**** BEGIN LOGGING AT Thu Jul 30 02:59:58 2015 Jul 30 04:27:14 trine: No. Jul 30 04:27:29 trine: that is not me Jul 30 05:27:25 build #69 of ppc40x is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/ppc40x/builds/69 Jul 30 06:42:01 kaloz r46520 trunk/package/kernel/mwlwifi/Makefile * mwlwifi: upgrade to latest rev with fixed firmware Jul 30 06:42:17 kaloz r46521 branches/chaos_calmer/package/kernel/mwlwifi/Makefile * mwlwifi: upgrade to latest rev with fixed firmware (backport of r46520) Jul 30 07:10:18 xiangfu, ok thanks for letting me know Jul 30 07:14:53 xiangfu, I was hoping it was you as I am trying to persuade COMFAST to provide the openwrt source code they are using in their products, at the moment they are not doing so and that information (jiangyaohui@software.comfast) was something I extracted from the device as it was booting up over my serial connection and it very loosely looked a bit like your nick Jul 30 07:15:40 sorry if it caused any confusion Jul 30 07:26:02 :) Jul 30 07:26:11 a lot of China company using openwrt. Jul 30 07:26:15 like the mi.com Jul 30 07:27:09 a very big company in China, this one: https://en.wikipedia.org/wiki/Xiaomi Jul 30 07:29:31 the China unicom. China mobile. also using OpenWrt in their product. Jul 30 08:17:06 xiangfu, yes I guess they can see the benefits of openwrt but they should also if they use it provide any GPL'ed source code but COMFAST are at the moment refusing sadly to do so Jul 30 08:20:46 xiangfu, I have bought one of their products but as it stands at the moment although it does have openwrt installed by COMFAST it has many features disabled Jul 30 08:21:42 xiangfu, this makes it little more than a piece of junk at the moment although the hardware looks quite good Jul 30 08:53:58 Yes Jul 30 08:54:18 cyrus r46522 trunk/package/network/ services/dnsmasq/patches/100-fix-dhcp-no-address-warning.patch services/dnsmasq/patches/210-dnssec-improve-timestamp-heuristic.patch services/dnsmasq/Makefile * dnsmasq: Bump to dnsmasq2.74 Jul 30 11:49:31 hello Jul 30 11:49:54 does anybody know if nbd keeps a feature list/todo list on the mt76 driver? Jul 30 11:54:51 i don't Jul 30 11:55:09 at the moment i'm primarily debugging the code, looking for performance/stability issues Jul 30 11:55:12 and fixing those Jul 30 12:17:40 nbd, does it support the tx beamforming I'm reading about in the specs for these wifi cards? Jul 30 12:20:17 it doesn't support that yet Jul 30 12:20:27 not sure if beamforming is very effective with 2x2 chipsets anyway Jul 30 12:21:21 it probably does not make a significant difference Jul 30 12:21:28 I see Jul 30 12:21:56 thanks Jul 30 12:48:11 nbd: hey! :) Jul 30 12:48:24 do you need any more info from this WD MyNet 750? Jul 30 12:48:36 I gotta take the kit back tomorrow for RS232 <-> TTL Jul 30 12:48:48 until my bits of kit arrive from China Jul 30 13:02:38 not at the moment. haven't had time to look into it some more Jul 30 14:19:07 kk Jul 30 14:19:11 I'll take the kit back then Jul 30 18:51:08 build #67 of avr32 is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/67 Jul 30 19:39:12 Guys, I've a problem where the realtime connections display in luci isn't resolving addresses to hosts. nslookups of the same addresses work correctly. Any ideas on how to investigate this? Jul 30 19:40:01 KevinDB: lua -lnixio -e 'print(nixio.getnameinfo("8.8.8.8"))' Jul 30 19:40:10 KevinDB: is this working? Jul 30 19:40:23 jow_laptop: Will check Jo Jul 30 19:42:08 returns: nil -6 Unrecognized address family or invalid length Jul 30 19:42:42 thats the reason then Jul 30 19:42:49 likely a musl problem Jul 30 19:42:59 since it used to work just fine with uclibc Jul 30 19:43:40 jow_laptop: I didn't think it was my imagination that it used to work....though I was beginning to doubt myself! Jul 30 19:45:40 jow_laptop: Should I log a ticket for this? Jul 30 19:46:31 will look into it right away Jul 30 19:48:00 jow_laptop: Fantastic! Thank you so much. If one day I get to be 1% of you guys I'll be pleased :-) Jul 30 19:49:48 musl libc indeed behaves differently Jul 30 19:49:59 it errors if the passed buffer is too large Jul 30 19:50:09 while other libcs only bail out if its too small Jul 30 19:53:57 jow_laptop: Is that the call to getnameinfo ? Jul 30 19:56:08 yeah Jul 30 20:08:35 jow_laptop: Anything I can do to help? Jul 30 20:14:02 KevinDB: you could test a patch Jul 30 20:14:57 jow_laptop: The compiler awaits :-) Jul 30 20:18:01 http://luci.subsignal.org/~jow/0001-luci-lib-nixio-pass-exact-sockaddr-length-to-getname.patch Jul 30 20:18:08 you need to apply that to feeds/luci/ Jul 30 20:18:49 jow_laptop: Will get onto it right now. Jul 30 20:19:15 then make package/luci-lib-nixio/{clean,compile}; scp bin/*/packages/luci/luci-lib-nixio*.ipk root@192.168.1.1:/tmp; ssh root@192.168.1.1; rm /tmp/opkg-lists/*; opkg install /tmp/luci-lib-nixio*.ipk Jul 30 20:19:39 finally repeat the getnameinfo() lua test case above Jul 30 20:24:41 jow_laptop: Sorry Jo, I'm being stupid. Where should I put that patch file and how to apply? I'm assuming a patch dir in package/feeds/luci/luci-lib-nixio ? Jul 30 20:27:21 cd feeds/luci/; wget -O- http://luci.subsignal.org/~jow/0001-luci-lib-nixio-pass-exact-sockaddr-length-to-getname.patch | git am Jul 30 20:29:09 jow_laptop: Thanks - working on it. Jul 30 20:34:39 jow_laptop: Aha! Names instead of addresses! Jul 30 20:35:17 ok, I'll push it then. Jul 30 20:35:31 on your end, cd feeds/luci/ and git reset --hard HEAD^ Jul 30 20:35:39 to throw away that premature commit Jul 30 20:36:24 jow_laptop: Ok, understand. Thanks for fabulous service as usual! Jul 30 20:37:49 you can now run scripts/feeds update luci Jul 30 20:37:59 I pushed the change Jul 30 20:38:06 * KevinDB grins Jul 30 20:39:29 jow_laptop: Done, fantastic. Thank you so much. Jul 30 23:49:13 build #71 of ar71xx.nand is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx.nand/builds/71 **** ENDING LOGGING AT Fri Jul 31 02:59:58 2015