**** BEGIN LOGGING AT Sun Dec 25 02:59:57 2011
Dec 25 05:06:38 <builder_> build #83 of rdc is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/rdc/builds/83
Dec 25 05:58:18 <builder_> build #119 of at91 is complete: Failure [failed compile_4]  Build details are at http://buildbot.openwrt.org:8010/builders/at91/builds/119
Dec 25 06:16:32 <builder_> build #84 of gemini is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/gemini/builds/84
Dec 25 06:51:06 <builder_> build #120 of s3c24xx is complete: Failure [failed compile_10]  Build details are at http://buildbot.openwrt.org:8010/builders/s3c24xx/builds/120
Dec 25 06:51:17 <builder_> build #117 of ubicom32 is complete: Failure [failed compile_3]  Build details are at http://buildbot.openwrt.org:8010/builders/ubicom32/builds/117
Dec 25 07:23:28 <builder_> build #83 of adm5120 is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/adm5120/builds/83
Dec 25 11:35:36 <nunojpg> I'm updating a package but I get a error when calling AR
Dec 25 11:35:48 <nunojpg> on the PC it is called as "ar rcv blas.a dnrm2.o daxpy.o ddot.o dscal.o"
Dec 25 11:36:11 <nunojpg> but cross-compiling the "rcv" isn't there
Dec 25 11:36:25 <nunojpg> mips-openwrt-linux-uclibc-ar blas.a dnrm2.o daxpy.o ddot.o dscal.o
Dec 25 11:36:51 <nunojpg> do I have to define the $(AR) variable somewhere?
Dec 25 12:23:08 <CIA-79> nbd * r29608 /branches/backfire/package/mac80211/Makefile: mac80211: disable b43 bcma support, fixes a compile error
Dec 25 12:40:41 <_trine> there seems to be a problem with using sysupgrade on my wndr3700 using the newest trunk
Dec 25 12:40:54 <_trine> root@OpenWrt:/tmp# sysupgrade openwrt-ar71xx-generic-wndr3700-squashfs-sysupgrade.bin
Dec 25 12:40:54 <_trine> Sysupgrade is not yet supported on unknown.
Dec 25 12:40:54 <_trine> Image check 'platform_check_image' failed.
Dec 25 12:41:38 <_trine> seems odd
Dec 25 12:57:45 <jow_laptop> __trine: juhosg reworked the flashing procedures and image tags recently, you should be able to just comment out the check image call in /sbin/sysupgrade
Dec 25 12:58:03 <jow_laptop> after you're on the new version it should work as expected again
Dec 25 12:58:53 <jow_laptop> #/topic Backfire 10.03.1 released, feedback welcome :) | Please contribute to the new OpenWrt wiki: http://wiki.openwrt.org
Dec 25 12:58:56 <jow_laptop> oops
Dec 25 13:00:33 <__trine> ok let me try
Dec 25 13:10:08 <jow_laptop> finally
Dec 25 13:10:10 <_trine> I have flashed the new code now
Dec 25 13:10:14 <jow_laptop> weak irc-fu
Dec 25 13:10:40 <_trine> thanks for the info Jow
Dec 25 13:28:41 <Weedy> jow_laptop: any idea how to query the status or state of a b43 card?
Dec 25 13:29:13 <Weedy> my wifi seems to lock up now and then, i would like to know when to run wifi instead of just croning it
Dec 25 13:29:54 <jow_laptop> Weedy: this depends on how it locked up. I suppose the on-chip firmware hangs or something, so you should see some dmesg entries
Dec 25 13:30:35 <jow_laptop> tbh I don't know how I would aproach that, one would need to debug such a hang yet
Dec 25 13:31:03 <jow_laptop> for example testing whether restarting hostapd alone is enough, or whether deleting and readding the vap kick the driver back into operation
Dec 25 13:31:26 <jow_laptop> s/yet/first/
Dec 25 13:39:16 <CIA-79> jow * r29609 /trunk/include/netfilter.mk: [netfilter] package CT target
Dec 25 13:39:41 <CIA-79> jow * r29610 /trunk/target/linux/generic/ (9 files): [generic] add missing config symbol
Dec 25 13:42:35 <Weedy> jow_laptop: so uhh, how do i restart hostapd?
