**** BEGIN LOGGING AT Thu Jul 16 02:59:57 2009 Jul 16 06:22:38 nico * r16855 / (2 files in 2 dirs): [packages] tiff: add a tiff-utils subpackage Jul 16 07:04:46 nico * r16856 /branches/8.09/target/linux/generic-2.6/patches-2.6.23/ (4 files): [kernel] add 2.6.23 security fixes for rb532 missed in 8.09.1 Jul 16 07:13:03 nico * r16857 /branches/8.09/target/linux/generic-2.6/ (4 files in 4 dirs): Jul 16 07:13:03 [kernel] fix buffer overflow in RTL8169 NIC driver Jul 16 07:13:03 - CVE-2009-1389 Jul 16 07:53:56 <[florian]> fofware: pong Jul 16 07:54:43 Hello <[florian]> I did a script to build images based on env setting Jul 16 07:55:12 I guess can be good to include in buildroot Jul 16 07:57:21 <[florian]> yes that's probably a good idea, could you send it to the mailing-list with a short description of what it does Jul 16 07:57:33 http://openwrt.pastebin.com/d6ebbca02 Jul 16 07:57:51 ok I will do Jul 16 07:58:17 Now I'm testing and I need write the help Jul 16 08:00:20 <[florian]> ./scripts/env is new or how long time is in buildroot Jul 16 08:00:47 <[florian]> it's been there for a couple of months now Jul 16 08:00:54 <[florian]> but few people used it before Jul 16 08:01:25 ah!!! because a long time i was waiting for something like it Jul 16 08:01:47 It's a great script Jul 16 09:35:17 lars * r16858 /trunk/target/linux/s3c24xx/patches-2.6.30/056-pcf50633.patch: [s3c24xx] Implement voltage listing for pc50633. Jul 16 09:39:59 well I managed to brick my POS.. Jul 16 09:43:23 JTAG FTW Jul 16 09:49:04 everybody: should there be a nuwiki announcement on the openwrt mainpage? Jul 16 09:49:34 danage: actually no, I shorted the flash pins 9&10 and apparently I found out the hard way you can set the nvram default IP to whatever you want and it'll stick indefinitely Jul 16 09:49:45 a whatwiki? Jul 16 09:49:56 re: channel topic Jul 16 10:00:29 lars * r16859 /trunk/target/linux/s3c24xx/files-2.6.30/arch/arm/mach-s3c2442/mach-gta02.c: Jul 16 10:00:29 [s3c24xx] mach_gta02: Add some missing always_on regulator constraints and Jul 16 10:00:29 declare regulator constraints to be fully specified. Jul 16 10:00:59 CIA-29: huh? Jul 16 10:19:17 florian * r16860 /trunk/ (14 files in 8 dirs): (log message trimmed) Jul 16 10:19:17 This patch adds a target for the IBM PowerXCell Accelerator Board. Jul 16 10:19:17 (aka mvXCell-8i from MatrixVision or GigaAccel 180 from Fixstars) Jul 16 10:19:17 http://us.fixstars.com/products/gigaaccel/ Jul 16 10:19:17 This build will create zImage suitable for TFTP boot image. Jul 16 10:19:18 * v2 Jul 16 10:19:22 - add axonram device driver Jul 16 10:35:37 #!/usr/bin/env bash Jul 16 10:35:37 BLD_BASEDIR="$PWD" Jul 16 10:35:37 BLD_BRANCHLIST=$(ls env/.git/refs/heads 2>/dev/null | grep -vE '^master') 2>/dev/null Jul 16 10:35:37 BLD_TARGETLIST=${BLD_BRANCHLIST:-""} Jul 16 10:35:37 BLD_FEEDLIST="-a" Jul 16 10:35:38 BLD_FEEDFLAG=0 Jul 16 10:35:39 BLD_BASEIMAGEDIR="$PWD/images" Jul 16 10:35:41 BLD_CLEANFLAG= Jul 16 10:35:43 BLD_CLEANLIST= Jul 16 10:35:45 clean_check () { Jul 16 10:35:47 [ "$BLD_CLEANFLAG" = "1" ] && { Jul 16 10:35:49 local CLEAN= Jul 16 10:35:51 for CLEAN in $BLD_CLEANLIST; do Jul 16 10:35:53 [ "$CLEAN" = "$MYTARGET" ] && make clean V=99 && break Jul 16 10:35:55 done Jul 16 10:35:57 } Jul 