**** BEGIN LOGGING AT Sat Mar 27 02:59:56 2010 Mar 27 03:06:08 build #32 of orion is complete: Failure [failed compile_7] Build details are at http://tksite.gotdns.org:8010/builders/orion/builds/32 Mar 27 05:03:50 build #32 of rb532 is complete: Failure [failed shell_3 compile_6] Build details are at http://tksite.gotdns.org:8010/builders/rb532/builds/32 Mar 27 05:21:54 build #42 of sibyte is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/sibyte/builds/42 Mar 27 09:53:50 hauke * r20502 /trunk/package/mac80211/patches/014-add_iw_handler.patch: mac80211: fix build of ipw2200 Mar 27 12:42:38 nbd, with your latest mac80211 updates: Mar 27 12:42:43 nbd, http://openwrt.pastebin.com/XXdEqEhh Mar 27 12:42:44 :( Mar 27 12:42:51 what i'm missing? Mar 27 12:44:37 dogge10: ignore the ioctl[SIOCSIWENCODEEXT]: Invalid argument Mar 27 12:44:43 dogge10: they're harmless Mar 27 12:45:00 dogge10: run /sbin/wifi -x &> /tmp/log and read why they happen Mar 27 12:45:20 but iwconfig -> wlan0 IEEE 802.11bgn Mode:Managed Access Point: Not-Associated Mar 27 12:46:02 logread, dmesg, ... tcpdump on an independent machine with monitor mode interface Mar 27 12:46:19 with the info you give, nobody can help you anyway Mar 27 12:46:58 and try #openwrt as this looks more like a user problem, imho Mar 27 12:47:28 that's to much for me. ok. then i've to wait a few more month until it gets fixed by someone else with more knowledge on debugging such stuff... Mar 27 12:48:14 xMff, btw. uhttpd still dies after a few seconds :( Mar 27 12:48:21 ath9k works fine in client mode Mar 27 12:49:22 well, at least not here Mar 27 12:49:32 client mode with ath9k will always use wpa_supplicant, check if wpa_supplicant is running, if the config looks OK, ... Mar 27 12:49:47 or wpad Mar 27 12:49:54 wpad = wpa_supplicant Mar 27 12:50:14 ahh, right, it was a multicall binary Mar 27 12:50:19 the program is still called wpa_supplicant on the router Mar 27 12:50:24 seems wpad isn't running Mar 27 12:50:45 wpad isn't even a binary, afaik Mar 27 12:51:00 wpad is just a pseudo package for both hostapd+wpa_supplicant, afaik Mar 27 12:51:06 ok. that's fine: Mar 27 12:51:07 1323 root 1660 S wpa_supplicant -B -P /var/run/wifi-wlan0.pid -D wext Mar 27 12:51:37 stintel: no, i'm pretty sure it is a binary Mar 27 12:52:03 stintel: it's hostapd and wpa_supplicant in one mutlicall binary Mar 27 12:52:06 the command line isnt' complete because it's too long, but you can get that from /proc/[PID]/cmdline Mar 27 12:52:10 rtz2: ic Mar 27 12:52:21 rtz2: still, wpa_supplicant is running instead of wpad Mar 27 12:52:29 same if you use ap mode, hostapd is used, not wpad Mar 27 12:52:34 maybe that's incomplete Mar 27 12:52:44 they should be symlinks Mar 27 12:52:51 stintel: like busybox Mar 27 12:53:12 hrrmmm Mar 27 12:53:26 wpa_supplicant-B-P/var/run/wifi-wlan0.pid-Dwext-iwlan0-c/var/run/wpa_supplicant-wlan0.conf Mar 27 12:53:29 you're right, my bad Mar 27 12:53:58 dogge10: well, add spaces where needed, stop the already running one, then start it again with "-dd" as extra option Mar 27 12:54:33 try to find hints where it goes wrong Mar 27 12:54:41 if you can't find those nobody will be able to help you Mar 27 12:55:34 your config is correct, and ath9k works fine in client mode here, so that means something else must be wrong, up to you to figure out what. Mar 27 12:57:04 http://openwrt.pastebin.com/YQ106HKg Mar 27 12:57:08 bad :( Mar 27 12:57:36 ah, my bad, remove the -B Mar 27 12:57:42 to prevent it going in backgroudn Mar 27 12:59:44 dogge10: no details no fix Mar 27 12:59:53 btw I had uhttpd crashing