Dec 25 13:44:14 <jow_laptop> killall -9 hostapd; hostapd -P /var/run/wifi-phy0.pid -B /var/run/hostapd-phy0.conf  ?
Dec 25 13:47:25 <Weedy> point
Dec 25 14:04:56 <Weedy> jow_laptop: yeah that works
Dec 25 14:05:13 <Weedy> so .mon or hostap is locking up
Dec 25 14:05:39 <CIA-79> jow * r29611 /trunk/package/usbreset/ (. Makefile src/ src/usbreset.c): [package] add usbreset - a small simple utility to send port rests to selected usb devices (#10394)
Dec 25 14:06:48 <jow_laptop> Weedy: well hostapd performs certain initializations upon startup, so maybe the driver indeed hangs
Dec 25 14:32:40 <written> jow_laptop, ping
Dec 25 14:48:51 <danitool> the driver ath9k cannot find any valid MAC address in the eeprom
Dec 25 14:49:42 <danitool> as a result of this it generates a random MAC, and openwrt uses that mac in /etc/config/wireless in radio0
Dec 25 14:51:03 <danitool> but when i restart the board, it generates again another random MAC address, and openwrt creates another radio1 with radio0 using the new MAC
Dec 25 14:51:30 <russell--> cool
Dec 25 14:51:45 <danitool> and everytime I restart the board, i'm getting new radios with random MAC adresses...
Dec 25 14:51:50 <russell--> what hardware is that?
Dec 25 14:51:59 <danitool> bcm6358
Dec 25 14:52:29 <danitool> but, spite of my board support isn't it a bug into the ath9 drivers itself?
Dec 25 14:54:30 <danitool> I think the drivers should use the previous generated random MAC...
Dec 25 14:54:52 <danitool> or the one at /etc/config/wireless..
Dec 25 14:55:14 <russell--> better yet, find a valid one in eeprom?
Dec 25 14:55:27 <danitool> yes, it would be better
Dec 25 14:55:41 <danitool> but it does find none
Dec 25 14:56:10 <russell--> do you know what it is supposed to be?  like written on the case?
Dec 25 14:57:09 <danitool> well, I can install the OEM firmware, and get from there, that's not a problem
Dec 25 14:57:41 <danitool> but once I get the MAC, what do I do with it?
Dec 25 14:57:42 <danitool> :P
Dec 25 14:57:47 <russell--> on the wgt, the macaddrs are stored on flash, and the bootloader or something snags it.  maybe you can grep and find where it's stored.
Dec 25 14:58:12 <danitool> maybe
Dec 25 14:59:03 <danitool> but then how do I pass that mac to the ath9k driver?
Dec 25 14:59:16 <russell--> read the source? ;-)
Dec 25 14:59:20 <danitool> the eeprom is stored into the flash
Dec 25 14:59:40 <danitool> yes... reading the source.. and writting into it :P
Dec 25 15:00:12 <danitool> but don't have enough skills...
Dec 25 15:00:27 * russell-- doesn't know any of the details, but presumes it works in a particular way, your task is just to figure out what that is and to arrange things so it works. ;-)
Dec 25 15:01:14 <danitool> sure, that assuming I have a valid mac anywhere
Dec 25 15:01:24 <danitool> somewhere*
Dec 25 15:01:45 <russell--> nothing printed on the case?
Dec 25 15:02:59 <danitool> yeah, the mac of the eth0 :D
Dec 25 15:03:16 <danitool> but I presume this is not the mac of the wlan0...
Dec 25 15:04:17 <russell--> this is an off-the-shelf device?  what is the brand/model?
Dec 25 15:04:44 <danitool> btw, I solved the problem changing random mac generation with a fixed mac in the code
Dec 25 15:05:38 <danitool> it's not yet supported by openwrt
Dec 25 15:05:51 <danitool> it's a huawei hg556a
Dec 25 15:06:04 <russell--> cool
Dec 25 15:06:09 <danitool> bcm63xx
Dec 25 15:07:23 <russell--> "The nvram data embedded in the bootloader and the WLAN calibration data in ath_data are specific for your device and difficult to restore otherwise."
Dec 25 15:08:49 <danitool> maybe, I compared the the ath_data with another one, a person who owns a lantiq router, and it is exactly the same..
