**** BEGIN LOGGING AT Fri Apr 17 02:59:58 2015 Apr 17 03:03:57 build #905 of x86 is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/x86/builds/905 Apr 17 03:52:41 build #265 of ar71xx.mikrotik is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx.mikrotik/builds/265 Apr 17 04:14:19 build #896 of rb532 is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/896 Apr 17 04:37:51 <_bazz> Hello, this topic is outdated.. Apr 17 04:40:15 <_bazz> I'd like to invite someone to please make a response to this post of mine. It details a bunch of.. improvements I've made to Netgear WNR2000v4 support, but .. I'm a little in the dark on how to create a patch and test it before submitting... Also since I have a lot of patch content. Not sure what counts as "logical patch" as asked for in the Submission Guidelines. Thanks. Link: Apr 17 04:40:17 <_bazz> https://forum.openwrt.org/viewtopic.php?pid=272797#p272797 Apr 17 04:41:00 <_bazz> To expound on the logical patch bit, I'm not sure if I should break these improvements up (requiring extra work)... or whatnot.. Please inform and guide me thank you Apr 17 04:41:45 <_bazz> This will be my first submission to a big open source project I'd like to make it count Apr 17 05:01:04 build #916 of ramips is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/ramips/builds/916 Apr 17 05:18:27 build #365 of adm8668 is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/adm8668/builds/365 Apr 17 05:22:43 baz: i think you separated the patches by describing each one with a bold line in your initial post - those logical units there seem good enough Apr 17 05:25:46 bazz: however there is a signed off requirement if you send patches and i am not sure if using some large forum patch is right without asking for a sign off Apr 17 05:28:26 bazz: instead of initial post i mean the post you have linked in irc, sry Apr 17 05:46:42 rmilecki r45472 trunk/target/linux/brcm47xx/base-files/lib/upgrade/platform.sh * brcm47xx: fix handling absolute paths in sysupgrade (for vendor formats) Apr 17 05:51:11 build #762 of mcs814x is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/mcs814x/builds/762 Apr 17 06:03:56 rmilecki r45473 trunk/target/linux/bcm53xx/base-files/lib/upgrade/platform.sh * bcm53xx: fix handling absolute paths in sysupgrade (for vendor formats) Apr 17 06:26:54 <_bazz> plntyk: thanks for your response. I plan on including a sign off when I am ready to submit to the official patch/devel mailing list.. I first need help learning how to create the patch. ie. the proper command from which directory.. where to put the patch.. how to apply it and test it... and since I have multiple patches.. how to apply and test them in sequence. Unless I'm missing something, with that I Apr 17 06:26:56 <_bazz> should have a good patch to submit.. or set :) Apr 17 06:27:03 build #907 of uml is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/907 Apr 17 06:29:26 _bazz: you could use git, and then you can create a new branch with your changes in it and edit/split/combine/play with the patch series (and you have automatic sign-off there) Apr 17 06:31:28 <_bazz> Hm, can you elaborate on the edit/split/etc part? Because right now I am most comfortable with my patch being 1 big patch.. but if segmenting it is better to do, I'm not sure how I would achieve that under git as the devs expect it to be done.. Apr 17 06:31:39 _bazz: the concept behind the "logical patch" is to have the changes necessary to fix one thing in one commit. it makes it easier to verify that change set works correctly, and to revert if not Apr 17 06:32:26 _bazz: also, if you included everything you wanted to fix in a single changeset but one part needed to be tweaked, it would be really hard to separate out later Apr 17 06:32:50 <_bazz> Fishman: thanks. So please correct anything that follows: Apr 17 06:32:53 _bazz: git can even send the patches as email - because most users dont have their mail clients configured in the correct way to send text mail Apr 17 06:33:47 _bazz: if you have a single patch /changeset you can use some mechanics from git add interactive (http://git-scm.com/book/be/v2/Git-Tools-Interactive-Staging) Apr 17 06:34:00 <_bazz> I will git clone openwrt.git, then i will create a new branch, then I will make one logical patch edit, which might apply to several files.. then I commit. I rinse repeat until all logical patches have been made, which will be something like 7 commits. Correct so far? then what? Apr 17 06:34:42 _bazz: from my simple understanding, that's probably correct. i'm unable to help with the details, though Apr 17 06:34:44 _bazz: yes - the add tool can split you change to (extreme) single line commits Apr 17 06:35:09 <_bazz> Well I'm not going to do single line commits (lol wtf) Apr 17 06:35:59 _bazz: well - the led changes are probably in that region adding 1-2 lines or so Apr 17 06:36:11 <_bazz> this is true :D Apr 17 06:36:33 <_bazz> but why use an add tool? Wouldn't I use my own judgement? Apr 