00:02:07  <piscisaureus_>hmm
00:02:36  <piscisaureus_>stripe level 3 === dead
00:02:56  <piscisaureus_>back to zombo.com then
00:05:20  * mikealquit (Quit: Leaving.)
00:14:50  * brsonquit (Ping timeout: 268 seconds)
00:16:44  <tjfontaine>welcome to zombo com
00:17:41  <piscisaureus_>actually
00:17:47  <piscisaureus_>I'm going to zombo now
00:17:49  <piscisaureus_>later guys
00:17:52  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
00:24:47  * mikealjoined
00:26:45  * brsonjoined
00:49:21  * erickt_joined
00:51:00  * ericktquit (Ping timeout: 244 seconds)
00:54:11  * erickt_quit (Ping timeout: 260 seconds)
00:54:33  * dapquit (Quit: Leaving.)
01:00:44  * brsonquit (Quit: leaving)
01:00:59  * brsonjoined
01:13:33  * dshaw_joined
01:18:51  * joshthecoderquit (Quit: Leaving...)
01:26:16  * blackorzarquit (Ping timeout: 260 seconds)
01:33:24  * mikealquit (Quit: Leaving.)
01:34:43  * brsonquit (Ping timeout: 245 seconds)
01:37:48  * mikealjoined
01:39:10  * mikealquit (Client Quit)
01:41:51  * dshaw_quit (Quit: Leaving.)
01:42:36  * TooTallNatequit (Quit: ["Textual IRC Client: www.textualapp.com"])
01:53:37  * joshthecoderjoined
01:55:03  * ericktjoined
02:22:37  * lohkeyquit (Ping timeout: 260 seconds)
02:36:16  * abraxasjoined
03:07:30  * loladirojoined
03:23:54  * AvianFlujoined
03:54:25  * brsonjoined
03:58:55  * loladiroquit (Quit: loladiro)
04:00:05  * ericktquit (Quit: erickt)
04:26:02  * abraxas_joined
04:29:53  * abraxasquit (Ping timeout: 265 seconds)
04:34:11  * blackorzarjoined
04:57:59  * brsonquit (Quit: leaving)
05:03:04  * abraxas_quit (*.net *.split)
05:11:57  * perezdjoined
05:22:52  * perezdquit (Quit: perezd)
05:34:28  * dshaw_joined
06:05:59  * blackorzar_joined
06:06:51  * blackorzarquit (Ping timeout: 256 seconds)
06:09:32  * blackorzarjoined
06:14:03  * blackorzar_quit (Ping timeout: 272 seconds)
06:52:09  * dshaw_quit (Quit: Leaving.)
06:56:12  * paddybyersjoined
07:01:04  * joshthecoderquit (Quit: Linkinus - http://linkinus.com)
07:01:58  * AndreasMadsenjoined
07:15:10  * abraxas_joined
07:15:13  * filirom1joined
07:15:32  * filirom1quit (Changing host)
07:15:32  * filirom1joined
07:27:41  * AvianFluquit (Quit: AvianFlu)
07:56:42  <CIA-131>v8: [email protected] * r12356 /trunk/ (102 files in 11 dirs): Version 3.13.3
07:56:42  <CIA-131>v8: [email protected] * r12357 /tags/3.13.3/trunk: Tagging version 3.13.3
07:56:44  <CIA-131>v8: [email protected] * r12358 /branches/bleeding_edge/src/mips/lithium-codegen-mips.cc:
07:56:44  <CIA-131>v8: MIPS: Fix write barrier for StoreKeyedFastElements on ARM.
07:56:44  <CIA-131>v8: Port r12238 (300f3dac)
07:56:44  <CIA-131>v8: BUG=
07:56:44  <CIA-131>v8: TEST=
07:56:44  <CIA-131>v8: Review URL: https://chromiumcodereview.appspot.com/10828216
07:56:45  <CIA-131>v8: Patch from Akos Palfi <[email protected]>.
07:56:47  <CIA-131>v8: [email protected] * r12359 /branches/3.11/ (test/cctest/test-api.cc src/version.cc src/ic.cc):
07:56:48  <CIA-131>v8: Back port r12247 to 3.11 (check holder, not receiver for dict mode).
07:56:48  <CIA-131>v8: This also back ports the improved tests from r12264, r12271 and r12311.
