| 00:51.20 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 00:51.20 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 02:26.25 | *** join/#asterisk-dev sruffell (~Adium@asterisk/the-kernel-guy/sruffell) |
| 02:26.25 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 02:59.53 | *** join/#asterisk-dev rrb3942 (~rbullock@cpe-67-242-215-62.rochester.res.rr.com) |
| 04:38.00 | *** join/#asterisk-dev jameswf (~james@unaffiliated/jameswf-home) |
| 04:57.01 | *** join/#asterisk-dev erinspice (~erin@207.98.195.107) |
| 04:58.50 | *** join/#asterisk-dev sruffell (~Adium@asterisk/the-kernel-guy/sruffell) |
| 04:58.50 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 05:42.22 | *** join/#asterisk-dev bbryant (~brett@c-76-26-221-79.hsd1.sc.comcast.net) |
| 05:42.22 | *** mode/#asterisk-dev [+o bbryant] by ChanServ |
| 06:46.22 | *** join/#asterisk-dev philipp64|laptop (~chatzilla@71-34-114-78.ptld.qwest.net) |
| 07:52.33 | *** join/#asterisk-dev asterisk-learner (~chatzilla@77.42.241.114) |
| 07:59.59 | *** join/#asterisk-dev lftsy (~lftsy@install.deckpoint.ch) |
| 08:29.22 | *** join/#asterisk-dev hehol (~hehol@2001:1438:1009:200:21b:63ff:feb5:e13b) |
| 08:37.14 | *** join/#asterisk-dev tamiel (~tamiel@213.30.183.226) |
| 08:43.26 | *** join/#asterisk-dev felipe_ (~felipe@unaffiliated/felipe) |
| 09:08.35 | *** join/#asterisk-dev hehol (~hehol@2001:1438:1009:200:21b:63ff:feb5:e13b) |
| 10:19.47 | *** join/#asterisk-dev guilhermebr (~Guilherme@200.103.96.98) |
| 10:28.02 | *** join/#asterisk-dev Pitel (~pitel@ip-94-113-20-150.net.upcbroadband.cz) |
| 10:28.54 | Pitel | Hi, can someone help me with https://issues.asterisk.org/view.php?id=18521 I'd like to know how to use valgrind too detect leaks in libraries used by asterisk. |
| 11:19.25 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 11:19.25 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 11:33.33 | snuff-home | mornin kp |
| 11:37.07 | kpfleming | morning |
| 11:37.43 | kpfleming | snuff-home: how's summer going for you? |
| 11:41.42 | *** join/#asterisk-dev msetim (~setim@187.59.70.173) |
| 12:17.47 | snuff-home | not bad |
| 12:18.12 | snuff-home | glad i'm not in queensland.. kinda wet up there.. too wet |
| 12:18.47 | snuff-home | the CEL userfield or whatever that was talked about on the mailing list a little.. i had a patch to add the field to the other CEL backends.. |
| 12:19.23 | snuff-home | i added it for the guy that wanted it for postgres one.. i just need to find which pc i did the mods to the rest of them :) |
| 12:20.01 | snuff-home | still playin too much gt5 |
| 12:21.21 | snuff-home | been good down in victoria.. weather between 20 and 35C (not F) very little rain |
| 12:21.36 | snuff-home | went to armin only mirage concert for NYE.. wasn't bad either |
| 12:21.54 | kpfleming | sounds nice |
| 12:22.11 | kpfleming | we got quite a bit of snow (4-5") for christmas, made for a nice weekend |
| 12:23.02 | snuff-home | sees murf already beat me to putting the full patch up ;) |
| 12:24.39 | snuff-home | i see NY got lil more snow than they wanted.. |
| 12:26.07 | kpfleming | yes |
| 12:27.41 | snuff-home | links issues 18559/18560 and 18462 |
| 12:29.16 | snuff-home | anymore demolision jobs?.. i just did a small one over the w/end.. great to use sledgehammer / saws / claw hammers to tear stuff down |
| 12:31.15 | kpfleming | have some renovation work scheduled for the 15th, a group of us going to help Habitat for Humanity work on a house |
| 12:33.55 | snuff-home | always feels good to be able to 'see' physically something u've done.. |
| 12:34.35 | kpfleming | yep |
| 12:35.06 | kpfleming | we actually have worked scheduled three weekends in a row, and then our group will try to be the organizers for one of the complete houses they will build during the spring as well |
