--- Log opened Sat Oct 20 00:00:39 2012
05:58 <@ryah> groan where's ben when i need him
05:58 <@ryah> human documentation... where are you?
11:00  * isaacs is in ireland at node copter
11:21 <@isaacs> ircretary: tell TooTallNate to ping me when he wakes up.  We're at node copter trying to cross compile stuff for arm
11:21 < ircretary> isaacs: I'll be sure to tell tootallnate
11:21 <@isaacs> ircretary: thanks
11:21 < ircretary> isaacs: You're welcome :)
12:30 < bnoordhuis> away for three days == 80 issues backlog :(
12:31 < mmalecki> bnoordhuis: you're alive!
12:31 < bnoordhuis> mmalecki: i am :)
12:32 < mmalecki> bnoordhuis: it's great to hear, I can't imagine how a backlog would look like if you didn't
13:05 < MI6> joyent/node: Ben Noordhuis master * cb6d084 : node: update description of --print - http://git.io/0ObmjA
13:16 < txdv> bnoordhuis: how is it going? xD
13:36 < bnoordhuis> txdv: sup
13:36 < txdv> how is the rm-ev going on?
13:37 < bnoordhuis> txdv: didn't get a chance to work on it this week
13:37 < bnoordhuis> but i'll finish it up next week
13:39 < txdv> o that real life always hindering plans and stuff
13:39 < txdv> hindering is actually a germanic word, funny
14:40 <@indutny> bnoordhuis: hoya
14:40 <@indutny> bnoordhuis: uv_wait() ?
14:40 < bnoordhuis> indutny: heya
14:41 <@indutny> btw, I'm finally back in Moscow!
14:41 < bnoordhuis> indutny: uv_wait can wait (until monday)
14:41 < bnoordhuis> ah, i was just about to ask if you were still here
14:41 <@indutny> hahahaha
14:43 < bnoordhuis> indutny: good to be back home?
14:43 <@indutny> yeah
14:43 <@indutny> finally, I've figured out my military service problems
14:43 <@indutny> mraleph: ^^
14:43 <@indutny> ;)
14:43 <@indutny> now I'm completely free
14:43 < bnoordhuis> you still have the draft in russia?
14:43 < mraleph> indutny: congrats.
14:43 <@indutny> yes
14:43 < mraleph> bnoordhuis: yeah, we have "draft".
14:43 <@indutny> mraleph: thanks
14:44 < bnoordhuis> mraleph: quotes meaning it's not quite the same thing?
14:44 < mraleph> indutny: at least you did not have to eat a lot of food to escape that.
14:44 < mraleph> bnoordhuis: well, it's not a draft if everybody tries to cheat as much as possible to get out of it.
14:45 <@indutny> haha
14:45 < bnoordhuis> i think that's universal across countries :)
14:45 <@indutny> bnoordhuis: well, army isn't really a great place to spend time
14:45 < mraleph> bnoordhuis: well I did not have to cheat myself but it was a kind of random event that I did not for see. 
14:45 <@indutny> bnoordhuis: at least, not in russia
14:46 <@indutny> for example, there were one guy that were killed in train in Omsk
14:46 < mraleph> bnoordhuis: I don't think e.g. in Denmark everybody cheats.
14:46 <@indutny> at the time when he joined the army
14:47 < bnoordhuis> mraleph: my father-in-law - convicted squatter and hippie - got out because he liked to shoot guns a little too much to his drill instructor's liking
14:47 < mraleph> yeah, you don't want to go there… it's pretty rough if you are in the wrong place with wrong people and you are not into this bdsm system.
14:48 < bnoordhuis> the draft got abandoned about 15 years ago here but before that, people tried to game it - a lot
14:48 < mraleph> I see.
14:48 <@indutny> bnoordhuis: interesting
14:48 < bnoordhuis> mraleph: doesn't denmark have a system similar to germany where you can choose between the army or civil service?
14:49 < mraleph> that is not what people tell me in Denmark. well, draft here was randomized so most people were not drafted due to chance.
