00:05:27  * plutoniixjoined
00:14:41  * bobmcw_joined
00:15:56  * bobmcwquit (Ping timeout: 240 seconds)
00:38:05  * caitp-joined
00:42:56  * caitp-quit (Ping timeout: 240 seconds)
01:08:30  * caitpjoined
01:11:40  * ofrobotsquit (Quit: My Mac has gone to sleep. ZZZzzz…)
01:22:24  * ofrobotsjoined
01:23:08  * ofrobotsquit (Client Quit)
01:39:23  * caitp-joined
01:43:34  * ofrobotsjoined
01:48:07  * ofrobotsquit (Ping timeout: 255 seconds)
01:57:28  * ofrobotsjoined
02:05:46  * ofrobotsquit (Quit: My Mac has gone to sleep. ZZZzzz…)
02:13:39  * caitp-quit (Ping timeout: 276 seconds)
02:51:22  * ofrobotsjoined
03:10:10  * caitp-joined
03:15:11  * caitp-quit (Ping timeout: 265 seconds)
03:51:07  * ofrobotsquit (Quit: My Mac has gone to sleep. ZZZzzz…)
04:10:53  * caitp-joined
04:16:08  * caitp-quit (Ping timeout: 272 seconds)
04:38:30  * ofrobotsjoined
05:11:27  * ofrobotsquit (Quit: My Mac has gone to sleep. ZZZzzz…)
05:12:06  * caitp-joined
05:18:09  * caitp-quit (Ping timeout: 250 seconds)
05:56:49  * ofrobotsjoined
06:02:49  * caitpquit (Ping timeout: 244 seconds)
06:14:50  * caitp-joined
06:19:12  * ofrobotsquit (Quit: My Mac has gone to sleep. ZZZzzz…)
06:19:28  * caitp-quit (Ping timeout: 255 seconds)
06:37:25  * enaqxquit (Remote host closed the connection)
06:44:42  * esasquit (Ping timeout: 276 seconds)
06:45:05  * mostynbjoined
06:45:53  * muellijoined
06:59:16  * caitpjoined
07:00:37  * enaqxjoined
07:03:47  * caitpquit (Ping timeout: 244 seconds)
07:15:36  * caitp-joined
07:21:42  * caitp-quit (Ping timeout: 272 seconds)
07:23:34  * plutoniixquit (Quit: จรลี จรลา)
08:15:37  * BobGneujoined
08:16:11  * Bob_Gneuquit (Ping timeout: 256 seconds)
08:17:46  * caitp-joined
08:23:46  * caitp-quit (Ping timeout: 272 seconds)
08:31:45  * caitpjoined
08:36:16  * caitpquit (Ping timeout: 244 seconds)
08:40:52  * caitpjoined
08:45:17  * rendarjoined
08:51:15  * caitpquit (Ping timeout: 244 seconds)
09:20:01  * caitp-joined
09:24:56  * caitp-quit (Ping timeout: 265 seconds)
09:37:43  * caitpjoined
09:42:55  * caitpquit (Ping timeout: 244 seconds)
09:46:30  * Bob_Gneujoined
09:46:32  * BobGneuquit (Ping timeout: 244 seconds)
10:04:31  <trungl-bot>Tree closed by buildbot@chromium.org: Tree is closed (Automatic: "Check" on http://build.chromium.org/p/client.v8/builders/V8%20Win64%20-%20debug/builds/3119 "V8 Win64 - debug" from 1630253f937f200bd67c05d39d503ba1144440cb: jochen@chromium.org (:jochen__))
10:14:34  <trungl-bot>Tree closed by machenbach@google.com: Tree is closed (maintenance)
10:21:08  * caitp-joined
10:24:36  <trungl-bot>Tree opened by machenbach@google.com: Tree is open
10:26:49  * caitp-quit (Ping timeout: 264 seconds)
10:32:42  * caitpjoined
10:32:52  * plutoniixjoined
10:34:14  * jjrvquit (Ping timeout: 272 seconds)
10:34:21  * jjrvjoined
10:39:45  * caitpquit (Ping timeout: 244 seconds)
11:22:26  * caitp-joined
11:27:34  * caitp-quit (Ping timeout: 264 seconds)
12:18:35  * StephenLynxjoined
12:23:23  * caitp-joined
12:23:35  <trungl-bot>Tree closed by machenbach@google.com: Tree is closed (msan compile)
12:28:28  * caitp-quit (Ping timeout: 255 seconds)
12:33:40  <trungl-bot>Tree opened by machenbach@google.com: Tree is open
