**** BEGIN LOGGING AT Thu Dec 13 03:00:02 2018 Dec 13 06:48:26 build #394 of mvebu/cortexa53 is complete: Failure [failed cleanold] Build details are at http://phase1.builds.lede-project.org/builders/mvebu%2Fcortexa53/builds/394 blamelist: Martin Schiller , Stijn Tintel , Jo-Philipp Wich , INAGAKI Hiroshi Dec 13 07:22:58 Hauke: Thanks for the info :) Dec 13 07:29:21 build #1151 of x86/geode is complete: Failure [failed cleanold] Build details are at http://phase1.builds.lede-project.org/builders/x86%2Fgeode/builds/1151 blamelist: Martin Schiller , Stijn Tintel , Jo-Philipp Wich , INAGAKI Hiroshi Dec 13 08:10:35 build #1173 of brcm47xx/generic is complete: Failure [failed cleanold] Build details are at http://phase1.builds.lede-project.org/builders/brcm47xx%2Fgeneric/builds/1173 blamelist: Martin Schiller , Stijn Tintel , Jo-Philipp Wich , INAGAKI Hiroshi Dec 13 08:20:25 lede 17.01: any plan for a kernel bump? Dec 13 08:21:18 xes: isn't it called 18.06 or something? Dec 13 08:22:19 18.06.1 Dec 13 08:52:03 xes: you mean a major bump? no Dec 13 09:01:18 xes: I started a week ago trying to bump the kernel to the latest 4.4. but it's a huge amount of work due to all changes up to the very latest which i currently dont have the time for :) Dec 13 09:04:32 build #817 of sunxi/cortexa53 is complete: Failure [failed] Build details are at http://phase1.builds.lede-project.org/builders/sunxi%2Fcortexa53/builds/817 blamelist: Martin Schiller , Stijn Tintel , Jo-Philipp Wich , INAGAKI Hiroshi Dec 13 09:07:04 xback: no doubt you've noticed the latest CVE mentioning bump already :-) Dec 13 09:08:58 :) Dec 13 09:12:36 ldir: it's already running here :) Dec 13 09:13:38 xback: could you make it run harder 'cos it's cold here this morning and the birds are all fluffy ;-) Dec 13 09:14:18 ldir: do you know the Nemo project, the new high voltage cable between GB and Belgium? Dec 13 09:15:25 oh my god.... you mean there are european electrons migrating here!!!! Dec 13 09:15:48 As soon as my server bumps the kernel, the current flow reverses :p Dec 13 09:16:18 lol Dec 13 09:31:35 ldir: I bumped the musl update patch yesterday to include the latest, but the patch naming is identical. make sure you replace it in your staging :) Dec 13 09:32:01 noted :-) Dec 13 09:43:29 laughs at strncpy(dest, source, sizeof(source)); somewhat missing the point Dec 13 09:58:49 ldir: :-) Dec 13 10:01:09 jow: i found that samba patch... i was looking at wrong place (samba4 in packages repository) Dec 13 10:01:11 it's https://git.openwrt.org/?p=openwrt/openwrt.git;a=commitdiff;h=fd569e5e9d0a46ea957cb253e97a4b3ea8c2c540 Dec 13 10:01:30 i didn't know we have samba4 in packages repository and samba36 in pacakges/network/services/ Dec 13 10:05:05 rmilecki: and the samba4 version doesn't build on mac os which is a bit frustrating. Dec 13 10:05:57 currently has other battles - involving qnap - and no doubt code quality of the strncpy standard exhibited above. Dec 13 10:10:11 ldir: good luck :) Dec 13 10:10:18 i'll try to work on that samba36 thing Dec 13 10:26:59 jow: xback: yep, I was refering to latest kernel in 4.4 serie.. Dec 13 10:32:20 xes: is anything holding you back to switch to 18.06 ? Dec 13 10:47:12 xback: a few instability/resets on dwc2 driver with ethernet IF devices that i didn't managed to solve Dec 13 10:49:14 I should check it again since my last try was a few months ago.. But you know, when something is so rock solid like 17.01 on Lantiq.. Dec 13 11:00:05 ldir: kernel bumps pushed to my staging Dec 13 11:08:44 jow: is there also a buildbot taking care of 18.06? Dec 13 11:09:21 jow: I would push all kernel bumps to that branch, but would like to keep an eye out afterwards ensuring all is building well :) Dec 13 11:17:32 xback: http://release-builds.openwrt.org/18.06/images/waterfall Dec 13 12:10:48 jow: do you have a moment to discuss https://git.openwrt.org/?p=openwrt/openwrt.git;a=commitdiff;h=fd569e5e9d0a46ea957cb253e97a4b3ea8c2c540 ? Dec 13 12:11:03 jow: did you start new work already? or are you having few days off? :) Dec 13 12:11:23 jow: i want to improve that samba36 solution and i see two ways: Dec 13 12:11:49 1) keep storing entries in /etc/config/samba but they should be verified on boot() Dec 13 12:12:15 2) store those dynamic entries in e.g. /var/run/... and modify samba init.d script to also check for that file Dec 13 12:45:00 https://patchwork.ozlabs.org/patch/1005502/ can be closed if anyone has those rights, the original author agreed that ynezz's method was perfecetly suitable Dec 13 12:46:26 jow: https://patchwork.ozlabs.org/patch/967011/ is superseded by the rpcd exec time changes you did too. Dec 13 12:56:33 hi folks, is there a reason the commits like 0692f0fb55ca5bd2426701b3a9e64e3b7487a7d did not get backportet to branch 18.06? since commit 514ad059ef40b747c93c363403cb49b45dd86e41 mt76 supports the 5ghz interface of those devices and right now its not working because of the missing/wrong package selection. Dec 13 13:14:13 rmilecki: I'd favor to store thme in /var/run or so Dec 13 13:14:36 jow: to don't wear flash too easily? or why? Dec 13 13:15:36 rmilecki: because random hotplug scripts should not write into uswerconfig and because /etc/ is not the proper place for autogenerated ephemeral config Dec 13 13:16:03 gak, "perf" is trying to build a host tool, that needs linux/list.h, Dec 13 13:16:10 jow: ok, thanks Dec 13 13:16:39 rmilecki: another consideration is whether we really want to expose any discovered filesystem as samba share by default Dec 13 13:16:53 jow: it's not by default Dec 13 13:16:56 it's a separated package Dec 13 13:17:00 so I think it's ok Dec 13 13:17:09 ok Dec 13 13:17:45 achterin: no particular reason Dec 13 14:04:13 wrt sdcard image generation, (for sunxi in particular, haven't looked at the others much) it seems to expect a matching board _defconfig in uboot sources. Dec 13 14:04:29 is there a way to just say, "this other uboot config is perfectly compatible" ? Dec 13 14:27:20 ok, perf's hostcompile is ... wonky, and I have no idea how to fix it. the target_cflags aren't used, because it's building for the "host" even though that doesn't seem to make sense for this jevents library, and telling it to use the target cc as the hostcc fails for the _other_ things it builds. Dec 13 16:16:24 did anyone get a chance to try out the new ath10k-ct driver/firmware via patches that were posted a few days ago? Dec 13 16:21:06 greearb__: I've been using it. It works well, thanks! Dec 13 16:24:42 mamarley, thanks, what chipset are you using? Dec 13 16:25:50 greearb__: QCA988x Dec 13 18:00:31 greearb__: I will test as soon as it is pushed to master Dec 13 23:11:07 build #395 of mvebu/cortexa53 is complete: Success [build successful] Build details are at http://phase1.builds.lede-project.org/builders/mvebu%2Fcortexa53/builds/395 Dec 14 00:02:38 build #1152 of x86/geode is complete: Success [build successful] Build details are at http://phase1.builds.lede-project.org/builders/x86%2Fgeode/builds/1152 Dec 14 00:31:52 is the website broken? i can't load the supported devices table: https://openwrt.org/toh/start (no table) Dec 14 00:56:52 build #1174 of brcm47xx/generic is complete: Success [build successful] Build details are at http://phase1.builds.lede-project.org/builders/brcm47xx%2Fgeneric/builds/1174 Dec 14 01:50:10 build #818 of sunxi/cortexa53 is complete: Success [build successful] Build details are at http://phase1.builds.lede-project.org/builders/sunxi%2Fcortexa53/builds/818 **** ENDING LOGGING AT Fri Dec 14 03:00:01 2018