17 06:37:27 <_bazz> I think it's safe to say that I determine which lines go into which patch.. but then what will the add tool do for me? Apr 17 06:37:31 with that workflow you can split one large patch into smaller logical pieces instead of doing the manual edits Apr 17 06:37:34 might save time Apr 17 06:37:57 <_bazz> eh, but it's another tool to learn Apr 17 06:38:11 the add is a part of the normal git command Apr 17 06:38:15 * _bazz *shrugs* Apr 17 06:38:39 <_bazz> I only understand git add to add files.. I didn't know it was versatile Apr 17 06:38:47 <_bazz> *whole files Apr 17 06:40:06 <_bazz> hm.. Ok but I'm hoping now for a broader conceptual image of the whole patching process.. Let's say I have my 7 commits in my branch.. Reflecting 7 logical patches... So are there no patch files to test with compiling the kernel?? Or what do I do next? Apr 17 06:40:10 i only discovered the interactive option because I was trying to split patches - there are many way to do things Apr 17 06:41:09 _bazz: "git format-patch" creates a series of .patch files Apr 17 06:41:45 <_bazz> Must I do it after each commit or at the end? Will it matter? Apr 17 06:42:48 <_bazz> Is there a pre-existing OpenWrt page that details these things?? Would be super useful!! If not we or me ought to create one based off of this knowledge Apr 17 06:45:45 _bazz: well this is "normal" git workflow that is described in the official documentation (branching for example: http://git-scm.com/book/en/v2/Git-Branching-Branching-Workflows // full book: http://git-scm.com/book/en/v2) Apr 17 06:46:30 there is even a gamification available: http://pcottle.github.io/learnGitBranching/ Apr 17 06:53:02 <_bazz> Yes but even I am already familiar with branching, I use it in another project quit frequently.. but I still didn't know enough I need a guide.. I still can't put the puzzle together.. you haven't shown me enough Apr 17 06:54:22 <_bazz> you've gotten me hypothetically branching, commiting, making the patch files..Then what do I do with them? Will the patch files be made with the proper -p1 requirement? How do I test the patches? I think I'll just submit them via email afterwards :) Apr 17 06:54:40 <_bazz> The test patches is the biggest question I want the answer to Apr 17 06:55:23 <_bazz> Is there a directory I put the patches in, and they automatically get invoked when I build the kernel? Apr 17 06:56:30 <_bazz> Can I use this method: http://wiki.openwrt.org/doc/devel/patches#adding_or_editing_kernel_patches Apr 17 06:58:54 _bazz: have you already seen https://dev.openwrt.org/wiki/SubmittingPatches ? Apr 17 06:59:00 <_bazz> Maybe to simplify things, you or someone, can imaginarily explain the whole process as you would have done it. Apr 17 06:59:20 <_bazz> Fishman: yes I've seen it. I can't say I've read everything Apr 17 06:59:35 ok. Apr 17 07:00:14 I think you are mixing different concepts atm : read http://git-scm.com/docs/git-format-patch the first sentence in description Apr 17 07:00:49 experiment: do a " git format-patch master" when you are in a branch different from master Apr 17 07:00:52 <_bazz> Yes I am. I don't have a firm grasp of it Apr 17 07:01:21 <_bazz> plntyk: I'm scared *laughs* Apr 17 07:02:14 <_bazz> OK I will do a small patch first to get started.. I'll see you in 10 minutes, if you're here Apr 17 07:02:32 <_bazz> is there a style for the branch name? Apr 17 07:03:29 no - because you use a local git you are totally free to use any style you like Apr 17 07:04:56 _bazz: i believe many patches go through the mailing list. you could browse the devel list to see a lot of examples Apr 17 07:04:59 as for email submission of patches : you can either copy/paste the patch by hand into a correctly (!!!) configured email program (most first time commits fail at this stage) or you use git to send the email too Apr 17 07:05:17 <_bazz> OK my 1st pet peeve, is since I'm new to all of this, I've been having buildroot make itself a toolchain to use.. It's really annoying when I clone a fresh copy of the repo to rebuild the toolchain.. Any tips? Apr 17 07:05:30 blogpost: https://burzalodowa.wordpress.com/2013/10/05/how-to-send-patches-with-git-send-email/ - you have to be subscribed to the mailing list to send patches Apr 17 07:05:56 <_bazz> plntyk: already subscribed thank you. I also have configured gmail to be text-only but I'll double check or use git Apr 17 07:06:29 afaik gmail web is broken for patches Apr 17 07:06:40 <_bazz> why's that? Apr 17 07:06:58 iirc, google always reformats Apr 17 07:07:34 <_bazz> I guess you're right.. I'll avoid it Apr 17 07:08:14 <_bazz> Hopefully you guys/girls ensured that my patch will be successful :) Apr 17 07:08:48 <_bazz> any help on the toolchain question I just posted above? Apr 17 07:09:04 * _bazz goes to google Apr 17 07:09:34 _bazz: http://kernelnewbies.org/FirstKernelPatch has "it line-wraps the mail (even when in plain text mode)" Apr 17 07:09:39 i'm in the "smarter than the average user" category, but