14:49 < mraleph> I have no idea how it worked. 
14:50 < bnoordhuis> indutny: for the record, my father-in-law is not a gun nut, he simply knew how to work the system :)
14:50 <@indutny> :)
14:52 < mraleph> bnoordhuis: actually in Russia they have very good shrinks there so simple gaming does not usually work.
14:52 < mraleph> bnoordhuis: they'll send you to a couple of weeks talking with shrinks in ward if you try e.g. to simulate suicidal tendencies.
14:53 < mraleph> so unless your are suicidal better not try anything funny :-) or they'll debug you until they find the truth.
14:54 < mraleph> it's not the first time they are dealing with people trying to simulate and they've become exceedingly efficient at that.
14:54 < mraleph> anyways that has nothing to do with #libuv, unless of course you are planning to draft developers.
14:55 < bnoordhuis> mraleph: i only want an army of volunteers
14:55 < mraleph> army of lovers.
14:55 < bnoordhuis> well... depends on who the volunteers are
14:55 <@indutny> oh gosh
14:56 < mraleph> indutny: http://www.youtube.com/watch?v=vYT2aWavXlc
14:56 <@indutny> hahahaha
15:13 <@indutny> oh crap
15:13 <@indutny> I've just tried to renew domain on domain.com via paypal
15:13 <@indutny> and it failed
15:13 <@indutny> but money came out
15:14 < mmalecki> indutny: if it charged your credit card you can initiate a chargeback
15:14 <@indutny> yeah
15:14 <@indutny> I know
15:14 <@indutny> but
15:14 <@indutny> I really would like to renew that domain
15:14 <@indutny> :)
15:15 < mmalecki> pay with a credit card?
15:15 < mmalecki> paypal is shit
15:15 <@indutny> meh
16:44 < Yorhel> Hmm. When is it safe to free() a uv_timer_t? I'm currently freeing the object immediately after calling uv_timer_stop(), but valgrind suggests that that isn't a good idea
16:46 < Yorhel> Or wait, am I supposed to do uv_timer_stop(), then uv_close() and free the memory in the callback?
16:47  * Yorhel is still getting used to libuv
16:57 <@piscisaureus_> Yorhel: you must free handles from the callback, always
16:57 <@piscisaureus_> Yorhel: btw - uv_close implicitly stop
16:57 <@piscisaureus_> ... s the timer as well, so you don't have to call it explicitly
16:57 < Yorhel> piscisaureus_: Nice, thanks!
16:58 <@piscisaureus_> Yorhel: also, if you malloc'ed the timer, you can shortcut it with uv_close(timer, (uv_close_cb) free)
16:59 < Yorhel> Even better :D
17:39 <@ryah> bnoordhuis: how does http_parser_pause work?
17:40 <@ryah> i presume that i call http_parser_pause(p, 1) and it will drop out of the http_parser_execute as soon as it can
17:40 <@ryah> but what happens with http_parser_pause(p, 0) ?
17:40 <@ryah> does it automatically reenter _execute? or do i need to call that again?
17:55 < mmalecki> ryah: doesn't seem like it reenters
17:55 < mmalecki> (looking at the code)
17:55 < mmalecki> it only sets state to HPE_OK
18:04 <@ryah> mmalecki[plane]: thanskf ro looking at the code for me :)
18:04 <@ryah> that's how lazy i am
20:58 <@piscisaureus_> when did we decide that uv_cpu_info should report the actual (scaled) cpu frequency?
20:58 <@piscisaureus_> it makes no sense to me :-(
20:59 <@piscisaureus_> bnoordhuis: ^
21:00 < bnoordhuis> piscisaureus_: non-scaled doesn't make sense when it's only running at half speed
21:00 < bnoordhuis> i.e. my laptop in battery mode
21:00 < bnoordhuis> or e.g.
