**** BEGIN LOGGING AT Sun Nov 10 02:59:58 2013 Nov 10 03:55:33 build #384 of avr32 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/384 Nov 10 03:58:27 build #323 of iop32x is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/323 Nov 10 09:57:04 kaloz r38702 trunk/target/linux/cns3xxx/files/arch/arm/mach-cns3xxx/laguna.c * [cns3xxx]: add pushbutton GPIO for GW2380/82/83 Nov 10 09:58:50 kaloz r38703 trunk/target/linux/cns3xxx/image/Makefile * [cns3xxx]: add uImage image build Nov 10 10:03:17 nbd r38704 trunk/package/network/config/netifd/files/etc/hotplug.d/iface/00-netstate * netifd: remove connect_time from /var/state, it is unused Nov 10 10:55:37 build #362 of kirkwood is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/kirkwood/builds/362 Nov 10 14:23:33 nbd: ping Nov 10 14:25:23 DonkeyHotei: pong Nov 10 14:25:59 nbd: i notice the atheros target does not use devicetree. is that planned? Nov 10 14:28:58 yes, just nobody's actively working on it at the moment Nov 10 14:30:47 nbd: i also notice /etc/diag.sh is empty. i can make one for fon2201 but idk how to tell it apart programmatically from devices with different led's Nov 10 14:31:32 as is, the power led is unused on the fon2201 Nov 10 14:31:51 in backfire, it was used Nov 10 14:36:20 nbd: how different are the led gpio's among atheros devices? Nov 10 14:59:22 updated openwrt/upstream, https://home.comcast.net/~sdwalker/uscan/index.html Nov 10 15:08:12 nbd r38705 trunk/package/system/procd/files/hotplug.json Nov 10 15:08:12 procd: reduce the number of unnecessary calls to /sbin/hotplug-call for tty devices Nov 10 15:19:34 what is the new "make kernel_menuconfig" for current trunk? Nov 10 15:19:52 it's still 'make kernel_menuconfig' Nov 10 15:20:13 nbd: it opens the regular menuconfig... Nov 10 15:22:32 huh? Nov 10 15:22:34 it doesn't do that here Nov 10 15:26:31 probably I messed with the .config file. I am trying distclean and if not just checkout again. Sorry Nov 10 15:28:18 hmm i can see the wifi toggle switch in /sys/kernel/debug/gpio but there is no hotplug.d action when i toggle it, yet there is action for all the other buttons. i wonder what that is all about Nov 10 15:34:39 JyZyXEL: most likely becuase the hotplug handler does not know the button Nov 10 15:34:52 as in on the kernel side Nov 10 15:35:38 hi Nov 10 15:35:58 JyZyXEL: did you ask in #mipslinux about insmod? Nov 10 15:36:05 could someone please apply my openssl patch with enabled ripemd/160 in openssl - i allready sent a mail to the ml Nov 10 15:36:41 tripolar just forward? Nov 10 15:37:27 DonkeyHotei: nope, i haven't asked yet or even made a bug report Nov 10 15:38:09 did someone notice OpenWrt-Commit bot is as Guest? Nov 10 15:38:27 Devastator: ? Nov 10 15:38:52 tripolar sorry, I've read send instead of apply Nov 10 15:39:49 nbd: how different are the led gpio's among atheros devices? Nov 10 15:55:13 hmm for example my wps buttons in debug/gpio shows up as gpio-239 (wps ) in hi Nov 10 15:55:27 and hotplug.d reports it as BTN_2 when using gpio_button_hotplug Nov 10 15:55:56 JyZyXEL: that would be a dts file issue Nov 10 15:56:45 i was looking at gpio-button-hotplug.c where linux/input.h is included that defines: #define BTN_2 0x102 Nov 10 15:57:08 look at the dts file Nov 10 15:57:16 DonkeyHotei: ? Nov 10 15:57:28 JyZyXEL: what code is givvn to the toggle switch ? Nov 10 15:58:00 looks like in the dts file the wps button is mapped to the: linux,code = <0x102>; Nov 10 15:58:17 it shouldn't be Nov 10 15:58:40 im not sure