that's only because i search and read a lot. and observe what happens here :-) Apr 17 07:10:14 _bazz: you'll likely get the most useful responses from actual devs than from me Apr 17 07:10:27 <_bazz> I'm reading this: http://wiki.openwrt.org/doc/howto/external_toolchain Apr 17 07:10:29 ...just gotta catch one :-) Apr 17 07:10:38 * _bazz is fishing for one lol Apr 17 07:11:11 * _bazz catches "the spirit of the dev-master" A.K.A. the wiki of openwrt Apr 17 07:11:26 <_bazz> should pluralize that Apr 17 07:11:29 well normally you dont need to rebuild the toolchain if you are building for one architecture only Apr 17 07:11:38 <_bazz> yay for being smart Fishman Apr 17 07:12:07 _bazz: only in some things. my comment was of a *relative* intelligence ;-) Apr 17 07:12:29 you have to pay some attention what you can/should delete to make your source tree clean again etc. Apr 17 07:12:35 _bazz: another possibility is to submit patches to tickets, but i think that's not the preferred method Apr 17 07:12:39 <_bazz> plntyk: but I never made a branch... I suppose I could still use this directory... copy everything somewhere else.. then get revert or something back to normal.. then add in the changes Apr 17 07:13:01 Fishman: patches attached to tickets are not accepted - they miss signed-off Apr 17 07:13:37 plntyk: understandable. but i got a typo fixed with a ticket :-P Apr 17 07:14:25 _bazz: you can use git format-patch origin to create a series against the original openwrt web repo Apr 17 07:14:37 * Fishman wanders off to watch a bit of tv before heading to bed. best of luck, _bazz, hope you're able to get your improvements in! Apr 17 07:14:59 ...easily enough! :-) Apr 17 07:15:05 <_bazz> plntyk: any advice on source code formatting as per openwrt standards.. My file is ugly as f*** Apr 17 07:15:12 <_bazz> are either of you devs? Apr 17 07:15:26 <_bazz> Thanks Fishman! Apr 17 07:19:49 sort of - i submitted some patches for router support, buildroot itself and packages . but there are more involved devs that know about project internals like the next stable release date (which is soon-ish) Apr 17 07:24:40 <_bazz> *thumbs up* Apr 17 07:26:31 <_bazz> I'm going to copy the changed files. do a a git stash, and make a branch, then apply the changes as commits, then do the git format-patch. How does that sound Apr 17 07:30:30 sounds good Apr 17 07:31:22 <_bazz> I wouldn't mind an explanation on what kind of patches http://wiki.openwrt.org/doc/devel/patches#adding_or_editing_kernel_patches refers to.. Since we've recently acknowledged this a different concept Apr 17 07:32:54 _bazz: look at the content of target/linux/generic/patches-3.18 and target/linux/ar71xx/patches-31.8 Apr 17 07:33:50 the patches in these 2 directories are appplied to the vanilla kernel that is downloaded from kernel org Apr 17 07:47:54 _bazz: code formatting should match the project you are writing patches for, so patches for the kernel should match kernel codestyle etc Apr 17 07:54:25 <_bazz> KanjiMonster: Thanks. My patches are looking good now no worries Apr 17 07:58:50 build #895 of ar7 is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/ar7/builds/895 Apr 17 08:13:05 rmilecki r45474 trunk/target/linux/brcm47xx/patches-3.18/320-MIPS-BCM47XX-Devices-database-update-for-4.x.patch * brcm47xx: add buttons support for WRT310N v2 Apr 17 08:26:06 oh gcc 5.1 might be released next week Apr 17 08:28:57 build #277 of x86.kvm_guest is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/x86.kvm_guest/builds/277 Apr 17 08:53:33 build #131 of netlogic is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/netlogic/builds/131 Apr 17 09:44:01 build #10 of ath25 is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/ath25/builds/10 Apr 17 09:49:37 build #368 of cns3xxx is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/cns3xxx/builds/368 Apr 17 09:54:14 hardest thing I found to get my head around wsa having to work with quilt to make your patch patches first :| Apr 17 10:00:53 build #754 of gemini is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/gemini/builds/754 Apr 17 10:03:41 build #935 of cobalt is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/935 Apr 17 10:07:56 build #386 of cns21xx is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/cns21xx/builds/386 Apr 17 10:10:51 build #932 of orion is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/932 Apr 17 10:15:56 build #942 of at91 is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/at91/builds/942 Apr 17 10:21:15 build #458 of mvebu is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/mvebu/builds/458 Apr 17 10:26:11 build #279 of ar71xx.nand is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx.nand/builds/279 Apr 17 10:28:34 build #385 of pxa is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/pxa/builds/385 Apr 17 11:03:11 build #865 of xburst is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/xburst/builds/865 Apr 17 11:03:13 build #580 of mpc85xx