| 15:59.58 | *** join/#asterisk-dev infobot (i=ibot@rikers.org) |
| 15:59.58 | *** topic/#asterisk-dev is Topic for #asterisk-dev is: Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- Tier 2 and 3.14159265 support is in #asterisk -=- ViewVC for SVN is currently down for the forseeable future |
| 16:07.33 | *** join/#asterisk-dev marv[work] (n=timr@router.asteriasgi.com) |
| 16:16.26 | *** join/#asterisk-dev zeroHalo (n=zeroHalo@173.13.92.17) |
| 16:37.47 | *** join/#asterisk-dev lesouvage (n=lesouvag@82.73.69.76) |
| 16:49.56 | *** join/#asterisk-dev errotan (n=errotan@62.201.123.1) |
| 17:00.26 | kpfleming | seanbright: wanna bet it's the 'No timing source found' message that makes that person think their TC400 is not working? |
| 17:05.20 | seanbright | kpfleming: that is what i am thinking too |
| 17:05.33 | seanbright | kpfleming: but it's about the journey |
| 17:05.34 | seanbright | :-) |
| 17:16.28 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 17:19.47 | *** join/#asterisk-dev jpeeler (n=jpeeler@asterisk/digium-software-dev/jpeeler) |
| 17:19.47 | *** mode/#asterisk-dev [+o jpeeler] by ChanServ |
| 17:23.40 | seanbright | someone might want to tell him that the "bogo" in bogomips means "bogus" |
| 17:25.50 | Corydon76-lap | seanbright: isn't that redundant, though? MIPS isn't a particularly good method of measuring anything about processor performance |
| 17:27.21 | seanbright | Corydon76-lap: you're right |
| 17:27.30 | seanbright | the point i was really trying to make is: |
| 17:27.34 | seanbright | <PROTECTED> |
| 17:27.39 | seanbright | ^ space added for emphasis |
| 17:28.46 | Corydon76-lap | Space added to prevent it from appearing in the logs, you mean |
| 17:28.53 | seanbright | well yes. |
| 17:28.53 | seanbright | heh |
| 17:51.32 | kpfleming | M15770 |
| 17:51.34 | MuffinMan | [closed] [Asterisk] Channels/chan_sip/T.38 0015770: faxing with T.38 fails reported by Had (Karma: neutral) https://issues.asterisk.org/view.php?id=15770 |
| 17:51.58 | kpfleming | that sort of thing really irks me... the poster knows his provider was the one that hung up the call, but he opens an issue against asterisk |
| 18:21.27 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 18:21.27 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 18:24.35 | *** join/#asterisk-dev bbryant (n=brett@68.208.65.50) |
| 18:24.35 | *** mode/#asterisk-dev [+o bbryant] by ChanServ |
| 18:26.06 | *** join/#asterisk-dev malcolmd (n=malcolmd@pdpc/sponsor/digium/malcolmd) |
| 18:26.06 | *** mode/#asterisk-dev [+o malcolmd] by ChanServ |
| 18:30.51 | *** join/#asterisk-dev denon (i=denon@synapse.subneural.net) |
| 18:30.59 | *** join/#asterisk-dev errotan (n=errotan@62.201.123.1) |
| 18:32.47 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 18:59.39 | kpfleming | a new oej branch |
| 18:59.44 | kpfleming | wonders what we'll see there |
| 19:01.56 | putnopvut | valgrind warning for a jump based on uninitialized variable in chan_sip. In trunk, there's a line in process_sdp: if (p->trtp && tsin.sin_port) {. On this line, tsin.sin_port is uninitialized since I received no text offers, but p->trtp is non-NULL since I have textsupport enabled. If someone wants to jump on that, go for it. |
| 19:03.14 | kpfleming | it's not a bug |
| 19:03.24 | kpfleming | i just had a discussion on gcc-help with the crew there about this issue |
| 19:03.43 | putnopvut | Why is that not a bug? |
| 19:03.55 | putnopvut | I guess you could answer by directing me to that discussion. |
