : saghul): The build passed.
14:42:57  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/0387c23759e7...431d61af31ef
14:42:57  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/1702032
14:42:57  * travis-cipart
14:46:58  <isaacs>bnoordhuis: i'm landing that fix directly in node for this release.
14:47:12  <isaacs>bnoordhuis: but if it lgty, can you pull it into libuv, and we'll get it on the next update?
14:47:41  <isaacs>bnoordhuis: i'm not sure what specifically has to be the case tomake ti break like that. it's certainly not *all* smartos systems, but the only one i test on is some gigantor 24-CPU thing
14:47:53  <isaacs>so maybe there's some edge case it's hitting with a smaller machine, or in the gz, whatever.
14:48:30  * dvv-androidquit (Remote host closed the connection)
14:48:31  <isaacs>uploading new binaries in just a sec
14:48:42  <isaacs>then blog post etc.
14:48:46  * dvv-androidjoined
14:53:11  <indutny>whoa
14:54:49  <CIA-108>libuv: saghul master * r431d61a / (src/win/error.c test/test-fs.c): win: map ERROR_PRIVILEGE_NOT_HELD to UV_EPERM - http://git.io/WYZj-A
14:54:49  <CIA-108>libuv: Ben Noordhuis master * rd0816aa / src/unix/sunos.c : sunos: fix free() of non-malloc'd pointer - http://git.io/pZLMXA
14:54:49  <CIA-108>libuv: Ben Noordhuis master * ra416a58 / (src/unix/sunos.c src/win/error.c test/test-fs.c): Merge branch 'v0.8' - http://git.io/ZMngww
14:54:51  <CIA-108>libuv: Ben Noordhuis v0.8 * rd0816aa / src/unix/sunos.c : sunos: fix free() of non-malloc'd pointer - http://git.io/pZLMXA
14:55:19  <bnoordhuis>^ applied
14:55:26  <pfox__>so i know you guys are on the cusp of releasing 0.8, for which i commend you. but i have (yet another) uv_getaddrinfo question..
14:55:40  <indutny>pfox__: shoot :)
14:55:46  <pfox__>im testing running it against localhost and it returns three distinct addrinfo*'s, all of which are ipv4 addresses and all point to 127.0.0.1
14:55:50  <pfox__>in 64but ubuntu
14:55:58  <pfox__>is that expected?
14:55:59  <tjfontaine>blame your /etc/hosts
14:56:02  * pfox__nods
14:56:39  <isaacs>bnoordhuis: thanks
14:57:08  * dvv-androidquit (Remote host closed the connection)
14:57:25  * dvv-androidjoined
14:59:06  * travis-cijoined
14:59:07  <travis-ci>[travis-ci] joyent/libuv#459 (master - a416a58 : Ben Noordhuis): The build passed.
14:59:07  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/34fb759508fe...a416a585b25c
14:59:07  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/1702119
14:59:07  * travis-cipart
14:59:37  * travis-cijoined
14:59:37  <travis-ci>[travis-ci] joyent/libuv#460 (v0.8 - d0816aa : Ben Noordhuis): The build passed.
14:59:37  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/431d61af31ef...d0816aae5267
14:59:37  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/1702121
14:59:37  * travis-cipart
15:00:20  <CIA-108>node: isaacs v0.8 * r7823474 / doc/blog/release/node-v0.8.0.md : blog post for v0.8.0 (+6 more commits...) - http://git.io/s1d08Q
15:00:41  * benviequit
15:03:48  <piscisaureus_>How do compiled addons retrieve the node version?
15:04:02  <indutny>yeaaaaaaah
15:04:04  <tjfontaine>from the defines?
15:04:08  <piscisaureus_>yes
15:04:12  <piscisaureus_>but they are in node_version.h
15:04:15  <piscisaureus_>which is not publiv
15:04:17  <piscisaureus_>*public
15:04:28  <isaacs>piscisaureus_: yeah, that sounds not so great :)
15:04:29  <indutny>piscisaureus_: it's public
15:04:45  <indutny>piscisaureus_: tools/installer.js
15:04:52  <isaacs>alright, we're live
15:04:54  <piscisaureus_>it's also lame, actually :-)
15:04:58  <indutny>haha
15:05:03  <indutny>latter argument is more real
15:05:14  <piscisaureus_>indutny: the problem here is that we guarantee api stability
15:05:21  * isaacstopic: node.js v0.8.0 http://blog.nodejs.org/2012/06/25/node-v0-8-0/
15:05:32  <indutny>piscisaureus_: well, because we released it - we should guarantee it's stability now :P
15:05:41  <indutny>piscisaureus_: where you was two days before? :)
15:05:46  <piscisaureus_>so modules shouldn't deal with the patch version
15:06:02  <piscisaureus_>if they need to do that anyway, they should be using a runtime function
15:06:03  * dvv-androidquit (Remote host closed the connection)
15:06:07  <isaacs>piscisaureus_: oh, right.
