**** BEGIN LOGGING AT Fri Apr 23 02:59:56 2010 Apr 23 03:31:43 nico * r21099 /packages/net/strongswan4/ (Config.in Makefile): Apr 23 03:31:43 [packages] strongswan4: move in VPN submenu, add 2 more config options and changes from #2538: Apr 23 03:31:43 * add a strongswan4-default package (matching upstream autoconf defaults) Apr 23 03:31:43 * add dependency on kmod-crypto-authenc Apr 23 03:31:43 * fix 2 typos Apr 23 03:37:47 nico * r21100 /packages/net/strongswan4/Makefile: [packages] stronswan4: remove copy/paste leftovers (oops) Apr 23 04:01:04 nico * r21101 /trunk/package/kernel/modules/ (001-depends.mk fs.mk netsupport.mk other.mk): package/kernel: introduce a new file that will be sourced first and holding complex dependencies spread over multiple files, use it for kmod-crc16 Apr 23 04:08:28 nico * r21102 /trunk/package/kernel/modules/ (001-depends.mk other.mk usb.mk): package/kernel: use macros for kmod-rfkill depends Apr 23 06:04:52 gmorning Apr 23 10:53:03 jow * r21103 /packages/ipv6/6scripts/ (Makefile files/6tunnel.init): [packages] 6scripts: fix incorrect route command (#7203) Apr 23 10:55:28 jow * r21104 /trunk/package/kernel/modules/usb.mk: [package] kernel: package cypress m8 usb-to-serial module (#7210) Apr 23 10:56:52 jow * r21105 /branches/backfire/package/kernel/modules/usb.mk: [backfire] merge r21104 Apr 23 10:56:52 nico * r21106 /packages/net/strongswan4/Makefile: [packages] strongswan4: mark padlock plugin as x86 only Apr 23 11:27:32 jow * r21107 /trunk/package/ (11 files in 11 dirs): [package] cleanup Makefiles (#7212) Apr 23 11:32:54 hi, I get the following problem on openwrt trunk with a wr741nd: https://dev.openwrt.org/ticket/7131 Apr 23 11:33:24 I would like to try to solve the problem, but I would need some help getting started Apr 23 11:36:45 sKAApGIF:find the place, where the warning comes from and then try to understand the code Apr 23 11:45:15 <[florian]> rtz2: ping? Apr 23 11:46:43 [florian]: pong Apr 23 11:48:23 <[florian]> rtz2: have you had a chance to post your switch suggestions on netdev? Apr 23 11:49:10 [florian]: not yet Apr 23 11:49:29 acoul * r21108 /trunk/target/linux/brcm47xx/ (3 files in 3 dirs): brcm47xx: fix phy isolation bit on Toshiba WRC-1000, Siemens SE505 v1, Askey RT-210W, RT-220W and alike boards (closes #6789) Apr 23 11:49:42 [florian]: I will need to work out something reasonable solid and I have my thermodynamics exam on monday Apr 23 11:49:53 so I didn't have that much time Apr 23 11:50:00 <[florian]> rtz2: no issue Apr 23 11:58:56 [florian]: by the way, did you try gcc 4.5 on rdc? Apr 23 11:59:16 <[florian]> rtz2: not yet, neither did I try 2.6.32 actually Apr 23 11:59:47 ok Apr 23 12:00:13 [florian]: I have some problems with gcc 4.5 I can't track down Apr 23 12:00:39 sed crashes during module loading and other things break at random times :/ Apr 23 12:02:56 @rtz2: are you using: CONFIG_CC_OPTIMIZE_FOR_SIZE ? Apr 23 12:03:47 I think so Apr 23 12:05:30 I don't think it's the default setting that's why I asked Apr 23 12:05:54 in a background script, how do I log error messages from a long running command? I've been using "ssh (...) 