11:19:07  * piscisaureus_joined
11:37:04  * bnoordhuisjoined
12:09:21  * AvianFlujoined
12:31:19  * paddybyersquit (Ping timeout: 244 seconds)
12:33:05  * dileepquit (Quit: Page closed)
12:53:27  * c4milojoined
13:08:49  * paddybyersjoined
13:12:42  * wavded_joined
13:13:37  * hzjoined
13:18:33  * piscisaureus_quit (Ping timeout: 264 seconds)
13:32:26  * dapjoined
13:33:00  * piscisaureus_joined
13:43:45  <piscisaureus_>bnoordhuis: heyu
13:44:21  <piscisaureus_>bnoordhuis: do you know how to re-export symbols with the solaris linker
13:46:07  <AvianFlu>piscisaureus_: http://docs.oracle.com/cd/E19683-01/817-3677/817-3677.pdf
13:46:13  <AvianFlu>"the solaris linker and libraries guide"
13:46:17  <piscisaureus_>ah
13:46:19  <piscisaureus_>thanks AvianFlu
13:46:27  <piscisaureus_>that stuff is not google-able
13:46:34  <AvianFlu>yeah, tell me about it XD
13:46:40  <piscisaureus_>I think joyent should do some SEO :-)
13:50:56  <mmalecki>I'd rather have them make it sane
13:51:42  <tjfontaine>I'd rather them switch to the llvm/clang stack :)
13:51:43  <AvianFlu>I'd rather the solaris linker die in a fire, personally
13:51:50  <AvianFlu>but either of your suggestions are also acceptable XD
13:55:40  <bnoordhuis>piscisaureus_: hoya
13:55:47  <bnoordhuis>with a symbol map?
13:56:06  <piscisaureus_>bnoordhuis: well, I don't care
13:56:13  <bnoordhuis>i think sunos ld calls it an export map
13:56:18  <bnoordhuis>but the basic principle is the same
13:56:22  <piscisaureus_>bnoordhuis: I could not find it :-(
13:56:35  <piscisaureus_>bnoordhuis: but it does support something akin to --whole-archive
13:56:42  <piscisaureus_>(-z allexctract)
13:56:47  <piscisaureus_>so I suppose that should be fine
13:57:05  <piscisaureus_>bnoordhuis: the only problem is that you can't specify this stuff in gyp
13:57:19  <piscisaureus_>and windows, conveniently, does not have --whole-archive
13:57:24  <piscisaureus_>:-(
14:12:15  <piscisaureus_>bnoordhuis: I think that we also need to re-export libuv
14:12:30  <piscisaureus_>bnoordhuis: e.g., in the windows version pthread_condvar_init is missing
14:12:36  <piscisaureus_>er, uv_condvar_init :-)
14:12:39  <bnoordhuis>piscisaureus_: heh
14:12:45  <bnoordhuis>but yes, that seems plausible
14:13:52  <piscisaureus_>I wonder if this process could be somehow automated
14:16:33  <indutny>isaacs: http://validator.w3.org/nu/?doc=http%3A%2F%2Fnodejs.org
14:28:59  * loladirojoined
14:40:14  * hzquit
14:43:08  <tjfontaine>piscisaureus_: why not just export all static dependencies, like how osx works today :)
14:47:32  <txdv>what other langauges have that inheritance model isaacs indutny?
14:47:58  <indutny>good question
14:48:06  <piscisaureus_>yeah
14:48:09  <piscisaureus_>let me try that
14:48:12  <indutny>http://en.wikipedia.org/wiki/Prototype-based_programming
14:48:19  <piscisaureus_>it seems reasonable to re-export v8+uv+openssl
14:48:30  <piscisaureus_>the other parts are probably very small anyway
14:50:02  <piscisaureus_>ok, building with opt:noref
14:54:30  * benoitcquit (Excess Flood)
14:55:57  <piscisaureus_>6,687,232 node-opt-ref.exe
14:55:57  <piscisaureus_>7,534,080 node.exe
14:56:02  <piscisaureus_>^-- acceptable right?
