00:11:49  * rendarquit (Quit: Leaving)
00:41:09  * abraxas_joined
00:45:34  * abraxas_quit (Ping timeout: 255 seconds)
00:51:49  * esasjoined
01:03:46  * esasquit (Ping timeout: 264 seconds)
01:10:54  * esasjoined
01:33:23  * seventhquit (Ping timeout: 240 seconds)
01:37:50  * abraxas_joined
02:15:36  * plutoniixquit (Ping timeout: 256 seconds)
03:12:58  * phpnode_joined
03:16:26  * phpnodequit (Ping timeout: 244 seconds)
03:33:26  * esasquit
05:01:58  * caitp-joined
05:03:53  * caitpquit (Ping timeout: 240 seconds)
05:48:49  * caitp-quit (Ping timeout: 258 seconds)
07:20:27  * muellijoined
07:22:15  * mostynbjoined
07:50:42  * muelliquit (Ping timeout: 245 seconds)
07:51:07  * muellijoined
07:58:41  * mostynbquit (Ping timeout: 264 seconds)
08:05:59  <trungl-bot`>Tree closed by [email protected]: Tree is closed (maintenance - no commits please)
08:07:00  <trungl-bot`>Tree opened by [email protected]: Tree is open
08:10:12  * mostynbjoined
08:23:56  * muelliquit (Ping timeout: 256 seconds)
08:45:00  * Lethalmanjoined
08:49:16  * marja___joined
08:54:34  * muellijoined
09:24:33  <trungl-bot`>Tree closed by [email protected]: Tree is closed (Automatic: "Check" on http://build.chromium.org/p/client.v8/builders/V8%20Linux/builds/1405 "V8 Linux" from b8c667729cd3abc02a367ee587e2723e1392def1: [email protected] (:jochen__))
09:38:40  <trungl-bot`>Tree opened by [email protected]: Tree is open ([email protected] is investigating)
09:51:13  * marja___quit (Quit: Leaving)
10:21:57  <trungl-bot`>Tree closed by [email protected]: Tree is closed (Automatic: "bot_update" on http://build.chromium.org/p/client.v8/builders/V8-Blink%20Win/builds/1506 "V8-Blink Win" from 82d0f800a2427f9b45f892e4c7b48038d9280242: [email protected])
10:28:00  <trungl-bot`>Tree opened by [email protected]: Tree is open (Automatic: Hodně štěstí!)
10:32:38  * abraxas_quit (Remote host closed the connection)
10:33:15  * abraxas_joined
10:38:12  * abraxas_quit (Ping timeout: 250 seconds)
10:38:29  * abraxas_joined
10:40:00  * abraxas_quit (Remote host closed the connection)
10:41:32  * mostynbquit (Ping timeout: 245 seconds)
10:54:36  * mostynbjoined
11:05:50  * mostynbquit (Ping timeout: 258 seconds)
11:15:20  * dpinojoined
11:19:37  * dpinoquit (Ping timeout: 252 seconds)
11:23:05  * phpnode_quit (Remote host closed the connection)
11:42:30  * marja___joined
11:45:39  <trungl-bot`>Tree closed by [email protected]: Tree is closed (Automatic: "Check" on http://build.chromium.org/p/client.v8/builders/V8%20Linux%20-%20nosse3/builds/1459 "V8 Linux - nosse3" from 50c4d8826b838596cda892d76ec84eb7a0eab9c1: [email protected])
11:47:40  <trungl-bot`>Tree opened by [email protected]: Tree is open
12:03:15  * rfquit (Ping timeout: 272 seconds)
12:04:10  * bnoordhuisjoined
12:06:45  * rfjoined
12:29:34  * abraxas_joined
12:34:24  * abraxas_quit (Ping timeout: 256 seconds)
13:24:17  * c4milojoined
13:29:37  * bnoordhuisquit (Ping timeout: 240 seconds)
13:37:15  * caitp-joined
13:54:37  * unixpicklejoined
13:59:05  * marja___quit (Remote host closed the connection)
14:02:34  * marja___joined
14:10:05  * c4miloquit (Remote host closed the connection)
14:10:15  * bnoordhuisjoined
14:18:16  * abraxas_joined
14:19:02  * marja___quit (Ping timeout: 258 seconds)
14:20:00  * marja___joined
14:23:17  * abraxas_quit (Ping timeout: 264 seconds)
14:33:30  * c4milojoined
14:59:37  * C-Manjoined
15:59:21  * rendarjoined
15:59:27  <rendar>i can't build v8 n linux, any help?
