00:00:09  * ircretaryjoined
00:05:43  * defunctzombiechanged nick to defunctzombie_zz
00:07:16  * qardquit (Quit: Leaving.)
00:14:41  * wolfeidauquit (Remote host closed the connection)
00:18:38  * kazuponjoined
00:23:03  * kazuponquit (Ping timeout: 245 seconds)
00:27:12  * st_lukequit (Remote host closed the connection)
00:27:26  * wolfeidaujoined
00:29:39  <MI6>joyent/node: Ben Noordhuis v0.10 * 806e300 : src: fix memory leak in UsingDomains() - http://git.io/1HxrcA
00:32:00  <tjfontaine>interesting
00:36:00  * abraxasjoined
00:38:32  <MI6>nodejs-v0.10: #269 UNSTABLE smartos-x64 (1/593) http://jenkins.nodejs.org/job/nodejs-v0.10/269/
00:41:50  <MI6>joyent/node: Ben Noordhuis master * 30e7d08 : Merge remote-tracking branch 'origin/v0.10' (+10 more commits) - http://git.io/lsh1Rw
00:43:30  * jmar777quit (Remote host closed the connection)
00:43:59  * jmar777joined
00:46:17  <bnoordhuis>okay, off to bed. enjoy your independence day, yanks!
00:48:04  * jmar777quit (Ping timeout: 240 seconds)
00:48:46  * wolfeidauquit (Remote host closed the connection)
00:48:49  <tjfontaine>heh
00:50:24  * bnoordhuisquit (Ping timeout: 240 seconds)
00:51:32  <MI6>nodejs-v0.10-windows: #97 UNSTABLE windows-x64 (9/593) windows-ia32 (8/593) http://jenkins.nodejs.org/job/nodejs-v0.10-windows/97/
00:52:27  * dominictarrjoined
00:52:49  <MI6>nodejs-master: #284 UNSTABLE linux-ia32 (1/610) smartos-ia32 (2/610) osx-x64 (1/610) osx-ia32 (2/610) linux-x64 (1/610) smartos-x64 (11/610) http://jenkins.nodejs.org/job/nodejs-master/284/
00:55:31  <MI6>nodejs-master-windows: #91 UNSTABLE windows-x64 (18/609) windows-ia32 (17/609) http://jenkins.nodejs.org/job/nodejs-master-windows/91/
00:58:46  * amartensquit (Quit: Leaving.)
00:59:51  * wolfeidaujoined
01:06:15  * st_lukejoined
01:19:13  * kazuponjoined
01:24:26  * kazuponquit (Ping timeout: 268 seconds)
01:29:16  <MI6>nodejs-master-windows: #92 UNSTABLE windows-x64 (18/610) windows-ia32 (31/610) http://jenkins.nodejs.org/job/nodejs-master-windows/92/
01:29:50  * defunctzombie_zzchanged nick to defunctzombie
01:36:03  * amartensjoined
01:38:46  * wolfeidauquit (Remote host closed the connection)
01:44:28  * wolfeidaujoined
01:49:23  * DrPizzaquit (Ping timeout: 264 seconds)
01:52:56  * txdv_joined
01:55:46  * indutny_joined
01:57:27  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
01:57:35  * M28_joined
01:58:26  * mraleph1joined
01:58:35  <mordy_>btw; src/unix/thread.c:55: error: 'PTHREAD_MUTEX_ERRORCHECK' undeclared (first use in this function)
01:58:41  <mordy_>on EL5 with current master
01:59:21  * rphillipsquit (*.net *.split)
01:59:23  * txdvquit (*.net *.split)
01:59:24  * chiltsquit (*.net *.split)
01:59:26  * indutnyquit (*.net *.split)
01:59:26  * indutny_changed nick to indutny
01:59:47  * rphillipsjoined
02:02:00  * chiltsjoined
02:03:41  * CAPSLOCKBOT_joined
02:03:41  * LOUDBOT_joined
02:05:01  * niska`joined
02:05:16  * juliangruber_joined
02:05:30  * hij1nxquit (*.net *.split)
02:05:30  * M28quit (*.net *.split)
02:05:30  * mralephquit (*.net *.split)
02:05:30  * juliangruberquit (*.net *.split)
02:05:32  * kenperkinsquit (*.net *.split)
02:05:33  * rvaggquit (*.net *.split)
02:05:34  * niskaquit (*.net *.split)
02:05:35  * CAPSLOCKBOTquit (*.net *.split)
02:05:35  * LOUDBOTquit (*.net *.split)
02:05:36  * Denmadquit (*.net *.split)
02:06:50  * rvaggjoined
02:14:42  * groundwaterjoined
02:22:43  * st_lukequit (Remote host closed the connection)
02:31:30  * Denmadjoined
02:34:14  * groundwaterquit (Quit: groundwater)
02:39:54  * kazuponjoined
02:45:06  * kazuponquit (Ping timeout: 264 seconds)
03:00:28  * defunctzombiechanged nick to defunctzombie_zz
03:05:42  * AvianFluquit (Remote host closed the connection)
03:06:01  * qardjoined
03:06:32  * qardquit (Client Quit)
03:22:25  * groundwaterjoined
03:38:15  * TooTallNatequit (Quit: ["Textual IRC Client: www.textualapp.com"])
03:59:59  * kazuponjoined
04:27:18  * dominictarrquit (Quit: dominictarr)
04:56:27  * timoxleyjoined
05:00:59  * kenperkinsjoined
05:42:11  * kazuponquit (Read error: Connection reset by peer)
05:42:32  * kazuponjoined
05:47:19  * kazuponquit (Ping timeout: 264 seconds)
05:48:03  * kazupon_joined
06:03:05  * timoxleyquit (Quit: Computer has gone to sleep.)
06:04:26  * dsantiagojoined
06:11:29  * timoxleyjoined
06:12:18  * defunctzombie_zzchanged nick to defunctzombie
06:13:24  * kazuponjoined
06:16:29  * hij1nxjoined
06:16:44  * kazupon_quit (Ping timeout: 268 seconds)
06:20:28  * rendarjoined
06:29:26  * defunctzombiechanged nick to defunctzombie_zz
06:30:51  * `3E|Zzzchanged nick to `3rdEden
06:47:29  * wolfeidauquit (Remote host closed the connection)
07:30:22  * defunctzombie_zzchanged nick to defunctzombie
07:37:20  * stagasjoined
07:40:20  * defunctzombiechanged nick to defunctzombie_zz
07:55:08  * amartensquit (Quit: Leaving.)
07:58:05  * loladiroquit (Quit: loladiro)
07:58:42  * stagas_joined
08:00:57  * stagasquit (Ping timeout: 268 seconds)
08:01:07  * stagas_changed nick to stagas
08:07:36  * loladirojoined
08:09:24  * loladiroquit (Client Quit)
08:18:05  * wolfeidaujoined
08:28:31  * stagas_joined
08:28:36  * kellabytequit (*.net *.split)
08:28:37  * mburnsquit (*.net *.split)
08:29:45  * stagasquit (Ping timeout: 256 seconds)
08:31:44  * timoxleyquit (Quit: Computer has gone to sleep.)