why its mapped to that, when there is an actual code for it in input.h Nov 10 15:58:48 #define KEY_WPS_BUTTON 0x211 Nov 10 16:00:06 the wifi toggle switch has in the dts file: linux,input-type = <5>; /* EV_SW */ Nov 10 16:08:09 toggle switches are not supported yet Nov 10 16:08:18 seems so Nov 10 16:09:14 though it seems gpio-button-hotplug.c has a case for EV_SW Nov 10 16:09:42 it is not passed to that Nov 10 16:14:55 nbd: this is what i came up with for /etc/diag.sh for the fon2201, but it needs to differentiate from other devices somehow: http://pastebin.com/7HGmdUcT Nov 10 16:54:28 Hi everyone, Nov 10 16:54:55 I usually spend my time in the non dev openwrt irc channel. However I'd like to write my own kernel patch so I thought I'd ask here. Nov 10 16:55:22 I'm a bit confused about quilt and when and how to use it. Nov 10 16:56:14 I've already applied patches from other people using the patch program something similar to "patch -p0 < patchfile" Nov 10 16:56:29 but the wiki says I need to use quilt for patching Nov 10 16:56:59 You do not need to use quilt imho. Nov 10 16:56:59 I've briefly read what quilt is, and I'm confused why I can't just use svn (or git, but my repository is in svn ) for patching. Why do I need to use this quilt? Nov 10 16:57:01 its confusing me Nov 10 16:57:55 slachta: what do I use then? Nov 10 16:58:04 or what do you use? Nov 10 16:58:33 I have written several kmods, several packages and all I needed is to know about diff, patch, and the same from git view. Nov 10 16:59:11 Sorry for slower responses, /me typing on tablet Nov 10 17:00:22 hi slachta no probelm , sorry I've gotta to go for 5 mins .. can I ask you some questions when I get back on how to patch please Nov 10 17:00:30 but its not kernel modules, but the actual kernel itself Nov 10 17:00:38 I need to make small change to netfilter Nov 10 17:00:39 brbv Nov 10 17:04:43 b Nov 10 17:05:32 so slachta how do you actual change code in the kernel, compiler it and then commit it Nov 10 17:05:40 using svn or git Nov 10 17:06:31 DomeMaster: the quilt way is a bit easier and makes it all easier. My way of patching any sources was a bit arduous. I just duplicate the source, make change and call diff -ruN on the changed files. Nov 10 17:07:29 ahh ok, could you explain to me what is the purpose of quilt please Nov 10 17:07:34 in the context of openwrt Nov 10 17:09:21 DomeMaster: there is no difference between quilt outside of openwrt buildroot and the one inside. It is just integrated into build system. Nov 10 17:09:25 http://wiki.openwrt.org/doc/devel/patches Nov 10 17:09:37 This link describes it nicely. Nov 10 17:10:08 I've read that slachta I just never understood why they used quilt over svn or git which they are already using Nov 10 17:10:13 thanks for all the info btw Nov 10 17:10:53 Because kernel sources are not versioned in git openwrt repositories Nov 10 17:11:38 It is just build systen with a bunch of macros, patches and everything else, but not linux kernel sources. Nov 10 17:12:27 ok I guess that makes sense, I have two follow up questions then :) Nov 10 17:12:39 So yes, you can do that with git/svn, but outside of openwrt buildroot if the sources are fetched from tarball Nov 10 17:12:42 1) I thought git was built by linus to be used specfically in the kernel, so why is quilt still being used Nov 10 17:13:28 2) So is my thinking correct. Quilt basically has the base version of the kernel for the version you have. Then just applies the patches in order according to the