07:56:48  <CIA-131>v8: Review URL: https://chromiumcodereview.appspot.com/10860062
07:56:49  <CIA-131>v8: [email protected] * r12360 /tags/3.11.10.20: Tagging 3.11.10.20
07:56:51  <CIA-131>v8: [email protected] * r12361 /branches/bleeding_edge/src/ (globals.h log.cc):
07:56:51  <CIA-131>v8: Introduce DISABLE_ASAN macro to disable AddressSanitizer instrumentation for given functions.
07:56:51  <CIA-131>v8: Disable AddressSanitizer instrumentation for StackTracer::Trace.
07:56:51  <CIA-131>v8: [email protected]
07:56:51  <CIA-131>v8: BUG=chromium:134722
07:56:58  * mikealjoined
07:57:22  <CIA-131>v8: While finding the valueOf descriptor, only look at the key entries.
07:57:22  <CIA-131>v8: Review URL: https://chromiumcodereview.appspot.com/10870012
07:57:22  <CIA-131>v8: [email protected] * r12367 /branches/bleeding_edge/ (28 files in 5 dirs): (log message trimmed)
07:57:22  <CIA-131>v8: Allow uint32 value on optimized frames if they are consumed by safe operations.
07:57:22  <CIA-131>v8: Safe operations are those that either do not observe unsignedness or have special support for uint32 values:
07:57:23  <CIA-131>v8: - all binary bitwise operations: they perform ToInt32 on inputs;
07:57:23  <CIA-131>v8: - >> and << shifts: they perform ToInt32 on left hand side and ToUint32 on right hand side;
07:57:24  <CIA-131>v8: - >>> shift: it performs ToUint32 on both inputs;
07:57:24  <CIA-131>v8: - stores to integer external arrays (not pixel, float or double ones): these stores are "bitwise";
07:57:25  <CIA-131>v8: [email protected] * r12368 /branches/bleeding_edge/src/deoptimizer.cc:
07:57:26  <CIA-131>v8: Fix order of conversions in ObjectToInt32 and ObjectToUint32 helpers.
07:57:26  <CIA-131>v8: [email protected]
07:57:27  <CIA-131>v8: BUG=
07:57:27  <CIA-131>v8: Review URL: https://chromiumcodereview.appspot.com/10878012
07:57:28  <CIA-131>v8: [email protected] * r12369 /branches/bleeding_edge/src/hydrogen.cc:
07:57:28  <CIA-131>v8: When processing collected uint32 instructions skip those with non-Integer32 representation.
08:04:54  * AndreasMadsenquit (Remote host closed the connection)
08:07:49  * `3rdEdenjoined
08:41:41  * ArmyOfBrucequit (Excess Flood)
08:42:12  * ArmyOfBrucejoined
08:47:37  * zippquit (Remote host closed the connection)
09:08:19  * pooyajoined
09:15:47  <CIA-131>v8: [email protected] * r12376 /branches/bleeding_edge/src/mips/ (lithium-codegen-mips.cc lithium-codegen-mips.h):
09:15:47  <CIA-131>v8: MIPS: Check that index and length are Smi in bounds check.
09:15:47  <CIA-131>v8: Port r12362 (cd39337c)
09:15:47  <CIA-131>v8: BUG=
09:15:47  <CIA-131>v8: TEST=
09:15:48  <CIA-131>v8: Review URL: https://chromiumcodereview.appspot.com/10876053
09:15:48  <CIA-131>v8: Patch from Akos Palfi <[email protected]>.
09:28:24  * pooyaquit (Quit: pooya)
09:37:01  * bnoordhuis_joined
09:40:20  * mmaleckijoined
09:44:44  * mmaleckiquit (Ping timeout: 240 seconds)
09:53:00  * mmaleckijoined
09:53:42  * mmalecki_joined
09:56:46  * blackorzarquit (Ping timeout: 252 seconds)
10:21:31  * stagasjoined
10:22:12  * rendarjoined
10:24:49  * loladirojoined
10:58:40  <indutny>em...
10:58:44  <indutny>what's happening there?