| 19:03.58 | kpfleming | believe it or not, the way the compiler optimizes that code causes valgrind to see that uninitialized memory is being read, but the actual jump instruction is not affected by its value |
| 19:04.02 | kpfleming | let me find the thread |
| 19:04.23 | putnopvut | Why isn't the jump instruction affected by the value? That makes no sense to me. |
| 19:04.48 | kpfleming | http://gcc.gnu.org/ml/gcc-help/2009-08/msg00116.html |
| 19:05.13 | oej | kpfleming: Life is full of surprises. So is this branch :-) |
| 19:05.41 | oej | Just some experiments while testing servers. Got more than 10000 channels running today. |
| 19:06.03 | kpfleming | putnopvut: in this case, it does not matter whether tsin.sin_port is non-zero or not, because unless p->trtp is non-zero, the && won't be satisifed |
| 19:06.09 | putnopvut | kpfleming: (After reading only the initial message of the thread) In my case, though, short-circuit rules don't affect the if. The first condition is true, so the second condition is evaluated. |
| 19:06.37 | putnopvut | kpfleming: re-read the last part of my first message :) |
| 19:06.42 | kpfleming | well, scratch that, my analysis was wrong |
| 19:07.51 | kpfleming | so really, it comes down to how the compiler turned it into assembly code |
| 19:08.07 | kpfleming | oh wait... duh. |
| 19:08.16 | putnopvut | All I'm trying to say is that in the case I saw the error, it is a completely legitimate warning. |
| 19:08.21 | kpfleming | yes |
| 19:08.28 | kpfleming | so... i'll go back to my hole now |
| 19:08.43 | putnopvut | Since I'm working on other things though, I'm leaving the fixing of it as an exercise to someone else though. It shouldn't be difficult to fix. |
| 19:09.13 | kpfleming | well, posting it here probably won't result in anyone recording it for future work |
| 19:09.24 | putnopvut | Yeah, I could open a bug report. |
| 19:10.09 | putnopvut | Now, there is another jump based on uninitialized value warning from my valgrind run that seems incorrect to me, so I'm ignoring that one. |
| 19:10.40 | russellb | oej: testing which branch? |
| 19:10.55 | oej | Plain 1.4 |
| 19:11.05 | russellb | 10k channels with 1.4?! |
| 19:11.06 | oej | So 1.6 should do better I guess, with hash tables |
| 19:11.08 | putnopvut | wow |
| 19:11.19 | russellb | I was sure it was some later 1.6 version ... |
| 19:11.40 | russellb | ref counted objects and hash tables in both chan_sip and ast_channels, reworked scheduler, among other things |
| 19:11.54 | putnopvut | oej: as an experiment with 1.6, you can also try turning on extenpatternmatchnew in extensions.conf if you have a really huge dialplan that you're using to test. |
| 19:12.08 | russellb | yeah, and new dialplan code in 1.6, too |
| 19:12.09 | putnopvut | If you're using a small dialplan, though, that won't be very helpful. |
| 19:12.13 | oej | Well, that's not part of the test. But thanks for the advice. |
| 19:12.17 | seanbright | i know what he did... |
| 19:12.21 | seanbright | #define ast_mutex_lock(x) |
| 19:12.27 | putnopvut | lol |
| 19:12.28 | seanbright | :-) |
| 19:12.39 | russellb | cd 1.4 ; svn merge .../branches/1.4 .../trunk . |
| 19:12.42 | seanbright | that's great. can't wait to see the branch. |
| 19:12.43 | oej | seanbright: STAY AWAY from my systems! |
| 19:12.45 | russellb | it's 1.4 with some patche s:-p |
| 19:12.47 | oej | :-) |
| 19:13.16 | Corydon76-lap | seanbright: I have a way for you to reduce memory usage: #define struct union |
| 19:13.21 | seanbright | heh |
| 19:13.24 | kpfleming | ouch |
| 19:13.34 | seanbright | make sure to put that in _private.h |