15:06:08  <piscisaureus_>and not a define
15:06:15  <isaacs>we should've been having this conversation an hour ago :)
15:06:18  <piscisaureus_>I';
15:06:19  * dvv-androidjoined
15:06:20  <piscisaureus_>haha
15:06:23  <indutny>_X_NODEJS >= 080
15:06:40  <piscisaureus_>it can wait
15:06:47  <indutny>for 0.9.0 :)
15:06:51  <isaacs>yeah
15:07:08  <piscisaureus_>isaacs: I was just considering how I would do libuv versioning
15:07:22  <piscisaureus_>isaacs: that should be a runtime function, except that you want to have the API version in a define
15:07:38  <isaacs>piscisaureus_: why not do both?
15:08:00  <isaacs>oh, wait, no, i see.
15:08:17  <piscisaureus_>isaacs: well, then your binary addons are tied to a particular patch version
15:08:17  <indutny>isaacs: lets make a bets, how fast we'll release 0.8.1
15:08:24  <bnoordhuis>piscisaureus_: you should look at how libfuse does versioning
15:08:27  <indutny>I bet on 3-5 days
15:08:30  <bnoordhuis>piscisaureus_: it's horrible to maintain though
15:09:34  <piscisaureus_>bnoordhuis: executive summary?
15:09:59  <piscisaureus_>isaacs: io benchmarks look more serious now :-)
15:10:13  <bnoordhuis>piscisaureus_: you #define LIB_API_VERSION 0x800, then #include "lib.h"
15:10:22  <piscisaureus_>ugh
15:10:31  <bnoordhuis>piscisaureus_: and then the library bends over backwards to accommodate old versions of the api
15:10:35  <isaacs>indutny: well, if it's not in the next 4 days, it won't be until after nodeconf
15:10:36  <indutny>bnoordhuis: a lot a libc like
15:10:47  <indutny>isaacs: haha, because we'll drink to much here
15:11:01  <indutny>to release anything
15:11:06  <indutny>at least node
15:11:08  <bnoordhuis>indutny: libc's approach is more linker magic-like
15:11:09  * dvv-androidquit (Remote host closed the connection)
15:11:23  <piscisaureus_>the node 0.8 installer is much bigger than the 0.6 one
15:11:26  * dvv-androidjoined
15:11:35  <indutny>piscisaureus_: images added?
15:11:50  <indutny>or because of openssl?
15:11:58  <piscisaureus_>indutny: neither I suppose
15:12:02  <indutny>hm...
15:12:09  <piscisaureus_>indutny: 0.6 was 3MB, 0.8 is 9
15:12:34  <piscisaureus_>I think NPM got bigger :-)
15:13:41  <isaacs>piscisaureus_: hm. that seems odd.
15:13:52  <isaacs>piscisaureus_: how does it relate to the 0.7.x versions?
15:14:03  <bnoordhuis>heh, that guy on HN was *fast*
15:14:26  <piscisaureus_>7.8 is is much smaller
15:14:58  <piscisaureus_>ah
15:15:05  <piscisaureus_>I suspect it could be @japj's patch :-)
15:15:11  <piscisaureus_>Not that it hurts a lot
15:15:23  <isaacs>hikes
15:15:25  <isaacs>[node@ryannode ~]$ ls -laF dist/{v0.7.12,v0.8.0}/{,x64/}*.msi
15:15:25  <isaacs>-rw-r--r-- 1 node other 3887104 Jun 19 21:37 dist/v0.7.12/node-v0.7.12-x86.msi
15:15:28  <isaacs>-rw-r--r-- 1 node other 4239360 Jun 19 21:37 dist/v0.7.12/x64/node-v0.7.12-x64.msi
15:15:31  <isaacs>-rw-r--r-- 1 node other 9039872 Jun 25 14:55 dist/v0.8.0/node-v0.8.0-x86.msi
15:15:34  <isaacs>-rw-r--r-- 1 node other 10346496 Jun 25 14:55 dist/v0.8.0/x64/node-v0.8.0-x64.msi
15:16:03  <indutny>eichs!
15:16:04  <isaacs>piscisaureus_: def a change between 0.7.12 and 0.8.0
15:16:42  <isaacs>bnoordhuis: i know, he must've been just reloading the blog page or something
15:17:34  <creationix>stripping the node binary installed in osx makes the file go from 18Mb to 12Mb
15:18:04  <piscisaureus_>isaacs: japj's patches were in 0.7.12
15:18:16  <piscisaureus_>isaacs: so it most likely is blog or npm :-)
15:19:53  * dvv-androidquit (Remote host closed the connection)
15:20:10  * dvv-androidjoined
15:22:17  <isaacs>piscisaureus_: that's kind of weird.
15:22:39  <isaacs>piscisaureus_: the blog output isn't included in the msi, and it's just some markdown text files
15:22:47  <isaacs>piscisaureus_: maybe tools/blog
15:22:52  <isaacs>that has a bit more stuff in it...
15:23:14  <piscisaureus_>isaacs: actually, the installed size didn't really changed size
15:23:43  <piscisaureus_>isaacs: so either there is stuff in the msi that shouldn't be there - or the compression level got much worse
15:23:58  <isaacs>hm. tools/blog is smaler than wafadmin or closure-linter
15:24:00  <isaacs>i doubt that's it
15:24:56  * toothrquit (Ping timeout: 244 seconds)
15:27:25  <creationix>hmm, howtonode doesn't like node 0.8.0, probably my bug though, I'll check http://howtonode.org:8080/
15:27:59  <creationix>yeah, it was my bug
15:28:04  <creationix>:)
15:29:28  <piscisaureus_>isaacs: ha!
