**** BEGIN LOGGING AT Wed May 11 02:59:58 2011
May 11 03:28:40 <builder_> build #33 of brcm63xx is complete: Exception [exception interrupted]  Build details are at http://buildbot.openwrt.org:8010/builders/brcm63xx/builds/33
May 11 03:28:40 <builder_> build #31 of cobalt is complete: Exception [exception interrupted]  Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/31
May 11 03:28:42 <builder_> build #30 of lantiq is complete: Exception [exception interrupted]  Build details are at http://buildbot.openwrt.org:8010/builders/lantiq/builds/30
May 11 03:28:44 <builder_> build #30 of ps3 is complete: Exception [exception interrupted]  Build details are at http://buildbot.openwrt.org:8010/builders/ps3/builds/30
May 11 04:11:30 <builder_> build #30 of pxcab is complete: Failure [failed compile_3]  Build details are at http://buildbot.openwrt.org:8010/builders/pxcab/builds/30
May 11 08:13:37 <CIA-83> florian * r26871 /packages/net/faifa/Makefile: [package] update faifa to r82
May 11 11:15:56 <KanjiMonster> nbd: ping
May 11 12:17:25 <KanjiMonster> [florian]: ping
May 11 12:23:04 <[florian]> KanjiMonster: pong
May 11 12:25:39 <KanjiMonster> [florian]: have you had time to look at my dsl-274x and speedport 303v patches?
May 11 12:26:10 <[florian]> KanjiMonster: not yet sorry
May 11 12:26:13 <[florian]> KanjiMonster: I will do this now
May 11 12:26:57 <KanjiMonster> thanks :)
May 11 12:27:16 <[florian]> spw303v looks good to me
May 11 12:27:19 <[florian]> I will commit it
May 11 12:27:57 <KanjiMonster> yay \o/
May 11 12:28:49 <[florian]> looks equally good
May 11 12:30:08 <KanjiMonster> the dsl-2741b is my current bcm63xx dev device ;)
May 11 12:50:37 <[florian]> KanjiMonster: I am rebuilding my toolchain atm, just to make sure it works
May 11 12:50:59 <KanjiMonster> [florian]: sure, take your time :)
May 11 12:52:33 <[florian]> KanjiMonster: that's what I do all the time ;)
May 11 13:35:41 <CIA-83> florian * r26872 /trunk/target/linux/generic/config-2.6.38: [kernel] add missing 2.6.38 touchscreen config symbols
May 11 13:35:45 <CIA-83> florian * r26873 /trunk/target/linux/ep93xx/ (7 files in 2 dirs): [ep93xx] switch to 2.6.38
May 11 13:35:53 <CIA-83> florian * r26874 /trunk/target/linux/brcm63xx/image/Makefile:
May 11 13:35:53 <CIA-83> bcm63xx: Decouple Board ID and image name for fixup
May 11 13:35:53 <CIA-83> Make the fixup board ID independent from the image name to allow using
May 11 13:35:53 <CIA-83> the same board ID for similar devices.
May 11 13:35:54 <CIA-83> Signed-off-by: Jonas Gorski <jonas.gorski+openwrt@gmail.com>
May 11 13:36:01 <CIA-83> florian * r26875 /trunk/target/linux/brcm63xx/base-files/etc/diag.sh:
May 11 13:36:01 <CIA-83> bcm63xx: Accept any led with power in its name as power led
May 11 13:36:01 <CIA-83> Signed-off-by: Jonas Gorski <jonas.gorski+openwrt@gmail.com>
May 11 13:36:16 <CIA-83> florian * r26876 /trunk/target/linux/brcm63xx/ (7 files in 5 dirs): (log message trimmed)
May 11 13:36:16 <CIA-83> bcm63xx: Update support for the DSL-274xB revision C
May 11 13:36:16 <CIA-83> Rename the DSL-274xB rev C2's board ID to "AW4139" to match the one used
May 11 13:36:16 <CIA-83> for the rev C3 (same hardware) and add a fixup for C2. Also generate
May 11 13:36:17 <CIA-83> images for C3.
May 11 13:36:17 <CIA-83> While we are at it, rename the LEDs to conform to the Linux LED naming
May 11 13:36:18 <CIA-83> conventions.
May 11 13:37:06 <CIA-83> florian * r26877 /trunk/tools/firmware-utils/ (Makefile src/spw303v.c src/xorimage.c):
May 11 13:37:06 <CIA-83> bcm63xx: Add firmware helpers for T-Home Speedport W 303V
May 11 13:37:06 <CIA-83> Add two helpers for generating correctly tagged images for the T-Home
May 11 13:37:06 <CIA-83> Speedport W 303V Typ B as well as xor'd ones as required by CFE.
May 11 13:37:07 <CIA-83> Signed-off-by: Jonas Gorski <jonas.gorski+openwrt@gmail.com>
May 11 13:37:08 <CIA-83> florian * r26878 /trunk/target/linux/brcm63xx/ (5 files in 4 dirs):
May 11 13:37:09 <CIA-83> bcm63xx: Add support for the T-Home Speedport W 303V
May 11 13:37:09 <CIA-83> Add the required board definition, default config and image generation for
May 11 13:37:09 <CIA-83> the T-Home Speedport W 303V Typ B.