| 19:13.46 | Corydon76-lap | and speed up the program: #define while if |
| 19:14.04 | seanbright | #define private public |
| 19:14.12 | seanbright | but i digress |
| 19:14.28 | Corydon76-lap | although if you really want to screw somebody up: #define main Main |
| 19:14.36 | putnopvut | Hmm, #define while if may cause compilation problems. |
| 19:14.37 | russellb | #define malloc(n) ({ static void buf[1024]; (buf);}) |
| 19:14.39 | russellb | that'll do |
| 19:14.44 | seanbright | how many channels can you get without your changes? |
| 19:14.52 | seanbright | oej: ^^^ |
| 19:14.54 | seanbright | 5k? |
| 19:15.34 | oej | I have no changes yet... Plain 1.4. |
| 19:15.39 | seanbright | oooh |
| 19:15.43 | seanbright | i missed that part |
| 19:15.47 | seanbright | thought it was based on 1.4 |
| 19:16.02 | Qwell | oej: what kind of server is this on? |
| 19:16.18 | oej | Current issue is that we only have a poor gigabit switch that can't take the traffic |
| 19:16.42 | oej | Will document properly soon... Still testing. |
| 19:16.48 | seanbright | cool |
| 19:18.13 | oej | We're currently testing the effect of running many meetme's. |
| 19:20.12 | oej | What really surprised me was the network interface. In all previous tests, the network gave up around 1.000 channels. I need to find out much more here. |
| 19:21.16 | Qwell | oej: newish kernel? there may have been very significant changes there.. |
| 19:21.33 | oej | Maybe. |
| 19:21.46 | oej | (Can't access the servers now) |
| 19:23.04 | kpfleming | Quote: "___________" doesn't work properly, this is a major severity issue. |
| 19:23.25 | seanbright | falves? |
| 19:23.30 | putnopvut | What is that a quote from, Mr. Fleming? |
| 19:24.04 | kpfleming | mantis 15773 |
| 19:24.08 | kpfleming | M15773 |
| 19:24.10 | MuffinMan | [assigned] [Asterisk] Applications/app_fax 0015773: SendFax without reliable transmission. reported by ChrisDE (Karma: -8) https://issues.asterisk.org/view.php?id=15773 |
| 19:24.38 | seanbright | kpfleming: you are not allowed on mantis for the rest of the day |
| 19:25.18 | kpfleming | some stuff is auto-assigned to me now |
| 19:25.22 | kpfleming | on porpoise |
| 19:25.44 | seanbright | unacceptable |
| 19:25.45 | seanbright | heh |
| 19:26.28 | Qwell | kpfleming: there was another that seanbright changed a bit ago. |
| 19:26.36 | file | kpfleming: porpoises! |
| 19:26.41 | Qwell | "This warning message shows up because I'm doing something wrong. MAJOR!" |
| 19:27.06 | kpfleming | nice |
| 19:27.06 | seanbright | i silently changed from major to minor |
| 19:27.08 | Qwell | $20 says you guess the reporter in 1. :P |
| 19:27.17 | kpfleming | i know which issue it is |
| 19:27.19 | seanbright | we need a checkbox that says "don't send update e-mail on change" |
| 19:27.32 | *** join/#asterisk-dev Marquis42 (n=bfhbmw0@65-127-126-34.dia.static.qwest.net) |
| 19:34.56 | tzafrir_laptop | any svn equivalent to 'git am' (merge to a branch from a mailbox of patches, with proper commit messages and such) |
| 19:34.59 | tzafrir_laptop | ? |
| 19:35.53 | kpfleming | nope |
| 19:36.01 | Corydon76-lap | tzafrir_laptop: honestly, that sounds like feature creep |
| 19:36.08 | Qwell | mailbox? |
| 19:36.15 | seanbright | a place where mail is stored |
| 19:36.26 | Corydon76-lap | but that isn't important now |
| 19:36.33 | Qwell | like...an Mbox file? that's a bit absurd, heh |
| 19:36.58 | *** part/#asterisk-dev zeroHalo (n=zeroHalo@173.13.92.17) |
| 19:37.02 | tzafrir_laptop | Qwell, or whatever alternative container of patches |
| 19:37.20 | Qwell | woah, an Asterisk from tzafrir |