is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/mpc85xx/builds/580 Apr 17 11:23:05 build #372 of sunxi is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/sunxi/builds/372 Apr 17 11:26:59 build #387 of x86_64 is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/x86_64/builds/387 Apr 17 11:35:51 build #771 of adm5120 is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/adm5120/builds/771 Apr 17 11:36:31 build #395 of malta is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/malta/builds/395 Apr 17 11:41:59 <_bazz> plntyk: Hey man, are you there? Apr 17 11:42:53 <_bazz> I was wondering if I should include the "Signed off" bit in each git commit.. also noticed that in git-log every previous commit has an svn-id but I assume that's because they are direct commits and not like me going to be sent via email Apr 17 11:43:24 each git commit needs a sign off Apr 17 11:43:34 <_bazz> :D yay u responded Apr 17 11:44:42 <_bazz> and each git commit should also have at the top [PATCH] ar71xx : netgear wnr2000v4 ... at least I'm planning on doing this, maybe foregoing the PATCH since it's a bit obvious Apr 17 11:45:12 include patch because afaik the bot would not catch it Apr 17 11:45:21 <_bazz> Okey Apr 17 11:45:58 the mailinglist is tracked and [patch] makes it so they appear at http://patchwork.ozlabs.org/project/openwrt/list/ Apr 17 11:46:55 don't put [PATCH] in your git commit itself though. Apr 17 11:47:00 <_bazz> I'm worried that I'm segmenting my patches too much.. Is it normal if I end having PATCH 1/27 lol Apr 17 11:47:03 you might want to look at "git send-email" Apr 17 11:47:28 <_bazz> or acceptable rather... Apr 17 11:47:50 27 is fine if they make sense. Apr 17 11:47:50 <_bazz> it won't be that bad, it will probably be like 16 *shrugs* Apr 17 11:48:00 it depends. is this series only to add support for wnr2000v4 ? Apr 17 11:48:01 <_bazz> Yes they all make sense! Apr 17 11:49:30 _bazz: one thing to look out for: the tree should remain buildable with each patch added, it isn't enough that it is buildable when all are applied Apr 17 11:49:38 <_bazz> Yes some of the commits are major, like correcting Ethernet support and adding support for a whole section of LEDS.. The rest are bug fixes of existing code such as "buttons need to be active low" ... and then finally some of them are subjective like "led prefix should be netgear instead of wnr2000v4 to keep in flow with all other Boards" Apr 17 11:50:26 <_bazz> KanjiMonster: Thanks I'll be positive to do this Apr 17 11:51:34 Using default firewall config 'fw3 -4 print' reports several 'type is unknown' lines. Any ideas on how to narrow down the cause please? Apr 17 11:52:21 _bazz: are the fixes for your own patches or for already existing code? fixes for your own patches should be merged into the patch that introduces the code Apr 17 11:53:10 <_bazz> fixes for already existing code Apr 17 11:53:19 <_bazz> code that is in latest snapshot Apr 17 11:53:37 <_bazz> I'm working off openwrt.git Apr 17 11:55:19 kevindb: can you give a more specific example? Apr 17 11:55:25 _bazz: if you have patches for things unreleated to wnr2000v4 (at least it sounds like that) then you should submit them separately Apr 17 11:57:13 <_bazz> they are all related directly to wnr2000v4.. O_O Apr 17 11:57:25 nbd r45475 trunk/package/kernel/linux/modules/usb.mk * kernel: mark kmod-usb-bcma/ssb as hidden, they are only selected from ohci/ehci Apr 17 11:57:38 <_bazz> is this a good example of a commit message? Apr 17 11:57:43 <_bazz> ar71xx: netgear wnr200v4 Apr 17 11:57:45 <_bazz> port masks are all off. example: LAN4 led comes on when LAN1 cable is plugged in. Apr 17 11:58:14 <_bazz> Signed off by: Michael J. Bazzinotti Apr 17 11:58:45 <_bazz> just an example heheh Apr 17 11:59:09 bad example i would say Apr 17 11:59:20 <_bazz> OK please correct myself Apr 17 11:59:22 jow_laptop: Frustratingly I don't think so. Archer C7 CC r45474, was restarting firewall manually & noticed the message. Cp'd default firewall config in, reboot. Still the same. Currently rebuilding after dirclean in case some leftovers are causing strangeness. Apr 17 12:00:26 _bazz: read https://www.kernel.org/doc/Documentation/SubmittingPatches part 2) "Describe your changes" Apr 17 12:00:40 kevindb: well the exact wording of the message would be useful Apr 17 12:01:18 _bazz: ar71xx: fix LAN portmasks Apr 17 12:01:35 on wnr2000v4 Apr 17 12:02:21 jow_laptop type is unknown * 6 then the usual iptables -t filter builds Apr 17 12:02:27 hi, got a simple question. Does sysupgrade check a crc sum which is the image header? Or does it only check for a magic number? Apr 17 12:02:50 which is in the image header ^^ Apr 17 12:04:09 jow_laptop: http://pastebin.com/d4DNC86X Apr 17 12:05:18 <_bazz> thanks for the magic touch plntyk Apr 17 12:08:40 kevindb: ah - that means that you have some "option type" in your firewall config which are not expected Apr 17 12:10:06 fw3 warns about options it does not know about, mainly to aid