14:56:13  <piscisaureus_>adds 900k to the binary
14:57:14  <tjfontaine>doesn't seem that bad
14:57:39  * janjongboomquit (Quit: janjongboom)
14:57:44  <piscisaureus_>but...
14:57:50  <piscisaureus_>openssl is not exported still..
14:57:57  * mikealjoined
14:58:11  <indutny>oh crap
14:58:12  <indutny>:)
14:58:19  <indutny>btw, how does compression affect resulting size?
14:58:28  <indutny>is there much difference after it?
14:58:29  <indutny>piscisaureus_: ^
14:58:35  * mikeal1quit (Read error: Connection reset by peer)
14:58:37  <piscisaureus_>indutny: look up
14:58:40  <piscisaureus_>but I think it's not all good
14:58:45  <piscisaureus_>libuv is also not exported
14:58:47  <indutny>well
14:58:52  <indutny>that's without compression, I suppose
14:58:59  <piscisaureus_>ah
14:59:01  <indutny>it's just with stripped symbols, right?
14:59:44  <piscisaureus_>weird
14:59:55  <piscisaureus_>actually the exports table of both files have the same lenght
15:00:14  * loladiroquit (Quit: loladiro)
15:00:40  <piscisaureus_>hmmmm
15:01:02  <piscisaureus_>wait, what
15:01:02  <piscisaureus_>fuck
15:01:15  <tjfontaine>o0
15:01:32  <piscisaureus_>forget about all this :-)
15:01:35  * benoitcjoined
15:02:02  <tjfontaine>forget about what?
15:02:15  <piscisaureus_>it doesn't make any differene
15:02:24  <piscisaureus_>the binary just gets larger
15:02:30  <piscisaureus_>and we will still need the def file on windows
15:02:55  <piscisaureus_>because the only symbols that are visible in the .exe must be in either a def file or declared as dllexport
15:03:06  <piscisaureus_>so infact libuv is included entirely already on windows
15:03:18  <tjfontaine>right I was going to ask that, NOREF is more like disabling dead code striping
15:03:24  <piscisaureus_>ya
15:03:27  <indutny>:)
15:03:36  <piscisaureus_>so we still need openssl.def on windows
15:03:36  <indutny>C++ gurus
15:03:37  <kohai>C has 23 beers
15:03:49  <indutny>what's the value of sizeof(int[123])
15:03:54  <indutny>on x86 platform
15:03:56  <piscisaureus_>thanks, kohai. That was the only reasonable response
15:04:03  <indutny>piscisaureus_: haha
15:04:03  <tjfontaine>sizeof pointer?
15:04:08  <indutny>pointer?
15:04:16  <piscisaureus_>indutny: 4*123
15:04:21  <indutny>ok
15:04:26  <tjfontaine>interesting
15:04:32  <indutny>that's the only difference between int* and int[x], right?
15:04:38  <piscisaureus_>I'm not sure
15:04:44  <indutny>(and yeah, I know it's plain C)
15:04:50  <indutny>(I'm just doing hard C++ action in background)
15:05:33  <indutny>that's actually pretty interesting question
15:05:40  <indutny>about difference between int* and int[x]
15:05:55  <indutny>it seems that there're no any other differencies, but I'm not sure
15:05:57  <indutny>bnoordhuis: ^^
15:07:23  * ericktjoined
15:09:22  <piscisaureus_>I wonder how chromium devs get anything done
15:09:27  <piscisaureus_>if compiling takes half a day
15:09:33  <tjfontaine>ninja
15:10:09  <tjfontaine>also I think a lot of them have 32core workstations
15:10:19  <piscisaureus_>I also wonder how much C02 compiling developers generate
15:10:19  <indutny>piscisaureus_: ++
15:10:20  <kohai>piscisaureus_ has 19 beers
15:10:31  <indutny>piscisaureus_: I think they're using development builds
15:10:36  <indutny>piscisaureus_: Release compiles for ages on my mac
15:10:56  <indutny>and yeah, build v8 on 16 core machine lasts for about 40 seconds
15:11:07  <indutny>I think chromium may build in like 4-5 minutes here
15:22:47  * dapquit (Quit: Leaving.)