11:01:01  <deoxxa>hot javascript on javascript action
11:03:41  * piscisaureus_joined
11:29:48  * mmalecki_quit (Quit: leaving)
11:44:52  * abraxas_quit (Remote host closed the connection)
11:46:54  * loladiroquit (Quit: loladiro)
11:51:23  * loladirojoined
12:00:16  * hzjoined
12:08:31  * loladiroquit (Quit: loladiro)
12:15:26  <CIA-131>v8: [email protected] * r12377 /trunk/ (97 files in 10 dirs):
12:15:26  <CIA-131>v8: Revert to code state of 3.13.1 plus r12350
12:15:26  <CIA-131>v8: [email protected]
12:15:26  <CIA-131>v8: Review URL: https://chromiumcodereview.appspot.com/10878047
12:15:27  <CIA-131>v8: [email protected] * r12378 /tags/3.13.3.1: Tagging version 3.13.3.1
12:43:28  * stagasquit (*.net *.split)
12:53:03  * AvianFlujoined
13:08:27  <CIA-131>v8: [email protected] * r12379 /branches/bleeding_edge/ (src/api.cc include/v8.h):
13:08:27  <CIA-131>v8: Add check to external string resource.
13:08:27  <CIA-131>v8: Extensions are a source of external strings where we don't perform check yet.
13:08:27  <CIA-131>v8: [email protected]
13:08:27  <CIA-131>v8: BUG=
13:08:28  <CIA-131>v8: Review URL: https://chromiumcodereview.appspot.com/10878025
13:18:25  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
13:24:19  * mmaleckiquit (Ping timeout: 256 seconds)
13:34:55  * wavded_quit (Quit: Connection closed for inactivity)
13:37:56  * piscisaureus_joined
13:39:17  <piscisaureus_>bnoordhuis_: hey, do you know if shigeki hangs out on irc?
13:42:28  * `3rdEdenchanged nick to `3E|Shop
13:48:25  * loladirojoined
13:50:02  <CIA-131>node: Brandon Wilson v0.8 * r5901144 / doc/pipe.css : doc: more uniform and sleek buttons - http://git.io/NGvXew
13:52:42  * c4milojoined
13:57:08  <bnoordhuis_>piscisaureus_: he doesn't afaik
13:57:22  <bnoordhuis_>hey, we're tracking v8 commits now?
13:59:02  <piscisaureus_>I don't know
13:59:10  <piscisaureus_>I don't mind but CIA is too noisy for this
13:59:24  <piscisaureus_>it would be okay if it displayed one line per commit
14:10:08  <tjfontaine>it looks like in advanced filtering you can convince cia to trim the lines
14:10:35  * bradleymeckjoined
14:13:10  <tjfontaine>http://cia.vc/doc/rulesets/ and linelimit of 2 or 3 might suffice
14:14:12  * blackorzarjoined
14:15:32  * mmaleckijoined
14:15:53  * blackorzar_joined
14:19:02  * blackorzarquit (Ping timeout: 264 seconds)
14:26:53  <piscisaureus_>y a
14:31:24  <tjfontaine>ryah has control, so it's up to him
14:40:12  * mmaleckiquit (Ping timeout: 276 seconds)
14:44:07  <piscisaureus_>wpoll_t* wpoll_create();
14:44:07  <piscisaureus_>int wpoll_close();
14:44:07  <piscisaureus_>int wpoll_ctl(wpoll_t wpoll_handle, int op, SOCKET sock, struct epoll_event *event);
14:44:08  <piscisaureus_>int epoll_wait(wpoll_t wpoll_handle, struct epoll_event *events, int maxevents, int timeout);
15:07:22  * ericktjoined
15:17:45  * ericktquit (Ping timeout: 260 seconds)
15:22:01  <CIA-131>node: koichik v0.8 * r752ac32 / (3 files in 3 dirs): https: make https.get() accept a URL - http://git.io/4ysFpA
15:23:22  * mmaleckijoined
15:29:19  <CIA-131>node: Brian White master * rc78678b / (lib/http.js test/simple/test-http-client-parse-error.js): http: bubble up parser errors to ClientRequest - http://git.io/QbdGhw
15:35:33  * ArmyOfBrucequit (Excess Flood)
15:36:04  * ArmyOfBrucejoined
15:41:05  <ryah>?
15:41:23  <ryah>tjfontaine: CIA?
