**** BEGIN LOGGING AT Fri Mar 22 02:59:58 2013 Mar 22 04:26:19 build #216 of ramips is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/ramips/builds/216 Mar 22 07:38:18 builder_: wondering why a full/fresh build needs '10 hrs, 57 mins, 48 secs'? here: 800 seconds (gnu compiler farm) 8-) Mar 22 09:16:38 hi Mar 22 09:17:07 anyone tried owrt on a wag54gp2? Mar 22 09:23:42 dlooh: yes, how much ram yours have? 8mb? Mar 22 09:24:10 dlooh: oops, misread: i read wap, so: no Mar 22 10:04:46 florian r36104 packages/libs/ uclibc++/patches uclibc++ uclibc++/patches/001-do-not-set-make.patch * uclibc++: fix build on FreeBSD Mar 22 10:04:48 florian r36105 packages/ ipv6/send/Makefile send send/patches send/patches/007-use-flex.patch * send: use flex instead of lex Mar 22 10:04:49 florian r36106 packages/utils/zsh/Makefile * zsh: disable gdbm Mar 22 10:06:39 florian r36107 trunk/package/base-files/ files/lib/upgrade/common.sh Makefile * base-files: sysupgrade fail with eglibc Mar 22 10:06:43 florian r36108 trunk/toolchain/gdb/Makefile * gdb: update to 7.5-2012.12-1 Mar 22 11:27:52 nbd or anyone, I'm having an issue with this patch: https://dev.openwrt.org/browser/trunk/target/linux/atheros/patches-3.8/210-reset_button.patch Mar 22 11:28:02 on backfire, the reset button works and I get hotplug events Mar 22 11:28:09 but in AA I don't get any event at all. Mar 22 11:28:27 I suspect this is related to the gpiolib changes in AA for atheros, but I'm not sure what to look at Mar 22 11:28:39 any ideas? Mar 22 11:35:50 "gpioctl get 6" still shows the reset button state changing when I press the button, Mar 22 11:35:59 so it's getting _somewhere_ Mar 22 11:47:31 <[florian]> anarsoul: does your same build work fine on real HW? Mar 22 11:48:33 [florian]: AFAIR old build worked properly on real hw Mar 22 11:48:56 I don't have z2 with me, will check tonight Mar 22 11:56:51 [florian]: can you backport https://dev.openwrt.org/changeset/36107/ Mar 22 11:57:04 it's also broken in AA Mar 22 11:57:10 thanks in advance Mar 22 11:57:12 <[florian]> champtar: sure, thanks for the reminder Mar 22 11:59:37 [florian], on brcm63 wifi, does you have tried b43 and wl, what are the main difference between these? Mar 22 12:38:34 http://pastebin.com/NUrkjLub Mar 22 12:41:46 last effort Mar 22 12:52:51 nbd r36109 trunk/rules.mk * build: add a HOSTCXX variable pointing at the c++ compiler Mar 22 12:52:54 nbd r36110 trunk/tools/upx/Makefile * tools/upx: use HOSTCXX to fix build errors on mac os x Mar 22 13:10:50 are there any guides or tips on debugging the kernel with gdb for me to try and find what's missing for hotplug? Mar 22 13:11:00 I'm happy with gdbserver for userspace, but never tried it on the kernel before Mar 22 13:13:33 cat /sys/kernel/debug/gpio shows the reset button going hi/low when I prss it too. Mar 22 13:14:24 karlp: whats your actual problem atm? that you receive no hotplug events? Mar 22 13:14:58 correct Mar 22 13:15:06 pushing the button on backfire gives a hotplug button event Mar 22 13:15:08 but not on AA Mar 22 13:15:41 is kmod-gpio-button-hotplug installed? Mar 22 13:16:17 no, but it'ðs not on backfire either Mar 22 13:16:29 backfire has a different implementation Mar 22 13:16:54 ok,I'll rebuild and check that, Mar 22 13:17:06 that then needs to get added into the Kconfig for this platform presumably? Mar 22 13:17:22 maybe, I lost track of what state the legacy atheros target is in Mar 22 13:17:42 yeah, I know it's old and uncool :) Mar 22 13:17:55 thats not what I meant Mar 22 13:18:16 but I do not know whether it still uses its own gpio stuff implementation or the generic one all other targets use Mar 22 13:18:18 atheros got moved to gpiolib, Mar 22 13:18:23 between backfire and AA Mar 22 13:18:31 I did that, it gave me the nice sysfs/gpio stuff Mar 22 