| 12:35.19 | snuff-home | best thing is it wont suffer from race conditions and lock contention :) |
| 12:35.20 | kpfleming | it's lots of fun, very rewarding and a great way to relieve stress |
| 12:35.23 | kpfleming | indeed |
| 12:35.45 | snuff-home | sounds good |
| 12:37.29 | snuff-home | house cleaning is not so fun.. deciding what to keep and what to throw between 2-4 ppl and actually getting vote to trash it |
| 12:38.33 | snuff-home | is very happy with freenode having an AU server again.. only been without one for almost 2 years i think.. |
| 12:41.58 | snuff-home | back to work 5th.. then.. next sunday off to brisbane for a week to install server n storage.. |
| 12:46.36 | *** join/#asterisk-dev sruffell (~sruffell@asterisk/the-kernel-guy/sruffell) |
| 12:46.36 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 12:50.43 | Pitel | Can someone help me with https://issues.asterisk.org/view.php?id=18521 I'd like to know how to use valgrind too detect leaks in libraries used by asterisk. |
| 12:53.14 | *** join/#asterisk-dev guilhermebr (~Guilherme@200.103.96.98) |
| 12:53.14 | *** join/#asterisk-dev micols (~mio@rlogin.dk) |
| 12:53.14 | *** join/#asterisk-dev paravoid (~paravoid@zebra.cslab.ece.ntua.gr) |
| 13:00.33 | *** join/#asterisk-dev leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 13:00.33 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 13:00.59 | *** join/#asterisk-dev rrb3942 (~rbullock@208.34.105.161) |
| 13:01.27 | *** join/#asterisk-dev russellb (~russellb@asterisk/digium-open-source-team-lead/russellb) |
| 13:01.27 | *** mode/#asterisk-dev [+o russellb] by ChanServ |
| 13:09.34 | *** join/#asterisk-dev jsmith (~jsmith@fedora/jsmith) |
| 13:09.35 | *** mode/#asterisk-dev [+o jsmith] by ChanServ |
| 13:10.03 | jsmith | Happy New Year, all! |
| 13:13.09 | kpfleming | jsmith: you too! |
| 13:14.10 | jsmith | is getting used to driving on the right-hand side of the road (and left-hand side of the car) after driving almost 1300 miles in Ireland/England/Northern Ireland |
| 13:21.48 | snuff-home | jsmith: avoid the closed heathrow etc during christmas? |
| 13:22.12 | jsmith | snuff-home: I did! I did plenty of driving in the snow, though :-p |
| 13:22.23 | jsmith | snuff-home: I flew into Bristol, and out of Luton |
| 13:23.39 | snuff-home | aah.. avoiding heathrow altogether.. smart if possible :) |
| 13:24.18 | sruffell | jsmith: what had you driving around the UK so much? |
| 13:24.55 | jsmith | sruffell: Vacation! |
| 13:25.40 | sruffell | ahh...nice! |
| 13:26.49 | snuff-home | calls it a night.. |
| 14:08.00 | *** join/#asterisk-dev bklang_ (~bklang@tesla.alkaloid.net) |
| 14:21.18 | *** join/#asterisk-dev jeffg (~jeffg@pdpc/supporter/professional/jeffg) |
| 14:34.33 | *** join/#asterisk-dev jameswf (~james@unaffiliated/jameswf-home) |
| 14:34.46 | jameswf | jsmith mind if I pm |
| 14:35.47 | jsmith | jameswf: No worries... pm away! |
| 14:44.07 | *** join/#asterisk-dev putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 14:44.07 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:58.05 | *** join/#asterisk-dev datachomper (~Adium@nat/digium/x-hwdlyilctgsrnrbn) |
| 15:01.08 | pabelanger | Corydon76-home: Did you ever make any headway with the dlclose() issue in main/loader.c? Last I remember, I though you may have had a work around? |
| 15:20.52 | Pitel | Can someone please help me with https://issues.asterisk.org/view.php?id=18521 I'd like to know how to use valgrind too detect leaks in libraries used by asterisk. |
| 15:22.11 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 15:41.20 | *** join/#asterisk-dev elguero (~miguel323@dsl092-072-112.bos1.dsl.speakeasy.net) |
