00:01:46 <|3b|> is there some point during a GC where 1 thread interrupts other threads to tell them about the GC? 00:01:55 |3b|: yup. 00:02:19 gc_stop_the_world, iirc. 00:02:33 <|3b|> and the thread doing the interruptions has itself been interrupted? 00:03:08 it's grabbed the GC lock and is in the C runtime. 00:04:02 <|3b|> by 'interrupt' in both cases i mean the stuff that would end up calling run-interruption in lisp, not just stopping things in general 00:05:50 it's not sigpipe, fwiw. 00:06:01 <|3b|> ok 00:07:33 *|3b|* suspects the (:timer :threaded-stress) is same root problem as the other deadlock, not much luck figuring out how though 00:17:32 <|3b|> ah, maybe i see it now... i wonder if the test was intended to keep running the timers after the threads exit 00:19:16 <|3b|> or no, it is a special so timers should stop first... hmm 00:35:39 <|3b|> is run-expired-timers (called from sigalrm-handler) run in any particular thread? 00:36:13 *|3b|* suspects that run-interrupts isn't the only thing that needs that slot cleared 00:37:45 <|3b|> possibly invoke-interrupt should do it? 00:47:24 I think we want to add a callback to with-interrupts. 00:47:54 no clue how it'd be implemented, though. 00:52:36 *|3b|* wonders if it would be cleaner to distinguish locks held before a wait from locks held while waiting 00:53:50 or, we could associate SP to the lock 00:54:19 you end up with a thread-local stack of (sp . lock-waiting-for) pairs 00:54:42 only the topmost, when SP matches, is relevant. 00:55:21 <|3b|> i think topmost isn't enough, you really want all waits after the lock was acquired 00:55:55 <|3b|> but not the ones below that 00:56:45 can still be supported with the same overhead 00:58:46 <|3b|> could be 00:59:29 that only needs a stack of waiting-for, actually. 01:00:06 store top-of-stack when a lock is acquired, that way you know when to stop looking. 01:00:19 <|3b|> yeah, that sounds like what i was thinking 01:08:17 yup, your thing passes my basic test of: does it work right for recursive locking 01:09:56 *|3b|* is better at the 'break stuff and figure out how it broke' than the 'make it work right' part :) 01:11:00 on the topic of breaking stuff, i need someone with ppc-fu. 01:12:13 nyef? 01:13:40 it's 3 lines of asm (: 01:27:12 hargettp [~hargettp@pool-71-184-177-187.bstnma.east.verizon.net] has joined #sbcl 01:31:35 -!- sbalousek [~sbalouse@ip174-67-214-212.oc.oc.cox.net] has quit [Read error: Connection reset by peer] 01:32:38 sbalousek [~sbalouse@ip174-67-214-212.oc.oc.cox.net] has joined #sbcl 02:05:29 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 02:20:26 -!- hargettp [~hargettp@pool-71-184-177-187.bstnma.east.verizon.net] has quit [Quit: Leaving...] 05:07:52 good luck! 05:26:45 Hope I did not make too much work for you. Do the good hack. 05:27:54 cpc26: what are you talking about? 05:29:32 http://www.sbcl.org/ 05:30:09 you are Colin Carr? 05:30:15 yes 05:30:30 i see 05:31:51 that's some hefty contribution 05:32:20 yeah, for greater good. 05:34:46 I really hate Kristian Hammond! 05:34:51 for real 05:34:59 POS 05:35:12 what's he famous for? 05:35:24 being an asshole. 05:36:48 so if threading API can be Kristian Hammond is an asshole, that would be great, 05:39:20 otherwise I am estatic to contribute to SBCL. 05:41:21 besides I make a fortune from SBCL 05:41:45 glad to have a place to give "back" 05:45:20 good to hear that real people are using it ;) 05:45:34 slyrus [~chatzilla@99-28-163-38.lightspeed.miamfl.sbcglobal.net] has joined #sbcl 05:45:59 yeah all the time 05:46:18 can't imagine using something else 05:48:39 anway I need better; elephant and screamer support plus mcclim 05:49:00 you should just wrap mclcim under sbcl 05:49:35 work with abcl to make a mcclim too? = we maybe 05:50:09 whatever you all are golden anyway 05:52:39 love what you do well except when I use GBB but I guess they are using some obscure common lisp etc. etc. 05:53:36 maybe not CL but Lisp "traditions" 05:57:33 but honestly; we just need to know how to work with 05:57:39 JMS 05:57:47 and proxies 05:58:02 java message service? 