15:59:40  <caitp->what's the issue?
16:00:45  <rendar>this http://nopaste.info/b4597c47b2.html
16:05:41  <rendar>caitp-, any idea on how to solve that?
16:07:00  * abraxas_joined
16:11:51  * abraxas_quit (Ping timeout: 252 seconds)
16:13:24  <caitp->looks like you're missing deps
16:13:30  <caitp->have you run `gclient sync`?
16:20:47  * c4miloquit (Remote host closed the connection)
16:21:55  <rendar>caitp-, indeed, it seems working after a 'make dependencies' on the v8 main tree, btw i also had another error i solved with 'apt-get install gcc-multilib'
16:22:10  <caitp->sweet
16:22:13  <caitp->now you're an expert :D
16:24:04  <rendar>caitp-, lol, another question. i saw than v8 APIs changed very much since 2.x and 3.x, starting a v8 project today, should i use the newest APIs? i also noticed that in the newest APIs hello world there is a call to a 'libplatform' ...is that include since v8 3.x?
16:26:20  <bnoordhuis>rendar: yes, since 3.27 or 3.28. look at include/v8-platform.h to see what you need to implement
16:26:46  <rendar>bnoordhuis, hey ben...i see thanks
16:26:57  <rendar>bnoordhuis, i was wondering what that libplatform was :-)
16:27:10  <bnoordhuis>rendar: btw, you can also steal it from here: https://github.com/iojs/io.js/blob/master/src/node_v8_platform.cc
16:27:21  <bnoordhuis>and np :)
16:27:26  <rendar>oh, cool! thanks man
16:30:24  * RT|Chatzillaquit (Quit: ChatZilla 0.9.86.1 [Firefox 2.0.0.22pre/2009081014])
16:33:05  <rendar>is v8 supporting amd64?
16:35:47  <bnoordhuis>rendar: yes. also arm64 if that is what you mean
16:36:19  <bnoordhuis>if you're asking how to build for amd64 / x86_64: `make x64.release`
16:36:34  <bnoordhuis>(v8 calls it x64. ia32 is i386.)
16:37:02  * unixpicklequit (Quit: Peace out)
16:38:56  <rendar>bnoordhuis, hmmm
16:39:36  <rendar>bnoordhuis, i ask this because i get this error http://nopaste.info/5ce0df6a03.html -- why it says "skipping incompatible stdc++" ? shoulnd't those be compatible?
16:41:23  <caitp->is this after you installed g++-multilib?
16:41:39  <rendar>caitp-, yep of course
16:42:00  <rendar>caitp-, without that the compiling didn't even start
16:42:05  <caitp->but it says you installed gcc-multilib
16:42:19  <rendar>i installed it
16:42:36  <rendar>apt-get successfully installed gcc-multilib package
16:42:50  <caitp->I mean there are two separate debian packages here, I'm not sure gcc-multilib would solve the libc++ issue
16:43:05  <caitp->try grabbing g++-multilib too
16:43:19  <caitp->I don't use gcc though, so i'm really just guessing
16:43:32  <rendar>caitp-, but why it says "skipping incompatible"? why are those incompatible with v8?
16:43:45  <caitp->it's not v8 saying it's incompatible
16:45:04  <rendar>yes, its 'ld', but i can't get why
16:45:09  <rendar>i always use C++ code, and it works
16:45:29  <rendar>caitp-, btw, when i type 'make dependencies' on v8 i also get this error /bin/sh: 1: gclient: not found
16:45:29  <rendar>make: *** [dependencies] Error 127
16:45:29  <rendar>[rendar v8]$
16:45:35  <rendar>what is 'gclient' there?
16:45:56  <caitp->gclient is from the depot_tools package, used to manage modular dependencies in various google projects
16:46:13  <caitp->and keep everything in a sort of inter-workable state
16:47:03  <rendar>oh
16:47:11  <caitp->http://www.chromium.org/developers/how-tos/install-depot-tools there's instructions there, you need to grab those and put them in your PATH
16:48:26  <caitp->however, I'm not sure you really need that final `gclient sync` line in `make dependencies`
16:48:47  <bnoordhuis>you can also build with `make 18nsupport=off`; i think that removes the need to run `make dependencies`
16:48:50  <caitp->so it probably doesn't matter
16:49:29  <bnoordhuis>rendar: that 'skipping incompatible stdc++' warning probably means that libc++ was compiled for another architecture
16:49:36  <caitp->well it's still going to want to get gtest and gmock
16:50:02  <rendar>bnoordhuis, hmm, so maybe i told v8 to compile for ia32, while i'm in amd64 arch?