15:24:12  <piscisaureus_>-rwxrwxr-x 1 bert bert 11269839 Oct 26 17:22 node-reexport
15:24:12  <piscisaureus_>-rwxrwxr-x 1 bert bert 10788921 Oct 26 17:10 node-no-reexport
15:24:22  <piscisaureus_>this is after re-exporting v8, openssl and libuv
15:24:23  <piscisaureus_>so
15:24:35  <piscisaureus_>let's just do this on windows
15:24:40  <piscisaureus_>er, on unices
15:24:45  <piscisaureus_>and use the def file on windows
15:24:47  <piscisaureus_>and be happy
15:25:36  <piscisaureus_>so we need to teach gyp to wrap libs in -Wl,--whole-archive -lmylib -Wl,--no-whole-archive
15:25:59  <indutny>nice
15:26:44  <tjfontaine>better you than me
15:32:12  * ericktquit (Quit: erickt)
15:33:01  * mmaleckichanged nick to mmalecki[brb]
15:34:25  <txdv>the only difference is that the size of the array is associated with the type
15:37:24  <piscisaureus_>$(LINK.$(TOOLSET)) -shared $(GYP_LDFLAGS) $(LDFLAGS.$(TOOLSET)) -Wl,-soname=$(@F) -o $@ -Wl,--whole-archive $(LD_INPUTS) -Wl,--no-whole-archive $(LIBS)
15:43:25  * AvianFluquit (Remote host closed the connection)
15:44:51  <isaacs>indutny: yeah, we're missing an alt attrib. i think you have commit access to that file :0
15:45:03  <indutny>haha
15:45:09  <indutny>you want my opinion?
15:45:13  <isaacs>indutny: pleaes :)
15:45:13  <indutny>fuck it :)
15:45:16  <isaacs>yes.
15:45:21  <isaacs>html validation is for losers.
15:45:30  <isaacs>in fact, we should put some MORE invalid shit in there, just for kicks.
15:45:34  <piscisaureus_>it's valid html6
15:45:37  <piscisaureus_>that's all that matters
15:45:42  <isaacs>html9000!
15:45:50  <isaacs>it's actually valid html5, even
15:46:04  <isaacs>it's just that the validator's an asshat
15:48:02  <indutny>it's not good for SEO
15:48:11  <indutny>google can't get alt name for that image
15:48:18  <isaacs>hahaha
15:48:18  <indutny>and people are unable to search us
15:48:27  <isaacs>the mighty google! thwarted by invalid html!
15:48:30  <isaacs>i feel so powerful!
15:48:38  <indutny>actually
15:48:46  <indutny>considering that it is <img src="http://nodejs.org/images/ryan-speaker.jpg">
15:48:53  <indutny>noone will know that node was created by ryah
15:49:04  <indutny>at least from search results
15:50:59  <CoverSlide>node was made by magical elves
15:54:02  <piscisaureus_>GROM
15:54:42  <indutny>CoverSlide: indeed
15:57:18  * mikealquit (Quit: Leaving.)
16:01:39  <CoverSlide>oh wow. i just realized how rarely i ever use twitter
16:03:10  * mjr_joined
16:04:10  * mmalecki[brb]changed nick to mmalecki
16:06:58  * kazuponjoined
16:11:56  * loladirojoined
16:25:54  * pooyajoined
16:26:27  <piscisaureus_>guys
16:26:51  <piscisaureus_>what do you think of the uv_custom api I proposed here? https://groups.google.com/d/msg/libuv/IO06T7I25Rc/tX2Llz9yuSQJ
16:27:00  <piscisaureus_>indutny? bnoordhuis?
