00:00:00  * ircretaryquit (Remote host closed the connection)
00:00:08  * ircretaryjoined
00:00:39  <othiym23>I have a test runner that will take a range of Node versions and run a given test suite against all of them and compile the results
00:00:49  <othiym23>seems like it would be nearly trivial to adapt to run on something like that
00:01:18  <groundwater>othiym23: i was going to mention that
00:01:49  <groundwater>one-command and you run yoru test suite across all versions of node
00:02:57  <tjfontaine>othiym23: yup, I did a 537 job, which were all commits from the v0.10/v0.11 branch
00:03:14  <tjfontaine>othiym23: then eventually went back and refactored so I could push it all through jenkins so I got windows, linux, osx as well
00:03:20  <tjfontaine>othiym23: that part of course was much slower :)
00:04:12  <groundwater>how many troll-demerits do i get for suggesting a windows manta environment
00:04:24  <tjfontaine>a lot, I did suggest it, when I suggested linux :)
00:04:37  <tjfontaine>I was willing to take the demerits based on how much I hate jenkins :P
00:06:46  <tjfontaine>anyway: mfind /commits | mjob create -s /gitcore --init 'tar xzf /assets/gitcore' -m 'cd gitcore && git checkout $(basename $MANTA_INPUT_FILE) && ./configure && make -j12 binary && find . -name \*tar.gz | xargs mput -f /$(basename $MANTA_INPUT_FILE)/smartos/'
00:06:50  <tjfontaine>more or less
00:07:24  <othiym23>why -j12?
00:07:43  <othiym23>(as opposed to -j4 or -j65536)
00:07:50  <tjfontaine>well, anything to parallelize the cpu time, but you could do 32 if you wanted, by default you only get 1024mb memory
00:08:01  <tjfontaine>32 makes sense because that's the logical cores available
00:08:06  * dshaw_quit (Quit: Leaving.)
00:08:07  <othiym23>oh right, that's per node
00:08:10  <othiym23>"node"
00:08:19  <othiym23>Node node
00:08:23  <tjfontaine>right
00:08:30  <othiym23>sorry, I've had an exciting day and am kind of braindead
00:08:37  <tjfontaine>exciting good day?
00:16:36  * dshaw_joined
00:19:09  * indexzerojoined
00:20:38  <Raynos>tjfontaine: if it helps we can hack on it somewhere this weekend. I can supply beer, you can supply code. Teamwork.
00:21:14  * dshaw_quit (Ping timeout: 256 seconds)
00:22:20  <othiym23>Raynos: sorry about that rando e-mail from my coworker, we're getting ready to reopen the beta and I don't think she realized it was still connected to everybody
00:22:35  <tjfontaine>Raynos: hehe, I may take you up on it
00:22:46  <Raynos>othiym23: It's cool I replied back to let her know it was live :D
00:23:28  <Raynos>tjfontaine: I assume your in SF. I can also be helpful in other means like actually ask you questions & write it myself whilst you do other stuff
00:24:04  <tjfontaine>hehe ya, moral support :)
00:31:39  <Raynos>tjfontaine: Where is a good place to ask questions like "I'm debugging core dumps & using ::findjsobjects and I can't see any differences"
00:32:15  * TooTallNatequit (Quit: Computer has gone to sleep.)
00:32:26  <Raynos>tjfontaine: Could ::findjsobjects be extended to return total no of objects and aggregate size in bytes for all those objects ?
00:32:44  <tjfontaine>well you can see total no of objections, use ::findjsobjects -v
00:33:20  <tjfontaine>but, #smartos/joyent are probably the best bets, or to use me
00:33:58  <Raynos>yeah I can see total no of objects but can't see total number of bytes of memory used for those objects (which I care more about)
00:34:20  <tjfontaine>right, I wonder if there is a way to get that
00:34:25  <tjfontaine>dap: ping
00:34:37  <dap>Hi
00:34:45  <tjfontaine>ohi
00:35:01  <tjfontaine>so Raynos is wondering if findjsobjects can report the total memory usage?