15:41:45  <tjfontaine>yes, did you add v8 to #libuv's cia settings?
15:43:21  * ericktjoined
15:44:14  <ryah>hm...
15:44:20  <ryah>i dont think so
15:45:06  <ryah>oh it was
15:45:08  <ryah>i removed it
15:45:23  * mmaleckiquit (Ping timeout: 268 seconds)
15:45:28  <tjfontaine>ok, we were just discussing being able to track v8 while also making it less verbose
15:46:18  * ericktquit (Client Quit)
15:47:55  * ericktjoined
15:51:27  * loladiroquit (Quit: loladiro)
15:54:41  * `3E|Shopchanged nick to `3E
16:08:12  * bradleymeckquit (Quit: bradleymeck)
16:11:59  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
16:16:36  * dapjoined
16:17:45  * piscisaureus_joined
16:18:07  * piscisaureus_quit (Client Quit)
16:18:38  * piscisaureus_joined
16:23:09  * piscisaureus_quit (Ping timeout: 248 seconds)
16:27:38  * filirom1quit (Quit: Page closed)
16:28:24  * ArmyOfBrucequit (Excess Flood)
16:28:52  * ArmyOfBrucejoined
16:31:01  * ericktquit (Quit: erickt)
16:58:40  * ericktjoined
17:06:46  * TooTallNatejoined
17:19:47  * arlolrajoined
17:27:00  * dshaw_joined
17:44:07  * stagasjoined
17:55:29  * brsonjoined
17:55:49  * dshaw_quit (Quit: Leaving.)
17:55:59  * TooTallNatequit (Read error: Operation timed out)
17:56:28  * bradleymeckjoined
17:58:57  * brsonquit (Read error: Connection reset by peer)
17:58:58  * TooTallNatejoined
17:59:45  * brsonjoined
18:02:13  * joshthecoderjoined
18:03:41  * dshaw_joined
18:12:59  * mikealquit (Quit: Leaving.)
18:13:17  * lohkeyjoined
18:24:54  * mscdexjoined
18:27:04  <mscdex>ok, so i'm back with the same libuv problem (i think) that i had earlier this month, different binding though
18:27:07  <mscdex>:-)
18:27:49  <mscdex>this time i've compiled debug builds and get an assertion error in libuv
18:29:12  <mscdex>consistently after about 10 seconds on Windows, i get this assertion: Assertion failed: 0, file c:\node\deps\uv\src\win\req-inl.h, line 220
18:31:08  <tjfontaine>what did you do to that poor request
18:31:30  <mscdex>it's not a request ;-)
18:31:36  <tjfontaine>clearly.
18:31:37  <mscdex>well, i guess to uv it is
18:32:04  <mscdex>but it's just an idling connected socket
18:34:37  * AvianFluquit (Quit: AvianFlu)
18:36:21  <tjfontaine>mscdex: I guess it would be helpful to know something about the req it's switching on there, at the very least what it says the type is
18:36:37  <mscdex>the type is some large negative number
18:36:45  * bradleymeck_joined
18:36:52  <mscdex>i think there is something fishy with uninitialized pointers or something
18:37:29  <tjfontaine>is there code available for this?
18:38:59  * arlolraquit (Quit: Linkinus - http://linkinus.com)
18:40:40  * bradleymeckquit (Ping timeout: 245 seconds)
18:40:40  * bradleymeck_changed nick to bradleymeck
18:46:31  <mscdex>there is, but the debug build of the library i depend on for Windows isn't up there
18:47:00  <tjfontaine>this is the maria stuff?
18:47:05  <mscdex>yep
18:49:15  * mikealjoined
18:50:27  <mscdex>i have what i believe was the same issue with a libssh2 binding i was writing earlier this month
18:50:52  <tjfontaine>and the solution there was?
18:51:03  <mscdex>i scrapped the binding and just started doing it pure js ;-)
18:51:16  <tjfontaine>heh
18:51:50  <tjfontaine>does the error exhibit itself on linux?