08:33:28  * kellabytejoined
08:33:38  * stagas_quit (Ping timeout: 268 seconds)
08:42:34  * timoxleyjoined
08:44:15  * mburnsjoined
08:55:27  * amartensjoined
08:58:56  * raph_joined
09:00:09  * amartensquit (Ping timeout: 268 seconds)
09:06:28  <raph_>hi, there. I'm working my way through the nikhilm code samples but I'm stuck at the wrappiungpattern to write to pipes
09:06:36  <raph_>*wrapping pattern
09:07:04  <raph_>cfr the "write to pipe" section at http://nikhilm.github.io/uvbook/filesystem.html
09:08:09  <raph_>in particular I don't get why the write_req_t has to be dereferenced at
09:08:11  <raph_>write_req_t *req = (write_req_t*) malloc(sizeof(write_req_t));
09:08:13  <raph_>and
09:08:24  <raph_>write_req_t *wr = (write_req_t*) req;
09:08:59  <raph_>I know it's more of a general C question, but maybe anyone here could explain this
09:09:01  <raph_>thanks!
09:46:33  <kuplatupsu>I don't know why there is an explicit cast to write_req_t* in the malloc-line, it shouldn't need it. Maybe it tries to be more readable, since malloc returns void*
09:47:05  <kuplatupsu>raph_: and in the second line, req isn't write_req_t* but different type uv_write_t*
09:53:35  * bnoordhuisjoined
09:54:25  * raph_quit (Ping timeout: 268 seconds)
09:55:45  * amartensjoined
09:56:33  * raph_joined
09:56:55  <raph_>yup, thanks. the wrapping is needed because the buffer is out of scope on the callbacks...
09:58:36  <MI6>joyent/node: Tim Oxley v0.10 * 2d6d461 : doc: remove obsolete spawn() stdio options - http://git.io/Ilzq1g
10:00:42  * amartensquit (Ping timeout: 264 seconds)
10:08:06  <MI6>nodejs-v0.10: #270 UNSTABLE linux-x64 (2/593) osx-x64 (1/593) http://jenkins.nodejs.org/job/nodejs-v0.10/270/
10:15:41  * abraxasquit (Remote host closed the connection)
10:16:10  <MI6>nodejs-v0.10-windows: #98 UNSTABLE windows-x64 (7/593) windows-ia32 (8/593) http://jenkins.nodejs.org/job/nodejs-v0.10-windows/98/
10:16:40  <MI6>joyent/libuv: Ben Noordhuis v0.10 * 37d0209 : unix: fix build when !defined(PTHREAD_MUTEX_ERRORCHECK) - http://git.io/LT4D5Q
10:16:42  <MI6>joyent/libuv: Ben Noordhuis master * 13f73fa : Merge remote-tracking branch 'origin/v0.10' (+1 more commits) - http://git.io/IgFRMw
10:16:52  <bnoordhuis>mordy_: ^
10:24:24  <MI6>libuv-master-gyp: #82 UNSTABLE windows-x64 (3/192) smartos-ia32 (2/191) smartos-x64 (2/191) windows-ia32 (4/192) http://jenkins.nodejs.org/job/libuv-master-gyp/82/
10:26:10  <MI6>libuv-master: #144 UNSTABLE windows (3/192) smartos (2/191) http://jenkins.nodejs.org/job/libuv-master/144/
10:26:12  <MI6>libuv-v0.10: #102 UNSTABLE smartos (2/187) windows (5/188) http://jenkins.nodejs.org/job/libuv-v0.10/102/
10:28:30  <MI6>libuv-v0.10-gyp: #65 UNSTABLE windows-ia32 (3/188) smartos-x64 (4/187) windows-x64 (5/188) smartos-ia32 (2/187) http://jenkins.nodejs.org/job/libuv-v0.10-gyp/65/
10:29:47  * dominictarrjoined
10:44:48  <MI6>libuv-node-integration: #133 UNSTABLE linux-x64 (2/610) smartos-x64 (12/610) smartos-ia32 (3/610) osx-x64 (1/610) osx-ia32 (1/610) linux-ia32 (1/610) http://jenkins.nodejs.org/job/libuv-node-integration/133/
10:52:23  * timoxleyquit (Quit: Computer has gone to sleep.)
10:55:42  <MI6>libuv-node-integration: #134 UNSTABLE smartos-x64 (1/593) linux-ia32 (1/593) http://jenkins.nodejs.org/job/libuv-node-integration/134/
10:56:05  * amartensjoined
10:58:32  * timoxleyjoined
11:00:42  * amartensquit (Ping timeout: 264 seconds)
11:00:46  * kazuponquit (Remote host closed the connection)
11:01:14  * kazuponjoined
11:04:09  * dominictarrquit (Quit: dominictarr)
11:05:35  * kazuponquit (Ping timeout: 256 seconds)
11:18:45  * st_lukejoined
11:22:16  * hzjoined
11:39:03  * bnoordhu1sjoined
11:42:09  * st_lukequit (*.net *.split)
11:42:10  * bnoordhuisquit (*.net *.split)
11:52:17  * kazuponjoined
11:56:25  * amartensjoined
12:01:21  * amartensquit (Ping timeout: 264 seconds)
12:22:19  * bnoordhu1schanged nick to bnoordhuis
12:29:05  * kazuponquit (Remote host closed the connection)
12:29:37  * kazuponjoined
12:34:21  * kazuponquit (Ping timeout: 264 seconds)
12:46:55  * groundwaterquit (*.net *.split)
12:46:55  * rvaggquit (*.net *.split)
12:46:56  * txdv_quit (*.net *.split)
12:47:00  * saghulquit (*.net *.split)
12:47:02  * othiym23quit (*.net *.split)
12:47:03  * pquernaquit (*.net *.split)
12:47:03  * chobie4quit (*.net *.split)
12:52:38  * defunctzombie_zzquit (*.net *.split)
12:52:40  * nsmquit (*.net *.split)
12:52:42  * jez0990quit (*.net *.split)
12:52:43  * KiNgMaRquit (*.net *.split)
12:53:37  * rendarquit (*.net *.split)
12:53:39  * mikealquit (*.net *.split)
12:53:40  * philipsquit (*.net *.split)
12:53:43  * mmaleckiquit (*.net *.split)
12:53:44  * tjfontainequit (*.net *.split)
12:53:44  * isaacsquit (*.net *.split)
12:53:44  * chrisdickinsonquit (*.net *.split)
12:56:02  * KiNgMaRjoined
12:56:02  * jez0990joined
12:56:02  * nsmjoined
12:56:02  * defunctzombie_zzjoined
12:56:29  * groundwaterjoined
12:56:29  * rvaggjoined
12:56:29  * txdv_joined
12:56:29  * saghuljoined
12:56:29  * othiym23joined
12:56:29  * pquernajoined
12:56:29  * chobie4joined
12:56:37  * mikealjoined
12:56:37  * philipsjoined
12:56:37  * mmaleckijoined
12:56:37  * tjfontainejoined
12:56:37  * isaacsjoined
12:56:37  * chrisdickinsonjoined
12:59:54  * rendarjoined
13:11:54  * isaacschanged nick to Guest97044
13:11:54  * tjfontainechanged nick to Guest52282
13:13:08  * mmaleckiquit (Quit: Reconnecting)
13:14:08  * rendarquit (*.net *.split)
13:14:11  * niska`quit (*.net *.split)
13:14:12  * mraleph1quit (*.net *.split)
13:15:03  * mmaleckijoined
13:15:03  * rendarjoined
13:15:03  * niska`joined
13:15:03  * mraleph1joined
13:18:19  * timoxleyquit (*.net *.split)