00:35:44  <dap>It doesn't today…I'm not sure if that's easy or not. ISTR there's something in the V8 C++ object that reports the size of the object, but I'm not sure if it's the right thing or not
00:36:28  <tjfontaine>ya, I wasn't sure how much information we have, we could certainly make some assumptions purely on structure size and count, but that's not the whole story of course
00:36:48  <Raynos>my use case is I compare two core dumps with a difference of 150mb and the number of objects hasn't grown noticably
00:37:04  <tjfontaine>well it could be external memory
00:37:12  <tjfontaine>do you run with UMEM_DEBUG=default
00:37:19  <Raynos>its all in [heap] as shown by pmap
00:37:31  <dap>What's "pmap COREFILE" show for the "heap" section?
00:37:36  <dap>Oh, that's the native heap, not JS.
00:37:44  <dap>I believe.
00:38:51  <tjfontaine>you should be able to get some insight from the umem stats
00:39:02  * dshaw_joined
00:39:11  <Raynos>pmap says 230mb / 350mb. previous core shows 146 / 244mb
00:39:19  <Raynos>also pmap COREFILE is super convenient :)
00:40:33  * dshaw_quit (Read error: Connection reset by peer)
00:40:53  * dshaw_joined
00:45:42  * dshaw_quit (Ping timeout: 264 seconds)
00:48:39  * st_lukequit (Remote host closed the connection)
00:49:15  <Raynos>tjfontaine, dap: thanks btw.
00:49:28  <dap>Oops, sorry for disappearing there
00:49:33  <dap>So yeah, it sounds like your native heap is growing
00:51:02  <dap>Raynos: can you gist up the ::umastat output, and the output of !uname -v?
00:53:19  * dshaw_joined
00:53:57  <Raynos>dap: ::umastat is an invalid command
00:53:59  <Raynos>in mdb
00:54:20  <dap>Crap. Do you know how this node was build?
00:54:22  <dap>built*?
00:54:39  <Raynos>its v0.8.4 build from source
00:54:53  <Raynos>I have an old version of mdb though
00:55:33  <Raynos>dap: https://gist.github.com/Raynos/ae299094077df0228a50
00:56:04  <tjfontaine>mlogin to the rescue? :)
00:56:06  <dap>::umastat has been around for a while, but it requires that your node program was linked against libumem (see libumem(3LIB))
00:57:40  * dshaw_quit (Ping timeout: 246 seconds)
00:57:49  <Raynos>I don't know about that. are more recent versions of node automatically linked? or do I have to do that explicitely
00:57:56  <dap>libumem is our better userland memory allocator, which keeps stats and cheap debugging information as well. If you're able to rerun, you can LD_PRELOAD it. If you link against that, you can actually use ::findleaks to find native leaks directly. From ::umastat, we might also be able to tell if it's heap fragmentation vs. an actual leak.
00:58:05  <dap>I believe more recent versions of Node are automatically linked against libumem.
00:58:18  * indexzeroquit (Quit: indexzero)
00:58:21  <dap>You can just use "-lumem" if you want to rebuild, or you can just LD_PRELOAD it
00:59:09  * indexzerojoined
00:59:15  <tjfontaine>I don't remember exactly when the change happened for -lumem, but I would have assumed all of 0.8 had it
00:59:29  <tjfontaine>is it possible the umem dmod for mdb didn't load?
00:59:35  <Raynos>that might be it
01:00:31  <dap>That's unlikely
01:00:40  <tjfontaine>k
01:00:45  <dap>But MDB prints the modules it loads on startup
01:00:52  <dap>You can use pmap to see if libumem was loaded into the process
01:01:11  <Raynos>yeah that was the issue.
01:01:32  * dappart
01:01:33  * dapjoined
01:01:35  <dap>which was?
01:01:54  <Raynos>dap: https://gist.github.com/Raynos/9742d850bf2e73ea72e2 I had to load libumem.so.1
01:02:13  <Raynos>I like how the only google hit for that error is the source code ( https://bitbucket.org/illumos/illumos-gate/src/36d116935c7e/usr/src/cmd/mdb/common/modules/libumem/umem.c )
01:02:46  * dshaw_joined
01:03:18  <dap>Sorry, what I really meant was: MDB will only load the libumem module (which is what defines the ::umastat command) if the process (or core file) you're debugging had libumem loaded. So while you were able to load the module explicitly, that doesn't help if the process didn't actually loaded libumem. Does that make sense?
01:04:13  <dap>I would check with pmap whether libumem was in the process. I suspect not, and in that case, there's nothing more I know how to help with.