16 10:35:59 } Jul 16 10:36:01 while getopts 'f:c:b:h' OPTION Jul 16 10:36:03 do Jul 16 10:36:05 case $OPTION in Jul 16 10:36:07 f) BLD_FEEDFLAG=1 Jul 16 10:36:09 BLD_FEEDLIST=${OPTARG:-"-a"} Jul 16 10:36:11 ;; Jul 16 10:36:13 c) BLD_CLEANFLAG=1 Jul 16 10:36:15 BLD_CLEANLIST=${OPTARG:-$BLD_BRANCHLIST} Jul 16 10:36:17 ;; Jul 16 10:36:19 b) BLD_TARGETLIST=${OPTARG:-$BLD_BRANCHLIST} Jul 16 10:36:21 ;; Jul 16 10:36:23 h|?) printf "Usage: [options]\n" $(basename $0) >&2 Jul 16 10:36:25 printf " -f feed : -a | feed_name(1) [feed_name(2) [feed_name(n)]]\n" Jul 16 10:36:27 printf " -c make clean: branch_name(1) [branch_name(2) [branch_name(n)]]\n" Jul 16 10:36:29 printf " -b make : branch_name(1) [branch_name(2) [branch_name(n)]]\n" Jul 16 10:36:31 exit 2 Jul 16 10:37:29 [florian]: what is the address of mail list to send the script and if you can take a look http://openwrt.pastebin.com/d1a171d2b Jul 16 10:37:41 I'm sorry wrong paste Jul 16 11:28:56 nbd * r16861 /trunk/package/madwifi/patches/441-fix_ibss_node_handling.patch: madwifi: clean up / fix ibss node handling Jul 16 11:30:04 nbd * r16862 /trunk/package/madwifi/files/lib/wifi/madwifi.sh: madwifi: turn software based tsf merging on by default, get rid of some iwpriv errors on wifi restart Jul 16 11:37:32 Why the hell are people spreading FUD that mmc-over-gpio is broken? It does work properly, if you fix up your config. I did test it recently and it works just fine. Jul 16 11:43:11 mb__: pu that into nuwiki please :) Jul 16 11:44:03 I did it twice now Jul 16 11:44:22 well, forum Jul 16 11:44:38 Do we really need to put something into the wiki if it is NOT broken? That sounds weird Jul 16 11:54:52 i would agree to that, maybe a write up on how to configure it properly? Jul 16 11:56:02 mb__: no but perhaps a general description so people know what they deal with and don't make the mistakes that lead them to think it's broken Jul 16 11:58:16 Well, it's obvious: Use the correct GPIOs and don't use the GPIO twice. Jul 16 11:58:30 That should really be obvious to anybody poking with hardware Jul 16 11:59:37 mb__: yeah, well so should the concept of signaling before changing over 5 lanes of traffic, lol Jul 16 12:00:17 ;) Jul 16 12:00:19 mb__: if you've ever had the joy of living on the East side of Los Angeles... dont.. Jul 16 12:06:51 lars * r16863 /trunk/target/linux/ (12 files in 5 dirs): [s3c24xx] Cleanup glamo platform_data. Jul 16 12:10:53 dStruct: haha Jul 16 12:11:04 me == westside Jul 16 12:11:22 danage: no really, i've never wanted to get out of my car and physically yell at somebody due to their inability to drive a vehicle Jul 16 12:11:36 danage: well I live in Tucson, AZ now, but i'm from Orange County Jul 16 12:13:09 hehe i been to tucson, you probably frying over there right now Jul 16 12:16:09 danage: it's not too bad, it was 108 the other day heh, but I was nice and cool inside Jul 16 12:16:40 danage: it's really not that bad with no humidity, if it was humid even a little bit and 108 you'd fry for sure Jul 16 12:17:36 hey why isn't Ash Optimize for Size enabled by default? Jul 16 12:19:14 lol at migloo's quit msg heh Jul 16 12:21:17 _migloo_: i trust everything.. came out ok lol Jul 16 12:23:32 danage: if I use psk2 in /etc/config/wireless will