15:29:29  * dvv-androidquit (Remote host closed the connection)
15:29:39  <piscisaureus_>isaacs: the msi is embedded in the msi :-(
15:29:42  * toothrjoined
15:29:45  * dvv-androidjoined
15:29:59  <isaacs>piscisaureus_: wtf?
15:30:23  <isaacs>piscisaureus_: did i accidentally something in the build?
15:30:27  <piscisaureus_>isaacs: within the msi I find another msi that is 5mb in size
15:30:41  <isaacs>that is weird...
15:31:02  <piscisaureus_>isaacs: and within that one, there's another one :-)
15:31:24  <tjfontaine>inception msi
15:31:27  <piscisaureus_>babushka msi
15:31:37  <isaacs>whoaaa
15:32:03  <isaacs>piscisaureus_: how can that happen, though? I built the msi from the tarball
15:32:11  <piscisaureus_>isaacs: good question...
15:32:12  <isaacs>some change in it adding it to itself multiple times or something?
15:32:21  <piscisaureus_>maybe that
15:32:23  <piscisaureus_>I will dig
15:32:25  <piscisaureus_>this is very odd
15:32:26  <isaacs>k, thanks
15:32:32  <isaacs>it must be recent (like, since 0.7.12)
15:32:40  <isaacs>let's leave it as-is for now, and fix on 0.8.1
15:32:55  <isaacs>then we can brag that we made the MSI 60% smaller. #speeduploops
15:34:25  <piscisaureus_>definitely something really funny
15:34:33  <piscisaureus_>lemme try to build the (unsigned) msi on my machone
15:35:05  <piscisaureus_>isaacs: you build with nosnapshot, right?
15:35:29  * dvv-androidquit (Remote host closed the connection)
15:35:45  <isaacs>piscisaureus_: yes
15:35:45  * dvv-androidjoined
15:35:46  <isaacs>C:\node-v0.8.0-x64>vcbuild release msi nosnapshot x64
15:36:01  <piscisaureus_>ok, cool
15:36:02  <isaacs>C:\node-v0.8.0>vcbuild release msi nosnapshot
15:36:06  <piscisaureus_>I will try that
15:36:13  <piscisaureus_>maybe it's something obvious
15:36:20  <isaacs>and only one instance found:
15:36:21  <isaacs>bash-3.1$ /usr/bin/find.exe . -name *.msi
15:36:22  <isaacs>./Release/node-v0.8.0-x86.msi
15:44:11  * pieternjoined
15:45:55  * isufyjoined
15:46:07  <piscisaureus_>isaacs: mine is also too big
15:46:08  * dvv-androidquit (Remote host closed the connection)
15:46:15  <piscisaureus_>isaacs: so you did nothing wrong
15:46:26  * dvv-androidjoined
15:49:05  * isaacsfeels better
15:49:16  <isaacs>piscisaureus_: but that also means it must be a relatively recent bug
15:49:18  <isaacs>bisect?
15:56:30  * stephankjoined
15:56:36  * dvv-androidquit (Remote host closed the connection)
15:56:53  * dvv-androidjoined
16:00:29  <piscisaureus_>isaacs: sorry, standup
16:00:41  <piscisaureus_>isaacs: yes, will bisect. I can confirm that 0.7.12 also builds small here
16:06:06  * theColequit (Quit: theCole)
16:08:22  * japjjoined
16:08:57  <CIA-108>node: koichik v0.8 * rd7825b3 / (doc/api/_toc.markdown doc/api/all.markdown): docs: add punycode to TOC - http://git.io/CU9Lwg
16:09:35  * dvv-androidquit (Remote host closed the connection)
16:09:48  * dvv-androidjoined
16:09:58  <japj>gz on the node 0.8.0 release!
16:10:01  * dapjoined
16:12:23  <isaacs>thanks :)
16:12:31  * isaacsassumes "gz" is some kind of compliment :)
16:12:37  <creationix>hmm, childProcess.spawn is acting weird in 0.8.0
16:12:49  <tjfontaine>isaacs: or he's saying it's time to gzip the msi's
16:13:09  <isaacs>creationix: can you elaborate?
16:13:14  <creationix>I'm still trying to narrow it down
16:13:21  <isaacs>creationix: k.
16:13:24  * isaacsrapt in anticipation
16:13:29  <piscisaureus_>creationix: close vs exit event?
16:13:44  <japj>and just for the record: I still think it is Berts new icon that is too big in the installer ;)
16:14:00  <creationix>but basically in wheat (for howtonode), when I spawn ("git", ["--git-dir=/home/tim/howtonode.git", "show", "0fe2404e7d5265c6e47c68d764b8191e5f153ab5:articles"])
16:14:20  <creationix>it exits right away with code 0 and no data events on either stdout or stderr
16:14:28  <creationix>if I run the command manually in bash, it works
16:14:32  <creationix>but only for some commands
16:14:45  <creationix>(only broken for some in node 0.8.0)
16:15:11  <isaacs>creationix: s/on('exit'/on('close'
16:15:11  * dvv-androidquit (Remote host closed the connection)
16:15:19  <creationix>"exit"
16:15:20  <piscisaureus_>japj: the icon is 150 kb in size, and I agree that's too big
16:15:23  <isaacs>creationix: git exits without flushing its output buffers.