01:04:58  * EhevuTovquit (Quit: This computer has gone to sleep)
01:07:16  <Raynos>dap: that makes sense
01:08:43  <dap>… but if you are able to preload libumem and rerun your program, you can likely find any native leaks directly with ::findleaks
01:08:45  <Raynos>dap: pmap core/api.core.2013.09.18~1 | grep libumem gives me nothing. That means libumem was not loaded?
01:08:49  <dap>Yeah.
01:08:53  <Raynos>Ok, thanks.
01:08:53  <dap>If that's remotely possible, it's worth a shot.
01:09:35  * `3rdEdenquit (Ping timeout: 260 seconds)
01:10:43  * Domenic_quit (Ping timeout: 256 seconds)
01:11:02  <dap>I've got to run for now. Good luck!
01:11:41  <tjfontaine>Raynos: I'm going as well for now, but I will probably be back later this evening but certainly tomorrow to help with any other questions
01:11:56  <Raynos>tjfontaine, dap: thanks again
01:12:02  <tjfontaine>I have already started a blog on the basics of this, I described some of it in one of those level* issues
01:12:23  <dap>No problem!
01:13:13  * dapquit (Quit: Leaving.)
01:21:29  * TooTallNatejoined
01:24:20  * jmar777joined
01:26:23  * dshaw_quit (Quit: Leaving.)
01:30:53  * indexzeroquit (Quit: indexzero)
01:31:24  * abraxasjoined
01:34:12  * julian_duquejoined
01:37:06  * julianduquequit (Ping timeout: 245 seconds)
01:55:46  * TooTallNatequit (Quit: Computer has gone to sleep.)
02:02:29  * amartensquit (Quit: Leaving.)
02:03:16  * julian_duquechanged nick to julianduque
02:20:32  * toothrotquit (Read error: Connection reset by peer)
02:21:58  * jmar777quit (Remote host closed the connection)
02:27:40  * groundwaterquit (Quit: groundwater)
02:34:19  * c4miloquit (Remote host closed the connection)
02:59:42  * c4milojoined
03:10:11  * Domenic_joined
03:11:14  * `3E|DINNERjoined
03:25:58  * Domenic_quit (Ping timeout: 256 seconds)
03:27:27  * `3E|DINNERquit (Ping timeout: 260 seconds)
03:37:55  * ecrjoined
03:39:57  * ecrquit (Client Quit)
03:45:58  * c4miloquit (Remote host closed the connection)
03:51:17  * st_lukejoined
04:01:40  * st_lukequit (Remote host closed the connection)
04:03:23  * st_lukejoined
04:42:04  * felixgejoined
04:42:04  * felixgequit (Changing host)
04:42:04  * felixgejoined
05:21:55  * Domenic_joined
05:25:45  * `3E|DINNERjoined
05:29:14  * c4milojoined
05:49:24  * brsonquit (Quit: leaving)
06:05:30  * Benviejoined
06:13:35  * amartensjoined
06:20:54  * rendarjoined
06:24:39  * amartensquit (Quit: Leaving.)
06:31:06  * bajtosjoined
06:33:11  * julianduquequit (Quit: leaving)
06:36:03  * julianduquejoined
06:38:22  * julianduquequit (Client Quit)
06:41:36  <MI6>nodejs-v0.10-windows: #219 UNSTABLE windows-ia32 (7/600) windows-x64 (7/600) http://jenkins.nodejs.org/job/nodejs-v0.10-windows/219/
06:50:18  * `3E|DINNERchanged nick to `3rdEden
06:50:48  * `3rdEdenchanged nick to Guest93877
06:51:05  * Guest93877changed nick to `3E
07:15:54  * c4miloquit (Remote host closed the connection)
07:29:11  * st_lukequit (Remote host closed the connection)
07:50:59  * dominictarrjoined
07:52:56  * bnoordhuisjoined
08:08:08  * xakajoined
08:08:45  * xakaquit (Client Quit)
08:13:56  * xakajoined
08:16:05  * xakaquit (Client Quit)
08:48:59  * bnoordhuisquit (Ping timeout: 248 seconds)
08:58:54  * hzjoined
09:01:33  * bnoordhuisjoined
09:17:50  * bnoordhuisquit (Ping timeout: 240 seconds)
09:48:16  * Kakerajoined
09:51:21  <indutny>hey people
10:35:24  * bajtosquit (Quit: bajtos)
10:45:59  <MI6>nodejs-v0.10: #1491 UNSTABLE linux-ia32 (1/600) smartos-x64 (3/600) http://jenkins.nodejs.org/job/nodejs-v0.10/1491/
11:20:51  * roxlu_quit (Ping timeout: 245 seconds)
11:27:44  * bnoordhuisjoined
12:12:29  * inolenquit (Quit: Leaving.)