05:58:13 under the radar but atill working 05:58:28 *still 05:59:31 once that is accompished then the object model cahnges. 05:59:48 so you shift from java to CL 06:00:07 the C++ is really asm with templates 06:00:21 anywy... 06:01:18 any way Kristian Hammond is an issue/problem. 06:03:54 anyway - good luck 06:04:27 -!- slyrus [~chatzilla@99-28-163-38.lightspeed.miamfl.sbcglobal.net] has quit [Read error: Connection reset by peer] 06:04:37 Hope I did not make too much work for you. Do the good hack. 06:04:49 slyrus [~chatzilla@99-28-163-38.lightspeed.miamfl.sbcglobal.net] has joined #sbcl 06:06:00 seriously, do what you need too Kristian Hammond is a seperate issue. 06:06:24 * to do 06:10:14 but srsl Kristian Hammond is a half wit. 06:10:50 and that is a great GIT commit et. al. 06:11:30 Kristian Hammond absolute idoiot, no joke 06:11:51 * whoop idiot 06:15:30 Kristian Hammond oh shit. did I mentio him? 06:17:15 anywway you have a plan and enough of this: goood luuck! 06:17:58 Kristian Hammond suks BTW 06:19:15 cfy [~cfy@unaffiliated/chenfengyuan] has joined #sbcl 06:20:23 JMS, yeah 08:21:36 nikodemus [~nikodemus@cs181063174.pp.htv.fi] has joined #sbcl 08:21:36 -!- ChanServ has set mode +o nikodemus 09:17:16 attila_lendvai [~attila_le@catv-89-132-188-166.catv.broadband.hu] has joined #sbcl 09:17:16 -!- attila_lendvai [~attila_le@catv-89-132-188-166.catv.broadband.hu] has quit [Changing host] 09:17:16 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 09:37:52 loke [~elias@bb115-66-87-235.singnet.com.sg] has joined #sbcl 10:10:20 hargettp [~hargettp@pool-71-184-177-187.bstnma.east.verizon.net] has joined #sbcl 10:36:37 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has left #sbcl 11:22:11 -!- tsuru [~charlie@adsl-74-240-217-227.bna.bellsouth.net] has quit [Remote host closed the connection] 11:22:30 tsuru [~charlie@adsl-74-240-217-227.bna.bellsouth.net] has joined #sbcl 12:10:41 homie [~levgue@xdsl-84-44-179-39.netcologne.de] has joined #sbcl 12:11:41 -!- cpc26 [cpc26@pilot.trilug.org] has quit [Remote host closed the connection] 12:29:41 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 250 seconds] 12:30:03 -!- stassats` [~stassats@wikipedia/stassats] has quit [Ping timeout: 276 seconds] 12:58:34 stassats [~stassats@wikipedia/stassats] has joined #sbcl 14:06:58 -!- Kryztof [~user@csrhodes.plus.com] has quit [Remote host closed the connection] 15:37:15 pchrist_ [~spirit@gentoo/developer/pchrist] has joined #sbcl 15:39:41 -!- pchrist [~spirit@gentoo/developer/pchrist] has quit [Ping timeout: 240 seconds] 15:43:41 -!- nikodemus [~nikodemus@cs181063174.pp.htv.fi] has quit [Quit: This computer has gone to sleep] 15:58:54 -!- cfy [~cfy@unaffiliated/chenfengyuan] has quit [Remote host closed the connection] 17:06:35 Vivitron [~user@pool-71-174-61-33.bstnma.fios.verizon.net] has joined #sbcl 17:42:33 -!- Vivitron [~user@pool-71-174-61-33.bstnma.fios.verizon.net] has quit [Ping timeout: 250 seconds] 18:09:31 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 252 seconds] 18:43:04 -!- homie [~levgue@xdsl-84-44-179-39.netcologne.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 18:47:39 homie [~levgue@xdsl-84-44-179-39.netcologne.de] has joined #sbcl 18:49:40 -!- homie [~levgue@xdsl-84-44-179-39.netcologne.de] has quit [Remote host closed the connection] 18:55:04 homie [~levgue@xdsl-84-44-179-39.netcologne.de] has joined #sbcl 19:53:27 Kryztof [~user@csrhodes.plus.com] has joined #sbcl 19:53:27 -!- ChanServ has set mode +o Kryztof 20:01:48 stassats [~stassats@wikipedia/stassats] has joined #sbcl 20:01:49 lol, LC_ALL strikes again 20:46:27 -!- scymtym [~user@2001:638:504:2093:21a:a0ff:fe34:2d7d] has quit [Remote host closed the connection] 21:03:55 scymtym [~user@2001:638:504:2093:21a:a0ff:fe34:2d7d] has joined #sbcl 21:10:58 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 264 seconds] 21:16:50 -!- hargettp [~hargettp@pool-71-184-177-187.bstnma.east.verizon.net] has quit [Quit: Leaving...] 23:34:29 homie` [~levgue@xdsl-78-35-156-115.netcologne.de] has joined #sbcl 23:36:35 -!- homie [~levgue@xdsl-84-44-179-39.netcologne.de] has quit [Ping timeout: 260 seconds]