quilt stack right? Nov 10 17:14:09 is the answer to 1, that buildroot uses quilt not the linux kernel? Nov 10 17:17:33 Ad 1) I think you are mixing thins together. Even if Linus participated on git, it is not intended to be only scm for linux kernel sources Nov 10 17:18:22 Imho quilt is just a tool which tracks modifications to the upstream Nov 10 17:19:10 Ad 2) yes Nov 10 17:19:46 It is annoying to type on tablet. I hate this keyboard. Nov 10 17:21:04 ok thanks so much man.. Especially considering the typing effort Nov 10 17:21:18 so my final question I guess, everything else seems pretty clear Nov 10 17:21:32 is what on earth was the "patch -p0 < patchfile" command doing in the past Nov 10 17:21:39 I've done that to apply other peoples patches Nov 10 17:21:47 why didn't I need to use quilt for that? Nov 10 17:24:10 The thing is that quilt is for tracking changes and creating patches, the same as diff do. Nov 10 17:24:35 Patch utility is for applying patches against sources Nov 10 17:25:10 ahhhhh Nov 10 17:25:17 got it, so quilt creates the patch itself Nov 10 17:25:22 it makes sense now man Nov 10 17:25:23 thanks Nov 10 17:25:34 so it sort of does the annoying work of tracking when to run diff for you Nov 10 17:26:31 The p option for patch is related to pathnames, the p0 is full path, p1 removes the leading slashes Nov 10 17:26:41 It is described in man pages Nov 10 17:27:14 Yes, the quilt does the hard work I usually do because I am lazy to use it :-) Nov 10 17:27:21 yeah sure slachta I understood how patch worked Nov 10 17:27:44 thanks, I just didn't get what it was patching. But my main confusion was what quilt did. Now I get it generates the patch thanks so much man Nov 10 17:27:45 :) Nov 10 17:28:08 You are welcome. Nov 10 18:06:38 tripolar r38706 packages/libs/gdbm/Makefile * [packages] libgdbm: update to 1.10 Nov 10 18:09:00 tripolar r38707 packages/libs/ libnet-1.1.x/Makefile libnet-1.1.x/patches/150-portability.patch libnet-1.1.x/patches/100-debian-subset.patch libnet-1.1.x/patches/160-endian_cache.patch * [packages] libnet1: update to 1.1.6 Nov 10 18:10:04 nbd: ping Nov 10 18:11:06 tripolar r38708 packages/libs/ libnids/patches/001-no_asm_for_i386.patch libnids/Makefile libnids/patches/002-configure.patch * [packages] libnids: update to 1.24 Nov 10 18:14:59 tripolar r38709 packages/libs/libnids/patches/003-missing_else.patch * [packages] libnids: also remove not needed patch Nov 10 18:19:58 tripolar r38710 packages/net/dsniff/ Makefile patches/011-noint32.patch * [packages] dsniff: fix build Nov 10 18:26:07 build #403 of rb532 is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/403 Nov 10 18:31:04 build #403 of ppc44x is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/403 Nov 10 18:43:44 DonkeyHotei: pong Nov 10 18:43:59 nbd: this is what i came up with for /etc/diag.sh for the fon2201, but it needs to differentiate from other devices somehow: http://pastebin.com/7HGmdUcT Nov 10 18:44:35 [Sun 2013-11-10 06:30:47 AM PST] nbd: i also notice /etc/diag.sh is empty. i can make one for fon2201 but idk how to tell it apart programmatically from devices with different led's Nov 10 18:44:35 [Sun 2013-11-10 06:31:31 AM PST] as is, the power led is unused on the fon2201 Nov 10 18:44:36 [Sun 2013-11-10 06:31:51 AM PST] in backfire, it was used Nov 10 18:44:37 [Sun 2013-11-10 06:36:20 AM PST] nbd: how different are the led gpio's among atheros devices? Nov 10 18:44:50 i have no idea about differences between different devices using that target Nov 10 