18:52:57  <mscdex>haven't tried yet
18:55:06  <mscdex>segfault on linux :-\
18:55:16  <mscdex>still about the same amount of time though, 10 seconds
18:59:05  * AvianFlujoined
19:08:31  * AvianFluquit (Quit: AvianFlu)
19:22:20  * hzquit
19:34:10  * brsonquit (Ping timeout: 268 seconds)
19:34:48  <mscdex>ok, i fixed the segfault, but it's exiting without any errors now
19:34:51  <mscdex>:S
19:35:39  * AvianFlujoined
19:47:59  * `3Equit (Quit: worked to hard, brains === tired)
20:05:40  * EhevuTovjoined
20:07:52  <bnoordhuis_>mscdex: what's the issue?
20:08:50  <mscdex>still trying to find out the root cause
20:09:44  <mscdex>with the gc idle notification enabled, my script lasts 10 seconds, without it enabled it seems to last much longer but i haven't tested to see just how long
20:10:50  <bnoordhuis_>mscdex: is the offending handle in any way tied to a js object?
20:11:19  <mscdex>yes, a poll handle
20:11:52  <bnoordhuis_>mscdex: how are you storing / associating it? is there code somewhere?
20:14:49  <mscdex>bnoordhuis_: library is here: https://github.com/mscdex/node-mariasql test code is here: https://gist.github.com/186202a3fc32d4ff6f6c
20:15:01  <mscdex>bnoordhuis_: connection to a mysql server required
20:15:14  <mscdex>i'm just storing the handle in the instance
20:16:45  <mscdex>i can see the object get gc'ed after 10 seconds by looking at the bt, but still trying to find out why
20:24:24  <bnoordhuis_>mscdex: i assume the relevant code is in src/client.cc?
20:24:30  <mscdex>yep
20:25:26  <bnoordhuis_>mscdex: it's probably a use after free bug
20:25:51  <bnoordhuis_>it works like this
20:25:58  <bnoordhuis_>you new Client(), then delete it
20:26:09  <bnoordhuis_>~Client() calls close() which calls uv_close()
20:26:24  <bnoordhuis_>uv_close() takes at least one event loop tick to complete
20:26:38  <bnoordhuis_>but by that time the memory that contains the poll handle has been deallocated
20:26:45  <mscdex>yeah, i fixed that part
20:26:54  <bnoordhuis_>did you push it?
20:26:59  <mscdex>yeah
20:27:08  <mscdex>i guess what i don't quite understand is why the Client instance is being gc'ed
20:27:57  <bnoordhuis_>mscdex: the bug is still there
20:28:29  <bnoordhuis_>here's how you fix it: make the poll_handle field a uv_poll_t* and new it
20:28:51  <bnoordhuis_>in your destructor call uv_close((uv_handle_t*)poll_handle, delete_cb)
20:28:59  <mscdex>oops, i did forget to push to github
20:29:10  <bnoordhuis_>where delete_cb does a delete reinterpret_cast<uv_poll_t*>(handle)
20:29:34  <bnoordhuis_>or cheat and malloc it, then call uv_close((uv_handle_t*)poll_handle, (uv_close_cb)free)
20:42:28  * piscisaureus_joined
20:43:59  <piscisaureus_>uv_close(handle, (uv_close_cb) free) <--bnoordhuis: nice one!
20:44:58  * loladirojoined
20:45:27  * brsonjoined
20:46:07  * mmaleckijoined
20:46:30  * saghulquit (Ping timeout: 245 seconds)
20:47:18  <bnoordhuis_>piscisaureus_: nasty, isn't it? :)
20:47:29  <piscisaureus_>bnoordhuis_: I like it :-)
20:47:42  <piscisaureus_>bnoordhuis_: if you want to do the poor man's solution, it's very easy :-0
20:47:57  * saghuljoined
20:55:36  * pooyajoined
20:56:05  <pooya>hi! it's me again.
20:56:16  <pooya>is there anyway to flush queue_work before exit?
20:58:57  <bnoordhuis_>pooya: no. you need to keep track of if all requests came back
20:59:09  <pooya>and use AtExt?
20:59:12  <pooya>AtExit
20:59:31  <CIA-131>node: Ben Noordhuis master * rc4fa77a / tools/js2c.py : tools: fix missing initializer warning in js2c.py - http://git.io/y0t92g
20:59:32  <bnoordhuis_>pooya: how do you mean?
21:00:02  <pooya>i mean adding a blocking callback to AtExit making blocking until all requests comeback before exiting?