12:23:33  * toothrjoined
12:27:58  <bnoordhuis>hey look, another c++-vs-fp discussion on HN
12:28:11  <bnoordhuis>they were interesting the first 100 times
12:36:43  * AvianFlujoined
12:39:50  <isaacbw>what a load of idiocy: https://news.ycombinator.com/reply?id=6416248&whence=item%3fid%3d6414162
12:39:53  <isaacbw>WRITE ALL THINGS WITH C
12:39:53  <LOUDBOT>NOW I HAVE A 2W ANTENNA, LETS GO FRY SOME NEIGHBOURS!
12:40:01  <isaacbw>lol
12:42:48  * inolenjoined
12:50:16  * inolenquit (Ping timeout: 240 seconds)
13:45:39  * bajtosjoined
13:45:46  * piscisaureus_joined
13:45:49  * bajtosquit (Client Quit)
13:47:39  * contrahaxquit (Read error: Connection reset by peer)
13:47:39  * emilsedghquit (Ping timeout: 245 seconds)
13:47:39  * LOUDBOTquit (Ping timeout: 245 seconds)
13:48:10  * emilsedgh_joined
13:48:26  * contrahaxjoined
13:52:54  * LOUDBOTjoined
14:04:14  * c4milojoined
14:04:18  * AvianFluquit (Remote host closed the connection)
14:04:48  * jmar777joined
14:08:52  * c4miloquit (Remote host closed the connection)
14:30:56  * julianduquejoined
14:37:05  * bnoordhuisquit (Ping timeout: 245 seconds)
14:37:30  * bradleymeckjoined
14:42:06  * austojoined
14:47:27  * inolenjoined
14:48:01  * c4milojoined
14:48:44  * dapjoined
14:48:47  * paulfryzeljoined
14:49:50  * c4miloquit (Remote host closed the connection)
14:50:45  * c4milojoined
14:51:40  * inolenquit (Ping timeout: 245 seconds)
14:57:13  * AvianFlujoined
15:03:01  * inolenjoined
15:17:10  * indexzerojoined
15:17:10  <MI6>nodejs-master: #564 UNSTABLE osx-ia32 (1/640) linux-x64 (3/640) smartos-x64 (6/640) http://jenkins.nodejs.org/job/nodejs-master/564/
15:17:47  * paulfryzelquit (Remote host closed the connection)
15:24:46  * jmar777quit (Remote host closed the connection)
15:41:23  * bradleymeckquit (Quit: bradleymeck)
15:45:49  * indexzeroquit (Quit: indexzero)
15:48:56  * bajtosjoined
15:53:22  * amartensjoined
16:00:27  <tjfontaine>hey party people
16:06:56  * TooTallNatejoined
16:20:04  * felixgequit (Quit: felixge)
16:35:39  * TooTallNatequit (Ping timeout: 260 seconds)
16:38:41  * inolenquit (Quit: Leaving.)
16:40:27  <MI6>joyent/node: Ben Noordhuis master * 0373818 : dns: fix c-ares error reporting regression (+1 more commits) - http://git.io/2TrNfg
16:42:42  * bnoordhuisjoined
16:47:16  * groundwaterjoined
16:48:06  * bnoordhuisquit (Ping timeout: 264 seconds)
16:49:18  <MI6>nodejs-master: #565 UNSTABLE linux-x64 (1/641) smartos-x64 (6/641) http://jenkins.nodejs.org/job/nodejs-master/565/
17:00:13  * dshaw_joined
17:01:39  * TooTallNatejoined
17:03:18  <MI6>nodejs-master-windows: #358 UNSTABLE windows-x64 (21/641) windows-ia32 (20/641) http://jenkins.nodejs.org/job/nodejs-master-windows/358/
17:11:08  * dshaw_quit (Quit: Leaving.)
