**** BEGIN LOGGING AT Sat Jul 09 02:59:58 2016 Jul 09 03:43:03 build #8 of socfpga is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/socfpga/builds/8 Jul 09 04:05:10 build #261 of ep93xx is complete: Failure [failed shell_13] Build details are at http://buildbot.openwrt.org:8010/builders/ep93xx/builds/261 Jul 09 05:03:14 build #331 of lantiq is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/lantiq/builds/331 Jul 09 05:44:34 build #328 of malta is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/malta/builds/328 Jul 09 06:20:31 build #285 of xburst is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/xburst/builds/285 Jul 09 06:31:50 build #272 of lantiq.xrx200 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/lantiq.xrx200/builds/272 Jul 09 08:03:59 build #311 of ramips.mt7620 is complete: Failure [failed shell_17] Build details are at http://buildbot.openwrt.org:8010/builders/ramips.mt7620/builds/311 Jul 09 08:46:31 https://github.com/blogic what is it with the ringleaders of lede all being queer german developers? I'm just saying, I might be seeing things that aren't there. Please guys - put one dev on constantly merging / qa-ing everything that LEDE does. Make it known that openwrt is the real bleeding edge. Jul 09 08:47:24 https://lwn.net/Articles/650816/ That's what FFMPEG vs LIBAV did, and they won, libav ran out of steam always behing "behind" ffmpeg for the most part. The goal is to keep everything compatible with lede and previous openwrt releases to an extent. Jul 09 08:47:34 Where is libav in debian now? Jul 09 08:47:44 Where will lede be next year? hopefully the same place. Jul 09 08:47:48 history. Jul 09 08:49:33 hopefully i'll have another 2 pull requests done today. Jul 09 09:43:16 hojuruku: slow down Jul 09 09:43:34 we're trying to work out the differences, solve/resolve things in openwrt as well Jul 09 09:44:25 it's great you have the opinion on openwrt being the bleeding edge still, but currently that's not really the case - i'm merging the appropriate patches from lede onto the openwrt github tree, and try to add things where can Jul 09 09:44:47 saying that lede devs are german queers doesn't really help :P Jul 09 09:45:15 it's a bit unprofessional to say the least ;p Jul 09 09:45:58 so the best would be to stay away from these comments, keep calm and keep coding ;p Jul 09 09:48:08 yeah i was coding hence the delay replying Jul 09 09:48:37 nearly finished, should be a good reference script to help all the other package maintainers upgrade to procd. I've worked out how to do it absolutely right. Jul 09 09:48:52 10k boot script, and it's clean with only necessary comments. Jul 09 09:49:13 add it to the procd wiki page when im done. Jul 09 09:50:05 wigyori: make your hard work be known so people don't defect to lede. try and get a LKML announcement published saying that Jul 09 09:52:30 wigyori: I talked about loosing the family home, due to gays saying you must not disrespect those who sexually molest infants in two court cases, my 72 year old father breaks it down on a live interview here: truthnews.com.au/web/radio/story/cultural_marxism_101 - i was very offended when jow said it was homophobic to critize those who have sex with babies. Anyone who has sex with children is a pedophile and deserves to be abused even Jul 09 09:52:31 if they had on national tv http://archive.org/details/TwoDadsAreBetterThanOne identify as gay and after their conviction. My last ever mention of the matter will be the only thing I said to jow about this since the ban " i never vilfiied homosexuals, i only vilified homosexuals who have sex wtih babies (Mark Newton and Peter Truong). Anyone who says you can't hate baby sex anything is a baby sex lover. So the question is have you acted Jul 09 09:52:31 on your desire to have sex with children yet?" Jul 09 09:53:54 that,s the logic behind how I got pissed off been banned and labeled falsely. anti-pedophilia activism is a good thing that needs to be done when the police aren't picking up the slack. it's not a label to make you an outcast or wrong. Jul 09 09:54:35 https://en.wikipedia.org/wiki/Anti-pedophile_activism wikipedia doesn't agree with me though. ok back to work... Jul 09 09:56:20 I don't talk off topic too much, but with me being a political refugee http://bit.ly/aurefugee2 you can guess why it came out, not everyone is a snowden or assange style political refugee. Jul 09 09:56:38 i'll commit my changes... Jul 09 10:03:06 https://github.com/hojuruku/openwrt-packages/blob/mini_snmpd/net/mini_snmpd/files/mini_snmpd.init now to test the thing. It's even more fancy than igmpproxy's script Jul 09 10:03:21 next to fix some issues with siproxd Jul 09 10:16:06 hojuruku: pedophile and similar topics will get anyone a ban here as well Jul 09 10:41:47 wigyori: what do you think of lede and tsinghua teaming up? can you imagine the american government doing the same what they did to huwei and banning any openwork derivative