**** BEGIN LOGGING AT Mon Jul 06 02:59:56 2009 Jul 06 03:56:09 mini_fo: build_sto_structure: failed to create storage dir [1]. Jul 06 03:56:11 mini_fo: get_neg_sto_dentry: ERROR building sto structure. Jul 06 03:56:24 what does this mean? when i reboot the trunk i saw this Jul 06 06:20:13 mirko * r16704 /packages/libs/libIDL2/Makefile: add as missing host-build-dependency for libIDL2 Jul 06 08:21:48 xMff: ping, phaidros ping Jul 06 10:52:17 florian * r16705 /trunk/target/linux/ (4 files in 4 dirs): [kernel] avoid overriding CONFIG_TUN with a couple of targets Jul 06 11:14:22 florian * r16706 /packages/utils/collectd/Makefile: [package] fix zlib dependency typo in collectd Makefile (#5457) Jul 06 11:16:05 florian * r16707 /trunk/package/i2c-gpio-custom/Makefile: [package] make i2c-gpio-custom selectable (#5458) Jul 06 11:21:23 florian * r16708 /trunk/target/linux/brcm63xx/profiles/103-No-WiFi.mk: [brcm63xx] add a profile with no Wi-Fi drivers enabled, useful for bcm6338 Jul 06 11:28:36 florian * r16709 /trunk/target/linux/brcm63xx/patches-2.6.27/500-lzma_initramfs.patch: [brcm63xx] add bcm47xx lzma initramfs patch which is useful to workaround some CFE limitations when ramdisk kernels are too big Jul 06 11:33:53 florian * r16710 /packages/net/smap/ (Makefile patches/02-segfault-and-toofast-bugfixes.patch): [package] fix a couple of bugs in smap (#5373) Jul 06 11:36:47 florian * r16711 /packages/net/netdiscover/ (Makefile patches/ patches/00-add-machine-ouput.patch): [package] print machines, patch from Daniel Dickinson (#5373) Jul 06 11:38:03 florian * r16712 /packages/net/mac-to-devinfo/ (11 files in 2 dirs): [package] add mac-to-devinfo (#5373) Jul 06 11:48:57 hcg * r16713 /trunk/target/linux/generic-2.6/config-2.6.31: Add config file for 2.6.31 kernels in preparation for 2.6.31 Jul 06 12:47:36 florian * r16714 /packages/net/remotefs/ (. Makefile files/ files/rfsd): [package] add remotefs (#5280) Jul 06 13:16:39 florian * r16715 /packages/net/openswan/Makefile: [package] fix openswan compilation (#5261) Jul 06 15:39:33 hello everyone Jul 06 15:39:58 I don't understand something about buildroot Jul 06 15:40:37 my ask is arround /target Jul 06 15:42:29 is possible make a multiple platform build, or make a build for one platform but with different selected packages at ones? Jul 06 15:43:34 or for do that i need a script to do a build for each case Jul 06 15:53:01 it might be possible to use one buildroot for multiple 2.6 arches Jul 06 15:53:09 but you need to move away .config between the builds Jul 06 15:53:20 mixing 2.4 and 2.6 does not work, this will result in weird errors Jul 06 16:00:46 ok, thanks xMff, so for exmple to build one image with xxxpackage installed and other without it, first steep is create .config for eachone Jul 06 16:01:05 yes Jul 06 16:03:08 xMff: and with a bulding script make a loop for build and make one, two, ..... , n Jul 06 16:03:50 yes Jul 06 16:04:12 xMff: and i need move each image file to other dir betwenn builds Jul 06 16:04:26 yeah or just move away bin/ Jul 06 16:06:09 ok, xMff, I did guess that buildroot can do that, but now i see is like do it by hand Jul 06 16:06:39 ok, thanks very much Jul 06 16:10:23 juhosg * r16716 /trunk/package/kernel/modules/other.mk: Jul 06 16:10:23 Add the gpio_keys input module to menuconfig Jul 06 16:10:23 Signed-off-by: Ferenc