16:27:07  <piscisaureus_>everyone?
16:31:25  <saghul>piscisaureus_ I do like the idea
16:32:31  <saghul>but uv_expect looks a bit weird (as a name) to me
16:33:14  <indutny>piscisaureus_: not sure Ifollow
16:35:23  <piscisaureus_>saghul: yeah I agree about uv_expect
16:36:09  * kazuponquit (Remote host closed the connection)
16:38:57  * sblomjoined
16:39:58  * kazuponjoined
16:40:27  <piscisaureus_>indutny: it's about inserting a custom callback into the libuv callback queue
16:41:29  <indutny>ok
16:41:47  <indutny>I should really take in more careful
16:41:53  <indutny>but I'm so excited about lir allocator
16:41:57  <indutny>ttyl
16:43:49  <isaacs>bnoordhuis: we already have a eio shim layer, don't we?
16:43:52  <isaacs>bnoordhuis: or was that only for ev?
16:44:17  * isaacstoo lazy to rtfs
16:44:30  <isaacs>ah, yes, src/eio-emul.h
16:44:31  <isaacs>nvm
16:44:33  <isaacs>bnoordhuis: ^
16:48:04  * mikealjoined
16:49:03  * loladiroquit (Quit: loladiro)
16:49:28  * mikealquit (Client Quit)
16:54:32  * Ralt_joined
16:54:52  * Ralt_quit (Read error: Connection reset by peer)
16:55:12  * Ralt_joined
16:55:33  * Ralt_quit (Read error: Connection reset by peer)
16:55:47  * Ralt_joined
16:57:02  <saghul>http://www.rasterbar.com/libtorrent_blog/2012/10/asynchronous-disk-io/
16:58:05  * loladirojoined
17:01:30  * TooTallNatejoined
17:06:03  * loladiroquit (Quit: loladiro)
17:09:52  * pooyaquit (Quit: pooya)
17:13:17  <piscisaureus_>heh
17:13:19  <piscisaureus_>so lame
17:13:40  <piscisaureus_>and it doesn't even mention that iocp actually doesnt work reliably unless you O_DIRECT on windows
17:16:19  <piscisaureus_>I should probably check it again
17:16:33  <piscisaureus_>although I don't have any spinning platters anymore so it's gonna be hard to verify
17:25:15  * loladirojoined
17:27:49  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
17:28:01  * `3rdEdenjoined
17:28:12  * ericktjoined
17:28:39  * kazuponquit (Remote host closed the connection)
17:30:27  * loladiroquit (Ping timeout: 276 seconds)
17:32:11  * tomshredsjoined
17:33:03  * paddybyersquit (Ping timeout: 244 seconds)
17:34:25  * AndreasMadsenjoined
17:35:27  * AvianFlujoined
17:36:41  * lohkeyjoined
17:46:50  * c4miloquit (Remote host closed the connection)
17:47:53  * V1joined
17:51:11  * `3rdEdenquit (Ping timeout: 245 seconds)
17:51:21  * joshthecoderjoined
17:52:06  * AvianFluquit (Remote host closed the connection)
17:55:42  * V1quit (Remote host closed the connection)
17:59:28  * loladirojoined
18:00:03  * AvianFlujoined
18:01:43  * loladiroquit (Client Quit)
18:01:43  * AvianFluquit (Read error: Connection reset by peer)
18:01:49  * AvianFlu_joined
18:03:37  * sblomquit (Ping timeout: 240 seconds)
18:03:51  * AvianFlu_changed nick to AvianFlu
18:08:48  * benoitcquit (Excess Flood)
18:08:57  * AvianFluquit (Read error: Connection reset by peer)
18:11:35  * AvianFlujoined
18:13:06  * loladirojoined
18:14:08  * sblomjoined
18:15:07  * brsonjoined
18:16:06  * benoitcjoined
18:17:13  * `3rdEdenjoined
18:17:13  * AvianFluquit (Read error: Connection reset by peer)