12:41:18  * esasjoined
13:02:52  * caitpjoined
13:11:00  * bobmcw_changed nick to bobmcw
13:11:00  * bobmcwquit (Changing host)
13:11:01  * bobmcwjoined
13:24:04  <trungl-bot>Tree closed by machenbach@google.com: Tree is closed (gnumbd outage)
13:24:39  * caitp-joined
13:29:23  * caitp-quit (Ping timeout: 264 seconds)
13:31:07  * WakiMikojoined
13:34:24  * srl295_quit (Ping timeout: 245 seconds)
13:35:48  * srl295joined
13:43:13  <trungl-bot>Tree opened by machenbach@google.com: Tree is open
13:43:56  * Gettyquit (Ping timeout: 240 seconds)
13:44:08  * Gettyjoined
13:54:10  * rendar_joined
13:56:54  * rendarquit (Ping timeout: 272 seconds)
14:09:51  * ofrobotsjoined
14:20:09  * mostynbquit (Quit: Leaving)
14:22:05  * ofrobotsquit (Quit: My Mac has gone to sleep. ZZZzzz…)
14:25:32  * caitp-joined
14:27:14  * C-Manjoined
14:31:19  * caitp-quit (Ping timeout: 255 seconds)
14:49:33  * bradleymeckjoined
15:09:12  * Bob_Gneuquit
15:19:50  * ofrobotsjoined
15:27:52  * caitp-joined
15:30:53  * muelliquit (Ping timeout: 250 seconds)
15:34:11  * caitp-quit (Ping timeout: 264 seconds)
15:43:42  * benglquit (Ping timeout: 256 seconds)
15:44:35  * bengljoined
15:44:52  * benglchanged nick to Guest14060
15:54:03  * RT|Chatzillaquit (Quit: ChatZilla 0.9.86.1 [Firefox 2.0.0.22pre/2009081014])
16:11:23  * muellijoined
16:13:25  * bradleymeckquit (Quit: bradleymeck)
16:31:11  * caitp-joined
16:36:05  * caitp-quit (Ping timeout: 248 seconds)
17:07:35  * bradleymeckjoined
17:15:51  * bradleymeck_joined
17:16:02  * bradleymeckquit (Ping timeout: 244 seconds)
17:16:02  * bradleymeck_changed nick to bradleymeck
17:31:01  * muelliquit (Ping timeout: 264 seconds)
17:32:34  * caitp-joined
17:37:40  * caitp-quit (Ping timeout: 250 seconds)
18:33:45  * caitp-joined
18:40:15  * caitp-quit (Ping timeout: 244 seconds)
18:41:34  * bradleymeckquit (Quit: bradleymeck)
18:44:32  * enaqxquit (Remote host closed the connection)
18:53:25  * xiinotulpjoined
18:56:36  * plutoniixquit (Ping timeout: 240 seconds)
19:35:29  * ncthom91joined
19:36:41  * ofrobotsquit (Quit: My Mac has gone to sleep. ZZZzzz…)
19:37:32  * caitp-joined
19:40:11  * rendar_quit (Ping timeout: 244 seconds)
19:42:59  * caitp-quit (Ping timeout: 256 seconds)
19:46:50  * rendarjoined
19:46:59  <ncthom91>hi all. Would anybody here be able to help me with a Nodejs addon? I'm struggling with the v8<->c++ boundary (because i'm very inexperienced in c++)
19:47:07  <ncthom91>particularly... I'm trying to write a simple node addon which wraps boost::interprocess::message_queue to exchange JS objects between different processes. Putting JS objects into the queue seems to work fine, but taking them back out is getting messy
19:54:22  <caitp>that's probably going to be messy for other reasons
19:59:25  * Guest14060changed nick to bengl
20:07:36  * bradleymeckjoined
20:07:51  * C-Manquit (Quit: Connection reset by beer)
20:09:56  * ofrobotsjoined
20:23:43  <ncthom91>caitp why is that?
20:24:16  <caitp>moving v8 objects across contexts and across processes
20:24:32  <caitp>it's doable, but I mean
20:25:22  <ncthom91>caitp dangerous because of what it means for GC?