21:00 <@piscisaureus_> bnoordhuis: huh are you kidding me
21:01 <@piscisaureus_> bnoordhuis: the frequency should increase when the cpu is busy
21:02 < bnoordhuis> piscisaureus_: i configured it that way, way back when
21:02 < bnoordhuis> gave me an additional 45 minutes of battery time
21:02 <@piscisaureus_> bnoordhuis: ghe
21:03 <@piscisaureus_> of which there are now 0 left
21:03 < bnoordhuis> anyway, i don't really care one way or the other
21:03 <@piscisaureus_> anyway
21:03 <@piscisaureus_> neither do I
21:03 <@piscisaureus_> but I don't want to land a planet into libuv to make this work
21:03 <@piscisaureus_> since this is just a stupid feature
21:03 < bnoordhuis> oh, right
21:04 < bnoordhuis> i can remove the cpufreq thing
21:04 <@piscisaureus_> bnoordhuis: it'll break node
21:04 <@piscisaureus_> er, it'll change node
21:04 < bnoordhuis> i don't think anyone is really going to care
21:05 < bnoordhuis> because that patch that introduced the change got back-ported to v0.8
21:05 < bnoordhuis> and no one complained
21:05 <@piscisaureus_> hehe
21:05 <@piscisaureus_> indeed
21:06 < bnoordhuis> so... remove?
21:06 <@piscisaureus_> I am doubtful about the irq/sys/idle time stuff too, but I don't see a compelling reason to remove
21:06 <@piscisaureus_> bnoordhuis: I am +1 on it
21:06 < bnoordhuis> yeah, irq... it doesn't get more pointless than that
21:07 <@piscisaureus_> bnoordhuis: it's also the per-cpu time
21:07 <@piscisaureus_> bnoordhuis: i mean, if it were the sys time of this program it'd be okay
21:08 < bnoordhuis> piscisaureus_: i think irq is just that, irq overhead
21:08 < bnoordhuis> i.e. dma operations and timers and whatnot
21:08 <@piscisaureus_> bnoordhuis: i wonder what the nice value is supposed to be
21:09 <@piscisaureus_> bnoordhuis: it's not filled out on windows
21:09 < bnoordhuis> piscisaureus_: you know i don't really know? :)
21:10 <@piscisaureus_> bnoordhuis: what does it say for you?
21:10 <@piscisaureus_> 5?
21:10 <@piscisaureus_> -39457349853?
21:10 <@piscisaureus_> 12.4 ?
21:11 < bnoordhuis> piscisaureus_: no, they're largish values, e.g. 77200 or 104200
21:11 <@piscisaureus_> ggg
21:13 < bnoordhuis> piscisaureus_: oh, i think i know what it is
21:13 <@piscisaureus_> bnoordhuis: ah it is similar to "user" but counting processes that were niced
21:14 < Gottox> Hi! Small question: can I uv_mutex_lock in one thread and uv_mutex_unlock in another?
21:14 <@piscisaureus_> Gottox: no
21:14 < bnoordhuis> Gottox: no
21:14 < bnoordhuis> damn, beaten to the punch
21:15 < Gottox> I suspected so... :/
21:15 <@piscisaureus_> unix and windows are remarkably compatible here I think :-p
21:15 <@piscisaureus_> Gottox: use a semaphore.
21:15 < bnoordhuis> Gottox: you probably want a semaphore or a condvar
21:15 <@piscisaureus_> bnoordhuis: too late again
21:15 < bnoordhuis> mine was more complete
21:15 < Gottox> semaphores, thank you! :)
22:06 < MI6> joyent/libuv: Ben Noordhuis master * 1bb1ba2 : unix: fix compiler warning in async.c  Include missing <string.h> header. - http://git.io/ywTMOg
22:09 < travis-ci> [travis-ci] joyent/libuv#814 (master - 1bb1ba2 : Ben Noordhuis): The build was broken.
22:09 < travis-ci> [travis-ci] Change view : https://github.com/joyent/libuv/compare/209abbab27ab...1bb1ba27ddfa
22:09 < travis-ci> [travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/2869947
--- Log closed Sun Oct 21 00:00:45 2012