from being deployed on government environments. Here's one of their great dev books : https://www.amazon.com/Reverse-engineering-tutorial-examples-Paperback/dp/7302262837?ie=UTF8&*Version*=1&*entries*=0 Jul 09 10:42:14 ^openwork^openwrt Jul 09 12:17:08 who said you have to be a developer to submit a patch? is that a new rule? Jul 09 12:17:36 oh sorry wrong channel someone said that on #openwrt Jul 09 13:09:39 EEE 802.11: hostapd_public_action - action=0 IEEE 802.11: handle_action - unknown action category 4 or invalid frame - google isn't giving me much answers about this, relates to an openwrt patch and happens when the wifi goes down. Jul 09 14:52:12 https://github.com/openwrt/packages/pull/2941 <- well that took longer than expected. All i can say is thank god procd isn't as bad as systemd ;) Jul 09 15:18:10 https://github.com/openwrt/packages/pull/2938#issuecomment-231539272 hope i helped. is there a better way of re-doing commits with signed off tags? (this is not for me) Jul 09 15:46:28 How to recompile the toolchain after patching the GCC makefiles (https://gist.github.com/0x0c/45bb3a2fc802b4b6a8af9c5d188a7b46) I've tried every "make toolchain" clean, prepare, install I could think of to no avail... Jul 09 15:52:03 FergusL: do you know what I do for a fresh build? Jul 09 15:53:39 FergusL: make sure my target/myplatform/linux/config-4.4 has all my changes in it, i think make kernel_menuconfig does it these days... and just rm -rf bin tmp build_dir staging_dir; ./scripts/feeds install -a (or make menuconfig will do it); and make -j4 Jul 09 15:53:55 you don't get fresher than rm -rf but it takes a bit longer :) Jul 09 15:54:51 hojuruku: well the thing is that I absolutely do NOT want to have to redo a fresh build from start Jul 09 15:55:02 FergusL: also update the PKG_RELEASE that might help next time but if you are changing gcc, it bootstraps the compiler, so the rm method is good Jul 09 15:55:48 FergusL: well if you added patch files to the gcc patches directory where the openwrt Buildroot Makefile is (not the gcc one) it makes no difference. Jul 09 15:56:25 but meanwhile I ruined it with a "make toolchain clean", that removed everything :/ so I went for "make toolchain/install" Jul 09 15:56:27 FergusL: i do that all the time, and I don't loose anything. I don't even used ccache, that would speed it up though. Jul 09 15:56:55 good thing I kept a clean backup of everything I had compiled before! Jul 09 15:57:40 FergusL: trust me, tgz your gcc working directory after you made patches in that dir and try total clean rebuild trick. there are a whole bunch of state files in $BUILDROOT/tmp too. Jul 09 15:58:21 hm I didn't put the patch in the patch folder, I just patched using the "patch" command from the root directory :s then ran "make menuconfig" to enable GCC 4.9 and then just wanted to rebuild the host toolchain, to continue with cross-compiling simple programs that need GCC 4.9 Jul 09 15:58:32 those directories i named, tmp bin staging_dir build_dir is used duing the build. in the instructions to apply patches, get sources etc of course is in the packages directory that you don't delete. Jul 09 15:59:02 ok Jul 09 15:59:06 FergusL: try adding a patch using quilt, or one that looks like it was made by quilt like what I do :P Jul 09 15:59:48 FergusL: at some point if you want to share your work, your patch has to build from a clean buildroot, and just what's in the package directory (it's makefile and /files / patches) Jul 09 15:59:54 so it must be faster next time but I'm afraid I lost everything with "make toolchain clean" Jul 09 16:00:21 I forgot about quilt when I wanted to apply the patch Jul 09 16:01:11 https://blogs.gentoo.org/lu_zero/2015/02/20/demotivation-fud-and-why-i-still-contribute-to-libav/ <- reminds me of lede, if they don't come back here's how they'll complain and say we are victims of FUD. Still spiting them with merging everything that do worthwile will turn out successful in the long term as it did for ffmpeg. Jul 09 16:01:14 but it seemed weird to me that nothing like "make toolchain/gcc/clean" or prepare, or install or compile worked, it is documented somewhere but definitely didn't work Jul 09 16:01:56 https://lwn.net/Articles/650816/ that guy above most definitely doesn't speak for #gentoo ask them what they think of libav :P Jul 09 16:03:20 FergusL: i just told you what works for me. if you are changing something that's as big as gcc, and it's bootstrapped by the build root and set's up a cross-compiler i'd let all the makefiles do their things from scatch after making your changes. Jul 09 16:03:34 https://wiki.openwrt.org/about/toolchain#openwrt_build_system_make_targets like here Jul 09 16:03:43 hojuruku: I believe you're right, indeed Jul 09 16:05:11 FergusL: if you are going to make changes in the buildroot build_dir/ back them up as you go with patches diff -Naur vs a tree that was unpacked before you started making changes and moved somewhere