**** BEGIN LOGGING AT Thu Jan 10 02:59:59 2013
Jan 10 06:16:40 <russell--> confirmed that line is the blocker (commenting out the call, it builds to completion) ... i guess i figure out how to convert from kernel_thread() to kthread_* now.
Jan 10 06:46:02 <OpenWrt-Commits> blogic r35077 trunk/target/linux/lantiq/image/ARV7518PW.dts * [lantiq] fixes USB reset on ARV7518
Jan 10 07:17:57 <builder> build #135 of rb532 is complete: Failure [failed shell_11]  Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/135
Jan 10 07:20:32 <builder> build #135 of ppc44x is complete: Failure [failed shell_11]  Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/135
Jan 10 07:36:34 <builder> build #125 of sibyte is complete: Failure [failed shell_11]  Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/125
Jan 10 08:00:02 <builder> build #152 of brcm63xx is complete: Failure [failed shell_14]  Build details are at http://buildbot.openwrt.org:8010/builders/brcm63xx/builds/152
Jan 10 09:29:02 <builder> build #147 of orion is complete: Failure [failed compile_8]  Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/147
Jan 10 11:49:58 <sungami> somewhere between r35062 and r35069 wireless got broken on wndr3700. The interfaces seem to come up, but the logs show entries like "IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready". Is this a known problem?
Jan 10 12:29:08 <blogic> dwmw2_gone: ping
Jan 10 13:13:27 <builder> build #140 of lantiq is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/lantiq/builds/140
Jan 10 13:24:36 <OpenWrt-Commits> kaloz r35078 trunk/target/linux/cns3xxx/files/arch/arm/mach-cns3xxx/gpio.c * [cns3xxx]: clean up PCI bus topology
Jan 10 13:26:15 <OpenWrt-Commits> kaloz r35079 trunk/target/linux/cns3xxx/patches-3.3/066-pcie_bus_topology.patch
Jan 10 13:26:15 <OpenWrt-Commits> [cns3xxx]: clean up PCI bus topology
Jan 10 13:26:15 <OpenWrt-Commits>  
Jan 10 13:26:15 <OpenWrt-Commits> This makes the PCI bus topology more standard for devices behind a bridge
Jan 10 13:26:15 <OpenWrt-Commits>  
Jan 10 13:26:17 <OpenWrt-Commits> Signed-off-by: Tim Harvey <tharvey@gateworks.com>
Jan 10 14:06:21 <OpenWrt-Commits> hauke r35080 trunk/target/linux/ (38 files in 6 dirs) * kernel: update bcma and ssb to master-2013-01-09 from wireless-testing
Jan 10 14:10:53 <OpenWrt-Commits> nbd r35081 trunk/package/mac80211/patches/300-pending_work.patch * ath9k: add some calibration fixes from wireless-testing
Jan 10 14:11:44 <OpenWrt-Commits> nbd r35082 branches/attitude_adjustment/package/mac80211/patches/300-pending_work.patch * ath9k: add some calibration fixes from wireless-testing
Jan 10 14:23:22 <OpenWrt-Commits> hauke r35083 trunk/target/linux/brcm47xx/patches-3.6/750-bgmac.patch * brcm47xx: update bgmac driver to the version in mainline kernel
Jan 10 15:47:42 <OpenWrt-Commits> obsy r35084 packages/net/transmission/Makefile * [packages] transmission: update to 2.76
Jan 10 16:20:25 <OpenWrt-Commits> acinonyx r35085 packages/ utils/rtl-sdr/Makefile utils/rtl-sdr * [packages] rtl-sdr: New package, software defined radio with RTL2832U dongle
Jan 10 17:19:20 <Pteridium> When compiling bcm63xx targets i add -mtune=4kc, but i don't know if there are better flags to pass to gcc in order to optimize the images.
Jan 10 17:19:54 <[florian]> Pteridium: I don't think brcm63xx targets need this
Jan 10 17:20:00 <[florian]> using -mips32 should be enough already
Jan 10 17:20:11 <[florian]> and technically they are not using a real mips 4kc compatible cpu
Jan 10 17:20:37 <[florian]> I don't know what could be the side effects on instruction ordering/pipeling compared to just using -mips32
Jan 10 17:21:28 <OpenWrt-Commits> florian r35086 trunk/target/linux/ (8 files in 3 dirs) * mvebu: update to 3.8-rc3
Jan 10 17:21:32 <OpenWrt-Commits> florian r35087 trunk/include/netfilter.mk * netfilter: xt_NOTRACK is incorporated in xt_CT as of 3.8-rc3
Jan 10 17:23:04 <Pteridium> Then, compiling with different cflags and benchmark will be the best option.
