**** BEGIN LOGGING AT Tue Apr 28 02:59:59 2020 Apr 28 03:35:33 ynezz et al: https://github.com/openwrt/openwrt/commit/0b894193737c957eddbdff47ae8a01276d08bc06.patch fixes the elfutils build problem Apr 28 03:35:55 build/run tested on a meraki mr24 Apr 28 05:48:26 russell--: thanks for test Apr 28 06:33:11 build #194 of mediatek/mt7629 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org/master/images/builders/mediatek%2Fmt7629/builds/194 Apr 28 08:36:46 blocktrron: thanks for fixing the SFP issues! Apr 28 09:07:19 f00b4r0: My request is being handled by my contact at MikroTik currently to obtain the bitmask values. It was passed on to SW dept. Apr 28 09:07:35 while at it, I also requested SFP details for boards like rb922 etc .. Apr 28 09:15:54 what's the policy in OpenWrt regarding BUILD_PATENTED? Apr 28 09:17:30 a fair amount of patented stuff is in the packages feed currently. Apr 28 09:22:37 while not sued; do build; done Apr 28 09:22:39 :) Apr 28 09:24:21 anyway, BUILD_PATENTED defaults to n Apr 28 09:28:23 in other words, I would simply assume, that the patented stuff should be put behind this config option and shouldn't be built/distributed by default Apr 28 09:28:44 mangix: care to provide some examples? Apr 28 09:35:58 ynezz: ffmpeg is the biggest one Apr 28 09:37:27 Why should users (who download official binaries) in the whole world suffer from the USA-only patents? Apr 28 09:37:46 They're not USA only Apr 28 09:42:16 Anyway, I'm wondering whether to put BUILD_PATENTED for ffmpeg. This would make minidlna unavailable. Last I looked the other packages can build without ffmpeg. Apr 28 09:43:10 I pushed a minidlna replacement, but it's relatively huge. Apr 28 09:51:52 default builds of ffmpeg should be avialable Apr 28 09:52:11 maybe disable whatever is problematic by default and add a full build variant Apr 28 09:52:52 seeing minidlna go would be proplematic... miniupnpd has been neutered too already Apr 28 09:53:01 so after loosing proper upnp support, we'd loose dlna support too Apr 28 09:56:53 xback: thanks! I'll have a question for you in a few minutes (in a call atm) Apr 28 09:58:01 jow: gerbera is a replacement that I pushed Apr 28 09:58:10 it's huge though Apr 28 09:59:33 yes, and has no luci support Apr 28 09:59:36 I asked in ffmpeg-dev about this. Their stance is that there are no guarentees on ffmpeg being patent-free, even with --disable-all Apr 28 09:59:54 so we losse ui and availability on lower end devices Apr 28 10:00:14 yes. And that is an issue Apr 28 10:44:08 build #348 of mediatek/mt7623 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org/master/images/builders/mediatek%2Fmt7623/builds/348 Apr 28 10:48:36 miniupnpd neutered? Apr 28 10:57:57 Seems people have issues with it. Haven't looked too much into it. Apr 28 10:58:10 at some point it stopped working properly with playstations etc. Apr 28 10:58:20 yeah because of IGDv2 iirc Apr 28 10:58:35 you can configure that in the config file nowadays Apr 28 10:58:38 afaik Apr 28 10:58:51 option igdv1 '1' Apr 28 11:01:53 ah good to know! Apr 28 11:02:08 I had indeed noticed that it failed to work with recent devices (including playstation) Apr 28 11:03:12 the problem is that what you call recent devices don't implement igdv2 and you need to downgrade to igdv1 Apr 28 11:03:48 yeah Apr 28 11:04:11 noted, thanks Apr 28 11:04:13 it should be made defualt Apr 28 11:04:25 curious to see if ps5 or xbox series x will implement v2 but I doubt it Apr 28 11:04:43 iirc someone enabled igdv2 eventually, broke the majority of device and thne refused to fix it Apr 28 11:05:08 and it is still not working properly ootb and that extra config option was never added to the ui Apr 28 11:05:22 because as usually, nobody cares Apr 28 11:05:46 on 18.06 I have a "Enabled IGDv1 mode" though? Apr 28 11:05:56 in Luci Apr 28 11:06:20 (still running 18.06 for now, mission critical confinement situation etc ;) Apr 28 11:07:14 yes because ldir added support for it eventually Apr 28 11:07:55 https://github.com/miniupnp/miniupnp/pull/282 Apr 28 11:09:01 thanks Apr 28 11:23:06 It seems it makes sense to compile it with igdv2 enabled but faking v1 by default. Apr 28 12:02:21 oooh? did I actually fix / do something useful ? Apr 28 12:07:13 build #343 of mediatek/mt7622 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org/master/images/builders/mediatek%2Fmt7622/builds/343 Apr 28 12:51:57 ynezz: ping Apr 28 12:52:08 xback: sup :) Apr 28 12:52:17 ynezz: as you also