| 15:48.59 | *** join/#asterisk-dev jpeeler (~jpeeler@asterisk/developer/jpeeler) |
| 15:48.59 | *** mode/#asterisk-dev [+o jpeeler] by ChanServ |
| 15:52.50 | Corydon76-home | pabelanger: I do not |
| 15:54.02 | Corydon76-home | pabelanger: My attempted workaround failed |
| 16:00.40 | *** join/#asterisk-dev The_Boy_Wonder (~vossel@asterisk/batman-developer/dvossel) |
| 16:00.40 | *** mode/#asterisk-dev [+o The_Boy_Wonder] by ChanServ |
| 16:18.52 | pabelanger | Corydon76-home: Okay, thanks for the update. So, the problem is within dlclose(), right? |
| 16:19.27 | Corydon76-home | pabelanger: the problem is that the error codes for dlclose() are unspecified |
| 16:20.33 | *** join/#asterisk-dev jsmith-away (~jsmith@fedora/jsmith) |
| 16:20.33 | *** mode/#asterisk-dev [+o jsmith-away] by ChanServ |
| 16:20.45 | Qwell | seanbright: ... |
| 16:20.46 | Qwell | ... |
| 16:24.35 | putnopvut | Qwell: would you expect anything less? |
| 16:24.52 | Qwell | Nope. I would've been disappointed had he not gotten it. |
| 16:25.08 | Corydon76-home | Another possible code fix to the DEBUG_THREADS issue would be to use a structure versioning element, then ensure that the extra elements for DEBUG_THREADS are moved to the end |
| 16:25.33 | Corydon76-home | then only access the extra elements if the structure version matches |
| 16:26.56 | Corydon76-home | Qwell: since you coded the original fix, I'd like to hear your input, too |
| 16:27.37 | Qwell | what is this supposed to be solving? I haven't seen any related discussion. |
| 16:28.26 | Corydon76-home | 8 months ago, a fix went in (after going through reviewboard) that made sure that DEBUG_THREADS did not change ABI |
| 16:28.43 | Corydon76-home | The result is that every mutex is now 3016 bytes |
| 16:28.56 | Corydon76-home | Not a problem if there's only a few per module, BUT |
| 16:29.33 | Corydon76-home | Each ao2 object includes a mutex, and chan_iax2 allocates 32767 ao2 object at load time. This consumes 100MB of memory |
| 16:29.44 | Qwell | why? |
| 16:29.51 | Qwell | sounds like a "Don't do that then." |
| 16:31.09 | seanbright | Qwell: yessir? |
| 16:31.10 | Corydon76-home | I did not look very detailed into that part. |
| 16:31.29 | Corydon76-home | It's something that David Vossel coded |
| 16:31.56 | russellb | chan_iax2 has always been that way |
| 16:32.19 | Corydon76-home | Well, the conversion to ao2 objects was just a recent thing |
| 16:33.01 | Corydon76-home | Although, you're right that it's always had 32767 mutexes |
| 16:33.49 | Corydon76-home | Just before, they were all in an array |
| 16:38.01 | Kobaz | yay, i got my commit passphrase |
| 17:02.01 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 17:02.01 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 17:30.50 | JunK-Y | lo guys |
| 17:31.48 | file | meow |
| 17:33.03 | JunK-Y | how are ya M. Colp? |
| 17:34.33 | file | good |
| 17:34.34 | file | you? |
| 17:34.48 | JunK-Y | im great |
| 18:00.29 | sruffell | tzafrir: Do you need me to wait on the chan_list commit still? |
| 18:02.07 | tzafrir | sruffell, I have a few minor fixes for that branch. But no, go ahead |
| 18:02.23 | sruffell | do you want me to add your ack? |
| 18:02.27 | sruffell | or just commit as is? |
| 18:05.21 | sruffell | btw: I had 4 changes to squash with some of the other ones...they are the "fixup!" commits at the head of https://github.com/sruffell/dahdi-linux/commits/chan_list_tocommit/ Fix a couple of bugs that I found in testing before leaving break. |
| 18:08.26 | *** join/#asterisk-dev luckman212 (~quassel@pool-96-246-172-198.nwrknj.fios.verizon.net) |
| 18:09.50 | *** join/#asterisk-dev The_Boy_Wonder (~vossel@asterisk/batman-developer/dvossel) |