Jan 10 17:23:40 <[florian]> probably yes
Jan 10 17:24:27 <[florian]> brcm63xx uses bmips cores, bmips3300 (6345), bmips32 (6338, 6348), bmips4350 (bcm6358, bcm6368)
Jan 10 17:27:06 <Pteridium> A lot of thanks Florian. Now understand why, with openssl benchmark the bcm6328 outperforms by a little margin the Lantiq Danube ( 24KEc an dsp extensions).
Jan 10 17:27:31 <[florian]> you are welcome
Jan 10 17:34:25 <Pteridium> This is not exactly my right place. I'm electronic technician, a linux newbie and without programming skills, but, since the BCM63xx target appeared i've learned a lot thanks to your wonderful job.
Jan 10 17:35:37 <Pteridium> In addition, always try to optimize my compilations as far as their can be.
Jan 10 17:38:23 <Pteridium> For example, with Lantiq Danube i use -mtune=24kec and -mdsp, and observed some improvements.
Jan 10 17:40:24 <stintel> cyrusff: hi, jow_laptop told me I should talk to you about ipv6-support, 6distributed etc
Jan 10 17:40:39 <cyrusff> stintel: go ahead
Jan 10 17:40:47 <stintel> cyrusff: most important one, is have a switch that makes it log all its actions
Jan 10 17:41:09 <stintel> like "I am now announcing a:b:c:d::/64 to br-lan", for example
Jan 10 17:41:24 <cyrusff> well you can grep syslog for 6distributed
Jan 10 17:41:34 <stintel> but openwrt doesnt syslog ?
Jan 10 17:41:39 <stintel> I didnt find in logread ?
Jan 10 17:41:50 <cyrusff> it should be in logread
Jan 10 17:41:54 <stintel> hmmm
Jan 10 17:42:08 <jow_laptop> cyrusff: I don't have any logread entries either
Jan 10 17:42:21 <stintel> maybe because it isn't announcing anything ?
Jan 10 17:42:27 <stintel> Jan 10 18:29:20 wrt0 daemon.notice 6distributed: delprefix for wanv6
Jan 10 17:42:31 <stintel> Jan 10 18:29:21 wrt0 daemon.notice 6distributed: Invoked deliface for wanv6 with desired prefix length 0
Jan 10 17:42:51 <cyrusff> jow_laptop: sure?
Jan 10 17:42:53 <stintel> there is something. could it include the prefix that the action is about ?
Jan 10 17:43:08 <cyrusff> well delprefix is done when the interface is deconfigured
Jan 10 17:43:23 <cyrusff> if there are prefixes they are displayed there
Jan 10 17:43:30 <stintel> aha.
Jan 10 17:43:40 <stintel> yes, now it makes sense. I disabled announcing prefixes
Jan 10 17:43:47 <jow_laptop> stintel: you can also use "ubus call 6distributed status" to see which prefixes it currently manages
Jan 10 17:43:51 <stintel> because bird6 does that for me. so there are no prefxies in the log now :)
Jan 10 17:44:05 <stintel> ok, that is interesting info, the ubus call
Jan 10 17:44:20 <stintel> ok, sorry for the noise guys.
Jan 10 17:44:36 <cyrusff> stintel: however be aware that ipv6-support and 6distributed will go away in the near future
Jan 10 17:44:41 <jow_laptop> stintel: http://pastebin.com/B9CYcJtt
Jan 10 17:44:57 <cyrusff> i am currently in the process of rewriting all this stuff
Jan 10 17:45:01 <stintel> jow_laptop: yeah, for me it shows the prefix I manuallly specified
Jan 10 17:45:08 <cyrusff> the prefix-stuff will go into netifd directly
Jan 10 17:45:12 <stintel> cyrusff: cool.
Jan 10 17:45:23 <jow_laptop> bbl
Jan 10 17:45:23 <cyrusff> so you can query it there
Jan 10 17:45:25 <stintel> this is actually pretty cool
Jan 10 17:45:50 <cyrusff> things will be less clutchy and more clean then
Jan 10 17:46:18 <stintel> does/will it support grabbing LAN prefixes from DHCP PD ?