21:00:18  <pooya>s/making blocking/blocking it/
21:00:37  <bnoordhuis_>pooya: oh, like that. in node, the event loop doesn't exit until all handles have been closed and all requests have finished
21:00:58  <bnoordhuis_>the keeping track thing is only relevant if you call exit()
21:01:04  <pooya>but not on process.exit, right?
21:01:18  <pooya>this for node-syslog module
21:01:58  <bnoordhuis_>pooya: process.exit() exits immediately, it doesn't run the AtExit handlers
21:01:58  <pooya>let say the app logs an error and does process.exit(1)
21:02:07  <pooya>ah
21:02:11  <pooya>good point
21:02:20  <pooya>maybe i should intact make the whole syslog thing blocking...
21:02:42  <bnoordhuis_>what's the worst that could happen if the request doesn't finish?
21:03:14  <pooya>well, i guess people expect all their console.log (or syslog message here) to appear, no matter what
21:03:16  <mscdex>bnoordhuis_: i made that modification locally, but it seems it's not the poll_handle that it is segfaulting on, it's the objectwrap handle_ i think
21:03:40  <bnoordhuis_>mscdex: what does the backtrace look like?
21:03:46  <bnoordhuis_>pooya: yeah. no way around that i guess
21:03:48  <mscdex>v8::Object::Get(v8::Handle<v8::Value>) ()
21:03:53  <mscdex>that's the top of the bt
21:03:54  <pooya>bnoordhuis_: how fast do we expect a method call be in node?
21:04:01  <mscdex>inside my close callback
21:04:10  * bradleymeckquit (Quit: bradleymeck)
21:04:24  <bnoordhuis_>pooya: instantaneous, i.e. it shouldn't block
21:04:32  <bnoordhuis_>mscdex: can you gist the full bt?
21:05:00  <mscdex>https://gist.github.com/acdecb42c61709c0bbd9
21:06:11  <mscdex>line 133 is the one that points to: Local<Function> Emit = Local<Function>::Cast(obj->handle_->Get(emit_symbol));
21:07:06  <bnoordhuis_>no doubt the handle's been gc'ed
21:07:14  <mscdex>right, but how can i check that?
21:07:26  <bnoordhuis_>which kind of makes sense because close() is called from the destructor
21:07:33  <bnoordhuis_>and close() calls uv_close()
21:07:56  <bnoordhuis_>i bet you a sixpack that valgrind complains
21:08:10  <pooya>bnoordhuis_: no rush, but take a look at this too. this broke all other syslog modules… https://github.com/bnoordhuis/node-unix-dgram/issues/6
21:08:28  <piscisaureus_>bnoordhuis_: it is a known problem... on windows (where child stdio pipes are nonblocking) writes don't always flush when process.exit() is called
21:09:37  <bnoordhuis_>pooya: what version of node?
21:10:41  <pooya>0.8.6
21:10:47  <pooya>should i try newer?
21:11:04  <bnoordhuis_>no, just wanted to make sure i'm testing it against the same version
21:11:10  <bnoordhuis_>test passes for me on linux btw
21:12:34  * ericktquit (Ping timeout: 240 seconds)
21:12:44  <pooya>oh, interesting
21:13:01  * brsonquit (Ping timeout: 252 seconds)
21:13:46  <pooya>bnoordhuis_: do you want me to test anything on mac?
21:14:51  <bnoordhuis_>pooya: i'm trying to deduce why you get a 20 byte slowbuffer
21:14:57  <bnoordhuis_>is it always a 20 byte slowbuffer?
21:16:32  <pooya>yeah, running the test it is always 20
21:16:50  <bnoordhuis_>pooya: can you gist the output of `dtruss node test/test-unix-dgram.js`?
21:16:53  <pooya>although the dump changes sometimes
21:17:14  <pooya>but i guess that's my memory dump
21:17:34  <bnoordhuis_>yeah, the buffer is allocated on the heap, it contains random garbage
21:18:59  <mscdex>ok, so i have a fix in place (pushed to gh) that keeps it from segfaulting. i still have the problem with the js object being gc'ed
21:20:55  <bnoordhuis_>mscdex: Ref() the ObjectWrap
21:21:21  * brsonjoined
21:21:41  <piscisaureus_>@bos31337: Lance Armstrong takes drugs and wins the hardest bike race in the world 7 times. Programmers take drugs and what do we get? Node.js. <-- hahaha
21:22:02  <pooya>bnoordhuis_: https://gist.github.com/3455746
21:22:58  <bnoordhuis_>haha
21:25:24  <mscdex>nice.. Ref() was it! thanks again! :-)
21:25:54  <bnoordhuis_>mscdex: no problem
21:26:01  * dshaw_quit (Quit: Leaving.)