13:18:20  * kenperkinsquit (*.net *.split)
13:18:22  * Denmadquit (*.net *.split)
13:18:24  * kellabytequit (*.net *.split)
13:18:26  * chiltsquit (*.net *.split)
13:18:27  * M28_quit (*.net *.split)
13:18:27  * indutnyquit (*.net *.split)
13:18:30  * hij1nxquit (*.net *.split)
13:18:35  * joshthecoder_quit (*.net *.split)
13:18:36  * trevnorrisquit (*.net *.split)
13:18:38  * ikquit (*.net *.split)
13:18:47  * mordy_quit (*.net *.split)
13:18:48  * dscape__quit (*.net *.split)
13:18:49  * `3rdEdenquit (*.net *.split)
13:18:49  * skebcioquit (*.net *.split)
13:18:50  * raph_quit (*.net *.split)
13:18:52  * CAPSLOCKBOT_quit (*.net *.split)
13:18:52  * LOUDBOT_quit (*.net *.split)
13:18:52  * rphillipsquit (*.net *.split)
13:18:58  * MI6quit (*.net *.split)
13:19:01  * Raltquit (*.net *.split)
13:19:02  * creationixquit (*.net *.split)
13:19:05  * ralquit (*.net *.split)
13:19:06  * kuplatupsuquit (*.net *.split)
13:19:10  * timoxleyjoined
13:19:10  * kenperkinsjoined
13:19:10  * Denmadjoined
13:19:14  * raph_joined
13:19:14  * LOUDBOT_joined
13:19:14  * CAPSLOCKBOT_joined
13:19:14  * rphillipsjoined
13:19:28  * MI6joined
13:19:28  * Raltjoined
13:19:28  * creationixjoined
13:19:28  * raljoined
13:19:28  * kuplatupsujoined
13:20:39  * kellabytejoined
13:20:39  * chiltsjoined
13:20:39  * M28_joined
13:20:39  * indutnyjoined
13:20:49  * mordy_joined
13:20:49  * dscape__joined
13:20:49  * `3rdEdenjoined
13:20:49  * skebciojoined
13:21:38  * M28_quit (Read error: Connection reset by peer)
13:22:46  * M28joined
13:25:34  * c4milojoined
13:26:56  * hij1nxjoined
13:26:56  * joshthecoder_joined
13:26:56  * trevnorrisjoined
13:26:56  * ikjoined
13:37:53  * mmaleckiquit (Quit: Reconnecting)
13:38:02  * mmaleckijoined
13:40:09  * kazuponjoined
13:44:30  <bnoordhuis>the netsplit is strong today
13:44:50  * kazuponquit (Ping timeout: 256 seconds)
13:56:51  * c4miloquit (Remote host closed the connection)
13:57:04  * amartensjoined
13:59:37  * jmar777joined
14:01:41  * amartensquit (Ping timeout: 248 seconds)
14:02:56  <indutny>heya
14:04:20  * st_lukejoined
14:08:06  <bnoordhuis>indutny: hoya
14:08:15  <indutny>how are you doing?
14:08:48  <bnoordhuis>great. it's friday 4 pm, a.k.a. beer time
14:09:06  <bnoordhuis>what about you, fedor?
14:18:02  * hzquit (Disconnected by services)
14:18:06  * hzjoined
14:20:12  * bnoordhuisquit (Ping timeout: 268 seconds)
14:23:40  * mmaleckiquit (Ping timeout: 276 seconds)
14:23:40  * ircretaryquit (Ping timeout: 276 seconds)
14:23:56  * hzquit (Disconnected by services)
14:23:59  * hzjoined
14:25:15  * mburnsquit (Ping timeout: 276 seconds)
14:25:50  * timoxleyquit (Quit: Computer has gone to sleep.)
14:28:29  * kellabytequit (Changing host)
14:28:29  * kellabytejoined
14:30:36  * timoxleyjoined
14:33:20  * mburnsjoined
14:33:43  * mmaleckijoined
14:36:54  * ircretaryjoined
14:39:20  * piscisaureus_joined
14:45:25  * st_lukequit (Remote host closed the connection)
14:45:59  * st_lukejoined
14:48:10  * pachetjoined
14:50:30  * st_lukequit (Ping timeout: 264 seconds)
14:54:52  <piscisaureus_>https://gist.github.com/anonymous/308c99b78364f3ba5f24
14:57:24  * amartensjoined
15:02:08  * amartensquit (Ping timeout: 268 seconds)
15:12:24  * bnoordhuisjoined
15:16:07  <bnoordhuis>piscisaureus_: ping bertje
15:16:18  <piscisaureus_>noordhuis
15:16:43  <piscisaureus_>bnoordhuis: sup?
15:16:45  <bnoordhuis>did you get a chance to look at my errno branch?
15:17:20  <piscisaureus_>let's se
15:17:45  <bnoordhuis>also this: http://www.nrc.nl/nieuws/2013/07/05/eerste-wolf-in-anderhalve-eeuw-dood-langs-de-weg/
15:18:55  <piscisaureus_>right
15:19:01  <piscisaureus_>bnoordhuis: where's that branch again?
15:19:19  <piscisaureus_>https://github.com/bnoordhuis/libuv/compare/errno ?
15:19:27  <bnoordhuis>yep
15:19:37  <bnoordhuis>last commit is the windows work
15:24:54  * defunctzombie_zzchanged nick to defunctzombie
15:28:22  <piscisaureus_>bnoordhuis: I find it quite disappointing that you made fs__capture_path return a windows error
15:29:59  <creationix>so it v0.11.5 the latest libuv release?
15:30:04  <creationix>s/it/is/
15:30:25  <Guest52282>bnoordhuis: re #5798 yes, that's exactly how the conversation went, it does one cb per tick
15:31:03  <bnoordhuis>Guest52282: identify yourself
15:31:08  <Guest52282>damnit
15:31:14  * Guest52282changed nick to tjfontaine
15:31:17  <bnoordhuis>:)
15:31:23  <bnoordhuis>creationix: yes
15:31:41  <creationix>there is a little interest in my luv library, so I wanted to update it to the latest libuv
15:31:47  <creationix>https://github.com/creationix/luv/commit/b3ff71acb295e34cca8a8787b355adcb8d2d6f61
15:31:53  <bnoordhuis>piscisaureus_: how so? i've made it so everything uses windows errors internally and only converts at the edges
15:32:01  <bnoordhuis>well, as much as reasonably possible anyway
15:32:20  * loladirojoined
15:32:34  <bnoordhuis>creationix: you may want to hold off a little, bert and i are just discussing a patch that will change _everything_
15:32:40  * jmar777quit (Remote host closed the connection)
15:32:41  <creationix>bnoordhuis, but the 0.11.x series is still considered unstable right?
15:32:46  <bnoordhuis>(i like hyperbole)
15:32:56  <bnoordhuis>yes, odd versions are unstable
15:33:13  <creationix>bnoordhuis, if it's the changed we talked about at nodeconf, I'll happily update to it
15:33:18  <piscisaureus_>bnoordhuis: so I find it confusing that you store windows errors in an int
15:33:34  <piscisaureus_>bnoordhuis: since we always use DWORD for that
15:33:58  <bnoordhuis>piscisaureus_: we talked about that, didn't we? they're always in the range 0-15,000ish right?