16:15:28  <isaacs>creationix: as do many programs
16:15:28  * dvv-androidjoined
16:15:37  <piscisaureus_>japj: however, it's odd that it adds 6mb to the msi
16:15:43  <isaacs>creationix: if you care about the output, you need to listen to close (which is, "the thing is over, AND output is closed")
16:15:58  <isaacs>creationix: or just pipe the stdout stream somewhere yourself. you'll continue to get data after the 'exit' event
16:16:26  <creationix>sure is, I got the data event after "exit"
16:16:38  <creationix>ok, so this is a know behavior change in 0.8 then?
16:16:48  <isaacs>creationix: yeagh, that was one case where we could not correct underlying problems without a semantics change.
16:16:58  <creationix>no problem, sorry for the trouble
16:17:03  <isaacs>creationix: there are cases (for instance, shared stdio fd's) where close will *never* happen
16:17:11  <creationix>right
16:17:13  <isaacs>creationix: or where you want to know the exit code, but don't care about the data.
16:17:16  <isaacs>etc.
16:17:37  <isaacs>so, we made 'exit' actually mean 'exit', and added 'close' for 'closed all the stdio'
16:17:46  <piscisaureus_>japj: gotcha.
16:17:48  * creationixshould have read the changes more carefully
16:17:48  <piscisaureus_>japj: https://github.com/joyent/node/commit/2dccebd4b829fbc53358568c348272f1e56522ba
16:18:36  <isaacs>creationix: that's the one that i think people will be bit by the most, honestly.
16:18:41  <tjfontaine>the icon of the exe is the exe?
16:18:45  <isaacs>creationix: but there's really no way around it
16:18:46  * dvv-androidquit (Remote host closed the connection)
16:18:54  <isaacs>creationix: or rather, the other ways around it were worse :)
16:19:01  * dvv-androidjoined
16:19:24  <isaacs>creationix: the other option was that we add a processExit event, and add close event, and have 'exit' continue to mean the same as 'close'
16:19:35  <isaacs>creationix: but really... i dunno. 'exit' should mean 'exit'
16:19:44  <creationix>isaacs: yeah, my code is basically doing CP.execFile, but was from before node had that
16:19:50  <creationix>isaacs: yeah, simple and correct is best
16:21:04  <japj>piscisaureus_: serious?
16:21:09  <piscisaureus_>japj: yes
16:21:15  <japj>piscisaureus_: I actually only meant that as a joke
16:21:17  <piscisaureus_>japj: I think it added node.exe twice
16:21:19  <japj>duh
16:21:34  <japj>so it adds the node.exe twice in the installer?
16:21:54  <piscisaureus_>japj: laaame
16:21:55  <tjfontaine>and doesn't add the .ico?
16:22:27  <piscisaureus_>I expected that it would extract the icon from the file
16:22:31  <piscisaureus_>instead, it just adds the file
16:22:33  <piscisaureus_>fail
16:22:39  <tjfontaine>indeed
16:23:17  <tjfontaine>also it took me a few times to parse ARPPRODUCTICON, I thought it was a typo on production
16:23:28  <piscisaureus_>ghe
16:23:35  <piscisaureus_>it's weird tho
16:23:43  * sh1mmerquit (Quit: sh1mmer)
16:23:47  <piscisaureus_>when I revert only that commit the msi is still 6,7 mb
16:24:00  <piscisaureus_>well
16:24:12  <piscisaureus_>I didn't revert - I pointed it a the ico file
16:25:26  <piscisaureus_>ah hmm, merge fail
16:26:41  * dvv-androidquit (Remote host closed the connection)
16:26:58  * dvv-androidjoined
16:27:36  <isaacs>piscisaureus_: how do you feel about adding userinfo lookup functionality to libuv?
16:27:55  <isaacs>like getpwnam stuff.
16:28:28  <creationix>alright, http://howtonode.org/ is running on 0.8.0!
16:28:29  <piscisaureus_>isaacs: I wouldn't mind - but I have to think about what a good interface would be
16:28:58  * creationixshould update wheat, there is code from the 0.2.x days still in there
16:29:44  <isaacs>piscisaureus_: well, of course, you know what my vote is: make it look like {get,set}pw{nam,uid}, and shim for whwatever the equivalent is on windows
16:30:37  <isaacs>it can incur IO, though, if it's not for the current uid's info
16:31:08  <isaacs>and if you're doing something silly like using active directories or whatnot, it can be pretty costly.
16:31:10  * dvv-androidquit (Remote host closed the connection)
16:31:25  * loladiroquit (Ping timeout: 252 seconds)
16:31:27  * dvv-androidjoined
16:31:43  * AndreasMadsenjoined
16:31:56  <piscisaureus_>isaacs: what part do you need, other than the user's home dir?
16:32:14  <isaacs>piscisaureus_: well, i'd like everything :)
16:32:26  <isaacs>piscisaureus_: i give you a name, you give me a uid, home dir, full name, whatever you've got.
