**** BEGIN LOGGING AT Thu Feb 11 02:59:57 2010 Feb 11 07:19:47 build #8 of octeon is complete: Failure [failed compile_15] Build details are at http://tksite.gotdns.org:8010/builders/octeon/builds/8 Feb 11 08:30:15 kaloz * r19576 /trunk/target/linux/iop32x/ (Makefile config-default): [iop32x]: upgrade to 2.6.32.7, only build squashfs images by default Feb 11 08:35:35 kaloz * r19577 /trunk/include/kernel-version.mk: we use per-target kernel versions for 2.6 for a long time now Feb 11 08:38:01 kaloz * r19578 /trunk/ (5 files in 4 dirs): with no users left, remove support for 2.6.21 Feb 11 08:40:21 kaloz * r19579 /trunk/target/linux/generic-2.6/ (config-2.6.21 patches-2.6.21/): remove 2.6.21 patches and config, too Feb 11 08:55:17 kaloz * r19580 /trunk/target/linux/ (29 files in 26 dirs): re-sort all target and subtarget config files Feb 11 08:56:12 what's wrong with this command? Feb 11 08:56:15 root@OpenWrt:/mnt/newpackages# opkg remove freeswitch_1.0.4-1_ifx_easy_50812.ipk No packages removed. Feb 11 08:56:38 just do opkg remove freeswitch afaik Feb 11 09:02:00 i did it thx Feb 11 09:02:04 now i have root@OpenWrt:/mnt/newpackages# opkg list | grep free freeswitch-mod-sofia - 1.0.4-1 - Feb 11 09:02:18 how am i supposed to demove that last freeswitch package? Feb 11 09:03:24 opkg remove freeswitch-mod-sofia ? Feb 11 09:03:30 just do opkg remove packagename Feb 11 09:03:40 without version or file extention Feb 11 09:06:39 <_trine> opkg remove freeswitch-mod-sofia Feb 11 09:10:14 root@OpenWrt:/mnt/newpackages# opkg remove freeswitch-mod-sofia No packages removed. Feb 11 09:15:29 nico are you here? Feb 11 09:15:49 {Nico}: i would like to have you speaking a little bit with mikej Feb 11 09:17:45 <_trine> Lalloso, do have that package installed Feb 11 09:17:53 <_trine> ^^you Feb 11 09:18:14 i think i have it because i see it in opkg list Feb 11 09:18:18 le me try with -force Feb 11 09:18:33 <_trine> opkg list_installed Feb 11 09:23:43 mmm Feb 11 09:23:56 on my system if i run that commands Feb 11 09:24:12 i see also packages that are not real packages but were installed in the squashfs+jffs2 image Feb 11 09:24:19 and if i try to remove them Feb 11 09:24:21 root@OpenWrt:/mnt/newpackages# opkg remove zlib Segmentation fault Feb 11 09:24:24 it segz Feb 11 09:24:34 ^_^ Feb 11 09:25:53 * Only have 1060 available blocks on filesystem /, pkg freeswitch-mod-sofia needs 1679 Feb 11 09:26:06 bad me Feb 11 09:26:40 i have 1M free and the ipk file is 600k Feb 11 09:26:51 what's wrong here? Feb 11 09:33:18 kaloz * r19581 /trunk/target/linux/ (45 files in 30 dirs): clean up handling of some kernel symbols Feb 11 10:07:45 _trine: do you know how the openwrt build system work? Feb 11 10:08:15 <_trine> not specifically Feb 11 10:11:16 Lalloso: just ask your question, perhaps somebody with knowledge sees it and answers it Feb 11 10:11:46 it's not a question Feb 11 10:11:59 not a specific one at least Feb 11 10:12:09 at the moment i'm reading this https://forum.openwrt.org/viewtopic.php?id=6809 Feb 11 10:12:44 i'm interested in better understanding how the initial svn checkout of openwrt interact with feeds Feb 11 10:13:10 and therefore how the packages differ from the main checkout Feb 11 10:32:57 nbd: ping Feb 11 10:36:21 which is the favourite webserver + scripting language on openwrt? Feb 11 10:50:10 * Packages were found, but none compatible with the architectures configured Feb 11 10:50:23 could i hack the ipk file to bypass this blokcing error? Feb 11 11:08:42 ehi my friends Feb 11 11:08:46 instead of using opkg Feb 11 11:09:04 can i just untar manually the ipkgs on the filesystem? Feb 11 11:09:14 i know i will mess the whole database but i'll reflash then Feb 11 11:09:37 yes