**** BEGIN LOGGING AT Fri Feb 25 02:59:57 2011 Feb 25 03:22:50 build #74 of avr32 is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/avr32/builds/74 Feb 25 03:36:57 HOLY SHIT Feb 25 03:37:18 build #68 of etrax is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/etrax/builds/68 Feb 25 03:37:19 I use teh INTERWEBS Feb 25 03:39:05 i suppose neither [florian] nor whip are around :\ Feb 25 03:51:50 build #78 of gemini is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/gemini/builds/78 Feb 25 04:12:53 current missing symbol(s) bustage: ppc44x/2.6.32.29 (RTC_DRV_GENERIC), ep93xx/2.6.36.4 (RTC_DRV_EP93XX), mpc52xx/2.6.30.10 (RTC_DRV_GENERIC), adm5120_router_le/2.6.32.29 (RTC_DRV_PL030), rdc/2.6.32.29 (IRQ_TIME_ACCOUNTING) Feb 25 04:23:42 build #67 of ep93xx is complete: Failure [failed compile_4] Build details are at http://tksite.gotdns.org:8010/builders/ep93xx/builds/67 Feb 25 04:42:07 build #66 of rdc is complete: Failure [failed compile_4] Build details are at http://tksite.gotdns.org:8010/builders/rdc/builds/66 Feb 25 05:09:46 build #67 of adm5120 is complete: Failure [failed compile_4] Build details are at http://tksite.gotdns.org:8010/builders/adm5120/builds/67 Feb 25 05:28:38 build #88 of ubicom32 is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/ubicom32/builds/88 Feb 25 06:39:58 :s Feb 25 07:59:54 hmm Feb 25 08:00:16 guess i'll have to check that out with 2.6.37 Feb 25 09:46:16 gmorning Feb 25 13:00:14 build #71 of ixp4xx is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ixp4xx/builds/71 Feb 25 13:04:24 build #70 of xburst is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/xburst/builds/70 Feb 25 13:10:11 build #86 of orion is complete: Failure [failed shell shell_4 compile_1] Build details are at http://tksite.gotdns.org:8010/builders/orion/builds/86 Feb 25 14:22:35 hey, I have to compile owrt with OABI, (not EABI). but every *libc Im trying its failing while compiling... (with different error messages). Feb 25 14:22:35 what last environment is known to work with OABI ? Feb 25 16:17:32 build #79 of ps3 is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/ps3/builds/79 Feb 25 16:52:39 kaloz * r25705 /trunk/ (rules.mk toolchain/Config.in toolchain/gcc/common.mk): Feb 25 16:52:39 We use different toolchain directories for different Feb 25 16:52:39 ARM archs, so we should set the default arch of gcc Feb 25 16:52:39 to reflect this. Feb 25 16:52:39 This enables EABI support for armv4 CPUs. Feb 25 16:52:40 Signed-off-by: Jochen Friedrich Feb 25 17:02:33 does anyone use OABI build? what is right combination of toolchains ? gcc+*libc+binutils ? I cannot make it work with OABI :( Feb 25 17:09:04 why do you need OABI? Feb 25 17:26:39 kaloz * r25706 /trunk/target/linux/gemini/ (8 files in 2 dirs): [target/gemini]: upgrade to 2.6.37.1 Feb 25 17:31:42 nbd: becasue I have some old modules which I cant for now use with EABI Feb 25 17:32:05 I HAVE to use oabi. :-/ Feb 25 17:33:15 kaloz * r25707 /trunk/target/linux/kirkwood/Makefile: [target/kirkwood]: switch back to xscale mtune, marvell_f is unlikely to get upstreamed Feb 25 17:43:30 kaloz * r25708 /trunk/toolchain/gcc/Config.version: [toolchain/gcc]: use the Linaro GCC for arm and mips targets Feb 25 17:44:31 kaloz * r25709 /trunk/toolchain/gcc/patches/ (4.5.2/995-fa526.patch linaro/995-fa526.patch): [toolchain/gcc]: backport fa526 optimization for gcc 4.5+ Feb 25 17:45:56 kaloz * r25710 /trunk/target/linux/gemini/Makefile: [gemini]: optimize for the fa526 core Feb 25 17:46:14 kaloz * r25711 /trunk/target/linux/cns21xx/Makefile: [cns21xx]: optimize for the fa526 core Feb 25 18:01:06 kaloz * r25712 /trunk/target/linux/gemini/patches/003-missing_from_upstream.patch: [target/gemini]: register gmac, too Feb 25 18:27:17 voyo: You may have to reconfigure and recompile the Linux kernel to include a support for "Allow old ABI binaries to run with this kernel (EXPERIMENTAL) " under the Kernel Features. Feb 25 18:32:59 Hi. Anyone noticed OABI builds are broken in backfile for some months now? Feb 25 18:33:04 Or is it just me? Feb 25 19:23:44 Wipster there? Feb 25 19:24:32 yeh Feb 25 19:24:45 i got the thing to work :D Feb 25 19:25:01 nice, how? Feb 25 19:25:26 deleted the last two patches Feb 25 19:25:50 ah yeh back to fixed phy Feb 25 19:25:52 and switched to 2.6.32 kernel Feb 25 19:26:30 will try with .37 Feb 25 19:26:59 since i can't seem able to use ap mode (acx) Feb 25 19:27:21 infrastructure works, client too... just not ap Feb 25 19:27:50 using acx mac from git Feb 25 19:28:45 maybe it's currently broken? :\ Feb 25 19:29:28 not idea about wireless, can you apply a patch when you go to kernel .37? think I found the issue Feb 25 19:30:37 anyway, i'll let you know if it was that 972 cpmac kernel patch after compilation is done Feb 25 19:30:58 (compiling for .37 atm) Feb 25 19:31:29 and sure, i'll test the patch out Feb 25 19:32:31 just upload it somewhere :) Feb 25 19:34:24 currently 972 will be issue for you, it moves away from fixed phy to properly detecting the switch in doing so needs to setup things properly. http://pastebin.com/sjiLshw6 Feb 25 19:34:52 hmm Feb 25 19:35:10 so the plan is to setup the thing properly huh Feb 25 19:36:28 thats the idea, so you can control the switches Feb 25 19:37:33 so the patch just skips some titan only bits if the h/w is different huh? Feb 25 19:38:02 mazilo: yeah I know that. I did this already. but seems that its owrt build what is broken, not the kernel. Feb 25 19:38:18 xl0: yeah, not only you, unfortunatelly... Feb 25 19:39:41 well I mistakenly thought that if an switch was detected I should MII remap, but that breaks sangams and should only be applied to ohio chips. Titan is still work in progress Feb 25 19:40:03 i see Feb 25 19:40:06 voyo: Sorry about that. Feb 25 19:40:14 wait Feb 25 19:40:34 you tested that on a dg834gv3? Feb 25 19:40:57 before the fixup i mean Feb 25 19:41:03 and it worked? Feb 25 19:41:12 yeh, which has an ohio chip the v2 has a sangam Feb 25 19:41:43 ahh, so there was more to the v3 syncing at better speedsz Feb 25 19:41:56 and v2 syncing easier on bad lines Feb 25 19:43:08 anyway with that modification to 972 you should be working fine and able to configure the 88E6060, let me know how it goes Feb 25 19:43:58 so you can commit to trunk Feb 25 19:43:59 heh Feb 25 19:45:15 damn Feb 25 19:45:31 doesn't v3 overheat or something? Feb 25 19:47:34 ah wait Feb 25 19:47:37 something more Feb 25 19:47:56 for the patches to be applied Feb 25 19:48:07 will i have to clean the whole distr? Feb 25 19:48:20 or a directory clean will