16:50:19  <bnoordhuis>caitp-: maybe i have those installed system-wide because they're not installed locally and building still works
16:50:21  <caitp->what exactly did you tell it to compile =)
16:50:32  <rendar>caitp-, nothing just 'make'
16:50:39  <caitp->ah
16:50:47  <bnoordhuis>right, that's an alias for `make native` which tries to guess your arch
16:50:47  <caitp->I believe `make` will go through each arch
16:50:52  <caitp->oh is that right
16:51:11  <bnoordhuis>um, i'm starting to doubt myself now
16:51:20  <rendar>bnoordhuis, so with 'make native' make should do that automatically
16:51:29  <caitp->just try `make x64.release -j32`, it's probably what you want
16:52:35  <bnoordhuis>caitp-: no, i think you're right; just make or `make all` tries to build release and debug builds
16:52:36  <caitp->you can add `i18nsupport=off` too, if you still have issues with deps
16:53:55  <rendar>and with 'make x64.debug' i compile the debug version right?
16:53:59  <bnoordhuis>correct
16:54:16  <bnoordhuis>you can add extrachecks=on for even more asserts
16:54:32  <bnoordhuis>i always test against extrachecks=on debug builds
16:55:14  <rendar>cool
16:55:39  <rendar>bnoordhuis, indeed it seems very useful
16:57:24  <caitp->moreso if you have more than just a macbook to work on :( I really need to get a real computer some time
16:57:55  <caitp->(tests take forever when DCHECKs and extrachecks are used :()
16:59:11  <bnoordhuis>unfortunately true. if it's any consolance, it's not much faster on my beefy desktop machine than it is on my laptop
17:04:58  <rendar>i see
17:05:05  <bnoordhuis>*consolation... i suck at english
17:05:09  <rendar>compiling v8-threads.o lol, is v8 multithreaded now?
17:05:24  <bnoordhuis>yes. it uses it for background compilation
17:05:34  <rendar>oh cool
17:05:47  <rendar>but i can only execute my script in only 1 thread, i guess
17:06:21  <caitp->well, the threading story is a bit more complicated than that
17:06:31  <rendar>caitp-, how was that?
17:06:54  <bnoordhuis>in only 1 thread at one time, really
17:06:57  <caitp->you can find lots of examples of multithreaded stuff, from node clusters, web workers, and threaded XHR requests
17:06:59  <caitp->among other things
17:07:13  <caitp->(in blink/chromium, node.js, etc)
17:07:14  <bnoordhuis>node cluster isn't multi-threaded, it's multi-process
17:07:20  <caitp->yes, but that's the same idea
17:08:01  <caitp->and in either case, they're both pretty heavyweight compared to things like rust or go tasks
17:08:11  <rendar>i see
17:08:33  <rendar>caitp-, so in theory i can run multiple scripts in multiple thread, the only restriction is that 1script:1thread
17:08:50  <rendar>caitp-, i mean, i can't run the same "context" or state of a particular script in 2 threads
17:08:53  <bnoordhuis>scripts run in an isolate and an isolate is single-threaded
17:09:04  <rendar>ok
17:09:05  <bnoordhuis>but you can pass around isolates using v8::Locker and v8::Unlocker objects
17:09:10  <caitp->you won't want to move your isolates between threads if you can avoid it
17:09:12  <rendar>but i can run N isolates in N threads
17:09:18  <bnoordhuis>yes
17:09:27  <rendar>caitp-, why not?
17:09:45  <rendar>caitp-, aren't there just like processes and i'm the scheduler?
17:09:48  <caitp->well, for one thing, what would be the use case?
17:09:49  <rendar>they*
17:10:27  <rendar>caitp-, just for example, let's for instance i need to run A.js and B.js 3 times each one, i have 2 threads T1 and T2
17:10:34  <rendar>T1 starts running A.js and T2 B.js
17:10:49  <rendar>if T2 finish first, it can run A.js for the second time, because its free now
17:10:51  <bnoordhuis>caitp-: work stealing. that's on the middle/long term roadmap for node actually :)
17:11:18  <rendar>bnoordhuis, what you mean with "work stealing" ? what i have just described?