15:34:18  <piscisaureus_>bnoordhuis: oh yes that's true
15:34:36  <piscisaureus_>bnoordhuis: but I thought that you were going to do some crazy packing scheme
15:35:03  <bnoordhuis>ah no, turned out there was no need
15:35:07  <piscisaureus_>bnoordhuis: but really what you should normally do is DWORD x = GetLastError()
15:48:21  * kazuponjoined
15:53:36  * timoxleyquit (Quit: Computer has gone to sleep.)
15:57:46  * amartensjoined
15:59:23  <piscisaureus_>I like the new ReturnValue thing in v8
16:01:34  * dapjoined
16:02:28  * amartensquit (Ping timeout: 276 seconds)
16:02:32  <tjfontaine>ya, I do as well
16:05:05  * TooTallNatejoined
16:09:16  <bnoordhuis>43 files changed, 2141 insertions(+), 3513 deletions(-) <- nearing completion
16:10:28  <bnoordhuis>tjfontaine: re setImmediate, shouldn't callbacks scheduled from the same tick all run on the next tick?
16:10:42  <bnoordhuis>that can't starve the event loop, right? it always cycles
16:11:36  * arlolrajoined
16:11:37  <bnoordhuis>recursive setImmediate() shouldn't be an issue if you copy the queue and empty it before invoking the callbacks
16:13:12  <tjfontaine>depends how many you've scheduled prior to the execution, and what starve means for your application
16:17:36  <piscisaureus_>bnoordhuis: I was wondering... this situation where uv_idle runs on every tick
16:17:54  <bnoordhuis>piscisaureus_: yeah, but setImmediate() uses a check handle
16:18:07  <piscisaureus_>bnoordhuis: could that be why accept-backoff didn't work that well in v0.6+
16:18:18  <piscisaureus_>on unix :)
16:19:18  <bnoordhuis>maybe. that's ancient history by now
16:19:29  <bnoordhuis>the nanosleep(1) approach didn't work great either
16:19:42  <piscisaureus_>yeah that's true
16:20:23  <piscisaureus_>one of the major assumptions behind accept-backoff was that uv_idle did the same as ev_idle
16:20:57  <piscisaureus_>funny that I never realized unix was different for 1.5 years
16:29:30  <mordy_>hrm; i might just copy those windows headers wholesale from libuv -- anything to be aware of?
16:30:05  <mordy_>for some reason mingw doesn't want to provide GetQueuedCompletionStatusEx as an exported symbol
16:30:05  <trevnorris>bnoordhuis: thanks for catching that memory leak.
16:35:38  <piscisaureus_>mordy_: #define _WIN32_WINNT _WIN32_WINNT_VISTA
16:38:48  <mordy_>piscisaureus_: in my own project? -- i've defined WINVER=0x6000 (good enough for VC)
16:40:16  * qardjoined
16:40:36  <piscisaureus_>mordy_: ah it's that pesky ripoff of the sdk that doesn't have it
16:40:38  <piscisaureus_>oh well
16:41:19  <piscisaureus_>mordy_: if you only care about running on vista+ you might just copy the declaration from the msdn page. And add __declspec(dllimport)
16:42:08  <tjfontaine>trevnorris: so that 5799 is if they use the smalloc::alloc on an arbtrary jsobj from c++ *and* set a cb?
16:45:30  <TooTallNate>that's probably for, like, me :P
16:45:55  <tjfontaine>I'm just curious on the flow of how someone manages to do that
16:46:10  <TooTallNate>tjfontaine: seems like any time a C++ addon wants to pass around a pointer in JS-land
16:46:22  <tjfontaine>because Buffer::New + cb makes sense, but smalloc::alloc + cb is weird
16:46:59  * joshthecoder_changed nick to joshthecoder
16:47:00  <TooTallNate>tjfontaine: ya i agree... i don't think it's actually allocing anything when you pass a callback
16:49:13  <tjfontaine>I think the pattern he's describing is: smalloc::alloc+cb, js-land someone calls .dispose(), cb doesn't fire it free()'s it
16:49:30  <tjfontaine>it may be also that ::Dispose has the same flaw of course
16:50:18  * TooTallNatequit (Quit: Computer has gone to sleep.)
16:52:21  * mikealquit (Quit: Leaving.)
17:06:18  * TooTallNatejoined
17:11:36  * mikealjoined
17:12:02  * defunctzombiechanged nick to defunctzombie_zz
17:16:59  * st_lukejoined
17:21:45  * st_lukequit (Ping timeout: 264 seconds)
17:22:30  * pachetquit (Quit: leaving)
17:24:25  * pooyajoined
17:24:34  * amartensjoined
17:25:39  <trevnorris>tjfontaine: since the callback needs to be called in the WeakCallback, it's necessary that smalloc::Alloc can accept a callback.
17:26:21  <trevnorris>tjfontaine: so if a user allocates memory using smalloc::Alloc in c++ and passes a callback, then uses smalloc::Dispose or Buffer.dispose() on the object the memory in the object will be free'd.
17:26:40  <trevnorris>tjfontaine: but instead it should be passed to the callback that the user passed.
17:27:08  <tjfontaine>right, I thought the cb issue had been solved, I remember asking about it
17:27:22  <trevnorris>that was another issue.
17:27:40  <trevnorris>relating specifically for Buffers.
17:27:51  <trevnorris>and since I don't allow Buffer's to be disposed it's not an issue.
17:27:53  <tjfontaine>well, this is what backs buffers :)
17:28:08  <trevnorris>heh
17:28:25  <tjfontaine>the question was meant to be interpreted broadly, "what happens if I pass a cb and someone calls dispose, does the cb get called" the answer is now clearly "no" :)
17:28:31  * hzquit
17:28:40  <trevnorris>yes, the callback will get called.
17:28:47  <trevnorris>but the data passed will be NULL
17:29:15  <tjfontaine>but you're still calling free before sending it to them?
17:29:25  <tjfontaine>s/sending it to them/calling it/
17:29:36  <trevnorris>in the case of the callback I send the memory to the callback.
17:29:49  <trevnorris>but since Dispose() free's the memory, it instead sends NULL
17:30:40  <tjfontaine>right this conversation was all in context of dispose, ok yes, that needs a fix asap :)
17:31:07  <trevnorris>you mean in case of Buffer.dispose()?
17:31:24  <tjfontaine>I mean you should *never* free() if someone passes a cb, that's breaking the contract
17:31:24  <trevnorris>you can't pass a callback to Buffer.alloc(), so from js land, there's no bug.
17:31:30  <tjfontaine>right
17:31:38  <tjfontaine>it only makes sense from cc, but
17:31:42  <tjfontaine>you let people call dispose from js
17:31:46  <trevnorris>yeah. there's a TODO in the code about this.
17:32:30  <trevnorris>there's a bigger issue that I can't figure out a nice way to handle.
17:32:40  <trevnorris>TooTallNate: maybe you know a way to do this
17:32:51  <trevnorris>I can't check if an Object is an instance of Buffer in cc
17:33:01  <trevnorris>so instead all it checks is if there's external memory set
17:34:55  <trevnorris>the only way I can figure is to call to js and have it check instanceof there. I haven't found an instanceof type check in the cc api yet.
17:38:39  <tjfontaine>it's more or less caching the constructor and checking the source objects constructor and seeing if they match, right?