too, but that seems fixed after another make dirclean (make package/uhttpd/clean package/luci/clean did'nt fix) Mar 27 13:00:05 xMff: :-) Mar 27 13:00:29 ah no Mar 27 13:00:32 still crashes here too Mar 27 13:00:45 nothing in logread, nothing in dmesg Mar 27 13:00:53 can you strace it? Mar 27 13:01:51 I'll test it on x86 Mar 27 13:02:09 I think I have strace, yes Mar 27 13:02:22 select(): Interrupted system call Mar 27 13:02:28 when I run it manually in foreground Mar 27 13:02:36 but that's probably not very helpful Mar 27 13:03:08 itan indicator, I need the strace to see why Mar 27 13:03:15 *thats an Mar 27 13:03:27 xMff: on the way Mar 27 13:03:46 http://openwrt.pastebin.ca/1853846 Mar 27 13:04:50 dogge10: 27|13:57:34 < stintel> ah, my bad, remove the -B 27|13:57:41 < stintel> to prevent it going in backgroudn Mar 27 13:05:19 juhosg * r20503 /trunk/target/linux/ar71xx/files/drivers/net/ag71xx/ag71xx_main.c: Mar 27 13:05:19 ar71xx: ag71xx: Add netpoll support Mar 27 13:05:19 This patch adds support for polling interrupts, used by Mar 27 13:05:19 netconsole and the likes. Mar 27 13:05:19 Tested on d-link dir-615 C1 using netconsole. Mar 27 13:05:20 Signed-off-by: Pat Erley Mar 27 13:05:23 juhosg * r20504 /trunk/target/linux/ar71xx/files/drivers/net/ag71xx/ag71xx_main.c: ar71xx: ag71xx: reorganize ag71xx_rx_packet function Mar 27 13:05:26 juhosg * r20505 /trunk/target/linux/ar71xx/files/drivers/net/ag71xx/ (ag71xx.h ag71xx_ar8216.c ag71xx_main.c): ar71xx: ag71xx: introduce ag71xx_has_ar8216() helper Mar 27 13:05:28 juhosg * r20506 /trunk/target/linux/ar71xx/files/drivers/net/ag71xx/ag71xx_main.c: ar71xx: ag71xx: copy received packet to avoid unaligned access in the IP stack Mar 27 13:05:46 stintel: ah of course Mar 27 13:06:19 stupid late night hacking Mar 27 13:06:22 xMff: :) Mar 27 13:06:33 better sleep at night Mar 27 13:06:52 look at my v3 patch series, already v3 and didn't even receive comments from other people :P Mar 27 13:07:26 hmmmm I hope these change to ag71xx don't break ar8316 switch support ! :) Mar 27 13:08:22 dogge10: why sleep if you can do something useful Mar 27 13:09:01 sleeping is useful :) Mar 27 13:09:14 dogge10: so any more output when you run with -dd ? Mar 27 13:09:16 ehr Mar 27 13:09:20 without -B ? Mar 27 13:10:12 sleeping is useless, I only do it because I have to :) Mar 27 13:10:24 if I didn't need to I wouldn't ever sleep :) Mar 27 13:10:27 no idea. maybe someone else fixes it Mar 27 13:11:26 dogge10: that's a bad attitude Mar 27 13:11:28 dogge10: wth dude, just run wpa_supplicant again like before but without -B, and paste the output Mar 27 13:12:12 dogge10: you are constantly hilighting people, expect them to drop everything to help you, now it's too much for you to do some debugging ? Mar 27 13:12:16 you're going on my /ignore Mar 27 13:12:44 sorry for my rant Mar 27 13:16:56 but the guy needs a cluepon :) Mar 27 13:18:15 well, I have to say, you are right Mar 27 13:18:31 he is kinda annoying and if he doesn't even want to help ... Mar 27 13:18:47 although I probably shouldn't have spoken, I do my share of highlighting too, and I know can be annoying :) Mar 27 13:19:30 i'm too, but you are willing to debug stuff and look for solutions yourself :) Mar 27 13:20:33 I probably shouldn't have bought rspro and be running trunk if that wasn't the case ;-) Mar 27 13:22:00 stintel: care to test a patch? Mar 27 13:25:01 xMff: sure Mar 27 13:25:35 http://openwrt.pastebin.com/wyKZUxv7 Mar 27 13:26:21 I "overlooked" this when doing the sigchld stuff and haven't seen it on brcm-2.4 even if it is kinda obvious Mar 27 