18:17:20  * AvianFlu_joined
18:28:30  * c4milojoined
18:28:59  * kazuponjoined
18:29:40  * AvianFlu_quit (Read error: Connection reset by peer)
18:29:53  * AvianFlujoined
18:35:36  * `3rdEdenquit (Remote host closed the connection)
18:37:08  * kazuponquit (Ping timeout: 248 seconds)
18:37:09  * paddybyersjoined
18:37:45  * AvianFluquit (Ping timeout: 264 seconds)
18:38:03  * brsonquit (Quit: leaving)
18:38:21  * brsonjoined
18:40:24  * `3rdEdenjoined
18:40:40  * AvianFlujoined
18:42:06  * V1joined
18:45:07  * sh1mmerjoined
18:45:21  * `3rdEdenquit (Ping timeout: 245 seconds)
18:45:33  * `3rdEdenjoined
18:47:31  * `3rdEden_joined
18:48:27  * V1quit (Ping timeout: 276 seconds)
18:48:30  * `3rdEde__joined
18:51:06  * loladiroquit (Quit: loladiro)
18:51:11  * `3rdEdenquit (Ping timeout: 245 seconds)
18:52:07  * `3rdEden_quit (Ping timeout: 256 seconds)
18:52:18  * `3rdEdenjoined
18:54:12  * benoitcquit (Excess Flood)
18:55:21  * `3rdEde__quit (Ping timeout: 245 seconds)
18:55:59  * V1joined
18:59:06  * `3rdEdenquit (Ping timeout: 245 seconds)
18:59:09  * `3rdEden_joined
18:59:49  * loladirojoined
19:00:36  * benoitcjoined
19:01:10  * V1quit (Ping timeout: 260 seconds)
19:02:39  * AndreasMadsenquit (Remote host closed the connection)
19:04:01  * kazuponjoined
19:06:00  * loladiroquit (Quit: loladiro)
19:07:01  * paddybyersquit (Ping timeout: 245 seconds)
19:12:28  * kazuponquit (Ping timeout: 255 seconds)
19:27:25  * paddybyersjoined
19:38:57  * kazuponjoined
19:39:26  * mmaleckiquit (Quit: leaving)
19:42:17  * sgallaghquit (Remote host closed the connection)
19:45:46  * `3rdEden_quit (Remote host closed the connection)
19:47:09  * kazuponquit (Ping timeout: 240 seconds)
19:58:55  * pooyajoined
20:05:29  * loladirojoined
20:08:18  * loladiroquit (Client Quit)
20:14:04  * kazuponjoined
20:19:16  * loladirojoined
20:19:24  * loladiroquit (Client Quit)
20:22:46  * kazuponquit (Ping timeout: 272 seconds)
20:31:05  * sblomquit (Ping timeout: 245 seconds)
20:32:46  * c4miloquit (Remote host closed the connection)
20:43:43  * paddybyersquit (Ping timeout: 252 seconds)
20:46:59  * hzjoined
20:48:48  * txdvquit (Read error: Connection reset by peer)
20:48:56  * txdvjoined
20:49:07  * kazuponjoined
20:56:09  * `3rdEdenjoined
20:57:21  * kazuponquit (Ping timeout: 245 seconds)
21:00:17  * `3rdEdenquit (Ping timeout: 240 seconds)
21:04:01  * mikealjoined
21:15:56  * AvianFluquit (Remote host closed the connection)
21:20:49  * rendarquit
21:24:17  * kazuponjoined
21:32:30  * kazuponquit (Ping timeout: 256 seconds)
21:32:36  * c4milojoined
21:36:40  * c4miloquit (Remote host closed the connection)
21:45:27  * c4milojoined
21:46:52  * c4miloquit (Remote host closed the connection)
21:55:15  * AvianFlujoined
21:57:19  * mmaleckijoined
21:59:30  * kazuponjoined
22:08:01  * kazuponquit (Ping timeout: 265 seconds)
22:10:07  * sblomjoined
22:13:20  * paddybyersjoined
22:20:15  * hzquit
22:21:11  * paddybyersquit (Read error: Connection reset by peer)
22:31:08  <bnoordhuis>TooTallNate: how do i get node-gyp to build a -m32 .node?