it pull hostapd-mini and do it that way, or does it need wpa-supplicant installed? Jul 16 12:31:39 tag 8.09 is still using b43 not bcm43xx? Jul 16 12:49:38 florian * r16864 /packages/net/aoetools/Makefile: [package] update aoetools to 30 (#5523) Jul 16 12:51:58 florian * r16865 /packages/net/netdiscover/patches/01-initfix+end-fix.patch: [package] fix bugs in netdiscover (#5529) Jul 16 13:13:27 lars * r16866 /trunk/target/linux/s3c24xx/files-2.6.30/arch/arm/mach-s3c2442/mach-gta02.c: Jul 16 13:13:27 [s3c24xx] Reverese mangle_pmu logic. Initalize regulator constrains to those Jul 16 13:13:27 used for all but GTA02v1 and changed them for GTA02v1 in mangle_pmu. Jul 16 13:39:38 jow * r16867 /trunk/target/linux/x86/base-files/lib/upgrade/platform.sh: Jul 16 13:39:38 [PATCH] sysupgrade-ext2 sync before writing image, not after Jul 16 13:39:38 Sysupgrade for ext2 currently flushes the filesystem buffers after an Jul 16 13:39:38 image is written to the disk. This should happen before the image is Jul 16 13:39:38 written. Jul 16 13:40:52 jow * r16868 /branches/8.09/target/linux/x86/base-files/lib/upgrade/platform.sh: Merge r16867 to 8.09 Jul 16 15:21:59 can someone link me to a copy of the default .config for tag 8.09 please? Jul 16 15:22:23 kernel 2.6 Jul 16 15:24:25 <{Nico}> dStruct: the .config for each target in 8.09.1 is available in it's download folder (i.e. http://downloads.openwrt.org/kamikaze/8.09.1/atheros/atheros.config) Jul 16 15:24:58 {Nico}: how about for broadcom? Jul 16 15:25:20 well that's 8.09.1 Jul 16 15:25:48 i need the config that was used to build http://downloads.openwrt.org/kamikaze/8.09/brcm47xx/openwrt-brcm47xx-squashfs.trx Jul 16 15:27:24 i downloaded tag 8.09, added nano, wpa-supplicant, and a couple other things I use, targeted brcm47xx 2.6 and left everything else alone and I get a read-only rootfs_data, do I need to target jffs2 instead of the default squashfs? Jul 16 15:28:42 /dev/root on /rom type squashfs (ro) Jul 16 15:29:37 <[florian]> rootfs_data is mounted as jffs2 and should be writable Jul 16 15:29:41 <[florian]> rootfs is squashfs and ro Jul 16 15:30:07 <[florian]> if you want a full jffs2 image, you should choose jffs2 in make menuconfig -> Targets Images Jul 16 15:30:59 well when I boot i get errors on the mappings, saying not enough erasable something, and I can't save any files, like on reboot they default back Jul 16 15:31:11 <[florian]> how much space do you have? Jul 16 15:31:15 but everything works on the default image Jul 16 15:31:24 4mb flash my .trx is 3.3mb Jul 16 15:31:45 i thought my upper limit was closer to 3.69mb+ Jul 16 15:32:41 rootfs 2.4M 2.4M 0 100% / Jul 16 15:32:42 tmpfs 6.7M 76.0k 6.7M 1% /tmp Jul 16 15:32:42 tmpfs 512.0k 0 512.0k 0% /dev Jul 16 15:32:42 mini_fo:/tmp/root 2.4M 2.4M 0 100% / Jul 16 15:33:53 i wish I could paste dmesg, it scrolled out of the buffer when I was trying to reconfigure my network and wireless to get back online Jul 16 15:34:42 <[florian]> you can paste dmesg at openwrt.pastebin.com Jul 16 15:35:12 well i'd have to reboot and reconfigure everything manually, i'll be back in a few minutes so i can link you to it Jul 16 15:36:00 looks like you have to setup your partitions properly dStruct Jul 16 15:36:36 danage: i just went