**** BEGIN LOGGING AT Sat Apr 19 02:59:59 2014 Apr 19 03:16:06 nbd: I have a fairly reproducible crash in the current netifd version Apr 19 03:16:28 nbd: It is caused by a proto 'none' section which is used to set the MAC address of a device Apr 19 03:17:52 reproducible across a wide variety of devices? Apr 19 03:19:48 nbd: It crashes in netifd_dump_status() (at ubus.c:653) because iface->l3_dev.dev is not set Apr 19 03:19:54 DonkeyHotei, just checked on one device Apr 19 03:20:03 A TL-WR1043ND v1 Apr 19 03:20:49 I can provide the core file and netifd binary if further information is needed Apr 19 03:20:55 then it could be a device-specific issue Apr 19 03:21:35 I doubt that, it looks more like a issue caused by my special config Apr 19 03:22:16 needs testing on other hardware, then Apr 19 03:26:33 I think the commit "interface: always delete l3 dev on proto down" might have caused this, the following commits fixed this up for proto-shell and proto-static, but not for proto none... Apr 19 03:27:51 plausible hypothesis Apr 19 03:28:13 you could bisect Apr 19 03:31:48 Bisecting is a bit overkill for 5 commits ;) I think I'll wait for a comment from nbd as it were his commits, but for now I'll be afk for a while Apr 19 03:35:52 he hasn't responded to a ping from someone else close to 24h ago Apr 19 04:20:17 build #548 of uml is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/548 Apr 19 04:33:40 build #583 of orion is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/583 Apr 19 06:22:21 build #32 of pxa is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/pxa/builds/32 Apr 19 07:41:09 build #32 of omap is complete: Failure [failed shell compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/omap/builds/32 Apr 19 09:01:15 build #34 of x86_64 is complete: Failure [failed compile_1] Build details are at http://buildbot.openwrt.org:8010/builders/x86_64/builds/34 Apr 19 09:54:59 neoraider: thanks, i will look into this Apr 19 09:59:36 nbd r40529 trunk/package/network/config/netifd/Makefile Apr 19 09:59:36 netifd: update to the latest version, fixes some more issues related to l3_dev handling Apr 19 09:59:38 neoraider: ^ Apr 19 10:21:35 nbd: morning Apr 19 10:21:50 nbd: do you think you could be the one pushing my ntfs-3g patch? Apr 19 10:21:56 i mean patch updating package Apr 19 12:31:24 zajec: hm, that package needs some cleanup Apr 19 12:32:22 btw. i'm currently writing the ucwmp code that generates the cwmp object tree from the json data Apr 19 13:12:20 nbd: what adri was pinging you about was two-fold Apr 19 13:12:41 i haven't been following irc much this week Apr 19 13:12:47 didn't spend much time at the computer Apr 19 13:13:14 1) ath9k has no way to set perm_addr after boot, and you maintain it Apr 19 13:13:49 2) the atheros (non-ar71xx) target needs reworking Apr 19 13:15:22 1 is something that needs to be handled in either mac80211 or cfg80211 Apr 19 13:15:23 not ath9k Apr 19 13:15:52 2: correct, though somebody mentioned he was working on upstreaming the target Apr 19 13:16:12 i have no time to do any work on the legacy atheros target Apr 19 13:16:41 but you can accept patches, right? Apr 19 13:16:57 yes Apr 19 13:17:07 though patches for 1 should go through upstream Apr 19 13:17:28 i can backport patches that were submitted Apr 19 13:17:34 the main thing wrong with 2 is there is no way to differentiate different boards Apr 19 13:17:52 right Apr 19 13:17:57 and the way to fix that is to start using DT Apr 19 13:18:08 i'd like to avoid infecting that target with the mips-machine stuff as well Apr 19 13:19:12 that will be a lot of work Apr 19 13:20:19 right, but adding mips-machine mess will make it even harder to push for upstream inclusion Apr 19 13:20:47 as for 1, ath9k sets a perm_addr on boot and it' Apr 19 13:21:01 as for 1, ath9k sets a perm_addr on boot and it's never touched again Apr 19 13:22:04 on my board, that's a random macaddr Apr 19 13:22:24 what board is that? Apr 19 13:22:27 blogic insists on having it set in userspace Apr 19 13:22:38 netgear dgn3500 Apr 19 13:23:38 and the eeprom data is loaded how? Apr 19 13:24:32 there is a config partition in the flash, but the macaddr is not in it, but in the uboot env instead Apr 19 13:25:34 