| 19:37.25 | Qwell | Asterisk branch too |
| 19:37.36 | Corydon76-lap | tzafrir_laptop: if someone proposed that to SVN, I think it'd be rejected as something with limited usefulness that would better be implemented outside of the tool, with scripting |
| 19:38.05 | kpfleming | even for git, it's a script |
| 19:38.16 | kpfleming | the 'git' command is really just a wrapper around various lower level tools |
| 19:38.28 | Corydon76-lap | Ah, that makes sense |
| 19:38.30 | tzafrir_laptop | Frankly I don't care that it is a script. It works :-) |
| 19:38.53 | Corydon76-lap | svnmerge works, too |
| 19:39.23 | tzafrir_laptop | svnmerge (or any other script around) happens to do that kind of merge? |
| 19:39.42 | kpfleming | probably, but i've never had a use for one so i've never looked |
| 19:43.24 | seanbright | i did download svn bisect from cpan |
| 19:43.28 | seanbright | that is pretty damn cool |
| 19:43.39 | Qwell | what's that? |
| 19:43.52 | seanbright | helps you find a commit using binary search |
| 19:44.20 | seanbright | you initialize it on a check out and check for a change in a file (like changing an if statement) |
| 19:44.26 | seanbright | then you say "no, it was before this" |
| 19:44.30 | seanbright | and it gets the middle rev |
| 19:44.35 | seanbright | rinse and repeat |
| 19:45.03 | Qwell | neat |
| 19:45.07 | seanbright | only used it a couple times, but it's paid for itself the few times i have used it |
| 19:45.57 | russellb | paid for itself... you're so witty |
| 19:46.17 | *** join/#asterisk-dev leifmadsen (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 19:46.17 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 19:46.48 | Corydon76-lap | feeds grape bread to seanbright |
| 19:46.56 | russellb | O.o |
| 19:47.03 | putnopvut | bisecting is a great method of finding the target revision. When I saw that git had a standard bisect method, I thought that was pretty nifty. |
| 19:47.10 | Corydon76-lap | russellb: want some? |
| 19:47.25 | russellb | nah, not right now, full from lunch |
| 19:49.11 | *** join/#asterisk-dev hugorebelo (n=hugorebe@200-171-132-124.completo.com.br) |
| 19:51.32 | Corydon76-lap | tzafrir_laptop: you going to tell us in which header file those functions are in, for #15017 ? |
| 19:52.07 | tzafrir_laptop | M15017 |
| 19:52.09 | MuffinMan | [feedback] [Asterisk] General 0015017: armv5tel: ast_expr2.c op_func.c implicit declerations reported by tzafrir (Karma: +121.50) https://issues.asterisk.org/view.php?id=15017 |
| 19:52.29 | tzafrir_laptop | Once I re-assemble that SheevaPlug machine |
| 19:52.42 | Corydon76-lap | Heh |
| 19:53.51 | Corydon76-lap | You're not cross-compiling for that? |
| 19:55.38 | tzafrir_laptop | no |
| 20:03.43 | *** join/#asterisk-dev mvanbaak (n=michiel@asterisk/contributor-and-bug-marshal/mvanbaak) [NETSPLIT VICTIM] |
| 20:03.43 | *** mode/#asterisk-dev [+o mvanbaak] by ChanServ |
| 20:03.57 | *** join/#asterisk-dev JunK-Y (n=junky@modemcable206.69-20-96.mc.videotron.ca) [NETSPLIT VICTIM] |
| 20:04.37 | *** join/#asterisk-dev errotan (n=errotan@62.201.123.1) |
| 20:04.38 | *** join/#asterisk-dev Laureano (n=Laureano@190.245.108.131) [NETSPLIT VICTIM] |
| 20:09.10 | kpfleming | someone on -users says he's setting up a small VOIP setup, with "roughly 5 users" |
| 20:09.10 | kpfleming | just how hard it is to count to 5? |
| 20:10.31 | snuff-home | hopes to get some input on a patch in 3 hrs.. just a few compiler warnings and my brain isn't working :) |
| 20:17.54 | Corydon76-lap | kpfleming: doesn't that usually depend upon whether a VP commits to being part of the test or not? |