with a default image, i thought it would just work, lemme reboot and get dmesg up Jul 16 15:36:43 brb Jul 16 15:46:46 ok here it is http://openwrt.pastebin.com/d35eacb14 Jul 16 15:47:00 yeah, i wouldn't paste more then 4 lines in here, that's my limit heh Jul 16 15:52:00 it's a .config issue right? do I just need to enable jffs2 or ramdisk, or is it more complicated then that? Jul 16 16:00:23 sorry lost my connection Jul 16 16:01:27 sorry, was having connection issues Jul 16 16:01:34 i missed anything that was said Jul 16 16:35:02 florian * r16869 /trunk/target/linux/generic-2.6/config-2.6.25: [kernel] add missing CONFIG_FIXED_PHY symbol to 2.6.25 Jul 16 16:42:49 well that sucked.. Jul 16 16:43:15 so 2.4 works fine Jul 16 16:43:32 i still get this msg tho even in 2.4 mtd: partition "rootfs" doesn't start on an erase block boundary -- force read-only Jul 16 16:43:40 but I'm able to save files like normal Jul 16 16:44:12 <[florian]> with a jffs2 image? Jul 16 16:44:24 <[florian]> whith a squashfs image this is normal Jul 16 16:45:12 using a squashfs Jul 16 16:45:24 <[florian]> then you will always get this message Jul 16 16:45:39 <[florian]> since the rootfs partition will never start on an erase block boundary Jul 16 16:45:41 but why would a clean svn or 8.09 2.6 not let me save files? Jul 16 16:46:00 s/or/of Jul 16 16:46:13 <[florian]> files should be save on the rootfs_data partition and you will see them in / thanks to the overlay Jul 16 16:47:05 i svn co'd a clean copy of tag 8.09, added a few feeds and left luci out, compiled it, flashed it, and got this dmesg http://openwrt.pastebin.com/d35eacb14 and couldn't save any files Jul 16 16:50:09 <[florian]> humm Jul 16 16:51:37 it's really weird, i can provide my .config too if that would help, I took the clean svn or tag 8.09, and menuconfig selected 2.6 bcrm47xx and left the default target of squashfs, added a few feeds, left all the major stuff alone, compiled and i couldn't save files Jul 16 16:51:48 s/or/of Jul 16 16:54:46 well I could save files, but on reboot they were default, like I have a new ssh key each time I reboot.. Jul 16 16:57:37 i even tried trunk because I heard the broadcom drivers were working good, and they were, but my stupid router kept rebooting every 10 minutes Jul 16 16:59:23 [florian]: if I svn co 8.09.1 can i enable the broadcom drivers? i'm told they dont work but i'm not having issues there, i just want a 2.6 kernel that includes wpa2 Jul 16 16:59:54 <[florian]> no they are not included in 8.09.1 afair Jul 16 17:00:04 that's what I heard Jul 16 17:00:42 well why does the 8.09 openwrt-brcm47xx-squashfs.trx work just fine, shouldn't tag 8.09 be the same files? Jul 16 17:01:25 this works 100% minus wpa http://downloads.openwrt.org/kamikaze/8.09/brcm47xx/openwrt-brcm47xx-squashfs.trx Jul 16 17:01:49 have you tried trunk? Jul 16 17:02:01 loswillios: yeah, my router reboots itself every 10 minutes =( Jul 16 17:02:28 do you have a serial cable attached? Jul 16 17:02:37 loswillios: nope, it's a Dell Truemobile 2300 Jul 16 17:02:39 v2 Jul 16 17:02:42 bummer Jul 16 17:03:09 loswillios: yeah, I have no idea why trunk would be rebooting, but it works perfectly up until that point Jul 16 17:03:40 you can bisect it, or just try older revision