Jan 10 17:46:55 <stintel> I think this is the way my ISP will give IPv6 addresses to the LAN
Jan 10 17:47:10 <cyrusff> yes but this is also supported by the current setup
Jan 10 17:47:12 <stintel> but I have a modem, I didn't want their managed router. my router is the openwrt device
Jan 10 17:47:28 <stintel> this is good news, thanks for the info
Jan 10 17:47:46 <cyrusff> sure you're welcome
Jan 10 17:47:48 <stintel> since it's 2013, trials for native v6 should start soon with my isp
Jan 10 17:47:55 <cyrusff> yeah
Jan 10 17:48:25 <cyrusff> we have a new small dhcpv6 client for that task
Jan 10 17:48:41 <cyrusff> its already in the repo
Jan 10 17:48:46 <stintel> ic. does it handle only the ipv6 address?
Jan 10 17:48:51 <stintel> or does it do both v4 and v6?
Jan 10 17:49:06 <cyrusff> well it does the v6 stuff + PD
Jan 10 17:49:09 <cyrusff> v4 is extra
Jan 10 17:49:27 <stintel> so I'll just have a dhcpv4 and dhcpv6 client running then, for dualstack ?
Jan 10 17:49:32 <cyrusff> yes
Jan 10 17:49:44 <stintel> ok. fair enough
Jan 10 17:49:44 <cyrusff> or pppd for IPv4 and the dhcpv6 client for IPv6
Jan 10 17:49:51 <cyrusff> its a bit different
Jan 10 17:50:11 <cyrusff> for things like ppp you didn't need dhcp before but for dhcpv6 you need it to get prefixes etc.
Jan 10 17:50:15 <stintel> yeah it depends on the wan connection indeed
Jan 10 17:51:47 <stintel> I smell a few new openwrt routers @ customers when we get native IPv6 in Belgium :)
Jan 10 17:52:06 <cyrusff> hehe
Jan 10 18:03:50 <OpenWrt-Commits> nbd r35088 branches/attitude_adjustment/ toolchain/uClibc/patches-0.9.33.2/010-backport_sscanf_alloc.patch toolchain/uClibc/patches-0.9.33.2/011-dlsym_rtld_next_fix.patch * uClibc: backport useful fixes from trunk (#12777)
Jan 10 18:25:35 <DonkeyHotei> luka, Delboy: help! i had an image booting, but it was overlaying tmpfs instead of rootfs_data, and then i used sysupgrade, and now i have a brick :(
Jan 10 18:47:39 <OpenWrt-Commits> nbd r35089 trunk/package/mac80211/patches/300-pending_work.patch
Jan 10 18:47:39 <OpenWrt-Commits> ath9k: add updated tx gain tables, fixes stability issues on TL-WDR4300 and other devices
Jan 10 18:48:07 <OpenWrt-Commits> nbd r35090 branches/attitude_adjustment/package/ mac80211/patches/300-pending_work.patch mac80211/patches/561-ath9k_revert_initval_change.patch
Jan 10 18:48:07 <OpenWrt-Commits> ath9k: add updated tx gain tables, fixes stability issues on TL-WDR4300 and other devices
Jan 10 19:23:42 <Pteridium> blogic: ping
Jan 10 19:25:59 <blogic> hi
Jan 10 19:34:10 <DonkeyHotei> blogic: is jtag at all possible on lantiq boards?
Jan 10 21:14:14 <Pteridium> Wich are the best GCC otimizations for the MIPS 24KEc core?
Jan 10 22:27:48 <DonkeyHotei> luka, Delboy: i repeat my request for help
Jan 10 22:44:29 <luka> DonkeyHotei: you need to have working serial, you wont get nowhere without it
Jan 10 22:44:50 <DonkeyHotei> luka: i did have it working, but one-way only
Jan 10 22:45:09 <DonkeyHotei> i used the UART boot mode to debrick it last time
Jan 10 22:45:10 <luka> that is not a working serial then ;)
Jan 10 22:45:28 <DonkeyHotei> but now, even the lan led's don't light up no matter what i do
Jan 10 22:45:45 <DonkeyHotei> the Tx pin never had the 3.3V
Jan 10 22:45:56 <DonkeyHotei> it must have been a manufacturing error'
Jan 10 22:46:26 <DonkeyHotei> did you ever have a board where the lan lights no longer went on?