16:32:28  <piscisaureus_>isaacs: what do you need the users pwd for :-p
16:32:30  <AndreasMadsen>isaacs: cluster.autoFork() should be replaced with cluster.fork() // https://github.com/joyent/node/issues/3532#issuecomment-6551996
16:32:42  <isaacs>piscisaureus_: i give you a name, you give me uid, home dir, ful name, etc.
16:33:07  <isaacs>piscisaureus_: don't need password. that's not actually what the passwd functions do. just stupid legacy unix naming confusing the issues, as usual.
16:33:23  <piscisaureus_>struct passwd {
16:33:23  <piscisaureus_> char *pw_name; /* username */
16:33:23  <piscisaureus_> char *pw_passwd; /* user password */
16:33:23  <piscisaureus_> uid_t pw_uid; /* user ID */
16:33:23  <piscisaureus_> gid_t pw_gid; /* group ID */
16:33:24  <piscisaureus_> char *pw_gecos; /* user information */
16:33:24  <piscisaureus_> char *pw_dir; /* home directory */
16:33:25  <piscisaureus_> char *pw_shell; /* shell program */
16:33:25  <piscisaureus_>};
16:33:29  <isaacs>piscisaureus_: once upon a time, /etc/passwd actually contained passwords. now the actual passwords are in /etc/shadow
16:33:39  <isaacs>piscisaureus_: right, but if you do getpwnam on another user, you don't get their password.
16:33:44  <piscisaureus_>phew
16:33:46  <piscisaureus_>:-p
16:33:47  <isaacs>piscisaureus_: you can SET it if you are them, or if you are rot.
16:33:49  <isaacs>*root
16:37:35  * dvv-androidquit (Remote host closed the connection)
16:37:52  * dvv-androidjoined
16:44:32  <CIA-108>node: Bert Belder v0.8 * r3f932c5 / tools/msvs/msi/product.wxs : windows msi: don't use .exe file as icon - http://git.io/Rh00tw
16:44:32  <CIA-108>node: Bert Belder v0.8 * r41a4b73 / src/res/node.ico : windows: remove 256x256 version of the icon - http://git.io/0ej9RA
16:44:39  <piscisaureus_>^-- japj: happy now ?
16:45:03  <piscisaureus_>My installer is now 3897 kb
16:45:09  <tjfontaine>excellent
16:45:19  * AndreasMadsenquit (Remote host closed the connection)
16:46:15  * dvv-androidquit (Remote host closed the connection)
16:46:32  * dvv-androidjoined
16:50:19  * dshaw_joined
16:51:42  <piscisaureus_>isaacs: I am checking out how to map pwd functions on windows.
16:52:11  <piscisaureus_>isaacs: I mean, if most of the fields have no equivalent anyway, there's not much reason to use libuv for this stuff right?
16:53:19  <isaacs>piscisaureus_: sure, i mean, we could just do it in node, too.
16:53:28  * dapquit (Quit: Leaving.)
16:53:33  <piscisaureus_>isaacs: yes :-)
16:53:40  <isaacs>piscisaureus_: but it's really a pita to not have this functionality in a server/systems programming environment.
16:54:01  <isaacs>piscisaureus_: i'm not suggesting we go whole-hog and do all the crazy ~-matching bs.
16:54:03  <piscisaureus_>isaacs: there are some concepts that map well, like user name, real name and home dir
16:54:05  <isaacs>that can be donein userland
16:54:39  <isaacs>piscisaureus_: but, there IS the fact that getpwnam can do IO, and be extremely slow, so we should have a uv_queue_work version and a blahSync version
16:54:40  * dvv-androidquit (Remote host closed the connection)
16:54:56  * dvv-androidjoined
16:55:17  <isaacs>we actually do this in a few cases internally, so that you can do `setuid('nobody')` instead of looking up the uid yourself
16:55:30  <piscisaureus_>isaacs: windows has users and groups too (although a guid is used for their ID), but their semantics are so wildly different that it probably makes to sense to make libuv return that
16:55:39  * TheJHjoined
16:56:22  <piscisaureus_>isaacs: what I would be interested in: "are we running with root/admin caps" (which actually is not really tied to the user on windows)
16:56:32  <isaacs>piscisaureus_: does windows let you change the user/group ever?
16:57:00  <piscisaureus_>isaacs: change, in what context?
16:57:13  <isaacs>piscisaureus_: if so, the result of require('pwd').groupSync() should be valid as an argument to process.setgid()
16:57:15  <piscisaureus_>isaacs: a program cannot change its uid but it can spawn other programs as another user
16:57:22  <isaacs>piscisaureus_: oh, wild.
16:57:23  <isaacs>ok
16:57:55  <isaacs>so, in theory at least, we would be able to do spawn(c, a, { uid: blah }) but not process.setiud(blah)
16:58:13  <isaacs>(neither are supported on windows atm, i believe)
16:58:15  <piscisaureus_>isaacs: yes. that wouldn't work for "setgid" btw - programs don't run in a particular group context
16:58:30  <isaacs>k
16:59:15  <piscisaureus_>isaacs: but I can see the use case; in windows you would sometimes start a node process as elevated admin, but it should spawn child processes with less priviliges
16:59:22  <piscisaureus_>like, how chrome works
16:59:29  <isaacs>piscisaureus_: or how npm works :)
16:59:31  <piscisaureus_>and sort of the only portable use case of setuid in windows
16:59:31  <piscisaureus_>yes
16:59:36  * dvv-androidquit (Remote host closed the connection)
16:59:42  <piscisaureus_>I think we can do that
16:59:53  * dvv-androidjoined
17:00:09  <isaacs>but the paths thing... it's a bit vexing that there's just no way to expand ~ paths properly in node programs
17:00:30  <piscisaureus_>isaacs: I agree - although it was very nice for portability :-)
17:00:49  <piscisaureus_>isaacs: so getting the home dir is somewhing we definitely want
17:01:08  <isaacs>piscisaureus_: if we're going to do it, i think we ought to do it properly.