May 11 13:37:10 <CIA-83> Signed-off-by: Jonas Gorski <jonas.gorski+openwrt@gmail.com>
May 11 14:31:10 <builder_> build #33 of s3c24xx is complete: Failure [failed compile_10]  Build details are at http://buildbot.openwrt.org:8010/builders/s3c24xx/builds/33
May 11 14:37:31 <KanjiMonster> nbd: unping
May 11 17:23:01 <builder_> build #22 of etrax is complete: Failure [failed compile_3]  Build details are at http://buildbot.openwrt.org:8010/builders/etrax/builds/22
May 11 18:09:10 <builder_> build #21 of ep93xx is complete: Failure [failed compile_4]  Build details are at http://buildbot.openwrt.org:8010/builders/ep93xx/builds/21
May 11 18:12:12 <builder_> build #20 of mpc52xx is complete: Failure [failed compile_4]  Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/20
May 11 20:26:10 <cshore> ping xMff
May 11 20:26:25 <xMff> pong cshore
May 11 20:27:15 <cshore> regarding https://dev.openwrt.org/ticket/9395 it sounds an awful lot like a bug that was fixed ages ago
May 11 20:27:24 <cshore> maybe even in 9.something
May 11 20:28:11 <xMff> cshore: maybe, will keep an eye on it. I had another thing, we should integrate mounting of usbfs into preinit
May 11 20:28:22 <xMff> cshore: and get rid of /e/i/usb
May 11 20:28:37 <cshore> ok
May 11 20:28:58 <cshore> makes sense since we're doing usb mounts in preinit for one thing
May 11 22:14:22 <builder_> build #30 of orion is complete: Failure [failed compile_7]  Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/30
May 11 22:30:12 <^Willie^> evening everyone.
May 11 22:31:30 <^Willie^> [mbm]: i have a ip7160 ubicom router here and ubicom sdk with some build results.. may you can help/advice build looks like going fine upload also. http://pastebin.com/bycMhBjw
May 11 22:32:59 <^Willie^> afther loading the firmware the ethernet interfaces look like stay offline.. gpio leds work but no ping reply nor dhcp.
May 11 23:00:05 <builder_> build #31 of atheros is complete: Failure [failed shell_4]  Build details are at http://buildbot.openwrt.org:8010/builders/atheros/builds/31
May 11 23:07:44 <[mbm]> ^Willie^: you'll probably need to check the console or try to gdb it
May 11 23:08:15 * [mbm] isn't going to dig through a thousand line pastebin showing the entire build
May 11 23:08:44 <^Willie^> [mbm]: jup need to check if it is 3.3 or 5v mosthly 3.3v based then i can probe the serial port if i can't find any data
May 11 23:09:39 <^Willie^> hmm for the build i had one question have you ever used that ubicom sdk software package and modified boardprofiles ?
May 11 23:10:34 <^Willie^> since my dlink firmware builds work fine.. and the ancient openwrt implementation from ubicom is pretty old and half the patches are removed so there was some work left todo before the openwrt part started building without errors.
May 11 23:10:51 <^Willie^> hmm have seen one error left in the last build...
May 11 23:10:54 * [mbm] hasn't built a dir652
May 11 23:11:25 <^Willie^> it is basicly a 855 ;) -> http://pastebin.com/GvSPB4wM
May 11 23:11:35 <^Willie^> or almosth the same.
May 11 23:11:42 <[mbm]> saw the gpl release, looked to be a very old version of openwrt
May 11 23:12:21 <^Willie^> jup it is.. some packages where already removed from upstream ftp mirrors and ubicoms svn server is located at there old scenix domain.. must be old yes
May 11 23:12:23 <[mbm]> heavily forked since the ip7k doesn't have an mmu
May 11 23:12:56 <^Willie^> hmm i think it is ppc40x nommu .. noticed some lines about that in the pre stages of the build.
May 11 23:13:09 <^Willie^> if so then i do wonder why somutch work...
May 11 23:13:40 <^Willie^> that can't be logic in my vision.. but i'll already be happy with this old openwrt build.
May 11 23:14:08 <^Willie^> on the stock dlink firmware the switch chip is going mad afther a few weeks uptime on 1gbit link 15kB/s troughput http..
May 11 23:14:18 <[mbm]> it's not ppc; ubicom actually makes processors; they have their own instruction set
May 11 23:14:18 <^Willie^> from lan client to lan client.
May 11 23:14:34 <^Willie^> hmm cross compiler identified as ppc40x :S
May 11 23:14:40 <^Willie^> later ubicom32..
May 11 23:15:14 <[mbm]> that doesn't sound right
May 11 23:15:31 <[mbm]> could be why your build failed ;)
May 11 23:15:35 <^Willie^> hmm v=99 in konsole..
May 11 23:15:40 <^Willie^> and then scroll..