Jan 10 22:46:28 <DonkeyHotei> luka: ?
Jan 10 22:46:39 <__trine> TX pins dont always have 3.3v it depends on what state its in
Jan 10 22:47:15 <__trine> and you need an oscilloscope to see what it's doing
Jan 10 22:47:40 <DonkeyHotei> __trine: it did pretty much nothing serial, from the very beginning
Jan 10 22:47:53 <__trine> find a nice friendly radio ham who has a scope
Jan 10 22:48:02 <__trine> so you can check it
Jan 10 22:48:24 <DonkeyHotei> i think the Tx pin from the SoC somehow never got connected well to the board. i've seen this on one other board before
Jan 10 22:48:52 <DonkeyHotei> but i don't think it even gets to serial now
Jan 10 22:49:09 <__trine> well thinking is not good enough you need to be able to properly measure it
Jan 10 22:49:15 <DonkeyHotei> it seems to be stuck in some alternate boot selection mode
Jan 10 22:49:23 <DonkeyHotei> luka: ^
Jan 10 22:50:12 <DonkeyHotei> when i used the UART boot mode before, the lan lights did not go on until after the .asc file was received
Jan 10 22:50:22 <DonkeyHotei> now they don't go on, period
Jan 10 22:51:08 <luka> DonkeyHotei: fix your serial or get another board
Jan 10 22:52:03 <DonkeyHotei> i don't think the serial was ever fixable; i think the serial issue is on a lower layer of the board to begin with
Jan 10 22:52:41 <DonkeyHotei> is there anything more useful i can do with this board now than send it to be recycled?
Jan 10 22:52:51 <DonkeyHotei> luka: ?
Jan 10 22:52:54 <__trine> set fire to it
Jan 10 22:53:06 <__trine> and know you have won over it :)
Jan 10 22:54:24 <__trine> can you not find where else on the board the serial lines are terminated
Jan 10 22:56:33 <DonkeyHotei> __trine: only luka or Delboy might have that info
Jan 10 22:56:44 <DonkeyHotei> and setting fire is not winning
Jan 10 22:57:19 <__trine> DonkeyHotei, do you not have a schematic of the circuit
Jan 10 22:57:40 <__trine> or does Delboy have one
Jan 10 22:58:15 <DonkeyHotei> i have a friend porting her number from POTS to SIP who lives nearly 100 miles away, and i have no car. i'm going there early tomorrow morning, the initial idea being to deploy the sx762 in place of the fon2100 running backfire
Jan 10 22:59:05 <DonkeyHotei> the FCC website has a letter from siemens requesting they keep the schematics confidential and not accessible to the public
Jan 10 23:15:49 <__trine> DonkeyHotei, were you intending putting openwrt on this box
Jan 10 23:16:05 <DonkeyHotei> yes
Jan 10 23:16:25 <__trine> and why would it not be ok with the original firmware for the purpose you were intending to use it for
Jan 10 23:16:29 <DonkeyHotei> otherwise, it was locked to a provider that does not serve this area
Jan 10 23:16:37 <__trine> I see
Jan 10 23:16:59 <__trine> but they are not all locked are they
Jan 10 23:17:07 <__trine> did you get it cheap
Jan 10 23:17:12 <DonkeyHotei> the voip part was locked
Jan 10 23:17:28 <DonkeyHotei> plus, i doubt the stock fw was ipv6-aware
Jan 10 23:18:01 <DonkeyHotei> also, i wanted to add print-server capability to the usb port, which i think was only for storage
Jan 10 23:18:02 <__trine> http://www.google.co.uk/url?sa=t&rct=j&q=sx762&source=web&cd=1&sqi=2&ved=0CDsQFjAA&url=http%3A%2F%2Fwww.voipon.co.uk%2Fdocuments%2Fsiemens_sx762_datasheet.pdf&ei=TUvvUNrnM8eY0QWesICQDw&usg=AFQjCNFuXnuKQWcMCLeB97iA3rGyQSbxPA&bvm=bv.1357700187,d.d2k&cad=rja
Jan 10 23:19:43 <DonkeyHotei> i got it as cheap as i could find it. i would have preferred something that didn't have just 8M flash, because i also wanted to provide wireless scanning, but sane-libs alone is 3MB, and i was not able to find any other lantiq voip devices anywhere, period
Jan 10 23:20:34 <__trine> I just use my dockstar and an SPA 3000 which does ok for me
Jan 10 23:22:32 <DonkeyHotei> i was hoping for an all-in-one because it would already be an unsightly mess attached to the docsis device providing the uplink
Jan 10 23:25:39 <DonkeyHotei> jtag appears to be disabled on these boards, too
Jan 10 23:26:40 <__trine> well I would guess one thing is for certain you wont get any help with it unless you are able to provide some measurable information
Jan 10 23:26:56 <DonkeyHotei> measurable in what way?