Wagner Jul 06 16:10:25 juhosg * r16717 /trunk/package/kernel/modules/fs.mk: Jul 06 16:10:25 Baltic language support in kernel Jul 06 16:10:25 Add Baltic langugae support in kernel package Jul 06 16:10:27 Signed-off-by: zintis.petersons@e-mail.lv Jul 06 18:39:16 in /lib/wifl/mac80211.sh the mode "mesh" was added a few days ago(it used to be only adhoc), but iwconfig still reports Ad-hoc after i did 'option mode mesh' in /etc/config/wireless, is this the way it should work? Jul 06 18:39:38 <[florian]> hey xxiao Jul 06 18:39:49 <[florian]> mesh is not adhoc Jul 06 18:39:59 [florian]: yes understand that. Jul 06 18:40:03 <[florian]> you are supposed to install iw for that Jul 06 18:40:36 <[florian]> and a sample configuration file would look like this: Jul 06 18:40:39 <[florian]> config wifi-iface option device "wlan0" option network lan option mode "mesh" option mesh_id "OpenWrt" Jul 06 18:40:45 [florian]: i set 'option mode adhoc' and i can use 'iw' to set up 11s, it worked fine. just curious on the 'mesh' option added to mac80211.sh recently Jul 06 18:40:46 <[florian]> sorry for the missing line breaks Jul 06 18:41:26 <[florian]> by working you mean, starting to beacon ? Jul 06 18:41:47 mesh is uci for iw + 11s Jul 06 18:42:10 [florian]: it's fine. i see i forget to do that 'option mesh_id "Openwrt", i still used 'option ssid "openwrt" Jul 06 18:42:50 let me try that quick Jul 06 18:42:50 <[florian]> ah Jul 06 18:43:09 by working i mean it beacons, and i can iperf 22.1mbps between two 11s nodes using ath9k and openwrt trunk Jul 06 18:43:17 <[florian]> excellent Jul 06 18:45:08 i hacked ibss.c yesterday to bump it up to all 11n speed, but whenever a second batman-adv is up it's forced back to 11g speed, so i'm trying luck with mesh now, see if it works with >54mbps at all Jul 06 18:49:59 [florian]: can you have a look at this strange output after I set up mesh in config/wireless: http://pastebin.com/d7c6b1887 Jul 06 18:51:34 [florian]: the mesh_id option is ignored, and was set up msh0 directly. Jul 06 18:52:24 if i do iw dev wlan0 interface add mesh type mesh mesh_id mesh i can use iwconfig to find new mesh interface "mesh" though Jul 06 18:54:14 <[florian]> xxiao: mesh_id applies to msh0 Jul 06 18:55:31 <[florian]> xxiao: basically what openwrt does, is setup a msh then trigger the interface to beacon to workaround some bugs that we had before with ath5k/ath9k Jul 06 18:55:42 <[florian]> xxiao: maybe the way we setup meshing is not adequate now Jul 06 18:56:47 if i use 'mesh_id' in config, the value behind it does not matter, it will always be msh0. however without mesh_id in config, it will be in managed mode. so you need mesh_id there, and set up a random string to it Jul 06 18:57:20 if you put 'option ssid' there, with 'option mode mesh' it will still be set as ad-hoc Jul 06 18:58:00 <[florian]> mesh_id is not supposed to define the interface number, but the mesh identifier, which can be compared to the cell id in case of ad-hoc Jul 06 18:58:07 <[florian]> msh is computed automatically Jul 06 18:58:59 <[florian]> I followed this howto to integrate the 802.11s bits into OpenWrt: http://www.open80211s.org/trac/wiki/HOWTO Jul 06 19:01:20 yes that's what i used too :) Jul 06 19:01:50 anyway i just set mesh_id to a random string(though all mesh nodes will use the same mesh_id) and iperf is now 