2>&1 | logger -p (...) -t (...)", but his has the inconvinient of keeping also one instance of logger always opened Apr 23 12:06:38 any error message would go to the log anyway, but with a wrong tag if I don't specify it... Apr 23 12:07:12 the logger instance should die if the script piped to exits Apr 23 12:07:18 *to it Apr 23 12:08:06 jow_laptop: the problem is that it runs for a long time...and during that time it keeps living(of course!) Apr 23 12:08:44 only way I see is to send everthing to a log file, and if it quits, then send the log file to logger Apr 23 12:10:20 this is used in a /etc/init.d/ script, otherwise error messages are logged anyway, but with the wrong tag.. Apr 23 12:11:14 nunojpg: whats so bad about the logger instance? Apr 23 12:12:40 [florian]: what is package broadcom-57xx all about? it looks like it is a driver for 5325 switch on spi, which would be what i need, but the depends are "Depends on: (TARGET_brcm_2_4 || TARGET_brcm47xx || TARGET_brcm63xx) && TARGET_brcm_2_4" Apr 23 12:13:23 this is for a ssh tunneling package with UCI config that I'm about to commit. Each tunnel have currently one process for the script itself, one for ssh, and one for logger. That is 3x Nr of tunnels. 2x would be a lot better Apr 23 12:14:07 also it won't compile... Apr 23 12:14:13 nunojpg: logger is part of busybox, it is resident in memory anyway, there should be not much overhead Apr 23 12:14:28 is trunk/Rules.make the same is trunk/rules.mk? Apr 23 12:14:46 jow_laptop: makes sense... Apr 23 12:15:21 <[florian]> ali12341: broadcom-57xx is the switch driver for certain bcm47xx based board with gigabit interfaces Apr 23 12:15:43 yeah? in the code it says it supports 5325E which is what i have... Apr 23 12:34:18 <[florian]> ali12341: the switch part can certainly work with more than just a tg3 adapter Apr 23 12:34:43 it seems to be quite dependent on 2.4 kernel, i can't get it to compile Apr 23 12:35:47 <[florian]> it basically is doing spi over gpio Apr 23 12:36:06 <[florian]> I would start with a fresh implementation Apr 23 12:36:12 <[florian]> which uses the linux spi API instead Apr 23 12:36:21 i'm looking at spidev now Apr 23 12:36:52 <[florian]> also take a look at spi-ks8995 which is a good example of how things work Apr 23 13:12:47 build #66 of ppc44x is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ppc44x/builds/66 Apr 23 13:39:26 build #68 of ppc40x is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ppc40x/builds/68 Apr 23 13:40:56 jow * r21109 /trunk/package/dnsmasq/ (Makefile files/dnsmasq.init): Apr 23 13:40:56 [package] dnsmasq: Apr 23 13:40:56 - when processing domain entries, append searchdomain to host entries if they are not fully qualified Apr 23 13:40:56 - create only one reverse record for address entries with multiple names Apr 23 13:43:59 jow * r21110 /branches/backfire/package/dnsmasq/ (Makefile files/dnsmasq.init): [backfire] merge r21109 Apr 23 13:53:17 ah this ^ is also maybe my problem, in which i needed to power cycle my cable ISP modem to get ip from it on router.. (and thats only needed if i reboot openwrt router) Apr 23 13:53:59 build #65 of atheros is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/atheros/builds/65 Apr 23 13:54:35 crow: nope Apr 23 14:03:12 jow_laptop ok to bad :) Apr 23 14:13:16 crow: but you get the same mac each time? Apr 23 14:13:49 jow_laptop my mac is same each time, and that mac address is registred with modem to isp... Apr 23 14:14:09 and it only happens with openwrt? Apr 23 14:14:32 jow_laptop well pc direkt (and cloned mac from router) passed every reboot... and power cycle Apr 23 14:48:33 nunojpg * r21111 /packages/net/sshtunnel/ (11 files in 7 dirs): [packages] add sshtunnel, Manages Local and Remote openssh ssh(1) tunnels Apr 23 15:07:06 jow_laptop but if i reboot openwrt router i cant get internet working (even worked before reboot) Apr 23 15:28:30 jow * r21112 /branches/backfire/package/hotplug2/patches/ (5 files): [backfire] merge r20679-r20685 Apr 23 15:32:25 jow * r21113 /branches/backfire/package/base-files/files/ (etc/hotplug2-common.rules