users spotting typos Apr 17 12:10:52 jow_laptop: Even the one cp /rom/etc/config/firewall ? Apr 17 12:11:03 can you pastebin the firewall config? Apr 17 12:11:09 <_bazz> What about formatting improvements.. additional comments, and spelling corrections.. does count as a patch in itself? Apr 17 12:11:34 maybe something changed in the libuci api... Apr 17 12:13:19 jow_laptop: I thought I saw a few uci updates recently...like past few days. Apr 17 12:17:22 _bazz: yes, in case you need to revert a patch because it breaks stuff you don't also want to also revert the formatting/spelling improvements Apr 17 12:18:29 <_bazz> yay :D Apr 17 12:18:30 kevindb: can you pastebin the firewall config even if you think its standard? Apr 17 12:19:23 jow_laptop: Will do... Apr 17 12:21:05 kevindb: nvm, I traced the occurence down Apr 17 12:21:16 kevindb: its caused by firewall rules announced via ubus Apr 17 12:21:35 kevindb: do you run mdns or something else non standard? ipv6 protocol handlers or something? Apr 17 12:25:20 jow_laptop: Avahi for mdns, samba (maybe?) igmpproxy Apr 17 12:26:55 jow_laptop: http://pastebin.com/42yvZ53q Apr 17 12:32:54 jow_laptop: And the default http://pastebin.com/Tg1Vx1ed Apr 17 12:33:18 kevindb: the error you see is not triggered by the uci configuration but by rules announced via ubus Apr 17 12:33:57 kevindb: there are two kinds of things that can announce such rules - one is protocol handlers and others are procd enabled init scripts Apr 17 12:36:52 jow_laptop: Ok, so is there a way to see who's doing all the rule making/breaking,or is it just shut processes down to see when it starts working again :-) Apr 17 12:38:42 either "ubus call network.interface dump" or "ubus call service list" Apr 17 12:38:55 one of both should have an entity describing firewall rules Apr 17 12:41:22 more specifically: ubus call service get_data '{ "type": "firewall" }' Apr 17 12:41:37 this will list rules requested by services Apr 17 12:41:50 if that is empty the rules triggering the problem are caused by a proto handler Apr 17 12:42:01 the "is unknown" error is un critical btw, its a false positove Apr 17 12:45:35 jow_laptop: Thank you very much. I shall investigate anyway 'cos I'm interested Apr 17 12:46:53 okay. I'll look into fixing that arror in the meanwhile (and make it print some more information about the rules origin) Apr 17 12:53:07 jow_laptop: http://pastebin.com/A90UATBF Would suggest I don't need to manually add those rules as igmpproxy does it anyway Apr 17 12:56:07 kevindb: yep, correct Apr 17 12:56:18 and those "type: rule" things are triggering the error messages Apr 17 12:56:31 but thats due to a bug in fw3 which is uncritical Apr 17 13:07:27 jow_laptop: Well that's 2 successes today then - bug in fw3 spotted and now my dishwasher is fixed as well! Result. Apr 17 13:08:34 jow_laptop: Thanks for your help a patience, much appreciated (as is everyone involved with OpenWrt) Apr 17 13:10:31 jow_laptop: as a suggestion, the ubus rules appear with comments like '/* @rule[0] */' Maybe the caller process name could be included in the comment also? Apr 17 13:11:15 cyrus r45476 trunk/package/network/ (6 files in 6 dirs) * network: shorten names of generated interfaces Apr 17 13:11:55 kevindb: yep, I thought about initializing the comment with ubus info, like "igmpproxy[pid] rule" Apr 17 13:12:42 and for proto handlers something like "map46[wan6] rule" Apr 17 13:12:58 well "rule" is redundant, so "ubus:igmpproxy[1234]" Apr 17 13:13:06 that would be clearer I guess Apr 17 13:13:36 jow_laptop: Excellent! It would have saved me a lot of confusion. Apr 17 13:14:05 cyrusff: since you implemented the ubus fw3 rule announcement; do you have further input? Apr 17 13:14:21 jow_laptop: igmpproxy[pid] rule no. ? Apr 17 13:15:00 cyrusff: what we also lack imho is a global setting "option ubus 0" or something. I think some people do not want not explicitely configured rules to appear in their config Apr 17 13:15:12 s/config/ruleset/ Apr 17 13:16:10 yeah maybe Apr 17 13:16:16 for some stuff its essential though Apr 17 13:16:41 i.e. dslite and map need to add custom nat rules Apr 17 13:17:01 otherwise they will not work as expected or at all Apr 17 13:17:08 as for the comments Apr 17 13:17:11 thats probably a good idea Apr 17 13:18:49 The key thing for me as an end user are sensible/obviously sourced rule comments. I thought something was wrong with my system 'cos a) the type warnings from fw3 and b) I was confused by some mystery rules, with numbers that didn't correspond with uci's take on firewall config. Apr 17 13:29:53 <_bazz> I'm concerned that my patch commit message is too long.. is this a concern to you as openwrt developer? Apr 17 13:30:45 <_bazz> http://pastebin.com/cx1LuLPm Apr 17 13:36:42 build #261 of lantiq.xrx200 is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/lantiq.xrx200/builds/261 Apr 17 13:39:32 <_bazz> Whatever moving on Apr 17 13:48:04 luka r45477 trunk/target/linux/ kirkwood/profiles/115-router.mk