do? Feb 25 19:51:49 if I put an init script in /trunk/package/base-files/files/etc/init.d, how do I make sure it gets enabled? (symlink in /etc/rc.d?) Feb 25 19:51:57 or should I go about this another way? Feb 25 19:54:03 PsyMan, just do a clean and put 972 back in place then apply that patch Feb 25 19:59:19 ehm Feb 25 19:59:25 that won't work Feb 25 19:59:38 ping nbd Feb 25 19:59:51 ? Feb 25 19:59:52 clean keeps some objects Feb 25 20:00:02 kernel ones included Feb 25 20:02:02 well I get away without distcleaning to much time to make the toolchain again Feb 25 20:02:19 it will end up skipping the whole part and use the previous objects Feb 25 20:02:27 heh Feb 25 20:02:46 i didn't the last couple of times Feb 25 20:02:47 netprince: by making sure that it is executable on the host Feb 25 20:03:10 netprince: don't source any target specific files outside of boot(), start() or stop(), don't call target commands either Feb 25 20:04:30 ahh, almost done Feb 25 20:04:33 xMff: ok, thanks. I was looking for a command in the install section of the Makefile or something, but didn't see it. This seems easy enough... Feb 25 20:04:40 compiling packages Feb 25 20:05:08 PsyMan, let me know how it works out when you test that Feb 25 20:05:17 netprince: the buildroot will do it automatically, source all initscripts and call their enable() procedure Feb 25 20:05:19 it will take some hours Feb 25 20:05:32 netprince: therfore its important that it does not fail during a simple evaluation Feb 25 20:05:43 i'll have to test the non fixed .37 build i make Feb 25 20:06:04 xMff: ok, that makes sense, thanks Feb 25 20:06:05 then if necessary the older one (i doubt i'll have to) Feb 25 20:06:24 then rebuild and test the patched one Feb 25 20:06:49 oh, and bring back those patches i deleted before doing so Feb 25 20:07:54 revert any local changes and svn up it to be sure Feb 25 20:09:48 hmm Feb 25 20:10:51 gotta remember to run the config cmds after that Feb 25 20:11:21 oh wait, same kernel, no need to Feb 25 20:12:35 now how was the syntax of that file editor Feb 25 20:12:59 aa to edit and wq to exit saving Feb 25 20:13:22 or was it x something to edit :\ Feb 25 20:14:13 esc for command mode, i for edit mode Feb 25 20:14:26 in command mode :x or :wq to save & quit Feb 25 20:15:01 ny the way, did it have to be THAT editor? Feb 25 20:15:28 why not use something not forcing you to remember a trillion of cmds? Feb 25 20:15:29 xD Feb 25 20:17:45 this is bloat disguised as something minimalistic Feb 25 20:18:23 vi ? Feb 25 20:18:35 would u prefer emacs ? :) Feb 25 20:19:07 ehm Feb 25 20:19:09 actually Feb 25 20:23:13 what was the name of that other editor Feb 25 20:24:45 joe Feb 25 20:25:33 pico! Feb 25 20:28:18 ah, build done too Feb 25 20:41:03 k Feb 25 20:41:08 applied the patch Feb 25 20:41:11 now to build Feb 25 20:42:46 PsyMan: it is actually one of the smallest editors busybox offers I think Feb 25 20:42:50 apart from cat Feb 25 20:47:42 heh Feb 25 20:49:37 problem with the easy alternatives like nano or pico is that they need ncurses Feb 25 20:52:24 makes sense Feb 25 20:52:37 you have limited space in there :\ Feb 25 20:53:17 the problem is not just "easyness" Feb 25 20:54:15 it's just a file editor so all those extras are not needed most of the time Feb 