21:26:02  <bnoordhuis_>pooya: can you try this patch and see what it prints? https://gist.github.com/4dc9cb24d80cf569c040
21:27:37  * dshaw_joined
21:30:10  <TooTallNate>isaacs: bnoordhuis_: do we want to document this Console class or no?
21:30:27  <bnoordhuis_>TooTallNate: i guess not
21:30:41  * gicodequit (Read error: Operation timed out)
21:30:44  <pooya>bnoordhuis_: ioctl(FIONREAD) = 20
21:30:48  <bnoordhuis_>TooTallNate: let me rephrase that, is there any reason to document it?
21:31:00  <bnoordhuis_>pooya: right, i was kind of expecting that
21:31:15  <TooTallNate>bnoordhuis_: it has minimal value to instantiate one in your own code i'd think
21:31:18  <TooTallNate>so i agree
21:31:19  <bnoordhuis_>pooya: so the question is why os x reports that there are 20 bytes to read when there are only 4
21:32:03  <pooya>or could be something wrong with the send code/
21:32:09  <pooya>that sends 20
21:32:48  <CIA-131>node: Nathan Rajlich master * r025f53c / (lib/console.js test/simple/test-console-instance.js): console: refactor the console module to be reusable - http://git.io/6UDsEQ
21:32:49  <CIA-131>node: Nathan Rajlich master * r0285dae / (lib/repl.js test/simple/test-repl-console.js): repl: create a new Console instance for the repl when "useGlobal" is off - http://git.io/oiwhTw
21:34:15  <bnoordhuis_>pooya: possible. what does https://gist.github.com/4a987af5c3ba69951e8a print?
21:37:03  * c4miloquit (Remote host closed the connection)
21:38:21  <pooya>"length=4, iov.iov_len=4" also offset=0 and length=4 and Buffer::length(buf)=4
21:39:08  <bnoordhuis_>well, that's okay then
21:44:12  <pooya>btw ipheader length=20. maybe a coincidence
21:44:49  <pooya>oh no
21:44:52  <pooya>minimum packet length is 20
21:45:16  <pooya>see total length here: http://en.wikipedia.org/wiki/IPv4#Header
21:46:24  <pooya>but then if header was 20, at least in our case it should have been 24 (20header+4 data)
21:46:45  <bnoordhuis_>pooya: does https://gist.github.com/61b7a521d96f20e0e7d6 fix it?
21:48:14  <pooya>bnoordhuis_: it did
21:48:21  <bnoordhuis_>\o/
21:50:27  * dshaw_quit (Quit: Leaving.)
21:52:21  <bnoordhuis_>pooya: test it out for a bit. if it works okay, ping me in a few days and i'll publish a new version to npm
21:53:15  * dshaw_joined
21:53:41  * arlolrajoined
21:54:03  <pooya>bnoordhuis_: thanks
21:54:12  <pooya>bnoordhuis_: will do :)
21:54:16  <bnoordhuis_>cool, thanks
21:57:04  * brsonquit (Ping timeout: 240 seconds)
22:00:57  * brsonjoined
22:06:13  * mikealquit (Quit: Leaving.)
22:08:54  * arlolraquit (Quit: Linkinus - http://linkinus.com)
22:09:02  * bradleymeckjoined
22:09:05  * mikealjoined
22:10:43  * mscdexpart ("Leaving")
22:20:50  * brsonquit (Ping timeout: 264 seconds)
22:26:33  * paddybyersquit (Quit: paddybyers)
22:32:18  * dshaw_quit (Quit: Leaving.)