17:43:50  * st_lukejoined
17:44:02  <trevnorris>i'll get a fix in by the end of the day.
17:44:27  <trevnorris>at the least I'll just make smalloc::Dispose not-public.
17:48:53  * loladiroquit (Quit: loladiro)
17:51:27  * st_lukequit (Remote host closed the connection)
17:55:55  * st_lukejoined
17:57:53  * st_lukequit (Remote host closed the connection)
18:02:40  * loladirojoined
18:08:49  * c4milojoined
18:10:56  * defunctzombie_zzchanged nick to defunctzombie
18:14:13  <piscisaureus_>https://groups.google.com/d/msg/nodejs/1AFAmL14TpA/M4KNWNh5Sy8J
18:14:40  * piscisaureus_cries
18:14:53  <piscisaureus_>someone on the internet is doing it wrong
18:15:05  <tjfontaine>I'm so happy this is now on the mailing list and not on the issue tracker
18:18:43  * wavdedjoined
18:22:19  <TooTallNate>http://imgs.xkcd.com/comics/duty_calls.png
18:22:58  <tjfontaine>:P
18:23:30  <tjfontaine>piscisaureus_: btw, do you have a blog/tutorial/example for consuming ETW trace points?
18:23:46  <piscisaureus_>Not really. Doesn't sblom?
18:24:04  <piscisaureus_>tjfontaine: thttp://tomasz.janczuk.org/2011/09/using-event-tracing-for-windows-to.html
18:24:12  <piscisaureus_>er
18:24:13  <piscisaureus_>http://tomasz.janczuk.org/2011/09/using-event-tracing-for-windows-to.html
18:24:19  <tjfontaine>he's not on here atm, you are, just finishing writing a chapter on Tracing for our docs
18:24:38  <tjfontaine>piscisaureus_: thanks
18:25:45  <piscisaureus_>tjfontaine: also https://github.com/tjanczuk/iisnode/blob/master/src/scripts/etw.bat
18:26:40  <piscisaureus_>just replace iisnode with node
18:26:50  <tjfontaine>piscisaureus_: excellent, hopefully I'll be able to cull that into a small blurb that can fit in these docs
18:26:59  <piscisaureus_>cool
18:27:02  <tjfontaine>piscisaureus_: the dtrace and systemtap stuff are roughly oneliners :)
18:27:12  <piscisaureus_>tjfontaine: so for node it is too
18:27:22  * LeftWingjoined
18:27:22  <tjfontaine>*etw
18:30:15  <piscisaureus_>something like "
18:30:16  <piscisaureus_>logman start node -ow -p {1040DFC4-61DB-484A-9530-584B2735F7F7} -ets -o output.etl
18:30:16  <piscisaureus_>--- do stuff with node ---
18:30:16  <piscisaureus_>logman stop node -ets
18:30:16  <piscisaureus_>tracerpt output.etl -o output.xml
18:30:17  <piscisaureus_>"
18:30:19  <piscisaureus_>^-- tjfontaine
18:30:28  <piscisaureus_>let me actually try that
18:30:37  <tjfontaine>:)
18:34:06  <bnoordhuis>TooTallNate: deps/npm/node_modules/node-gyp/gyp/test/mac/postbuild-fail/postbuild-fail.sh:1:#!/usr/bin/bash <- /usr/bin/bash?
18:34:33  <TooTallNate>bnoordhuis: well that's weird
18:34:33  <TooTallNate>haha
18:34:38  <TooTallNate>bnoordhuis: it's from gyp though
18:34:50  <bnoordhuis>oh, okay
18:35:07  <TooTallNate>bnoordhuis: and is that an old node-gyp? I don't think i publish their test/ dir anymore
18:35:18  <bnoordhuis>it's from v0.10
18:35:31  <bnoordhuis>no wait, it's in master as well
18:35:32  <TooTallNate>ahh
18:35:44  <TooTallNate>i'm not sure when the last time isaacs' updated npm's copy
18:38:04  <bnoordhuis>right
18:38:16  <bnoordhuis>`make install` installs a bunc of test/ crap that's in deps/npm
18:38:27  <bnoordhuis>i guess we ought to fix that
18:38:45  * dominictarrjoined
18:41:57  <tjfontaine>trevnorris: heh, I guess you deleted the previous comment and then recommented?
18:42:24  <trevnorris>tjfontaine: yeah. was doing that while feeding my kid. didn't come across right.
18:42:38  <tjfontaine>hehe
18:42:57  <trevnorris>mozilla closed the offices today so trying to get some work done from home.
18:43:08  <trevnorris>problem is my home doesn't actually have a place to work.
18:43:12  <tjfontaine>hehe I've WFH all week because of BART
18:43:39  <trevnorris>hah.
18:44:04  <tjfontaine>do you actually want to have a conversation on breadth/depth or are you ok with that not ever being the case?
18:44:40  <trevnorris>I personally don't think it can be done. after working through the nextTick logic for 2 weeks I'm convinced there's no safe way to do it.
18:44:48  <trevnorris>but i'm wrong quite often. :P
18:45:18  <tjfontaine>no, that's pretty much what I was going to say, the order of execution is what defines the insertion in the list :)
18:45:58  <trevnorris>yeha
18:46:02  <trevnorris>yeah
18:47:54  <trevnorris>and it gets really tricky once users start to schedule multiple bredth in multiple depth.
18:48:22  <trevnorris>there's no easily definable way to determine when to break execution.
18:49:30  <piscisaureus_>tjfontaine: https://gist.github.com/piscisaureus/5936473
18:50:38  <tjfontaine>that uuid is our installer id, i.e. is that portable?
18:50:58  <piscisaureus_>It's the ID of our event set
18:51:46  <tjfontaine>ok, does that change when we regenerate?
18:51:47  <piscisaureus_>it's defined here: https://github.com/joyent/node/commits/66f64ae072271ef36e9f684a965f1fc410d86ebc/src/node_win32_etw_provider-inl.h
18:51:51  <piscisaureus_>not afaict
18:51:53  <tjfontaine>k
18:52:28  <piscisaureus_>stupid
18:52:32  <tjfontaine>ya, it's in the manifest
18:52:32  <piscisaureus_>here of course: https://github.com/joyent/node/commits/66f64ae072271ef36e9f684a965f1fc410d86ebc/src/res/node_etw_provider.man
18:53:25  <tjfontaine>ok I'm going to include that example mostly verbatim in the docs
18:59:35  <piscisaureus_>tjfontaine: maybe also link to http://blogs.msdn.com/b/interoperability/archive/2012/12/03/ms-open-tech-contributes-support-for-windows-etw-and-perf-counters-to-node-js.aspx :)
19:00:39  <tjfontaine>heh, I might do a separate blog post that talks about doing the same trace operation on each platform, not sure we want to link to much else besides general documentation for each supported tracing platform
19:01:20  <piscisaureus_>I don't see the problem
19:01:23  <piscisaureus_>but ok
19:01:27  <piscisaureus_>Someone should do a
19:01:34  <tjfontaine>we dont' have many external links in our docs at the moment
19:01:52  <trevnorris>well this is annoying. Handle<Object>->Get(uint32_t) will get the data if it's been assigned external array data.