17:01:32  <isaacs>piscisaureus_: like, get whatever parts of the pwd struct that can be made portable
17:01:56  * dapjoined
17:06:26  * TooTallNatejoined
17:08:15  * dvv-androidquit (Remote host closed the connection)
17:08:32  * dvv-androidjoined
17:15:58  <piscisaureus_>isaacs: ok. lemme dig into this stuff for a while.
17:16:24  <isaacs>kewl
17:20:00  * dshaw_quit (Quit: Leaving.)
17:20:07  * dvv-androidquit (Remote host closed the connection)
17:20:24  * dvv-androidjoined
17:20:33  * japjquit (Ping timeout: 252 seconds)
17:26:08  * isufyquit
17:27:30  * isufyjoined
17:28:22  * isufyquit (Client Quit)
17:30:35  * dvv-androidquit (Remote host closed the connection)
17:30:51  * dvv-androidjoined
17:32:33  * isufyjoined
17:32:44  * japjjoined
17:39:42  * dvv-androidquit (Remote host closed the connection)
17:39:59  * dvv-androidjoined
17:40:45  * japjquit (Read error: Connection reset by peer)
17:47:26  * mjr__joined
17:50:23  * dvv-androidquit (Remote host closed the connection)
17:50:40  * dvv-androidjoined
17:51:11  * brsonjoined
17:58:40  * dvv-androidquit (Remote host closed the connection)
17:58:57  * dvv-androidjoined
18:03:29  * c4milojoined
18:04:12  * dvv-androidquit (Remote host closed the connection)
18:04:29  * dvv-androidjoined
18:09:39  * dshaw_joined
18:10:33  * `3rdEdenjoined
18:16:40  * dvv-androidquit (Remote host closed the connection)
18:16:54  * dvv-androidjoined
18:31:47  * hzquit (Ping timeout: 272 seconds)
18:32:32  * dvv-androidquit (Remote host closed the connection)
18:32:49  * dvv-androidjoined
18:46:03  * dvv-androidquit (Remote host closed the connection)
18:46:19  * dvv-androidjoined
18:59:57  * dvv-androidquit (Remote host closed the connection)
19:00:14  * dvv-androidjoined
19:01:20  * ryahjoined
19:01:32  * theColejoined
19:01:47  <ryah>congratulations everyone especially piscisaureus_, bnoordhuis, and isaacs
19:02:11  * ryahvirtually toasts a beer
19:02:17  <indutny>hm
19:02:28  <piscisaureus_>ryah: thanks :-)
19:02:29  <indutny>ryah: congrats! :)
19:02:38  <indutny>ryah: this would be impossible without you
19:02:40  * mikealquit (Quit: Leaving.)
19:02:42  <pfox__>yeah, big congrats libuv crowd
19:02:50  <pfox__>going to rebase our libuv submodule soon
19:02:52  <pfox__>hilarity ensues
19:03:02  <ryah>indutny: i think you mean, it would have been impossible with me :)
19:03:12  <indutny>hahaha
19:03:24  <indutny>well, you know better
19:03:43  <indutny>how can I claim
19:03:51  <ryah>https://github.com/igrigorik/node-spdyproxy <-- is this using node-spdy?
19:03:57  <indutny>yeees
19:04:02  <ryah>sweeet :D
19:04:08  <indutny>yeeees it is :)
19:04:15  <indutny>looks like people enjoy API
19:04:39  <indutny>and that's most pleasant part of anything I make
19:04:43  <indutny>when people enjoy API
19:04:56  <indutny>because underlying techonology may change, evolve and be replaced
19:04:56  * dvv-androidquit (Remote host closed the connection)
19:05:13  * dvv-androidjoined
19:05:14  <indutny>API is what makes node different from others, actually!