Dec 25 15:09:46 <danitool> I backed up the entire flash before start installing openwrt in the device
Dec 25 15:10:03 <danitool> I think, i didn't miss anything
Dec 25 15:10:25 <danitool> the wireless works ok, I only have this little issue with the mac
Dec 25 15:10:37 <russell--> do you have a serial console on it?
Dec 25 15:10:42 <danitool> yes
Dec 25 15:10:44 <danitool> and jtag
Dec 25 15:11:00 <russell--> does the bootloader tell you anything?
Dec 25 15:11:33 <danitool> no
Dec 25 15:11:46 <danitool> just reports the eth0 mac address
Dec 25 15:13:46 <danitool> if the mac is into the ath_data partition I don't know where is it, is there any informationa about the structure of the ath9k eeproms?
Dec 25 15:13:53 <russell--> my wgt634u's (which are admittedly weird) can set the macaddrs like so:
Dec 25 15:13:56 <russell--> CFE> setenv -p et0macaddr 00-0f-b5-3d-58-b6
Dec 25 15:13:57 <russell--> CFE> setenv -p et1macaddr 00-0f-b5-3d-58-b7
Dec 25 15:14:26 <russell--> one of those is the wireless macaddr iirc
Dec 25 15:14:53 <russell--> printenv might show you something similar
Dec 25 15:15:46 <danitool> invalid command
Dec 25 15:16:13 <russell--> there is probably a "help" that will show you what commands are valid
Dec 25 15:16:20 <russell--> getenv maybe?
Dec 25 15:16:30 <russell--> show or something
Dec 25 15:17:31 <danitool> p
Dec 25 15:17:46 <danitool> "print boot line and board parameters"
Dec 25 15:18:02 <russell--> anything interesting?
Dec 25 15:18:07 <danitool> no
Dec 25 15:18:14 <danitool> only the eth0 address
Dec 25 15:18:25 <russell--> any other choices out of help listing?
Dec 25 15:18:47 <danitool> no, it has not very much options..
Dec 25 15:19:12 <russell--> what does the eth0 address varaible look like?
Dec 25 15:19:52 <russell--> i mean, the name, not the value
Dec 25 15:20:11 <danitool> Base MAC address
Dec 25 15:20:33 <russell--> maybe they just add one to get the wlan macaddr
Dec 25 15:20:43 <danitool> maybe
Dec 25 15:21:03 <russell--> maybe reflash the stock f/w and see what they use
Dec 25 15:21:35 <danitool> I can initialize the board assigning "Base MAC addres + 1" to the wifi
Dec 25 15:22:34 <russell--> "base" does sound like they are using it for multiple addresses
Dec 25 15:23:02 <russell--> "base" and "offset" being commonly used together ;-)
Dec 25 15:23:08 <danitool> but it won't change the fact of assigning new radio if I change the mac
Dec 25 15:24:52 <russell--> oh, i think you can specify phy phy0 or something instead of the mac
Dec 25 15:25:10 <russell--> i think i did that once upon a time
Dec 25 15:26:54 <russell-->         option phy      phy0
Dec 25 15:26:57 <danitool> well, btw that won't explain why the driver ath9k assing new random mac address on every boot, or why it can't find any valid mac address in the rom
Dec 25 15:27:19 <russell--> it does say "work-in-progress" ;-)
Dec 25 15:27:26 <danitool> :D
Dec 25 15:27:54 <danitool> work-in-progress means "it will be ready in 2020" :P
Dec 25 15:28:16 <russell--> it will be ready faster if you fix it! :-)
Dec 25 15:28:28 <danitool> of course
Dec 25 15:28:55 <russell--> but i'm just wasting your time, since i don't know the answer or the right way to do it
Dec 25 15:29:10 <danitool> if I had all the information about the structure of ath9k eeprom from the atheros people :P
Dec 25 15:29:53 <danitool> well, all info is always wellcome
Dec 25 15:30:16 * russell-- is just applying the general problem solving algorithm
Dec 25 15:30:17 <danitool> I did take the note about "         option phy      phy0"
Dec 25 15:30:31 <russell--> yeah, that's handy
Dec 25 15:30:48 <russell--> remove or comment out the macaddr line
Dec 25 15:31:28 <russell--> in the config wifi-device  radio0 line
Dec 25 15:31:33 <russell--> or section, rather
Dec 25 15:33:24 <danitool> ok, I'll install the OEM firmware again to copy the MAC address, and see if it exists somewhere in my flash backup
Dec 25 15:33:55 <russell--> i'm guessing it'll be the eth0 + 1
Dec 25 15:34:30 <danitool> i guess it won't share any hex value :P
Dec 25 15:34:46 <russell--> that their init scripts are finding the base macaddr and configuring it with that plus one
Dec 25 15:36:22 * russell-- assumes they are running linux too?  maybe you can login on the serial console and see how they boot up.