13:26:39 hrrrm, patch failed, maybe I need some white-space ignore option? Mar 27 13:26:48 ah Mar 27 13:27:45 patch failed? Mar 27 13:28:27 xMff: http://openwrt.pastebin.com/HHG5ct1K Mar 27 13:28:32 am I doing something wrong ? Mar 27 13:29:30 maybe svn up is needed? Mar 27 13:30:15 or maybe its pastebin again... http://luci.subsignal.org/~jow/uhttpd-4.patch Mar 27 13:30:24 xMff: don't think so, using git, git-svn-id: svn://svn.openwrt.org/openwrt/trunk@20506 3c298f89-4303-0410-b956-a3cf2f4a3e73 Mar 27 13:30:47 xMff: yeah, pastebin definitely Mar 27 13:30:51 florian * r20507 /trunk/package/avila-wdt/src/avila-wdt.c: [ixp4xx] fix avila-wdt compilation failure (#6948) Mar 27 13:31:19 florian * r20508 /packages/net/openser/patches/007-modules_build.patch: [package] exclude postgresql from openser, not packaged and does not build Mar 27 13:31:23 florian * r20509 /packages/libs/libnids/patches/003-missing_else.patch: [package] fix libnids compilation failure (#6947) Mar 27 13:31:28 florian * r20510 /packages/utils/gnupg/patches/ (. 001-mips_gcc4.4.patch): [package] fix gnupg compilation failure with gcc-4.4 (#6951) Mar 27 13:32:02 xMff: make package/uhttpd/clean world should be fine right Mar 27 13:32:09 yes Mar 27 13:32:41 or just make package/uhttpd/{clean,compile} maybe and upload the ipk :) Mar 27 13:32:48 better indeed Mar 27 13:33:02 getting better at this :) Mar 27 13:35:59 xMff: still seems to crash, http://openwrt.pastebin.com/Xhedtiww Mar 27 13:38:51 ok Mar 27 13:41:16 stintel: actually the hwclock applet is *enabled* in openwrt ;-) -> https://dev.openwrt.org/browser/trunk/package/busybox/config/util-linux/Config.in#L284 Mar 27 13:41:22 KanjiMonster: heh Mar 27 13:41:53 KanjiMonster: strange, I rm .config, make menuconfig, select target and rspro profile and it was disabled for me ? Mar 27 13:42:13 maybe some other leftovers in tmp/, dunno, oh well Mar 27 13:42:44 not that big of a deal :) Mar 27 13:54:28 stintel: new approach, same url Mar 27 13:55:09 florian * r20511 /packages/utils/bash/Makefile: [package] also install rbash (#6927) Mar 27 13:55:11 florian * r20512 /packages/ipv6/dibbler/ (Makefile files/ files/dibbler.init): [package] fix dibbler-server pid creation (#6906) Mar 27 13:57:04 xMff: just a sec Mar 27 13:57:14 looks like I have some other problem: http://openwrt.pastebin.com/12gwRLtf Mar 27 13:57:38 when scp'ing firmware over wireless to the rspro Mar 27 13:57:58 ouch Mar 27 13:58:53 and it happens 3 times in a row :( Mar 27 14:05:23 xMff: looks good now, but I made an error during testing the previous patch so it could have worked too, sorryy Mar 27 14:05:30 s/y$// Mar 27 14:06:08 you want me to retry the previous patch too, or do you like this approach better ? Mar 27 14:06:12 stintel: enable kernel debug symbols ;p Mar 27 14:06:33 KanjiMonster: I was afraid something like that would follow :P Mar 27 14:07:21 imo the rs pro has enough flash to have them always enabled Mar 27 14:07:23 CONFIG_KERNEL_KALLSYMS ? Mar 27 14:07:38 KanjiMonster: the same reason I prefer real package over busybox applet :D Mar 27 14:07:53 build #52 of ppc40x is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ppc40x/builds/52 Mar 27 14:08:00 thats the one Mar 27 14:08:23 stintel: I like this approach better but it still has races, its a piti that uclibc does not support pselect(), that would solve this problem cleanly Mar 27 14:10:11 xMff: ok. fyi, it's still running, so it's a lot better already Mar 27 14:12:24 xMff: it does Mar 27 14:12:33 rtz2: you sure? Mar 27 14:13:04 xMff: well, I think I used ppoll, but i'm 100% sure I had one of