19:05:25  * mikealjoined
19:07:29  <TooTallNate>tjfontaine: i've made ref@0.0.19 use node-bindings
19:07:58  <tjfontaine>TooTallNate: groovy, thanks greatly, I also got the gyp cc
19:08:11  <TooTallNate>yup yup, that's 1/2 done
19:08:22  <TooTallNate>windows needs a tweak apparently
19:08:37  <tjfontaine>go figure :)
19:10:36  * ryahquit (Quit: leaving)
19:12:09  * dvv-androidquit (Remote host closed the connection)
19:12:25  * dvv-androidjoined
19:12:33  <tjfontaine>TooTallNate: I have this crazy idea to add a .toString to ref-struct, but I'll probably do that initially as monkey patch
19:15:07  * theColequit (Quit: theCole)
19:18:39  * dvv-androidquit (Remote host closed the connection)
19:18:55  * dvv-androidjoined
19:19:22  <TooTallNate>tjfontaine: ya totally, i've been meaning to add something like that
19:19:26  <TooTallNate>and/or .inspect()
19:21:10  * mmaleckijoined
19:23:54  * CoverSlidequit (Read error: No route to host)
19:24:29  * hzjoined
19:25:14  * dvv-androidquit (Remote host closed the connection)
19:25:28  * dvv-androidjoined
19:31:56  * dvv-androidquit (Remote host closed the connection)
19:32:12  * dvv-androidjoined
19:32:25  * isufyquit (Ping timeout: 244 seconds)
19:33:37  * isufyjoined
19:36:18  * mmaleckiquit (Ping timeout: 265 seconds)
19:40:03  <isaacs>piscisaureus_: https://twitter.com/3rdeden/status/217338276086939649
19:40:12  <isaacs>piscisaureus_: don't we have something like this now? that shows what's keeping the event loop open?
19:40:13  * dvv-androidquit (Remote host closed the connection)
19:40:29  * dvv-androidjoined
19:40:31  <piscisaureus_>isaacs: process._getActiveHandles() and process._getActiveReqs()
19:40:38  <isaacs>right. thanks :)
19:40:47  <piscisaureus_>isaacs: er, _getActiveRequests()
19:41:30  <piscisaureus_>`3rdEden: ^ that's for you, btw
19:44:23  <`3rdEden>ooh
19:44:38  <`3rdEden>thanks isaacs / piscisaureus_
19:44:55  <piscisaureus_>np :-)
19:48:51  * dvv-androidquit (Remote host closed the connection)
19:49:08  * dvv-androidjoined
19:52:06  <`3rdEden>works like a charm, awesome stuff. it seems like I have 9 leaking socket connections that are not yet closed
19:55:12  <isaacs>`3rdEden: it's still internal/private api, of course, as evidenced by the _
19:55:23  <isaacs>`3rdEden: we haven't figured out the best way to expose it nicely yet.
19:56:33  * isufyquit
20:03:17  * dvv-androidquit (Remote host closed the connection)
20:03:33  * dvv-androidjoined
20:05:30  <`3rdEden>isaacs yes, I understand but having this kind of information is pretty powerfull
20:05:46  <`3rdEden>(enough it seems pretty costy to gather the information)
20:05:51  * sh1mmerjoined
20:05:58  <`3rdEden>even though* it seems..
20:06:02  * sh1mmerquit (Remote host closed the connection)
20:06:15  * sh1mmerjoined
20:11:31  * isufyjoined
20:13:53  * dvv-androidquit (Remote host closed the connection)
20:14:09  * dvv-androidjoined
20:23:57  * dvv-androidquit (Remote host closed the connection)
20:24:13  * dvv-androidjoined
20:29:33  * mmaleckijoined
20:34:41  * dvv-androidquit (Remote host closed the connection)
20:34:57  * dvv-androidjoined
20:47:19  * dvv-androidquit (Remote host closed the connection)
20:47:35  * dvv-androidjoined
20:52:29  * dvv-androidquit (Remote host closed the connection)
20:52:45  * dvv-androidjoined
20:56:38  * brsonquit (Ping timeout: 244 seconds)
21:00:34  * sh1mmerquit (Read error: Connection reset by peer)
21:01:01  * sh1mmerjoined
21:03:14  * dvv-androidquit (Remote host closed the connection)
21:03:31  * dvv-androidjoined
21:05:46  * sh1mmerquit (Read error: Connection reset by peer)
21:06:14  * sh1mmerjoined
21:10:58  * dvv-androidquit (Remote host closed the connection)
21:11:14  * dvv-androidjoined
21:13:44  * brsonjoined
21:15:43  * `3rdEdenquit (Quit: Linkinus - http://linkinus.com)
21:18:57  <CIA-108>node: Andreas Madsen v0.8 * r3d51646 / doc/api/cluster.markdown : doc: replace references to cluster.autoFork - http://git.io/mlHurQ
21:20:10  * TheJHquit (Read error: Operation timed out)
21:22:02  * dvv-androidquit (Remote host closed the connection)
21:22:18  * dvv-androidjoined
21:23:46  * theColejoined
21:31:24  * dvv-androidquit (Remote host closed the connection)
21:31:40  * dvv-androidjoined
21:36:33  * theColequit (Ping timeout: 260 seconds)
21:37:53  * dvv-androidquit (Remote host closed the connection)
21:38:11  * dvv-androidjoined
21:40:14  * c4miloquit (Read error: Connection reset by peer)
21:40:28  * c4milojoined
21:40:47  * benviejoined
21:44:26  * dvv-androidquit (Remote host closed the connection)
21:44:43  * dvv-androidjoined
21:48:40  * sh1mmerquit (Quit: sh1mmer)
21:49:26  * rendarquit
21:53:43  * dvv-androidquit (Remote host closed the connection)
21:54:00  * dvv-androidjoined
21:55:59  * brsonquit (Ping timeout: 265 seconds)
21:56:39  * brsonjoined
22:02:43  * dvv-androidquit (Remote host closed the connection)
22:02:59  * dvv-androidjoined
22:12:05  * dvv-androidquit (Remote host closed the connection)
22:12:22  * dvv-androidjoined
22:19:34  * bbenviejoined
22:20:17  * benviequit (Read error: Connection reset by peer)
22:21:08  * dvv-androidquit (Remote host closed the connection)
22:21:24  * dvv-androidjoined
22:30:46  * dvv-androidquit (Remote host closed the connection)
22:31:02  * dvv-androidjoined
22:41:47  * dvv-androidquit (Remote host closed the connection)
22:42:04  * dvv-androidjoined
22:43:15  * hzquit
22:43:48  <CIA-108>libuv: Charlie McConnell master * r7573f4a / test/test-fs-event.c : test: improve clean-up in test-fs-event - http://git.io/F2O6Ow
22:43:48  <CIA-108>libuv: Charlie McConnell master * rf3fd8e3 / test/runner-unix.c : test: prevent running the tests as root - http://git.io/Ih0B0w
22:45:46  * travis-cijoined
22:45:46  <travis-ci>[travis-ci] joyent/libuv#461 (master - f3fd8e3 : Charlie McConnell): The build passed.