19.3mbps Jul 06 19:02:07 now looking at how to let it support 11n speed, rcstat shows only 11g speed Jul 06 19:04:06 <[florian]> what I do not fully understand is whether you had to hack OpenWrt /lib/wifi/mac80211.sh to get mesh working or not ? Jul 06 19:04:09 nbd * r16718 /trunk/package/uci/trigger/ (apply_config lib/trigger.lua): ucitrigger: add options to force enable/disable specific triggers Jul 06 19:05:12 [florian]: no i did not need to, it all works for me after the compat-wirelss 06-25 is added and the mac80211.sh is updated. before mac80211.sh added 'mesh' option, i can also use iw to get mesh working Jul 06 19:05:27 nbd * r16719 /trunk/package/ (17 files in 8 dirs): upgrade to the new version of wprobe - includes reconfigurable layer 2 statistics, remote access, more configuration options and many bugfixes Jul 06 19:05:42 <[florian]> xxiao: ok,cool then :) Jul 06 19:52:56 florian * r16720 /trunk/toolchain/uClibc/patches-0.9.30.1/400-gcc4.4-fixes.patch: [uclibc] add more gcc-4.4 patches from uclibc's mailing-list Jul 06 19:55:08 ping xMff, ping phaidros Jul 06 19:55:23 and hi [florian] :) Jul 06 19:55:54 danage: pong Jul 06 19:56:09 i missed the meeting. had an unexpected personal emergency Jul 06 19:56:51 <[florian]> danage: hey ! Jul 06 19:57:26 hi [florian] sorry for being absent but i guess no patches were missed Jul 06 19:57:35 xMff: is there a log somewhere perhaps? Jul 06 19:58:06 <[florian]> danage: indeed, I did not investigate more on the bcm6345 recently Jul 06 19:58:22 <[florian]> danage: but do not worry, it is still in the todo list, with a fairly high priority ;) Jul 06 19:58:44 yeah i just had a bad conscience for leaving without notice Jul 06 19:59:11 but good to hear no harm done in terms of bcm6345 Jul 06 19:59:15 <[florian]> no trouble Jul 06 20:00:14 danage: uhm, let me find my chatlogs Jul 06 20:06:10 florian * r16721 /packages/libs/faad2/Makefile: [package] workaround gcc-3.4.6 not handling -iquote parameter (#5446) Jul 06 20:07:45 danage: see pm Jul 06 20:08:40 thanks xMff Jul 06 20:19:56 Hi guys one question WRT54G (TM) is the same of WRT54GS V 3 hardward? Jul 06 22:45:20 jow * r16722 /branches/8.09/package/madwifi/files/lib/wifi/madwifi.sh: [8.09] madwifi: merge r15713 and r15824 Jul 07 00:07:21 nico * r16723 /trunk/package/busybox/patches/901-df_human_readable.patch: fix df -h output (closes: #5290) Jul 07 00:38:44 nbd * r16724 /trunk/package/wprobe/files/ (wprobe.config wprobe.init): update the init script and config for the new wprobe Jul 07 01:10:05 nico * r16725 /branches/packages_8.09/net/asterisk-1.4.x/ (3 files in 2 dirs): add 2 fixes backported from asterisk svn related to pickup & transfer (should close: #5451), bump release number Jul 07 02:11:14 nico * r16726 / (2 files in 2 dirs): use .$(LINUX_KMOD_SUFFIX) instead of .ko in AutoLoad (closes: #5452) Jul 07 02:22:49 nbd * r16727 /trunk/package/wprobe/src/ (exporter/wprobe-export.c user/wprobe-lib.c): fix wprobe-export Jul 07 02:35:01 nico * r16728 /trunk/ (Config.in rules.mk toolchain/Config.in): add a config option to select the binary stripping method to use (between none, strip & sstrip) Jul 07 03:00:05 nico * r16729 /trunk/package/kexec-tools/Makefile: fix kexec-tools builds when libc != uClibc, strip quotes from KEXEC_TARGET_NAME config option **** ENDING LOGGING AT Tue Jul 07 03:00:06 2009