| 18:09.50 | *** mode/#asterisk-dev [+o The_Boy_Wonder] by ChanServ |
| 18:17.16 | sruffell | tzafrir: unless you catch me here in the next 5 minutes...I just added your ack to the one commit that touched xpp.....if you want me to add it to the other ones, let me know. |
| 19:01.55 | *** join/#asterisk-dev angler (~angler@pdpc/sponsor/digium/angler) |
| 19:39.58 | Qwell | Corydon76-home: If we're going to change ABI again to facilitate this, why not just make the biggest part of the struct (ast_lock_track) a pointer? |
| 19:41.01 | Corydon76-home | Qwell: I think the main problem it was designed for was to make it a single allocation |
| 19:41.19 | kpfleming | that ast_lock_track stuff would still have to be allocated, and the same amount of memory would be consumed |
| 19:41.40 | Corydon76-home | Qwell: That's why I was thinking if we moved the DEBUG_THREADS stuff to the end, then used a version for the struct... |
| 19:42.07 | kpfleming | also, when the structure is inside another structure, it would require allocation of the lock tracking data later... somehow |
| 19:42.50 | Qwell | kpfleming: it would only have to be allocated if Asterisk was compiled with DEBUG_THREADS, no? the only time it's used is when Asterisk itself has that compile option. stuff is just empty otherwise |
| 19:42.50 | Corydon76-home | Not really a structure version, but really just a boolean indicating whether the extended stuff was "there" or not |
| 19:42.51 | kpfleming | right |
| 19:43.55 | Qwell | kpfleming: and there is already an init function for the tracking stuff |
| 19:44.03 | kpfleming | alright |
| 19:44.15 | Corydon76-home | That works |
| 19:51.29 | Corydon76-home | Qwell: guess we can get rid of the boolean, then, too, right? |
| 19:51.43 | Qwell | probably so |
| 19:51.59 | Qwell | it's used a bit funkily, but that shouldn't be too difficult |
| 19:53.12 | Corydon76-home | Change rwlock at the same time, even though it wasn't a problem (keep the two parallel implementations similar)? |
| 19:53.30 | Qwell | no reason not to |
| 19:53.39 | Qwell | the effort required would be minimal |
| 19:54.17 | Corydon76-home | There was one thing that bugged the crap out of me... see if it's worth the effort to fix. The debug arguments are tacked onto the beginning for mutexes but on the end for rwlocks |
| 19:55.31 | Corydon76-home | PITA to switch from mutexes to rwlocks when you're using the debug versions in utility code (such that the code blames the caller) |
| 19:56.42 | Corydon76-home | Oh, it's even worse. rwlock isn't consistent |
| 19:57.24 | Qwell | yeah... I didn't want to change that. It was painful to keep. |
| 21:00.29 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 21:45.21 | *** join/#asterisk-dev Daviey (~Daviey@ubuntu/member/daviey) |
| 21:54.31 | *** join/#asterisk-dev philipp64|laptop (~chatzilla@192.55.64.175) |
| 21:56.16 | *** join/#asterisk-dev outtolunc (~me@adsl-66-218-53-172.dslextreme.com) |
| 22:20.11 | *** join/#asterisk-dev aimka (~aimka@unaffiliated/aimka) |
| 22:20.45 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 22:35.06 | *** join/#asterisk-dev msetim (~setim@189.26.31.114.dynamic.adsl.gvt.net.br) |
| 22:38.39 | *** join/#asterisk-dev Qwell (~north@pdpc/sponsor/digium/Qwell) |
| 22:38.39 | *** mode/#asterisk-dev [+o Qwell] by ChanServ |
| 22:58.55 | *** join/#asterisk-dev file (~file@asterisk/developer-and-muffin-lover/file) |
| 22:58.55 | *** mode/#asterisk-dev [+o file] by ChanServ |
| 23:04.25 | *** join/#asterisk-dev jthurman42 (~jthurman@66.154.131.25) |
| 23:10.02 | jthurman42 | Anyone able to take a look at a patch for issue 18403 (sip deadlock on refer)? I'd like to see that issue resolved before 1.8.2 comes. |