**** BEGIN LOGGING AT Thu Feb 14 03:00:00 2013 Feb 14 04:46:53 build #170 of rb532 is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/170 Feb 14 04:49:58 build #170 of ppc44x is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/170 Feb 14 05:39:10 build #175 of uml is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/175 Feb 14 07:25:18 build #160 of sibyte is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/160 Feb 14 08:27:00 has anybody else seen weirdness with make defconfig? Feb 14 08:50:27 lol, now i can't get it to fail ... /me checking to rehab clinic Feb 14 09:26:36 hi, I've been having trouble compiling packages which are /not/ prepared for cross-compilation (e.g. gutenprint, ghostscript) Feb 14 09:26:52 shocking ;) Feb 14 09:27:18 I was wondering if there are rules in "include" already set up to help with that Feb 14 09:27:25 yes Feb 14 09:27:33 jow_laptop, now we are talking Feb 14 09:28:04 jow_laptop, those would be? Or if you can point me to a package that uses them Feb 14 09:28:20 build #184 of orion is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/184 Feb 14 09:28:46 fborges: too broad question Feb 14 09:28:53 jow_laptop, ok Feb 14 09:29:00 jow_laptop, I already compiled gutenprint Feb 14 09:29:00 there are 1000 different ways cross compilation could be fucked up Feb 14 09:29:07 and thew fuckups differ with each package Feb 14 09:29:12 jow_laptop, fair point Feb 14 09:29:33 jow_laptop, the problem I had with gutenprint and encountered so far with ghostscript is Feb 14 09:29:53 jow_laptop, they will compile some binary which they expect to run during their build process Feb 14 09:30:09 ah yes Feb 14 09:30:12 two approaches Feb 14 09:30:16 jow_laptop, my solution has been to compile locally and place the binaries somewhere Feb 14 09:30:52 1) sometimes that is just due to a crappy autoconfig macro and stupid developers that use AC_TRY_RUN() instead of AC_TRY_LINK() or similar Feb 14 09:31:22 2) if host-built binaries are indeed needed you can define a host build phase in your openwrt makefile, see e.g. glib2 Feb 14 09:31:43 jow_laptop, what I encountered so far seems more like .2. Feb 14 09:31:52 that has a host-build phase to compile the glib-genmarshal utility Feb 14 09:33:17 jow_laptop, thanks, I'll check it out Feb 14 09:33:40 jow_laptop, probably an improved version of the hack I did with gutenprint Feb 14 09:54:57 after some detox therapy, it looks like just a problem of feeds install -a not being run first, or something vaguely related to that, and thus the build system is ignorant of my packages and so defconfig naturally ignores them under those conditions. Feb 14 10:04:45 build #157 of ixp4xx is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/ixp4xx/builds/157 Feb 14 10:28:42 hey guys… I'm currently working on a project regarding 802.11p and I've received a pair of devices with .11p modules and OpenWRT, but regulatory domain does not allow me to set frequencies high enough (5.85 GHz)…what is the easiest way to edit regulatory domain on openwrt? I was unable to find the public key, the drivers have some patches for 802.11p support, overall the software inside the device seems to be custom buil Feb 14 10:31:54 ath9k driver? Feb 14 10:32:00 ath5k Feb 14 10:32:12 you need to patch mac80211 then Feb 14 10:32:22 regdb is compiled into cfg80211.ko Feb 14 10:34:29 ok, thanks Feb 14 11:27:56 build #144 of iop32x is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/144 Feb 14 11:34:39 build #141 of gemini is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/gemini/builds/141 Feb 14 11:58:53 nbd: seems uci inspired some other projects https://fedoraproject.org/wiki/Features/SystemConfigurationShell Feb 14 12:10:22 someone should inspire them to kill systemd Feb 14 12:36:27 hauke r35595 trunk/package/switch/src/switch-core.c * switch: export cpuport, ports and vlans via /proc Feb 14 12:37:03 hauke r35596 trunk/package/switch/src/ switch-core.h switch-core.c switch-robo.c switch-adm.c * switch: export name of device found via /proc Feb 14 12:38:26 hauke r35597 trunk/target/linux/brcm47xx/base-files/etc/init.d/netconfig Feb 14 12:38:26 brcm47xx: if the cpuport of the switch is port 8 assume vlan0ports: "1 2 3 4 8" and vlan0ports: "0 8" Feb 14 12:43:48 luka r35598 packages/net/mosquitto/Makefile * [packages] mosquitto: upgrade to version 1.1.2 Feb 14 12:50:54 thanks luka :) Feb 14 12:59:22 florian r35599 packages/net/openconnect/Makefile * openconnect: update openconnect to 4.08 (CVE-2012-6128) Feb 14 13:00:14 florian r35600 trunk/package/libs/ openssl/Makefile openssl/patches/120-cisco-dtls-fix.patch * openssl: update OpenSSL to 1.0.1e, fix Cisco DTLS. Feb 14 13:00:17 florian r35601 trunk/target/linux/x86/config-3.7 * x86: update config-3.7 for x86 (turn off CONFIG_GPIO_ICH) Feb 14 14:01:35 juhosg r35602 trunk/target/linux/generic/files/drivers/net/phy/ar8216.c * generic: ar8216: register the switch from the probe routine Feb 14 14:55:51 juhosg r35603 trunk/target/linux/generic/files/drivers/net/phy/ar8216.c * generic: ar8216: move PHY4 RGMII workaround code out of the loop Feb 14 16:02:50 karlp: np Feb 14 16:04:29 build #131 of octeon is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/octeon/builds/131 Feb 14 18:39:36 build #153 of ar71xx is complete: Failure [failed shell_14] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx/builds/153 Feb 14 19:00:53 build #146 of ep93xx is complete: Failure [failed shell_14] Build details are at http://buildbot.openwrt.org:8010/builders/ep93xx/builds/146 Feb 14 19:33:20 hi, I have a silly question about using "host-build.mk", is there a straightforward way to only apply patches to the cross-compiled target? Feb 14 19:34:05 by that I mean, the correct way to make patches skip the "host" build? Feb 14 19:39:58 nevermind... figured it out Feb 15 02:06:28 build #166 of avr32 is complete: Failure [failed compile_9] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/166 Feb 15 02:50:17 build #171 of rb532 is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/171 Feb 15 02:53:59 build #171 of ppc44x is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/171 **** ENDING LOGGING AT Fri Feb 15 02:59:58 2013