them running Mar 27 14:13:43 I got implicit declaration + undefined reference, if it is there, it should have been provided by the same headers, right? Mar 27 14:14:07 maybe I need one of those funny feature macro defines Mar 27 14:15:15 xMff: ok, I used ppoll Mar 27 14:16:28 xMff: and no macro or whatever trickery Mar 27 14:16:39 xMff: it worked out of the box Mar 27 14:17:10 xMff: fyi, I seem to get no graphs in Statistics > Graphs, while there are images in /tmp/rrdimg/OpenWrt/*/ Mar 27 14:17:47 rtz2: I see. Well I'll do a userspace emulation of it now, still prone to races but the probability is low enough to ignore it Mar 27 14:18:54 the Mar 27 14:19:00 are there though Mar 27 14:19:35 xMff: are you really sure, pselect isn't there? Mar 27 14:20:04 rtz2: not anymore :P but when I tried it on brcm-2.4 it didn't work out - could be a 2.6 only thing Mar 27 14:21:18 stintel: yes, I am aware of this, the image directory is symlinked into the docroot and since the paths are normalized with realpath() it is detected as out-of-docroot and thus forbidden Mar 27 14:21:43 stintel: will look into it next Mar 27 14:22:06 ok Mar 27 14:23:03 gahhhh Mar 27 14:23:26 the 4716 seems to need a completly new network driver Mar 27 14:23:41 this will become ugly ./ Mar 27 14:23:43 :/ Mar 27 14:26:30 do e.g. the asus sources contain one in source form? Mar 27 14:27:42 KanjiMonster: yes, but the code for the sonics silicon backplane is completly different Mar 27 14:28:03 from the on currently in the kernel Mar 27 14:28:29 I either have to use everything or do a complete rewrite :/ Mar 27 14:31:37 jow * r20513 /trunk/package/uhttpd/ (6 files in 2 dirs): [package] uhttpd: block SIGCHLD until it is expected (#6957) Mar 27 14:34:11 KanjiMonster: I don't seem to get anything more after enabling CONFIG_KERNEL_KALLSYMS :/ Mar 27 14:34:27 have you cleaned the kernel? Mar 27 14:34:28 hmmm Mar 27 14:34:37 I am really messing up everything :/ Mar 27 14:34:48 stintel: that's not the right option Mar 27 14:34:52 :P Mar 27 14:35:07 stintel: there should be one in openwrt menuconfig Mar 27 14:35:27 that's where I get it :) Mar 27 14:36:01 stintel: ohh, right Mar 27 14:36:08 Global build settings > Kernel build options > compile the kernel with symbol table info Mar 27 14:36:11 stintel: I only know it under the disaply name Mar 27 14:36:35 :) Mar 27 14:36:48 stintel: try to clean out the kernel tree then Mar 27 14:37:04 stintel: you can also look up the addresses in your System.map Mar 27 14:37:23 well not anymore after a recompile Mar 27 14:40:16 seems I either just didn't enable it, or didn't save config when exiting menuconfig Mar 27 14:40:37 like running a patch with --dry-run then don't do it again without Mar 27 14:40:45 I'm gonna make some coffee Mar 27 14:42:54 :D Mar 27 14:59:25 xMff: when I switch luci feed from 0.9 to trunk or back, should make package/luci/clean be enough, or not really ? Mar 27 14:59:36 clean should be enough Mar 27 15:02:36 Any ideas, are the in-kernel atheros drivers now workin on mips4k cpus? Mar 27 15:25:37 heh, now I got my image with kallsyms enabled on the test rspro, my main rspro ap crashed :P Mar 27 15:28:19 jow * r20514 /trunk/target/linux/x86/base-files/lib/upgrade/platform.sh: [x86] sysupgrade: Dump image to sda device instead of hda (patch by Acinonyx) Mar 27 15:29:23 jow * r20515 /trunk/target/linux/x86/ (5 files in 4 dirs): [x86] Remove support for older kernels (patch by Acinonyx) Mar 27 15:30:47 jow * r20516 /trunk/target/linux/x86/config-default: [x86] Change default processor family to 486, refresh config (patch by Acinonyx) Mar 