22:36:21  * paddybyersjoined
22:36:51  * dshaw_joined
22:46:51  * AvianFluquit (Quit: AvianFlu)
22:52:56  <piscisaureus_>http://a8.sphotos.ak.fbcdn.net/hphotos-ak-ash4/298729_398256620229423_142254267_n.jpg
22:59:10  * bradleymeckquit (Quit: bradleymeck)
23:00:46  * hzjoined
23:03:55  * philips_quit (Excess Flood)
23:07:00  * philips_joined
23:10:12  * paddybyersquit (Quit: paddybyers)
23:13:32  <CIA-131>node: Nathan Rajlich v0.8 * r2d0c1da / doc/api/https.markdown : docs: fix syntax error in "https" example - http://git.io/381dFg
23:18:40  <piscisaureus_>bnoordhuis_: hey
23:18:47  <piscisaureus_>bnoordhuis_: when does one get EPOLLHUP?
23:19:01  <piscisaureus_>bnoordhuis_: I can't figure it out from that man page ....
23:28:07  * rendarquit
23:35:59  * hzquit
23:38:16  * dshaw_quit (Quit: Leaving.)
23:38:23  <bnoordhuis_>piscisaureus_: in the same cases when you get POLLHUP from select/poll
23:38:40  * c4milojoined
23:39:02  <piscisaureus_>bnoordhuis_: so that is when you shutdown the socket?
23:39:38  <bnoordhuis_>piscisaureus_: no, when the other end has disconnected
23:39:50  <piscisaureus_>bnoordhuis_: I though that POLLRDHUP was for that
23:40:48  <bnoordhuis_>piscisaureus_: i think that's what you when the other end called shutdown(SHUT_WR)
23:41:29  <piscisaureus_>hmm
23:41:30  <piscisaureus_>too bad
23:41:38  <piscisaureus_>in that case I cant emulate POLLRDHUP on windows
23:41:41  <piscisaureus_>:-p
23:41:46  <piscisaureus_>I was so close :-)
23:43:24  <bnoordhuis_>piscisaureus_: did a quick check, the kernel only sets POLLRDHUP on RCV_SHUTDOWN
23:43:42  <piscisaureus_>right, ok
23:43:50  <piscisaureus_>I suppose it doesn't matter much
23:44:05  <piscisaureus_>the man page says it's very useful for edge triggered but that's not going to work anyway
23:46:41  <bnoordhuis_>piscisaureus_: in case it helps, you get POLLRDHUP once a FIN has been received
23:47:05  <piscisaureus_>bnoordhuis_: yeah, that's what I suspected now
23:47:14  <piscisaureus_>bnoordhuis_: that's not emulatable
23:47:23  <piscisaureus_>whatever, who cares
23:47:34  <piscisaureus_>this is just for educational purposes anyway
23:47:59  <bnoordhuis_>i've never seen it used in real code anyway
23:49:33  <piscisaureus_>bnoordhuis_: while we're at it... when is EPOLLPRI reported?
23:49:41  <piscisaureus_>bnoordhuis_: it's not OOB data right?
23:49:49  <piscisaureus_>bnoordhuis_: does that actually ever happen?
23:50:00  <bnoordhuis_>piscisaureus_: not with tcp / udp
23:50:24  <piscisaureus_>it's weird
23:50:46  <piscisaureus_>windows has separate flags for pollrdnorm and pollrdoob
23:50:55  <piscisaureus_>but apparently linux puts normal and oob data in the same box
23:51:08  <bnoordhuis_>maybe i lied
23:51:16  <piscisaureus_>well
23:51:21  <piscisaureus_>windows also has the POLLPRI flag
23:51:31  <piscisaureus_>but the docs say it doesn't do anything
23:52:01  <bnoordhuis_>i think a tcp packet with the URG flag set may raise POLLPRI
23:52:33  * tomshredsjoined
23:52:53  <piscisaureus_>http://linux.die.net/man/2/poll <-- the flags are different there
23:52:59  <bnoordhuis_>i wonder if that ever happens on today's internet
23:53:13  <piscisaureus_>ah right
23:53:31  <piscisaureus_>the poll man page says that pollpri is used for oob data and POLLRDBAND is "generally not used"
23:55:56  <piscisaureus_>@kraih: Would really like to see libuv support in rakudo #perl6. https://t.co/eEBPetqV <-- Of course, who doesn't
23:58:28  <bnoordhuis_>hah, make it so
23:59:24  <bnoordhuis_>it's something of a worthwhile goal - there are no perl5 / perl6 bindings afaik