May 11 23:16:38 <[mbm]> might have inherited some obscure variable from your shell
May 11 23:17:21 <^Willie^> hmm i was already surpriced that my recompiled dlink config worked compiled from a amd64 ;)
May 11 23:17:43 <^Willie^> but that can be a reason for failure also in my xp..
May 11 23:19:28 <^Willie^> also ehm i know it is posible adding a gdb module and remote read the data over lan with this modem and can flush&upload firmware on the fly as long u-boot won't fail i guess.
May 11 23:20:55 <^Willie^> my failed wrt build was pretty easy fixed by powerup with the reset button 10 sec on hold then goto http:192.168.0.1 and the webpage ask upload firmware directly.. not tested if the firmware is really flushed at that stage by a extra reboot..
May 11 23:21:35 <^Willie^> so if i can help i'll be happy
May 11 23:22:52 <^Willie^> ahh i think i have the reason of failure found... wrong target arch
May 11 23:23:30 <^Willie^> Collecting target info: doneing...ux/ppc40xs4
May 11 23:23:32 <^Willie^> Warning! Found recursive dependency: (null) TARGET_ubicom32 DEVEL (null)
May 11 23:24:14 <^Willie^> [mbm]: any idea if there is any defaults for the rs-232 pins ? iirc same as on the dev boards i hope...
May 11 23:24:15 <xMff> that has nothing to do with the toolchain
May 11 23:24:33 <xMff> ppc40x just happens to be one of openwrts supported targets
May 11 23:24:44 <xMff> ubicom didn't strip that from their copy
May 11 23:24:47 <^Willie^> that was the last few lines of make menuconfig(openwrt)
May 11 23:25:13 <^Willie^> moment check my ubicom firm.. made with the same config script..
May 11 23:31:48 <^Willie^> hmm o must have done something wrong at the menuconfig or must do extra checking..
May 11 23:32:38 <xMff> ^Willie^: I think we can conclude here that the stuff that ubicom ships is quite ancient and apparently you do need the dongle to properly debug the board
May 11 23:32:49 <^Willie^> logs from dlink build all CROSS_COMPILE="ubicom32-elf-" ARCH="ubicom32" and CC="ubicom32-elf-gcc"
May 11 23:33:03 <xMff> ^Willie^: ... so I am not sure if you will get anywhere simply by test-flashing the generated openwrt iamge
May 11 23:33:34 <^Willie^> xMff: indeed but i can debug..
May 11 23:33:34 <xMff> ^Willie^: and again, you're chasing a ghost with the ppc4xx thing, it just was enumerated when openwrt collected its target info
May 11 23:34:47 <^Willie^> jup i guess so.. but will clean the build and start from the bgin tomorrow.. afther fixing the serial port that dumps output by default and posible even trying including there gdb module so that i can use gdb over ip..
May 11 23:35:01 <^Willie^> and start with a distclean and then redo all steps i have done ..
May 11 23:35:13 <^Willie^> hmm rm -rf and untar..
May 11 23:35:19 <^Willie^> that way restart.
May 11 23:38:20 <^Willie^> http://developer.ubicom.com/wiki/index.php/Debugging
May 11 23:40:03 <^Willie^> will first look how far i can come with serial access. befor start thinking about further tweaking.
May 11 23:40:12 <xMff> yes
May 11 23:41:59 <^Willie^> those pl2303 usb-232 cords will do as 3.3v serial.. will lookup that info .. else i can start shopping
May 11 23:44:26 <^Willie^> xMff: are the openwrt builds from svn mutch like this sdk build ?
May 11 23:44:40 <xMff> they work similar
May 11 23:44:55 <xMff> but I do not know what d-link / ubicom did change
May 11 23:46:13 <^Willie^> hmm indeed but if i can manage this then posible i can also try the current openwrt trunk..
May 11 23:46:38 <^Willie^> as long they did not change the kernel drivers in a weird way.
May 11 23:47:29 <^Willie^> anyway it is almosth 2:00 am again here in holland ..
May 11 23:48:03 * xMff is in the same tz
May 11 23:48:19 <^Willie^> hmm also a night type :p
May 11 23:48:43 <xMff> you have to keep in mind that this ubicom32 architecture is entirely new
May 11 23:48:48 <xMff> its not just some new arm or mips flavor
May 11 23:49:06 <xMff> it needs patches in binutils, patches in gcc, patches in uclibc, patches in the kernel
May 11 23:49:52 <xMff> its a very hard target for somebody who never did anything like that
May 11 23:50:20 <^Willie^> hmm indeed it is but my broadcom boards fail easyer then this ..
May 11 23:50:33 <^Willie^> if it is about bricking and noway back or forwards
May 11 23:53:02 <^Willie^> xMff: also oxford based or pogoplug clone is not easy but pogoplug and running glibc based arm or mips(el) is more like regulair linux setups on x86 indeed.
May 11 23:53:51 <^Willie^> hmm dovecot or clamav back in 2005 was not jet availible for mipsel 32bit but manual build worked straight forward without any patching..
May 11 23:54:06 <^Willie^> guess that was just luck ..
**** ENDING LOGGING AT Thu May 12 02:59:58 2011