13:18:32 then you probably need kmod-gpio-button-hotplug Mar 22 13:18:38 but I barely know what I'm doing. Mar 22 13:18:41 thats what e.g. ar71xx uses to get hotplug for gpio button presses Mar 22 13:19:09 ok, so that's something I missed when I moved it from it's own gpio to the genericgpiolib Mar 22 13:19:53 can I just add it to target/linux/atheros/Makefile:DEFAULT_PACKAGES if this fixes it? is that the right place for it? Mar 22 13:19:53 also note that the /etc/config/system button stuff is atheros specifc Mar 22 13:20:01 as far as I can see no other target implements this Mar 22 13:20:32 oh, this is with just a plain 02-test file in hotplug.d/button that just does logger $BUTTON $ACTION Mar 22 13:20:39 ok Mar 22 13:21:07 do other platforms have anything for reboot/factory default on button press at all? Mar 22 13:21:31 no Mar 22 13:22:02 there never was any button-triggered reset facility in openwrt by default Mar 22 13:22:33 that makes sense. the atheros default config has logger reboot and logger factory defauult, but it just logs that, it doesn't actually do anything anyway Mar 22 13:22:57 hotplug doesn't need to be restarted or anything does it? just editing the files takes effect immediately? Mar 22 13:23:04 what files? Mar 22 13:23:16 hotplug2.rules and hotplug2-common.rules Mar 22 13:23:31 for these you need to restart hotplugd Mar 22 13:23:33 like removing the ^ from button as noted here: http://wiki.openwrt.org/doc/howto/hardware.button Mar 22 13:23:44 there's no init.d script for hotplug* Mar 22 13:23:49 nope Mar 22 13:23:55 grep hotplug /etc/init.d/* Mar 22 13:24:01 should give you the cmdline Mar 22 13:24:05 ok, thanks. Mar 22 13:24:24 ah, it's in boot, excellent Mar 22 13:24:48 that will be faster than rebooting :) Mar 22 13:25:05 is target/linux/atheros/Makefile:DEFAULT_PACKAGES the right place to add kmod's like this? Mar 22 13:26:05 yes Mar 22 14:11:04 jow r36111 trunk/package/network/config/firewall3/Makefile * firewall3: update to git head Mar 22 14:21:08 jow_laptop: still no hotplug events with that module. Mar 22 14:21:48 I've uncommented the "uncomment me to log hotplug events" in hotplug2-common.rules and still nothing Mar 22 14:53:19 karlp: you have to restart hotplug2 (or reboot) for it to have any effect (or somehow force hotplug to re-read it) Mar 22 14:59:43 rebooting hasn't helped. Mar 22 14:59:56 I've tried the section from init.d/boot to restart and also rebooting Mar 22 15:03:46 http://pastebay.net/1191960 has parts of the hotplug snippet Mar 22 15:04:02 lsmod on AA shows gpio_button_hotplug is loaded Mar 22 15:06:09 <__trine> trunk isn't compiling http://dpaste.com/1031367/ Mar 22 15:06:32 <__trine> and I have done make oldconfig Mar 22 15:06:59 <[florian]> __trine: that package is not technically in trunk Mar 22 15:07:17 <__trine> yes I understand that Mar 22 15:08:37 <__trine> but it's still stopping my trunk compiling though Mar 22 15:09:43 <[florian]> well complain to the freeswitch package maintainer Mar 22 16:18:34 <__trine> [florian], I dont wish to complain to anyone I am simply pointing out trunk is not compiling, in this way I believe I help if that's not your view then I am sorry Mar 22 16:20:33 <[florian]> __trine: by complaining I meant talking to mazilo and telling him that freeswitch does not build for you Mar 22 16:22:19 what does this mean: Mar 22 16:22:28 Package bcmkernel is missing dependencies for the following libraries: libnvram.so Mar 22 16:22:29 <__trine> [florian], I directed the information to the dev channel so anyone who was interested might be made aware Mar 22 16:22:52 how does a package "invent" a dependency ? Mar 22 16:23:19 <[florian]> it does not, openwrt analyses the "NEEDED" section of the ELF binary and detects that this library is not in your package's DEPENDS Mar 22 