kirkwood/Makefile kirkwood/profiles/120-plug.mk kirkwood/profiles/110-nas.mk * kirkwood: put uboot-envtools in DEFAULT_PACKAGES Apr 17 13:48:08 luka r45478 trunk/target/linux/ imx6/profiles/110-wandboard.mk imx6/Makefile imx6/profiles/100-generic.mk * imx6: put uboot-envtools in DEFAULT_PACKAGES Apr 17 14:00:18 build #349 of omap is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/omap/builds/349 Apr 17 14:00:29 build #488 of mxs is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/mxs/builds/488 Apr 17 14:07:08 build #718 of octeon is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/octeon/builds/718 Apr 17 14:12:26 kevindb: are in the mood of testing a patch for fw3 ? I can't atm Apr 17 14:14:53 kevindb: http://luci.subsignal.org/~jow/0001-ubus-store-rule-origin-as-comment.patch & http://luci.subsignal.org/~jow/0002-ubus-print-rule-name-when-reporting-errors.patch Apr 17 14:15:10 kevindb: should be enough to put that into package/network/config/firewall/patches/ Apr 17 14:18:00 jow_laptop: I am and I shall :-) Apr 17 14:44:14 jow_laptop: http://pastebin.com/nn5qFHi0 But most likely I'm an idiot. Apr 17 14:44:48 kevindb: nope, I forgot to update the function prototype in the header file# Apr 17 14:46:07 kevindb: I regenerated patch 0002 Apr 17 14:48:11 cyrus r45479 trunk/package/network/ (5 files in 4 dirs) * network: also shorten virtual interface names of ppp and 3g/4g connections Apr 17 14:48:55 jow_laptop: Let me have another try :-) Apr 17 15:03:33 jow_laptop: It compiles, so let me get flashing. If not back in 10 minutes......... ;-) Apr 17 15:03:57 kevindb: you cal also scp the fw3 binary out of the build directory to your routers /tmp Apr 17 15:04:08 and execute it as /tmp/fw3 print Apr 17 15:04:14 that should be enough already Apr 17 15:09:53 jow_laptop: It's wonderful! Apr 17 15:11:30 jow_laptop: rule source displays well in luci, no type errors at the top of 'fw3 print'. Thank you so much. Apr 17 15:11:50 great Apr 17 15:14:43 jow_laptop: now when mystery rules appear you know where to point the finger. That will save many questions I think. Apr 17 15:23:18 yeah, thats important I think Apr 17 15:32:28 <3 from the forum: "I've used both Tomato and DDWRT but OpenWRT is by far the least user friendly software I've ever used." Apr 17 16:17:38 plntyk: the thread is funny. It also looks like the guy never even tried to click the button Apr 17 16:18:05 maybe dd-wrt is user friendlier by resetting itself or something Apr 17 16:49:51 build #832 of kirkwood is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/kirkwood/builds/832 Apr 17 16:53:11 build #374 of mpc83xx is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/mpc83xx/builds/374 Apr 17 17:36:06 build #831 of ar71xx is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx/builds/831 Apr 17 17:54:18 build #839 of ixp4xx is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/ixp4xx/builds/839 Apr 17 18:01:01 <_bazz> Hello, I am making my first patch submission EVER. So please consider looking at my rough draft and please give me some good advice on it :D hehe : http://pastebin.com/3Xzs1NG1 Apr 17 18:07:02 just send it to the mailinglist Apr 17 18:12:11 _bazz: drop things like depends on patch 1/11 Apr 17 18:13:22 _bazz: we usually don't pick patches from the middle randomly ;) Apr 17 18:13:28 _bazz: so it doesn't help about anything Apr 17 18:14:44 _bazz: i guess you may not be native English, but you could work on bit nicer descriptions ;) Apr 17 18:14:49 originally, pressing a button would trigger a release state, Apr 17 18:14:50 and releasing would trigger a press. (Can be observed from hotplug.d) Apr 17 18:15:11 Starting with a big letter, no "," before "and", using 80 chars where possible Apr 17 18:15:34 This (...) at the end (after .) alos looks a bit funny ;) Apr 17 18:28:45 build #271 of brcm63xx.smp is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/brcm63xx.smp/builds/271 Apr 17 18:35:07 <_trine> glp, I have sent you several Emails which you have not replied to,, is there any reason why I have had no reply? Apr 17 18:46:24 <_trine> I have just asked this question on #openwrt without any response so I will ask it here as well, it was also the subject of a enquiry to glp as I believe he is the relevant person Apr 17 18:46:30 <_trine> is there any written code of conduct or protocol regarding how one openwrt maintainer can make additions to the git repo of another maintainers project without informing the actual maintainer they have done so? Apr 17 18:52:16 _trine: for dev.openwrt.org, we have this -> https://dev.openwrt.org/wiki/CommitPolicy Apr 17 18:52:26 build #120 of ramips.mt7620 is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/ramips.mt7620/builds/120 Apr 17 18:53:13 nbd r45480 trunk/target/linux/bcm53xx/patches-3.18/340-ARM-BCM5301X-Add-profiling-support.patch * bcm53xx: add profiling support Apr 17 18:53:19 nbd r45481 trunk/package/network/utils/iptables/Makefile * iptables: remove layer7 leftovers (#19506) Apr 17 18:53:23 nbd r45482 trunk/package/network/config/qos-scripts/Makefile * qos-scripts: drop obsolete depdendency on iptabes-mod-filter (#19506) Apr 17 18:54:43 <_trine> KanjiMonster, in the case of that code being broken what is supposed to happen Apr 17 18:55:09 build #780 of ep93xx is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/ep93xx/builds/780 Apr 17 18:56:26 _trine: start with the direct way; poke the offender and say "hey, don't do that, that's rude" Apr 17 18:57:01 or maybe ask why they did Apr 17 18:57:14 they might have had a reason for doing so Apr 17 18:58:45 <_trine> KanjiMonster, I don't think that is a particularly good idea and from my perspective I cannot imagine a situation where one maintainer would make additions to another maintainers code with the decency of even telling them Apr 17 18:59:00 <_trine> KanjiMonster, I don't think that is a particularly good idea and from my perspective I cannot imagine a situation where one maintainer would make additions to another maintainers code without the decency of even telling them Apr 17 19:00:01 <_trine> to do so would be totally inappropriate and there should be a code of conduct to cover this in the event it happens Apr 17 19:00:39 <_trine> but it appears there isn't Apr 17 19:01:52 <_trine> I have emailed glp and asked him to confirm receipt of the email but it has been met with silence Apr 17 19:03:26 <_trine> as he appears to be the #openwrt Relationship manager, communication and layer8 Apr 17 19:04:04 <_trine> according to https://dev.openwrt.org/wiki/people Apr 17 19:09:05 <_bazz> KanjiMonster: Aren't I supposed to have somwhere a patch-set summary message? Where I ought put that? Apr 17 19:09:56 _bazz: afaik when you send via git send-email you can create a cover leter (summary) Apr 17 19:12:10 _bazz: --cover-letter for format-patch will give you a patch 0/ Apr 17 19:14:15 <_bazz> KanjiMonster: The Patch Submission Guide links explicitly warns againsr using a Patch 0/X Apr 17 19:14:34 <_bazz> I feel like I enforce that thing more than you do :P lol Apr 17 19:14:47 _trine: seriously, talk with him/her. Ask for the reason of doing so. Be polite. Try to get it resolved by yourselves. And only if that fails, try to get a third party involved like glp. Apr 17 19:15:12 _bazz: yes, if the patch 0/x would actually contain a diff/changes that are supposed to be applied Apr 17 19:15:23 but for cover letters it's the accepted solution Apr 17 19:15:30 <_bazz> Is that so! :D Apr 17 19:15:45 as patchwork will then ignore it Apr 17 19:15:55 (well, also the missing diff in it might be rhe cause) Apr 17 19:16:20 <_trine> KanjiMonster, I am always polite and I don't need lessons in being so however when this happens there should be a code of conduct to cover it Apr 17 19:16:48 <_trine> otherwise it could all decay into chaos Apr 17 19:17:25 <_bazz> KanjiMonster: not so! http://www.ozlabs.org/~akpm/stuff/tpp.txt search for "patch 0" Apr 17 19:18:03 "Updated 4 March 2006" <- that's like 7 years old Apr 17 19:18:26 <_bazz> It's directly linked from Patch Submissions :P Apr 17 19:18:41 <_bazz> at least a dead link. I had the cojones to find a mirror Apr 17 19:18:42 <_trine> KanjiMonster, perhaps you have in mind a set of circumstances where one maintainer could alter or make changes to another maintainers code without even telling them never about asking their permission. If you have then please share it with me Apr 17 19:19:03 <_trine> ^^^ mind Apr 17 19:19:29 <_bazz> Anyways, are you sure I should do this Patch 0 business? Is there an alternative? Apr 17 19:20:09 _bazz: you don't have to add a cover letter; feel free not to do one. But it is the accepted practice in openwrt and kernel development Apr 17 19:21:00 <_bazz> that's not what I asked.. Please be helpful I'm looking for an alternative to Patch 0. I would be most delighted :) lol Apr 17 19:21:05 _bazz: https://www.google.de/search?q=%22patch+0%22+site%3Alkml.org - 14,700 results Apr 17 19:21:32 alternatives are to add some comments under the tearline in the individual patches Apr 17 19:21:45 (the -- where the diffstat is) Apr 17 19:21:52 this will be dropped by git Apr 17 19:22:06 <_bazz> https://www.google.de/search?q=%22patch+0%22+site%3Alkml.org#q=%22patch+0%22+site:openwrt.org hehe :) Apr 17 19:22:48 <_trine> is there a current up-to-date list of #openwrt devs Apr 17 19:23:04 it's the old problem, documentation is always outdated Apr 17 19:24:07 <_bazz> I'm not very good at summaries.. "Welp, I basically did a lot of improvements to get base line functionality that we should expect from this model :)" how's that Apr 17 19:24:31 <_bazz> I'll take out the word "a lot of" Apr 17 19:24:35 _trine: as usual, https://dev.openwrt.org/wiki/people is the most up to date there is for svn Apr 17 19:24:36 <_bazz> and welp lol Apr 17 19:24:56 <_trine> KanjiMonster, you are not in that list Apr 17 19:25:25 _bazz: well, you don't need to add one. If all of the patches mention wnr2000v4 