17:12:08  * AvianFluquit (Remote host closed the connection)
17:13:58  * AvianFlujoined
17:14:24  * dshaw_joined
17:15:00  * bnoordhuisjoined
17:16:15  * dshaw_1joined
17:16:31  * dshaw_quit (Read error: Connection reset by peer)
17:17:55  * c4miloquit (Remote host closed the connection)
17:20:05  * indexzerojoined
17:20:05  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
17:21:25  * inolenjoined
17:36:47  * c4milojoined
17:53:38  <MI6>libuv-master: #248 UNSTABLE windows (4/195) smartos (9/194) http://jenkins.nodejs.org/job/libuv-master/248/
17:57:15  * M28quit (Ping timeout: 248 seconds)
18:04:09  <trevnorris>morning
18:04:13  <tjfontaine>hola
18:06:24  <MI6>libuv-node-integration: #233 UNSTABLE smartos-ia32 (1/641) smartos-x64 (5/641) http://jenkins.nodejs.org/job/libuv-node-integration/233/
18:06:33  <trevnorris>bnoordhuis: well, that was a fast fix :)
18:07:20  * bnoordhuisquit (Ping timeout: 256 seconds)
18:08:26  * felixgejoined
18:10:45  * bradleymeckjoined
18:15:47  * felixgequit (Quit: felixge)
18:19:47  * jmar777joined
18:20:44  * c4miloquit (Remote host closed the connection)
18:25:09  * c4milojoined
18:25:38  * dshaw_1quit (Quit: Leaving.)
18:28:10  * dshaw_joined
18:30:01  * c4miloquit (Ping timeout: 245 seconds)
18:33:06  * dshaw_quit (Ping timeout: 264 seconds)
18:36:47  * brsonjoined
18:38:05  * c4milojoined
18:41:18  * defunctzombie_zzchanged nick to defunctzombie
18:45:49  * dshaw_joined
18:45:54  <trevnorris>tjfontaine: can https://github.com/joyent/node/pull/6248 be closed?
18:46:24  <tjfontaine>could have sworn I pressed the comment and close
18:46:36  <tjfontaine>thanks for finding it though
18:46:42  <trevnorris>:)
18:48:05  <trevnorris>tjfontaine: oh, and you should love my latest issue ;P
18:48:12  <tjfontaine>ya I saw it :)
18:48:17  <tjfontaine>and promptly ignored for now
18:48:29  <trevnorris>haha
18:48:48  <tjfontaine>some of the phrasing is odd
18:49:01  * felixgejoined
18:49:02  <trevnorris>probably, didn't even read over it twice.
18:49:19  <tjfontaine>and I doubt we'll really be able to get away with things not inheriting EE
18:49:34  <MI6>nodejs-master-windows: #359 UNSTABLE windows-x64 (22/641) windows-ia32 (21/641) http://jenkins.nodejs.org/job/nodejs-master-windows/359/
18:50:44  <tjfontaine>returning functions for calling is cute, but how does that work from an async mechanism?
18:51:09  <trevnorris>still figuring that out.
18:51:26  * EhevuTovjoined
18:51:50  <tjfontaine>maybe doing something like a nextTick/setImmediate where cbs are pushed into an array and popped and executed in js land
18:51:52  <trevnorris>i'm wondering if there's just a really quick way to dump into some point of JS execution. like possibly a v8 natives type thing.
18:52:11  <trevnorris>problem is getting into JS from MakeCallback w/o needing to call a function.
18:52:30  <tjfontaine>well, if you can coallesce a group of cb's your life gets easier at least
18:52:43  <tjfontaine>anyway, it's an interesting though experiment
18:52:53  <tjfontaine>*thought
18:53:24  <trevnorris>yeah, if we even did half that list before v1.0 i'd be impressed.
18:53:32  <tjfontaine>heh
18:54:16  * bajtosquit (Quit: bajtos)
18:57:19  * bnoordhuisjoined
19:01:31  * EhevuTovquit (Quit: This computer has gone to sleep)
19:01:38  * dshaw_quit (Quit: Leaving.)
19:02:12  * EhevuTovjoined
19:07:47  <trevnorris>bnoordhuis: have a sec?