Feb 11 11:13:22 who is in charge of create the symbolic links inside /usr/lib? Feb 11 11:13:40 is there some control script or is it inside the tar? Feb 11 11:16:19 ? Feb 11 11:16:31 you mean the symlinks to *.so files? Feb 11 11:16:34 yes Feb 11 11:16:42 this happens during compilation Feb 11 11:17:00 okay so they should be inside the ipkgs and merely copied over the filesystem right? Feb 11 11:17:01 the packaging just copies *.so* into the package Feb 11 11:17:04 yes Feb 11 11:17:07 okay Feb 11 11:17:19 why is there a debian-binary in every .ipk? Feb 11 11:17:30 legacy stuff Feb 11 11:17:37 k thx a lot jow_laptop Feb 11 11:17:43 it was supposed to be dpkg compatible at some point Feb 11 11:18:01 gh :) Feb 11 11:18:25 in the end the only interesting part is inside data.tar.gz as far as i've understood up to now Feb 11 11:21:49 correct Feb 11 11:22:57 strange Feb 11 11:23:05 the binary seg faults Feb 11 11:23:24 in the end opkg was right telling me that architecture were not compatible Feb 11 11:23:35 what i really don't understand is why Feb 11 11:23:48 the binary compiled for A target works on B target Feb 11 11:24:04 but the binary compiled for B target segfaults on A target Feb 11 11:24:22 maybe B is a superset of A, feature-wise Feb 11 11:24:22 i would prefer both segfaults for harmony of the unisverse Feb 11 11:25:07 B is a MIPS 24Kc while A is a 34K Feb 11 11:25:30 you mean a superset in terms of hardware/OS right? Feb 11 11:25:38 cpu instructions Feb 11 11:26:08 thx for the hint jow_laptop Feb 11 11:26:12 i will investigate this with mips Feb 11 11:26:37 was the binary compiled with soft-float? Feb 11 11:26:46 i also have to say that B is a vanilla openwrt with proper toolchain compiled with openwrt Feb 11 11:26:52 and if not, do both targets have the Kernel-FPU emulator enabled? Feb 11 11:27:09 while A is compiled with openwrt but with external toolchain arrived from Infineon Feb 11 11:28:06 could you please give me an hint on how to verify the two things? Feb 11 11:28:20 soft-float and kernel-fpu? Feb 11 11:29:29 checkung for the kernel fpu emulator is easy, just read dmesg and check for algorithmics bla bla emulator enabled Feb 11 11:29:48 as for checking the binary, I don't know Feb 11 11:30:23 you should check the exact signal the binary receives when it crashes Feb 11 11:31:12 okay thx Feb 11 11:31:20 i thought sigsev is just one signal Feb 11 11:31:24 but maybe it's not so Feb 11 11:31:37 e.g. SIGFPE for floating point exeception or SIGSEGV for segmentation fault Feb 11 11:32:49 mmm Feb 11 11:33:55 there is also a signal for bad instructions Feb 11 11:34:01 sigill or so Feb 11 11:34:15 but wouldn't these signals provide a different output rather than Segmentation Fault? Feb 11 11:34:30 I think ive seen sometimes FPU exception message somewhere Feb 11 11:35:31 as far as concerned the first check i think i have fpu emulator on both A and B Feb 11 11:35:32 Algorithmics/MIPS FPU Emulator v1.5 Feb 11 11:37:00 ok, then it shouldn't matter whether the binary was build with or without soft-float Feb 11 11:37:59 okay so the reason for which it crashes must be in somewhere else Feb 11 11:40:52 changing the subject... is there an openwrt device with ADSL also? Feb 11 11:57:25 can i issue a make package-compile without doing the full make before? Feb 11 12:01:20 well you need a complete toolchain at least Feb 11 12:04:34 okay Feb 11 12:04:48 i thought openwrt understands that and build what is needed :) Feb 11 12:35:52 @[florian]: in between got kernel lzma compressed out of original fw image, uncompressed and tried netbooting it. wag54g2 freezes, still don't know what bootloader expects... Feb 11 12:36:20 would be possible bootloader disassembly or an /dev/kmem dump analisys? Feb 11 12:44:05 btw, cpu is arm1026ej, mith MMU/PMU support Feb 11 13:07:12 juhosg * r19582 /trunk/include/kernel-build.mk: sort Kconfig symbols automatically to make Kaloz happy :p Feb 11 13:15:01 is there any known issue regarding ubi-media.h? Feb 11 13:26:16 ah ha here it is https://forum.openwrt.org/viewtopic.php?pid=101856 Feb 11 13:31:42 juhosg * r19583 /trunk/target/linux/generic-2.6/patches-2.6.32/ (028-module_exports.patch 270-scci_sd_section_fix.patch): Feb 11 13:31:42 kernel: 2.6.32: move discarding of SYMTAB entries to the right place. Feb 11 13:31:42 This helps to avoid linker errors on various targets. Also remove the Feb 11 13:31:42 scsi_sd section fix patch, because it is not needed now. Feb 11 13:31:46 juhosg * r19584 /trunk/target/linux/mpc83xx/patches/001-rb600.patch: mpc83xx: fix device registration on the RB600 Feb 11 13:31:48 juhosg * r19585 /trunk/target/linux/mpc83xx/config-default: mpc83xx: sync kernel config Feb 11 13:31:50 juhosg * r19586 /trunk/target/linux/mpc83xx/config-default: mpc83xx: enable jffs2 Feb 11 13:31:54 juhosg * r19587 /trunk/target/linux/mpc83xx/ (5 files in 2 dirs): mpc83xx: add 2.6.32 support Feb 11 13:49:40 is there a target in openwrt for omap (arm thingie in androids) already? Feb 11 13:51:19 or better: whats the goldfish for? Feb 11 14:02:36 yeah, I just try it :) Feb 11 14:14:33 phaidros: goldfish is andriod :-) Feb 11 14:14:43 android* Feb 11 14:14:44 thepeople: yay! Feb 11 14:15:10 thats kewl, so I see if I can get eg busybox bnary running on my droid :) Feb 11 14:17:39 well I should be clearer, it is android interface on openwrt :-) Feb 11 14:20:13 thepeople: but that means that the build chain is arm-eabi, right? Feb 11 14:20:29 or isn't it the arm platform then? Feb 11 14:21:37 not sure, I haven't followed it that closely. Feb 11 14:23:04 k, I test and report back :) Feb 11 14:56:06 is ftp://ftp.all.kernel.org/pub/linux/kernel/v2.6/linux-2.6.30.10.tar.bz2 down? Feb 11 14:56:39 and? Feb 11 14:56:54 i mean is it down also for you or just for me? Feb 11 15:16:15 workshere Feb 11 15:16:24 try http:// instead of ftp:// Feb 11 15:16:31 maybe your firewall is blocking stuff Feb 11 15:21:09 it works but very slow thx Feb 11 15:23:20 i'm having problem with make v=99 Feb 11 15:23:24 when it compiles mtd-utils Feb 11 15:23:25 i think Feb 11 15:23:36 is there a way to disable such compilation in make menuconfig or is it needed? Feb 11 15:23:45 i've already removed "mtd" Feb 11 15:40:10 how can i know which gcc version is being compiled in the toolchain created by openwrt? Feb 11 15:54:24 i've found it it's under advanced configuration options for developers :D Feb 11 17:23:33 build #6 of iop32x is complete: Failure [failed compile_7 compile_15] Build details are at http://tksite.gotdns.org:8010/builders/iop32x/builds/6 Feb 11 19:11:23 cshore: ping Feb 11 19:28:33 ping thepeople Feb 11 19:29:48 tripolar_: see pm Feb 11 19:44:24 thepeople: ok, doesnt work :/ Feb 11 19:44:45 phaidros: :-( Feb 11 19:44:56 I know it was working on the android emulator Feb 11 19:45:00 would have been to easy ;) Feb 11 19:45:04 but that was as far as it made it Feb 11 19:45:47 phaidros: what fun is working, working provides no challenge :-P Feb 11 19:46:04 nah, now I have to setup the android toolchain .. Feb 11 19:46:18 openwrt feels more home though Feb 11 19:46:24 or work on the openwrt port :-) Feb 11 19:46:46 nah, not capable of this .. I consider myself more as advanced user ;) Feb 11 20:16:17 rtz2: ping Feb 11 20:56:29 kaloz can u add me to the list? Feb 11 20:56:38 the wiki löist u just edited Feb 11 20:57:25 just tell me in private what you want to have there ("what" should be the list of packages) Feb 11 