in the subject, it's clear that you are fixing support for it Apr 17 19:25:33 _trine: I am, with my svn handle Apr 17 19:25:51 note the list is not ordered Apr 17 19:26:05 <_trine> ah yes Apr 17 19:27:24 <_trine> it would appear that some other people with commit access are not in that list though Apr 17 19:28:11 note this list does not apply to github Apr 17 19:28:38 for example on github my handle is actually KanjiMonster Apr 17 19:29:05 nbd r45483 trunk/package/network/config/netifd/Makefile Apr 17 19:29:05 netifd: update to the latest version, fixes issues in handling device config from interfaces Apr 17 19:29:17 <_trine> KanjiMonster, I get it jogo Apr 17 19:29:41 <_trine> but there are others who are not included Apr 17 19:30:17 <_bazz> Thanks for the help all yall, especially KanjiMonster and plntyk who've been very helpful to me.. I'm just going to tidy up the patch ordering and I may call upon you once more if I may, to take a look at the final final Apr 17 19:32:18 "others"? Apr 17 19:32:32 <_trine> KanjiMonster, is there a list of #openwrt devs firstclass and a list of #openwrt second class? Apr 17 19:32:54 I'm not quite sure what you mean by that Apr 17 19:33:00 <_trine> well Apr 17 19:33:27 <_trine> some devs don't seem to appear in that published list Apr 17 19:34:01 <_trine> or is there a distinction between maintainers and devs Apr 17 19:34:58 the list at dev.openwrt.org is only people with commit access Apr 17 19:35:37 <_trine> then it is definitely not a complete list Apr 17 19:36:59 AFAICT only zajec is missing Apr 17 19:37:43 <_bazz> KanjiMonster: Um, I'm a bit bothered, after I reordered the commits and resolved the conflict.. I got this nasty message on several of my commits that show up in the patches too :( ... Apr 17 19:37:46 <_trine> well I do not wish to name names but there are more Apr 17 19:37:48 <_bazz> Signed-off-by: Michael J. Bazzinotti Apr 17 19:37:50 <_bazz> Conflicts: Apr 17 19:37:52 <_bazz> target/linux/ar71xx/files/arch/mips/ath79/mach-wnr2000-v4.c Apr 17 19:37:54 <_bazz> --- Apr 17 19:38:16 huh Apr 17 19:38:20 well, edit it out Apr 17 19:38:26 <_bazz> LOL Apr 17 19:38:32 <_bazz> lolol Apr 17 19:38:38 that's my usual tactic with git ;P Apr 17 19:38:41 <_bazz> I guess a part of me wished it was handled auto Apr 17 19:38:58 never had it, but maybe they added a new default for that Apr 17 19:39:01 <_bazz> because then if changes need to be, the process must be recycled.. *scripts* Apr 17 20:37:41 build #120 of ramips.mt7628 is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/ramips.mt7628/builds/120 Apr 17 20:59:36 build #947 of brcm47xx is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/brcm47xx/builds/947 Apr 17 21:17:48 patch submission process looks rather over elaborate for a first timer given all the scrollback Apr 17 22:18:48 build #494 of imx6 is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/imx6/builds/494 Apr 17 22:38:38 build #271 of brcm47xx.mips74k is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/brcm47xx.mips74k/builds/271 Apr 17 22:41:23 build #385 of realview is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/realview/builds/385 Apr 17 22:46:05 russell--: fyi for iproute-4.0.0 910-sanitize_headers_for_musl.patch needs updated for https://git.kernel.org/cgit/linux/kernel/git/shemminger/iproute2.git/commit/include/linux/xfrm.h?id=e46efaed0ff49b475d50ee8ffba2984e65076ba7 :/ Apr 17 23:15:17 build #280 of ar71xx.nand is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx.nand/builds/280 Apr 17 23:21:30 build #930 of lantiq is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/lantiq/builds/930 Apr 17 23:23:38 nico r45484 packages/utils/xz * xz: moved to github Apr 17 23:23:40 nico r45485 packages/libs/ldns * ldns: moved to github Apr 17 23:34:21 {Nico}_: can you kill github's libnetfilter-cttimeout libnetfilter-queue libnetfilter-cthelper? they're in trunk as of r45077 Apr 18 00:10:38 build #971 of brcm63xx is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/brcm63xx/builds/971 Apr 18 00:13:34 build #922 of ppc40x is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/ppc40x/builds/922 Apr 18 00:44:42 build #118 of oxnas is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/oxnas/builds/118 Apr 18 01:37:06 build #300 of brcm47xx.legacy is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/brcm47xx.legacy/builds/300 Apr 18 01:58:01 build #118 of ipq806x is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/ipq806x/builds/118 Apr 18 02:05:55 build #917 of ramips is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/ramips/builds/917 Apr 18 02:27:09 build #271 of ramips.rt3883 is complete: Exception [exception MasterShellCommand] Build details are at http://buildbot.openwrt.org:8010/builders/ramips.rt3883/builds/271 Apr 18 02:30:43 build #743 of mpc52xx is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/743 Apr 18 02:48:12 swalker: thanks, i'll work on it this weekend **** ENDING LOGGING AT Sat Apr 18 02:59:59 2015