| 00:23.26 | *** join/#asterisk-dev infobot (ibot@208.53.50.136) |
| 00:23.26 | *** topic/#asterisk-dev is Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- Tier 2 and 3.14159265 support is in #asterisk -=- Check out our blog! blogs.asterisk.org -=- Follow on Twitter at @AsteriskDev |
| 05:54.13 | snuff-work | mm.. your guys asterisk SSL cert.. it has the 'bad' signing key from the CAs.. |
| 05:54.39 | snuff-work | most nightly / canary builds now complain about *.asterisk.org cert being 'insecure' =) |
| 05:57.30 | snuff-work | also.. gerrit doesn't appear very 'happy' atm |
| 06:09.17 | *** join/#asterisk-dev CELYA_ (~Thunderbi@2a01:e0a:d:1111:5ce9:cac0:410f:f992) |
| 07:44.58 | *** join/#asterisk-dev jkroon (~jkroon@165.16.203.126) |
| 08:28.00 | *** join/#asterisk-dev Tim_Toady (~fuzzy@snf-33276.vm.okeanos.grnet.gr) |
| 09:33.02 | file | it's been brought up twice before, it's in IT's hands |
| 09:33.22 | file | as for Gerrit can you be specific? seems fine here and I didn't get any monitoring alerts |
| 11:56.54 | *** join/#asterisk-dev CELYA_ (~Thunderbi@2a01:e0a:d:1111:7c52:2f21:5761:c4eb) |
| 14:13.41 | *** join/#asterisk-dev kharwell (kharwell@nat/digium/x-wgbhoanxsrvrcucy) |
| 14:13.41 | *** mode/#asterisk-dev [+o kharwell] by ChanServ |
| 14:17.36 | *** join/#asterisk-dev jkroon (~jkroon@165.16.204.173) |
| 14:50.34 | *** join/#asterisk-dev DivideBy0 (~DivideBy0@unaffiliated/divideby0x0) |
| 14:50.34 | *** mode/#asterisk-dev [+o DivideBy0] by ChanServ |
| 14:57.43 | *** join/#asterisk-dev bford (uid283514@gateway/web/irccloud.com/x-biniajryorzfvtti) |
| 14:57.43 | *** mode/#asterisk-dev [+o bford] by ChanServ |
| 15:07.44 | *** join/#asterisk-dev rmudgett (rmudgett@nat/digium/x-ynlfrezxtlagcdsa) |
| 15:07.44 | *** mode/#asterisk-dev [+o rmudgett] by ChanServ |
| 16:10.10 | *** join/#asterisk-dev jameswf (uid27319@gateway/web/irccloud.com/x-lvkvlhbtysfwmjya) |
| 16:42.42 | *** join/#asterisk-dev CELYA_ (~Thunderbi@2a01:e0a:d:1111:59d6:ce9:7c92:d344) |
| 18:40.06 | seanbright | rmudgett: i would like your review on https://gerrit.asterisk.org/#/c/asterisk/+/10013/ when you have some free time |
| 18:40.19 | seanbright | i promise not to abandon it if you say something i don't like :D |
| 18:41.07 | rmudgett | Did you have a following review that was going to use those conversion routines? |
| 18:41.22 | seanbright | i didn't feel there was a point to doing that until this was accepted |
| 19:02.33 | rmudgett | seanbright: There. Feel better? |
| 19:02.44 | seanbright | rmudgett: very slightly |
| 19:05.33 | jkroon | seanbright, i'd drop the memset, you're already returning an error so the caller should not trust the result, however, it's probably safer to leave the memsets. |
| 19:06.12 | seanbright | jkroon: heh... that was a rollercoaster of emotions |
| 19:06.23 | seanbright | change it! don't change it! |
| 19:06.38 | seanbright | my reasoning for the memset is that we also return error conditions from ast_sockaddr_parse() |
| 19:06.47 | seanbright | but we rarely check it |
| 19:07.44 | jkroon | as I say ... it's safer to rather just zero it anyway, just in case. |
| 19:08.55 | seanbright | you're talking specifically about the memset() in the not IPv4, not IPv6 case, correct? not the ones from the previous patch sets? |