20:59:00 Kaloz did u get my text? Feb 11 20:59:29 yeah Feb 11 21:05:53 ping nbd Feb 11 21:23:58 build #6 of ixp4xx is complete: Failure [failed compile_7 compile_15] Build details are at http://tksite.gotdns.org:8010/builders/ixp4xx/builds/6 Feb 11 22:02:31 ping Kaloz Feb 11 22:07:28 ping thepeople Feb 11 22:12:38 ping larsc Feb 11 22:29:27 nbd: ping Feb 12 00:26:27 thepeople * r19588 /packages/net/chillispot/patches/150-chillispot-rmtctrl.patch: add a internal socket for chilli, so you can control connection from external script, thanks fofware Feb 12 00:36:14 thepeople * r19589 /packages/net/remotectrl/ (16 files in 2 dirs): commit package to go along with the last commit, thanks fofware Feb 12 00:38:16 a pretty generic package name ;) Feb 12 00:38:33 yep Feb 12 00:38:50 i keep forgetting to update my router Feb 12 00:39:47 xMff: maybe :-P Feb 12 00:40:03 I'll call my next package "package" :D Feb 12 00:40:16 xMff: i dare you Feb 12 00:41:46 hehe Feb 12 00:41:56 I will call my next package OpenWrt :-) Feb 12 00:42:04 either that or something insanely specific Feb 12 00:43:25 thispackageallowscontrolofchillispotfromexternalscripts Feb 12 00:43:29 :-P Feb 12 00:43:49 thepeople * r19590 /packages/net/remotectrl/Makefile: fix category Feb 12 00:44:11 thispackageallowscontrolofchillispotfromexternalscriptsondaysendingindayduringmonthswithmorethen29days Feb 12 00:44:25 now make that a nice abbreviation Feb 12 00:45:06 tpaccfesdeddmwmt29d? Feb 12 00:45:14 It's probably easier to run md5sum over it Feb 12 00:45:16 yep :-D Feb 12 00:45:54 i'm still picturing one of you actually doing this and i can't stop lauging Feb 12 00:46:15 just wait it will happen Feb 12 00:46:43 trolling the openwrt userbase = hilarity Feb 12 00:47:46 trolling the developers trolling the openwrt userbase = awsomeness Feb 12 00:48:21 trolling the internet = awesome Feb 12 00:48:26 hmm Feb 12 00:48:32 wonder if i can find that pic Feb 12 00:49:27 http://forums.xkcd.com/download/file.php?id=19928 Feb 12 00:52:53 lol Feb 12 00:53:41 I hope his dad had backups.... Feb 12 01:03:52 thepeople: as in now supprts sending remote data to chilli daemon ? Feb 12 01:05:03 dingo_: yes, I believe so Feb 12 01:05:14 ping thepeople Feb 12 01:05:14 dingo_: fofware would be the one to ask for sure Feb 12 01:05:20 tripolar_: pong Feb 12 01:05:43 thepeople can u commit the 2.6.32.8 update i send to the mailing list? Feb 12 01:05:49 tripolar_: yea Feb 12 01:06:06 updating brcm47xx too would also be nice Feb 12 01:06:13 to 2.6.32 Feb 12 01:08:26 tripolar_: you should ask nbd or [florian] about that Feb 12 01:09:12 thepeople i asked nbd but he had no time and ment he will look at it .. this was about one week ago Feb 12 01:10:00 tripolar_: have you tested it? Feb 12 01:10:42 yeah for some weeks by now Feb 12 01:10:49 without any problem Feb 12 01:10:57 lol but remotecontrol is a dummy package :) was my mistake and my bad english to explain to thepeople what i need :) Feb 12 01:11:35 nbd also ment he will disable devtmpfs in 2.6.32 because its not used by now .. but he didnt Feb 12 01:11:51 hello xMff, dingo_, Weedy and all Feb 12 01:11:57 fofware: what package was I supposed to grab? Feb 12 01:12:13 fofware: so whats tis remote control Feb 12 01:12:13 thepeople * r19591 /trunk/ (12 files in 12 dirs): update kernel from 2.6.32.7 to 2.6.32.8, thanks tripolar Feb 12 01:14:01 tripolar_: patch? Feb 12 01:14:01 remote control it a package that i did to try make a package with c programs and to try sockets handlle before put it in chillispot Feb 12 01:14:10 thepeople its also in the mail Feb 12 01:14:13 fofware: so I can remove it? Feb 12 01:14:17 for brcm47xx Feb 12 01:14:25 yeah thepeople Feb 12 01:14:31 thepeople: or do u mena