17:11:25  <caitp->bnoordhuis the cost of moving them may get in the way of the benefits though
17:11:40  <bnoordhuis>caitp-: sure, but then again, it may not
17:11:47  <caitp->the best reason to avoid it is of course, threads are basically fountains of bugs
17:11:53  <caitp->so it's just nice to avoid that mess if you can :D
17:11:58  <bnoordhuis>rendar: sorry no, i was replying to caitp-'s "what would be the use case?"
17:12:14  <rendar>bnoordhuis, oh! i see :) what you mean with 'work stealing'?
17:12:28  <caitp->sort of what you were saying
17:12:34  <rendar>ok
17:12:36  <caitp->one thread has free time, grab work from another
17:12:40  <bnoordhuis>yep
17:12:51  <rendar>like the OS does with processes, right
17:12:54  <bnoordhuis>indeed
17:13:14  <caitp->it is basically what the OS does --- but the thing is
17:13:21  <rendar>yes that's basically what i meant before
17:13:50  <caitp->I think you'd see that work a lot better if there were more of a microtask sort of architecture (for v8 and js), but there just isn't
17:14:05  <caitp->that fine-grained scheduling with these details is a bit hard for the os
17:14:13  <caitp->and it's a bit hard with platform thread apis
17:14:19  <caitp->to do it efficiently
17:14:27  <rendar>i see
17:14:39  <caitp->but if node can pull it off, then that's interesting
17:15:03  <rendar>caitp-, btw when i run a js script with v8, basically i have an isolate which represent the machine state of my script, right?
17:15:15  <bnoordhuis>caitp-: node's approach would basically be standalone isolates, web worker style, that only communicate through message passing, like the cluster module does now
17:15:38  <bnoordhuis>but the fact that it shares an address space, makes it possible to do message passing more efficiently
17:15:59  <bnoordhuis>and work scheduling too
17:16:08  <rendar>bnoordhuis, i see, but that would change a lot of node architecture, or not?
17:16:54  <bnoordhuis>rendar: yes, but we're halfway there already
17:17:01  <bnoordhuis>okay, gotta run, dinner. biab
17:18:18  <rendar>bye
17:19:10  <rendar>caitp-, where are the executable once compiled? i should have libv8.so and libplatform.so right?
17:19:51  <caitp->you built with `make x64.debug`?
17:19:56  <rendar>release
17:20:22  <caitp->should be out/x64.release/ by default
17:20:47  <rendar>there i found [rendar x64.release]$ ls
17:20:47  <rendar>cctest d8 lineprocessor mksnapshot obj obj.host obj.target parser-shell process shell unittests
17:21:29  * bnoordhuisquit (Ping timeout: 264 seconds)
17:21:38  <caitp->I've got libgmock, libgtest, libicudata, libicui18n, libicuuc, libv8_base, libv8_libbase, libv8_libbplatform, libv8_nosnapshot, libv8_snapshot
17:21:42  <caitp->you don't have any of those?
17:21:47  <rendar>nop
17:21:58  <caitp->huh
17:22:43  <caitp->where'd you get the source from?
17:22:48  <rendar>git
17:22:54  <rendar>i've just cloned the git repo
17:22:57  <caitp->no but which one :p
17:23:13  <rendar>github? or you mean the version
17:23:23  <caitp->https://github.com/v8/v8-git-mirror this thing?
17:25:04  <rendar> git clone git://github.com/v8/v8.git v8
17:26:33  <rendar>caitp-, is that making any difference?
17:26:52  <caitp->moved to v8-git-mirror a while ago, around the git migration
17:27:18  <caitp->so there's newer stuff in v8-git-mirror, but I can't think of a reason why you'd be missing lib outputs
17:27:30  <rendar>yeah..
17:27:31  <rendar>me too
17:27:56  <caitp->try `find out/x64.release/ -name 'lib*' ?
17:29:09  <rendar>here we go http://nopaste.info/51b90f5009.html
17:29:15  <rendar>are they all the files, right?
17:29:40  <caitp->those look like what you need
17:29:50  <rendar>caitp-, why it compiled .a files and not .so?
17:29:53  <caitp->not sure why they're in that subdirectory for you
17:30:07  <caitp->you need `library=shared` to build shared libs
17:30:16  <rendar>ok
17:30:26  <rendar>make x64.release library=shared ?
17:30:29  * muelliquit (Ping timeout: 244 seconds)
17:30:33  <caitp->(-Dcomponent=shared_library using gyp)
17:30:37  <caitp->yeah
17:30:53  <rendar>i see, thanks
17:32:22  <rendar>caitp-, 'make install' doesn't work, if i want to install that, i simply have to copy all lib* executables in /usr/lib and home/v8/include in /usr/include manually?