20:25:55  <ncthom91>(mind you, super newb here, but I'm curious to learn)
20:28:34  <caitp>i wouldn't say that necessarily, but I feel like moving between processes, you probably need to deserialize it and re-serialize it
20:28:53  <caitp>i'm not sure exactly how the blink bindings accomplish it, but that's one place where it's done
20:29:05  <ncthom91>ah, yea. Luckily, my objects are super simple... really just a set of strings
20:29:24  <ncthom91>right now I'm just copying the string contents into a struct and shoving that into the queue
20:29:27  <caitp>node has that whole clustering thing, but I think they're using just threads so it might be slightly simpler (just lots of extra locking and stuff)
20:31:05  <ncthom91>caitp node's clustering is just a load balancer in front of a couple child processes afaik
20:31:33  <caitp>i'm not sure if they're actually child processes yet, i remember there was discussion about that
20:32:20  <bradleymeck>caitp: node cluster is json serialization w/ some magic for buffers and fds
20:32:38  <bradleymeck>node doesn't support multiple isolates at all currently
20:32:55  <bradleymeck>but several attempts have been made with varying issues, eventually it will get threads
20:35:17  * enaqxjoined
20:35:44  <ncthom91>caitp so, as far as I've read, shared memory is way faster than dgram packets over UDP, which is how I understand node's cluster module to work
20:35:55  <caitp>oh really? so just one isolate and multiple contexts?
20:35:58  <ncthom91>hence my attempt at writing a node addon to wrap boost's message queue
20:37:20  <ncthom91>I am by all means open to correction if that sounds dumb :)
20:37:43  <bradleymeck>caitp: normally just 1 context
20:37:58  <caitp>i think realistically you'd need to deserialize and just share the deserialized code
20:38:28  <caitp>but some more embedder-headed people could have better ideas
20:38:42  <ncthom91>was that directed at me? Really all I need to share via this queue is strings, so I'm not too worried about serialization
20:39:11  * caitp-joined
20:39:31  <bradleymeck>Buffers in node can point to shared memory if you make them via C++, maybe just do that
20:39:41  <bradleymeck>they won't be strings, but they can be shared
20:41:15  <ncthom91>bradleymeck interesting... So, it might be helpful for me to flesh out the details of my idea a little bit: basically I want to do file transformation across processes, so the main process reads a bunhc of files and puts each file contents in this shared queue, then worker processes read from the queue, perform the transformation on the string, and put the result in a results queue back to the main process
20:42:03  <bradleymeck>sure, as long as the transformation takes up the same space that would be easy
20:42:24  <ncthom91>I can't make that guarantee unfortunately
20:43:35  <ncthom91>given that extra bit of context, do you think shared mem buffers is easier, or more appropriate, than boost's mq?
20:43:46  <bradleymeck>well then you will need to pass the address of the shared mem between processes
20:44:13  <bradleymeck>i generally just serialize things, even under decent load it does fine
20:44:45  <ncthom91>that's kind of why I was thinking a queue (or at least an array) might simplify things, and that involves applying a limit to the size of the file I can queue in either direction
20:45:43  * caitp-quit (Ping timeout: 255 seconds)
20:46:44  <ncthom91>plus, if I manage my own shared mem buffers, race conditions come into play - something I think boost's mq takes care of safely :P
20:47:08  <bradleymeck>if you are going to go for a queue go for a persistent message queue is what I would say
20:47:19  <bradleymeck>rabbitmq or something
20:47:32  <ncthom91>why do you suggest that?
20:48:14  <ncthom91>i suppose there's the case where a worker process pops off the shared queue and errors out before queueing the result...
20:49:11  <ncthom91>for one example of why persistence is nice, lol
20:49:15  <bradleymeck>all sorts of reasons, but mostly once you start queueing you take responsibility when you accept the work
20:49:27  <bradleymeck>so if your process dies mid xform it can pick up the work again
20:49:36  <ncthom91>right
20:50:42  <ncthom91>hmm ok. thanks for that.
20:51:40  <bradleymeck>a lot of message queues can even avoid the network layer, but also modern OS will skip the full network layer once a connection is established anywho
20:51:51  <bradleymeck>if to the loopback address*
20:52:43  <ncthom91>interesting..