sbin/hotplug-call): [backfire] merge r20683 Apr 23 15:50:31 acoul * r21114 /trunk/package/kernel/modules/usb.mk: [package] fix kmod-nls-base dependency with usb-core (#6753) Apr 23 17:08:00 <_trine> there a small bug in the newest trunk Apr 23 17:11:08 <_trine> in the default /etc/config/network file Apr 23 17:11:28 <_trine> option 'gateway' ' Apr 23 17:11:34 <_trine> and it should be Apr 23 17:11:37 <_trine> option 'gateway' '' Apr 23 17:11:59 <_trine> ie a missing ' Apr 23 17:16:28 which platform? Apr 23 18:39:08 [florian]: i got spidev working, now the module hangs inside wait_for_completion() - not sure if it is a IRQ problem, or just because whatever is connected on spi doesn't respond... Apr 23 20:22:53 [florian]: spi seems dead, no interrupts. but looking in the registers, i see what looks like valid data in the rx buffer. so CFE must be using it successfully Apr 23 20:45:26 hello, i'm trying to enable the intel iwlwifi driver in mac80211, i'ts all working except iwl4965 because there is a patchfile missing from compat-wireless tarball Apr 23 20:46:12 the md5sum of the compat-wireless tarball from mirror2.openwrt.org does not match the one from kernel.org Apr 23 20:46:52 and at least patches/08-rename-iwl4965-config.patch is missing.. so what am i missing about how its supposed to work? Apr 23 20:47:18 i checked both compat-wireless-2010-04-16.tar.bz2 and compat-wireless-2010-04-21.tar.bz2 Apr 23 20:48:19 this is the url mentioned in the mac80211/Makefile - http://wireless.kernel.org/download/compat-wireless-2.6/ Apr 23 20:49:11 but it's commented out and the tarball comes from - http://mirror2.openwrt.org/sources Apr 23 20:49:48 md5s do not match and at least one file is missing... that's not a mirror Apr 23 20:54:16 hmm it's true, http://mirror2.openwrt.org/sources/compat-wireless-2010-04-21.tar.bz2 != http://wireless.kernel.org/download/compat-wireless-2.6/compat-wireless-2010-04-21.tar.bz2 Apr 23 20:59:00 ok i see now that nbd switched to wireless-testing a couple months ago Apr 23 20:59:20 still sorting it out... Apr 23 21:00:22 my issue is the patch that went missing is breaking iwlagn build with support for iwl4965 Apr 23 21:10:54 ok so the log entry for the revison to use compat-wireless-2010-04-21.tar.bz2 says update to wireless-testing 2010-04-21 Apr 23 21:21:31 i dont see where it comes from.. out of time for now. Apr 23 21:26:32 ok, https://dev.openwrt.org/changeset/19110... so it's creation by nbd Apr 23 21:27:03 nbd: any idea what happened to patches/08-rename-iwl4965-config.patch in the unoffcial compat-wireless snapshots? Apr 23 21:52:26 <_trine> I'm having problems using ipv6 from my pc Apr 23 21:53:31 <_trine> ipv6 works on the router but not from the PC I have just been chatting to the very helpful people on #ipv6 but they seem to think my configs all look ok Apr 23 21:53:38 <_trine> so I'm stuck Apr 23 21:58:17 <_trine> this was all working not long ago so I am wondering if anything has changed that could account for this Apr 23 23:09:20 nunojpg * r21115 /packages/utils/open2300/ (. Makefile patches/ patches/x1.patch patches/x2.patch): [packages] add open2300, WS23XX weather station utilities Apr 24 00:30:01 craigc * r21116 /packages/net/multiwan/ (Makefile files/etc/config/multiwan files/usr/bin/multiwan): Updated for better use with ppp and dns requests Apr 24 00:53:10 framer99: will fix that on the next wireless-testing sync Apr 24 02:05:35 nbd: ok thanks, I was very confused for a while there. I'll submit the patch to mac80211/Makefile for enabling iwlwifi tonight or tomorrow **** ENDING LOGGING AT Sat Apr 24 02:59:56 2010