19:01:59  <piscisaureus_>run-node-and-make-flamegraph-on-whatever-platform module
19:02:47  <tjfontaine>ya, though getting the js stack on not-dtrace is still a bit of a problem
19:03:03  <piscisaureus_>etw can do it just fine
19:03:08  <piscisaureus_>but nobody knows how
19:03:11  <tjfontaine>heh
19:03:22  <tjfontaine>ok, I didn't think it could from my conversation with sblom
19:03:37  <tjfontaine>but looking at that graph it certainly appears to
19:04:32  * arlolraquit (Quit: Leaving...)
19:05:06  <piscisaureus_>the prolem is that it takes 20 steps and installing 1 SDK and 2 NPM command line modules
19:05:33  <tjfontaine>ya
19:06:19  <tjfontaine>I spent some time working on an httptop tool yesterday, which at the moment just is a driver for dtrace, 1/2 of the screen shows your popular routes, and as you select different routes shows you the distribution of latencies
19:06:35  <tjfontaine>https://dsz91cxz97a03.cloudfront.net/GfVRIJfwfc-1200x1200.png
19:13:34  * Guest97044changed nick to isaacs
19:17:28  * M28_joined
19:17:54  * M28quit (Read error: Connection reset by peer)
19:18:05  * M28_quit (Client Quit)
19:18:14  * M28joined
19:21:35  * mikealquit (Quit: Leaving.)
19:24:09  * mikealjoined
19:28:48  <trevnorris>bnoordhuis: so ->Has(uint32_t) is a lot faster than ->Has(Handle<Value>), but the former also retrieves the indexes from an external data array.
19:29:04  <tjfontaine>bnoordhuis: I've already agreed in principle with your idea in #5798, just a matter of getting consensus on the need for a change
19:29:15  <trevnorris>bnoordhuis: so i'm wondering if it's safe enough to set a value via ->Has(-1), since it will flow past the max size of 0x3fffffff?
19:32:09  * pooya_joined
19:32:25  * c4miloquit (Remote host closed the connection)
19:34:05  * pooyaquit (Ping timeout: 268 seconds)
19:34:05  * pooya_changed nick to pooya
19:40:33  <bnoordhuis>trevnorris: you mean external as in External<T>?
19:41:19  <trevnorris>bnoordhuis: I was just looking at this api: Set (uint32_t index, Handle< Value > value)
19:41:22  * dominictarrquit (Quit: dominictarr)
19:42:36  <bnoordhuis>trevnorris: sorry, i don't think i'm following. what is the question?
19:44:04  <trevnorris>bnoordhuis: wait. it seems that <Object> allows to set values like <Array>. bummer. can't have it exposed to the user. nm. :)
19:47:11  * M28quit (Read error: Connection reset by peer)
19:48:28  * M28joined
19:50:59  <M28>trevnorris what are you trying to do?
19:51:01  <M28>hidden fields?
19:51:12  <M28>theres ::SetHiddenField...
19:52:36  <trevnorris>M28: yeah, except that operation is so freakishly slow. not using operations like that was a large part of the Buffer rewrite
19:53:10  <M28>->Set("_private_plsdontmesswithme_pls_plz_pls", ...) it is then
19:53:10  <M28>:p
19:53:48  <trevnorris>M28: any ->Set* is super freakin slow
19:54:08  <M28>probably
19:54:26  <M28>can't you just have a struct pointer in an internal field?
19:54:36  <trevnorris>slower
19:54:37  <M28>like, you'll only be getting/setting it every once in a while
19:54:43  <M28>...
19:54:44  <trevnorris>and I'd have to create the object in cc instead of js
19:54:49  <trevnorris>which is slower as well
19:54:52  <M28>wat
19:54:55  <trevnorris>yeah
19:54:57  <M28>...
19:55:07  <trevnorris>https://github.com/trevnorris/talks
19:55:27  <trevnorris>only folder there right now.
19:55:33  <trevnorris>check the commit log
19:56:10  <M28>what are you using those fields for?
19:56:39  <trevnorris>in the repo, or in my core changes i'm working on?
19:56:43  <M28>also, shouldn't you be using String::NewSymbol for the name of each property?
19:56:52  <M28>(instead of newfromutf8)
19:57:05  <M28>in the core changes
19:58:38  * dominictarrjoined
20:06:21  <bnoordhuis>String::NewSymbol is only useful if you're creating the same string over and over
20:06:30  <bnoordhuis>in which case you should probably stuff it in a Persistent
20:06:54  <bnoordhuis>we're abusing it an awful lot in node core, people thought it was a good idea at the time :-/
20:07:28  <tjfontaine>who could have envisioned that Persistent would be going away :P
20:07:35  <tjfontaine>well
20:07:52  * pooyaquit (Quit: pooya)
20:08:02  <tjfontaine>that v8 would change internals as wide as they are
20:09:08  <M28>PERSISTENT IS GOING AWAY?!
20:09:08  <LOUDBOT_>I AM THE EBAY KILLER I SELL PEOPLE THINGS ON EBAY SO THAT I CAN GET THEIR HOME ADDRESS TO GO FUCK THEM UP
20:09:14  <M28>WHAT DO I DO NOW
20:09:15  <LOUDBOT_>AMERICAN GIRLS BY COUNTING CROWS IS A SONG ABOUT AN STD THAT TURNS MEN INTO PREOP TRANSEXUALS BY FORCE AND MAKES THEM GO OUT IN DRAG ON FRIDAY NIGHTS
20:09:24  <M28>lol
20:09:39  <bnoordhuis>M28: not quite going away but it's changed
20:09:47  <M28>what has changed?
20:09:57  <bnoordhuis>it no longer inherits from Handle, you can no longer deref the Handle it points to, etc.
20:10:23  <bnoordhuis>if you want to get the handle, you have to recreate it with Local<T>::New(isolate, persistent)
20:10:26  <M28>was that the thing that it needs something like UNSAFE_HANDLES enabled?
20:10:52  <bnoordhuis>yep. or rather V8_USE_UNSAFE_HANDLES gives you the old api
20:11:07  <bnoordhuis>i've been working on upgrading node to the new api for the last three days
20:11:13  <bnoordhuis>massive, massive pain
20:11:20  <M28>yeah it broke a lot of my stuff
20:11:24  <M28>I'm using unsafe handles atm
20:11:49  <tjfontaine>piscisaureus_: in your etw sample, can I limit that to a specific trace point?
20:12:00  <tjfontaine>piscisaureus_: or does that filtering always happen in the post processing for tracerpt
20:13:03  <M28>is those api changes documented anywhere? I couldn't find it when I was mad that I updated v8 and had to add that #define
20:13:03  <tjfontaine>hmm maybe that's what [-n] name Name of the target object. means
20:13:05  <M28>are*
20:13:22  <tjfontaine>they'are least in the codereview's they've done
20:14:11  <M28>trevnorris, you work at mozilla? :o
20:14:28  <M28>do you know dvander?
20:15:08  <tjfontaine>ah I can be more explicit in -p
20:16:17  * st_lukejoined
20:16:23  <M28>https://github.com/trevnorris/cbuffer/blob/master/cbuffer.js#L42
20:16:44  <M28>also, you should probably delete the item or assign it to null, in case it's a circular buffer of objects, so it can be garbage collected
20:31:52  * pooyajoined
20:31:59  <M28>I submitted a pull request
20:35:30  <bnoordhuis>yay, i have #5800
20:35:34  <tjfontaine>yay, thanks!