| 20:18.24 | kpfleming | eh? |
| 20:18.30 | Corydon76-lap | many times, they're still waffling on that question after the test is over... |
| 20:19.16 | kpfleming | i have no idea what you are talking about... |
| 20:19.39 | Corydon76-lap | Think PHB. |
| 20:19.46 | kpfleming | oh |
| 20:19.48 | kpfleming | ok |
| 20:21.14 | file | thinking is so last week |
| 20:21.14 | Corydon76-lap | The kind of people who only want to be part of a test group, if the test is successful... |
| 20:23.36 | Corydon76-lap | It's more amusing when you're consulting and don't have to deal with that same person 8 hours a day |
| 20:41.08 | *** join/#asterisk-dev bbryant (n=brett@68.208.65.50) |
| 20:41.08 | *** mode/#asterisk-dev [+o bbryant] by ChanServ |
| 21:41.06 | *** join/#asterisk-dev korihor (n=korihor@190.205.251.61) |
| 21:46.29 | *** join/#asterisk-dev itamarjp (n=itamar@189-015-69-079.xd-dynamic.ctbcnetsuper.com.br) |
| 21:49.34 | *** join/#asterisk-dev tzafrir_laptop (n=tzafrir@bzq-179-75-202.static.bezeqint.net) |
| 21:49.34 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 22:19.09 | kpfleming | quote: "I'm not sure if I needed to make and make install after applying that patch." |
| 22:24.40 | leifmadsen | o.O |
| 22:24.46 | kpfleming | yeah |
| 22:24.58 | kpfleming | if we wrote asterisk in perl, then you could patch your running copy |
| 22:25.21 | leifmadsen | that'd be hawt |
| 22:25.31 | leifmadsen | runs off to have a walk with the g/f and to get some frozen yogurt! |
| 22:28.59 | Corydon76-lap | kpfleming: you can already patch the running copy of Asterisk, as long as you aren't trying to patch the core |
| 22:29.15 | Corydon76-lap | just unload the module, load it again |
| 22:29.43 | Corydon76-lap | has done it before with app_voicemail.so |
| 22:35.56 | tzafrir_laptop | you'd still have to run 'make' |
| 22:46.47 | *** join/#asterisk-dev ZX81 (n=Matt_Rid@121.74.12.42) |
| 22:47.42 | *** join/#asterisk-dev ZX81 (n=Matt_Rid@121.74.12.42) |
| 22:48.23 | *** join/#asterisk-dev ZX81 (n=Matt_Rid@121.74.12.42) |
| 22:50.09 | *** part/#asterisk-dev ZX81 (n=Matt_Rid@121.74.12.42) |
| 22:54.20 | *** join/#asterisk-dev ZX81 (n=Matt_Rid@121.74.12.42) |
| 23:04.41 | philipp64 | moy: what's the status of app_bundle for 1.6 and dahdi? |
| 23:05.18 | moy | philipp64: hu? why should I know that? |
| 23:05.35 | philipp64 | moy: 15175 |
| 23:06.22 | kpfleming | M15175 |
| 23:06.23 | moy | oh, that crap ... |
| 23:06.24 | MuffinMan | [new] [Asterisk] Applications/NewFeature 0015175: [patch] v.110 dialin support for ISDN channels reported by dwmw2 (Karma: +23.00) https://issues.asterisk.org/view.php?id=15175 |
| 23:07.13 | moy | well the guy with the link to test never got back, sorry about not updating the bug tracker, so, I have no way to test it anymore, we should close it |
| 23:07.57 | philipp64 | if you can get me all the files, I can line up a tester... |
| 23:09.53 | moy | app_v110.c in the bug report is all needed ... and I'd need access to the tester box to probably teak libpri and dahdi |
| 23:10.25 | philipp64 | ok. |
| 23:12.56 | moy | the trunk-rev patch of course, the other is the original outdated code |
| 23:27.45 | snuff-work | mm anyone access pastebin.ca ? |
| 23:29.41 | tzafrir_laptop | snuff-home, not me. |
| 23:30.19 | snuff-work | k i'm not going mad then :) |
| 23:30.42 | philipp64 | and all the rest of app_bundle.tar.gz compiles and works fine? |
| 23:31.38 | philipp64 | wait, isn't app_waitfordigits ... nope, still not part of trunk... |
| 23:33.56 | *** join/#asterisk-dev Mw3 (i=mw3@mw3.hu) |