19:12:05  <bnoordhuis>trevnorris: sure. shoot
19:13:59  <trevnorris>bnoordhuis: say you need to run a JS function from C++, it's much faster to return the function and run it in JS. But in an async scenario that's not possible. So i'm looking for a quick way to dump into JS, and was thinking maybe a v8 native might help? or you have any other ideas?
19:14:40  <bnoordhuis>trevnorris: i'm not sure what you mean
19:15:03  <bnoordhuis>what does 'async scenario' mean in this context?
19:15:20  <trevnorris>say, coming from a libuv callback into MakeCallback
19:15:26  * felixgequit (Ping timeout: 245 seconds)
19:16:06  <trevnorris>for an example case, check what i'm doing in basics.cc and run.js: https://github.com/trevnorris/v8-basics
19:16:12  <bnoordhuis>oh, i see what you mean
19:16:26  <bnoordhuis>i don't think that's really possible now
19:16:39  <bnoordhuis>what you'd want is an event pull mechanism whereas libuv is push
19:17:14  <bnoordhuis>you could collect events with uv_run(UV_RUN_ONCE), then dispatch them afterwards
19:17:39  <bnoordhuis>but i'm not sure if it would be faster. it will change the behavior of some things as well
19:18:03  <bnoordhuis>unless i'm completely missing the point, of course :)
19:18:11  <trevnorris>sounds complicated. more I was wondering if there was a quicker way to enter JS from C++ so that JS could execute a function, without needing to use Function::Call()
19:18:16  * TooTallNatequit (Quit: Computer has gone to sleep.)
19:18:58  <trevnorris>like, if im already in JS and I need to call a native method and the native method needs to call a callback. it's much faster to have the native method return the callback and run that callback from JS
19:19:35  <bnoordhuis>ah, right. so what's stopping you?
19:20:46  <trevnorris>the callback passed to MakeCallback needs to be called with Function::Call(), which is the slow part. don't know if this is the correct terminology, but there's no js stack setup to call the MakeCallback callback.
19:21:45  <bnoordhuis>right, but you can just return the Function to js land in that case, right?
19:22:14  <trevnorris>but return it to what? in an async scenario there's no js function waiting for a return value
19:22:40  <bnoordhuis>okay, i guess we were talking about the same thing after all :)
19:22:47  <bnoordhuis>that's what i meant with push vs pull
19:22:53  <trevnorris>ah, ok.
19:23:39  <trevnorris>ok, so i'll just have to eat the cost of the initial Function::Call() in MakeCallback.
19:23:46  <bnoordhuis>yes, for now
19:24:03  <bnoordhuis>we could avoid it for some things, e.g. the timer callback
19:24:33  <bnoordhuis>by rewriting the inner loop from uv_run(UV_RUN_DEFAULT) to uv_run(UV_RUN_ONCE) and run timers (and check/idle/etc. handles) before and afterwards
19:24:45  <trevnorris>ooh, no. don't want to go there.
19:25:02  <trevnorris>and least, I don't :)
19:25:16  <bnoordhuis>it's not terribly complicated - but it might not have a measurable impact
19:25:25  <trevnorris>yeah, makes sense.
19:25:35  <bnoordhuis>i'll put it on my 'rainy day' list
19:26:20  <trevnorris>think there's some magical v8::internal that would allow us to call a JS function faster? (you mentioned something about a different internal array that's faster)
19:26:24  <trevnorris>something like that.
19:26:40  * dshaw_joined
19:27:32  <trevnorris>oh well, this is all v0.13 material. just been on my mind a lot.
19:28:20  <trevnorris>since MakeCallback can run upwards of 100k/sec in high I/O and it's 5x's faster to pass a js callback to js then run (i.e. savings of 60 ns :P
19:28:36  <trevnorris>per call)
19:31:18  * dshaw_quit (Ping timeout: 264 seconds)
19:32:33  * dshaw_joined
19:36:16  <bnoordhuis>trevnorris: v8 has this concept of 'runtime functions', which are js->c++ calls but without the usual overhead
19:36:33  <bnoordhuis>usual overhead as in setting up stack frames, checking for exceptions, etc.
19:36:40  * dshaw_quit (Ping timeout: 245 seconds)
19:37:15  <bnoordhuis>they're buried deep inside v8, you can't create your own without patching v8
19:37:38  <trevnorris>bnoordhuis: interesting. think they'd accept a patch to allow them to be created?