20:36:46  <tjfontaine>etw generates a *lot* of data by default
20:45:39  * kazuponquit (Remote host closed the connection)
20:46:06  * kazuponjoined
20:51:10  * kazuponquit (Ping timeout: 268 seconds)
20:56:00  * pachetjoined
20:56:00  * pachetquit (Changing host)
20:56:00  * pachetjoined
21:05:05  <wavded>is there a benefit to being able to close fd you didn't open, e.g. i can mess up a node prog quick with fs.closeSync(1)?
21:06:56  * mikealquit (Quit: Leaving.)
21:07:19  * mikealjoined
21:09:21  * mikealquit (Client Quit)
21:10:12  <wavded>was thinking also that an attacker could just throw a bunch of crap data just by trying a bunch of numbers without having to access any object reference (fs.writeSync(i,'asdasd'))
21:11:10  * st_lukequit (Remote host closed the connection)
21:16:03  <piscisaureus_>tjfontaine: dunno about filtering. Probably post-processing only
21:16:18  <bnoordhuis>wavded: a process can inherit file descriptors, c.f. systemd
21:16:27  <piscisaureus_>tjfontaine: also instead of the GUID one could specify NodeJS-ETW-Provider but I'd rather rename that to just "node" :)
21:16:30  * kazuponjoined
21:16:38  * defunctzombiechanged nick to defunctzombie_zz
21:17:56  <tjfontaine>piscisaureus_: already did that one :)
21:18:05  <piscisaureus_>tjfontaine: you renamed it?
21:18:11  <tjfontaine>no, used the !guid
21:18:32  <piscisaureus_>Let's rename it later
21:18:35  <tjfontaine>I also changed it to generate csv, because all that xml pains me
21:18:39  <tjfontaine>yes we can change that in master
21:18:46  <piscisaureus_>ha! you did research
21:18:57  <piscisaureus_>tjfontaine: now document performance counters >:)
21:18:58  * defunctzombie_zzchanged nick to defunctzombie
21:19:03  <tjfontaine>I should know how to do this shit :)
21:19:16  <tjfontaine>piscisaureus_: heh, ya I'm not sure how to address perfctrs just yet
21:19:24  <tjfontaine>probably be an addendum
21:20:41  <tjfontaine>isaacs: before I get too far into this do you know why ### subheading fails in the first ## subheading, but works in the second ## subheading? https://gist.github.com/tjfontaine/5937328
21:22:05  * mikealjoined
21:23:02  <wavded>bnoordhuis: agh ok, i'll look into that more, was just surprised, also played with closing a req using fs.write and fs.close instead of req.write and end
21:24:31  <tjfontaine>oh this seems to be because I named it Examples
21:24:55  <tjfontaine>haha, that's hilarious
21:25:31  * kazuponquit (Ping timeout: 276 seconds)
21:40:49  * wavdedquit (Quit: Hasta la pasta)
21:41:53  <bnoordhuis>piscisaureus_: btw, was that DWORD thing the only comment you had on the errno work?
21:42:10  <bnoordhuis>and is it actually something that needs addressing?
21:44:05  <trevnorris>M28: heh, didn't realize people were actually using that.
21:44:15  <M28>I was just taking a look at it :p
21:48:07  * rendarquit
21:48:45  <piscisaureus_>bnoordhuis: yes
21:49:12  <piscisaureus_>bnoordhuis: well that the internal error routing is confusing now. I don't really mind how you fix it.
21:49:28  <piscisaureus_>(e.g. you could also *immediately* translate vs translate on the edges)
21:53:49  <bnoordhuis>piscisaureus_: that's harder to deal with really. there's quite a few places where you need the native error code a few stack frames down
21:53:59  * st_lukejoined
21:54:13  <piscisaureus_>bnoordhuis: do whatever you think is easiest
21:54:26  <bnoordhuis>immediately translating was my first approach actually. didn't work out :)
21:54:34  <piscisaureus_>bnoordhuis: but ... interesting, I can't imagine where we'd need it actually
21:56:04  <bnoordhuis>piscisaureus_: when you say "fix", do you mean that there is something to actually fix?
21:56:18  <bnoordhuis>if yes, can you point me to a few places where you think it can be improved?
21:58:07  <tjfontaine>sigh, systemtap why do you hate me so
21:58:46  <tjfontaine>https://gist.github.com/tjfontaine/5937511
21:58:56  <tjfontaine>what's an alternate way to split that over multiple lines?
21:59:58  <bnoordhuis>tjfontaine: no backslashes, just newlines
22:00:11  <tjfontaine>mk suppose that's fine because of '
22:00:19  <bnoordhuis>exactly
22:00:54  <tjfontaine>I just have a pattern man, why does it have to cramp my style :P
22:01:00  <bnoordhuis>haha
22:01:45  <tjfontaine>also the process("") is an interesting pattern, I always forget that it needs to match the pathing
22:06:49  <tjfontaine>https://github.com/tjfontaine/node/compare/trace-docs where it stands at the moment, I still need to discuss perf counters I guess
22:11:46  <TooTallNate>anyone know what this means?
22:11:47  <TooTallNate>Error: 140735310340880:error:140770FC:SSL routines:SSL23_GET_SERVER_HELLO:unknown protocol:../deps/openssl/openssl/ssl/s23_clnt.c:766:
22:12:02  <tjfontaine>something didn't do the handshake properly?
22:12:24  <TooTallNate>tjfontaine: is it a server problem or a client problem?
22:12:32  <TooTallNate>tjfontaine: (this script is using node as the http client FYI)
22:12:52  <tjfontaine>so you're seeing this as a client?
22:13:01  <bnoordhuis>TooTallNate: the other end probably doesn't speak ssl/tls
22:13:05  <tjfontaine>indeed
22:13:10  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
22:13:14  <tjfontaine>or a version advertisement mismatch?
22:13:25  <TooTallNate>bnoordhuis: is there an openssl command to run to test the server?
22:13:58  <tjfontaine>openssl s_client -connect host:443
22:14:05  <TooTallNate>tjfontaine: bnoordhuis https://cloudup.com/c2hnxOvK2ks
22:14:19  <tjfontaine>if that passes probably need to tweak supported methods
22:14:24  <MI6>joyent/node: trevnorris created branch maxtickdepth-doc-remove - http://git.io/44Mqag
22:14:25  <trevnorris>review anyone ^
22:14:45  <trevnorris>can't believe that hung around so long :-/
22:15:43  <bnoordhuis>trevnorris: lgtm
22:15:47  <trevnorris>thanks
22:16:21  <tjfontaine>TooTallNate: looks like it's relatively happy with tlsv1, there shouldn't be a problem with the node client then I wouldn't think
22:16:28  <MI6>joyent/node: Trevor Norris master * 71ade1c : doc: remove mention of maxTickDepth - http://git.io/LW_p_Q
22:21:39  * kazuponjoined
22:25:12  <trevnorris>bnoordhuis: ok. so Persistent<>::New is gone. So we're left with Persistenet<> val(Isolate*,...) ?
22:26:33  * kazuponquit (Ping timeout: 264 seconds)
22:28:34  <trevnorris>oooh. wait. I see what you were talking about earlier.
22:30:10  <trevnorris>using "Reset" before the value has ever been set seemed strange. oh well.
22:30:40  <bnoordhuis>trevnorris: the persistent starts out empty now until you reset it
22:31:03  <bnoordhuis>upside is you no longer have to call Dispose() before reassigning
22:31:14  <trevnorris>ah, nice.