22:31:21  <TooTallNate>bnoordhuis: --arch=ia32
22:31:21  <kohai>bnoordhuis has 24 beers
22:31:29  <TooTallNate>haha, whoop
22:31:32  <TooTallNate>bnoordhuis++
22:31:33  <kohai>bnoordhuis has 25 beers
22:31:41  <bnoordhuis>haha
22:31:43  <bnoordhuis>thanks nathan
22:31:51  <TooTallNate>np
22:34:38  * kazuponjoined
22:34:40  <bnoordhuis>/build/buildd/gdb-7.4-2012.04/gdb/regcache.c:1000: internal-error: regcache_raw_supply: Assertion `regnum >= 0 && regnum < regcache->descr->nr_raw_registers' failed.
22:34:40  <sblom>++kohai
22:34:42  <bnoordhuis>A problem internal to GDB has been detected,
22:34:44  * bnoordhuissighs
22:37:40  * TheJHjoined
22:38:16  * TooTallNatequit (Ping timeout: 245 seconds)
22:40:32  * TooTallNatejoined
22:43:09  * kazuponquit (Ping timeout: 264 seconds)
22:54:40  * pooyaquit (Quit: pooya)
22:57:42  * loladirojoined
22:57:50  * loladiroquit (Client Quit)
23:09:40  * kazuponjoined
23:14:17  <tjfontaine>bnoordhuis: impressive
23:15:11  <bnoordhuis>tjfontaine: i seem to have a knack for finding gdb bugs, unfortunately...
23:17:18  <tjfontaine>bnoordhuis: how reliably can you reproduce that?
23:17:32  <bnoordhuis>tjfontaine: 100% with this one core dump
23:17:59  <tjfontaine>excellent
23:18:05  * kazuponquit (Ping timeout: 252 seconds)
23:18:25  <bnoordhuis>tjfontaine: you're gonna ask me to open a bug report?
23:18:27  <tjfontaine>how much time are you going to spend trying to figure out why vs creating a new core?
23:18:33  <tjfontaine>bnoordhuis: quite the contrary :)
23:18:52  <tjfontaine>bnoordhuis: if you were on a mac I'd have you try lldb though
23:20:08  <bnoordhuis>tjfontaine: well... it i wasn't trying to fix something else, i'd probably spend a couple of hours in `gdb gdb`
23:20:47  <tjfontaine>bnoordhuis: that's a rabbit hole I would be afraid to fall into
23:22:18  <bnoordhuis>tjfontaine: rightfully so. it's usually not a very rewarding rabbit hole either. still... it's like a scab, you know?
23:22:36  <tjfontaine>heh indeed
23:24:19  * lohkeypart
23:25:53  <bnoordhuis>tjfontaine: that libtorrent blog post was interesting btw
23:26:30  <bnoordhuis>nothing quite new but it's some consolance knowing we're not the only one struggling with aio :)
23:26:40  <tjfontaine>bnoordhuis: that one didn't come from me though, but I'll take the credit
23:27:12  <bnoordhuis>oh right, sorry - it was saghul
23:38:23  <tjfontaine>hmm the article asserts that hfs+ doesn't support sparse, but I'm pretty sure you can make dmgs that are sparse
23:38:59  <tjfontaine>ah different type of file
23:44:38  * kazuponjoined
23:47:39  * sblomquit (Ping timeout: 260 seconds)
23:50:07  * sblomjoined
23:52:48  * kazuponquit (Ping timeout: 252 seconds)
23:55:36  * AvianFluquit (Remote host closed the connection)
23:56:09  * ericktquit (Ping timeout: 240 seconds)