19:38:34  <bnoordhuis>i very much doubt that. for one, the list of runtime functions is static, generated at compile time
19:39:26  <trevnorris>ah, ok.
19:40:35  <bnoordhuis>trevnorris: it's in deps/v8/src/runtime.{cc,h} in case you're interested
19:40:52  <trevnorris>well, just calling a C++ function from JS only costs 8ns, and generally things running faster than that are better in js.
19:40:58  <trevnorris>bnoordhuis: thanks. i'll take a look.
19:43:34  <trevnorris>bnoordhuis: do you remember what was that faster internally array thing that doesn't do a bunch of checks?
19:43:44  <trevnorris>mentioned it a while back.
19:44:15  * dshaw_joined
19:45:17  <bnoordhuis>trevnorris: FixedArray
19:46:10  <trevnorris>bnoordhuis: thanks for the chat.
19:49:32  <bnoordhuis>np
19:50:13  * julianduquequit (Quit: leaving)
19:54:04  * bradleymeckquit (Quit: bradleymeck)
20:22:22  * jmar777quit (Read error: Connection reset by peer)
20:22:39  * jmar777joined
20:37:51  * bradleymeckjoined
20:38:23  * bradleymeckquit (Client Quit)
20:38:27  * jmar777quit (Remote host closed the connection)
20:58:28  * dshaw_quit (Quit: Leaving.)
21:02:38  * jmar777joined
21:05:16  <MI6>joyent/libuv: Timothy J Fontaine master * 717e0e5 : build: use -pthread on sunos - http://git.io/-Du6BQ
21:06:37  * kellabytequit (Quit: Quit)
21:06:56  <tjfontaine>bnoordhuis: it wasn't an issue in node because the gyp builds already used it of course
21:14:37  <MI6>libuv-master: #249 UNSTABLE windows (4/195) smartos (2/194) linux (1/194) http://jenkins.nodejs.org/job/libuv-master/249/
21:16:58  <MI6>libuv-master-gyp: #187 UNSTABLE windows-x64 (3/195) smartos-ia32 (2/194) windows-ia32 (4/195) smartos-x64 (2/194) http://jenkins.nodejs.org/job/libuv-master-gyp/187/
21:34:03  <MI6>libuv-node-integration: #234 UNSTABLE smartos-x64 (7/641) http://jenkins.nodejs.org/job/libuv-node-integration/234/
21:37:05  * AvianFluquit (Ping timeout: 245 seconds)
21:45:56  * jmar777quit (Remote host closed the connection)
21:49:01  <tjfontaine>Raynos: how goes the battle?
21:49:36  <Raynos>tjfontaine: which one? the one against mdb?
21:50:10  <tjfontaine>yes
21:50:10  <Raynos>it can be roughly described as memory leaks are not important, ship more features, write more web pages
21:50:31  <tjfontaine>hehe ok, do you have plans for making sure you're using umem going forward?
21:51:30  <Raynos>tjfontaine: not really
21:51:51  <tjfontaine>ok
21:51:53  <Raynos>tjfontaine: well I do. I heard node 0.12 is coming in 2 weeks so ill just upgrade to that
21:52:08  <tjfontaine>haha, I don't know where this 2 weeks thing keeps coming from :)
22:14:53  <Domenic_>it's always 2 weeks
22:16:16  <tjfontaine>5 mins turkish
22:18:35  * emilsedgh_quit (Changing host)
22:18:35  * emilsedgh_joined
22:21:02  * indexzeroquit (Quit: indexzero)
22:30:44  <othiym23>it comes from this dude named Isaac Schlueter
22:30:49  <othiym23>some of you know him, I think
22:34:58  * bnoordhuisquit (Ping timeout: 252 seconds)
22:36:13  <tjfontaine>I've never seen this person
22:41:13  * dshaw_joined
22:43:01  * dominictarrquit (Quit: dominictarr)
22:51:54  * rendarquit
23:10:26  * TooTallNatejoined
23:14:12  * jmar777joined
23:15:58  * M28joined
23:19:55  * Kakeraquit (Ping timeout: 240 seconds)
23:35:59  * brsonquit (Quit: Lost terminal)
23:36:39  * brsonjoined
23:40:56  * bnoordhuisjoined
23:45:38  * bnoordhuisquit (Ping timeout: 256 seconds)