else. Or do it properly and learn how to use quilt. Jul 09 16:05:25 (which i haven't :P - too many things to do :P) Jul 09 16:05:50 FergusL: https://wiki.openwrt.org/doc/devel/patches Jul 09 16:07:10 change the two directory names to a and b - and add a signed off if you are going to submit a patch to the dev mailing list. Or then there's github. Jul 09 16:09:11 FergusL:the toolchain targets work - there are some other target like prepare that will built toolchain too Jul 09 16:18:12 plntyk: probably it didn't work because it didn't know about the changes, and again I only patched makefiles, so probably the make command doesn't check the makefiles themselves ! Jul 09 16:18:30 hence why it's on the buildroot part that I must have done that? like adding patches that patch the tooclahin makefiles Jul 09 16:20:27 you could do make dirclean to clean all stuff except .config files Jul 09 16:20:55 the patch you posted / revert to have the linaro 4.9 gcc back looked ok Jul 09 16:22:58 plntyk: .config files are kernel related, right? so it cleans host toolchain + packages ? Jul 09 16:23:20 still trying to understand all those systems for building! Jul 09 16:24:10 the .config files normally contain only package configuration for the target and some other stuff like selected compiler version, libc etc Jul 09 16:25:13 it does contain the kernel package selection but the kernel symbols are either defined via package makefiles or in target/linux/generic/config- and target/linux/target-arch//config- Jul 09 16:25:38 package/kernel/linux/modules/*.mk do contain the config symbols for modules Jul 09 16:26:51 diclean clean host toolchain and packages Jul 09 16:26:55 *dirclean* Jul 09 16:44:18 build #279 of kirkwood is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/kirkwood/builds/279 Jul 09 17:16:36 Thanks a lot for the details plntyk Jul 09 17:40:25 FergusL: check kernel config after setting up your .config (make menuconfig) by doing make kernel_menuconfig Jul 09 18:30:16 https://github.com/hojuruku/lede-source/tree/p_procd/package/system/procd when I finished the procd patch I'm working on should I give it to openwrt, lede or both? Jul 09 18:30:44 (openwrt devs reply only) Jul 09 18:31:04 afaik there is only one upstream for procd Jul 09 18:31:45 lede and owrt are both downstream Jul 09 18:38:23 it's implemented now as patch files :) Jul 09 18:39:57 i'm going to patch zram-swapfs to play nice with procd_zram_tmpfs and fix up the broken makefile deps for procd using ZRAM, and of course the patch to use busybox instead. I been running it as a week. someone upstream at procd put through a patch to gobble up half your ram with zram on tmp. 4mb is enough for me, so i'll have options to let people choose size allocated by procd Jul 09 18:40:34 the patch is already done, just the makefile and zram-swap to go. Jul 09 18:45:00 hojuruku if I understand you correctly, a few days ago you said you won't work with lede devs or I didn't understand? Jul 09 18:45:28 hojuruku: procd maintainers are part of lede, so it should definitely be on the lede list. you can cc openwrt as well if you want to, but it won't make much of a difference Jul 09 19:11:21 Devastator: we'll i dont' want to but i know who wrote procd. Are they only updating procd upstream on lede or on openwrt's project git? Jul 09 19:11:28 i don't know i can't ask them Jul 09 19:14:39 hojuruku you seem to have good skills, I just don't understand your problems with them.. anyway, this is offtopic Jul 09 19:15:47 yeah way too of topic. Jul 09 19:16:19 i'm making a version of openwrt to do what the fon pinapple did, but not over wifi. it's my hobby project Jul 09 19:16:25 hence the need to work on rp-ppoe ;) Jul 09 19:17:00 anyway it's late i'll go lurk guys later. Jul 09 19:18:59 hojuruku: lede git automatically replicates trees like procd to git.openwrt.org Jul 09 19:26:56 there was someone else working on rp-pppoe too Jul 09 19:27:06 nighty Jul 09 19:27:31 idk how far he got Jul 09 19:46:21 build #271 of mcs814x is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/mcs814x/builds/271 Jul 09 22:20:41 build #353 of realview is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/realview/builds/353 Jul 09 23:57:21 build #269 of brcm47xx.mips74k is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/brcm47xx.mips74k/builds/269 Jul 10 00:45:35 build #329 of malta is complete: Failure [failed shell_18 compile_9] Build details are at http://buildbot.openwrt.org:8010/builders/malta/builds/329 Jul 10 00:45:56 build #321 of rb532 is complete: Failure [failed shell_18 compile_9] Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/321 Jul 10 00:46:16 build #271 of octeon is complete: Failure [failed shell_18 compile_9] Build details are at http://buildbot.openwrt.org:8010/builders/octeon/builds/271 Jul 10 02:33:56 build #243 of mpc85xx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/mpc85xx/builds/243 **** ENDING LOGGING AT Sun Jul 10 02:59:58 2016