run imx6 boards, maybe you have a clue about this one: https://pastebin.com/raw/6n217iSf Apr 28 12:52:43 as soon as pcie ports get populated .. duplicate sysfs entries are trying to be made Apr 28 12:53:00 this is clean master on a gw5400 Ventana Apr 28 12:53:48 this is some regression on 5.4.35 ? Apr 28 12:54:27 I haven't tested yet on 4.19. Just fired up my imx6's again today Apr 28 12:54:36 currently building master on 4.19 in the background here Apr 28 12:54:46 but maybe you already knew about this or so :-) Apr 28 12:54:55 seeing it for the first time Apr 28 12:55:09 It only occurs when placing devices in the pcie sockets Apr 28 13:00:43 I've just booted 5.4.34 with `01:00.0 Network controller: Qualcomm Atheros AR958x 802.11abgn Wireless Network Adapter (rev 01)` no issue Apr 28 13:01:27 IIRC ventana has some pcie controller Apr 28 13:04:33 here is my bootlog http://paste.ubuntu.com/p/rTDRbw5xFV/ Apr 28 13:05:39 thnx Apr 28 13:05:48 I'll try to dig a bit Apr 28 13:22:34 5.4.35 is fine as well Apr 28 13:30:24 ynezz: does your imx hw use dwc as pcie host? Apr 28 13:52:32 xback: arch/arm/boot/dts/imx6q-apalis-ixora.dts -> arch/arm/boot/dts/imx6qdl.dtsi -> compatible = "fsl,imx6q-pcie", "snps,dw-pcie" Apr 28 14:22:50 hm, how is this supposed to work "dnsmasq: found already running DHCP-server on interface 'eth0' refusing to start, use 'option force 1' to override" ? Apr 28 14:23:14 I understand, that disabled dhcp makes sense in this case Apr 28 14:23:24 but why I'm without dns as well? :) Apr 28 14:33:44 ynezz: https://pastebin.com/raw/kdX7XpN1 ventana on kernel 4.19 Apr 28 14:35:10 xback: seems like a bisect afternoon :) Apr 28 14:35:17 indeed :p Apr 28 14:53:40 ynezz: it is supposed to skip the dhcp setup for this particular interface, dnsmasq itself should still start normally Apr 28 14:53:40 ynezz: mangix: If you know of a component which uses patented technologie we should make it depend on BUILD_PATENTED Apr 28 14:54:05 the patents for some video and audio codecs already expired Apr 28 14:54:44 we should probably not build parts where we known that it violates patents and someone also enforces them in the build bots Apr 28 14:55:18 but shipping the source code should be fine, as far as I understood this source code does not violate a patent only the binary, but I am not a lawyer Apr 28 14:56:00 jow: yeah, it starts, but dns doesnt work Apr 28 14:56:02 I am not aware that a community driven open source project was suied for patent violation Apr 28 14:56:09 ynezz: then it seems bugged Apr 28 15:00:18 `service dnsmasq restart` fixed it :p Apr 28 15:00:43 ynezz: maybe some dnssec chicken egg timing cosmic ray issue? Apr 28 15:04:28 or sunray Apr 28 15:24:48 ynezz: found it .. https://patchwork.kernel.org/patch/11213957/ Apr 28 15:24:55 same was needed in pcie dwc controller Apr 28 15:25:29 this is for the irq error Apr 28 15:30:27 mangix: thanks a lot. installing 'libjson-c-dev' did the trick. I just had to additionally do: '$ sudo ln -s /usr/include/json-c /usr/include/json' Apr 28 18:16:41 build #389 of ath25/generic is complete: Exception [exception interrupted] Build details are at http://buildbot.openwrt.org/master/images/builders/ath25%2Fgeneric/builds/389 blamelist: Petr ?tetiar , ?lvaro Fern?ndez Rojas , Luiz Angelo Daros de Luca , Adrian Schmutzler Apr 28 18:16:41 , Antonio Quartulli , DENG Qingfang , Pawel Dembicki Apr 28 18:17:20 build #195 of mediatek/mt7629 is complete: Exception [exception interrupted] Build details are at http://buildbot.openwrt.org/master/images/builders/mediatek%2Fmt7629/builds/195 blamelist: Petr ?tetiar , ?lvaro Fern?ndez Rojas , Luiz Angelo Daros de Luca , Adrian Schmutzler Apr 28 18:17:20 , Antonio Quartulli , DENG Qingfang , Pawel Dembicki Apr 28 18:17:50 build #279 of ramips/mt7621 is complete: Exception [exception interrupted] Build details are at http://buildbot.openwrt.org/master/images/builders/ramips%2Fmt7621/builds/279 blamelist: Petr ?tetiar , ?lvaro Fern?ndez Rojas , Luiz Angelo Daros de Luca , Adrian Schmutzler Apr 28 18:17:50 , Antonio Quartulli , DENG Qingfang , Pawel Dembicki Apr 28 22:30:16 build #276 of ipq40xx/generic is complete: Failure [failed] Build details are at http://buildbot.openwrt.org/master/images/builders/ipq40xx%2Fgeneric/builds/276 blamelist: Petr ?tetiar , ?lvaro Fern?ndez Rojas , Luiz Angelo Daros de Luca , Adrian Schmutzler , Antonio Apr 28 22:30:16 Quartulli , DENG Qingfang , Pawel Dembicki **** ENDING LOGGING AT Wed Apr 29 02:59:57 2020