25 20:55:44 extras? Feb 25 20:55:52 there are no extras Feb 25 20:56:00 its how vi works Feb 25 20:56:06 save to tmp, edit from left, from right, etc Feb 25 20:56:07 :p Feb 25 20:56:27 well if that is too much bloat for you I suggest to learn "cat" Feb 25 20:56:43 hah Feb 25 22:00:42 nano has issues with locking. pico's no better Feb 25 22:00:52 nano, pico and joe all have issues with line endings and spacing. Feb 25 22:01:47 what zile or mg? Feb 25 22:01:50 *about Feb 25 22:02:45 assuming you configure a .zile to not drop stray backup files that is Feb 25 22:44:52 jow * r25713 /trunk/package/base-files/files/etc/hotplug.d/iface/10-routes: [package] base-files: fix sysctl handling, do not react on fake ifaces in route hotplug handler Feb 25 23:23:52 jow * r25714 /trunk/package/relayd/ (Makefile files/ files/relay.hotplug files/relay.sh): [package] relayd: add uci integration Feb 25 23:26:00 nbd * r25715 /trunk/target/linux/ar71xx/Makefile: ar71xx: update to 2.6.37.1 Feb 25 23:26:46 nbd * r25716 /trunk/target/linux/ar71xx/ (config-2.6.36 patches-2.6.36/): ar71xx: remove support for 2.6.36 Feb 26 00:05:58 jow * r25717 /branches/backfire/package/base-files/files/etc/hotplug.d/iface/10-routes: [backfire] merge r25713 Feb 26 00:12:58 jow * r25718 /branches/backfire/package/ (6 files in 4 dirs): [backfire] backport relayd, libubox and libjson-c Feb 26 00:18:18 nbd * r25719 /packages/libs/libjson-c/: packages: remove libjson-c, it is now in backfire and trunk Feb 26 00:30:40 nbd * r25720 /trunk/target/linux/atheros/Makefile: atheros: update to 2.6.37.1 Feb 26 00:30:45 nbd * r25721 /trunk/target/linux/atheros/ (10 files in 6 dirs): atheros: remove support for older kernels Feb 26 00:35:35 nbd * r25722 /trunk/include/netfilter.mk: move nf_{conntrack,nat}_tftp to ipt-nathelper-extra, most people don't need this Feb 26 00:39:25 Wipster Feb 26 00:39:30 it works now :) Feb 26 00:47:05 nbd * r25723 /trunk/package/redboot-ar231x/ (. Makefile): add redboot-ar231x (LZMA compressed redboot for a few devices) Feb 26 01:23:55 PsyMan, excellent thanks Feb 26 01:24:13 only problem was Feb 26 01:24:33 i made the image too large that there was no free space left Feb 26 01:24:37 only 3 blocks Feb 26 01:24:57 it was a disaster xD Feb 26 01:25:29 it could not even save its settings Feb 26 01:35:32 lost an 3 hours for nothing :~( Feb 26 01:35:59 ok, i didn't, wasn't around while compiling so meh Feb 26 01:36:00 :p Feb 26 01:36:23 well, changing the config and compiling again doesn't take much time Feb 26 01:37:11 it does when you clean everything up and being ready to delete everything Feb 26 01:37:46 yeah Feb 26 01:41:06 was like Feb 26 01:41:27 "if it doesn't work, i'll just revert to the previous ones i made" Feb 26 01:42:05 so i ended up with 3 images that left only 3 blocks of the flash free Feb 26 01:43:02 damn the number 3 xD Feb 26 02:05:35 so Feb 26 02:06:01 how many free blocks are needed for a jffs partition? Feb 26 02:13:38 5 Feb 26 02:15:26 hmm Feb 26 02:15:31 ahh Feb 26 02:15:34 lucky me Feb 26 02:16:45 so all needed is to replace wpa authentication with wpad Feb 26 02:17:02 and delete hmm Feb 26 02:18:31 ahh Feb 26 02:18:50 the extra language Feb 26 02:19:02 on luci Feb 26 02:19:09 simle stuff **** ENDING LOGGING AT Sat Feb 26 02:59:57 2011