Jul 16 17:03:47 eh? Jul 16 17:04:16 so far i've tried trunk, and both branch and tag 8.09 Jul 16 17:04:38 ie, find a revision past 8.09 and prior trunk that works Jul 16 17:04:59 8.09 was tagged around r9000 or something Jul 16 17:05:07 loswillios: is there a way to find a more stable one? Jul 16 17:05:33 loswillios: or is there a way to tell exactly when http://downloads.openwrt.org/kamikaze/8.09/brcm47xx/openwrt-brcm47xx-squashfs.trx was made or tagged, maybe tag 8.09 isn't exact? Jul 16 17:05:49 loswillios: or could I be .config'ing something wrong? Jul 16 17:06:22 it was tagged in september 2008 Jul 16 17:06:44 well that file i linked was up loaded feb 09 Jul 16 17:07:08 yeah, it was stabilized in between Jul 16 17:08:13 loswillios: hmm good to know, so i'm most likely looking for something around feb 15th then Jul 16 17:16:30 loswillios: well what the hell i'll give trunk another whirl Jul 16 17:25:58 what's a broadcom/SiByte SB-1? Jul 16 17:30:17 oh a cpu I dont have.. so the broadcom drivers in trunk, are they 'fixed' on both 2.4 and 2.6? Jul 16 17:41:53 nbd * r16870 /trunk/package/madwifi/patches/441-fix_ibss_node_handling.patch: madwifi: fix a race condition in the ibss merge Jul 16 17:44:07 loswillios: can I use something like syslog-ng3 to dump my logs remotely? Jul 16 17:46:20 probably, dunno how much it catches though Jul 16 17:46:38 loswillios: well i just svn up'd so we'll see.. Jul 16 17:59:46 well here goes, I really hope this works Jul 16 18:57:50 x86-alix build problem Jul 16 18:58:09 cp: cannot stat `/aux/src/openwrt/staging_dir/host/usr/lib/grub/i386-*/stage1': No such file or directory Jul 16 18:58:09 cp: cannot stat `/aux/src/openwrt/staging_dir/host/usr/lib/grub/i386-*/stage2': No such file or directory Jul 16 18:58:09 cp: cannot stat `/aux/src/openwrt/staging_dir/host/usr/lib/grub/i386-*/e2fs_stage1_5': No such file or directory Jul 16 19:01:47 lars * r16871 /trunk/target/linux/s3c24xx/ (4 files in 2 dirs): [s3c24xx] Get rid of glamo-lcm-spi. It's not used and probably never will be. Jul 16 20:16:14 [florian]: ping Jul 16 20:20:15 when I have defined env.... how I compile one package, in others words, I'm writing some programs in c and modified some files, I did switched for branch where i modified files and did make package/chillisocket/compile V=99 Jul 16 20:21:12 that compile the package but I don't change nothing in build_dir Jul 16 20:21:27 that compile the package but It don't change nothing in build_dir Jul 16 20:25:20 I did found where it's the thing happend Jul 16 20:25:28 thanks anyway Jul 16 20:32:37 is it possible to compile all the lua files in LuCI as stripped bytecode? My device is pretty darn slow and the interpreter owns my CPU Jul 16 20:40:48 aport`_: try Libraries --> Build Target Jul 16 20:42:26 xMff, in menuconfig? Jul 16 20:42:45 nm found it Jul 16 20:42:52 oh snap Jul 16 20:42:56 hello xMff, where is the mailing-list of openwrt or to what address I can send script that I think can be good include in openwrt buildroot Jul 16 20:42:57 awesome Jul 16 20:43:13 fofware: lists.openwrt.org Jul 16 20:43:23 thanks xMff Jul 16 20:44:01 xMff, do you use ./scripts/env in your system? Jul 16 20:44:18 fofware: no, separate buildroots Jul 16 20:45:03 ok, so you can't test