17:32:58  <caitp->I guess --- there isn't really infrastructure for system v8 libs
17:33:15  <rendar>i see
17:33:22  <caitp->which makes sense
17:33:30  <rendar>why?
17:33:44  <caitp->canary will want a different v8 from chromium which will want a different v8 from node 0.11 which will want a different v8 from node 0.10
17:33:58  <rendar>oh
17:34:00  <rendar>i see!
17:34:26  <rendar>so for those is better to have their own libv8.so in their directories, and not a system-wide one
17:34:38  <rendar>or even including v8 with a static compile
17:54:18  <rendar>caitp-, i've included all headers and added -Lpath and -lv8_stuff.a but if i compile i still get a lot of undefined symbols, what am i missing?
17:55:18  * Lethalmanquit (Quit: Sto andando via)
17:55:27  <caitp->a good guide for linking would be to look at the rules for d8
17:55:48  <caitp->it's configured with just gyp, but you should be able to use the gypfile as a guide
17:55:58  * abraxas_joined
17:56:56  <rendar>caitp-, i see, what is d8 exactly?
17:57:23  <caitp->d8 is a javascript shell application that is part of v8
17:57:35  <rendar>cool
17:57:40  <caitp->a lot of the unit and perf tests are run through it
18:01:02  * abraxas_quit (Ping timeout: 265 seconds)
18:02:00  <rendar>caitp-, ok i've solved all undefined symbol except 1, icudt52_dat, i've included all icu libraries created by v8, but that one there isn't...it seems some kind of statically binary datafile for ICU
18:09:26  * c4milojoined
18:11:55  * plutoniixjoined
18:14:30  * c4miloquit (Ping timeout: 250 seconds)
18:23:07  <trungl-bot`>Tree closed by [email protected]: Tree is closed (Automatic: "bot_update" on http://build.chromium.org/p/client.v8/builders/Auto-roll%20-%20v8rel/builds/9341 "Auto-roll - v8rel" from None: )
18:25:07  <trungl-bot`>Tree opened by [email protected]: Tree is open (Automatic: ☃)
18:27:37  * bnoordhuisjoined
18:31:58  * bnoordhuisquit (Ping timeout: 245 seconds)
18:38:19  * bnoordhuisjoined
18:43:31  * bnoordhuisquit (Ping timeout: 272 seconds)
19:22:34  * eseideljoined
19:24:00  <eseidel>jochen__: curious what if any interest v8 has in gn? looks like it has a partially workign gn build. I'm considering adding samples/build.gn (for shell) and eventually need to fix the snapshotting building for arm in the gn build (currently the snapshot is somehow built wrong)
19:44:58  * abraxas_joined
19:49:11  * bnoordhuisjoined
19:49:28  * abraxas_quit (Ping timeout: 258 seconds)
19:52:40  * C-Manquit (Quit: Connection reset by beer)
19:53:53  * bnoordhuisquit (Ping timeout: 264 seconds)
19:58:28  * c4milojoined
20:00:32  * bnoordhuisjoined
20:03:13  * c4miloquit (Ping timeout: 252 seconds)
20:36:09  * octetcloudjoined
20:59:39  * muellijoined
21:19:32  * bnoordhu1sjoined
21:20:53  * muelliquit (Ping timeout: 264 seconds)
21:26:04  * muellijoined
21:28:25  * trevnorr1schanged nick to trevnorris
21:29:26  * c4milojoined
21:30:14  * rmcilroy_joined
21:33:43  * abraxas_joined
21:35:05  * bnoordhuisquit (*.net *.split)
21:35:06  * rmcilroyquit (*.net *.split)
21:35:06  * guorquit (*.net *.split)
21:35:06  * rosseauxquit (*.net *.split)
21:35:07  * katuquit (*.net *.split)
21:38:17  * abraxas_quit (Ping timeout: 240 seconds)
22:03:55  * rendarquit (Quit: Leaving)
22:14:45  * plutoniixquit (Quit: จรลี จรลา)
22:19:58  * muelliquit (Ping timeout: 264 seconds)
22:30:01  * guorjoined
22:31:31  * RT|Chatzillajoined
22:32:56  * c4miloquit (Remote host closed the connection)
22:36:32  * katujoined
22:37:41  * rosseauxjoined
23:22:40  * abraxas_joined
23:27:12  * abraxas_quit (Ping timeout: 256 seconds)