16:23:24 <[florian]> while it will actually use it at runtime Mar 22 16:23:55 ic Mar 22 16:25:13 how do I turn it off :) Mar 22 16:25:48 just add whatever provides nvram.so as a dependency to your package ;) Mar 22 16:26:03 <[florian]> you can't turn it off without patching include/package-ipkg.mk Mar 22 16:26:51 that's what I get for creating Frankenstein Mar 22 16:27:24 ok, tnx Mar 22 16:29:08 <[florian]> bcmkernel you mean right :p Mar 22 16:34:22 [florian]: yeah :) Mar 22 16:49:08 how do I stop a package from getting wiped and getting a new copy being unpacked from the dl directory ? Mar 22 16:49:25 by overriding Build/Preprare Mar 22 16:55:53 nbd ping Mar 22 17:29:34 build #217 of ramips is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/ramips/builds/217 Mar 22 17:29:35 build #202 of rb532 is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/202 Mar 22 17:30:38 build #202 of ppc44x is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/202 Mar 22 17:30:39 build #212 of uml is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/212 Mar 22 17:31:43 build #205 of x86 is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/x86/builds/205 Mar 22 17:31:43 build #201 of ar7 is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/ar7/builds/201 Mar 22 17:32:47 build #192 of sibyte is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/192 Mar 22 17:32:47 build #199 of avr32 is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/199 Mar 22 17:33:51 build #197 of xburst is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/xburst/builds/197 Mar 22 17:33:52 build #193 of ixp4xx is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/ixp4xx/builds/193 Mar 22 17:34:56 build #193 of kirkwood is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/kirkwood/builds/193 Mar 22 17:34:56 build #187 of ar71xx is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx/builds/187 Mar 22 17:36:00 build #190 of au1000 is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/au1000/builds/190 Mar 22 17:36:00 build #172 of iop32x is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/172 Mar 22 17:37:04 build #175 of ep93xx is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/ep93xx/builds/175 Mar 22 17:37:04 build #168 of gemini is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/gemini/builds/168 Mar 22 17:38:08 build #174 of mcs814x is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/mcs814x/builds/174 Mar 22 17:38:08 build #179 of adm5120 is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/adm5120/builds/179 Mar 22 17:39:12 build #162 of mpc52xx is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/162 Mar 22 17:39:13 build #155 of octeon is complete: Failure [failed svn compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/octeon/builds/155 Mar 22 18:10:11 build #217 of lantiq is complete: Failure [failed compile_3] Build details are at http://buildbot.openwrt.org:8010/builders/lantiq/builds/217 Mar 22 20:42:26 [florian]: hi, tested on real hw, and there's no opkg crash Mar 22 20:42:30 probably some qemu issue Mar 22 20:42:46 <[florian]> anarsoul: looks like, but still it's pretty strange Mar 22 20:43:12 would be really nice to fix it with qemu :) Mar 22 20:44:55 it's not convenient to test build on real hw, NFS over usb networking is slow (pxa27x udc gives only ~400kb/s), and changing SD cards is even worse Mar 22 20:45:44 <[florian]> yeah I guess so Mar 22 21:22:14 [florian]: it works on my laptop, but not on PC Mar 22 21:22:28 I wonder what's difference between laptop and PC, hardware is pretty close Mar 22 21:22:33 and OS is same Mar 22 21:22:59 * anarsoul 's completely confused Mar 23 02:15:51 build #226 of orion is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/226 **** ENDING LOGGING AT Sat Mar 23 02:59:58 2013