27 15:30:52 I'll upgrade my main rspro first, enable kallsyms on it while at it, and see if I still hit kernel bugs Mar 27 15:31:21 jow * r20517 /trunk/target/linux/x86/olpc/config-default: [x86] Add OLPC kernel config (patch by Acinonyx) Mar 27 16:35:44 k seems with both client and ap running r20517 I no longer have the oops Mar 27 17:13:35 jow * r20518 /trunk/package/base-files/ (3 files in 3 dirs): [package] base-files: add a hotplug handler for rdate Mar 27 17:17:03 xMff luci frontend for rdate? Mar 27 17:19:34 dogge10: patience for you? Mar 27 17:20:51 hehe Mar 27 17:21:00 can rdate disabled? is it optional? Mar 27 17:21:04 florian * r20519 /packages/utils/fowsr/ (. Makefile): [package] add fowsr (#6952) Mar 27 17:21:07 florian * r20520 /packages/net/nocatauth/patches/004-fw_detect.patch: [package] fix nocatauth check for iptables (#6929) Mar 27 17:21:09 florian * r20521 /packages/net/openser/Makefile: [package] fix openser module linking issues (#6932) Mar 27 17:30:11 yes, no Mar 27 17:30:30 just remove all servers and it is disabled Mar 27 17:38:28 juhosg * r20522 /trunk/tools/firmware-utils/src/trx.c: Mar 27 17:38:28 firmware-utils/trx: add relative offset parameter Mar 27 17:38:28 * will be used for the WRT160NL board Mar 27 17:38:28 * patch by Bernhard Loos Mar 27 17:38:28 * note: rejected parts has been fixed Mar 27 17:38:31 juhosg * r20523 /trunk/package/mtd/src/ (crc32.h trx.c): Mar 27 17:38:32 package/mtd: reuse existing crc32 stuff in the trx code Mar 27 17:38:32 * patch by Bernhard Loos Mar 27 17:38:32 * note: fixed compiler error on brcm47xx Mar 27 17:38:35 juhosg * r20524 /trunk/package/mtd/src/ (mtd.c trx.c): Mar 27 17:38:35 package/mtd: add fixtrx command Mar 27 17:38:35 * patch by Bernhard Loos Mar 27 17:38:35 * note: trailing whitespaces are removed Mar 27 17:38:38 juhosg * r20525 /trunk/target/linux/ar71xx/ (files/drivers/mtd/wrt160nl_part.c image/Makefile): Mar 27 17:38:38 ar71xx: update image generation and trx header parsing for the WRT160NL Mar 27 17:38:38 * patch by Bernhard Loos Mar 27 17:38:38 * note: trailing whitespaces has been removed Mar 27 17:38:42 juhosg * r20526 /trunk/target/linux/ar71xx/base-files/etc/uci-defaults/wrt160nl: Mar 27 17:38:42 ar71xx: add uci defaults to fix trx header on the WRT160NL boards Mar 27 17:38:42 * patch by Otto Solares Mar 27 17:57:00 xMff rdate frontend. nice. so, ntpclient is no longer required :) Mar 27 19:54:26 how is the gui going for uhttpd, xMff ? Mar 27 19:56:34 I have mor important stuff on my todo, get the stuff working thats already there for example Mar 27 19:56:44 dogge10 what was biggest image you flashed on your wl500gpv1 ? i am looking at what max is. Mar 27 19:57:07 can't remember, sorry Mar 27 19:57:53 dogge10: I am sure patches to add the funtionality are welcome Mar 27 20:00:18 xMff i did some more tcpdumps on router and pc, and i also got real hub "netgeard ds-108" was realy cheep, just big :) Mar 27 20:08:13 cshore: ping Mar 27 20:16:48 mirko * r20527 /packages/Xorg/lib/vte/Makefile: vte depends on gdkx and therefore libX11 Mar 27 20:40:33 thepeople: pong Mar 27 20:41:18 cshore: do you want to be listed as maintainer of freeswitch (I think you were working on it) Mar 27 20:41:36 https://dev.openwrt.org/wiki/packages Mar 27 20:41:49 I think {Nico} is the primary maintainer Mar 27 20:46:11 build #28 of x86 is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/x86/builds/28 Mar 27 20:48:48 build #34 of iop32x is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/iop32x/builds/34 Mar 28 00:34:51 build #47 of ar71xx is complete: Exception [exception