18:44:56 i haven't done anything with that target in years Nov 10 18:45:29 were there non-fon devices in that target? Nov 10 18:47:33 yes Nov 10 18:47:48 many? Nov 10 18:48:20 no idea how many Nov 10 18:48:30 but quite a few at least Nov 10 18:48:47 you think that /etc/diag.sh above is safe for the target? Nov 10 18:48:53 no idea Nov 10 18:50:14 how is it that the led was used in backfire but not anymore? Nov 10 18:52:52 nbd: ^ Nov 10 18:53:11 what led? Nov 10 18:53:15 power Nov 10 18:54:20 no idea Nov 10 18:54:56 maybe ask your questions on the list or something Nov 10 18:55:07 i don't plan on starting to work on the atheros target any time soon Nov 10 18:56:27 nbd: would you be willing to commit the above /etc/diag.sh in the interim? Nov 10 19:04:26 send stuff to the list Nov 10 19:04:33 instead of bugging me directly Nov 10 19:04:48 i'm working on completely different stuff at the moment Nov 10 19:08:39 tripolar r38711 packages/net/dsniff/Makefile Nov 10 19:08:39 [packages] dsniff: depends on libnet0, error didn't showup because both libs where build Nov 10 19:10:49 JyZyXEL: https://dev.openwrt.org/ticket/14455 Nov 10 19:10:54 do you add all this tickets ? Nov 10 19:11:04 *these Nov 10 19:12:58 yeah i thought it would be good to have 1 thing per ticket instead of a list of things Nov 10 19:13:31 can you not add 1 ticket / hour Nov 10 19:13:42 and then 1 bug with 3 version of a patch in 2 tickets Nov 10 19:13:51 and generally try to learn to make full sentences please Nov 10 19:14:07 https://dev.openwrt.org/ticket/14455 Nov 10 19:14:10 nbd if you don't mind me asking, what are you working for atm? Nov 10 19:14:29 Devastator: stuff that you are distracting him from right now Nov 10 19:14:37 heh Nov 10 19:14:39 sorry Nov 10 19:14:42 i am serious Nov 10 19:15:15 JyZyXEL: https://dev.openwrt.org/ticket/14455 stuff like that .. the led is not even wired how should it work ? Nov 10 19:15:34 and it would be nice if in the bug report, the actual report was written in a coherent sentence Nov 10 19:15:43 rather than leaving it up to me to figure out what you mean Nov 10 19:16:19 you expect me to invest a pile of time o fix the problem so at least be so decent to take the time to make a proper report Nov 10 19:16:35 those reports look like oyu were in a hurry or so Nov 10 19:17:13 thxx ! Nov 10 19:24:41 actually my motivation was to just get visibility for those small things, hoping that someone other would notice it and possible send a patch in the comments Nov 10 20:30:09 build #438 of orion is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/438 Nov 10 23:31:44 hauke r38712 trunk/target/linux/ generic/files/drivers/net/phy/b53/b53_priv.h generic/files/drivers/net/phy/b53/b53_common.c * kernel: b53: add initial support for switches in BCM5301X Nov 11 00:09:53 hauke r38713 trunk/target/linux/ generic/patches-3.10/025-bcma_backport.patch generic/patches-3.12/025-bcma_backport.patch * kernel: bcma: detect pci ID 0x4313 Nov 11 00:15:15 hauke r38714 trunk/target/linux/ (8 files in 2 dirs) * kernel: bgmac: make it send and receive some packages on BCM4708 Nov 11 00:16:44 hauke r38715 trunk/target/linux/ bcm53xx/patches-3.10/122-bcma-add-arm-support.patch bcm53xx/config-3.10 bcm53xx/patches-3.10/121-bcma-fix-dma-mask.patch bcm53xx/patches-3.10/202-bgmac-make-bgmac-work-on-systems-without-nvram.patch * bcm53xx: make bgmac work Nov 11 02:37:09 build #416 of uml is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/416 **** ENDING LOGGING AT Mon Nov 11 02:59:58 2013