20:54:34  <ncthom91>I wasn't planning on distributing the work across machines, mostly just looking to gain the parallelization benefit on one machine... I'd be curious to see if the added parallelization going across several machines is worth the network overhead
20:55:17  * StephenLynxquit (Remote host closed the connection)
20:55:29  <ncthom91>an exploration that will have to wait until I've built this project enough to perform such a test hahaha
20:55:57  <ncthom91>thanks bradleymeck & caitp, that was super helpful
21:03:42  <bradleymeck>caitp: is there a fast serializer in v8 for doing things like moving objects w/ internal fields or do you just walk the object tree yourself I guess?
21:05:15  <caitp>i'm not sure what all is exposed to the api, i just hack on the engine
21:05:28  <caitp>i really need to get more involved with the api side
21:35:17  * bradleymeckquit (Quit: bradleymeck)
21:42:38  * caitp-joined
21:42:52  * ncthom91quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
21:44:25  * enaqxquit (Remote host closed the connection)
21:47:27  * caitp-quit (Ping timeout: 250 seconds)
21:47:32  * ofrobotsquit (Quit: My Mac has gone to sleep. ZZZzzz…)
21:53:58  * enaqxjoined
22:02:24  * ofrobotsjoined
22:02:33  * ofrobotsquit (Client Quit)
22:07:53  * enaqxquit (Remote host closed the connection)
22:08:58  * xiinotulpquit (Quit: จรลี จรลา)
22:09:10  * ofrobotsjoined
22:12:01  * rendarquit
22:14:13  * enaqxjoined
22:14:28  * RT|Chatzillajoined
22:28:26  * Renacorjoined
22:28:56  <Renacor>hi
22:29:00  <Renacor>trying to install v8 on centos
22:29:09  <Renacor>running fetch v8 after i got depot tools
22:29:16  <Renacor>and I am getting this error: WARNING: subprocess '"git" "-c" "core.deltaBaseCacheLimit=2g" "clone" "--no-checkout" "--progress" "https://chromium.googlesource.com/v8/v8.git" "/root/tmp/_gclient_v8_9KBPF8"' in /root/tmp failed; will retry after a short nap...
22:29:59  <Renacor>what am I doing wrong?
22:30:56  * ofrobotsquit (Quit: My Mac has gone to sleep. ZZZzzz…)
22:31:27  * ofrobotsjoined
22:32:32  <caitp>maybe a permissions issue? dunno
22:32:51  <caitp>might try running fetch with -v to get a bit more info
22:33:22  <Renacor>i am root
22:33:50  <Renacor>-v is a bad argument
22:33:55  <caitp>really
22:34:21  <Renacor>[root@app02 tmp]# fetch v8 -v
22:34:21  <Renacor>Error: Got bad arguments ['-v']
22:34:22  <Renacor>usage: fetch.py [options] <recipe> [--property=value [--property2=value2 ...]]
22:34:23  <Renacor>]
22:34:27  <caitp>yeah you're right
22:34:40  <caitp>weird, gclient is a bit more helpful
22:34:47  <caitp>try running the command manually?
22:35:37  <Renacor>hmm
22:35:42  <Renacor>running "git" "-c" "core.deltaBaseCacheLimit=2g" "clone" "--no-checkout" "--progress" "https://chromium.googlesource.com/v8/v8.git" "/root/tmp/_gclient_v8_XLv0OH"
22:35:53  <Renacor>freaks out: Uknown option: -c
22:36:37  <caitp>could be your version of git
22:36:52  <Renacor>what version is required?
22:37:03  <Renacor>[root@app02 tmp]# git --version
22:37:04  <Renacor>git version 1.7.1
22:40:33  <caitp>assuming debian, install-build-deps.sh will get git-core if you didn't have it
22:40:55  <caitp>maybe very old version of debian is not supported?
22:41:01  <Renacor>centos
22:41:05  <caitp>oh right
22:41:47  <Renacor>k i think you are right
22:42:03  <Renacor>my friend is on 2.3.2 and he has the -c option
22:43:53  * caitp-joined
22:44:04  <caitp>yeah, the switch to git wasn't that long ago so there was probably never a time where support for older gits was needed
22:44:15  <caitp>mystery solved
22:44:19  <Renacor>word,
22:44:23  <Renacor>thanks lemme try that out!
22:47:01  * ofrobotsquit (Quit: My Mac has gone to sleep. ZZZzzz…)
22:52:20  * ofrobotsjoined
23:17:24  * caitp-quit (Ping timeout: 250 seconds)
23:52:41  * caitpquit (Ping timeout: 250 seconds)