the devtmpfs patch? Feb 12 01:14:43 tripolar_: yea the devtmpfs Feb 12 01:14:58 thepeople * r19592 /trunk/target/linux/brcm47xx/Makefile: update brcm47xx to .32, thanks tripolar Feb 12 01:15:00 remote control do nothing for openwrt it is a simple chat Feb 12 01:15:15 betwing programs Feb 12 01:15:31 fofware: specifically chillispot Feb 12 01:16:15 thepeople: http://dpaste.com/157706/ Feb 12 01:16:21 thepeople * r19593 /packages/net/remotectrl/: woops, this was unneeded Feb 12 01:17:20 tripolar_ should dcc u the patch? Feb 12 01:17:36 sorry thepeople, and thanks Feb 12 01:18:43 fofware: np Feb 12 01:18:53 thepeople * r19594 /trunk/target/linux/generic-2.6/config-2.6.32: disable devtmpfs in 2.6.32 because its not used, thanks tripolar Feb 12 01:18:59 dingo_: in patch for chillispot I add to it, a socket so with external script you can check connections, autorize and reject Feb 12 01:19:25 thepeople: thanks Feb 12 01:19:30 tripolar_: np Feb 12 01:19:50 dingo_: with this socket connectio is more easy show in webpage what happend inside chillispot Feb 12 01:20:10 fofware: yupp i follow you Feb 12 01:20:19 thepeopleif i make more patches over time will it be possible to get more rights later in time? Feb 12 01:21:00 tripolar_: it is possible, it is something the core developers will need to consider when the time comes Feb 12 01:21:10 okay Feb 12 01:23:19 thepeople i'll go now have a nice day Feb 12 01:23:20 bye Feb 12 01:37:04 thepeople * r19595 / (8 files in 5 dirs): (log message trimmed) Feb 12 01:37:04 Here is LVM support for OpenWRT. It introduces 3 new packages: Feb 12 01:37:04 - kmod-dm for the device mapper kernel module. Feb 12 01:37:04 - kmod-libdevmapper for the device-mapper library. Feb 12 01:37:04 - kmod-lvm2 for the lvm tool. Feb 12 01:37:04 Note that it currently does not create the symlinks for all the lvm Feb 12 01:37:05 tools, so you have to use lvm pvs rather than just pvs. Feb 12 01:38:28 ping netprince Feb 12 01:40:14 netprince, remember the udhcp commit (r19530) ? What do you think about this solution: http://openwrt.pastebin.com/m25b2197 Feb 12 01:40:59 thepeople * r19596 /trunk/ (3 files in 3 dirs): (log message trimmed) Feb 12 01:41:00 [tools/wrt350nv2-builder] Update to v2.1 Feb 12 01:41:00 Stays closer to the stock firmware by using wrt350n.bin for the Feb 12 01:41:00 created image. Feb 12 01:41:00 Adds the possibility to create the image in two separate steps (-b / -z). Feb 12 01:41:00 Makefiles now support the possibility to include a u-boot image for a Feb 12 01:41:01 100%-stock-like image. Feb 12 01:42:05 a complete revert of the whole section would probably work too Feb 12 01:47:18 thepeople: ping Feb 12 01:48:23 fofware: pong Feb 12 01:49:01 ok, patch work ok, that I did about package, is for chillispot package Feb 12 01:49:48 thepeople: I don't know if i can do a patch of makefile or only send a new one Feb 12 01:51:19 thepeople: to add a uci confile for chilli and new init.file that read configuration form /etc/config/chillispot and start chillispot with proper settings Feb 12 01:51:29 fofware: send me a patch, that way I know that what I have is correct Feb 12 01:51:45 ok Feb 12 01:52:36 thepeople: the patch for chillipackage I do the same way to patch the source ? Feb 12 01:53:00 yep Feb 12 01:53:08 thepeople: thanks Feb 12 01:53:13 or you can do a svn diff Feb 12 01:53:19 when everything is in place Feb 12 01:54:58 ok I will read how do a svn diff if i don't understand do a simple diff Feb 12 02:06:58 jow * r19597 /trunk/package/base-files/ (3 files in 3 dirs): [package] base-files: revert dhcp state information on ifdown and restore pre-r19530 behaviour in udhcp default.script **** ENDING LOGGING AT Fri Feb 12 02:59:56 2010