Jan 10 23:27:32 <__trine> so you need to figure out how you are going to achieve that such as getting the serial working so you can see what is happening
Jan 10 23:27:52 <__trine> otherwise its just one big guessing game
Jan 10 23:28:08 <DonkeyHotei> i gave up on the serial because it was clear there was simply no access to Tx
Jan 10 23:28:29 <__trine> then its probably scrap
Jan 10 23:28:33 <DonkeyHotei> but working Rx was pretty useful in guessing the first few times i bricked it
Jan 10 23:35:50 <__trine> any cheap old scope would do for this kind of think and they would be very cheap to buy on ebay
Jan 10 23:36:00 <__trine> /s/think/thing
Jan 10 23:38:16 <DonkeyHotei> i don't get what you'd hope to accomplish with the scope. we already know the supposed Tx did not work, and if there were another access point, luka or Delboy would have found it, having gone through 50 of these boards
Jan 10 23:40:00 <__trine> http://www.ebay.co.uk/itm/Scopex-4D-10-Oscilloscope-/170969150859?pt=UK_BOI_Electrical_Test_Measurement_Equipment_ET&hash=item27ce8e358b
Jan 10 23:40:09 <__trine> there you go DonkeyHotei
Jan 10 23:40:54 <DonkeyHotei> that's beside the point
Jan 10 23:41:42 <__trine> just look for a very cheap scope so you can have a poke around
Jan 10 23:43:11 <DonkeyHotei> there are just too many possible points to try, and now that the board is stuck in some alternate boot mode, i'm not sure there even is any serial
Jan 10 23:43:46 <__trine> but you would be able to know for sure if the TX was working or not if you had a cheapo scope
Jan 10 23:45:17 <DonkeyHotei> not if i can't find it
Jan 10 23:45:36 <DonkeyHotei> the Tx that was supposed to work DEFINITELY didn't
Jan 10 23:48:06 <OpenWrt-Commits> heil r35091 packages/net/ (25 files in 2 dirs)
Jan 10 23:48:06 <OpenWrt-Commits> package: haproxy
Jan 10 23:48:06 <OpenWrt-Commits>  - add missing patches
Jan 10 23:48:06 <OpenWrt-Commits>  - add patch number to version
Jan 10 23:50:55 <OpenWrt-Commits> heil r35092 packages/net/ haproxy/patches/0023-BUG-MINOR-http-don-t-process-abortonclose-when-1.4.22.diff haproxy/Makefile
Jan 10 23:50:56 <OpenWrt-Commits> package: haproxy
Jan 10 23:50:56 <OpenWrt-Commits>  - refresh patches
Jan 10 23:51:07 <__trine> DonkeyHotei, http://www.picotech.com/picoscope-oscilloscope-software.html
Jan 10 23:51:28 <__trine> there is a free limited version
Jan 10 23:53:13 <DonkeyHotei> requires windows and their weird little usb dongle
Jan 10 23:53:28 <__trine> http://www.ledametrix.com/oscope/index.html
Jan 10 23:54:39 <__trine> and this one http://zelscope.com/
Jan 10 23:55:01 <__trine> anything like that would probably do for what you want
Jan 11 01:53:02 <builder> build #153 of brcm63xx is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/brcm63xx/builds/153
Jan 11 02:19:06 <russell--> buffalo wzr-600dhp arrived today
Jan 11 02:50:32 <builder> build #129 of avr32 is complete: Failure [failed compile_9]  Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/129
**** ENDING LOGGING AT Fri Jan 11 02:59:58 2013