| 00:01.41 | *** join/#asterisk-dev superscrat (~asanders@173-21-89-217.client.mchsi.com) |
| 00:19.14 | *** part/#asterisk-dev kharwell (kharwell@nat/digium/x-qtmjvfzqncyofsbb) |
| 01:54.21 | stefan27 | C still puzzles me, when I add a simple line to rtp_engine.c 'int wx3_testf(int a) { return a; }' and recompile, then the function wx3_testf is still not in defined in global scope in /usr/sbin/asterisk , that is, /usr/bin/nm /usr/sbin/asterisk outputs 000000000058df20 t wx3_testf (not ... T wx3_testf); it should be in global scope ? what am I missing |
| 01:58.53 | file | stefan27, the build system for Asterisk requires symbols to be explicitly defined in exports file (either by name OR using a wildcard such as ast_*) otherwise they aren't globally exported |
| 01:59.26 | stefan27 | Oh, so some magic in the make process adds static modifiers? |
| 01:59.28 | file | in the case of the main binary this is main/asterisk.exports.in which gets turned into main/asterisk.exports by the build system with some variable substitution |
| 02:00.53 | file | sorta |
| 02:01.23 | file | the linker is told about it |
| 02:03.22 | stefan27 | thanks, prefixing with ast_ worked, you've saved me some trouble. |
| 02:03.58 | file | darn - I'm only supposed to cause trouble! |
| 02:07.12 | stefan27 | I'm trying do that multiple libsrtp-sessions hack. I managed to get in contact with the guy you discussed that with 1 ½ years ago, he opted to make his own webrtc gateway (called Janus) instead of trying to fix all the problems with dtls and ice by patching asterisk |
| 02:09.35 | file | nods |
| 02:09.58 | file | it's easy to start fresh, trying to maintain something that already exists without breaking or changing things too much is hard |
| 02:10.35 | stefan27 | his gateway seemed to work fine, i tested it against our 13.5 installation and vp8 video was a lot better since FIRs were communicated properly, but our webrtc client is tightly coupled to sipml and asterisk, I'm not sure what's best for us |
| 02:11.13 | stefan27 | yeah, well for this experimental branch of asterisk13.5 we don't have to worry as much about breaking stuff as you would, since our use cases are very limited |
| 02:11.20 | file | yup |
| 02:12.00 | stefan27 | this patch will be the ugliest ever, I could post it under the JIRA but it would never get pulled by anyone sane |
| 06:51.06 | *** join/#asterisk-dev goddva (~goddva@77.40.154.242) |
| 06:57.44 | *** join/#asterisk-dev bof22 (~Thunderbi@185.13.183.107) |
| 07:17.17 | *** join/#asterisk-dev hehol (~hehol@gatekeeper.loca.net) |
| 07:55.07 | *** join/#asterisk-dev pchero_work (~pchero@109.70.54.56) |
| 08:18.40 | *** join/#asterisk-dev tzafrir (~tzafrir@local.xorcom.com) |
| 08:18.40 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 13:11.13 | file | ponders |
| 13:12.04 | *** join/#asterisk-dev skrusty (~skrusty@168.63.14.171) |
| 13:43.12 | *** join/#asterisk-dev bof22 (~Thunderbi@185.13.183.107) |
| 13:47.47 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@216-191-106-163.dedicated.allstream.net) |
| 14:37.07 | *** join/#asterisk-dev gtjoseph (~gtj0@unaffiliated/gtj) |
| 15:03.39 | *** join/#asterisk-dev acidfoo (45a5a03c@gateway/web/freenode/ip.69.165.160.60) |
| 15:15.50 | *** join/#asterisk-dev kharwell (kharwell@nat/digium/x-riipfpzwxajfcknw) |
| 15:18.10 | *** part/#asterisk-dev kharwell (kharwell@nat/digium/x-riipfpzwxajfcknw) |
| 15:18.33 | *** join/#asterisk-dev kharwell (kharwell@nat/digium/x-ppbddvebnangkiqn) |
| 15:48.37 | *** join/#asterisk-dev elguero (cf9a22a3@gateway/web/freenode/ip.207.154.34.163) |
| 15:57.10 | *** join/#asterisk-dev rmudgett (rmudgett@nat/digium/x-dizthmayprfkqfgv) |
| 15:57.28 | *** join/#asterisk-dev rmudgett (rmudgett@nat/digium/x-npoxcovthbgzimvd) |
| 16:25.42 | *** join/#asterisk-dev zerohalo (~zerohalo@c-174-62-142-83.hsd1.ma.comcast.net) |
| 17:40.21 | *** join/#asterisk-dev matesstar (d4478a32@gateway/web/freenode/ip.212.71.138.50) |
| 18:18.28 | coreyfarrell | last night I had a crazy idea about pbx.c in master. is it actually required that 'dialplan reload' take effect for existing channels? or would it be acceptable if the updated dialplan was for new channels only? |
| 18:22.26 | rmudgett | Implementing that would require multiple dialplan's in memory at the same time controlled by ao2 ref counting. Dialplans can be quite large. |
| 18:22.32 | file | insanely large. |
| 18:22.48 | file | and it's a drastic behavior change. |
| 18:23.39 | coreyfarrell | right so frequent reloads on a busy system would cause huge memory usage |
| 18:25.34 | file | yes |
| 18:27.36 | coreyfarrell | so for 'ACO' style immutable dialplan structures channels would still need to grab updated versions of dialplan / drop old versions frequently. |
| 18:29.57 | file | oui |
| 18:39.28 | *** join/#asterisk-dev tzafrir (~tzafrir@bzq-179-40-172.cust.bezeqint.net) |
| 18:39.28 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 18:51.56 | *** join/#asterisk-dev elguero (cf9a22a3@gateway/web/freenode/ip.207.154.34.163) |
| 19:06.08 | *** join/#asterisk-dev goddva (~goddva@2a02:fe0:c510:e300:ae9e:17ff:fe91:a8c9) |
| 19:32.45 | *** join/#asterisk-dev goddva (~goddva@77.40.154.242) |
| 20:55.48 | *** join/#asterisk-dev acidfoo (45a5a03c@gateway/web/freenode/ip.69.165.160.60) |
| 22:04.39 | *** join/#asterisk-dev Synthase__ (uid63346@gateway/web/irccloud.com/x-vzqtbjdmkhabzduz) |
| 22:05.24 | *** join/#asterisk-dev Qwell (~north@asterisk/developer/Qwell) |
| 22:05.24 | *** mode/#asterisk-dev [+o Qwell] by ChanServ |
| 22:21.43 | *** join/#asterisk-dev gtjoseph (~gtj0@unaffiliated/gtj) |
| 22:30.33 | *** join/#asterisk-dev snuff-work (~blatus@203.87.61.22) |
| 22:30.33 | *** mode/#asterisk-dev [+o snuff-work] by ChanServ |
| 23:47.44 | *** join/#asterisk-dev [TK]D-Fender (~joe@64.235.216.2) |