actually maybe i should grep the partition to make sure Apr 19 13:28:04 so you could try to submit a patch to linux-wireless to allow cfg80211 or mac80211 to change perm_addr via sysfs Apr 19 13:28:25 the field is stored in the wiphy struct anyway Apr 19 13:28:43 and the driver does not need to be notified of changes to it Apr 19 13:28:53 because the effective mac address is per-vif anyway Apr 19 13:29:56 once the patch gets an ack from johannes berg, i will merge it into openwrt Apr 19 13:37:17 i grepped, it's not in the partition Apr 19 13:37:50 i suggested sysfs myself Apr 19 13:38:33 [Fri 2014-04-18 02:39:03 PM PDT] Well, I dunno if sysfs is going to get accepted in Apr 19 13:38:33 [Fri 2014-04-18 02:42:23 PM PDT] as i said i'm open to ideas Apr 19 13:40:22 nbd: ^ Apr 19 13:41:48 thing is, it's the driver that reports the address via sysfs, not 80211 Apr 19 13:43:38 nvm, i misremembered Apr 19 14:47:02 nbd: nice! :) Apr 19 14:47:21 nbd: what's wrong with ntfs-3g? Apr 19 14:48:05 nbd: and does it inteference with version update? Apr 19 14:48:22 nbd: could we bump version anyway, since it probably doesn't make things worse? Apr 19 14:48:45 nbd: ntfs-3g seems to be just a Makefile... i guess you don't like sth about it. what is that? Apr 19 14:57:11 DonkeyHotei: cool, well, do it via sysfs then. I don't mind being wrong. : Apr 19 14:57:15 DonkeyHotei: :P Apr 19 15:37:52 zajec: i'm talking about the crap for old versions in the package Apr 19 15:38:38 okay :) Apr 19 15:42:19 after it gets cleaned up, i'm considering moving it to trunk Apr 19 15:52:18 hello world Apr 19 15:53:23 i own my wd net 900 - ubicom ip8 based router Apr 19 15:53:59 i know same IP8 was used in dlink 827 and 857 Apr 19 15:54:30 and I have found that dlink released sources that include openwrt.tar.gz based on kamikaze Apr 19 15:59:43 could anyone with openwrt experience try to compile those sources? It fails for me http://sourceforge.net/projects/officialdlinkfirmware/files/DIR-857/ Apr 19 16:00:58 build #50 of sunxi is complete: Failure [failed shell] Build details are at http://buildbot.openwrt.org:8010/builders/sunxi/builds/50 Apr 19 17:18:20 hi. since some week the wifi command (on bb-trunk) kills my wifi, only sometimes it repairs it http://pastebin.com/ULG3cfCr Apr 19 17:25:14 rundfreifunk: what version did you test last? Apr 19 17:26:39 my last build is 3-4 days old Apr 19 17:27:07 nbd r40530 trunk/feeds.conf.default * feeds.conf.default: the telephony feed is now on git.openwrt.org Apr 19 17:28:49 DISTRIB_REVISION="r40512" Apr 19 17:28:56 rundfreifunk: i committed a fix earlier today Apr 19 17:29:00 rundfreifunk: please test that Apr 19 17:29:14 ok, will do that Apr 19 17:44:11 build #33 of mpc83xx is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/mpc83xx/builds/33 Apr 19 18:23:46 build #32 of adm8668 is complete: Failure [failed shell compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/adm8668/builds/32 Apr 19 18:28:00 build #32 of cns3xxx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/cns3xxx/builds/32 Apr 19 18:38:51 build #539 of rb532 is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/539 Apr 19 18:39:36 build #539 of ppc44x is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/539 Apr 19 18:39:42 build #508 of sibyte is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/508 Apr 19 19:36:22 build #572 of lantiq is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/lantiq/builds/572 Apr 19 21:40:04 luka r40531 trunk/target/linux/imx6/profiles/120-gateworks.mk * imx6-gateworks: enable new canbus packages Apr 19 21:40:05 luka r40532 trunk/package/boot/kobs-ng/ Makefile patches/002-add-init-size-param.patch * kobs-ng: Add support for overriding size Apr 19 21:40:06 luka r40533 trunk/tools/flex/Makefile * [tools] flex: upgrade to version 2.5.39 Apr 19 21:40:07 luka r40534 trunk/tools/mpc/Makefile * [tools] mpc: upgrade to 1.0.2 Apr 19 21:40:08 luka r40535 trunk/tools/scons/Makefile * [tools] scons: upgrade to 2.3.1 Apr 19 22:11:00 build #518 of avr32 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/518 Apr 19 22:14:10 build #446 of iop32x is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/446 **** ENDING LOGGING AT Sun Apr 20 02:59:58 2014