failed slave lost shell_19 compile_15] Build details are at http://tksite.gotdns.org:8010/builders/ar71xx/builds/47 Mar 28 00:34:59 build #33 of kirkwood is complete: Exception [exception failed slave lost shell_19 compile_15] Build details are at http://tksite.gotdns.org:8010/builders/kirkwood/builds/33 Mar 28 00:35:16 build #33 of ixp4xx is complete: Exception [exception failed slave lost shell_19 compile_15] Build details are at http://tksite.gotdns.org:8010/builders/ixp4xx/builds/33 Mar 28 00:35:45 nbd * r20528 /trunk/target/linux/ar71xx/files/arch/mips/ (ar71xx/devices.c include/asm/mach-ar71xx/ar71xx.h): ar71xx: reset the mdio bus on ar7241/ar7242 Mar 28 00:36:05 nbd * r20529 /trunk/target/linux/ar71xx/files/arch/mips/ar71xx/mach-pb92.c: ar71xx: fix the wan port on pb92 Mar 28 00:36:38 nbd * r20530 /trunk/target/linux/ar71xx/files/arch/mips/pci/pci-ar724x.c: ar71xx: fix pci on ar7241/ar7242 Mar 28 01:27:09 nico * r20531 /trunk/package/opkg/patches/009-remove-upgrade-all.patch: [package] opkg: fix usage to reflect change in the upgrade command Mar 28 02:18:19 hey does that mean there no command to upgrade everything in one fell swoop anymore using opkg? Mar 28 02:18:48 right, it is to dangerous Mar 28 02:19:05 even on platforms with a fully writeable fs, it cannot upgrade the kernel image Mar 28 02:19:29 it will happily update kmods and the kernel meta package while leaving the old linux image around Mar 28 02:19:47 in the worst case the system won't boot anymore due to lacking disk support Mar 28 02:20:03 hm Mar 28 02:20:10 right....should we dot with extroot then...since it's going to have many packages that will not updating after a flash Mar 28 02:20:14 actually, i think opkg upgrade is a good idea in regular scenarios Mar 28 02:20:26 for instance if you only upgrade packages within a stable release it's fine Mar 28 02:20:28 not=need Mar 28 02:20:50 just upgrades to a different release are a bad idea Mar 28 02:20:54 we thought about using opkg in the buildroot, this would give us the ability to flag certain packages as hold Mar 28 02:20:56 maybe we can come up with a way of enforcing that Mar 28 02:21:03 then we can reenable upgrade all Mar 28 02:21:04 such as marking packages with a dependency on the release Mar 28 02:21:12 automatically at build time Mar 28 02:21:17 yes this is already possible Mar 28 02:21:32 that way we could re-enable upgrades and use them for security fixes and stuff Mar 28 02:21:34 but the ipkg.sh script we use right now is legacy Mar 28 02:21:58 it cannot flag stuff and creates the package list slightly wrong Mar 28 02:22:03 is that a problem? Mar 28 02:22:03 oh Mar 28 02:22:28 imho we should fix that in the shell script, possibly removing all stuff that we don't need Mar 28 02:22:37 i don't think we need to build a full-blown opkg for the host Mar 28 02:22:49 well I'd rewrite it in perl Mar 28 02:22:54 fine with me Mar 28 02:23:10 seriously, the stuff this ipkg.sh does can be done with 20 lines of perl or something Mar 28 02:23:17 sounds good Mar 28 02:27:54 or just replace ipkg.sh with opkg Mar 28 02:28:22 i think replacing it with a simple script is better Mar 28 02:28:37 who knows what portability issues we'd have to deal with when compiling it for the host Mar 28 02:29:08 nbd: it is on my list to do after 10.03 Mar 28 02:29:11 updated openwrt/upstream, http://pastehtml.com/view/5tzq30c.html, wip Mar 28 02:35:47 jow * r20532 /trunk/package/uhttpd/ (Makefile src/uhttpd.c): [package] uhttpd: fix parameter parsing issue on avr32 and most likely ixp4xx **** ENDING LOGGING AT Sun Mar 28 02:59:56 2010