22:31:31  <bnoordhuis>you don't have to call Clear() anymore either when destroying the persistent
22:32:12  <trevnorris>that's nice.
22:32:27  <MI6>nodejs-master: #285 UNSTABLE linux-ia32 (2/610) smartos-ia32 (5/610) osx-x64 (2/610) osx-ia32 (2/610) linux-x64 (4/610) smartos-x64 (9/610) http://jenkins.nodejs.org/job/nodejs-master/285/
22:33:15  <TooTallNate>tjfontaine: well a standalone script connecting to the ssl server seems to work fine :\
22:33:29  <TooTallNate>tjfontaine: so that leads me to believe it's a problem with mikeal/request :\
22:33:56  <tjfontaine>TooTallNate: you could be mutating your advertised protocols some way
22:37:00  <TooTallNate>tjfontaine: seems like mikeal/request doesn't do SSL proxying correctly
22:40:21  <trevnorris>bnoordhuis: this may sounds stupid, but couldn't a dev randomly assign pointers to the return value of ClearAndLeak?
22:42:54  * Domenic_quit (Remote host closed the connection)
22:43:00  * Raynosquit (Write error: Broken pipe)
22:43:01  * dscape__quit (Remote host closed the connection)
22:43:01  * `3rdEdenquit (Write error: Broken pipe)
22:44:38  <bnoordhuis>trevnorris: you cannot do anything useful with the return value of ClearAndLeak()
22:44:49  <bnoordhuis>it returns T*, i.e. a raw value pointer
22:45:09  <bnoordhuis>and you can't turn that T* pointer into a handle again :-/
22:45:37  <trevnorris>oh, and I assume v8 will eventually reusee that pointer value
22:45:59  <trevnorris>strange. wonder why they even bothered to return it then.
22:49:52  <bnoordhuis>yeah, me too. maybe work in progress, maybe an oversight
22:51:26  <trevnorris>bnoordhuis: nice. actually compiles and most the tests are running.
22:52:57  * groundwaterquit (Quit: groundwater)
22:55:32  <MI6>nodejs-master-windows: #93 UNSTABLE windows-x64 (22/610) windows-ia32 (23/610) http://jenkins.nodejs.org/job/nodejs-master-windows/93/
22:55:53  <tjfontaine>oof
22:56:03  <tjfontaine>that was just a documentation change jenkins
22:56:33  <tjfontaine>sigh I have some of those done, I just need to wrap it up
23:02:43  * arlolrajoined
23:04:30  <trevnorris>so it's safe to pass Persistent<> by reference, correct?
23:05:02  <bnoordhuis>trevnorris: yes
23:05:10  <trevnorris>cool. makes this much easier.
23:05:41  <bnoordhuis>yes. i kind of agree with the change, it's just a pain to update all the code
23:14:25  * arlolraquit (Quit: Linkinus - http://linkinus.com)
23:14:36  <trevnorris>only part I'm really a fan of is args.GetIsolate(). but yeah. really a pain.
23:15:02  <trevnorris>feel bad for all the module devs that'll be upgrading this.
23:15:41  * pachetquit (Quit: leaving)
23:16:42  <trevnorris>tjfontaine: fyi, simple/test-http-exit-delay.js is failing on my box. haven't investigated why, but wanted to check if it's a known issue.
23:22:27  * kazuponjoined
23:27:01  * kazuponquit (Ping timeout: 248 seconds)
23:27:32  <trevnorris>tjfontaine, TooTallNate: here you go: https://github.com/joyent/node/pull/5802
23:32:22  <rvagg>node debug build on smartos, master & v0.11.3 branch, I can't get past this:
23:32:23  <rvagg> ACTION _home_rvagg_git_node_node_gyp_node_dtrace_ustack_target_node_dtrace_ustack /home/rvagg/git/node/out/Debug/obj.target/node/src/node_dtrace_ustack.o
23:32:24  <rvagg>dtrace: failed to compile script src/v8ustack.d: line 224: failed to resolve V8DBG_OFF_FP_CONTEXT: Unknown variable name
23:33:02  <TooTallNate>trevnorris: lgtm i think
23:33:19  <trevnorris>TooTallNate: cool. have one more small commit to throw on
23:33:33  <rvagg>any ideas? I'm assuming that there are people here that are doing debug builds on smartos
23:34:41  <trevnorris>TooTallNate: ok, second commit on: https://github.com/joyent/node/pull/5802
23:36:02  <TooTallNate>trevnorris: cool, good catch
23:38:08  <trevnorris>TooTallNate: thanks. so you're good w/ it?
23:39:08  <trevnorris>oh, and added advantage, you'll now be able to get access to the Persistent<>
23:41:27  <TooTallNate>trevnorris: oh nice!
23:41:37  <TooTallNate>trevnorris: and ya it lgtm, though i haven't pulled the code
23:42:40  <trevnorris>bnoordhuis: awake enough for a quick review of a bug fix :) https://github.com/joyent/node/pull/5802
23:45:22  <tjfontaine>trevnorris: on master you mean, i saw that when ben merged
23:45:50  <tjfontaine>rvagg: which platform image are you on?
23:46:08  * stagasjoined
23:46:12  <trevnorris>ok
23:46:42  <rvagg>no idea
23:46:48  <rvagg>base something
23:46:53  <tjfontaine>uname -a please
23:46:56  <rvagg>vmadm list isn't helpful...
23:47:01  <tjfontaine>trevnorris: in lib/timers.js change the Date.now to Timer.now
23:47:05  <tjfontaine>trevnorris: lemme know if that fixes it
23:47:11  <rvagg>right, SunOS 6f2d4797-fb39-45aa-b7e4-2d5703a923e0 5.11 joyent_20130529T211010Z i86pc i386 i86pc Solaris
23:47:15  <trevnorris>tjfontaine: will do.
23:47:23  <bnoordhuis>trevnorris: do you hate me?
23:47:26  <tjfontaine>rvagg: oh this is your smartos-live?
23:47:51  <rvagg>tjfontaine: yeah, running on vmware
23:48:01  <trevnorris>bnoordhuis: heh, no. didn't actually expect you to get to this tonight. :)
23:48:08  <tjfontaine>rvagg: ok thanks
23:48:34  <bnoordhuis>trevnorris: well, i was thinking of having to rebase my upgrade-v8 after that
23:48:42  <tjfontaine>trevnorris: ben's probably asking because you're touching stuff he is :P
23:48:51  <trevnorris>lol, hey I had to do that for over a month w/ my buffer changes. :P
23:49:05  <tjfontaine>comcast is hating me today
23:49:28  <indutny>hoya
23:49:43  <indutny>just took a look at smalloc
23:49:46  <indutny>trevnorris: good job
23:49:51  <indutny>btw, wanna review this https://github.com/joyent/node/pull/5803 ?
23:50:02  <indutny>bnoordhuis: may be you too? :)
23:50:10  <trevnorris>indutny: thanks dude. I try to pay attention. :)
23:52:18  <trevnorris>indutny: lgtm, also SetupBufferJS in node_buffer.cc could use those SET_METHOD macros.
23:52:29  <trevnorris>a lot of unnecessary I left in there.
23:53:31  <trevnorris>tjfontaine: didn't work.
23:54:35  <tjfontaine>ok
23:55:46  <indutny>good tip
23:55:53  <indutny>I'll add commit to that PR