Dec 25 15:37:07 <danitool> yes, they use a linux based system
Dec 25 15:37:26 <danitool> and I can access through telnet to the busybox too
Dec 25 15:37:59 <danitool> sptie of this they didn't publish the source code
Dec 25 15:40:06 <russell--> 64 meg of ram, nice!
Dec 25 15:41:14 <danitool> and 3 usb ports :P
Dec 25 15:41:46 <russell--> how much was it?
Dec 25 15:42:23 <danitool> how much money?
Dec 25 15:42:57 <danitool> well, 0, since it was a donation
Dec 25 15:43:14 <russell--> always the best price!
Dec 25 15:43:41 <danitool> btw, you can't buy it, it's an ISP hardware
Dec 25 15:43:48 <russell--> looks like it was a vodafone cpe, yeat
Dec 25 15:43:51 <russell--> yeah*
Dec 25 15:44:28 <danitool> perhaps at ebay it's about 20
Dec 25 15:45:04 <russell--> also a dsl modem
Dec 25 15:45:49 <russell--> sexy
Dec 25 15:48:24 <russell--> http://www.ebay.com/itm/Huawei-HG556a-ADSL-Modem-3G-Wireless-Router-Pri-nt-/170563007993?pt=COMP_EN_Routers&hash=item27b658f5f9
Dec 25 15:50:35 <russell--> "Item will be marked as "Gift" and declared low value to minimize the custom tax." LOL
Dec 25 15:51:46 <russell--> "also, the dessicant packs are really crack cocaine!"
Dec 25 16:05:31 <danitool> ok, russell-- you won:  ath0     Link encap:Ethernet  HWaddr 5C:4C:A9:6E:4A:A3
Dec 25 16:05:44 <danitool> eth0            Link encap:Ethernet  HWaddr 5C:4C:A9:6E:4A:A2
Dec 25 17:53:23 <russell--> danitool: /me rocks!
Dec 25 17:53:33 <danitool> sure
Dec 25 17:53:36 <russell--> lol
Dec 25 17:56:24 <danitool> ok, now I can assign a MAC to the wifi based on the mac address, still generating a random MAC address, this is harmless and has no sense, but the wifi works ok
Dec 25 20:18:36 <builder_> build #98 of sibyte is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/98
Dec 25 20:51:56 <builder_> build #99 of ar7 is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/ar7/builds/99
Dec 25 21:03:04 <swalker> updated openwrt/upstream, https://home.comcast.net/~sdwalker/uscan/index.html
Dec 25 21:08:06 <blogic> swalker: ping
Dec 25 21:10:38 <swalker> blogic: pong
Dec 25 21:10:58 <blogic> how about i put a file online with current latest driver release you can read ?
Dec 25 21:11:34 <blogic> and then ltq-* can get a different status than indereminable
Dec 25 21:11:40 <blogic> i think i even asked tis before
Dec 25 21:12:13 <swalker> I think so iirc
Dec 25 21:12:16 <blogic> hehe
Dec 25 21:12:34 <blogic> i am getting old
Dec 25 21:14:57 <swalker> <blogic> i could offer a tct file on my server with the latest versions that you can harvest ?! <swalker> I don't think uscan has the ability to parse a txt file, only http/ftp directory listings and html pages looking for "<a href=" tags <swalker> I could point the packages to look at mirror*.openwrt.org but I don't generally classify that as an upstream
Dec 25 21:21:15 <blogic> swalker: http://dev.phrozen.org/ltq-mirror/
Dec 25 21:21:28 <blogic> could you add ltq-dsl-fw-0.1.tar.bz2 to the list please
Dec 25 21:52:59 <swalker> blogic: ltq-* added; I haven't tried adding sub-packages yet
Dec 25 21:57:59 <blogic> ok
Dec 25 21:58:18 <blogic> thx !
**** ENDING LOGGING AT Mon Dec 26 02:59:56 2011