22:45:46  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/a416a585b25c...f3fd8e3c1bf9
22:45:46  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/1705995
22:45:46  * travis-cipart
22:50:39  * thr4shquit (Ping timeout: 246 seconds)
22:52:26  * dvv-androidquit (Remote host closed the connection)
22:52:42  * paddybyersquit (Quit: paddybyers)
22:52:43  * dvv-androidjoined
22:56:45  * c4miloquit (Remote host closed the connection)
23:00:47  * dvv-androidquit (Remote host closed the connection)
23:01:03  * dvv-androidjoined
23:11:41  * dvv-androidquit (Remote host closed the connection)
23:11:58  * dvv-androidjoined
23:16:02  <piscisaureus_>https://chromiumcodereview.appspot.com/10640021
23:16:09  <piscisaureus_>^-- just landed on 3.11
23:16:16  <piscisaureus_>looks like a security bug
23:24:38  * dvv-androidquit (Remote host closed the connection)
23:24:55  * dvv-androidjoined
23:29:56  * einarosquit (Ping timeout: 240 seconds)
23:30:10  * einarosjoined
23:30:26  * saghulquit (Ping timeout: 240 seconds)
23:31:02  * saghuljoined
23:31:56  * TooTallNatequit (Ping timeout: 240 seconds)
23:33:23  * dvv-androidquit (Remote host closed the connection)
23:33:40  * dvv-androidjoined
23:34:15  <piscisaureus_>bnoordhuis: https://gist.github.com/86ebb3ed4674516bf442
23:34:51  <bnoordhuis>polymorphic, degree 4. sounds sexy
23:35:41  <bnoordhuis>piscisaureus_: is it supposed to crash or silently 0wn my machine?
23:35:53  <piscisaureus_>bnoordhuis: crash, or silently own
23:35:58  <piscisaureus_>bnoordhuis: this should crash the debug version
23:35:59  <mmalecki>it sends your ~ to google
23:36:17  <bnoordhuis>ah, right
23:36:18  <bnoordhuis># Fatal error in ../deps/v8/src/x64/assembler-x64.cc, line 470
23:36:18  <bnoordhuis># CHECK(is_int8(disp)) failed
23:36:24  <piscisaureus_>ah
23:36:30  <piscisaureus_>that's another crash site as I get actually
23:36:38  <piscisaureus_>but I am running ia32
23:39:53  * dvv-androidquit (Remote host closed the connection)
23:40:09  * dvv-androidjoined
23:41:22  <bnoordhuis>piscisaureus_: is the idea that it erroneously generates a near jump with an address that's < -128 || > 127?
23:41:27  * isufyquit (Ping timeout: 244 seconds)
23:41:58  <piscisaureus_>bnoordhuis: yes
23:42:03  * isufyjoined
23:42:38  <bnoordhuis>piscisaureus_: right. i see in the release build that it generates a backwards jump which should probably be a forward jump
23:42:51  <bnoordhuis>it doesn't crash however
23:43:15  <piscisaureus_>bnoordhuis: well, the jump is probably just wrong
23:43:52  <piscisaureus_>bnoordhuis: the jump is based on type feedback that is no longer valid
23:43:59  <bnoordhuis>ah right
23:46:31  * dvv-androidquit (Remote host closed the connection)
23:46:38  <piscisaureus_>bnoordhuis: the jump is based on the fact that the objects generated by the "base" constructor have a particular hidden class
23:46:47  * dvv-androidjoined
23:47:01  <piscisaureus_>bnoordhuis: so v8 tries to optimize the "obj.foo" lookup that the crash() function does
23:47:39  <piscisaureus_>bnoordhuis: but after warming up the type feedback suddenly the object type changes (it no longer has a hidden class, since it now has indexed properties)
23:47:56  <piscisaureus_>bnoordhuis: this change however is not properly propagated into the optimized code
23:48:03  <bnoordhuis>right
23:48:03  <piscisaureus_>hence, splat
23:51:08  * benviejoined
23:51:26  * bbenviequit (Read error: Connection reset by peer)
23:56:00  * dvv-androidquit (Remote host closed the connection)
23:56:16  * dvv-androidjoined
23:59:25  * sh1mmerjoined