the script Jul 16 20:48:55 I'm seeing to development Is more confortable separate buildroot Jul 16 22:06:58 xMff, sorry one questipn about Data type, int have 4 bytes in all platform right? Jul 16 22:07:14 fofware: no Jul 16 22:07:29 8 on amd64 at least Jul 16 22:07:32 iirc Jul 16 22:07:35 ah! bad my Jul 16 22:09:43 so to my socket I need use a char* and conver they to int in each side Jul 16 22:13:35 you can use datatypes like uint32_t to ensure a 32bit int on all systems Jul 16 22:14:07 xMff: and number in lua have 8 bytes or it's have a variable lenght Jul 16 22:14:20 all 32bit int Jul 16 22:14:58 for all platform? Jul 16 22:15:24 yes Jul 16 22:15:38 including Commadore 64? Jul 16 22:15:49 so that you say uint32_t is my solution Jul 16 22:15:54 thanks Jul 16 22:16:17 maligor: s/all/most/ Jul 16 22:16:46 you should be aware of endianness also Jul 16 22:16:55 when converting from char* to int Jul 16 22:17:57 or other idea is use 4byte and send exadecimal values Jul 16 22:19:01 but in lua that reqire extra convertion funtions Jul 16 22:24:55 make dirclean fixed my x86 problem Jul 16 23:19:45 my wifi adapter goes to promiscuous mode when network starts but doesn't go to forwarding state untill i restarts the wifi adapter Jul 16 23:19:53 any clue Jul 16 23:46:37 iptables build broken? Jul 16 23:47:20 bzcat /src_archive/openwrt/dl/iptables-1.4.4.tar.bz2 | /bin/tar -C /aux/src/openwrt/build_dir/linux-x86_generic/iptables-1.4.4/.. -xf - Jul 16 23:47:20 ls: cannot access ./patches/1.4.4: No such file or directory Jul 16 23:48:40 is ther another 1.4.x in that dir? Jul 16 23:48:57 if yes, try symlinking that Jul 16 23:50:43 trying Jul 16 23:51:22 heh. no. patches don't apply Jul 16 23:51:32 that might not even be the problem. /me looks closer Jul 16 23:55:01 config.log sees some mips32 polution in the x86 build Jul 16 23:56:36 configure:2920: i486-openwrt-linux-uclibc-gcc -Os -pipe -mips32 -mtune=mips32 -funit-at-a-time -fhonour-copts -fpic -I/aux/src/openwrt/build_dir/linux-x86_generic/iptables-1.4.4/include -I/aux/src/openwrt/staging_dir/target-i386_uClibc-0.9.30.1/usr/include -I/aux/src/openwrt/staging_dir/target-i386_uClibc-0.9.30.1/include -L/aux/src/openwrt/staging_dir/toolchain-i386_gcc-4.1.2_uClibc-0.9.30.1/usr/lib -L/aux/src/openwrt/staging_dir/toolcha Jul 16 23:56:38 in-i386_gcc-4.1.2_uClibc-0.9.30.1/lib -L/aux/src/openwrt/staging_dir/target-i386_uClibc-0.9.30.1/usr/lib -L/aux/src/openwrt/staging_dir/target-i386_uClibc-0.9.30.1/lib conftest.c >&5 Jul 16 23:56:41 cc1: error: unrecognized command line option "-mips32" Jul 16 23:56:43 conftest.c:1: error: bad value (mips32) for -mtune= switch Jul 16 23:57:02 where do you supposed that's coming from? Jul 16 23:58:22 aha Jul 16 23:58:25 found it Jul 16 23:58:29 in my .config Jul 17 00:05:54 i thought i'd be clever and just modify a .config from a wgt634u for x86-alix. i guess it doesn't fix everything ;-) Jul 17 00:24:59 How does one cause a package to be selected if another package is selected, but *not* create a dependancy. E.g. you want a default B to be selected if A is selected, but you want it to be possible to deselect B and select C without losing A Jul 17 02:10:01 xMff: can I bother you for a minute? **** ENDING LOGGING AT Fri Jul 17 02:59:56 2009