| 19:09.10 | jkroon | seanbright, yes, in the final patch. |
| 19:09.47 | jkroon | seeing that i don't think we deal with anything other than inet or inet6 it shouldn't ever execute, but bar that one day when we have inet8 ... |
| 19:09.56 | seanbright | what is the phrase everyone uses in reference to code that is not often hit? |
| 19:09.59 | seanbright | non-optimal? |
| 19:10.12 | seanbright | off-nominal |
| 19:10.23 | seanbright | it's the off-nominal code path, so i think it's fine |
| 19:10.48 | jkroon | agreed. |
| 19:11.06 | rmudgett | off-broadway? |
| 19:11.17 | seanbright | let me look it up |
| 19:11.56 | seanbright | yes, off-nominal |
| 19:12.07 | seanbright | sometimes hyphenated, sometimes not |
| 19:12.16 | coreyfarrell | if the error path could leave addr uninitialized then we would want attribute_warn_unused_result on the function declarations, force callers to check for -1. |
| 19:13.00 | rmudgett | This would be an error return that is unlikely to be checked normally. |
| 19:13.36 | seanbright | in the case of this patch, if we return an error, the passed structure is zero'd out |
| 19:13.45 | seanbright | which should be fine |
| 19:13.55 | seanbright | unless i am missing something |
| 19:14.07 | coreyfarrell | yes I think that's best for those functions. |
| 19:15.04 | seanbright | is now known as off-nominal |
| 19:16.54 | off-nominal | and no one else had it registered... fantastic |
| 19:17.06 | rmudgett | off-nominal: Wednesday seems pretty nominal to me. :) |
| 19:47.27 | *** join/#asterisk-dev CELYA_ (~Thunderbi@2a01cb0583d31d0094933db4924becf5.ipv6.abo.wanadoo.fr) |
| 20:19.35 | off-nominal | is calling a function through a function pointer dramatically more expensive then calling a function directly? |
| 20:20.15 | off-nominal | i assume you lose the benefit of potential compiler optimizations |
| 20:22.34 | rmudgett | I wouldn't expect it to be any more expensive than other function calls. |
| 20:30.41 | off-nominal | rmudgett: thanks |
| 21:57.39 | *** join/#asterisk-dev sh_smith (~sh_smith@cpe-76-174-26-91.socal.res.rr.com) |
| 21:58.43 | cresl1n | IIRC it is more expensive |
| 21:59.26 | cresl1n | There can be a lot of memory cache misses when calling through a function pointer |
| 22:00.02 | cresl1n | You also lose any potential code condensing that the compiler can make pre-runtime to get the code tighter/closer together and more closely optimized |
| 22:00.13 | file | he speaks, albeit delayed |
| 22:01.08 | cresl1n | yeah, sorry |
| 22:01.11 | cresl1n | long day |
| 22:01.59 | file | cresl1n: tomorrow is a new one! and I'm sure it too will be long! you can't escape, muahahahahaha! |
| 22:08.17 | rmudgett | coreyfarrell: I thought you had added some NULL pointers to struct ast_channel_tech so additional callbacks could be added for new features. |
| 22:08.29 | rmudgett | Apparently that didn't happen. |
| 22:08.56 | coreyfarrell | rmudgett: no that was AST_MODULE_INFO |
| 22:19.36 | *** join/#asterisk-dev sh_smith (~sh_smith@cpe-76-174-26-91.socal.res.rr.com) |
| 22:25.37 | *** join/#asterisk-dev sh_smith (~sh_smith@cpe-76-174-26-91.socal.res.rr.com) |
| 22:56.57 | *** part/#asterisk-dev kharwell (kharwell@nat/digium/x-wgbhoanxsrvrcucy) |